17 - PART 10 Disorders of the Gastrointestinal System

01 - SECTION 1 Disorders of the Alimentary Tract
SECTION 1 Disorders of the Alimentary Tract
Disorders of the Gastrointestinal System
PART 10
Section 1	 Disorders of the Alimentary
Tract
William L. Hasler, Chung Owyang
Approach to the Patient
with Gastrointestinal
Disease
ANATOMIC CONSIDERATIONS
The gastrointestinal (GI) tract extends from the mouth to the anus
and is composed of organs with distinct functions. Sphincters assist
in gut compartmentalization to separate the organs. Distinct gut wall
layers contribute to regional activities. The mucosa regulates fluid
and nutrient transfer but is also a barrier to prevent entry of some
luminal contents. Smooth muscle and enteric nerves mediate pro­
pulsion. Many organs possess a serosal layer to provide a supportive
foundation and permit external input. The gut interacts with many
other systems. Pancreaticobiliary conduits deliver bile and enzymes
into the duodenum. Lymphatic channels assist in gut immune
activities. Intrinsic nerves control propulsion and fluid regulation.
Extrinsic neural input provides volitional or involuntary control
specific for each gut region.
FUNCTIONS OF THE GI TRACT
The GI tract serves two main functions—assimilating nutrients and
eliminating waste. In the mouth, food is processed, mixed with salivary
amylase, and delivered to the pharynx. Aborally propagating esopha­
geal contractions propel the ingested bolus into the stomach in coor­
dination with relaxation of the upper and lower esophageal sphincters.
Basal tone in the lower esophageal sphincter prevents gastroesophageal
reflux.
The stomach triturates and mixes food with pepsin and acid. The
proximal stomach serves a storage function by relaxing to accommo­
date the meal. Phasic contractions in the distal stomach propel food
residue against the pylorus, where it is ground and thrust proximally
for further mixing prior to emptying into the duodenum.
Most nutrient absorption occurs in the small intestine. Mucosal
villi provide maximal surface area and possess specialized enzymes
and transporters. Triturated food from the stomach mixes with
pancreatic secretions and bile in the duodenum. Pancreatic juice
contains both enzymes for digestion and bicarbonate to optimize
luminal pH for enzyme activation. Bile secreted by the liver and
stored in the gallbladder is needed for lipid digestion. Most nutri­
ents and minerals are rapidly absorbed in the proximal intestine,
while bile acids and vitamin B12 are absorbed in the ileum. Bile
contains byproducts of erythrocyte degradation, toxins, medica­
tions, and cholesterol that are emptied into the colon along with
indigestible residue. The ileocecal junction is a sphincter that limits
coloileal reflux of colonic microbes.
The colon prepares waste for evacuation by to-and-fro contractions
that dehydrate the stool. As a consequence, daily luminal volumes
decrease from 1000–1500 mL in the ileum to 100–200 mL that are
expelled from the rectum. Esophageal transit takes seconds, stomach
and small intestine emptying times range from minutes to a few hours,
but colon propagation requires >1 day in most individuals. A dense
colonic microbiome ferments undigested carbohydrates and shortchain fatty acids and modulates immune activity. The colon terminates
in the anus, which possesses volitional and involuntary controls to
retain stool in the rectum until it can be released in a convenient
setting.
EXTRINSIC MODULATION OF GUT
FUNCTION
GI function is modified by influences outside the gut. Unlike other
systems, the gut is in continuity with the outside environment. Several
mechanisms protect against injury from foods, medications, toxins,
and microbes. Mucosal immune pathways include epithelial and
lamina propria lymphocytes and plasma cells and lymph node chains
to prevent noxious agents from entering the circulation. Antimicrobial
peptides secreted by Paneth cells defend against pathogens. Drugs
and toxins absorbed into the portal venous circulation are filtered
and detoxified in the liver. Many GI reflexes involve extrinsic vagus
or splanchnic nerve pathways. The brain-gut axis alters function in
gut regions not under volitional regulation. Stress disrupts normal gut
motor, secretory, and sensory activities.
OVERVIEW OF GI DISEASES
GI diseases range in severity from conditions that produce mild symp­
toms and no long-term morbidity to those with intractable symptoms
or adverse outcomes. Diseases may be localized to one gut organ or
exhibit diffuse involvement at many sites.
■
■CLASSIFICATION OF GI DISEASES
GI diseases are manifestations of alterations in nutrient and fluid
assimilation, transit and sensation, immune function, and vascular
supply. Neoplastic degeneration may affect any GI site. Genetic factors
underlie some disorders.
Impaired Digestion and Absorption 
Diseases of the stomach,
small intestine, biliary tree, and pancreas can disrupt digestion and
absorption. The most common maldigestion syndrome, lactase defi­
ciency, produces gas and diarrhea after ingesting dairy products but
has no adverse outcomes. Intestinal sucrase-isomaltase deficiency
produces similar symptoms after consuming sucrose. Disorders that
affect digestion and/or absorption more diffusely, including celiac
disease, bacterial overgrowth, infectious enteritis, Crohn’s ileitis, and
radiation enteritis produce anemia, dehydration, electrolyte disorders,
or malnutrition. Gastric acid hypersecretory conditions impair diges­
tion by damaging the intestinal mucosa, impairing pancreatic enzyme
activation, and accelerating transit. Benign or neoplastic biliary
obstruction impairs fat digestion. Impaired pancreatic enzyme release
in chronic pancreatitis or pancreatic cancer impairs digestion and leads
to malnutrition.
Altered Secretion 
Some GI diseases result from alteration of gut
secretion. Gastric acid hypersecretion occurs in gastrinoma, G-cell
hyperplasia, retained antrum, and some patients with duodenal ulcers.
Gastric acid is reduced in atrophic gastritis and pernicious anemia.
Intestinal or colonic hypersecretion and/or impaired absorption causes
diarrhea with acute bacterial or viral infection, chronic Giardia or
cryptosporidia infections, small-intestinal bacterial overgrowth, bile
salt diarrhea, microscopic colitis, diabetic diarrhea, and endocrine
neoplasias with tumor overproduction of secretagogue transmitters
such as vasoactive intestinal polypeptide.
Altered Gut Transit 
Mechanical obstruction can block transit
of gut contents. Esophageal occlusion most often is due to stricture
(from acid exposure or eosinophilic esophagitis) or neoplasm. Gastric
obstruction develops from ulcer disease or cancer. Small-intestinal
obstruction commonly results from adhesions but also occurs with
Crohn’s disease, radiation- or drug-induced strictures, and malig­
nancy. The most common cause of colonic obstruction is colon cancer,
although inflammatory strictures develop with inflammatory bowel
disease (IBD), after diverticulitis, or with some medications.
Delayed propulsion can develop in any gut region. Achalasia is
characterized by impaired esophageal body peristalsis and incomplete
lower esophageal sphincter relaxation. Gastroparesis is the delay in
gastric emptying resulting from impaired antral motility or spastic

03 - 333 Gastrointestinal Endoscopy
333 Gastrointestinal Endoscopy
for nausea in first-trimester pregnancy. Peppermint oil and caraway
seed oil products and herbal preparations such as STW 5 (a nineherb mixture) are useful for functional dyspepsia and IBS. Lowpotency pancreatic enzyme preparations are sold as digestive aids
but have little evidence to support their efficacy.
THERAPIES TARGETING GUT DYSBIOSIS
Antibiotics are prescribed to treat H. pylori–induced ulcers, infec­
tious diarrhea, complicated diverticulitis, and small intestinal
bacterial overgrowth. Some cases of diarrhea-predominant IBS
respond to the nonabsorbable antibiotic rifaximin. Probiotics con­
taining bacterial cultures and prebiotics that selectively nourish
nonnoxious commensal bacteria have been given as adjunctive
therapy for infectious diarrhea and IBS, with limited evidence of
efficacy. Postbiotics are metabolites made by probiotic organisms
that inhibit pathogenic luminal bacteria. Transplanting donor feces
into the colon by colonoscopy or enema is effective treatment for
recurrent Clostridioides difficile colitis. Commercial oral and rectal
microbiota-based products have been approved or are in develop­
ment for this indication.
THERAPEUTIC ENDOSCOPY
In addition to its diagnostic role, endoscopy has numerous thera­
peutic capabilities. Cautery techniques and injection of vasocon­
strictor substances can stop hemorrhage from ulcers and vascular
malformations. Endoscopically placed clips can occlude arterial
bleeding sites, while hemostatic powder sprays can stop brisk per­
sistent GI bleeding. Endoscopic encirclement of esophageal varices
and hemorrhoids with constricting bands stops hemorrhage from
these sites. Bleeding gastric varices can be injected with thrombin or
cyanoacrylate. Endoscopy can remove polyps in the stomach, small
bowel, or colon. Decompressive colonoscopy withdraws excess
gas in some cases of acute colonic pseudoobstruction. Endoscopic
mucosal resection, submucosal dissection, and radiofrequency
techniques ablate some cases of Barrett’s esophagus with dysplasia
and resect superficial cancers or subepithelial tumors elsewhere in
the gut. Obstructions of the GI lumen and pancreaticobiliary tree
are relieved by endoscopic dilation or placing plastic or expandable
metal stents. Endoscopic sphincterotomy of the ampulla of Vater
treats choledocholithiasis. Cholangioscopy can facilitate stone lith­
otripsy in the common bile duct, ablation of small ductal tumors,
and placement of gallbladder stents to facilitate drainage in nonop­
erative candidates. Interventional EUS is used for pancreatic cyst
gastrostomy using lumen-apposing metal stents, pancreatic necro­
sectomy, and placement of fiducial markers to direct pancreatic and
rectal radiotherapy. EUS also can facilitate endoscopic access to the
excluded distal stomach in patients who have undergone bariatric
gastric bypass surgery using similar stents so that ERCP can be
done for pancreaticobiliary conditions. EUS-directed stent place­
ment can manage postsurgical stenoses after pancreatic resection.
Endoscopy is used to insert gastric feeding tubes. Peroral endo­
scopic myotomy is performed on the lower esophageal sphincter
in achalasia, the pylorus in gastroparesis, and the upper esophageal
sphincter for Zenker’s diverticulum. Endoscopic treatments for acid
reflux including transoral incisionless fundoplication have been
developed as potential alternatives to surgery. Endoscopic bariatric
methods, including intragastric balloons, sleeve gastroplasty, and
duodenal resurfacing and diversion, have been devised.
PART 10
Disorders of the Gastrointestinal System
INTERVENTIONAL RADIOLOGY
Interventional radiology techniques offer benefits in selected set­
tings. Angiographic embolization or vasoconstriction decreases
bleeding from gut sites not amenable to endoscopic intervention.
Angiographic embolectomy, stent placement, and thrombolysis
also manage mesenteric ischemia. Dilatation or stenting under
fluoroscopic guidance relieves luminal strictures. Contrast enemas
can reduce colon volvulus. CT- and ultrasound-directed drainage of
abdominal fluid collections can obviate the need for surgery. Per­
cutaneous transhepatic cholangiography relieves biliary obstruction
when ERCP is contraindicated. Percutaneous cholecystostomy treats
acute cholecystitis in patients unable to undergo cholecystectomy.
Transjugular intrahepatic portosystemic shunts are performed for
variceal hemorrhage not amenable to endoscopic therapy. Litho­
tripsy is rarely performed to fragment gallstones in patients who
are not surgical candidates. Radiologic approaches are often chosen
over endoscopy for gastroenterostomy placement. Radiographic
assistance is sometimes needed for placement of central venous
catheters for parenteral nutrition.
SURGERY
Roles of surgery in GI conditions include disease cure, symptom
control, maintenance of nutrition, and palliation of unresectable
neoplasm. Surgery cures medication-unresponsive ulcerative coli­
tis, diverticulitis, cholecystitis, appendicitis, and intraabdominal
abscess, but only reduces symptoms and treats complications in
Crohn’s disease. Surgery is performed for ulcer complications like
bleeding, obstruction, or perforation and intestinal obstructions
that persist after conservative care. Gastroesophageal fundoplica­
tion is performed for refractory acid reflux. Acid exposure time
on pH testing helps select candidates for fundoplication. Achalasia
responds to operations to reduce lower esophageal sphincter tone.
Operations for motor disorders include implanted electrical stimu­
lators for gastroparesis and electrical devices and artificial sphinc­
ters for fecal incontinence. Surgery can place a jejunostomy for
long-term enteral feedings. Other common indications for surgery
include hernias, hemorrhoids, and nonhealing anal fissures.
PSYCHOLOGICAL APPROACHES AND PHYSICAL THERAPY
Psychological therapies, including psychotherapy, cognitive behav­
ioral therapy, and hypnosis, show efficacy in DGBIs and are most
beneficial for patients with significant psychological dysfunction.
Behavioral therapists provide instruction in diaphragmatic breath­
ing for belching or rumination. Biofeedback methods administered
by physical therapists can treat refractory fecal incontinence or
constipation secondary to dyssynergia.
■
■FURTHER READING
Benech N et al: Update on microbiota-derived therapies for recurrent
Clostridioides difficile infections. Clin Microbiol Infect 30:462, 2024.
Gergely M et al: Management of refractory inflammatory bowel
disease. Curr Opin Gastroenterol 38:347, 2022.
Hossain B et al: Prevalence and impact of gastrointestinal manifesta­
tions in COVID-19 patients: A systematic review. J Community Hosp
Intern Med Perspect 13:39, 2023.
Jain S et al: Optimal strategies for colorectal cancer screening. Curr
Treat Options Oncol 23:474, 2022.
Orpen-Palmer J et al: Update on the management of upper gastroin­
testinal bleeding. BMJ Med 1:e000202, 2022.
Shakir SM et al: Updates to the diagnosis and clinical management of
Helicobacter pylori infections. Clin Chem 69:869, 2023.
Louis Michel Wong Kee Song, Vinay
Chandrasekhara, Mark Topazian
Gastrointestinal
Endoscopy
Gastrointestinal endoscopy has been attempted for over 200 years,
but the introduction of semirigid and flexible gastroscopes in the
mid-twentieth century marked the dawn of the modern endoscopic
era. Since then, rapid advances in endoscopic technology have led to
FIGURE 333-1  Gastrointestinal endoscope. Shown here is a conventional
colonoscope with control knobs for tip deflection, push buttons for suction and
air insufflation (single arrows), and a working channel for passage of accessories
(double arrows).
dramatic changes in the diagnosis and treatment of many digestive
diseases. Innovative endoscopic devices and new endoscopic treatment
modalities continue to expand the use of endoscopy in patient care.
Flexible endoscopes provide an electronic video image generated
by a charge-coupled device (CCD) or a complementary metal oxide
semiconductor (CMOS) chip in the tip of the endoscope. Operator
controls permit deflection of the endoscope tip; fiberoptic bundles
or light-emitting diodes provide light at the tip of the endoscope; and
working channels allow washing, suctioning, and the passage of instru­
ments (Fig. 333-1). Progressive changes in the diameter and stiffness
of endoscopes have improved the ease and patient tolerance of endos­
copy. High-resolution and high-definition endoscopes equipped with
electronic and optical magnification capabilities enable acquisition
of images with a high level of detail. Advanced imaging techniques,
including narrow-band imaging (Fig. 333-2) and real-time imageprocessing enhancement algorithms, aid in tissue characterization or
differentiation.
ENDOSCOPIC PROCEDURES
■
■UPPER ENDOSCOPY
Upper gastrointestinal endoscopy, also referred to as esophagogastro­
duodenoscopy (EGD), is performed by passing a flexible endoscope
through the mouth into the esophagus, stomach, and duodenum. The
procedure is the best method for examining the upper gastrointestinal
mucosa (Fig. 333-3). While the upper gastrointestinal radiographic
series has similar accuracy for diagnosis of duodenal ulcer (Fig. 333-4),
EGD is superior for detection of gastric ulcers (Fig. 333-5) and flat
mucosal lesions, such as Barrett’s esophagus (Fig. 333-6), and it per­
mits directed biopsy and endoscopic therapy. Intravenous sedation
is given to most patients in the United States to ease the anxiety and
discomfort of the procedure, although in many countries, EGD is rou­
tinely performed with topical pharyngeal anesthesia only. Patient toler­
ance of unsedated EGD is improved by the use of an ultrathin, 5-mm
diameter endoscope that can be passed transorally or transnasally.
■
■COLONOSCOPY
Colonoscopy is performed by passing a flexible colonoscope through
the anal canal into the rectum and colon. The cecum is reached in
95% of cases, and the terminal ileum (Fig. 333-7) can usually be
examined. Colonoscopy is the gold standard for imaging the colonic
mucosa (Fig. 333-8). Colonoscopy has greater sensitivity than barium
enema for the detection of colitis (Fig. 333-9), colon polyps (Fig. 333-10),
and colorectal cancer (Fig. 333-11). Computed tomography (CT) colo­
nography rivals the accuracy of colonoscopy for the detection of some
A
CHAPTER 333
Gastrointestinal Endoscopy
B
FIGURE 333-2  Flat colon polyp. A. White-light imaging. B. Corresponding narrowband imaging enhances mucosal features and lesion delineation.
polyps and cancer, although it is not as sensitive for the detection of flat
lesions, such as serrated polyps (Fig. 333-12). Intravenous sedation is
usually given before colonoscopy in the United States, although a will­
ing patient and a skilled examiner can complete the procedure without
sedation in many cases.
■
■FLEXIBLE SIGMOIDOSCOPY
Flexible sigmoidoscopy is akin to colonoscopy, but it visualizes only the
rectum and a variable portion of the left colon, typically to 60 cm from
the anal verge. This procedure may result in abdominal cramping and
discomfort, but it is brief and thus can be performed without sedation.
Flexible sigmoidoscopy is primarily used for evaluation of diarrhea and
rectal outlet bleeding.
■
■SMALL-BOWEL ENDOSCOPY
Three endoscopic techniques are currently used to evaluate the small
intestine, most often in patients presenting with presumed smallbowel bleeding. For capsule endoscopy, the patient swallows a dispos­
able capsule that contains a CMOS chip camera. Color still images
(Fig. 333-13) are transmitted wirelessly to an external receiver at sev­
eral frames per second until the capsule’s battery is exhausted or it is
passed into the toilet. Capsule endoscopy enables visualization of the
small-bowel mucosa beyond the reach of a conventional endoscope,
A
PART 10
Disorders of the Gastrointestinal System
C
E
FIGURE 333-3  Normal upper endoscopic examination. A. Esophagus. B. Gastroesophageal junction. C. Gastric fundus. D. Gastric body. E. Gastric antrum. F. Pylorus.
G. Duodenal bulb. H. Second portion of the duodenum.
B
D
F
G
FIGURE 333-3  (Continued)
A
FIGURE 333-4  Duodenal ulcers. A. Ulcer with a small, flat, pigmented spot in its base. B. Ulcer with a visible vessel (arrow) in a patient with recent hemorrhage.
A
FIGURE 333-5  Gastric ulcers. A. Benign gastric ulcer in the antrum. B. Malignant gastric ulcer involving greater curvature of stomach.
H
CHAPTER 333
Gastrointestinal Endoscopy
B
B
A
C
FIGURE 333-6  Barrett’s esophagus. A. Salmon-colored Barrett’s mucosa extending proximally from the gastroesophageal junction. B. Barrett’s esophagus with a suspicious
nodule (arrow) identified during endoscopic surveillance. C. Histologic finding of intramucosal adenocarcinoma in the endoscopically resected nodule. Tumor extends into
the esophageal submucosa (arrow). D. Barrett’s esophagus with locally advanced adenocarcinoma.
PART 10
Disorders of the Gastrointestinal System
and at present, it is solely a diagnostic procedure. Patients with a
history of prior intestinal surgery or Crohn’s disease are at risk for
capsule retention at the site of a clinically unsuspected small-bowel
stricture, and ingestion of a “patency capsule” composed of radiologi­
cally opaque biodegradable material may be indicated before capsule
endoscopy in such patients.
Push enteroscopy is generally performed using a variable-stiffness
pediatric or adult colonoscope, or a dedicated enteroscope with or
without the assistance of a stiffening overtube that extends from the
mouth to the small intestine. The proximal to mid-jejunum is usu­
ally reached, and the instrument channel of the endoscope allows for
biopsy or endoscopic therapy.
A
FIGURE 333-7  Colonoscopic view of terminal ileum. A. Normal-appearing terminal ileum (TI). B. View of normal villi of TI enhanced by examination under water immersion.
B
D
Deeper intubation of the small bowel can be accomplished by single-
or double-balloon enteroscopy, which enables pleating of the small
intestine onto an overtube (Fig. 333-14, Video V5-1). With balloonassisted enteroscopy, the entire small intestine can be visualized in
some patients when both the oral and anal routes of insertion are used.
Biopsies and endoscopic therapy, such as thermal ablation of vascular
ectasias and polypectomy, can be performed throughout the visualized
small bowel (Fig. 333-15).
■
■ENDOSCOPIC RETROGRADE
CHOLANGIOPANCREATOGRAPHY
During endoscopic retrograde cholangiopancreatography (ERCP), a
side-viewing endoscope is passed through the mouth to the duodenum,
B
A
C
FIGURE 333-8  Normal colonoscopic examination. A. Cecum with view of appendiceal orifice. B. Ileocecal valve. C. Normal-appearing colon. D. Rectum (retroflexed view).
the bile duct and/or pancreatic duct is cannulated with a thin plastic
catheter, and radiographic contrast material is injected under fluo­
roscopic guidance (Fig. 333-16). When indicated, the major papilla
can be incised using the technique of endoscopic sphincterotomy
(Fig. 333-17). Stones can be retrieved from the ducts, biopsies can be
performed, strictures can be dilated and/or stented (Fig. 333-18), and
ductal leaks can be treated (Fig. 333-19). ERCP is usually performed
for therapy but is also important diagnostically as it facilitates tissue
sampling of biliary or pancreatic ductal strictures.
■
■ENDOSCOPIC ULTRASOUND
Endoscopic ultrasound (EUS) utilizes ultrasound transducers incor­
porated into the tip of a flexible endoscope. Ultrasound images are
obtained of the gut wall and adjacent organs, vessels, lymph nodes, and
other structures. High-resolution images are obtained by bringing a
high-frequency ultrasound transducer close to the area of interest via
endoscopy. EUS provides the most accurate preoperative local stag­
ing of esophageal, pancreatic, and rectal malignancies (Fig. 333-20),
but it does not detect distant metastases that are beyond its imaging
range. EUS is also useful for diagnosis of bile duct stones, gallbladder
disease, subepithelial gastrointestinal lesions, and chronic pancreatitis.
Fine-needle aspirates and core biopsies of organs, masses, and lymph
nodes in the posterior mediastinum, abdomen, retroperitoneum, and
B
CHAPTER 333
Gastrointestinal Endoscopy
D
pelvis can be obtained under EUS guidance (Fig. 333-21). EUS-guided
therapeutic procedures are increasingly performed, including drainage
of abscesses, pseudocysts, and pancreatic necrosis into the gut lumen
(Video V5-2); celiac plexus neurolysis for treatment of pancreatic
pain; ethanol ablation of pancreatic neuroendocrine tumors; treatment
of gastric and intestinal varices; biliary or pancreatic drainage; and
endoscopic gastrojejunostomy for palliation of malignant gastric outlet
obstruction (Video V5-3).
■
■NATURAL ORIFICE TRANSLUMINAL
ENDOSCOPIC SURGERY
Natural orifice transluminal endoscopic surgery (NOTES) represents a
collection of endoscopic methods that entail passage of an endoscope
or its accessories into or through the wall of the gastrointestinal tract
to perform diagnostic or therapeutic interventions. Some NOTES
procedures, such as percutaneous endoscopic gastrostomy (PEG) or
endoscopic necrosectomy of pancreatic necrosis, are well-established
clinical procedures (Video V5-2); others such as peroral endoscopic
myotomy (POEM) for achalasia (Fig. 333-22) and gastroparesis,
peroral endoscopic tumorectomy (POET) (Fig. 333-23), and endo­
scopic full-thickness resection (EFTR) of gastrointestinal mural lesions
(Fig. 333-24, Video V5-4), are newer minimally invasive therapeutic
options.
A
PART 10
Disorders of the Gastrointestinal System
C
FIGURE 333-9  Causes of colitis. A. Chronic ulcerative colitis with diffuse inflammation. B. Severe Crohn’s colitis with deep ulcers. C. Pseudomembranous colitis with yellow,
adherent pseudomembranes. D. Ischemic colitis with patchy mucosal edema, subepithelial hemorrhage, superficial ulcerations, and cyanosis.
A
FIGURE 333-10  Colonic polyps. A. Pedunculated polyp on a stalk. B. Sessile polyp.
B
D
B
FIGURE 333-11  Ulcerated colon adenocarcinoma narrowing the colonic lumen.
A
B
C
FIGURE 333-12  Flat serrated polyp in the cecum. A. Appearance of the lesion under
conventional white-light imaging. B. Mucosal patterns and boundary of the lesion
enhanced with narrow-band imaging. C. Submucosal lifting of the lesion with dye
(methylene blue) injection before resection.
FIGURE 333-13  Capsule endoscopy. Image of a jejunal vascular ectasia.
■
■ENDOSCOPIC RESECTION AND CLOSURE
TECHNIQUES
Endoscopic mucosal resection (EMR) (Fig. 333-25, Video V5-5) and
endoscopic submucosal dissection (ESD) (Fig. 333-26, Video V5-6)
are the two commonly used techniques for the resection of benign and
early-stage malignant gastrointestinal neoplasms. In addition to pro­
viding larger specimens for more accurate histopathologic assessment
and diagnosis, these techniques may be curative for some dysplastic
lesions and superficial carcinomas involving the esophagus, stomach,
and colon.
CHAPTER 333
Gastrointestinal Endoscopy
Several devices are available for closure of mucosal defects created
by EMR and ESD, as well as gastrointestinal fistulas and perfora­
tions. Endoscopic clips deployed through the working channel of an
FIGURE 333-14  Double-balloon enteroscopy. Radiograph of the orally inserted
instrument deep in the small intestine.
A
PART 10
Disorders of the Gastrointestinal System
B
C
FIGURE 333-15  Nonsteroidal anti-inflammatory drug (NSAID)–induced proximal
ileal stricture managed via double-balloon enteroscopy. A. High-grade ileal
stricture causing obstructive symptoms. B. Balloon dilation of the ileal stricture.
C. Appearance of the stricture after dilation.
A
B
FIGURE 333-16  Endoscopic retrograde cholangiopancreatography (ERCP) for bile
duct stones. A. Faceted bile duct stones are demonstrated in the common bile duct
and common hepatic duct. B. After endoscopic sphincterotomy, the stones are
extracted with a stone extraction balloon.
A
C
FIGURE 333-17  Endoscopic sphincterotomy. A. A normal-appearing ampulla of Vater (arrow). B. Biliary endoscopic sphincterotomy is performed with electrosurgery. C. Bile
duct stones are extracted with a balloon catheter.
endoscope have been used for many years to treat bleeding lesions,
and the development of larger over-the-scope clips has facilitated
endoscopic closure of gastrointestinal fistulas and perforations not
previously amenable to endoscopic therapy (Video V5-7). Endoscopic
suturing can be used to close some perforations and large defects (Fig.
333-27), anastomotic leaks, and fistulas. Endoscopic suturing may also
be used to prevent stent migration (Fig. 333-28, Video V5-8) and to
perform endoscopic bariatric procedures. These technologies are play­
ing an expanding role in patient care.
RISKS OF ENDOSCOPY
Medications used during sedation may cause respiratory depression or
allergic reactions. All endoscopic procedures carry some risk of bleed­
ing and gastrointestinal perforation. The risk is small with diagnostic
B
CHAPTER 333
Gastrointestinal Endoscopy
upper endoscopy, flexible sigmoidoscopy, and colonoscopy (<1:1000
procedures), but it ranges from 0.5 to 5% when therapeutic maneuvers
such as polypectomy, EMR, ESD, control of hemorrhage, or stricture
dilation are performed. The risk of adverse events for diagnostic EUS
(without needle aspiration) is similar to that for diagnostic upper
endoscopy.
Infectious complications are uncommon with most endoscopic
procedures. Some procedures carry a higher incidence of postproce­
dural bacteremia, and prophylactic antibiotics may be indicated (Table
333-1). Management of antithrombotic agents before endoscopic pro­
cedures should take into account the procedural risk of hemorrhage,
the agent, and the patient condition, as summarized in Table 333-2.
ERCP carries additional risks. Pancreatitis occurs in ~5% of patients
undergoing the procedure, and young, anicteric patients with normal
A
PART 10
Disorders of the Gastrointestinal System
C
FIGURE 333-18  Endoscopic diagnosis, staging, and palliation of hilar cholangiocarcinoma. A. Endoscopic retrograde cholangiopancreatography (ERCP) in a patient with
obstructive jaundice demonstrates a malignant-appearing stricture of the biliary confluence extending into the left and right intrahepatic ducts. B. Per oral cholangioscopy
demonstrating a stricture with dilated, tortuous vessels with a malignant appearance. C. Intraductal biopsy obtained during ERCP demonstrates malignant cells infiltrating
the submucosa of the bile duct wall (arrow). (Image courtesy of Dr. Thomas Smyrk.) D. Endoscopic placement of multiple plastic stents draining the right anterior, right
posterior, and left systems relieves the biliary obstruction.
ducts are at increased risk (up to 25%). Post-ERCP pancreatitis is
usually mild and self-limited, but it may result in prolonged hospital­
ization, surgery, diabetes, or death when severe. Significant bleeding
occurs after endoscopic sphincterotomy in ~1% of cases. Ascending
cholangitis, pseudocyst infection, duodenal perforation, and abscess
formation may occur as a result of ERCP.
PEG tube placement during EGD is associated with a 10–15% inci­
dence of adverse events, most often wound infections. Fasciitis, pneu­
monia, bleeding (Fig. 333-29), buried bumper syndrome (Fig.
333-30), and colonic injury may result from PEG tube placement.
B
D
URGENT ENDOSCOPY
■
■ACUTE GASTROINTESTINAL HEMORRHAGE
Endoscopy is the primary diagnostic and therapeutic procedure for
patients with acute gastrointestinal hemorrhage. Although gastroin­
testinal bleeding stops spontaneously in most cases, some patients will
have persistent or recurrent hemorrhage that may be life-threatening.
Clinical predictors of rebleeding help identify patients most likely to
benefit from urgent endoscopy and endoscopic, angiographic, or surgi­
cal hemostasis.
A
FIGURE 333-19  Bile leak. A. Site of leak (arrow) from the cystic duct after laparoscopic cholecystectomy. B. Contrast leaks from the cystic duct stump across surgical clips
into the gallbladder fossa (arrow).
A
FIGURE 333-20  Local staging of gastrointestinal cancers with endoscopic ultrasound. In each example, the arrowhead marks the primary tumor and the arrow indicates
the muscularis propria (MP) of the intestinal wall. A. T2 gastric cancer. The tumor invades the MP. B. Submucosal gastric tumor. The tumor is confined to the submucosal
space without invasion into the MP.
A
FIGURE 333-21  Endoscopic ultrasound (EUS)–guided tissue sampling. A. Ultrasound image of a 22-gauge needle (arrow) passed through the gastric wall and positioned in
a hypoechoic pancreatic neck mass. B. Touch preparation demonstrating aspirated malignant cells.
B
CHAPTER 333
Gastrointestinal Endoscopy
B
B
A
B
D
E
PART 10
Disorders of the Gastrointestinal System
H
G
FIGURE 333-22  Peroral endoscopic myotomy (POEM) for achalasia. A. Dilated aperistaltic esophagus with retained secretions. B. Hypertonic lower esophageal sphincter
(LES) region. C. Mucosal incision (mucosotomy) 10 cm proximal to the LES. D. Submucosal dissection using an electrosurgical knife following endoscope entry through the
mucosotomy site into the submucosal space. E. Completion of submucosal tunnel to the cardia. F. Initiation of myotomy of the muscularis propria distal to the mucosotomy
site. G. Completion of myotomy to the cardia. H. Closure of mucosotomy site with clips. I. Patulous gastroesophageal junction following myotomy.
Initial Evaluation 
The initial evaluation of the bleeding patient
focuses on the severity of hemorrhage as reflected by the presence of
supine hypotension or tachycardia, postural vital sign changes, and
the frequency of hematemesis or melena. Decreases in hematocrit and
hemoglobin lag behind the clinical course and are not reliable gauges
of the magnitude of acute bleeding. Nasogastric tube aspiration and
lavage can also be used to judge the severity of bleeding, but these are
no longer routinely performed for this purpose. The bedside initial
evaluation, completed well before the bleeding source is confidently
identified, guides immediate supportive care of the patient; triage to
outpatient follow-up, a hospital ward, or an intensive care unit; and
timing of endoscopy. The severity of the initial hemorrhage is the most
important indication for urgent endoscopy, since a large initial bleed
increases the likelihood of ongoing or recurrent bleeding. Patients
with resting hypotension or orthostatic change in vital signs, repeated
hematemesis, bloody nasogastric aspirate that does not clear with
large-volume lavage, or those requiring blood transfusions should be
considered for urgent endoscopy within 12–24 h of presentation. In
addition, patients with cirrhosis, coagulopathy, or respiratory or renal
failure and those >70 years old are more likely to have significant
rebleeding and to benefit from prompt evaluation and treatment.
C
F
I
Bedside evaluation also suggests an upper or lower gastrointestinal
source of bleeding in most patients. Over 90% of patients with melena
are bleeding proximal to the ligament of Treitz, and ~85% of patients
with hematochezia are bleeding from the colon. Melena can result from
bleeding in the small bowel or right colon, especially in older patients
with slow colonic transit. Conversely, some patients with massive
hematochezia may be bleeding from an upper gastrointestinal source,
with rapid intestinal transit. An urgent upper endoscopy should be
considered in such patients.
Endoscopy should be performed after the patient has been resus­
citated with intravenous fluids and transfusions, as necessary. Marked
coagulopathy or thrombocytopenia is usually treated before endoscopy,
since correction of these abnormalities may lead to resolution of bleed­
ing, and techniques for endoscopic hemostasis are limited in such
patients. Metabolic derangements should also be addressed. Tracheal
intubation for airway protection should be considered before upper
endoscopy in patients with repeated recent hematemesis, particularly in
those with suspected variceal hemorrhage. A single dose of erythromy­
cin (3–4 mg/kg or 250 mg) administered intravenously 30–90 min before
upper endoscopy increases gastric emptying and may clear blood and
clots from the stomach to improve endoscopic visualization.
A
C
E
G
FIGURE 333-23  Peroral endoscopic tumorectomy (POET). A. Mid-esophageal subepithelial lesion (arrow). B. Mucosal incision (mucosotomy) 5 cm proximal to the
lesion. C. Submucosal dissection and tunneling to the site of the lesion. D. Dissection of the lesion from its attachment to the muscularis propria. E. Postresection defect
through the muscularis propria. F. Mucosotomy site. G. Closure of mucosotomy site with clips. H. Resected specimen (leiomyoma).
B
CHAPTER 333
Gastrointestinal Endoscopy
D
F
H
A
C
D
PART 10
Disorders of the Gastrointestinal System
FIGURE 333-24  Endoscopic full-thickness resection (EFTR) of a gastrointestinal stromal tumor. A. Subepithelial lesion in the proximal stomach. B. Hypoechoic lesion arising
from the fourth layer (muscularis propria) at endoscopic ultrasound. C. Full-thickness resection defect. D. Closure of defect using an over-the-scope clip.
Most patients with hematochezia who are otherwise stable can
undergo semi-elective colonoscopy. Controlled trials have not shown a
benefit to urgent colonoscopy (within 24 h of presentation) in patients
hospitalized with hematochezia, although selected patients with mas­
sive or recurrent large-volume episodes of hematochezia should prob­
ably undergo urgent colonoscopy after both upper endoscopy and a
rapid colonic purge with an oral polyethylene glycol solution. Colo­
noscopy has a higher diagnostic yield than radionuclide bleeding scans
or catheter-based angiography in lower gastrointestinal bleeding, and
endoscopic therapy can be applied in some cases. Urgent colonoscopy
A
FIGURE 333-25  Endoscopic mucosal resection (EMR). A. Large sessile polypoid fold in the transverse colon. B. Lifting of lesion following submucosal fluid injection.
C. Piecemeal hot snare resection. D. Initial resection site. E. Resection defect following completion of piecemeal EMR.
B
can be hindered by poor visualization due to persistent vigorous bleed­
ing with recurrent hemodynamic instability, and other techniques (e.g.,
angiography or even emergent subtotal colectomy) must be employed.
The anal and rectal mucosa should also be visualized endoscopically
early in the course of massive rectal bleeding, as bleeding lesions in or
close to the anal canal may be identified that are amenable to endo­
scopic or surgical transanal hemostatic techniques.
Peptic Ulcer 
The endoscopic appearance of peptic ulcers provides
useful prognostic information and guides the need for endoscopic
B
C
E
FIGURE 333-25  (Continued)
A
FIGURE 333-26  Endoscopic submucosal dissection (ESD). A. Large, flat, distal rectal adenoma. B. Circumferential incision following submucosal fluid injection at the
periphery of the lesion. C. ESD using an electrosurgical knife. D. Rectal defect following ESD. E. Specimen resected en bloc.
D
CHAPTER 333
Gastrointestinal Endoscopy
B
PART 10
Disorders of the Gastrointestinal System
FIGURE 333-27  Closure of large defect using an endoscopic suturing device. A. Ulcerated inflammatory fibroid polyp in the antrum. B. Large defect following endoscopic
submucosal dissection of the lesion. C. Closure of the defect using endoscopic sutures (arrows). D. Resected specimen.
C
D
E
A
B
FIGURE 333-26  (Continued)
C
FIGURE 333-27  (Continued)
therapy in patients with acute hemorrhage (Fig. 333-31). A cleanbased ulcer is associated with a low risk (3–5%) of rebleeding; patients
with melena and a clean-based ulcer may be discharged home from
the emergency room or endoscopy suite if they are young, reliable,
otherwise healthy, and able to return as needed. Flat pigmented spots
and adherent clots covering the ulcer base have a 10% and 20% risk of
rebleeding, respectively. Flat pigmented spots do not require treatment,
but endoscopic therapy is often applied to an ulcer with an adherent
clot. When a fibrin plug is seen protruding from a vessel wall in the
base of an ulcer (so-called sentinel clot or visible vessel), the risk of
rebleeding from the ulcer approximates 40%. This finding typically
leads to endoscopic therapy to decrease the rebleeding rate. When
A
C
FIGURE 333-28  Prevention of stent migration using endoscopic sutures. A. Esophagogastric anastomotic stricture refractory to balloon dilation. B. Temporary placement of
a covered esophageal stent. C. Endoscopic suturing device to anchor the stent to the esophageal wall. D. Stent fixation with endoscopic sutures (arrows).
D
active spurting from an ulcer is seen, there is a 90% risk of ongoing
bleeding without endoscopic or surgical therapy.
Endoscopic therapy of ulcers with high-risk stigmata typically
lowers the rebleeding rate to 5–10%. Several hemostatic techniques
are available, including injection of epinephrine or a sclerosant into
and around the vessel (Fig. 333-32), “coaptive coagulation” of the
vessel in the base of the ulcer using a thermal probe that is pressed
against the site of bleeding (Fig. 333-33), placement of through-thescope clips (Fig. 333-34) or an over-the-scope clip (Fig. 333-35),
application of a hemostatic powder or gel, or a combination of these
modalities (Video V5-9). Epinephrine injection can slow or stop
active bleeding, but it is not a stand-alone technique for definitive
CHAPTER 333
Gastrointestinal Endoscopy
B
D
TABLE 333-1  Antibiotic Prophylaxis for Endoscopic Procedures
PATIENT CONDITION
PROCEDURE CONTEMPLATED
GOAL OF PROPHYLAXIS
All cardiac conditions
Any endoscopic procedure
Prevention of infective endocarditis
Not recommended
Bile duct obstruction in the absence of
cholangitis
ERCP with complete drainage
Prevention of cholangitis
Not recommended
Bile duct obstruction in the absence of
cholangitis
ERCP with anticipated incomplete
drainage (e.g., sclerosing cholangitis,
hilar strictures)
Sterile pancreatic fluid collection (e.g.,
pseudocyst, necrosis), which communicates
with pancreatic duct
ERCP
Prevention of cyst infection
Recommended; continue antibiotics
after the procedure
Sterile pancreatic fluid collection
Transmural drainage
Prevention of cyst infection
Recommended
Solid lesion along upper GI tract
EUS-FNA or FNB
Prevention of local infection
Not recommendeda
Solid lesion along lower GI tract
EUS-FNA or FNB
Prevention of local infection
Not recommendeda
Cystic lesions along GI tract (including
mediastinum and pancreas)
EUS-FNA or FNB
Prevention of cyst infection
Recommended
All patients
Percutaneous endoscopic feeding tube
placement
Cirrhosis with acute GI bleeding
Recommended for all such patients,
regardless of endoscopic procedures
Continuous peritoneal dialysis
Lower GI tract endoscopy
Prevention of bacterial peritonitis
Recommended
Synthetic vascular graft and other
nonvalvular cardiovascular devices
Any endoscopic procedure
Prevention of graft and device infection
Not recommendedd
Prosthetic joints
Any endoscopic procedure
Prevention of septic arthritis
Not recommendedd
aLow rates of bacteremia and local infection. bCefazolin or an antibiotic with equivalent coverage of oral and skin flora. cRisk for bacterial infection associated with cirrhosis
and GI bleeding is well established; ceftriaxone or a quinolone antibiotic recommended. dVery low risk of infection.
Abbreviations: ERCP, endoscopic retrograde cholangiopancreatography; EUS-FNA, endoscopic ultrasound–fine-needle aspiration; EUS-FNB, endoscopic ultrasound–fine
needle biopsy; GI, gastrointestinal.
Source: Reproduced with permission from MA Kashab et al: Antibiotic prophylaxis for GI endoscopy. Gastrointest Endosc 81:81, 2015.
PART 10
Disorders of the Gastrointestinal System
hemostasis. In conjunction with endoscopic therapy, the administra­
tion of a proton pump inhibitor decreases the risk of rebleeding and
improves patient outcome.
Varices 
Two complementary strategies guide therapy of bleeding
varices: local treatment of the bleeding varices and treatment of the
underlying portal hypertension. Local therapies, including endoscopic
variceal band ligation, endoscopic variceal sclerotherapy, stent place­
ment, and tamponade with a stent or Sengstaken-Blakemore tube,
effectively control acute hemorrhage in most patients, although thera­
pies that decrease portal pressure (pharmacologic treatment, surgical
shunts, or radiologically placed transjugular intrahepatic portosys­
temic shunts) also play an important role.
Endoscopic variceal ligation (EVL) is indicated for the prevention
of a first bleed (primary prophylaxis) from large esophageal varices
(Fig. 333-36), particularly in patients in whom nonselective beta
blockers are contraindicated or not tolerated. EVL is also the preferred
endoscopic therapy for control of active esophageal variceal bleeding
and for subsequent eradication of esophageal varices (secondary pro­
phylaxis). During EVL, a varix is suctioned into a cap fitted at the tip
of the endoscope, and a rubber band is released from the cap, ligating
the varix (Fig. 333-37, Video V5-10). EVL controls acute hemorrhage
in up to 90% of patients. Complications of EVL, such as postligation
ulcer bleeding and esophageal stenosis, are uncommon. Endoscopic
variceal sclerotherapy (EVS) involves the injection of a sclerosing,
thrombogenic solution into or next to esophageal varices. EVS also
controls acute hemorrhage in most patients, but due to its higher
complication rate, it is generally used as salvage therapy when band
ligation fails. Bleeding from large gastric fundal varices (Fig. 333-38)
is best treated with endoscopic cyanoacrylate (“glue”) injection (Video
V5-11) or EUS-guided coil placement and cyanoacrylate injection,
since EVL or EVS of these varices is associated with a high rebleeding
rate. Complications of cyanoacrylate injection include infection and
glue embolization to other organs, such as the lungs, brain, and spleen.
After treatment of the acute hemorrhage, an elective course of
endoscopic therapy can be undertaken with the goal of eradicating
PERIPROCEDURAL ANTIBIOTIC
PROPHYLAXIS
Prevention of cholangitis
Recommended; continue antibiotics
after the procedure
Prevention of peristomal infection
Recommendedb
Prevention of infectious complications
and reduction of mortality
Recommended, upon admissionc
esophageal varices and preventing rebleeding months to years later.
However, this chronic therapy is less successful, preventing long-term
rebleeding in ~50% of patients. Pharmacologic therapies that decrease
portal pressure have similar efficacy. The preferred strategy, however,
for secondary prophylaxis of variceal bleeding is the combination of
EVL with a nonselective beta blocker.
Dieulafoy’s Lesion 
This lesion, also called persistent caliber
artery, is a large-caliber arteriole that runs immediately beneath the
gastrointestinal epithelium and bleeds through a focal mucosal ero­
sion (Fig. 333-39). Dieulafoy’s lesion commonly involves the lesser
curvature of the proximal stomach, causes impressive arterial hemor­
rhage, and may be difficult to diagnose when not actively bleeding; it is
often recognized only after repeated endoscopy for recurrent bleeding.
Endoscopic therapy, such as thermal coagulation, band ligation, clip
placement, or endoscopic suturing, is typically effective for control
of bleeding and sealing of the underlying vessel once the lesion has
been identified (Video V5-12). Rescue therapies, such as angiographic
embolization or surgical oversewing, are considered in situations
where endoscopic therapy has failed.
Mallory-Weiss Tear 
A Mallory-Weiss tear is a linear mucosal rent
near or across the gastroesophageal junction that is often associated
with retching or vomiting (Fig. 333-40). When the tear disrupts a
submucosal arteriole, brisk hemorrhage may result. Endoscopy is the
best method for diagnosis, and an actively bleeding tear can be treated
endoscopically with coaptive coagulation, band ligation, or clip place­
ment, with or without epinephrine injection (Video V5-13). Unlike
peptic ulcer, a Mallory-Weiss tear with a nonbleeding sentinel clot in its
base rarely rebleeds and thus does not necessitate endoscopic therapy.
Vascular Ectasias 
Vascular ectasias are flat mucosal vascular
anomalies that are best diagnosed by endoscopy. They usually cause
slow intestinal blood loss and occur either in a sporadic fashion or in a
well-defined pattern of distribution (e.g., gastric antral vascular ectasia
[GAVE] or “watermelon stomach”) (Fig. 333-41). Cecal vascular ecta­
sias, GAVE, and radiation-induced rectal ectasias are often responsive
TABLE 333-2  Management of Antithrombotic Drugs Prior to Endoscopic Procedures
BLEEDING RISK
OF PROCEDURE
MANAGEMENT
DRUG
Warfarin
Lowa
Continue
N/A
Ensure that INR is not supratherapeutic
 
Highb
Discontinue
3–7 days (usually 5), INR should
be ≤1.5 for procedure
Dabigatran, rivaroxaban,
apixaban, edoxaban
Lowa
Hold morning dose on
day of procedure
Dabigatran
Highb
Discontinue
2–3 days if GFR is ≥50 mL/min,
4–5 days if GFR is 30–49 mL/min
Rivaroxaban, apixaban,
edoxaban
Higha
Discontinue
2 days if GFR is ≥60 mL/min, 3
days if GFR is 30–59 mL/min, 4
days if GFR is <30 mL/min
Heparin
Lowa
Continue
N/A
 
 
Highb
Discontinue
4–6 h for unfractionated heparin Skip one dose if using low-molecular-weight heparin
Aspirin
Any
Continue
N/A
Low-dose aspirin does not substantially increase the risk
of endoscopic procedures
Aspirin with dipyridamole
Lowa
Continue
N/A
 
 
Highb
Discontinue
2–7 days
Consider continuing aspirin monotherapy
P2Y12 receptor antagonists
(clopidogrel, prasugrel,
ticlopidine, ticagrelor,
cangrelor)
Lowa
Continue
Coronary stent in place:
discuss with cardiologist
No coronary stent:
discontinue, consider
substituting aspirin
Highb
aLow-risk endoscopic procedures include esophagogastroduodenoscopy (EGD) or colonoscopy with or without biopsy, endoscopic ultrasound (EUS) without fine-needle
aspiration (FNA), and endoscopic retrograde cholangiopancreatography (ERCP) with stent exchange. bHigh-risk endoscopic procedures include EGD or colonoscopy with
dilation, polypectomy, or thermal ablation; percutaneous endoscopic gastrostomy (PEG); EUS with FNA; and ERCP with sphincterotomy or pseudocyst drainage. cBridging
therapy with low-molecular-weight heparin should be considered for patients discontinuing warfarin who are at high risk for thromboembolism, including those with
(1) prosthetic metal heart valve, (2) atrial fibrillation with a CHA2DS2-VASc score ≥3, mitral stenosis, prosthetic valve or history of stroke or transient ischemic attack; (3)
mechanical mitral valve; (4) mechanical aortic valve with other thromboembolic risk factors or older-generation mechanical aortic valve; or (5) venous thromboembolism
(VTE) within the past 3 months.
Abbreviations: GFR, glomerular filtration rate; GI, gastrointestinal; INR, international normalized ratio; N/A, not applicable.
Source: Adapted from RD Acosta et al: Gastrointest Endosc 83:3, 2016; and AM Veitch et al: Gut 70:1611, 2021.
A
FIGURE 333-29  Bleeding from percutaneous endoscopic gastrostomy (PEG) tube placement. A. Patient with melena from a recently placed PEG tube. B. Loosening of the
internal disk bumper of the PEG tube revealed active bleeding from within the PEG tract.
INTERVAL BETWEEN LAST
DOSE AND PROCEDURE
COMMENTS
Consider bridging therapy with low-molecular-weight
heparin for patients at high risk of thrombosisc; usually
safe to resume warfarin on the same or next day
For life-threatening GI hemorrhage, consider reversal
with unactivated prothrombin complex concentrate
N/A
 
Bridging therapy not recommended; resume drug when
bleeding risk is low
For life-threatening GI hemorrhage, consider use of a
reversal agent
Bridging therapy not recommended; resume drug when
bleeding risk is low
For life-threatening GI hemorrhage, consider use of a
reversal agent
N/A
5 days (clopidogrel or
ticagrelor), 7 days (prasugrel),
10–14 days (ticlopidine)
Risk of stent thrombosis for at least 12 months after
insertion of drug-eluting coronary stent or 1 month after
insertion of bare metal coronary stent
CHAPTER 333
Gastrointestinal Endoscopy
B
A
FIGURE 333-30  Buried bumper syndrome. A. Migration of the internal disk bumper of a percutaneous endoscopic gastrostomy (PEG) tube through the gastric wall.
B. Close-up view of the disk bumper (arrow) buried in the gastric wall.
to local endoscopic ablative therapy, such as argon plasma coagulation
(Video V5-14). Patients with diffuse small-bowel vascular ectasias
(associated with chronic renal failure and with hereditary hemorrhagic
telangiectasia) may continue to bleed despite endoscopic treatment
of easily accessible lesions by conventional endoscopy. These patients
may benefit from deep enteroscopy with endoscopic hemostasis, or
pharmacologic therapy, such as octreotide or low-dose thalidomide, in
those who continue to bleed despite endoscopic therapy.
PART 10
Disorders of the Gastrointestinal System
Colonic Diverticula 
Diverticula form where nutrient arteries
penetrate the muscular wall of the colon en route to the colonic mucosa
(Fig. 333-42). The artery found in the base of a diverticulum may
bleed, causing painless and impressive hematochezia. Colonoscopy
is indicated in patients with hematochezia and suspected diverticular
hemorrhage, since other causes of bleeding (e.g., vascular ectasias,
colitis, and colon cancer) must be excluded. In addition, an actively
A
FIGURE 333-31  Stigmata of hemorrhage in peptic ulcers. A. Gastric antral ulcer with a clean base. B. Duodenal ulcer with flat pigmented spots (arrows). C. Duodenal ulcer
with a dense adherent clot. D. Duodenal ulcer with a pigmented protuberance/visible vessel (arrow). E. Duodenal ulcer with active spurting (arrow).
B
bleeding diverticulum may be seen and treated during colonoscopy
(Fig. 333-43, Video V5-15).
■
■GASTROINTESTINAL OBSTRUCTION AND
PSEUDOOBSTRUCTION
Endoscopy is useful for evaluation and treatment of some forms of
gastrointestinal obstruction. An important exception is small-bowel
obstruction due to surgical adhesions, which is generally not diagnosed
or treated endoscopically. Esophageal, gastroduodenal, and colonic
obstruction or pseudoobstruction can all be diagnosed and often man­
aged endoscopically.
Acute Esophageal Obstruction 
Esophageal obstruction by
impacted food (Fig. 333-44) or an ingested foreign body (Fig. 333-45) is
a potentially life-threatening event and represents an endoscopic emer­
gency. Left untreated, the patient may develop esophageal ulceration,
B
C
D
E
FIGURE 333-31  (Continued)
FIGURE 333-32  Injection therapy for ulcer hemostasis. Epinephrine injection into a
duodenal ulcer with visible vessel (arrow) and adherent clot.
CHAPTER 333
Gastrointestinal Endoscopy
ischemia, and perforation. Patients with persistent esophageal obstruc­
tion often have hypersalivation and are usually unable to swallow
water. Sips of a carbonated beverage, sublingual nifedipine or nitrates,
or intravenous glucagon may resolve an esophageal food impaction,
but in many patients, an underlying web, ring, or stricture is present,
and endoscopic removal of the obstructing food bolus is necessary.
Endoscopy is generally the best initial test in such patients since endo­
scopic removal of the obstructing material is usually possible, and the
presence of an underlying esophageal pathology can often be deter­
mined. Radiographs of the chest and neck should be considered before
endoscopy in patients with fever, obstruction for ≥24 h, or ingestion of
a sharp object, such as a fishbone. Radiographic contrast studies inter­
fere with subsequent endoscopy and are not advisable in most patients
with a clinical picture of esophageal obstruction.
Gastric Outlet Obstruction 
Obstruction of the gastric outlet is
commonly caused by gastric, duodenal, or pancreatic malignancy or
chronic peptic ulceration with stenosis of the pylorus (Fig. 333-46).
Patients vomit partially digested food many hours after eating. Gas­
tric decompression with a nasogastric tube and subsequent lavage for
removal of retained material is the first step in treatment. Endoscopy is
useful for diagnosis and treatment. Patients with benign pyloric steno­
sis may be treated with endoscopic balloon dilation of the pylorus, and
a course of endoscopic dilation results in long-term relief of symptoms
PART 10
Disorders of the Gastrointestinal System
A
B
C
A
B
FIGURE 333-33  Contact coagulation for ulcer hemostasis. A. Duodenal ulcer with a visible vessel (arrow). B. Coagulation of the vessel with a contact thermal probe.
C. Obliteration of the treated vessel (arrow).
FIGURE 333-34  Through-the-scope clip placement for ulcer hemostasis. A. Superficial duodenal ulcer with visible vessel (arrow). B. Hemostasis secured following
placement of multiple through-the-scope clips.
A
B
FIGURE 333-35  Over-the-scope clip placement for ulcer hemostasis. A. Pyloric
channel ulcer with visible vessel (arrow). B. Hemostasis secured following
placement of an over-the-scope clip.
in ~50% of patients. Removable, fully covered lumen-apposing metal
stents (LAMS) may also be used to treat benign pyloric stenosis
(Video V5-16). Malignant gastric outlet obstruction can be relieved
with endoscopically placed expandable stents across the obstruction in
patients with inoperable malignancy (Video V5-17) or by EUS-guided
gastroenterostomy to bypass the obstruction (Video V5-3).
Colonic Obstruction and Pseudoobstruction 
These con­
ditions both present with abdominal distention and discomfort,
tympany, and a dilated colon on plain abdominal radiography.
The radiographic appearance may be characteristic of a particular
condition, such as sigmoid volvulus (Fig. 333-47). Both obstruc­
tion and pseudoobstruction may lead to colonic perforation if left
untreated. Acute colonic pseudoobstruction is a form of colonic
FIGURE 333-36  Esophageal varices.
A
CHAPTER 333
Gastrointestinal Endoscopy
B
FIGURE 333-37  Endoscopic variceal ligation. A. Esophageal varices with red wale
marks. B. Band ligation of varices.
ileus that is usually attributable to electrolyte disorders, narcotic
and anticholinergic medications, immobility (as after surgery), or
retroperitoneal hemorrhage or mass. Multiple causative factors are
often present. Colonoscopy, water-soluble contrast enema, or CT
may be used to assess for an obstructing lesion and differentiate
obstruction from pseudoobstruction. One of these diagnostic studies
should be strongly considered if the patient does not have clear risk
factors for pseudoobstruction, if radiographs do not show air in the
rectum, or if the patient fails to improve when underlying causes of
pseudoobstruction have been addressed. The risk of cecal perforation
in pseudoobstruction rises when the cecal diameter exceeds 12 cm,
and decompression of the colon may be achieved using intravenous
neostigmine or via colonoscopic decompression (Fig. 333-48). Most
patients should receive a trial of conservative therapy (with correc­
tion of electrolyte disorders, discontinuation of offending medica­
tions, and increased mobilization) before undergoing an invasive
decompressive procedure for colonic pseudoobstruction.
Acute colonic obstruction is an indication for urgent intervention.
In the past, emergent diverting colostomy was usually performed
with a subsequent second operation after bowel preparation to treat
the underlying cause of obstruction. Colonoscopic placement of an
PART 10
Disorders of the Gastrointestinal System
A
B
C
A
B
FIGURE 333-39  Dieulafoy’s lesion. A. Actively spurting gastric Dieulafoy’s lesion. B. Coagulation of the lesion using a contact thermal probe. C. Hemostasis secured
following contact coagulation (arrow). D. Histology of a gastric Dieulafoy’s lesion. A persistent caliber artery (arrows) is present in the gastric submucosa, immediately
beneath the mucosa.
FIGURE 333-38  Gastric varices. A. Large gastric fundal varices with stigmata of recent bleeding (arrow). B. Injection of cyanoacrylate (glue) into the culprit gastric
varix. C. Obliterated varix following glue injection on endoscopic follow-up at 1 month (arrow).
C
FIGURE 333-39  (Continued)
expandable stent is an alternative treatment option that can relieve
malignant colonic obstruction without emergency surgery and permit
bowel preparation for an elective one-stage operation (Fig. 333-49,
Video V5-18).
■
■ACUTE BILIARY OBSTRUCTION
The steady, severe pain that occurs when a gallstone acutely obstructs
the common bile duct often brings patients to seek medical atten­
tion. The diagnosis of a ductal stone is suspected when the patient is
jaundiced or when serum liver tests or pancreatic enzyme levels are
elevated; it is confirmed by transabdominal ultrasound, EUS, magnetic
resonance cholangiopancreatography (MRCP), or direct cholangiog­
raphy (performed endoscopically, percutaneously, or during surgery).
ERCP is the primary means of treating common bile duct stones (Figs.
333-16 and 333-17), although they can also be removed by bile duct
exploration at the time of cholecystectomy. Radiologic percutaneous
biliary drainage may be required in some cases.
Bile Duct Imaging 
While transabdominal ultrasound diagnoses
only a minority of bile duct stones, MRCP and EUS are >90% accurate
and have an important role in diagnosis. Examples of these modalities
are shown in Fig. 333-50.
FIGURE 333-40  Mallory-Weiss tear with an adherent clot at the gastroesophageal
junction following forceful retching and vomiting.
D
If the suspicion for a bile duct stone is high and urgent treatment is
required (as in a patient with obstructive jaundice and biliary sepsis),
ERCP is the procedure of choice since it remains the gold standard
for diagnosis and allows for immediate treatment (Video V5-19). If
a persistent bile duct stone is relatively unlikely (as in a patient with
gallstone pancreatitis), ERCP may be supplanted by less invasive imag­
ing techniques, such as EUS, MRCP, or intraoperative cholangiography
performed during cholecystectomy, sparing some patients the risk and
discomfort of ERCP.
CHAPTER 333
Gastrointestinal Endoscopy
Ascending Cholangitis 
Charcot’s triad of jaundice, abdominal
pain, and fever is present in ~70% of patients with ascending cholan­
gitis and biliary sepsis. These patients are managed initially with fluid
resuscitation and intravenous antibiotics. Abdominal ultrasound is
often performed to assess for gallbladder stones and bile duct dila­
tion. However, the bile duct may not be dilated early in the course of
acute biliary obstruction. Medical management usually improves the
patient’s clinical status, providing a window of ~24 h during which
biliary drainage should be established, typically by ERCP. Undue
delay can result in recrudescence of overt sepsis and increased mor­
bidity and mortality rates. In addition to Charcot’s triad, the presence
of shock and confusion (Reynolds’s pentad) is associated with a high
mortality rate and should prompt urgent intervention to restore biliary
drainage.
Gallstone Pancreatitis 
Gallstones may cause acute pancreatitis
as they pass through the ampulla of Vater. The occurrence of gallstone
pancreatitis usually implies passage of a stone into the duodenum, and
only ~20% of patients harbor a persistent stone in the ampulla or the
common bile duct. Retained stones are more common in patients with
jaundice, rising serum liver tests following hospitalization, severe pan­
creatitis, or superimposed ascending cholangitis.
Urgent ERCP decreases the morbidity rate of gallstone pancreati­
tis in a subset of patients with retained bile duct stones. It is unclear
whether the benefit of ERCP is mainly attributable to treatment and
prevention of ascending cholangitis or to relief of pancreatic duc­
tal obstruction. ERCP is warranted early in the course of gallstone
pancreatitis if ascending cholangitis is suspected, especially in a
jaundiced patient. Urgent ERCP may also benefit patients predicted
to have severe pancreatitis using a clinical index of severity, such as
the Glasgow or Ranson score. Since the benefit of ERCP is limited to
patients with a retained bile duct stone, a strategy of initial MRCP or
EUS for diagnosis decreases the utilization of ERCP in gallstone pan­
creatitis and improves clinical outcomes by limiting the occurrence of
ERCP-related adverse events.
PART 10
Disorders of the Gastrointestinal System
A
B
C
FIGURE 333-41  Gastrointestinal vascular ectasias. A. Gastric antral vascular ectasia (“watermelon stomach”) characterized by stripes of prominent flat or raised vascular
ectasias. B. Cecal vascular ectasia. C. Radiation-induced vascular ectasias of the rectum in a patient previously treated for prostate cancer.
ELECTIVE ENDOSCOPY
■
■DYSPEPSIA
Dyspepsia is a chronic or recurrent burning discomfort or pain in
the upper abdomen that may be caused by diverse processes, such as
gastroesophageal reflux, peptic ulcer disease, and “nonulcer dyspepsia,”
a heterogeneous category that includes disorders of motility, sensa­
tion, and somatization. Gastric and esophageal malignancies are less
common causes of dyspepsia. Careful history-taking allows accurate
differential diagnosis of dyspepsia in only about half of patients. In
the remainder, endoscopy can be a useful diagnostic tool, especially
in patients whose symptoms are not resolved by Helicobacter pylori
treatment or an empirical trial of acid-reducing therapy. Endoscopy
should be performed at the outset in patients with dyspepsia and alarm
features, such as weight loss, obstructive symptoms, or iron-deficiency
anemia.
■
■GASTROESOPHAGEAL REFLUX DISEASE
When classic symptoms of gastroesophageal reflux are present, such
as water brash and substernal heartburn, presumptive diagnosis and
empirical treatment are often sufficient. Endoscopy is a sensitive test
for diagnosis of esophagitis (Fig. 333-51), but it will miss nonero­
sive reflux disease (NERD) since some patients have symptomatic
reflux without esophagitis. The most sensitive test for diagnosis of
gastroesophageal reflux disease (GERD) is 24-h ambulatory pH and
impedance monitoring. Endoscopy is indicated in patients with reflux
symptoms refractory to antisecretory therapy; in those with alarm
symptoms, such as dysphagia, weight loss, or gastrointestinal bleed­
ing; and in those with recurrent dyspepsia after treatment that is not
clearly due to reflux on clinical grounds alone. Endoscopy should be
considered in patients with long-standing GERD, as they have a sixfold
increased risk of harboring Barrett’s esophagus compared to patients
with <1 year of reflux symptoms.
FIGURE 333-42  Colonic diverticula.
Barrett’s Esophagus and Esophageal Squamous Dysplasia 
Barrett’s esophagus is specialized columnar metaplasia that replaces
the normal squamous mucosa of the distal esophagus in some persons
with GERD. Barrett’s epithelium is a major risk factor for adenocarci­
noma of the esophagus and is readily detected endoscopically, due to
proximal displacement of the squamocolumnar junction (Fig. 333-6).
A screening EGD for Barrett’s esophagus should be considered in
patients with a chronic (≥10 year) history of GERD symptoms, even
if their symptoms have been mild. Endoscopic biopsy is the gold
standard for confirmation of Barrett’s esophagus and for dysplasia or
cancer arising in Barrett’s mucosa.
Periodic EGD with biopsies is recommended for surveillance of
patients with Barrett’s esophagus. Endoscopic resection (EMR or ESD)
and/or ablation are treatment options when high-grade dysplasia or
intramucosal cancer are found in the Barrett’s mucosa. Both endo­
scopic therapy and periodic surveillance are acceptable options in
patients with Barrett’s esophagus and low-grade dysplasia. Radiofre­
quency ablation (RFA) is the most common ablative modality used
for endoscopic treatment of Barrett’s esophagus, and other modalities,
such as cryotherapy, are also available.
Esophageal squamous dysplasia is the precursor lesion of esophageal
squamous cell cancer (ESCC), the most common type of esophageal
malignancy worldwide. Endoscopic detection of esophageal squa­
mous dysplasia often requires specialized imaging methods, such as
chromoendoscopy with Lugol’s iodine solution. Once detected, it can
be treated endoscopically with EMR, ESD, or RFA (Fig. 333-52). Pop­
ulation-based screening for esophageal squamous dysplasia has been
shown to decrease the occurrence of ESCC in high-incidence regions.
■
■PEPTIC ULCER
Peptic ulcer classically causes epigastric gnawing or burning, often
occurring nocturnally and promptly relieved by food or antacids.
Although endoscopy is the most sensitive diagnostic test for peptic
ulcer, it is not a cost-effective strategy in young patients with ulcer-like
dyspeptic symptoms unless endoscopy is available at low cost. Patients
with suspected peptic ulcer should be evaluated for H. pylori infection.
Serology (which diagnoses past or current infection), urea breath test­
ing (current infection), and stool tests (current infection) are noninva­
sive and less costly than endoscopy with biopsy. Patients aged >50 and
those with alarm symptoms or persistent symptoms despite treatment
should undergo endoscopy to exclude malignancy.
■
■NONULCER DYSPEPSIA
Nonulcer dyspepsia may be associated with bloating and, unlike peptic
ulcer, tends not to remit and recur. Most patients describe persis­
tent symptoms despite acid-reducing, prokinetic, or anti-Helicobacter
A
CHAPTER 333
Gastrointestinal Endoscopy
B
C
FIGURE 333-43  Diverticular hemorrhage. A. Actively bleeding sigmoid diverticulum.
B. Treatment of the bleeding vessel at the dome of the diverticulum with a contact
thermal probe. C. Hemostasis secured following contact coagulation with tattoo
injection to aid future localization.
FIGURE 333-44  Esophageal food impaction. Meat bolus impacted in the distal
esophagus.
therapy and are referred for endoscopy to exclude a refractory ulcer
and assess for other causes. Although endoscopy is useful for excluding
other diagnoses, its impact on the treatment of patients with nonulcer
dyspepsia is limited.
PART 10
Disorders of the Gastrointestinal System
■
■DYSPHAGIA
About 50% of patients presenting with difficulty swallowing have a
mechanical obstruction; the remainder have a motility disorder, such
as achalasia or diffuse esophageal spasm, or an inflammatory disorder,
such as eosinophilic esophagitis. Careful history-taking often points to
a presumptive diagnosis and leads to the appropriate use of diagnostic
tests. Esophageal strictures (Fig. 333-53) typically cause progres­
sive dysphagia, first for solids, then for liquids; motility disorders
often cause intermittent dysphagia for both solids and liquids. Some
underlying disorders have characteristic historic features: Schatzki’s
ring (Fig. 333-54) causes episodic dysphagia for solids, typically at the
beginning of a meal; oropharyngeal motor disorders typically present
with difficulty initiating deglutition (transfer dysphagia) and nasal
reflux or coughing with swallowing; and achalasia may cause nocturnal
regurgitation of undigested food from the esophagus.
FIGURE 333-45  Esophageal foreign body. Intentionally ingested toothbrush
impacted in the esophageal lumen.
A
B
C
FIGURE 333-46  Gastric outlet obstruction due to pyloric stenosis. A. Nonsteroidal
anti-inflammatory agent–induced ulcer disease with severe stenosis of the pylorus
(arrow). B. Balloon dilation of the stenosis. C. Appearance of pyloric ring after
dilation.
When mechanical obstruction is suspected, endoscopy is a useful
initial diagnostic test, since it permits immediate biopsy and/or dilation
of strictures, masses, or rings. The presence of linear furrows and multiple
corrugated rings throughout a narrowed esophagus should raise sus­
picion for eosinophilic esophagitis, an increasingly recognized cause
of recurrent dysphagia and food impaction (Fig. 333-55). Blind or
forceful passage of an endoscope may lead to perforation in a patient
with stenosis of the cervical esophagus or a Zenker’s diverticulum
(Fig. 333-56), but gentle passage of an endoscope under direct visual
guidance is reasonably safe. Endoscopy can miss a subtle stricture or
ring in some patients.
When transfer dysphagia is evident or an esophageal motility disor­
der is suspected, esophageal radiography and/or a video-swallow study
are the best initial diagnostic tests. The oropharyngeal swallowing
mechanism, esophageal peristalsis, and the lower esophageal sphincter
A
B
C
FIGURE 333-47  Sigmoid volvulus. A. Abdominal x-ray showing characteristic
radiologic appearance of a “bent inner tube.” B. Site of sigmoid torsion identified
at colonoscopy C. Significantly dilated colonic lumen proximal to the sigmoid twist
with mild ischemic-appearing mucosal changes.
A
B
FIGURE 333-48  Acute colonic pseudoobstruction. A. Acute colonic dilation
occurring in a patient soon after knee surgery. B. Colonoscopic placement of
decompression tube with marked improvement in colonic dilation.
CHAPTER 333
can all be assessed. In some disorders, subsequent esophageal manom­
etry is required for diagnosis.
Various causes of dysphagia are amenable to endoscopic therapy.
Benign strictures, rings, and webs can be dilated using a through-thescope balloon (Fig. 333-57) or a tapered polyvinyl dilator passed over
a guide wire. In some instances, fibrotic strictures may respond to
needle-knife electroincision (Fig. 333-58) when they prove refractory
to dilation. Self-expanding esophageal stents can be used to palliate
dysphagia from malignant obstruction (Fig. 333-59), and flexible
endoscopic cricopharyngeal myotomy is an option for Zenker’s diver­
ticulum (Video V5-20). Recent advances in third-space (submucosal)
endoscopy have enabled the development of procedures, such as
POEM (Video V5-21) and POET (Video V5-22), for the management
of achalasia and select subepithelial esophageal tumors, respectively.
Gastrointestinal Endoscopy
■
■ENDOSCOPIC TREATMENT OF OBESITY
A significant proportion of Americans are overweight or obese, and
obesity-associated diabetes has become a major public health prob­
lem. Bariatric surgery is the most effective weight-loss intervention,
decreasing long-term mortality in obese persons, but many patients
choose not to undergo surgery. Endoscopic treatments for obesity have
been developed and include insertion of an intragastric balloon or duo­
denojejunal bypass liner, placement of a percutaneous gastric tube for
aspiration of gastric contents after meals, duodenal mucosal electro­
poration, or endoscopic sleeve gastroplasty, which utilizes endoscopic
suturing to narrow the lumen of the gastric body (Video V5-23).
Prospective trials show that these treatments induce total-body weight
loss of 7–20% and provide varying degrees of glycemic control and
improvement in other diseases linked to obesity, such as metabolicassociated fatty liver disease (MAFLD). Additional endoscopic modali­
ties are undergoing clinical trials. The efficacy of bariatric endoscopy
may approach that of bariatric surgery in the short term. Long-term
outcomes are still under evaluation.
■
■TREATMENT OF MALIGNANCIES
Endoscopy plays an important role in the treatment of gastrointestinal
malignancies. Early-stage malignancies limited to the mucosal and
superficial submucosal layers may be resected using the techniques
PART 10
Disorders of the Gastrointestinal System
A
B
C
FIGURE 333-49  Obstructing colonic carcinoma. A. Colonic adenocarcinoma causing marked luminal narrowing of the distal transverse colon. B. Endoscopic placement of
a self-expandable metal stent. C. Radiograph of expanded stent across the obstructing tumor with a residual waist (arrow).
A
B
C
FIGURE 333-50  Methods of bile duct imaging. Arrows mark bile duct stones. A. Endoscopic ultrasound (EUS) demonstrating a hyperechoic stone with acoustic shadowing
(arrowhead). B. Magnetic resonance cholangiopancreatography (MRCP). C. Helical computed tomography (CT).
A
C
FIGURE 333-51  Causes of esophagitis. A. Severe reflux esophagitis with mucosal ulceration and friability. B. Cytomegalovirus esophagitis. C. Herpes simplex virus
esophagitis with target-type shallow ulcerations. D. Candida esophagitis with white plaques adherent to the esophageal mucosa.
of EMR (Video V5-5) or ESD (Video V5-6). RFA and cryotherapy are
effective modalities for ablative treatment of high-grade dysplasia and
intramucosal cancer in Barrett’s esophagus (Video V5-24). Gastroin­
testinal stromal tumors can be removed en bloc by endoscopic fullthickness resection (Video V5-4). In general, endoscopic techniques
offer the advantage of a minimally invasive approach to treatment but
rely on other imaging techniques (such as CT, magnetic resonance
imaging [MRI], positron emission tomography [PET], and EUS) to
exclude distant metastases or locally advanced disease better treated by
surgery or other modalities. The decision to treat an early-stage gas­
trointestinal malignancy endoscopically is often made in collaboration
with a surgeon and/or oncologist.
A
FIGURE 333-52  Early squamous cell cancer. A. Nodularity in the distal esophagus due to T1 esophageal squamous cell cancer. B. The nodular lesion remains unstained
following chromoendoscopy with Lugol’s solution without additional unstained areas. C. Circumferential mucosal incision around the lesion. D. Resection defect following
en bloc removal of the lesion via endoscopic submucosal dissection.
B
D
CHAPTER 333
Endoscopic palliation of gastrointestinal malignancies relieves
symptoms and, in many cases, prolongs survival. Malignant obstruc­
tion can be relieved by endoscopic stent placement (Figs. 333-18,
333-49, 333-59, and 333-60; Videos V5-17 and V5-18), and malignant
gastrointestinal bleeding can be palliated endoscopically as well. EUSguided celiac plexus neurolysis may relieve pancreatic cancer pain.
Gastrointestinal Endoscopy
■
■ANEMIA AND OCCULT BLOOD IN THE STOOL
Iron-deficiency anemia may be attributed to poor iron absorption
(as in celiac sprue) or, more commonly, chronic blood loss. Intestinal
bleeding should be strongly suspected in men and postmenopausal
women with iron-deficiency anemia, and colonoscopy is indicated in
B
C
FIGURE 333-52  (Continued)
such patients, even in the absence of detectable occult blood in the
stool. Approximately 30% will have large colonic polyps or colorectal
cancer, and a few patients will have colonic vascular lesions. When a
convincing source of blood loss is not found in the colon, upper gastrointestinal endoscopy should be considered; if no lesion is found,
duodenal biopsies should be obtained to exclude sprue (Fig. 333-61).
Small-bowel evaluation with capsule endoscopy (Fig. 333-62), CT or
magnetic resonance (MR) enterography, or deep enteroscopy may be
appropriate if both EGD and colonoscopy are unrevealing.
PART 10
Disorders of the Gastrointestinal System
Tests for occult blood in the stool detect hemoglobin or the heme
moiety and are most sensitive for colonic blood loss, although they will
also detect larger amounts of upper gastrointestinal bleeding. Patients
with occult blood in the stool should undergo colonoscopy to diagnose
or exclude colorectal neoplasia, especially if they are >50 years old or
have a family history of colonic neoplasia. Whether upper endoscopy
is also indicated depends on the patient’s symptoms.
The small intestine may be the source of chronic intestinal bleeding, especially if colonoscopy and upper endoscopy are not diagnostic. The utility of small-bowel evaluation varies with the clinical
setting and is most important in patients in whom bleeding causes
A
B
FIGURE 333-53  Esophageal stricture. A. Peptic stricture associated with esophagitis. B. Balloon dilation of peptic stricture.
D
chronic or recurrent anemia. In contrast to the low diagnostic yield
of small-bowel radiography, positive findings on capsule endoscopy
are seen in 50–70% of patients with suspected small intestinal bleeding. The most common finding is mucosal vascular ectasia. CT and
MR enterography accurately detect small-bowel masses and Crohn’s
disease and are also useful for initial small-bowel evaluation. Deep
enteroscopy may follow capsule endoscopy for biopsy of lesions or to
provide specific therapy, such as argon plasma coagulation of vascular ectasias (Fig. 333-63).
■
■COLORECTAL CANCER SCREENING
The majority of colon cancers develop from preexisting colonic adenomas, and colorectal cancer can be largely prevented by the detection
and removal of adenomatous polyps (Video V5-25). The choice of
screening strategy for an asymptomatic person depends on personal
and family history. Individuals with inflammatory bowel disease, a history of colorectal polyps or cancer, family members with adenomatous
polyps or cancer, or certain familial cancer syndromes (Fig. 333-64)
are at increased risk for colorectal cancer. An individual without these
factors is generally considered at average risk.
FIGURE 333-54  Schatzki’s ring at the gastroesophageal junction.
Screening strategies are summarized in Table 333-3. While fecal
immunochemical tests (FITs) for heme or stool tests for occult blood
have been shown to decrease the mortality rate from colorectal cancer,
they do not detect some cancers and many polyps. FIT-DNA multitar­
geted stool DNA tests appear to be more sensitive, but direct visualiza­
tion of the colon is the gold standard method for detection of polyps
and cancers and remains a preferred screening strategy. Sigmoidoscopy
is also used for colorectal cancer screening. However, the distribution
of colon cancers has changed in the United States over time, with
proportionally fewer rectal and left-sided cancers than in the past.
Large American studies of colonoscopy for screening of average-risk
individuals show that cancers are roughly equally distributed between
the left and right colon and half of patients with right-sided lesions
have no polyps in the left colon. Visualization of the entire colon thus
appears to be the optimal strategy for colorectal cancer screening and
prevention.
Computed tomography colonography (CTC) is a radiologic tech­
nique that images the colon with CT following rectal insufflation of
FIGURE 333-55  Eosinophilic esophagitis. Multiple circular rings of the esophagus
creating a corrugated appearance and an impacted grape at the narrowed
esophagogastric junction. The diagnosis requires biopsy with histologic finding of
≥15 eosinophils/high-power field.
CHAPTER 333
A
Gastrointestinal Endoscopy
B
FIGURE 333-56  Zenker’s diverticulum. A. Contrast esophagography demonstrates
a moderate-sized Zenker’s diverticulum. B. Endoscopic view of the Zenker’s
diverticulum (left) relative to the true esophageal lumen (right) separated by the
diverticular septum. C. Flexible endoscopic diverticulotomy using an electrosurgical
knife. D. Appearance after diverticulotomy.
the colonic lumen. Computer rendering of CT images generates an
electronic display of a virtual “flight” along the colonic lumen, simulat­
ing colonoscopy (Fig. 333-65). Findings detected during CTC often
require subsequent conventional colonoscopy for confirmation and
treatment.
■
■DIARRHEA
Most cases of diarrhea are acute, self-limited, and due to infections
or medication. Chronic diarrhea (lasting >4–6 weeks) is more often
due to a primary inflammatory, malabsorptive, or motility disorder; is
less likely to resolve spontaneously; and generally, requires diagnostic
evaluation. Patients with chronic diarrhea or severe, unexplained acute
diarrhea often undergo endoscopy if stool tests for pathogens are
C
PART 10
Disorders of the Gastrointestinal System
D
FIGURE 333-56  (Continued)
unrevealing. The choice of endoscopic testing depends on the clinical
setting.
Patients with colonic symptoms and findings such as bloody
diarrhea, tenesmus, fever, or leukocytes in stool generally undergo
sigmoidoscopy or colonoscopy to assess for colitis (Fig. 333-9). Sig­
moidoscopy is an appropriate initial test in most patients. Conversely,
patients with symptoms and findings suggesting small-bowel disease,
such as large-volume watery stools, substantial weight loss, and malab­
sorption of iron, calcium, or fat, may undergo upper endoscopy with
duodenal aspirates for assessment of bacterial overgrowth and biopsies
for assessment of mucosal diseases, such as celiac sprue.
Many patients with chronic diarrhea do not fit either of these pat­
terns. In the setting of a long-standing history of altered bowel habits
dating to early adulthood, without findings such as blood in the stool
or anemia, a diagnosis of irritable bowel syndrome may be made
without direct visualization of the bowel and by relying on appropri­
ate blood tests (including complete blood count, C-reactive protein
and antibody tests for celiac disease) to screen for other diagnoses.
Steatorrhea and upper abdominal pain may prompt evaluation of
the pancreas rather than the gut. Patients whose chronic diarrhea is
not easily categorized often undergo initial colonoscopy to examine
the entire colon and terminal ileum for inflammatory or neoplastic
disease (Fig. 333-66).
A
B
C
FIGURE 333-57  Endoscopic management of peptic stricture. A. Peptic stricture.
B. Through-the-scope balloon dilation of stricture. C. Improvement in luminal
diameter after dilation.
■
■MINOR HEMATOCHEZIA
Bright red blood passed with or on formed brown stool usually has an
anal, rectal, or sigmoid source (Fig. 333-67). Even trivial amounts of
hematochezia should be investigated with colonoscopy and/or flexible
sigmoidoscopy together with anoscopy to exclude polyps or cancers,
especially in patients >40 years old and those with a personal or fam­
ily history of colorectal polyps or cancer. Patients reporting red blood
on the toilet tissue only, without blood in the toilet or on the stool,
are generally bleeding from a lesion in the anal canal; careful external
inspection, digital examination, and sigmoidoscopy with anoscopy
may be sufficient for diagnosis in such cases.
■
■PANCREATITIS
About 20% of patients with pancreatitis have no identified cause after
routine clinical investigation (including a review of medication and
alcohol use; measurement of serum triglyceride, calcium, and immu­
noglobulin G subclass 4 [IgG4] levels; abdominal ultrasonography;
and CT or MRI). Endoscopic assessment leads to a specific diagnosis
A
B
FIGURE 333-58  Endoscopic management of an esophagogastric anastomotic stricture. A. Recurrent anastomotic stricture despite periodic balloon dilation. B. Needle-knife
electroincision of stricture. C. Improvement in luminal opening after therapy.
in the majority of such patients, often altering clinical management.
Endoscopic investigation is particularly appropriate if the patient has
had more than one episode of pancreatitis.
Microlithiasis, or the presence of microscopic crystals in bile, is
a leading cause of previously unexplained acute pancreatitis and is
sometimes seen during abdominal ultrasonography as layering sludge
or flecks of floating, echogenic material in the gallbladder. EUS may
identify microlithiasis or gallstones not seen on transabdominal
ultrasound.
Previously undetected chronic pancreatitis, pancreatic malignancy,
or pancreas divisum may be diagnosed by either ERCP or EUS. Auto­
immune pancreatitis is often suspected based on CT, MRI, or serologic
findings, but it may first become apparent during EUS and may require
EUS-guided pancreatic biopsy for histologic diagnosis.
Severe pancreatitis often results in pancreatic fluid collections.
Symptomatic pseudocysts and areas of walled-off pancreatic necro­
sis can be drained into the stomach or duodenum endoscopically,
using transpapillary and transmural endoscopic techniques. Pancreatic
necrosis can be debrided by direct endoscopic necrosectomy (Video
V5-2) via an endoscopically created transmural drainage site.
■
■CANCER STAGING
Local staging of esophageal, gastric, pancreatic, bile duct, and rectal
cancers can be obtained with EUS (Fig. 333-20). EUS with fine-needle
aspiration (Fig. 333-21) or biopsy currently provides the most accurate
preoperative assessment of local tumor and nodal staging, but it does
not detect many distant metastases. Details of the local tumor stage
can guide treatment decisions, including resectability and need for
neoadjuvant therapy. EUS with transesophageal needle biopsy may
A
FIGURE 333-59  Palliation of malignant dysphagia. A. Obstructing distal esophageal cancer. B. Palliative stent placement.
C
also be used to assess the presence of non-small-cell lung cancer in
mediastinal nodes.
OPEN-ACCESS ENDOSCOPY
Direct scheduling of endoscopic procedures by providers without
preceding gastroenterology consultation, or open-access endoscopy,
is common. When the indications for endoscopy are clear-cut and
appropriate, the procedural risks are low, and the patient understands
what to expect, open-access endoscopy streamlines patient care and
decreases costs.
CHAPTER 333
Patients referred for open-access endoscopy should have a recent
history, physical examination, and medication list that are available
for review when the patient comes to the endoscopy suite. Patients
with unstable or symptomatic cardiovascular or respiratory conditions
should not be referred directly for open-access endoscopy. Those with
particular conditions who are undergoing certain procedures should
be prescribed prophylactic antibiotics before endoscopy (Table 333-1).
In addition, patients taking anticoagulants and/or antiplatelet drugs
may require adjustment of these agents before endoscopy based on the
procedural risk for bleeding and their underlying risk for a thrombo­
embolic event (Table 333-2).
Gastrointestinal Endoscopy
Common indications for open-access EGD include dyspepsia
resistant to a trial of appropriate therapy, dysphagia, gastrointestinal
bleeding, and persistent anorexia or early satiety. Open-access colo­
noscopy is often requested in men or postmenopausal women with
iron-deficiency anemia, in patients with hematochezia or occult blood
in the stool, in patients with a previous history of colorectal adeno­
matous polyps or cancer, and for colorectal cancer screening. Flexible
sigmoidoscopy is commonly performed as an open-access procedure.
B
A
PART 10
Disorders of the Gastrointestinal System
C
FIGURE 333-60  Placement of biliary and duodenal self-expanding metal stents (SEMS) for obstruction caused by pancreatic cancer. A. Endoscopic retrograde
cholangiopancreatography (ERCP) demonstrates a distal bile duct stricture (arrow). B. A biliary SEMS is placed. C. Contrast injection demonstrates a duodenal stricture
(arrow). D. Biliary and duodenal SEMS in place.
FIGURE 333-61  Celiac sprue. Scalloped duodenal folds in a patient with celiac
sprue.
B
D
When patients are referred for open-access colonoscopy, the pri­
mary care provider may need to choose a colonic preparation. Com­
monly used oral preparations include polyethylene glycol lavage
solution, with or without citric acid. A “split-dose” regimen improves
the quality of colonic preparation. Osmotic purgative preparations
(such as sodium phosphate or magnesium citrate) are also effective
but may cause fluid and electrolyte abnormalities and renal toxicity,
especially in patients with renal failure or congestive heart failure and
those >70 years of age.
FIGURE 333-62  Capsule endoscopy. Images of a mildly scalloped jejunal fold (left)
and an ileal tumor (right) in a patient with celiac sprue. (Images courtesy of Dr.
Elizabeth Rajan; with permission.)
A
C
FIGURE 333-63  Small-bowel vascular ectasia. A. Actively bleeding mid-jejunal vascular ectasia identified by double-balloon enteroscopy. B. Ablation of vascular ectasia
with argon plasma coagulation (APC). C. Hemostasis secured following APC.
FIGURE 333-64  Familial adenomatous polyposis. Numerous colon polyps in a patient with familial adenomatous polyposis syndrome.
B
CHAPTER 333
Gastrointestinal Endoscopy
TABLE 333-3  Colorectal Cancer Screening Strategies
 
CHOICES/RECOMMENDATIONS
COMMENTS
Average-Risk Patients
Asymptomatic individuals between 45 and 75 years of age
Colonoscopy every 10 yearsa
Gold standard cancer prevention strategy
 
Multitargeted stool DNA test every 1–3 years 
FIT or HSgFOBT every year, with or without flexible
sigmoidoscopy every 10 years
 
CT colonography every 5 years
Colonoscopy if results are positive
 
Flexible sigmoidoscopy every 5 years
Does not detect proximal colon polyps and cancers;
colonoscopy if an adenomatous polyp is found
Asymptomatic individuals > 75 years of age
Selective screening
Consider patient’s overall health, results of previous
screening exams, and preferences
Personal History of Polyps or CRC
1–2 small (<10 mm) tubular adenomas
Repeat colonoscopy in 7–10 yearsa
Assuming complete polyp resection. Interval may vary
based on prior personal history and family history
3–4 tubular adenomas <10 mm
Repeat colonoscopy in 3–5 yearsa; subsequent
colonoscopy based on findings
5–10 tubular adenomas <10 mm
Repeat colonoscopy in 3 yearsa
Assuming complete polyp resection
10 adenomas on a single exam
Repeat colonoscopy in 1 yeara
Consider evaluation for FAP or HNPCC; see
recommendations below
Adenoma ≥10 mm, or adenoma with tubulovillous or
villous histology, or high-grade dysplasia
Repeat colonoscopy in 3 yearsa
Assuming complete polyp resection
Piecemeal removal of any sessile polyp
Exam in 6 months to verify complete removal
 
Small (<1 cm) hyperplastic polyps of sigmoid and
rectum
Repeat colonoscopy in 10 yearsa
Those with hyperplastic polyposis syndrome merit more
frequent follow-up
Hyperplastic polyp ≥10 mm
Repeat colonoscopy in 3–5 yearsa
 
PART 10
Disorders of the Gastrointestinal System
1–2 SSPs <10 mm
Repeat colonoscopy in 5–10 yearsa
Assuming complete polyp resection
3–4 SSPs <10 mm
Repeat colonoscopy in 3–5 yearsa
Assuming complete polyp resection
5–10 SSPs <10 mm, or any single SSP ≥10 mm, or any
SSP with dysplasia
Repeat colonoscopy in 3 yearsa
Assuming complete polyp resection. Serrated polyposis
syndrome merits more frequent follow-up
Piecemeal removal of serrated polyp ≥1 cm
Exam in 2–6 months to verify complete removal
 
Colon cancer
Evaluate entire colon around the time of resection,
then repeat colonoscopy in 1 yeara
Subsequent colonoscopy in 3 years if the 1-year
examination is normal
Inflammatory Bowel Disease
Long-standing (>8 years) ulcerative pancolitis or
Crohn’s colitis, or left-sided ulcerative colitis
of >15 years’ duration
Colonoscopy with biopsies every 1–2 years
Consider chromoendoscopy or other advanced imaging
techniques for detection of flat dysplasia during
colonoscopy
Family History of Polyps or CRC
First-degree relatives with only small tubular adenomas
Same as average risk
 
One first-degree relative with CRC or advanced
adenoma at age ≥60 years
Begin screening starting at age 40, tests and
intervals per average-risk recommendations
One first-degree relative with CRC or advanced
adenoma at age <60 years, or two first-degree relatives
with CRC or advanced adenomas at any age
Colonoscopy every 5 years beginning at age
40 years or 10 years before the age at diagnosis of
the youngest affected relative, whichever is earlier
Familial adenomatous polyposis (FAP)
Sigmoidoscopy or colonoscopy annually, beginning
at age 10–12 years
Hereditary nonpolyposis colorectal cancer (HNPCC;
Lynch syndrome)   
 
 
Serrated polyposis syndrome (SPS)   
 
Colonoscopy every 2 years beginning at age 20–25
years (or 10 years younger than the youngest firstdegree relative was when diagnosed with CRC) until
age 40, then annually thereafter   
Colonoscopy at age 40 (or the same age at which
the youngest first-degree relative was when
diagnosed with SPS, or 10 years younger than the
youngest first-degree relative was when diagnosed
with CRC), then every 1–2 years thereafter 
Colonoscopy every 3–5 years beginning 10 years
before the age at diagnosis of the youngest affected
relative
 
 
Family colon cancer syndrome X
aAssumes good colonic preparation and complete examination to cecum. bHigh-risk adenoma: any adenoma ≥1 cm in size or containing high-grade dysplasia or villous features.
Abbreviations: CRC, colorectal cancer; CT, computed tomography; FIT, fecal immunochemical test; HSgFOBT, high-sensitivity guaiac fecal occult blood test; SSP, sessile
serrated polyp.
Sources: Adapted from U.S. Preventative Services Task Force Draft Guidelines finalized May 18, 2021 (https://www.uspreventiveservicestaskforce.org/uspstf/
recommendation/colorectal-cancer-screening) and American Cancer Society Guidelines (https://www.cancer.org/cancer/types/colon-rectal-cancer/detection-diagnosisstaging/acs-recommendations.html), both accessed on September 15, 2023, as well as the following articles: DK Rex et al: Colorectal cancer screening: Recommendations
for physicians and patients from the U.S. Multi-Society Task Force on Colorectal Cancer. Gastroenterology 153:307, 2017; S Gupta et al: Recommendations for follow-up after
colonoscopy and polypectomy: A consensus update by the US Multi-Society Task Force on Colorectal Cancer. Gastroenterology 158:1131, 2020; G Mankaney et al: Serrated
polyposis syndrome. Clin Gastroenterol Hepatol 18:777, 2020.
Less sensitive than colonoscopy; colonoscopy if results
are positive
Less sensitive than colonoscopy; colonoscopy if results
are positive
Assuming complete polyp resection
 
 
Consider genetic counseling and testing; consider
screening family members
Consider histologic evaluation for microsatellite
instability in tumor specimens of patients who meet
modified Bethesda criteria; consider genetic counseling
and testing, consider screening family members 
Consider screening family members, even of patients
with multiple serrated polyps who do not meet SPS
criteria

04 - 334 Diseases of the Esophagus
334 Diseases of the Esophagus
FIGURE 333-65  Virtual colonoscopy image of a colon polyp (arrow). (Image
courtesy of Dr. Jeff Fidler; with permission.)
FIGURE 333-66  Crohn’s ileitis. Edema, erythema, ulcers, and exudates involving the
terminal ileum.
FIGURE 333-67  Internal hemorrhoids with bleeding stigmata (arrow) as seen on
retroflexed view of the rectum.
■
■FURTHER READING
Ahmed O et al: AGA clinical practice update on the optimal manage­
ment of the malignant alimentary tract obstruction: expert review.
Clin Gastroenterol Hepatol 19:1780, 2021.
ASGE Standards of Practice Committee et al: Antibiotic prophy­
laxis for GI endoscopy. Gastrointest Endosc 81:81, 2015.
ASGE Standards of Practice Committee et al: Open-access endos­
copy. Gastrointest Endosc 81:1326, 2015.
Kaplan DE et al: AASLD practice guidance on risk stratification and
management of portal hypertension and varices in cirrhosis. Hepatol­
ogy 79:1180, 2024.
Laine L et al: ACG clinical guideline: Upper gastrointestinal and ulcer
bleeding. Am J Gastroenterol 116:899, 2021.
Sengupta N et al: Management of patients with acute lower gastro­
intestinal bleeding: An updated ACG guideline. Am J Gastroenterol
118:208, 2023.
Shaheen NJ et al: Guideline to practice: diagnosis and management of
Barrett’s esophagus: An updated ACG guideline. Am J Gastroenterol
117:1177, 2022.
Shaukat A et al: ACG clinical guidelines: colorectal cancer screening
2021. Am J Gastroenterol 116:458, 2021.
Peter J. Kahrilas, Ikuo Hirano*
Diseases of the
Esophagus
CHAPTER 334
Diseases of the Esophagus
ESOPHAGEAL STRUCTURE AND FUNCTION
The esophagus is a hollow, muscular tube coursing through the pos­
terior mediastinum joining the hypopharynx to the stomach with a
sphincter at each end. It functions to transport food and fluid between
these ends, otherwise remaining empty. The physiology of swallowing,
esophageal motility, and oral and pharyngeal dysphagia are described
in Chap. 47. Esophageal diseases can be manifested by impaired func­
tion or pain. Key functional impairments are swallowing disorders and
excessive gastroesophageal reflux. Pain, sometimes indistinguishable
from cardiac chest pain, can result from inflammation, infection, dys­
motility, or neoplasm.
SYMPTOMS OF ESOPHAGEAL DISEASE
The clinical history remains central to the evaluation of esophageal
symptoms. A thoughtfully obtained history will often expedite manage­
ment. Important details include weight gain or loss, gastrointestinal
bleeding, dietary habits including the timing of meals, smoking, and
alcohol consumption. The major esophageal symptoms are heartburn,
regurgitation, chest pain, dysphagia, odynophagia, and globus sensation.
Heartburn (pyrosis), the most common esophageal symptom, is
characterized by a discomfort or burning sensation behind the ster­
num that arises from the epigastrium and may radiate toward the
neck. Heartburn is usually an intermittent symptom, most commonly
experienced after eating, during exercise, and while lying recumbent.
The discomfort is relieved with drinking water or taking an antacid but
can occur frequently, interfering with normal activities including sleep.
The association between heartburn and gastroesophageal reflux dis­
ease (GERD) is so strong that empirical therapy for GERD has become
accepted management. However, the term heartburn is often misused
and/or referred to using other terms such as indigestion or repeating,
making it important to clarify the intended meaning.
*Deceased.
Regurgitation is the effortless return of food or fluid into the phar­
ynx without nausea or retching. Patients report a sour or burning fluid
in the throat or mouth that may also contain undigested food par­
ticles. Bending, belching, or maneuvers that increase intraabdominal
pressure can provoke regurgitation. A clinician needs to discriminate
among regurgitation, vomiting, and rumination. Vomiting is preceded
by nausea and accompanied by retching. Rumination is a behavior in
which recently swallowed food is regurgitated and then reswallowed
repetitively for up to an hour. Although there is some linkage between
rumination and cognitive deficiency, the behavior is also exhibited by
unimpaired individuals.
Chest pain is a common esophageal symptom with characteristics
similar to cardiac pain, sometimes making this distinction difficult.
Esophageal pain is usually experienced as a pressure-type sensation in
the mid chest, radiating to the mid back, arms, or jaws. The similarity
to cardiac pain is likely because the two organs share a nerve plexus
and the nerve endings in the esophageal wall have poor discriminative
ability among stimuli. Esophageal distention or even chemostimulation
(e.g., with acid) will often be perceived as chest pain. Gastroesophageal
reflux is the most common cause of esophageal chest pain.
Esophageal dysphagia (Chap. 47) is often described as a feeling of
food “sticking” or even lodging in the chest. Important distinctions are
between uniquely solid food dysphagia as opposed to liquid and solid,
episodic versus constant dysphagia, and progressive versus static dys­
phagia. If the dysphagia is for liquids as well as solid food, it suggests
a motility disorder such as achalasia. Conversely, uniquely solid food
dysphagia is suggestive of a stricture, web, ring, or tumor. Of note, a
patient’s localization of food hang-up in the esophagus is notoriously
imprecise. Approximately 30% of distal esophageal obstructions are
perceived as cervical dysphagia. In such instances, the absence of con­
comitant symptoms generally associated with oropharyngeal dysphagia
such as aspiration, nasopharyngeal regurgitation, cough, drooling, or
obvious neuromuscular compromise should suggest an esophageal
etiology.
PART 10
Disorders of the Gastrointestinal System
Odynophagia is pain either caused by or exacerbated by swallowing.
Although typically considered distinct from dysphagia, odynophagia
may manifest concurrently with dysphagia. Odynophagia is more
common with pill or infectious esophagitis than with reflux esophagitis
and should prompt a search for these entities. When odynophagia does
occur in GERD, it is likely related to an esophageal ulcer or extensive
erosions.
Globus sensation, also known as globus pharyngeus, is the per­
ception of a lump or fullness in the throat that is felt irrespective of
swallowing. Although such patients are frequently referred for an
evaluation of dysphagia, globus sensation is often relieved by the act
of swallowing. As implied by its alternative name, “globus hystericus,”
globus sensation often occurs in the setting of anxiety or obsessivecompulsive disorders. Clinical experience teaches that it is often attrib­
utable to GERD.
Water brash is excessive salivation resulting from a vagal reflex trig­
gered by acidification of the esophageal mucosa. This is not a common
symptom. Afflicted individuals will describe the unpleasant sensation
of the mouth rapidly filling with salty thin fluid, often in the setting of
concomitant heartburn.
DIAGNOSTIC STUDIES
■
■ENDOSCOPY
Endoscopy, also known as esophagogastroduodenoscopy (EGD), is
the most useful test for the evaluation of the proximal gastrointesti­
nal tract. Modern instruments produce high-quality, color images of
the esophageal, gastric, and duodenal lumen. Endoscopes also have
an instrumentation channel through which biopsy forceps, injection
catheters for focal delivery of therapeutic agents, balloon dilators, or
devices for hemostasis or removal of mucosal lesions can be used.
The key advantages of endoscopy over barium radiography are as fol­
lows: (1) increased sensitivity for the detection of mucosal lesions; (2)
vastly increased sensitivity for the detection of abnormalities mainly
identifiable by color, such as Barrett’s metaplasia or vascular lesions;
(3) the ability to obtain biopsy specimens for histologic examination of
suspected abnormalities; and (4) the ability to dilate strictures during
the examination. Submucosal (“third space”) endoscopy has emerged
as a diagnostic modality for assessment of subepithelial lesions, resec­
tion of superficial dysplastic areas, and therapy of esophageal motility
disorders. The main disadvantages of endoscopy are low sensitivity
for detection of very proximal esophageal strictures or narrow caliber
esophagus without focal stricturing, cost, and the need for sedatives or
anesthetics.
■
■RADIOGRAPHY
Contrast radiography of the esophagus, stomach, and duodenum
can demonstrate reflux of the contrast media, hiatal hernia, mucosal
granularity, erosions, ulcerations, and strictures. The sensitivity of
radiography compared with endoscopy for detecting reflux esophagitis
reportedly ranges from 22 to 95%, with higher grades of esophagitis
(i.e., ulceration or stricture) exhibiting greater detection rates. Con­
versely, the sensitivity of barium radiography for detecting esophageal
strictures is greater than that of endoscopy, especially when the study
is done in conjunction with a 13-mm barium tablet. Barium studies
also provide an assessment of esophageal function and morphology
that may be undetected on endoscopy. Tracheoesophageal fistula,
altered postsurgical anatomy, and extrinsic esophageal compression
are conditions where radiographic imaging complements endoscopic
assessment. Hypopharyngeal pathology and disorders of the cricopha­
ryngeus muscle are better appreciated on radiographic examination
than with endoscopy, particularly with rapid sequence or video fluoro­
scopic recording. The major shortcoming of barium radiography is that
it rarely obviates the need for endoscopy. Either a positive or a negative
study is usually followed by an endoscopic evaluation to obtain biop­
sies, provide therapy, clarify findings in the case of a positive examina­
tion, or add a level of certainty in the case of a negative examination.
■
■ENDOSCOPIC ULTRASOUND
Endoscopic ultrasound (EUS) instruments combine an endoscope with
an ultrasound transducer to create a transmural image of the tissue sur­
rounding the endoscope tip. The key advantage of EUS over alternative
radiologic imaging techniques is much greater resolution attributable
to the proximity of the ultrasound transducer to the area being exam­
ined. Available devices can provide either radial imaging (360-degree,
cross-sectional) or a curved linear image that can guide fine-needle
aspiration of imaged structures such as lymph nodes or tumors. Major
esophageal applications of EUS are to stage esophageal cancer, to evalu­
ate dysplasia in Barrett’s esophagus, and to assess submucosal lesions.
■
■ESOPHAGEAL MANOMETRY
Esophageal manometry, or motility testing, entails positioning a cath­
eter with multiple pressure sensors within the esophagus and then
observing the contractility following test swallows. The upper esopha­
geal sphincter and lower esophageal sphincter (LES) appear as zones
of high pressure that relax on swallowing, whereas the intersphinc­
teric esophagus exhibits peristaltic contractions. Manometry is used
to diagnose motility disorders (achalasia, diffuse esophageal spasm
[DES]) and to assess peristaltic integrity prior to the surgery for reflux
disease. Technologic advances have enhanced esophageal manom­
etry as high-resolution esophageal pressure topography (Fig. 334-1).
Manometry can also be combined with intraluminal impedance moni­
toring. Impedance recordings use a series of paired electrodes added
to the manometry catheter. Esophageal luminal contents in contact
with the electrodes decrease (liquid) or increase (air) the impedance
signal, allowing detection of anterograde or retrograde esophageal
bolus transit.
■
■FUNCTIONAL LUMEN IMAGING PROBE
The functional lumen imaging probe (FLIP) is a catheter-based tech­
nology that utilizes high-resolution impedance planimetry during
volume-controlled esophageal distension to measure esophageal crosssectional area and distensibility (i.e., cross-sectional area in relation to
distension pressure) along a 16-cm length of the esophagus. These data
FIGURE 334-1  High-resolution esophageal pressure topography (left) and conventional manometry (right) of a normal swallow. E, esophageal body; LES, lower esophageal
sphincter; UES, upper esophageal sphincter.
are displayed in real time on a computer monitor both as a cylinder of
varied diameter along the 16-cm length of the probe or as real-time
diameter topography akin to high-resolution manometry. FLIP studies
are performed with sedation in conjunction with upper endoscopy or
intraoperatively; the probe is passed transorally. Assessing esophageal
motility with FLIP is based on quantifying esophagogastric junction
opening during volumetric distention and characterizing distensioninduced esophageal contractility, which is, in essence, secondary peri­
stalsis. These findings are classified according to a scheme known as
FLIP panometry. There is excellent correlation between FLIP panom­
etry and high-resolution manometry in the detection of both achalasia
and normal contractility, the two most important potential findings.
Hence, FLIP can be used to evaluate patients with esophageal dyspha­
gia, often in lieu of manometry.
REFLUX TESTING
GERD is often diagnosed in the absence of endoscopic signs of esopha­
gitis, which would otherwise define the disease. This occurs in the
settings of partially treated disease, an abnormally sensitive esophageal
mucosa, or, most commonly, in nonerosive reflux disease. In such
instances, reflux testing can demonstrate excessive esophageal expo­
sure to refluxed gastric fluid, the physiologic abnormality of GERD.
This can be done by ambulatory 24- to 96-h esophageal pH recording
using either a wireless pH-sensitive transmitter that is affixed to the
esophageal mucosa or a transnasally positioned wire electrode with
the tip stationed in the distal esophagus. Either way, the outcome is
expressed as the percentage of the day that the pH was <4 (indicative
of recent acid reflux), with values exceeding 6% indicative of GERD.
Reflux testing is useful in the evaluation of patients presenting with
atypical symptoms or an inexplicably poor response to therapy. Intralu­
minal impedance monitoring can be added to pH monitoring to detect
reflux events irrespective of whether or not they are acidic, potentially
increasing the sensitivity of the study.
STRUCTURAL DISORDERS
■
■HIATAL HERNIA
Hiatal hernia is a herniation of viscera, most commonly the stomach,
into the mediastinum through the esophageal hiatus of the diaphragm.
Four types of hiatal hernia are distinguished, with type I, or sliding
hiatal hernia, comprising about 95% of the overall total. A sliding hiatal
hernia is one in which the gastroesophageal junction and gastric cardia
translocate cephalad as a result of weakening of the phrenoesophageal
ligament attaching the gastroesophageal junction to the diaphragm at
the hiatus and dilatation of the diaphragmatic hiatus. The incidence
of sliding hernia increases with age. True to its name, sliding hernias
enlarge with increased intraabdominal pressure, swallowing, and res­
piration. Conceptually, sliding hernias are the result of wear and tear:
increased intraabdominal pressure from abdominal obesity, pregnancy,
etc., along with hereditary factors predisposing to the condition. The
main significance of sliding hernias is the propensity of affected indi­
viduals to develop GERD.
Type II, III, and IV hiatal hernias are all subtypes of paraesophageal
hernia in which there is herniation into the mediastinum of viscera
above the gastric cardia. With type II and III paraesophageal hernias,
the gastric fundus herniates, with the distinction being that in type II, the
gastroesophageal junction remains fixed at the hiatus, whereas type III
is a combined sliding and paraesophageal hernia. With type IV hiatal
hernias, viscera other than the stomach herniate into the mediasti­
num, most commonly the colon. With type II and III paraesophageal
hernias, the stomach progressively inverts as it herniates, and large
paraesophageal hernias can lead to an “upside down stomach,” gastric
volvulus, and even strangulation of the stomach. Because of this risk,
surgical repair is often advocated for large paraesophageal hernias,
particularly when they are symptomatic.
CHAPTER 334
Diseases of the Esophagus
■
■RINGS AND WEBS
A lower esophageal mucosal ring, also called a B ring, is a thin membra­
nous narrowing at the squamocolumnar mucosal junction (Fig. 334-2).
Tubular
esophagus
Esophageal
vestibule
Phrenic
ampulla
Sliding hiatal
hernia
A ring
B ring
squamo-columnar
junction
Rugal folds
traversing hiatus
Diaphragmatic
impression
FIGURE 334-2  Radiographic anatomy of the gastroesophageal junction.
Its origin is unknown, but B rings are demonstrable in ~10–15% of the
general population and are usually asymptomatic. When the lumen
diameter is <13 mm, distal rings are usually associated with episodic
solid food dysphagia and are called Schatzki rings. Patients typically
present older than 40 years, consistent with an acquired rather than
congenital origin. Schatzki ring is a common cause of intermittent
food impaction, also known as “steakhouse syndrome” because meat is
a typical instigator. Symptomatic rings are readily treated by dilation.
Web-like constrictions higher in the esophagus can be of congeni­
tal or inflammatory origin. Asymptomatic cervical esophageal webs
are demonstrated in ~10% of people and typically originate along
the anterior aspect of the esophagus. Depending on the degree of
impingement, they can cause intermittent dysphagia to solids similar to
Schatzki rings and are similarly treated with dilation. The combination
of symptomatic proximal esophageal webs and iron-deficiency anemia
in middle-aged women constitutes Plummer-Vinson or Paterson-Kelly
syndrome.
■
■DIVERTICULA
Esophageal diverticula are categorized by location, with the most com­
mon being epiphrenic, hypopharyngeal (Zenker’s), and midesopha­
geal. Epiphrenic and Zenker’s diverticula are false diverticula involving
herniation of the mucosa and submucosa through the muscular layer
of the esophagus. These lesions result from increased intraluminal
pressure associated with distal obstruction. In the case of Zenker’s, the
obstruction is a stenotic cricopharyngeus muscle (upper esophageal
sphincter), and the hypopharyngeal herniation occurs in an area of
natural weakness proximal to the cricopharyngeus known as Killian’s
triangle (Fig. 334-3). Small Zenker’s diverticula are usually asymptom­
atic, but when they enlarge sufficiently to retain food and saliva, they
can be associated with dysphagia, halitosis, and aspiration. Treatment
is by surgical diverticulectomy and cricopharyngeal myotomy or tran­
soral, endoscopic myotomy.
PART 10
Disorders of the Gastrointestinal System
Epiphrenic diverticula are often associated with achalasia or esopha­
geal hypercontractile disorders. Midesophageal diverticula may be
caused by traction from adjacent inflammation (tuberculosis, histo­
plasmosis), in which case they are true diverticula involving all lay­
ers of the esophageal wall, or by pulsion associated with esophageal
motility disorders. Midesophageal and epiphrenic diverticula are often
asymptomatic; symptoms tend to correlate more with the underlying
esophageal disorder. Large epiphrenic diverticula can be removed
surgically, usually in conjunction with a myotomy. Diffuse intramural
esophageal pseudodiverticulosis is a rare entity resulting from dilata­
tion of the excretory ducts of submucosal esophageal glands (Fig. 334-4).
A
B
C
FIGURE 334-3  Examples of small (A) and large (B, C) Zenker’s diverticula arising from Killian’s triangle in the distal hypopharynx. Smaller diverticula are evident only
during the swallow, whereas larger ones retain food and fluid.
FIGURE 334-4  Intramural esophageal pseudodiverticulosis associated with
chronic obstruction. Invaginations of contrast into the esophageal wall outline deep
esophageal glands.
Esophageal candidiasis or strictures are commonly found in associa­
tion with pseudodiverticulosis.
■
■TUMORS
The American Cancer Society’s 2023 estimates for esophageal can­
cer in the United States are about 21,560 new cases (79% men) and
16,120 deaths. It is ~7 times less common than colorectal cancer but
kills about a third as many patients, emphasizing both the rarity and
lethality of esophageal cancer. One notable trend is the shift of domi­
nant esophageal cancer type from squamous cell to adenocarcinoma,
strongly linked to reflux disease and Barrett’s metaplasia. Other dis­
tinctions between cell types are the predilection for adenocarcinoma
to affect the distal esophagus in white males and for squamous cell
carcinoma to affect the more proximal esophagus in black males with
the added risk factors of smoking, alcohol consumption, caustic injury,
and human papillomavirus infection (Chap. 85).
The typical presentation of esophageal cancer is of progressive solid
food dysphagia and weight loss. Associated symptoms may include
odynophagia, iron deficiency, cough from tracheoesophageal fistula,
and hoarseness from left recurrent laryngeal nerve injury. Generally,
respiratory symptoms are manifestations of locally invasive or even
metastatic disease. Even when detected as a small lesion, esophageal
cancer has poor survival because of the abundant esophageal lymphat­
ics leading to early regional lymph node metastases.
Benign esophageal tumors are uncommon and usually discovered
incidentally. They include gastrointestinal stromal tumors, leiomyoma,
fibrovascular polyps, squamous papilloma, granular cell tumors, lipo­
mas, mesenchymal neoplasms, and inflammatory fibroid polyps.
CONGENITAL ANOMALIES
The most common congenital esophageal anomaly is esophageal
atresia, occurring in ~2.5 per 10,000 live births. Atresia can occur in
several permutations, the common denominator being developmental
failure of fusion between the proximal and distal esophagus associated
with a tracheoesophageal fistula, most commonly with the distal seg­
ment excluded. Alternatively, there can be an H-type configuration in
which esophageal fusion has occurred, but with a tracheoesophageal
fistula. Esophageal atresia is usually recognized and corrected surgi­
cally within the first few days of life. Later life complications include
dysphagia from anastomotic strictures or absent peristalsis and reflux,
which can be severe. Less common developmental anomalies include
congenital esophageal stenosis, webs, and duplications.
Dysphagia can also result from vascular congenital abnormalities
that cause extrinsic compression of the esophagus, a condition called
dysphagia lusoria. Most commonly, the esophagus is compressed by an
aberrant right subclavian artery arising from the descending aorta and
passing behind the esophagus. Alternatively, vascular rings may sur­
round and constrict the esophagus.
Heterotopic gastric mucosa, also known as an esophageal inlet
patch, is a focus of gastric-type epithelium in the proximal cervi­
cal esophagus; the estimated prevalence is 4–5%. The inlet patch is
thought to result from incomplete replacement of embryonic columnar
epithelium with squamous epithelium. The majority of inlet patches
are asymptomatic, but acid production along with associated symp­
toms and even ulceration can occur as most contain fundic-type gastric
epithelium with parietal cells.
ESOPHAGEAL MOTILITY DISORDERS
Esophageal motility disorders are diseases attributable to abnormal
esophageal neuromuscular dysfunction commonly associated with
dysphagia, chest pain, or heartburn. The major entities are achalasia,
distal esophageal spasm (DES), and GERD. Motility disorders can also
be secondary to systemic disease processes, as is the case with pseu­
doachalasia, Chagas’ disease, and scleroderma. Not included in this
discussion are diseases affecting the pharynx and proximal esophagus,
the impairment of which is almost always part of a more global neuro­
muscular disease process.
■
■ACHALASIA
Achalasia is a rare disease with a population incidence estimated to
be 1–3 per 100,000 and presentation usually occurring between age
25 and 60 years. Although there is some degree of heterogeneity in its
pathogenesis, the classic cases are caused by autoimmune-mediated
death of ganglion cells within the esophageal myenteric plexus; with
long-standing disease, aganglionosis is noted. The disease involves
both excitatory (cholinergic) and inhibitory (nitric oxide) ganglionic
neurons, which mediate deglutitive LES relaxation and the sequential
propagation of peristalsis. Their absence leads to impaired deglutitive
LES relaxation and absent peristalsis. Increasing evidence suggests that
the ultimate cause of ganglion cell degeneration in achalasia is an auto­
immune process attributable to a latent infection with human herpes
simplex virus 1 combined with genetic susceptibility.
Long-standing achalasia is characterized by progressive dilatation
and sigmoid deformity of the esophagus, sometimes associated with
hypertrophy of the LES. Clinical manifestations may include dysphagia,
regurgitation, chest pain, and weight loss. Most patients report solid
and liquid food dysphagia. Regurgitation occurs when food, fluid, and
secretions are retained in the dilated esophagus. Patients with advanced
achalasia are at risk for bronchitis, pneumonia, or lung abscess from
chronic regurgitation and aspiration. Chest pain may manifest early
in the course of achalasia. Patients describe a squeezing, pressure-like
retrosternal pain, sometimes radiating to the neck, arms, jaw, and back.
Paradoxically, some patients complain of heartburn that may be a chest
pain equivalent. Treatment of achalasia is less effective at alleviating
chest pain than it is in relieving dysphagia or regurgitation.
The differential diagnosis of achalasia includes DES, Chagas’ dis­
ease, opioid-induced esophageal dysmotility, and pseudoachalasia.
Chagas’ disease is endemic in areas of central Brazil, Venezuela, and
northern Argentina and spread by the bite of the reduviid (kissing) bug
that transmits the protozoan Trypanosoma cruzi. The chronic phase
of the disease develops years after infection and results from destruc­
tion of autonomic ganglion cells throughout the body, including the
heart, gut, urinary tract, and respiratory tract. Manometric features of
achalasia have been described in patients on chronic opioids and may
be confused with primary achalasia. Tumor infiltration, most com­
monly seen with carcinoma in the gastric cardia or distal esophagus,
can also mimic primary achalasia. The resultant “pseudoachalasia”
accounts for up to 5% of suspected cases and is more likely with
advanced age, abrupt onset of symptoms (<1 year), and weight loss.
Hence, endoscopy is a necessary part of the evaluation of achalasia.
When the clinical suspicion for pseudoachalasia is high and endoscopy
is nondiagnostic, computed tomography (CT) scanning or EUS may
be of value. Rarely, pseudoachalasia can result from a paraneoplastic
syndrome with circulating antineuronal antibodies.
CHAPTER 334
Diseases of the Esophagus
Achalasia is diagnosed by barium swallow x-ray and/or esopha­
geal manometry. Endoscopy excludes tumors or benign mechanical
strictures of the esophagogastric junction. The barium swallow x-ray
appearance is of a dilated esophagus with poor emptying, an air-fluid
level, and tapering at the LES giving it a beak-like appearance (Fig. 334-5).
Occasionally, an epiphrenic diverticulum is observed. In longstanding achalasia, the esophagus may assume a sigmoid configura­
tion. The diagnostic criteria for achalasia with esophageal manometry
FIGURE 334-5  Achalasia with esophageal dilatation, tapering at the
gastroesophageal junction, and an air-fluid level within the esophagus. The
example on the left shows sigmoid deformity with very advanced disease.
A. Classic achalasia
0 Pharynx
cm
–10
mmHg
35 Stomach
Seconds
B. Achalasia with compression
0 Pharynx
cm
–5
35 Stomach
–10
mmHg
PART 10
Disorders of the Gastrointestinal System
Seconds
mmHg
C. Spastic achalasia
0 Pharynx
cm
35 Stomach
–10
Seconds
FIGURE 334-6  Three subtypes of achalasia: classic (A), with esophageal
compression (B), and spastic achalasia (C) imaged with high-resolution manometry.
All are characterized by impaired lower esophageal sphincter (LES) relaxation and
absent peristalsis. However, classic achalasia has minimal pressurization of the
esophageal body, whereas substantial fluid pressurization is observed in achalasia
with esophageal compression, and spastic esophageal contractions are observed
with spastic achalasia.
are impaired LES relaxation and absent peristalsis. High-resolution
manometry has somewhat advanced this diagnosis; three subtypes of
achalasia are differentiated based on the pattern of pressurization in the
nonperistaltic esophagus (Fig. 334-6). Because manometry identifies
early disease before esophageal dilatation and food retention, it is the
most sensitive diagnostic test.
No method of preventing or “curing” achalasia is known. Therapy is
thus directed at reducing LES pressure so that gravity and esophageal
pressurization permit esophageal emptying. While peristalsis does
not recover, remnants of peristalsis masked by esophageal pressuriza­
tion and dilatation prior to therapy may be demonstrable following
effective treatment. LES pressure can be reduced by pharmacologic
therapy, pneumatic balloon dilation, or LES myotomy by means
of submucosal (third space) endoscopy or laparoscopic surgery.
Pharmacologic therapies are relatively ineffective but can be offered
as temporizing therapies. Nitrates or calcium channel blockers are
administered before eating but should be used with caution because of
their effects on blood pressure. Botulinum toxin, injected into the LES
under endoscopic guidance, inhibits acetylcholine release from nerve
endings and improves dysphagia in about two-thirds of cases for at
least 6 months. Sildenafil and alternative phosphodiesterase inhibitors
effectively decrease LES pressure, but practicalities limit their clinical
use in achalasia.
The only durable therapies for achalasia are pneumatic dilation and
LES myotomy. Pneumatic dilation, with a reported efficacy ranging
widely from 60 to 90%, is an endoscopic technique using a noncompli­
ant, cylindrical balloon dilator positioned across the LES and inflated
to a diameter of 3–4 cm. The major complication is perforation, with
a reported incidence of 0.5–5%. The most common surgical procedure
for achalasia is laparoscopic Heller myotomy, usually performed in
conjunction with an antireflux procedure (partial fundoplication);
good to excellent results are reported in 62–90% of cases. A European
randomized controlled trial demonstrated an equivalent response rate
of ~90% for both pneumatic dilation and laparoscopic Heller myotomy
at 5-year follow-up. Occasionally, patients with advanced disease fail
to respond to pneumatic dilation or Heller myotomy or relapse years
after response to primary therapy. In such refractory cases, esophageal
resection with gastric pull-up or interposition of a segment of trans­
verse colon may be the only option other than gastrostomy feeding.
An endoscopic approach to LES myotomy is increasingly available,
referred to as peroral endoscopic (or esophageal) myotomy (POEM).
This technique involves endoscopically incising the esophageal mucosa
and creating a tunnel in the submucosa of the esophageal wall through
which the circular muscle of the LES and a calibrated length of distal
esophagus are transected with electrocautery. GERD is common after
POEM but managed effectively with medications. Potential advantages
over the conventional laparoscopic approach include avoidance of
surgical disruption of the diaphragmatic hiatus and more rapid recov­
ery. An international, multicenter, randomized trial of POEM and
pneumatic dilation demonstrated greater symptom relief with POEM
compared to dilation at 2 and 5 years. A European, multicenter, ran­
domized trial of POEM and Heller myotomy reported similar efficacy
for symptom relief, exceeding 80% with either.
In untreated or inadequately treated achalasia, esophageal dilatation
predisposes to stasis esophagitis. Prolonged stasis esophagitis is the
likely explanation for the association between achalasia and esophageal
squamous cell cancer. Tumors develop after years of achalasia, usually
in the setting of extreme esophageal dilatation, with the overall squa­
mous cell cancer risk increased 17-fold compared to controls.
■
■DISTAL ESOPHAGEAL SPASM
DES is manifest by episodes of dysphagia and chest pain attribut­
able to abnormal esophageal contractions with normal deglutitive
LES relaxation. The pathophysiology and natural history of DES are
poorly defined. Radiographically, DES has been characterized by
tertiary contractions or a “corkscrew esophagus” (Fig. 334-7), but in
many instances, these abnormalities are indicative of achalasia. Mano­
metrically, a variety of defining features have been proposed including
uncoordinated (“spastic”) activity in the distal esophagus, spontaneous
and repetitive contractions, or high-amplitude and prolonged contrac­
tions. High-resolution manometry has defined DES by the occurrence
of contractions in the distal esophagus with short latency relative to the
time of the pharyngeal contraction, a dysfunction indicative of impair­
ment of inhibitory myenteric plexus neurons. When defined with this
restrictive criterion (Fig. 334-8), DES is substantially less common
than achalasia.
Esophageal chest pain closely mimics angina pectoris. Features sug­
gesting esophageal pain include pain that is nonexertional, prolonged,
meal-related, relieved with antacids, accompanied by heartburn, dys­
phagia, or regurgitation, and interrupts sleep. However, all of these fea­
tures exhibit overlap with cardiac pain, which still must be the primary
consideration. Furthermore, even within the spectrum of esophageal
diseases, both chest pain and dysphagia are also characteristic of peptic
FIGURE 334-7  Distal esophageal spasm. The characteristic “corkscrew”
esophagus results from spastic contraction of the circular muscle in the esophageal
wall; more precisely, this is actually a helical array of muscle. These findings are
also seen with spastic achalasia.
or infectious esophagitis. Only after these more common entities have
been excluded by evaluation and/or treatment should a diagnosis of
DES be pursued.
Although DES is diagnosed by manometry, endoscopy is useful to
identify alternative structural and inflammatory lesions that may cause
chest pain. Radiographically, a “corkscrew esophagus,” “rosary bead
esophagus,” pseudodiverticula, or curling can be indicative of DES,
but these are also found with spastic achalasia. Given these vagaries of
defining DES and the resultant heterogeneity of patients identified for
inclusion in therapeutic trials, it is not surprising that trial results have
been disappointing. Only small, uncontrolled trials exist, reporting
response to nitrates, calcium channel blockers, hydralazine, botulinum
toxin, and anxiolytics. POEM with distal esophageal myotomy or sur­
gical myotomy should be considered only with severe weight loss or
intractable pain. These indications are extremely rare.
■
■NONSPECIFIC MANOMETRIC FINDINGS
Manometric studies done to evaluate chest pain and/or dysphagia
often report minor abnormalities (e.g., hypertensive or hypotensive
peristalsis, hypertensive LES) that are insufficient to diagnose either
mmHg
Jackhammer esophagus
Time (s)
Time (s)
Normal latency with hypercontractility
Short latency, premature contraction
FIGURE 334-8  Esophageal pressure topography of the two major variants of esophageal spasm: hypercontractile
esophagus (left) and distal esophageal spasm (right). Hypercontractile esophagus is defined by the extraordinarily
vigorous, sometimes repetitive contractions with normal peristaltic onset and normal latency of the contraction.
Distal esophageal spasm is similar but primarily defined by a short latency (premature) contraction.
achalasia or DES. These findings are of unclear significance. Reflux
and psychiatric diagnoses, particularly anxiety and depression, are
common among such individuals. A lower visceral pain threshold and
symptoms of irritable bowel syndrome are noted in more than half of
such patients. Consequently, therapy for these individuals should tar­
get either the most common esophageal disorder, GERD, or cognitive
disorders that may be present.
GASTROESOPHAGEAL REFLUX DISEASE
The current concept of GERD is that it encompasses a family of
conditions with the commonality that they are caused by gastro­
esophageal reflux resulting in either troublesome symptoms or an
array of potential esophageal and extraesophageal manifestations. It
is estimated that 10–15% of adults in the United States are affected
by GERD, although such estimates are based on population studies of
self-reported chronic heartburn. With respect to the esophagus, the
spectrum of injury includes esophagitis, stricture, Barrett’s esophagus,
and adenocarcinoma (Fig. 334-9). Of particular concern is the rising
incidence of esophageal adenocarcinoma, an epidemiologic trend that
parallels the increasing incidence of GERD. About 9200 incident cases
of esophageal adenocarcinoma were noted in the United States in 2020
(estimated as half of all esophageal cancers); this disease burden has
increased sixfold in the past 20 years.
■
■PATHOPHYSIOLOGY
The best-defined subset of GERD patients, albeit a minority overall,
have esophagitis. Esophagitis occurs when refluxed gastric acid and
pepsin induce inflammation of the esophageal mucosa that leads to
microscopic injury and macroscopic erosions or ulcers. Experimental
evidence supports a cytokine-mediated inflammatory pathway rather
than direct caustic injury to the esophageal epithelium. Note that some
degree of gastroesophageal reflux is normal, physiologically inter­
twined with the mechanism of belching (transient LES relaxation),
but esophagitis results from excessive reflux, often accompanied
by impaired clearance/neutralization of the refluxed gastric juice.
Restricting reflux to that which is physiological levels depends on the
anatomic and physiologic integrity of the esophagogastric junction, a
complex valvular mechanism functionally dependent on the LES, the
surrounding crural diaphragm, and the native subdiaphragmatic endto-side architecture of the gastroesophageal junction. Pathologic reflux
is a consequence of the interplay between progressive anatomic distor­
tion of the native architecture of the gastroesophageal junction and
physiology. Relevant factors include (1) widening of the diaphragmatic
hiatus and diminished crural diaphragm sphincteric function; (2) loss
of an intraabdominal esophageal segment with complete disabling of
the native flap valve architecture; (3) axial hiatus hernia, which causes
CHAPTER 334
Diseases of the Esophagus
LES hypotension and loss of the ability to
prevent reflux during swallow-induced LES
relaxation, inspiration, or straining; and
(4) increased compliance of gastroesopha­
geal junction leading to wider sphincter
opening during transient LES relaxations
(the belch reflex) with loss of the ability to
prevent liquid escaping along with the gas.
Factors tending to exacerbate reflux regard­
less of mechanism are abdominal obesity,
pregnancy, gastric hypersecretory states,
delayed gastric emptying, disruption of
esophageal peristalsis, and gluttony.
Diffuse esophageal spasm
Latency= 3.5 s
After acid reflux, peristalsis returns the
refluxed fluid to the stomach, and acid
clearance is completed by titration of the
residual acid by bicarbonate contained in
swallowed saliva. Consequently, two causes
of prolonged acid clearance are impaired
peristalsis and reduced salivation. Impaired
peristaltic emptying can be attributable
to disrupted peristalsis or superimposed
reflux associated with a hiatal hernia. With
Esophageal stricture with chronic
erosive esophagitis
A
B
Erosive esophagitis
PART 10
Disorders of the Gastrointestinal System
Barrett’s esophagus
Esophageal adenocarcinoma
with Barrett’s esophagus
C
D
FIGURE 334-9  Endoscopic appearance of (A) peptic esophagitis, (B) a peptic stricture, (C) Barrett’s
metaplasia, and (D) adenocarcinoma developing within an area of Barrett’s esophagus.
superimposed reflux, fluid retained within a sliding hiatal hernia
refluxes back into the esophagus during swallow-related LES relax­
ation, a phenomenon that does not normally occur.
Inherent in the pathophysiologic model of GERD is that gastric
juice is harmful to the esophageal epithelium. However, gastric acid
hypersecretion is usually not a dominant factor in the development
of esophagitis. An obvious exception is with Zollinger-Ellison syn­
drome, which is associated with severe esophagitis in ~50% of patients.
Another caveat is with chronic Helicobacter pylori gastritis, which may
have a protective effect by inducing atrophic gastritis with concomitant
hypoacidity. Pepsin, bile, and pancreatic enzymes within gastric secre­
tions can also injure the esophageal epithelium, but their noxious prop­
erties are either lessened without an acidic environment or dependent
on acidity for activation. Bile warrants attention because it persists in
refluxate despite acid-suppressing medications. Bile can traverse the
cell membrane, imparting severe cellular injury in a weakly acidic envi­
ronment, and has also been invoked as a cofactor in the pathogenesis
of Barrett’s metaplasia and adenocarcinoma. Hence, the causticity of
gastric refluxate extends beyond hydrochloric acid.
■
■SYMPTOMS
Heartburn and regurgitation are the typical symptoms of GERD.
Somewhat less common are dysphagia and chest pain. In each case,
multiple potential mechanisms for symptom genesis operate that
extend beyond the basic concepts of mucosal erosion and activation of
afferent sensory nerves. Specifically, visceral hypersensitivity is increas­
ingly recognized as a cofactor. Nonetheless, the dominant clinical
strategy is empirical treatment with acid inhibitors, reserving further
evaluation for those who fail to respond. Important exceptions to this
are patients with chest pain or persistent dysphagia, each of which may
be indicative of GERD or alternative diagnoses. With chest pain, car­
diac disease must be considered. With dysphagia, eosinophilic esopha­
gitis (EoE) should be considered as an alternative diagnosis along with
peptic stricture, motility disorders, or esophageal cancer.
Extraesophageal syndromes associated with
GERD include chronic cough, laryngitis, asthma,
and dental erosions. Other conditions including
pharyngitis, chronic bronchitis, pulmonary fibrosis,
chronic sinusitis, cardiac arrhythmias, sleep apnea,
and recurrent aspiration pneumonia have proposed
associations with GERD. However, in both cases, it
is important to emphasize the word association as
opposed to causation. In many instances, the disor­
ders likely coexist without causality. Potential mech­
anisms for extraesophageal GERD manifestations
are either regurgitation with direct contact between
the refluxate and supraesophageal structures or via a
vagovagal reflex wherein reflux activation of esopha­
geal afferent nerves triggers efferent vagal reflexes
such as bronchospasm, cough, or arrhythmias.
■
■DIFFERENTIAL DIAGNOSIS
Although generally quite characteristic, symptoms
from GERD need to be distinguished from symp­
toms related to infectious or pill esophagitis, EoE,
peptic ulcer disease, dyspepsia, biliary colic, coronary
artery disease, and esophageal motility disorders. It
is especially important that coronary artery disease
be given early consideration because of its potentially
lethal implications. The remaining elements of the
differential diagnosis can be addressed by endos­
copy, upper gastrointestinal series, or esophageal
manometry as appropriate. Erosive esophagitis at the
esophagogastric junction is the endoscopic hallmark
of GERD, but it is identified in only about one-third of
GERD patients. The distinction among etiologies of
esophagitis is readily made by endoscopic appear­
ance, but mucosal biopsies are necessary to evaluate
for infectious or eosinophilic inflammation. In terms
of endoscopic appearance, the ulcerations seen in peptic esophagitis
are usually few and distal, whereas infectious ulcerations are numer­
ous, punctate, and diffuse. EoE characteristically exhibits multiple
esophageal edema, rings, linear furrows, white punctate exudate, and
strictures. Esophageal ulcerations from pill esophagitis are usually
singular and deep at points of luminal narrowing, especially near the
carina, with sparing of the distal esophagus.
■
■COMPLICATIONS
The complications of GERD are related to chronic esophagitis (bleed­
ing and stricture) and the relationship between GERD and esophageal
adenocarcinoma. However, both erosive esophagitis and peptic stric­
tures have become increasingly rare in the era of potent antisecretory
medications. Conversely, the most severe histologic consequence of
GERD is Barrett’s metaplasia with the associated risk of esophageal
adenocarcinoma, and the incidence of these lesions has increased, not
decreased, in the era of potent acid suppression. Barrett’s metapla­
sia, recognized endoscopically by salmon-colored mucosa extending
proximally from the gastroesophageal junction (Fig. 334-9) or his­
topathologically by the finding of intestinal metaplasia, is associated
with a significantly increased risk for development of esophageal
adenocarcinoma.
Barrett’s metaplasia can progress to adenocarcinoma through the
intermediate stages of low- and high-grade dysplasia (Fig. 334-10).
Owing to this risk, areas of Barrett’s metaplasia and especially any
included areas of mucosal irregularity should be carefully inspected
and extensively biopsied. The rate of cancer development is estimated
at 0.1–0.3% per year, but vagaries in definitional criteria and of the
extent of Barrett’s metaplasia requisite to establish the diagnosis have
contributed to variability and inconsistency in this risk assessment.
The group at greatest risk is obese white males in their sixth decade
of life. However, despite common practice, the utility of endoscopic
screening and surveillance programs intended to control the adenocar­
cinoma risk remains an open question. Also of note, although in a large
Barrett’s metaplasia
High-grade dysplasia
Alcian blue stain
H&E stain
FIGURE 334-10  Histopathology of Barrett’s metaplasia and Barrett’s metaplasia with high-grade dysplasia (arrows). H&E, hematoxylin and eosin.
randomized, controlled trial of chemoprevention in Barrett’s patients,
high-dose proton pump inhibitor therapy along with aspirin did sig­
nificantly better at achieving the composite endpoint of delaying allcause mortality or Barrett’s progression, the effect was driven mainly
by improved overall survival rather than reduced Barrett’s dysplasia or
adenocarcinoma development.
Although the management of Barrett’s esophagus remains contro­
versial, the finding of dysplasia in Barrett’s, particularly high-grade
dysplasia, mandates further intervention. In addition to the high rate
of progression to adenocarcinoma, there is also a high prevalence of
unrecognized coexisting cancer with high-grade dysplasia. Treatment
recommendations for Barrett’s esophagus with high-grade dysplasia
have evolved over the past several years. Historically, esophagectomy
was the gold standard treatment for high-grade dysplasia. However,
esophagectomy has a mortality ranging from 3 to 10%, along with
substantial morbidity. Prospective studies have demonstrated the
efficacy of endoscopic mucosal ablation therapy with substantially less
morbidity and essentially no mortality. Consequently, current societal
guidelines endorse endoscopic mucosal ablation therapies for the man­
agement of high-grade dysplasia.
TREATMENT
Gastroesophageal Reflux Disease
Lifestyle modifications are routinely advocated as GERD therapy.
Broadly speaking, these fall into three categories: (1) avoidance
of foods that reduce LES pressure, making them “refluxogenic”
(these commonly include fatty foods, alcohol, spearmint, pep­
permint, and possibly coffee and tea); (2) avoidance of acidic
foods that are inherently irritating (citrus fruits, tomato-based
foods); and (3) adoption of behaviors to minimize reflux and/or
heartburn. In general, minimal evidence supports the efficacy of
these measures. However, clinical experience dictates that subsets
of patients benefit from specific recommendations based on their
individual history and symptom profile. A patient with sleep dis­
turbance from nighttime heartburn is more likely to benefit from
elevation of the head of the bed and avoidance of eating before
retiring. The most broadly applicable recommendation is for
weight reduction. Even though the benefit with respect to reflux
cannot be assured, the strong epidemiologic relationship between
body mass index and GERD and the secondary health gains of
weight reduction is beyond dispute.
The dominant pharmacologic approach to GERD management
is with inhibitors of gastric acid secretion, and abundant data sup­
port the effectiveness of this approach. Pharmacologically reducing
the acidity of gastric juice does not prevent reflux, but it ameliorates
reflux symptoms and allows esophagitis to heal. The hierarchy
of effectiveness among pharmaceuticals for healing esophagitis
parallels their antisecretory potency. Potassium competitive acid
blockers (PCABs) are more efficacious than proton pump inhibi­
tors (PPIs), which are more efficacious than histamine-2 receptor
antagonists (H2RAs). The differences are most evident with severe
esophagitis, much less so with mild disease.
Paradoxically, the perceived frequency and severity of heartburn
correlate poorly with the presence or severity of esophagitis. When
GERD treatments are assessed in terms of resolving heartburn, both
efficacy and differences among pharmaceuticals are less clear-cut
than with the objective of healing esophagitis. Although the same
overall hierarchy of effectiveness exists between PPIs and H2RAs
(minimal data are available for PCABs), observed efficacy rates are
lower and vary widely, likely reflecting patient heterogeneity.
Reflux symptoms tend to be chronic, irrespective of esophagitis.
Thus, a common management strategy is indefinite treatment with
PPIs or H2RAs as necessary for symptom control. The side effects of
PPI therapy are generally minimal. Rare cases of interstitial nephri­
tis and severe, reversible hypomagnesemia have been reported.
Vitamin B12 and iron absorption may be compromised and suscep­
tibility to enteric infections, particularly Clostridioides difficile coli­
tis, increased with treatment. Observational data have also noted
an association between PPI exposure and renal disease, dementia,
and cardiovascular disease, but the hazard ratios reported in these
studies were small, and the potential for unrecognized residual
confounding bias was substantial. Population studies have also sug­
gested a slight increased risk of bone fracture with chronic PPI use
suggesting an impairment of calcium absorption, but prospective
studies have failed to corroborate this. Nonetheless, as with any
medication, PPI dosage should be minimized to that necessary for
the clinical indication.
CHAPTER 334
Diseases of the Esophagus
Laparoscopic fundoplication, wherein the proximal stomach is
wrapped around the distal esophagus to create an antireflux bar­
rier, is a surgical alternative to the management of chronic GERD.
Just as with PPI therapy, evidence on the utility of fundoplication
is strongest for treating esophagitis, and controlled trials suggest
similar efficacy to PPI therapy. However, the benefits of fundoplica­
tion must be weighed against potential deleterious effects, including
surgical morbidity and mortality, postoperative dysphagia, failure
or breakdown requiring reoperation, an inability to belch, and
increased bloating, flatulence, and bowel symptoms after surgery.
■
■EOSINOPHILIC ESOPHAGITIS
EoE is increasingly recognized in adults and children around the
world. Current prevalence estimates in the United States identified
4–8 cases per 10,000 with a predilection for white males between 30
and 40 years of age. The increasing prevalence of EoE is attributable
to a combination of an increasing incidence and a growing recogni­
tion of the condition. There is also an incompletely understood, but
important, interaction between EoE and GERD that may confound the
diagnosis of the disease. Genome-wide analysis studies demonstrated
susceptibility elements at 5q22 (thymic stromal lymphopoietin) and
2p23 (CAPN14) in EoE.
EoE is diagnosed by the combination of esophageal symptoms and
esophageal mucosal biopsies demonstrating eosinophil-predominant
inflammation. Alternative etiologies of esophageal eosinophilia include
GERD, drug hypersensitivity, connective tissue disorders, hypereo­
sinophilic syndrome, Crohn’s disease, eosinophilic gastroenteritis, and
infection. EoE is an immunologic disorder induced by antigen sensiti­
zation in susceptible individuals. Food allergens are the dominant trig­
gers, although aeroallergens may also contribute. The natural history
of EoE is incompletely understood, but an increased risk of esophageal
stricture development paralleling the duration of untreated disease has
been noted.
EoE should be strongly considered in children and adults with
dysphagia and esophageal food impactions. In preadolescent chil­
dren, symptom presentations of EoE include chest or abdominal
pain, nausea, vomiting, and food aversion. Other symptoms in adults
may include atypical chest pain and heartburn. An atopic history of
IgE-mediated food allergy, asthma, eczema, and/or allergic rhinitis is
present in the majority of patients. Peripheral blood eosinophilia is
demonstrable in 25–50% of patients, but the specificity of this finding
is problematic in the setting of concomitant atopy. The characteristic
endoscopic esophageal findings include loss of vascular markings
(edema), multiple esophageal rings, longitudinally oriented furrows,
whitish exudate, and strictures (Fig. 334-11). Histologic confirmation
is made with the demonstration of esophageal mucosal eosinophilia
(peak density ≥15 eosinophils per high-power field) (Fig. 334-12).
Complications of EoE in adolescents and adults include food impac­
tion, esophageal stricture, narrow-caliber esophagus, and (rarely)
esophageal perforation. In children, complications include feeding
difficulties and weight loss.
Since the complications of EoE are a consequence of ongoing
eosinophilic inflammation, preventing them is dependent on follow
up endoscopy with biopsies to verify a selected therapy’s effectiveness
in controlling eosinophilic inflammation. Primary therapy often starts
with a PPI, which is effective at improving eosinophilic inflammation
in 30–50% of patients. Additional first-line therapies include elimina­
tion diets or swallowed topical glucocorticoids. Elemental formula
diets devoid of allergenic protein are a highly effective therapy but
are limited by palatability. Notably, allergy testing by means of either
serum IgE or skin prick testing has demonstrated poor sensitivity
and specificity in the identification of foods responsible for EoE in
an individual patient. Empiric elimination of common food aller­
gies (milk, wheat, egg, soy, nuts, and seafood) followed by systematic
PART 10
Disorders of the Gastrointestinal System
A
B
C
D
FIGURE 334-11  Endoscopic features of (A) eosinophilic esophagitis (EoE), (B)
Candida esophagitis, (C) giant ulcer associated with HIV, and (D) a Schatzki ring.
FIGURE 334-12  Histopathology of eosinophilic esophagitis (EoE) showing
infiltration of the esophageal squamous epithelium with eosinophils. Additional
features of basal cell hyperplasia and lamina propria fibrosis are present.
Eosinophilic inflammation can also be seen with gastroesophageal reflux disease.
reintroduction has been an effective diet therapy in both children and
adults. The intent of the elimination diet approach is the identification
and long-term avoidance of specific food trigger(s). Swallowed, topical
glucocorticoids (e.g., fluticasone propionate or budesonide) are effec­
tive in 50–80% of patients. Systemic glucocorticoids are not generally
recommended due to side effects and lack of proven benefit beyond
that achieved with topical glucocorticoids. The first U.S. Food and
Drug Administration–approved treatment for EoE is the interleukin
4 and 13–blocking biologic dupilumab, injected subcutaneously once
a week. Additional therapies targeting other immune pathways and
allergic cytokine mediators have also shown promise in initial clinical
trials. Esophageal dilation is highly effective at relieving dysphagia in
patients with fibrostenosis but does not address the underlying inflam­
matory process. Dilation should be approached cautiously because
of the risk of deep, esophageal mural laceration or perforation in the
characteristic stiff-walled esophagus. Once an effective therapy is
identified, maintenance therapy is advocated due to the chronicity of
EoE and nearly universal disease recurrence upon cessation of therapy.
INFECTIOUS ESOPHAGITIS
With the increased use of immunosuppression for organ transplanta­
tion and chronic inflammatory diseases, use of chemotherapy, and
the AIDS epidemic, infections with Candida species, herpesvirus, and
cytomegalovirus (CMV) have become relatively common. Although
rare, infectious esophagitis also occurs among the non-immunocom­
promised, with herpes simplex and Candida albicans being the most
common pathogens. Among AIDS patients, infectious esophagitis
becomes more common as the CD4 count declines; cases are rare with
a CD4 count >200 and common when <100. HIV itself may also be
associated with a self-limited syndrome of acute esophageal ulceration
with oral ulcers and a maculopapular skin rash at the time of sero­
conversion. Additionally, some patients with advanced disease have
deep, persistent esophageal ulcers treated with oral glucocorticoids or
thalidomide. However, with the widespread use of highly effective anti­
viral therapies, these HIV complications have become less common.
Regardless of the infectious agent, odynophagia is a characteristic
symptom of infectious esophagitis; dysphagia, chest pain, and hem­
orrhage are also common. Odynophagia is uncommon with reflux
esophagitis, so its presence should always raise suspicion of an alterna­
tive etiology.
■
■CANDIDA ESOPHAGITIS
Candida is normally found in the throat but can become pathogenic
and produce esophagitis in a compromised host; C. albicans is most
common. Candida esophagitis also occurs with esophageal stasis
secondary to esophageal motor disorders and diverticula. Patients
complain of odynophagia and dysphagia. If oral thrush is present,
empirical therapy is appropriate, but co-infection is common, and per­
sistent symptoms should lead to prompt endoscopy with biopsy, which
is the most useful diagnostic evaluation. Candida esophagitis has a
characteristic appearance of white plaques or exudate with friability.
Oral fluconazole (400 mg on the first day, followed by 200 mg daily) for
7–14 days is the preferred initial treatment. Patients refractory to that
may respond to higher dose fluconazole, voriconazole, itraconazole, or
posaconazole. Alternatively, poorly responsive patients or those who
cannot swallow medications can be treated with an intravenous echi­
nocandin or amphotericin B.
■
■HERPETIC ESOPHAGITIS
Herpes simplex virus type 1 or 2 may cause esophagitis. Vesicles on
the nose and lips may coexist and are suggestive of a herpetic etiology.
Varicella-zoster virus can also cause esophagitis in children with chick­
enpox or adults with zoster. The characteristic endoscopic findings
are vesicles and small, superficial ulcerations. Because herpes simplex
infections are limited to squamous epithelium, biopsies from the ulcer
margins are most likely to reveal the characteristic ground-glass nuclei,
eosinophilic Cowdry’s type A inclusion bodies, and giant cells. Cul­
ture or polymerase chain reaction (PCR) assays are helpful to identify
acyclovir-resistant strains. Acyclovir (200 mg orally five times a day for
7–10 days) can be used for immunocompetent hosts, although the dis­
ease is typically self-limited after a 1- to 2-week period in such patients.
Immunocompromised patients are treated with acyclovir (400 mg orally
five times a day for 14–21 days), famciclovir (500 mg orally three times
a day), or valacyclovir (1 g orally three times a day). In patients with
severe odynophagia, intravenous acyclovir, 5 mg/kg every 8 h for
7–14 days, reduces this morbidity.
■
■CYTOMEGALOVIRUS
CMV esophagitis occurs primarily in immunocompromised patients,
particularly those with HIV, malignancy, and recipients of bone mar­
row or organ transplants. CMV is usually reactivated from a latent
infection. Endoscopically, CMV lesions appear as large serpiginous
ulcers in an otherwise normal mucosa, particularly in the distal esoph­
agus. Biopsies from the ulcer bases have the greatest diagnostic yield
for finding the pathognomonic large nuclear or cytoplasmic inclusion
bodies. Immunohistology with monoclonal antibodies to CMV and in
situ hybridization tests are useful for early diagnosis. Data on therapy
for CMV esophagitis are limited. Treatment studies of CMV gastroin­
testinal disease have demonstrated effectiveness of both ganciclovir
(5 mg/kg every 12 h IV) and valganciclovir (900 mg orally every 12 h).
Therapy is continued until healing, which may take 3–6 weeks. Mainte­
nance therapy may be needed for patients with relapsing disease.
MECHANICAL TRAUMA AND
IATROGENIC INJURY 
■
■ESOPHAGEAL PERFORATION
Most cases of esophageal perforation are from instrumentation of
the esophagus or trauma. Alternatively, forceful vomiting or retch­
ing can lead to spontaneous rupture at the gastroesophageal junction
(Boerhaave’s syndrome). More rarely, corrosive esophagitis or neo­
plasms lead to perforation. Instrument perforation from endoscopy
or nasogastric tube placement typically occurs in the hypopharynx
or at the gastroesophageal junction. Perforation may also occur at the
site of a stricture in the setting of endoscopic food disimpaction or
therapeutic esophageal dilation. Esophageal perforation causes pleu­
ritic retrosternal pain often associated with pneumomediastinum and
subcutaneous emphysema. Mediastinitis is a major complication of
esophageal perforation, and prompt recognition is key to optimizing
outcome. CT of the chest is most sensitive in detecting mediastinal
air. Esophageal perforation is confirmed by a contrast swallow, usually
Gastrografin followed by thin barium. Treatment includes nasogastric
suction and parenteral broad-spectrum antibiotics with prompt surgi­
cal drainage and repair in noncontained leaks. Conservative therapy
with nasogastric suction, NPO status, and antibiotics without surgery
may be appropriate in cases of contained perforation that are detected
early. Endoscopic clipping or stent placement may be indicated in
nonoperated iatrogenic perforations or nonoperable cases such as
perforated tumors.
■
■MALLORY-WEISS TEAR
Vomiting, retching, or vigorous coughing can cause a nontransmural
tear at the gastroesophageal junction that is a common cause of upper
gastrointestinal bleeding. Most patients present with hematemesis.
Antecedent vomiting is the norm but not always evident. Bleeding
usually abates spontaneously, but protracted bleeding may respond
to local epinephrine or cauterization therapy, endoscopic clipping, or
angiographic embolization. Surgery is rarely needed.
■
■RADIATION ESOPHAGITIS
Radiation esophagitis can complicate treatment for thoracic cancers,
especially breast and lung cancers, with the risk proportional to radia­
tion dosage. Radiosensitizing drugs such as doxorubicin, bleomycin,
cyclophosphamide, and cisplatin increase the risk. Dysphagia and
odynophagia may last weeks to months after therapy. The esophageal
mucosa becomes erythematous, edematous, and friable. Submucosal
fibrosis and degenerative tissue changes and stricturing may occur
years after the radiation exposure. Radiation exposure in excess of 5000
cGy has been associated with increased risk of esophageal stricture.
Treatment for acute radiation esophagitis is supportive. Chronic stric­
tures are managed with esophageal dilation.
CHAPTER 334
Diseases of the Esophagus
■
■CORROSIVE ESOPHAGITIS
Caustic esophageal injury from ingestion of alkali or, less commonly,
acid can be accidental or from attempted suicide. Absence of oral
injury does not exclude possible esophageal involvement. Thus, early
endoscopic evaluation is recommended to assess and grade injury to
the esophageal mucosa. Severe corrosive injury may lead to esopha­
geal perforation, bleeding, stricture, and death. Glucocorticoids have
not been shown to improve the clinical outcome of acute corrosive
esophagitis and are not recommended. Healing of more severe grades
of caustic injury is commonly associated with severe stricture forma­
tion and often requires life-long repeated dilation.
■
■PILL ESOPHAGITIS
Pill-induced esophagitis occurs when a swallowed pill fails to traverse
the entire esophagus and lodges within the lumen. Generally, this
is attributed to poor “pill-taking habits”: inadequate liquid with the
pill or lying down immediately after taking a pill. The most common
location for the pill to lodge is in the mid-esophagus near the crossing
of the aorta or carina. Extrinsic compression from these structures
halts the movement of the pill or capsule. Since initially reported in 1970,
thousands of cases of pill esophagitis have been reported, suggesting
that this is not an unusual occurrence. A wide variety of medications
are implicated, with the most common being doxycycline, tetracycline,
quinidine, phenytoin, potassium chloride, ferrous sulfate, nonsteroidal
anti-inflammatory drugs (NSAIDs), and bisphosphonates.
Typical symptoms of pill esophagitis are the sudden onset of chest
pain and odynophagia. Characteristically, the pain will develop over
a period of hours or will awaken the individual from sleep. A classic
history in the setting of ingestion of a recognized pill offender obviates

05 - 335 Peptic Ulcer Disease and Related Disorders
335 Peptic Ulcer Disease and Related Disorders
the need for diagnostic testing in most patients. When endoscopy is
performed, localized ulceration or inflammation is evident. Histologi­
cally, acute inflammation is typical. Chest CT imaging will sometimes
reveal esophageal thickening consistent with transmural inflammation.
Although the condition usually resolves within days to weeks, symp­
toms may persist for months and stricture can develop in severe cases.
No specific therapy is known to hasten the healing process, but antise­
cretory medications are frequently prescribed to remove concomitant
reflux as an aggravating factor. When healing results in stricture forma­
tion, dilation is indicated.
■
■FOREIGN BODIES AND FOOD IMPACTION
Food or foreign bodies may lodge in the esophagus, causing complete
obstruction, which in turn can cause an inability to handle secretions
(foaming at the mouth) and severe chest pain. Food impaction may
occur due to peptic stricture, carcinoma, Schatzki ring, EoE, achala­
sia, or simply inattentive eating. If it does not resolve spontaneously,
impacted food should be removed endoscopically. Use of meat ten­
derizer enzymes to facilitate passage of a meat bolus is discouraged
because of potential esophageal injury. Glucagon (1 mg IV) is some­
times tried before endoscopic dislodgement. After emergent treatment,
patients should be evaluated for potential causes of the impaction with
treatment rendered as indicated.
ESOPHAGEAL MANIFESTATIONS OF
SYSTEMIC DISEASE
■
■SCLERODERMA AND CONNECTIVE
TISSUE DISORDERS
Scleroderma esophagus (hypotensive LES and absent esophageal con­
tractility) was initially described as a manifestation of scleroderma or
other collagen vascular diseases and thought to be specific for these
disorders. However, this nomenclature subsequently has been dis­
carded because an estimated half of qualifying patients do not have an
identifiable rheumatologic disease, and reflux disease is often the only
identifiable association. When scleroderma esophagus occurs as a man­
ifestation of a connective tissue disorder, the histopathologic findings
are of infiltration and destruction of the esophageal muscularis propria
with collagen deposition and fibrosis and reduction in the number of
interstitial cells of Cajal. The pathogenesis of absent peristalsis and LES
hypotension in the absence of a connective tissue disorder is unknown.
Regardless of the underlying cause, the manometric abnormalities pre­
dispose patients to severe GERD due to inadequate LES barrier function
combined with poor esophageal clearance of refluxed acid. Dysphagia
may also be manifest but is generally mild and alleviated by eating in an
upright position and using liquids to facilitate solid transit.
PART 10
Disorders of the Gastrointestinal System
■
■DERMATOLOGIC DISEASES
A host of dermatologic disorders (lichen planus, pemphigus vulgaris,
bullous pemphigoid, cicatricial pemphigoid, Behçet’s syndrome, and
epidermolysis bullosa) can affect the oropharynx and esophagus,
particularly the proximal esophagus, with blisters, bullae, ulceration,
webs, and strictures. Topical or systemic anti-inflammatory therapy
is effective for mucosal healing. Stevens-Johnson syndrome and graftversus-host disease can also involve the esophagus. Esophageal dilation
may be necessary to treat strictures.
■
■FURTHER READING
Hirano I et al: American Gastroenterological Institute and the joint
task force on allergy-immunology practice parameters clinical guide­
lines for the management of eosinophilic esophagitis. Gastroenterol­
ogy 158:1776, 2020.
Kahrilas PJ et al: Advances in the management of oesophageal motil­
ity disorders in the era of high-resolution manometry: A focus on
achalasia syndromes. Nat Rev Gastroenterol Hepatol 15:323, 2018.
Katzka DA, Kahrilas PJ: Advances in the diagnosis and manage­
ment of gastroesophageal reflux disease. BMJ 23:371, 2020.
Katzka DA et al: Phenotypes of gastroesophageal reflux disease:
Where Rome, Lyon, and Montreal meet. Clin Gastroenterol Hepatol
18:767, 2020.
Shaheen NJ et al: Diagnosis and management of Barrett’s esophagus:
An updated ACG guideline. Am J Gastroenterol 117:559, 2022.
Straumann A, Katzka DA: Diagnosis and treatment of eosinophilic
esophagitis. Gastroenterology 154:346, 2018.
Von Arnim U et al: Monitoring patients with eosinophilic esophagitis
in routine clinical practice – international expert recommendations.
Clin Gastroenterol Hepatol 21:2526, 2023.
John Del Valle
Peptic Ulcer Disease
and Related Disorders
PEPTIC ULCER DISEASE
A peptic ulcer is defined as disruption of the mucosal integrity of the
stomach and/or duodenum leading to a local defect or excavation due
to active inflammation. Although burning epigastric pain exacerbated
by fasting and improved with meals is a symptom complex associated
with peptic ulcer disease (PUD), it is now clear that >90% patients with
this symptom complex (dyspepsia) do not have ulcers and that the
majority of patients with peptic ulcers may be asymptomatic. Ulcers
occur within the stomach and/or duodenum and are often chronic in
nature. Acid peptic disorders are very common in the United States,
with 4 million individuals (new cases and recurrences) affected per
year. Lifetime overall prevalence of PUD in the United States is ~8.4%
with a slightly higher prevalence in men. PUD significantly affects
quality of life by impairing overall patient well-being and contribut­
ing substantially to work absenteeism. Moreover, an estimated 15,000
deaths per year occur as a consequence of complicated PUD. The
financial impact of these common disorders has been substantial, with
an estimated burden on direct and indirect health care costs of ~$6 bil­
lion per year in the United States, with $3 billion spent on hospitaliza­
tions, $2 billion on physician office visits, and $1 billion in decreased
productivity and days lost from work.
■
■GASTRIC PHYSIOLOGY
Gastric Anatomy 
The gastric epithelial lining consists of rugae
that contain microscopic gastric pits, each branching into four or
five gastric glands made up of highly specialized epithelial cells. The
makeup of gastric glands varies with their anatomic location. Glands
within the gastric cardia comprise <5% of the gastric gland area and
contain mucous and endocrine cells. The 75% of gastric glands are
found within the oxyntic mucosa and contain mucous neck, parietal,
chief, endocrine, enterochromaffin, and enterochromaffin-like (ECL)
cells (Fig. 335-1). Highly specialized tuft cells are located in the neck
region of the gastric gland. These specialized cells are thought to
sample luminal contents, which in turn may be important in regulating
gastric acid secretion. Pyloric glands contain mucous and endocrine
cells (including gastrin cells) and are found in the antrum.
The parietal cell, also known as the oxyntic cell, is usually found in
the neck or isthmus or in the oxyntic gland. The resting, or unstimu­
lated, parietal cell has prominent cytoplasmic tubulovesicles and intra­
cellular canaliculi containing short microvilli along its apical surface
(Fig. 335-2). H+,K+-adenosine triphosphatase (ATPase) is expressed
in the tubulovesicle membrane; upon cell stimulation, this membrane,
along with apical membranes, transforms into a dense network of api­
cal intracellular canaliculi containing long microvilli. Acid secretion, a
process requiring high energy, occurs at the apical canalicular surface.
Numerous mitochondria (30–40% of total cell volume) generate the
energy required for secretion.
Corpus gland
Human
Corpus
Antrum
Mouse
Antral gland
Corpus
Antrum
FIGURE 335-1  Diagrammatic representation of the oxyntic gastric gland. Cellular constituents of the stomach. The human stomach is divided into three components the
cardia, the corpus and the antrum. The oxyntic gland which contains the majority of the acid producing cells are located in the corpus.  Glands located in the antrum secrete
predominantly mucus and also contain the important gastrin producing cells.  It is also important to note that key progenitor cells are located within both the gastric corpus
and antral glands. (Reproduced with permission from AC Engevik et al: The physiology of the gastric parietal cell. Physiol Rev 100:573, 2020, Figure 1.)
Resting
Stimulated
Canaliculus
HCl
H+,K+–ATPase
KCl
Tubulovesicles
KCl
Active
pump
H3O+
Active
pump
Ca
–
cAMP
Gastrin
ACh
Histamine
FIGURE 335-2  Gastric parietal cell undergoing transformation after secretagoguemediated stimulation. cAMP, cyclic adenosine monophosphate. (Reproduced with
permission from SJ Hersey, G Sachs: Gastric acid secretion. Am Physiol Soc 75:155,
1995.)
Surface cells
(MUC5AC)
Progenitor cells
Pit (foveolus)
Mucous neck cells
(MUC6, TFF2)
Parietal cells
(Н/К-АТРаsе)
Isthmus
D cells
(Somatostatin)
ECL cells
(Histidine decarboxylase)
Neck
EC cells
(Serotonin)
X Cell
(Ghrelin)
Base
Chief cells
(MIST1, PGC)
Surface cells
(MUC5AC)
CHAPTER 335
G cells
(Gastrin)
D cells
(Somatostatin)
Pit
ECL cells
(Histidine decarboxylase)
Peptic Ulcer Disease and Related Disorders
EC cells
(Serotonin)
Progenitor zone
Progenitor cells
(LRIG1, LGR5)
Base
Deep mucous
gland cells
(MUC6, TFF2, CD44v9)
Gastroduodenal Mucosal Defense 
The gastric epithelium is
under constant assault by a series of endogenous noxious factors,
including hydrochloric acid (HCl), pepsinogen/pepsin, and bile salts.
In addition, a steady flow of exogenous substances such as medications,
alcohol, and bacteria encounter the gastric mucosa. A highly intricate
biologic system is in place to provide defense from mucosal injury and
to repair any injury that may occur.
The mucosal defense system can be envisioned as a three-level
barrier, composed of preepithelial, epithelial, and subepithelial ele­
ments (Fig. 335-3). The first line of defense is a mucus-bicarbonatephospholipid layer, which serves as a physicochemical barrier to
multiple molecules, including hydrogen ions. Mucus is secreted in a
regulated fashion by gastroduodenal surface epithelial cells. It consists
primarily of water (95%) and a mixture of phospholipids and glyco­
proteins (mucin). The mucous gel functions as a nonstirred water layer
impeding diffusion of ions and molecules such as pepsin. Bicarbonate,
secreted in a regulated manner by surface epithelial cells of the gastro­
duodenal mucosa into the mucous gel, forms a pH gradient ranging
PART 10
Disorders of the Gastrointestinal System
FIGURE 335-3  Components involved in providing gastroduodenal mucosal defense and repair. CCK, cholecystokinin; CRF, corticotropin-releasing factor; EGF, epidermal
growth factor; HCl, hydrochloride; IGF, insulin-like growth factor; TGFα, transforming growth factor α; TRF, thyrotropin releasing factor. (Republished with permission of
John Wiley and Son’s Inc, from Bioregulation and Its Disorders in the Gastrointestinal Tract, T Yoshikawa, T Arakawa [eds]: 1998; permission conveyed through Copyright
Clearance Center, Inc.)
from 1 to 2 at the gastric luminal surface and reaching 6–7 along the
epithelial cell surface.
Surface epithelial cells provide the next line of defense through several
factors, including mucus production, epithelial cell ionic transporters
that maintain intracellular pH and bicarbonate production, and intracel­
lular tight junctions. Surface epithelial cells generate heat shock proteins
that prevent protein denaturation and protect cells from certain factors
such as increased temperature, cytotoxic agents, or oxidative stress. Epi­
thelial cells also generate trefoil factor family peptides and cathelicidins,
which also play a role in surface cell protection and regeneration. If the
preepithelial barrier is breached, gastric epithelial cells bordering a site
of injury can migrate to restore a damaged region (restitution). This
process occurs independent of cell division and requires uninterrupted
blood flow and an alkaline pH in the surrounding environment. Several
growth factors, including epidermal growth factor (EGF), transform­
ing growth factor (TGF) α, and basic fibroblast growth factor (FGF),
modulate the process of restitution. Larger defects that are not effectively
repaired by restitution require cell proliferation. Epithelial cell regenera­
tion is regulated by prostaglandins and growth factors such as EGF and
TGF-α. In tandem with epithelial cell renewal, formation of new vessels
(angiogenesis) within the injured microvascular bed occurs. Both FGF
and vascular endothelial growth factor (VEGF) are important in regulat­
ing angiogenesis in the gastric mucosa. In addition, the gastric peptide
gastrin (see below) has been found to stimulate cell proliferation, migra­
tion, invasion, angiogenesis, and autophagy. Finally, gastric parietal cells
(see below) express sonic hedgehog, a family of proteins important in
regulating cell lineage in multiple organs. This latter finding suggests
that parietal cells may also have the ability to regulate gastric stem cells.
Membrane phospholipids
Phospholipase A2
Arachidonic acid
Stomach
Kidney
Platelets
Endothelium
Macrophages
Leukocytes
Fibroblasts
Endothelium
COX-1
housekeeping
COX-2
inflammation
PGI2, PGE2
Inflammation
Mitogenesis
Bone formation
Other functions?
TXA2, PGI2, PGE2
Gastrointestinal mucosal integrity
Platelet aggregation
Renal function
FIGURE 335-4  Schematic representation of the steps involved in synthesis of
prostaglandin E2 (PGE2) and prostacyclin (PGI2). Characteristics and distribution of
the cyclooxygenase (COX) enzymes 1 and 2 are also shown. TXA2, thromboxane A2.
An elaborate microvascular system within the gastric submucosal
layer is the key component of the subepithelial defense/repair system,
providing HCO3
−, which neutralizes the acid generated by the parietal
cell. Moreover, this microcirculatory bed provides an adequate supply
of micronutrients and oxygen while removing toxic metabolic byproducts. Several locally produced factors including nitric oxide (NO)
(see below), hydrogen sulfide, and prostacyclin contribute to the vas­
cular protective pathway through vasodilation of the microcirculation.
Prostaglandins play a central role in gastric epithelial defense/
repair (Fig. 335-4). The gastric mucosa contains abundant levels of
prostaglandins that regulate the release of mucosal bicarbonate and
mucus, inhibit parietal cell secretion, and are important in maintaining
mucosal blood flow and epithelial cell restitution. Prostaglandins are
derived from esterified arachidonic acid, which is formed from phos­
pholipids (cell membrane) by the action of phospholipase A2. A key
enzyme that controls the rate-limiting step in prostaglandin synthesis
is cyclooxygenase (COX), which is present in two isoforms (COX-1,
COX-2), each having distinct characteristics regarding structure, tissue
distribution, and expression. COX-1 is expressed in a host of tissues,
including the stomach, platelets, kidneys, and endothelial cells. This
isoform is expressed in a constitutive manner and plays an important
role in maintaining the integrity of renal function, platelet aggregation,
and gastrointestinal (GI) mucosal integrity. In contrast, the expression
of COX-2 is inducible by inflammatory stimuli, and it is expressed in
macrophages, leukocytes, fibroblasts, and synovial cells. The beneficial
effects of nonsteroidal anti-inflammatory drugs (NSAIDs) on tissue
inflammation are due to inhibition of COX-2; the toxicity of these
drugs (e.g., GI mucosal ulceration and renal dysfunction) is related to
inhibition of the COX-1 isoform. The highly COX-2–selective NSAIDs
have the potential to provide the beneficial effect of decreasing tis­
sue inflammation while minimizing toxicity in the GI tract. Selective
COX-2 inhibitors have had adverse effects on the cardiovascular (CV)
system, leading to increased risk of myocardial infarction. Therefore,
the U.S. Food and Drug Administration (FDA) has removed two of
these agents (valdecoxib and rofecoxib) from the market (see below).
NO is important in the maintenance of gastric mucosal integrity.
The key enzyme NO synthase is constitutively expressed in the mucosa
and contributes to cytoprotection by stimulating gastric mucus,
increasing mucosal blood flow, and maintaining epithelial cell barrier
function. The central nervous system (CNS) and hormonal factors also
play a role in regulating mucosal defense through multiple pathways
(Fig. 335-3).
Since the discovery of Helicobacter pylori and its impact on gastric
pathology, it has become clear that the stomach has an elaborate and
complex inherent immunologic system in place. Although a detailed
description of the gastric immune system is beyond the scope of this
chapter, several features are worth highlighting. The gastric immune
response to certain pathogens such as H. pylori (see below) involves
extensive interplay between innate (dendritic cells, epithelial cells,
neutrophils, and macrophages) and adaptive (B and T cells) compo­
nents. Helper T cells (TH and TH regulatory cells) have been extensively
studied and appear to play an important role in a broad array of gastric
physiology extending from gastric secretion to epithelial cell turnover
via production of a number of cytokines.
The discovery of H. pylori has also led to the understanding that
the stomach, once thought to be devoid of microorganisms due to its
highly adverse environment (acid and pepsin), can serve as host for
bacterial communities consisting of hundreds of phylotypes, otherwise
known as its microbiota. The conceptual framework of the microbiome
has been receiving extensive attention in light of its importance in
human health and disease. The overall relevance of the gastric microbi­
ome and its impact on gastric pathology remain to be established, but it
is likely that alteration of microorganism homeostasis will play a role in
aspects of certain disorders such as PUD, gastritis, and gastric cancer.
Physiology of Gastric Secretion 
Understanding of the physiol­
ogy of gastric secretion is important when considering the pathophysi­
ology of PUD and the therapeutic options available. A detailed review of
the many components of gastric acid secretion is beyond the scope of
this chapter, and the reader is referred to Engevik et al (see Further
Reading) for a comprehensive discussion on this very important topic.
For the purposes of this chapter, a brief review of acid secretion will
follow. HCl and pepsinogen are the two principal gastric secretory
products capable of inducing mucosal injury. Gastric acid and pepsino­
gen play a physiologic role in protein digestion; absorption of iron, cal­
cium, magnesium, and vitamin B12; and killing ingested bacteria. Acid
secretion should be viewed as occurring under basal and stimulated
conditions. Basal acid production occurs in a circadian pattern, with
the highest levels occurring during the night and lowest levels during
the morning hours. Cholinergic input via the vagus nerve and hista­
minergic input from local gastric sources are the principal contribu­
tors to basal acid secretion. Stimulated gastric acid secretion occurs
primarily in three phases based on the site where the signal originates
(cephalic, gastric, and intestinal). Sight, smell, and taste of food are the
components of the cephalic phase, which stimulates gastric secretion
via the vagus nerve. The gastric phase is activated once food enters the
stomach. This component of secretion is driven by nutrients (amino
acids and amines) that directly (via peptone and amino acid receptors)
and indirectly (via stimulation of intramural gastrin-releasing peptide
neurons) stimulate the G cell to release gastrin, which in turn activates
the parietal cell via direct and indirect mechanisms. Distention of the
stomach wall also leads to gastrin release and acid production. The last
phase of gastric acid secretion is initiated as food enters the intestine
and is mediated by luminal distention and nutrient assimilation. A
series of pathways that inhibit gastric acid production are also set into
motion during these phases. The GI hormone somatostatin is released
from endocrine cells found in the gastric mucosa (D cells) in response
to HCl. Somatostatin can inhibit acid production by both direct (pari­
etal cell) and indirect mechanisms (decreased histamine release from
ECL cells, ghrelin release from Gr cells, and gastrin release from G cells).
Additional neural (central and peripheral) and humoral (amylin,
atrial natriuretic peptide [ANP], cholecystokinin, ghrelin, interleukin
11 [IL-11], obestatin, secretin, and serotonin) factors play a role in
counterbalancing acid secretion. Under physiologic circumstances,
these phases occur simultaneously. Ghrelin, the appetite-regulating
hormone expressed in Gr cells in the stomach, and its related peptide
motilin (released from the duodenum) may increase gastric acid secre­
tion through stimulation of histamine release from ECL cells, but this
remains to be confirmed.
CHAPTER 335
Peptic Ulcer Disease and Related Disorders
The acid-secreting parietal cell is located in the oxyntic gland,
adjacent to other cellular elements (ECL cell, D cell) important in the
gastric secretory process (Fig. 335-5). This unique cell also secretes
intrinsic factor (IF) and IL-11. The parietal cell expresses receptors for
several stimulants of acid secretion, including histamine (H2), gastrin
(cholecystokinin 2/gastrin receptor), and acetylcholine (muscarinic,
M3). Binding of histamine to the H2 receptor and activation of the
CGRP
PACAP
ACh
ACh
ACh
ACh
ACh
GRP
VIP
+
+
+
–
–
–
–
+
EC Cell
(ANP)
D Cell
(SST)
G Cell
(Gastrin)
D Cell
(SST)
–
+
+
–
HP
(Chronic Antrum)
Acid
HP (Acute)
FIGURE 335-5  Regulation of gastric acid secretion at the cellular level. ACh, acetylcholine; ANP, atrial natriuretic peptide; CGRP, calcitonin gene-related peptide; EC,
enterochromaffin; ECL, enterochromaffin-like; GRP, gastrin-releasing peptide; PACAP, pituitary adenylate-cyclase activating peptide; SST, somatostatin; VIP, vasoactive
intestinal peptide.
gastrin and muscarinic receptors result in stimulation of downstream
signaling pathways, which in turn regulates the acid-secreting pump,
H+,K+-ATPase. Parietal cells also express receptors for ligands that
inhibit acid production (glucagon-like peptide 1, prostaglandins,
somatostatin, EGF, neurotensin, and urocortin). Histamine also stimu­
lates gastric acid secretion indirectly by activating the histamine H3
receptor on D cells, which inhibits somatostatin release.
PART 10
Disorders of the Gastrointestinal System
The enzyme H+,K+-ATPase is responsible for generating the large
concentration of H+ ions. This enzyme uses the chemical energy of
adenosine triphosphate (ATP) to transfer H+ ions from parietal cell
cytoplasm to the secretory canaliculi in exchange for K+. The H+,K+-
ATPase is located within the secretory canaliculus and in nonsecretory
cytoplasmic tubulovesicles. The tubulovesicles are impermeable to K+,
which leads to an inactive pump in this location. The distribution of
pumps between the nonsecretory vesicles and the secretory canaliculus
varies according to parietal cell activity (Fig. 335-2). Proton pumps are
recycled back to the inactive state in cytoplasmic vesicles once parietal
cell activation ceases. In addition, acid secretion requires a number of
apical and basolateral parietal cell membrane chloride and potassium
channels. Parietal cells also express members of the sonic hedgehog
(Shh) family proteins, which play an important role in regulating cell
types in multiple organs. This family of proteins may also regulate cell
differentiation as well as restitution of mucosal defense in the gastric
epithelium.
The chief cell, found primarily in the gastric fundus, synthesizes and
secretes pepsinogen, the inactive precursor of the proteolytic enzyme
pepsin. The acid environment within the stomach leads to cleavage of
the inactive precursor to pepsin and provides the low pH (<2) required
for pepsin activity. Pepsin activity is significantly diminished at a pH of
4 and irreversibly inactivated and denatured at a pH of ≥7. Many of the
secretagogues that stimulate acid secretion also stimulate pepsinogen
release. The precise role of pepsin in the pathogenesis of PUD remains
to be established.
■
■PATHOPHYSIOLOGIC BASIS OF PUD
PUD encompasses both gastric ulcers (GUs) and duodenal ulcers
(DUs). Ulcers are defined as breaks in the mucosal surface >5 mm in
size, with depth to the submucosa. DUs and GUs share many common
features in terms of pathogenesis, diagnosis, and treatment, but several
factors distinguish them from one another. H. pylori and NSAIDs are
the most common risk factors for PUD, with estimated odds ratios in
the United States of 3.7 and 3.3, respectively. Additional risk factors
(odds ratio) include chronic obstructive lung disease (2.34), chronic
renal insufficiency (2.29), current tobacco use (1.99), former tobacco
use (1.55), older age (1.67), three or more doctor visits in a year (1.49),
Vagus
+
+
–
Parietal
Cell
+
+
H3
H2
+
–
–
+
ECL Cell
(Histamine)
Fundus
Antrum
coronary heart disease (1.46), former alcohol use (1.29), AfricanAmerican race (1.20), obesity (1.18), and diabetes (1.13). Selective
serotonin reuptake inhibitors (SSRIs) and gastric bypass surgery are
also associated with an increased incidence of PUD. A rise in idio­
pathic PUD has also been noted. The mechanisms by which some of
these risk factors lead to ulcer disease are highlighted below.
Epidemiology 
• 
DUODENAL ULCERS  DUs are estimated to
occur in 6–15% of the Western population. The incidence of DUs
declined steadily from 1960 to 1980 and has remained stable since
then. The death rates, need for surgery, and physician visits have
decreased by >50% over the past 30 years. The reason for the reduction
in the frequency of DUs is likely related to the decreasing frequency of
H. pylori in turn associated with overall improved sanitary conditions
across the world. Before the discovery of H. pylori, the natural history
of DUs was typified by frequent recurrences after initial therapy. Eradi­
cation of H. pylori has reduced these recurrence rates by >80%.
GASTRIC ULCERS  GUs tend to occur later in life than duodenal
lesions, with a peak incidence reported in the sixth decade. More than
one-half of GUs occur in males and are less common than DUs, per­
haps due to the higher likelihood of GUs being silent and presenting
only after a complication develops. Autopsy studies suggest a similar
incidence of DUs and GUs.
Pathology 
• 
DUODENAL ULCERS  DUs occur most often in the
first portion of the duodenum (>95%), with ~90% located within 3 cm
of the pylorus. They are usually ≤1 cm in diameter but can occasionally
reach 3–6 cm (giant ulcer). Ulcers are sharply demarcated, with depth
at times reaching the muscularis propria. The base of the ulcer often
consists of a zone of eosinophilic necrosis with surrounding fibrosis.
Malignant DUs are extremely rare.
GASTRIC ULCERS  In contrast to DUs, GUs can represent a malignancy
and should be biopsied upon discovery. Benign GUs are most often
found distal to the junction between the antrum and the acid secre­
tory mucosa. Benign GUs are quite rare in the gastric fundus and are
histologically similar to DUs. Benign GUs associated with H. pylori are
also associated with antral gastritis. In contrast, NSAID-related GUs
are not accompanied by chronic active gastritis but may instead have
evidence of a chemical gastropathy, typified by foveolar hyperplasia,
edema of the lamina propria, and epithelial regeneration in the absence of
H. pylori. Extension of smooth-muscle fibers into the upper portions of
the mucosa, where they are not typically found, may also occur.
Pathophysiology 
• 
DUODENAL ULCERS  H. pylori and NSAIDinduced injuries account for the majority of DUs. Many acid secretory
abnormalities have been described in DU patients. Of these, average
basal and nocturnal gastric acid secretion appears to be increased in
DU patients as compared to controls; however, the level of overlap
between DU patients and control subjects is substantial. The reason
for this altered secretory process is unclear, but H. pylori infection may
contribute. Bicarbonate secretion is significantly decreased in the duo­
denal bulb of patients with an active DU as compared to control sub­
jects. H. pylori infection may also play a role in this process (see below).
GASTRIC ULCERS  As in DUs, the majority of GUs can be attributed to
either H. pylori or NSAID-induced mucosal damage. Prepyloric GUs or
those in the body associated with a DU or a duodenal scar are similar
in pathogenesis to DUs. Gastric acid output (basal and stimulated)
tends to be normal or decreased in GU patients. When GUs develop
in the presence of minimal acid levels, impairment of mucosal defense
factors may be present.
H. PYLORI AND ACID PEPTIC DISORDERS  Although gastric infection
with the bacterium H. pylori was once thought to account for the
majority of PUD (Chap. 168), more recent studies suggest that only
one-fourth of DUs and one-sixth of GUs were associated with H. pylori
infection. These changes are likely due to successful rates of eradica­
tion of the organism coupled with improved sanitary conditions. It
appears that NSAIDs have become the most common cause of PUD
(see below). This organism also plays a role in the development of
gastric mucosa-associated lymphoid tissue (MALT) lymphoma and
gastric adenocarcinoma. Although the entire genome of H. pylori has
been sequenced, it is still not clear how this organism, which resides in
the stomach, causes ulceration in the duodenum. H. pylori eradication
efforts may lead to a decrease in gastric cancer in high-risk popula­
tions, particularly in individuals who have not developed chronic
atrophic gastritis and gastric metaplasia.
The Bacterium  The bacterium, initially named Campylobacter pylori­
dis, is a gram-negative microaerophilic rod found most commonly in
the deeper portions of the mucous gel coating the gastric mucosa or
between the mucous layer and the gastric epithelium. It may attach to
gastric epithelium but under normal circumstances does not appear
to invade cells. It is strategically designed to live within the aggressive
environment of the stomach. It is S-shaped (~0.5–3 μm in size) and
contains multiple sheathed flagella. Initially, H. pylori resides in the
antrum but, over time, migrates toward the more proximal segments
of the stomach. The organism is capable of transforming into a coccoid
form, which represents a dormant state that may facilitate survival in
adverse conditions. The genome of H. pylori (1.65 million base pairs)
encodes ~1500 proteins. Among this multitude of proteins, some fac­
tors are essential determinants of H. pylori–mediated pathogenesis
and colonization such as the outer membrane protein (Hop proteins),
urease, and the vacuolating cytotoxin (Vac A). Moreover, the majority
of H. pylori strains contain a genomic fragment that encodes the cag
pathogenicity island (cag-PAI). Several of the genes that make up cagPAI encode components of a type IV secretion island that translocates
Cag A into host cells. Once in the cell, Cag A activates a series of cellu­
lar events important in cell growth and cytokine production. H. pylori
also has extensive genetic diversity that in turn enhances its ability to
promote disease. The first step in infection by H. pylori is dependent on
the bacteria’s motility and its ability to produce urease. Urease produces
ammonia from urea, an essential step in alkalinizing the surround­
ing pH. Additional bacterial factors include but are not limited to,
catalase, lipase, adhesins, platelet-activating factor, and pic B (induces
cytokines). Multiple strains of H. pylori exist and are characterized by
their ability to express several of these factors (Cag A, Vac A, etc.). It is
possible that the different diseases related to H. pylori infection can be
attributed to different strains of the organism with distinct pathogenic
features.
Epidemiology  The prevalence of H. pylori varies throughout the world
and depends largely on the overall standard of living in the region,
with overall global prevalence decreasing. The global prevalence of
H. pylori infection in adults has declined from 50–55% to 43% dur­
ing 2014 to 2020. Contributing factors to these changes likely include
improvement of socioeconomic status and living standards as well as
enhanced hygiene conditions. It is also possible that the increased use
of antibiotics specifically with eradication therapies in individuals with
infection could be a contributing factor to the overall decrease in global
prevalence with a high variability in H. pylori infection throughout the
world. In developing parts of the world, 80% of the population may be
infected by the age of 20, whereas the prevalence is 20–50% in industri­
alized countries. In contrast, in the United States, this organism is rare
in childhood. The overall prevalence of H. pylori in the United States is
~30%, with individuals born before 1950 having a higher rate of infec­
tion than those born later. About 10% of Americans <30 years of age
are colonized with the bacteria. The rate of infection with H. pylori in
industrialized countries has decreased substantially in recent decades.
The steady increase in the prevalence of H. pylori noted with increasing
age is due primarily to a cohort effect, reflecting higher transmission
during a period in which the earlier cohorts were children. It has been
calculated through mathematical models that improved sanitation
during the latter half of the nineteenth century dramatically decreased
transmission of H. pylori. Moreover, with the present rate of interven­
tion, the organism ultimately will be eliminated from the United States.
Two factors that predispose to higher colonization rates include poor
socioeconomic status and less education about the organism. These
factors, not race, are responsible for the rate of H. pylori infection in
blacks and Hispanic Americans being double the rate seen in whites
of comparable age. Other risk factors for H. pylori infection are (1)
birth or residence in a developing country, (2) domestic crowding, (3)
unsanitary living conditions, (4) unclean food or water, and (5) expo­
sure to gastric contents of an infected individual.
CHAPTER 335
Transmission of H. pylori occurs from person to person, following
an oral-oral or fecal-oral route. The risk of H. pylori infection is declin­
ing in developing countries. The rate of infection in the United States
has fallen by >50% when compared to 30 years ago.
Pathophysiology  H. pylori infection is virtually always associated with
a chronic active gastritis, but only 10–15% of infected individu­
als develop frank peptic ulceration. The basis for this difference is
unknown but is likely due to a combination of host and bacterial fac­
tors, some of which are outlined below. Initial studies suggested that
90% of all DUs were associated with H. pylori, but H. pylori is present
in only one-sixth of individuals with GUs and one-fourth of patients
with DUs. The pathophysiology of ulcers not associated with H. pylori
or NSAID ingestion (or the rare Zollinger-Ellison syndrome [ZES])
is becoming more relevant as the incidence of H. pylori is dropping,
particularly in the Western world (see below).
Peptic Ulcer Disease and Related Disorders
The particular end result of H. pylori infection (gastritis, PUD,
gastric MALT lymphoma, gastric cancer) is determined by a complex
interplay between bacterial and host factors (Fig. 335-6).
Bacterial factors
Structure
Adhesins
Porins
Enzymes
(urease, vac A, cag A, etc.)
Host factors
Duration
Location
Inflammatory response
Genetics??
Chronic gastritis
Peptic ulcer disease
Gastric MALT lymphoma
Gastric cancer
FIGURE 335-6  Outline of the bacterial and host factors important in determining
H. pylori–induced gastrointestinal disease. MALT, mucosal-associated lymphoid
tissue.
Bacterial factors: H. pylori is able to facilitate gastric residence,
induce mucosal injury, and avoid host defense. Different strains of
H. pylori produce different virulence factors including γ-glutamyl
transpeptidase (GGT), cytotoxin-associated gene A (Cag A) product,
and virulence components vacuolating toxin (Vac A), in addition to
pathogen-associated molecular patterns (PAMPs) such as flagella and
lipopolysaccharide (LPS). A specific region of the bacterial genome,
the pathogenicity island (cag-PAI), encodes the virulence factors Cag
A and pic B. Vac A also contributes to pathogenicity, although it is not
encoded within the pathogenicity island. These virulence factors, in
conjunction with additional bacterial constituents, can cause muco­
sal damage, in part through their ability to target the host immune
cells. For example, Vac A targets human CD4 T cells, inhibiting their
proliferation, and in addition can disrupt normal function of B cells,
CD8 T cells, macrophages, and mast cells. Multiple studies have dem­
onstrated that H. pylori strains that are cag-PAI positive are associated
with a higher risk of PUD, premalignant gastric lesions, and gastric
cancer than are strains that lack the cag-PAI. In addition, H. pylori
may directly inhibit parietal cell H+,K+-ATPase activity through a Cag
A–dependent mechanism, leading in part to the low acid production
observed after acute infection with the organism. Urease, which allows
the bacteria to reside in the acidic stomach, generates NH3, which can
damage epithelial cells. The bacteria produce surface factors that are
chemotactic for neutrophils and monocytes, which in turn contribute
to epithelial cell injury (see below). H. pylori makes proteases and
phospholipases that break down the glycoprotein lipid complex of
the mucous gel, thus reducing the efficacy of this first line of mucosal
defense. H. pylori expresses adhesins (outer membrane proteins like
BabA), which facilitate attachment of the bacteria to gastric epithelial
cells. Although LPS of gram-negative bacteria often plays an important
role in the infection, H. pylori LPS has low immunologic activity com­
pared to that of other organisms. It may promote a smoldering chronic
inflammation.
PART 10
Disorders of the Gastrointestinal System
Host factors: Studies in twins suggest that there may be genetic
predisposition to acquire H. pylori. The inflammatory response to H.
pylori includes recruitment of neutrophils, lymphocytes (T and B),
macrophages, and plasma cells. The pathogen leads to local injury by
binding to class II major histocompatibility complex (MHC) molecules
expressed on gastric epithelial cells, leading to cell death (apoptosis).
Moreover, bacterial strains that encode cag-PAI can introduce Cag
A into the host cells, leading to further cell injury and activation of
cellular pathways involved in cytokine production and repression of
tumor-suppressor genes. Elevated concentrations of multiple cytokines
are found in the gastric epithelium of H. pylori–infected individuals,
including interleukin (IL) 1α/β, IL-2, IL-6, IL-8, tumor necrosis fac­
tor (TNF) α, and interferon (IFN) γ. H. pylori infection also leads to
both a mucosal and a systemic humoral response, which does not lead
to eradication of the bacteria but further compounds epithelial cell
injury. Additional mechanisms by which H. pylori may cause epithe­
lial cell injury include (1) activated neutrophil-mediated production
of reactive oxygen or nitrogen species and enhanced epithelial cell
turnover and (2) apoptosis related to interaction with T cells (T helper
1 [TH1] cells) and IFN-γ. Finally, the human stomach is colonized
by a host of commensal organisms that may affect the likelihood of
H. pylori infection and subsequent mucosal injury. Moreover, coloniza­
tion of the stomach with H. pylori likely alters the composition of the
gastric microbiota. The impact of the latter on gastric pathophysiology
remains unknown but some studies suggest a potential increase in the
development of gastric cancer. H. pylori also appears to regulate NO
formation via different mechanisms that in turn may contribute to the
organism’s cytotoxic effects. Specifically, H. pylori–derived factors, such as
urease, or the bacterium itself, stimulate NO synthase (NOS2) expression
in macrophages and in gastric epithelial cells leading to NO release and
subsequent cytotoxic effect on surrounding cells. H. pylori also leads to
the formation of 8-nitroguanine (8-NO2-Gua), which in conjunction with
oncoprotein Cag A, may contribute to the development of gastric cancer.
The basis for H. pylori–mediated duodenal ulceration remains
unclear. Studies suggest that H. pylori associated with duodenal ulcer­
ation may be more virulent. In addition, certain specific bacterial
Parietal cell
FUNDUS
Vagus
Canaliculus
Acetylcholine
Histamine
+
+
H, K ATPase
Tubulovesicles
ECL cell
+
Histamine
+
–
–
Somatostatin
Somatostatin
ECL cell
D cell
+
Gastrin
ANTRUM
Blood vessel
Gastrin
D cell
G cell
–
Somatostatin
FIGURE 335-7  Summary of potential mechanisms by which H. pylori may lead
to gastric secretory abnormalities. D, somatostatin cell; ECL, enterochromaffinlike cell; G, G cell. (Reproduced with permission from J Calam et al: How does
Helicobacter pylori cause mucosal damage? Its effect on acid and gastrin
physiology. Gastroenterology 113:543, 1997.)
factors such as the DU-promoting gene A (dupA) may be associated
with the development of DUs. Another potential contributing factor
is that gastric metaplasia in the duodenum of DU patients, which may
be due to high acid exposure (see below), permits H. pylori to bind to
it and produce local injury secondary to the host response. Another
hypothesis is that H. pylori antral infection could lead to increased acid
production, increased duodenal acid, and mucosal injury. Basal and
stimulated (meal, gastrin-releasing peptide [GRP]) gastrin release is
increased in H. pylori–infected individuals, and somatostatin-secreting
D cells may be decreased. H. pylori infection might induce increased
acid secretion through both direct and indirect actions of H. pylori and
proinflammatory cytokines (IL-8, TNF, and IL-1) on G, D, and parietal
cells (Fig. 335-7). GUs, in contrast, are associated with H. pylori–
induced pangastritis and normal or low gastric acid secretion. The
H. pylori–mediated decrease in gastric acid secretion after long-term
infection may be due to the bacterium’s ability to inhibit H+,K+-ATPase
expression. H. pylori infection has also been associated with decreased
duodenal mucosal bicarbonate production. Data supporting and con­
tradicting each of these interesting theories have been demonstrated.
Thus, the mechanism by which H. pylori infection of the stomach leads
to duodenal ulceration remains to be established. The development of
in vitro organoids, a unique tool that replicates in part the multicellular
structure of the intact organ, provides a more physiologic model for
experimentation in an in vitro system. Moreover, the development of
advanced microscopic optical imaging techniques will lead to increased
understanding of parietal cell adaptation to H. pylori infection.
In summary, the final effect of H. pylori on the GI tract is variable
and determined by microbial and host factors. The type and distri­
bution of gastritis correlate with the ultimate gastric and duodenal
pathology observed. Specifically, the presence of antral-predominant
gastritis is associated with DU formation; gastritis involving primarily
the corpus predisposes to the development of GUs, gastric atrophy, and
ultimately gastric carcinoma (Fig. 335-8).
NSAID-INDUCED DISEASE 
Epidemiology  NSAIDs represent a group of the most commonly used
medications in the world and the United States. It is estimated that
7 billion dollars per year are spent on NSAIDs worldwide, with >30 bil­
lion over-the-counter tablets sold. More than 30 million individuals take
NSAIDs, with >100 million prescriptions sold yearly in the United States
High level of acid production
Duodenal ulcer
MALT
lymphoma
Antralpredominant
gastritis
Chronic
H. pylori
infection
Asymptomatic
H. pylori
infection
Nonatrophic
pangastritis
Normal gastric
mucosa
Corpuspredominant
atrophic
gastritis
Gastric ulcer
Acute
H. pylori
infection
Intestinal metaplasia
Dysplasia
Gastric
cancer
Low level of acid production
Childhood
Advanced age
FIGURE 335-8  Natural history of H. pylori infection. MALT, mucosal-associated
lymphoid tissue. (From The New England Journal of Medicine, Medical progress:
Helicobacter pylori infection, S Suerbaum, P Michetti: 347:1175-1186. Copyright @2002
Massachusetts Medical Society. Reprinted with permission from Massachusetts
Medical Society.)
alone. In fact, after the introduction of COX-2 inhibitors in the year
2000, the number of prescriptions written for NSAIDs was >111
million at a cost of $4.8 billion. Side effects and complications due to
NSAIDs are considered the most common drug-related toxicities in
the United States. The spectrum of NSAID-induced morbidity ranges
from nausea and dyspepsia (prevalence reported as high as 50–60%)
to a serious GI complication such as endoscopy-documented peptic
ulceration (15–30% of individuals taking NSAIDs regularly), which
is complicated by bleeding or perforation in as many as 1.5% of users
per year. It is estimated that NSAID-induced GI bleeding accounts
for 60,000–120,000 hospital admissions per year, and deaths related
to NSAID-induced toxicity may be as high as 16,000 per year in the
United States. Approximately 4–5% of patients develop symptomatic
ulcers within 1 year. Unfortunately, dyspeptic symptoms do not correlate with NSAID-induced pathology. Over 80% of patients with serious NSAID-related complications did not have preceding dyspepsia.
In view of the lack of warning signs, it is important to identify patients
who are at increased risk for morbidity and mortality related to NSAID
usage. Even 75 mg/d of aspirin may lead to serious GI ulceration; thus, no
dose of NSAID is completely safe. In fact, the incidence of mucosal injury
(ulcers and erosions) in patients taking low-dose aspirin (75–325 mg) has
been estimated to range from as low as 8 to as high as 60%. It appears
that H. pylori infection increases the risk of PUD-associated GI bleeding in chronic users of low-dose aspirin. Established risk factors include
advanced age, history of ulcer, concomitant use of glucocorticoids, highdose NSAIDs, multiple NSAIDs, concomitant use of anticoagulants or
clopidogrel, and serious or multisystem disease. Possible risk factors
include concomitant infection with H. pylori, cigarette smoking, and
alcohol consumption. SSRIs have a synergistic effect on the induction of
GI bleeding believed to be due in part to this agent’s ability to decrease
platelet aggregation by decreasing serotonin content in platelets.
Pathophysiology  Prostaglandins play a critical role in maintaining
gastroduodenal mucosal integrity and repair. It therefore follows that
interruption of prostaglandin synthesis can impair mucosal defense
and repair, thus facilitating mucosal injury via a systemic mechanism.
Animal studies have demonstrated that neutrophil adherence to the
gastric microcirculation plays an essential role in the initiation of
NSAID-induced mucosal injury. A summary of the pathogenetic pathways by which systemically administered NSAIDs may lead to mucosal
injury is shown in Fig. 335-9. Single nucleotide polymorphisms (SNPs)
have been found in several genes, including those encoding certain subtypes of cytochrome P450 (see below), IL-1β (IL-1β), angiotensinogen
Gastrointestinal mucosal injury
Mitochondrial uncoupling
Reactive prooxidants
MOS
ATP
Mitochondrial fission
Mucosal PGHS-1
PGE 2
Mucosal defense
Intestinal mucosal barrier function
Mucosal inflammation
Apoptosis
FIGURE 335-9  Effect of nonsteroidal anti-inflammatory drugs (NSAIDs) on different
target organs. The action of NSAIDs on major organs including stomach, small
intestine, heart, liver, kidney, respiratory tract, and brain is mainly mediated through
prostaglandin endoperoxide synthase (PGHS)–dependent prostanoid modulation
and alteration of mitochondrial functional integrity leading to mitochondrial
oxidative stress (MOS) generation, depolarization of mitochondrial transmembrane
potential (ΔΨm), and consequent cell death. However, in heart, low-dose aspirin
actually offers cardioprotection through antithrombotic effect. Upward arrows
indicate upregulation/elevation; downward arrows indicate downregulation/
depletion. (From S Bindu et al: Non-steroidal anti-inflammatory drugs (NSAIDs) and
organ damage: A current perspective. Biochem Pharmacol 180:114147, 2020.)
(AGT), and an organic ion transporting polypeptide (SLCO1B1), but
these findings need confirmation in larger-scale studies.
CHAPTER 335
Injury to the mucosa also occurs as a result of the topical use of
NSAIDs, leading to increased epithelial surface permeability. Aspirin
and many NSAIDs are weak acids that remain in a nonionized lipophilic form when found within the acid environment of the stomach.
Under these conditions, NSAIDs migrate across lipid membranes of
epithelial cells, leading to cell injury once trapped intracellularly in an
ionized form. Topical NSAIDs can also alter the surface mucous layer,
permitting back diffusion of H+ and pepsin, leading to further epithelial cell damage. Moreover, enteric-coated or buffered preparations are
also associated with risk of peptic ulceration. NSAIDs can also lead to
mucosal injury via production of additional proinflammatory mediators such as TNF and leukotrienes through simultaneous activation of
the lipoxygenase pathway.
Peptic Ulcer Disease and Related Disorders
The interplay between H. pylori and NSAIDs in the pathogenesis
of PUD is complex. Meta-analysis supports the conclusion that each
of these aggressive factors is an independent and synergistic risk factor for PUD and its complications such as GI bleeding. For example,
eradication of H. pylori reduces the likelihood of GI complications in
high-risk individuals to levels observed in individuals with average risk
of NSAID-induced complications.
In summary, NSAID-induced mucosal injury is a multifaceted process involving the interaction of multiple, often synergistic pathophysiologic processes at the epithelium and surrounding interfaces.
PATHOGENETIC FACTORS UNRELATED TO H. PYLORI AND NSAIDS IN
ACID PEPTIC DISEASE  Cigarette smoking has been implicated in
the pathogenesis of PUD. Not only have smokers been found to have
ulcers more frequently than do nonsmokers, but smoking appears to
decrease healing rates, impair response to therapy, and increase ulcerrelated complications such as perforation. The mechanism responsible
for increased ulcer diathesis in smokers is unknown. Theories have
included altered gastric emptying, decreased proximal duodenal
bicarbonate production, increased risk for H. pylori infection, and cigarette-induced generation of noxious mucosal free radicals. Genetic predisposition may play a role in ulcer development. First-degree relatives
of DU patients are three times as likely to develop an ulcer; however,
the potential role of H. pylori infection in contacts is a major consideration. Increased frequencies of blood group O and of the nonsecretor status have also been implicated as genetic risk factors for peptic
diathesis. However, H. pylori preferentially binds to group O antigens.
Additional genetic factors have been postulated to predispose certain
individuals to developing PUD and/or upper GI bleeding. Specifically,
genes encoding the NSAID-metabolizing enzymes cytochrome P450
2C9 and 2C8 (CYP2C9 and CYP2C8) are potential susceptibility genes
for NSAID-induced PUD, but unfortunately, the studies have not been
consistent in demonstrating this association. In a United Kingdom
study, the CYP2C19*17 gain-of-function polymorphism was associated
with PUD in a Caucasian cohort, irrespective of ulcer etiology. These
findings need to be confirmed in broader studies. Psychological stress
has been thought to contribute to PUD, but studies examining the role
of psychological factors in its pathogenesis have generated conflicting
results. Although PUD is associated with certain personality traits
(neuroticism), these same traits are also present in individuals with
nonulcer dyspepsia (NUD) and other functional and organic disorders.
Diet has also been thought to play a role in peptic diseases. Certain
foods and beverages can cause dyspepsia, but no convincing studies
indicate an association between ulcer formation and a specific diet.
Specific chronic disorders have been shown to have a strong asso­
ciation with PUD: (1) advanced age, (2) chronic pulmonary disease,
(3) chronic renal failure, (4) cirrhosis, (5) nephrolithiasis, (6) α1antitrypsin deficiency, and (7) systemic mastocytosis. Disorders with
a possible association are (1) hyperparathyroidism, (2) coronary artery
disease, (3) polycythemia vera, (4) chronic pancreatitis, (5) former
alcohol use, (6) obesity, (7) African-American race, and (8) three or
more doctor visits in a year.
Multiple factors play a role in the pathogenesis of PUD. The two
predominant causes are H. pylori infection and NSAID ingestion. PUD
not related to H. pylori or NSAIDs is increasing. Other less common
causes of PUD are shown in Table 335-1. These etiologic agents should
be considered as the incidence of H. pylori is decreasing. Independent
of the inciting or injurious agent, peptic ulcers develop as a result of an
imbalance between mucosal protection/repair and aggressive factors.
Gastric acid plays an important role in mucosal injury.
PART 10
Disorders of the Gastrointestinal System
■
■CLINICAL FEATURES
History 
Abdominal pain is common to many GI disorders, includ­
ing DU and GU, but it has a poor predictive value for the presence
TABLE 335-1  Causes of Ulcers Not Caused by Helicobacter pylori
and NSAIDs
Pathogenesis of Non-Hp and Non-NSAID Ulcer Disease
Infection
  Cytomegalovirus
  Herpes simplex virus
  Helicobacter heilmannii
Drug/Toxin
  Bisphosphonates
  Checkpoint inhibitor
  Chemotherapy
  Clopidogrel
  Crack cocaine
  Glucocorticoids (when combined with NSAIDs)
  Mycophenolate mofetil
  Potassium chloride
Miscellaneous
  Basophilia in myeloproliferative disease
  Duodenal obstruction (e.g., annular pancreas)
  Infiltrating disease
  Ischemia
  Radiation therapy
  Eosinophilic infiltration
  Sarcoidosis
  Crohn’s disease
  Idiopathic hypersecretory state
Abbreviations: Hp, H. pylori; NSAIDs, nonsteroidal anti-inflammatory drugs.
of either DU or GU. Approximately two-thirds of patients with PUD
do not have abdominal pain, and up to 87% of patients with NSAIDinduced mucosal disease can present with a complication (bleeding,
perforation, and obstruction) without antecedent symptoms. Despite
this poor correlation, a careful history and physical examination are
essential components of the approach to a patient suspected of having
peptic ulcers.
Epigastric pain described as a burning or gnawing discomfort can
be present in both DU and GU. The discomfort is also described as an
ill-defined, aching sensation or as hunger pain. The typical pain pat­
tern in DU occurs 90 min to 3 h after a meal and is frequently relieved
by antacids or food. Pain that awakes the patient from sleep (between
midnight and 3 A.M.) is the most discriminating symptom, with twothirds of DU patients describing this complaint. Unfortunately, this
symptom is also present in one-third of patients with NUD (see below).
Elderly patients are less likely to have abdominal pain as a manifesta­
tion of PUD and may instead present with a complication such as
ulcer bleeding or perforation. The pain pattern in GU patients may be
different from that in DU patients, where discomfort may actually be
precipitated by food. Nausea and weight loss occur more commonly in
GU patients. Endoscopy detects ulcers in <30% of patients who have
dyspepsia.
The mechanism for development of abdominal pain in ulcer patients
is unknown. Several possible explanations include acid-induced activa­
tion of chemical receptors in the duodenum, enhanced duodenal sen­
sitivity to bile acids and pepsin, and altered gastroduodenal motility.
Variation in the intensity or distribution of the abdominal pain,
as well as the onset of associated symptoms such as nausea and/or
vomiting, may be indicative of an ulcer complication. Dyspepsia that
becomes constant, is no longer relieved by food or antacids, or radi­
ates to the back may indicate a penetrating ulcer (pancreas). Sudden
onset of severe, generalized abdominal pain may indicate perforation.
Pain worsening with meals, nausea, and vomiting of undigested food
suggest gastric outlet obstruction. Tarry stools or coffee-ground emesis
indicate bleeding.
Physical Examination 
Epigastric tenderness is the most frequent
finding in patients with GU or DU. Pain may be found to the right of
the midline in 20% of patients. Unfortunately, the predictive value of
this finding is low. Physical examination is critically important for dis­
covering evidence of ulcer complication. Tachycardia and orthostasis
suggest dehydration secondary to vomiting or active GI blood loss. A
severely tender, board-like abdomen suggests a perforation. Presence
of a succussion splash indicates retained fluid in the stomach, suggest­
ing gastric outlet obstruction.
PUD-Related Complications 
• 
GASTROINTESTINAL BLEEDING 
GI bleeding is the most common complication observed in PUD.
Bleeding is estimated to occur in 19.4–57 per 100,000 individuals in a
general population or in ~15% of patients. Bleeding and complications
of ulcer disease occur more often in individuals >60 years of age. The
30-day mortality rate is as high as 2.5–10%. The higher incidence in
the elderly is likely due to the increased use of NSAIDs in this group.
In addition, up to 80% of the mortality in PUD-related bleeding is due
to nonbleeding causes such as multiorgan failure (24%), pulmonary
complications (24%), and malignancy (34%).
Greater than 50% of patients with ulcer-related hemorrhage bleed
without any preceding warning signs or symptoms.
PERFORATION  The second most common ulcer-related complication
is perforation, being reported in as many as 6–7% of PUD patients
with an estimated 30-day mortality of >20%. Acute abdominal pain,
tachycardia, and abdominal rigidity compose the classic triad associ­
ated with this complication. It is essential to remember that elderly
patients or individuals who are immunosuppressed may not have this
classic presentation. As in the case of bleeding, the incidence of perfo­
ration in the elderly appears to be increasing secondary to increased
use of NSAIDs. Perforation of DUs has become less common in light
of the increased rates of H. pylori eradication, with NSAID-induced
GUs leading to perforation occurring more commonly. Penetration is
a form of perforation in which the ulcer bed tunnels into an adjacent
organ. DUs tend to penetrate posteriorly into the pancreas, leading to
pancreatitis, whereas GUs tend to penetrate into the left hepatic lobe.
Gastrocolic fistulas associated with GUs have also been described.
Mortality for this complication can be >20% within 30 days.
GASTRIC OUTLET OBSTRUCTION  Gastric outlet obstruction is the
least common ulcer-related complication, occurring in 1–2% of
patients. A patient may have relative obstruction secondary to ulcerrelated inflammation and edema in the peripyloric and duodenal
region. This process often resolves with ulcer healing. A fixed, mechan­
ical obstruction secondary to scar formation in the peripyloric areas
is also possible. The latter requires endoscopic (balloon dilation with
or without placement of a biodegradable stent) or surgical interven­
tion with a stricturoplasty or gastrojejunostomy. Signs and symptoms
relative to mechanical obstruction may develop insidiously. New onset
of early satiety, nausea, vomiting, increase of postprandial abdominal
pain, and weight loss should make gastric outlet obstruction a possible
diagnosis.
Differential Diagnosis 
The list of GI and non-GI disorders that
can mimic ulceration of the stomach or duodenum is quite extensive.
The most commonly encountered diagnosis among patients seen for
upper abdominal discomfort is functional dyspepsia (FD) or essential
dyspepsia, which refers to a group of heterogeneous disorders typified
by upper abdominal pain without the presence of an ulcer. The symp­
toms can range from postprandial fullness and early satiety to epigas­
tric burning pain. The dichotomy of this symptom complex has led to
the identification of two subcategories of FD including postprandial
distress syndrome (PDS) and epigastric pain syndrome (EPS). Dyspep­
sia has been reported to occur in up to 30% of the U.S. population. Up
to 80% of patients seeking medical care for dyspepsia have a negative
diagnostic evaluation. The etiology of FD is not established, but postin­
fectious states, certain foods, and H. pylori infection may contribute to
the pathogenesis of this common disorder.
Several additional disease processes that may present with “ulcerlike” symptoms include proximal GI tumors, gastroesophageal reflux,
vascular disease, pancreaticobiliary disease (biliary colic, chronic pan­
creatitis), and gastroduodenal Crohn’s disease.
Diagnostic Evaluation 
In view of the poor predictive value of
abdominal pain for the presence of a gastroduodenal ulcer and the
multiple disease processes that can mimic this disease, the clinician
is often confronted with having to establish the presence of an ulcer.
Documentation of an ulcer requires either a radiographic (barium
study, rarely done in today’s environment) or an endoscopic procedure.
However, a large percentage of patients with symptoms suggestive of an
A
B
FIGURE 335-10  Barium study demonstrating (A) a benign duodenal ulcer and (B) a benign gastric ulcer.
ulcer have NUD; testing for H. pylori and antibiotic therapy (see below)
are appropriate for individuals who are otherwise healthy, without
red flag signs such as nausea, vomiting, weight loss, or evidence of GI
bleeding and <60 years of age, before embarking on a diagnostic evalu­
ation (Chap. 48).
Barium studies of the proximal GI tract are rarely used as a first
test for documenting an ulcer (Fig. 335-10). Endoscopy provides the
most sensitive and specific approach for examining the upper GI
tract (Fig. 335-11). In addition to permitting direct visualization of
the mucosa, endoscopy facilitates photographic documentation of
a mucosal defect and tissue biopsy to rule out malignancy (GU) or
H. pylori. Endoscopic examination is particularly helpful in identifying
lesions too small to detect by radiographic examination, for evaluation
of atypical radiographic abnormalities, or to determine if an ulcer is a
source of blood loss.
Although the methods for diagnosing H. pylori are outlined in
Chap. 168, a brief summary will be included here (Table 335-2).
Several biopsy urease tests have been developed (PyloriTek, CLOtest,
Hpfast, Pronto Dry) that have a sensitivity and specificity of >90–95%.
Several noninvasive methods for detecting this organism have been
developed. Three types of studies routinely used include serologic
testing, the 13C-urea breath test, and the fecal H. pylori (Hp) antigen
test (monoclonal antibody test). A urinary Hp antigen test and a home
breath test appear promising.
Occasionally, specialized testing such as serum gastrin and gastric
acid analysis may be needed in individuals with complicated or refrac­
tory PUD (see “Zollinger-Ellison Syndrome,” below). Screening for
aspirin or NSAIDs (blood or urine) may also be necessary in refractory
H. pylori–negative PUD patients.
CHAPTER 335
TREATMENT
Peptic Ulcer Disease
Peptic Ulcer Disease and Related Disorders
Before the discovery of H. pylori, the therapy of PUD was centered
on the old dictum by Schwartz of “no acid, no ulcer.” Although acid
secretion is still important in the pathogenesis of PUD, eradication
of H. pylori and therapy/prevention of NSAID-induced disease is
the mainstay of treatment. A summary of commonly used drugs for
treatment of acid peptic disorders is shown in Table 335-3.
ACID-NEUTRALIZING/INHIBITORY DRUGS
Antacids  Before we understood the important role of histamine
in stimulating parietal cell activity, neutralization of secreted acid
with antacids constituted the main form of therapy for peptic ulcers.
They are now rarely, if ever, used as the primary therapeutic agent
A
B
FIGURE 335-11  Endoscopy demonstrating (A) a benign duodenal ulcer and (B) a benign gastric ulcer.
but instead are often used by patients for symptomatic relief of
dyspepsia. The most commonly used agents are mixtures of alumi­
num hydroxide and magnesium hydroxide. Aluminum hydroxide
can produce constipation and phosphate depletion; magnesium
hydroxide may cause loose stools. Many of the commonly used ant­
acids (e.g., Maalox, Mylanta) have a combination of both aluminum
and magnesium hydroxide in order to avoid these side effects. The
magnesium-containing preparation should not be used in chronic
renal failure patients because of possible hypermagnesemia, and
aluminum may cause chronic neurotoxicity in these patients.
PART 10
Disorders of the Gastrointestinal System
Calcium carbonate and sodium bicarbonate are potent antacids
with varying levels of potential problems. The long-term use of
calcium carbonate (converts to calcium chloride in the stomach)
can lead to milk-alkali syndrome (hypercalcemia and hyperphos­
phatemia with possible renal calcinosis and progression to renal
insufficiency). Sodium bicarbonate may induce systemic alkalosis.
H2 Receptor Antagonists  Four of these agents are presently avail­
able (cimetidine, ranitidine, famotidine, and nizatidine), and their
structures share homology with histamine. Although each has
different potency, all will significantly inhibit basal and stimulated
acid secretion to comparable levels when used at therapeutic doses.
Moreover, similar ulcer-healing rates are achieved with each drug
when used at the correct dosage. Presently, this class of drug is often
used for treatment of active ulcers (4–6 weeks) in combination with
antibiotics directed at eradicating H. pylori (see below).
TABLE 335-2  Tests for Detection of Helicobacter pylori
SENSITIVITY/
SPECIFICITY, % COMMENTS
TEST
Invasive (Endoscopy/Biopsy Required)
Rapid urease
80–95/95–100
Simple, false negative with recent use of
PPIs, antibiotics, or bismuth compounds
Histology
60–90/>95
Requires pathology processing and staining;
provides histologic information
Culture
76-90/100
Time-consuming, expensive, dependent
on experience; allows determination of
antibiotic susceptibility
Noninvasive
Serology
74.4/59
Inexpensive, convenient; not useful for early
follow-up; cannot distinguish active and
prior infection
Urea breath test
95/>95
Simple, rapid; useful for early follow-up;
false negatives with recent therapy (see
rapid urease test)
Stool antigen
95/>95
Inexpensive, convenient
Abbreviation: PPIs, proton pump inhibitors.
Cimetidine was the first H2 receptor antagonist used for the
treatment of acid peptic disorders. Cimetidine may have weak
antiandrogenic side effects resulting in reversible gynecomastia and
impotence, primarily in patients receiving high doses for prolonged
periods of time (months to years). In view of cimetidine’s ability
to inhibit cytochrome P450, careful monitoring of drugs such as
warfarin, phenytoin, and theophylline is indicated with long-term
usage. Other rare reversible adverse effects reported with cimeti­
dine include confusion and elevated levels of serum aminotransfer­
ases, creatinine, and serum prolactin. Famotidine is a more potent
H2 receptor antagonists than cimetidine. It can be used once a day
at bedtime for ulcer prevention, which was commonly done before
the discovery of H. pylori and the development of proton pump
inhibitors (PPIs). Patients may develop tolerance to H2 blockers,
a rare event with PPIs (see below). Comparable nighttime dosing
regimens are cimetidine 800 mg and famotidine 40 mg.
Additional rare, reversible systemic toxicities reported with H2
receptor antagonists include pancytopenia, neutropenia, anemia,
and thrombocytopenia, with a prevalence rate varying from 0.01 to
0.2%. Cimetidine can bind to hepatic cytochrome P450; famotidine
does not.
Proton Pump (H+,K+-ATPase) Inhibitors  Omeprazole, esomepra­
zole, lansoprazole, rabeprazole, and pantoprazole are substituted
TABLE 335-3  Drugs Used in the Treatment of Peptic Ulcer Disease
DRUG TYPE/
MECHANISM
EXAMPLES
DOSE
Acid-Suppressing Drugs
Antacids
Mylanta, Maalox, Tums,
Gaviscon
100–140 meq/L 1 and 3 h
after meals and hs
H2 receptor antagonists
Cimetidine
400 mg bid
 
Famotidine
40 mg hs
Proton pump inhibitors
Omeprazole
20 mg/d
 
Lansoprazole
30 mg/d
 
Rabeprazole
20 mg/d
 
Pantoprazole
40 mg/d
 
Esomeprazole
20 mg/d
 
Dexlansoprazole
30 mg/d
Mucosal Protective Agents
Sucralfate
Sucralfate
1 g qid
Prostaglandin analogue
Misoprostol
200 μg qid
Bismuth-containing
compounds
Bismuth subsalicylate
(BSS)
See anti–H. pylori
regimens (Table 335-4)
Abbreviation: hs, at bedtime (hora somni).
Peptic Ulcer Disease and Related Disorders
CHAPTER 335
benzimidazole derivatives that covalently bind and irreversibly
inhibit H+,K+-ATPase. Esomeprazole is the S-enantiomer of omepra­
zole, which is a racemic mixture of both S- and R-optical isomers.
The R-isomer of lansoprazole, dexlansoprazole, is the most recent
PPI approved for clinical use. Its reported advantage is a dual
delayed-release system aimed at improving treatment of gastro­
esophageal reflux disease (GERD). These are the most potent acid
inhibitory agents available. Omeprazole and lansoprazole are the
PPIs that have been used for the longest time. Both are acid-labile
and are administered as enteric-coated granules in a sustainedrelease capsule that dissolves within the small intestine at a pH of
6. Lansoprazole is available in an orally disintegrating tablet that
can be taken with or without water, an advantage for individuals
who have significant dysphagia. Absorption kinetics are similar
to the capsule. In addition, a lansoprazole-naproxen combination
preparation that has been made available is targeted at decreasing
NSAID-related GI injury (see below). Omeprazole is available as
non–enteric-coated granules mixed with sodium bicarbonate in a
powder form that can be administered orally or via gastric tube.
The sodium bicarbonate has two purposes: to protect the omepra­
zole from acid degradation and to promote rapid gastric alkaliniza­
tion and subsequent proton pump activation, which facilitates rapid
action of the PPI. Pantoprazole and rabeprazole are available as
enteric-coated tablets. Pantoprazole is also available as a parenteral
formulation for intravenous use. These agents are lipophilic com­
pounds; upon entering the parietal cell, they are protonated and
trapped within the acid environment of the tubulovesicular and
canalicular system. These agents potently inhibit all phases of gas­
tric acid secretion. Onset of action is rapid, with a maximum acid
inhibitory effect between 2 and 6 h after administration and dura­
tion of inhibition lasting up to 72–96 h. With repeated daily dosing,
progressive acid inhibitory effects are observed, with basal and
secretagogue-stimulated acid production being inhibited by >95%
after 1 week of therapy. The half-life of PPIs is ~18 h; thus, it can take
between 2 and 5 days for gastric acid secretion to return to normal
levels once these drugs have been discontinued. Because the pumps
need to be activated for these agents to be effective, their efficacy
is maximized if they are administered before a meal (except for the
immediate-release formulation of omeprazole) (e.g., in the morn­
ing before breakfast). Mild to moderate hypergastrinemia has been
observed in patients taking these drugs. Carcinoid tumors devel­
oped in some animals given the drugs chronically; however, exten­
sive experience has failed to demonstrate gastric carcinoid tumor
development in humans. Serum gastrin levels return to normal
levels within 1–2 weeks after drug cessation. Rebound gastric acid
hypersecretion has been described in H. pylori–negative individuals
after discontinuation of PPIs. It occurs even after relatively shortterm usage (2 months) and may last for up to 2 months after the PPI
has been discontinued. The mechanism involves gastrin-induced
hyperplasia and hypertrophy of histamine-secreting ECL cells. The
clinical relevance of this observation is that individuals may have
worsening symptoms of GERD or dyspepsia upon stopping the
PPI. Gradual tapering of the PPI and switching to an H2 receptor
antagonist may prevent this from occurring. H. pylori–induced
inflammation and concomitant decrease in acid production may
explain why this does not occur in H. pylori–positive patients. IF
production is also inhibited, but vitamin B12 deficiency anemia is
uncommon, probably because of the large stores of the vitamin. As
with any agent that leads to significant hypochlorhydria, PPIs may
interfere with absorption of drugs such as ketoconazole, ampicillin,
iron, and digoxin. Hepatic cytochrome P450 can be inhibited by the
earlier PPIs (omeprazole, lansoprazole). Rabeprazole, pantoprazole,
and esomeprazole do not appear to interact significantly with drugs
metabolized by the cytochrome P450 system. The overall clinical
significance of this observation is not definitely established. Cau­
tion should be taken when using theophylline, warfarin, diazepam,
atazanavir, and phenytoin concomitantly with PPIs.
The list of potential side effects with long-term PPI use has
steadily grown over the years. These agents are commonly used
since several formulations have become available as over-the-coun­
ter medications. Moreover, up to 70% of current prescriptions for
long-term PPIs may be unwarranted and between 35 and 60% of
in-hospital use of PPIs may be inappropriate. Interpretation of the
multiple studies should take into consideration that the vast major­
ity were retrospective observational studies in which confounding
factors could not be accounted for entirely.
Long-term acid suppression, especially with PPIs, has been asso­
ciated with a higher incidence of community-acquired pneumonia
as well as community- and hospital-acquired Clostridioides difficile–
associated disease. A meta-analysis showed a 74% increased risk of
C. difficile infection and a 2.5-fold higher risk of reinfection as com­
pared to nonusers. In light of these concerns, the FDA published a
safety alert regarding the association between C. difficile infection
and PPI use. Although the risk of spontaneous bacterial peritonitis
in cirrhotics was thought to be increased, the data here are less
supportive. The impact of PPI-induced changes in the host micro­
biome is postulated to play a role in the increased risk of infection,
but this theory needs to be confirmed. These observations require
confirmation but should alert the practitioner to take caution when
recommending these agents for long-term use, especially in elderly
patients at risk for developing pneumonia or C. difficile infection.
Diarrhea is also associated with PPI use, which in some cases has
been associated with the development of collagenous colitis (hazard
ratio of 4.5), particularly with lansoprazole. The mechanism for
PPI-induced collagenous colitis is unclear, but in vitro studies dem­
onstrate that PPIs may induce collagen gene expression. The colitis
usually resolves with cessation of the PPI.
A population-based study revealed that long-term use of PPIs was
associated with the development of hip fractures in older women.
The absolute risk of fracture remained low and may be zero despite
an observed increase associated with the dose and duration of acid
suppression. The mechanism for this observation is not clear, and
this finding must be confirmed before making broad recommenda­
tions regarding the discontinuation of these agents in patients who
benefit from them. Long-term use of PPIs has also been implicated
in the development of iron, vitamin B12, and magnesium deficiency.
A meta-analysis of nine observational studies found a 40% increase
in hypomagnesemia in PPI users as compared to nonusers. One
approach to consider in patients needing to take PPIs long term is to
check a complete blood count looking for evidence of anemia due to
iron or vitamin B12 deficiency, vitamin B12 level, and a magnesium
level after 1–2 years of PPI use, but these recommendations are not
evidence based or recommended by expert opinion. PPIs may exert
a negative effect on the antiplatelet effect of clopidogrel. Although
the evidence is mixed and inconclusive, a small increase in mortal­
ity and readmission rate for coronary events was seen in patients
receiving a PPI while on clopidogrel in earlier studies. Subsequently,
three meta-analyses reported an inverse correlation between clopi­
dogrel and PPI use; therefore, the influence of this drug interaction
on mortality is not clearly established. The mechanism involves the
competition of the PPI and clopidogrel with the same cytochrome
P450 (CYP2C19). Whether this is a class effect of PPIs is unclear;
there appears to be at least a theoretical advantage of pantoprazole
over the other PPIs, but this has not been confirmed. This drug
interaction is particularly relevant in light of the common use of
aspirin and clopidogrel for prevention of coronary events and the
efficacy of PPIs in preventing GI bleeding in these patients. The
FDA has made several recommendations while awaiting further
evidence to clarify the impact of PPI therapy on clopidogrel use.
Health care providers should continue to prescribe clopidogrel to
patients who require it and should reevaluate the need for starting
or continuing treatment with a PPI. From a practical standpoint,
additional recommendations to consider include the following:
Patients taking clopidogrel with aspirin, especially with other GI
risk factors for bleeding, should receive GI protective therapy.
Although high-dose H2 blockers have been considered an option,
these do not appear to be as effective as PPIs. If PPIs are to be given,
some have recommended that there be a 12-h separation between
Bone fracture
C. difficile infection
Community-acquired
pneumonia
(2 studies)
Side effect
Hypomagnesemia
Acute interstitial
nephritis
Acute kidney disease
Chronic
kidney disease
2.00
2.50
3.00
1.50
1.00
0.50
5.00
5.50
6.00
4.50
4.00
3.50
0.00
Adjusted odds ratio (95% confidence interval)
FIGURE 335-12.  Evidence supporting the potential adverse effects of proton pump
inhibitor drugs. (Adapted from AJ Schoenfeld, D Grady: Adverse effects associated
with proton pump inhibitors. JAMA Intern Med 176:172, 2016.)
administration of the PPI and clopidogrel to minimize competition
of the two agents with the involved cytochrome P450. One option is
to give the PPI 30 min before breakfast and the clopidogrel at bed­
time. Insufficient data are available to firmly recommend one PPI
over another. Additional concerning side effects with long-term PPI
use include increased cardiac risks independent of clopidogrel use,
dementia, and acute and chronic kidney injury. Again, the data are
often retrospective and confounding variables were not consistently
eliminated, thus making it difficult to establish a definitive association
between PPIs and the toxicities outlined. A summary of the side
effects with the corresponding relative risks is shown in Fig. 335-12.
Ultimately, heightened awareness of inappropriate long-term use of
PPIs is paramount. Patients aged ≥65 years of age have a higher risk
for some of the long-term side effects of PPIs highlighted above, in
part due to the higher prevalence of concomitant chronic diseases.
It is therefore essential to carefully select individuals, especially
among the elderly, who need long-term PPI therapy and discon­
tinue it in those individuals who do not need it. Abrupt withdrawal
of a PPI in a long-term user may result in a component of rebound
hyperacidity; thus, this agent should be tapered gradually over the
course of 1–2 weeks with possible transition to an H2 blocker for a
short period of time.
PART 10
Disorders of the Gastrointestinal System
Development of novel acid inhibitory agents continues in an
attempt to primarily address the need for better agents to treat
GERD. For example, modified H2 blockers with greater potency
and duration as well as novel PPIs with longer half-life and potency
are under study. For example, tenatoprazole is a PPI containing
an imidazopyridine ring instead of a benzimidazole ring, which
promotes irreversible proton pump inhibition. This agent has a
longer half-life than the other PPIs and may be beneficial for inhib­
iting nocturnal acid secretion, which has significant relevance in
GERD. Additional PPIs with longer half-life and combined with
other agents are being studied, but the details are beyond the scope
of this chapter. A second new class of agents is the potassiumcompetitive acid pump antagonists (P-CAPs). These compounds
inhibit gastric acid secretion via potassium competitive binding
of the H+,K+-ATPase. Revaprazan, vonoprazan and tegoprazan are
agents approved for use in Korea and Japan, and vonoprazan has
been approved by the FDA for use in the United States. Vonoprazan
may be superior to PPIs when combined with antibiotics for the
treatment of H. pylori, and this novel agent was awarded Fast Track
status by the FDA for the treatment of H. pylori in combination with
both amoxicillin and clarithromycin and with amoxicillin alone.
CYTOPROTECTIVE AGENTS
Sucralfate  Sucralfate is a complex sucrose salt in which the
hydroxyl groups have been substituted by aluminum hydroxide and
sulfate. This compound is insoluble in water and becomes a viscous
paste within the stomach and duodenum, binding primarily to sites
of active ulceration. Sucralfate may act by several mechanisms:
serving as a physicochemical barrier, promoting a trophic action
by binding growth factors such as EGF, enhancing prostaglan­
din synthesis, stimulating mucus and bicarbonate secretion, and
enhancing mucosal defense and repair. Toxicity from this drug
is rare, with constipation being most common (2–3%). It should
be avoided in patients with chronic renal insufficiency to prevent
aluminum-induced neurotoxicity. Hypophosphatemia and gastric
bezoar formation have also been reported rarely. Standard dosing
of sucralfate is 1 g qid.
Bismuth-Containing Preparations  Sir William Osler considered
bismuth-containing compounds the drug of choice for treating
PUD. The resurgence in the use of these agents is due to their effect
against H. pylori. Colloidal bismuth subcitrate (CBS) and bismuth
subsalicylate (BSS; Pepto-Bismol) are the most widely used prepara­
tions. The mechanism by which these agents induce ulcer healing
is unclear. Adverse effects with short-term use include black stools,
constipation, and darkening of the tongue. Long-term use with high
doses, especially with the avidly absorbed CBS, may lead to neuro­
toxicity. These compounds are commonly used as one of the agents
in an anti–H. pylori regimen (see below).
6.50
Prostaglandin Analogues  In view of their central role in main­
taining mucosal integrity and repair, stable prostaglandin analogues
were developed for the treatment of PUD. The mechanism by which
this rapidly absorbed drug provides its therapeutic effect is through
enhancement of mucosal defense and repair. The most common
toxicity noted with this drug is diarrhea (10–30% incidence).
Other major toxicities include uterine bleeding and contractions;
misoprostol is contraindicated in women who may be pregnant,
and women of childbearing age must be made clearly aware of this
potential drug toxicity. The standard therapeutic dose is 200 μg qid.
Miscellaneous Drugs  A number of drugs including anticholiner­
gic agents and tricyclic antidepressants were used for treating acid
peptic disorders, but in light of their toxicity and the development
of potent antisecretory agents, these are rarely, if ever, used today.
Newer agents such as teprenone, an acyclic polyisoprenoid com­
pound used as a gastric mucosal protector that is employed to treat
gastritis and GUs outside of the United States; plant-based thera­
pies; and CCK2 receptor antagonists are intriguing therapies but
require further evaluation.
THERAPY OF H. PYLORI
The physician’s goal in treating PUD is to provide relief of symp­
toms (pain or dyspepsia), promote ulcer healing, and ultimately
prevent ulcer recurrence and complications. The greatest influence
of understanding the role of H. pylori in peptic disease has been the
ability to prevent recurrence. Documented eradication of H. pylori
in patients with PUD is associated with a dramatic decrease in ulcer
recurrence to <10–20% as compared to 59% in GU patients and
67% in DU patients when the organism is not eliminated. Eradica­
tion of the organism may lead to diminished recurrent ulcer bleed­
ing. The effect of its eradication on ulcer perforation is unclear.
Extensive effort has been made in determining who of the many
individuals with H. pylori infection should be treated. The common
conclusion arrived at by multiple consensus conferences around the
world is that H. pylori should be eradicated in patients with docu­
mented PUD. This holds true independent of time of presentation
(first episode or not), severity of symptoms, presence of confound­
ing factors such as ingestion of NSAIDs, or whether the ulcer is in
remission. Some have advocated treating patients with a history of
documented PUD who are found to be H. pylori positive by stool
antigen or breath testing. Between 60 and 90% of patients with
gastric MALT lymphoma experience complete remission of the
tumor in response to H. pylori eradication. The Maastricht VI/Flor­
ence Consensus Report recommends a test-and-treat approach for
patients with uninvestigated dyspepsia if the local incidence of H.
pylori is >20%. The American College of Gastroenterology (ACG)
clinical guidelines (developed for North America) recommend that
individuals aged <60 years with uninvestigated dyspepsia should
be tested and treated for H. pylori. In addition, recommendations
from this consensus report and the ACG clinical guidelines include
testing and offering eradication of H. pylori in patients who will be
using NSAIDs (including low-dose aspirin) on a long-term basis,
especially if there is a prior history of PUD. These individuals will
require continued PPI treatment as well as eradication treatment,
because eradication of the organism alone does not eliminate the
risk of gastroduodenal ulcers in patients already receiving longterm NSAIDs. Moreover, it appears that eradication of H pylori
decreases NSAID-induced GI bleeding. Treating patients with
NUD to prevent gastric cancer or patients with GERD requiring
long-term acid suppression remains controversial. Guidelines from
the ACG suggest eradication of H. pylori in patients who have
undergone resection of early gastric cancer. The Maastricht VI/
Florence Consensus Report also evaluated H. pylori treatment in
gastric cancer prevention and recommends that eradication should
be considered in the following situations: first-degree relatives of
family members with gastric cancer; patients with previous gastric
neoplasm treated by endoscopic or subtotal resection; individuals
with a risk of gastritis (severe pangastritis or body-predominant
gastritis) or severe atrophy; patients with gastric acid inhibition
for >1 year; individuals with strong environmental risk factors for
gastric cancer (heavy smoking; high exposure to dust, coal, quartz,
or cement; and/or work in quarries); and H. pylori–positive patients
with a fear of gastric cancer. Finally, the ACG clinical guidelines
recommend testing and offering H. pylori eradication to patients
with unexplained iron-deficiency anemia and idiopathic thrombo­
cytopenic purpura. The Maastricht VI/Florence Consensus Report
concurs with the ACG with the aforementioned recommendations
and in addition recommends eradicating H. pylori in patients
TABLE 335-4  Recommended First-Line Therapies for H. pylori Infection
REGIMEN
DRUGS (DOSES)
DOSING FREQUENCY
DURATION (DAYS)
FDA APPROVAL
Clarithromycin triple (Only in
patients without prior exposure
to macrolides, incidence of
clarithromycin <15%, or either of
the above unknown) 
PPI (standard or double dose)
bid
Yesa
Clarithromycin (500 mg)
 
 
 
Amoxicillin (1 g) or metronidazole (500 mg tid)
 
 
 
Bismuth quadruple (Now often
recommended as first-line
therapy)
PPI (standard dose)
bid
10–14
Nob
Bismuth subcitrate (120–300 mg) or subsalicylate (300 mg)
qid
 
 
Tetracycline (500 mg)
qid
 
 
Metronidazole (250–500 mg)
qid (250 mg)
tid to qid (500 mg)
Concomitant
PPI (standard dose)
bid
10–14
No
 
Clarithromycin (500 mg)
 
 
 
 
Amoxicillin (1 g)
 
 
 
 
Nitroimidazole (500 mg)c
 
 
 
Sequential
PPI (standard dose)
bid
5–7
No
 
PPI, clarithromycin (500 mg) + nitroimidazole (500 mg)c
bid
5–7
 
Hybrid
PPI (standard dose) + amoxicillin (1 g)
bid
No
 
PPI, amoxicillin, clarithromycin (500 mg), nitroimidazole (500 mg)c
bid
Levofloxacin triple
PPI (standard or double dose) + amoxicillin (1 g)
bid
5–7
No
 
Levofloxacin (500 mg)
qd
 
 
 
Amoxicillin (1 g)
bid
 
 
Levofloxacin sequential
PPI (standard or double dose) + amoxicillin (1 g)
bid
5–7
No
 
PPI, amoxicillin, levofloxacin (500 mg qd), nitroimidazole (500 mg)c
bid
5–7
 
LOAD
Levofloxacin (250 mg)
qd
7–10
No
 
PPI (double dose)
qd
 
 
 
Nitazoxanide (500 mg)
bid
 
 
 
Doxycycline (100 mg)
qd
 
 
aSeveral PPI, clarithromycin, and amoxicillin combinations have achieved FDA approval. The regimen of a PPI, clarithromycin, and metronidazole is not an FDA-approved
treatment regimen. bThe regimen of a PPI, bismuth, tetracycline, and metronidazole combined with a PPI for 10 days is an FDA-approved treatment regimen. cMetronidazole
or tinidazole.
Abbreviations: bid, twice daily; FDA, Food and Drug Administration; PPI, proton pump inhibitor; tid, three times daily; qd, once daily; qid, four times daily.
Source: Reproduced with permission from WD Chey et al: ACG clinical guideline: Treatment of Helicobacter pylori infection. Am J Gastroenterol 112:212, 2017.
with unexplained vitamin B12 deficiency. Despite this, concerns
have been raised about the widespread use of antibiotics for the
therapy of all cases of H. pylori positivity, including the potential
for increased bacterial resistance rates, reported weight gain, and
alteration of the microbiome. In fact, the increasing incidence of
bacterial resistance to antibiotics coupled with variability in general
prevalence of the organism worldwide and limitations in develop­
ing personalized H. pylori–driven sensitivity therapies have led to
increasing challenges in effectively treating this organism.
Multiple drugs have been evaluated in the therapy of H. pylori.
No single agent is effective in eradicating the organism. Combina­
tion therapy for 14 days provides the greatest efficacy, although
regimens based on sequential administration of antibiotics also
appear promising (see below). A shorter administration course
(7–10 days), although attractive, has not proved as successful as
the 14-day regimens. The agents used with the greatest frequency
include amoxicillin, metronidazole, tetracycline, clarithromycin,
and bismuth compounds. Moreover, the steady increase in H pylori
resistance to antibiotics, in particular clarithromycin and metro­
nidazole, has significantly impacted the approach to eradication
of this organism (see below). In addition, increasing resistance to
antibiotics has led to a higher incidence of H. pylori infections that
are refractory to first-line therapies.
Suggested treatment regimens for H. pylori are outlined in
Table 335-4. Choice of a particular regimen will be influenced by
several factors, including efficacy, patient tolerance, existing antibi­
otic resistance, prior antibiotic use, and cost of the drugs. The aim
for initial eradication rates should be 85–90%. The combination
of bismuth, metronidazole, and tetracycline was the first triple
regimen found effective against H. pylori. The combination of two
CHAPTER 335
Peptic Ulcer Disease and Related Disorders
antibiotics plus either a PPI, H2 blocker, or bismuth compound has
comparable success rates. Addition of acid suppression assists in
providing early symptom relief and enhances bacterial eradication.
Triple therapy, although effective, has several drawbacks, includ­
ing the potential for poor patient compliance and drug-induced side
effects. Compliance is being addressed by simplifying the regimens
so that patients can take the medications twice a day. Simpler (dual
therapy) and shorter regimens (7 and 10 days) are not as effective as
triple therapy for 14 days. One anti–H. pylori regimen is available in
prepackaged formulation: Prevpac (lansoprazole, clarithromycin, and
amoxicillin).The contents of the Prevpac are to be taken twice per day
for 14 days, understanding that clarithromycin-based triple therapy
should be avoided in settings where H. pylori resistance to this agent
exceeds 15% or the patient has been recently exposed to macrolide
antibiotics or if neither of these factors is known (see below).
Side effects have been reported in up to 20–30% of patients on
triple therapy. Bismuth may cause black stools, constipation, or
darkening of the tongue. The most feared complication with amoxi­
cillin is pseudomembranous colitis, but this occurs in <1–2% of
patients. Amoxicillin can also lead to antibiotic-associated diarrhea,
nausea, vomiting, skin rash, and allergic reaction. Concomitant use
of probiotics may ameliorate some of the antibiotic side effects (see
below). Tetracycline has been reported to cause rashes and, very
rarely, hepatotoxicity and anaphylaxis.
One important concern with treating patients who may not
need therapy is the potential for development of antibiotic-resistant
strains. The incidence and type of antibiotic-resistant H. pylori
strains vary worldwide. Strains resistant to metronidazole, clar­
ithromycin, amoxicillin, and tetracycline have been described, with
the latter two being uncommon. Antibiotic-resistant strains are the
most common cause for treatment failure in compliant patients.
Unfortunately, in vitro resistance does not predict outcome in
patients. Culture and sensitivity testing of H. pylori is not performed
routinely. In light of this, culture-independent methods are needed
to determine antibiotic resistance. Detection of resistance toward
several antibiotics can be achieved by detecting different H. pylori
mutations or other genetic changes, but the accuracy of the molecu­
lar tools available vary widely between different antibiotics. There­
fore, a standardized approach to determine antibiotic resistance that
is easy to use is still not readily available. In addition, the standard
method of obtaining tissue through endoscopy is cost prohibitive
when considering the widespread presence of this organism in
certain parts of the world; thus, new efforts are aimed at develop­
ing techniques that capitalize on DNA extraction and performing
polymerase chain reaction (PCR) on stool samples from patients
infected with H. pylori. These diagnostic modalities are still in early
phases of development. In light of this, most recommendations
for treatment of H. pylori outline that if individual susceptibility
testing is not available, first-line therapy in areas of a high inci­
dence of clarithromycin resistance (>15%) or in situations where
resistance to clarithromycin is not known that bismuth quadruple
therapy should be the first line of treatment. If not available, non­
bismuth concomitant quadruple therapy may be a consideration.
An algorithmic approach to the therapy of H. pylori as outlined in
the Maastricht VI/Florence Consensus Report is provided in Fig.
335-13. Although resistance to metronidazole has been found in as
many as 30% of isolates in North America and 80% in developing
countries, triple therapy is effective in eradicating the organism in
50% of patients infected with a resistant strain. Clarithromycin
resistance is seen in 13–16% of individuals in the United States,
with resistance to amoxicillin being <1% and resistance to both
metronidazole and clarithromycin in the 5% range. Resistance to
tetracycline and rifabutin (see below) is reported to be <2% in the
United States. In light of the paucity of H. pylori antibiotic real-time
resistance data, asking the patient about prior antibiotic exposure
should be included in the decision-making and used as a surrogate for
potential antibiotic resistance, especially when it comes to prior
macrolide use. Clarithromycin use should be excluded in patients
with prior macrolide usage.
PART 10
Disorders of the Gastrointestinal System
Failure of H. pylori eradication with triple therapy in a compliant
patient is usually due to infection with a resistant organism. Addi­
tional important factors in treatment failure include inadequate acid
suppression and inadequate adherence to the recommended regi­
men. Addressing both of these latter issues is critical before embark­
ing on selecting alternative antibiotics for eradication of H. pylori.
An in-depth review of the approach to patients with refractory H.
pylori is beyond the scope of this chapter, and readers are referred to
the American Gastroenterological Association 2021 guidelines (see
Further Reading) for in-depth discussion of this important topic. A
series of salvage therapies for H. pylori are shown in Table 335-5.
The combination of PPI, amoxicillin, and rifabutin for 10 days has
also been used successfully (86% cure rate) in patients infected with
resistant strains. Additional regimens considered for second-line
therapy include levofloxacin-based triple therapy (levofloxacin,
amoxicillin, PPI) for 10 days and furazolidone-based triple therapy
(furazolidone, amoxicillin, PPI) for 14 days. Unfortunately, no
treatment regimen is universally accepted for patients in whom two
courses of antibiotics have failed. If eradication is still not achieved
in a compliant patient, then culture and sensitivity of the organism
should be considered. One challenge with this approach is that
culture and sensitivity testing is cumbersome and not widely avail­
able; thus, H. pylori resistance data within specific communities are
often not available. Non-culture-based approaches using molecular
markers to determine potential resistance through stool testing are
being developed but are not widely available. Additional factors that
may lower eradication rates include the patient’s country of origin
(higher in Northeast Asia than other parts of Asia or Europe) and
cigarette smoking. In addition, meta-analysis suggests that even the
most effective regimens (quadruple therapy including PPI, bismuth,
tetracycline, and metronidazole and triple therapy including PPI,
clarithromycin, and amoxicillin) may have suboptimal eradication
rates (<80%), thus demonstrating the need for the development of
more efficacious treatments.
In view of the observation that 15–25% of patients treated with
first-line therapy may still remain infected with the organism,
new approaches to treatment have been explored. One promising
approach is sequential therapy. Regimens examined consist of 5 days
of amoxicillin and a PPI, followed by an additional 5 days of PPI
plus tinidazole and clarithromycin or levofloxacin. One promising
regimen that has the benefit of being shorter in duration, easier
to take, and less expensive is 5 days of concomitant therapy (PPI
twice daily, amoxicillin 1 g twice daily, levofloxacin 500 mg twice
daily, and tinidazole 500 mg twice daily). Initial studies have dem­
onstrated eradication rates of >90% with good patient tolerance.
Confirmation of these findings and applicability of this approach
in the United States are needed, although some experts are recom­
mending abandoning clarithromycin-based triple therapy in the
United States for the concomitant therapy or the alternative sequen­
tial therapies highlighted above.
Innovative non-antibiotic-mediated approaches have been
explored in an effort to improve eradication rates of H. pylori.
Pretreatment of patients with N-acetylcysteine as a mucolytic agent
to destroy the H. pylori biofilm and therefore impair antibiotic
resistance has been examined, but more studies are needed to con­
firm the applicability of this approach. In vitro studies suggest that
certain probiotics like Lactobacillus or its metabolites can inhibit
H. pylori. Administration of probiotics has been attempted in sev­
eral clinical studies in an effort to maximize antibiotic-mediated
eradication with varying results. Overall, it appears that the use of
certain probiotics, such as Lactobacillus spp., Saccharomyces spp.,
Bifidobacterium spp., and Bacillus clausii, did not alter eradication
rates but importantly decreased antibiotic-associated side effects
including nausea, dysgeusia, diarrhea, and abdominal discomfort/
pain, resulting in enhanced tolerability of H. pylori therapies.
Additional studies are needed to confirm the potential benefits
of probiotics in this setting. Statins, specifically atorvastatin, have
been used with some success as an adjunct to quadruple therapy
in patients with NUD. Additional nonpharmacologic therapies are
Low (<15%)
clarithromycin resistance
Bismuth 4-drug
1st line
If fails
Levofloxacin 4- or
3-drug
OR
2nd line
If fails
If fails
If fails
Clarithromycin 3- or
4-drug
Levofloxacin 4- or
3-drug
Bismuth 4-drug
3rd line
If fails
If fails
Rifabutin 3-drug
Rifabutin 3-drug
Rifabutin 3-drug
4th line
A
High (>15%)
or unknown clarithromycin resistance
1st option
Bismuth 4-drug
1st line
If fails
Levofloxacin 4- or
3-drug
2nd line
If fails
Rifabutin 3-drug
3rd line
4th line
B
FIGURE 335-13  Empirical approach to Helicobacter pylori eradication if individual antibiotic susceptibility testing is not avaliable. A. Low <15% clarithromycin resistance;
B. High >15% or unknown clarithromycin resistance. Treatment regimens: Bismuth 4-drug—proton pump inhibitor (PPI), bismuth, tetracycline, metronidazole; Clarithromycin
3-drug—PPI, clarithromycin, amoxicillin (use only if clarithromycin sensitivity is known); Nonbismuth 4-drug—PPI, clarithromycin, amoxicillin, metronidazole; Levofloxacin
4-drug—PPI, levofloxacin, amoxicillin, bismuth; Levofloxacin 3-drug—same as levofloxacin 4-drug omitting bismuth; Rifabutin 3-drug in the setting of high fluoroquinolone
resistance—rifabutin, high dose PPI, amoxicillin. (modified from Malfetheiner P, et al: Gut 71:9, 2022). (Adapted from permission from P Malfertheiner et al: Management of
Helicobacter pylori infection-the Maastricht V/Florence Consensus Report. Gut 71:9, 2022.)
also being explored including oxygen-enriched environment or
hyperbaric oxygen therapy, antimicrobial photodynamic therapy,
nanomaterials, antimicrobial peptides, phase therapy, and modified
lysins. Readers are referred to a recent comprehensive review on
this topic (Luo et al in Further Reading).
Reinfection after successful eradication of H. pylori is rare in the
United States (<1% per year). If recurrent infection occurs within
OR
Clarithromycin 3-drug
If fails
Levofloxacin 4- or
3-drug
Bismuth 4-drug
If fails
CHAPTER 335
2nd option (bismuth 4-drug
locally unavailable)
Non-bismuth 4-drug
(concomitant)
Peptic Ulcer Disease and Related Disorders
OR
Levofloxacin 4- or
3-drug
Bismuth 4-drug
If fails
If fails
Levofloxacin 4- or
3-drug
Bismuth 4-drug
If fails
If fails
Rifabutin 3-drug
Rifabutin 3-drug
the first 6 months after completing therapy, the most likely explana­
tion is recrudescence as opposed to reinfection.
THERAPY OF NSAID-RELATED GASTRIC OR
DUODENAL INJURY
Medical intervention for NSAID-related mucosal injury includes
treatment of an active ulcer and primary prevention of future
TABLE 335-5  Salvage Therapies for H. pylori Infection
REGIMEN
DRUGS (DOSES)
DOSING FREQUENCY
DURATION (DAYS)
FDA APPROVAL
Bismuth quadruple
PPI (standard dose)
bid
Noa
 
Bismuth subcitrate (120–300 mg) or subsalicylate (300 mg)
qid
 
 
 
Tetracycline (500 mg)
qid
 
 
 
Metronidazole (500 mg)
tid or qid
 
 
Levofloxacin triple
PPI (standard dose)
bid
No
 
Levofloxacin (500 mg)
qd
 
 
 
Amoxicillin (1 g)
bid
 
 
Concomitant
PPI (standard dose)
bid
10–14
No
 
Clarithromycin (500 mg)
bid
 
 
 
Amoxicillin (1 g)
bid
 
 
 
Nitroimidazole (500 mg)
bid or tid
 
 
Rifabutin triple
PPI (standard dose)
bid
No
 
Rifabutin (300 mg)
qd
 
 
 
Amoxicillin (1 g)
bid
 
 
High-dose dual
PPI (standard to double dose)
tid or qid
No
 
Amoxicillin (1 g tid or 750 mg qid)
tid or qid
 
 
aPPI, bismuth, tetracycline, and metronidazole prescribed separately is not an FDA-approved treatment regimen. However, Pylera, a combination product containing bismuth
subcitrate, tetracycline, and metronidazole, combined with a PPI for 10 days is an FDA-approved treatment regimen.
Abbreviations: bid, twice daily; FDA, Food and Drug Administration; PPI, proton pump inhibitor; tid, three times daily; qd, once daily; qid, four times daily.
Source: Reproduced with permission from WD Chey et al: ACG clinical guideline: Treatment of Helicobacter pylori infection. Am J Gastroenterol 112:212, 2017.
injury. Recommendations for the treatment and primary preven­
tion of NSAID-related mucosal injury are listed in Table 335-6.
Ideally, the injurious agent should be stopped as the first step in the
therapy of an active NSAID-induced ulcer. If that is possible, then
treatment with one of the acid inhibitory agents (H2 blockers, PPIs)
is indicated. Cessation of NSAIDs is not always possible because of
the patient’s severe underlying disease. Only PPIs can heal GUs or
DUs, independent of whether NSAIDs are discontinued.
PART 10
Disorders of the Gastrointestinal System
The widespread use of NSAIDs has created some concern due to
the increasing likelihood of GI and CV side effects associated with
these agents. The approach to primary prevention has included
avoiding the agent, using the lowest possible dose of the agent for
the shortest period of time possible, using NSAIDs that are theoret­
ically less injurious, using newer topical NSAID preparations, and/
or using concomitant medical therapy to prevent NSAID-induced
injury. Several nonselective NSAIDs that are associated with a lower
likelihood of GI and CV toxicity include naproxen and ibuprofen,
although the beneficial effect may be eliminated if higher dosages
of the agents are used. Primary prevention of NSAID-induced
ulceration can be accomplished by a PPI and, if not tolerated,
misoprostol (200 μg qid). High-dose H2 blockers (famotidine 40 mg
bid) have also shown some promise in preventing endoscopically
documented ulcers, although PPIs are superior. The highly selec­
tive COX-2 inhibitors, celecoxib and rofecoxib, are 100 times more
selective inhibitors of COX-2 than standard NSAIDs, leading to
gastric or duodenal mucosal injury that is comparable to placebo.
TABLE 335-6  Recommendations for Treatment of NSAID-Related
Mucosal Injury
CLINICAL SETTING
RECOMMENDATION
Active ulcer
 
  NSAID discontinued
H2 receptor antagonist or PPI
  NSAID continued
PPI
Prophylactic therapy
Misoprostol
 
PPI
 
Selective COX-2 inhibitor
H. pylori infection
Eradication if tests positive for H. pylori
Abbreviations: COX-2, isoenzyme of cyclooxygenase; NSAID, nonsteroidal antiinflammatory drug; PPI, proton pump inhibitor.
Their utilization led to an increase in CV events initially leading
to both celecoxib and rofecoxib being withdrawn from the market.
After continued analysis of available data, the FDA concluded that
celecoxib is no less safe than several nonopioid pain killers and has
been allowed to remain on the market with a warning label that
it was associated with increased risk of heart attacks and strokes.
Additional caution was engendered when the CLASS study demon­
strated that the advantage of celecoxib in preventing GI complica­
tions was offset when low-dose aspirin was used simultaneously.
Therefore, gastric protection therapy is required in individuals
taking COX-2 inhibitors and aspirin prophylaxis. Finally, much of
the work demonstrating the benefit of COX-2 inhibitors and PPIs
on GI injury has been performed in individuals of average risk; it
is unclear if the same level of benefit will be achieved in high-risk
patients. For example, concomitant use of warfarin and a COX-2
inhibitor was associated with rates of GI bleeding similar to those
observed in patients taking nonselective NSAIDs. A combination of
factors, including withdrawal of the majority of COX-2 inhibitors
from the market, the observation that low-dose aspirin appears to
diminish the beneficial effect of COX-2–selective inhibitors, and
the growing use of aspirin for prophylaxis of CV events, has signifi­
cantly altered the approach to gastric protective therapy during the
use of NSAIDs. A set of guidelines for the approach to the use of
NSAIDs was published by the ACG and is shown in Table 335-7.
TABLE 335-7  Guide to NSAID Therapy
 
NO/LOW NSAID GI RISK
NSAID GI RISK
No CV risk (no
aspirin)
Traditional NSAID
Coxib or
Traditional NSAID + PPI or
misoprostol
Consider non-NSAID therapy
CV risk
(consider
aspirin)
Traditional NSAID + PPI or
misoprostol if GI risk warrants
gastroprotection
Consider non-NSAID therapy
A gastroprotective agent must
be added if a traditional NSAID
is prescribed
Consider non-NSAID therapy
Abbreviations: CV, cardiovascular; GI, gastrointestinal; NSAID, nonsteroidal antiinflammatory drug; PPI, proton pump inhibitor.
Source: Reproduced with permission from AM Fendrick: Am J Manag Care 10:740,
2004.
Individuals who are not at risk for CV events, do not use aspirin,
and are without risk for GI complications can receive nonselective
NSAIDs without gastric protection. In those without CV risk fac­
tors but with a high potential risk (prior GI bleeding or multiple
GI risk factors) for NSAID-induced GI toxicity, cautious use of a
selective COX-2 inhibitor and co-therapy with high-dose PPI or
misoprostol are recommended. Individuals at moderate GI risk
without cardiac risk factors can be treated with a COX-2 inhibi­
tor alone or with a nonselective NSAID with PPI or misoprostol.
Individuals with CV risk factors, who require low-dose aspirin and
have low potential for NSAID-induced toxicity, should be consid­
ered for a non-NSAID agent or use of a traditional NSAID such
as naproxen (lower CV side effects) in combination with gastric
protection, if warranted. Finally, individuals with CV and GI risks
who require aspirin must be considered for non-NSAID therapy,
but if that is not an option, then gastric protection with any type of
NSAID must be considered. Any patient, regardless of risk status,
who is being considered for long-term traditional NSAID therapy
should also be considered for H. pylori testing and treatment if
positive. Assuring the use of GI protective agents with NSAIDs is
difficult, even in high-risk patients. This is in part due to under­
prescribing of the appropriate protective agent; other times, the
difficulty is related to patient compliance. The latter may be due to
patients forgetting to take multiple pills or preferring not to take the
extra pill, especially if they have no GI symptoms. Several NSAID
gastroprotective-containing combination pills are now commer­
cially available, including double-dose famotidine with ibuprofen,
diclofenac with misoprostol, and naproxen with esomeprazole.
Although initial studies suggested improved compliance and a
cost advantage when taking these combination drugs, their clinical
benefit over the use of separate pills has not been established. One
additional concern with NSAID-induced GI complications is the
relatively low rate of primary care provider compliance with estab­
lished guidelines outlining preventative measures. An intervention
including professional education, informatics to facilitate review,
and financial incentives for practices to review patients’ charts to
assess appropriateness showed a reduced rate of high-risk prescrib­
ing of antiplatelet medications and NSAIDs with a tendency toward
improved clinical outcomes. Efforts continue toward developing
safer NSAIDs, including topical NSAIDs, NSAID formulations that
are rapidly absorbed (diclofenac potassium powder mixed with a
buffering agent, Prosorb and SoluMatrix technology), NO-releasing
NSAIDs, hydrogen sulfide–releasing NSAIDs, dual COX/5-LOX
inhibitors, NSAID prodrugs, and agents that can effectively seques­
ter unbound NSAIDs without interfering with their efficacy.
APPROACH AND THERAPY: SUMMARY
Controversy continues regarding the best approach to the patient
who presents with dyspepsia (Chap. 48). The discovery of H. pylori
and its role in pathogenesis of ulcers has added a new variable to
the equation. Previously, if a patient <60 years of age presented with
dyspepsia and without alarming signs or symptoms suggestive of
an ulcer complication or malignancy, an empirical therapeutic trial
with acid suppression was commonly recommended; today, how­
ever, a test-and-treat approach with a noninvasive diagnostic tool
for H. pylori and eradication of the organism, if positive, is recom­
mended (Fig. 335-14).
Once an ulcer (GU or DU) is documented, the main issue to
consider is whether H. pylori or an NSAID is involved. With H.
pylori present, independent of the NSAID status, triple or qua­
druple therapy is recommended for 14 days, followed by continued
acid-suppressing drugs (H2 receptor antagonist or PPIs) for a
total of 4–6 weeks. H. pylori eradication should be documented
4 weeks after completing antibiotics. The test of choice for docu­
menting eradication is the laboratory-based validated monoclonal
stool antigen test or a urea breath test (UBT). The patient must be
off antisecretory agents for at least 7 days when being tested for
eradication of H. pylori with UBT or stool antigen. Serologic testing
is not useful for the purpose of documenting eradication because
New-Onset Dyspepsia
40 years old
Alarm symptoms
Exclude by history GERD, biliary
pain, IBS, aerophagia,
medication-related
–
Noninvasive Hp testing
–
+
or
Anti-Hp
therapy
Refer to
gastroenterologist
Empiric trial
H2 blocker
4 weeks
after therapy
Confirm eradication UBT
Symptoms remain or recur
FIGURE 335-14  Overview of new-onset dyspepsia. GERD, gastroesophageal reflux
disease; Hp, Helicobacter pylori; IBS, irritable bowel syndrome; UBT, urea breath
test. (Reproduced with permission from BS Anand, DY Graham: State-of-the-Art:
Ulcer and Gastritis, Endoscopy 31:215, 1999. © Georg Thieme Verlag KG.)
CHAPTER 335
antibody titers fall slowly and often do not become undetectable.
Some recommend that patients with complicated ulcer disease or
who are frail should be treated with long-term acid suppression,
thus making documentation of H. pylori eradication a moot point.
In view of this discrepancy in practice, it would be best to discuss
with the patient the different options available.
Peptic Ulcer Disease and Related Disorders
Several issues differentiate the approach to a GU versus a DU.
GUs, especially of the body and fundus, have the potential of being
malignant. Multiple biopsies of a GU should be taken initially; even
if these are negative for neoplasm, repeat endoscopy to document
healing at 8–12 weeks should be performed, with biopsy if the ulcer
is still present. About 70% of GUs eventually found to be malignant
undergo significant (usually incomplete) healing. Repeat endos­
copy is warranted in patients with DU if symptoms persist despite
medical therapy or a complication is suspected.
The majority (>90%) of GUs and DUs heal with the conventional
therapy outlined above. A GU that fails to heal after 12 weeks and
a DU that does not heal after 8 weeks of therapy should be con­
sidered refractory. Once poor compliance and persistent H. pylori
infection have been excluded, NSAID use, either inadvertent or sur­
reptitious, must be excluded. In addition, cigarette smoking must be
eliminated. For a GU, malignancy must be meticulously excluded.
Next, consideration should be given to a gastric acid hypersecretory
state such as ZES (see “Zollinger-Ellison Syndrome,” below) or the
idiopathic form, which can be excluded with gastric acid analysis.
Although a subset of patients has gastric acid hypersecretion of
unclear etiology as a contributing factor to refractory ulcers, ZES
should be excluded with a fasting gastrin or secretin stimulation
test (see below). More than 90% of refractory ulcers (either DUs
or GUs) heal after 8 weeks of treatment with higher doses of PPI
(omeprazole 40 mg/d; lansoprazole 30–60 mg/d). This higher dose
is also effective in maintaining remission. Surgical intervention
may be a consideration at this point; however, other rare causes of
refractory ulcers must be excluded before recommending surgery.
Rare etiologies of refractory ulcers that may be diagnosed by gastric
or duodenal biopsies include ischemia, Crohn’s disease, amyloido­
sis, sarcoidosis, lymphoma, eosinophilic gastroenteritis, smoking
crack cocaine, or infection (cytomegalovirus [CMV], tuberculosis,
or syphilis).
SURGICAL THERAPY
Surgical intervention in PUD can be viewed as being either elec­
tive, for treatment of medically refractory disease, or as urgent/
emergent, for the treatment of an ulcer-related complication. The
development of pharmacologic and endoscopic approaches for
the treatment of peptic disease and its complications has led to a
substantial decrease in the number of operations needed for this
disorder with a drop of >90% for elective ulcer surgery over the past
four decades. Refractory ulcers are an exceedingly rare occurrence.
Surgery is more often required for treatment of an ulcer-related
complication.
Hemorrhage is the most common ulcer-related complication,
occurring in ~15–25% of patients. Bleeding may occur in any age
group but is most often seen in older patients (sixth decade or
beyond). The majority of patients stop bleeding spontaneously,
but endoscopic therapy (Chap. 333) is necessary in some. Paren­
terally and orally administered PPIs also decrease ulcer rebleed­
ing in patients who have undergone endoscopic therapy. Patients
unresponsive or refractory to endoscopic intervention will require
angiographic intervention or surgery (~5% of transfusion-requiring
patients).
Free peritoneal perforation occurs in ~2–3% of DU patients,
with NSAID-induced GU perforations occurring more commonly.
Sudden onset of severe abdominal pain with peritoneal signs and
evidence of pneumoperitoneum on abdominal imaging is the clas­
sic presentation of a perforated viscous, but this presentation occurs
in only two-thirds of patients. The latter is especially true in elderly
patients (>70 years old), obese individuals, and immunocompro­
mised patients. It is important to keep in mind that, as in the case
of bleeding, up to 10% of these patients will not have antecedent
ulcer symptoms. Delay in diagnosis clearly leads to higher mortal­
ity; thus, early suspicion and intervention with nasogastric suction,
intravenous PPI, antibiotics, and surgical consultation are essential.
Concomitant bleeding may occur in up to 10% of patients with
perforation, with mortality being increased substantially. Peptic
ulcer can also penetrate into adjacent organs, especially with a
posterior DU, which can penetrate into the pancreas, colon, liver,
or biliary tree.
PART 10
Disorders of the Gastrointestinal System
Pyloric channel ulcers or DUs can lead to gastric outlet obstruc­
tion in ~2–3% of patients. This can result from chronic scarring
or from impaired motility due to inflammation and/or edema
with pylorospasm. Patients may present with early satiety, nausea,
vomiting of undigested food, and weight loss. Conservative man­
agement with nasogastric suction, intravenous hydration/nutrition,
and antisecretory agents is indicated for 7–10 days with the hope
that a functional obstruction will reverse. If a mechanical obstruc­
tion persists, endoscopic intervention with balloon dilation may be
effective. Surgery should be considered if all else fails.
Specific Operations for Duodenal Ulcers  Surgical treatment was
originally designed to decrease gastric acid secretion. Operations
most commonly performed include (1) vagotomy and drainage
(by pyloroplasty, gastroduodenostomy, or gastrojejunostomy), (2)
highly selective vagotomy (which does not require a drainage pro­
cedure), and (3) vagotomy with antrectomy. The specific procedure
performed is dictated by the underlying circumstances: elective
versus emergency, the degree and extent of duodenal ulceration,
the etiology of the ulcer (H. pylori, NSAIDs, malignancy), and the
expertise of the surgeon. Moreover, the trend has been toward a
dramatic decrease in the need for surgery for treatment of refrac­
tory PUD, and when needed, minimally invasive and anatomypreserving operations are preferred.
Vagotomy is a component of each of these procedures and is
aimed at decreasing acid secretion through ablating cholinergic
input to the stomach. Unfortunately, both truncal and selective
vagotomy (preserves the celiac and hepatic branches) result in gastric
atony despite successful reduction of both basal acid output (BAO;
decreased by 85%) and maximal acid output (MAO; decreased
by 50%). Drainage through pyloroplasty or gastroduodenostomy
is required in an effort to compensate for the vagotomy-induced
gastric motility disorder. This procedure has an intermediate com­
plication rate and a 10% ulcer recurrence rate. To minimize gastric
dysmotility, highly selective vagotomy (also known as parietal cell,
super-selective, or proximal vagotomy) was developed. Only the
vagal fibers innervating the portion of the stomach that contains
parietal cells are transected, thus leaving fibers important for regu­
lating gastric motility intact. Although this procedure leads to an
immediate decrease in both BAO and stimulated acid output, acid
secretion recovers over time. By the end of the first postoperative
year, basal and stimulated acid output are ~30 and 50%, respec­
tively, of preoperative levels. Ulcer recurrence rates are higher with
highly selective vagotomy (≥10%), although the overall complica­
tion rates are the lowest of the three procedures.
The procedure that provides the lowest rates of ulcer recurrence
(1%) but has the highest complication rate is vagotomy (trun­
cal or selective) in combination with antrectomy. Antrectomy is
aimed at eliminating an additional stimulant of gastric acid secre­
tion, gastrin. Two principal types of reanastomoses are used after
antrectomy: gastroduodenostomy (Billroth I) or gastrojejunostomy
(Billroth II) (Fig. 335-15). Although Billroth I is often preferred
over II, severe duodenal inflammation or scarring may preclude its
performance. Prospective, randomized studies confirm that partial
gastrectomy followed by Roux-en-Y reconstruction leads to a sig­
nificantly better clinical, endoscopic, and histologic outcome than
Billroth II reconstruction.
Of these procedures, highly selective vagotomy may be the pro­
cedure of choice in the elective setting, except in situations where
ulcer recurrence rates are high (prepyloric ulcers and those refrac­
tory to medical therapy). Selection of vagotomy and antrectomy
may be more appropriate in these circumstances.
These procedures have been traditionally performed by standard
laparotomy. The advent of laparoscopic surgery has led several
surgical teams to successfully perform highly selective vagotomy,
truncal vagotomy/pyloroplasty, and truncal vagotomy/antrectomy
Antrum
Fundus
Duodenum
Billroth I
Billroth II
FIGURE 335-15  Schematic representation of Billroth I and II procedures.
through this approach. An increase in the number of laparoscopic
procedures for treatment of PUD has occurred. Laparoscopic repair
of perforated peptic ulcers is safe, feasible for the experienced sur­
geon, and associated with decreased postoperative pain, although it
does take longer than an open approach. Moreover, no difference
between the two approaches is noted in postoperative complica­
tions or length of hospital stay.
Specific Operations for GUs  The location and presence of a
concomitant DU dictate the operative procedure performed for
a GU. Antrectomy (including the ulcer) with a Billroth I anasto­
mosis is the treatment of choice for an antral ulcer. Vagotomy is
performed only if a DU is present. Although ulcer excision with
vagotomy and drainage procedure has been proposed, the higher
incidence of ulcer recurrence makes this a less desirable approach.
Ulcers located near the esophagogastric junction may require a
more radical approach, a subtotal gastrectomy with a Roux-en-Y
esophagogastrojejunostomy (Csendes’ procedure). A less aggres­
sive approach, including antrectomy, intraoperative ulcer biopsy,
and vagotomy (Kelling-Madlener procedure), may be indicated in
fragile patients with a high GU. Ulcer recurrence approaches 30%
with this procedure.
Surgery-Related Complications  Complications seen after surgery
for PUD are related primarily to the extent of the anatomic modi­
fication performed. Minimal alteration (highly selective vagotomy)
is associated with higher rates of ulcer recurrence and less GI
disturbance. More aggressive surgical procedures have a lower
rate of ulcer recurrence but a greater incidence of GI dysfunction.
Overall, morbidity and mortality related to these procedures are
quite low. Morbidity associated with vagotomy and antrectomy or
pyloroplasty is ≤5%, with mortality ~1%. Highly selective vagotomy
has lower morbidity and mortality rates of 1 and 0.3%, respectively.
In addition to the potential early consequences of any intraab­
dominal procedure (bleeding, infection, thromboembolism), gas­
troparesis, duodenal stump leak, and efferent loop obstruction can
be observed.
Recurrent Ulceration  The risk of ulcer recurrence is directly
related to the procedure performed. Ulcers that recur after partial
gastric resection tend to develop at the anastomosis (stomal or
marginal ulcer). Epigastric abdominal pain is the most frequent
presenting complaint (>90%). Severity and duration of pain tend to
be more progressive than observed with DUs before surgery.
Ulcers may recur for several reasons, including incomplete
vagotomy, inadequate drainage, retained antrum, and, less likely,
persistent or recurrent H. pylori infection. ZES should have been
excluded preoperatively. Surreptitious use of NSAIDs is an impor­
tant reason for recurrent ulcers after surgery, especially if the initial
procedure was done for an NSAID-induced ulcer. Once H. pylori
and NSAIDs have been excluded as etiologic factors, the ques­
tion of incomplete vagotomy or retained gastric antrum should
be explored. For the latter, fasting plasma gastrin levels should be
determined. If elevated, retained antrum or ZES (see below) should
be considered. Incomplete vagotomy can be ruled out by gastric
acid analysis coupled with sham feeding. In this test, gastric acid
output is measured while the patient sees, smells, and chews a
meal (without swallowing). The cephalic phase of gastric secretion,
which is mediated by the vagus, is being assessed with this study.
An increase in gastric acid output in response to sham feeding is
evidence that the vagus nerve is intact. A rise in serum pancreatic
polypeptide >50% within 30 min of sham feeding is also suggestive
of an intact vagus nerve.
Medical therapy with H2 blockers will heal postoperative ulcer­
ation in 70–90% of patients. The efficacy of PPIs has not been fully
assessed in this group, but one may anticipate greater rates of ulcer
healing compared to those obtained with H2 blockers. Repeat oper­
ation (complete vagotomy, partial gastrectomy) may be required in
a small subgroup of patients who have not responded to aggressive
medical management.
Afferent Loop Syndromes  Although rarely seen today as a result
of the decrease in the performance of Billroth II anastomosis, two
types of afferent loop syndrome can occur in patients who have
undergone this type of partial gastric resection. The more common
of the two is bacterial overgrowth in the afferent limb secondary
to stasis. Patients may experience postprandial abdominal pain,
bloating, and diarrhea with concomitant malabsorption of fats and
vitamin B12. Cases refractory to antibiotics may require surgical
revision of the loop. The less common afferent loop syndrome can
present with severe abdominal pain and bloating that occur 20–60 min
after meals. Pain is often followed by nausea and vomiting of bilecontaining material. The pain and bloating may improve after eme­
sis. The cause of this clinical picture is theorized to be incomplete
drainage of bile and pancreatic secretions from an afferent loop that
is partially obstructed. Cases refractory to dietary measures may
need surgical revision or conversion of the Billroth II anastomosis
to a Roux-en-Y gastrojejunostomy.
Dumping Syndrome  Dumping syndrome consists of a series of
vasomotor and GI signs and symptoms and occurs in patients who
have undergone vagotomy and drainage (especially Billroth pro­
cedures). Two phases of dumping, early and late, can occur. Early
dumping takes place 15–30 min after meals and consists of crampy
abdominal discomfort, nausea, diarrhea, belching, tachycardia, pal­
pitations, diaphoresis, light-headedness, and, rarely, syncope. These
signs and symptoms arise from the rapid emptying of hyperosmolar
gastric contents into the small intestine, resulting in a fluid shift into
the gut lumen with plasma volume contraction and acute intestinal
distention. Release of vasoactive GI hormones (vasoactive intestinal
polypeptide, neurotensin, motilin) is also theorized to play a role in
early dumping.
CHAPTER 335
The late phase of dumping typically occurs 90 min to 3 h after
meals. Vasomotor symptoms (light-headedness, diaphoresis, palpi­
tations, tachycardia, and syncope) predominate during this phase.
This component of dumping is thought to be secondary to hypogly­
cemia from excessive insulin release.
Peptic Ulcer Disease and Related Disorders
Dumping syndrome is most noticeable after meals rich in simple
carbohydrates (especially sucrose) and high osmolarity. Ingestion
of large amounts of fluids may also contribute. After vagotomy and
drainage, up to 50% of patients will experience dumping syndrome
to some degree early on. Signs and symptoms often improve with
time, but a severe protracted picture can occur in up to 1% of
patients.
Dietary modification is the cornerstone of therapy for patients
with dumping syndrome. Small, multiple (six) meals devoid of
simple carbohydrates coupled with elimination of liquids during
meals is important. Antidiarrheals and anticholinergic agents are
complementary to diet. Guar and pectin, which increase the viscos­
ity of intraluminal contents, may be beneficial in more symptomatic
individuals. Acarbose, an α-glucosidase inhibitor that delays diges­
tion of ingested carbohydrates, has also been shown to be beneficial
in the treatment of the late phases of dumping. The somatostatin
analogue octreotide has been successful in diet-refractory cases.
This drug is administered subcutaneously (50 μg tid), titrated
according to clinical response. A long-acting depot formulation of
octreotide can be administered once every 28 days and provides
symptom relief comparable to the short-acting agent. In addition,
patient weight gain and quality of life appear to be superior with the
long-acting form.
Postvagotomy Diarrhea  Up to 10% of patients may seek medical
attention for the treatment of postvagotomy diarrhea. This compli­
cation is most commonly observed after truncal vagotomy, which
is rarely performed today. Patients may complain of intermittent
diarrhea that occurs typically 1–2 h after meals. Occasionally, the
symptoms may be severe and relentless. This is due to a motility
disorder from interruption of the vagal fibers supplying the luminal
gut. Other contributing factors may include decreased absorption
of nutrients (see below), increased excretion of bile acids, and
release of luminal factors that promote secretion. Diphenoxylate or
loperamide is often useful in symptom control. The bile salt–binding
agent cholestyramine may be helpful in severe cases. Surgical rever­
sal of a 10-cm segment of jejunum may yield a substantial improve­
ment in bowel frequency in a subset of patients.
Bile Reflux Gastropathy  A subset of post–partial gastrectomy
patients who present with abdominal pain, early satiety, nausea,
and vomiting will have mucosal erythema of the gastric remnant
as the only finding. Histologic examination of the gastric mucosa
reveals minimal inflammation but the presence of epithelial cell
injury. This clinical picture is categorized as bile or alkaline reflux
gastropathy/gastritis. Although reflux of bile is implicated as the
reason for this disorder, the mechanism is unknown. Prokinetic
agents, cholestyramine, and sucralfate have been somewhat effec­
tive treatments. Severe refractory symptoms may require using
either nuclear scanning with 99mTc-HIDA to document reflux.
Surgical diversion of pancreaticobiliary secretions away from the
gastric remnant with a Roux-en-Y gastrojejunostomy consisting of
a long (50–60 cm) Roux limb has been used in severe cases. Bilious
vomiting improves, but early satiety and bloating may persist in up
to 50% of patients.
Maldigestion and Malabsorption  Weight loss can be observed
in up to 60% of patients after partial gastric resection. Patients can
experience a 10% loss of body weight, which stabilizes 3 months post­
operatively. A significant component of this weight reduction is due
to decreased oral intake. However, mild steatorrhea can also develop.
Reasons for maldigestion/malabsorption include decreased gastric
acid production, rapid gastric emptying, decreased food dispersion in
the stomach, reduced luminal bile concentration, reduced pancreatic
secretory response to feeding, and rapid intestinal transit.
PART 10
Disorders of the Gastrointestinal System
Decreased serum vitamin B12 levels can be observed after partial
gastrectomy. This is usually not due to deficiency of IF, since a
minimal amount of parietal cells (source of IF) is removed during
antrectomy. Reduced vitamin B12 may be due to competition for
the vitamin by bacterial overgrowth or inability to split the vitamin
from its protein-bound source due to hypochlorhydria.
Iron-deficiency anemia may be a consequence of impaired
absorption of dietary iron in patients with a Billroth II gastroje­
junostomy. Absorption of iron salts is normal in these individuals;
thus, a favorable response to oral iron supplementation can be
anticipated. Folate deficiency with concomitant anemia can also
develop in these patients. This deficiency may be secondary to
decreased absorption or diminished oral intake.
Malabsorption of vitamin D and calcium resulting in osteopo­
rosis and osteomalacia is common after partial gastrectomy and
gastrojejunostomy (Billroth II). Osteomalacia can occur as a late
complication in up to 25% of post–partial gastrectomy patients.
Bone fractures occur twice as commonly in men after gastric
surgery as in a control population. It may take years before x-ray
findings demonstrate diminished bone density. Elevated alkaline
phosphatase, reduced serum calcium, bone pain, and pathologic
fractures may be seen in patients with osteomalacia. The high
incidence of these abnormalities in this subgroup of patients justi­
fies treating them with vitamin D and calcium supplementation
indefinitely. Therapy is especially important in females. Copper
deficiency has also been reported in patients undergoing surgeries
that bypass the duodenum, where copper is primarily absorbed.
Patients may present with a rare syndrome that includes ataxia,
myelopathy, and peripheral neuropathy.
Gastric Adenocarcinoma  The incidence of adenocarcinoma in
the gastric stump is increased 15 years after resection. Some have
reported a four- to fivefold increase in gastric cancer 20–25 years
after resection. The pathogenesis is unclear but may involve alka­
line reflux, bacterial proliferation, or hypochlorhydria. The role
of endoscopic screening is not clear, and most guidelines do not
support its use.
Additional Complications  Reflux esophagitis and a higher inci­
dence of gallstones and cholecystitis have been reported in patients
undergoing subtotal gastrectomy. The latter is thought to be due to
decreased gallbladder contractility associated with vagotomy and
bypass of the duodenum, leading to decreased postprandial release
of cholecystokinin.
RELATED CONDITIONS
■
■ZOLLINGER-ELLISON SYNDROME
Severe peptic ulcer diathesis secondary to gastric acid hypersecretion due
to unregulated gastrin release from a non–β-cell, often well-differentiated
neuroendocrine tumor (NET; gastrinoma) defines the components of
ZES. Initially, ZES was typified by aggressive and refractory ulceration
in which total gastrectomy provided the only chance for enhancing
survival. Today, it can be cured by surgical resection in up to 40% of
patients with the sporadic form of the disease (see below).
Epidemiology 
The true incidence of ZES is unknown, but esti­
mates suggest that it varies from 0.1 to 1% of individuals presenting
with PUD, with 0.1–3 individuals per year having this rare diagnosis.
Others have estimated an incidence of 0.5–3 per million population.
Females are slightly more commonly affected than males, and the
majority of patients are diagnosed between ages 30 and 65. In addition,
the time of diagnosis is estimated to be between 4 and 7 years after the
onset of symptoms. Gastrinomas are classified into sporadic tumors
(80%) and those associated with multiple endocrine neoplasia (MEN)
type 1 (see below). The widespread availability and use of PPIs have
led to a decreased patient referral for gastrinoma evaluation, delay in
diagnosis, and an increase in false-positive diagnoses of ZES. In fact,
diagnosis may be delayed for ≥6 years after symptoms consistent with
ZES are displayed.
Pathophysiology 
Hypergastrinemia originating from an autono­
mous neoplasm is the driving force responsible for the clinical mani­
festations in ZES. Gastrin stimulates acid secretion through gastrin
receptors on parietal cells and by inducing histamine release from ECL
cells. Gastrin also has a trophic action on gastric epithelial cells. Longstanding hypergastrinemia leads to markedly increased gastric acid
secretion through both parietal cell stimulation and increased parietal
cell mass. The increased gastric acid output leads to peptic ulcer dia­
thesis, erosive esophagitis, and diarrhea.
Tumor Distribution 
Although early studies suggested that the
vast majority of gastrinomas occurred within the pancreas, a signifi­
cant number of these lesions are extrapancreatic. Between 60 and 90%
of these tumors are found within the hypothetical gastrinoma triangle
(confluence of the cystic and common bile ducts superiorly, junction
of the second and third portions of the duodenum inferiorly, and junc­
tion of the neck and body of the pancreas medially). Duodenal tumors
constitute the most common nonpancreatic lesion; between 60 and
100% of gastrinomas are found here. Duodenal tumors are smaller,
slower growing, and less likely to metastasize than pancreatic lesions
and occur primarily in the first and second portion of the duodenum
(90%). Less common extrapancreatic sites include stomach, bones,
ovaries, heart, liver, and lymph nodes. More than 60% of tumors are
considered malignant (determined by local invasion and/or evidence
of metastasis), with up to 30–50% of patients having multiple lesions
or metastatic disease at presentation. Histologically, gastrin-producing
cells appear well-differentiated (grade 1 or 2 histologically), expressing
markers typically found in endocrine neoplasms (chromogranin, neuronspecific enolase). Although not clearly established in gastrinomas,
histologic grade in pancreatic NETs generally is an important predictor
of survival in these rare neoplasms (Chap. 89).
Clinical Manifestations 
Gastric acid hypersecretion is respon­
sible for the signs and symptoms observed in patients with ZES.
The most common clinical presentation for gastrinoma patients is
abdominal pain in the presence of acid peptic disorders. Peptic ulcer
is the most common clinical manifestation, occurring in >90% of
gastrinoma patients. Initial presentation and ulcer location (duodenal
bulb) may be indistinguishable from common PUD. Clinical situations
that should create suspicion of gastrinoma are ulcers in unusual loca­
tions (second part of the duodenum and beyond), ulcers refractory to
standard medical therapy, ulcer recurrence after acid-reducing surgery,
ulcers presenting with frank complications (bleeding, obstruction, and
perforation), or ulcers in the absence of H. pylori or NSAID ingestion.
Symptoms of esophageal origin are present in up to two-thirds of
patients with ZES, with a spectrum ranging from mild esophagitis to
frank ulceration with stricture and Barrett’s mucosa.
Diarrhea, the next most common clinical manifestation, is found in
up to 70% of patients. Although diarrhea often occurs concomitantly
with acid peptic disease, it may also occur independent of an ulcer and
classically will abate with PPI therapy. Etiology of the diarrhea is mul­
tifactorial, resulting from marked volume overload to the small bowel,
pancreatic enzyme inactivation by acid, and damage of the intestinal
epithelial surface by acid. The epithelial damage can lead to a mild
degree of maldigestion and malabsorption of nutrients. The diarrhea
may also have a secretory component due to the direct stimulatory
effect of gastrin on enterocytes or the co-secretion of additional hor­
mones from the tumor such as vasoactive intestinal peptide.
Gastrinomas can develop in the presence of MEN 1 syndrome
(Chaps. 89 and 400) in ~25% of patients. This autosomal dominant
disorder involves primarily three organ sites: the parathyroid glands
(80–90%), pancreas (40–80%), and pituitary gland (30–60%). The
syndrome is caused by inactivating mutations of the MEN1 tumorsuppressor gene found on the long arm of chromosome 11q13. The
gene encodes for menin, which has an important role in DNA replica­
tion and transcriptional regulation. A genetic diagnosis is obtained by
sequencing of the MEN1 gene, which can reveal mutations in 70–90%
of typical MEN 1 cases. A family may have an unknown mutation,
making a genetic diagnosis impossible, and therefore, certain individu­
als will require a clinical diagnosis, which is determined by whether a
patient has tumors in two of the three endocrine organs (parathyroid,
pancreas/duodenum, or pituitary) or has a family history of MEN 1
and one of the endocrine organ tumors. In view of the stimulatory
effect of calcium on gastric secretion, the hyperparathyroidism and
hypercalcemia seen in MEN 1 patients may have a direct effect on ulcer
disease. Resolution of hypercalcemia by parathyroidectomy reduces
gastrin and gastric acid output in gastrinoma patients. An additional
distinguishing feature in ZES patients with MEN 1 is the higher inci­
dence of gastric carcinoid tumor development (as compared to patients
with sporadic gastrinomas). ZES presents and is diagnosed earlier in
MEN 1 patients, and they have a more indolent course as compared
to patients with sporadic gastrinoma. Gastrinomas tend to be smaller,
multiple, and located in the duodenal wall more often than is seen in
patients with sporadic ZES. Establishing the diagnosis of MEN 1 is
critical in order to provide genetic counseling to the patient and their
family and also to determine the recommended surgical approach.
Therefore, gastrinoma patients should be screened for MEN 1 by per­
forming a detailed family history and obtaining several serum markers
including calcium, parathyroid, prolactin, and pancreatic polypeptide
levels.
Diagnosis 
Establishing an early diagnosis is important in order to
minimize the long-term sequelae of gastric acid hypersecretion, pre­
vent metastatic disease, and counsel family members if a diagnosis of
MEN 1 is established. Biochemical measurements of gastrin and acid
secretion in patients suspected of having ZES play an important role
is establishing this rare diagnosis. Often, patients suspected of having
ZES will be treated with a PPI in an effort to ameliorate symptoms
and decrease the likelihood of possible acid-related complications.
The presence of the PPI, which will lower acid secretion and poten­
tially elevate fasting gastrin levels in normal individuals, will make
the diagnostic approach in these individuals somewhat difficult.
Significant morbidity related to peptic diathesis has been described
when stopping PPIs in gastrinoma patients; therefore, a systematic
approach in stopping these agents is warranted (see below). The first
step in the evaluation of a patient suspected of having ZES is to obtain
a fasting gastrin level. A list of clinical scenarios that should arouse
TABLE 335-8  When to Obtain a Fasting Serum Gastrin Level
Multiple ulcers
Ulcers in unusual locations; associated with severe esophagitis; resistant
to therapy with frequent recurrences; in the absence of nonsteroidal antiinflammatory drug ingestion or Helicobacter pylori infection
Ulcer patients awaiting surgery
Severe or refractory GERD
GERD associated with diarrhea
Extensive family history for peptic ulcer disease
Postoperative ulcer recurrence
Basal hyperchlorhydria
Unexplained diarrhea or steatorrhea
Diarrhea improved with PPI
Hypercalcemia
Family history of pancreatic islet, pituitary, or parathyroid tumor
Prominent gastric or duodenal folds
suspicion regarding this diagnosis is shown in Table 335-8. Fasting
gastrin levels obtained using a dependable assay are usually <150 pg/
mL. A normal fasting gastrin, on two separate occasions, especially
if the patient is on a PPI, virtually excludes this diagnosis. Virtually
all gastrinoma patients will have a gastrin level >150–200 pg/mL.
Measurement of fasting gastrin should be repeated to confirm the
clinical suspicion. Some of the commercial biochemical assays used
for measuring serum gastrin may be inaccurate. Variable specificity of
the antibodies used have led to both false-positive and false-negative
fasting gastrin levels, placing in jeopardy the ability to make an accu­
rate diagnosis of ZES.
CHAPTER 335
Multiple processes can lead to an elevated fasting gastrin level, the
most frequent of which are gastric hypochlorhydria and achlorhydria,
with or without pernicious anemia. Gastric acid induces feedback
inhibition of gastrin release. A decrease in acid production will subse­
quently lead to failure of the feedback inhibitory pathway, resulting in
net hypergastrinemia. Gastrin levels will thus be high in patients using
antisecretory agents for the treatment of acid peptic disorders and
dyspepsia. H. pylori infection can also cause hypergastrinemia. Addi­
tional causes of elevated gastrin include retained gastric antrum; G-cell
hyperplasia; gastric outlet obstruction; renal insufficiency; massive
small-bowel obstruction; and conditions such as rheumatoid arthritis,
vitiligo, diabetes mellitus, and pheochromocytoma. Although a fasting
gastrin >10 times normal is highly suggestive of ZES, two-thirds of
patients will have fasting gastrin levels that overlap with levels found
in the more common disorders outlined above, especially if a PPI is
being taken by the patient. The effect of the PPI on gastrin levels and
acid secretion will linger several days after stopping the PPI; therefore,
it should be stopped for a minimum of 7 days before testing. During
this period, the patient should be placed on a histamine H2 antagonist,
such as famotidine, twice to three times per day. Although this type of
agent has a short-term effect on gastrin and acid secretion, it needs to
be stopped 24 h before repeating fasting gastrin levels or performing
some of the tests highlighted below. The patient may take antacids for
the final day, stopping them ~12 h before testing is performed. Height­
ened awareness of complications related to gastric acid hypersecretion
during the period of PPI cessation is critical.
Peptic Ulcer Disease and Related Disorders
Historically the next study typically recommended for establish­
ing a biochemical diagnosis of gastrinoma is to assess acid secretion.
Unfortunately, few centers perform this type of testing; thus, only
a brief summary will be provided here. Typically, nothing further
needs to be done if decreased acid output in the absence of a PPI is
observed. A pH can be measured on gastric fluid obtained either dur­
ing endoscopy or through nasogastric aspiration; a pH <3 is suggestive
of a gastrinoma, but a pH >3 is not helpful in excluding the diagnosis.
In those situations where the pH is >3, formal gastric acid analysis
should be performed if available. Normal BAO in nongastric surgery
patients is typically <5 meq/h. A BAO >15 meq/h in the presence of
hypergastrinemia is considered pathognomonic of ZES, but up to 12%
of patients with common PUD may have elevated BAO to a lesser
degree that can overlap with levels seen in ZES patients. In an effort
to improve the sensitivity and specificity of gastric secretory studies, a
BAO/MAO ratio was established using pentagastrin infusion as a way
to maximally stimulate acid production, with a BAO/MAO ratio >0.6
being highly suggestive of ZES. Pentagastrin is no longer available in
the United States, making measurement of MAO virtually impossible.
An endoscopic method for measuring gastric acid output has been
developed but requires further validation.
Gastrin provocative tests have been developed in an effort to dif­
ferentiate between the causes of hypergastrinemia and are especially
helpful in patients with indeterminate acid secretory studies. The tests
are the secretin stimulation test and the calcium infusion study; the
latter is rarely, if ever, utilized in our current environment due to the
cumbersome nature of the test and its lower sensitivity and specificity
than secretin stimulation. The most sensitive and specific gastrin pro­
vocative test for the diagnosis of gastrinoma is the secretin study. An
increase in gastrin of ≥120 pg within 15 min of secretin injection has a
sensitivity and specificity of >90% for ZES. PPI-induced hypochlorhy­
dria or achlorhydria may lead to a false-positive secretin test; thus, this
agent must be stopped for 1 week before testing.
In light of the limited availability of the biochemical studies outlined
above, more studies make a diagnosis of gastrinoma based on the pres­
ence of elevated gastrin and low gastric pH in the right clinical setting
coupled with tumor localization tests outlined below and positive
histology by biopsy (difficult to obtain). Revised guidelines for the best
approach to establishing a diagnosis of gastrinoma taking into consid­
eration the above outlined limitations are being considered, but none
have replaced the established guidelines outlined earlier in this section
(see Jensen et al in Further Reading).
PART 10
Disorders of the Gastrointestinal System
Tumor Localization 
Once the biochemical diagnosis of gastri­
noma has been confirmed (if possible), the tumor must be located.
Multiple imaging studies have been used in an effort to enhance
tumor localization (Table 335-9). The broad range of sensitivity is
due to the variable success rates achieved by the different investiga­
tive groups. Endoscopic ultrasound (EUS) permits imaging of the
pancreas with a high degree of resolution (<5 mm). This modality is
particularly helpful in excluding small neoplasms within the pancreas
and in assessing the presence of surrounding lymph nodes and vascu­
lar involvement, but it is not very sensitive (43%) for finding duode­
nal lesions. This latter observation has led some to not include EUS in
the routine preoperative evaluation of a patient suspected of having
a gastrinoma. Several types of endocrine tumors express cell-surface
receptors for somatostatin, in particular the subtype 2 (SSTR2).
This permits the localization, staging, and prediction of therapeutic
response to somatostatin analogues (see below) by gastrinomas.
The original functional scinitigraphic tool developed measuring
TABLE 335-9  Sensitivity of Imaging Studies in Zollinger-Ellison
Syndrome
 
SENSITIVITY, %
PRIMARY
GASTRINOMA
HEPATIC METASTATIC
GASTRINOMA
STUDY
Ultrasound
0–28
15–77
CT scan
0–59
99–100
Selective angiography
35–68
96–100
Portal venous sampling
70–90
N/A
SASI
55–78
N/A
MRI
20–25
88–100
OctreoScan
55–77
90–100
EUS
28–86
N/A
Abbreviations: CT, computed tomography; EUS, endoscopic ultrasonography;
MRI, magnetic resonance imaging; N/A, not applicable; OctreoScan, imaging with
111In-pentetreotide; SASI, selective arterial secretin injection.
the uptake of the stable somatostatin analogue 111In-pentetreotide
(OctreoScan) has demonstrated sensitivity and specificity rates of
80%. Positron emission tomography (PET)–computed tomography
(CT) with 68Ga-DOTATATE has been developed and is superior
than OctreoScan for assessing tumor presence in patients with
well-differentiated NETs such as gastrinomas, with sensitivity and
specificity of >90%, making it the functional imaging study of choice
when available. 18F-Fluordeoxyglucose (18F-FDG) PET imaging has
been found to be useful in pancreatic NETs, including gastrinomas,
particularly as a prognostic marker.
Up to 50% of patients have metastatic disease at diagnosis. Success
in controlling gastric acid hypersecretion has shifted the emphasis of
therapy toward providing a surgical cure. Detecting the primary tumor
and excluding metastatic disease are critical in view of this paradigm
shift. Once a biochemical diagnosis has been confirmed, the patient
should first undergo an abdominal CT scan, magnetic resonance imag­
ing (MRI), or OctreoScan/PET-CT with 68Ga-DOTATATE (depending
on availability) to exclude metastatic disease. Once metastatic disease
has been excluded, an experienced endocrine surgeon may opt for
exploratory laparotomy with intraoperative ultrasound or transillumi­
nation. In other centers, careful examination of the peripancreatic area
with EUS, accompanied by endoscopic exploration of the duodenum
for primary tumors, will be performed before surgery. Selective arterial
secretin injection may be a useful adjuvant for localizing tumors in a
subset of patients. The extent of the diagnostic and surgical approach
must be carefully balanced with the patient’s overall physiologic condi­
tion and the natural history of a slow-growing gastrinoma.
TREATMENT
Zollinger-Ellison Syndrome
Treatment of functional endocrine tumors is directed at ameliorat­
ing the signs and symptoms related to hormone overproduction,
curative resection of the neoplasm, and attempts to control tumor
growth in metastatic disease.
PPIs are the treatment of choice and have decreased the need
for total gastrectomy. Initial PPI doses tend to be higher than those
used for treatment of GERD or PUD. The initial dose of omeprazole,
lansoprazole, rabeprazole, or esomeprazole should be in the range of
60 mg in divided doses in a 24-h period. When gastric acid analysis
was more widely available, dosing was adjusted to achieve a BAO
<10 meq/h (at the drug trough) in surgery-naive patients and to
<5 meq/h in individuals who have previously undergone an acidreducing operation. Close monitoring of clinical symptoms when
starting PPIs and increasing the dose accordingly are paramount.
Although the somatostatin analogue has inhibitory effects on gastrin
release from receptor-bearing tumors and inhibits gastric acid secre­
tion to some extent, PPIs have the advantage of reducing parietal
cell activity to a greater degree. Despite this, octreotide or lanreotide
may be considered as adjunctive therapy to the PPI in patients with
tumors that express somatostatin receptors and have peptic symp­
toms that are difficult to control with high-dose PPI.
The ultimate goal of surgery would be to provide a definitive
cure. Improved understanding of tumor distribution has led to
immediate cure rates as high as 33% with 10-year disease-free inter­
vals as high as 95% in sporadic gastrinoma patients undergoing
surgery. A positive outcome is highly dependent on the experience
of the surgical team treating these rare tumors. Surgical therapy of
gastrinoma patients with MEN 1 remains controversial because of
the difficulty in rendering these patients disease-free with surgery.
In contrast to the encouraging postoperative results observed in
patients with sporadic disease, <5% of MEN 1 patients are diseasefree 5 years after an operation. Moreover, in contrast to patients
with sporadic ZES, the clinical course of MEN 1 patients tends to be
benign and rarely leads to disease-related mortality, recommending
that early surgery be deferred. Some groups suggest surgery only if
a clearly identifiable, nonmetastatic lesion is documented by struc­
tural studies. Others advocate a more aggressive approach, where
all patients free of hepatic metastasis are explored and all detected
tumors in the duodenum are resected; this is followed by enucle­
ation of lesions in the pancreatic head, with a distal pancreatectomy
to follow. The outcome of the two approaches has not been clearly
defined. Laparoscopic surgical interventions may provide attractive
approaches in the future but currently seem to be of some limited
benefit in patients with gastrinoma because a significant percentage
of the tumors may be extrapancreatic and difficult to localize with a
laparoscopic approach. Finally, patients selected for surgery should
be individuals whose health status would lead them to tolerate a
more aggressive operation and obtain the long-term benefits from
such aggressive surgery, which are often witnessed after 10 years.
Therapy of metastatic endocrine tumors in general remains sub­
optimal; gastrinomas are no exception. In light of the observation
that in many instances tumor growth is indolent and that many
individuals with metastatic disease remain relatively stable for
significant periods of time, many advocate not instituting systemic
tumor-targeted therapy until evidence of tumor progression or
refractory symptoms not controlled with PPIs are noted. Medical
approaches, including biologic therapy (IFN-α, long-acting soma­
tostatin analogues, and peptide receptor radionuclides), systemic
chemotherapy (streptozotocin, 5-fluorouracil, and doxorubicin),
and hepatic artery embolization, may lead to significant toxicity
without a substantial improvement in overall survival. Use of temo­
zolomide with capecitabine has demonstrated radiographic regres­
sion and progression-free survival in patients with well-differentiated
NETs in the range of 70% and 18 months, respectively. Systemic
therapy with radiolabeled somatostatin analogues (peptide receptor
radiotherapy [PRRT]) has been used in the therapy of metastatic
NETs and appears to be very promising in terms of radiographic
regression, symptoms, and progression-free survival, but addi­
tional studies are warranted. Several promising therapies are being
explored, including radiofrequency ablation or cryoablation of liver
lesions and use of agents that block the VEGF receptor pathway
(sunitinib, surufatinib), the mammalian target of rapamycin, and
immune checkpoint inhibitors (Chap. 89).
Surgical approaches, including debulking surgery and liver
transplantation for hepatic metastasis, have also produced limited
benefit.
The overall 5- and 10-year survival rates for gastrinoma patients
are 62–75% and 47–53%, respectively. Individuals with the entire
tumor resected or those with a negative laparotomy have 5- and
10-year survival rates >90%. Patients with incompletely resected
tumors have 5- and 10-year survival rates of 43 and 25%, respec­
tively. Patients with hepatic metastasis have <20% survival at 5 years.
Favorable prognostic indicators include primary duodenal wall
tumors, isolated lymph node tumor, the presence of MEN 1, and
undetectable tumor upon surgical exploration. Poor outcome is
seen in patients with shorter disease duration; female sex; older age
at diagnosis; higher gastrin levels (>10,000 pg/mL); poor histologic
differentiation; high proliferative index; certain tumor molecular
changes including chromosome 1qLOH and chromosome XLOH;
large pancreatic primary tumors (>2–3 cm); metastatic disease
to lymph nodes, liver, and bone; and Cushing’s syndrome. Rapid
growth of hepatic metastases is also predictive of poor outcome.
■
■STRESS-RELATED MUCOSAL INJURY
Patients suffering from shock, sepsis, massive burns, severe trauma, or
head injury can develop acute erosive gastric mucosal changes or frank
ulceration with bleeding. Classified as stress-induced gastritis or ulcers,
injury is most commonly observed in the acid-producing (fundus
and body) portions of the stomach. The most common presentation
is GI bleeding, which is usually minimal but can occasionally be lifethreatening. Respiratory failure requiring mechanical ventilation and
underlying coagulopathy are risk factors for bleeding, which tends to
occur 48–72 h after the acute injury or insult.
Histologically, stress injury does not contain inflammation or H.
pylori; thus, “gastritis” is a misnomer. Although elevated gastric acid
secretion may be noted in patients with stress ulceration after head
trauma (Cushing’s ulcer) and severe burns (Curling’s ulcer), mucosal
ischemia, breakdown of the normal protective barriers of the stomach,
systemic release of cytokines, poor GI motility, and oxidative stress
also play an important role in the pathogenesis. Acid must contribute
to injury in view of the significant drop in bleeding noted when acid
inhibitors are used as prophylaxis for stress gastritis.
Improvement in the general management of intensive care unit
patients has led to a significant decrease in the incidence of GI bleed­
ing due to stress ulceration. The estimated decrease in bleeding is from
20–30% to <5%. This improvement has led to some debate regarding
the need for prophylactic therapy. The high mortality associated with
stress-induced clinically important GI bleeding (>40%) and the limited
benefit of medical (endoscopic, angiographic) and surgical therapy in a
patient with hemodynamically compromising bleeding associated with
stress ulcer/gastritis support the use of preventive measures in high-risk
patients (mechanically ventilated, coagulopathy, multiorgan failure, or
severe burns). Meta-analysis comparing H2 blockers with PPIs for the
prevention of stress-associated clinically important and overt GI bleed­
ing demonstrates superiority of the latter without increasing the risk
of nosocomial infections, increasing mortality, or prolonging intensive
care unit length of stay. Therefore, PPIs are the treatment of choice for
stress prophylaxis. Oral PPI is the best option if the patient can toler­
ate enteral administration. Pantoprazole is available as an intravenous
formulation for individuals in whom enteral administration is not
possible. If bleeding occurs despite these measures, endoscopy, intra­
arterial vasopressin, and embolization are options. If all else fails, then
surgery should be considered. Although vagotomy and antrectomy
may be used, the better approach would be a total gastrectomy, which
has an exceedingly high mortality rate in this setting. Concerns with
the effect of PPIs on the immune system coupled with the high cost of
this agent have led to several comparative studies of PPIs and H2 recep­
tor antagonists for stress prophylaxis in patients requiring mechanical
ventilation. Although the PEPTIC trial demonstrated comparative effi­
cacy between the two agents regarding mortality, technical aspects of
the study led to some limitation in the final interpretation of the results.
A meta-analysis performed in neurocritical patients did not reach
strong clinical recommendations about the utility of ulcer prophylaxis.
Reasons for this conclusion included the overall high or unclear risk of
bias of individual trials, the low event rates, and the modest sample size
examined. Currently, a trial to determine safety and efficacy of panto­
prazole compared to placebo for preventing stress-induced erosions is
ongoing in mechanically ventilated patients.
CHAPTER 335
Peptic Ulcer Disease and Related Disorders
■
■GASTRITIS
The term gastritis should be reserved for histologically documented
inflammation of the gastric mucosa. Gastritis is not the mucosal
erythema seen during endoscopy and is not interchangeable with
“dyspepsia.” The etiologic factors leading to gastritis are broad and het­
erogeneous. Gastritis has been classified based on time course (acute
vs chronic), histologic features, and anatomic distribution or proposed
pathogenic mechanism (Table 335-10).
The correlation between the histologic findings of gastritis, the clini­
cal picture of abdominal pain or dyspepsia, and endoscopic findings
noted on gross inspection of the gastric mucosa is poor. Therefore,
there is no typical clinical manifestation of gastritis.
Acute Gastritis 
The most common causes of acute gastritis are
infectious. Acute infection with H. pylori induces gastritis. However, H.
pylori acute gastritis has not been extensively studied. It is reported as
presenting with sudden onset of epigastric pain, nausea, and vomiting,
and limited mucosal histologic studies demonstrate a marked infiltrate
of neutrophils with edema and hyperemia. If not treated, this picture
will evolve into one of chronic gastritis. Hypochlorhydria lasting for up
to 1 year may follow acute H. pylori infection.
Bacterial infection of the stomach or phlegmonous gastritis is a rare,
potentially life-threatening disorder characterized by marked and dif­
fuse acute inflammatory infiltrates of the entire gastric wall, at times
accompanied by necrosis. Elderly individuals, alcoholics, and AIDS
TABLE 335-10  Classification of Gastritis
I.	 Acute gastritis
A.	 Acute Helicobacter pylori infection
B.	 Other acute infectious gastritides
Bacterial (other than H. pylori)
Helicobacter heilmannii
Phlegmonous
Mycobacterial
Syphilitic
Viral
Parasitic
Fungal
II.	 Chronic atrophic gastritis
A.	 Type A: Autoimmune, body-predominant
B.	 Type B: H. pylori–related, antral-predominant
C.	 Indeterminate
III.	 Uncommon forms of gastritis
A.	 Lymphocytic
B.	 Eosinophilic
C.	 Crohn’s disease
D.	 Sarcoidosis
E.	 Isolated granulomatous gastritis
F.	 Russell body gastritis
patients may be affected. Potential iatrogenic causes include polypec­
tomy and mucosal injection with India ink. Organisms associated with
this entity include streptococci, staphylococci, Escherichia coli, Proteus,
and Haemophilus species. Failure of supportive measures and antibiot­
ics may result in gastrectomy.
PART 10
Disorders of the Gastrointestinal System
Other types of infectious gastritis may occur in immunocompro­
mised individuals such as AIDS patients. Examples include herpetic
(herpes simplex) or CMV gastritis. The histologic finding of intra­
nuclear inclusions would be observed in the latter.
Chronic Gastritis 
Chronic gastritis is identified histologically by
an inflammatory cell infiltrate consisting primarily of lymphocytes
and plasma cells, with very scant neutrophil involvement. Distribution
of the inflammation may be patchy, initially involving superficial and
glandular portions of the gastric mucosa. This picture may progress
to more severe glandular destruction, with atrophy and metaplasia.
Chronic gastritis has been classified according to histologic character­
istics. These include superficial atrophic changes and gastric atrophy.
The association of atrophic gastritis with the development of gastric
cancer has led to the development of endoscopic and serologic markers
of severity. Some of these include gross inspection and classification
of mucosal abnormalities during standard endoscopy, magnification
endoscopy, endoscopy with narrow band imaging and/or autofluores­
cence imaging, and measurement of several serum biomarkers includ­
ing pepsinogen I and II levels, gastrin-17, and anti–H. pylori serologies.
The clinical utility of these tools is currently being explored.
The early phase of chronic gastritis is superficial gastritis. The
inflammatory changes are limited to the lamina propria of the surface
mucosa, with edema and cellular infiltrates separating intact gastric
glands. The next stage is atrophic gastritis. The inflammatory infil­
trate extends deeper into the mucosa, with progressive distortion and
destruction of the glands. The final stage of chronic gastritis is gastric
atrophy. Glandular structures are lost, and there is a paucity of inflam­
matory infiltrates. Endoscopically, the mucosa may be substantially
thin, permitting clear visualization of the underlying blood vessels.
Gastric glands may undergo morphologic transformation in chronic
gastritis. Intestinal metaplasia denotes the conversion of gastric glands
to a small intestinal phenotype with small-bowel mucosal glands con­
taining goblet cells. The metaplastic changes may vary in distribution
from patchy to fairly extensive gastric involvement. Intestinal metapla­
sia is an important predisposing factor for gastric cancer (Chap. 85).
Chronic gastritis has also been classified according to the predomi­
nant site of involvement. Type A refers to the body-predominant form
(autoimmune), and type B is the antral-predominant form (H. pylori–
related). This classification is artificial in view of the difficulty in dis­
tinguishing between these two entities. The term AB gastritis has been
used to refer to a mixed antral/body picture. Histologic classification is
the most accurate way to approach chronic gastritis.
TYPE A GASTRITIS  The less common of the two forms involves pri­
marily the fundus and body, with antral sparing. Traditionally, this
form of gastritis has been associated with pernicious anemia (Chap.
104) in the presence of circulating antibodies against parietal cells and
IF; thus, it is also called autoimmune gastritis. H. pylori infection can
lead to a similar distribution of gastritis.
Antibodies to parietal cells have been detected in >90% of patients
with pernicious anemia and in up to 50% of patients with type A
gastritis. The parietal cell antibody is directed against H+,K+-ATPase.
T cells are also implicated in the injury pattern of this form of gastritis.
A subset of patients infected with H. pylori develop antibodies against
H+,K+-ATPase, potentially leading to the atrophic gastritis pattern
seen in some patients infected with this organism. The mechanism
is thought to involve molecular mimicry between H. pylori LPS and
H+,K+-ATPase.
Parietal cell antibodies and atrophic gastritis are observed in fam­
ily members of patients with pernicious anemia. These antibodies
are observed in up to 20% of individuals aged >60 and in ~20% of
patients with vitiligo and Addison’s disease. About one-half of patients
with pernicious anemia have antibodies to thyroid antigens, and
~30% of patients with thyroid disease have circulating anti–parietal
cell antibodies. Anti-IF antibodies are more specific than parietal cell
antibodies for type A gastritis, being present in ~40% of patients with
pernicious anemia. Another parameter consistent with this form of
gastritis being autoimmune in origin is the higher incidence of specific
familial histocompatibility haplotypes such as HLA-B8 and HLA-DR3.
Low pepsinogen levels have also been observed; thus, this marker has
been used as an additional diagnostic tool in autoimmune gastritis.
The parietal cell–containing gastric gland is preferentially targeted
in this form of gastritis, and achlorhydria results. Parietal cells are the
source of IF, the lack of which will lead to vitamin B12 deficiency and its
sequelae (megaloblastic anemia, neurologic dysfunction).
Gastric acid plays an important role in feedback inhibition of gastrin
release from G cells. Achlorhydria, coupled with relative sparing of the
antral mucosa (site of G cells), leads to hypergastrinemia. Gastrin levels
can be markedly elevated (>500 pg/mL) in patients with pernicious
anemia. ECL cell hyperplasia with frank development of gastric carci­
noid tumors may result from gastrin trophic effects. Hypergastrinemia
and achlorhydria may also be seen in nonpernicious anemia–associated
type A gastritis.
TYPE B GASTRITIS  Type B, or antral-predominant, gastritis is the
more common form of chronic gastritis. H. pylori infection is the cause
of this entity. Although described as “antral-predominant,” this is likely
a misnomer in view of studies documenting the progression of the
inflammatory process toward the body and fundus of infected individ­
uals. The conversion to a pangastritis is time dependent and estimated
to require 15–20 years. This form of gastritis increases with age, being
present in up to 100% of persons aged >70. Histology improves after
H. pylori eradication. The number of H. pylori organisms decreases
dramatically with progression to gastric atrophy, and the degree of
inflammation correlates with the level of these organisms. Early on,
with antral-predominant findings, the quantity of H. pylori is highest
and a dense chronic inflammatory infiltrate of the lamina propria is
noted, accompanied by epithelial cell infiltration with polymorpho­
nuclear leukocytes (Fig. 335-16).
Multifocal atrophic gastritis, gastric atrophy with subsequent
metaplasia, has been observed in chronic H. pylori–induced gastritis.
This may ultimately lead to development of gastric adenocarcinoma
(Fig. 335-8; Chap. 85). H. pylori infection is now considered an
independent risk factor for gastric cancer. Worldwide epidemiologic
studies have documented a higher incidence of H. pylori infection in
FIGURE 335-16  Chronic gastritis and H. pylori organisms. Steiner silver stain of
superficial gastric mucosa showing abundant darkly stained microorganisms
layered over the apical portion of the surface epithelium. Note that there is no tissue
invasion.
patients with adenocarcinoma of the stomach as compared to control
subjects. Seropositivity for H. pylori is associated with a three- to
sixfold increased risk of gastric cancer. This risk may be as high as
ninefold after adjusting for the inaccuracy of serologic testing in the
elderly. The mechanism by which H. pylori infection leads to cancer
is unknown, but it appears to be related to the chronic inflamma­
tion induced by the organism. Eradication of H. pylori as a general
preventative measure for gastric cancer is being evaluated but is not
yet recommended.
Infection with H. pylori is also associated with development of a
low-grade B-cell lymphoma, gastric MALT lymphoma (Chap. 113).
The chronic T-cell stimulation caused by the infection leads to produc­
tion of cytokines that promote the B-cell tumor. The tumor should be
initially staged with a CT scan of the abdomen and EUS. Tumor growth
remains dependent on the presence of H. pylori, and its eradication is
often associated with complete regression of the tumor. The tumor
may take more than a year to regress after treating the infection. Such
patients should be followed by EUS every 2–3 months. If the tumor is
stable or decreasing in size, no other therapy is necessary. If the tumor
grows, it may have become a high-grade B-cell lymphoma. When the
tumor becomes a high-grade aggressive lymphoma histologically, it
loses responsiveness to H. pylori eradication.
TREATMENT
Chronic Gastritis
Treatment in chronic gastritis is aimed at the sequelae and not the
underlying inflammation. Patients with pernicious anemia will
require parenteral vitamin B12 supplementation on a long-term
basis. Eradication of H. pylori is often recommended even if PUD or
a low-grade MALT lymphoma is not present. Expert opinion sug­
gests that patients with atrophic gastritis complicated by intestinal
metaplasia without dysplasia should undergo surveillance endos­
copy every 3 years.
Miscellaneous Forms of Gastritis 
Lymphocytic gastritis is char­
acterized histologically by intense infiltration of the surface epithelium
with lymphocytes. The infiltrative process is primarily in the body of
the stomach and consists of mature T cells and plasmacytes. The etiol­
ogy of this form of chronic gastritis is unknown. It has been described
in patients with celiac sprue, but whether there is a common factor
associating these two entities is unknown. No specific symptoms sug­
gest lymphocytic gastritis. A subgroup of patients has thickened folds
noted on endoscopy. These folds are often capped by small nodules
that contain a central depression or erosion; this form of the disease
is called varioliform gastritis. H. pylori probably plays no significant
role in lymphocytic gastritis. Therapy with glucocorticoids or sodium
cromoglycate has obtained unclear results.
Marked eosinophilic infiltration involving any layer of the stom­
ach (mucosa, muscularis propria, and serosa) is characteristic of
eosinophilic gastritis. Affected individuals will often have circulating
eosinophilia with clinical manifestation of systemic allergy. Involve­
ment may range from isolated gastric disease to diffuse eosinophilic
gastroenteritis. Antral involvement predominates, with prominent
edematous folds being observed on endoscopy. These prominent antral
folds can lead to outlet obstruction. Patients can present with epigastric
discomfort, nausea, and vomiting. Treatment with glucocorticoids has
been successful.
Several systemic disorders may be associated with granulomatous
gastritis. Gastric involvement has been observed in Crohn’s disease.
Involvement may range from granulomatous infiltrates noted only on
gastric biopsies to frank ulceration and stricture formation. Gastric
Crohn’s disease usually occurs in the presence of small-intestinal dis­
ease. Several rare infectious processes can lead to granulomatous gas­
tritis, including histoplasmosis, candidiasis, syphilis, and tuberculosis.
Other unusual causes of this form of gastritis include sarcoidosis, idio­
pathic granulomatous gastritis, and eosinophilic granulomas involving
the stomach. Establishing the specific etiologic agent in this form of
gastritis can be difficult, at times requiring repeat endoscopy with
biopsy and cytology. Occasionally, a surgically obtained full-thickness
biopsy of the stomach may be required to exclude malignancy.
Russell body gastritis (RBG) is a mucosal lesion of unknown etiol­
ogy that has a pseudotumoral endoscopic appearance. Histologically, it
is defined by the presence of numerous plasma cells containing Russell
bodies (RBs) that express kappa and lambda light chains. Only 10 cases
have been reported, and 7 of these have been associated with H. pylori
infection. The lesion can be confused with a neoplastic process, but it
is benign in nature, and the natural history of the lesion is not known.
There have been cases of resolution of the lesion when H. pylori was
eradicated.
CHAPTER 335
Peptic Ulcer Disease and Related Disorders
Immune checkpoint inhibitor–induced enterocolitis and gastritis
are recognized sequelae of these oncologic therapies. The gastritis
typically occurs later in the course of therapy. The diagnosis is made
by the histologic findings on gastric mucosal biopsies obtained endo­
scopically. This is an important diagnosis to make since therapy with
glucocorticoids and potentially IL-6 receptor blockers will be required.
Moreover, this side effect will have an effect on the oncologic therapy
prescribed.
■
■MÉNÉTRIER’S DISEASE
Ménétrier’s disease (MD) is a very rare gastropathy characterized
by large, tortuous mucosal folds. MD has an average age of onset of
40–60 years with a male predominance. The differential diagnosis of
large gastric folds includes ZES, malignancy (lymphoma, infiltrat­
ing carcinoma), infectious etiologies (CMV, histoplasmosis, syphilis,
tuberculosis), gastritis polyposa profunda, and infiltrative disorders
such as sarcoidosis. MD is most commonly confused with large or
multiple gastric polyps (prolonged PPI use) or familial polyposis syn­
dromes. The mucosal folds in MD are often most prominent in the
body and fundus, sparing the antrum. Histologically, massive foveolar
hyperplasia (hyperplasia of surface and glandular mucous cells) and a
marked reduction in oxyntic glands and parietal cells and chief cells are
noted. This hyperplasia produces the prominent folds observed. The
pits of the gastric glands elongate and may become extremely dilated
and tortuous. Although the lamina propria may contain a mild chronic
inflammatory infiltrate including eosinophils and plasma cells, MD is
not considered a form of gastritis. The etiology of this unusual clinical
picture in children is often CMV, but the etiology in adults is unknown.
Overexpression of the growth factor TGF-α has been demonstrated in
patients with MD. The overexpression of TGF-α in turn results in over­
stimulation of the epidermal growth factor receptor (EGFR) pathway
and increased proliferation of mucus cells, resulting in the observed
foveolar hyperplasia.
The clinical presentation in adults is usually insidious and progres­
sive. Epigastric pain, nausea, vomiting, anorexia, peripheral edema,

06 - 336 Disorders of Absorption
336 Disorders of Absorption
and weight loss are signs and symptoms in patients with MD. Occult GI
bleeding may occur, but overt bleeding is unusual and, when present,
is due to superficial mucosal erosions. In fact, bleeding is more often
seen in one of the common mimics of MD, gastric polyposis. Twenty
to 100% of patients (depending on time of presentation) develop a pro­
tein-losing gastropathy due to hypersecretion of gastric mucus accom­
panied by hypoalbuminemia and edema. Gastric acid secretion is
usually reduced or absent because of the decreased parietal cells. Large
gastric folds are readily detectable by either radiographic (barium
meal) or endoscopic methods. Endoscopy with deep mucosal biopsy,
preferably full thickness with a snare technique, is required to establish
the diagnosis and exclude other entities that may present similarly. A
nondiagnostic biopsy may lead to a surgically obtained full-thickness
biopsy to exclude malignancy. Although MD is considered premalig­
nant by some, the risk of neoplastic progression is not defined. Com­
plete blood count, serum gastrin, serum albumin, CMV and H. pylori
serology, and pH testing of gastric aspirate during endoscopy should be
included as part of the initial evaluation of patients with large gastric
folds. A retrospective case-control study suggested an increased risk of
developing gastric adenocarcinoma and decreased 5- and 10-year sur­
vival compared to a control cohort. Additional studies will be required
to confirm these initial findings, but this may be difficult to achieve in
light of the rarity of this unusual diagnosis.
TREATMENT
Ménétrier’s Disease
Medical therapy with anticholinergic agents, prostaglandins, PPIs,
prednisone, somatostatin analogues (octreotide), and H2 receptor
antagonists yields varying results. Ulcers should be treated with
a standard approach. The discovery that MD is associated with
overstimulation of the EGFR pathway has led to the successful
use of the EGF inhibitory antibody, cetuximab, in these patients.
Specifically, four of seven patients who completed a 1-month trial
with this agent demonstrated near-complete histologic remission
and improvement in symptoms. Cetuximab is now considered the
first-line treatment for MD, leaving partial or total gastrectomy for
severe disease with persistent and substantial protein loss despite
therapy with this agent.
PART 10
Disorders of the Gastrointestinal System
■
■FURTHER READING
Abrignani MG et al: Proton pump inhibitors and gastroprotection in
patients treated with antithrombotic drugs: A cardiologic point of
view. World J Cardiology 15:8, 2023.
Bindu S et al: Non-steroidal anti-inflammatory drugs (NSAIDs)
and organ damage: A current perspective. Biochem Pharmacol
180:114147, 2020.
Bjarnason I et al: Mechanisms of damage to the gastrointestinal
tract from nonsteroidal anti-inflammatory drugs. Gastroenterology
154:500, 2018.
Brandi ML et al: Multiple endocrine neoplasia type 1: Latest insights.
Endocr Rev 42:133, 2021.
Chey WD et al: ACG clinical guideline: Treatment of Helicobacter
pylori infection. Am J Gastroenterol 112:212, 2017.
Engevik A et al: The physiology of the gastric parietal cell. Physiol
Rev 100:2, 2020.
Jensen RT, Ito T: Gastrinoma, in Endotext [Internet]. Feingold KR,
Anawalt B, Blackman MR, et al (eds). South Dartmouth (MA) 2000-.
Updated Sept. 2023. Available from MDText.com https://www.ncbi.
nlm.nih.gov/books/NBK279075/.
Kavitt RT et al: Diagnosis and treatment of peptic ulcer disease. Am
J Med 132:447, 2019.
Luo Q et al: A review on the research progress on non-pharmaco­
logical therapy of Helicobacter pylori. Front Microbiol 14:1134254,
Malfertheiner P et al: Management of Helicobacter pylori infection:
The Maastricht V/Florence Consensus Report. Gut 71:9, 2022.
Malfertheiner P et al: Helicobacter pylori infection. Nat Rev 9:19,
2023.
Pennelli G et al: Gastritis: Update on etiological features and histo­
logical practice approach. Pathologica 112:153, 2020.
Rao S et al: Survival times of patients with Menetrier’s disease and risk
of gastric cancer. Clin Gastroenterol Hepatol 19:10, 2021.
Savarino V et al: Proton pump inhibitors: Use and misuse in the clini­
cal setting. Expert Rev Clin Pharmacol 11:1123, 2018.
Shah SC et al: AGA Clinical Practice Update on the management of
refractory Helicobacter pylori infection: Expert review. Gastroenterol­
ogy 160:5, 2021.
Yao X, Smolka AJ: Gastric parietal cell physiology and Helicobacter
pylori-induced disease. Gastroenterology 156:2158, 2019.
Deborah C. Rubin
Disorders of Absorption
A wide range of diseases affect gastrointestinal (GI) absorptive func­
tion and may result in malabsorption syndromes. These disorders
affect one or more of the three phases of enteral nutrient processing.
Luminal digestion is initiated by lingual and gastric lipase and gastric
pepsin and continues in the small bowel by the actions of pancreatic
enzymes and bile salts. Small-intestinal mucosal digestion and absorp­
tion are mediated by enterocyte brush border enzymes including
disaccharidases, enterokinases, and peptidases, which digest nutri­
ents upon contact, and by mixed micelles containing lipids and bile
salts. Protein and carbohydrate digestive products are transported
into the enterocyte by carriers and transporters, and lipids enter by
diffusion mediated by micelles. Once in the enterocyte, nutrients may
be reprocessed for postmucosal absorption and entry into lymphatics
(long-chain triglycerides as part of chylomicrons) or are transported
into the bloodstream. Malabsorptive diseases or syndromes can be
classified by their effects on one or more of these three phases of
absorption (Table 336-1).
Disorders of absorption also have diverse clinical presentations.
For example, the deficiency of a single brush border membrane
protein such as lactase causes symptoms of diarrhea by affecting the
absorption of one nutrient, lactose. Celiac sprue may be localized
to the duodenum and present with isolated iron deficiency or may
cause diffuse intestinal mucosal disease, affecting the absorption of
multiple nutrients and causing a constellation of symptoms and clini­
cal presentations.
Definition of Diarrhea 
Diarrhea is the most common symptom
associated with disorders of absorption. For most patients, diarrhea as
a symptom is defined as an increase in stool number or frequency or a
change in consistency. Because normal bowel patterns may vary from
as many as two to four bowel movements per day to one stool per week,
it is critical to use an objective measure of diarrhea to help direct evalu­
ation. In health, stool volume or weight is <200 mL or <200 g, respec­
tively, in 24 h. Collection of stool for weight/volume determination is
one of the most useful tools for an evaluation of diarrhea. In particular,
a 72-h collection for weight/volume and fecal fat determination is the
gold standard for documenting the presence of steatorrhea, or fatty
stool. Steatorrhea, defined as increased stool fat excretion to >7% of
dietary fat, is a common manifestation of malabsorption. Steatorrhea
often results in large, bulky, and malodorous stools. Malabsorption
of single nutrients like lactose may result in an osmotic diarrhea, in
which the osmotically active unabsorbed nutrient causes fluid to be
drawn into the GI tract lumen. Malabsorptive diarrhea frequently is
precipitated by eating and resolves or significantly decreases at night,
TABLE 336-1  Classification of Malabsorption Syndromes
Inadequate digestion
  Postgastrectomya
  Deficiency or inactivation of pancreatic lipase
    Exocrine pancreatic insufficiency
      Chronic pancreatitis
      Pancreatic carcinoma
      Cystic fibrosis
      Pancreatic insufficiency—congenital or acquired
  Gastrinoma—acid inactivation of lipase
  Drugs—orlistat
Reduced intraduodenal bile-acid concentration/impaired micelle formation
  Liver disease
    Parenchymal liver disease
    Cholestatic liver disease
  Bacterial overgrowth in small intestine:
    Anatomic stasis
Functional stasis
      Afferent loop
  Diabetesa
      Stasis/blind
  Sclerodermaa
      Loop/strictures/fistulae
  Intestinal pseudo-obstruction
  Interrupted enterohepatic circulation of bile salts
    Ileal resection
    Crohn’s disease
  Drugs (binding or precipitating bile salts)—neomycin, cholestyramine, calcium
carbonate
  Impaired mucosal absorption/mucosal loss or defect
    Intestinal resection or bypassa
    Inflammation, infiltration, or infection:
      Crohn’s diseasea
Celiac disease
      Amyloidosis
Collagenous sprue
      Sclerodermaa
Whipple’s diseasea
      Lymphomaa
Radiation enteritisa
      Eosinophilic enteritis
Folate and vitamin B12 deficiency
      Mastocytosis
Infections—giardiasis
      Tropical sprue
Graft versus host disease
  Genetic disorders
    Disaccharidase deficiency
    Agammaglobulinemia
    Abetalipoproteinemia
    Hartnup’s disease
    Cystinuria
Impaired nutrient delivery to and/or from intestine:
  Lymphatic obstruction
Circulatory disorders
    Lymphomaa
  Congestive heart failure
    Lymphangiectasia
  Constrictive pericarditis
  Mesenteric artery atherosclerosis
  Vasculitis
Endocrine and metabolic disorders
  Diabetesa
  Hypoparathyroidism
  Adrenal insufficiency
  Hyperthyroidism
  Carcinoid syndrome
aMalabsorption caused by more than one mechanism.
with fasting, and thus can frequently be distinguished from secre­
tory diarrheas, for example from infectious causes such as bacterial
enterotoxigenic Escherichia coli. In this circumstance, intestinal fluid
and electrolyte secretion is stimulated by enterotoxin and will continue
even during fasting.
OVERVIEW: NUTRIENT DIGESTION
AND ABSORPTION
Luminal digestive processes begin in the mouth and proceed through­
out the GI tract, mediated by salivary amylase, lingual and gastric
lipases, gastric acid, pancreatic enzymes, and bile salts. As nutrients are
digested in the lumen of the proximal GI tract, they are further pro­
cessed by enterocyte brush border enzymes including disaccharidases
such as lactase and sucrase-isomaltase, which produce monosaccha­
rides, and peptidases, which hydrolyze polypeptides into tripeptides
and dipeptides and amino acids. Lipids in mixed micelles are then
absorbed into enterocytes.
The surface area of the small bowel, which is normally 600 cm
long, is further enhanced by circular folds, villi, and microvilli. Fol­
lowing uptake into enterocytes, nutrients are further processed and
transported into the lymphatics or into the portal circulation for use
by other cells throughout the body. The intestine is also presented with
7–9 L of fluid daily, a volume comprising dietary fluid intake (1–2 L/day)
and salivary, gastric, pancreatic, biliary, and intestinal fluid (6–7 L/day).
In health, almost all of this fluid is reabsorbed by the small bowel and
colon, resulting in a normal stool volume of <200 mL or stool weight
of <200 g.
■
■SPECIFIC NUTRIENTS
Lipids 
Lipid absorption is a complex process that requires hydroly­
sis by pancreatic enzymes and bile salts for physiochemical dispersion
of fats, followed by absorption of processed lipid nutrients dispersed
in bile salt–mixed micelles across the intestinal epithelium. Bile acids
are synthesized in the liver, secreted into the intestinal lumen, and
constantly recirculated by absorption in the ileum. The ileum expresses
fibroblast growth factor 19 (FGF19), which is a physiologic bile acid
sensor. FGF19 is secreted from the ileum into the bloodstream in
response to bile acid flux and negatively regulates hepatic bile acid
synthesis by affecting the transcription of hepatic CYP7A1.
CHAPTER 336
Disorders of Absorption
Thus, assimilation of dietary lipid requires three integrated pro­
cesses: an intraluminal or digestive phase, a mucosal or absorptive
phase, and a delivery or postabsorptive phase (Table 336-2).
Gastric lipases begin the lipolytic process. Following entry into
the small bowel, long-chain triglycerides, with carbon lengths >12
and that are the major component of dietary lipid, are hydrolyzed by
pancreatic lipases into fatty acids and monoglyceride during a process
called lipolysis (Fig. 336-1). Long-chain free fatty acids are dispersed
by bile salts into mixed micelles, which contact the brush border and
permit fatty acid absorption into enterocytes across this specialized
apical membrane. The other two types of fatty acids that compose fats,
medium-chain and short-chain fatty acids, are soluble in the unstirred
water layer. Medium-chain triglycerides with carbon chain lengths of
8–12 are found in coconut oil. Long-chain fatty acids are re-esterified
to triglycerides in enterocytes, packaged into chylomicrons that con­
tain apolipoproteins on the surface, which are subsequently secreted
into the extracellular space, and because of their size, are excluded
TABLE 336-2  Defects in Lipid Digestion and Absorption in Steatorrhea
PATHOPHYSIOLOGIC
DEFECT
DISEASE EXAMPLE
PHASE, PROCESS
Digestive
Lipolysis formation
Decreased lipase
secretion
Chronic pancreatitis
Micelle formation
Decreased intraduodenal
bile acids
Absorptive
Mucosal uptake and
re-esterification
Mucosal dysfunction
Celiac disease
Postabsorptive
Chylomicron formation
Absent β-lipoproteins
Abetalipoproteinemia
Delivery from intestine
Abnormal lymphatics
Intestinal
lymphangiectasia
Pancreas
Liver
Jejunal Mucosa
Lipolysis
Micellar
Solubilization
with Bile Acid
Absorption
(1) Esterification
Fatty acids
Fatty acids
To tissues
for utilization
of fat
Cholesterol
Phospholipid
β–Lipoprotein
Triglycerides
Triglycerides
β-Monoglyceride
β-Monoglyceride
(2) Chylomicron
formation
FIGURE 336-1  Schematic representation of lipid digestion and absorption. Dietary lipid is in the form of
long-chain triglycerides. The overall process can be divided into (1) a digestive phase that includes both
lipolysis and micelle formation requiring pancreatic lipase and conjugated bile acids, respectively, in the
duodenum; (2) an absorptive phase for mucosal uptake and re-esterification; and (3) a postabsorptive phase
that includes chylomicron formation and exit from the intestinal epithelial cell via lymphatics. (Courtesy of
John M. Dietschy, MD; with permission.)
from capillaries and enter the lymphatics. Medium-chain triglycerides
do not require micelle formation or pancreatic lipolysis as they are
directly absorbed intact from the small bowel into the bloodstream,
and short-chain fatty acids (carbon length <8) are produced by and
absorbed in the colon.
Carbohydrates 
Dietary carbohydrate consists of starch, sucrose,
lactose, maltose, and monosaccharides such as glucose and fructose.
Starch is digested by salivary α-amylase in the mouth, followed by
pancreatic amylase. The main products include maltotriose, maltose,
and α-dextrins. These are further digested on the brush border mem­
brane by disaccharidases such as glucoamylase and sucrase-isomaltase.
Dietary lactose is digested by brush border lactase, sucrose by sucrase,
and trehalose by trehalase. The final digested products are glucose,
fructose, and galactose, which are transported into the enterocyte by
transporters such as SLCA5 (formerly SGLT-1), which transports glu­
cose or galactose in a sodium-dependent manner, and GLUT-5, which
transports fructose by facilitated diffusion. Glucose, galactose, and
fructose exit the cell via GLUT-2.
PART 10
Disorders of the Gastrointestinal System
Proteins 
Dietary protein digestion begins in the stomach by pepsin.
Pancreatic proteases including endopeptidases, exopeptidases, and
trypsin are activated in the small-bowel lumen. Trypsinogen is acti­
vated by brush border enterokinase to generate active trypsin. Trypsin
in turn activates chymotrypsinogen to chymotrypsin, proelastase to
elastase, and procarboxypeptidases to carboxypeptidases A and B.
These enzymes digest protein into dipeptides, tripeptides, larger poly­
peptides, or free amino acids. At the brush border, peptidases digest
larger peptides into dipeptides and tripeptides or free amino acids,
which enter the enterocyte via specialized carriers. Most dipeptides
and tripeptides are further metabolized intracellularly by cytoplasmic
peptidase into amino acids, which directly enter the bloodstream via
carriers in the basolateral membrane. Small amounts of dipeptides and
tripeptides may also enter the bloodstream.
■
■LUMINAL PHASE OF DIGESTION
The luminal phase of digestion begins in the mouth, starting with
mastication and lipase secretion by the tongue and salivary glands.
The stomach continues the luminal digestive process, via gastric acid,
gastric lipase, and pepsin secretion as well as mechanical trituration
of contents. In the small-bowel lumen, pancreatic enzymes (amylase,
lipases, carboxypeptidase, trypsin, and other endopeptidases) contrib­
ute to carbohydrate, lipid, and protein digestion, respectively. Bile salts
produced by the liver are secreted into the intestinal lumen (and reab­
sorbed in the ileum via the enterohepatic circulation) and are required
for efficient lipid absorption.
Disorders That Affect the Luminal Phase of Digestion 
The
luminal phase may be disrupted by disorders of gastric and intestinal
motility including the sequelae of gastric surgery,
systemic diseases such as scleroderma, or endo­
crine disorders such as diabetes mellitus; pancreatic
diseases leading to pancreatic insufficiency with
reduced pancreatic enzyme secretion; or luminal
bile salt deficiency caused by hepatobiliary disease,
ileal disease, or small-bowel bacterial overgrowth.
Lymphatics
Delivery
Gastric Resection 
Surgical procedures that
remove or bypass part of the stomach and duo­
denal bulb such as Roux-en-Y gastric bypass for
weight loss, or resection of the gastric antrum
and duodenal bulb with creation of a Billroth II
anastomosis for treatment of peptic ulcer disease,
result in rapid gastric emptying into the jejunum,
which leads to diarrhea and weight loss due to
inadequate mixing of luminal nutrients with bile
and pancreatic secretions.
Disordered Intestinal Motility 
Hyperthy­
roidism may cause diarrhea and malabsorption
due to increased intestinal motility with rapid
transit, also resulting in inadequate nutrient mixing with pancre­
aticobiliary secretions. Long-standing diabetes mellitus may result in
damage to the enteric nervous system resulting in increased motility
and diarrhea, or reduced motility and constipation. Disorders that
affect the intestinal smooth muscle such as connective tissue disorders
including scleroderma may have profound effects on GI motility.
Pancreatic Disorders 
Chronic pancreatitis (see Chap. 359) may
result in a marked reduction in pancreatic enzyme secretion and pancre­
atic insufficiency, with subsequent fat, protein, and carbohydrate malab­
sorption. Patients with chronic pancreatitis present with steatorrhea, or
fatty stools, which are often voluminous, bulky, and malodorous. Patients
with steatorrhea also develop deficiency of fat-soluble vitamins including
vitamins A, E, and most commonly, vitamins D and K, which depend on
the same lipid absorption mechanisms and thus are malabsorbed along
with dietary fat. Weight loss is common. For a discussion of causes of
acute and chronic pancreatitis, please see Chap. 359.
Disorders That Result in Luminal Bile Salt Deficiency 
Bile
acid synthesis and the enterohepatic circulation are shown in Fig. 336-2.
NORMAL
Cholesterol
Bile acids
0.5 g synthesized
per day
[Bile acids]
4 mM
Bile acid
pool size
4.0 g
Jejunum
Ileum
Na
COLON
0.5 g
Bile acids
excreted per day
FIGURE 336-2  Schematic representation of the enterohepatic circulation of bile
acids. Bile-acid synthesis is cholesterol catabolism and occurs in the liver. Bile
acids are secreted in bile and are stored in the gallbladder between meals and
at night. Food in the duodenum induces the release of cholecystokinin, a potent
stimulus for gallbladder contraction resulting in bile-acid entry into the duodenum.
Bile acids are primarily absorbed via an Na-dependent transport process that is
located only in the ileum. A relatively small quantity of bile acids (~500 mg) is not
absorbed in a 24-h period and is lost in stool. Fecal bile acid losses are matched by
bile-acid synthesis. The bile acid pool (the total amount of bile acids in the body) is
~4 g and is circulated twice during each meal or six to eight times in a 24-h period.
TABLE 336-3  Defects in Enterohepatic Circulation of Bile Acids
PATHOPHYSIOLOGIC
DEFECT
DISEASE EXAMPLE
PROCESS
Synthesis
Decreased hepatic
function
Cirrhosis
Biliary secretion
Altered canalicular
function
Primary biliary
cholangitis
Maintenance of
conjugated bile acids
Bacterial overgrowth
Jejunal diverticulosis
Reabsorption
Abnormal ileal function
Crohn’s disease
Bile acids are synthesized from cholesterol in the liver. The two pri­
mary bile acids are cholic acid and chenodeoxycholic acid. These are
conjugated in the liver to taurine and glycine and are secreted into bile
ducts, stored in the gallbladder, and then delivered to the intestinal
lumen. Conjugation prevents bile acids from passive diffusion in the
small-bowel lumen, retaining bile acid concentrations required for
lipid absorption. Bile acids emulsify fats and fat-soluble vitamins to
facilitate their absorption. Bile acids are efficiently reabsorbed in the
ileum into the portal circulation and are extracted by the liver in a
process called enterohepatic circulation (Fig. 336-2). Small amounts are
deconjugated in the ileum by bacteria or pass into the colon and are
deconjugated and metabolized by colonic bacteria to become second­
ary bile acids. The two major secondary bile acids are lithocholic acid
and deoxycholic acid.
Processes that affect any of the above pathways may result in luminal
bile salt deficiency and malabsorption. Thus, hepatobiliary diseases,
intestinal ileal resection, extensive disease such as Crohn’s disease, and
small-bowel bacterial overgrowth may result in luminal bile salt defi­
ciency and malabsorption (Table 336-3).
Hepatobiliary Disease 
Hepatic disorders that result in decreased
bile acid synthesis due to hepatocyte dysfunction or reduced secretion
of bile into the gut lumen caused by diseases of the bile ducts such
as primary sclerosing cholangitis or primary biliary cholangitis may
result in luminal bile salt deficiency and fat malabsorption. These are
discussed in Chap. 357.
Ileal Resection or Ileal Disease 
Diseases that involve the ileal
mucosa or that result in ileal resection may lead to reduced recycling
of bile acids by the enterohepatic circulation and increased entry into
and concentration of bile acids in the colon, which produces a secre­
tory diarrhea, or may result in malabsorption due to inadequate bile
acid concentrations in the small-bowel lumen. In general, resection or
disease involving <100 cm of ileum results in bile acid spillage into the
colon; resections of >100 cm result in loss of bile acids that exceed liver
synthetic capacity, and malabsorption becomes the dominant patho­
physiologic mechanism for diarrhea, due to bile acid deficiency (Table
336-4). The most common disorder of the GI tract that targets the
ileum is Crohn’s disease (Chap. 337), which is a chronic inflammatory
disorder that may involve the entire GI tract, but most commonly the
TABLE 336-4  Comparison of Bile Acid and Fatty Acid Diarrhea
BILE-ACID DIARRHEA
FATTY ACID DIARRHEA
Extent of ileal disease
Limited
Extensive
Ileal bile acid absorption
Reduced
Reduced
Fecal bile acid excretion
Increased
Increased
Fecal bile acid loss
compensated by hepatic
synthesis
Yes
No
Bile acid pool size
Normal
Reduced
Intraduodenal (bile acid)
Normal
Reduced
Steatorrhea
None or mild
20 g
Response to
cholestyramine
Yes
No
Response to low fat diet
No
Yes
ileum and colon. If severe or refractory to treatment, Crohn’s disease
may lead to chronic inflammation, marked epithelial dysfunction, and
stricturing and fibrosis, and surgical resection may be required to treat
small-bowel obstruction or refractory disease.
Primary Bile Acid Diarrhea 
A subset of patients with functional
diarrhea or irritable bowel syndrome with diarrhea have been shown to
have bile acid malabsorption. Although the mechanisms are still being
elucidated, reduced FGF19 secretion by ileal enterocytes has been
observed. FGF19 regulates serum 7-alpha-hydroxy-4-cholesten-3-one
(C4) levels; reductions in circulating FGF19 lead to increased hepatic
bile acid synthesis via increased C4 expression. Chronic diarrhea
results from increased bile acid spillage into the colon, which induces a
secretory diarrhea and exerts prokinetic effects via TGR5.
Treatment 
Bile acid sequestrants are effective in reducing diarrhea
by binding bile acids to prevent spillage into the colon. Hepatic synthe­
sis of bile acids is sufficient to maintain intraluminal concentrations
that are adequate for fat absorption.
Small-Intestinal Bacterial Overgrowth 
The intestine contains
a rich microbiome. Bacterial titers increase along the horizontal axis
of the gut from duodenum to ileum. However, intestinal disorders
affecting motility or causing stasis of bowel contents may lead to smallintestinal bacterial overgrowth. These include scleroderma bowel,
chronic intestinal pseudo-obstruction, the creation of blind surgi­
cal loops such as Roux-en-Y gastric bypass, Billroth II anastomosis,
small-bowel strictures, or fibrosis from inflammatory disorders such
as Crohn’s disease and diffuse diverticulosis (Fig. 336-3). Surgical
resection of the ileocecal valve increases ileal bacterial counts from the
colon. Hypochlorhydria from chronic proton pump inhibitor use also
increases the risk of small-intestinal bacterial overgrowth. Bacterial
overgrowth causes deconjugation of bile acids, which facilitates their
absorption in the proximal bowel and results in luminal bile acid defi­
ciency, which in turn causes malabsorptive diarrhea with steatorrhea.
Bacterial overgrowth may also damage the brush border and result in
carbohydrate maldigestion and short-chain fatty acid production in
the colon, with diarrhea and gas. These patients are also at risk for B12
deficiency due to bacterial metabolism of B12 resulting in macrocytic
anemia and peripheral neuropathy. In contrast, elevated serum folate
levels may also be observed, derived from bacterial synthesis of folate.
CHAPTER 336
Disorders of Absorption
Small-intestinal bacterial overgrowth has also been observed in
patients with diarrhea-predominant irritable bowel syndrome. The
underlying mechanisms are unclear, but treatment of bacterial over­
growth leads to resolution of symptoms in a subset of irritable bowel
syndrome patients.
Diagnosis 
Duodenal aspirate for bacterial titers is the gold standard
but is not generally available to most practitioners. Breath hydrogen
testing with administration of lactulose, a nondigestible disaccharide,
or glucose is widely available but must be interpreted carefully to avoid
false-positive results. Many clinicians choose to treat empirically with
antibiotics (see “Treatment”) and observe for resolution of symptoms.
Treatment 
When possible, surgical correction of blind loops,
endoscopic or surgical treatment of strictures, and removal of large
diverticula can be pursued for definitive therapy, in addition to treat­
ment of underlying disorders such as Crohn’s disease to avoid recurrent
stricture formation or fibrosis. Other disorders such as scleroderma or
other diffuse motility disorders may not be easily treated. In these cir­
cumstances, treatment with the nonabsorbable antibiotic, rifaximin, or
with other antibiotics such as metronidazole, doxycycline, amoxicillinclavulanic acid, or cephalosporins for several weeks is often pursued.
Patients may require retreatment or even chronic therapy with rotating
antibiotics depending on the severity of symptoms.
■
■MUCOSAL PHASE OF DIGESTION
AND ABSORPTION
The intestinal epithelium (also known as the mucosa) plays a critical
role in continued digestion of nutrients and absorption from the intes­
tinal lumen into the bloodstream and lymphatics.
A
B
PART 10
Disorders of the Gastrointestinal System
C
D
FIGURE 336-3  Barium contrast small-intestinal radiologic examinations. A. Normal individual. B. Celiac disease. C. Jejunal diverticulosis. D. Crohn’s disease. (Courtesy of
Morton Burrell, MD, Yale University; with permission.)
The small-bowel epithelial or mucosal digestive and absorptive
phase is mediated by enterocytic brush border enzymes, including
peptidases and hydrolases. Brush border enterokinase is required for
the conversion of pancreatic trypsinogen to trypsin, which further
activates trypsinogen and other pancreatic protease proenzymes. The
brush border membrane of the small-bowel epithelium expresses a
wide variety of disaccharidases, peptidases, and other hydrolases that
continue the digestive process for carbohydrates and proteins, with
enzymatic digestion of disaccharides to monosaccharides and dipepti­
dases to amino acids, which are then absorbed by specific transporters.
Long-chain fatty acids are re-esterified to triglycerides in enterocytes,
packaged into chylomicrons with apolipoproteins on the surface, which
are subsequently secreted into the extracellular space, and because of
their size, are excluded from capillaries and enter the lymphatics.
INTESTINAL MUCOSAL DISORDERS
■
■DISORDERS OF ENTEROCYTE CARBOHYDRATE
TRANSPORTERS AND ENZYME DEFICIENCIES
Lactose Intolerance Due to Lactase Deficiency 
This is the
most common brush border disaccharidase deficiency and is a frequent
cause of diarrhea, abdominal pain, gassiness, and bloating. Lactose is
present in many dairy products but is also a “hidden” component of a
vast number of processed foods.
Lactose malabsorption can result from lactase deficiency, which is
regulated by primary genetic mechanisms (adult-type hypolactasia) or
secondary due to damage to the epithelial (mucosal) lining of the gut,
from infections (viral, bacterial, or parasitic) or from intestinal muco­
sal diseases. Congenital lactase deficiency is very rare and is an auto­
somal recessive disorder. Hypolactasia in adulthood is very common
throughout the world and is considered to be the genetic wild-type;
lactase persistence results from a C to T mutation (LACTASE LCT13910CT and LCT-13910TT), and adults with hypolactasia have
absence of this “persistence” allele. Lactose is metabolized by lactase
into glucose and galactose, which are both absorbed by transporters at
the enterocyte surface. Patients who are lactase deficient have elevated
luminal lactose levels upon ingestion of lactose. The mechanism for
diarrhea in lactase deficiency is complex. Undigested lactose acts as an
osmotic substance to draw fluid into the small-bowel lumen. In addi­
tion, when unabsorbed lactose enters the colon, luminal bacteria fer­
ment lactose, producing intestinal gas (hydrogen, carbon dioxide, and
methane), bloating, and abdominal pain. Luminal lactose is metabo­
lized by bacteria into short-chain fatty acids that can be absorbed by
the colon, but watery diarrhea may occur when a large lactose load
exceeds the colon’s absorptive capacity.
Diagnosis 
When lactose intolerance is suspected, a common initial
approach is to institute a lactose-exclusion diet and assess for resolu­
tion of symptoms. This is a rapid and generally effective diagnostic and
therapeutic method. Patients are provided with a list of lactose-containing
foods and lactose-free alternatives. Patients are also counseled on
alternative calcium sources, because dairy-containing foods are a
major source of dietary calcium, which is important for osteoporosis
prevention.
Should the results of dietary exclusion be ambiguous, a lactosetolerance test or breath hydrogen test may prove useful. For the lactosetolerance test, patients ingest a standardized liquid lactose solution
(usually 50 g of lactose) followed by timed measurements of serum
glucose for 90 min. If lactose digestion is normal, glucose levels should
rise by >20 mg/L. Serum glucose rise <20 mg/L plus the presence of
symptoms of lactose intolerance (abdominal discomfort, gassiness,
and diarrhea) is considered a positive test. A breath hydrogen test is
performed by measuring breath hydrogen levels following ingestion
of a standardized lactose load. Breath hydrogen levels should not
exceed >20 ppm above the fasting baseline. Generally, the peak occurs
between 2 and 4 h. Both methods may be inaccurate if the patient
has abnormal gastric emptying or abnormal intestinal transit. Breath
hydrogen measurements may be abnormal in the setting of bacterial
overgrowth, which may cause very similar symptoms.
Treatment 
Patients may elect to completely eliminate lactose from
their diets. It is very important to consider calcium and vitamin D
supplementation because elimination of milk and soft cheeses removes
important dietary sources. They also may need to consult a dietitian for
guidance about hidden lactose in prepared or other foods. An alterna­
tive is to consider using lactase supplementation, which is available
over the counter but which may need to be titrated to avoid symptoms.
Glucose-Galactose Malabsorption 
This rare congenital disor­
der is an autosomal recessive disease in which mutations occur in the
SLC5A1 gene (also known as SGLT1). SLC5A1 is a brush border pro­
tein and member of the sodium-dependent glucose transporter family;
mutations in this gene result in malabsorption of glucose and galactose.
Gene sequencing has shown that most patients have loss-of-function
single-nucleotide variations. SLC5A1 actively transports glucose or
galactose coupled to sodium cotransport; patients who are homozy­
gous for these loss-of-function variants have severe congenital diarrhea
and death if unrecognized. Treatment focuses on eliminating glucose-
and galactose-containing foods and substituting fructose-containing
foods. Fructose is absorbed by the brush border transporter GLUT5 by
facilitated diffusion and is not dependent on SLC5A1.
Sucrase-Isomaltase Deficiency  
This is another rare congenital
autosomal recessive disorder, caused by mutations in the sucraseisomaltase gene. Infants present with diarrhea that begins with
ingestion of sucrose coincident with the introduction of table foods.
Sucrase-isomaltase gene variants have been identified that cause symp­
toms later in life and may be associated with increased risk of irritable
bowel syndrome.
Abetalipoproteinemia 
Abetalipoproteinemia is a rare disorder
of lipid metabolism associated with abnormal erythrocytes (acan­
thocytes), neurologic symptoms, and steatorrhea (see Chap. 419).
Lipolysis, micelle formation, and lipid uptake are all normal in patients
with abetalipoproteinemia, but the re-esterified triglyceride cannot exit
the epithelial cell because of the failure to produce chylomicrons. This
disorder results from mutation of microsomal triglyceride transfer pro­
tein (MTP), which catalyzes the transfer of triglyceride onto nascent
apolipoprotein B–containing particles. Mutations in MTP decrease
this transfer and decrease formation of chylomicrons. Small-intestinal
biopsy samples obtained from these rare patients in the postprandial
state reveal lipid-laden small-intestinal epithelial cells that become
normal in appearance after a 72- to 96-h fast.
INTESTINAL MUCOSAL DISORDERS THAT
RESULT IN MALABSORPTION OF MULTIPLE
NUTRIENTS
■
■CELIAC DISEASE
Celiac disease, also known as celiac sprue or gluten-sensitive enter­
opathy, is a small-intestinal enteropathy that results from an immune
response to gluten ingestion and is characterized by autoantibodies
to tissue transglutaminase. Gluten is found in foods produced from
wheat, rye, barley, and some varieties of oats, and it is a common addi­
tive to prepared foods and pharmaceuticals. Tissue transglutaminase
is involved in the pathogenesis of this disorder, as it deamidates gluta­
mine residues of gluten-derived peptides, facilitating their presentation
by antigen-presenting cells.
Epidemiology and Genetics 
The incidence and prevalence of
celiac disease have been increasing worldwide. Increased awareness
among clinicians and patients has led to increases in detection, but
there is evidence that the true incidence appears to be increasing as
well. Global prevalence has been measured at 1.4%. In the United States,
data from the National Health and Nutrition Examination survey
showed seroprevalence of 0.2% in non-Hispanic black populations,
0.3% in Hispanic individuals, and 1.0% in white populations.
The prevalence of celiac disease is 10–15% in first-degree rela­
tives. Host genetic factors include histocompatibility locus antigens
HLADQ2 and DQ8; the presence of one of the two haplotypes is
necessary but not sufficient for developing celiac disease. HLADQ2
and DQ8 are found in 25–35% of the general population; because most
carriers never develop celiac disease, detection of these alleles is not
useful for diagnosis. However, a negative test is very useful for ruling
out celiac disease, with a negative predictive value of >99%. This is
particularly helpful in patients who self-discontinued gluten ingestion
prior to serologic or endoscopic testing.
Presentation 
Patients with celiac disease have a wide variety of
disease manifestations, ranging from being asymptomatic, to having
isolated iron-deficiency anemia due to duodenal disease, to severe
diarrhea, weight loss, and malabsorption of multiple nutrients with
more diffuse disease. Celiac disease primarily affects the proximal
small intestine; it may involve the duodenum only or may cause wide­
spread jejunal disease resulting in severe symptoms.
Diarrhea, weight loss, and growth failure in children are common
presenting complaints, but additional signs and symptoms have become
increasingly recognized to be associated with celiac disease, including
bloating and irregular bowel habits, migraine headaches, and ataxia. In
addition, patients may be identified after presenting with osteoporosis,
iron-deficiency anemia, or detection of abnormal liver enzymes.
Mechanism of Diarrhea 
Patients with celiac disease have villus
atrophy in the proximal small intestine and thus develop steatorrhea
from mucosal malabsorption and may have lactase deficiency. How­
ever, they also develop a secretory component due to crypt hyperplasia
and fluid hypersecretion from the crypt epithelium.
CHAPTER 336
Associated Diseases 
Patients with celiac disease have a higher
incidence of other autoimmune disorders such as type 1 diabetes mel­
litus and autoimmune thyroid disease. Dermatitis herpetiformis is a skin
disorder that is highly associated with celiac disease, characterized by a
vesicular rash mediated by IgA deposits in the skin. Down syndrome and
Turner syndrome patients also have an increased risk of celiac disease.
Disorders of Absorption
Diagnosis 
Patients are screened for celiac disease first by testing
for serum antibodies, including the screening test of choice, which
is tissue transglutaminase IgA (TTG-IgA). Serum IgA levels should
also be measured to detect false-negative results from IgA deficiency.
If a patient is IgA deficient, IgG-based testing including TTG-IgG or
anti- deamidated gliadin peptide (DGP-IgG) antibody testing can be
performed. The diagnosis in adults with positive antibody levels is con­
firmed by endoscopy with small-intestinal biopsy; biopsy is required
in most patients to confirm the diagnosis. Biopsies typically show
characteristic villus blunting, crypt hyperplasia, and inflammation,
including increased intraepithelial lymphocytes. The Marsh classifica­
tion categorizes different types of celiac disease–related lesions and is
currently used to quantify severity of disease involvement.
Family members of patients with celiac disease are screened if symp­
tomatic; recommendations regarding screening asymptomatic family
members are still controversial.
Complications 
Complications of celiac disease include refractory
celiac disease, enteropathy-associated T-cell lymphoma, hyposplenism,
and small-bowel adenocarcinoma.
Refractory Celiac Disease 
This complication is most common
in patients with ongoing active celiac disease, found in about 10% of
patients with persistent active disease. Patients have ongoing diarrhea
and weight loss with persistent villus atrophy on biopsy after 1 year of
following a strict gluten-free diet. These patients also have negative
celiac serology, confirming their adherence to the gluten-free diet.
Type 1 refractory celiac disease has a normal intraepithelial lympho­
cyte population, whereas type 2 disease has clonal expansion of CD3+
intraepithelial lymphocytes that also contain a monoclonal rearrange­
ment of the gamma chain of the T-cell receptor. Type 2 refractory
celiac disease has a worse prognosis due to its association with T-cell
lymphoma, which occurs in 33–50% of cases after 5 years. The therapy
for celiac disease–related lymphoma is intense and includes high-dose
chemotherapy and sometimes stem cell transplantation.
Small-bowel adenocarcinoma is a very rare cancer in the general
population but is increased in celiac disease patients.
Therapy and Follow-Up 
The mainstay of celiac disease treatment
is institution of a strict gluten-free diet. This is challenging for patients
because of the widespread presence of gluten in both raw and prepared
foods, inaccurate food labeling, and cross-contamination during food
preparation. Patients must receive rigorous dietary instruction from a
dietitian and adhere lifelong to a gluten-free diet.
For patients whose symptoms resolve, serologic follow-up is gener­
ally recommended to confirm compliance with a gluten-free diet. A
follow-up biopsy to document complete healing of villus atrophy is also
generally recommended. However, subsequent biopsies are not recom­
mended unless symptoms recur. For patients without symptom resolu­
tion, a biopsy is required to determine the degree of disease activity and
to rule out other causes of persistent diarrhea and complications such
as refractory celiac disease or T-cell lymphoma. The most common
cause of residual disease activity is dietary nonadherence or inadver­
tent gluten exposure. These patients pursue repeat consultation with
a dietitian and efforts to reduce restaurant or other out-of-the-home
exposure or cross-contamination at home. If biopsies are negative but
symptoms persist, other causes of abdominal pain and diarrhea that are
associated with celiac disease are considered, including irritable bowel
syndrome, microscopic colitis, small-bowel bacterial overgrowth, and
lactose or fructose intolerance.
Nonceliac Gluten Sensitivity 
These patients have symptoms
consistent with celiac disease but have negative serology and nega­
tive biopsies and no evidence for wheat allergy. Upon discontinuation
of gluten, they have relief of abdominal pain, diarrhea, headaches/
migraines, and other celiac disease–type symptoms. The etiology of
this disorder remains unknown.
PART 10
Disorders of the Gastrointestinal System
■
■WHIPPLE’S DISEASE
Whipple’s disease is a chronic, multiorgan disease caused by Troph­
eryma whipplei, a gram-positive non-acid-fast, periodic acid–Schiff
(PAS)–positive rod, which is ubiquitous in the environment. Whipple’s
disease most commonly occurs in middle-aged men. Classic Whipple’s
disease is defined by the presence of arthralgias, weight loss, diarrhea,
and abdominal pain. Other manifestations including central nervous
system (CNS) and cardiac involvement are common and occur later in
the disease. T. whipplei can be detected by polymerase chain reaction
on involved tissue and is difficult to detect in the bloodstream. The
intestinal lesion is also characterized by PAS-positive macrophages.
Clinical Presentation 
Arthralgias and arthritis are present for an
average of 6 years before the GI symptoms begin, consistent with a per­
sistent and substantial lag in diagnosis, which is still a problem today.
Joint disease is present in >80% of patients. GI manifestations include
diarrhea, abdominal pain, and weight loss from malabsorption. CNS
involvement is common and may include symptoms such as psychi­
atric manifestations or memory problems. Dementia and encephalitis
may occur in later stages. Cardiac involvement may include endocar­
ditis, pericarditis, and myocarditis.
Diagnosis 
For patients with GI manifestations, endoscopy with
biopsies is performed and tissue is tested for T. whipplei by polymerase
chain reaction. Tissue is also stained for PAS-positive macrophages, and
immunohistochemistry may also be performed to detect T. whipplei.
Treatment 
Prolonged antibiotics are recommended, although the
optimal regimen is still uncertain. Relapses are common and often
associated with the first manifestations of CNS involvement.
■
■TROPICAL SPRUE
Tropical sprue is a poorly understood syndrome that is manifested by
chronic diarrhea, steatorrhea, weight loss, and nutritional deficiencies,
including both folate and vitamin B12. Malabsorption of two unrelated
substances is required for diagnosis. This disease occurs in 8–20% of
people who have had an attack of infectious gastroenteritis in India
and is considered by some to be a postinfectious complication. It is
prevalent in some but not all tropical areas, including southern India,
Pakistan, the Philippines, Puerto Rico, Haiti, and Cuba. It occurs in
residents of as well as visitors to these areas.
Chronic diarrhea in a tropical environment is most often caused
by infectious agents, including Giardia lamblia, Yersinia enterocolitica,
Entamoeba histolytica, Clostridioides difficile, Cryptosporidium parvum,
Isospora belli, Strongyloides stercoralis, and Cyclospora cayetanensis.
Tropical sprue should not be entertained as a possible diagnosis until
the presence of cysts and trophozoites has been excluded in three stool
samples. Infections of the GI tract and diarrhea are discussed in
Chaps. 49, 138, 139, 169–173, and 230.
Etiology 
Because tropical sprue responds to antibiotics, the con­
sensus is that it may be caused by one or more infectious agents. None­
theless, the etiology and pathogenesis of tropical sprue are uncertain.
Its occurrence is not evenly distributed in all tropical areas; it is rarely
observed in Africa, Jamaica, or Southeast Asia. The incidence of tropi­
cal sprue appears to have decreased substantially during the past two
or three decades, perhaps in relation to improved sanitation in many
tropical countries during this time. Some have speculated that the
reduced occurrence is attributable to the wider use of antibiotics in
acute diarrhea, especially in travelers to tropical areas from temper­
ate countries. Folic acid is absorbed exclusively in the duodenum and
proximal jejunum, and most patients with tropical sprue have evidence
of folate malabsorption and depletion.
The clinical pattern of tropical sprue varies in different areas of
the world (e.g., India vs Puerto Rico). Not infrequently, individuals in
southern India initially report the occurrence of acute enteritis before
the development of steatorrhea and malabsorption. In contrast, in
Puerto Rico, a more insidious onset of symptoms and a more dramatic
response to antibiotics are seen compared with some other locations.
Tropical sprue in different areas of the world may not be the same dis­
ease, and similar clinical entities may have different etiologies.
Diagnosis 
The diagnosis of tropical sprue is based on an abnormal
small-intestinal mucosal biopsy in an individual with chronic diarrhea
and evidence of malabsorption who is either residing or has recently
lived in a tropical country. The small-intestinal biopsy in tropical sprue
does not reveal pathognomonic features but resembles, and can often
be indistinguishable from, that seen in celiac disease (Fig. 336-4). The
biopsy sample in tropical sprue has less villous architectural alteration
and more mononuclear cell infiltrate in the lamina propria. In contrast
to those of celiac disease, the histologic features of tropical sprue mani­
fest with a similar degree of severity throughout the small intestine,
and a gluten-free diet does not result in either clinical or histologic
improvement in tropical sprue.
Treatment 
Broad-spectrum antibiotics and folic acid are most
often curative, especially if the patient leaves the tropical area and does
not return. Tetracycline should be used for up to 6 months and may
be associated with improvement within 1–2 weeks. Folic acid alone
induces hematologic remission as well as improvement in appetite,
weight gain, and some morphologic changes in small-intestinal biopsy.
Because of marked folate deficiency, folic acid is most often given
together with antibiotics.
■
■ENVIRONMENTAL ENTERIC DYSFUNCTION
Environmental enteric dysfunction (EED; formerly known as tropical
enteropathy) occurs in low- and middle-income countries and is char­
acterized by altered crypt villus morphology with crypt atrophy and
mucosal inflammation and barrier dysfunction, resulting in chronic
low-grade malabsorption. It is associated with childhood stunting, a
major worldwide health problem. Gut microbiota and adverse envi­
ronmental and economic conditions contribute to its pathogenesis.
Although the gold standard for diagnosis is small-bowel biopsy, this is
often limited by lack of resources. Therapies for EED are the subject of
intense investigation.
A
B
C
D
E
F
G
FIGURE 336-4  Small-intestinal mucosal biopsies. A. Normal individual. B. Untreated celiac disease. C. Treated celiac disease. D. Intestinal lymphangiectasia. E. Whipple’s
disease. F. Lymphoma. G. Giardiasis. (Courtesy of Marie Robert, MD, Yale University; with permission.)
■
■SHORT-BOWEL SYNDROME
■
■OVERVIEW
Short-bowel syndrome results from intestinal resection to treat a mul­
titude of disorders including Crohn’s disease, vascular diseases such
as mesenteric arterial or venous thrombosis resulting in intestinal
ischemia, volvulus, trauma, internal herniation, radiation enteritis,
and diffuse carcinoma, among others. In children, the most common
causes of short-bowel syndrome are necrotizing enterocolitis, intestinal
atresias, volvulus, and malrotation. Short-bowel syndrome is defined
as extensive removal of small intestine resulting in <200 cm of remain­
ing small bowel. Intestinal failure is functionally defined as persistent
parenteral nutrition dependence, generally found in patients who have
<100 cm of remaining small bowel and no residual colon in continuity.
Clinical Features 
Loss of small-bowel surface area in short-bowel
syndrome results in severe diarrhea, weight loss, and malabsorption
of multiple nutrients, including fat, protein, and carbohydrate. The
severity of symptoms and ultimate dependence on parenteral nutrition
are generally related to the extent of resection, presence or absence of
CHAPTER 336
Disorders of Absorption
residual colon in continuity, retention of the ileocecal valve, and sever­
ity of the underlying disease. The intestine has a remarkable capacity
to adapt to loss of small-bowel surface area, but this adaptive process
is variable from patient to patient. Following resection, the adapting
residual intestine exhibits an increase in crypt cell proliferation result­
ing in epithelial hyperplasia. The adaptive process generally continues
for up to 2 years after resection, but improvements in nutrient, fluid,
and electrolyte absorptive capacity have been reported even as late as
3–5 years after surgery. Massive diarrhea generally occurs in the first
three postoperative months, associated with increased gastric acid
secretion and malabsorption. Gradually, patients show enhanced func­
tional capacity and reduced diarrhea. Specific nutrient deficiencies are
dependent upon which segment of gut has been removed. For example,
resection of the ileum results in loss of B12 absorptive and bile salt
reabsorptive capacity. Malabsorbed bile salts reach the colon and cause
a secretory diarrhea. In addition, resection of >100 cm of ileum results
in such severe bile salt malabsorption that the liver cannot compen­
sate by increased synthesis, thus precipitating fat malabsorption due
to bile salt insufficiency/deficiency. Substantial resection of the colon
also results in fluid and electrolyte loss and imbalance. The colon also
plays a role in nutrient absorption because it metabolizes malabsorbed
carbohydrate into short-chain fatty acids that can be absorbed by the
colon and can contribute several hundred additional calories per day.
Long-Term Complications 
Because massive resection often leads
to severe fat malabsorption, fat-soluble vitamin deficiency is common,
and vitamin D deficiency can be very difficult to treat even with highdose oral vitamin D supplementation, resulting in an increased risk of
osteoporosis. Patients with a history of multiple surgeries often have
extensive adhesive disease, and the residual intestine may have mark­
edly abnormal motility or areas of stricturing and narrowing, resulting
in recurrent bacterial overgrowth. The frequency of renal calcium
oxalate stones increases in patients with a shortened small bowel with
an intact colon in continuity; calcium is saponified in the intestinal
luminal contents that contain fatty acids, freeing oxalate to be absorbed
in the colon resulting in hyperoxaluria.
Treatment 
The major focus of treatment for short-bowel syndrome
is to control diarrhea and normalize nutrient, fluid, and electrolyte
absorption so that patients can maintain their weight and have a
healthy nutritional status without the support of parenteral nutrition.
Medications include opiates and derivatives including loperamide
and diphenoxylate-atropine, which slow intestinal motility to allow
for more contact time between luminal nutrients and the small-bowel
mucosal surface. In the first year following resection, acid-blocking
medications are used to treat gastric hypersecretion, including proton
pump inhibitors or histamine 2 antagonists. Small-bowel bacterial
overgrowth is common and is treated with antibiotics if suspected.
The only medication that is specific for short-bowel syndrome but
limited for use in parenteral nutrition or intravenous fluid–dependent
patients is teduglutide, a glucagon-like peptide 2 (GLP-2) analogue that
enhances crypt cell proliferation and villus hyperplasia and increases
nutrient and fluid and electrolyte absorption. Patients treated with
teduglutide have an average reduction of 20% of their parenteral nutri­
tion requirements. Greater efficacy has been noted for patients without
a residual colon, likely due to lower circulating endogenous GLP-2
levels compared to those with a colon in continuity.
Dietary Therapy 
Patients with short-bowel syndrome must con­
sume three to four times their normal caloric intake to maintain their
weight. The presence of luminal nutrients is required for the adaptive
process to occur, so early feeding is recommended, even if parenteral
nutrition is also required. These effects are most likely mediated by
direct contact with the mucosa as well as stimulation of secretion of gut
hormones such as GLP-2.
PART 10
Disorders of the Gastrointestinal System
If the patient has all or part of their colon remaining in continuity,
a low-fat diet is instituted to reduce the concentration of malabsorbed
fatty acids that induce a secretory diarrhea. High complex carbohy­
drates are encouraged because when malabsorbed and present in the
colon, they are converted to short-chain fatty acids and are absorbed,
contributing several hundred additional kilocalories per day. All
patients are asked to take a high-potency multivitamin on a daily basis.
Patients in whom oral nutrition fails are fed with parenteral nutrition.
Monitoring 
Patients with short-bowel syndrome are at high risk for
osteoporosis due to dietary calcium and vitamin D malabsorption, so
they are periodically monitored for vitamin D deficiency and calcium
levels and with dual x-ray absorptiometry (DEXA) studies to assess
bone density. Malabsorption of vitamins and minerals is common;
therefore, fat-soluble vitamins, vitamin B12, folic acid, iron, magnesium,
and zinc are monitored periodically. More unusual deficiencies include
copper, selenium, and chromium, but these are usually seen in paren­
teral nutrition–dependent patients and can be corrected by adjusting
daily intravenous dosages. Signs and symptoms of vitamin and mineral
deficiency are also carefully monitored (e.g., hair loss, skin and nail
changes, neurologic symptoms such as peripheral neuropathy).
■
■DISORDERS OF POSTMUCOSAL ABSORPTION
Following uptake into enterocytes, nutrients are further processed and
transported into the lymphatics or into the portal circulation for use
by other cells throughout the body. Primary or secondary disorders of
the lymphatics may result in significant diarrhea and malabsorption.
Primary disorders of the intestinal lymphatics include intestinal lym­
phangiectasia, which may be congenital or acquired. Secondary causes
of intestinal lymphatic damage or blockage include retroperitoneal
fibrosis, fibrosing mesenteritis, and lymphoma. Circulatory causes of
impaired delivery of nutrients from the intestine include Fontan physi­
ology, congestive heart failure, and constrictive pericarditis. The end
result of damage to lymphatic channels is malabsorption and diarrhea
with concomitant protein-losing enteropathy.
■
■PROTEIN-LOSING ENTEROPATHY
Protein-losing enteropathy refers to a large group of GI and non-GI
disorders characterized by hypoproteinemia and edema in the absence
of liver disease with reduced protein synthesis, or kidney disease with
proteinuria. These diseases are characterized by excess protein loss in
the GI tract. Diseases that may result in increased protein loss into the
GI tract can be classified into three groups: (1) mucosal ulceration,
such that the protein loss primarily represents exudation across dam­
aged mucosa (e.g., ulcerative colitis, GI carcinomas); (2) nonulcerated
mucosa, but with evidence of mucosal damage so that the protein loss
represents loss across epithelia with altered permeability (e.g., celiac
disease and Ménétrier’s disease [hypertrophic gastropathy] in the small
intestine and stomach, respectively); and (3) lymphatic dysfunction,
representing either primary lymphatic disease or lymphatic disease
secondary to partial lymphatic obstruction that may occur as a result
of enlarged lymph nodes or cardiac disease. These result in increased
lymphatic pressure, causing exudation of protein into the GI tract
lumen.
Diagnosis 
The diagnosis of protein-losing enteropathy is suggested
by diarrhea, peripheral edema, and low serum albumin and globulin
levels in the absence of renal and hepatic disease. An individual with
protein-losing enteropathy rarely has selective loss of only albumin or
only globulins. Therefore, marked reduction of serum albumin with
normal serum globulins should suggest renal and/or hepatic disease.
Likewise, reduced serum globulins with normal serum albumin lev­
els are more likely a result of reduced globulin synthesis rather than
enhanced globulin loss into the intestine. Alpha-1 antitrypsin, a pro­
tein that accounts for ~4% of total serum proteins and is resistant to
proteolysis, can be used to detect enhanced rates of serum protein loss
into the intestinal tract but cannot be used to assess gastric protein
loss because of its degradation in an acid milieu. Alpha-1 antitrypsin
can be measured in a spot or 24-h stool collection and, if elevated, is
diagnostic. A more accurate determination is alpha-1 antitrypsin clear­
ance, measured by determining stool volume as well as both stool and
plasma alpha-1 antitrypsin concentrations. In addition to the loss of
protein via abnormal and distended lymphatics, peripheral lympho­
cytes may be lost via lymphatics, with consequent relative lymphopenia
and specifically loss of CD3+ T cells. Thus, lymphopenia in a patient
with hypoproteinemia indicates increased loss of protein into the GI
tract.
Patients with increased protein loss into the GI tract from lymphatic
obstruction often have steatorrhea and diarrhea. The steatorrhea is
a result of altered lymphatic flow as lipid-containing chylomicrons
exit from intestinal epithelial cells via intestinal lymphatics (Table
336-2; Fig. 336-4). In the absence of mechanical or anatomic lym­
phatic obstruction, intrinsic intestinal lymphatic dysfunction—with
or without lymphatic dysfunction in the peripheral extremities—has
been designated intestinal lymphangiectasia. Similarly, ~50% of indi­
viduals with intrinsic peripheral lymphatic disease (Milroy’s disease)
also have intestinal lymphangiectasia and hypoproteinemia. Other
than steatorrhea and enhanced protein loss into the GI tract, all other
aspects of intestinal absorptive function are normal in intestinal
lymphangiectasia.
Endoscopy and Imaging 
Endoscopy (including double-balloon
enteroscopy) with biopsy and video capsule endoscopy may be per­
formed to rule out mucosal disease. Magnetic resonance enterography
or computed axial tomography may provide additional insight into the
underlying cause and can detect specific lesions, and lymphangiogra­
phy can be used for diagnosis and potential therapeutic intervention.
Other Causes 
Ménétrier’s disease (also called hypertrophic gas­
tropathy) is an uncommon entity that involves the body and fundus of
the stomach and is characterized by large gastric folds, reduced gastric
acid secretion, and, at times, enhanced protein loss into the stomach.
Patients who have idiopathic protein-losing enteropathy without
evidence of GI disease should be examined for cardiac disease. As more
patients with congenital heart disease reach adulthood, Fontan physiol­
ogy has become a more common cause of protein-losing enteropathy.
Other cardiac causes include right-sided valvular disease and chronic
pericarditis (Chaps. 277, 278, 279 and 281).
Treatment 
As excess protein loss into the GI tract is most often sec­
ondary to a specific disease, treatment should be directed primarily to
the underlying disease process and not to the hypoproteinemia. When
enhanced protein loss is secondary to lymphatic obstruction, it is criti­
cal to establish the nature of this obstruction. Identification of mesen­
teric nodes or lymphoma may be possible by imaging studies. Similarly,
it is important to exclude cardiac disease as a cause of protein-losing
enteropathy. Patients with congenital heart disease may be examined
by intranodal lymphangiography or noncontrast magnetic resonance
lymphangiography and may undergo embolization of the site of lym­
phatic leakage or surgical lymphatic interventions to decompress the
lymphatic system or target exclusion of abnormal lymphatic channels.
The increased protein loss that occurs in intestinal lymphangiectasia
is a result of distended lymphatics associated with lipid malabsorption.
The hypoproteinemia is treated with a low-fat, high-protein diet and
the administration of medium-chain triglycerides, which do not exit
from the intestinal epithelial cells via lymphatics but are delivered to
the body via the portal vein. Other medical therapies including octreo­
tide, a somatostatin analogue, intravenous heparin, budesonide and
other glucocorticoids, and sirolimus have been studied but have gener­
ally been ineffective. Supportive care includes repletion of nutritional
deficiencies and support stockings for leg edema control.
APPROACH TO THE PATIENT
Evaluation of the Patient with Suspected
Malabsorption
The evaluation of patients with malabsorption is often challenging
due to the large number of underlying disorders and the wide array
of available tests. Thus, an extensive history and careful physical
examination are essential to develop a more limited differential
diagnosis and thereby avoid extensive and unnecessary testing.
HISTORY
A careful history should include questions about symptoms includ­
ing abdominal pain, diarrhea, weight loss, bloating, symptoms or
signs of selective nutrient deficiency including iron-deficiency ane­
mia, bone fracture, or osteoporosis suggesting vitamin D and/or cal­
cium deficiency, peripheral neuropathy resulting from vitamin B12
deficiency, hair loss that may result from generalized protein defi­
ciency, predisposing disorders such as chronic pancreatitis or liver
disease particularly involving the bile ducts such as primary biliary
cholangitis or primary sclerosing cholangitis, history of small-bowel
resection (due to Crohn’s disease, trauma, ischemic bowel disease,
etc.), and travel history. A multitude of nonspecific symptoms such
as fatigue and weakness may also be reported. The protean mani­
festations of malabsorption and the underlying pathophysiology of
clinical manifestations are summarized in Table 336-5.
PHYSICAL EXAMINATION
A careful physical examination may provide clues to underlying
nutrient deficiencies and help assess severity of the malabsorptive
process. For example, evidence of significant weight loss may be
detected by bitemporal wasting and reduced arm circumference,
TABLE 336-5  Pathophysiology of Clinical Manifestations of
Malabsorption Disorders
SYMPTOM OR SIGN
MECHANISM
Weight loss/malnutrition
Anorexia, malabsorption of nutrients
Diarrhea
Impaired absorption or secretion of water and
electrolytes; colonic fluid secretion secondary to
unabsorbed dihydroxy bile acids and fatty acids
Flatus
Bacterial fermentation of unabsorbed
carbohydrate
Glossitis, cheilosis, stomatitis
Deficiency of iron, vitamin B12, folate, and
vitamin A
Abdominal pain
Bowel distention or inflammation, pancreatitis
Bone pain
Calcium, vitamin D malabsorption, protein
deficiency, osteoporosis
Tetany, paresthesia
Calcium and magnesium malabsorption
Weakness
Anemia, electrolyte depletion (particularly K+)
Azotemia, hypotension
Fluid and electrolyte depletion
Amenorrhea, decreased
libido
Protein depletion, decreased calories, secondary
hypopituitarism
Anemia
Impaired absorption of iron, folate, vitamin B12
Bleeding
Vitamin K malabsorption, hypoprothrombinemia
Night blindness/
xerophthalmia
Vitamin A malabsorption
Peripheral neuropathy
Vitamin B12.thiamine, vitamin E, pyridoxine, and
niacin deficiency
Dermatitis
Deficiency of vitamin A, zinc, and essential fatty
acid
CHAPTER 336
iron deficiency may cause nail spooning, and vitamin B12 deficiency
may result in significant peripheral neuropathy resulting in sensory
reduction with tingling or numbness.
Disorders of Absorption
LABORATORY EXAMINATION
Diseases that exclusively affect the proximal small intestine
(e.g., celiac disease limited to the duodenum) may result in irondeficiency anemia. Resection or disease of the terminal ileum
frequently results in B12 deficiency since B12 absorption occurs
exclusively in the ileum, causing a macrocytic anemia. Disorders
that cause steatorrhea are almost invariably associated with fatsoluble vitamin deficiency, specifically vitamin D (very common),
vitamin E, vitamin A, and vitamin K. The functional result of
vitamin K deficiency is an elevated prothrombin time/international
normalized ratio (INR), so this blood test is frequently measured
instead of vitamin K levels. Serum carotene levels can suggest fat
malabsorption but may decrease simply due to poor dietary con­
sumption of leafy vegetables.
To diagnose steatorrhea, a spot stool can be submitted for Sudan
III staining, which is specific for fecal fat. This is a useful qualitative
but not quantitative test. Stool for elastase is helpful for diagnosing
pancreatic insufficiency. A 24-h assessment of stool volume/weight
is useful to establish the presence of clinically significant absorp­
tive or secretory diarrhea versus diarrhea from other causes such
as proctitis, which causes frequent, small, low-volume stools. The
gold standard for documenting steatorrhea is the 72-h fecal fat col­
lection, which is performed in concert with the patient’s consump­
tion of a 100-g fat diet. This test is highly accurate but difficult to
obtain due to patient reluctance to collect stool. Also patients with
fat malabsorption may poorly tolerate a 100-g fat diet. A diet with
strictly quantified albeit reduced fat calories may be substituted.
Finally, the calculation of the stool osmotic gap is a very useful and
easy way to diagnose an osmotic diarrhea. A spot stool sample is
sent to the lab for quantitation of fecal sodium and potassium con­
centration. Although stool osmolality can also be measured in the
lab, measurements are often inaccurate due to bacterial degradation
of nonabsorbed carbohydrate as the stool sits prior to examination.
Because normal stool osmolality reflects serum osmolality at 290
mOsm/kg H2O, the osmotic gap may be calculated as follows:
290 – 2 (stool [Na+] + stool [K+]).
If >50–100, a stool osmotic gap is present indicating the presence
of unmeasured osmoles (e.g., malabsorbed lactose), and osmotic
diarrhea can be diagnosed. If <50, one can presume a secretory
component. Of note, malabsorbed fatty acids may also cause a
secretory diarrhea by inducing secretion in the colon, so a malab­
sorptive diarrhea may have both an osmotic and secretory compo­
nent. Extensive celiac disease may cause both osmotic diarrhea due
to malabsorbed carbohydrate and also secretory diarrhea due to
crypt hyperplasia.
Urinary D-xylose Test  The urinary d-xylose test for carbohydrate
absorption provides a measure of proximal small-bowel absorp­
tive function. d-Xylose, a pentose, is absorbed almost exclusively
in the proximal small intestine and is excreted in the urine. The
d-xylose test is usually performed by administering 25 g of d-xylose
and collecting urine for 5 h. An abnormal test (excretion of <4.5 g)
primarily reflects duodenal/jejunal mucosal disease. The ease of
obtaining a mucosal biopsy of the small intestine by endoscopy and
the false-negative rate of the d-xylose test have led to its diminished
use. When small-intestinal mucosal disease is suspected, a smallintestinal mucosal biopsy should be performed. The d-xylose test
can also be abnormal in patients with delayed gastric emptying,
impaired renal function, and sequestration in patients with large
collections of fluid in a third space (i.e., ascites, pleural fluid).
PART 10
Disorders of the Gastrointestinal System
Radiologic Examination  A small-bowel follow-through barium
examination may be very useful for detecting evidence of smallbowel diseases such as celiac disease, jejunal diverticulosis that
predisposes to small-bowel bacterial overgrowth, or Crohn’s dis­
ease (Fig. 336-3). Magnetic resonance enterography and computed
tomography enterography are commonly used for diagnosis and
management of inflammatory, stricturing disorders such as Crohn’s
disease and as an initial assessment of malabsorption, providing a
means to visualize the entire luminal GI tract as well as the hepato­
biliary tree and pancreas.
Endoscopic Evaluation and Small-Bowel Biopsies  Endoscopy
with small-bowel biopsy is essential in the evaluation of patients
with documented steatorrhea or chronic diarrhea, as well as to
evaluate abnormalities detected by radiologic imaging or by cap­
sule endoscopy. In patients with documented steatorrhea and no
evidence of pancreatic or hepatobiliary disease, an upper endoscopy
and possible small-bowel enteroscopy are required to examine the
small-bowel mucosa and to take biopsies for analysis. An upper
endoscopy will visualize the stomach and duodenum; the maxi­
mum reach of the typical upper endoscopy scope is the ligament
of Treitz. Small-bowel enteroscopy using a longer scope such as a
pediatric colonoscope can be used to visualize the jejunum. Single-
and double-balloon enteroscopy provide a means for examining
much more of the jejunum and, if successful, will reach the ileum.
Capsule endoscopy provides another means for visualizing the
entire small bowel. Colonoscopy can be used for a retrograde view
and biopsy of the terminal ileum.
Biopsy Analysis  Small-bowel pathology may be divided into the
three groups (Table 336-6) described below.
Diffuse histopathologic findings involving the entire or majority of
the mucosa that are specific for a particular disease entity; these
include agammaglobulinemia (e.g., combined variable immuno­
deficiency) and abetalipoproteinemia. Immune globulin deficiency
is associated with a variety of histopathologic findings on smallintestinal mucosal biopsy. The characteristic feature is the absence of
or substantial reduction in the number of plasma cells in the lamina
propria; the mucosal architecture may be either perfectly normal or
flat (i.e., villous atrophy). Abetalipoproteinemia is characterized by
TABLE 336-6  Diseases That Can Be Diagnosed by Small-Intestinal
Mucosal Biopsies
LESIONS
PATHOLOGIC FINDINGS
Diffuse, Specific
Agammaglobulinemia
No plasma cells; either normal or absent villi (“flat
mucosa”)
Abetalipoproteinemia
Normal villi; epithelial cells vacuolated with fat
postprandially
Patchy, Specific
Intestinal lymphoma
Malignant cells in lamina propria and submucosa
Intestinal lymphangiectasia
Dilated lymphatics; clubbed villi
Eosinophilic gastroenteritis
Eosinophil infiltration of lamina propria and
mucosa
Amyloidosis
Amyloid deposits
Crohn’s disease
Noncaseating granulomas
Infection by one or more
microorganisms (see text)
Specific organisms
Mastocytosis
Mast cell infiltration of lamina propria
Whipple’s disease
Lamina propria includes macrophages containing
material positive on periodic acid–Schiff staining;
can be diffuse or patchy.
Diffuse, Nonspecific
Celiac disease
Short or absent villi; mononuclear infiltrate;
epithelial cell damage; hypertrophy of crypts
Tropical sprue
Similar to celiac disease
Bacterial overgrowth
Patchy damage to villi; lymphocyte infiltration
Folate deficiency
Short villi; decreased mitosis in crypts;
megalocytosis
Vitamin B12 deficiency
Similar to folate deficiency
Radiation enteritis
Similar to folate deficiency
Zollinger-Ellison syndrome
Mucosal ulceration and erosion from acid
Protein-calorie malnutrition
Villous atrophy; secondary bacterial overgrowth
Drug-induced enteritis
Variable histology
a normal mucosal appearance except for the presence of mucosal
absorptive cells that contain lipid postprandially and disappear after
a prolonged period of either fat-free intake or fasting.
2.	 Patchy lesions that are specific for a disease entity include, for example,
intestinal lymphoma or intestinal lymphangiectasia. Several diseases
feature an abnormal small-intestinal mucosa with a patchy distri­
bution. As a result, biopsy samples obtained randomly or in the
absence of endoscopically visualized abnormalities may not reveal
diagnostic features. Intestinal lymphoma can at times be diagnosed
on mucosal biopsy by the identification of malignant lymphoma cells
in the lamina propria and submucosa (Chap. 113). Dilated lymphat­
ics in the submucosa and sometimes in the lamina propria indicate
lymphangiectasia associated with hypoproteinemia secondary to
protein loss into the intestine. Eosinophilic gastroenteritis comprises
a heterogeneous group of disorders with a spectrum of presentations
and symptoms, with an eosinophilic infiltrate of the lamina propria,
and with or without peripheral eosinophilia. The patchy nature of
the infiltrate and its presence in the submucosa often lead to an
absence of histopathologic findings on mucosal biopsy. Amyloid
deposition can be identified by Congo red staining in some patients
with amyloidosis involving the duodenum (Chap. 117). Whipple’s
disease exhibits PAS-positive macrophages, and immunohistochemi­
cal analysis can detect the pathogenic organism. Although in severe
cases the lesions are diffuse, patchy disease may also occur, requiring
enteroscopy and sampling of multiple mucosal sites.
3.	 Diffuse nonspecific lesions may be found in more than one disorder.
For example, villus atrophy/absence may be found in celiac disease,
tropical sprue, or bacterial overgrowth, among other disorders.
Several microorganisms can be identified in small-intestinal biopsy
samples, establishing a correct diagnosis. At times, the biopsy is

07 - 337 Inflammatory Bowel Disease
337 Inflammatory Bowel Disease
performed specifically to diagnose infection (e.g., Whipple’s disease
or giardiasis). In most other instances, the infection is detected
incidentally during the workup for diarrhea or other abdominal
symptoms. Many of these infections occur in immunocompromised
patients with diarrhea; the etiologic agents include Cryptosporidium,
Isospora belli, microsporidia, Cyclospora, Toxoplasma, cytomega­
lovirus, adenovirus, Mycobacterium avium-intracellulare, and G.
lamblia. In immunocompromised patients, when Candida, Aspergil­
lus, Cryptococcus, or Histoplasma organisms are seen on duodenal
biopsy, their presence generally reflects systemic infection. Apart
from Whipple’s disease and infections in the immunocompromised
host, small-bowel biopsy is seldom used as the primary mode of
diagnosis of infection. Even giardiasis is more easily diagnosed by
stool antigen studies and/or duodenal aspiration than by duodenal
biopsy.
SUMMARY
The evaluation and management of patients with disorders of absorp­
tion are challenging due to the complexity of the underlying patho­
physiology and the large number of associated diseases. A diagnostic
approach based on the information summarized in Tables 336-1 and
336-5 should prove useful for guiding the care of these challenging
patients.
Acknowledgments
Henry Binder wrote this chapter in prior editions and some material from
his chapter has been retained.
■
■FURTHER READING
Bering J, DiBaise JK: Short bowel syndrome: Complications and
management. Nutr Clin Pract 38:S46, 2023.
Boumaza A et al: Whipple’s disease and Tropheryma whipplei infec­
tions: From bench to bedside. Lancet Infect Dis 22:e280, 2022.
Bousaba J et al: Bile acid diarrhea: As bad as it gets? Curr Opin
Gastroenterol 39:184, 2023.
Cowardin CA et al: Environmental enteric dysfunction: Gut and
microbiota adaptation in pregnancy and infancy. Nat Rev Gastroenterol
Hepatol 20:223, 2023.
Fernández-Bañares F et al: Carbohydrate maldigestion and intolerance.
Nutrients 14:1923, 2022.
Gregory M, Rubin DC: Small bowel disorders, in The Washington
Manual Gastroenterology Subspecialty Consult, 4th ed, Gyawali CP
et al (eds). Walters Kluwer, 2021, pp.142–155.
Jansson-Knodell CL, Rubio-Tapia A: Gluten-related disorders from
bench to bedside. Clin Gastroenterol Hepatol 22:693, 2024.
Johnson LR: Digestion and absorption of nutrients, in Gastrointestinal
Physiology, 9th ed. Philadelphia, Elsevier, 2019, pp 102–120.
Ozen A, Lenardo MJ: Protein-losing enteropathy. N Engl J Med
389:733, 2023.
Zafar H et al: Small intestinal bacterial overgrowth: current update.
Curr Opin Gastroenterol 39:522, 2023.
Sonia Friedman, Richard S. Blumberg
Inflammatory Bowel
Disease
Inflammatory bowel disease (IBD) is a chronic idiopathic inflammatory
disease of the gastrointestinal tract. Ulcerative colitis (UC) and Crohn’s
disease (CD) are the two major types of IBD.
■
■GLOBAL CONSIDERATIONS: EPIDEMIOLOGY
UC and CD have emerged as global diseases in the twenty-first century.
The disease burden is high, with a prevalence of >0.3% in North America,
Oceania, and most countries in Europe. In newly industrialized
countries in Africa, Asia, and South America where there is increased
urbanization and Westernization, the incidence of IBD has been ris­
ing and mirrors the prior increase of IBD in the Western world in the
twentieth century. For example, in Brazil, the annual percent change is
+11.1% (95% confidence interval [CI], 4.8–17.8%) for CD and +14.9%
(95% CI, 10.4–19.6%) for UC, whereas in Taiwan, the annual percent
change is +4.0% (95% CI, 1.0–7.1%) for CD and +4.8% (95% CI,
1.8–8.0%) for UC. In a study of newly diagnosed IBD cases between
2011 and 2013 from 13 countries or regions in the Asia Pacific, the
mean annual IBD incidence per 100,000 was 1.50 (95% CI, 1.43–1.57).
India (9.31; 95% CI, 8.38–10.31) and China (3.64; 95% CI, 2.97–4.42)
had the highest IBD incidences in Asia. The highest reported preva­
lence values were in Europe (UC, 505 per 100,000 in Norway; CD, 322
per 100,000 in Germany) and North America (UC, 286 per 100,000 in
the United States; CD, 319 per 100,000 in Canada). The most recent
U.S. study showed that 2.39 million Americans are diagnosed with
IBD. The most likely factors that explain the geographic variability
of IBD rates, especially the rising incidence in developing countries
and urban areas, are environmental variables including changes in
diet (with downstream effects on the intestinal microbiota), exposure
to sunlight or temperature differences, and socioeconomic status and
hygiene (Table 337-1).
Peak incidence of UC and CD is in the second to fourth decades,
with 78% of CD studies and 51% of UC studies reporting the highest
incidence among those aged 20–29 years old. A second modest rise in
incidence occurs between the seventh and ninth decades of life. The
female-to-male ratio ranges from 0.51 to 1.58 for UC studies and 0.34
to 1.65 for CD studies, suggesting that the diagnosis of IBD is not gen­
der-specific. Pediatric IBD (patients <17 years old) composes ~20–25%
of all IBD patients, and ~5% of all IBD patients are <10 years old.
Children with IBD are also grouped as those with early-onset (EO) IBD
(patients <10 years old), very-early-onset (VEO) IBD (patients <6 years
old), and infantile IBD (patients <2 years old). VEOIBD and infantile
IBD mainly affect the colon and are resistant to standard medications,
and patients often have a strong family history of IBD, with at least one
first-degree relative affected. In infantile IBD or VEOIBD, a number of
rare, single genetic mutations have been identified as the basis for this
susceptibility in up to 10% of patients, suggesting a simple Mendelian
origin of the disease in these cases.
CHAPTER 337
Inflammatory Bowel Disease
Although the greatest prevalence of IBD is among Whites, the preva­
lence of IBD in Latinx, Black, and Asian people is increasing. Urban areas
have a higher prevalence of IBD than rural areas, and high socioeconomic
classes have a higher prevalence than lower socioeconomic classes.
Epidemiologic studies have identified a number of potential envi­
ronmental factors that are associated with disease risk (Fig. 337-1).
TABLE 337-1  Epidemiology of IBD
 
ULCERATIVE COLITIS
CROHN’S DISEASE
Age of onset
Second to fourth decades
and seventh to ninth
decades
Second to fourth decades
and seventh to ninth
decades
Ethnicity
White > Black > Latinx > Asian
Female-to-male ratio
0.51–1.58
0.34–1.65
Smoking
May prevent disease
(odds ratio 0.58)
May cause disease
(odds ratio 1.76)
Oral contraceptives
No increased risk
Hazard ratio 2.82
Appendectomy
Protective (risk reduction
13–26%)
Not protective
Monozygotic twins
6–18% concordance
38–58% concordance
Dizygotic twins
0–2% concordance
4% concordance
Infections in the first
year of life
1.6 and 3 times the risk of developing IBD by age 10
and 20 years
Abbreviation: IBD, inflammatory bowel disease.
Genetic susceptibility
TLR4
XBP1
DLG5
ECM1
ITLN1
SLC22A5
DMBT1
PTGER4
XBP1
NOD2
ATG16L1
Microbial flora
Enteropathogens
Antibiotics
Diet, hygiene
NSAIDs, smoking
PART 10
Disorders of the Gastrointestinal System
Environmental factors
FIGURE 337-1  Pathogenesis of inflammatory bowel disease (IBD). In IBD, the tridirectional relationship between the commensal flora (microbiota), intestinal epithelial
cells (IECs), and mucosal immune system is dysregulated, leading to chronic inflammation. Each of these three factors is affected by genetic and environmental factors
that determine risk for the disease. NSAIDs, nonsteroidal anti-inflammatory drugs. (Republished with permission Annual Review of Immunology from Inflammatory Bowel
Disease, A Kaser et al: 28:573, 2010. Permission conveyed through Copyright Clearance Center, Inc.)
Smoking is an important risk factor in IBD with opposite effects on
UC (odds ratio [OR] 0.58) and CD (OR 1.76) that may be influenced
by genetic risk factors and ethnic origin. Previous appendectomy with
confirmed appendicitis (risk reduction of 13–26%), particularly at a
young age, has a protective effect on the development of UC across dif­
ferent geographical regions and populations. Appendectomy is mod­
estly associated with the development of CD, but this may be due to
diagnostic bias. Oral contraceptive use is associated with an increased
risk of CD, with a reported hazard ratio as high as 2.82 among current
users and 1.39 among past users. The association between oral con­
traceptive use and UC is limited to women with a history of smoking.
Breast-feeding may protect against the development of IBD, which
supports the potential importance of early-life exposures to later IBD
development. Infections in the first year of life are associated with
development of IBD, especially before the ages of 10 and 20 years.
Infectious gastroenteritis with pathogens (e.g., Salmonella, Shigella,
Campylobacter spp., Clostridioides difficile) increases IBD risk by two-
to threefold. Diets high in animal protein, sugars, sweets, oils, fish and
shellfish, and dietary fat, especially ω-6 fatty acids, and low in ω-3 fatty
acids have been implicated in increasing the risk of IBD. A protective
effect of vitamin D on the risk of CD has been reported.
IBD is a familial disease in 5–10% of patients (Fig. 337-2), and the
strongest risk factor for the development of IBD is a first-degree rela­
tive with the disease. The children of mothers and fathers with UC have
an approximately fourfold increased risk of UC, and the children of
mothers and fathers with CD have an almost eightfold increased risk of
CD. Some of these patients may exhibit early-onset disease during the
first decade of life and, in CD, a concordance of anatomic site and clini­
cal type within families. In twin studies, 38–58% of monozygotic twins
are concordant for CD, and 6–18% are concordant for UC, whereas 4%
IL23R, IL12B, JAK2, STAT3, CCR6,
NOD2, TLR4, CARD9, IRF5,
ATG16L1, IRGM, LRRK2
TNFSF15, TNFRSF6B
TNFAIP3, PTPN2/22
NLRP3, IL18RAP
ICOSL, ARPC2, STAT3, IL10
Immune dysregulation
IEC
Stress
Diet, hygiene
of dizygotic twins are concordant for CD, and 0–2% are concordant for
UC in Swedish and Danish cohorts. In the remainder of patients, IBD
is observed in the absence of a family history (i.e., sporadic disease).
GLOBAL CONSIDERATIONS:
IBD PHENOTYPES
IBD location and behavior show racial differences that may reflect
underlying genetic variations and have important implications for
diagnosis and management of disease. Blacks and Latinxs tend to
have an ileocolonic CD distribution. Data from East Asia show that
Monogenic
Oligogenic
Polygenic
Environment
Undiagnosed
infections?
Early onset
Genetics
Familial
(10%)
Sporadic
FIGURE 337-2  A model for the syndromic nature of inflammatory bowel disease
(IBD). Genetic and environmental factors variably influence the development and
phenotypic manifestations of IBD. At the one extreme, IBD is exemplified as a simple
Mendelian disorder as observed in early-onset IBD due to single-gene defects such
as IL10, IL10RA, and IL10RB; and at the other extreme, it may be exemplified by as yet
to be described emerging infectious diseases. (Reproduced with permission from A
Kaser et al: Genes and environment: How will our concepts on the pathophysiology
of IBD develop in the future? Dig Dis 28:395, 2010.)
ileocolonic CD is the most common CD phenotype (50.5–71%) and
perianal disease is more common in East Asian patients (30.3–58.8%)
than Whites (25.1–29.6%). Pancolonic disease is more common than
left-sided colitis or proctitis among Black, Latinx, and Asian patients
with UC. Older Asian patients with UC (age >60) tend to have a more
aggressive disease course. Among Blacks, joint involvement is the
predominant extraintestinal manifestation (EIM) reported and ranges
from 15.7 to 29.6%. Ocular involvement is also common in African
Americans and ranges from 7.1 to 13%. Dermatologic manifestations
are the most common EIMs reported in Latinxs (10–13%). These
ethnic variations indicate the importance of different genetic and/or
environmental factors in the pathogenesis of this disorder.
ETIOLOGY AND PATHOGENESIS
Under physiologic conditions, homeostasis normally exists between
the commensal microbiota, epithelial cells that line the interior of the
intestines (intestinal epithelial cells [IECs]), and immune cells within
the tissues (Fig. 337-1). A consensus hypothesis is that each of these
three major host compartments that function together as an integrated
“supraorganism” (microbiota, IECs, and immune cells) are affected
by specific environmental (e.g., smoking, antibiotics, enteropatho­
gens) and genetic factors that, in a susceptible host, cumulatively and
interactively disrupt homeostasis during the course of one’s life and, in
so doing, culminate in a chronic state of dysregulated inflammation;
i.e., IBD. Although chronic activation of the mucosal immune system
may represent an appropriate response to an infectious agent, a search
for such an agent has thus far been unrewarding in IBD. As such, IBD is
currently considered an inappropriate immune response to the endog­
enous (autochthonous) commensal microbiota within the intestines,
with or without some component of autoimmunity. Importantly, the
normal, uninflamed intestines contain a large number of immune cells
that are in a unique state of activation, in which the gut is restrained
from full immunologic responses to the commensal microbiota and
dietary antigens by very powerful regulatory pathways that function
within the immune system (e.g., T regulatory cells that express the
FoxP3 transcription factor and suppress inflammation). Maintenance
of homeostasis also involves oversight from local parenchymal cells
including nerve, endothelial, and stromal cells, as well as the com­
mensal microbiota that provide essential remedial factors necessary for
health and serve as a target of the immune response. During the course
of infections or other environmental stimuli in the normal host, full
activation of the lymphoid tissues in the intestines occurs but is rapidly
superseded by dampening of the immune response and tissue repair. In
IBD, such processes may not be regulated normally.
GENETIC CONSIDERATIONS
The genetic underpinning of IBD is known from its concordance
in identical twins, its occurrence in the context of several genetic
syndromes, and the development of severe, refractory IBD in
early life in association with single-gene defects that affect the immune
system (Table 337-2). More than 60 different gene defects have been
identified in patients with VEOIBD and infantile IBD by whole-exome
sequencing (WES), in whom most of the monogenic mutations have
been discovered. These include mutations in genes encoding, for
example, interleukin (IL) 10, the IL-10 receptor (IL-10R), cytotoxic
T-lymphocyte-associated protein-4 (CTLA4), neutrophil cytosolic fac­
tor 2 protein (NCF2), X-linked inhibitor of apoptosis protein (XIAP),
lipopolysaccharide responsive and beige-like anchor protein (LRBA),
and tetratricopeptide repeat domain 7A protein (TTC7), among many
other genes that are involved in host-commensal interactions. A mono­
genic etiology may also be possible in a small subset of adult patients
with IBD. In addition, IBD has a familial origin in at least 10% of
afflicted adults, consistent with an inherited basis for this disease (Fig.
337-2). However, the majority of pediatric and adult IBD cases are
multigenic (or polygenic) in origin, suggesting a syndromic nature of
this disease that gives rise to multiple clinical subgroups beyond the
simple classification as UC and CD. The polygenic nature of the disease
has been elucidated through a variety of genetic approaches, including
candidate gene studies, linkage analysis, and genome-wide association
TABLE 337-2  Primary Genetic Disorders Associated with IBD
NAME
GENETIC ASSOCIATION
PHENOTYPE
Turner’s syndrome
Loss of part or all of X
chromosome
Associated with UC and
colonic CD
Hermansky-Pudlak
syndrome
Autosomal recessive
disorder genes involved in
the biogenesis of lysosomerelated organelles or
adaptor protein-3 complex
Granulomatous colitis,
oculocutaneous albinism,
platelet dysfunction, and
pulmonary fibrosis
Wiskott-Aldrich
syndrome (WAS)
X-linked recessive disorder,
loss of WAS protein
function
Colitis, immunodeficiency,
severely dysfunctional
platelets, and
thrombocytopenia
Glycogen storage
disease type 1b
Autosomal recessive
disorder of SLC37A4
resulting in deficiency of
the glucose-6-phosphate
translocase
Granulomatous colitis,
presents in infancy with
hypoglycemia, growth
failure, hepatomegaly, and
neutropenia
Immune dysregulation
polyendocrinopathy,
enteropathy X-linked
(IPEX)
Loss of FoxP3 transcription
factor and T regulatory cell
function
UC-like autoimmune
enteropathy, with
endocrinopathy (neonatal
type 1 diabetes or
thyroiditis), dermatitis
Early-onset IBD
Deficient IL-10 and IL-10
receptor function
Severe, refractory IBD in
early life
Abbreviations: CD, Crohn’s disease; IBD, inflammatory bowel disease; IL, interleukin;
UC, ulcerative colitis.
CHAPTER 337
studies (GWAS) that focus on the identification of disease-associated
single nucleotide polymorphisms (SNPs) within the human genome
and WES and whole-genome sequencing to elucidate the specific
mutations potentially involved. GWAS have identified ~240 genetic
loci with multiple potential candidate genes; two-thirds of these loci are
associated with both disease phenotypes, with the remainder being
specific for either CD or UC (Table 337-3). These genetic similarities
account for the overlapping immunopathogenesis and consequently
epidemiologic observations of both diseases in the same families and
similarities in response to therapies. Because the specific causal vari­
ants for each identified gene or locus are mostly unknown as most risk
loci are contained within regulatory (noncoding) regions of the associ­
ated genes, it is not clear whether the similarities in the genetic risk
factors associated with CD and UC are shared at a structural or func­
tional level. The risk conferred by each identified gene or locus is
unequal and generally small, such that only ~20% of the disease risk is
considered to be explained by the current genetic information. Further,
many of the genetic risk factors identified are also observed to be asso­
ciated with risk for other immune-mediated diseases, suggesting that
related immunogenetic pathways are involved in the pathogenesis of
multiple different disorders, accounting for the common responsive­
ness to similar types of biologic therapies (e.g., anti–tumor necrosis
factor [TNF] therapies) and possibly the simultaneous occurrence of
these disorders. The diseases and the genetic risk factors that are shared
with IBD include, for example, rheumatoid arthritis (TNFAIP3), pso­
riasis (IL23R, IL12B), ankylosing spondylitis (IL23R), type 1 diabetes
mellitus (IL10, PTPN2), asthma (ORMDL3), and systemic lupus ery­
thematosus (TNFAIP3, IL10), among others.
Inflammatory Bowel Disease
The genetic factors that are recognized to mediate risk for IBD
have highlighted the importance of shared mechanisms of disease that
variably affect CD and/or UC (Table 337-3). These include the fol­
lowing: those genes that are associated with fundamental cell biologic
processes such as the unfolded protein response due to endoplasmic
reticulum stress, autophagy, and metabolism that regulate the ability
of cells to manage the physiologic needs of the intestinal environment;
those associated with innate immunity associated with nonlymphoid
cells that function in responses to and control of microbes; those
associated with the regulation of adaptive immunity that control the
balance between inflammatory and anti-inflammatory cellular path­
ways associated with lymphocytes; and, finally, those that are involved
in the development and resolution of inflammation associated with
PART 10
Disorders of the Gastrointestinal System
TABLE 337-3  Some Genetic Loci Associated with Crohn’s Disease and/or Ulcerative Colitis
CHROMOSOME
PUTATIVE
GENE
GENE NAME
PROTEIN FUNCTION
CD
UC
Unfolded Protein Response, Autophagy, and Metabolism
2q37
ATG16L1
ATG16 autophagy related 16-like 1
Autophagy
+
5q31
SLC22A5
Solute carrier family 22, member 5
b-Carnitine transporter
+
5q33
IRGM
Immunity-related GTPase family, M
Autophagy
+
7p21
AGR2
Anterior gradient 2
Unfolded protein response
+
+
12q12
LRRK2
Leucine-rich repeat kinase 2
Autophagy
+
13q14
LACC1
Laccase domain containing 1 (FAMIN)
Immunometabolic regulator
+
17q21
ORMDL3
Orosomucoid related member 1-like 3
Unfolded protein response and lipid synthesis
+
+
22q12
XBP1
X-box binding protein 1
Unfolded protein response
+
+
Innate Immunity
1q23
ITLN1
Intelectin 1
Bacterial binding
+
16q12
NOD2
Nucleotide-binding oligomerization domain containing 2
Bacterial sensing and autophagy activation
+
Adaptive Immunity
1p31
IL23R
Interleukin 23 receptor
TH17 cell stimulation
+
+
1q32
IL10
Interleukin 10
Treg-associated cytokine
+
5q33
IL12B
Interleukin 12 subunit beta
IL-12 p40 chain of IL-12/IL-23
+
+
18p11
PTPN2
Protein tyrosine phosphatase, nonreceptor type 2
T-cell regulation
+
Inflammation and Healing
3p21
MST1
Macrophage Stimulating 1
Macrophage activation
+
+
5p13
PTGER4
Prostaglandin E receptor 4
PGE2 receptor
+
+
6q23
TNFAIP3
Tumor necrosis factor, alpha-induced protein 3 (A20)
Toll-like receptor regulation
+
6q27
CCR6
Chemokine (C-C motif) receptor 6
Dendritic cell migration
+
9p24
JAK2
Janus kinase 2
IL-6R and IL-23R signaling
+
+
9q32
TNFSF15
Tumor necrosis factor–like cytokine 1A (TL1A)
Promotes inflammation and fibrosis
+
+
17q21
STAT3
Signal transducer and activator of transcription 3
IL-6R, IL-23R, and IL-10R signaling
+
+
Abbreviations: CD, Crohn’s disease; GTPase, guanosine triphosphatase; IL, interleukin; PGE2, prostaglandin E2; Treg, T regulatory cell; UC, ulcerative colitis.
Source: Adapted from A Kaser et al: Ann Rev Immunol 28:573, 2010; Graham DB, Xavier RJ: Nature 578:527, 2020.
healing that control leukocyte recruitment and inflammatory media­
tor production or the development of fibrosis. Each of these genetic
susceptibilities contributes in an incremental manner to IBD risk,
variably affects the activities of virtually all subtypes of immune and
nonimmune cells within the intestines, and encodes mutations (poly­
morphisms) that promote or protect from IBD. Some of these loci are
associated with specific subtypes of disease such as the association
between NOD2 polymorphisms and fibrostenosing CD or ATG16L1
and fistulizing disease, especially within the ileum. However, the clini­
cal utility of these genetic risk factors for the diagnosis or determina­
tion of prognosis and therapeutic responses remains to be defined.
■
■COMMENSAL MICROBIOTA AND IBD
The endogenous commensal microbiota within the intestines plays a
central role in the pathogenesis of IBD. Humans are born with sterile
guts and acquire their commensal microbiota initially from the mother
during egress through the birth canal and subsequently from environ­
mental sources. A stable configuration of up to 1000 species of bacteria
that achieves a biomass of ~1012 colony-forming units per gram of feces
is achieved by 3 years of age, which likely persists into adult life, with
each individual human possessing a unique combination of species.
In addition, the intestines contain other microbial life forms including
fungi, archaea, viruses, and protists. The microbiota is thus considered
as a critical and sustaining component of the human organism. The
establishment and maintenance of the intestinal microbiota compo­
sition and function are under the control of host (e.g., immune and
epithelial responses), environmental (e.g., diet and antibiotics), and
likely genetic (e.g., NOD2) factors (Fig. 337-1). In turn, the microbiota,
through its structural components and metabolic activity, has major
influences on the epithelial and immune function of the host, which,
through epigenetic effects, may have durable consequences. During
early life when the commensal microbiota is being established, these
microbial effects on the host may be particularly important in deter­
mining later life risk for IBD. Specific components of the microbiota
can promote or protect from disease. The commensal microbiota in
patients with both UC and CD is demonstrably different from that of
nonafflicted individuals, a state of dysbiosis suggesting the presence
of microorganisms that drive disease (e.g., Proteobacteria such as
enteroinvasive and adherent Escherichia coli) and to which the immune
response is directed and/or the loss of microorganisms that hinder
inflammation (e.g., Firmicutes such as Faecalibacterium prausnitzii).
Many of the changes in the commensal microbiota occur as a conse­
quence of the inflammation and are thus potential secondary drivers
of disease. In addition, agents that alter the intestinal microbiota such
as metronidazole, ciprofloxacin, and elemental diets, may improve CD.
CD may also respond to fecal diversion, demonstrating the ability of
luminal contents to exacerbate disease.
■
■DEFECTIVE IMMUNE REGULATION IN IBD
The mucosal immune system does not normally elicit an inflammatory
immune response to luminal contents due to oral (mucosal) tolerance.
Administration of soluble antigens orally, rather than subcutaneously
or intramuscularly, leads to antigen-specific control of the response
and the host’s ability to tolerate the antigen. Multiple mechanisms are
involved in the induction of oral tolerance and include deletion or
anergy (nonresponsiveness) of antigen-reactive T cells or induction
of CD4+ T cells that suppress gut inflammation (e.g., T regulatory
cells expressing the FoxP3 transcription factor) and that secrete antiinflammatory cytokines such as IL-10, IL-35, and transforming growth
factor β (TGF-β). Oral tolerance may be responsible for the lack of
immune responsiveness to dietary antigens and the commensal micro­
biota in the intestinal lumen. In IBD, this suppression of inflammation
is altered, leading to uncontrolled inflammation. The mechanisms of
this regulated immune suppression are incompletely known.
Inflammatory Bowel Disease
CHAPTER 337
Gene knockout (–/–) or transgenic (Tg) mouse models of IBD, includ­
ing those that are directed at genes associated with risk for the human
disease, have revealed that deleting specific cytokines (e.g., IL-2, IL-10,
TGF-β) or their receptors, deleting molecules associated with T-cell
antigen recognition (e.g., T-cell antigen receptors), or interfering with
IEC barrier function and the regulation of responses to commensal
bacteria (e.g., mucus glycoproteins or nuclear factor-κB [NF-κB]) leads
to spontaneous colitis or enteritis. In the majority of circumstances,
intestinal inflammation in these animal models requires the presence
of the commensal microbiota. However, in some cases, activation of
certain elements of the intestinal immune system may be exacerbated
by the absence of bacteria, resulting in severe colitis and emphasizing
the presence of protective properties of the commensal microbiota.
Thus, a variety of specific alterations in either the microbiota or host
can lead to uncontrolled immune activation and inflammation directed
at the intestines in mice. How these relate to human IBD remains to
be defined, but they are consistent with inappropriate responses of the
genetically susceptible host to the commensal microbiota.
■
■THE INFLAMMATORY CASCADE IN IBD
In both UC and CD, inflammation likely emerges from the genetic
predisposition of the host in the context of yet-to-be-defined envi­
ronmental factors. Once initiated in IBD by abnormal innate immune
sensing of bacteria by parenchymal cells (e.g., IECs) and hematopoietic
cells (e.g., macrophages, dendritic cells), the immune inflammatory
response is perpetuated by T-cell and B-cell activation when coupled
together with inadequate regulatory pathways. A sequential cascade
of inflammatory mediators extends the response, making each step
a potential target for therapy. Inflammatory cytokines from innate
immune cells such as IL-1, IL-6, IL-12, IL-23, and TNF have diverse
effects on tissues. They promote fibrogenesis, collagen production,
activation of tissue metalloproteinases, and/or the production of other
inflammatory mediators; they also activate the coagulation cascade in
local blood vessels (e.g., increased production of von Willebrand fac­
tor). These cytokines are normally produced in response to infection
but are usually turned off or inhibited by cytokines such as IL-10 and
TGF-β at the appropriate time to limit tissue damage. In IBD, their
activity is not regulated, resulting in an imbalance between the proin­
flammatory and anti-inflammatory mediators. Some of these cytokines
activate other inflammatory cells (macrophages and B cells), and others
such as chemokines act indirectly to recruit other lymphocytes, inflam­
matory leukocytes, and mononuclear cells from the bloodstream into
the gut through interactions between homing receptors on leukocytes
(e.g., α4β7 integrin) and addressins on vascular endothelium (e.g.,
MadCAM1). Lymphocytes such as CD4+ T helper (TH) cells emerge
from the lymph nodes under the influence of sphingosine-1-phosphate
(S1P) gradients that act on S1P receptors (S1PR) expressed on the lym­
phocyte and endothelium. CD4+ TH cells that promote inflammation
are of four major types, all of which may be associated with colitis in
animal models and perhaps humans: TH1 cells (secrete IL-2, interferon
[IFN] γ), TH2 cells (secrete IL-4, IL-5, IL-13), TH9 cells (secrete IL-9),
and TH17 cells (secrete IL-17, IL-21, IL-22). TH17 cells may also provide
protective functions. Innate immune-like cells (ILCs) that lack T-cell
receptors are also present in intestines, polarize to the same functional
fates, and may similarly participate in IBD. TH1 cells induce transmu­
ral granulomatous inflammation that resembles CD; TH2 cells and
related natural killer T cells that secrete IL-4, IL-5, and IL-13 induce
superficial mucosal inflammation resembling UC in animal models;
TH9 cells promote allergic-like inflammation; and TH17 cells may be
responsible for neutrophilic recruitment. Each of these T-cell subsets
cross-regulates each other. The TH1 cytokine pathway is initiated by
IL-12, a key cytokine in the pathogenesis of experimental models of
mucosal inflammation. IL-4 and IL-23, together with IL-6 and TGF-β,
induce TH2 and TH17 cells, respectively, and IL-23 inhibits the suppres­
sive function of regulatory T cells. Activated macrophages secrete TNF
and IL-6. Cytokines produced by innate immune cells and lymphocytes
exert their effects on target cells through cytokine-specific receptors,
which precisely engage intracellular Janus kinases (JAK) 1, 2, and/or 3
for transmission of intracellular activating signals.
These characteristics of the immune response in IBD explain the
beneficial therapeutic effects of antibodies to block proinflammatory
cytokines or the signaling by their receptors (e.g., anti-TNF, anti-IL-12,
anti-IL-23, JAK inhibitors), molecules that antagonize the activity
of S1PR-induced emigration of lymphocytes from the lymph nodes
(ozanimod), or antibodies that impede leukocyte recruitment into the
intestines (e.g., anti-α4β7). They also highlight the potential usefulness
of cytokines that inhibit inflammation and promote regulatory T cells
or promote intestinal barrier function (e.g., IL-10) in the treatment of
IBD. Therapies such as the 5-aminosalicylic acid (5-ASA) compounds
and glucocorticoids are also potent inhibitors of these inflammatory
mediators through inhibition of transcription factors such as NF-κB
that regulate their expression.
PATHOLOGY
■
■ULCERATIVE COLITIS: MACROSCOPIC FEATURES
UC is a mucosal disease that usually involves the rectum and extends
proximally to involve all or part of the colon. About 40–50% of patients
have disease limited to the rectum and rectosigmoid, 30–40% have dis­
ease extending beyond the sigmoid but not involving the whole colon,
and 20% have a pancolitis. Proximal spread occurs in continuity with­
out areas of uninvolved mucosa. When the whole colon is involved,
the inflammation extends 2–3 cm into the terminal ileum in 10–20%
of patients. The endoscopic changes of backwash ileitis are superficial
and mild and are of little clinical significance. Although variations in
macroscopic activity may suggest skip areas, biopsies from normalappearing mucosa are usually abnormal. Thus, it is important to obtain
multiple biopsies from apparently uninvolved mucosa, whether proxi­
mal or distal, during endoscopy. One caveat is that effective medical
therapy can change the appearance of the mucosa such that either skip
areas or the entire colon can be microscopically normal.
With mild inflammation, the mucosa is erythematous and has a
fine granular surface that resembles sandpaper. In more severe disease,
the mucosa is hemorrhagic, edematous, and ulcerated (Fig. 337-3). In
long-standing disease, inflammatory polyps (pseudopolyps) may be
present as a result of epithelial regeneration. The mucosa may appear
normal in remission, but in patients with many years of disease, it
appears atrophic and featureless, and the entire colon becomes nar­
rowed and shortened. Patients with fulminant disease can develop a
toxic colitis or megacolon where the bowel wall becomes thin and the
mucosa is severely ulcerated; this may lead to perforation.
■
■ULCERATIVE COLITIS: MICROSCOPIC FEATURES
Histologic findings correlate well with the endoscopic appearance
and clinical course of UC. The process is limited to the mucosa and
FIGURE 337-3  Ulcerative colitis. Diffuse (nonsegmental) mucosal disease,
with broad areas of ulceration. The bowel wall is not thickened, and there is no
cobblestoning. (Courtesy of Dr. R. Odze, Division of Gastrointestinal Pathology,
Department of Pathology, Brigham and Women’s Hospital, Boston, Massachusetts;
with permission.)
FIGURE 337-4  Medium-power view of colonic mucosa in ulcerative colitis
showing diffuse mixed inflammation, basal lymphoplasmacytosis, crypt atrophy and
irregularity, and superficial erosion. These features are typical of chronic active
ulcerative colitis. (Courtesy of Dr. R. Odze, Division of Gastrointestinal Pathology,
Department of Pathology, Brigham and Women’s Hospital, Boston, Massachusetts;
with permission.)
superficial submucosa, with deeper layers unaffected except in fulmi­
nant disease. In UC, two major histologic features suggest chronicity
and help distinguish it from infectious or acute self-limited colitis.
First, the crypt architecture of the colon is distorted; crypts may be
bifid and reduced in number, often with a gap between the crypt bases
and the muscularis mucosae. Second, some patients have basal plasma
cells and multiple basal lymphoid aggregates. Mucosal vascular con­
gestion, with edema and focal hemorrhage, and an inflammatory cell
infiltrate of neutrophils, lymphocytes, plasma cells, and macrophages
may be present. The neutrophils invade the epithelium, usually in
the crypts, giving rise to cryptitis and, ultimately, to crypt abscesses
(Fig. 337-4). Ileal changes in patients with backwash ileitis include
villous atrophy and crypt regeneration with increased inflammation,
increased neutrophil and mononuclear inflammation in the lamina
propria, and patchy cryptitis and crypt abscesses.
PART 10
Disorders of the Gastrointestinal System
■
■CROHN’S DISEASE: MACROSCOPIC FEATURES
CD can affect any part of the gastrointestinal (GI) tract from the mouth
to the anus. Some 30–40% of patients have small-bowel disease alone,
40–55% have disease involving both the small and large intestines, and
15–25% have colitis alone. In the 75% of patients with small-intestinal
disease, the terminal ileum is involved in 90%. Unlike UC, which
almost always involves the rectum, the rectum is often spared in CD.
CD is often segmental with skip areas throughout the diseased intes­
tine (Fig. 337-5). Perianal disease, manifesting as perirectal fistulas,
fissures, abscesses, and anal stenosis, is present in one-third of patients
with CD, particularly those with colonic involvement. Rarely, CD may
also involve the liver and the pancreas.
Unlike UC, CD is a transmural process. Endoscopically, aphthous or
small superficial ulcerations characterize mild disease; in more active dis­
ease, stellate ulcerations fuse longitudinally and transversely to demarcate
islands of mucosa that frequently are histologically normal. This “cobble­
stone” appearance is characteristic of CD, both endoscopically and by
barium radiography. As in UC, pseudopolyps can form in CD.
Active CD is characterized by focal inflammation and formation
of fistula tracts, which resolve by fibrosis and stricturing of the bowel.
The bowel wall thickens and becomes narrowed and fibrotic, leading
to chronic, recurrent bowel obstructions. Projections of thickened
mesentery known as “creeping fat” encase the bowel, and serosal and
mesenteric inflammation promotes adhesions and fistula formation.
■
■CROHN’S DISEASE: MICROSCOPIC FEATURES
The earliest lesions are aphthoid ulcerations and focal crypt abscesses
with loose aggregations of macrophages, which form noncaseating
FIGURE 337-5  Crohn’s disease of the colon showing thickening of the wall, with
stenosis, linear serpiginous ulcers, and cobblestoning of the mucosa. (Courtesy
of Dr. R. Odze, Division of Gastrointestinal Pathology, Department of Pathology,
Brigham and Women’s Hospital, Boston, Massachusetts; with permission.)
granulomas in all layers of the bowel wall (Fig. 337-6). Granulomas are
a characteristic feature of CD and are less commonly found on muco­
sal biopsies than on surgical resection specimens. Other histologic
features of CD include submucosal or subserosal lymphoid aggregates,
particularly away from areas of ulceration, gross and microscopic skip
areas, and transmural inflammation that is accompanied by fissures
that penetrate deeply into the bowel wall and sometimes form fistulous
tracts or local abscesses.
CLINICAL PRESENTATION
■
■ULCERATIVE COLITIS
Signs and Symptoms 
The major symptoms of UC are diarrhea,
rectal bleeding, tenesmus, passage of mucus, and crampy abdominal
pain. The severity of symptoms correlates with the extent of disease.
Although UC can present acutely, symptoms usually have been present
for weeks to months.
Patients with proctitis usually pass fresh blood or blood-stained
mucus, either mixed with stool or streaked onto the surface of a normal
or hard stool. They also have tenesmus, or urgency with a feeling of
incomplete evacuation, but rarely have abdominal pain. With proctitis
FIGURE 337-6  Medium-power view of Crohn’s colitis showing mixed acute and
chronic inflammation, crypt atrophy, and multiple small epithelioid granulomas
in the mucosa. (Courtesy of Dr. R. Odze, Division of Gastrointestinal Pathology,
Department of Pathology, Brigham and Women’s Hospital, Boston, Massachusetts;
with permission.)
TABLE 337-4  Montreal Classification of Extent and Severity of
Ulcerative Colitis (UC)
EXTENT
ANATOMY
E1: Ulcerative proctitis
Involvement limited to the rectum
E2: Left-sided UC (distal UC)
Involvement limited to the colorectum distal to
the splenic flexure
E3: Extensive UC (pancolitis)
Involvement extends proximal to the splenic
flexure
SEVERITY
DEFINITION
S0: Clinical remission
Absence of symptoms
S1: Mild disease activity
≤4 stools/d (with or without blood), absence of
systemic illness, normal inflammatory markers
(ESR)
S2: Moderate disease activity ≥4 stools/d but minimal signs of systemic toxicity
S3: Severe disease activity
≥6 bloody stools/d, pulse ≥90 beats/min,
temperature ≥37.5°C, hemoglobin <10.5 g/100 mL,
and ESR ≥30 mm/h
Abbreviation: ESR, erythrocyte sedimentation rate.
Source: C Gasche et al: A simple classification of Crohn’s disease: Report of
the Working Party for the World Congresses of Gastroenterology, Vienna 1998.
Inflamm Bowel Dis 6:8, 2000; and J Satsangi et al: The Montreal classification
of inflammatory bowel disease: Controversies, consensus, and implications.
Gut 55:749, 2006.
or proctosigmoiditis, proximal transit slows, which may account for
the constipation commonly seen in patients with distal disease.
When the disease extends beyond the rectum, blood is usually
mixed with stool or grossly bloody diarrhea may be noted. Colonic
motility is altered by inflammation with rapid transit through the
inflamed intestine. When the disease is severe, patients pass a liquid
stool containing blood, pus, and fecal matter. Diarrhea is often noc­
turnal and/or postprandial. Although severe pain is not a prominent
symptom, some patients with active disease may experience lower
abdominal discomfort or mild central abdominal cramping. Severe
cramping and abdominal pain can occur with severe attacks of the dis­
ease. Other symptoms in moderate to severe disease include anorexia,
nausea, vomiting, fever, and weight loss.
Physical signs of proctitis include a tender anal canal and blood
on rectal examination. With more extensive disease, patients have
tenderness to palpation directly over the colon. Patients with a toxic
colitis have severe pain and bleeding, and those with megacolon have
hepatic tympany. Both may have signs of peritonitis if a perforation has
occurred. The classification of disease activity is shown in Table 337-4.
Laboratory, Endoscopic, and Radiographic Features 
Active
disease can be associated with a rise in acute-phase reactants (C-reactive
protein [CRP]), platelet count, and erythrocyte sedimentation rate
(ESR) and a decrease in hemoglobin. Fecal lactoferrin, a glycoprotein
present in activated neutrophils, is a highly sensitive and specific
marker for detecting intestinal inflammation. Fecal calprotectin is
present in neutrophils and monocytes, and the levels correlate well
with histologic inflammation, predict relapses, and detect pouchitis.
Both fecal lactoferrin and calprotectin are an integral part of IBD
management and are used frequently to rule out active inflammation
versus symptoms of irritable bowel or bacterial overgrowth. In severely
ill patients, the serum albumin level will fall rather quickly. Leukocy­
tosis may be present but is not a specific indicator of disease activity.
Diagnosis relies on the patient’s history, clinical symptoms, negative
stool and/or tissue examination for bacteria, C. difficile toxin, ova and
parasites, and viruses depending on epidemiologic considerations and
clinical presentation; sigmoidoscopic appearance (see Fig. 333-4A);
and histology of rectal or colonic biopsy specimens.
Sigmoidoscopy is used to assess disease activity and is usually per­
formed before treatment. If the patient is not having an acute flare,
colonoscopy is used to assess disease extent and activity (Fig. 337-7).
Endoscopically mild disease is characterized by erythema, decreased
vascular pattern, and mild friability. Moderate disease is characterized
by marked erythema, absent vascular pattern, friability, and erosions,
FIGURE 337-7  Colonoscopy with acute ulcerative colitis: severe colon
inflammation with erythema, friability, and exudates. (Courtesy of Dr. M. Hamilton,
Gastroenterology Division, Department of Medicine, Brigham and Women’s
Hospital, Boston, Massachusetts; with permission.)
and severe disease is characterized by spontaneous bleeding and ulcer­
ations. Histologic features change more slowly than clinical features
but can also be used to grade disease activity. Intestinal ultrasound is
a newer tool to assess UC activity and mucosal healing and correlates
well with endoscopy and inflammatory markers.
CHAPTER 337
Complications 
Only 15% of patients with UC present initially
with severe disease. Massive hemorrhage occurs in 1% of patients, and
treatment for the disease usually stops the bleeding. Toxic megacolon is
defined as a transverse or right colon with a diameter of >6 cm, with
loss of haustration in patients with severe attacks of UC. It occurs rarely
and can be triggered by electrolyte abnormalities and narcotics. About
50% of acute dilations will resolve with conservative management
alone, but urgent colectomy is required for those who do not improve.
Perforation is the most dangerous of the local complications, and the
physical signs of peritonitis may not be obvious, especially if the patient
is receiving glucocorticoids. Although perforation is rare, the mortality
rate for perforation complicating a toxic megacolon is ~15%. In addi­
tion, patients can develop a toxic colitis and such severe ulcerations
that the bowel may perforate without first dilating.
Inflammatory Bowel Disease
Strictures occur in ~5% of patients and are always a concern in UC
because of the possibility of underlying neoplasia. Although benign
strictures can form from the inflammation and fibrosis of UC, stric­
tures that are impassable with the colonoscope should be presumed
malignant until proven otherwise. A stricture that prevents passage
of the colonoscope is an indication for surgery. UC patients occasion­
ally develop anal fissures, perianal abscesses, or hemorrhoids, but the
occurrence of extensive perianal lesions should suggest CD.
■
■CROHN’S DISEASE
Signs and Symptoms 
Although CD usually presents as acute or
chronic bowel inflammation, the inflammatory process evolves toward
one of two patterns of disease: a fibrostenotic obstructing pattern or
a penetrating fistulous pattern, each with different treatments and
prognoses. The site of disease influences the clinical manifestations
(Table 337-5).
ILEOCOLITIS  Because the most common site of inflammation is the
terminal ileum, the usual presentation of ileocolitis is a chronic his­
tory of recurrent episodes of right lower quadrant pain and diarrhea.
Sometimes the initial presentation mimics acute appendicitis with
pronounced right lower quadrant pain, a palpable mass, fever, and
leukocytosis. Pain is usually colicky; it precedes and is relieved by def­
ecation. A low-grade fever is usually noted. High-spiking fever suggests
intraabdominal abscess formation. Weight loss is common—typically
TABLE 337-5  Vienna and Montreal Classifications of Crohn’s Disease
 
VIENNA
MONTREAL
Age at diagnosis
A1: <40 years
A2: >40 years
A1: <16 years
A2: Between 17 and 40 years
A3: >40 years
Location
L1: Ileal
L2: Colonic
L3: Ileocolonic
L4: Upper
L1: Ileal
L2: Colonic
L3: Ileocolonic
L4: Isolated upper diseasea
Behavior
B1: Nonstricturing,
nonpenetrating
B2: Stricturing
B3: Penetrating
B1: Nonstricturing,
nonpenetrating
B2: Stricturing
B3: Penetrating
p: Perianal disease modifierb
aL4 is a modifier and can be added to L1–L3 when there is concomitant foregut
disease. bp is added to B1–B3 when there is concomitant perianal disease.
10–20% of body weight—and develops as a consequence of diarrhea,
anorexia, and fear of eating.
An inflammatory mass may be palpated in the right lower quadrant
of the abdomen. The mass is composed of inflamed bowel, induration of
the mesentery, and enlarged abdominal lymph nodes. The “string sign”
on radiographic studies results from a severely narrowed loop of bowel,
which makes the lumen resemble a frayed cotton string. It is caused by
incomplete filling of the lumen as the result of edema, irritability, and
spasms associated with inflammation and ulcerations. The sign may be
seen in both nonstenotic and stenotic phases of the disease.
PART 10
Disorders of the Gastrointestinal System
Bowel obstruction may take several forms. In the early stages of
disease, bowel wall edema and spasm produce intermittent obstructive
manifestations and increasing symptoms of postprandial pain. Over
several years, persistent inflammation gradually progresses to fibroste­
notic narrowing and stricture. Diarrhea will decrease and be replaced
by chronic bowel obstruction. Acute episodes of obstruction occur as
well, precipitated by bowel inflammation and spasm or sometimes by
impaction of undigested food or medication. These episodes usually
resolve with intravenous fluids and gastric decompression.
Severe inflammation of the ileocecal region may lead to localized
wall thinning, with microperforation and fistula formation to the
adjacent bowel, the skin, or the urinary bladder, or to an abscess cav­
ity in the mesentery. Enterovesical fistulas typically present as dysuria
or recurrent bladder infections or, less commonly, as pneumaturia
or fecaluria. Enterocutaneous fistulas follow tissue planes of least
resistance, usually draining through abdominal surgical scars. Entero­
vaginal fistulas are rare and present as dyspareunia or as a feculent or
foul-smelling, often painful vaginal discharge. They are unlikely to
develop without a prior hysterectomy.
JEJUNOILEITIS  Extensive inflammatory disease is associated with a
loss of digestive and absorptive surface, resulting in malabsorption and
steatorrhea. Nutritional deficiencies can also result from poor intake
and enteric losses of protein and other nutrients. Intestinal malab­
sorption can cause anemia, hypoalbuminemia, hypocalcemia, hypo­
magnesemia, coagulopathy, and hyperoxaluria with nephrolithiasis in
patients with an intact colon. Many patients need to take intravenous
iron since oral iron is poorly tolerated and often ineffective. Verte­
bral fractures are caused by a combination of vitamin D deficiency,
hypocalcemia, and prolonged glucocorticoid use. Pellagra from niacin
deficiency can occur in extensive small-bowel disease, and malabsorp­
tion of vitamin B12 can lead to megaloblastic anemia and neurologic
symptoms. Other important nutrients to measure and replete if low are
folate and vitamins A, E, and K. Levels of minerals such as zinc, sele­
nium, copper, and magnesium are often low in patients with extensive
small-bowel inflammation or resections, and these should be repleted
as well. Most patients should take daily multivitamin, calcium, and
vitamin D supplements.
Diarrhea is characteristic of active disease; its causes include (1)
bacterial overgrowth in obstructive stasis or fistulization, (2) bile acid
malabsorption due to a diseased or resected terminal ileum, (3) intes­
tinal inflammation with decreased water absorption and increased
secretion of electrolytes, and (4) enteroenteric fistula(e).
COLITIS AND PERIANAL DISEASE  Patients with colitis present with
low-grade fevers, malaise, diarrhea, crampy abdominal pain, and
sometimes hematochezia. Gross bleeding is not as common as in UC
and appears in about one-half of patients with exclusively colonic dis­
ease. Only 1–2% exhibit massive bleeding. Pain is caused by passage of
fecal material through narrowed and inflamed segments of the large
bowel. Decreased rectal compliance is another cause for diarrhea in
Crohn’s colitis patients.
Stricturing can occur in the colon in 4–16% of patients and produce
symptoms of bowel obstruction. If the endoscopist is unable to traverse
a stricture in Crohn’s colitis, surgical resection should be considered,
especially if the patient has symptoms of chronic obstruction. Colonic
disease may fistulize into the stomach or duodenum, causing feculent
vomiting, or to the proximal or mid-small bowel, causing malab­
sorption by “short circuiting” the absorptive surface and bacterial
overgrowth. Ten percent of women with Crohn’s colitis will develop a
rectovaginal fistula.
Perianal disease affects about one-third of patients with Crohn’s colitis
and is manifested by incontinence, large hemorrhoidal tags, anal stric­
tures, anorectal fistulas, and perirectal abscesses. Not all patients with
perianal fistula will have endoscopic evidence of colonic inflammation.
GASTRODUODENAL DISEASE  Symptoms and signs of upper GI tract
disease include nausea, vomiting, and epigastric pain. Patients usually
have a Helicobacter pylori–negative gastritis. The second portion of the
duodenum is more commonly involved than the bulb. Fistulas involv­
ing the stomach or duodenum arise from the small or large bowel and
do not necessarily signify the presence of upper GI tract involvement.
Patients with advanced gastroduodenal CD may develop a chronic gas­
tric outlet obstruction. About 30% of children diagnosed with CD have
esophagogastroduodenal involvement. The classification of disease
activity is shown in Table 337-5.
Laboratory, Endoscopic, and Radiographic Features 
Laboratory
abnormalities include elevated ESR and CRP. In more severe disease, find­
ings include hypoalbuminemia, anemia, and leukocytosis. Fecal calprotec­
tin and lactoferrin levels have been used to distinguish IBD from irritable
bowel syndrome (IBS), to assess whether CD is active, and to detect post­
operative recurrence of CD. Fecal calprotectin is a more sensitive marker of
ileocolonic or colonic inflammation rather than isolated ileal inflammation.
Endoscopic features of CD include rectal sparing, aphthous ulcer­
ations, fistulas, and skip lesions. Colonoscopy allows examination and
biopsy of mass lesions or strictures and biopsy of the terminal ileum.
Upper endoscopy is useful in diagnosing gastroduodenal involvement
in patients with upper tract symptoms. An ileal or colonic stricture may
be dilated with a balloon introduced through the colonoscope. Strictures
≤4 cm in length and those at anastomotic sites respond better to endo­
scopic dilation. The perforation rate is as high as 10%. Most endoscopists
dilate only fibrotic strictures and not those associated with active inflam­
mation. Wireless capsule endoscopy (WCE) allows direct visualization
of the entire small-bowel mucosa (Fig. 337-8). The diagnostic yield
of detecting lesions suggestive of active CD is higher with WCE than
computed tomography (CT) or magnetic resonance (MR) enterography.
WCE should not be used in the setting of a small-bowel stricture. Cap­
sule retention occurs in <1% of patients with suspected CD, but retention
rates of 4–6% are seen in patients with established CD. It is helpful to
give the patient with CD a patency capsule, which is made of barium and
starts to dissolve 30 h after ingestion. An abdominal x-ray can be taken
at around 30 h after ingestion to see if the capsule is still present in the
small bowel, which would indicate a stricture.
In CD, early radiographic findings in the small bowel include
thickened folds and aphthous ulcerations. “Cobblestoning” from lon­
gitudinal and transverse ulcerations most frequently involves the small
bowel. In more advanced disease, strictures, fistulas, inflammatory
masses, and abscesses may be detected. The earliest macroscopic find­
ings of colonic CD are aphthous ulcers. These small ulcers are often
FIGURE 337-8  Wireless capsule endoscopy image in a patient with Crohn’s disease
of the ileum shows ulcerations and narrowing of the intestinal lumen. (Courtesy
of Dr. S. Reddy, Gastroenterology Division, Department of Medicine, Brigham and
Women’s Hospital, Boston, Massachusetts; with permission.)
multiple and separated by normal intervening mucosa. As the disease
progresses, aphthous ulcers become enlarged, deeper, and occasionally
connected to one another, forming longitudinal stellate, serpiginous,
and linear ulcers (see Fig. 333-4B).
The transmural inflammation of CD leads to decreased luminal
diameter and limited distensibility. As ulcers progress deeper, they can
lead to fistula formation. The segmental nature of CD results in wide
gaps of normal or dilated bowel between involved segments.
CT enterography and MR enterography have been shown to be
equally accurate in the identification of active small-bowel inflamma­
tion. MRI is thought to offer superior soft tissue contrast and has the
added advantage of avoiding radiation exposure changes (Figs. 337-9
and 337-10). The lack of ionizing radiation is particularly appealing
in younger patients and when monitoring response to therapy where
serial images will be obtained. Pelvic MRI is superior to pelvic CT
for demonstrating pelvic lesions such as ischiorectal abscesses and
perianal fistulas (Fig. 337-11). An underutilized resource for assessing
small-bowel CD is small-bowel ultrasound (SBUS). SBUS is at least
as sensitive as MR enterography and CT enterography for detecting
small-bowel CD, with a sensitivity of 94%, specificity of 97%, positive
predictive value of 97%, and negative predictive value of 94%. Use of
oral contrast medium can increase the sensitivity and specificity to
detect small-bowel lesions to 100%. SBUS is best suited for distal smallbowel assessment, as the sensitivity of detecting lesions within the duo­
denum and proximal jejunum may be lower due to anatomic position.
The limitations of SBUS include availability and operator dependence.
Complications 
Because CD is a transmural process, serosal adhe­
sions develop that provide direct pathways for fistula formation and
reduce the incidence of free perforation. Perforation occurs in 1–2%
of patients, usually in the ileum but occasionally in the jejunum or
as a complication of toxic megacolon. The peritonitis of free perfo­
ration, especially colonic, may be fatal. Intraabdominal and pelvic
abscesses occur in 10–30% of patients with CD at some time in the
course of their illness. CT-guided percutaneous drainage of the abscess
is standard therapy. Despite adequate drainage, most patients need
resection of the offending bowel segment. Percutaneous drainage has
an especially high failure rate in abdominal wall abscesses. Systemic
glucocorticoid therapy increases the risk of intraabdominal and pelvic
FIGURE 337-9  A coronal magnetic resonance image was obtained using a half
Fourier single-shot T2-weighted acquisition with fat saturation in a 27-year-old
pregnant (23 weeks’ gestation) woman. The patient had Crohn’s disease and was
maintained on mercaptopurine and prednisone. She presented with abdominal pain,
distension, vomiting, and small-bowel obstruction. The image reveals a 7- to 10-cm
long stricture at the terminal ileum (white arrows) causing obstruction and significant
dilatation of the proximal small bowel (white asterisk). A fetus is seen in the uterus
(dashed white arrows). (Courtesy of Drs. J. F. B. Chick and P. B. Shyn, Abdominal
Imaging and Intervention, Department of Radiology, Brigham and Women’s Hospital,
Harvard Medical School, Boston, Massachusetts; with permission.)
CHAPTER 337
Inflammatory Bowel Disease
abscesses in CD patients who have never had an operation. Other com­
plications include intestinal obstruction in 40%, massive hemorrhage,
malabsorption, and severe perianal disease.
Serologic Markers 
Clinical factors described at diagnosis are
more helpful than serologies at predicting the natural history of IBD.
The positive and negative predictive values of serologic tests such
as anti-Saccharomyces cerevisiae antibody (ASCA) and perinuclear
antineutrophil cytoplasmic antibody (pANCA) vary depending on
the prevalence of IBD in different populations, such that their clinical
usefulness is unclear.
DIFFERENTIAL DIAGNOSIS OF UC AND CD
Once a diagnosis of IBD is made, distinguishing between UC and
CD is impossible initially in up to 15% of cases. These are termed
indeterminate colitis. Fortunately, in most cases, the true nature of the
underlying colitis becomes evident later in the course of the patient’s
disease. Approximately 5% of colon resection specimens are difficult to
classify as either UC or CD because they exhibit overlapping histologic
features.
■
■INFECTIOUS DISEASES
Infections of the small intestines and colon can mimic CD or UC. They
may be bacterial, fungal, viral, or protozoal in origin (Table 337-6).
Campylobacter colitis can mimic the endoscopic appearance of severe
UC and can cause a relapse of established UC. Salmonella can cause
watery or bloody diarrhea, nausea, and vomiting. Shigellosis causes
watery diarrhea, abdominal pain, and fever followed by rectal tenesmus
and by the passage of blood and mucus per rectum. All three are usu­
ally self-limited, but 1% of patients infected with Salmonella become
asymptomatic carriers. Yersinia enterocolitica infection occurs mainly
in the terminal ileum and causes mucosal ulceration, neutrophil
FIGURE 337-10  A coronal balanced, steady-state, free precession, T2-weighted
image with fat saturation was obtained in a 32-year-old man with Crohn’s disease
and prior episodes of bowel obstruction, fistulas, and abscesses. He was being
treated with mercaptopurine and presented with abdominal distention and diarrhea.
The image demonstrates a new gastrocolic fistula (solid white arrows). Multifocal
involvement of the small bowel and terminal ileum is also present (dashed white
arrows). (Courtesy of Drs. J. F. B. Chick and P. B. Shyn, Abdominal Imaging and
Intervention, Department of Radiology, Brigham and Women’s Hospital, Harvard
Medical School, Boston, Massachusetts; with permission.)
PART 10
Disorders of the Gastrointestinal System
FIGURE 337-11  Axial T2-weighted fat-saturated image obtained in a 39-year-old
male with Crohn’s disease shows a defect in the internal sphincter at the 6 o’clock
position of the mid anal canal (open white arrow) communicating with a 1.1-cm
intersphincteric collection (black arrow). Wide defect in the external sphincter at the
7 o’clock position (solid white arrow) leads to a moderate-sized perianal abscess in
the ischioanal fossa (asterisk). (Courtesy of Drs. J.S. Quon and P.B. Shyn, Abdominal
Imaging and Intervention, Department of Radiology, Brigham and Women’s Hospital,
Harvard Medical School, Boston, Massachusetts; with permission.)
TABLE 337-6  Diseases That Mimic IBD
Infectious Etiologies
Bacterial
Mycobacterial
Viral
  Salmonella
  Tuberculosis
  Cytomegalovirus
  Shigella
  Mycobacterium avium
  Herpes simplex
  Toxigenic
Parasitic
  HIV
  Escherichia coli
  Amebiasis
Fungal
  Campylobacter
  Isospora
  Histoplasmosis
  Yersinia
  Trichuris trichiura
  Candida
  Clostridioides difficile
  Hookworm
  Aspergillus
  Gonorrhea
  Strongyloides
 
  Chlamydia trachomatis
 
Noninfectious Etiologies
Inflammatory
Neoplastic
Drugs and Chemicals
  Appendicitis
  Lymphoma
  NSAIDs
  Diverticulitis
  Metastatic
  Phosphosoda
  Diversion colitis
  Carcinoma
  Cathartic colon
  Collagenous/lymphocytic
  Carcinoma of the ileum
  Carcinoid
  Familial polyposis
  Gold
  Oral contraceptives
  Cocaine
  Immune checkpoint
colitis
  Ischemic colitis
  Radiation colitis/enteritis
inhibitor colitis
  Mycophenolate
  Solitary rectal ulcer
syndrome
mofetil
  Eosinophilic
gastroenteritis
  Neutropenic colitis
  Behçet’s syndrome
  Graft-versus-host disease
Abbreviations: IBD, inflammatory bowel disease; NSAIDs, nonsteroidal antiinflammatory drugs.
invasion, and thickening of the ileal wall. Other bacterial infections
that may mimic IBD include C. difficile, which presents with watery
diarrhea, tenesmus, nausea, and vomiting; and E. coli, three categories
of which can cause colitis. These are enterohemorrhagic, enteroinva­
sive, and enteroadherent E. coli, all of which can cause bloody diarrhea
and abdominal tenderness. Gonorrhea, Chlamydia, and syphilis can
also cause proctitis.
GI involvement with mycobacterial infection occurs primarily in
the immunosuppressed patient but may occur in patients with nor­
mal immunity. Distal ileal and cecal involvement predominates, and
patients present with symptoms of small-bowel obstruction and a ten­
der abdominal mass. The diagnosis is made most directly by colonos­
copy with biopsy and culture. Although most of the patients with viral
colitis are immunosuppressed, cytomegalovirus (CMV) and herpes
simplex proctitis may occur in immunocompetent individuals. CMV
occurs most commonly in the esophagus, colon, and rectum but may
also involve the small intestine. Symptoms include abdominal pain,
bloody diarrhea, fever, and weight loss. With severe disease, necrosis
and perforation can occur. Diagnosis is made by identification of char­
acteristic intranuclear inclusions in mucosal cells on biopsy. Herpes
simplex infection of the GI tract is limited to the oropharynx, anorec­
tum, and perianal areas. Symptoms include anorectal pain, tenesmus,
constipation, inguinal adenopathy, difficulty with urinary voiding, and
sacral paresthesias. Diagnosis is made by rectal biopsy with identifica­
tion of characteristic cellular inclusions and viral culture. HIV itself
can cause diarrhea, nausea, vomiting, and anorexia. Small-intestinal
biopsies show partial villous atrophy; small-bowel bacterial overgrowth
and fat malabsorption may also be noted.
Protozoan parasites include Isospora belli, which can cause a selflimited infection in healthy hosts but causes a chronic profuse, watery
diarrhea and weight loss in AIDS patients. Entamoeba histolytica
or related species infect ~10% of the world’s population; symptoms
include abdominal pain, tenesmus, frequent loose stools containing
blood and mucus, and abdominal tenderness. Colonoscopy reveals
focal punctate ulcers with normal intervening mucosa; diagnosis is
made by biopsy or serum amebic antibodies. Fulminant amebic colitis
is rare but has a mortality rate of >50%.
Other parasitic infections that may mimic IBD include hookworm
(Necator americanus), whipworm (Trichuris trichiura), and Strongyloi­
des stercoralis. In severely immunocompromised patients, Candida or
Aspergillus can be identified in the submucosa. Disseminated histoplas­
mosis can involve the ileocecal area.
■
■NONINFECTIOUS DISEASES
Diverticulitis can be confused with CD clinically and radiographically.
Both diseases cause fever, abdominal pain, tender abdominal mass, leu­
kocytosis, elevated ESR, partial obstruction, and fistulas. Perianal disease
or ileitis on small-bowel series favors the diagnosis of CD. Significant
endoscopic mucosal abnormalities are more likely in CD than in diver­
ticulitis. Endoscopic or clinical recurrence following segmental resection
favors CD. Diverticular-associated colitis is similar to CD, but mucosal
abnormalities are limited to the sigmoid and descending colon.
Ischemic colitis is commonly confused with IBD. The ischemic
process can be chronic and diffuse, as in UC, or segmental, as in CD.
Colonic inflammation due to ischemia may resolve quickly or may per­
sist and result in transmural scarring and stricture formation. Ischemic
bowel disease should be considered in the elderly following abdominal
aortic aneurysm repair or when a patient has a hypercoagulable state or
a severe cardiac or peripheral vascular disorder. Patients usually present
with sudden onset of left lower quadrant pain, urgency to defecate, and
the passage of bright red blood per rectum. Endoscopic examination
often demonstrates a normal-appearing rectum and a sharp transition
to an area of inflammation in the descending colon and splenic flexure.
The effects of radiotherapy on the GI tract can be difficult to distin­
guish from IBD. Acute symptoms can occur within 1–2 weeks of start­
ing radiotherapy. When the rectum and sigmoid are irradiated, patients
develop bloody, mucoid diarrhea and tenesmus, as in distal UC. With
small-bowel involvement, diarrhea is common. Late symptoms include
malabsorption and weight loss. Stricturing with obstruction and bacte­
rial overgrowth may occur. Fistulas can penetrate the bladder, vagina,
or abdominal wall. Flexible sigmoidoscopy reveals mucosal granularity,
friability, numerous telangiectasias, and occasionally discrete ulcer­
ations. Biopsy can be diagnostic.
Solitary rectal ulcer syndrome is uncommon and can be con­
fused with IBD. It occurs in persons of all ages and may be caused
by impaired evacuation and failure of relaxation of the puborectalis
muscle. Single or multiple ulcerations may arise from anal sphincter
overactivity, higher intrarectal pressures during defecation, and digital
removal of stool. Patients complain of constipation with straining and
pass blood and mucus per rectum. Other symptoms include abdominal
pain, diarrhea, tenesmus, and perineal pain. Ulceration, which may be
as large as 5 cm in diameter, is usually observed anteriorly or anterolat­
erally 3–15 cm from the anal verge. Biopsies can be diagnostic.
Several types of colitis are associated with nonsteroidal antiinflammatory drugs (NSAIDs), including de novo colitis, reactivation
of IBD, and proctitis caused by use of suppositories. Most patients with
NSAID-related colitis present with diarrhea and abdominal pain, and
complications include stricture, bleeding, obstruction, perforation,
and fistulization. Withdrawal of these agents is crucial, and in cases of
reactivated IBD, standard therapies are indicated.
Colitis secondary to immune checkpoint inhibitors (ICIs), termed
ICI-related colitis, has emerged as these agents have found use in a
wide variety of cancers. Immune checkpoint proteins such as CTLA-4
and programmed cell death protein 1 (PD-1) are receptors expressed
on the surface of effector T cells that interact with their ligands CD80/
CD86 (CTLA-4) and programmed death ligand 1 (PD-L1) on antigenpresenting cells and normally function as inhibitors of immune
responses. ICIs block these inhibitory pathways and promote the acti­
vation and proliferation of the native adaptive T-cell response against
malignant cells as their mechanism of antitumor activity. While very
effective at enhancing antitumor T-cell activity, ICIs also activate global
T-cell responses that induce several autoimmune-related adverse
events. Although immune-related adverse events of ICIs occur in mul­
tiple organ systems, the GI tract is affected in 21–44% of patients. The
most common clinical presentation is self-limited diarrhea that can be
associated with frank colitis and can lead to significant morbidity and
mortality if not managed appropriately. Treatment is generally based
on symptom severity. Moderate to severe symptoms usually require
glucocorticoids, whereas biologics such as anti-TNF agents and integ­
rin inhibitors are used in steroid-refractory cases.
■
■THE ATYPICAL COLITIDES
Two atypical colitides—collagenous colitis and lymphocytic colitis—
have normal endoscopic appearances. Collagenous colitis has two main
histologic components: increased subepithelial collagen deposition and
colitis with increased intraepithelial lymphocytes. The female-to-male
ratio is 9:1, and most patients present in the sixth or seventh decade of
life. The main symptom is chronic watery diarrhea. Risk factors include
smoking; use of NSAIDs, proton pump inhibitors, or beta blockers; and
a history of autoimmune disease.
Lymphocytic colitis has features similar to collagenous colitis, includ­
ing age at onset and clinical presentation, but it has almost equal inci­
dence in men and women and no subepithelial collagen deposition on
pathologic section. However, intraepithelial lymphocytes are increased.
Use of sertraline (but not beta blockers) is an additional risk factor. The
frequency of celiac disease is increased in lymphocytic colitis and ranges
from 9 to 27%. Celiac disease should be excluded in all patients with
lymphocytic colitis, particularly if diarrhea does not respond to conven­
tional therapy. Treatments for both microscopic colitides vary depending
on symptom severity and include, antidiarrheals (e.g., loperamide and
diphenoxylate), bismuth, aminosalicylates, budesonide, systemic gluco­
corticoids, and biologics for refractory disease.
CHAPTER 337
Diversion colitis is an inflammatory process that arises in segments
of the large intestine that are not continuous with the fecal stream. It
usually occurs in patients with ileostomy or colostomy when a mucus
fistula or a Hartmann’s pouch has been created. Clinically, patients
have mucus or bloody discharge from the rectum. Erythema, granu­
larity, friability, and, in more severe cases, ulceration can be seen on
endoscopy. Histopathology shows areas of active inflammation with
foci of cryptitis and crypt abscesses. Crypt architecture is normal,
which differentiates it from UC but not necessarily CD. Short-chain
fatty acid enemas may help in diversion colitis, but the definitive
therapy is surgical reanastomosis.
Inflammatory Bowel Disease
EXTRAINTESTINAL MANIFESTATIONS
Up to one-third of IBD patients have at least one extraintestinal disease
manifestation. Please see Table 337-7 for a summary of IBD EIMs.
■
■DERMATOLOGIC
Erythema nodosum (EN) occurs in up to 15% of CD patients and
10% of UC patients. Attacks usually correlate with bowel activity;
skin lesions develop after the onset of bowel symptoms, and patients
frequently have concomitant active peripheral arthritis. The lesions of
EN are hot, red, tender nodules measuring 1–5 cm in diameter and are
found on the anterior surface of the lower legs, ankles, calves, thighs,
and arms. Therapy is directed toward the underlying bowel disease.
Pyoderma gangrenosum (PG) is seen in 1–12% of UC patients and
less commonly in Crohn’s colitis. Although it usually presents after
the diagnosis of IBD, PG may occur years before the onset of bowel
symptoms, run a course independent of the bowel disease, respond
poorly to colectomy, and even develop years after proctocolectomy. It
is usually associated with severe disease. Lesions are commonly found
on the dorsal surface of the feet and legs but may occur on the arms,
chest, stoma, and even the face. PG usually begins as a pustule and
then spreads concentrically to rapidly undermine healthy skin. Lesions
then ulcerate, with violaceous edges surrounded by a margin of ery­
thema. Centrally, they contain necrotic tissue with blood and exudates.
Lesions may be single or multiple and grow as large as 30 cm. They are
sometimes very difficult to treat and often require IV antibiotics, IV
glucocorticoids, dapsone, azathioprine, thalidomide, IV cyclosporine
(CSA), infliximab, or adalimumab.
TABLE 337-7  Extraintestinal Manifestations
CATEGORY
CLINICAL COURSE
TREATMENT
Rheumatologic Disorders (5–20%)
Peripheral arthritis
Asymmetric, migratory
Parallels bowel activity
Sacroiliitis
Symmetric: spine and hip joints
Independent of bowel activity
Ankylosing spondylitis
Gradual fusion of spine
Independent of bowel activity
Two-thirds have HLA-B27 antigen
Metabolic Bone Disorders (up to 40% of patients)
Osteoporosis
Risk increased by glucocorticoids, cyclosporine, methotrexate, total
parenteral nutrition, malabsorption, and inflammation
Fracture rates highest in the elderly (age >60)
Osteonecrosis
Death of osteocytes and adipocytes and eventual bone collapse; affects hips
more than knees or shoulders; risk factor is steroid use
Dermatologic Disorders (10–20%)
Erythema nodosum
Hot, red, tender, nodules/extremities
Parallels bowel activity
Pyoderma gangrenosum
Ulcerating, necrotic lesions on extremities, trunk, face, stoma
Independent of bowel activity
Psoriasis
Unrelated to bowel activity
Topical steroids, light therapy, methotrexate, infliximab,
adalimumab, ustekinumab, risankizumab, JAK inhibitors
Pyoderma vegetans
Intertriginous areas
Parallels bowel activity
PART 10
Disorders of the Gastrointestinal System
Pyostomatitis vegetans
Mucous membranes
Parallels bowel activity
Metastatic Crohn’s
disease (CD)
CD of the skin
Parallels bowel activity
Sweet syndrome
Neutrophilic dermatosis
Parallels bowel activity
Aphthous stomatitis
Oral ulcerations
Parallels bowel activity
Ocular Disorders (1–11%)
Uveitis
Ocular pain, photophobia, blurred vision, headache
Independent of bowel activity
Episcleritis
Mild ocular burning
Parallels bowel activity
Hepatobiliary Disorders (10–35%)
Fatty liver
Secondary to chronic illness, malnutrition, steroid therapy
Improve nutrition, reduce steroids
Cholelithiasis
Patients with ileitis or ileal resection
Malabsorption of bile acids, depletion of bile salt pool, secretion of lithogenic bile
Primary sclerosing
cholangitis (PSC)
Intrahepatic and extrahepatic
Inflammation and fibrosis leading to biliary cirrhosis and hepatic failure
7–10% cholangiocarcinoma
Small-duct PSC involves small-caliber bile ducts and has a better prognosis
Urologic
Nephrolithiasis (10–20%)
CD patients following small-bowel resection; calcium oxalate stones most
common
Less Common Extraintestinal Manifestations
Thromboembolic disorders
Increased risk of venous and arterial thrombosis; factors responsible include
abnormalities of the platelet-endothelial interaction, hyperhomocysteinemia,
alterations in the coagulation cascade, impaired fibrinolysis, involvement of
tissue factor–bearing microvesicles, disruption of the normal coagulation
system by autoantibodies, and a genetic predisposition
Cardiopulmonary
Endocarditis, myocarditis, pleuropericarditis, interstitial lung disease
Treatment is varied; stop 5-ASA agents as they can rarely
cause interstitial lung disease
Systemic amyloidosis
Secondary (reactive) in long-standing IBD, especially CD
Colchicine and referral to specialty center
Pancreatitis
Duodenal fistulas, ampullary CD, gallstones, PSC, drugs (MP, azathioprine,
5-ASAs), autoimmune, primary CD of the pancreas
Abbreviations: 5-ASA, 5-aminosalicylic acid; DEXA, dual-energy x-ray absorptiometry; ERCP, endoscopic retrograde cholangiopancreatography; IBD, inflammatory bowel
disease; IL, interleukin; MP, mercaptopurine; TNF, tumor necrosis factor.
Reduce bowel inflammation
Steroids, injections, methotrexate, anti-TNF
Physical therapy, steroids, injections, methotrexate, antiTNF, IL-17 inhibitors, JAK inhibitors
Screening with DEXA scan, check vitamin D levels, treat if
osteoporosis or osteopenia on long-term corticosteroids
Pain control, injections, joint replacement
Reduce bowel inflammation
Antibiotics, steroids, cyclosporine, infliximab, dapsone,
azathioprine, intralesional steroids; not debridement or
colectomy
Evanescent; resolves without progression
Evanescent; resolves without progression
Reduce bowel inflammation
Reduce bowel inflammation
Reduce bowel inflammation/topical therapy
Topical or systemic steroids
Topical corticosteroids
Reduce bowel inflammation; cholecystectomy in
symptomatic patients
Cholecystectomy in patients with gallbladder polyps due
to the high incidence of malignancy
Low-oxalate diet; control of bowel inflammation; surgical
intervention
Anticoagulation; control of inflammation
Treatment is varied; stop offending medication; diagnose
and treat with ERCP and/or cholecystectomy
Other dermatologic manifestations include pyoderma vegetans,
which occurs in intertriginous areas; pyostomatitis vegetans, which
involves the mucous membranes; Sweet syndrome, a neutrophilic
dermatosis; and metastatic CD, a rare disorder defined by cutaneous
granuloma formation. Psoriasis affects 5–10% of patients with IBD
and is unrelated to bowel activity, consistent with the potential shared
immunogenetic basis of these diseases. Perianal skin tags are found in
75–80% of patients with CD, especially those with colon involvement.
Oral mucosal lesions, seen often in CD and rarely in UC, include aph­
thous stomatitis and “cobblestone” lesions of the buccal mucosa.
■
■RHEUMATOLOGIC
Peripheral arthritis develops in 15–20% of IBD patients, is more
common in CD, and worsens with exacerbations of bowel activity. It
is asymmetric, polyarticular, and migratory and most often affects
large joints of the upper and lower extremities. Treatment is directed
at reducing bowel inflammation. In severe UC, colectomy frequently
cures the arthritis.
Ankylosing spondylitis (AS) occurs in ~10% of IBD patients and
is more common in CD than UC. About two-thirds of IBD patients
with AS express the HLA-B27 antigen. The AS activity is not related to
bowel activity and does not remit with glucocorticoids or colectomy.
It most often affects the spine and pelvis, producing symptoms of dif­
fuse low-back pain, buttock pain, and morning stiffness. The course
is continuous and progressive, leading to permanent skeletal damage
and deformity. Anti-TNF therapy reduces spinal inflammation and
improves functional status and quality of life.
Sacroiliitis is symmetric, occurs equally in UC and CD, is often
asymptomatic, does not correlate with bowel activity, and does not
always progress to AS. Other rheumatic manifestations include hyper­
trophic osteoarthropathy, pelvic/femoral osteomyelitis, and relapsing
polychondritis.
■
■OCULAR
The incidence of ocular complications in IBD patients is 1–10%. The
most common are conjunctivitis, anterior uveitis/iritis, and episcleritis.
Uveitis is associated with both UC and Crohn’s colitis, may be found
during periods of remission, and may develop in patients following
bowel resection. Symptoms include ocular pain, photophobia, blurred
vision, and headache. Prompt intervention, sometimes with systemic
glucocorticoids, is required to prevent scarring and visual impairment.
Episcleritis is a benign disorder that presents with symptoms of mild
ocular burning. It occurs in 3–4% of IBD patients, more commonly in
Crohn’s colitis, and is treated with topical glucocorticoids.
■
■HEPATOBILIARY
Hepatic steatosis is detectable in about one-half of the abnormal liver
biopsies from patients with CD and UC; patients usually present
with hepatomegaly. Fatty liver usually results from a combination of
chronic debilitating illness, malnutrition, and glucocorticoid therapy.
Cholelithiasis occurs in 10–35% of CD patients with ileitis or ileal
resection. Gallstone formation is caused by malabsorption of bile
acids, resulting in depletion of the bile salt pool and the secretion of
lithogenic bile.
Primary sclerosing cholangitis (PSC) is a disorder characterized by
both intrahepatic and extrahepatic bile duct inflammation and fibro­
sis, frequently leading to biliary cirrhosis and hepatic failure; ~5% of
patients with UC have PSC, but 50–75% of patients with PSC have IBD.
PSC occurs less often in patients with CD. Although it can be recog­
nized after the diagnosis of IBD, PSC can be detected earlier or even
years after proctocolectomy. Consistent with this, the immunogenetic
basis for PSC appears to be overlapping but distinct from UC based on
GWAS, although both IBD and PSC are commonly pANCA positive.
Most patients have no symptoms at the time of diagnosis; when symp­
toms are present, they consist of fatigue, jaundice, abdominal pain,
fever, anorexia, and malaise. The traditional gold standard diagnostic
test is endoscopic retrograde cholangiopancreatography (ERCP), but
magnetic resonance cholangiopancreatography (MRCP) is sensitive,
specific, and safer. MRCP is reasonable as an initial diagnostic test in
children and adults and can visualize irregularities, multifocal stric­
tures, and dilatations of all levels of the biliary tree. In patients with
PSC, both ERCP and MRCP demonstrate multiple bile duct strictures
alternating with relatively normal segments.
Gallbladder polyps in patients with PSC have a high incidence of
malignancy, and cholecystectomy is recommended, even if a mass
lesion is <1 cm in diameter. Gallbladder surveillance with ultrasound
should be performed annually. Endoscopic stenting may be palliative
for cholestasis secondary to bile duct obstruction. Patients with symp­
tomatic disease develop cirrhosis and liver failure over 5–10 years and
eventually require liver transplantation. PSC patients have a 10–15%
lifetime risk of developing cholangiocarcinoma and then cannot be
transplanted. Patients with IBD and PSC are at increased risk of colon
cancer and should be surveyed yearly by colonoscopy and biopsy.
In addition, cholangiography is normal in a small percentage of
patients who have a variant of PSC known as small duct primary scle­
rosing cholangitis. This variant (sometimes referred to as “pericholan­
gitis”) is probably a form of PSC involving small-caliber bile ducts. It
has similar biochemical and histologic features to classic PSC. It has a
significantly better prognosis than classic PSC, although it may evolve
into classic PSC. Granulomatous hepatitis and hepatic amyloidosis are
much rarer EIMs of IBD.
■
■UROLOGIC
The most frequent genitourinary complications are calculi, ureteral
obstruction, and ileal bladder fistulas. The highest frequency of neph­
rolithiasis (10–20%) occurs in patients with CD following small-bowel
resection. Calcium oxalate stones develop secondary to hyperoxaluria,
which results from increased absorption of dietary oxalate. Normally,
dietary calcium combines with luminal oxalate to form insoluble cal­
cium oxalate, which is eliminated in the stool. In patients with ileal
dysfunction, however, nonabsorbed fatty acids bind calcium and leave
oxalate unbound. The unbound oxalate is then delivered to the colon,
where it is readily absorbed, especially in the presence of inflammation.
CHAPTER 337
Inflammatory Bowel Disease
■
■METABOLIC BONE DISORDERS
Low bone mass occurs in 14–42% of IBD patients. The risk is increased
by glucocorticoids, CSA, methotrexate (MTX), and total parenteral
nutrition (TPN). Malabsorption and inflammation mediated by IL-1,
IL-6, TNF, and other inflammatory mediators also contribute to low
bone density. An increased incidence of hip, spine, wrist, and rib
fractures has been noted: 36% in CD and 45% in UC. The absolute
risk of an osteoporotic fracture is ~1% per person per year. Fracture
rates, particularly in the spine and hip, are highest among the elderly
(age >60). One study noted an OR of 1.72 for vertebral fracture and an
OR of 1.59 for hip fracture. The disease severity predicted the risk of
a fracture. Only 13% of IBD patients who had a fracture were on any
kind of antifracture treatment. Up to 20% of bone mass can be lost per
year with chronic glucocorticoid use. The effect is dose-dependent.
Budesonide may also suppress the pituitary-adrenal axis and thus car­
ries a risk of causing osteoporosis.
Osteonecrosis is characterized by death of osteocytes and adipocytes
and eventual bone collapse. The pain is aggravated by motion and
swelling of the joints. It affects the hips more often than knees and
shoulders, and in one series, 4.3% of patients developed osteonecrosis
within 6 months of starting glucocorticoids. Diagnosis is made by bone
scan or magnetic resonance imaging (MRI), and treatment consists of
pain control, cord decompression, osteotomy, and joint replacement.
■
■THROMBOEMBOLIC DISORDERS
Patients with IBD have an increased risk of both venous and arterial
thrombosis even if the disease is not active. Factors responsible for
the hypercoagulable state have included abnormalities of the plateletendothelial interaction, hyperhomocysteinemia, alterations in the
coagulation cascade, impaired fibrinolysis, involvement of tissue
factor–bearing microvesicles, disruption of the normal coagulation
system by autoantibodies, and a genetic predisposition. A spectrum of
vasculitides involving small, medium, and large vessels has also been
observed.
■
■OTHER DISORDERS
More common cardiopulmonary manifestations include endocarditis,
myocarditis, pleuropericarditis, and interstitial lung disease. A second­
ary or reactive amyloidosis can occur in patients with long-standing
IBD, especially in patients with CD. Amyloid material is deposited sys­
temically and can cause diarrhea, constipation, and renal failure. The
renal disease can be successfully treated with colchicine. Pancreatitis is
a rare EIM of IBD and results from duodenal fistulas; ampullary CD;
gallstones; PSC; drugs such as mercaptopurine, azathioprine, or, very
rarely, 5-ASA agents; autoimmune pancreatitis; and primary CD of the
pancreas.
TREATMENT
Inflammatory Bowel Disease
5-ASA AGENTS
These agents are effective at inducing and maintaining remission
in UC. Peroxisome proliferator–activated receptor γ (PPAR-γ) may
mediate 5-ASA therapeutic action by decreasing nuclear localiza­
tion of NF-κB. Sulfa-free aminosalicylate formulations include
alternative azo-bonded carriers, 5-ASA dimers, and delayed-release
and controlled-release preparations. Each has the same efficacy as
sulfasalazine when equimolar concentrations are used.
Sulfasalazine is effective treatment for mild to moderate UC, but
its high rate of side effects limits its use. Although sulfasalazine is
more effective at higher doses, at 6 or 8 g/d, up to 30% of patients
experience allergic reactions or intolerable side effects such as
headache, anorexia, nausea, and vomiting that are attributable to
the sulfapyridine moiety that is attached to 5-ASA. Hypersensitivity
reactions, independent of sulfapyridine levels, include rash, fever,
hepatitis, agranulocytosis, hypersensitivity pneumonitis, pancre­
atitis, worsening of colitis, and reversible sperm abnormalities.
Sulfasalazine can also impair folate absorption, and patients should
be given folic acid supplements.
PART 10
Disorders of the Gastrointestinal System
Balsalazide contains an azo bond binding mesalamine to the car­
rier molecule 4-aminobenzoyl-β-alanine; it is effective in the colon.
Delzicol and Asacol HD (high dose) are enteric-coated forms
of mesalamine with the 5-ASA being released at pH >7. They dis­
integrate with complete breakup of the tablet occurring in many
different parts of the gut ranging from the small intestine to the
splenic flexure; they have increased gastric residence when taken
with a meal. Lialda is a once-a-day formulation of mesalamine
(Multi-Matrix System [MMX]) designed to release mesalamine
in the colon. The MMX technology incorporates mesalamine into
a lipophilic matrix within a hydrophilic matrix encapsulated in a
polymer resistant to degradation at a low pH (<7) to delay release
TABLE 337-8  Oral 5-Aminosalicylic Acid (5-ASA) Preparations
PREPARATION
FORMULATION
DELIVERY
DOSING PER DAY
Azo-Bond
Sulfasalazine (500 mg) (Azulfidine)
Sulfapyridine-5-ASA
Balsalazide (750 mg) (Colazal)
Aminobenzoyl-alanine–5-ASA
Delayed-Release
Mesalamine (400, 800 mg) (Delzicol,
Asacol HD)
Mesalamine (1.2 g) (Lialda)
Eudragit S (pH 7)
MMX mesalamine (SPD476)
Controlled-Release
Mesalamine (250, 500, 1000 mg)
(Pentasa)
Ethylcellulose microgranules
Stomach-colon
2–4 g (acute)
1.5–4 g (maintenance)
Delayed- and Extended-Release
Mesalamine (0.375 g) (Apriso)
Intellicor extended-release mechanism
Ileum-colon
1.5 g (maintenance)
Abbreviation: MMX, Multi-Matrix System.
throughout the colon. The safety profile appears to be comparable
to other 5-ASA formulations.
Apriso is a formulation containing encapsulated mesalamine gran­
ules that delivers mesalamine to the terminal ileum and colon via a
proprietary extended-release mechanism (Intellicor). The outer coat­
ing of this agent (Eudragit L) dissolves at a pH >6. In addition, there
is a polymer matrix core that aids in sustained release throughout the
colon. Because Lialda and Apriso are given once daily, an anticipated
benefit is improved compliance compared with two to four daily
doses required for other mesalamine preparations.
Pentasa is another mesalamine formulation that uses an ethylcel­
lulose coating to allow water absorption into small beads containing
the mesalamine. Water dissolves the 5-ASA, which then diffuses out
of the bead into the lumen. Disintegration of the capsule occurs in
the stomach. The microspheres then disperse throughout the entire
GI tract from the small intestine through the distal colon in both
fasted and fed conditions.
Salofalk Granu-Stix, an unencapsulated version of mesalamine,
has been in use in Europe for induction and maintenance of remis­
sion for several years.
Appropriate doses of the 5-ASA compounds are shown in
Table 337-8. Some 50–75% of patients with mild to moderate
UC improve when treated with 5-ASA doses equivalent to 2 g/d
of mesalamine; the dose response continues up to at least 4.8 g/d.
More common side effects of the 5-ASA medications include
headaches, nausea, hair loss, and abdominal pain. Rare side effects
of the 5-ASA medications include renal impairment, hematuria,
pancreatitis, and paradoxical worsening of colitis. Renal function
tests and urinalysis should be checked yearly.
Topical Rowasa enemas are composed of mesalamine and are
effective in mild-to-moderate distal UC. Combination therapy with
mesalamine in both oral and enema form is more effective than
either treatment alone for both distal and extensive UC.
Canasa suppositories composed of mesalamine are effective in
treating proctitis.
GLUCOCORTICOIDS
The majority of patients with moderate to severe UC benefit from
oral or parenteral glucocorticoids. Prednisone is usually started at
doses of 40–60 mg/d for active UC that is unresponsive to 5-ASA
therapy. Parenteral glucocorticoids may be administered as hydro­
cortisone, 300 mg/d, or methylprednisolone, 40–60 mg/d. A newer
glucocorticoid for UC, budesonide (Uceris), is released entirely in
the colon and has minimal to no glucocorticoid side effects. The
dose is 9 mg/d for 8 weeks, and no taper is required. Topically
applied glucocorticoids (hydrocortisone enemas or budesonide
foam) are also beneficial for distal colitis and may serve as an
adjunct in those who have rectal involvement plus more proximal
Colon
3–6 g (acute)
2–4 g (maintenance)
6.75–9 g
Colon
Distal ileum-colon
2.4–4.8 g (acute)
1.6–4.8 g (maintenance)
2.4–4.8 g
Ileum-colon
disease. Hydrocortisone enemas are significantly absorbed from
the rectum and can lead to adrenal suppression with prolonged
administration. Topical 5-ASA therapy is more effective than topi­
cal steroid therapy in the treatment of distal UC.
Glucocorticoids are also effective for treatment of moderate to
severe CD. Controlled-ileal-release budesonide has been nearly
equal to prednisone for ileocolonic CD with fewer glucocorticoid
side effects. Budesonide is used for 2–3 months at a dose of 9 mg/d
and then tapered. Glucocorticoids play no role in maintenance
therapy in either UC or CD. Once clinical remission has been
induced, they should be tapered according to the clinical activity,
normally at a rate of no more than 5–10 mg/week. The side effects
are numerous, including fluid retention, abdominal striae, fat redis­
tribution, hyperglycemia, subcapsular cataracts, osteonecrosis,
osteoporosis, myopathy, emotional disturbances, and withdrawal
symptoms. Most of these side effects, aside from osteonecrosis, are
related to the dose and duration of therapy.
ANTIBIOTICS
Antibiotics have no role in the treatment of active or quiescent
UC. However, pouchitis, which occurs in ~30–50% of UC patients
after colectomy and ileal pouch anal anastomosis (IPAA), usu­
ally responds to treatment with a variety of antibiotics including
metronidazole and ciprofloxacin. Some patients require long-term
treatment with antibiotics for chronic pouchitis.
AZATHIOPRINE AND MERCAPTOPURINE
Azathioprine and mercaptopurine (MP) are purine analogues used
concomitantly with biologic therapy or, much less often, as the sole
immunosuppressants. Azathioprine is rapidly absorbed and con­
verted to MP, which is then metabolized to the active end product,
thioinosinic acid, an inhibitor of purine ribonucleotide synthesis
and cell proliferation. Efficacy can be seen as early as 3–4 weeks but
can take up to 4–6 months. Adherence can be monitored by mea­
suring the levels of 6-thioguanine and 6-methylmercaptopurine,
end products of MP metabolism. The doses used range from 2 to
3 mg/kg per day for azathioprine and 1 to 1.5 mg/kg per day for MP.
Although azathioprine and MP are usually safe, pancreatitis occurs
in 3–4% of patients, typically presents within the first few weeks
of therapy, and is completely reversible when the drug is stopped.
Other side effects include nausea, fever, rash, and hepatitis. Bone
marrow suppression (particularly leukopenia) is dose-related and
often delayed, necessitating regular monitoring of the complete blood
cell count (CBC). Additionally, 1 in 300 individuals lacks thiopurine
methyltransferase, the enzyme responsible for drug metabolism to
inactive end products (6-methylmercaptopurine); an additional 11%
of the population are heterozygotes with intermediate enzyme activ­
ity. Both are at increased risk of toxicity because of increased accumu­
lation of active 6-thioguanine metabolites. Although 6-thioguanine
and 6-methylmercaptopurine levels can be followed to determine
correct drug dosing and reduce toxicity, weight-based dosing is an
acceptable alternative. CBCs and liver function tests should be moni­
tored frequently regardless of dosing strategy.
One meta-analysis demonstrated a fourfold risk of lymphoma
in IBD patients on azathioprine and MP. The highest risk for thio­
purine-associated lymphoma is in patients >65 years old actively
using thiopurines (yearly incidence rate per 1000 patient-years of
5.41), with a moderate risk in those between the ages of 50 and
65 (incidence rate of 2.58 compared to an incidence rate of 0.37 in
patients <50 years old). Patients using thiopurines also have a two-
to threefold increased risk of nonmelanoma skin cancers.
METHOTREXATE
MTX inhibits dihydrofolate reductase, resulting in impaired DNA
synthesis. Additional anti-inflammatory properties may be related
to decreases in the production of IL-1. It is used most often con­
comitantly with biologic therapy to decrease antibody formation
and improve disease response. Intramuscular (IM) or subcutane­
ous (SC) doses range from 15 to 25 mg/week. Potential toxici­
ties include leukopenia and hepatic fibrosis, necessitating periodic
evaluation of CBCs and liver enzymes. The role of liver biopsy in
patients on long-term MTX is uncertain but is probably limited to
those with increased liver enzymes. Hypersensitivity pneumonitis is
a rare but serious complication of therapy.
CYCLOSPORINE
CSA is a lipophilic peptide with inhibitory effects on both the cel­
lular and humoral immune systems. CSA blocks the production of
IL-2 by T helper lymphocytes. CSA binds to cyclophilin, and this
complex inhibits calcineurin, a cytoplasmic phosphatase enzyme
involved in the activation of T cells. CSA also indirectly inhibits
B-cell function by blocking helper T cells. CSA has a more rapid
onset of action than MP and azathioprine.
CSA is most effective when given at 2–4 mg/kg per day IV
in severe UC that is refractory to IV glucocorticoids, with 82%
of patients responding. CSA can be an alternative to infliximab
for steroid-refractory UC before colectomy. Levels as measured
by monoclonal radioimmunoassay or by the high-performance
liquid chromatography assay should be maintained between 150
and 350 ng/mL.
CSA may cause significant toxicity; renal function should be
monitored frequently. Hypertension, gingival hyperplasia, hyper­
trichosis, paresthesias, tremors, headaches, and electrolyte abnor­
malities are common side effects. Creatinine elevation calls for dose
reduction or discontinuation. Seizures may also complicate therapy,
especially if the patient is hypomagnesemic or if serum cholesterol
levels are <3.1 mmol/L (<120 mg/dL). Opportunistic infections,
most notably Pneumocystis jirovecii pneumonia, may occur with
combination immunosuppressive treatment; antibiotic prophylaxis
with trimethoprim-sulfamethoxazole should be given.
CHAPTER 337
TACROLIMUS
Tacrolimus is a macrolide antibiotic with immunomodulatory prop­
erties similar to CSA but 100 times as potent and not dependent on
bile or mucosal integrity for absorption. Thus, tacrolimus has good
oral absorption despite proximal small-bowel Crohn’s involvement.
Tacrolimus is effective in children with refractory IBD and in adults
with extensive involvement of the small bowel. Tacrolimus use is
decreasing due to side effects and the expanding choice of biologic
and small-molecule therapies.
Inflammatory Bowel Disease
BIOLOGIC THERAPIES
Biologic therapy is now commonly given as an initial therapy for
patients with moderate to severe CD and UC to prevent future
complications of IBD. High-risk patients with UC who are more
likely to require biologics include those with moderate to severe
disease, steroid-dependent or steroid-refractory disease, and refrac­
tory pouchitis. High-risk patients with CD who are more likely to
require biologics include those who are <30 years old, with exten­
sive disease, perianal or severe rectal disease and/or deep ulcer­
ations in the colon, and stricturing or penetrating disease behavior.
The current goal of IBD treatment with biologics is to treat early
in the disease course, treat aggressively with appropriate therapies,
check drug and drug metabolite levels, administer dual therapy
with immunomodulators and biologics in appropriate patients,
and aim for deep remission (endoscopic and histologic remission).
Patients who respond to biologic therapies enjoy an improvement
in clinical symptoms; a better quality of life; less disability, fatigue,
and depression; and fewer surgeries and hospitalizations.
Anti-TNF Therapies  TNF is a proinflammatory cytokine that
regulates immune cells to coordinate a systemic immune response.
Dysregulation of TNF production has been associated with
immune-mediated disorders including IBD, and inhibition of
TNF signaling is used in the treatment of IBD. Four TNF inhibi­
tors are currently approved for the treatment of IBD: infliximab,
adalimumab, certolizumab pegol, and golimumab. Infliximab,
a chimeric IgG1 antibody against TNF-α, was the first biologic
therapy approved for moderately to severely active inflammatory
and fistulizing CD and UC.
Randomized trials support combination therapy with infliximab
and azathioprine for moderate to severe CD and UC. For moder­
ate to severe CD, combination therapy has been shown to be more
effective than either infliximab or azathioprine alone. Similarly,
combination therapy has been shown to be more effective for mod­
erate to severe UC than either infliximab or azathioprine alone.
Hospitalized patients with acute severe glucocorticoid-refractory
UC have a high inflammatory burden and may develop a proteinlosing enteropathy, leading to an accelerated consumption, exces­
sive fecal wasting, and low serum concentrations of infliximab.
Given a clear exposure–response relationship for infliximab in
patients with IBD, intensive infliximab dosing regimens have been
used in these patients.
Adalimumab (ADA) is a recombinant human monoclonal IgG1
antibody containing only human peptide sequences and is injected
subcutaneously. ADA binds TNF and neutralizes its function by
blocking the interaction between TNF and its cell-surface receptor.
Therefore, it seems to have a similar mechanism of action to inflix­
imab but with less immunogenicity. ADA is approved for treatment
of moderate to severe CD and UC. For CD and UC, results with ADA
are better in patients who are naïve to anti-TNF than in patients who
have previously been treated with infliximab. In clinical practice, the
remission rate in both CD and UC patients taking ADA increases
with a dose increase to 40 mg weekly instead of every other week.
Certolizumab pegol is a pegylated form of an anti-TNF Fab
portion of an antibody administered SC once monthly. SC cer­
tolizumab pegol is effective for induction of clinical response in
patients with active inflammatory CD.
Golimumab is another fully human IgG1 antibody against TNF-α
and is currently approved for the treatment of moderately to severely
active UC. Like ADA and certolizumab, golimumab is injected SC.
PART 10
Disorders of the Gastrointestinal System
Side Effects of Anti-TNF Therapies 
Development of Antibodies and Drug Levels 
The develop­
ment of antibodies to infliximab is associated with an increased risk
of infusion reactions and a decreased response to treatment. Cur­
rent practice does not include giving on-demand or episodic infu­
sions in contrast to scheduled periodic infusions because patients
are most likely to develop antibodies. Anti-infliximab antibodies
are generally present when the quality of response or the response
duration to infliximab infusion decreases. Commercial assays can
detect both infliximab and ADA antibodies and measure trough
levels to determine optimal dosing. If a patient has high antiinfliximab antibodies and a low trough level of infliximab, it is best
to switch to another anti-TNF therapy. If a patient has a therapeutic
anti-TNF level and active inflammatory symptoms, the drug should
be switched to a different class of biologic. Most acute infusion reac­
tions and serum sickness can be managed with glucocorticoids and
antihistamines. Some reactions can be serious and would neces­
sitate a change in therapy, especially if a patient has anti-infliximab
antibodies. It is now common practice to add an immunomodula­
tor such as azathioprine, MP, or MTX to anti-TNF therapy to help
prevent antibody formation.
Non-Hodgkin’s Lymphoma (NHL) 
The baseline risk of NHL in
CD patients is 2 in 10,000, slightly higher than in the general popu­
lation. Azathioprine and/or MP therapy increases the risk to ~4 in
10,000. It is difficult to assess whether anti-TNF medications are
associated with lymphoma because many patients are also receiving
thiopurines. After adjustment for co-treatments, no excess risk of
lymphoma was found in a Danish study of a cohort of IBD patients
exposed to anti-TNF medications.
Hepatosplenic T-Cell Lymphoma (HSTCL) 
HSTCL is a nearly
universally fatal lymphoma in patients with or without CD. In
patients with CD, a total of 37 unique cases have been reported.
Eighty-six percent of the patients were male, and the median age
was 26 years. Patients had CD for a mean of 10 years before the
diagnosis of HSTCL. Thirty-six patients had used either MP or
azathioprine, and 28 patients had used infliximab.
Skin Lesions 
New-onset psoriasiform skin lesions develop in
nearly 5% of IBD patients treated with anti-TNF therapy. Most
often, these can be treated topically, and occasionally, anti-TNF
therapy must be decreased, switched, or stopped. Patients with IBD
may have a slight, unexplained, intrinsic higher risk of developing
melanoma. The risk of melanoma is increased almost twofold with
anti-TNF and not thiopurine use. The risk of nonmelanoma skin
cancer is increased with thiopurines and biologics, especially with
≥1 year of follow-up. Patients on these medications should have a
skin check at least once a year.
Infections 
All of the anti-TNF drugs are associated with an
increased risk of infections, particularly reactivation of latent tuber­
culosis and opportunistic fungal infections including disseminated
histoplasmosis and coccidioidomycosis. Patients should have a
purified protein derivative (PPD) or a QuantiFERON-TB Gold test
before initiation of anti-TNF therapy. Patients >65 years old have
a higher rate of infections and death on infliximab or ADA than
those <65 years old.
Other 
Acute liver injury due to reactivation of hepatitis B virus
and to autoimmune effects and cholestasis has been reported.
Rarely, infliximab and the other anti-TNF drugs have been associ­
ated with optic neuritis, seizures, new onset or exacerbation of clini­
cal symptoms, and radiographic evidence of central nervous system
demyelinating disorders, including multiple sclerosis. They may
exacerbate symptoms in patients with New York Heart Association
functional class III/IV heart failure.
ANTI-INTEGRINS
Integrins are expressed on the cell surface of leukocytes and serve as
mediators of leukocyte adhesion to vascular endothelium and entry
into the intestines. α4-Integrin along with its β1 or β7 subunit inter­
act with endothelial ligands termed adhesion molecules. Interaction
between α4β7 and mucosal addressin cellular adhesion molecule
(MAdCAM-1) is important in lymphocyte trafficking to gut mucosa.
Natalizumab is a recombinant humanized IgG4 antibody against
α4-integrin and is effective in induction and maintenance of patients
with CD. However, natalizumab is no longer used for CD due to the
risk of progressive multifocal leukoencephalopathy (PML) and the
development of alternative biologic and small-molecule therapies.
Vedolizumab (VDZ), another leukocyte trafficking inhibitor, is
a monoclonal antibody directed against α4β7-integrin specifically
and has the ability to convey gut-selective immunosuppression.
Unlike natalizumab, it inhibits adhesion of a discrete gut-homing
subset of T lymphocytes to MAdCAM-1, but not to vascular adhe­
sion molecule 1. VDZ decreases GI inflammation without inhibit­
ing systemic immune responses or affecting T-cell trafficking to
the central nervous system. It may be prescribed as a first-line
biologic or after failure of a TNF antagonist in patients with CD or
UC. Vedolizumab is more effective than adalimumab for first-line
biologic therapy in moderate to severe UC.
Ustekinumab, a fully human IgG1 monoclonal antibody, blocks
the biologic activity of IL-12 and IL-23 through their common p40
subunit by inhibiting the interaction of these cytokines with their
receptors on T cells, natural killer cells, and antigen-presenting
cells. It is as equally effective as adalimumab for first-line biologic
therapy for moderate to severe CD and is another option for treat­
ment of moderate to severe UC. It is particularly appealing for use
in patients with concomitant psoriatic arthritis.
Risankizumab is a monoclonal antibody that selectively binds
to the IL-23 p19 subunit, inhibiting its interaction with the IL-23R
complex It is the first selective IL-23 inhibitor approved for moder­
ate to severe CD and provides an additional therapeutic option for
patients, particularly those who have been previously treated with
other advanced IBD therapies.
SMALL MOLECULES
Small molecules (drugs with molecular weight <1 kDa) are a new
class of orally administered medications developed for IBD that lack
the immunogenicity associated with monoclonal antibodies. The
advantage of small molecules is their ability to diffuse through cell
membranes into the intracellular space and inhibit cytokine signaling
pathways. This mechanism of action may be more efficacious com­
pared to monoclonal antibodies that inhibit specific targets because
several cytokine pathways are involved in IBD pathogenesis and
inhibiting numerous cytokines may be synergistic. A key regulatory
pathway of many cytokines associated with IBD is the JAK/STAT
pathway that activates transcription and translation of proteins that
mediate the immune response. Janus kinase (JAK) is a family of intra­
cellular, nonreceptor tyrosine kinases that regulate cytokine signal­
ing via the JAK/STAT pathway, ultimately suppressing the immune
response and inflammation. The JAK family members include JAK1,
JAK2, JAK3, and tyrosine kinase 2 (TYK2).
Tofacitinib is a reversible and competitive JAK inhibitor used for
the treatment of moderate to severe UC refractory to conventional
therapy. It competes with ATP to bind to the ATP-docking site of the
kinase domain of JAK. By competing with ATP, tofacitinib inhibits
phosphorylation and activation of JAK, leading to downstream
reduction of cytokine production and alteration of the immune
response. Although tofacitinib is a pan-JAK inhibitor, it has higher
specificity for JAK1 and JAK3 than for JAK2 and TYK2. The panJAK inhibition is concerning for adverse events and overall safety.
Upadacitinib is an oral, selective, and reversible JAK inhibitor
that is approved for both UC and CD. It potently inhibits JAK1 and
is less potent against the other isoforms, JAK2, JAK3, and TYK2.
Upadacitinib demonstrates rapid clinical and endoscopic improve­
ment at the end of induction, which is sustained to the end of main­
tenance, with a positive benefit/risk profile.
The FDA review concluded that there is an increased risk of
serious adverse events including heart attack, stroke, cancer, blood
clots, and death in patients who are prescribed JAK inhibitors.
Patients who are at risk for cardiovascular disease, have a history of
blood clots, are current or past smokers, and/or are over the age of
50 should consider alternative therapies. The risk of herpes zoster is
higher with JAK inhibitors, and patients should receive the shingles
vaccine (Shingrix). Tofacitinib and upadacitinib are FDA approved
for use as second-line agents after failure of anti-TNF therapy.
Ozanimod is a potent sphingosine-1-phosphate (S1P1) receptor
modulator that binds selectively with high affinity to the S1P recep­
tor subtypes S1P1 and S1P5, both of which are involved in immune
regulation. By preventing trafficking of disease-exacerbating lym­
phocytes from the lymph nodes to the gut, ozanimod may provide
immunomodulatory effects and moderate disease processes.
Ozanimod is approved for the treatment of moderate to severe
ulcerative colitis. It is administered as a daily capsule.
The biologic and small-molecule therapies used in daily practice
are detailed in Table 337-9.
NUTRITIONAL THERAPIES
Diet has long been thought to contribute to the pathogenesis of IBD
and may also be an avenue for managing disease activity. Diet plays
a significant role in shaping the gut microbiome, and dietary com­
ponents may interact with the microbiome and stimulate a mucosal
immune response. In fact, active CD responds to exclusive enteral
nutrition (EEN) or bowel rest with TPN, interventions as effective
as glucocorticoids in inducing remission but not as effective for
maintenance therapy. In contrast to CD, active UC is not effectively
treated by elemental diets or TPN.
Dietary approaches to maintenance therapy in CD have largely
been adapted from epidemiologic studies; however, significant
heterogeneity is noted among research study outcomes. In general,
low fiber, refined carbohydrates (especially sweetened beverages),
animal fats, red meat, and processed meat have been associated with
onset of IBD. Therefore, the overall dietary approach is to maximize
fiber intake, particularly from fruits and vegetables, and to limit
consumption of higher-risk foods. Several defined diets adhere
to these principles with some variation, including the Mediterra­
nean diet pattern, specific carbohydrate diet, semi-vegetarian diet,
and IBD anti-inflammatory diet (IBD-AID). However, it remains
unclear whether diet studies will eventually lead to evidence-based
nutrition guidelines.
Standard medical management of UC and CD is shown in
Fig. 337-12.
SURGICAL THERAPY
Ulcerative Colitis  Nearly one-half of patients with extensive
chronic UC undergo surgery within the first 10 years of their illness.
The indications for surgery are listed in Table 337-10. Morbidity is
~20% for elective, 30% for urgent, and 40% for emergency procto­
colectomy. The risks are primarily hemorrhage, contamination and
sepsis, and neural injury. The operation of choice is an IPAA.
Because UC is a mucosal disease, the rectal mucosa can be dis­
sected and removed down to the dentate line of the anus or ~2 cm
proximal to this landmark. The ileum is fashioned into a pouch
that serves as a neorectum. This ileal pouch is then sutured cir­
cumferentially to the anus in an end-to-end fashion. If performed
carefully, this operation preserves the anal sphincter and maintains
continence. The overall operative morbidity is 10%, with the major
complication being bowel obstruction. Pouch failure necessitating
conversion to permanent ileostomy occurs in 5–10% of patients.
Some inflamed rectal mucosa is usually left behind, and thus,
endoscopic surveillance is necessary. Primary dysplasia of the ileal
mucosa of the pouch has occurred rarely.
Patients with IPAA usually have ~6–10 bowel movements a
day. On validated quality-of-life indices, they report better per­
formance in sports and sexual activities than ileostomy patients.
The most frequent complication of IPAA is pouchitis in ~30–50%
of patients with UC. This syndrome consists of increased stool
frequency, watery stools, cramping, urgency, nocturnal leakage of
stool, arthralgias, malaise, and fever. Pouch biopsies may distin­
guish true pouchitis from underlying CD. Although pouchitis usu­
ally responds to antibiotics, 3–5% of patients remain refractory and
may require chronic antibiotic therapy, biologics, or even pouch
removal.
CHAPTER 337
Inflammatory Bowel Disease
Crohn’s Disease  The majority of patients with CD will require
at least one operation in their lifetime. The need for surgery is
related to duration of disease and the site of involvement. Patients
with small-bowel disease have an 80% chance of requiring surgery.
Those with colitis alone have a 50% chance. Surgery is an option
only when medical treatment has failed or complications dictate its
necessity. The indications for surgery are shown in Table 337-10.
Small-Intestinal Disease 
Because CD is chronic and recur­
rent, with no clear surgical cure, as little intestine as possible is
resected. Current surgical alternatives for treatment of obstructing
CD include resection of the diseased segment and strictureplasty.
Surgical resection of the diseased segment is the most frequently
performed operation, and in most cases, primary anastomosis can
be done to restore continuity. An end-to-end anastomosis may
provide the best opportunity for an optimal functional outcome,
compared to an antiperistaltic side-to-side anastomosis, which cre­
ates a functional block to motility leading to distention and pain
at the anastomotic site in a subgroup of patients. If much of the
small bowel has already been resected and the strictures are short,
with intervening areas of normal mucosa, strictureplasties should
be done to avoid a functionally insufficient length of bowel. The
strictured area of intestine is incised longitudinally and the incision
sutured transversely, thus widening the narrowed area. Complica­
tions of strictureplasty include prolonged ileus, hemorrhage, fistula,
abscess, leak, and restricture.
Risk factors for early recurrence of disease include cigarette
smoking, penetrating disease (internal fistulas, abscesses, or other
evidence of penetration through the wall of the bowel), early recur­
rence since a previous surgery, multiple surgeries, and a young age
at the time of the first surgery. Aggressive postoperative treatment
with biologics should be considered for this group of patients. It is
also recommended to evaluate for endoscopic recurrence of CD via
a colonoscopy, if possible, 3–6 months after surgery.
TABLE 337-9  Biologic and Small-Molecule Agents in the Treatment of Inflammatory Bowel Disease
MEDICATION
DOSAGE
INDICATION
SERIOUS TOXICITIES
Infliximab
5 mg/kg at 0, 2, and
6 weeks, then every
8 weeks; may increase
dose to 10 mg/kg every
4 weeks depending on
trough levels
Intensive dosing
for hospitalized
corticosteroidrefractory patients
Moderate to severe
Crohn’s disease and
ulcerative colitis
Fistulizing Crohn’s
disease
Increased risk of infections (bacterial
and fungal), tuberculosis (TB)
reactivation, hepatitis B reactivation,
lymphoma (controversial), psoriasis,
melanoma and nonmelanoma skin
cancers, drug-induced lupus
Contraindicated in multiple sclerosis,
class III/IV congestive heart failure
Adalimumab
160 mg day 0, 80 mg
day 14 and then 40 mg
every 14 days; may
increase to 40 mg
every 7 days depending
on trough levels
Moderate to severe
Crohn’s disease and
ulcerative colitis
Fistulizing Crohn’s
disease
As above
Injection site
reactions (better
with citrate-free
preparation)
Certolizumab
400 mg on days 0 and
14, then 400 mg every
28 days
Moderate to severe
Crohn’s disease
As above
As above
As above
Golimumab
200 mg on day 0, 100
mg on day 14, then 100
mg every 28 days
Moderate to severe
ulcerative colitis
As above
As above
As above
Vedolizumab
300 mg at 0, 2, and
6 weeks, then every
8 weeks; may increase
dose to 300 mg every
4 weeks
Moderate to severe
ulcerative colitis
(more effective than
adalimumab as firstline therapy in one
study) Moderate
to severe Crohn’s
disease.
No increased risk of serious systemic
or opportunistic infections
No increased risk of malignancy
PART 10
Disorders of the Gastrointestinal System
Ustekinumab
6 mg/kg IV, then 90 mg
every 8 weeks; may
increase dose to 90 mg
every 4 weeks
Moderate to severe
Crohn’s disease and
ulcerative colitis
Reversible posterior
leukoencephalopathy syndrome
(presents with headaches, seizures,
confusion, and visual disturbances),
anaphylaxis, and angioedema
Risankizumab
600 mg IV every 4
weeks × 3 doses then
180 mg or 360 mg SC
every 8 weeks
Moderate to severe
Crohn’s disease
Reversible posterior
leukoencephalopathy syndrome
(presents with headaches, seizures,
confusion, and visual disturbances),
anaphylaxis, and angioedema
Tofacitinib
10 mg bid; can
decrease to 5 mg
bid when patient in
remission
Moderate to severe
ulcerative colitis
Increased risk of heart attack,
stroke, cancer, blood clots, and
death. Patients who are at risk for
cardiovascular disease, are current
or past smokers, and/or are over the
age of 50 should consider alternative
therapies.
Increased risk of viral infections,
including herpes zoster, and bacterial
and invasive fungal infections
Upadacitinib
45 mg PO qd × 8
weeks for ulcerative
colitis; 45 mg PO qd ×
12 weeks for Crohn’s
disease, then 15 or 30
mg qd
Moderate to severe
ulcerative colitis or
Crohn’s disease
Increased risk of heart attack,
stroke, cancer, blood clots, and
death. Patients who are at risk for
cardiovascular disease, are current
or past smokers, and/or are over the
age of 50 should consider alternative
therapies.
Increased risk of viral infections,
including herpes zoster, and bacterial
and invasive fungal infections
Ozanimod
0.23 mg PO days 1–4,
0.46 mg PO days 5–7,
0.92 mg PO thereafter
Moderate to severe
UC
Bradycardia, atrioventricular
conduction delay, increased blood
pressure, macular edema, posterior
reversible encephalopathy syndrome
OTHER COMMON
SIDE EFFECTS
TESTING
Infusion reactions Prior to infusion:
TB testing
Hepatitis B testing (HBsAb, HBsAg, HBcAb)
Maintenance:
Skin check yearly
Influenza, pneumococcal vaccinations
Hepatitis B vaccine if not immune
As above
Nasopharyngitis,
headache,
arthralgias,
nausea
Prior to infusion:
TB testing
Hepatitis B testing (HBsAb, HBsAg, HBcAb)
Maintenance:
Influenza, pneumococcal vaccinations
Hepatitis B vaccine if not immune
Nasopharyngitis,
upper respiratory
tract infection,
fatigue, headache
Prior to infusion:
TB testing
Hepatitis B testing (HBsAb, HBsAg, HBcAb)
Maintenance:
Influenza, pneumococcal vaccinations
Hepatitis B vaccine if not immune
Nasopharyngitis,
upper respiratory
tract infection,
fatigue, headache
TB testing
Hepatitis B testing (HBsAb, HBsAg, HBcAb)
Maintenance:
Influenza, pneumococcal vaccinations
Hepatitis B vaccine if not immune
Elevated lipids,
neutropenia,
anemia, elevated
liver enzymes
Prior to infusion:
First dose of Shingrix recommended, TB
testing
Hepatitis B testing (HBsAb, HBsAg, HBcAb)
Maintenance:
Influenza, pneumococcal vaccinations
Hepatitis B vaccine if not immune
Elevated lipids,
neutropenia,
anemia, elevated
liver enzymes
Prior to infusion:
First dose of Shingrix recommended, TB
testing
Hepatitis B testing (HBsAb, HBsAg, HBcAb)
Maintenance:
Influenza, pneumococcal vaccinations
Hepatitis B vaccine if not immune
Infections,
elevated liver
enzymes
Prior to infusion:
First dose of Shingrix recommended, TB
testing
Hepatitis B testing (HBsAb, HBsAg, HBcAb)
Maintenance:
Influenza, pneumococcal vaccinations
Hepatitis B vaccine if not immune
Tofacitinib/
upadacitinib
Biologic +/– MP/AZA/
MTX or ozanimod
Prednisone oral
(induction of remission only)
Hydrocortisone rectal
Budesonide oral and/or rectal
5-ASA oral and/or rectal
Mild to Moderate Ulcerative Colitis
Moderate to Severe Ulcerative Colitis
Prednisone oral (induction of remission only)
Total
parenteral or
exclusive enteral
nutrition and bowel
rest
Mild to Moderate Crohn’s Disease
Upadacitinib
Biologic +/– MP/AZA/MTX
Hydrocortisone or solumedrol IV
(induction of remission only)
Prednisone oral (induction of remission only)
Moderate to Severe Crohn’s Disease
Fistulizing Crohn’s Disease
FIGURE 337-12  Medical management of inflammatory bowel disease. 5-ASA, 5-aminosalicylic acid; AZA, azathioprine; MP, mercaptopurine; MTX, methotrexate.
Colorectal Disease 
A greater percentage of patients with
Crohn’s colitis require surgery for intractability, fulminant disease,
and anorectal disease. Several alternatives are available, ranging
from the use of a temporary loop ileostomy to resection of segments
of diseased colon or even the entire colon and rectum. For patients
with segmental involvement, segmental colon resection with pri­
mary anastomosis can be performed. In 20–25% of patients with
extensive colitis, the rectum is spared sufficiently to consider rectal
preservation. Most surgeons believe that an IPAA is contraindicated
in CD due to the high incidence of pouch failure. A diverting colos­
tomy may help heal severe perianal disease or rectovaginal fistulas,
but disease almost always recurs with reanastomosis. These patients
often require a total proctocolectomy and ileostomy.
TABLE 337-10  Indications for Surgery
ULCERATIVE COLITIS
CROHN’S DISEASE
Intractable disease
Small Intestine
Fulminant disease
  Stricture and obstruction
Toxic megacolon
  unresponsive to medical therapy
Colonic perforation
  Massive hemorrhage
Massive colonic hemorrhage
  Refractory fistula
Extracolonic disease
  Abscess
Colonic obstruction
Colon and rectum
Colon cancer prophylaxis
  Intractable disease
Colon dysplasia or cancer
  Fulminant disease
 
  Perianal disease unresponsive to
medical therapy
 
  Refractory fistula
 
  Colonic obstruction
 
  Cancer prophylaxis
 
  Colon dysplasia or cancer
Cyclosporine IV
Tofacitinib/upadacitinib
Biologic +/– MP/AZA/MTX
Hydrocortisone or solumedrol IV
(induction of remission only)
Upadacitinib
Prednisone oral (induction of remission only)
Biologic +/– MP/AZA/MTX
Total
parenteral
nutrition and
bowel rest
Budesonide oral
Biologic +/– MP/AZA/MTX
Abscess drainage and antibiotics
CHAPTER 337
Inflammatory Bowel Disease
■
■IBD AND PREGNANCY
Patients with quiescent UC and CD have normal fertility rates; the
fallopian tubes can be scarred by the inflammatory process of CD,
especially on the right side because of the proximity of the terminal
ileum. In addition, perirectal, perineal, and rectovaginal abscesses and
fistulas as well as pelvic surgery can result in dyspareunia. Infertility
in men can be caused by sulfasalazine but reverses when treatment is
stopped. Women with an IPAA have decreased fertility due to scarring
or occlusion of the fallopian tubes secondary to pelvic inflammation
and adhesions, although studies have shown that fertility is improved
with laparoscopic versus open IPAA.
Mild or quiescent UC or CD has no effect on birth outcomes. The
courses of CD and UC during pregnancy mostly correlate with disease
activity at the time of conception. Patients should be in remission
for 3–6 months before conceiving. Most CD patients can deliver vagi­
nally, but cesarean delivery may be the preferred route of delivery for
patients with anorectal and perirectal abscesses and fistulas to reduce
the likelihood of fistulas developing or extending into the episiotomy
scar. Unless they desire multiple children, UC patients with an IPAA
may consider a cesarean delivery due to an increased risk of future fecal
incontinence.
Sulfasalazine and all mesalamines are safe for use in pregnancy and
nursing with the caveat that additional folate supplementation must be
given with sulfasalazine. Topical 5-ASA agents are safe during pregnancy
and nursing. Glucocorticoids are generally safe for use during pregnancy
and are indicated for patients with moderate to severe disease activity.
The amount of glucocorticoids received by the nursing infant is minimal.
The safest antibiotics to use for CD in pregnancy for short periods of
time (weeks, not months) are ampicillin and cephalosporins. Metroni­
dazole can be used in the second or third trimester. Ciprofloxacin causes
cartilage lesions in immature animals and should be avoided because of
the absence of data on its effects on growth and development in humans.
MP and azathioprine pose minimal or no risk during pregnancy.
Breast milk has been shown to contain negligible levels of MP/azathio­
prine when measured in a limited number of patients.
MTX is teratogenic and should be discontinued at least 3 months
before conception.
In a large prospective and multiple retrospective studies, no increased
risk of stillbirths, miscarriages, or spontaneous abortions was seen with
infliximab, ADA, certolizumab, vedolizumab, or ustekinumab. These
biologics, with the exception of certolizumab, are IgG1 antibodies and
are actively transported across the placenta in the late second and third
trimesters. Infants can have serum levels of infliximab, ADA, vedoli­
zumab, and ustekinumab well into the first year of life, and live vac­
cines should be avoided until 12 months of age. Certolizumab crosses
the placenta by passive diffusion, and infant serum and cord blood
levels are minimal. The biologics are safe in nursing. Miniscule levels
of infliximab, ADA, certolizumab, vedolizumab, and ustekinumab
have been reported in breast milk. These levels are of no clinical sig­
nificance. It is recommended that drugs should not be switched during
pregnancy unless necessitated by the medical condition of the IBD.
Tofacitinib and upadacitinib should not be used during pregnancy.
Animal studies show teratogenic effects with both of these drugs, and
data in humans are limited. A washout period of at least 4 weeks is
recommended before conception. Surgery in UC should be performed
only for emergency indications, including severe hemorrhage, perfora­
tion, and megacolon refractory to medical therapy. Total colectomy
and ileostomy carry a 50% risk of postoperative spontaneous abortion.
The best time to perform surgery is in the second trimester if necessary.
Patients with IPAAs have increased nighttime stool frequency during
pregnancy that resolves postpartum. Transient small-bowel obstruc­
tion or ileus has been noted in up to 8% of patients with ileostomies.
CANCER IN IBD
PART 10
Disorders of the Gastrointestinal System
■
■ULCERATIVE COLITIS
Patients with long-standing UC are at increased risk for developing
colonic epithelial dysplasia and carcinoma (Fig. 337-13).
The risk of neoplasia in chronic UC increases with duration and
extent of disease. In contrast to the relatively high risk in one large
meta-analysis (2% after 10 years, 8% after 20 years, and 18% after 30 years
of disease), a decrease in the risk of colorectal cancer has been noted
over time potentially due to better control of inflammation and better
colonoscopic surveillance. The rates of colon cancer are still about
1.5–2 times higher than in the general population, and colonoscopic
surveillance is the standard of care.
Annual or biennial colonoscopy with multiple biopsies is recom­
mended for patients with >8–10 years of extensive colitis (greater
than one-third of the colon involved) or 12–15 years of proctosig­
moiditis (less than one-third but more than just the rectum) and has
been widely used to screen and survey for subsequent dysplasia and
carcinoma. International guideline societies have recommended chro­
moendoscopy for dysplasia surveillance in IBD. Chromoendoscopy
enhances the visualization of the surface and pit pattern of the mucosa,
FIGURE 337-13  Medium-power view of low-grade dysplasia in a patient
with chronic ulcerative colitis. Low-grade dysplastic crypts are interspersed
among regenerating crypts. (Courtesy of Dr. R. Odze, Division of Gastrointestinal
Pathology, Department of Pathology, Brigham and Women’s Hospital, Boston,
Massachusetts; with permission.)
as well as borders of lesions, in order to better define areas of dysplasia
compared to standard-definition white light endoscopy. In real-life
settings, the practice has been to use standard-definition white light
endoscopy with surveillance biopsies in patients with chronic colitis at
average risk and chromoendoscopy in higher-risk patients including
those with a history of dysplasia, PSC, or family history of colorectal
cancer.
Risk factors for cancer in UC include long-duration disease, exten­
sive disease, family history of colon cancer, PSC, a colon stricture, and
the presence of postinflammatory pseudopolyps on colonoscopy.
■
■CROHN’S DISEASE
Risk factors for developing cancer in Crohn’s colitis are long-duration
and extensive disease, bypassed colon segments, colon strictures, PSC,
and family history of colon cancer. The cancer risks in CD and UC are
probably equivalent for similar extent and duration of disease. In the
CESAME study, a prospective observational cohort of IBD patients
in France, the standardized incidence ratios of colorectal cancer were
2.2 for all IBD patients (95% CI, 1.5–3.0; p < .001) and 7.0 for patients
with long-standing extensive colitis (both Crohn’s and UC) (95% CI,
4.4–10.5; p < .001). Thus, the same endoscopic surveillance strategy
used for UC is recommended for patients with chronic Crohn’s colitis.
A pediatric colonoscope can be used to pass narrow strictures in CD
patients, but surgery should be considered in symptomatic patients
with impassable strictures.
■
■MANAGEMENT OF DYSPLASIA AND CANCER
Dysplasia can be flat or polypoid. If flat high-grade dysplasia is
encountered on colonoscopic surveillance, the usual treatment is
colectomy for UC and either colectomy or segmental resection for
CD. If flat low-grade dysplasia is found (Fig. 337-13), most inves­
tigators recommend immediate colectomy. Adenomas may occur
coincidently in UC and CD patients with chronic colitis and can be
removed endoscopically provided that biopsies of the surrounding
mucosa are free of dysplasia.
IBD patients are also at greater risk for other malignancies. Patients
with CD may have an increased risk of NHL, leukemia, and myelodys­
plastic syndromes. Severe, chronic, complicated perianal disease in CD
patients may be associated with an increased risk of cancer in the lower
rectum and anal canal (squamous cell cancers). Although the absolute
risk of small-bowel adenocarcinoma in CD is low (2.2% at 25 years in
one study), patients with long-standing, extensive, small-bowel disease
should be considered for screening.
COVID-19 AND IBD
COVID-19, caused by SARS-CoV-2, was first reported in December
2019 and has rapidly spread throughout the world, leading to an inter­
national pandemic. Glucocorticoids, immunomodulators (thiopurines,
MTX), biologics, and JAK inhibitors, commonly used to treat IBD,
are associated with higher rates of serious viral and bacterial infec­
tions, and patients with IBD using these medications are potentially
at increased risk of a serious COVID-19 infection. Yet, it is also pos­
sible that some forms of immune suppression may blunt the excessive
immune response/cytokine storm characteristic of severe COVID-19
infection and consequently reduce mortality. Using data from the Sur­
veillance Epidemiology of Coronavirus Under Research Exclusion for
Inflammatory Bowel Disease, it was found that increasing age (adjusted
OR 1.04; 95% CI, 1.01–1.02), two or more comorbidities (adjusted
OR 2.9; 95% CI, 1.1–7.8), and systemic glucocorticoids (adjusted OR
6.9; 95% CI, 2.3–20.5) are associated with severe COVID-19 in IBD
patients. Anti-TNF treatment was not associated with severe COVID-19
(adjusted OR 0.9; 95% CI, 0.4–2.2).
■
■FURTHER READING
Barberio B et al: Efficacy of biological therapies and small molecules
in induction and maintenance of remission in luminal Crohn’s dis­
ease: Systematic review and network meta-analysis. Gut 72:264, 2023.
Chang JT: Pathophysiology of inflammatory bowel disease. N Eng J
Med 383:2652, 2020.

08 - 338 Irritable Bowel Syndrome
338 Irritable Bowel Syndrome
D’haens G et al: Risankizumab as induction therapy for Crohn’s dis­
ease: Results from the phase 3 ADVANCE and MOTIVATE induc­
tion trials. Lancet 399:2015, 2022.
Graham DB, Xavier RJ: Pathway paradigms revealed from the genet­
ics of inflammatory bowel disease. Nature 578:527, 2020.
Levine A et al: Crohn’s disease exclusion diet plus partial enteral nutri­
tion induces sustained remission in a randomized controlled trial.
Gastroenterology 157:440, 2019.
Lewis JD et al: A randomized trial comparing the specific carbohy­
drate diet to a Mediterranean diet in adults with Crohn’s disease.
Gastroenterology 161:837, 2021.
Lewis JD et al: Incidence, prevalence, and racial and ethnic distribu­
tion of inflammatory bowel disease in the United States. Gastroenter­
ology 165:1197, 2023.
Mahadevan U et al: Pregnancy and neonatal outcomes after fetal
exposure to biologics and thiopurines among women with inflamma­
tory bowel disease. Gastroenterology 160:1131, 2021.
Ng SC et al: Worldwide incidence and prevalence of inflammatory
bowel disease in the 21st century: A systematic review of populationbased studies. Lancet 390:2769, 2017.
Sands BE: Ustekinumab versus adalimumab for induction and mainte­
nance therapy in biologic-naive patients with moderately to severely
active Crohn’s disease: A multicentre, randomised, double-blind,
parallel-group, phase 3b trial. Lancet 399:2200, 2022.
Sands BE et al: VARSITY Study Group. Vedolizumab versus adali­
mumab for moderate-to-severe ulcerative colitis. N Engl J Med
381:1215, 2019.
Chung Owyang, Prashant Singh
Irritable Bowel
Syndrome
Irritable bowel syndrome (IBS) is a functional bowel disorder charac­
terized by abdominal pain or discomfort and altered bowel habits in
the absence of detectable structural abnormalities. No clear diagnostic
markers exist for IBS; thus, the diagnosis of the disorder is based on
clinical presentation. In 2016, the Rome III criteria for the diagno­
sis of IBS were updated to Rome IV (Table 338-1). Throughout the
world, ~10% of adults and adolescents have symptoms consistent with
IBS. IBS symptoms fluctuate over time and often overlap with other
functional disorders such as fibromyalgia, headache, backache, and
genitourinary symptoms. Severity of symptoms varies and can signifi­
cantly impair quality of life, resulting in high direct and indirect health
care costs. Altered gastrointestinal (GI) motility, visceral hyperalgesia,
disturbance of brain–gut interaction, abnormal central processing,
autonomic and hormonal events, microbial dysbiosis, genetic and envi­
ronmental factors, and psychosocial disturbances are variably involved,
depending on the individual.
TABLE 338-1  Rome IV Diagnostic Criteria for Irritable Bowel Syndromea
Recurrent abdominal pain, on average, at least 1 day per week in the last
3 months, associated with ≥2 of the following criteria:
Related to defecation
Associated with a change in frequency of stool
Associated with a change in form (appearance) of stool
aCriteria fulfilled for the last 3 months with symptom onset at least 6 months prior to
diagnosis.
■
■CLINICAL FEATURES
IBS is a disorder that affects all ages, although most patients have their
first symptoms before age 45. Women are diagnosed with IBS two to
three times as often as men and make up 80% of the population with
severe IBS. As indicated in Table 338-1, abdominal pain is essential
for diagnosing IBS. This should be associated with defecation and/or
have its onset associated with a change in frequency or form of stool.
In comparison to Rome III, the Rome IV criteria are more stringent,
requiring abdominal pain to occur at a minimum of once a week and
eliminating “discomfort” as one of the criteria. Painless diarrhea or
constipation does not fulfill the diagnostic criteria to be classified as
IBS. Supportive symptoms that are not part of the diagnostic criteria
include straining, urgency or a feeling of incomplete defecation, pass­
ing mucus, and bloating.
Abdominal Pain 
According to the current IBS diagnostic criteria,
abdominal pain is a prerequisite clinical feature of IBS. Abdominal pain
in IBS is highly variable in intensity and location and can occur any­
where in the abdomen. It is frequently episodic and crampy, but it may
be superimposed on a background of constant ache. Pain may be mild
enough to be ignored, or it may interfere with daily activities. Despite
this, malnutrition due to inadequate caloric intake is exceedingly rare
with IBS. Sleep deprivation is also unusual because abdominal pain is
almost uniformly present only during waking hours. Pain is often exac­
erbated by eating or emotional stress and improved by the passage of
flatus or stools. In addition, female patients with IBS commonly report
worsening symptoms during the premenstrual and menstrual phases.
Altered Bowel Habits 
Alteration in bowel habits is the most con­
sistent clinical feature in IBS. Based on the predominant bowel type,
IBS can be classified into IBS with predominant constipation (IBS-C),
IBS with predominant diarrhea (IBS-D), and IBS with both mixed con­
stipation and diarrhea (IBS-M). However, bowel pattern subtypes are
highly unstable. In a patient population with ~33% prevalence rates of
IBS-D, IBS-C, and IBS-M forms, 75% of patients change subtypes and
29% switch between IBS-C and IBS-D over 1 year.
CHAPTER 338
Irritable Bowel Syndrome
Patients with IBS-C have symptoms of hard stools (often with
narrow caliber), infrequent stools (less than three bowel movements
per week), incomplete evacuation, straining, sensation of anal block­
age, and digitalization. On the other hand, those with IBS-D usually
have small volumes of loose stools (generally <200 mL). This is often
associated with fecal urgency, the passage of mucus, and in some cases
incontinence. Diarrhea may be aggravated by emotional stress or eat­
ing. Weight loss, nocturnal symptoms, and bleeding per rectum are
not common in patients with IBS and, when present, should warrant
evaluation for other causes.
Gas and Flatulence 
Patients with IBS frequently complain of
bloating, abdominal distention and increased belching or flatulence, all
of which they attribute to increased gas. Although some patients with
these symptoms actually may have a larger amount of gas, quantitative
measurements reveal that most patients who complain of increased
gas generate no more than a normal amount of intestinal gas. Most
patients with IBS have impaired transit and intolerance of intestinal
gas loads. In addition, patients with IBS tend to reflux gas from the
distal to the more proximal intestine, which may explain the belching.
Some patients with bloating may also experience visible distention with
increase in abdominal girth.
Upper GI Symptoms 
Between 25 and 50% of patients with IBS
complain of dyspepsia, heartburn, nausea, and vomiting. This suggests
that other areas of the gut apart from the colon may be involved. Pro­
longed ambulant recordings of small-bowel motility in patients with
IBS show a high incidence of abnormalities in the small bowel during
the diurnal (waking) period; nocturnal motor patterns are not different
from those of healthy controls. The overlap between dyspepsia and IBS
is high. The prevalence of IBS is higher among patients with dyspepsia
(31.7%) than among those who reported no symptoms of dyspepsia
(7.9%). Conversely, among patients with IBS, 55.6% reported symp­
toms of dyspepsia. In addition, the functional abdominal symptoms
can change over time. Those with predominant dyspepsia or IBS can
flux between the two. Thus, it is conceivable that functional dyspepsia
and IBS are two manifestations of a single, more extensive digestive
system disorder. Furthermore, IBS symptoms are prevalent in other
functional GI disorders such as noncardiac chest pain patients and
esophageal hypersensitivity.
■
■PATHOPHYSIOLOGY
The pathogenesis of IBS is poorly understood, although roles of
abnormal gut motor and sensory activity, central neural dysfunction,
psychological disturbances, mucosal inflammation, stress, and luminal
factors such as bile acid malabsorption and gut dysbiosis have been
proposed (Fig. 338-1).
GI Motor Abnormalities 
Studies of colonic myoelectrical and
motor activity under unstimulated conditions have not shown consis­
tent abnormalities in IBS. In contrast, colonic motor abnormalities are
more prominent under stimulated conditions in IBS. IBS patients may
exhibit increased rectosigmoid motor activity for up to 3 h after eating.
Similarly, inflation of rectal balloons both in IBS-D and IBS-C patients
leads to marked and prolonged distention-evoked contractile activ­
ity. Recordings from the transverse, descending, and sigmoid colon
showed that the motility index and peak amplitude of high-amplitude
propagating contractions (HAPCs) in IBS-D patients were greatly
increased compared to those in healthy subjects and were associated
with rapid colonic transit and accompanied by abdominal pain.
Visceral Hypersensitivity 
IBS patients frequently exhibit exag­
gerated sensory responses to visceral stimulation. The frequency of
perceptions of food intolerance is at least twofold more common than
in the general population. On the other hand, prolonged fasting in IBS
patients is often associated with significant improvement in symptoms.
Rectal balloon inflation produces nonpainful and painful sensations
at lower volumes in IBS patients than in healthy controls without
altering rectal tension, suggestive of visceral afferent dysfunction in
IBS. Similar studies show gastric and esophageal hypersensitivity in
patients with nonulcer dyspepsia and noncardiac chest pain, raising the
possibility that these conditions have a similar pathophysiologic basis.
Lipids lower the thresholds for the first sensation of gas, discomfort,
and pain in IBS patients. Hence, postprandial symptoms in IBS patients
may be explained in part by a nutrient-dependent exaggerated sensory
component of the gastrocolonic response. In contrast to enhanced gut
sensitivity, IBS patients do not exhibit heightened sensitivity elsewhere
in the body. Thus, the afferent pathway disturbances in IBS appear to
be selective for visceral innervation with sparing of somatic pathways.
The mechanisms responsible for visceral hypersensitivity are still
under investigation.
PART 10
Disorders of the Gastrointestinal System
“Leaky” gut
dysbiosis
Motility
abnormalities
Visceral
hypersensitivity
Brain–gut
interactions
• HPA axis
• Autonomic
dysfunction
IBS
Psychological
• Anxiety/panic
• Depression
• Somatization
Genetic factors
• Twin studies
• SERT polymorphisms
Bile acid
malabsorption
FIGURE 338-1  Pathophysiology of irritable bowel syndrome (IBS). The cause of IBS is likely
to be multifactorial. Patients often show evidence of visceral hypersensitivity and motility
abnormalities. Many IBS patients have increased anxiety and/or depression, and their symptoms
are often exacerbated by mental or physical stress, suggesting abnormal brain–gut interaction.
Genetic studies suggest a few IBS patients may have genetic abnormalities affecting the
serotonin transport system in the enteric nerves. Up to 30% of IBS patients may have bile acid
malabsorption. Gut dysbiosis and impaired mucosa permeability also have been reported in many
IBS patients. This may lead to subclinical mucosa inflammation. HPA, hypothalamic-pituitaryadrenal; SERT, serotonin reuptake transporter.
Central Neural Dysregulation 
The role of central nervous sys­
tem (CNS) factors in the pathogenesis of IBS is strongly suggested by
the clinical association of emotional disorders and stress with symptom
exacerbation and the therapeutic response to therapies that act on cere­
bral cortical sites. Functional brain imaging studies such as magnetic
resonance imaging (MRI) have shown that in response to distal colonic
stimulation, the mid-cingulate cortex—a brain region concerned with
attention processes and response selection—shows greater activation
in IBS patients. Modulation of this region is associated with changes
in the subjective unpleasantness of pain. In addition, IBS patients also
show preferential activation of the prefrontal lobe, which contains a
vigilance network within the brain that increases alertness. These may
represent a form of cerebral dysfunction, leading to hypervigilance of
visceral pain.
Abnormal Psychological Features 
Abnormal psychiatric fea­
tures are recorded in up to 80% of IBS patients, especially in referral
centers; however, no single psychiatric diagnosis predominates. Most
of these patients demonstrated exaggerated symptoms in response to
visceral distention, and this abnormality persists even after exclusion
of psychological factors.
Psychological factors influence pain thresholds in IBS patients, as
stress alters sensory thresholds. An association between prior sexual
or physical abuse and development of IBS has been reported. Clinical
studies suggest that IBS has a strong developmental component that
involves interactions of genetic and epigenetic factors early in life.
These may modulate brain networks related to emotional arousal and/
or central autonomic control, salience, and somatosensory integration.
Abuse is associated with greater pain reporting, psychological distress,
and poor health outcome. Brain functional MRI studies show greater
activation of the posterior and middle dorsal cingulate cortex, which
is implicated in affect processing in IBS patients with a past history of
sexual abuse.
Postinfectious IBS 
A systematic review found that >10% of
patients with infectious enteritis develop IBS later and the risk was
four-fold higher compared to individuals without infectious enteri­
tis. Conversely, about a third of IBS patients experienced an acute
“gastroenteritis-like” illness at the onset of their chronic IBS symp­
tomatology. This group of “postinfective” IBS occurs more commonly
in females and affects younger rather than older patients. Other risk
factors include clinical severity of infection (e.g., prolonged duration,
requiring antibiotics), type of infection (risk is higher for parasitic and
bacterial infection compared to viral), and presence of psychological
comorbidities. Increased rectal mucosal enteroendocrine cells and
T lymphocytes following Campylobacter enteritis may contribute to
postinfective IBS. Persistent bacterial dysbiosis following
an infection causing abnormal suppression of host proteo­
lytic activity leading to altered intestinal barrier has also
been implicated in a subset of these patients.
Immune Activation and Mucosal Inflammation 
Some patients with IBS display persistent signs of lowgrade mucosal inflammation with activated lymphocytes,
mast cells, and enhanced expression of proinflammatory
cytokines. Other studies also indicate that peripheral blood
mononuclear cells (PBMCs) from IBS patients show abnor­
mal release of proinflammatory cytokines such as interleu­
kin (IL) 6, IL-1β, and tumor necrosis factor (TNF). These
abnormalities may contribute to abnormal epithelial secre­
tion, increased permeability, and visceral hypersensitivity.
Located at the host-environment interface, mast cells are
in close proximity to sensory nerves. Electromicroscopic
evidence of mast cell activation is commonly observed in
the colonic mucosa of IBS patients. The proximity of acti­
vated mast cells to submucosal nerve fibers correlates with
the frequency and severity of abdominal pain in patients
with IBS. Other studies report that the colonic mucosa of
IBS patients releases increased amounts of mast cell media­
tors, including histamine, proteases, and prostaglandin E2.
These findings, together with the observation that marked excitation of
visceral sensory nerves innervating the colon occurs after exposure to
IBS mucosal supernatant, support a prominent role for mast cells in the
pathogenesis of visceral hypersensitivity. Increasing evidence suggests
that some members of the superfamily of transient receptor potential
(TRP) cation channels such as TRPV1 (vanilloid) channels are central
to the initiation and persistence of visceral hypersensitivity. Mucosal
inflammation can lead to increased expression and/or sensitization of
TRPV1. Enhanced expression of TRPV1 channels in the sensory neu­
rons of the gut has been observed in IBS, and such expression appears
to correlate with visceral hypersensitivity and abdominal pain.
Altered Intestinal Permeability  
The gut epithelial lining repre­
sents a semipermeable barrier that allows nutrient and water absorp­
tion while being restrictive against pathogenic molecules, toxins, and
bacteria. The intestinal barrier is impaired in a subset of patients with
IBS. This is primarily described in patients with IBS-D and postinfec­
tious IBS. It is unclear whether or not those with IBS-C and IBS-M also
have increased intestinal permeability. There appears to be a correla­
tion between altered intestinal permeability and symptom severity in
IBS. Studies have not only shown structural changes in barrier func­
tion, such as a decrease in gene and protein expression of tight junction
proteins, but also functional consequences, such as the increased pas­
sage of macromolecules including bacteria across the epithelial barrier
in Ussing chamber studies. Although the exact etiology of the altered
mucosal barrier is not well understood, several factors such as stress,
diet, and microbial dysbiosis appear to play a key role. Several studies
have shown a key role of mast cell activation in mediating barrier dys­
function due to stress and diet-related dysbiosis.
Altered Gut Flora 
IBS patients have fourfold increased odds of
having small-intestinal bacterial overgrowth compared to healthy con­
trols based on positive lactulose or glucose hydrogen breath test. This
finding, however, has been challenged by a number of other studies
that found no increased incidence of bacterial overgrowth based on
jejunal aspirate culture. An abnormal H2 breath test can occur because
of rapid small-bowel transit and may lead to erroneous interpretation.
It is unclear if an abnormal breath test can predict the response to
available treatment options such as nonselective antibiotics, and there­
fore, the role of testing for small-intestinal bacterial overgrowth in IBS
patients remains unclear.
Many, but not all, studies using culture-independent approaches
such as 16S rRNA gene-based analysis found significant differences
between the molecular profile of the fecal microbiota of IBS patients
and that of healthy subjects. Although an IBS-specific microbial sig­
nature has not yet been identified, but overall, the diversity of fecal
microbiome appears to be significantly lower in IBS patients compared
to healthy controls. However, the functional significance of this find­
ing is not well understood. Despite a lack of consensus on the exact
microbial difference between IBS patients and controls, IBS patients
generally had decreased proportions of the genera Bifidobacterium and
Faecalibacterium and increased abundance of family Enterobacteria­
ceae (phylum Proteobacteria), family Lactobacillaceae, and genus Bac­
teroides (phylum Bacteroidetes) (Fig. 338-2). Experiments with animal
IBS vs Controls
IBS
Healthy controls
Lactobacillaceae
Bacteroides
Enterobacteriaceae
Bifidobacterium
Faecalibacterium
FIGURE 338-2  Changes in gut microbiota among patients with irritable bowel
syndrome (IBS). (Adapted from R Pittayanon et al: Gastroenterology 157:97, 2019.)
models have shown that colonization of germ-free animals with micro­
biota from patients with IBS can induce visceral hypersensitivity, alter
gut-brain axis, impair intestinal permeability, and alter GI transit time,
indicating the importance and possible etiologic role of the microbiota
in IBS. Gut dysbiosis acting in concert with genetic susceptibility and
environmental insults may alter mucosal permeability and increase
antigen presentation to the immune cells in the lamina propria. This
may result in mast cell activation and altered enteric neuronal and
smooth-muscle function causing IBS symptoms. In addition, release of
cytokines and chemokines from mucosal inflammation may generate
extra-GI symptoms such as chronic fatigue, muscle pain, and anxiety
(Fig. 338-3).
Abnormal Serotonin Pathways 
The serotonin-containing
enterochromaffin cells in the colon are increased in a subset of IBS-D
patients compared to healthy individuals. Furthermore, postpran­
dial plasma serotonin levels were significantly higher in this group
of patients compared to healthy controls. Tryptophan hydroxylase 1
(TPH1) is the rate-limiting enzyme in enterochromaffin cell serotonin
biosynthesis, and functional TPH1 polymorphism has been shown to
be associated with IBS subtypes. In addition, gut microbes promote
colonic serotonin production through an effect of short-chain fatty
acids on enterochromaffin cells. In IBS patients, the expression of
mucosal serotonin reuptake transporter (SERT) is downregulated due
to gram-negative gut dysbiosis, leading to abnormal mucosal sero­
tonin levels in IBS. Because serotonin plays an important role in the
regulation of GI motility and visceral perception, the increased release
of serotonin may contribute to the postprandial symptoms of these
patients.
CHAPTER 338
Bile Acids 
Up to 25% of patients with IBS-D are found to have
evidence of idiopathic bile acid diarrhea. Mechanisms leading to
increased hepatic synthesis and colonic bile acid levels in idio­
pathic bile acid diarrhea include reduced synthesis of fibroblast
growth factor 19 (FGF-19) by the ileal mucosa or genetic variation
in proteins involved in negative feedback of bile acid synthesis.
Increased colonic bile acid levels in turn influence bowel habit by
accelerating colonic transit, intestinal permeability, and visceral
hypersensitivity.
Irritable Bowel Syndrome
APPROACH TO THE PATIENT
Irritable Bowel Syndrome
Because IBS is a disorder for which no pathognomonic abnor­
malities have been identified, its diagnosis relies on recognition of
positive clinical features and elimination of other organic diseases.
Symptom-based criteria have been developed for the purpose of
differentiating patients with IBS from those with organic diseases.
These include the Manning, Rome I, Rome II, Rome III, and
Rome IV criteria. Rome IV criteria for the diagnosis of IBS were
published in 2016 (Table 338-1) and defined IBS on the basis of
abdominal pain and altered bowel habits that occur with sufficient
frequency in affected patients. A careful history and physical
examination are frequently helpful in establishing the diagnosis.
Clinical features suggestive of IBS include recurrence of lower
abdominal pain with altered bowel habits over a period of time
without progressive deterioration, onset of symptoms during
periods of stress or emotional upset, absence of other systemic
symptoms such as fever and weight loss, and small-volume stool
without any evidence of blood.
On the other hand, the appearance of the disorder for the first
time in old age, progressive course from time of onset, rectal bleed­
ing, significant weight loss persistent diarrhea after a 48-h fast, and
presence of nocturnal diarrhea or steatorrheal stools argue against
the diagnosis of IBS.
Because the major symptoms of IBS—abdominal pain, abdominal
bloating, and alteration in bowel habits—are common complaints
of many GI organic disorders, the list of differential diagnoses
is a long one. The quality, location, and timing of pain may be
-Dysbiosis
-Genetic
susceptibility
-Environmental
insults
Altered
permeability
Systemic
cytokines
& chemokines
Extra-GI
symptoms
FIGURE 338-3  Gut dysbiosis and irritable bowel syndrome (IBS). Gut dysbiosis acting in concert with genetic and environmental factors may alter intestinal permeability
and increase antigen presentation, resulting in mast cell activation. Products of mast cell degranulation may alter neuronal and smooth-muscle function causing IBS
symptoms. The cytokines and chemokines generated from mucosal inflammation may cause symptoms such as fibromyalgia, chronic fatigue, and mood changes.
GI, gastrointestinal. (Adapted from NJ Talley, AA Fodor: Gastroenterology 141:1555, 2011.)
helpful to suggest specific disorders. Pain due to IBS that occurs
in the epigastric or periumbilical area must be differentiated from
biliary tract disease, peptic ulcer disorders, intestinal ischemia, and
carcinoma of the stomach and pancreas. If pain occurs mainly in
the lower abdomen, the possibility of diverticular disease of the
colon, inflammatory bowel disease (including ulcerative colitis and
Crohn’s disease), and carcinoma of the colon must be considered.
Postprandial pain accompanied by bloating, nausea, and vomiting
suggests gastroparesis or partial intestinal obstruction. Patients with
small intestinal bacteria overgrowth can present with abdominal
pain, nausea, and bloating. Intestinal infestation with Giardia lamblia
or other parasites may cause similar symptoms. When diarrhea is
the major complaint, the possibility of lactase deficiency, laxative
abuse, malabsorption, celiac sprue, hyperthyroidism, inflammatory
bowel disease (IBD), and infectious diarrhea must be ruled out. On
the other hand, constipation may be a side effect of many different
drugs, such as anticholinergic, antihypertensive, and antidepressant
medications. Endocrinopathies such as hypothyroidism and hypo­
parathyroidism must also be considered in the differential diagnosis
of constipation. In addition, acute intermittent porphyria and lead
poisoning may present in a fashion similar to IBS, with painful con­
stipation as the major complaint. These possibilities are suspected
on the basis of their clinical presentations and are confirmed by
appropriate serum and urine tests.
PART 10
Disorders of the Gastrointestinal System
Few tests are required for patients who have typical IBS symp­
toms and no alarm features. Unnecessary investigations may be
costly and even harmful. The American Gastroenterological Asso­
ciation has delineated factors to be considered when determining
the aggressiveness of the diagnostic evaluation. These include the
duration of symptoms, the change in symptoms over time, the age
and sex of the patient, the referral status of the patient, prior diag­
nostic studies, a family history of colorectal malignancy, and the
degree of psychosocial dysfunction. Thus, a younger individual
with mild symptoms requires a minimal diagnostic evaluation,
while an older person or an individual with rapidly progressive
symptoms should undergo a more thorough exclusion of organic
disease. Most patients should have a complete blood count and
fecal calprotectin (or fecal lactoferrin), and/or C-reactive protein
should be checked in patients with IBS-D even in the absence of
alarm features to rule out IBD. Serologic testing for celiac disease
should be performed in patients with symptoms of IBS-D if the
prevalence of celiac disease in the general population is 1% or
higher. Stool examination for ova and parasites is not necessary
for all patients with IBS-D but could be helpful in regions where
parasitic infections are endemic/common or if history suggests a
possibility of parasitic infection (e.g., drinking untreated water).
Increased
antigen
presentation
Mast cell
activation
Altered
enteric
neuronal &
smooth
muscle
function
IBS
In the presence of alarm features or symptom onset after age 40,
or in the absence of response to traditional therapies such as
antidiarrheals, colonoscopy is warranted. Random biopsies to
rule out microscopic colitis should be performed if the patient has
persistent or chronic diarrhea. If the main symptoms are diarrhea
and increased gas, the possibility of lactase deficiency should be
ruled out with a hydrogen breath test or with evaluation after a
3-week lactose-free diet. Excessive gas with bloating also raises the
possibility of small-bowel bacteria overgrowth, and this should
be ruled out with a glucose hydrogen breath test. In patients with
concurrent symptoms of dyspepsia, upper GI radiographs or
esophagogastroduodenoscopy may be advisable. In patients with
postprandial right upper quadrant pain, an ultrasonogram of the
gallbladder should be obtained. Laboratory features that argue
against IBS include evidence of anemia, elevated sedimentation
rate, presence of leukocytes or blood in stool, and stool volume
200–300 mL/d. These findings would necessitate other diagnos­
tic considerations.
TREATMENT
Irritable Bowel Syndrome
PATIENT COUNSELING AND DIETARY ALTERATIONS
Reassurance and careful explanation of the functional nature of the
disorder are of primary importance, along with counseling on how
to identify triggers for the patient’s IBS symptoms. Stress manage­
ment may be helpful if stress is a trigger, and identifying obvious
food precipitants is an important first step in patient counseling and
dietary change. As a therapeutic trial, patients should be encour­
aged to eliminate any foodstuffs that appear to produce symptoms.
However, patients should avoid nutritionally depleted diets. It is
unclear if exercise improves overall IBS symptoms, but given its
low risk and overall health benefits, it should be encouraged in all
IBS patients.
STOOL-BULKING AGENTS
High-fiber diets and fiber supplementation are frequently used in
treating IBS. The water-holding action of fibers may contribute
to increased stool bulk, and fiber also speeds up colonic transit in
most people. Furthermore, because of its hydrophilic properties,
soluble fiber binds water by soaking up extra water in the colon
and therefore can be helpful with diarrhea in some IBS patients.
Furthermore, fiber supplementation with psyllium has been shown
to reduce perception of rectal distention, indicating that fiber may
have a positive effect on visceral afferent function.
Irritable Bowel Syndrome
CHAPTER 338
Controlled trials of dietary fiber in IBS patients have produced
variable results. This is not surprising since IBS is a heterogeneous
disorder, with some patients being constipated and other hav­
ing predominant diarrhea. Most investigations report increases in
stool weight, decreases in colonic transit times, and improvement
in constipation. Others have noted benefits in patients with alter­
nating diarrhea and constipation, pain, and bloating. However,
most studies observe no response in patients with diarrhea- or
pain-predominant IBS. While studies do not support the use of
insoluble dietary fiber such as wheat bran in IBS, soluble fibers such
as psyllium preparations are often found to be effective in manag­
ing IBS symptoms. Fiber should be started at a nominal dose and
slowly titrated up as tolerated over the course of several weeks to a
targeted dose of 20–30 g of total dietary and supplementary fiber
per day. Even when used judiciously, fiber can exacerbate bloating,
flatulence, constipation, and diarrhea.
ANTISPASMODICS
Clinicians have observed that anticholinergic drugs may provide
temporary relief for symptoms such as painful cramps related to
intestinal spasm. Although controlled clinical trials have produced
mixed results, evidence generally supports the beneficial effects
of anticholinergic drugs for pain. Physiologic studies demonstrate
that anticholinergic drugs inhibit the gastrocolic reflex; hence,
postprandial pain is best managed by giving antispasmodics 30 min
before meals so that effective blood levels are achieved shortly
before the anticipated onset of pain. Most anticholinergics contain
natural belladonna alkaloids, which may cause xerostomia, urinary
hesitancy and retention, blurred vision, and drowsiness. They
should be used in the elderly with caution. Some physicians prefer
to use synthetic anticholinergics such as dicyclomine that have less
effect on mucous membrane secretion and produce fewer unde­
sirable side effects. Peppermint oil appears to reduce abdominal
cramps by some undefined mechanism. In a meta-analysis of 10
double-blind, randomized controlled trials evaluating 1030 IBS
patients, peppermint oil was found to be significantly superior to
placebo for global improvement of IBS symptoms and reduction in
abdominal pain. The most commonly reported adverse event was
heartburn, which was mild and transient.
ANTIDIARRHEAL AGENTS
Peripherally acting opiate-based agents are the initial therapy of
choice for IBS-D. Physiologic studies demonstrate increases in
segmenting colonic contractions, delays in fecal transit, increases
in anal pressures, and reductions in rectal perception with these
drugs. When diarrhea is severe or the predominant symptom,
small doses of loperamide, 2–4 mg every 4–6 h up to a maximum
of 12 mg/d, can be prescribed. These agents are less addictive than
paregoric, codeine, or tincture of opium. In general, the intestines
do not become tolerant of the antidiarrheal effect of opiates,
and increasing doses are not required to maintain antidiarrheal
potency. These agents are most useful if taken before anticipated
stressful events that are known to cause diarrhea. However,
not infrequently, a high dose of loperamide may cause cramp­
ing because of increases in colonic contractions. Eluxadoline, a
locally acting, mixed µ-opioid receptor agonist, κ-opioid receptor
agonist, and δ-opioid receptor antagonist with minimal systemic
bioavailability, has been approved for treating abdominal pain
and diarrhea by the U.S. Food and Drug administration (FDA)
for patients with IBS-D. Given a small risk of pancreatitis associ­
ated with eluxadoline, it should not be prescribed for patients
with prior cholecystectomy or significant alcohol use or prior
history of pancreatitis. Other antidiarrheal agents that may be
used in IBS patients are the bile acid binders cholestyramine or
colesevelam because up to 30% of IBS-D patients may have bile
acid malabsorption.
OSMOTIC LAXATIVES
Osmotic agents such as magnesium citrate–based products, sodium
phosphate–based products, and nonabsorbable carbohydrates are
hypertonic products that, through osmosis, extract fluid into the
intestinal lumen to soften stool and enhance colonic transit. In
contrast, polyethylene glycol (PEG)–based solution is iso-osmotic
and induces bowel movement by high-volume lavage. The osmotic
laxatives were better than placebo in improving symptoms of
chronic constipation in clinical trials. However, chronic use of
magnesium hydroxide may result in severe hypermagnesemia in
patients with renal impairment. Frequent sodium phosphate–based
bowel cleansing should be avoided as this is associated with hyper­
phosphatasemia, hypocalcemia, and hypokalemia. In 19 trials,
PEG consistently induced more bowel movements than placebo.
A Cochrane review of 10 randomized trials showed that PEG was
superior to lactulose for improving stool frequency and abdomi­
nal pain. Among the nonabsorbable carbohydrates, lactulose and
sorbitol had similar laxative effects. However, bacterial metabolism
of unabsorbed carbohydrates often leads to gas production and
abdominal pain, which can limit long-term use.
ANTIFLATULENCE THERAPY
The management of excessive gas is seldom satisfactory, except
when there is obvious aerophagia or disaccharidase deficiency.
Patients should be advised to eat slowly and not chew gum or drink
carbonated beverages. Bloating may decrease if an associated gut
syndrome such as IBS or constipation is improved. If bloating is
accompanied by diarrhea and worsens after ingesting dairy prod­
ucts, fresh fruits, vegetables, or juices, further investigation or a
dietary exclusion trial may be worthwhile. Avoiding gas forming
foods, exercising, losing excess weight, and taking activated char­
coal are safe but unproven remedies. A low FODMAP (fermentable
oligosaccharides, disaccharides, monosaccharides, and polyols) diet
has been shown to be quite effective in reducing gas and bloating
(see “Low FODMAP Diet”) (Table 338-2). Data regarding the use
of surfactants such as simethicone are conflicting. Antibiotics may
help in a subgroup of IBS patients with predominant symptoms of
bloating. Beano, an over-the-counter oral β-glycosidase solution,
may reduce rectal passage of gas without decreasing bloating and
pain.
ANTIDEPRESSANT DRUGS
In addition to their mood-elevating effects, antidepressant medi­
cations have several physiologic effects that suggest they may be
beneficial in IBS. In IBS-D patients, the tricyclic antidepressant
imipramine slows jejunal migrating motor complex transit propa­
gation and delays orocecal and whole-gut transit, indicative of a
motor inhibitory effect. Some studies also suggest that tricyclic
agents may alter visceral afferent neural function.
Tricyclic antidepressants is effective in IBS patients of all
subtypes. The beneficial effects of the tricyclic compounds in
the treatment of IBS appear to be independent of their effects
on depression. The therapeutic benefits for the bowel symptoms
occur faster and at a lower dosage. The efficacy of antidepressant
agents from other chemical classes in the management of IBS is
less well evaluated. Selective serotonin reuptake inhibitors (SSRIs)
have also been studied in patients with IBS with conflicting results
and significant heterogeneity among individual trials.
SEROTONIN RECEPTOR MODULATORS
Serotonin 5-HT3 and 5-HT4 receptors are found throughout the
GI tract. Tegaserod, a 5-HT4 receptor agonist, exhibits prokinetic
activity by stimulating peristalsis. Clinical studies involving >4000
IBS-C patients reported reduction in abdominal discomfort and
improvements in constipation and bloating compared to placebo.
Diarrhea is the only major side effect. In 2007, the drug was vol­
untarily withdrawn from the market after a greater number of
cardiovascular complications were observed in a database of 18,000
patients receiving tegaserod (0.11 vs 0.01% in placebo). In 2019, the
FDA reviewed additional data and approved the use of tegaserod
in women younger than 65 years old who do not have a history of
ischemic cardiovascular disease and who have no more than one
risk factor for cardiovascular disease.
TABLE 338-2  Some Common Food Sources of FODMAPs
FOOD TYPE
FREE FRUCTOSE
LACTOSE
FRUCTANS
Fruits
Apple, cherry, mango,
pear, watermelon
Vegetables
Asparagus, artichokes,
sugar snap peas
Grains and cereals
Wheat, rye, barley
Nuts and seeds
Pistachios
Milk and milk products
Milk, yogurt, ice cream,
custard, soft cheeses
Legumes
Legumes, lentils, chickpeas
Legumes, chickpeas,
lentils
Other
Honey, high-fructose
corn syrup
Food additives
Inulin, FOS
Sorbitol, mannitol, maltitol,
xylitol, isomalt
Abbreviations: FODMAPs, fermentable oligosaccharides, disaccharides, monosaccharides, and polyols; FOS, fructo-oligosaccharides.
Source: Reproduced with permission from PR Gibson et al: Food choice as a key management strategy for functional gastrointestinal symptoms. Am J Gastroenterol 107:657, 2012.
SECRETAGOGUES
Lubiprostone, linaclotide, and plecanatide are secretagogues that
stimulate net efflux of ions and water into the intestinal lumen and
thus enhance transit and facilitate ease of defecation. By activating
channels on the apical (luminal) enterocyte surface, these secreta­
gogues increase intestinal chloride secretion. Other ion channels
and transporters secrete sodium into the intestine to maintain
electroneutrality, followed by the secretion of water. Lubiprostone
is a bicyclic fatty acid derived from prostaglandin E1 that activates
type 3 chloride channels in the apical membrane of intestinal
epithelial cells. Oral lubiprostone was effective in the treatment of
patients with IBS-C in large phase 3, randomized, double-blind,
placebo-controlled multicenter trials. The recommended daily
dose is 24 mcg twice daily. In general, the drug is well tolerated.
The major side effects are nausea and diarrhea. Linaclotide and
plecanatide are minimally absorbed 14-amino-acid peptide gua­
nylate cyclase-C (GC-C) agonists that bind to and activate GC-C
on the luminal surface of intestinal epithelium. The subsequent
increase in cyclic guanosine monophosphate activates the cyclic
fibrosis transmembrane regulator and induces fluid secretion
into the GI tract. These drugs are similar to endogenous peptides
secreted by the small intestine (uroguanylin) or colon (guany­
lin). In two 12-week, double-blind, randomized, controlled trials,
linaclotide (290 or 145 μg, once daily) reduced constipation and
pain. A lower dose (72 μg once daily) was also more effective than
placebo. Linaclotide has been approved by the FDA for treatment
of IBS-C. Plecanatide (3- and 6-mg doses) also has been shown
to be more effective than placebo in two phase 3 trials for IBS-C.
The 3-mg once-daily dose has been approved by the FDA. The
only significant side effect was diarrhea, which occurred in <5%
of patients. Tenapanor, a small-molecule inhibitor of GI sodiumhydrogen exchange-3, is more effective than placebo when given
at 50 mg twice daily in patients with IBS-C and has been approved
by the FDA for treatment of IBS-C.
MODULATION OF GUT FLORA
Because altered colonic flora (gut dysbiosis) may contribute to
the pathogenesis of IBS, this has led to great interest in using
antibiotics, prebiotics, probiotics, and dietary measures to treat
IBS.
PART 10
Disorders of the Gastrointestinal System
Antibiotics  Antibiotic treatment benefits a subset of IBS patients.
In a double-blind, randomized, placebo-controlled study, neomy­
cin dosed at 500 mg twice daily for 10 days was more effective
than placebo at improving symptom scores among IBS patients.
The nonabsorbed oral antibiotic rifaximin is the most thoroughly
GALACTOOLIGOSACCHARIDES
POLYOLS
Peach, persimmon, watermelon
Apple, apricot, pear, avocado,
blackberries, cherry, nectarine,
plum, prune
Artichokes, beetroot, Brussels
sprout, chicory, fennel, garlic,
leek, onion, peas
Cauliflower, mushroom, snow
peas
Chicory drinks
studied antibiotic for the treatment of IBS. In two identical doubleblind, placebo-controlled studies with >1200 nonconstipated IBS
patients, those receiving rifaximin at a dose of 550 mg three times
daily for 2 weeks were significantly more likely to report adequate
relief of IBS symptoms compared with those taking placebo. Cur­
rently, rifaximin is approved by the FDA for treatment of IBS-D.
Moreover, a placebo-controlled randomized clinical trial has shown
that patients with IBS-D who respond to rifaximin are more likely
to respond to subsequent courses of rifaximin compared to placebo
if they experience relapse of symptoms.
Prebiotics  These are nondigestible food ingredients that
stimulate growth and/or activity of bacteria in the GI tract.
There have been four randomized trials to examine the effects
of prebiotics. Three of the four studies reported that prebiot­
ics worsened or did not improve IBS symptoms. This is not
surprising given the adverse effects of a high-carbohydrate diet
on IBS symptoms.
Probiotics  These are defined as live microorganisms that when
administered in adequate amounts confer a health benefit on the
host. Multiple meta-analyses have suggested that probiotics might
be an effective strategy in improving global IBS symptoms includ­
ing abdominal pain and bloating; however, given the limitations of
existing studies (small sample size, short duration, lack of rigorous
endpoints, single-centered), it is unclear which probiotics to recom­
mend for patients with IBS. Given the lack of high-quality data, the
American Gastroenterological Association makes no recommenda­
tions for use of probiotics in patients with IBS. Large-scale studies
of well-phenotyped IBS patients are needed to establish the efficacy
of probiotics in IBS.
Low FODMAP Diet  A diet rich in FODMAPs often triggers
symptoms in IBS patients. FODMAPs are poorly absorbed by the
small intestine and fermented by bacteria in the colon to produce
gas and osmotically active carbohydrates (Fig. 338-4). At the same
time, on entering the colon, FODMAPs may serve as nutrients for
the colonic bacteria and promote the growth of gram-negative
commensal bacteria, which may induce barrier dysfunction and
subclinical mucosal inflammation. Fructose and fructans induce
IBS symptoms in a dose-dependent manner. In contrast, a low
FODMAP diet reduces IBS symptoms in 50–80% of patients with
IBS. A network meta-analysis of 13 randomized controlled trials
of IBS patients found that a low FODMAP diet was superior to all
other dietary interventions in reducing global IBS symptoms and
primarily improves abdominal pain and bloating in patients with
IBS. The current approach to low FODMAP diet implementation
High FODMAP diet
Acute effects
Osmotic effects
Bacterial fermentation
Gas and short-chain
fatty acid production
Colonic distention and acceleration
of colonic transit
IBS symptoms (pain, bloating, and diarrhea)
FIGURE 338-4  Pathogenesis of FODMAP-related symptoms. FODMAPs are poorly absorbed by the small intestine and fermented by gut bacteria to produce gas and
osmotically active carbohydrates. These events act in concert to cause bloating, flatulence, and diarrhea. FODMAPs may also serve as nutrients for colonic bacteria,
which may induce mucosal changes such as mast cell activation and barrier dysfunction. AGER, advanced glycosylation end product receptor; FODMAP, fermentable
oligosaccharides, disaccharides, monosaccharides, and polyols; LPS, lipopolysaccharide. (Adapted from P Singh et al: Am J Gastroenterol 117:947, 2022.)
is a three-step “step-down” process with an initial 2- to 6-week
restriction phase (which restricts all five FODMAP subgroups
[fructose, lactose, fructo- and galacto-oligosaccharides, and poly­
ols]), followed by a reintroduction phase in which food challenges
with individual FODMAPs are done to identify an individual’s
dietary triggers, and finally, a long-term maintenance phase
in which the diet is personalized based on the outcome of the
reintroduction phase. Physicians should be careful in not imple­
menting a low FODMAP diet among patients with a history of
eating disorders and those showing signs or symptoms of avoid­
ant restricting eating behavior. Given that between 20 and 50% of
patients do not respond to a low FODMAP diet, the identification
of patients who are more likely to respond to a low FODMAP diet
at baseline would be highly beneficial.
SUMMARY
The treatment strategy of IBS depends on the severity of the disorder
(Table 338-3). Most IBS patients have mild symptoms. They are usu­
ally cared for in primary care practices, have little or no psychosocial
difficulties, and do not seek health care often. Treatment usually
involves education, reassurance, and dietary/lifestyle changes. A
smaller portion have moderate symptoms that are usually intermit­
tent and correlate with altered gut physiology, e.g., worsened with
eating or stress and relieved by defecation. For IBS-D patients, treat­
ments include gut-acting pharmacologic agents such as antispas­
modics, antidiarrheals, and bile acid binders (Table 338-4). In IBS-C
TABLE 338-3  Spectrum of Severity in IBS
 
MILD
MODERATE
SEVERE
Clinical Features
Prevalence
70%
25%
5%
Correlations with gut physiology
+++
++
+
Symptoms constant
+++
Psychosocial difficulties
+++
Health care issues
+
++
+++
Practice type
Primary
Specialty
Referral
Chronic effects
Increase in gramnegative bacteria
Advanced glycation and
stimulation of AGER
Increase in luminal LPS
Mast cell activation and barrier dysfunction
Visceral hypersensitivity
CHAPTER 338
TABLE 338-4  Possible Drugs for a Dominant Symptom in IBS
SYMPTOM
DRUG
DOSE
Diarrhea
Loperamide
2–4 mg when necessary/maximum
12 g/d
Irritable Bowel Syndrome
Diphenoxylate
hydrochloride and
atropine sulfate (Lomotil)
1–2 tabs as needed or daily (can be
taken up to 4 times daily)
 
Cholestyramine resin
4 g with meals, can be started qd
and increased to tid
 
Eluxadoline
100 mg bid
Alosetrona
0.5–1 mg bid (for severe IBS,
women)
Constipation
Psyllium husk
3–4 g bid with meals, then adjust
Methylcellulose
2 g bid with meals, then adjust
Calcium polycarbophil
1 g qd to qid
Lactulose syrup
10–20 g bid
70% sorbitol
15 mL bid
Polyethylene glycol 3350
17 g in 250 mL water qd, can be
titrated to tid
Lubiprostone
8 or 24 µg bid
Magnesium hydroxide
15–60 mL qd
Linaclotide
72, 145, and 290 μg qd
 
Plecanatide
3 mg qd
Tenapanor
50 mg bid
Abdominal pain
Smooth-muscle relaxant
qd to qid ac
 
Tricyclic antidepressants
Start 25–50 mg hs, then adjust
Low FODMAP diet
 
Gas and
bloating
Low FODMAP diet
Probiotics
qd
Rifaximin
550 mg bid
aAvailable only in the United States.
Abbreviation: FODMAP, fermentable oligosaccharides, disaccharides,
monosaccharides, and polyols; IBS, irritable bowel syndrome.
Source: Reproduced with permission from GF Longstreth et al: Functional bowel
disorders. Gastroenterology 130:1480, 2006.

09 - 339 Diverticular Disease and Common Anorectal Disorders
339 Diverticular Disease and Common Anorectal Disorders
patients, increased fiber intake and the use of osmotic agents such as
PEG may achieve satisfactory results. For patients with more severe
constipation, a chloride channel opener (lubiprostone) or GC-C ago­
nist (linaclotide or plecanatide) or NHE3 inhibitor (tenapanor) may
be considered. For IBS patients with predominant gas and bloating,
a low FODMAP diet may provide significant relief. Some patients
may benefit from probiotics and rifaximin treatment. A small pro­
portion of IBS patients have severe and refractory symptoms, are
usually seen in referral centers, and frequently have constant pain
and psychosocial difficulties. This group of patients is best managed
with neuromodulators and other psychological treatments (Table
338-4). Clinical trials demonstrating success of a low FODMAP diet
in improving IBS symptoms and quality of life provide strong evi­
dence supporting the use of this dietary approach in the treatment
of IBS. These observations, if confirmed, may lead to the use of the
low FODMAP diet as the first line of treatment of IBS patients with
moderate to severe symptoms.
■
■FURTHER READING
Chang L et al: AGA Clinical Practice Guideline on the pharmaco­
logical management of irritable bowel syndrome with constipation.
Gastroenterology 163:118, 2022.
Dionne J et al: A systematic review and meta-analysis evaluating the
efficacy of a gluten-free diet and a low FODMAP diet in treating
symptoms of irritable bowel syndrome. Am J Gastroenterol 113:1290,
2018.
Drossman DA: Functional gastrointestinal disorders: History,
pathophysiology, clinical features, and Rome IV. Gastroenterology
150:1262, 2016.
Lembo A et al: AGA Clinical Practice Guideline on the pharmacologi­
PART 10
Disorders of the Gastrointestinal System
cal management of irritable bowel syndrome with diarrhea. Gastro­
enterology 163:137, 2022.
Mayer EA et al: Brain-gut microbiome interactions and functional
bowel disorders. Gastroenterology 146:1500; 2014.
Pittayanon R et al: Gut microbiota in patients with irritable bowel
syndrome: A systematic review. Gastroenterology 157:97, 2019.
Zhou SY et al: FODMAP diet modulates visceral nociception by lipo­
polysaccharide-mediated intestinal barrier dysfunction and intestinal
inflammation. J Clin Invest 128:267, 2018.
Susan L. Gearhart
Diverticular Disease
and Common Anorectal
Disorders
■
■DIVERTICULAR DISEASE
Incidence and Epidemiology 
In the United States, diverticulosis
affects one-half of the population aged >60 years, and the majority
of affected individuals will have no associated symptoms. However,
studies have shown that ~5% of individuals with diverticulosis will
develop acute diverticular disease. In addition, 10–25% of individu­
als with diverticular disease will experience recurrent symptoms, and
up to 10% will develop complications leading to surgery. Diverticular
disease has become the fifth most costly gastrointestinal disorder in the
United States and is the leading indication for elective colon resection.
The incidence of diverticular disease is on the rise and most prevalent
among middle-aged individuals. The majority of patients with diver­
ticular disease report a lower health-related quality of life and more
depression as compared to matched controls, thus adding to health
care costs. Formerly, diverticular disease was confined to developed
countries; however, with the adoption of westernized diets in develop­
ing countries, diverticulosis is on the rise across the globe. Immigrants
to the United States develop diverticular disease at the same rate as
U.S. natives. Although the prevalence among females and males is
similar, males tend to present at a younger age. Known risk factors for
the development of diverticular disease include the use of nonsteroidal
anti-inflammatory drugs (NSAIDS), aspirin, steroids, opioids, smok­
ing, and sedentary lifestyle.
Anatomy and Pathophysiology 
Two types of diverticula occur
in the intestine: true and false (or pseudodiverticula). A true diverticu­
lum is a saclike herniation of the entire bowel wall, whereas a pseudo­
diverticulum involves only a protrusion of the mucosa and submucosa
through the muscularis propria of the colon (Fig. 339-1). The type of
diverticulum most commonly seen in the colon is the pseudodiver­
ticulum. The diverticula occur at the point where the nutrient artery,
or vasa recta, penetrates through the muscularis propria, resulting in
a break in the integrity of the colonic wall. Diverticula are most com­
monly encountered in the sigmoid colon. This anatomic restriction
may be a result of the relative high-pressure zone within the muscular
sigmoid colon. Higher-amplitude contractions combined with consti­
pated, high-fat-content stool within the sigmoid lumen in an area of
weakness in the colonic wall result in the creation of these diverticula.
FIGURE 339-1  Gross and microscopic view of sigmoid diverticular disease. Arrows
mark an inflamed diverticulum with the diverticular wall made up only of mucosa.
Consequently, the vasa recta is either compressed or eroded, leading to
either perforation or bleeding.
As mentioned above, diverticula commonly affect the left and
sigmoid colon; the rectum is always spared. However, in Asian popula­
tions, 70% of diverticula are seen in the right colon and cecum. Diver­
ticulitis is inflammation of a diverticulum. Previous understanding of
the pathogenesis of diverticulosis attributed the disease to poor dietary
choices, and the onset of diverticulitis would occur acutely when these
diverticula become obstructed. However, evidence now suggests that
the pathogenesis is more complex and multifactorial. Better under­
standing of the gut microbiota suggests that dysbiosis is an important
aspect of disease. Chronic low-grade inflammation is thought to play
a key role in neuronal degeneration, leading to dysmotility and high
intraluminal pressure. As a consequence, pockets or outpouchings
develop in the colonic wall where it is weakest. Studies have also
shown that it is more common for patients with diverticular disease
to have abnormal collagen cross-linking. Researchers hypothesize that
abnormal collagen cross-linking leads to loss of intestinal compliance
and, therefore, higher intraluminal pressure, leading to pockets or
outpouchings.
A newer approach to examine causality in diseases such as diverticu­
litis is to examine genome-wide associations (GWAs) using a variety of
biobanks. Biobanks from Iceland, Denmark, and the United Kingdom
have been examined for GWAs in diverticular disease and identified
LAMB4 and TNFSF15 variants to be associate with early-onset, severe
diverticular disease in otherwise healthy families. Although the exact
function these genes play in the development of diverticular disease is
unknown, the protein encoded by the gene LAMB4 is part of the extra­
cellular matrix lamin family, whereas the protein encoded by the gene
TNFSF15 is a member of the tumor necrosis family, which has implica­
tions in inflammation.
Presentation, Evaluation, and Management of Diverticular
Bleeding 
Hemorrhage from a colonic diverticulum is the most
common cause of hematochezia in patients >60 years, yet only 20% of
patients with diverticulosis will have gastrointestinal bleeding. Patients
at increased risk for bleeding tend to be hypertensive, have atheroscle­
rosis, and regularly use anticoagulants and NSAIDs. Additional risk
factors include obesity and a history of diabetes mellitus. Of note, due
to increased use of anticoagulants in our aging population, there has
been a rise in the incidence of diverticular bleeding. Most bleeds are
self-limited and stop spontaneously with bowel rest. The lifetime risk
of rebleeding is 25%.
Initial localization of diverticular bleeding may include colonoscopy,
multiplanar computed tomography (CT) angiogram, or nuclear medi­
cine tagged red cell scan. If the patient is stable, ongoing bleeding is
best managed by angiography. If mesenteric angiography can localize
the bleeding site, the vessel can be occluded successfully with a coil in
80% of cases. The patient can then be followed closely with repetitive
colonoscopy, if necessary, looking for evidence of colonic ischemia.
However, with highly selective coil embolization, the rate of colonic
ischemia is <10%, and the risk of acute rebleeding is <25%. Longterm results (40 months) indicate that >50% of patients with acute
diverticular bleeds treated with highly selective angiography have had
definitive treatment. Alternatively, colonoscopic ligation with banding,
placement of a detachable snare, and over-the-scope hemoclip have
been shown to be effective methods to obtain hemostasis if the bleed­
ing site can be localized. These approaches have been shown to prevent
rebleeding or the requirement of emergent surgery. In the event that
these measures fail to achieve hemostasis, a segmental resection of the
colon may be undertaken. This may be advantageous in patients on
chronic anticoagulation and immunosuppression as delayed bleeding
and perforation have been reported in this subpopulation.
If the patient is refractory to angiographic or endoscopic treatments,
unstable, or has required a large-volume transfusion, current recom­
mendations are that surgery should be performed. If the bleeding has
been localized, a segmental resection can be performed. If the site
of bleeding has not been definitively identified, a subtotal colectomy
may be required. In patients without severe comorbidities, surgical
TABLE 339-1  Presentation of Diverticular Disease
Uncomplicated Diverticular Disease—75%
Abdominal pain
Fever
Leukocytosis
Anorexia/obstipation
Complicated Diverticular Disease—25%
Abscess 16%
Perforation 10%
Stricture 5%
Fistula 2%
resection can be performed with a primary anastomosis. A higher
anastomotic leak rate has been reported in patients who received
10 units of blood.
Presentation, Evaluation, and Staging of Diverticulitis 
Acute uncomplicated diverticulitis characteristically presents with
fever, anorexia, left lower quadrant abdominal pain, and obstipa­
tion (Table 339-1). The diagnosis of diverticulitis is best made on a
contrast-enhanced abdominal and pelvic CT scan demonstrating the
following findings: sigmoid diverticula, thickened colonic wall >4 mm,
and inflammation within the pericolic fat without the collection of
contrast material or fluid. Additional syndromes have been identified
that relate to the presence of diverticulosis or diverticular disease.
Symptomatic colitis-associated diverticulosis (SCAD) is uncommon
(<1% of patients with diverticulosis) and occurs when inflammation
is identified between the diverticula on endoscopic assessment. Symp­
toms of SCAD mimic irritable bowel disease. Symptomatic uncom­
plicated diverticular disease (SUDD) is a condition seen in patients
with known diverticular disease and ongoing abdominal pain without
evidence of overt inflammation on radiographic imaging.
CHAPTER 339
Diverticular Disease and Common Anorectal Disorders
Complicated diverticular disease is defined as diverticular disease
associated with an abscess or perforation and less commonly with a
fistula (Table 339-1). Symptoms of complicated diverticular disease
may be similar to uncomplicated disease, or patients may exhibit signs
of peritonitis indicating the presence of a diverticular perforation. If a
pericolonic abscess has formed, the patient may have abdominal dis­
tention and signs of localized peritonitis. Laboratory investigations
often demonstrate a leukocytosis. Rarely, a patient may present with
an air-fluid level in the left lower quadrant on plain abdominal film.
This should raise a concern for a giant diverticulum of the sigmoid
colon and is managed with resection to avoid impending perforation.
Perforated diverticular disease is staged using the Hinchey clas­
sification system (Fig. 339-2). This staging system was developed to
predict outcomes following the surgical management of complicated
diverticular disease. The Hinchey staging system has been modified
to include the development of a phlegmon or early abscess (Hinchey
stage Ia). In complicated diverticular disease with fistula formation,
common locations include cutaneous, vaginal, or vesicle fistulas. These
conditions present with either passage of stool through the skin or
vagina or the presence of air in the urinary stream (pneumaturia).
Colovaginal fistulas are more common in women who have undergone
a hysterectomy.
TREATMENT
Medical Management of Diverticular Disease
Asymptomatic diverticular disease discovered on imaging studies
or at the time of colonoscopy is best managed by lifestyle changes.
Although the data regarding dietary risks and symptomatic diver­
ticular disease are limited (Table 339-2), patients may benefit from
a fiber-enriched diet or supplements that include 30 g of fiber
each day. The use of fiber decreases colonic transit time and,
therefore, prevents increased intraluminal pressure, leading to the
Abscess
I
II
Feces
III
IV
FIGURE 339-2  Hinchey classification of diverticulitis. Stage I: Perforated
diverticulitis with a confined paracolic abscess. Stage II: Perforated diverticulitis
that has closed spontaneously with distant abscess formation. Stage III:
Noncommunicating perforated diverticulitis with fecal peritonitis (the diverticular
neck is closed off, and therefore, contrast will not freely expel on radiographic
images). Stage IV: Perforation and free communication with the peritoneum,
resulting in fecal peritonitis.
PART 10
Disorders of the Gastrointestinal System
development of diverticulosis. The incidence of complicated diver­
ticular disease appears to also be increased in patients who smoke
and are obese. Therefore, patients should be encouraged to refrain
from smoking and to join a weight loss program. The historical
recommendation to avoid eating nuts is based on no more than
anecdotal data.
ANTIBIOTICS
The routine use of antibiotics in uncomplicated diverticular disease
does not appear to reduce time to symptom resolution or reduce the
risk of complications or recurrence. Two large, randomized trials
(the AVOD trial and the Diabolo trial) and a large meta-analysis
demonstrated that immunocompetent patients with uncomplicated
diverticular disease had no difference in time to symptom resolu­
tion, recurrence rates, development of complicated diverticular
disease, or surgery if treated with or without antibiotics. Currently,
patients who are immunocompromised, have findings of extensive
inflammation on radiographic studies, are at risk for disease pro­
gression, or have computed tomography (CT) findings of compli­
cated diverticular disease should be treated with antibiotics. Known
risk factors for disease progression included the American Society
of Anesthesiologists (ASA) classification of III or IV, >5 days of
TABLE 339-2  The Use of Fiber in the Management of Diverticular Disease (DD)
JOURNAL, STUDY YEAR
PATIENTS (N)
INTERVENTION
STUDY LENGTH
FINDINGS
Lancet, 1977
Wheat or bran crisp bread
3 months
Significant reduction of symptoms score
BMJ, 1981
Bran, ispaghula, placebo
16 weeks
No difference
J Gastroenterol, 1977
Methylcellulose
3 months
Significant reduction in symptoms
BMJ, 2011
47,033
Vegetarian vs nonvegetarian
11.6 years
Vegetarians had a 31% lower risk of DD
Gastroenterology, 2012
Fiber consumption
12 years
Fiber associated with great risk of DD
JAMA, 2008
47,288
Nut, corn, popcorn consumption
18 years
Higher nut, corn, and popcorn had lower risk of recurrence
Ann R Coll Surg Engl, 1985
Fiber consumption
66 months
Higher fiber associated with 19% reduction in symptom
recurrence
Source: Modified from A Turis et al: Review article: The pathophysiology and medical management of diverticulosis and diverticular disease of the colon. Aliment Parmacol
Ther 42:664, 2015.
symptoms, and elevation of C-reactive protein (CRP) or white
blood cell (WBC) count.
If the use of antimicrobial therapy is desired, the current recom­
mended antimicrobial coverage for uncomplicated acute diverticu­
litis is a third-generation cephalosporin (or ciprofloxacin if there is
a known allergy to cephalosporins) and metronidazole targeting a
mixed flora. Alternatively, single-agent therapy with a third-generation
penicillin such as IV piperacillin or oral penicillin/clavulanic acid
may be effective. The usual course of antibiotics is 5 days. A
study compared the use of intravenous (IV) versus oral antibiot­
ics in uncomplicated diverticular disease and noted no difference
in recovery time or progression of the disease and recommended
that safe home treatment on oral antibiotics after a 6-h observation
in the emergency department is reasonable. Exclusion criteria for
home treatment included complicated diverticular disease, immu­
nocompromised patient, significant active comorbidities, poor
social support, or a decline during observation.
DIET AND OTHER MEDICAL THERAPIES
Patients should remain on a limited diet until their pain resolves.
The use of anti-inflammatory medications (mesalazine) in ran­
domized clinical trials has shown them to be beneficial at reducing
symptoms and disease recurrence in patients with SUDD. However,
when objective signs of inflammation such as CRP and computer­
ized imaging are taken into consideration, no benefit for the use of
mesalazine has been shown.
Probiotics are increasingly used by gastroenterologists for mul­
tiple bowel disorders and may prevent recurrence of diverticulitis.
Specifically, probiotics containing Lactobacillus acidophilus and
Bifidobacterium strains may be beneficial; however, a systematic
review was unable to show any benefit to the use of probiotics alone.
The addition of fiber or mesalazine with probiotics has shown some
promise in maintaining remission. Rifaximin (a poorly absorbed
broad-spectrum antibiotic), when compared to fiber alone for the
treatment of SUDD, is associated with 30% less frequent recurrent
symptoms from uncomplicated diverticular disease.
COLONOSCOPY
Colonoscopy should be considered ~6 weeks after the first episode
of uncomplicated diverticular disease and after the development
of complicated diverticular disease as the overall prevalence of
colon cancer is low in these patients (<2%) but higher incidence
has been seen with complicated diverticulitis (6–8%). The parallel
epidemiology of colorectal cancer and diverticular disease provides
enough concern for an endoscopic evaluation before operative
management.
SURGICAL MANAGEMENT OF DIVERTICULAR DISEASE
Preoperative risk factors influencing postoperative mortality rates
include higher ASA physical status class (Table 339-3) and pre­
existing organ failure. In patients who are low risk (ASA P1 and
P2), surgical therapy can be offered to those who do not rapidly
improve on medical therapy. For uncomplicated diverticular dis­
ease, medical therapy can be continued beyond two attacks without
an increased risk of perforation requiring a colostomy. However,
TABLE 339-3  American Society of Anesthesiologists Physical Status
Classification System
P1
A normal healthy patient
P2
A patient with mild systemic disease
P3
A patient with severe systemic disease
P4
A patient with severe systemic disease that is a constant threat to life
P5
A moribund patient who is not expected to survive without the operation
P6
A declared brain-dead patient whose organs are being removed for
donor purposes
patients on immunosuppressive therapy, in chronic renal failure,
or with a collagen-vascular disease have a fivefold greater risk of
perforation during recurrent attacks. A multicentered randomized
clinical trial (DIRECT trial) comparing surgery with conserva­
tive management for recurrent SUDD demonstrated that elective
surgical resection was associated with an improved quality of life
and was more cost-effective at 5 years following resection as com­
pared to conservative management. Surgical therapy is generally
indicated in all low-surgical-risk patients with complicated diver­
ticular disease. A randomized trial (LASER trial) examined the use
of laparoscopic surgery for refractory or complicated diverticular
disease and demonstrated improved quality of life and a reduction
in recurrent symptoms by 50% when compared to conservative
treatment.
The goals of surgical management of diverticular disease include
controlling sepsis, eliminating complications such as fistula or
obstruction, removing the diseased colonic segment, and restor­
ing intestinal continuity. Table 339-4 lists the operations most
commonly indicated based on the Hinchey classification and the
predicted postoperative outcomes. The current options for uncom­
plicated diverticular disease include an open or a minimally inva­
sive resection of the diseased area with reanastomosis to the
rectosigmoid. Preservation of portions of the sigmoid colon may
lead to early recurrence of the disease. The benefits of minimally
invasive resection over open surgical techniques include early
discharge (by at least 1 day), less narcotic use, less postoperative
complications, and an earlier return to work.
The options for the surgical management of complicated diver­
ticular disease (Fig. 339-3) include the following open or minimally
invasive procedures: (1) proximal diversion of the fecal stream with
an ileostomy or colostomy and sutured omental patch with drain­
age, (2) resection with colostomy and mucous fistula or closure of
distal bowel with formation of a Hartmann’s pouch (Hartmann’s
procedure), (3) resection with anastomosis (coloproctostomy), or
(4) resection with anastomosis and diversion (coloproctostomy
with loop ileostomy or colostomy). (5) Laparoscopic technique
of washout and drainage without diversion has been described
for Hinchey III patients; however, a threefold increased risk of
TABLE 339-4  Outcome Following Surgical Therapy for Complicated Diverticular Disease Based on Modified Hinchey Staging
HINCHEY STAGE
OPERATIVE PROCEDURE
ANASTOMOTIC LEAK RATE, %
OVERALL MORBIDITY RATE, %
Ia (pericolic phlegmon)
Laparoscopic or open colon resection
Ib (pericolic abscess)
Percutaneous drainage followed by laparoscopic or open
colon resection
II
Percutaneous drainage followed by laparoscopic or open colon
resection +/− proximal diversion with an ostomy
III
Laparoscopic washout and drainage
or
Laparoscopic or open resection with proximal diversion (ostomy)
or
Hartmann’s procedure
IV
Hartmann’s procedure
or
Washout with proximal diversion
recurrent peritonitis requiring reoperation with washout alone has
been reported. Robotic surgery resection for complicated diverticu­
lar disease is associated with a lower rate of conversion to an open
procedure.
Patients with Hinchey stage Ia may be managed with antibiotic
therapy only or followed by resection with anastomosis following
further workup and symptom resolution. Patients with Hinchey
stages Ib and II disease are managed with percutaneous drainage
followed by resection with anastomosis following further evalu­
ation and symptom resolution. Current guidelines put forth by
the American Society of Colon and Rectal Surgeons suggest, in
addition to antibiotic therapy, CT-guided percutaneous drainage
of diverticular abscesses that are >3 cm and have a well-defined
wall. Abscesses that are <5 cm may resolve with antibiotic therapy
alone. Contraindications to percutaneous drainage are no per­
cutaneous access route, pneumoperitoneum, and fecal peritonitis.
Drainage of a diverticular abscess is associated with a 20–25%
failure rate. Urgent operative intervention is undertaken if percu­
taneous drainage fails and patients develop generalized peritonitis,
and most will need to be managed with a Hartmann’s procedure
(resection of the sigmoid colon with end colostomy and stapling of
the rectosigmoid distal to the diseased segment). In selected cases,
nonoperative therapy may be considered.
The management of Hinchey stage III disease is under debate.
In this population of patients, no fecal peritonitis is present, and it
is presumed that the perforation has sealed. Historically, Hinchey
stage III has been managed with a Hartmann’s procedure or with
primary anastomosis and proximal diversion. Several studies
have examined short- and long-term outcomes for laparoscopic
peritoneal lavage to remove the peritoneal contamination and
place drainage catheters should a communication to the bowel
still exist. However, this procedure has been associated with an
increased risk of requiring reoperation for ongoing peritonitis.
Overall, ostomy rates are lower with the use of laparoscopic peri­
toneal lavage. No anastomosis of any type should be attempted in
Hinchey stage IV disease or in the presence of fecal peritonitis. A
limited approach to these patients is associated with a decreased
mortality rate.
CHAPTER 339
Diverticular Disease and Common Anorectal Disorders
Recurrent Symptoms 
Recurrent abdominal symptoms following
surgical resection for diverticular disease occur in 10% of patients.
Recurrent diverticular disease develops in patients following inad­
equate surgical resection. A retained segment of diseased rectosigmoid
colon is associated with twice the incidence of recurrence. The pres­
ence of irritable bowel syndrome may also cause recurrence of initial
symptoms. Patients undergoing surgical resection for presumed diver­
ticulitis and symptoms of chronic abdominal cramping and irregular
loose bowel movements consistent with irritable bowel syndrome have
poorer functional outcomes.
30% risk of peritonitis requiring
reoperation if no resection is
performed.
Overall morbidity 50%
Overall mortality 15%
—
Overall morbidity 50%
Overall mortality 15%
FIGURE 339-3  Methods of surgical management of complicated diverticular
disease. 1. Drainage, omental pedicle graft, and proximal diversion. 2. Hartmann’s
procedure. 3. Sigmoid resection with coloproctostomy. 4. Sigmoid resection with
coloproctostomy and proximal diversion.
PART 10
Disorders of the Gastrointestinal System
COMMON DISEASES OF THE ANORECTUM
■
■RECTAL PROLAPSE (PROCIDENTIA)
Incidence and Epidemiology 
Rectal prolapse is six times more
common in women than in men. The incidence of rectal prolapse
peaks in women >60 years. Women with rectal prolapse have a higher
incidence of associated pelvic floor disorders including urinary incon­
tinence, rectocele, cystocele, and enterocele. About 20% of children
with rectal prolapse will have cystic fibrosis. All children presenting
with prolapse should undergo a sweat chloride test. Less common
associations include Ehlers-Danlos syndrome, solitary rectal ulcer syn­
drome, congenital hypothyroidism, Hirschsprung’s disease, dementia,
cognitively impaired, and schizophrenia.
Anatomy and Pathophysiology 
Rectal prolapse (procidentia) is
a circumferential, full-thickness protrusion of the rectal wall through
the anal orifice. It is often associated with a redundant sigmoid colon,
pelvic laxity, and a deep rectovaginal septum (pouch of Douglas).
Initially, rectal prolapse was believed to be the result of early internal
rectal intussusception, which occurs in the upper to mid rectum. This
was considered to be the first step in an inevitable progression to
full-thickness external prolapse. However, only 1 of 38 patients with
internal prolapse followed for >5 years developed full-thickness pro­
lapse. Others have suggested that full-thickness prolapse is the result
of damage to the nerve supply to the pelvic floor muscles or pudendal
nerves from repeated stretching with straining to defecate. Damage to
the pudendal nerves would weaken the pelvic floor muscles, including
the external anal sphincter muscles. Bilateral pudendal nerve injury
is identified significantly more with full-thickness prolapse and fecal
incontinence than unilateral injury.
Presentation and Evaluation 
In external prolapse, patient com­
plaints include a palpable anal mass, bleeding per rectum, leakage of
blood and mucus, and poor perianal hygiene. Prolapse of the rectum
usually occurs following defecation and will spontaneously reduce
or require the patient to manually reduce the prolapse. Constipation
occurs in ~30–67% of patients with rectal prolapse. Differing degrees
A
C
B
D
FIGURE 339-4  Degrees of rectal prolapse. Mucosal prolapse only (A, B, sagittal
view). Full-thickness prolapse associated with redundant rectosigmoid and deep
pouch of Douglas (C, D, sagittal view).
of fecal incontinence occur in 50–70% of patients. Patients with inter­
nal rectal prolapse will present with symptoms of both constipation
and incontinence. Other associated findings include outlet obstruc­
tion (anismus) in 30%, colonic inertia in 10%, and solitary rectal ulcer
syndrome in 12%.
Office evaluation is best performed after the patient has been given
an enema, which enables the prolapse to protrude. An important dis­
tinction should be made between full-thickness rectal prolapse and
isolated mucosal prolapse associated with hemorrhoidal disease
(Fig. 339-4). Mucosal prolapse is known for radial grooves rather
than circumferential folds around the anus and is due to increased
laxity of the connective tissue between the submucosa and underlying
muscle of the anal canal. The evaluation of prolapse should also include
cystoproctography and colonoscopy. These examinations evaluate for
associated pelvic floor disorders and rule out a malignancy or a polyp
as the lead point for prolapse.
TREATMENT
Rectal Prolapse
The medical approach to the management of rectal prolapse is
limited and includes stool-bulking agents or fiber supplementa­
tion to ease the process of evacuation. Surgical correction of rectal
prolapse is the mainstay of therapy. Previously, the presence of
internal rectal prolapse identified on imaging studies has been con­
sidered a nonsurgical disorder, and biofeedback was recommended.
However, only one-third of patients will have successful resolution
of symptoms from biofeedback. Two approaches are commonly
considered: transabdominal and transperineal. Transabdominal
approaches have been associated with lower recurrence rates, but
some patients with significant comorbidities are better served by a
transperineal approach.
Common transperineal approaches include a transanal proctec­
tomy (Altmeier procedure), mucosal proctectomy (Delorme proce­
dure), or placement of a Tirsch wire encircling the anus. The goal of
the transperineal approach is to remove the redundant rectosigmoid
colon. Common transabdominal approaches include presacral suture
or mesh rectopexy (Ripstein) with (Frykman-Goldberg) or without
resection of the redundant sigmoid. Colon resection, in general, is
reserved for patients with constipation and outlet obstruction. Ven­
tral rectopexy is an effective method of abdominal repair of internal
and full-thickness prolapse that does not require sigmoid resection.
This repair has been shown to have improved pelvic floor functional
results over other abdominal repairs. Transabdominal procedures
can be performed effectively with laparoscopic and robotic tech­
niques. Short- and long-term recurrence rates are low (<10%), and
symptoms improved in more than three-fourths of patients.
■
■FECAL INCONTINENCE
Incidence and Epidemiology 
Fecal incontinence is the involun­
tary passage of fecal material or the inability to control the initiation
of defecation. The prevalence of fecal incontinence in adults in the
United States approaches 15% and is expected to increase given our
aging population. A higher incidence of incontinence is seen among
older parous women. One-half of patients with fecal incontinence also
suffer from urinary incontinence. The cause of fecal incontinence is
often multifactorial; however, the majority of women with fecal incon­
tinence are parous and may have experienced obstetrical injury to the
pelvic floor, either while carrying a fetus or during the delivery. An
anatomic sphincter defect may occur in up to 32% of women following
childbirth regardless of visible damage to the perineum. Risk factors at
the time of delivery include prolonged labor, the use of forceps, and the
need for an episiotomy. Symptoms of incontinence can present two or
more decades after obstetric injury. Medical conditions known to con­
tribute to the development of fecal incontinence are listed in Table 339-5.
Anatomy and Pathophysiology 
The anal sphincter complex
is made up of the internal and external anal sphincter. The internal
sphincter is smooth muscle and a continuation of the circular fibers of
the rectal wall. It is innervated by the intestinal myenteric plexus and
is therefore not under voluntary control. The external anal sphincter
is formed in continuation with the levator ani muscles and is under
voluntary control. The pudendal nerve supplies motor innervation to
the external anal sphincter. Obstetric injury may result in tearing of
the muscle fibers anteriorly at the time of the delivery. This results in
an obvious anterior defect on endoanal ultrasound. Injury may also be
the result of stretching of the pudendal nerves during pregnancy or
delivery of the fetus through the birth canal.
Presentation and Evaluation 
Patients may suffer with varying
degrees of fecal incontinence. Fecal incontinence is classified into three
categories based on the clinical presentation. These categories include
TABLE 339-5  Medical Conditions That Contribute to Symptoms of
Fecal Incontinence
Neurologic Disorders
• Dementia
• Brain tumor
• Stroke
• Multiple sclerosis
• Tabes dorsalis
• Cauda equina lesions
Skeletal Muscle Disorders
• Myasthenia gravis
• Myopathies, muscular dystrophy
Miscellaneous
• Hypothyroidism
• Irritable bowel syndrome
• Diabetes
• Severe diarrhea
• Scleroderma
passive incontinence (passage of stool without awareness), urge incon­
tinence (leakage of stool despite attempts at holding), and fecal seep­
age (seepage of residue generally following defecation associated with
normal continence). Beyond the immediate problems associated with
fecal incontinence, these patients are often withdrawn and suffer from
depression. For this reason, quality-of-life measures are an important
component in the evaluation of patients with fecal incontinence.
The evaluation of fecal incontinence should include a thorough
history and physical examination including digital rectal examination
(DRE). Weak sphincter tone on DRE and loss of the “anal wink” reflex
(S1-level control) may indicate a neurogenic dysfunction. Perianal
scars may represent surgical injury. Other studies helpful in the diag­
nosis of fecal incontinence include anal manometry, pudendal nerve
terminal motor latency (PNTML), and endoanal ultrasound. Centers
that care for patients with fecal incontinence will have an anorectal
physiology laboratory that uses standardized methods of evaluating
anorectal physiology. Pudendal nerve studies evaluate the function of
the nerves innervating the anal canal using a finger electrode placed in
the anal canal. Stretch injuries to these nerves will result in a delayed
response of the sphincter muscle to a stimulus, indicating a prolonged
latency. Finally, endoanal ultrasound will evaluate the extent of the
injury to the sphincter muscles before surgical repair. Unfortunately,
all of these investigations are user-dependent, and very few studies
demonstrate that these studies predict outcome following an interven­
tion. Magnetic resonance imaging (MRI) has also been utilized in the
evaluation of the sphincter muscle complex in treatment planning for
fecal incontinence, but its role has not been well-established.
Rarely does a pelvic floor disorder exist alone. The majority of
patients with fecal incontinence will have some degree of urinary
incontinence. Similarly, fecal incontinence is a part of the spectrum
of pelvic organ prolapse. For this reason, patients may present with
symptoms of obstructed defecation as well as fecal incontinence. Care­
ful evaluation including dynamic MRI or cinedefecography should
be performed to search for other associated defects. Surgical repair
of incontinence without attention to other associated defects may
decrease the success of the repair.
CHAPTER 339
Diverticular Disease and Common Anorectal Disorders
TREATMENT
Fecal Incontinence
Medical management of fecal incontinence includes strategies to
bulk up the stool, which help in increasing fecal sensation and
complete evacuation. Stool bulking agents include fiber supple­
mentation, loperamide, diphenoxylate, and bile acid binders. These
agents help to bind the stool, resulting in more complete evacuation
and decreasing the frequency of bowel movements. This can be
particularly helpful in patients with mild symptoms of fecal incon­
tinence. Patients may be offered a form of physical therapy called
biofeedback. This therapy helps strengthen the external sphincter
muscle while training the patient to relax with defecation to avoid
unnecessary straining and further injury to the sphincter muscles.
At a minimum, biofeedback is risk-free and most patients will have
some improvement. For this reason, it should be incorporated into
the initial recommendation to all patients with fecal incontinence.
It is important to note that there is no medical or surgical therapy to
reduce incontinence of flatus. Dietary alterations to reduce gas pro­
duction and the use of probiotics are the only antidotal remedies.
Historically, the “gold standard” for the treatment of fecal incon­
tinence with an isolated sphincter defect has been the overlapping
sphincteroplasty. The external anal sphincter muscle and scar tis­
sue, as well as any identifiable internal sphincter muscle, are dis­
sected free from the surrounding adipose and connective tissue. An
overlapping sphincter repair is performed in an attempt to rebuild
the muscular ring and restore its function. However, long-term
results following overlapping sphincteroplasty have been poor, with
a 50% failure rate over 5 years.
Alternative therapies for the treatment of fecal incontinence
included sacral nerve modulation, collagen-enhancing injectables,
and anal implantable self-expanding prostheses (THD Gatekeeper
and Sphinkeeper). Sacral nerve stimulation (SNS) was the first
implantable pacemaker and is a U.S. Food and Drug Administra­
tion (FDA)-approved adaptation of a procedure developed for the
management of urinary incontinence. SNS is indicated in patients
with two or more episodes of frank fecal incontinence per week.
Long-term results for SNS have been promising, with nearly 80% of
patients having a reduction in incontinence episodes by at least 50%.
There are inherent challenges with the SNS device, and to avoid these
challenges, percutaneous or transcutaneous tibial nerve stimulation
(PTNS or TTNS) is being offered to patients with fecal incontinence.
PTNS is offered in a clinical setting, while TTNS can be performed at
home. Both require weekly sessions of stimulation of the tibial nerve,
which serves to feedback to the third sacral nerve root. Collagenenhancing injectables and anal self-expanding implants have been
around for several years. Results from randomized trials have been
variable but support the use of injectables/implantables in patients
who are not candidates for sacral nerve modulation.
Finally, the use of stem cells to increase the bulk of the sphincter
muscles is currently being tested. Stem cells can be harvested from
the patient’s own muscle, grown, and then implanted into their
sphincter complex. Concern for cost and the need for an additional
procedure have dampened enthusiasm.
MIXED CONSTIPATION AND FECAL INCONTINENCE
It is estimated that up to 20% of patients presenting for evaluation of
fecal incontinence have mixed constipation and fecal incontinence.
Fecal incontinence, in this instance, is related to the large number
of laxatives required to produce stool evacuation. This condition
is more difficult to manage and often requires close collaboration
with different specialties. One promising intervention is the use of
a transanal irrigation system (Peristeen by Coloplast). This system
provides daily irrigations of the left side of the colon to assist with
evacuations and has shown promising results. The daily assistance
with evacuation leads to less need for oral laxatives and reduced
episodes of incontinence. One drawback is long-term adherence to
the irrigations.
PART 10
Disorders of the Gastrointestinal System
■
■HEMORRHOIDAL DISEASE
Incidence and Epidemiology 
Symptomatic hemorrhoids
affect >1 million individuals in the Western world per year. Hemor­
rhoidal disease can occur at any age and affects females and males
at similar rates. The prevalence of hemorrhoidal disease is less in
developing countries. The typical low-fiber, high-fat diet prominent
in developed countries is associated with constipation and straining
and the development of symptomatic hemorrhoids. Other risk fac­
tors for hemorrhoids include pregnancy and high-impact activities
(bike riding).
Anatomy and Pathophysiology 
Hemorrhoidal cushions are a
normal part of the anal canal. The vascular structures contained within
this tissue aid in continence by preventing damage to the sphincter
muscle. Three main hemorrhoidal complexes traversing the anal canal
include the left lateral, the right anterior, and the right posterior. Pro­
longed engorgement with straining leads to prolapse of this tissue into
the anal canal. Over time, the anatomic support system of the hemor­
rhoidal complex weakens, exposing this tissue to the outside of the
anal canal where it is susceptible to injury. Hemorrhoids are commonly
classified as external or internal. External hemorrhoids originate below
the dentate line and are covered with squamous epithelium and are not
associated with an internal component. Internal hemorrhoids originate
above the dentate line and are covered with mucosa and transitional
zone epithelium and represent the majority of hemorrhoids.
Presentation and Evaluation 
Patients commonly present to a
physician for two reasons: bleeding and protrusion. Pain is less com­
mon than with fissures and, if present, is described as a dull ache from
engorgement of the hemorrhoidal tissue. Severe pain may indicate a
thrombosed external hemorrhoid. Hemorrhoidal bleeding is described
as painless bright red blood seen either in the toilet or upon wiping.
Occasional patients can present with significant bleeding, which may
be a cause of anemia; however, the presence of a colonic neoplasm
must be ruled out in anemic patients, especially given the noted rise in
young patients with colorectal cancer. Patients who present with pro­
truding tissue complain about inability to maintain perianal hygiene
and anal itching (pruritis) and are often concerned about the presence
of a malignancy.
The diagnosis of hemorrhoidal disease is made on physical exami­
nation. Inspection of the perianal region for evidence of thrombosis
or excoriation is performed, followed by a careful digital examination.
Anoscopy is performed paying particular attention to the known posi­
tion of hemorrhoidal disease. The patient is asked to strain. It is impor­
tant to differentiate the circumferential appearance of a full-thickness
rectal prolapse from the radial nature of prolapsing hemorrhoids (see
“Rectal Prolapse,” above). The stage and location of the hemorrhoidal
complexes are defined.
TREATMENT
Hemorrhoidal Disease
The treatment for bleeding hemorrhoids is based on the stage of the
disease (Table 339-6). In all patients with bleeding, the possibility
of other causes must be considered, and further studies should
include endoscopic assessments.
With rare exceptions, the acutely thrombosed hemorrhoid can be
excised within the first 48 h by performing an elliptical excision and
clot extraction. This can often be performed at the bedside with local
anesthesia; sitz baths, fiber, and stool softeners are prescribed. Addi­
tional nonoperative therapies for bleeding hemorrhoids include rub­
ber band ligation, infrared coagulation, and sclerotherapy. Sensation
begins at the dentate line; therefore, all procedures can be performed
without discomfort either endoscopically or in the office. Bands are
placed around the engorged tissue, causing ischemia and fibrosis.
This aids in fixing the tissue proximally in the anal canal. Patients
may complain of a dull ache for 24 h following band application.
During sclerotherapy, 1–2 mL of a sclerosant (usually sodium tetra­
decyl sulfate) is injected using a 25-gauge needle into the submucosa
of the hemorrhoidal complex. Care must be taken not to inject the
anal canal circumferentially, or stenosis may occur.
For surgical management of hemorrhoidal disease, excisional
hemorrhoidectomy with sharp dissection (Milligan-Morgan
hemorrhoidectomy), bipolar electrocautery ligation (LigaSure),
transhemorrhoidal dearterialization (THD), and stapled hemor­
rhoidectomy (“the procedure for prolapse or hemorrhoids” [PPH])
are the procedures of choice. All surgical methods of management
TABLE 339-6  The Staging and Treatment of Hemorrhoids
DESCRIPTION OF
CLASSIFICATION
TREATMENT
STAGE
I
Enlargement with
bleeding
Fiber supplementation
Short course of cortisone suppository
Sclerotherapy
Infrared coagulation
II
Protrusion with
spontaneous reduction
Fiber supplementation
Short course of cortisone suppository
Sclerotherapy
Infrared coagulation
III
Protrusion requiring
manual reduction
Fiber supplementation
Short course of cortisone suppository
Rubber band ligation
Operative hemorrhoidectomy
IV
Irreducible protrusion
Fiber supplementation
Cortisone suppository
Operative hemorrhoidectomy
are equally effective in the treatment of symptomatic third- and
fourth-degree hemorrhoids. Current postprocedure outcomes fol­
lowing hemorrhoidectomy target improvements in bleeding, pain,
and recurrence rates. The sutured hemorrhoidectomy involves the
removal of redundant tissue down to the anal verge, and unpleasant
anal skin tags are removed as well. However, pain and postopera­
tive bleeding are more common. The stapled hemorrhoidectomy is
associated with less discomfort; however, this procedure does not
remove anal skin tags, and an increased number of complications
are associated with use of the stapling device. THD uses ultrasound
guidance to ligate the blood supply to the anal tissue, hence reduc­
ing hemorrhoidal engorgement. Ligation with electrocautery uses a
bipolar device (LigaSure) to remove the unwanted prolapsed tissue.
Both THD and ligation with electrocautery have demonstrated
similar short-term and long-term results; however, ligation with
electrocautery requires less operative time. No procedures on hem­
orrhoids should be done in patients who are immunocompromised
or who have active proctitis. Emergent hemorrhoidectomy for
bleeding hemorrhoids is associated with a higher complication rate.
Acute complications associated with the treatment of hemor­
rhoids include pain, infection, recurrent bleeding, and urinary
retention. Care should be taken to place bands properly and to
avoid overhydration in patients undergoing operative hemorrhoid­
ectomy. Late complications include fecal incontinence as a result
of injury to the sphincter during the dissection. Anal stenosis
may develop from overzealous excision, with loss of mucosal skin
bridges for reepithelialization. Finally, an ectropion (prolapse of
rectal mucosa from the anal canal) may develop. Patients with an
ectropion complain of a “wet” anus as a result of inability to prevent
soiling once the rectal mucosa is exposed below the dentate line.
■
■ANORECTAL ABSCESS
Incidence and Epidemiology 
The development of a perianal
abscess is more common in men than women by a ratio of 3:1. The
peak incidence is in the third to fifth decade of life. Perianal pain asso­
ciated with the presence of an abscess accounts for 15% of office visits
to a colorectal surgeon. The disease is more prevalent in immunocom­
promised patients such as those with diabetes, hematologic disorders,
or inflammatory bowel disease (IBD) and persons who are HIV posi­
tive. These disorders should be considered in patients with recurrent
perianal infections.
Anatomy and Pathophysiology 
An anorectal abscess is an
abnormal fluid-containing cavity in the anorectal region. Anorectal
abscess results from an infection involving the glands surrounding
the anal canal. Normally, these glands release mucus into the anal
canal, which aids in defecation. When stool accidentally enters the
anal glands, the glands become infected, and an abscess develops.
Anorectal abscesses are perianal in 40–50% of patients, ischiorectal in
20–25%, intersphincteric in 2–5%, and supralevator in 2.5% (Fig. 339-5).
Presentation and Evaluation 
Perianal pain and fever are the
hallmarks of an abscess. Patients may have difficulty voiding and have
blood in the stool. On physical examination, a large fluctuant area is
usually readily visible. Routine laboratory evaluation shows an elevated
WBC count. Diagnostic procedures are rarely necessary unless evalu­
ating a recurrent abscess. A CT scan or MRI has an accuracy of 80%
in determining incomplete drainage. If there is a concern about the
presence of IBD, a rigid or flexible sigmoidoscopic examination may
be done at the time of drainage to evaluate for inflammation within the
rectosigmoid region. A more complete evaluation for Crohn’s disease
would include a full colonoscopy and small-bowel enterography.
TREATMENT
Anorectal Abscess
As with all abscesses, the “gold standard” is drainage. Office drainage
of an uncomplicated anorectal abscess may suffice. A small incision
Fistula
tracts
Abscesses
Supralevator
Intersphincteric
Ischiorectal
Perianal
Intersphincteric
Extrasphincteric
Trans-sphincteric
Suprasphincteric
FIGURE 339-5  Common locations of anorectal abscess (left) and fistula in ano (right).
close to the anal verge is made, and a Mallenkot drain is advanced
into the abscess cavity. The Mallenkot catheter (a mushroom-shaped
catheter with grooves) can be maintained by the patient for up to
1 week before being removed to avoid tissue ingrowth. For patients
with a complicated abscess or who are diabetic or immunocompro­
mised, drainage should be performed in an operating room under
anesthesia. These patients are at greater risk for developing necrotiz­
ing fasciitis. The role of antibiotics in the management of anorectal
abscesses is limited. Antibiotics are only warranted in patients who
are immunocompromised or have obvious cellulitis on physical exam.
CHAPTER 339
■
■FISTULA IN ANO
Diverticular Disease and Common Anorectal Disorders
Incidence and Epidemiology 
The incidence of fistulizing peri­
anal disease parallels the incidence of anorectal abscess and is esti­
mated to be 1 in 10,000 individuals. Some 30–40% of abscesses will
give rise to an anal fistula. Although the majority of the fistulas are
cryptoglandular in origin, 10% are associated with IBD, tuberculosis,
malignancy, and radiation.
Anatomy and Pathophysiology 
An anal fistula is defined as
a communication of an abscess cavity with an identifiable internal
opening within the anal canal. This identifiable opening is commonly
located at the dentate line where the anal glands enter the anal canal.
Patients experiencing continuous drainage at 1 month following the
treatment of a perianal abscess likely have an anal fistula. These fistulas
are classified by their relationship to the anal sphincter muscles, with
70% being intersphincteric, 23% transsphincteric, 5% suprasphincteric,
and 2% extrasphincteric (Fig. 339-5).
Presentation and Evaluation 
A patient with an anal fistula will
complain of constant drainage from the perianal region associated with
tenderness. Examination under anesthesia is the best way to evaluate
a fistula. However, in cases with a complex fistula, a preoperative MRI
will identify tracts with an accuracy of 80%. During surgery, hydrogen
peroxide injected through the external opening will aid in identifying
the internal opening. Goodsall’s rule states that a posterior external
fistula will enter the anal canal in the posterior midline, whereas an
anterior fistula will enter at the nearest crypt. A fistula exiting >3 cm
from the anal verge may have a complicated upward extension and may
not obey Goodsall’s rule.
TREATMENT
Fistula in Ano
A newly diagnosed draining fistula is best managed with an opera­
tive intervention including initial placement of a seton catheter
(a vessel loop or silk tie placed through the fistula tract), which

10 - 340 Mesenteric Vascular Insufficiency
340 Mesenteric Vascular Insufficiency
maintains the tract patency and reduces the likelihood of recurrent
infection and inflammation. Once the inflammation is reduced
(4–6 weeks), the exact relationship of the fistula tract to the anal
sphincters can be ascertained. Two general approaches are com­
mon: sphincter-cutting procedures and sphincter-sparing proce­
dures. A simple fistulotomy or fistulectomy is a sphincter-cutting
procedure and is usually performed for intersphincteric and low
(less than one-third of the muscle) transsphincteric fistulas. In a
systematic review, sphincter-cutting procedures were associated
with a 94% success rate and a 13% rate of mild fecal incontinence.
Sphincter-sparing surgery is usually performed for a higher trans­
sphincteric fistula. These procedures include an anorectal advance­
ment flap, ligation of the internal fistula tract (LIFT procedure),
or insertion of fibrin or cyanoacrylate glue. Sphincter-preserving
procedures are associated with up to 78% success rates with no
incidence of fecal incontinence.
Fistulizing disease of the anus is common in Crohn’s disease,
and the use of mesenchymal stem cell therapy may improve heal­
ing rates of fistula associated with Crohn’s disease. The ADMIRE
study examined the use of allogeneic expanded adipose-derived
mesenchymal stem cells in the treatment of complex perianal fistula
in Crohn’s disease. The study included 212 patients randomized to
stem cell therapy or placebo. Fistula remission rates at 52 weeks
were significantly higher with the use of stem cell therapy over pla­
cebo (59 vs 42%, respectively). Further studies are currently being
performed to elucidate the benefits of stem cell therapy in perianal
fistula. As with any stem cell therapy, the cost can be prohibitive to
its standard use.
PART 10
Disorders of the Gastrointestinal System
■
■ANAL FISSURE
Incidence and Epidemiology 
Anal fissures occur at all ages but
are more common in the third through the fifth decades. The preva­
lence is equal in males and females. It is associated with constipation,
diarrhea, infectious etiologies, perianal trauma, and Crohn’s disease.
Anatomy and Pathophysiology 
Trauma to the anal canal occurs
following defecation. This injury occurs in the anterior or, more com­
monly, posterior anal canal. Irritation caused by the trauma to the anal
canal results in an increased resting pressure of the internal sphincter.
The blood supply to the sphincter and anal mucosa enters laterally.
Therefore, increased anal sphincter tone results in a relative ischemia
in the region of the fissure and leads to poor healing of the anal injury.
A fissure that is not in the posterior or anterior position should raise
suspicion for other causes, including tuberculosis, syphilis, Crohn’s
disease, and malignancy.
Presentation and Evaluation 
A fissure can be easily diagnosed
on history alone. The classic complaint is pain, which is strongly asso­
ciated with defecation and is relentless. The bright red bleeding that
can be associated with a fissure is less extensive than that associated
with hemorrhoids and is usually only noted on wiping. On examina­
tion, most fissures are located in either the posterior or anterior posi­
tion. A lateral fissure is worrisome because it may have a less benign
nature, and systemic disorders should be ruled out. A chronic fissure is
indicated by the presence of a hypertrophied anal papilla at the proxi­
mal end of the fissure and a sentinel pile or skin tag at the distal end.
Often the circular fibers of the hypertrophied internal sphincter are
visible within the base of the fissure.
TREATMENT
Anal Fissure
The management of the acute fissure is conservative. Stool softeners
for those with constipation, increased dietary fiber, topical anes­
thetics, glucocorticoids, and sitz baths are prescribed and will heal
60–90% of fissures. Chronic fissures are those present for >6 weeks.
These can be treated with modalities aimed at decreasing the anal
canal resting pressure including nifedipine ointment applied three
times a day and botulinum toxin type A, up to 20 units, injected into
the internal sphincter on each side of the fissure. Both treatments
are associated with a fissure healing rate of >80%. Surgical manage­
ment includes anal dilatation and lateral internal sphincterotomy.
Usually, one-third of the internal sphincter muscle is divided; it is
easily identified because it is hypertrophied. Recurrence rates from
medical therapy are higher, but this is offset by a risk of inconti­
nence following sphincterotomy. Lateral internal sphincterotomy
may lead to incontinence more commonly in women.
Acknowledgment
The author thanks Cory Sandore for providing some illustrations for this
chapter.
■
■FURTHER READING
Bharucha AE et al: Surgical interventions and the use of device-aided
therapy for the treatment of fecal incontinence and defecatory disor­
ders. Clin Gastroenterol Hepatol 15:1844, 2017.
Daniels L et al: Randomized clinical trial of observation versus antibi­
otic treatment for a first episode of CT-proven uncomplicated acute
diverticulitis (DIABOLO trial). BJS 104:52, 2017.
Guttenplan M: The evaluation and office management of hemor­
rhoids for the gastroenterologist. Curr Gastroenterol Rep 19:30, 2017.
Hall J et al: The American Society of Colon and Rectal Surgeons
Clinical Practice Guidelines for the treatment of left-sided colonic
diverticulitis. Dis Colon Rectum 63:728, 2020.
Panes J et al: Long-term efficacy and safety of stem cell therapy
(Cx601) for complex perianal fistulas in patients with Crohn’s disease.
Gastroenterology 154:1334, 2018.
Prichard D, Bharucha AE: Management of pelvic floor disorders:
Biofeedback and more. Curr Treat Options Gastroenterol 12:456,
2014.
Salfity HV et al: Minimally invasive incision and drainage technique
in the treatment of simple subcutaneous abscess in adults. Am Surg
83:699, 2017.
Strate LL, Morris AM: Epidemiology, pathophysiology, and treat­
ment of diverticulitis. Gastroetenterology 156:1282, 2019.
Sugrue J et al: Sphincter-sparing anal fistula repair: Are we getting
better? Dis Colon Rectum 60:1071, 2017.
Tursi A: Dietary pattern and colonic diverticulosis. Curr Opin Clin
Nutr Metab Care 20:409, 2017.
Daniel Willie-Permor, Mahmoud Malas
Mesenteric Vascular
Insufficiency
INTESTINAL ISCHEMIA
■
■INCIDENCE AND EPIDEMIOLOGY
Intestinal ischemia occurs when splanchnic perfusion fails to meet the
metabolic demands of the intestines, resulting in ischemic tissue injury.
Mesenteric ischemia affects 2–3 people per 100,000 with an increasing
incidence in the aging population. Mortality with acute presentation
remains high (between 50 and 80%), and early diagnosis and prompt
intervention are crucial in improving clinical outcomes. Intestinal isch­
emia is further classified into chronic mesenteric ischemia (CMI) and
acute mesenteric ischemia (AMI). CMI is secondary to multiple major
visceral arterio-occlusive disease with involvement of the superior
mesenteric artery (SMA) most worrisome. AMI is most associated with
(1) arterio-occlusive mesenteric ischemia, (2) nonocclusive mesenteric
ischemia, and (3) mesenteric venous thrombosis.
CMI is the failure to achieve normal postprandial hyperemic intes­
tinal blood flow. This occurs due to an imbalance between supply and
demand of oxygen metabolites to the intestinal tract similar to cardiac
angina. CMI occurs due to significant atherosclerotic disease leading to
the narrowing of the SMA and/or celiac artery origins.
AMI is the occurrence of an abrupt cessation of mesenteric blood
flow, usually embolic or thrombotic in nature. Approximately 50%
of AMI is due to embolus to the mid to distal SMA. The embolus
etiology includes atrial fibrillation, recent myocardial infarction, soft
atherosclerotic plaque, infective endocarditis, valvular heart disease,
and recent cardiac or vascular catheterization. Approximately 25–30%
of the cases are characterized by an acute-on-chronic thrombosis
in patients with preexisting mesenteric atherosclerosis. Thrombotic
occlusion most commonly occurs at areas of severe atherosclerotic
narrowing at the SMA and the celiac artery.
Nonocclusive mesenteric ischemia represents 20% of the cases and
is secondary to intestinal ischemia when subjected to acute hemody­
namic instability. Patients above the age of 50, especially those with
coexisting conditions like myocardial infarction, congestive heart
failure, aortic insufficiency, and renal or liver disease, who are also
undergoing cardiovascular surgery, face the highest risk. Hypovolemia,
shock, and use of vasoconstrictive agents (e.g., digoxin, α-adrenergic
agonists, cocaine) can precipitate ischemia in these patients. It is the
most prevalent gastrointestinal disease complicating cardiovascular
surgery. The incidence of ischemic colitis following elective aortic
repair is 5–9%, and the incidence triples in patients following emergent
repair.
Mesenteric venous thrombosis accounts for <10% of cases and is
generally precipitated by a hypercoagulable state due to an underlying
inherited disorder such as factor V Leiden, prothrombin mutation,
protein S deficiency, protein C deficiency, antithrombin deficiency, and
antiphospholipid syndrome. It may also occur as a result of acquired
thrombophilia in malignancies, hematologic disorders, and use of oral
contraceptives.
■
■ANATOMY AND PATHOPHYSIOLOGY
The blood supply to the intestines is supplied by the celiac artery, SMA,
and inferior mesenteric artery (IMA) (Fig. 340-1). Extensive collat­
eralization occurs between major mesenteric trunks and branches of
the mesenteric arcades. Collateral vessels within the small bowel are
numerous and meet within the duodenum and the bed of the pancreas.
Collateral vessels within the colon meet at the splenic flexure and
descending/sigmoid colon. These areas, which are inherently at risk for
decreased blood flow, are known as Griffiths’ point and Sudeck’s point,
respectively, and are the most common locations for colonic ischemia
(Fig. 340-1, shaded areas). The splanchnic circulation can receive up
to 30% of the cardiac output. Protective responses to prevent intestinal
ischemia include abundant collateralization, autoregulation of blood
flow, and the ability to increase oxygen extraction from the blood.
Occlusive ischemia is a result of disruption of blood flow by an
embolus or progressive thrombosis in a major artery supplying the
intestine. In >75% of cases, emboli originate from the heart and pref­
erentially lodge in the SMA just distal to the origin of the middle colic
artery. Progressive stenosis of typically two of the three major vessels
supplying the intestine is required for the development of chronic
intestinal angina. Involvement of the SMA is most worrisome. Non­
occlusive ischemia is disproportionate mesenteric vasoconstriction
(arteriolar vasospasm) in response to a severe physiologic stress such
as shock. If left untreated, early mucosal stress ulceration will progress
to full-thickness injury.
■
■PRESENTATION, EVALUATION,
AND MANAGEMENT
Patients with CMI typically present with insidious onset of symptoms
and classically present with recurrent episodes of acute dull, crampy,
postprandial epigastric pain, which has also been referred to as “intesti­
nal angina.” Patients also describe fear of eating resulting in weight loss.
Left phrenic a.
Aorta
Right phrenic a.
Splenic a.
Griffiths’
point
Celiac trunk
Pancreaticoduodenal a.
Arc of
Riolan
SMA
IMA
Marginal a.
IIA
Sudeck’s
point
Hemorrhoidal aa.
Superior
Middle
Inferior
CHAPTER 340
FIGURE 340-1  Blood supply to the intestines includes the celiac artery, superior
mesenteric artery (SMA), inferior mesenteric artery (IMA), and branches of the
internal iliac artery (IIA). Griffiths’ and Sudeck’s points, indicated by shaded areas,
are watershed areas within the colonic blood supply and common locations for
ischemia.
Mesenteric Vascular Insufficiency
Chronic diarrhea may also be noted. Duration of symptoms is typically
6–12 months. Physical examination will often reveal a malnourished
patient with other manifestations of atherosclerosis. Prompt diagnostic
investigation is necessary to rule out gastrointestinal malignancies
and other possible causes, and may include accelerated evaluation
via esophagogastroduodenoscopy, colonoscopy, abdominal computed
tomography (CT) scan, and abdominal ultrasound examination.
Duplex ultrasound has gained popularity as a screening tool for
evaluation of the mesenteric vessels due to high sensitivity and speci­
ficity. Mesenteric duplex scan demonstrating a high peak velocity of
flow in the SMA is associated with an ~80% positive predictive value
of mesenteric ischemia. More significantly, a negative duplex scan vir­
tually precludes the diagnosis of mesenteric ischemia. It is important
to perform the test while the patient is fasting because the presence
of increased bowel gas prevents adequate visualization of flow distur­
bances within the vessels or the lack of a vasodilation response to feed­
ing during the test. Thin-sliced CT angiography is the gold standard
diagnostic tool in assessing the degree of atherosclerotic disease of
the aortic and visceral vessels as well as evaluating the bowels. Venous
phase can also help diagnose mesenteric vein thrombosis.
The management of CMI includes aggressive medical therapy of
atherosclerotic disease including cessation of smoking and antiplatelet
and lipid-lowering medications. A full cardiac and vascular evaluation
should be performed before intervention on CMI.
Treatment, involving either endovascular, open surgical, or hybrid
revascularization, should be individualized based on the patient’s
comorbidities and anatomy. Endovascular revascularization involves
targeted vessel treatment with visceral stents with the SMA anatomy
being the key determinant. The revascularization of the celiac axis and
IMA represents secondary focal points, offering potential therapeutic
benefits. This approach becomes particularly relevant when the SMA
is deemed unsuitable for intervention or when technical outcomes
are deemed suboptimal. Open revascularization involves antegrade
bypass from the supraceliac aorta or retrograde bypass typically from
the common or external iliac arteries with a synthetic or autogenous
graft to the targeted vessels, usually the SMA and/or celiac artery. In
patients with suitable lesions requiring revascularization, an endovas­
cular approach is recommended as the first-line therapy. It is especially
favorable for short segment stenosis with minimal to moderate calci­
fication or thrombus. Angioplasty with endovascular stenting in the
treatment of CMI is associated with an 80% long-term success rate.
Open revascularization should be considered in patients with lesions
not amenable to endovascular treatment, such as severe calcification,
longer lesions, small vessel diameter, or failed endovascular interven­
tions, or in a specific subset of younger, healthier patients, in whom
the potential long-term advantages may outweigh the heightened
perioperative risks. Retrograde open mesenteric stenting (ROMS) is a
hybrid approach, combining aspects of both traditional open surgical
bypass and percutaneous endovascular therapy. ROMS is primarily
indicated for treating mesenteric ischemia, both acute and chronic,
particularly in cases where conventional endovascular or open surgical
approaches are not feasible or have been unsuccessful. This includes
scenarios where there is significant stenosis or occlusion in the mesen­
teric arteries that cannot be adequately addressed through less invasive
percutaneous methods from the aorta or in situations where immediate
direct visualization and potential resection of the bowel are necessary
due to the presence of necrosis or perforation. The technique involves
a laparotomy that allows for direct bowel assessment and the exposure
and stenting of the SMA using a retrograde approach through the
midsegment of the SMA, thus allowing for immediate revasculariza­
tion, assessment of bowel viability, and, if necessary, bowel resection.
ROMS also offers the advantage of faster operative times compared
to traditional bypass and avoids the placement of prosthetic material
in potentially contaminated peritoneal cavities, which is a significant
concern in the setting of bowel necrosis and peritonitis.
PART 10
Disorders of the Gastrointestinal System
Acute intestinal ischemia remains one of the most challenging
diagnoses. The mortality rate of AMI is >50%. The most significant
indicator of survival is the timeliness of diagnosis and treatment. An
overview of diagnosis and management of each form of intestinal isch­
emia is given in Table 340-1.
AMI resulting from arterial embolus or thrombosis presentation
is nonspecific and requires a high index of suspicion for the diag­
nosis. The most common complaint, occurring in 95% of cases, is
severe, acute, nonremitting abdominal pain that is strikingly out of
proportion to the physical findings. The reason behind the pain being
TABLE 340-1  Overview of the Management of Acute Intestinal Ischemia
KEY TO EARLY
DIAGNOSIS
TREATMENT OF
UNDERLYING CAUSE
TREATMENT OF SPECIFIC LESION
CONDITION
Arterio-occlusive
mesenteric ischemia
1.  Arterial embolus
Computed tomography
angiography (CTA)
Early laparotomy
Anticoagulation
Cardioversion
Thrombectomy
Broad-spectrum antibiotics
2.  Arterial thrombosis
Duplex ultrasound
CTA
Anticoagulation
Broad-spectrum antibiotics
Resuscitation
Mesenteric venous
thrombosis
Venous thrombosis
CTA with venous phase
Anticoagulation
Resuscitation
Nonocclusive
mesenteric ischemia
Vasospasm:
Hypoperfusion:
CT
Resuscitation
Support cardiac output
Avoid vasoconstrictors
Broad-spectrum antibiotics
Source: Modified from GB Bulkley, in JL Cameron (ed): Current Surgical Therapy, 2nd ed. Toronto, BC Decker, 1986.
disproportionate to the clinical findings is that ischemia initiates from
the mucosal layer and progresses toward the serosal layer. This may be
associated with nausea (44%), vomiting (35%), diarrhea (35%), and
blood per rectum (16%). Later findings will demonstrate peritonitis
and cardiovascular collapse. Specific clinical features can help differen­
tiate the underlying etiology, whether embolic or thrombotic. Patients
with embolic ischemia are typically older adults with underlying con­
ditions that predispose to embolism such as atrial fibrillation, prior
embolic event, or recent infective endocarditis. Thrombotic ischemia
typically presents as an acute occlusion in patients with underlying
atherosclerotic disease who may have been previously diagnosed with
CMI.
AMI is a surgical emergency requiring emergent admission to
a monitored bed or intensive care unit for resuscitation with fluids
and broad-spectrum antibiotics in addition to further evaluation. If
the diagnosis of intestinal ischemia is being considered, consultation
with a surgical service is necessary. Often the decision to operate
is made on a high index of suspicion from the history and physical
exam despite normal laboratory findings. In patients with suspected
AMI, a CT angiography with 1-mm or thinner cuts should be used to
detect mesenteric arterial occlusive disease most likely from embolic
or thrombotic etiology and is the gold standard. Additional diagnostic
modalities that may be useful in diagnosis, but should not delay surgi­
cal therapy, include electrocardiogram (ECG) and echocardiogram.
Patients with AMI should be given a heparin bolus immediately and
started on a therapeutic heparin drip. Correction of electrolyte abnor­
malities and empiric broad-spectrum antibiotic therapy should also be
initiated instantly.
If CT angiography verifies acute embolic occlusion of the SMA, sur­
gical exploration should not be delayed. The goal of operative explora­
tion is to resect compromised bowel, restore blood supply, and preserve
all viable bowel. The entire length of the small and large bowel begin­
ning at the ligament of Treitz should be evaluated. The SMA artery
should be localized, typically at the mesocolon of the transverse colon.
A transverse arteriotomy of the SMA should be made with removal of
embolus with an embolectomy Fogarty catheter passed in a retrograde
and antegrade manner to restore blood flow. In the case of SMA occlu­
sion where the embolus usually lies just proximal to the origin of the
middle colic artery, the proximal jejunum is often spared while the
remainder of the small bowel up to the transverse colon may become
ischemic. Nonviable bowel should be resected. Questionable bowel
TREATMENT OF SYSTEMIC
CONSEQUENCE
Laparotomy
Embolectomy
Assess viability and resect nonviable
bowel
Anticoagulation
Resuscitation
Broad-spectrum antibiotics
Emergent surgical intervention
Assessment of bowel
Endovascular approach: thrombolysis,
angioplasty, and stenting
Endarterectomy/thrombectomy or
vascular bypass
Assess viability and resect nonviable
bowel
Anticoagulation
Resuscitation
Broad-spectrum antibiotics
Emergent surgical intervention
Assessment of bowel
Anticoagulation
Hypercoagulable workup
Anticoagulation
Resuscitation
Broad-spectrum antibiotics
Support cardiac output
Avoid vasoconstrictors
Vasospasm
Intraarterial vasodilators
Hypoperfusion
Assess viability and resect dead bowel
Resuscitation
Broad-spectrum antibiotics
Support cardiac output
Avoid vasoconstrictors

11 - 341 Acute Intestinal Obstruction
341 Acute Intestinal Obstruction
should undergo a second-look laparotomy in a 24- to 48-h period.
After revascularization, peristalsis and return of a pink color of the
bowel wall should be observed. Palpation of major arterial mesenteric
vessels can be performed, as well as applying a Doppler flowmeter to
the antimesenteric border of the bowel wall, but neither is a definitive
indicator of viability.
Acute-on-chronic mesenteric ischemia typically involves the orifice
of the SMA. Therefore, the entire small bowel is compromised. Revas­
cularization with an endovascular, open, and/or a hybrid approach
should be individualized based the patient’s critical status, comorbidi­
ties, and anatomy. Endovascular stenting, suction thrombectomy, and/
or a thrombolysis catheter should be considered for intervention. The
bowel should be evaluated for viability typically via a laparoscopic or
exploratory laparotomy.
Nonocclusive or vasospastic mesenteric ischemia presents with gen­
eralized abdominal pain, anorexia, bloody stools, and abdominal dis­
tention. Often these patients are obtunded, and physical findings may
not assist in the diagnosis or be obscured by the underlying etiology.
The presence of leukocytosis, metabolic acidosis, and/or lactic acidosis
is useful in support of the diagnosis of advanced intestinal ischemia;
however, these markers may not be indicative of either reversible isch­
emia or frank necrosis.
Emergent admission to a monitored bed or intensive care unit is
recommended for resuscitation and further evaluation, and the patient
should be started on broad-spectrum antibiotics. Anticoagulation is
not recommended as the goal is resuscitation to maintain hemody­
namics. For select patients, intramesenteric infusion of vasodilators
such as papaverine, prostaglandins, or nitroglycerin can be used for
the reversal of spasm and mesenteric ischemia, but the priority should
be resuscitation and treatment of the underlying pathology. If ischemic
colitis is a concern, colonoscopy should be considered to assess the
integrity of the colonic mucosa. Ischemia of the colonic mucosa is
graded as mild with minimal mucosal erythema or as moderate with
pale mucosal ulcerations and evidence of extension to the muscular
layer of the bowel wall. Severe ischemic colitis presents with severe
ulcerations resulting in black or green discoloration of the mucosa,
consistent with full-thickness bowel-wall necrosis. Ischemic colitis is
optimally treated with resection of the ischemic bowel and formation
of a proximal stoma.
Onset of mesenteric venous thrombosis can be acute or subacute
based on the location of thrombosis in the splanchnic circulation.
Patients often present with vague abdominal pain associated with nau­
sea and vomiting. Physical examination findings include abdominal
distention with mild to moderate tenderness and signs of dehydra­
tion. Findings on CT delayed venous phase include diffuse bowel-wall
thickening and thrombus within the splanchnic system. IV therapeutic
anticoagulation, broad-spectrum antibiotics, and correction of electro­
lyte abnormalities should be performed. Surgical intervention is not
performed unless there is evidence of peritonitis and/or bowel perfora­
tion. If there is evidence of bowel compromise, an exploratory laparot­
omy should be performed, with resection of the compromised bowel.
Second-look laparotomy after 24–48 h should be attempted because
anticoagulation can help prevent resection of viable bowel. Hyperco­
agulability testing should be performed, and if underlying inherited
disorders are diagnosed, life-long anticoagulation is recommended.
Acknowledgments
Rizwan Ahmed contributed to the 19th edition and Maryam Khan and
Jaideep Das Gupta contributed to the 21st edition.
■
■FURTHER READING
Bala M et al: Acute mesenteric ischemia: Updated guidelines of the World
Society of Emergency Surgery. World J Emerg Surg 17:54, 2022.
Cirillo-Penn NC et al: Midterm clinical outcomes of retrograde open
mesenteric stenting for mesenteric ischemia. Ann Vasc Surg 89:20,
2023.
Deng QW et al: Risk factors for postoperative acute mesenteric isch­
emia among adult patients undergoing cardiac surgery: A systematic
review and meta-analysis. J Crit Care 42:294, 2017.
Oderich GS et al: Multicenter study of retrograde open mesenteric
artery stenting through laparotomy for treatment of acute and
chronic mesenteric ischemia. J Vasc Surg 68:470, 2018.
Salsano G et al: What is the best revascularization strategy for acute
occlusive arterial mesenteric ischemia: Systematic review and metaanalysis. Cardiovasc Intervent Radiol 41:27, 2018.
Sise MJ: Acute mesenteric ischemia. Surg Clin North Am 94:165, 2014.
Danny O. Jacobs
Acute Intestinal
Obstruction
■
■EPIDEMIOLOGY
Globally, although the incidence and prevalence of acute intestinal
obstruction have increased over the past two decades, morbidity and
mortality appear to be decreasing. Diagnosis remains challenging. The
extent of mechanical obstruction is typically described as partial, high
grade, or complete—generally correlating with the risk of complica­
tions and the urgency with which the underlying disease process must
be addressed. Obstruction is also commonly described as being either
“simple” or, alternatively, “strangulated” if vascular insufficiency and
intestinal ischemia are evident.
CHAPTER 341
Acute intestinal obstruction occurs either mechanically from block­
age or from intestinal dysmotility when there is no blockage. In
the latter instance, the abnormality is functional. Mechanical bowel
obstruction may be caused by extrinsic processes, intrinsic abnor­
malities of the bowel wall, or intraluminal abnormalities (Table 341-1).
Within each of these broad categories are many diseases that can
impede intestinal propulsion. Intrinsic diseases that can cause intes­
tinal obstruction are usually congenital, inflammatory, neoplastic, or
traumatic in origin, although intussusception and radiation injury can
also be etiologic.
Acute Intestinal Obstruction
Acute intestinal obstruction accounts for ~1–3% of all hospitaliza­
tions and a quarter of all urgent or emergent general surgery admis­
sions. Approximately 80% of cases involve the small bowel, and about
one-third of these patients show evidence of significant ischemia. The
mortality rate for patients with strangulation who are operated on within
24–30 h of the onset of symptoms is ~8% but triples shortly thereafter.
Extrinsic diseases most commonly cause mechanical obstruction of
the small intestine. In the United States and Europe, almost all cases
are caused by postoperative adhesions, carcinomatosis, or herniation
of the anterior abdominal wall. Carcinomatosis most often originates
from the ovary, but can originate from the pancreas, stomach, or colon.
Rarely, metastasis from distant organs like the breast and skin can also
occur.
Adhesions are responsible for the majority of cases of early post­
operative obstruction that require intervention. Approximately 20%
of patients who were treated conservatively and between 5 and 30%
of patients who were managed operatively will require readmission
within 10 years for recurrence.
Open operations of the lower abdomen, including appendectomy
and colorectal and gynecologic procedures, are especially likely to
create adhesions that can cause bowel obstruction (Table 341-2).
Although laparoscopic procedures may generate fewer postopera­
tive adhesions compared with open surgery, the risk of obstructive
adhesion formation is not eliminated. The risk of internal herniation
is increased by abdominal procedures such as laparoscopic or open
Roux-en-Y gastric bypass.
Volvulus, which occurs when bowel twists on its mesenteric axis,
can cause partial or complete obstruction and vascular insufficiency
TABLE 341-1  Most Common Causes of Acute Intestinal Obstruction
Extrinsic Disease
Adhesions (especially due to previous abdominal surgery), internal or external
hernias, neoplasms (including carcinomatosis and extraintestinal malignancies,
mostly commonly ovarian), endometriosis or intraperitoneal abscesses, and
idiopathic sclerosis
Intrinsic Disease
Congenital (e.g., malrotation, atresia, stenosis, intestinal duplication, cyst
formation, and congenital bands—the latter rarely in adults)
Inflammation (e.g., inflammatory bowel disease, especially Crohn’s disease,
but also diverticulitis, radiation, tuberculosis, lymphogranuloma venereum, and
schistosomiasis)
Neoplasia (note: primary small-bowel cancer is rare; obstructive colon cancer
may mimic small-bowel obstruction if the ileocecal valve is incompetent)
Traumatic (e.g., hematoma formation, anastomotic strictures)
Other, including intussusception (where the lead point is typically a polyp or
tumor in adults), volvulus, obstruction of duodenum by superior mesenteric
artery, radiation or ischemic injury, and aganglionosis, which is Hirschsprung’s
disease
Intraluminal Abnormalities
Bezoars, feces, foreign bodies including inspissated barium, gallstones (entering
the lumen via a cholecystoenteric fistula), enteroliths
affecting the sigmoid colon most commonly comprising approximately
two-thirds of all cases of volvulus and 4% of all cases of large-bowel
obstruction. The cecum and terminal ileum can also volvulize, or
the cecum alone may be involved as a cecal bascule. Risk factors
include institutionalization, the presence of neuropsychiatric condi­
tions requiring psychotropic medication, chronic constipation, and
aging; patients typically present in their seventies or eighties. Colonic
volvulus is more common in Eastern Europe, Russia, and Africa than
it is the United States. It is rare for adhesions or hernias to obstruct
the colon. Cancer of the descending colon and rectum is responsible
for approximately two-thirds of all cases, followed by diverticulitis.
Functional obstruction, also known as ileus and pseudo-obstruction,
is present when dysmotility prevents intestinal contents from being
propelled distally and no mechanical blockage exists. Ileus that occurs
after intraabdominal surgery is the most commonly known form of
functional bowel obstruction, but there are numerous other causes
(Table 341-3). Although postoperative ileus is most often transient,
it is a common reason why hospital discharge is delayed. Pseudoobstruction of the colon, also known as Ogilvie’s syndrome, is a rela­
tively rare disease. Some patients with Ogilvie’s syndrome have colonic
dysmotility due to abnormalities of their autonomic nervous system
that may be inherited.
PART 10
Disorders of the Gastrointestinal System
■
■PATHOPHYSIOLOGY
The manifestations of acute intestinal obstruction depend on the
nature of the underlying disease process, its location, and changes in
blood flow (Fig. 341-1). Increased intestinal contractility, which occurs
proximally and distal to the obstruction, is a characteristic response.
Subsequently, intestinal peristalsis slows as the intestine or stomach
proximal to the point of obstruction dilates and fills with gastrointesti­
nal secretions and swallowed air. Although swallowed air is a primary
TABLE 341-2  Acute Small-Intestinal and Colonic Obstruction
Incidences
CAUSE
INCIDENCE
Postoperative adhesions
50% overall
Neoplasms
~20%
Hernias (especially ventral or internal types, where the risk of
strangulation is increased)
~10%
Inflammatory bowel disease, other inflammation (obstruction
may resolve if acute inflammation and edema subside)
~5%
Intussusception, volvulus, other miscellaneous diseases
<15%
TABLE 341-3  Most Common Causes of Ileus (Functional or
Pseudo-Obstruction of the Intestine)
Intraabdominal procedures, lumbar spinal injuries, or surgical procedures on the
lumbar spine and pelvis
Metabolic or electrolyte abnormalities, especially hypokalemia and
hypomagnesemia, but also hyponatremia, uremia, and severe hyperglycemia
Drugs such as opiates, antihistamines, and some psychotropic (e.g., haloperidol,
tricyclic antidepressants) and anticholinergic agents
Intestinal ischemia
Intraabdominal or retroperitoneal inflammation or hemorrhage
Lower lobe pneumonias
Intraoperative radiation (likely due to smooth muscle damage)
Systemic sepsis
Hyperparathyroidism
Pseudo-obstruction (Ogilvie’s syndrome)
Ileus secondary to hereditary or acquired visceral myopathies and neuropathies
that disrupt myocellular neural coordination
Some collagen vascular diseases such as lupus erythematosus or scleroderma
source of intestinal distension, intraluminal air may also accumulate
from fermentation, local carbon dioxide production, and altered gas­
eous diffusion.
Intraluminal dilation also increases intraluminal pressure. When
luminal pressure exceeds venous pressure, venous and lymphatic
drainage is impeded. Edema ensues, and the bowel wall proximal to
the site of blockage may become hypoxemic. Epithelial necrosis can be
identified within 12 h of obstruction. Ultimately, arterial blood supply
may become so compromised that full-thickness ischemia, necrosis,
and perforation result. Stasis increases bacteria counts within the jeju­
num and ileum. Bacteria, such as Escherichia coli, Streptococcus faecalis,
and Klebsiella, and other pathogens may be recovered from intestinal
cultures, mesenteric lymph nodes, the bloodstream, and other sites.
Other manifestations depend on the degree of hypovolemia, the
patient’s metabolic response, and the presence or absence of associ­
ated intestinal ischemia. Inflammatory edema eventually increases the
production of reactive oxygen species and activates neutrophils and
macrophages, which accumulate within the bowel wall. Their accumu­
lation, along with changes in innate immunity, disrupts secretory and
neuromotor processes. Dehydration is caused by loss of the normal
intestinal absorptive capacity as well as fluid accumulation in the gas­
tric or intestinal wall and intraperitoneally.
Anorexia and emesis tend to exacerbate intravascular volume deple­
tion. In the worst-case scenario after high-grade distal obstruction,
emesis leads to losses of gastric potassium, hydrogen, and chloride,
while dehydration stimulates proximal renal tubule bicarbonate reab­
sorption. Intraperitoneal fluid accumulation, especially in patients with
severe distal bowel obstruction, may increase intraabdominal pressure
enough to elevate the diaphragm, inhibit respiration, impede systemic
venous return, and promote vascular instability. Severe hemodynamic
compromise may elicit a systemic inflammatory response and general­
ized microvascular leakage.
Closed-loop obstruction results when the proximal and distal open­
ings of a given bowel segment are both occluded, for example, due
to volvulus or a hernia. It is the most common precursor for stran­
gulation, but not every closed loop strangulates. The risk of vascular
insufficiency, systemic inflammation, hemodynamic compromise,
and irreversible intestinal ischemia is much greater in patients with
closed-loop obstruction. Pathologic changes may occur rapidly, such
that emergent intervention is indicated. Irreversible bowel ischemia
may progress to transmural necrosis even if obstruction is relieved.
The provider should remember that patients with high-grade distal
colonic obstruction who have competent ileocecal valves may pres­
ent with closed-loop obstruction. In this instance, the cecum may
progressively dilate such that ischemic necrosis results in perforation,
especially when the cecal diameter exceeds 10–12 cm. Patients with
distal colonic obstruction whose ileocecal valves are incompetent tend
Abnormal
bacteria
colonization
Patients with distal
obstruction may still
discharge intraluminal
contents
Note: intraluminal obstruction is displayed
FIGURE 341-1  Pathophysiologic changes of small-bowel obstruction.
to present later in the course of disease and mimic patients with distal
small-bowel obstruction.
■
■HISTORY AND PHYSICAL FINDINGS
Even though the presenting signs and symptoms can be misleading,
many patients with acute obstruction can be accurately diagnosed
after a thorough history and physical examination is performed
before imaging. Even though small-bowel obstruction with strangu­
lation can be especially difficult to diagnosis promptly, early recogni­
tion allows earlier treatment, which decreases the risk of morbidity
and mortality.
The cardinal signs are colicky abdominal pain, abdominal disten­
tion, emesis, and obstipation. More intraluminal fluid accumulates
in patients with distal obstruction, which typically leads to greater
distention, more discomfort, and delayed emesis. This emesis is fecu­
lent when there is bacterial overgrowth. Patients with more proximal
obstruction commonly present with less abdominal distention but
more pronounced vomiting. Elements of the history that might be
helpful include any prior history of surgery, including herniorrhaphy,
as well as any history of cancer or inflammatory bowel disease.
Most patients, even those with simple obstruction, appear to be crit­
ically ill. Many may be oliguric, hypotensive, and tachycardic because
of severe intravascular volume depletion. Fever is worrisome for stran­
gulation or systemic inflammation. Bowel sounds and bowel functional
activity are notoriously difficult to interpret. Classically, many patients
with early small-bowel obstruction will have high-pitched, “musical”
tinkling bowel sounds and peristaltic “rushes” known as borborygmi.
Later in the course of disease, the bowel sounds may be absent or hypo­
active as peristaltic activity decreases. This contrasts with the common
Inflammatory
mediators
released
Fluid and air
accumulate; bacteria
overgrowth may occur
Air
Epithelial
necrosis
Fluid
Proximal
bowel
dilatation
Inflammatory
wall edema
Point of obstruction
from extrinsic, intrinsic,
or intraluminal disease
CHAPTER 341
Collapsed
bowel distal
to obstruction
Acute Intestinal Obstruction
findings in patients with ileus or pseudo-obstruction where bowel
sounds are typically absent or hypoactive from the beginning. Lastly,
patients with partial blockage may continue to pass flatus and stool,
and those with complete blockage may even evacuate bowel contents
present downstream beyond their obstruction.
All surgical incisions should be examined, and the presence of a
tender abdominal or groin mass strongly suggests that an incarcerated
hernia may be the cause of obstruction. The presence of tenderness
should increase the concern about the presence of complications such
as ischemia, necrosis, or localized peritonitis. Severe pain with local­
ization or signs of peritoneal irritation is suspicious for strangulated
or closed-loop obstruction. It is important to remember that the dis­
comfort may be out of proportion to physical findings mimicking the
complaints of patients with acute mesenteric ischemia. Patients with
colonic volvulus present with the classic manifestations of closed-loop
obstruction: severe abdominal pain, vomiting, and obstipation. Asym­
metrical abdominal distension and a tympanic mass may be evident.
Patients with ileus or pseudo-obstruction may have signs and
symptoms like those with bowel obstruction. Although abdominal
distention is present, colicky abdominal pain is typically absent, and
patients may not have nausea or emesis. Ongoing, regular discharge
of stool or flatus can sometimes help distinguish patients with ileus
from those with complete mechanical bowel obstruction. The overall
risk of ileus appears to be less in patients who undergo laparoscopic
procedures.
■
■LABORATORY AND IMAGING STUDIES
Laboratory testing should include a complete blood count and serum
electrolyte and creatinine measurements. Serial assessments are often
useful. Mild hemoconcentration and slight elevation of the white blood
cell count commonly occur after simple bowel obstruction. Emesis and
dehydration may cause hypokalemia, hypochloremia, elevated blood
urea nitrogen–to–creatinine ratios, and metabolic alkalosis. Patients
may be hyponatremic on admission because many have attempted to
rehydrate themselves with hypotonic fluids. The presence of guaiacpositive stools and iron-deficiency anemia are strongly suggestive of
malignancy.
Higher white blood cell counts with the presence of immature
forms and the presence of metabolic acidosis are worrisome for severe
volume depletion or ischemic necrosis and sepsis. Presently, no labora­
tory tests are especially useful for identifying the presence of simple
or strangulated obstruction, although increases in serum d-lactate,
creatine kinase BB isoenzymes, or intestinal fatty acid binding protein
levels may be suggestive of the latter.
Recommendations for diagnostic imaging continue to evolve. In
all cases, the key is not to delay operative intervention unnecessarily
when the patient’s signs or symptoms strongly suggest that highgrade or complete obstruction or bowel compromise is present.
Abdominal radiography, which must include upright or cross-table
lateral views, can be completed quickly and may indicate the need
for emergency surgical intervention in patients who are not in the
immediate postoperative period. A “staircasing” pattern of dilated
air and fluid-filled small-bowel loops >2.5 cm in diameter with little
or no air seen in the colon are classical findings in patients with
small-bowel obstruction. Little bowel gas appears in patients with
proximal bowel obstruction or in patients whose intestinal lumens
are filled with fluid.
Upright plain films of the abdomen of patients with large-bowel
obstruction typically show colon dilatation. Small-bowel air-fluid levels
may not be obvious if the ileocecal valve is incompetent. Although it
can be difficult to distinguish from ileus, small-bowel obstruction is
more likely when air-fluid levels are seen without significant colonic
distension. Free air suggests that perforation has occurred in patients
who have not recently undergone surgical procedures. A gas-filled,
“coffee bean”–shaped dilated shadow may be seen in patients with
volvulus.
PART 10
Disorders of the Gastrointestinal System
More sophisticated imaging can be beneficial when the diagnosis
is unclear. Computed tomography (CT) is the most frequently used
imaging modality. Its sensitivity for detecting bowel obstruction is
~95% (78–100%) in patients with high-grade obstruction, with a
specificity of 96% and an accuracy of ≥95%. Its accuracy in diagnosing
closed-loop obstruction is much lower (60%). CT may also provide
useful information regarding location or to identify circumstances
where surgical intervention is urgently needed. Patients who have
evidence of contrast appearing within the cecum within 4–24 h of oral
administration of water-soluble contrast can be expected to improve
with high sensitivity and specificity (~95% each).
Contrast studies may demonstrate a “bird’s beak,” a “c-loop,” or
“whorl” deformity on CT imaging at the site where twisting obstructs
the lumen when a colonic volvulus is present. Abdominal radiography,
unlike CT imaging, may not accurately distinguish obstruction from
other causes of colonic dysmotility. Examples of some CT images are
provided in Fig. 341-2.
Ultrasonographic evaluations are especially difficult to interpret but
may be sensitive and appropriate studies to evaluate patients who are
pregnant or for whom x-ray exposure is otherwise contraindicated or
inappropriate.
CT imaging with enteral and IV contrast can also identify ischemia.
Altered bowel wall enhancement is the most specific early finding, but
its sensitivity is low. Mesenteric venous gas, pneumoperitoneum, and
pneumatosis intestinalis are late findings indicating the presence of
bowel necrosis. CT scanning after a water-soluble contrast enema may
help distinguish ileus or pseudo-obstruction from distal large-bowel
obstruction in patients who present with evidence of small-bowel and
colonic distention. CT enteroclysis, though now rarely performed, can
accurately identify neoplasia as a cause of bowel obstruction. Contrast
enemas or colonoscopies are almost always needed to identify causes
of acute colonic obstruction.
A
B
C
FIGURE 341-2  Computed tomography with oral and intravenous contrast
demonstrating (A) evidence of small-bowel dilatation with air-fluid levels consistent
with a small-bowel obstruction; (B) a partial small-bowel obstruction from an
incarcerated ventral hernia (arrow); and (C) decompressed bowel seen distal to
the hernia (arrow). (Reproduced with permission from D Longo et al: Harrison’s
Principles of Internal Medicine, 18th ed. New York: McGraw-Hill; 2012.)
Barium studies are generally contraindicated in patients with firm
evidence of complete or high-grade bowel obstruction, especially
when they present acutely. Barium should never be given orally to
a patient with possible obstruction until that diagnosis has been
excluded. In every other instance, such investigations should only
be performed in exceptional circumstances and with great caution
because patients with significant obstruction may develop barium
concretions as an additional source of blockage and some who would
have otherwise recovered will require operative intervention. Barium
opacification also renders cross-sectional imaging studies or angiogra­
phy uninterpretable.
TREATMENT
Acute Intestinal Obstruction
An improved understanding of the pathophysiology of bowel
obstruction and the importance of fluid resuscitation, electrolyte
repletion, intestinal decompression, and the selected use of antibi­
otics has likely contributed to a reduction in mortality from acute
bowel obstruction. Patients should be stabilized as quickly as possi­
ble. Nasogastric tube suction decompresses the stomach, minimizes
further distention from swallowed air, improves patient comfort,
and reduces the risk of aspiration. Urine output should be assessed
using a Foley catheter. In some cases, for example, in patients with
cardiac disease, central venous pressures should be monitored. The
use of antibiotics is controversial, although prophylactic adminis­
tration may be warranted if operation is anticipated.
Complete bowel obstruction is an indication for intervention.
Stenting may be possible and warranted for some patients with
high-grade obstruction due to unresectable stage IV malignancies.
Stenting may also allow elective mechanical bowel preparation
before operation. Because treatment options are so variable, it is
helpful to make as precise a diagnosis as possible preoperatively.
ILEUS
Patients with ileus are treated supportively with IV fluids and
nasogastric decompression while any underlying pathology is
treated, taking care to optimize the use of narcotics. Pharma­
cologic treatments continue to be evaluated with some studies
showing that treatment with peripherally active μ-opioid recep­
tor antagonists (e.g., alvimopan and methylnaltrexone) or 5-HT4
agonists that stimulate the release of acetylcholine from enteric
neurons (e.g., mosapride and prucalopride) may accelerate gas­
trointestinal recovery in some patients who have undergone
abdominal surgery.
COLONIC PSEUDO-OBSTRUCTION (OGILVIE’S DISEASE)
Neostigmine is an acetylcholinesterase inhibitor that increases cho­
linergic (parasympathetic) activity, which can stimulate colonic
motility. Some studies have shown it to be moderately effective in
alleviating acute colonic pseudo-obstruction. It is the most com­
mon therapeutic approach and can be used once it is certain that
there is no mechanical obstruction. Cardiac monitoring is required,
and atropine should be immediately available. Intravenous admin­
istration induces defecation and flatus within 10 min in the major­
ity of patients who will respond. Sympathetic blockade by epidural
anesthesia can successfully ameliorate pseudo-obstruction in some
patients.
VOLVULUS
Patients with sigmoid volvulus can often be decompressed using a
flexible tube inserted through a rigid proctoscope or using a flexible
sigmoidoscope. Successful decompression results in sudden release
of gas and fluid with evidence of decreased abdominal distension
and allows definitive correction to be scheduled electively. Cecal
volvulus most often requires laparotomy or laparoscopic correction.
INTRAOPERATIVE STRATEGIES
Approximately 60–80% of selected patients with mechanical bowel
obstruction can be successfully treated conservatively. Most cases of
radiation-induced obstruction should be managed nonoperatively
if possible. Early consultation with a surgeon is prudent when there
is concern about strangulation obstruction or other abnormality
that needs to be addressed urgently. Deterioration signifies a need
for intervention. At this time, the decision as to whether the patient
can continue to be treated nonoperatively can only be based on
clinical judgment, although, as described earlier, imaging studies
can sometimes be helpful. The frequency of major complications
after operation ranges from 12 to 47%, with greater risk being
attributed to resection therapies and the patient’s overall health.
Risk is increased for patients with American Society of Anesthesi­
ologists (ASA) physical status of class III or higher.
At operation, dilation proximal to the site of blockage with distal
collapse is a defining feature of bowel obstruction. Intraopera­
tive strategies depend on the underlying problem and range from
lysis of adhesions to resection with or without diverting ostomy to
primary resection with anastomosis. Resection is warranted when
there is concern about the bowel’s viability after the obstructive pro­
cess is relieved. Laparoscopic approaches can be useful for patients
with early obstruction when extensive adhesions are not expected
to be present. Some patients with high-grade obstruction secondary
to malignant disease that is not amendable to resection will benefit
from bypass procedures.
ADULT INTUSSUSCEPTION AND GALLSTONE ILEUS
Primary resection is prudent. Careful manual reduction of any
involved bowel may limit the amount of intestine that needs to be
removed. A proximal ostomy may be required if unprepped colon
is involved. The most common site of intestinal obstruction in
patients with gallstone “ileus” is the ileum (60% of patients). The
gallstone enters the intestinal tract most often via a cholecysto­
duodenal fistula. It can usually be removed by operative enteroli­
thotomy. Addressing gallbladder disease during urgent or emergent
surgery is not recommended.
CHAPTER 341
POSTOPERATIVE BOWEL OBSTRUCTION
Early postoperative mechanical bowel obstruction is that which
occurs within the first 6 weeks of operation. Although it tends
to respond and behave differently from classic mechanical bowel
obstruction and may be very difficult to distinguish from post­
operative ileus, most are partial and can be expected to resolve
spontaneously. A higher index of suspicion for a definitive site of
obstruction is warranted for patients who undergo laparoscopic
surgical procedures. Patients who first had ileus and then subse­
quently develop obstructive symptoms after an initial return of
normal bowel function are more likely to have true postoperative
small-bowel obstruction. The longer it takes for a patient’s obstruc­
tive symptoms to resolve after hospitalization, the more likely the
patient is to require surgical intervention.
Acute Intestinal Obstruction
Acknowledgment
The wisdom and expertise of Dr. William Silen are gratefully
acknowledged.
■
■FURTHER READING
Catena F et al: Adhesive small bowel adhesions obstruction: Evolu­
tions in diagnosis, management and prevention. World J Gastrointest
Surg 27:222, 2016.
Ferrada P et al: Surgery or stenting for colonic obstruction: A practice
management guideline from the Eastern Association for the Surgery
of Trauma. J Trauma Acute Care Surg 80:659, 2016.
Griffiths S, Glancy DG: Intestinal obstruction. Surgery (Oxford)
41:47, 2023.
Jaffe T, Thompson WM: Large-bowel obstruction in the adults: Clas­
sic radiographic and CT findings, etiology and mimics. Radiology
275:651, 2015.
Long B et al: Emergency medicine evaluation and management of
small bowel obstruction: Evidence-based recommendations. J Emerg
Med 56:166, 2019.
Paulson EK, Thompson WM: Review of small-bowel obstruction:
The diagnosis and when to worry. Radiology 275:332, 2015.
Perry H et al: Relative accuracy of emergency CT in adults with nontraumatic abdominal pain. Brit Inst Rad 89:20150416, 2016.
Taylor MR, Lalani N: Adult small bowel obstruction. Acad Emerg
Med 20:528, 2013.

12 - 342 Acute Appendicitis and Peritonitis
342 Acute Appendicitis and Peritonitis
Danny O. Jacobs
Acute Appendicitis and
Peritonitis
ACUTE APPENDICITIS
■
■INCIDENCE AND EPIDEMIOLOGY
Acute appendicitis is the most common acute general surgery emer­
gency affecting the abdomen, with a rate of ~10–11 cases per 10,000
people annually over the past several decades affecting biological males
and females equivalently. Appendicitis still occurs most commonly in
10- to 19-year-olds, although the worldwide incidence and average
age at diagnosis appear to be increasing gradually, whereas disability
from complications and mortality appear to be decreasing. Significant
differences in geographic risk variation are noted in the United States
and worldwide.
The risk of perforation is ~10–20% and is a significant cause of
excess morbidity for all patients, but the risk appears to be signifi­
cantly higher in patients <5 or >65 years of age. Patients with higher
socioeconomic status appear to have lower risk of appendicitis with or
without perforation. It is also important to note that the incidence of
appendiceal tumors also appears to be increasing.
How best to manage patients who present with an appendiceal
mass and small localized abscess without perforation is controversial,
especially whether interval or immediate appendectomy should be per­
formed when both can be appropriate depending on the clinical details.
PART 10
Disorders of the Gastrointestinal System
■
■PATHOGENESIS OF APPENDICITIS AND
APPENDICEAL PERFORATION
Appendicitis was first described in 1886 by Reginald Fitz. Its etiology is
still not completely understood. Fecaliths, incompletely digested food
residue, lymphoid hyperplasia, intraluminal scarring, tumors, bacteria,
viruses, and inflammatory bowel disease have all been associated with
inflammation of the appendix with potentially different outcomes
depending on pathogenesis.
Although not proven, obstruction of the appendiceal lumen is
believed to be an important step in the development of appendicitis—
at least in some cases. Here, obstruction leads to bacterial overgrowth
and luminal distension, with an increase in intraluminal pressure that
can inhibit the flow of lymph and blood. Then, vascular thrombosis
and ischemic necrosis with perforation of the distal appendix may
occur. For this reason, perforation that occurs near the base of the
appendix should raise concerns about another disease process. Most
patients who will perforate do so before they are evaluated by surgeons.
Appendiceal fecaliths (or appendicoliths) are found in ~50% of
patients with gangrenous appendicitis who perforate but are rarely
identified in those who have simpler disease. This suggests that the
underlying pathophysiologic processes are different. Uncomplicated
appendicitis (e.g., without gangrenous necrosis, appendicoliths, perfo­
ration, or tumors) does not always progress to perforation. It appears
that at least some cases of simple acute appendicitis may temporarily
resolve or be managed with antibiotic therapy, although at least onethird will require subsequent appendectomy. Nevertheless, appen­
dicitis does not invariably progress to perforation, and the use of
nonoperative therapies to treat uncomplicated appendicitis continues
to be studied intensively.
These findings highlight the importance of clinical decision-making
and risk assessment when deciding and discussing treatment options
with patients who presumably have simple disease, for example, decid­
ing who is an appropriate candidate for nonoperative management
and who is not without bias. The latter is especially pertinent given the
difficulty in assessing which patients might progress to perforation and
which will not.
When perforation occurs, the resultant leak may be contained by
the omentum or other surrounding tissues to form an abscess. Free
TABLE 342-1  Some Conditions That Mimic Appendicitis
Crohn’s disease
Cholecystitis or other gallbladder disease
COVID-19 infection occasionally co-incident
Diverticulitis
Ectopic pregnancy
Endometriosis
Gastroenteritis or colitis
Gastric or duodenal ulceration
Hepatitis
Kidney disease, including nephrolithiasis
Liver abscess
Meckel’s diverticulitis
Mittelschmerz
Mesenteric adenitis
Omental torsion
Pancreatitis
Lower lobe pneumonia
Pelvic inflammatory disease
Ruptured ovarian cyst or other
cystic disease of the ovaries
Small-bowel obstruction
Urinary tract infection
perforation normally causes severe peritonitis. These patients may also
develop infective suppurative thrombosis of the portal vein and its
tributaries along with intrahepatic abscesses. The prognosis of the very
unfortunate patients who develop this rare but dreaded complication
is very poor.
■
■CLINICAL MANIFESTATIONS
Improved diagnosis, supportive care, and surgical interventions are
likely responsible for the remarkable decrease in the risk of mortal­
ity from simple appendicitis to currently <1%. Nevertheless, it is still
important to identify patients who might have appendicitis as early
as possible. Patients who have persistent symptoms that have not
improved over 48 h may be more likely to perforate or develop other
complications.
Appendicitis should be included in the differential diagnosis of
abdominal pain for every patient in any age group unless it is certain
that the organ has been previously removed (Table 342-1).
The appendix’s anatomic location, which varies, may directly influ­
ence how the patient presents. Where the appendix can be “found”
ranges from local differences in how the appendiceal body and tip
lie relative to its attachment to the cecum (Figs. 342-1 and 342-2), to
where the appendix is actually situated in the peritoneal cavity—for
example, from its typical location in the right lower quadrant, to the
pelvis, right flank, right upper quadrant (as may be observed during
pregnancy), or even the left side of the abdomen for patients with mal­
rotation or who have severely redundant colons.
Because the differential diagnosis of appendicitis is so extensive,
deciding if a patient has appendicitis can be difficult (Table 342-2).
Many patients may not present with the classically described history or
physical findings, and some may not have any abdominal discomfort
early in the disease process. Soliciting an appropriate history requires
detecting and evaluating symptoms that might suggest alternative
diagnoses.
0.5%
64%
1%
32%
2%
FIGURE 342-1  Regional anatomic variations of the appendix.
FIGURE 342-2  Locations of the appendix and cecum.
What is the classic history? Nonspecific complaints occur first.
Patients may notice changes in bowel habits or malaise and vague,
perhaps intermittent, crampy abdominal pain in the epigastric or peri­
umbilical region. The pain subsequently migrates to the right lower
quadrant over 12–24 h, where it is sharper and can be definitively local­
ized as transmural inflammation when the appendix irritates the pari­
etal peritoneum. Parietal peritoneal irritation may be associated with
local muscle rigidity and stiffness. Patients with appendicitis will most
often observe that their nausea, if present, followed the development of
abdominal pain, which can help distinguish them from patients with
gastroenteritis, for example, in whom nausea occurs first. Emesis, if
present, also occurs after the onset of pain and is typically mild and
scant. Thus, timing of the onset of symptoms and the characteristics
of the patient’s pain and any associated findings must be rigorously
assessed. Anorexia is so common that the diagnosis of appendicitis
should be questioned in its absence.
Arriving at the correct diagnosis is even more challenging when
the appendix is not located in the right lower quadrant, in women of
childbearing age, and in the very young or elderly. Because the differ­
ential diagnosis of appendicitis is so broad, often the key question to
answer expeditiously is whether the patient has appendicitis or some
other condition that requires immediate operative intervention. A
major concern is that the likelihood of a delay in diagnosis is greater
TABLE 342-2  Relative Frequency of Common Presenting Symptoms
SYMPTOMS
FREQUENCY
Abdominal pain
95%
Anorexia
70%
Constipation
4–16%
Diarrhea
4–16%
Fever
10–20%
Migration of pain to right lower
quadrant
50–60%
Nausea
65%
Vomiting
50–75%
TABLE 342-3  Relative Frequency of Some Presenting Signs
SIGNS
FREQUENCY
Abdominal tenderness
95%
Right lower quadrant tenderness
90%
Rebound tenderness
30–70%
Rectal tenderness
30–40%
Cervical motion tenderness
30%
Rigidity
~10%
Psoas sign
3–5%
Obturator sign
5–10%
Rovsing’s sign
5%
Palpable mass
<5%
if the appendix is unusually positioned. All patients should undergo a
rectal examination. An inflamed appendix located behind the cecum
or below the pelvic brim may prompt very little tenderness of the ante­
rior abdominal wall.
Patients with pelvic appendicitis are more likely to present with
dysuria, urinary frequency, diarrhea, or tenesmus. They may only
experience pain in the suprapubic region on palpation or on rectal
or pelvic examination. A pelvic examination in women is mandatory
to rule out conditions affecting urogynecologic organs that can cause
abdominal pain and mimic appendicitis such as pelvic inflammatory
disease, ectopic pregnancy, and ovarian torsion. None of the currently
available decision tools yet appear to be able to circumvent or obviate
the need for expert clinical opinion. The relative frequencies of some
presenting signs are displayed in Table 342-3.
CHAPTER 342
Patients with simple appendicitis normally only appear mildly ill
with a pulse and temperature that are usually only slightly above nor­
mal. The provider should be concerned about other disease processes
beside appendicitis or the presence of complications such as perfora­
tion, phlegmon, or abscess formation if the temperature is >38.3°C
(~101°F) and if there are rigors.
Acute Appendicitis and Peritonitis
Patients with appendicitis will be found to lie quite still to avoid
peritoneal irritation caused by movement, and some will report dis­
comfort caused by a bumpy car ride on the way to the hospital or clinic,
coughing, sneezing, or other actions that replicate a Valsalva maneuver.
The entire abdomen should be examined systematically starting in an
area where the patient does not report discomfort if possible. Clas­
sically, maximal tenderness is identified where the appendix is most
often located—in the right lower quadrant at or near McBurney’s point,
which is approximately one-third of the way along a line originating at
the anterior iliac spine and running to the umbilicus. Gentle pressure
in the left lower quadrant may elicit pain in the right lower quadrant
if the appendix is located there. This is Rovsing’s sign (Table 342-4).
Evidence of parietal peritoneal irritation is often best elicited by gentle
abdominal percussion, jiggling the patient’s gurney or bed, or mildly
bumping the feet.
Atypical presentation and pain patterns are common, especially in
the very old or the very young. Diagnosing appendicitis in children can
be especially challenging because they tend to respond so dramatically
to stimulation and obtaining an accurate history may be difficult. In
addition, it is important to remember that the smaller omentum found
in children may be less likely to wall off an appendiceal perforation.
Observing the child in a quiet surrounding may be helpful.
TABLE 342-4  Classic Signs of Appendicitis in Patients with Abdominal
Pain
MANEUVER
FINDINGS
Rovsing’s sign
Palpating in the left lower quadrant causes pain in the right
lower quadrant
Obturator sign
Internal rotation of the hip causes pain, suggesting the
possibility of an inflamed appendix located in the pelvis
Iliopsoas sign
Extending the right hip causes pain along posterolateral
back and hip, suggesting retrocecal appendicitis
Signs and symptoms of appendicitis can be subtle in the elderly who
may not react as vigorously to appendicitis as younger people. Pain, if
noticed, may be minimal and have originated in the right lower quad­
rant or, otherwise, where the appendix is located. It may never have
been noticed to be intermittent, or there may only be significant dis­
comfort with deep palpation. Nausea, anorexia, and emesis may be the
predominant complaints. The rare patient may even present with signs
and symptoms of distal bowel obstruction secondary to appendiceal
inflammation and phlegmon or abscess formation.
■
■LABORATORY TESTING
Laboratory testing does not identify patients with appendicitis. The
white blood cell count is only mildly to moderately elevated in ~70%
of patients with simple appendicitis (with a leukocytosis of 10,000–
18,000 cells/μL). A “left shift” toward immature polymorphonuclear
leukocytes is present in >95% of cases. A sickle cell preparation may
be prudent to obtain in those of African, Spanish, Mediterranean, or
Indian ancestry. Serum amylase and lipase levels should be measured.
Urinalysis is indicated to help exclude genitourinary conditions that
may mimic acute appendicitis, but a few red or white blood cells may
be present as a nonspecific finding. An inflamed appendix that abuts
the ureter or bladder may cause sterile pyuria or hematuria. Every
woman of childbearing age should have a pregnancy test. Cervical cul­
tures are indicated if pelvic inflammatory disease is suspected. Anemia
and guaiac-positive stools should raise concern about the presence of
other diseases or complications such as cancer.
■
■IMAGING
Plain films of the abdomen are rarely helpful and so are not routinely
obtained unless the clinician is worried about other conditions such as
intestinal obstruction, perforated viscus, or ureterolithiasis. Less than
5% of patients will present with an opaque fecalith in the right lower
quadrant. The presence of a fecalith is not diagnostic of appendicitis,
although its presence in an appropriate location where the patient com­
plains of pain is suggestive and is associated with a greater likelihood
of complications.
PART 10
Disorders of the Gastrointestinal System
The effectiveness of ultrasonography as a tool to diagnosis appen­
dicitis is highly operator dependent. Even in very skilled hands, the
appendix may not be visualized. Its overall sensitivity is ~0.86, with a
specificity of 0.81. Ultrasonography, especially intravaginal techniques,
appears to be most useful for identifying pelvic pathology in women.
Ultrasonographic findings suggesting the presence of appendicitis
include wall thickening, an increased appendiceal diameter, and the
presence of free fluid. Current practice in some institutions is to first
perform ultrasonography and progress to other imaging studies only if
the findings are equivocal or complications are suspected.
The sensitivity and specificity of computed tomography (CT) are
at least 0.94 and 0.95, respectively. Thus, CT imaging, given its high
negative predictive value, especially with the safe use of oral and intra­
venous contrast, may be helpful if the diagnosis is in doubt, although
studies performed early in the course of disease may not have any typi­
cal radiographic findings. Overall, in patients in whom the diagnosis
is uncertain, delaying operation at the time of presentation to obtain
CT does not appear to increase the risk of perforation. CT scanning
is a superior method for assessing the severity of acute appendicitis in
the absence of peritoneal findings indicative of perforation, abscess, or
suspicion of an associated malignancy.
Suggestive findings on CT examination include dilatation >6 mm
with wall thickening, a lumen that does not fill with enteric contrast,
and fatty tissue stranding or air surrounding the appendix, which sug­
gests inflammation (Figs. 342-3 and 342-4). The presence of luminal
air or contrast is not consistent with a diagnosis of appendicitis. Fur­
thermore, nonvisualization of the appendix is a nonspecific finding
that should not be used to rule out the presence of appendiceal or
periappendiceal inflammation.
■
■SPECIAL PATIENT POPULATIONS
Appendicitis is the most common extrauterine general surgical emer­
gency observed during pregnancy. Early symptoms of appendicitis such
as nausea and anorexia may be overlooked. Diagnosing appendicitis
FIGURE 342-3  Computed tomography with oral and intravenous contrast of acute
appendicitis. There is thickening of the wall of the appendix and periappendiceal
stranding (arrow).
in pregnant patients may be especially difficult because as the uterus
enlarges the appendix may be pushed higher along the right flank even to
the right upper quadrant or because the gravid uterus may obscure typi­
cal physical findings. Ultrasonography may facilitate early diagnosis. A
high index of suspicion is required because of the effects of unrecognized
and untreated appendicitis on the fetus. For example, the fetal mortality
rate is four times greater (from 5 to 20%) in patients with perforation.
Immunocompromised patients may present with only mild tender­
ness and may have many other disease processes in their differential
diagnosis, including atypical infections from mycobacteria, Cytomega­
lovirus, or other fungi. Enterocolitis is a concern and may be present
in patients who present with abdominal pain, fever, and neutropenia
due to chemotherapy. CT imaging may be very helpful, although it is
important not to be overly cautious and delay operative intervention
for those patients who are believed to have appendicitis.
TREATMENT
Acute Appendicitis
In the absence of contraindications, most patients who have
strongly suggestive medical histories and physical examinations
FIGURE 342-4  Appendiceal fecalith (arrow).
with supportive laboratory findings are candidates for appendec­
tomy. Certainly, in some instances, imaging studies are not manda­
tory but are often obtained before surgical consultation is requested.
Imaging and close observation are appropriate in patients whose
evaluations are suggestive but not convincing.
Of course, CT may accurately indicate the other intraabdomi­
nal processes that warrant intervention. Whenever the diagnosis
is uncertain, it is prudent to observe the patient and repeat the
abdominal examination over 6–8 h. Any evidence of progression is
an indication for operation. Narcotics can be given to patients with
severe discomfort after an initial, thorough examination.
All patients should be fully prepared for surgery and have any
fluid and electrolyte abnormalities corrected. Either laparoscopic
or open appendectomy is a satisfactory choice for patients with
uncomplicated appendicitis, although most procedures are now
performed in a minimally invasive fashion to the patient’s benefit
in terms of recovery time and fewer overall potential complications.
Endoscopic treatment for patients with uncomplicated appendicitis
is being evaluated for efficacy.
Management of those who present with a mass representing a
phlegmon or abscess can be more difficult. Such patients are most
commonly treated with broad-spectrum antibiotics, percutaneous
drainage especially if an abscess is noted >3 cm in diameter, paren­
teral fluids, and bowel rest. If they appear to respond to conserva­
tive management, the appendix can then be more safely removed
6–12 weeks later when inflammation has diminished.
A laparoscopic approach may also be useful when the exact
diagnosis is uncertain. A laparoscopic approach may also facilitate
exposure in those who are very obese. Absent complications, most
patients can be discharged within 24–40 h of operation. The most
common postoperative complications are fever and leukocytosis.
Continuation of these findings beyond 5 days should raise concern
for the presence of an intraabdominal abscess. The mortality rate
for uncomplicated, nonperforated appendicitis is 0.1–0.5%, which
approximates the overall risk of general anesthesia. The mortal­
ity rate for perforated appendicitis or other complicated disease
is much higher, ranging from 3% overall to as high as 15% in the
elderly.
ACUTE PERITONITIS
Acute peritonitis, or inflammation of the visceral and parietal perito­
neum, is most often but not always infectious in origin, resulting from
perforation of a hollow viscus. This is called secondary peritonitis, as
opposed to primary or spontaneous peritonitis, when a specific intraab­
dominal source cannot be identified. In either instance, the inflamma­
tion can be localized or diffuse.
■
■ETIOLOGY
Infective organisms may contaminate the peritoneal cavity after
spillage from a hollow viscus, because of a penetrating wound of the
abdominal wall, or because of the introduction of a foreign object like
a peritoneal dialysis catheter or port that becomes infected. Secondary
peritonitis most commonly results from perforation of the appendix,
colonic diverticula, or the stomach and duodenum. It may also occur as
a complication of bowel infarction or incarceration, cancer, inflamma­
tory bowel disease, and intestinal obstruction or volvulus. Conditions
that may cause secondary bacterial peritonitis and their mechanisms
are listed in Table 342-5. Over 90% of the cases of primary or sponta­
neous bacterial peritonitis occur in patients with ascites or hypopro­
teinemia (<1 g/L).
Aseptic peritonitis is most commonly caused by the abnormal
presence of physiologic fluids such as gastric juice, bile, pancreatic
enzymes, blood, or urine. It can also be caused by the effects of nor­
mally sterile foreign bodies such as surgical sponges or instruments.
More rarely, it occurs as a complication of systemic diseases such as
lupus erythematosus, porphyria, and familial Mediterranean fever. The
chemical irritation caused by stomach acid and activated pancreatic
enzymes is extreme, and secondary bacterial infection may occur.
TABLE 342-5  Conditions Leading to Secondary Bacterial Peritonitis
Bowel perforation
  Appendicitis
  Anastomotic leakage
  Adhesion
  Diverticulitis
  Iatrogenic (including endoscopic
Perforation or leakage of other organs
  Biliary leakage (e.g., after liver
biopsy)
  Cholecystitis
  Intraperitoneal bleeding
  Pancreatitis
  Salpingitis
  Urinary bladder
Loss of peritoneal integrity
  Intraperitoneal chemotherapy
  Iatrogenic (e.g., postoperative
perforation)
  Ingested foreign body
  Inflammation
  Intussusception
  Neoplasms
  Obstruction
  Peptic ulcer disease
  Strangulated hernia
  Vascular (including ischemia or
foreign body)
  Perinephric abscess
  Peritoneal dialysis or other
indwelling devices
  Trauma
embolus)
  Trauma (blunt or penetrating)
■
■CLINICAL FEATURES
The cardinal signs and symptoms of peritonitis are acute, typically
severe, abdominal pain with tenderness and fever. How patients’ com­
plaints of pain are manifested depends on their overall physical health
and whether the inflammation is diffuse or localized. Elderly and
immunosuppressed patients may not respond as aggressively to the
irritation. Diffuse, generalized peritonitis is most often recognized as
diffuse abdominal tenderness with guarding, rigidity, and other evidence
of parietal peritoneal irritation. Physical findings may only be identified
in a specific region of the abdomen if the intraperitoneal inflammatory
process is limited or otherwise contained as may occur in patients with
uncomplicated appendicitis or diverticulitis. Bowel sounds are usually
absent to hypoactive but are not reliable as a physical finding.
CHAPTER 342
Acute Appendicitis and Peritonitis
Most patients present with tachycardia and signs of volume deple­
tion with hypotension. Laboratory testing typically reveals a significant
leukocytosis, and patients may be severely acidotic. Radiographic stud­
ies may show dilatation of the bowel and associated bowel wall edema.
Free air or other evidence of leakage requires attention and could
represent a surgical emergency. In stable patients in whom ascites is
present, diagnostic paracentesis is indicated, where the fluid is tested
for protein and lactate dehydrogenase and the cell count is measured.
■
■THERAPY AND PROGNOSIS
Whereas mortality rates can be <10% for reasonably healthy patients
with relatively uncomplicated, localized peritonitis, mortality rates
40% have been reported for the elderly or immunocompromised.
Successful treatment depends on correcting any electrolyte abnormali­
ties, restoration of fluid volume and stabilization of the cardiovascular
system, appropriate antibiotic therapy, and surgical correction of any
underlying abnormalities.
Acknowledgment
The wisdom and expertise of Dr. William Silen are gratefully acknowl­
edged in this updated chapter on acute appendicitis and peritonitis.
■
■FURTHER READING
Andersson RE: Short-term complications and long-term morbidity
of laparoscopic and open appendicectomy in a national cohort. Br J
Surg 101:1135, 2014.
Buckius MT et al: Changing epidemiology of acute appendicitis in
the United States: Study period 1993–2008. J Surg Res 175:185, 2012.
CODA Collaborative: A randomized trial comparing antibiotics
with appendectomy for appendicitis. N Engl J Med 383:1907, 2020.
Di Saverio S et al: Diagnosis and treatment of acute appendicitis:
2020 update of the WSES Jerusalem guidelines. World J Emerg Surg
15:27, 2020.
Drake FT et al: Time to appendectomy and risk of perforation in acute
appendicitis. JAMA Surg 149:837, 2014.

13 - SECTION 2 Nutrition
SECTION 2 Nutrition
Flum DR: Acute appendicitis—appendectomy of the “antibiotics first”
strategy. N Engl J Med 372:1937, 2015.
Khan S et al: Endoscopic retrograde appendicitis therapy: Is it really a
need of the hour. Ann Surg 277:e1, 2023.
Moris D et al: Diagnosis and management of acute appendicitis adults.
JAMA 326:2299, 2021.
Ohle R et al: The Alvarado score for predicting acute appendicitis: A
systematic review. BMC Med 9:139, 2011.
Talan DA, DiSaverio S: Treatment of acute uncomplicated appendi­
citis. N Engl J Med 385:1116, 2021.
Vons C et al: Amoxicillin plus clavulanic acid versus appendicectomy
for treatment of acute uncomplicated appendicitis: An open-label,
non-inferiority, randomised controlled trial. Lancet 377:1573, 2011.
Section 2	 Nutrition
Johanna T. Dwyer
Nutrient Requirements
and Dietary Assessment
PART 10
Disorders of the Gastrointestinal System
Nutrients are substances that are essential but not synthesized in suf­
ficient amounts in the body and therefore must be supplied by the
diet. Nutrient requirements for groups of healthy persons have been
determined experimentally. The absence of essential nutrients leads
to growth impairment, organ and metabolic dysfunction, and failure
to maintain nitrogen balance or adequate status of protein and other
nutrients. For good health, we require energy-providing nutrients (pro­
tein, fat, and carbohydrate), vitamins, minerals, and water. Require­
ments for organic nutrients include 9 essential amino acids, several
fatty acids, glucose, 4 fat-soluble vitamins, 10 water-soluble vitamins,
dietary fiber, and choline. The inorganic nutrients include 4 minerals,
7 trace minerals, 3 electrolytes, and the ultra trace elements that must
also be supplied by diet. Individuals of different ages and physiologic
states differ in the amounts of nutrients they require.
Conditionally essential nutrients are not required in the diet but
must be supplied to certain individuals such as those with genetic
defects; pathologies such as infection, disease, or trauma with nutri­
tional implications; and some developmentally immature infants
who need them because they do not synthesize inositol, taurine,
arginine, and/or glutamine in adequate amounts. Many other organic
and inorganic bioactive compounds that are present in foods and
dietary supplements may also have health effects, including pesticides,
heavy metals like lead, phytochemicals, zoochemicals, and microbial
products.
■
■ESSENTIAL NUTRIENT REQUIREMENTS
Energy 
The estimated energy requirement (EER) is the predicted
average requirement to maintain energy balance and health in an adult
of a defined age, sex, weight, height, level of physical activity, and life
stage. For weight to remain stable, energy intake must match total
energy expenditure (TEE). The major components of TEE are resting
energy expenditure (REE) and the physical activity level (PAL), and
minor components include the energy cost of metabolizing food (ther­
mic effect of food, or specific dynamic action) and cold-induced (shiv­
ering) thermogenesis. The average energy intake is ~2600 kcal/d for
American men and ~1800 kcal/d for American women, but individuals
vary with body size and activity level. Although estimates introduce
considerable error, energy intakes are usually calculated from formulas
rather than measuring TEE directly. In individuals whose weights are
stable, for males, REE = 900 + 10m, and for females, REE = 700 + 7m,
where m is mass in kilograms. The calculated REE is then multiplied
by the appropriate PAL to account for physical activity that ranges from
1.4 (inactive) for individuals who perform only essential activities of
daily living (e.g., 30 minutes walking and 90 minutes light to moderate
activity) to 1.6 (low active), 1.75 (active), and 2.05 (very active). The
result is the estimated energy (EER) or intake. When describing their
PALs, sedentary, inactive, and low active individuals tend to overes­
timate, and so in practice, use of PALs of 1.2 to 1.4 may be closer to
their actual activity levels. The EER provides a rough target for plan­
ning caloric needs for a person of a certain age, sex, weight, height,
and physical activity level who is neither gaining nor losing weight.
For further discussion of energy balance in health and disease, see
Chap. 345.
Protein 
Dietary protein consists of both “essential” (e.g., not
synthesized endogenously and must be supplied by diet) and “nones­
sential” (e.g., synthesized endogenously or obtained from diet) amino
acids that are all required for protein synthesis. The nine essential
amino acids are histidine, isoleucine, leucine, lysine, methionine/
cystine, phenylalanine/tyrosine, threonine, tryptophan, and valine.
Several amino acids, such as alanine, arginine, aspartic acid, glutamic
acid, glutamine, and glycine can also be converted to glucose and used
for energy and gluconeogenesis. When energy intake is inadequate,
protein intake must be increased, because ingested amino acids are
diverted into pathways of glucose synthesis and oxidation. In extreme
energy deprivation, protein-calorie malnutrition may ensue (Chap. 345).
For adults, the recommended dietary allowance (RDA) for protein is
~0.8 g/kg desirable body mass per day, assuming that energy needs are
met and that the protein is of relatively high biological value. Current
recommendations for a healthy diet call for at least 10–14% of calories
from protein. Most American diets provide at least those amounts. Bio­
logic value tends to be highest for animal proteins, followed by proteins
from legumes (beans), cereals (rice, wheat, corn), and roots. Combina­
tions of plant proteins that complement one another in their essential
amino acid profiles or combinations of animal and plant proteins can
increase biologic value and lower total protein intakes necessary to
meet requirements. In healthy people with adequate diets, the timing
of protein intake over the course of the day has little effect.
Protein needs increase during growth, pregnancy, lactation, and
rehabilitation after injury or undernutrition. Tolerance to dietary pro­
tein is decreased in renal insufficiency (with consequent uremia) and
in liver failure. Usual protein intakes can precipitate encephalopathy in
patients with cirrhosis of the liver.
Fat and Carbohydrate 
Carbohydrate (4 kcal/g), fat (9 kcal/g), and
protein (4 kcal/g) all provide energy. So does alcohol (ethanol) (7 kcal/g),
but it is not a nutrient. Fats are a concentrated source of energy and
constitute, on average, 34% of calories in U.S. diets. However, for
optimal health, fat intake should total no more than 30% of calories.
Saturated fat and trans-fat should be limited to <10% of calories and
polyunsaturated fats to <10% of calories, with monounsaturated fats
accounting for the remainder of fat intake. At least 45–55% of total
calories should be derived from carbohydrates, with <10% and prefer­
ably <6% from added sugars. The brain requires ~100 g of glucose per
day for fuel; other tissues use ~50 g/d. Some tissues (e.g., brain and
red blood cells) rely on glucose supplied either exogenously or from
muscle proteolysis. Over time, during hypocaloric states, adaptations
that lower carbohydrate needs are possible.
Water 
For adults, 1–1.5 mL of water per kilocalorie of energy
expenditure is sufficient under usual conditions to allow for normal
variations in physical activity, sweating, and the diet’s solute load.
Water losses include 50–100 mL/d in the feces; 500–1000 mL/d
by evaporation or exhalation; and, depending on the renal solute
load, ≥1000 mL/d in the urine. If external losses increase, intakes must
increase accordingly to avoid underhydration. Fever increases water
losses by ~200 mL/d per °C; diarrheal losses vary but may be as great
as 5 L/d in severe diarrhea. Heavy sweating, vigorous exercise, and

14 - 343 Nutrient Requirements and Dietary Assessment
343 Nutrient Requirements and Dietary Assessment
Flum DR: Acute appendicitis—appendectomy of the “antibiotics first”
strategy. N Engl J Med 372:1937, 2015.
Khan S et al: Endoscopic retrograde appendicitis therapy: Is it really a
need of the hour. Ann Surg 277:e1, 2023.
Moris D et al: Diagnosis and management of acute appendicitis adults.
JAMA 326:2299, 2021.
Ohle R et al: The Alvarado score for predicting acute appendicitis: A
systematic review. BMC Med 9:139, 2011.
Talan DA, DiSaverio S: Treatment of acute uncomplicated appendi­
citis. N Engl J Med 385:1116, 2021.
Vons C et al: Amoxicillin plus clavulanic acid versus appendicectomy
for treatment of acute uncomplicated appendicitis: An open-label,
non-inferiority, randomised controlled trial. Lancet 377:1573, 2011.
Section 2	 Nutrition
Johanna T. Dwyer
Nutrient Requirements
and Dietary Assessment
PART 10
Disorders of the Gastrointestinal System
Nutrients are substances that are essential but not synthesized in suf­
ficient amounts in the body and therefore must be supplied by the
diet. Nutrient requirements for groups of healthy persons have been
determined experimentally. The absence of essential nutrients leads
to growth impairment, organ and metabolic dysfunction, and failure
to maintain nitrogen balance or adequate status of protein and other
nutrients. For good health, we require energy-providing nutrients (pro­
tein, fat, and carbohydrate), vitamins, minerals, and water. Require­
ments for organic nutrients include 9 essential amino acids, several
fatty acids, glucose, 4 fat-soluble vitamins, 10 water-soluble vitamins,
dietary fiber, and choline. The inorganic nutrients include 4 minerals,
7 trace minerals, 3 electrolytes, and the ultra trace elements that must
also be supplied by diet. Individuals of different ages and physiologic
states differ in the amounts of nutrients they require.
Conditionally essential nutrients are not required in the diet but
must be supplied to certain individuals such as those with genetic
defects; pathologies such as infection, disease, or trauma with nutri­
tional implications; and some developmentally immature infants
who need them because they do not synthesize inositol, taurine,
arginine, and/or glutamine in adequate amounts. Many other organic
and inorganic bioactive compounds that are present in foods and
dietary supplements may also have health effects, including pesticides,
heavy metals like lead, phytochemicals, zoochemicals, and microbial
products.
■
■ESSENTIAL NUTRIENT REQUIREMENTS
Energy 
The estimated energy requirement (EER) is the predicted
average requirement to maintain energy balance and health in an adult
of a defined age, sex, weight, height, level of physical activity, and life
stage. For weight to remain stable, energy intake must match total
energy expenditure (TEE). The major components of TEE are resting
energy expenditure (REE) and the physical activity level (PAL), and
minor components include the energy cost of metabolizing food (ther­
mic effect of food, or specific dynamic action) and cold-induced (shiv­
ering) thermogenesis. The average energy intake is ~2600 kcal/d for
American men and ~1800 kcal/d for American women, but individuals
vary with body size and activity level. Although estimates introduce
considerable error, energy intakes are usually calculated from formulas
rather than measuring TEE directly. In individuals whose weights are
stable, for males, REE = 900 + 10m, and for females, REE = 700 + 7m,
where m is mass in kilograms. The calculated REE is then multiplied
by the appropriate PAL to account for physical activity that ranges from
1.4 (inactive) for individuals who perform only essential activities of
daily living (e.g., 30 minutes walking and 90 minutes light to moderate
activity) to 1.6 (low active), 1.75 (active), and 2.05 (very active). The
result is the estimated energy (EER) or intake. When describing their
PALs, sedentary, inactive, and low active individuals tend to overes­
timate, and so in practice, use of PALs of 1.2 to 1.4 may be closer to
their actual activity levels. The EER provides a rough target for plan­
ning caloric needs for a person of a certain age, sex, weight, height,
and physical activity level who is neither gaining nor losing weight.
For further discussion of energy balance in health and disease, see
Chap. 345.
Protein 
Dietary protein consists of both “essential” (e.g., not
synthesized endogenously and must be supplied by diet) and “nones­
sential” (e.g., synthesized endogenously or obtained from diet) amino
acids that are all required for protein synthesis. The nine essential
amino acids are histidine, isoleucine, leucine, lysine, methionine/
cystine, phenylalanine/tyrosine, threonine, tryptophan, and valine.
Several amino acids, such as alanine, arginine, aspartic acid, glutamic
acid, glutamine, and glycine can also be converted to glucose and used
for energy and gluconeogenesis. When energy intake is inadequate,
protein intake must be increased, because ingested amino acids are
diverted into pathways of glucose synthesis and oxidation. In extreme
energy deprivation, protein-calorie malnutrition may ensue (Chap. 345).
For adults, the recommended dietary allowance (RDA) for protein is
~0.8 g/kg desirable body mass per day, assuming that energy needs are
met and that the protein is of relatively high biological value. Current
recommendations for a healthy diet call for at least 10–14% of calories
from protein. Most American diets provide at least those amounts. Bio­
logic value tends to be highest for animal proteins, followed by proteins
from legumes (beans), cereals (rice, wheat, corn), and roots. Combina­
tions of plant proteins that complement one another in their essential
amino acid profiles or combinations of animal and plant proteins can
increase biologic value and lower total protein intakes necessary to
meet requirements. In healthy people with adequate diets, the timing
of protein intake over the course of the day has little effect.
Protein needs increase during growth, pregnancy, lactation, and
rehabilitation after injury or undernutrition. Tolerance to dietary pro­
tein is decreased in renal insufficiency (with consequent uremia) and
in liver failure. Usual protein intakes can precipitate encephalopathy in
patients with cirrhosis of the liver.
Fat and Carbohydrate 
Carbohydrate (4 kcal/g), fat (9 kcal/g), and
protein (4 kcal/g) all provide energy. So does alcohol (ethanol) (7 kcal/g),
but it is not a nutrient. Fats are a concentrated source of energy and
constitute, on average, 34% of calories in U.S. diets. However, for
optimal health, fat intake should total no more than 30% of calories.
Saturated fat and trans-fat should be limited to <10% of calories and
polyunsaturated fats to <10% of calories, with monounsaturated fats
accounting for the remainder of fat intake. At least 45–55% of total
calories should be derived from carbohydrates, with <10% and prefer­
ably <6% from added sugars. The brain requires ~100 g of glucose per
day for fuel; other tissues use ~50 g/d. Some tissues (e.g., brain and
red blood cells) rely on glucose supplied either exogenously or from
muscle proteolysis. Over time, during hypocaloric states, adaptations
that lower carbohydrate needs are possible.
Water 
For adults, 1–1.5 mL of water per kilocalorie of energy
expenditure is sufficient under usual conditions to allow for normal
variations in physical activity, sweating, and the diet’s solute load.
Water losses include 50–100 mL/d in the feces; 500–1000 mL/d
by evaporation or exhalation; and, depending on the renal solute
load, ≥1000 mL/d in the urine. If external losses increase, intakes must
increase accordingly to avoid underhydration. Fever increases water
losses by ~200 mL/d per °C; diarrheal losses vary but may be as great
as 5 L/d in severe diarrhea. Heavy sweating, vigorous exercise, and
vomiting also increase water losses. When renal function is normal and
solute intakes are adequate, the kidneys can adjust to increased water
intake by excreting up to 18 L of excess water per day (Chap. 393). How­
ever, obligatory urine outputs can compromise hydration status when
there is inadequate water intake or when losses increase in disease or
kidney damage.
Infants have high requirements for water because of their large sur­
face area to volume ratios, their inability to communicate their thirst,
and the limited capacity of the immature kidney to handle high renal
solute loads. Increased water needs during pregnancy are ~30 mL/d.
During lactation, milk production increases daily water requirements
so that ~1000 mL of additional water is needed, or 1 mL for each mil­
liliter of milk produced. Special attention must also be paid to the water
needs of the elderly, who have reduced total-body water, blunted thirst
sensation, and frequently use medications such as diuretics.
Other Nutrients 
See Chap. 344 for detailed descriptions of
vitamins and minerals.
■
■DIETARY REFERENCE INTAKES AND RDAS
Fortunately, human life and well-being can be maintained within a
fairly wide range of most nutrient intakes. However, the capacity for
adaptation is not infinite—too much, as well as too little, intake of
a nutrient can have adverse effects itself or decrease health benefits
conferred by another nutrient. Therefore, benchmark recommendations
regarding nutrient intakes have been developed to guide menu develop­
ment and clinical practice. They are referred to in the United States and
Canada as the dietary reference intakes (DRIs) and are set by National
Academy of Medicine’s (formerly Institute of Medicine [IOM]) Food
and Nutrition Board and Health Canada. The recommendations apply
to the general population, including those with or at risk for chronic
diseases, overweight, and obesity. Exceptions may be needed for
individuals with severe comorbidities, diseases, disabilities, or use of
medications known to alter energy or some nutrient requirements. The
DRIs supplanted the recommended dietary allowances (RDAs)—the
single reference values used in the United States until the early 1990s.
DRIs now include an estimated average requirement (EAR) as well as
other reference values used for dietary planning: the RDA, the adequate
intake (AI), the chronic disease risk reduction intake (CDRR), and the
tolerable upper level (UL). The DRIs also include acceptable macronu­
trient distribution ranges (AMDRs) for protein, fat, and carbohydrate.
The types of evidence and criteria used to establish nutrient require­
ments vary by nutrient, age, and physiologic group. The current DRIs
for vitamins and elements are provided in Tables 343-1 and 343-2,
respectively. Table 343-3 provides DRIs for water and macronutrients.
Energy has neither an RDA nor a UL because intakes above or below
requirements lead to weight gain or loss; EERs are used instead and
are discussed in Chap. 345 on energy balance in health and disease.
Estimated Average Requirement (EAR) 
When florid manifes­
tations of the classic dietary-deficiency diseases such as rickets (defi­
ciency of vitamin D and calcium), scurvy (deficiency of vitamin C),
xerophthalmia (deficiency of vitamin A), and protein-calorie malnutri­
tion were common, nutrient adequacy was inferred from the absence
of their clinical deficiency signs. Later, biochemical and other earlier
changes were used that appeared long before the deficiency was clini­
cally apparent. Today, criteria of adequacy are based on biologic mark­
ers when they are available, and to sensitive biochemical, physiologic,
or behavioral tests that reflect early changes in regulatory processes;
maintenance of body stores of nutrients; or, if available, the amount
of a nutrient that minimizes the risk of chronic degenerative disease.
However, relevant biomarkers of disease risk are often not available,
and the long time lags between intake and disease outcomes further
complicate the picture.
The EAR is the amount of a nutrient estimated to be adequate for
half of the healthy individuals of a specific age and sex. It is not an
effective estimate of nutrient adequacy in individuals because it is a
median requirement for a group, with 50% of individuals in a group
falling below and 50% falling above the requirement. Thus, a person
with a usual intake at the EAR has a 50% risk of inadequate intake. For
these reasons, the other standards described below are more useful for
clinical purposes.
Recommended Dietary Allowances 
The RDA is the nutrient
intake goal for planning diets of individuals. It is the average daily
dietary intake level that meets the nutrient requirements of nearly all
healthy persons of a specific sex, age, life stage, or physiologic condi­
tion (e.g., pregnancy or lactation). The RDA is set two standard devia­
tions above the EAR to ensure that the needs of virtually any individual
are met. The risk of dietary inadequacy increases as one’s intake falls
below the RDA. However, the RDA is an overly generous criterion for
evaluating nutrient adequacy because, by definition, the RDA exceeds
the actual requirements of all but ~2–3% of the population. Therefore,
many people whose intake falls below the RDA are still getting enough
of the nutrient.
The RDAs are used to formulate food guides such as the U.S.
Department of Agriculture’s (USDA) MyPlate Plan for individuals
(https://www.choosemyplate.gov/resources/MyPlatePlan), as well as to
create food-exchange lists for therapeutic diet planning. An online tool,
available at https://www.nal.usda.gov/fnic/dri-calculator/, allows health
professionals to calculate individualized daily nutrient recommenda­
tions for dietary planning based on the DRIs.
The RDAs are also used as a standard for labels describing the nutri­
tional content of foods and nutrient-containing dietary supplements.
On food labels, the nutrient content in a food is stated by weight and/
or as a percentage of the daily value (DV), a variant of the RDA used on
the nutrition facts panel that, for an adult, represents the highest RDA
for an adult consuming 2000 kcal.
CHAPTER 343
Adequate Intake (AI) 
Some nutrients lack an EAR and so an
RDA cannot be set. In this circumstance, the AI is used, based on
observed or experimentally determined approximations of nutrient
intakes in healthy people. In the DRIs, AIs rather than RDAs are pro­
posed for nutrients consumed by infants (up to age 1 year) as well as
for chromium, fluoride, manganese, sodium, potassium, pantothenic
acid, biotin, choline, fiber, and water consumed by persons of all ages.
Nutrient Requirements and Dietary Assessment
Tolerable Upper Levels (UL) 
Healthy individuals derive no
established benefit from consuming nutrient levels above the RDA or
AI. In fact, excessive nutrient intake can disturb body functions and
cause acute, progressive, or permanent disabilities. The tolerable UL
is the highest level of chronic (usually daily) nutrient intake that is
unlikely to pose a risk of adverse health effects for most of the popula­
tion. Data on the adverse effects of large amounts of many nutrients
are unavailable or too limited to establish a UL and so the lack of a
UL does not mean that the risk of adverse effects from high intake is
nonexistent. Nutrient levels in commonly eaten foods rarely exceed
the UL, although very highly fortified foods and dietary supplements
provide more concentrated amounts of nutrients per serving and thus
pose a potential risk of toxicity. Dietary supplements have Supplement
Facts panels on labels that express the amount of nutrients present in
absolute units or as the percentage of the DV provided per recom­
mended serving size. Total nutrient intakes, including that from foods,
supplements, and over-the-counter medications (e.g., antacids), should
not exceed RDA levels.
Chronic Disease Risk Reduction Intake (CDRR) 
The CDRR
is the level above which a reduction in intake is expected to lower
chronic disease risk is the CDRR. For example, the sodium CDRR for
adults is 2300 mg/d; this is the lowest level of intake for which there is
sufficiently strong evidence to characterize a CDRR. CDRRs are not
available for potassium or other nutrients, but note that the AI for
potassium was recently reduced to 2500 mg/d.
Acceptable Macronutrient Distribution Ranges (AMDRs) 
AMDRs are estimated ranges from epidemiologic data of energyproviding macronutrient intakes (protein, carbohydrate, and fat)
considered to be healthful. These ranges are 10–35% of calories for
protein, 20–35% of calories for fat, and 45–65% of calories for carbohy­
drate. Alcohol, which also provides energy, is not a nutrient; therefore,
no recommendations for it are provided. The latest DRI committee
(μg/d)
CHOLINE
(mg/d)g
Birth to 6 mo
400*
40*
4*
2.0*
0.2*
0.3*
2*
0.1*
65*
0.4*
1.7*
5*
125*
6–12 mo
500*
50*
5*
2.5*
0.3*
0.4*
4*
0.3*
80*
0.5*
1.8*
6*
150*
1–3 y
30*
0.5
0.5
0.5
0.9
2*
8*
200*
4–8 y
55*
0.6
0.6
0.6
1.2
3*
12*
250*
9–13 y
60*
0.9
0.9
1.0
1.8
4*
20*
375*
14–18 y
75*
1.2
1.3
1.3
2.4
5*
25*
550*
19–30 y
120*
1.2
1.3
1.3
2.4
5*
30*
550*
31–50 y
120*
1.2
1.3
1.3
2.4
5*
30*
550*
51–70 y
120*
1.2
1.3
1.7
2.4h
5*
30*
550*
70 y
120*
1.2
1.3
1.7
2.4h
5*
30*
550*
9–13 y
60*
0.9
0.9
1.0
1.8
4*
20*
375*
14–18 y
75*
1.0
1.0
1.2
400i
2.4
5*
25*
400*
ACID (mg/d)
BIOTIN
(lg/d)
PANTOTHENIC
(lg/d)f
VITAMIN B12
(mg/d)
FOLATE
(mg/d)e
VITAMIN B6
PART 10
Disorders of the Gastrointestinal System
(mg/d)
NIACIN
TABLE 343-1  Dietary Reference Intakes (DRIs): Recommended Dietary Allowances and Adequate Intakes for Vitamins
(mg/d)
RIBOFLAVIN
(lg/d)
THIAMIN
(mg/d)d
VITAMIN K
(lg/d)b,c
VITAMIN E
(mg/d)
VITAMIN D
(lg/d)a
VITAMIN C
GROUP
VITAMIN A
LIFE-STAGE
Children
Females
Infants
Males
aAs retinol activity equivalents (RAEs). 1 RAE = 1 μg retinol, 12 μg β-carotene, 24 μg α-carotene, or 24 μg β-cryptoxanthin. The RAE for dietary provitamin A carotenoids is twofold greater than the retinol equivalent (RE), whereas the RAE for
bound B12, it is advisable for those >50 years of age to meet their RDA mainly by consuming foods fortified with B12 or a supplement containing B12. iIn view of evidence linking inadequate folate intake with neural tube defects in the fetus, it
(SRR-, SSR-, SRS-, and SSS-α-tocopherol) also found in fortified foods and supplements. eAs niacin equivalents (NEs). 1 mg of niacin = 60 mg of tryptophan; 0–6 months = preformed niacin (not NE). fAs dietary folate equivalents
intake level sufficient to meet the nutrient requirements of nearly all healthy individuals (97–98%) in a group. The RDA is calculated from an estimated average requirement (EAR). If sufficient scientific evidence is not available to establish
is recommended that all women capable of becoming pregnant consume 400 μg of folate from supplements or fortified foods in addition to intake of food folate from a varied diet. jIt is assumed that women will continue consuming 400 μg
an EAR and thus to calculate an RDA, an AI is usually developed. For healthy breast-fed infants, an AI is the mean intake. The AI for other life-stage and sex-specific groups is believed to cover the needs of all healthy individuals in those
whether a dietary supply of choline is needed at all stages of the life cycle, and it may be that the choline requirement can be met by endogenous synthesis at some of these stages. hBecause 10–30% of older people may malabsorb foodpreformed vitamin A is the same as the RE. bAs cholecalciferol. 1 μg cholecalciferol = 40 IU vitamin D. cUnder the assumption of minimal sunlight. dAs α-tocopherol. α-Tocopherol includes RRR-α-tocopherol, the only form of α-tocopherol
(DFEs). 1 DFE = 1 μg food folate = 0.6 μg of folic acid from fortified food or as a supplement consumed with food = 0.5 μg of a supplement taken on an empty stomach. gAlthough AIs have been set for choline, there are few data to assess
Note: This table (taken from the DRI reports; see www.nap.edu) presents recommended dietary allowances (RDAs) in bold type and adequate intakes (AIs) in ordinary type followed by an asterisk (*). An RDA is the average daily dietary
that occurs naturally in foods, and the 2R-stereoisomeric forms of α-tocopherol (RRR-, RSR-, RRS-, and RSS-α-tocopherol) that occur in fortified foods and supplements. It does not include the 2S-stereoisomeric forms of a-tocopherol
Source: National Academies of Sciences, Engineering, and Medicine. 2019. Dietary Reference Intakes for Sodium and Potassium. https://doi.org/10.17226/25353. Adapted and reproduced with permission from the National Academy of
19–30 y
90*
1.1
1.1
1.3
400i
2.4
5*
30*
425*
31–50 y
90*
1.1
1.1
1.3
400i
2.4
5*
30*
425*
51–70 y
90*
1.1
1.1
1.5
2.4h
5*
30*
425*
70 y
90*
1.1
1.1
1.5
2.4h
5*
30*
425*
14–18 y
75*
1.4
1.4
1.9
600j
2.6
6*
30*
450*
19–30 y
90*
1.4
1.4
1.9
600j
2.6
6*
30*
450*
31–50 y
90*
1.4
1.4
1.9
600j
2.6
6*
30*
450*
14–18 y
75*
1.4
1.6
2.0
2.8
7*
35*
550*
19–30 y
90*
1.4
1.6
2.0
2.8
7*
35*
550*
31–50 y
90*
1.4
1.6
2.0
2.8
7*
35*
550*
from supplements or fortified food until their pregnancy is confirmed and they enter prenatal care, which ordinarily occurs after the end of the periconceptional period—the critical time for formation of the neural tube.
groups, but lack of data or uncertainty in the data makes it impossible to specify with confidence the percentage of individuals covered by this intake.
Sciences, Courtesy of the National Academies.
Lactating Women
Pregnant Women
(g/d)
CHLORIDE
Birth to 6 mo
200*
0.2*
200*
0.01*
110*
0.27*
30*
0.003*
2*
100*
15*
2*
0.4*
0.12*
0.18*
6–12 mo
260*
5.5*
220*
0.5*
130*
75*
0.6*
3*
275*
20*
0.7*
0.37*
0.57*
(g/d)
9–13 y
21*
2*
1.6*
4.5*
1.5*
2.3*
70 y
30*
4*
2.3*
4.7*
1.2*
1.8*
51–70 y
30*
4*
2.3*
4.7*
1.3*
2.0*
31–50 y
35*
4*
2.3*
4.7*
1.5*
2.3*
19–30 y
35*
4*
2.3*
4.7*
1.5*
2.3*
14–18 y
35*
3*
2.2*
4.7*
1.5*
2.3*
9–13 y
25*
2*
1.9*
4.5*
1.5*
2.3*
4–8 y
15*
1*
1.5*
3.8*
1.2*
1.9*
1–3 y
11*
0.7*
1.2*
3.0*
1.0*
1.5*
(g/d)
SODIUM
(mg/d)
POTASSIUM
(lg/d)
ZINC
(mg/d)
SELENIUM
(lg/d)
PHOSPHORUS
(mg/d)
MOLYBDENUM
TABLE 343-2  Dietary Reference Intakes (DRIs): Recommended Dietary Allowances and Adequate Intakes for Elements
(mg/d)
MANGANESE
(mg/d)
MAGNESIUM
(lg/d)
IRON
(mg/d)
IODINE
(lg/d)
FLUORIDE
(lg/d)
COPPER
(mg/d)
CHROMIUM
GROUP
CALCIUM
LIFE-STAGE
Children
Females
Infants
Males
intake level sufficient to meet the nutrient requirements of nearly all healthy individuals (97–98%) in a group. The RDA is calculated from an estimated average requirement (EAR). If sufficient scientific evidence is not available to establish
an EAR and thus to calculate an RDA, an AI is usually developed. For healthy breast-fed infants, an AI is the mean intake. The AI for other life-stage and sex-specific groups is believed to cover the needs of all healthy individuals in those
Note: This table (taken from the DRI reports; see www.nap.edu) presents recommended dietary allowances (RDAs) in bold type and adequate intakes (AIs) in ordinary type followed by an asterisk (*). An RDA is the average daily dietary
Source: National Academies of Sciences, Engineering, and Medicine. 2019. Dietary Reference Intakes for Sodium and Potassium. https://doi.org/10.17226/25353. Adapted and reproduced with permission from the National Academy of
31–50 y
45*
3*
2.6*
5.1*
1.5*
2.3*
19–30 y
45*
3*
2.6*
5.1*
1.5*
2.3*
14–18 y
44*
3*
2.6*
5.1*
1.5*
2.3*
31–50 y
30*
3*
2.0*
4.7*
1.5*
2.3*
19–30 y
30*
3*
2.0*
4.7*
1.5*
2.3*
14–18 y
29*
3*
2.0*
4.7*
1.5*
2.3*
70 y
20*
3*
1.8*
4.7*
1.2*
1.8*
51–70 y
20*
3*
1.8*
4.7*
1.3*
2.0*
31–50 y
25*
3*
1.8*
4.7*
1.5*
2.3*
19–30 y
25*
3*
1.8*
4.7*
1.5*
2.3*
14–18 y
24*
3*
1.6*
4.7*
1.5*
2.3*
groups, but lack of data or uncertainty in the data makes it impossible to specify with confidence the percentage of individuals covered by this intake.
CHAPTER 343
Nutrient Requirements and Dietary Assessment
Sciences, Courtesy of the National Academies.
Lactating Women
Pregnant Women
TABLE 343-3  Dietary Reference Intakes (DRIs): Recommended Dietary Allowances and Adequate Intakes for Total Water and Macronutrients
LIFE-STAGE
GROUP
TOTAL WATERa
(L/d)
CARBOHYDRATE
(g/d)
TOTAL FIBER
(g/d)
Infants
Birth to 6 mo
0.7*
60*
NDc
31*
4.4*
0.5*
9.1*
6–12 mo
0.8*
95*
ND
30*
4.6*
0.5*
11.0
Children
1–3 y
1.3*
19*
ND
7*
0.7*
4–8 y
1.7*
25*
ND
10*
0.9*
Males
9–13 y
2.4*
31*
ND
12*
1.2*
14–18 y
3.3*
38*
ND
16*
1.6*
19–30 y
3.7*
38*
ND
17*
1.6*
31–50 y
3.7*
38*
ND
17*
1.6*
51–70 y
3.7*
30*
ND
14*
1.6*
70 y
3.7*
30*
ND
14*
1.6*
Females
9–13 y
2.1*
26*
ND
10*
1.0*
14–18 y
2.3*
26*
ND
11*
1.1*
19–30 y
2.7*
25*
ND
12*
1.1*
31–50 y
2.7*
25*
ND
12*
1.1*
51–70 y
2.7*
21*
ND
11*
1.1*
70 y
2.7*
21*
ND
11*
1.1*
Pregnant Women
PART 10
Disorders of the Gastrointestinal System
14–18 y
3.0*
28*
ND
13*
1.4*
19–30 y
3.0*
28*
ND
13*
1.4*
31–50 y
3.0*
28*
ND
13*
1.4*
Lactating Women
14–18
3.8*
29*
ND
13*
1.3*
19–30 y
3.8*
29*
ND
13*
1.3*
31–50 y
3.8*
29*
ND
13*
1.3*
Note: This table (taken from the DRI reports; see www.nap.edu) presents recommended dietary allowances (RDAs) in bold type and adequate intakes (AIs) in ordinary type
followed by an asterisk (*). An RDA is the average daily dietary intake level sufficient to meet the nutrient requirements of nearly all healthy individuals (97–98%) in a group.
The RDA is calculated from an estimated average requirement (EAR). If sufficient scientific evidence is not available to establish an EAR and thus to calculate an RDA, an AI
is usually developed. For healthy breast-fed infants, an AI is the mean intake. The AI for other life-stage and sex-specific groups is believed to cover the needs of all healthy
individuals in those groups, but lack of data or uncertainty in the data make it impossible to specify with confidence the percentage of individuals covered by this intake.
aTotal water includes all water contained in food, beverages, and drinking water. bBased on grams of protein per kilogram of body weight for the reference body weight
(e.g., for adults: 0.8 g/kg body weight for the reference body weight). cNot determined.
Source: National Academies of Sciences, Engineering, and Medicine. 2019. Dietary Reference Intakes for Sodium and Potassium. https://doi.org/10.17226/25353. Adapted
and reproduced with permission from the National Academy of Sciences, Courtesy of the National Academies.
recommended that the CDRR method be used instead of the AMDR
for establishing DRIs for macronutrients in the future.
■
■FACTORS ALTERING NUTRIENT NEEDS
The DRIs are affected by age, sex, growth rate, pregnancy, lacta­
tion, physical activity level, concomitant diseases, drugs, and dietary
composition.
Physiologic Factors 
Growth, strenuous physical activity, preg­
nancy, and lactation all increase needs for energy and several essential
nutrients. Energy needs rise during pregnancy due to fetal growth
demands and increased energy required for milk production during
lactation. Energy needs decrease with loss of lean body mass, the major
determinant of REE. The energy needs of older persons, especially
those aged >70 years, tend to be lower than those of younger persons
because lean tissue, physical activity, and health often decline with age.
Dietary Composition 
Dietary composition affects the biologic
availability and use of nutrients. For example, iron absorption may be
impaired by large amounts of calcium or lead; likewise, non-heme iron
uptake may be impaired by a lack of ascorbic acid and amino acids in
the meal. Bodily protein may be decreased when essential amino acids
`-LINOLENIC ACID
(g/d)
FAT
(g/d)
LINOLEIC ACID
(g/d)
PROTEINb
(g/d)
are not present in sufficient amounts—a rare scenario in U.S. diets.
Animal foods, such as milk, eggs, and meat, have high biologic values,
with most of the needed amino acids present in adequate amounts.
Plant proteins in corn (maize), soy, rice, and wheat have lower biologic
values and must be combined with other plant or animal proteins or
fortified with the amino acids that are deficient to achieve optimal use
by the body.
Route of Intake 
The RDAs apply only to oral intakes. When nutri­
ents are administered parenterally, similar values can sometimes be
used for amino acids, glucose (carbohydrate), fats, sodium, chloride,
potassium, and most vitamins because their intestinal absorption rate
is nearly 100%. However, the oral bioavailability of most mineral ele­
ments may be only half that obtained by parenteral administration. For
some nutrients that are not readily stored in the body or that cannot be
stored in large amounts, timing of administration may also be impor­
tant. For example, amino acids cannot be used for protein synthesis if
they are not supplied together intravenously; instead, they will be used
for energy production. In contrast, among healthy individuals eating
adequate diets, the distribution of protein intake over the course of the
day has little effect on health.
Disease 
Dietary deficiency diseases include protein-calorie malnu­
trition, iron-deficiency anemia, goiter (due to iodine deficiency), rick­
ets and osteomalacia (vitamin D deficiency), xeropthalmia (vitamin A
deficiency), megaloblastic anemia (vitamin B12 or folic acid deficiency),
scurvy (vitamin C/ascorbic acid deficiency), beriberi (thiamin defi­
ciency), and pellagra (niacin and tryptophan deficiency) (Chaps. 344
and 345). Each deficiency disease is characterized by imbalances at the
cellular level between the supply of nutrients or energy and the body’s
nutritional needs for growth, maintenance, and other functions. Imbal­
ances and excesses in nutrient intakes are recognized as risk factors
for certain chronic degenerative diseases, such as saturated/trans-fat
and cholesterol in coronary artery disease; sodium in hypertension;
obesity in hormone-dependent cancers (endometrial and breast); and
ethanol in alcohol addiction. These disorders are multifactorial in their
etiology and pathogenesis, and diet is only one of the many risk factors
for their occurrence. Osteoporosis, for example, is sometimes associ­
ated with calcium deficiency secondary to vitamin D deficiency, as
well as with environment-related risk factors (e.g., smoking, sedentary
lifestyle), physiology (e.g., estrogen deficiency), genetic determinants
(e.g., defects in collagen metabolism), and drug use (chronic steroids
and aromatase inhibitors) (Chap. 423).
■
■DIETARY ASSESSMENT
Nutrition assessment in clinical situations is an iterative process
involving screening for malnutrition risk; assessing the diet and other
biomarkers to determine if malnutrition is present and if so its possible
causes; planning and implementing appropriate dietary intakes and
medical nutrition therapy; and reassessing to make sure that intakes
have been consumed. Disease states affecting the bioavailability,
requirements, use, or excretion of specific nutrients may require spe­
cific measurements of certain nutrients or their biomarkers to assess
status and ensure that nutrient replacement is adequate (Chap. 344).
The Joint Commission, which accredits and certifies health care
organizations in the United States, requires most health care facilities
to perform nutrition screening for identifying possible malnutrition
risk within 24 h after hospital admission. However, a single univer­
sally recognized or validated standard is not used yet. The Global
Leadership Initiative on Malnutrition (GLIM) proposes that in acute
care settings a tool be used that evaluates weight loss, low body mass
index (BMI), reduced muscle mass, reduced food intake or absorption,
and increased disease burden or inflammation. Other screens usually
include these and also abnormal weight for height or body mass index
(e.g., BMI <19 or >25); reported weight change (involuntary loss or
gain of >5 kg in the past 6 months) (Chap. 50); diagnoses with known
nutritional implications (e.g., metabolic disease, any disease affecting
the gastrointestinal tract, alcohol abuse); present therapeutic dietary
prescription; chronic poor appetite; presence of dental, chewing, and
swallowing problems; major food allergies or intolerances; need for
assistance with preparing or shopping for food, eating, or other aspects
of self-care; and social isolation. The nutritional status of hospitalized
patients should be reassessed periodically—at least once every week.
Patients who exhibit frank malnutrition or a high risk of it on nutri­
tional screening need more complete dietary assessment that varies
with the clinical setting, the severity of the patient’s illness, and the
stability of the patient’s condition.
Acute-Care Settings 
The goal is to identify and avoid inadequate
intake, assure intake is appropriate, and gather enough information
to determine if malnutrition is due to poor dietary intake or other
causes and whether nutritional therapy is indicated (Chap. 346). In
acute-care settings, anorexia, various diseases, test procedures, and
medications can further compromise dietary intake. Dietary assess­
ment focuses on what patients are currently eating, whether or not
they are able and willing to eat, and whether or not they experience
any problems with eating. Simple observations may suffice to suggest
inadequate oral intake. These include dietitians’ and nurses’ notes;
observation of a patient’s frequent refusal to eat or the amount of food
served versus that eaten on trays; the frequent performance of tests and
procedures that are likely to cause meals to be skipped; consumption
limited to nutritionally inadequate diet orders (e.g., clear liquids or full
liquids) for more than a few days; the occurrence of fever, dehydra­
tion, gastrointestinal distress, vomiting, diarrhea, or a comatose state;
and the presence of diseases or use of treatments that involve any part
of the alimentary tract. Acutely ill patients with diet-related diseases
such as diabetes and chronic renal failure are in particular need of
assessment because an inappropriate diet may exacerbate these condi­
tions and adversely affect other therapies. Abnormal biochemical values
(serum albumin levels <35 g/L [<3.5 mg/dL]; serum cholesterol levels
<3.9 mmol/L [<150 mg/dL]) are nonspecific but may indicate a need for
further nutritional assessment.
Most therapeutic diets offered in hospitals are calculated to meet
individual nutrient requirements and the RDA if they are eaten.
Exceptions include clear liquids, some full-liquid diets, and test diets
(such as those adhered to in preparation for gastrointestinal proce­
dures), which are inadequate in several nutrients and should not be
used, if possible, for >24 h. However, because as much as half of the
food served to hospitalized patients is not eaten, it cannot be assumed
that the intakes of hospitalized patients are adequate. Dietary assess­
ment should compare how much and what kinds of food the patient
has consumed with the diet that has been provided. Major deviations
in intakes of energy, protein, fluids, or other nutrients of special con­
cern for the patient’s illness should be noted and corrected, especially
for long-staying patients.
Nutritional monitoring is especially important for patients who
are very ill, who have extended lengths of hospital stay, or who are
discharged home with diagnoses involving the gastrointestinal tract,
end-stage renal disease, and insulin-dependent diabetes. Patients who
are fed by enteral and parenteral routes and those receiving home
chemotherapy infusions also require special nutritional assessment
and monitoring by physicians and/or dietitians with certification in
nutritional support (Chap. 346).
CHAPTER 343
Ambulatory Settings 
The aim of dietary assessment in the out­
patient setting is to determine whether or not the patient’s usual diet
is a health risk in itself, if it contributes to existing chronic disease–
related problems, and if medical nutrition therapy is required. The
information is used to plan an eating pattern that fulfills therapeutic
goals while ensuring patient adherence. Dietary assessment should
review the adequacy of present and usual food intakes, including
vitamin and mineral supplements, oral nutritional supplements,
medical foods, other dietary supplements, medications, and alcohol,
because all of these may affect the patient’s nutritional status. Assess­
ment should focus on the dietary constituents that are most likely to
be involved or compromised by the specific diagnosis as well as on
any comorbidities that are present. A diet history of several days of
intake or a food-frequency questionnaire can be used and reviewed to
provide a representation of the usual diet and assess the adequacy of
the patient’s habitual diet. The Automated Self-Administered 24-Hour
Dietary Assessment Tool (ASA24) (which is a web-based, self-admin­
istered, 24-h recall tool), a diet history questionnaire (DHQ), a daily
food checklist, and validated short dietary instruments are all available
at the National Institute of Health’s National Cancer Institute webpage
at https://epi.grants.cancer.gov/dietary-assessment/resources.html. The
Nutritools interactive website (www.nutritools.org) guides clinicians
and researchers to choose among suitable validated dietary assessment
tools used in North America and elsewhere. Assessment findings are
used to craft dietary recommendations for changes in healthier direc­
tions. These may be as simple as a food guide, a food-exchange list, or
more personalized and detailed recommendations by a dietitian that
are tailored to the patient’s food preferences and medical nutrition
therapy needs. Therapeutic dietary prescriptions and menu plans for
most diseases are available in the diet manuals of most hospitals and
from the Academy of Nutrition and Dietetics for many other diseases.
For persons on therapeutic diets, counseling using food-exchange
lists such as those available from the American Diabetes Association
for diabetes and from the Academy of Nutrition and Dietetics or the
National Kidney Foundation’s Council on Renal Nutrition for renal
disease is helpful. Patients who follow ethnic or unusual dietary
Nutrient Requirements and Dietary Assessment
TABLE 343-4  Choose My Plate: A Guide to Individualized
Dietary Planning
EXAMPLES OF STANDARD PORTION SIZES AT
 
INDICATED ENERGY LEVEL
DIETARY FACTOR, UNIT OF
MEASURE (ADVICE)
LOWER:
1600 kcal
MODERATE:
2200 kcal
HIGHER:
2800 kcal
Fruits, cups (Focus on fruits.)
1.5
2.5
Vegetables, cups (Vary
vegetables.)
3.5
Grains, oz eq (Make at least half
of grains whole.)a
Protein foods, oz eq (Go lean
with protein.)b
Dairy, cups or ozc (Choose
calcium-rich foods.)
“Empty” calories, kcald
Sodium, mg
<2300 at all
energy levels
 
 
Physical activity, min
At least 150 min vigorous physical activity per
week at all energy levels
Note: Oils (formerly listed with portions of 5, 6, and 8 teaspoons for the lower,
moderate, and higher energy levels, respectively) are no longer singled out in
Choose My Plate, but rather are included in the empty calories/added sugar
category with SOFAS (calories from solid fats and added sugars). The limit is
the remaining number of calories in each food pattern above after intake of the
recommended amounts of the nutrient-dense foods.
aFor example, 1 serving equals 1 slice bread, 1 cup ready-to-eat cereal, or 0.5 cup
cooked rice, pasta, or cooked cereal. bFor example, 1 serving equals 1 oz lean meat,
poultry, or fish; 1 egg; 1 tablespoon peanut butter; 0.25 cup cooked dry beans; or
0.5 oz nuts or seeds. cFor example, 1 serving equals 1 cup milk or yogurt, 1.5 oz
natural cheese, or 2 oz processed cheese. dFormerly called “discretionary calorie
allowance.” Portions are calculated as the number of calories remaining after all of
the above allotments are accounted for.
Abbreviation: oz eq, ounce equivalent.
Source: Data from U.S. Department of Agriculture (http://www.Choosemyplate.gov).
PART 10
Disorders of the Gastrointestinal System
patterns need extra instruction on how to categorize their foods and
on the appropriate portion sizes that constitute a serving.
For healthy persons who do not require medical nutrition therapy,
a rough guide for avoiding inadequate nutrient intakes as well as
excessive intakes of fat (especially saturated and trans fats), sodium,
sugar, and alcohol is the USDA’s Choose My Plate (Table 343-4). The
Choose My Plate graphic emphasizes a balance between calories and
nutritional needs, encouraging increased intake of fruits and veg­
etables, whole grains, and low-fat milk in conjunction with reduced
intake of sodium and high-calorie sugary drinks. The guide provides
a web-based calculator that tailors the number of servings suggested
for healthy people of different weights, sexes, ages, and life-cycle stages
to meet their needs while avoiding excess (https://www.myplate.gov/
myplate-plan and www.ChooseMyPlate.gov). Reviewing the guide with
patients helps them identify food groups eaten in insufficient amounts
or in excess and aids the transition to healthier dietary patterns.
■
■NUTRITIONAL STATUS ASSESSMENT
Full nutritional status assessment is reserved for seriously ill patients
and those at very high nutritional risk when the cause of malnutrition
is still uncertain after initial clinical evaluation and dietary assessment.
It involves multiple dimensions, including documentation of dietary
intake, health history, anthropometric measurements, biochemical
measurements of blood and urine, clinical examination, and functional
status evaluation. For further discussion of nutritional assessment,
see Chap. 345.
■
■GLOBAL CONSIDERATIONS
Several DRIs (e.g., the EAR, the UL, and energy needs) are estimates
of physiologic requirements based on experimental evidence that apply
globally to healthy individuals when adjusted for age, sex, body size,
and physical activity level. However, the AIs, AMDRs, and CDRRs are
not exportable and should be used with caution in other countries.
AIs for micronutrients are based on epidemiologic observations of
customary and adequate intakes in healthy U.S. and Canadian popu­
lations. The AMDRs are not based on direct experimental evidence
but represent expert opinion on potentially healthy intakes of energyproviding nutrients in North American populations. The CDRR may
also vary in other populations. Nutrient-based standards like the
DRIs have also been developed by the World Health Organization/
Food and Agricultural Organization of the United Nations. They are
available at https://www.who.int/activities/establishing-global-nutrientrequirements. The European Food Safety Authority (EFSA) Panel
on Dietetic Products, Nutrition, and Allergies periodically publishes
its recommendations in the EFSA Journal (https://efsa.onlinelibrary.
wiley.com/journal/18314732). Other countries have also promulgated
recommendations similar to the DRI. The different standards have
many similarities in their basic concepts, definitions, and nutrient rec­
ommendation levels, but some differ from the DRIs as a result of the
functional criteria chosen for nutrient requirements, environmental
differences, the timeliness of the evidence reviewed, and expert judg­
ment. There is a growing trend toward global harmonization of these
recommendations.
DISCLAIMER
The opinions expressed in this article are the author’s own and do not
reflect the views of the National Institutes of Health, the U.S. Depart­
ment of Health and Human Services, or the U.S. government.
■
■FURTHER READING
Alves LF et al: GLIM criteria to identify malnutrition in patients
in hospital settings: A systematic review. J Parenteral Enteral Nutr
47:702, 2023.
Lewis JL, Dwyer JT: Establishing nutrient intake values, in Present
Knowledge in Nutrition, Vol 2. BP Marriott, DF Birt, VA Stallings, AA
Yates, eds. London, Academic Press, 2020, pp 267–289.
National Academy of Sciences, Engineering, and Medicine:
Guiding Principles for Developing Dietary Reference Intakes Based on
Chronic Disease. Washington, DC, National Academies Press, 2017.
National Academy of Sciences, Engineering, and Medicine:
Dietary Reference Intakes for Sodium and Potassium. Washington,
DC, National Academies Press, 2019.
National Academy of Sciences, Engineering, and Medicine:
Harmonizing the Process for Establishing Nutrient Reference Values:
A Tool Kit. Washington, DC, National Academies Press, 2020.
National Academy of Sciences, Engineering, and Medicine:
Nutrition During Pregnancy and Lactation: Exploring New Evidence:
Proceedings of a Workshop—in Brief. Washington, DC, National
Academies Press, 2020.
National Academy of Sciences, Engineering, and Medicine:
Approaches to assessing intake of food and dietary supplements in
pregnant women and children 2 to 11 years of age: Proceedings of a
workshop series. Washington, DC, National Academies Press, 2022.
National Academy of Sciences, Engineering, and Medicine:
Defining populations for Dietary Reference Intake recommenda­
tions: A letter report. Washington, DC, National Academies Press,
2022.
National Academy of Sciences, Engineering, and Medicine:
Assessing intake of food and dietary supplements in older adults:
Proceedings of a workshop series. Washington, DC, National Acad­
emies Press, 2023.
National Academy of Sciences, Engineering, and Medicine:
Dietary Reference Intakes for Energy. Washington, DC, National
Academies Press, 2023.
National Academy of Sciences, Engineering, and Medicine:
Using systematic reviews to support future Dietary Reference Intakes:
A letter report. Washington, DC, National Academies Press, 2023.
US Department of Agriculture and U.S. Department of Health
and Human Services: Dietary Guidelines for Americans, 20202025, 9th ed. Washington, DC, US Government Printing Office,
2020.

15 - 344 Vitamin and Trace Mineral Deficiency and Excess
344 Vitamin and Trace Mineral Deficiency and Excess
Paolo M. Suter
Vitamin and Trace
Mineral Deficiency
and Excess
Vitamins are required constituents of the human diet because they are
synthesized inadequately or not at all in the human body. Only small
amounts of these substances are needed to carry out essential bio­
chemical reactions (e.g., by acting as coenzymes or prosthetic groups).
Overt vitamin or trace mineral deficiencies are rare in Western coun­
tries because of a plentiful, varied, and inexpensive food supply; food
fortification; and use of supplements. However, multiple nutrient
deficiencies may appear together in persons who are chronically ill,
alcoholic, on specific medications (see below), or living in poverty.
After bariatric surgery, patients are at high risk for multiple nutrient
deficiencies. Moreover, subclinical vitamin and trace mineral deficien­
cies (often designated as “hidden hunger”), as diagnosed by laboratory
testing, are quite common in the normal population, especially in the
geriatric age group and socioeconomically deprived individuals due to
the lack of nutrient-dense foods. Conversely, because of the widespread
use of nutrient supplements and food fortification, nutrient toxicities
are gaining pathophysiologic and clinical importance.
Victims of famine, emergency-affected and displaced populations,
refugees, and camp populations are at increased risk for protein-energy
malnutrition and classic micronutrient deficiencies (vitamin A, iron,
zinc, iodine) as well as for overt deficiencies in thiamine (beriberi),
riboflavin, vitamin C (scurvy), and niacin (pellagra).
Body stores of vitamins and minerals vary tremendously. For exam­
ple, stores of vitamins B12 and A are large, and an adult may not become
deficient until ≥1 year after beginning to eat a deficient diet. However,
folate and thiamine may become depleted within weeks among those
eating a deficient diet. Therapeutic modalities can deplete essential
TABLE 344-1  Principal Clinical Findings of Vitamin Malnutrition
NUTRIENT
CLINICAL FINDING
Thiamine
Beriberi: neuropathy, muscle weakness and wasting,
cardiomegaly, edema, ophthalmoplegia, confabulation
Riboflavin
Magenta tongue, angular stomatitis, seborrhea,
cheilosis, ocular symptoms, corneal vascularization
Niacin
Pellagra: pigmented rash of sun-exposed areas,
bright red tongue, diarrhea, apathy, memory loss,
disorientation
Vitamin B6
Seborrhea, glossitis, convulsions, neuropathy,
depression, confusion, microcytic anemia
Folate
Megaloblastic anemia, atrophic glossitis, depression,
↑ homocysteine
Vitamin B12
Megaloblastic anemia, loss of vibratory and position
sense, abnormal gait, dementia, impotence, loss of
bladder and bowel control, ↑ homocysteine,
↑ methylmalonic acid
Vitamin C
Scurvy: petechiae, ecchymosis, coiled hairs, inflamed
and bleeding gums, joint effusion, poor wound healing,
fatigue
Vitamin A
Xerophthalmia, night blindness, Bitot’s spots, follicular
hyperkeratosis, impaired embryonic development,
immune dysfunction
Vitamin D
Rickets: skeletal deformation, rachitic rosary, bowed
legs; osteomalacia
Vitamin E
Peripheral neuropathy, spinocerebellar ataxia, skeletal
muscle atrophy, retinopathy
Vitamin K
Elevated prothrombin time, bleeding
<10 μg/d
Fat malabsorption, liver disease, antibiotic use
nutrients from the body; for example, hemodialysis or diuretics remove
water-soluble vitamins, which must be replaced by supplementation.
Vitamins and trace minerals play several roles in diseases: (1) defi­
ciencies of vitamins and minerals may be caused by disease states such
as malabsorption; (2) either deficiency or excess of vitamins and min­
erals can cause disease in and of itself (e.g., vitamin A intoxication and
liver disease); and (3) vitamins and minerals in high doses may be used
as drugs (e.g., niacin for hypercholesterolemia). Since they are covered
elsewhere, the hematologic-related vitamins and minerals (Chaps. 102
and 104) either are not considered or are considered only briefly in
this chapter, as are the bone-related vitamins and minerals (vitamin D,
calcium, phosphorus, magnesium; Chap. 421).
VITAMINS
See also Table 344-1 and Fig. 344-1.
■
■THIAMINE (VITAMIN B1)
Thiamine was the first B vitamin to be identified and therefore is
referred to as vitamin B1. Thiamine functions in the decarboxylation of
α-ketoacids (e.g., pyruvate α-ketoglutarate) and branched-chain amino
acids and thus is essential for energy generation. In addition, thiamine
pyrophosphate acts as a coenzyme for a transketolase reaction that
mediates the conversion of hexose and pentose phosphates. It has been
postulated that thiamine plays a role in peripheral nerve conduction,
although the exact chemical reactions underlying this function are not
known.
Food Sources 
The median intake of thiamine in the United States
from food alone is ~2 mg/d. Primary food sources for thiamine include
yeast, organ meat, pork, legumes, beef, whole grains, and nuts. Milled
rice and grains contain little thiamine. Thiamine deficiency is there­
fore more common in cultures that rely heavily on a milled polished
rice-based diet. Certain foods contain antithiamine factors such as
heat-labile thiaminases (raw fish, shellfish), which destroy the vitamin,
or heat-stable polyhydroxyphenols (tannins; in coffee, tea, Brussels
sprouts, or betel nuts), which inactivate the vitamin. Thus, drinking
large amounts of tea or coffee could theoretically lower thiamine body
stores.
CHAPTER 344
Vitamin and Trace Mineral Deficiency and Excess
DIETARY LEVEL PER DAY ASSOCIATED
WITH OVERT DEFICIENCY IN ADULTS
CONTRIBUTING FACTORS TO DEFICIENCY
<0.3 mg/1000 kcal
Alcoholism, chronic diuretic use, bariatric surgery,
hyperemesis, thiaminases in food
<0.4 mg
Alcoholism, individuals with poor diets and low
intake of milk products
<9.0 niacin equivalents
Alcoholism, vitamin B6 deficiency, riboflavin
deficiency, tryptophan deficiency
<0.2 mg
Alcoholism, isoniazid
<100 μg/d
Alcoholism, sulfasalazine, pyrimethamine,
triamterene
<1.0 μg/d
Gastric atrophy (pernicious anemia), terminal ileal
disease, strict vegetarianism, acid-reducing drugs
(e.g., H2 blockers), metformin
<10 mg/d
Smoking, alcoholism
<300 μg/d
Fat malabsorption, infection, measles, alcoholism,
protein-energy malnutrition
<2.0 μg/d
Aging, lack of sunlight exposure, fat malabsorption,
deeply pigmented skin
Not described unless underlying
contributing factor is present
Occurs only with fat malabsorption or genetic
abnormalities of vitamin E metabolism/transport
Deficiency 
Most dietary deficiency of thiamine worldwide is the
result of poor dietary intake due to the lack of food or disproportion­
ate reliance on highly processed staple crops. Food processing removes
thiamine, and high-heat or long-duration cooking destroys it. In Western
countries, the primary causes of thiamine deficiency are alcohol­
ism and chronic illnesses such as cancer. Alcohol interferes directly
with the absorption of thiamine and with the synthesis of thiamine
Vitamin
Thiamine (B1)
NH2
CH3
N
N
N
S
CH2CH2OH
Riboflavin (B2)
Flavin mononucleotide
(FMN) and flavin
adenine dinucleotide
(FAD)
O
N
NH
O
N
N
Ribityl
Niacin
Nicotinamide adenine
dinucleotide phosphate
(NADP) and nicotinamide
adenine dinucleotide
(NAD)
PART 10
Disorders of the Gastrointestinal System
O
C
O
+
N
H
Vitamin B6
Pyridoxal phosphate
Cofactor for
enzymes of amino
acid metabolism
CH2OH
CH2OH
HO
N
Folate
Polyglutamate forms of
(5, 6, 7, 8)
tetrahydrofolate with
carbon unit
attachments
n
N
H
O
COOH
N
C
CH
O
H
CH2
N
CH2
CH2
H2N
N
N
C
CH
N
H
O
Vitamin B12
Methylcobalamine
Adenosylcobalamin
Coenzyme for
methionine
synthase
and
CH2CH2CONH2
CONH2
H3C
CH2CH2CONH2
CH3
CH2
CH2CH2CONH2
H3C
H3C
Co+
N
N
CONH2
N
N
CH3
CH3
CH2
CH3
NHCOCH2CH2
CH2CH2CONH2
CH3
CH2
CHCH3
N
CH3
O–
N
O
CH3
P
O
O
HO
O
HOCH2
OH
Cbl
FIGURE 344-1  Structures and principal functions of vitamins associated with human disorders.
pyrophosphate, and it increases urinary excretion. Thiamine should
always be replenished when a patient with alcoholism is being refed,
as carbohydrate repletion without adequate thiamine can precipitate
acute thiamine deficiency with lactic acidosis. Other at-risk popula­
tions are women with prolonged hyperemesis gravidarum, anorexia
(including eating disorders), patients with overall poor nutritional
status who are receiving parenteral glucose, patients who have had
Active derivative or
cofactor form
Principal
function
Thiamine pyrophosphate
Coenzyme for
cleavage of
carbon-carbon
bonds; amino
acid and
carbohydrate
metabolism
Cofactor for
oxidation,
reduction
reactions,
and covalently
attached
prosthetic groups
for some
enzymes
Coenzymes for
oxidation and
reduction
reactions
Coenzyme for one
carbon transfer in
nucleic acid and
amino acid
metabolism
COOH
CH2
CH2
C
OH
O
L-methylmalonyl-
CoA mutase
bariatric/metabolic surgery (bariatric Wernicke), and patients receiving
chronic diuretic therapy (e.g., in hypertension or systolic heart failure)
due to increased urinary thiamine losses. Different drugs (e.g., metfor­
min, verapamil) could inhibit intestinal thiamine transporters (ThTR2), thereby increasing the risk of deficiency for this vitamin. Maternal
thiamine deficiency can lead to infantile beriberi in breast-fed children.
Thiamine deficiency could be an underlying factor in motor vehicle
accidents and could be overlooked in the setting of head injury.
Thiamine deficiency in its early stage is challenging to identify. It
causes anorexia and nonspecific symptoms (e.g., irritability, decrease
in short-term memory). Prolonged thiamine deficiency causes beriberi,
which is classically categorized as wet or dry, although there is consid­
erable overlap between the two categories. In either form of beriberi,
patients may complain of pain and paresthesia. Wet beriberi presents
primarily with cardiovascular symptoms that are due to impaired myo­
cardial energy metabolism and dysautonomia; it can occur after 3 months
of a thiamine-deficient diet. Patients present with an enlarged heart,
tachycardia, high-output congestive heart failure, peripheral edema,
and peripheral neuritis. Patients with dry beriberi present with a sym­
metric peripheral neuropathy of the motor and sensory systems, with
diminished reflexes. The neuropathy affects the legs most markedly,
and patients have difficulty rising from a squatting position.
Alcoholic patients with chronic thiamine deficiency also may have
central nervous system (CNS) manifestations known as Wernicke’s
encephalopathy, which consists of horizontal nystagmus, ophthal­
moplegia (due to weakness of one or more extraocular muscles),
cerebellar ataxia, and mental impairment (Chap. 464). When there
Vitamin
Vitamin C
O
O
C
C
C
C
C
CH2OH
OH OH
OH
Vitamin A
Retinol, retinaldehyde,
and retinoic acid
Formation of
rhodopsin (vision)
and glycoproteins
(epithelial cell
function); also
regulates gene
transcription
(β-Carotene)
CH2OH
(Retinol)
Vitamin D
1,25-Dihydroxyvitamin D
Maintenance of
blood calcium
and phosphorus
levels;
antiproliferative
hormone
CH2
OH
HO
Vitamin E
Tocopherols and
tocotrienols
Antioxidants
O
CH2[CH2
CH2]3H
CH
CH2
HO
Vitamin K
Vitamin K hydroquinone
Cofactor for
posttranslation
carboxylation of
many proteins
including essential
clotting factors
O
O
R
FIGURE 344-1  (Continued)
is an additional loss of memory and a confabulatory psychosis, the
syndrome is known as Wernicke-Korsakoff syndrome. Despite the
typical clinical picture and history, Wernicke-Korsakoff syndrome is
underdiagnosed.
The laboratory diagnosis of thiamine deficiency usually is made by
a functional enzymatic assay of transketolase activity measured before
and after the addition of thiamine pyrophosphate. A >25% stimulation
in response to the addition of thiamine pyrophosphate (i.e., an activity
coefficient of 1.25) is interpreted as abnormal. Thiamine or the phos­
phorylated esters of thiamine in serum or blood also can be measured
by high-performance liquid chromatography to detect deficiency.
TREATMENT
Thiamine Deficiency
In acute thiamine deficiency with either cardiovascular or neu­
rologic signs, 200 mg of thiamine three times daily should be
given intravenously until there is no further improvement in acute
symptoms; oral thiamine (10 mg/d) should subsequently be given
until recovery is complete. Cardiovascular and ophthalmoplegic
improvement occurs within 24 h. Other manifestations gradually
clear, although psychosis in Wernicke-Korsakoff syndrome may
be permanent or may persist for several months. Other nutri­
ent deficiencies should be corrected concomitantly (e.g., magne­
sium, required for thiamine activation). In view of the widespread,
often unrecognized (subclinical) deficiency, a more generous
CHAPTER 344
Active derivative or
cofactor form
Principal
function
Ascorbic acid and
dehydroascorbic acid
Participation as a
redox ion in many
biologic
oxidation and
hydrogen transfer
reactions
Vitamin and Trace Mineral Deficiency and Excess
OH
CH3
supplementation of this vitamin in the emergency care setting and
in all patients with any alcohol use disorder is warranted.
Toxicity 
Although hypersensitivity/anaphylaxis has been reported
after high intravenous doses of thiamine, no adverse effects have been
recorded from either food or supplements at high doses.
■
■RIBOFLAVIN (VITAMIN B2)
Riboflavin is important for the metabolism of fat, carbohydrate, and
protein, acting as a respiratory coenzyme and an electron donor in
energy production. Enzymes that contain flavin adenine dinucleotide
(FAD) or flavin mononucleotide (FMN) as prosthetic groups are
known as flavoenzymes (e.g., succinic acid dehydrogenase, monoamine
oxidase, glutathione reductase). FAD is a cofactor for methyltetrahy­
drofolate reductase and therefore modulates homocysteine metabo­
lism. The vitamin also plays a role in drug and steroid metabolism,
including detoxification reactions.
Although much is known about the chemical and enzymatic reac­
tions of riboflavin, the clinical manifestations of riboflavin deficiency
are nonspecific and are similar to those of other deficiencies of B vita­
mins. Riboflavin deficiency is manifested principally by lesions of the
mucocutaneous surfaces of the mouth and skin. In addition, corneal
vascularization, anemia, and personality changes have been described
with riboflavin deficiency.
Deficiency and Excess 
Riboflavin deficiency is rare and almost
always due to dietary deficiency. Milk, other dairy products, and
enriched breads and cereals are the most important dietary sources of
riboflavin in the United States, although lean meat, fish, eggs, broccoli,
and legumes are also good sources. Riboflavin is extremely sensitive
to light, and milk should be stored in containers that protect against
photodegradation. Laboratory diagnosis of riboflavin deficiency can
be made by determination of red blood cell or urinary riboflavin con­
centrations or by measurement of erythrocyte glutathione reductase
activity, with and without added FAD. Because of the limited capacity
of the gastrointestinal tract to absorb riboflavin (~27 mg after one oral
dose or meal) as well as the instantaneous urinary excretion, riboflavin
toxicity has not been described.
PART 10
Disorders of the Gastrointestinal System
■
■NIACIN (VITAMIN B3)
The term niacin refers to nicotinic acid and nicotinamide and their
biologically active derivatives. Nicotinic acid and nicotinamide serve
as precursors of two coenzymes, nicotinamide adenine dinucleotide
(NAD) and NAD phosphate (NADP), which are important in numer­
ous oxidation and reduction reactions in the body. In addition, NAD
and NADP are active in adenine diphosphate–ribose transfer reactions
involved in DNA repair and calcium mobilization.
Metabolism and Requirements 
Nicotinic acid and nicotinamide
are absorbed well from the stomach and small intestine. The bioavail­
ability of niacin from beans, milk, meat, and eggs is high; bioavailability
from cereal grains is lower. Since flour is enriched with “free” niacin
(i.e., the non-coenzyme form), bioavailability is excellent. Median
intakes of niacin in the United States considerably exceed the recom­
mended dietary allowance (RDA).
The amino acid tryptophan can be converted to niacin with an
efficiency of 60:1 by weight. Thus, the RDA for niacin is expressed in
niacin equivalents. A lower-level conversion of tryptophan to niacin
occurs in vitamin B6, riboflavin, and/or iron deficiencies, and in the
presence of isoniazid. The urinary excretion products of niacin include
2-pyridone and 2-methyl nicotinamide, measurements of which are
used in the diagnosis of niacin deficiency.
Deficiency 
Niacin deficiency causes pellagra, which is found
mostly among people eating corn-based diets in parts of China, Africa,
and India. Pellagra in North America is found mainly among alcohol­
ics; among patients with congenital defects of intestinal and kidney
absorption of tryptophan (Hartnup disease); and among patients with
carcinoid syndrome, in which there is increased conversion of trypto­
phan to serotonin. The antituberculosis drug isoniazid is a structural
analogue of niacin and can precipitate pellagra. This also occurs in the
setting of mass scale-up of isoniazid tuberculosis preventive therapy in
low-income countries. In the setting of famine or population displace­
ment, pellagra results from the absolute lack of niacin but also from the
deficiency of micronutrients required for the conversion of tryptophan
to niacin (e.g., iron, riboflavin, and pyridoxine). The early symptoms
of pellagra include loss of appetite, generalized weakness and irritabil­
ity, abdominal pain, and vomiting. Bright red glossitis then ensues and
is followed by a characteristic skin rash that is pigmented and scaling,
particularly in skin areas exposed to sunlight. This rash is known as
Casal’s necklace because it forms a ring around the neck; it is seen in
advanced cases. Vaginitis and esophagitis also may occur. Diarrhea
(due in part to proctitis and in part to malabsorption), depression,
seizures, and dementia are also part of the pellagra syndrome. The pri­
mary manifestations of this syndrome are sometimes referred to as “the
four Ds”: dermatitis, diarrhea, and dementia leading to death. Aging is
characterized by a decline in cellular NAD+, and it seems plausible that
maintaining and/or reestablishing cellular NAD+, might have beneficial
effects (e.g., metabolic disorders).
TREATMENT
Pellagra
Treatment of pellagra consists of oral supplementation with 100–
200 mg of nicotinamide or nicotinic acid three times daily for up to
4 weeks. High doses of nicotinic acid (2 g/d in a time-release form)
may be used for the treatment of elevated cholesterol and triglyc­
eride levels and/or low high-density lipoprotein cholesterol levels,
but without proven evidence to prevent cardiovascular disease.
Nevertheless, nicotinic acid may be useful in patients with statin
intolerance or severe hypertriglyceridemia (Chap. 419).
Toxicity 
Prostaglandin-mediated flushing due to binding of the
vitamin to a G protein–coupled receptor has been observed at daily
nicotinic acid doses as low as 30 mg taken as a supplement or as
therapy for dyslipidemia. There is no evidence of toxicity from niacin
that is derived from food sources. Flushing always starts in the face
and may be accompanied by skin dryness, itching, paresthesia, and
headache. Flushing is subject to tachyphylaxis and often improves
with time; premedication with aspirin may alleviate these symptoms.
Nausea, vomiting, and abdominal pain also occur at similar doses of
niacin. Hepatic toxicity is the most serious toxic reaction caused by
sustained-release niacin and may present as jaundice with elevated
aspartate aminotransferase (AST) and alanine aminotransferase (ALT)
levels. A few cases of fulminant hepatitis requiring liver transplantation
have been reported at doses of 3–9 g/d. Other toxic reactions include
glucose intolerance, hyperuricemia, macular edema, and macular
cysts. The combination of nicotinic acid preparations for dyslipidemia
plus 3-hydroxy-3-methylglutaryl–coenzyme A (HMG-CoA) reductase
inhibitors may increase the risk of rhabdomyolysis. The upper limit for
daily (nontherapeutic) niacin intake has been set at 35 mg.
■
■PYRIDOXINE (VITAMIN B6)
Vitamin B6 refers to a family of compounds that includes pyridoxine,
pyridoxal, pyridoxamine, and their 5′-phosphate derivatives. 5′-Pyridoxal
phosphate (PLP) is a cofactor for >100 enzymes involved in amino acid
metabolism. Vitamin B6 also is involved in heme and neurotransmitter
synthesis and in the metabolism of glycogen, lipids, steroids, sphingoid
bases, and several vitamins, including the conversion of tryptophan to
niacin.
Dietary Sources 
Plants contain vitamin B6 in the form of pyridox­
ine, whereas animal tissues contain PLP and pyridoxamine phosphate.
The vitamin B6 contained in plants is less bioavailable than that in
animal tissues. Rich food sources of vitamin B6 include legumes, nuts,
wheat bran, and meat, although it is present in all food groups.
Deficiency 
Symptoms of vitamin B6 deficiency include epithelial
changes, as seen frequently with other B vitamin deficiencies. In
addition, severe vitamin B6 deficiency can lead to peripheral neuropa­
thy, abnormal electroencephalograms, and personality changes that
include depression and confusion. In infants, diarrhea, seizures, and
anemia have been reported. Microcytic hypochromic anemia is due
to diminished hemoglobin synthesis, since the first enzyme involved
in heme biosynthesis (aminolevulinate synthase) requires PLP as a
cofactor (Chap. 102). In some case reports, platelet dysfunction has
been reported. Since vitamin B6 is necessary for the conversion of
homocysteine to cystathionine, it is possible that chronic low-grade
vitamin B6 deficiency may result in hyperhomocysteinemia, which has
been associated with vascular dysfunction and an increased risk of car­
diovascular disease; however, so far, there is only limited randomized
controlled trial evidence (Chap. 431). Independent of homocysteine,
low levels of circulating vitamin B6 have been associated with inflam­
mation and elevated levels of C-reactive protein.
Certain medications, such as isoniazid, l-dopa, penicillamine, and
cycloserine, interact with PLP due to a reaction with carbonyl groups.
Pyridoxine should be given concurrently with isoniazid to avoid
neuropathy. The increased ratio of AST to ALT seen in alcoholic liver
disease reflects the relative vitamin B6 dependence of ALT. Vitamin B6
dependency syndromes that require pharmacologic doses of vitamin B6
are rare; they include cystathionine β-synthase deficiency, pyridoxineresponsive (primarily sideroblastic) anemias, and gyrate atrophy with
chorioretinal degeneration due to decreased activity of the mitochon­
drial enzyme ornithine aminotransferase. In these situations, 100–200 mg/d
of oral vitamin B6 is required for treatment.
Severe nausea and vomiting in pregnancy might respond to pyri­
doxine combined with doxylamine. High doses of vitamin B6 have
been used to treat carpal tunnel syndrome, premenstrual syndrome,
schizophrenia, autism, and diabetic neuropathy but have not been
found to be effective.
The laboratory diagnosis of vitamin B6 deficiency is generally based
on low plasma PLP values (<20 nmol/L). Vitamin B6 deficiency is
treated with 50 mg/d; higher doses of 100–200 mg/d are given if the
deficiency is related to medication use. Vitamin B6 should not be given
with l-dopa, since the vitamin interferes with the action of this drug.
Toxicity 
The safe upper limit for vitamin B6 has been set at 100 mg/d,
although no adverse effects have been associated with high intakes of
vitamin B6 from food sources only. When toxicity occurs, it causes
severe sensory neuropathy, leaving patients unable to walk; however, in
most cases, this is reversible upon cessation of the high intake. Medi­
cation safety monitoring suggests a rather high prevalence of vitamin
B6–induced neuropathy. Accordingly, long-term high-dose vitamin B6
supplementation should be discouraged. Some cases of photosensitiv­
ity and dermatitis have been reported.
■
■FOLATE (VITAMIN B12)
See Chap. 104.
■
■VITAMIN C
Both ascorbic acid (only the l-isomer) and its oxidized product dehy­
droascorbic acid are biologically active. Actions of vitamin C include
antioxidant activity, promotion of nonheme iron absorption, carnitine
biosynthesis, conversion of dopamine to norepinephrine, tyrosine
catabolism, histone and DNA demethylation, and synthesis of many
peptide hormones. Vitamin C is also important for connective tissue
metabolism and cross-linking (proline hydroxylation), and it is a com­
ponent of many drug-metabolizing enzyme systems, particularly the
mixed-function oxidase systems.
Absorption and Dietary Sources 
Vitamin C is almost com­
pletely absorbed if <100 mg is administered in a single dose; however,
only ≤50% is absorbed at doses >1 g. Enhanced degradation and fecal
and urinary excretion of vitamin C occur at higher intake levels.
Good dietary sources of vitamin C include citrus fruits, green veg­
etables (especially broccoli), tomatoes, and potatoes. Consumption
of five servings of fruits and vegetables a day provides vitamin C in
excess of the RDA of 90 mg/d for men and 75 mg/d for women. In
addition, ~40% of the U.S. population consumes vitamin C as a dietary
supplement in which “natural forms” of the vitamin are no more bio­
available than synthetic forms. Smoking (including “passive” smoking),
hemodialysis, pregnancy, lactation, and stress (e.g., infection, trauma)
appear to increase vitamin C requirements.
Deficiency 
Vitamin C deficiency causes scurvy. In the United States,
this condition is seen primarily among the poor and the elderly, in
alcoholics who consume <10 mg/d of vitamin C, and in young adults
who eat severely unbalanced diets. In addition to generalized fatigue,
symptoms of scurvy primarily reflect impaired formation of mature
connective tissue and include bleeding into the skin (petechiae, ecchy­
moses, perifollicular hemorrhages); inflamed and bleeding gums;
and manifestations of bleeding into joints, the peritoneal cavity, the
pericardium, and the adrenal glands. In children, vitamin C deficiency
may cause impaired bone growth. Laboratory diagnosis of vitamin C
deficiency is based on low plasma or leukocyte levels.
Administration of vitamin C (200 mg/d) improves the symptoms
of scurvy within several days. High-dose vitamin C supplementation
(e.g., 0.2 g up to several grams per day) may slightly decrease the symp­
toms and duration of upper respiratory tract infections. Vitamin C sup­
plementation has also been reported to be useful in Chédiak-Higashi
syndrome (Chap. 67) and osteogenesis imperfecta (Chap. 425). Diets
high in vitamin C have been claimed to lower the incidence of certain
cancers, particularly esophageal and gastric cancers. If proven, this
effect may be because vitamin C can prevent the conversion of nitrites
and secondary amines to carcinogenic nitrosamines. Emerging evi­
dence suggests a therapeutic effect of intravenous parenteral (not oral)
pharmacologic doses of up to 1 g/kg body weight of ascorbic acid in the
treatment of cancers (e.g., metastatic pancreatic, ovarian, glioblastoma,
and non-small-cell lung cancers). The mechanism of pharmacologic
ascorbate in cancer treatment (as a stand-alone agent or with other
therapeutic agents) appears to be pro-oxidative, either synergistic (e.g.,
gemcitabine, programmed cell death protein 1 [PD-1] inhibitors, radia­
tion) or additive with other agents.
CHAPTER 344
Vitamin and Trace Mineral Deficiency and Excess
Toxicity 
Taking >2 g of vitamin C in a single dose may result
in abdominal pain, diarrhea, and nausea. Since vitamin C may be
metabolized to oxalate, it is feared that chronic high-dose vitamin C
supplementation could result in an increased prevalence of kidney
stones. However, except in patients with preexisting renal disease, this
association has not been borne out in several trials. Nevertheless, it is
reasonable to advise patients with a history of kidney stones (especially
oxalate renal stones) and renal insufficiency not to take large doses of
vitamin C. There is also an unproven but possible risk that chronic high
doses of vitamin C could promote iron overload and iron toxicity (e.g.,
in patients with hemochromatosis or thalassemia major). High doses of
vitamin C can induce hemolysis in patients with glucose-6-phosphate
dehydrogenase deficiency, and doses >1 g/d can cause false-negative
guaiac reactions and interfere with tests for urinary glucose. High
doses may interfere with the activity of certain drugs and diagnostic
tests (e.g., false-negative results of guaiac-based fecal occult blood
tests). Parenteral high-dose vitamin C in patients with severe infections
or sepsis has no beneficial effects and may increase mortality.
■
■BIOTIN
Biotin (also known as vitamin B7 or vitamin H) is a water-soluble vita­
min that plays a role in gene expression, gluconeogenesis, and fatty acid
synthesis and serves as a carbon dioxide (CO2) carrier on the surface
of both cytosolic and mitochondrial carboxylase enzymes. The vitamin
also functions in the catabolism of specific amino acids (e.g., leucine)
and in gene regulation by histone biotinylation. Excellent food sources
of biotin include organ meat such as liver or kidney, soy and other
beans, yeast, and egg yolks; however, egg white contains the protein
avidin, which strongly binds the vitamin and reduces its bioavailability.
Biotin deficiency due to low dietary intake is rare; rather, deficiency
is due to inborn errors of metabolism (e.g., biotinidase deficiency).
Biotin deficiency has been induced by experimental feeding of egg
white diets and by biotin-free parenteral nutrition in patients with
short bowels. In adults, biotin deficiency results in mental changes
(depression, hallucinations), paresthesia, anorexia, and nausea. A
scaling, seborrheic, and erythematous rash may occur around the eyes,
nose, and mouth as well as on the extremities. In infants, biotin defi­
ciency presents as hypotonia, lethargy, and apathy. In addition, infants
may develop alopecia and a characteristic rash that includes the ears.
At present, evidence does not support a therapeutic role of high-dose
biotin in multiple sclerosis. The laboratory diagnosis of biotin defi­
ciency can be established on the basis of a decreased concentration of
urinary biotin (or its major metabolites), increased urinary excretion of
3-hydroxyisovaleric acid after a leucine challenge, or decreased activity
of biotin-dependent enzymes in lymphocytes (e.g., propionyl-CoA car­
boxylase). Treatment requires pharmacologic doses of biotin, that is, up
to 10 mg/d. No toxicity is known. High-dose biotin supplements could
interfere with different immunoassay platforms based on streptavidinbiotin technology (e.g., biotinylated antibodies), resulting in false-positive
(e.g., free T4 or T3) or false-negative tests (e.g., thyroid-stimulating
hormone, troponin, β-human chorionic gonadotropin pregnancy test).
■
■PANTOTHENIC ACID (VITAMIN B5)
Pantothenic acid is a component of coenzyme A and phosphopante­
theine, which are involved in fatty acid metabolism and the synthesis
of cholesterol, steroid hormones, and all compounds formed from
isoprenoid units. In addition, pantothenic acid is involved in the acety­
lation of proteins. The vitamin is excreted in the urine, and the labora­
tory diagnosis of deficiency is based on low urinary vitamin levels.
The vitamin is ubiquitous in the food supply. Liver, yeast, egg
yolks, whole grains, and vegetables are particularly good sources.
Human pantothenic acid deficiency has been demonstrated only by
experimental feeding of diets low in pantothenic acid or by admin­
istration of a specific pantothenic acid antagonist. The symptoms of
pantothenic acid deficiency are nonspecific and include gastrointes­
tinal disturbance, depression, muscle cramps, paresthesia, ataxia, and
hypoglycemia. Pantothenic acid deficiency is believed to have caused
the “burning feet syndrome” seen in prisoners of war during World
War II. No toxicity of this vitamin has been reported.
PART 10
Disorders of the Gastrointestinal System
■
■CHOLINE
Choline is a precursor for acetylcholine, phospholipids, and betaine.
Choline is necessary for the structural integrity of cell membranes,
cholinergic neurotransmission, lipid and cholesterol metabolism,
methyl-group metabolism, and transmembrane signaling. Recently,
a recommended adequate intake was set at 550 mg/d for men and
425 mg/d for women, although certain genetic polymorphisms can
increase an individual’s requirement. Choline is thought to be a “con­
ditionally essential” nutrient in that its de novo synthesis occurs in the
liver and results in lesser-than-used amounts only under certain stress
conditions (e.g., alcoholic liver disease). The dietary requirement for
choline depends on the status of other nutrients involved in methylgroup metabolism (folate, vitamin B12, vitamin B6, and methionine)
and thus varies widely. Choline is widely distributed in food (e.g., egg
yolks, wheat germ, organ meat, milk) in the form of lecithin (phospha­
tidylcholine). Choline deficiency has occurred only in experimental
conditions or in patients receiving parenteral nutrition devoid of
choline and rarely in specific inborn errors of choline metabolism.
Deficiency results in fatty liver, elevated aminotransferase levels, and
skeletal muscle damage with high creatine phosphokinase values. The
diagnosis of choline deficiency is currently based on low plasma levels,
although nonspecific conditions (e.g., heavy exercise) may also sup­
press plasma levels.
Toxicity from choline results in hypotension, increased sweating,
diarrhea, salivation, and a fishy body odor. The upper limit for choline
intake has been set at 3.5 g/d. Because of its ability to lower choles­
terol and homocysteine levels, choline treatment has been suggested
for patients with dementia and patients at high risk of cardiovascular
disease. However, the benefits of such treatment have not been firmly
documented; recently, signals for an increased cardiovascular risk have
been reported. Choline- and betaine-restricted diets are of therapeutic
value in trimethylaminuria (“fish odor syndrome”) or in decreasing the
production of the gut microbiome-derived trimethylamine N-oxide
(TMAO) as a potential cardiovascular risk modulator.
■
■FLAVONOIDS
Flavonoids constitute a large family of polyphenolic phytochemicals
that contribute to the aroma, taste, and color of fruits and vegetables.
Major groups of dietary flavonoids include anthocyanidins in ber­
ries; flavanols (catechins) in green tea and chocolate; flavonols (e.g.,
quercetin) in broccoli, kale, leeks, onions, and the skins of grapes and
apples; and isoflavones (e.g., genistein) in legumes. Isoflavones have a
low bioavailability and are partially metabolized by the intestinal flora.
The dietary intake of flavonoids is highly variable and estimated at
10–400 mg/d; this figure is almost certainly an underestimate attribut­
able to a lack of information on their concentrations in many foods.
Several flavonoids have antioxidant activity and affect cell signaling.
From observational epidemiologic studies and limited clinical (human
and animal) studies, flavonoids have been postulated to play a role
in the prevention of several chronic diseases, including neurodegen­
erative disease, diabetes, and osteoporosis. The ultimate importance
and usefulness of these compounds against human disease have not
been consistently demonstrated. Nevertheless, a dietary pattern with
high intake of fruits, vegetables, and legumes should be encouraged
to assure a higher intake of these and others nonnutritive bioactives.
■
■VITAMIN A
Vitamin A, in the strictest sense, refers to retinol and retinyl esters.
However, the oxidized metabolites retinaldehyde and retinoic acid
are also biologically active compounds. The term retinoids includes
all molecules (including synthetic molecules) that are chemically
related to retinol. Retinaldehyde (11-cis) is the form of vitamin A that
is required for normal vision, whereas retinoic acid is necessary for
normal morphogenesis, growth, and cell differentiation. Retinoic acid
does not function directly in vision and, in contrast to retinol, is not
involved in reproduction. Vitamin A also plays a role in iron utiliza­
tion, humoral immunity, T cell–mediated immunity, natural killer cell
activity, and phagocytosis.
Vitamin A is found in the human food supply in two forms: pre­
formed as retinyl esters and provitamin A carotenoids. There are >700
carotenoids in nature, ~50 of which can be metabolized to vitamin A.
β-Carotene is the most prevalent carotenoid with provitamin A activity
in the food supply. In humans, significant fractions of carotenoids are
absorbed intact and are stored in liver and fat. It is estimated that in
healthy humans ≥12 μg (range, 4–27 μg) of dietary all-trans β-carotene
is equivalent to 1 μg of retinol activity, whereas the figure is ≥24 μg
for other dietary provitamin A carotenoids (e.g., β-cryptoxanthin,
α-carotene). The vitamin A equivalency for a β-carotene supplement
in an oily solution is 2:1.
Metabolism 
The liver contains ~90% of the vitamin A reserves in
healthy individuals and secretes vitamin A in the form of retinol, which
is bound in the circulation to retinol-binding protein. Once binding
has occurred, the retinol-binding protein complex interacts with a
second protein, transthyretin. This trimolecular complex functions to
prevent vitamin A from being filtered by the kidney glomerulus, thus
protecting the body against the toxicity of retinol and allowing retinol
to be taken up by specific cell-surface receptors that recognize retinolbinding protein. A certain amount of vitamin A enters peripheral cells
even if it is not bound to retinol-binding protein. After retinol is inter­
nalized by the cell, it becomes bound to a series of cellular retinol-binding
proteins, which function as sequestering and transporting agents as
well as co-ligands for enzymatic reactions. Certain cells also contain
retinoic acid–binding proteins, which have sequestering functions but
also shuttle retinoic acid to the nucleus and enable its metabolism.
Vitamin A metabolites (retinoids) such as retinoic acid are potent
regulators of gene transcription through nuclear receptor signaling,
thus playing a key role in many cellular and metabolic pathways. Two
families of receptors (retinoic acid receptors [RARs] and retinoid X
receptors [RXRs]) are active in retinoid-mediated gene transcription.
Retinoid receptors regulate transcription by binding as dimeric com­
plexes to specific DNA sites—the retinoic acid response elements—in
target genes (Chap. 389). The receptors can either stimulate or repress
gene expression in response to their ligands. RARs bind all-trans
retinoic acid and 9-cis-retinoic acid, whereas RXRs bind only 9-cisretinoic acid.
The retinoid receptors play an important role in controlling cell
proliferation and differentiation. RXRs dimerize with other nuclear
receptors to function as coregulators of genes responsive to retinoids,
but also to thyroid hormone and calcitriol. RXR agonists induce insulin
sensitivity experimentally, perhaps because RXRs are cofactors for the
peroxisome proliferator-activated receptors, which also mediate fatty
acid and carbohydrate metabolism and are targets for different drugs
including thiazolidinedione drugs (e.g., rosiglitazone and pioglitazone)
(Chap. 416).
Dietary Sources 
The retinol activity equivalent (RAE) is used to
express the vitamin A value of food: 1 RAE is defined as 1 μg of retinol
(0.003491 mmol), 12 μg of β-carotene, and 24 μg of other provitamin
A carotenoids. In older literature, vitamin A often was expressed in
international units (IUs), with 1 μg of retinol equal to 3.33 IU of retinol
and 20 IU of β-carotene. Although these IUs are no longer in scientific
use, they can still be found in reports of the food industry and in public
health interventions in low-income countries.
Liver, fish, and eggs are excellent food sources for preformed
vitamin A; vegetable sources of provitamin A carotenoids include
dark green and deeply colored fruits and vegetables. Moderate cook­
ing of vegetables enhances carotenoid release for uptake in the gut.
Carotenoid absorption is also aided by some fat in a meal. Exclusive
breast-feeding can cover the vitamin A needs of infants if the mother
has an adequate vitamin A status and a large enough volume of milk.
If the nursing mother has inadequate vitamin A intake or concomitant
diseases or her infant was a preterm delivery, breast milk probably
will not supply enough vitamin A to prevent deficiency. In developing
countries, chronic dietary deficiency is the main cause of vitamin A
deficiency and is exacerbated by infection. In early childhood, low vita­
min A status results from inadequate intakes of animal food sources
and edible oils, both of which are expensive, coupled with seasonal
unavailability of vegetables and fruits and lack of marketed fortified
food products. Factors that interfere with vitamin A metabolism may
also affect status or function. For example, concurrent zinc deficiency
can interfere with the mobilization of vitamin A from liver stores. Alco­
hol interferes with the conversion of retinol to retinaldehyde in the eye
by competing for alcohol (retinol) dehydrogenase. Drugs that interfere
with the absorption of vitamin A include mineral oil, neomycin, and
bile acid sequestrants (e.g., cholestyramine).
Deficiency 
Vitamin A deficiency is endemic in areas where diets
are chronically poor, especially in southern Asia, sub-Saharan Africa,
some parts of Latin America, and the western Pacific, including parts
of China. Vitamin A status is usually assessed by measuring serum
retinol (normal range, 1.05–3.50 μmol/L [30–100 μg/dL]) or via doseresponse tests or tests of dark adaptation. To assure a correct biochemi­
cal assessment of vitamin A status, a simultaneous assessment of the
inflammatory status is needed (in analogy to the assessment of iron
status); not doing so may result in an overestimation of vitamin A
deficiency. Correction factors to adjust the measured plasma vitamin
A levels to account for the influence of C-reactive protein and α1-acid
glycoprotein are available. Stable isotopic or invasive liver biopsy meth­
ods are available to estimate total-body stores of vitamin A. As judged
by deficient serum retinol (<0.70 μmol/L [20 μg/dL]), vitamin A
deficiency worldwide is present in 190 million preschool-age children,
among whom >5 million have an ocular manifestation of deficiency
termed xerophthalmia. This condition includes milder stages of night
blindness and conjunctival xerosis (dryness) with Bitot’s spots (white
patches of keratinized epithelium appearing on the sclera) that may
affect 1–5% of children in deficient populations as well as rare, poten­
tially blinding corneal ulceration and necrosis. Keratomalacia (soften­
ing of the cornea) leads to corneal scarring that blinds an estimated
quarter of a million children each year and is associated with fatality
rates of 4–25%. However, vitamin A deficiency severe enough to cause
any clinical stage poses an increased risk of death from diarrhea, dys­
entery, measles, malaria, or respiratory disease. This is because vitamin
A deficiency can compromise barrier, innate, and acquired immune
defenses to infection. In areas where deficiency is widely prevalent,
vitamin A supplementation can markedly reduce the risk of childhood
mortality (by 23–34%, on average). About 10% of pregnant women
in undernourished settings also develop night blindness (assessed by
history) during the latter half of pregnancy; this level of moderate
to severe vitamin A deficiency is associated with an increased risk of
maternal infection and death. Maternal vitamin A deficiency may also
exacerbate already low vitamin A nutrition and associated risks for
the newborn. In South Asia, where maternal deficiency is prominent,
giving infants a single oral dose (50,000 IU) of vitamin A shortly after
birth has reduced infant mortality by ≥10%, whereas in African set­
tings less affected by maternal vitamin A deficiency, no effect has been
noted, revealing differences in risk of deficiency and benefit of supple­
mentation across regions. However, the World Health Organization
does not recommend high-dose supplementation to newborns.
TREATMENT
Vitamin A Deficiency
Vitamin A is commercially available for treatment and prevention
in esterified forms (e.g., acetate, palmitate), which are more stable
than other forms. Any stage of xerophthalmia should be treated
with 60 mg (or RAE) or 200,000 IU of vitamin A in oily solution,
usually contained in a soft-gel capsule. The same dose is repeated
1 and 14 days later. Doses should be reduced by half for patients
6–11 months of age. Mothers with night blindness or Bitot’s spots
should be given vitamin A orally 3 mg daily for at least 3 months.
These regimens are efficacious, and they are far less expensive and
more widely available than injectable water-miscible vitamin A. A
common approach to prevention is to provide vitamin A supple­
mentation every 4–6 months to young children 6 months to 5 years
of age (both HIV-positive and HIV-negative) in high-risk areas.
For prevention, infants 6–11 months of age should receive 30 mg of
vitamin A; children 12–59 months of age should receive 60 mg. For
reasons that are not clear, although early neonatal vitamin A may
reduce infant mortality, vitamin A given between 1 and 5 months
of age has not proven effective in improving survival in high-risk
settings.
CHAPTER 344
Vitamin and Trace Mineral Deficiency and Excess
Uncomplicated vitamin A deficiency is rare in industrialized
countries. One high-risk group—extremely low-birth-weight
(<1000-g) infants—is likely to be vitamin A deficient and should
receive a supplement of 1500 μg (or RAE) three times a week for 4
weeks. Severe measles in any society can lead to secondary vitamin
A deficiency. Children hospitalized with measles should receive
two 60-mg doses of vitamin A on 2 consecutive days. Vitamin A
deficiency most often occurs in patients with malabsorptive dis­
eases (e.g., celiac sprue, short-bowel syndrome) who have abnormal
dark adaptation or symptoms of night blindness without other
ocular changes. Typically, such patients are diagnosed in advanced
care settings where they are treated for 1 month with 15 mg/d of a
water-miscible preparation of vitamin A. This treatment is followed
by a lower maintenance dose, with the exact amount determined
by monitoring serum retinol. Finding application elsewhere in
medicine, retinoic acid is useful in the treatment of promyelocytic
leukemia (Chap. 109) and also is used in the treatment of cystic
acne because it inhibits keratinization, decreases sebum secretion,
and possibly alters the inflammatory reaction (Chap. 60).
No specific signs or symptoms result from carotenoid deficiency.
It was postulated that β-carotene would be an effective chemopre­
ventive agent for cancer because numerous epidemiologic studies
had shown that diets high in β-carotene were associated with lower
incidences of cancers of the respiratory and digestive systems.
However, intervention studies in smokers found that treatment with
high doses of β-carotene actually resulted in more lung cancers than
did treatment with placebo. Non–provitamin A carotenoids such
as lutein and zeaxanthin have been suggested to confer protection
against macular degeneration, and one large-scale intervention
study did not show a beneficial effect except in those with a low
lutein status. The use of the non–provitamin A carotenoid lycopene
to protect against prostate cancer has been proposed. However, the
effectiveness of these agents has not been proved by intervention
studies, and the mechanisms underlying these purported biologic
actions are unknown.
Selective plant-breeding techniques that lead to a higher provi­
tamin A carotenoid content in staple foods may decrease vitamin A
malnutrition in low-income countries. Moreover, a recently devel­
oped genetically modified food (Golden Rice) had a β-carotene–to–
vitamin A conversion ratio of ~3:1 in children.
Toxicity 
The acute toxicity of vitamin A was first noted in Arctic
explorers who ate polar bear liver and has also been seen after admin­
istration of 150 mg to adults or 100 mg to children. Acute toxicity is
manifested by increased intracranial pressure, vertigo, diplopia, bulg­
ing fontanels (in children), seizures, and exfoliative dermatitis; it may
result in death. Among children being treated for vitamin A deficiency
according to the protocols outlined above, transient bulging of fon­
tanels occurs in 2% of infants, and transient nausea, vomiting, and
headache occur in 5% of preschoolers. Chronic vitamin A intoxication
is largely a concern in industrialized countries and has been seen in
otherwise healthy adults who ingest 15 mg/d and children who ingest
6 mg/d over a period of several months. Manifestations include dry skin,
cheilosis, glossitis, vomiting, alopecia, bone demineralization and pain,
hypercalcemia, lymph node enlargement, hyperlipidemia, amenorrhea,
and features of pseudotumor cerebri with increased intracranial pres­
sure and papilledema. Liver fibrosis with portal hypertension may also
result from chronic vitamin A intoxication. Provision of vitamin A in
excess to pregnant women has resulted in spontaneous abortion and
in congenital malformations, including craniofacial abnormalities and
valvular heart disease. In pregnancy, the daily dose of vitamin A should
not exceed 3 mg. Also, topical retinoids should be avoided during
pregnancy. Commercially available retinoid derivatives are also toxic,
including 13-cis-retinoic acid, which has been associated with birth
defects. Thus, contraception should be continued for at least 1 year and
possibly longer in women who have taken 13-cis-retinoic acid.
PART 10
Disorders of the Gastrointestinal System
In malnourished children, vitamin A supplements (30–60 mg), in
amounts calculated as a function of age and given in several rounds
over 2 years, are considered to amplify nonspecific effects of vaccines.
However, for unclear reasons, in one African setting, there has been a
negative effect on mortality rates in incompletely vaccinated girls.
High doses of supplemental carotenoids do not result in toxic symp­
toms but should be avoided in smokers due to an increased risk of lung
cancer. Very high doses of β-carotene (~200 mg/d) have been used to
treat or prevent the skin rashes of erythropoietic protoporphyria. Caro­
tenemia, which is characterized by a yellowing of the skin (in creases
of the palms and soles) but not the sclerae, may follow ingestion
of >30 mg of β-carotene daily. Hypothyroid patients are particularly
susceptible to the development of carotenemia due to impaired break­
down of carotene to vitamin A. Reduction of carotenes in the diet
results in the disappearance of skin yellowing and carotenemia over a
period of 30–60 days.
■
■VITAMIN D
The metabolism of the fat-soluble vitamin D is described in detail
in Chap. 421. The biologic effects of this vitamin are mediated by
vitamin D receptors, which are found in most tissues; binding with
these receptors potentially expands vitamin D actions to many differ­
ent cell systems and organs (e.g., immune cells, brain, breast, colon,
and prostate) in addition to the classic endocrine effects on calcium
and phosphate metabolism and bone health. Vitamin D is thought to
be important for maintaining normal function of many nonskeletal
tissues such as muscle (including heart muscle), for immune func­
tion, and for inflammation as well as for cell proliferation and differ­
entiation. Older studies have shown that vitamin D may be useful as
adjunctive treatment for tuberculosis, psoriasis, and multiple sclerosis
or for the prevention of certain cancers. Vitamin D insufficiency may
increase the risk of type 1 diabetes mellitus, cardiovascular disease
(insulin resistance, hypertension, or low-grade inflammation), or brain
dysfunction (e.g., depression). However, the exact physiologic roles of
vitamin D in these nonskeletal diseases and the importance of these
roles have so far not been clarified. Recent placebo-controlled studies
did not show a therapeutic benefit of vitamin D for cancer prevention,
control of cardiovascular disease, or risk of type 2 diabetes, depression,
tuberculosis infection, or other respiratory infections. Presently, it is
not known whether these effects of vitamin D supplements (with or
without calcium) might be different according to the baseline status
(normal vs severely deficient) of patients.
The skin is a major source of vitamin D, which is synthesized upon
skin exposure to ultraviolet B radiation (UV-B; wavelength, 290–320 nm).
Except for fish, food (unless fortified) contains only limited amounts of
vitamin D. Vitamin D2 (ergocalciferol) is obtained from plant sources
and is the chemical form found in some supplements.
Deficiency 
Vitamin D status is assessed by measuring serum levels
of 25-dihydroxyvitamin D (25[OH] vitamin D); however, there is no
consensus on a uniform assay, on optimal serum levels, or on the real
benefit of biochemical screening in asymptomatic adults. The optimal
level might, in fact, differ according to the targeted disease entity. Epi­
demiologic and experimental data indicate that a 25(OH) vitamin D
level of >20 ng/mL (≥50 nmol/L; to convert ng/mL to nmol/L, multiply
by 2.496) is sufficient for good bone health. The latter 25(OH) vitamin
D plasma concentration would cover the requirements of 97.5% of
the population. Some experts, however, advocate higher serum levels
(e.g., >30 ng/mL) for other desirable endpoints of vitamin D action.
There is insufficient evidence to recommend combined vitamin D and
calcium supplementation as a primary preventive strategy (as opposed
to secondary prevention) for reduction of the incidence of fractures in
healthy men and premenopausal women.
Risk factors for vitamin D deficiency are old age, lack of sun expo­
sure, dark skin (especially among residents of northern latitudes), fat
malabsorption, and obesity; deficiency can also occur after gastric
bypass surgery. In addition, in African populations, the prevalence of
vitamin D deficiency might be high (especially in women, newborn
babies, urban populations, and those living in northern African coun­
tries). Rickets represents the classic disease of vitamin D deficiency.
Signs of deficiency are muscle soreness, weakness, and bone pain.
Some of these effects are independent of calcium intake. To prevent
glucocorticoid-induced osteoporosis, treatment with calcium (1000–
1200 mg/d) and vitamin D (600–800 IU/d) through diet and/or supple­
ments in combination with weight-bearing exercise is recommended.
The U.S. National Academy of Sciences recently advised that the
majority of adult North Americans should receive 600 IU/d of vitamin
D (RDA = 15 μg/d or 600 IU/d; Chap. 343). However, for people aged
70 years, the RDA is set at 20 μg/d (800 IU/d). The consumption of
fortified or enriched foods as well as suberythemal sun exposure should
be encouraged for people at risk for vitamin D deficiency. If adequate
intake is impossible, vitamin D supplements should be taken, espe­
cially during the winter months. Vitamin D deficiency can be treated
by oral administration of 50,000 IU/week for 6–8 weeks followed by a
maintenance dose of 800 IU/d (20 μg/d) from food and supplements
once normal plasma levels have been attained. There is still uncertainty
regarding the optimal therapeutic dosage (high vs low) for elderly at
risk of falls. The physiologic effects of vitamin D2 and vitamin D3 are
similar when these vitamins are ingested over long periods.
Toxicity 
The upper limit of intake has been set at 4000 IU/d. Con­
trary to earlier beliefs, acute vitamin D intoxication is rare and usually
is caused by the uncontrolled and excessive ingestion of supplements or
by faulty food fortification practices. High plasma levels of 1,25(OH)2
vitamin D and calcium are central features of toxicity and mandate
discontinuation of vitamin D and calcium supplements; in addition,
treatment of hypercalcemia may be required.
■
■VITAMIN E
Vitamin E is the collective designation for all stereoisomers of tocoph­
erols and tocotrienols, although only the α-tocopherols meet human
requirements. Vitamin E acts as a chain-breaking antioxidant and
is an efficient peroxyl radical scavenger that protects low-density
lipoproteins and polyunsaturated fats in membranes from oxidation.
A network of other antioxidants (e.g., vitamin C, glutathione) and
enzymes maintains vitamin E in a reduced state. Vitamin E also inhib­
its prostaglandin synthesis and the activities of protein kinase C and
phospholipase A2.
Absorption and Metabolism 
After absorption, vitamin E is
taken up from chylomicrons by the liver, and a hepatic α-tocopherol
transport protein mediates intracellular vitamin E transport and incor­
poration into very-low-density lipoprotein. The transport protein has
a particular affinity for the RRR isomeric form of α-tocopherol; thus,
this natural isomer has the most biologic activity.
Requirement 
Vitamin E is widely distributed in the food supply,
with particularly high levels in sunflower oil, safflower oil, and wheat
germ oil; γ-tocotrienols are notably present in soybean and corn oils.
Vitamin E is also found in meats, nuts, and cereal grains, and small
amounts are present in fruits and vegetables. Vitamin E pills contain­
ing doses of 50–1000 mg are ingested by ~10% of the U.S. population.
The RDA for vitamin E is 15 mg/d (34.9 μmol or 22.5 IU) for all adults.
Diets high in polyunsaturated fats may necessitate a slightly higher
intake of vitamin E.
Dietary deficiency of vitamin E does not exist in developed coun­
tries but can occur in developing countries due to inadequate intake.
Vitamin E deficiency is seen only in severe and prolonged malabsorp­
tive diseases, such as celiac disease, chronic cholestatic liver disease,
or after small-intestinal resection or bariatric surgery. Children with
cystic fibrosis or prolonged cholestasis may develop vitamin E defi­
ciency characterized by areflexia and hemolytic anemia. Children
with abetalipoproteinemia cannot absorb or transport vitamin E and
become deficient quite rapidly. A familial form of isolated vitamin E
deficiency also exists; it is due to a defect in the α-tocopherol transport
protein. Vitamin E deficiency causes axonal degeneration of the large
myelinated axons and results in posterior column and spinocerebellar
symptoms. Peripheral neuropathy is initially characterized by are­
flexia, with progression to an ataxic gait, and by decreased vibration
and position sensations. Ophthalmoplegia, skeletal myopathy, and
pigmented retinopathy may also be features of vitamin E deficiency. A
deficiency of either vitamin E or selenium in the host has been shown
to increase certain viral mutations and, therefore, virulence. The labo­
ratory diagnosis of vitamin E deficiency is based on low blood levels
of α-tocopherol (<5 μg/mL, or <0.8 mg of α-tocopherol per gram of
total lipids).
TREATMENT
Vitamin E Deficiency
Symptomatic vitamin E deficiency should be treated with 800–1200 mg
of α-tocopherol per day. Patients with abetalipoproteinemia may
need as much as 5000–7000 mg/d. Children with symptomatic vitamin E
deficiency should be treated orally with water-miscible esters
(400 mg/d); alternatively, 2 mg/kg per d may be administered intra­
muscularly. Vitamin E in high doses may protect against oxygeninduced retrolental fibroplasia and bronchopulmonary dysplasia as
well as intraventricular hemorrhage of prematurity. Vitamin E has
been suggested to increase sexual performance, treat intermittent
claudication, and slow the aging process, but convincing evidence
for these properties is lacking. When given in combination with
other antioxidants, vitamin E may help prevent macular degenera­
tion. Vitamin E may have favorable therapeutic effects in noncir­
rhotic nondiabetic patients with nonalcoholic steatohepatitis. High
doses (60–800 mg/d) of vitamin E have been shown in controlled
trials to improve parameters of immune function and reduce colds
in nursing home residents, but intervention studies using vitamin E
to prevent cardiovascular disease or cancer have not shown efficacy,
and at doses >400 mg/d, vitamin E may even increase all-cause
mortality rates and prostate cancer risk (especially in combination
with selenium supplements).
Toxicity 
All forms of vitamin E are absorbed and could contribute
to toxicity; however, the toxicity risk seems to be rather low as long as
liver function is normal. High doses of vitamin E (>800 mg/d) may
reduce platelet aggregation and interfere with vitamin K metabolism
and are therefore contraindicated in patients taking warfarin and anti­
platelet agents (such as aspirin or clopidogrel). Nausea, flatulence, and
diarrhea have been reported at doses >1 g/d.
■
■VITAMIN K
There are two natural forms of vitamin K: vitamin K1, also known as
phylloquinone, from vegetable sources, and vitamin K2, or menaqui­
nones, which are synthesized by bacterial flora and found in hepatic
tissue. Phylloquinone can be converted to menaquinone in some
organs.
Vitamin K is required for the posttranslational carboxylation of glu­
tamic acid, which is necessary for calcium binding to γ-carboxylated
proteins such as prothrombin (factor II); factors VII, IX, and X; pro­
tein C; protein S; and proteins found in bone (osteocalcin) and vascu­
lar smooth muscle (e.g., matrix Gla protein). However, the importance
of vitamin K for bone mineralization and prevention of vascular calcifi­
cation in different patient groups (including chronic kidney disease) is
unclear. Warfarin-type drugs inhibit γ-carboxylation by preventing the
conversion of vitamin K to its active hydroquinone form.
Dietary Sources 
Vitamin K is found in green leafy vegetables such
as kale and spinach, and appreciable amounts are also present in mar­
garine and liver. Vitamin K is present in vegetable oils; olive, canola,
and soybean oils are particularly rich sources. The average daily intake
by Americans is estimated to be ~100 μg/d.
CHAPTER 344
Deficiency 
The symptoms of vitamin K deficiency are due to hem­
orrhage; newborns are particularly susceptible because of low fat stores,
low breast milk levels of vitamin K, relative sterility of the infantile
intestinal tract, liver immaturity, and poor placental transport. Intra­
cranial bleeding as well as gastrointestinal and skin bleeding can occur
in vitamin K–deficient infants 1–7 days after birth. Thus, vitamin K
(0.5–1 mg IM) is given prophylactically at delivery.
Vitamin and Trace Mineral Deficiency and Excess
Vitamin K deficiency in adults may be seen in patients with chronic
small-intestinal disease (e.g., celiac disease, Crohn’s disease), in those
with obstructed biliary tracts, or after small-bowel resection. Broadspectrum antibiotic treatment can precipitate vitamin K deficiency by
reducing numbers of gut bacteria, which synthesize menaquinones,
and by inhibiting the metabolism of vitamin K. In patients with
warfarin therapy, the antiobesity drug orlistat can lead to changes in
international normalized ratio due to vitamin K malabsorption. The
assessment of the vitamin K status can be done by measurement of
phylloquinone (vitamin K1) concentration in serum (deficiency
<0.15 μg/L); the cellular utilization of vitamin K can be assessed by
the serum or plasma concentration of undercarboxylated prothrombin
(protein induced by vitamin K absence/antagonism [PIVKA-II]). An
elevated prothrombin time or activated partial thromboplastin time
or reduced clotting factors are useful markers in severe deficiency but
are otherwise nonspecific and lack sensitivity. Vitamin K deficiency is
treated with a parenteral dose of 10 mg. For patients with chronic mal­
absorption, 1–2 mg/d should be given orally or 1–2 mg per week can
be taken parenterally. Patients with liver disease may have an elevated
prothrombin time because of liver cell destruction as well as vitamin K
deficiency. If an elevated prothrombin time does not improve during
vitamin K therapy, it can be deduced that this abnormality is not the
result of vitamin K deficiency.
Toxicity 
Toxicity from dietary phylloquinones and menaquinones
has not been described. High doses of vitamin K can impair the actions
of oral vitamin K antagonist anticoagulants.
MINERALS
See also Table 344-2.
■
■CALCIUM
See Chap. 421.
TABLE 344-2  Deficiencies and Toxicities of Metals
ELEMENT
DEFICIENCY
TOXICITY
Boron
No biologic function determined
Developmental defects, male sterility, testicular atrophy
20 mg/d (extrapolated from animal data)
Calcium
Reduced bone mass, osteoporosis
Renal insufficiency (milk-alkali syndrome),
nephrolithiasis, impaired iron absorption, thiazide
diuretics
Copper
Anemia, growth retardation, defective
keratinization and pigmentation of hair,
hypothermia, degenerative changes in aortic
elastin, osteopenia, mental deterioration
Nausea, vomiting, diarrhea, hepatic failure, tremor,
mental deterioration, hemolytic anemia, renal
dysfunction
Chromium
Impaired glucose tolerance
Occupational: Renal failure, dermatitis, pulmonary
cancer
Fluoride
↑ Dental caries
Dental and skeletal fluorosis, osteosclerosis
10 mg/d (fluorosis)
Iodine
Thyroid enlargement, ↓ T4, cretinism
Thyroid dysfunction, acne-like eruptions
1100 μg/d (thyroid dysfunction)
Iron
Muscle abnormalities, koilonychia, pica,
anemia, ↓ work performance, impaired cognitive
development, premature labor, ↑ perinatal
maternal death
Gastrointestinal effects (nausea, vomiting, diarrhea,
constipation), iron overload with organ damage, acute
and chronic systemic toxicity, increased susceptibility to
malaria, increased risk association with certain chronic
diseases (e.g., diabetes)
Manganese
Impaired growth and skeletal development,
reproduction, lipid and carbohydrate metabolism;
upper body rash
General: Neurotoxicity, Parkinson-like symptoms
Occupational: Encephalitis-like syndrome, Parkinson-like
syndrome, psychosis, pneumoconiosis
Molybdenum
Severe neurologic abnormalities
Reproductive and fetal abnormalities
2 mg/d (extrapolated from animal data)
Selenium
Cardiomyopathy, heart failure, striated muscle
degeneration
General: Alopecia, nausea, vomiting, abnormal nails,
emotional lability, peripheral neuropathy, lassitude, garlic
odor to breath, dermatitis
Occupational: Lung and nasal carcinomas, liver necrosis,
pulmonary inflammation
PART 10
Disorders of the Gastrointestinal System
Phosphorus
Rickets (osteomalacia), proximal muscle
weakness, rhabdomyolysis, paresthesia, ataxia,
seizure, confusion, heart failure, hemolysis,
acidosis
Hyperphosphatemia
4000 mg/d
Zinc
Growth retardation, ↓ taste and smell, alopecia,
dermatitis, diarrhea, immune dysfunction,
failure to thrive, gonadal atrophy, congenital
malformations
General: Reduced copper absorption, gastritis, sweating,
fever, nausea, vomiting
Occupational: Respiratory distress, pulmonary fibrosis
■
■ZINC
Zinc is an integral component of many metalloenzymes in the body;
it is involved in the synthesis and stabilization of proteins, DNA, and
RNA and plays a structural role in ribosomes and membranes. Zinc
is necessary for the binding of steroid hormone receptors and several
other transcription factors to DNA. Zinc is essential for normal sper­
matogenesis, fetal growth, and embryonic development.
Absorption 
The absorption of zinc from the diet is inhibited by
dietary phytate, fiber, oxalate, iron, and copper as well as by certain
drugs, including penicillamine, sodium valproate, and ethambutol.
Protein-containing foods, i.e., meat, shellfish, nuts, and legumes, are
good sources of bioavailable zinc, whereas zinc in grains and legumes
is less available for absorption. Grains and legumes contain phytate that
binds zinc in the intestine and reduces its availability for absorption.
Deficiency 
Mild zinc deficiency has been described in many dis­
eases, including diabetes mellitus, HIV/AIDS, cirrhosis, alcoholism,
inflammatory bowel disease, malabsorption syndromes, and sickle
cell disease. In these diseases, mild chronic zinc deficiency can cause
stunted growth in children, decreased taste sensation (hypogeusia), and
impaired immune function. Severe chronic zinc deficiency has been
described as a cause of hypogonadism and dwarfism in several Middle
Eastern countries. In these children, hypopigmented hair is also part of
the syndrome. Acrodermatitis enteropathica is a rare autosomal reces­
sive disorder characterized by abnormalities in zinc absorption. Clinical
manifestations include diarrhea, alopecia, muscle wasting, depression,
irritability, and a rash involving the extremities, face, and perineum.
The rash is characterized by vesicular and pustular crusting with scaling
and erythema. Occasional patients with Wilson’s disease have developed
zinc deficiency as a consequence of penicillamine therapy (Chap. 427).
TOLERABLE UPPER (DIETARY)
INTAKE LEVEL
2500 mg/d (milk-alkali)
10 mg/d (liver toxicity)
Not determined
45 mg/d of elemental iron
(gastrointestinal side effects)
11 mg/d (neurotoxicity)
400 μg/d (hair, nail changes)
40 mg/d (impaired copper metabolism)
Zinc deficiency is prevalent in many developing countries and usu­
ally coexists with other micronutrient deficiencies (especially iron defi­
ciency). Zinc (20 mg/d until recovery) may be an effective adjunctive
therapeutic strategy for diarrheal disease and pneumonia in children
≥6 months of age.
The diagnosis of zinc deficiency is usually based on a serum zinc
level <12 μmol/L (<70 μg/dL). Pregnancy and birth control pills may
cause a slight depression in serum zinc levels, and hypoalbuminemia
from any cause can result in hypozincemia. In acute stress situations
(illness, but also postexercise recovery), zinc may be redistributed
from serum into tissues. Zinc deficiency may be treated with 60 mg
of elemental zinc taken by mouth twice a day. Zinc gluconate lozenges
(13 mg of elemental zinc every 2 h while awake) have been reported to
reduce the duration and symptoms of the common cold in adults, but
study results are conflicting.
Toxicity 
Acute zinc toxicity after oral ingestion causes nausea,
vomiting, and fever. Zinc fumes from welding may also be toxic and
cause fever, respiratory distress, excessive salivation, sweating, and
headache. Chronic large doses of zinc (ranging from 150 to 450 mg/d)
may depress immune function and cause hypochromic anemia as a
result of a secondary copper deficiency. Intranasal zinc preparations
should be avoided because they may lead to irreversible damage of the
nasal mucosa and anosmia.
■
■COPPER
Copper is an integral part of numerous enzyme systems, including
amine oxidases, ferroxidase (ceruloplasmin), cytochrome c oxidase,
superoxide dismutase, and dopamine hydroxylase. Copper is also a
component of ferroprotein, a transport protein involved in the baso­
lateral transfer of iron during absorption from the enterocyte. As such,

16 - 345 Malnutrition and Nutritional Assessment
345 Malnutrition and Nutritional Assessment
copper plays a role in iron metabolism, melanin synthesis, energy
production, neurotransmitter synthesis, and CNS function; the syn­
thesis and cross-linking of elastin and collagen; and the scavenging of
superoxide radicals. Dietary sources of copper include shellfish, liver,
nuts, legumes, bran, and organ meats.
Deficiency 
Dietary copper deficiency is relatively rare, although
it has been described in premature infants who are fed milk diets
and in infants with malabsorption (Table 344-2). Copper-deficiency
anemia (refractory to therapeutic iron) has been reported in patients
with malabsorptive diseases and nephrotic syndrome and in patients
treated for Wilson’s disease with chronic high doses of oral zinc, which
can interfere with copper absorption. Menkes kinky hair syndrome is
an X-linked metabolic disturbance of copper metabolism character­
ized by intellectual disability, hypocupremia, and decreased circulating
ceruloplasmin (Chap. 425). This syndrome is caused by mutations in
the copper-transporting ATP7A gene. Children with this disease often
die within 5 years because of dissecting aneurysms or cardiac rupture.
Aceruloplasminemia is a rare autosomal recessive disease character­
ized by tissue iron overload, mental deterioration, microcytic anemia,
and low serum iron and copper concentrations.
The diagnosis of copper deficiency is usually based on low serum
levels of copper (<65 μg/dL) and low ceruloplasmin levels (<20 mg/
dL). Serum levels of copper may be elevated in pregnancy or stress con­
ditions since ceruloplasmin is an acute-phase reactant and 90% of cir­
culating copper is bound to ceruloplasmin. It has been suggested that
mild or subclinical copper deficiency is more common than expected;
at-risk individuals include patients with cholestasis or chronic diar­
rheal diseases, dialysis patients, and people on long-term zinc supple­
ments. The role of copper in cardiovascular disease, immune function,
bone health, or neurodegenerative diseases is still unclear.
Toxicity 
Copper toxicity is usually accidental (Table 344-2). In
severe cases, kidney failure, liver failure, and coma may ensue. In Wil­
son’s disease, mutations in the copper-transporting ATP7B gene lead
to accumulation of copper in the liver and brain, with low blood levels
due to decreased ceruloplasmin (Chap. 427). A potential negative role
of copper in the pathogenesis of Alzheimer’s disease has been reported.
■
■SELENIUM
Selenium, in the form of selenocysteine, is a component of the enzyme
glutathione peroxidase, which serves to protect proteins, cell mem­
branes, lipids, and nucleic acids from oxidant molecules. As such,
selenium is being actively studied as a chemopreventive agent against
certain cancers, such as prostate cancer. However, it remains unclear
whether selenium is effective as a chemopreventive agent or whether
it increases cancer risk (e.g., prostate cancer). Convincing evidence for
a protective effect of selenium on cognitive decline or cardiovascular
disease risk is presently lacking. Selenocysteine is also found in the
deiodinase enzymes, which mediate the deiodination of thyroxine to
triiodothyronine (Chap. 394). Rich dietary sources of selenium include
seafood, muscle meat, and cereals, although the selenium content of
cereal is determined by the soil concentration. Countries with low soil
concentrations include parts of Scandinavia, China, and New Zealand.
Keshan disease is an endemic cardiomyopathy found in children and
young women residing in regions of China where dietary intake of
selenium is low (<20 μg/d). Concomitant deficiencies of iodine and
selenium may worsen the clinical manifestations of cretinism. Chronic
ingestion of large amounts of selenium leads to selenosis, character­
ized by hair and nail brittleness and loss, garlic breath odor, skin rash,
myopathy, irritability, and other abnormalities of the nervous system.
■
■CHROMIUM
Chromium potentiates the action of insulin in patients with impaired
glucose tolerance, presumably by increasing insulin receptor–medi­
ated signaling, although its usefulness in treating type 2 diabetes is
uncertain. In addition, improvement in blood lipid profiles has been
reported in some patients. The usefulness of chromium supplements
in muscle building has not been substantiated. Rich food sources of
chromium include yeast, meat, and grain products. Chromium in the
trivalent state is found in supplements and is largely nontoxic; however,
chromium-6 is a product of stainless steel welding and is a known pul­
monary carcinogen as well as a cause of liver, kidney, and CNS damage.
■
■MAGNESIUM
See Chap. 421.
■
■FLUORIDE, MANGANESE, AND ULTRATRACE
ELEMENTS
An essential function for fluoride in humans has not been described,
although it is useful for the maintenance of structure in teeth and
bones. Adult fluorosis results in mottled and pitted defects in tooth
enamel as well as brittle bone (skeletal fluorosis).
Manganese and molybdenum deficiencies have been reported in
patients with rare genetic abnormalities and in a few patients receiv­
ing prolonged total parenteral nutrition. Several manganese-specific
enzymes have been identified (e.g., manganese superoxide dismutase).
Deficiencies of manganese have been reported to result in bone demin­
eralization, poor growth, ataxia, disturbances in carbohydrate and lipid
metabolism, and convulsions.
Ultratrace elements are defined as those needed in amounts <1 mg/d.
Essentiality has not been established for most ultratrace elements,
although selenium, chromium, and iodine are clearly essential
(Chap. 394). Molybdenum is necessary for the activity of sulfite and
xanthine oxidase, and molybdenum deficiency may result in skeletal
and brain lesions.
■
■FURTHER READING
Combs GF Jr, Mcclung JP: The Vitamins: Fundamental Aspects in
CHAPTER 345
Nutrition and Health, 6th ed. London, Academic Press, 2022.
DeAngelo SL et al: Selenoproteins and tRNA-Sec: Regulators of can­
cer redox homeostasis. Trends Cancer 9:1006, 2023.
Fan D et al: Cell signaling pathways based on vitamin C and their appli­
cation in cancer therapy. Biomed Pharmacother 162:114695, 2023.
Imdad A et al: Vitamin A supplementation for preventing morbid­
Malnutrition and Nutritional Assessment
ity and mortality in children from six months to five years of age.
Cochrane Database Syst Rev 3:CD008524, 2022.
Lassi ZS et al: Zinc supplementation for the promotion of growth
and prevention of infections in infants less than six months of age.
Cochrane Database Syst Rev 4:CD010205, 2020.
Mechanick JI et al: Clinical practice guidelines for the perioperative
nutrition, metabolic, and nonsurgical support of patients undergoing
bariatric procedures–2019 update. Surg Obes Relat Dis 16:175, 2020.
Ota Y et al: Comprehensive review of Wernicke encephalopathy:
Pathophysiology, clinical symptoms and imaging findings. Jpn J
Radiol 38:809, 2020.
Staub E et al: Enteral zinc supplementation for prevention of morbid­
ity and mortality in preterm neonates. Cochrane Database Syst Rev
3:CD012797, 2021.
Charles Chin Han Lew, Charlene W. Compher
Malnutrition and
Nutritional Assessment
Malnutrition occurs in 30–50% of hospitalized patients depending
on the setting, the patient’s diagnosis, and the criteria that are used to
diagnose malnutrition. Notable adverse outcomes associated with mal­
nutrition include poor wound healing, compromised immune status,
and impaired organ function. These can lead to increased length of
hospital stay, readmissions, higher mortality, and increased health care
costs. It is now widely appreciated that acute or chronic inflammation
At risk for malnutrition
Use validated screening tools
Risk screening
Diagnostic
Assessment
Assessment criteria
Phenotypic
Non-volitional weight loss
Low body mass index
Reduced muscle mass
Etiologic
Reduced food intake or assimilationa
Disease burden/inflammatory conditionb
Diagnosis
Meets criteria for malnutrition diagnosis
Requires at least 1 Phenotypic criterion and
1 Etiologic criterion
Severity
Grading
Determine severity of malnutrition
Severity determined based on Phenotypic
criterion
FIGURE 345-1  Global Leadership Initiative on Malnutrition (GLIM) diagnostic
scheme for screening, assessment, diagnosis, and grading of malnutrition. a≤50%
of energy requirement >1 week, or any reduction for >2 weeks, or any chronic
gastrointestinal condition that adversely impacts food assimilation or absorption.
bAcute disease/injury (e.g., major infection, burns, trauma or closed head injury) or
chronic disease (e.g., malignant disease, chronic obstructive pulmonary disease,
congestive heart failure, chronic renal disease or any disease with chronic or
recurrent inflammation) or C-reactive protein may be used as a supportive laboratory
measure (mild inflammation: 3.0-9.9 mg/L, moderate inflammation: 10-50 mg/L, severe
inflammation: >50 mg/L). (Reproduced with permission from Jensen GL et al: GLIM
criteria for the diagnosis of malnutrition: A consensus report from the global clinical
nutrition community. JPEN J Parenter Enteral Nutr 43:32, 2019, Figure 1.)
PART 10
Disorders of the Gastrointestinal System
contributes to the pathophysiology of malnutrition. The presence
of inflammation can also render historic nutrition parameters, like
albumin and prealbumin, unreliable. In patients with high levels of
inflammation, nutrition care is supportive. At moderate or low levels
of inflammation, nutrition care may be therapeutic in reducing the
nutritional deficits and improving clinical outcomes.
■
■MALNUTRITION DIAGNOSIS IN CLINICAL
SETTINGS
Several tools have been used to diagnose malnutrition, such as the Sub­
jective Global Assessment, Mini Nutrition Assessment, and the Acad­
emy of Nutrition and Dietetics and American Society for Parenteral
and Enteral Nutrition Indicators of Malnutrition. These tools use com­
mon parameters such as dietary intake and weight history, muscle and
fat loss, body mass index (BMI), and/or physical function to diagnose
malnutrition. However, nuances between them lead to incomparable
malnutrition prevalence rates in the literature.
TABLE 345-1  Thresholds for Severity Grading of Malnutrition Into Stage 1 (Moderate) and Stage 2 (Severe) Malnutrition
WEIGHT LOSS (%)
LOW BODY MASS INDEX (kg/m2)b
REDUCED MUSCLE MASSc
Stage 1/moderate malnutrition (requires 1
phenotypic criterion that meets this grade)
5%–10% within the past 6 months, or
10%–20% beyond 6 months
Stage 2/severe malnutrition (requires 1
phenotypic criterion that meets this grade)
10% within the past 6 months, or
20% beyond 6 months
aSeverity grading is based on the noted phenotypic criteria, whereas the etiologic criteria described in the text and Figure 345-1 are used to provide the context to guide
intervention and anticipated outcomes. bFurther research is needed to secure consensus reference body mass index for Asian populations in clinical settings. cFor example,
appendicular lean mass index (kg/m2) by dual-energy absorptiometry or corresponding standards using other body composition methods such as bioelectrical impedance
analysis, computed tomography, or magnetic resonance imaging. When not available or by regional preference, physical examination or standard anthropometric measures
such as mid-arm muscle or calf circumferences may be used. Functional assessments such as hand-grip strength may be used as a supportive measure.
Source: Reproduced with permission from Jensen GL et al: GLIM criteria for the diagnosis of malnutrition: A consensus report from the global clinical nutrition community
JPEN J Parenter Enteral Nutr 43:32, 2019, Table 4.
In 2019, four international nutrition societies introduced the Global
Leadership Initiative on Malnutrition (GLIM) consensus criteria for
diagnosing malnutrition in adults (Fig. 345-1). GLIM incorporates
broadly available nutrition parameters that are commonly used in other
diagnostic tools to enable data from these earlier assessment methods
to be described in the GLIM framework. To diagnose malnutrition and
its severity, GLIM requires one etiologic criterion (inflammation/
disease burden or reduced food intake/assimilation) and one phenotypic
criterion (weight loss, low BMI, or reduced muscle mass) (Table 345-1).
To facilitate the diagnosis of malnutrition in settings with variable
availability and expertise of personnel and equipment across clinical
settings, several options are suggested for the phenotypic criterion.
The phenotypic criterion is used to define the severity of malnutrition.
Recent publications highlight GLIM’s widespread adoption in hospitals
and clinical settings, showing predictive value for adverse outcomes.
■
■NUTRITION ASSESSMENT FOR PROTEINCALORIE MALNUTRITION
The Nutrition Care Process in the hospital setting involves screening
for malnutrition risk, comprehensive nutrition assessment of patients
at malnutrition risk, development of a nutrition care plan, and moni­
toring for needed adjustments to the plan. Screening for the risk of
malnutrition is typically undertaken by the bedside nurse or by a clini­
cal dietitian using validated tools such as the Malnutrition Screening
Tool, Malnutrition Universal Screening Tool, Nutritional Risk Screen­
ing 2002, or other in-house tools. A comprehensive nutrition assess­
ment includes a review of medical, surgical, and social history; dietary
intake and weight history; medication profile; available laboratory,
diagnostic tests, body composition, and anthropometric measures; and
nutrition-focused physical examination. The clinical dietitian designs
an individualized intervention and monitoring plan for the patient.
The nutrition assessment findings and care plan are shared with the
physician who may make a malnutrition diagnosis and prescribe the
recommended nutrition care plan. The response to nutrition therapy is
monitored with adjustments to the care plan as needed. Micronutrient
deficiencies of clinical relevance may be detected in association with
malnutrition, but a detailed discussion of their assessment is beyond
the scope of this chapter (see Chap. 344).
In ambulatory clinic settings, physicians may have less access to
dietitian support than in inpatient settings. However, the GLIM criteria
are designed to be feasible even in an outpatient setting. Nurses can
obtain measures of height and weight (for BMI) and calf or arm cir­
cumference to assess muscle mass. A diagnosis of malnutrition can be
made in the clinic with referral of the patient to a dietitian for a detailed
nutrition assessment and development of a care plan.
■
■COMPONENTS OF A DETAILED NUTRITIONAL
ASSESSMENT
Medical/Surgical History and Clinical Diagnosis 
The risk
of malnutrition varies widely across medical/surgical diagnoses and
clinical settings. During the early stages of diseases, malnutrition may
be less likely than with disease progression. Knowledge of a patient’s
PHENOTYPIC CRITERIAa
<20 if <70 years, <22 if ≥70 years
Mild-to-moderate deficit (per validated
assessment methods; see below)
<18.5 if <70 years, <20 if ≥70 years
Severe deficit (per validated
assessment methods; see below)
medical/surgical history and associated clinical diagnoses is especially
helpful in discerning the likelihood of malnutrition and inflamma­
tion. Several conditions or diseases are characterized by severe acute
inflammatory response (e.g., critical illness), whereas others are more
typically associated with a chronic inflammatory response that is mild
to moderate in severity (e.g., chronic cardiometabolic, oncologic, or
gastrointestinal disease, or organ failure) and may be relapsing and
remitting. It is common for acute inflammatory events to be superim­
posed on patients with chronic conditions; for example, a patient with
chronic renal disease may be admitted to the hospital with sepsis, such
that the acute increase in inflammation should be time-limited while
the chronic inflammation will continue.
The inflammatory milieu, especially when severe, may modify
nutrient requirements by elevating resting energy expenditure and
promoting anabolic resistance and catabolism in muscle. Inflamma­
tion also promotes anorexia, decreases food intake, and further com­
promises nutritional status. A deteriorating course may result because
severe inflammation may reduce the benefit of nutritional interven­
tions to being supportive rather than therapeutic, and the associated
malnutrition may diminish the effectiveness of medical therapies.
It is also imperative to recognize medical/surgical conditions that
place patients at increased risk of becoming malnourished because
they have increased nutritional requirements or compromised intake
or assimilation. Chronic gastrointestinal disease, even if relapsing and
remitting, may limit nutrient assimilation. Therefore, patients with
seemingly adequate oral intake may still be at risk of malnutrition due
to compromised nutrient assimilation.
Clinical Signs and Physical Examination 
Nonspecific clinical
indicators of inflammation include fever, hypothermia, and tachy­
cardia. The nutrition-focused physical examination should identify
edema as well as physical signs of weight gain/loss. Physical findings
of weight loss associated with decreased muscle and subcutaneous fat
mass should be noted. If edema is present, an estimation of dry weight
may be needed to identify actual weight loss.
Anthropometric Data 
Nonvolitional weight loss is a well-val­
idated nutrition parameter associated with underlying disease or
inflammation. The degree and duration of nonvolitional weight loss
determine its clinical significance. A 5–10% loss of body weight over
6 months is clinically relevant, whereas >10% loss over the same dura­
tion is severe.
The trend in body weight may be obtained from patients, their
family or caregivers, and the medical record. Obtaining body weight
at each clinic, emergency department, and hospital visit will support
efforts to detect changes in this important parameter. Training staff
members to weigh patients consistently without shoes and heavy outer
garments and maintaining scales and stadiometers in good calibration
facilitate acquisition of trustworthy weight data. Clinical sites may
maintain chair or bed scales for patients who cannot stand. Height may
be estimated by doubling the arm span measurement from the sternal
notch to the end of the longest finger or by measurement of knee height
using a caliper.
BMI, defined as weight (kg)/height (m2), is used primarily to screen
for overweight or obesity. The National Institutes of Health/World
Health Organization BMI categories for adults are as follows: BMI
<18.5 kg/m2 = underweight, BMI 18.5–24.9 kg/m2 = desirable, BMI
25.0–29.9 kg/m2 = overweight, and BMI ≥30 kg/m2 = obese. Note that
patients of any BMI status can be malnourished. Overweight or obese
patients can be malnourished due to weight loss or loss of muscle
mass (e.g., sarcopenic obesity). Similarly, being underweight does not
equate to malnutrition. BMI <18.5 kg/m2 is infrequent in economically
developed countries.
Classical anthropometric measurements such as arm and calf cir­
cumference can be helpful. However, measuring the calf circumfer­
ence may be more accessible as it requires less specialized training in
comparison to measuring the arm circumference. Additionally, calf
circumference has established cutoff values validated for various BMI
ranges and ethnicities. Clinically available body composition mea­
sures include bioelectrical impedance analysis (BIA), dual-energy x-ray
absorptiometry (DXA), and air displacement plethysmography (Table
345-2). BIA has the advantage of being portable and is used for patients
in diverse clinical settings. Abdominal computed tomography (CT)
and magnetic resonance imaging (MRI) taken for disease diagnostic
purposes can be interpreted to measure muscle mass, though they are
not routinely used to aid in the diagnosis of malnutrition.
Laboratory Indicators 
There is, unfortunately, no biomarker
descriptive of comprehensive nutritional status. Laboratory findings
(Table 345-2) should be evaluated in combination with other assess­
ment parameters to appropriately diagnose malnutrition. Albumin and
prealbumin are often available in patients with suspected malnutrition;
however, these measures have poor sensitivity and specificity as indi­
cators of malnutrition. Albumin and prealbumin are negative acutephase reactants with levels that are reduced by the systemic response
to injury, disease, or inflammation. C-reactive protein, a positive acutephase reactant, may be useful to identify the presence of inflammation,
particularly if the clinical context is unclear about the extent of inflam­
mation. Trends in repeated levels of C-reactive protein over time may
help to clarify the extent and trajectory of the inflammatory challenge.
Research suggests that interleukin 6, and perhaps other cytokines, may
offer promise as indicators of inflammatory status. Nonspecific labora­
tory indicators that are often associated with inflammatory response
include leukocytosis and hyperglycemia. Additional tests that may
be obtained to help confirm the presence of inflammatory response
include 24-h urine urea nitrogen and indirect calorimetry to measure
energy expenditure. In the setting of a severe acute systemic inflamma­
tory response, negative nitrogen balance and elevated resting energy
expenditure are anticipated.
CHAPTER 345
Diet History 
Brief diet surveys are available to target specific
aspects of the diet, such as food patterns, to prevent disease or track
intake of nutrients of key importance such as calcium, vitamin D, and
phosphorus for bone disease management. However, more specific
information about the adequacy of intake of key nutrients or food
patterns relative to estimated requirements is most often obtained
by consultation with a clinical dietitian. Since patients often present
to health care practitioners with acute medical events superimposed
upon chronic health conditions, it is common for patients to have had
decreased food intake and progressive malnutrition for extended peri­
ods prior to assessment. Therefore, compromised dietary intake must
not be overlooked so that appropriate intervention may be undertaken.
Malnutrition and Nutritional Assessment
Assessment of the adequacy of nutritional intake must include oral
nutrition supplements, enteral tube feedings, and parenteral nutrition
both during hospitalization and as a component of home nutrition
support. Ongoing reassessment of actual intake received by the patient
is undertaken in hospital settings, most commonly by the clinical
dietitian, the bedside nurse, and the nutrition support team if available.
Changes in the adequacy of oral intake, body weight, or tolerance to
enteral or parenteral feedings would all require adjustment of the feed­
ings to optimize patient care.
A comprehensive assessment should also include a review of medi­
cations, vitamin/mineral supplements, or herbal remedies with atten­
tion to undesirable nutrition-sensitive side effects such as anorexia,
dysgeusia, oral mucositis, nausea, vomiting, diarrhea, and constipation.
Potential drug–nutrient interactions should also be identified to enable
alternative therapeutic options.
Functional Outcomes 
Advanced malnutrition is accompanied
by declines in muscle mass and function that can be detected by
various functional test outlined in Table 345-2. Among them, hand­
grip strength may have better clinical utility since it is easily assessed
and helps to identify which patients benefit most from individualized
nutritional support.
In summary, malnutrition puts patients at risk of adverse outcomes.
Therefore, timely diagnosis and treatment of malnutrition should be
provided to improve patient outcome. Robust evidence exists dem­
onstrating that a multipronged approach to malnutrition intervention
reduces mortality and holds promise to improve the quality of life of
patients.
PART 10
Disorders of the Gastrointestinal System
TABLE 345-2  Common Body Composition Studies, Laboratories, and Other Studies Used in Nutrition Assessment
TEST
NOTES
Body Composition Studies (Recommended)
Air plethysmography
May be used to assess body composition. It comprises a dual-chamber, sealed compartment containing an oscillating
diaphragm, which allows it to measure body volume using Poisson’s law, and compute fat mass and fat-free mass. However,
its validity in a racially diverse population needs further studies, and it is not readily available in most hospitals.
Bioelectrical impedance analysis (BIA)
A promising method to assess body composition as the equipment is easily portable. It uses the opposition of an electrical
current through body tissues (i.e., impedance) and population-specific equations to estimate total body water and body
composition. BIA equations are highly device and population specific. Fever, some medications, and fluid/electrolyte
disturbances can influence its precision and accuracy.
Calf circumference
A practical and valid clinical estimate of skeletal muscle mass, in which low calf circumference indicates muscle loss.
Ethnic specific reference ranges are available, and measurements should be adjusted in persons with body mass index
different from the normal range (underweight, overweight, or obese). The technique requires less training than mid-upper arm
circumference.
Dual energy x-ray absorptiometry
(DXA)
May be used to assess regional and total fat mass in a diverse group of adult clinical patients. However, its validity in
assessing muscle mass is unclear.
Imaging with computed tomography
(CT) or magnetic resonance imaging
(MRI)
May be used to quantify fat and fat-free mass when scans are taken for other diagnostic purposes. Both are costly, and CT
entails x-ray exposure.
Mid-upper arm circumference
Mid-upper arm circumference is the circumference of the upper arm measured at the midpoint between the olecranon
process and the acromion process. It estimates both muscle and subcutaneous fat stores, but references are population
specific. Take note that this measurement technique requires reliability training.
Ultrasound
A promising method to assess body composition as the equipment is easily portable. It uses high-frequency sound waves to
capture live images of muscle tissues. Since most studies examined a single muscle, it is unclear if results can be extrapolated
to reflect overall nutritional status. More research is needed to establish standardization of measurement protocol.
Laboratory Tests and Other Studies (Recommended)
Complete blood count with differential
May be used to screen for nutritional anemias (iron, B12, and folate deficiencies) and thrombocytopenia (vitamin C and folate
deficiencies).
C-reactive protein (CRP)
May be used to confirm systemic inflammation. While inflammation may be associated with malnutrition, CRP lacks specificity
as a biomarker for diagnosing malnutrition. Nevertheless, when combined with other nutrition assessment methods, CRP can
complement the diagnostic process for malnutrition. In addition, elevated levels may be associated with reduced food intake
and a lack of response to nutritional interventions.
Indirect calorimetry
May be used to determine resting energy expenditure (REE) to aid in determining caloric intake goal. Predictive equations that
were based on REE measures in specific populations are also used to estimate REE, especially when indirect calorimetry is
not available.
Nitrogen balance (NB)
May be used to reflect the degree of catabolism and adequacy of protein replacement delivery in patients with normal renal
and liver function. Method requires collection of 24-h urine. Nitrogen balance = (protein delivery [g]/6.25) − (urinary urea
nitrogen + 4 insensible losses).
Specific micronutrients
May be used to validate clinical symptoms of micronutrient deficiencies. Exercise caution with interpretation since levels can
be influenced by acute illness.
Laboratory Tests and Other Studies (Not Recommended)
Blood urea nitrogen (BUN), serum
creatinine
Not recommended as malnutrition indicators. Although low BUN and serum creatinine may reflect reduced muscle mass, their
levels are influenced by factors unrelated to malnutrition (e.g., renal and hepatic insufficiency).
Cholesterol
Not recommended as a malnutrition indicator. While becoming hypocholesterolemic (<120 mg/dL) during hospitalization
may be linked to poorer clinical outcomes, insufficient evidence supports its use for diagnosing malnutrition and evaluating
nutritional interventions.
Creatinine height index (CHI)
Not recommended as a malnutrition indicator. CHI = (24-h urinary creatinine excretion/ideal urinary creatinine for gender and
height) × 100. Although urinary creatinine excretion reflects muscle mass, it can be influenced by renal insufficiency, meat
consumption, physical activity, and trauma. Requires accurate 24-h urine collection.
Cytokines
Not recommended as malnutrition indicators. Their cutoff values and role as an indicator of inflammatory status are still being
studied.
Electrocardiogram
Not recommended as a malnutrition indicator. Although prolonged QT interval may be present in severely malnourished
patients, the former can be influenced by factors unrelated to malnutrition.
Fecal fat test
May be used to validate clinical symptoms of fat malabsorption. It reflects the percentage of dietary fat that the body does not
absorb. Requires stool samples collected over 72 h during ingestion of high-fat diet.
Serum proteins (albumin, prealbumin,
transferrin, and retinol-binding protein)
Not recommended as a malnutrition indicator. Their levels are influenced by factors other than malnutrition (e.g., systemic
inflammation, hepatic and renal insufficiency, protein-losing enteropathies, corticosteroids, hydration, and iron status),
reducing their specificity in diagnosing malnutrition and evaluating nutritional interventions. However, they remain valuable for
predicting clinical outcomes.
Skin testing—recall antigens
Not recommended as a malnutrition indicator. Although delayed hypersensitivity is associated with malnutrition, the former
can be influenced by factors unrelated to malnutrition.
Total lymphocyte count
Not recommended as malnutrition indicator. Low levels are influenced by factors unrelated to malnutrition.
Urine 3-methylhistidine
Not recommended as a malnutrition indicator. Although it reflects muscle mass, its level is influenced by meat intake and
poorly reflects changes in body protein stores. Requires accurate 24-h urine collection
(Continued)

17 - 346 Enteral and Parenteral Nutrition
346 Enteral and Parenteral Nutrition
TABLE 345-2  Common Body Composition Studies, Laboratories, and Other Studies Used in Nutrition Assessment
TEST
NOTES
Functional tests (Recommended)
Clinical Frailty Scale
A tool that measures frailty, which is defined as a state of increased vulnerability to adverse health outcomes due to reduced
physiologic reserves. It uses various self-reported factors (e.g., physical mobility, cognitive function, and the ability to perform
daily activities) to classify frailty into nine classifications, ranging from very fit (level 1) to severely frail (level 9). Patients who
score ≥5 are considered frail.
Sit-to-Stand Test
Assesses lower limb strength, balance control, and functional mobility by measuring the time required for a person to transfer
from a seated to a standing position and back to sitting five times.
Handgrip strength
A measure of muscular strength generated by one’s forearm muscles to screen for upper body strength.  Low grip strength
is associated with poorer clinical outcomes. Patients with low grip strength may benefit most from personalized nutritional
support.
SARC-F
A screening tool that determines the risk of sarcopenia, which is defined as a progressive and generalized loss of skeletal
muscle mass and strength. It uses self-reported deficiencies in strength, walking, rising from a chair, climbing stairs, and
experiencing falls to provide a score from 0 to 10, and a score ≥4 is considered at risk of sarcopenia.
6-Meter Walk Test
Assesses walking speed (meter/sec) over a 6-m distance to determine functional mobility, gait, and vestibular function.
Acknowledgments
The authors are grateful to Dr. Gordon L. Jensen for contributions to this
chapter in previous editions of Harrison’s.
■
■FURTHER READING
Cederholm T, Bosaeus I: Malnutrition in adults. N Engl J Med 391:155,
2024.
Gonzalez CC et al: Calf circumference: Cutoff values from NHANES
1999-2006. Am J Clin Nutr 113:1679, 2021.
Guerra RS et al: Usefulness of six diagnostic and screening measures
for undernutrition in predicting length of hospital stay: A compara­
tive analysis. J Acad Nutr Diet 115:927, 2015.
Jensen GL et al: GLIM criteria for the diagnosis of malnutrition: A
consensus report from the global clinical nutrition community. J
Parenter Enteral Nutr 43:32, 2019.
Lew CCL et al: Association between malnutrition and clinical out­
comes in the intensive care unit: A systematic review. J Parenter
Enteral Nutr 41:744, 2017.
Schuetz P et al: Individualized nutritional support in medical inpatients
at nutritional risk: A randomized clinical trial. Lancet 393:2312, 2019.
Uhl S et al: Interventions for malnutrition in hospitalized adults. J Hosp
Med 17:556, 2022.
White JV et al: Consensus statement: Academy of Nutrition and Dietetics
and American Society for Parenteral and Enteral Nutrition. Character­
istics recommended for the identification and documentation of adult
malnutrition (under-nutrition). J Parenter Enteral Nutr 36:275, 2012.
Wong A et al: An umbrella review and meta-analysis of interventions,
excluding enteral and parenteral nutrition, initiated in the hospital for
adults with or at risk of malnutrition. Am J Clin Nutr 118:672, 2023.
L. John Hoffer, Bruce R. Bistrian
Enteral and Parenteral
Nutrition
There are three kinds of specialized nutritional support (SNS): (1) optimized
voluntary nutritional support, which is indicated when a patient’s bar­
riers to adequate nutrition can be overcome by special attention to the
details of how their food is constituted, prepared, and served and its con­
sumption optimized; (2) instrumental enteral nutrition (EN), in which
a liquid nutrient formula is delivered through a tube placed in the
stomach or small intestine; and (3) parenteral nutrition (PN), in which
(Continued)
a nutritionally complete mixture of crystalline amino acids, glucose,
lipid emulsions, minerals, electrolytes, and micronutrients is infused
directly into the bloodstream.
When does a hospitalized adult patient need SNS, and when it’s
indicated, how should it be provided? This chapter summarizes the
physiologic principles that guide the correct use of SNS and offers prac­
tical information about the diagnosis and management of nutritional
disorders in adult hospitalized patients.
The management of in-hospital nutritional disorders follows three
steps: (1) screening and diagnosis; (2) determination of the severity and
urgency of treating a diagnosed nutritional disorder in its overall clinical
context; and (3) selection of the kind of SNS to provide, its composition,
and the details of providing it. To follow these steps properly, physicians
require a general understanding of nutritional physiology, nutrient
requirements, the pathophysiology and diagnosis of the nutritional
disorders, and familiarity with the indications, advantages, risks, and
administration of the different kinds of SNS. Because most physicians
lack formal clinical nutrition training, they should, in general, collabo­
rate with clinical dietitians and specialized pharmacists in this process.
CHAPTER 346
Enteral and Parenteral Nutrition
■
■NUTRITIONAL PHYSIOLOGY
(See Chaps. 343–345.)
Energy 
The total daily energy expenditure of a healthy sedentary
adult is ~36 kcal/kg. Resting energy expenditure accounts for ~75% of
this total. Resting energy expenditure can be measured by indirect cal­
orimetry or estimated using a variety of predictive equations that input
weight, height, age, sex, and sometimes, disease-related factors. Fever
and some diseases increase resting energy expenditure; prolonged
semi-starvation triggers an adaptive reduction in resting and total
energy expenditure. Total energy expenditure determines how much
dietary energy must be consumed to maintain an existing store of body
fat and protein. The amount of energy a patient actually requires may
be less than their total energy expenditure, as in obesity therapy or
during temporary periods of exogenous energy intolerance, or greater
than total energy expenditure during recovery from starvation disease.
Protein and Amino Acids 
Dietary protein provides the 23 essen­
tial and nonessential amino acids required for protein synthesis. Pro­
tein must be consumed throughout life, because endogenous protein
turnover entails a minimum obligatory rate of amino acid catabolism.
Amino acid catabolism is regulated, appropriately increasing and
decreasing in response to variations in protein intake, but it cannot fall
below a certain minimum rate—the “protein minimum”—that deter­
mines an individual’s minimum dietary protein requirement. The aver­
age daily minimum protein requirement for a healthy adult consuming
an otherwise adequate diet is 0.65 g/kg; the “safe” or “recommended”
intake is 0.80 g/kg. Average protein consumption in wealthy societies is
approximately twice the average minimum requirement.
Energy balance affects the minimum protein requirement. Positive
energy balance, when achieved by greater carbohydrate consumption,
can improve body protein retention. Negative energy balance, such as
during therapeutic weight reduction, reduces the efficiency of protein
turnover and increases the dietary protein requirement. Some diseases
(or their treatments) increase the protein requirement, by (1) increas­
ing amino acid loss from the body (as in malabsorption and protein
loss via wound exudates, fistulas, or inflammatory diarrhea), removing
amino acids from the circulation (renal replacement therapy), or (2)
increased muscle protein catabolism, which occurs as an adverse effect
of high-dose glucocorticoid therapy and, most commonly, as part of
the metabolic response to the systemic inflammation that accompanies
major injury, serious infections, and intense immune system activation.
A highly protein-catabolic patient may excrete 15 g N (nitrogen)/d or
more in their urine in the absence of dietary protein provision: this is
more than three times faster than occurs with simple fasting. Since 1 g N
lost from the body reflects the loss of 6.25 g formed protein, 15 g N
loss/d implies the loss of 15 × 6.25 = 94 g protein per day. Since the
body’s hydrated metabolically active tissue mass (also called body cell
mass) is ~20% protein, one may calculate that 94 g protein loss/d indi­
cates a loss from the body of ~470 g (1 lb) of active tissue mass per day.
Sufficiently generous protein provision can mitigate this kind of active
tissue loss. The extent to which protein-catabolic disease increases the
protein requirement is debated, but a common recommendation for
critically ill patients is 1.2–2.0 g protein/kg of normal body weight per
day—close to the habitual protein intake of healthy people in wealthy
societies.
Protein-Energy Interactions 
Both energy deficiency and sys­
temic inflammation increase the dietary protein requirement; how do
these factors interact? The loss of body protein caused by energy defi­
ciency can be prevented or minimized by increased protein consump­
tion. Systemic inflammation reduces the benefit of increased protein
provision under hypocaloric conditions but does not eliminate it. Some
evidence suggests that a minimum energy dose, such as 50% of total
energy expenditure, may help preserve the active tissue store during
severe systemic inflammation. It is clear, however, that energy doses
70% of total energy expenditure do not have any greater proteinsparing effect, and the additional amounts of glucose and fluid volume
required to deliver them can have adverse effects that include hypergly­
cemia, dyslipidemia, and extracellular volume expansion.
PART 10
Disorders of the Gastrointestinal System
Permissive Underfeeding and Hypocaloric Nutrition 
These
terms have different meanings, and they should not be conflated.
Permissive underfeeding refers to the deliberate underprovision of all
nutrients including protein, whereas hypocaloric nutrition refers to
reduced energy provision combined with increased protein to compen­
sate for the known protein-wasting effect of negative energy balance.
Micronutrients 
Minimum amounts of the nine water-soluble vita­
mins (the B vitamins and vitamin C), four fat-soluble vitamins (A, D,
E, and K), eight minerals (calcium, phosphorus, potassium, sodium,
chloride, magnesium, zinc, and iron), essential fatty acids, and several
essential trace elements are required to avoid deficiency diseases. Overt
deficiencies of potassium, sodium, magnesium, and phosphorus are so
common in hospitalized patients that it is standard practice to monitor
for and correct them. Some drugs cause renal potassium, magnesium,
or zinc losses that require increased provision. Gastrointestinal losses
from nasogastric drainage tubes or intestinal losses from fistulas or
diarrhea incur losses of potassium, sodium, calcium, magnesium, and
zinc.
Less investigated, but common, are subclinical deficiencies of zinc,
vitamin C, vitamin D, and possibly other micronutrients. Physicians
often assume that consumption of a regular hospital diet protects
patients from micronutrient deficiencies. This assumption is unwar­
ranted when the patient’s nutritional status is deficient upon admission
and remains so during their hospital stay.
Systemic inflammation can complicate the clinical laboratory evalu­
ation of micronutrient status. It increases the requirements for certain
micronutrients, partially redistributing them from the extracellular
compartment into tissue stores, as occurs with iron, selenium, zinc,
and perhaps other nutrients. This effect reduces the reliability of many
standard laboratory tests of micronutrient status. The evaluation of
a patient’s micronutrient status should begin with a determination
of their previous and current intake and health status, including the
identification of potential routes of excessive micronutrient loss. In
acute-care clinical settings, the intensity of systemic inflammation
should be determined using clinical criteria and by measuring serum
concentrations of albumin or C-reactive protein, an acute-phase pro­
tein whose wide dynamic range and short half-life make it particularly
useful in this situation,
■
■MACRONUTRIENT MALNUTRITION SYNDROMES
The decision to embark on SNS must be justified by a well-formulated
nutritional diagnosis and a clearly defined therapeutic goal. This
chapter focuses on the diagnosis, treatment, and prevention of inhospital adult starvation-related malnutrition (SRM) and two related
conditions: chronic disease–related malnutrition (CDM) and acute
disease–related malnutrition (ADM). As explained in Chap. 345, SRM
is the disease brought about by prolonged semi-starvation. Synonyms
for this disease are “starvation-induced protein-energy malnutrition”
and “starvation disease.” CDM can be understood as SRM (simple
starvation disease) that is combined and compounded by moderately
severe systemic inflammation. SRM and CDM are anatomically similar
but etiologically and metabolically different, as elaborated upon below.
ADM refers to an injury-induced metabolic condition that creates a
high risk of severe body protein deficiency, rather than to an alreadyexisting anatomic starvation disease.
Starvation-Related Malnutrition 
The pathologic features that
make SRM a disease, and distinguish it from the semi-starvation that
precedes it, emerge when the patient’s active tissue mass has been
depleted enough to impair specific physiologic functions.
The body normally adapts to sustained inadequate energy provision
by reducing energy expenditure and net protein catabolism, partly
through hormone- and nervous system–regulated alterations in cel­
lular metabolism, but mostly from loss of muscle, which is the major
generator of energy expenditure. (Visceral organ mass and function
have a higher priority for protection during starvation.) These adapta­
tions permit prolonged survival, although at a cost that includes muscle
loss (including cardiac and respiratory muscle), skin thinning, lethargy,
a tendency to hypothermia, and functional disability. The cardinal
anatomic features of SRM—generalized muscle loss and subcutane­
ous adipose tissue depletion—are easy to detect by simple physical
examination.
SRM always manifests as weight loss, but weight loss alone may
not reveal its full severity. Semistarvation increases the extracellular
fluid (ECF) volume, sometimes seriously enough to cause edema.
Starvation-associated edema may occur even in the absence of its
known common causes (it is termed “starvation edema”). In adults
with initially normal body composition, starvation-induced weight
loss linearly tracks the loss of active tissue mass; this occurs because
the weight loss caused by the depletion of adipose tissue is counterbal­
anced by increased ECF volume. A 25% reduction in body weight (and
active tissue mass) significantly compromises physiologic function; a
50% reduction places otherwise uncompromised young adults on the
cusp of thermodynamic survival; older people with comorbidities are
at even greater risk. People with SRM are frail, feel unwell, lack strength
and emotional range, and are at risk of hypothermia.
The main cause of SRM worldwide is involuntary food deprivation.
Its causes in hospitalized patients are many: inadvertent or physi­
cian-ordered food deprivation; psychological depression or distress;
anorexia nervosa; poorly controlled pain or nausea; badly presented
unappealing food; communication barriers; physical or sensory dis­
ability; dysphagia and other mechanical difficulties ingesting food;
partial obstruction of the esophagus, stomach, or intestines; thrush;
intestinal angina; and most commonly, combinations of these causes.
Chronic Disease–Related Malnutrition and Cachexia 
These
terms refer to SRM combined with and complicated by chronic sys­
temic inflammation. CDM is common among patients suffering from
chronic infection; inflammatory autoimmune disease; chronic severe
hepatic, renal, cardiac, or pulmonary disease; and neoplastic diseases
that induce a systemic inflammatory response or cause tissue injury.
CDM induces anorexia—a strong disinclination to eat even when there
is no physical barrier to it—and is worsened by it. Its features include
muscle loss, weakness, fatigue, and impaired adaption to starvation, all
of which contribute to a vicious cycle of worsening starvation disease.
Fortunately, the nutritional deficit on the input side of this equa­
tion (anorexia-driven inadequate food consumption) is usually the
strongest driver of the patient’s CDM, making it amenable to a wellorganized nutritional intervention while an effective treatment plan for
the primary disease is identified and implemented.
Cachexia is an older term now used with varying meanings. It
generally refers to a metabolic syndrome of moderate systemic inflam­
mation, unrelenting generalized muscle loss, and their associated
symptoms, including anorexia, fatigue, frailty, and mood disturbance.
For the purpose of this chapter, it may be considered as roughly syn­
onymous with CDM.
Acute Disease–Related Malnutrition 
Other terms for ADM are
“injury-induced malnutrition” and “protein-catabolic critical illness.” The
metabolic-inflammatory response to severe tissue injury and sepsis
mobilizes muscle amino acids and leads to rapid and severe general­
ized muscle loss and increased resting energy expenditure under
conditions in which voluntary food intake is usually impossible. SRM
or CDM may be absent at the onset of a patient’s critical illness, but
muscle loss will rapidly develop unless the inciting medical or surgical
disease is rapidly and effectively treated and SNS provided. Death from
simple starvation of nonobese adults typically occurs within ~8 weeks,
whereas death from untreated starvation in patients with sustained
ADM will occur much sooner.
■
■NUTRITIONAL DIAGNOSIS
The cardinal anatomic features of starvation disease (either SRM or
CDM) are generalized muscle loss and diminished body fat. Routine
physical examination will immediately reveal these features, but,
perhaps surprisingly, what ought to be an easy diagnosis is often over­
looked. This section explains the details and pitfalls of diagnosing SRM
and CDM.
Muscle Mass 
Muscle tissue normally accounts for ~80% of an
adult’s active tissue mass. It is easy to detect loss of muscle mass, and
its severity is determinable almost at a glance. But generalized muscle
loss has many causes: (1) old age–related muscle loss (sarcopenia); (2) dis­
use muscle atrophy; (3) high-dose glucocorticoid therapy and certain
endocrine diseases (uncontrolled diabetes mellitus, adrenocortical
insufficiency, hyperthyroidism, androgen deficiency, hypopituitarism);
and (4) primary muscle or neuromuscular diseases. The guiding prin­
ciple is that SRM and CDM are very common treatable causes of mus­
cle loss. Whenever generalized muscle loss is identified, its potential
causes should be considered and the treatable ones addressed. Old age
is irreversible, but adequate protein and energy provision, combined
with physical rehabilitation, can be lifesaving.
Generalized muscle loss of any cause is especially dangerous in
ADM, because the combination of ADM and preexisting muscle loss
places patients at the cliff edge of lethal depletion of their active tis­
sue store. A diminished muscle mass is less able to release adequate
amounts of amino acids into the circulation for protein synthesis at
sites of tissue injury and healing and to the central protein pool to
regulate the immunoinflammatory response.
Subcutaneous Adipose Tissue 
Severe adipose tissue deple­
tion almost always indicates starvation disease, but it is not required
to make the diagnosis. The current obesity epidemic has created a
population of obese patients with SRM or CDM in whom muscle loss
has outpaced fat loss. A conscientious physical examination will easily
identify the patient’s diminished muscle mass beneath their residual
subcutaneous fat.
ECF Volume 
The ECF compartment normally represents ~20%
of body weight. SRM moderately increases ECF volume. CDM is
associated with additional edema-promoting conditions, especially the
hypoalbuminemia associated with systemic inflammation. The effect
of increased ECF volume to increase body weight and distort physical
appearance should be borne in mind, for it may conceal the severity of
muscle loss.
Body Mass Index 
Body mass index (BMI) is defined as body
weight (kg) divided by the square of height (m2). The normal reference
range is 20–25 kg/m2. BMI <19–20 usually indicates nonpathologically
reduced muscle and fat mass. BMI <15 indicates severe starvation
disease, and BMI <13 is usually thermodynamically incompatible with
life, especially in older patients with comorbidities. Some guidelines
and clinical trial enrollment criteria have defined what they term as
“malnutrition”—meant as starvation disease—as a BMI <16 or 17. But
it is wrong to rely on BMI as the sole basis for diagnosing starvation
disease. A BMI <17 certainly implies starvation disease, for the body
composition required for such a BMI can only develop by jettisoning a
large fraction of the body’s active tissue and adipose tissue store. But a
BMI >17 does not rule out starvation disease. Many patients with star­
vation disease have a normal or above-normal BMI, despite important
muscle loss, because of residual obesity or an expanded ECF volume.
Visual BMI 
With some practice, clinicians can learn to accurately
predict the BMI of nonobese, nonedematous patients by systematically
examining their muscle groups. This skill enables them to evaluate the
severity of starvation disease even in obese or edematous patients—for
whom measured BMI is unreliable—simply by evaluating their muscle
groups while consciously discounting their subcutaneous fat and
edema. Visual BMI can be used to estimate a patient’s “normalized
dry body weight”—what their body weight would be without obesity,
edema, or ascites. For example, the normalized dry body weight of a
1.75-m adult with a visual BMI of 17 is 1.752 × 17 = 52 kg.
CHAPTER 346
Laboratory and Technical Assessment 
Clinical laboratory
measurements have three main purposes in the evaluation and man­
agement of starvation disease.
Enteral and Parenteral Nutrition
MUSCLE MASS  Bedside ultrasound is proving to be a valuable tech­
nique for quantifying muscle mass at specific body sites. However, it
need not and should not replace the immediate information provided
by the eyes, hands, and mind of an astute clinician.
SYSTEMIC INFLAMMATION  The absence or presence of systemic
inflammation is what fundamentally distinguishes SRM from CDM.
The most useful laboratory indicators of systemic inflammation are a
reduced serum albumin concentration and increased serum C-reactive
protein concentration. Systemic inflammation increases the perme­
ability of capillary walls to large molecules, and the resulting osmotic
shift increases ECF volume. As intravascular albumin redistributes
in this larger volume, its serum concentration decreases. Variably
increased albumin catabolism and reduced liver albumin synthesis
also contribute. Dietary protein deficiency and muscle loss perpetuate
inflammation-induced hypoalbuminemia, because the amino acids
required for hepatic albumin synthesis come from the diet and endog­
enous muscle protein.
Hypoalbuminemia and a reduced serum prealbumin concentration
are often considered as indicators of “malnutrition.” This is incorrect.
These serum proteins are negative acute-phase reactants that indicate
the presence of systemic inflammation. Systemic inflammation induces
anorexia and increases muscle catabolism, effects that increase the risk
of CDM, but the disease itself may or not exist at the time, and it may
never develop. Serum acute-phase reactants will remain abnormal
as long as systemic inflammation persists. Appropriate nutrition can
mildly improve serum albumin concentration, but it will not fully nor­
malize until after the systemic inflammation subsides. Even then, full
normalization may require several weeks because of albumin’s serum
half-life of at least 2 weeks.
PROTEIN-CATABOLIC INTENSITY  The defining feature of proteincatabolic disease is generalized muscle loss. Situations of greatly
increased protein loss can be identified by measuring body N loss.
Most N leaves the body in the urine (almost all of it in urea, ammo­
nium, and creatinine). Total N is not usually measured in hospital
laboratories, but urinary urea N (which normally accounts for ~85%
of urinary N) can be measured and is routinely available. A recent,
validated formula estimates daily total N loss as follows: N loss (g) = g
N in urinary urea/0.85 + 2.
Net muscle protein catabolism follows approximately first-order
kinetics; this means that the rate of N loss from muscle is proportional
to the amount of N available to be lost. Protein-catabolic muscledepleted patients will lose less body N per day than equally catabolic
patients with a preserved muscle store, but they are at greater risk of
succumbing to their disease because they are closer to lethal depletion.
Rate of N loss always has to be interpreted in the context of existing
muscle mass.
Instrumental Nutritional Assessment 
Many nutritional assess­
ment instruments claim to identify “malnutrition” by enumerating and
summing a list of risk factors, laboratory results, and physical findings.
These tools are often hindered by ambiguity about their definition of
malnutrition and by failure to distinguish between screening and diag­
nosis. Correct clinical diagnosis is the identification of a known patho­
logic entity—SRM or CDM, for example—by considering the patient’s
medical history, pertinent findings on physical examination, and labo­
ratory or imaging reports. Diagnosis also involves an estimation of the
probability that the diagnosis is correct and a judgment of its severity.
By contrast, screening is the application of a simple test that identifies
people at sufficiently high risk of a certain disease to warrant definitive
procedures to establish the diagnosis or rule it out or that identifies
people at sufficiently high risk of developing it to warrant specific pre­
ventive interventions. Screening tools and risk predictors are extremely
useful, but they should not be confused with clinical diagnosis.
PART 10
Disorders of the Gastrointestinal System
■
■SPECIALIZED NUTRITIONAL SUPPORT
Optimized Voluntary Nutritional Support 
When feasible, this
is the approach of choice because it engages and empowers the patient,
encourages mobilization and reconditioning, is consistent with the
objectives of patient-centered medicine, and is risk-free. Its disadvan­
tage is that it is time-consuming and labor-intensive, and it demands
astute attention to the specific needs of individual patients.
Enteral Nutrition 
This is nutrition provided through a feeding
tube placed through the nose into the stomach or beyond it into the
duodenum, by insertion of a tube through the abdominal wall into the
stomach or beyond it into the jejunum, or by a surgical approach to
access the stomach or small intestine. EN is usually the treatment of
choice when optimized voluntary nutritional support is impossible or
has failed. It is relatively simple, safe, and inexpensive, and it maintains
the digestive, absorptive, and immunologic barrier functions of the
gastrointestinal tract. EN is appropriate when optimized voluntary
nutrition is not feasible or has failed and the patient’s gastrointestinal
tract is functioning and can be accessed.
EN Products 
The most common forms of EN are commercially
manufactured formulas.
STANDARD POLYMERIC FORMULAS  These are the most widely used
EN products. They are available in a wide variety of formats designed
to meet the nutritional requirements of a normal, healthy person.
Carbohydrates provide most of the energy. The proteins are intact and
of high biologic value (casein, whey, or soy) and require adequate pan­
creatic and intestinal enzyme function for digestion and absorption.
They are isotonic or nearly so and provide 1000–2000 kcal and 50–70 g
protein/L.
POLYMERIC FORMULAS WITH FIBER  The addition of dietary fiber to
formulas sometimes improves bowel function and feeding tolerance.
Fermentable (soluble) fibers such as pectin and guar are metabolized
by colonic bacteria, yielding short-chain fatty acids that fuel colono­
cytes. Nonfermentable (insoluble) fibers increase fecal bulk, improve
peristalsis, and may improve diarrhea.
ELEMENTAL AND SEMI-ELEMENTAL FORMULAS  The macronutrients
in these formulas are partially or completely hydrolyzed. They were
designed for patients with known maldigestion and malabsorption but
are sometimes used empirically for patients intolerant of a standard
polymeric formula, especially when acutely ill, although without strong
evidence of superiority.
IMMUNE-ENHANCING FORMULAS  In addition to providing macronu­
trients and conventional amounts of micronutrients, these EN prod­
ucts contain large amounts of certain nutrients designed to favorably
modulate the immune response: arginine and n-3 fatty acids especially
have some clinical experimental support, but also various combina­
tions of glutamine, nucleotides, and antioxidants for which evidence of
benefit is limited or lacking.
PROTEIN-ENRICHED FORMULAS  Most EN products provide energy
and protein in a ratio appropriate for a healthy person. Proteinenriched formulas provide ~90 g protein and 1000 kcal/L. Originally
marketed to meet the increased protein requirement of weight-reducing
obese patients, these products are increasingly used to provide proteincatabolic patients with a more generous amount of protein without
energy overfeeding. EN can be further protein-enriched by adding
flushes of water-soluble powdered protein supplements.
OTHER FORMULAS  Various disease-specific EN products are available
for patients with diabetes or hepatic, renal, or pulmonary disease. Their
use can improve some metabolic endpoints, without clear evidence that
they improve clinical outcomes.
Parenteral Nutrition 
PN delivers a complete nutritional regi­
men directly into the bloodstream in the form of crystalline amino
acids, glucose, triglyceride emulsions, minerals (calcium, phosphate,
magnesium, and zinc), electrolytes, and micronutrients. Because of
its high osmolarity (>1200 mOsm/L) and often large volume, PN is
infused into a central vein in adults. Ready-to-use PN admixtures
typically containing 4–7% hydrated amino acids and 20–25% glucose
(with or without electrolytes) are available in two-chamber (amino
acids and glucose) or three-chamber (amino acids, glucose, and lipid)
bags that are intermixed with vitamins, trace minerals, and additional
electrolytes then added just prior to infusion. Although convenient
and cost-effective, these products have fixed nutrient compositions
and are dosed according to the volume required to meet a patient’s
energy requirement but not necessarily their protein requirement. In
some situations—especially ADM—a more sophisticated approach is
justified that uses a computer-controlled sterile compounder to create
combinations of amino acids and glucose that precisely meet the pro­
tein, energy, fluid, and acid-base regulatory requirements of individual
patients.
Amino Acids 
The amino acids in PN admixtures provide sufficient
amounts of the essential amino acids and total nonessential amino acid
N. Unlike protein-bound amino acids, which are connected by peptide
bonds, free amino acids are hydrated. This reduces their energy den­
sity from 4.0 (in formed protein) to 3.3 kcal/g and reduces the amount
of protein substrate they provide by ~17%. Thus, 100 g of free mixed
amino acids provides 83 g protein substrate.
Carbohydrate and Lipids 
The glucose in PN is dextrose mono­
hydrate; this hydration state reduces its energy concentration from
4.0 kcal/g (as in formed carbohydrate) to 3.4 kcal/g. Lipid emulsions
provide energy (~10 kcal/g) and the essential n-6 and n-3 fatty acids.
Traditional lipid emulsions based solely on soybean oil are giving way
to mixed emulsions that include medium-chain triglycerides, n-9
monounsaturated fatty acids, and n-3 fatty acids. Emulsions of pure
soybean oil; or a mixture of 80% olive oil and 20% soybean oil; or a
mixture of 30% soybean oil, 30% medium-chain triglycerides, 25%
olive oil, and 15% fish oil are available in the United States.
Fish oil (either as a component of a mixed emulsion or adminis­
tered separately) may reduce the risk of infections and length of stay
in critically ill patients. Some complex lipid emulsions are more highly
enriched in n-3 fatty acids and/or contain fewer n-6 polyunsaturated
fatty acids than soybean lipid, which is more prone to lipid peroxida­
tion and could promote the formation of the proinflammatory n-6
derivatives. Standard lipid infusion rates should not exceed 8 g/h,
equivalent to 175 g (1925 kcal)/d for a 70-kg patient. Because of their
longer chain length and slower tissue uptake, pure fish oil emulsions
are infused at lower rates.
Minerals, Micronutrients, and Trace Elements 
The default
concentrations of electrolytes, minerals, and micronutrients in PN
solutions are designed to meet the requirements of a healthy adult.
These starting doses must be adjusted to meet the frequently abnormal
and often-changing requirements of individual patients. Being unsta­
ble, multivitamin mixtures are injected into PN bags just prior to their
delivery to the medical unit. Parenteral water-soluble vitamin require­
ments are greater than standard oral requirements, because hospital­
ized patients often have vitamin deficiencies or increased requirements
and because intravenous administration of vitamins increases their loss
in the urine. Ascorbic acid degrades spontaneously in PN solutions,
even when protected from light.
APPROACH TO THE PATIENT
Indications, Selection, and Provision of Specialized
Nutritional Support
Most hospitalized patients do not require SNS because they can
and will eat enough with appropriate management of their primary
disease. Others may have a terminal disease whose downward
course will not be mitigated by SNS. Patients who are unable to eat
and digest enough of the standard hospital diet and have or are at
high risk of developing SRM or CDM are candidates for optimized
voluntary nutritional support. Instrumental SNS is indicated when
optimized voluntary nutritional support is inappropriate, impracti­
cal, or has failed. The decision to provide EN or PN is based on a
combination of four factors: (1) the determination that nutrient
ingestion will likely continue to be inadequate for many days;
(2) there is important muscle loss (of any cause); (3) the patient’s
nutrient requirements are increased (as from inflammatory diar­
rhea, enterocutaneous fistulas or exudates, or an inflammatory
protein-catabolic state); and (4) the reasoned judgment that SNS
has a reasonable prospect of improving the patient’s clinical out­
come or quality of life.
EN THERAPY
EN is indicated when the patient is unable to eat enough food and
is unlikely to do so for a long time, their gastrointestinal tract is
functional and accessible, and optimized voluntary nutrition is
impossible or inappropriate. EN is commonly used for patients with
impaired consciousness, severe dysphagia, or severe upper gastroin­
testinal tract dysfunction or obstruction, or who need mechanical
ventilation. Equally commonly, situations arise in which a patient’s
voluntary food intake is seriously curtailed by anorexia, unappeal­
ing food, nausea, vomiting, pain, distress, delirium, depression,
chewing difficulties, mild dysphagia, physical and sensory disability
(including dysgeusia), or undiagnosed thrush. In these complicated
and difficult situations, the clinical diagnosis of SRM or CDM will
tip the decision toward EN or PN.
EN is contraindicated by intestinal ischemia, mechanical
obstruction, peritonitis, and gastrointestinal hemorrhage. Highdose pressor therapy is a relative contraindication because of the
rare but lethal risk of intestinal ischemic injury. Severe coagulopa­
thy, esophageal varices, absent gag reflex, hypotension, paralytic
ileus, diarrhea, and nausea and vomiting are not absolute contra­
indications, but they increase the risk of complications and make it
less likely that EN will succeed in achieving its nutritional goal. In
general, intensive EN is best avoided under conditions of very grave
and unstable illness.
Initiation, Progression, and Monitoring  Nasogastric tube feed­
ing may proceed when the patient’s gastrointestinal function is
adequate with respect to gastric contractility (e.g., nasogastric
tube output <1200 mL/d) and intestinal function (no known or
suspected intraabdominal pathologic process and a nondistended
abdomen—mere absence of bowel sounds is not, by itself, a contra­
indication). After consent has been obtained and the appropriate
feeding tube (usually a nasogastric tube for short-term feeding) has
been placed and its position verified, the head of the patient’s bed
is raised to at least 30° and kept there to reduce the risk of regur­
gitation. The clinical dietitian ordinarily orders the formula and
adjusts its rate. When a standard polymeric formula is infused, it
normally commences at 50 mL/h and is advanced by 25 mL/h every
4–8 h until the goal rate is attained. Elemental formulas begin at a
slower rate and progress more slowly. Intragastric bolus feeding is
an option (200–400 mL feeding solution infused over 15–60 min
at regular intervals with verification of residual gastric contents
every 4 h).
Complications  The common complications of EN are aspiration
of regurgitated or vomited formula, diarrhea, fluid volume and
electrolyte derangements, hyperglycemia, nausea, abdominal pain,
constipation, and failure to achieve the nutritional goal.
Aspiration 
Patients with delayed gastric emptying, impaired gag
reflex, and ineffective cough are at high risk of aspiration pneumo­
nia. Ventilator-associated pneumonia is most often caused by aspi­
ration of microbial pathogens in the mouth and throat past the cuff
of an endotracheal or tracheostomy tube. Tracheal suctioning can
induce coughing and gastric regurgitation. Ventilator-associated
pneumonia can be prevented by elevating the head of the bed,
good mouth hygiene and gastrointestinal decontamination, nursedirected algorithms for formula advancement, and sometimes
postpyloric feeding. EN need not be suspended for gastric residual
volumes <300–400 mL in the absence of other signs of gastroin­
testinal intolerance (nausea, vomiting, abdominal pain, abdominal
distention). Continuous EN is often tolerated better than bolus
feeding and is the only option during jejunal feeding.
Diarrhea 
Diarrhea commonly occurs when the patient’s bowel
function is compromised by disease or drugs (most often, broadspectrum antibiotics). Once infection and inflammatory causes
have been ruled out, EN-associated diarrhea may be controlled by
using a fiber-containing formula or adding an antidiarrheal agent
to it. H2 blockers or proton pump inhibitors may help reduce the
net volume of fluid presented to the colon. Luminal nutrients exert
trophic effects on intestinal mucosa, so it is often appropriate to
continue tube feeding despite moderate but tolerable diarrhea, even
if it necessitates supplemental parenteral fluid support. Except in
cases of impaired small-intestinal absorptive function, there are no
well-established indications for elemental formulas; nevertheless,
they may be tested empirically when diarrhea persists despite the
use of fiber-enriched formulas and antidiarrheal drugs.
Gastrointestinal Intolerance 
Abnormally high gastric residual
volumes, abdominal distention, pain, and nausea are greatly dis­
tressing for patients, increase nursing workload, and slow the pro­
gression of EN. These problems can be avoided or minimized by
ensuring that the patient is in normal fluid and electrolyte balance,
by preventing severe hyperglycemia, and, when nausea, vomiting,
or abdominal distention develop, by the judicious prescription of
antiemetic and prokinetic drugs (and sometimes proton pump
inhibitors) on a regular—rather than as-needed—basis. Patients
with gastroparesis require postpyloric feeding.
Fluid Volume, Electrolyte, and Blood Glucose Abnormalities 
EN’s essential purpose is to provide macronutrients at an appropri­
ate rate. EN also provides standard amounts of fluid, electrolytes,
minerals, and micronutrients. Standard EN products are not
designed to manage abnormal fluid volume, electrolyte, and min­
eral requirements, which vary considerably and can change rapidly.
Blood glucose concentrations should be monitored regularly, and
additional measures—including intravenous fluid, electrolyte, and
insulin therapy—may be required to maintain homeostasis.
Failure to Reach the Nutritional Goal 
EN is frequently
delayed or interrupted by intolerance to the administered formula,
CHAPTER 346
Enteral and Parenteral Nutrition
by diagnostic tests and procedures (including dialysis), physical or
occupational therapy, or a clogged or pulled out tube. The result can
be a long delay in the progression of EN and ultimate failure to meet
the patient’s nutrient requirements.
EN in the Intensive Care Unit  Most critically ill patients cannot
eat anything and hence depend entirely on SNS. EN serves two pur­
poses in this setting. The first is to meet the patient’s macronutrient
requirements, especially their increased protein requirement. The
second purpose is to infuse sufficient nutrients into the intestines
to maintain their barrier and immunologic functions in the face
of a systemic inflammatory response that threatens them. Current
guidelines recommend starting EN soon after a critically ill patient
has been fluid resuscitated and stabilized. Once EN is underway,
its rate of delivery is gradually increased toward the nutritional
goal. EN frequently falls far short of its protein provision target,
even after a week or longer in the intensive care unit. Newer, highprotein EN products (and sometimes the addition of powdered
protein supplements) can correct this protein shortfall. There is
uncertainty about the timing and macronutrient composition of EN
early in ADM. We continue to recommend early EN as a platform
and preparation for later, more generous nutrient provision and to
maintain intestinal function.
PN THERAPY
PN is more resource-intensive and requires more expertise than
EN. It is used when invasive SNS is indicated and EN is impossible,
inappropriate, or insufficient to meet the patient’s nutritional needs.
The risks of PN are those of inserting and maintaining a central
venous catheter (traumatic injury from the insertion, serious infec­
tion, and venous thrombosis); allergy to some of its components;
glucose, electrolyte, magnesium, phosphate, and acid-base balance
abnormalities; and the adverse effects of the large intravenous fluid
volumes. When prolonged for many weeks—and especially when
it delivers excess energy—PN may cause or contribute to hepatic
dysfunction.
PART 10
Disorders of the Gastrointestinal System
Initiation, Progression, Monitoring, and Discontinuation  When
indicated, PN should begin soon after the patient has been hemo­
dynamically stabilized; glucose, electrolyte, and acid-base homeo­
stasis has been established; and they are able to tolerate the fluid
volume required to deliver it. Newer evidence indicates that the
provision of energy equal to total energy expenditure in the first
week of PN is not, in general, beneficial and can be harmful under
certain conditions. The high osmolarity of adult PN solutions and
the need for strict sterility require that they be infused through a
dedicated port in a central venous catheter. Jugular or femoral vein
catheters should not be used because of the difficulty maintaining
a dry, sterile dressing over the insertion site and the higher rate of
complications. The initial dose of glucose should not exceed 200 g/d
to avoid hyperglycemia (and—in susceptible patients with adapted
SRM—the refeeding syndrome).
Most non–critically ill patients do not require >500 g glucose
(1700 kcal)/d—assuming in this example a normalized dry body
weight of 70 kg. Patients with ADM should, in general, not receive
350 g glucose (1200 kcal)/d during the intense phase of their criti­
cal illness. A glucose infusion rate of ~200 g/d is physiologic and
rarely needs to be exceeded. When it eventually becomes appropri­
ate to provide energy equal to total energy expenditure, this may be
achieved by infusing a lipid emulsion.
We recommend hypocaloric nutrition (limited in glucose, lipid,
and fluid volume but generous in amino acids) for the first 2 weeks
of SNS in fat-sufficient or obese patients with ADM. Energy provi­
sion may increase, if indicated, after the catabolic storm has abated.
Lipids are commonly introduced after the first week of PN to
make up energy shortfalls. Serum triglyceride concentrations are
measured before commencing lipid infusions to detect preexisting
hypertriglyceridemia (>400 mg/dL), a relative contraindication.
Lipids may be infused daily or two or three times weekly. Lipid
infusions are not necessary to prevent essential fatty acid deficiency
during hypocaloric nutrition of patients with a normal or excessive
fat store, because the endogenous fat mobilized during energy defi­
ciency provides essential fatty acids to the rest of their body.
Capillary blood glucose concentrations are monitored several
times daily. Subcutaneous regular insulin is added to the PN admix­
ture as required to maintain average serum glucose concentrations
<140 mg/dL and >80 mg/dL. (Upper and lower limits of 180 and
100 mg/dL appear to be appropriate for critically ill patients with
diabetes mellitus.) The dose of regular insulin required on a given
day can be added to the following day’s PN solution. The insulin
dose increases roughly proportionately to the glucose dose. Cer­
tain benchmarks are useful. Basal endogenous insulin secre­
tion is ~30 units/d in normal people. When insulin is required
for nondiabetic, noncatabolic patients, 10 units of regular insulin
roughly cover 100 g infused glucose. Patients with non-insulindependent diabetes require ~20 units/100 g glucose. Noncatabolic
patients with insulin-dependent diabetes usually require approxi­
mately twice their at-home insulin dose, because parenteral glucose
stimulates insulin release more potently than oral carbohydrate and
because some insulin adheres to the infusion bag.
Biochemical Monitoring 
Serum urea, creatinine, electrolytes,
glucose, magnesium, phosphate, calcium, and albumin concentra­
tions are determined prior to commencing PN and most of them
(not albumin) monitored for the first few days, then twice weekly
or as required. Serum triglycerides and liver function indicators are
measured at baseline and after PN is underway to verify tolerance
to lipid infusions. An N balance determination may be useful at
the outset to evaluate the severity of protein catabolism in patients
with CDM or ADM, to identify patients who will benefit from more
generous amino acid provision, and during the course of PN to
determine whether N balance is improving with therapy.
Discontinuation 
PN is tapered and discontinued when the
patient can be adequately nourished by the enteral route. The PN
dose is proportionately reduced as food intake increases. Once a
patient can tolerate one-half to two-thirds of their requirement
by the enteral route and there is no mechanical or other barrier
to further increases, PN should be terminated. The transition to
oral nutrition can be difficult for some patients. In this situation,
optimized voluntary nutrition, although labor-intensive, is much
preferred to replacing PN with EN, for it is safe, effective, fosters
well-being, and prepares the patient for discharge home. It is tempt­
ing to discontinue PN in hopes of stimulating the patient’s appetite,
but this impulse should be resisted. PN does not cause anorexia nor
does discontinuing it stimulate appetite. Too-early discontinuation
of PN may delay a patient’s progression to full voluntary food con­
sumption by inducing anxiety and recreating starvation conditions.
A patient is most successfully weaned from PN by optimizing their
voluntary nutrition (including food brought from home), provid­
ing emotional support, encouraging physical activity, and being
patient. Some patients on the cusp of adequate oral nutrition will
benefit from discharge to the security and pleasure of home life and
homemade food; these patients are identified by observing, asking,
and listening.
Complications  Patients receiving PN are at higher risk of blood­
stream infections than other patients with central venous catheters.
Rigorously aseptic insertion technique, meticulous dressing care,
one port dedicated solely to PN, and careful glycemic control
reduce this risk. EN and PN carry equivalent overall risks and com­
plication rates, apart from gastrointestinal side effects of EN. PN
is much more prone to the error of inadvertent toxic overfeeding.
Hyperglycemia 
The most frequent metabolic complication
of PN is hyperglycemia in patients with insulin resistance due
to non-insulin-dependent diabetes mellitus, high-dose glucocor­
ticoid therapy, or severe systemic inflammation; the problem
is exacerbated by excessive rates of glucose provision. Glucose
concentrations are most easily kept <140 mg/dL and with the least
risk of hypoglycemia by infusing hypocaloric amounts of glucose
and, when necessary, meeting the patient’s energy requirement with
intravenous lipid. In ADM, the benefits of using the lowest pos­
sible insulin dose—minimal hyperinsulinemia and a reduced risk of
hypoglycemia—almost always outweigh the doubtful goal of rapidly
matching energy provision to energy expenditure.
Hypoglycemia 
Reactive hypoglycemia is uncommon but can
occur immediately after high-glucose, non-insulin-containing PN
is abruptly discontinued. It is prevented by slowing the PN infu­
sion rate to 50 mL/h for 1 or 2 h prior to stopping, replacing it with
10% glucose, or when the oral route is available, providing a snack.
More often, hypoglycemia occurs when the intensity of the patient’s
metabolic stress (or their glucocorticoid dose) decreases without
an appropriate downward adjustment of the insulin dose. This
problem is avoided by frequent capillary glucose determinations
and careful attention to medication doses and the patient’s general
condition.
Artefactual Hyperglycemia and Hyperkalemia 
Blood sam­
ples must be meticulously collected from a dual-port central venous
catheter. The intermixing of sample with even a tiny volume of
PN solution will falsely indicate hyperglycemia and hyperkalemia
and can result in treatment error. This error is identified when the
patient’s apparent serum glucose (and potassium) concentration
abruptly increases without reason, and the high glucose concentra­
tion is contradicted by a concurrent capillary glucose reading.
Volume Overload 
Volume overload causes substantial mor­
bidity and increases mortality. Total fluid provision should not
normally exceed 2 L/d in the absence of abnormally high losses.
Hypertonic intravenous glucose triggers a more intense insulin
response than oral glucose, and it can increase urinary sodium
and water retention. Volume overload can be prevented by using a
compounder to prepare PN solutions, infusing glucose at the lowest
appropriate rate, avoiding energy overfeeding, and monitoring the
patient’s volume balance.
Hypertriglyceridemia 
This complication occurs when the rate
of lipid infusion exceeds plasma triglyceride clearance capacity.
Sepsis, renal failure, diabetes mellitus, high-dose glucocorticoid
therapy, and multiple-organ failure reduce triglyceride clearance.
An impaired immune response, increased risk of acute pancreatitis,
and altered pulmonary hemodynamics are potential, but not well
documented, complications of PN-induced severe hypertriglyceri­
demia. Lipid infusion rates should not usually exceed ~50 g (500 kcal)/d
in ADM.
Liver Disease 
Mild elevations of serum liver enzyme concen­
trations may occur within 2–4 weeks of initiating PN, but in most
cases, they will return to normal even when PN is continued. Clini­
cally important hepatic dysfunction, although common in children,
is uncommon in adults for whom energy overfeeding with resultant
fatty liver is avoided. Intrahepatic cholestasis occasionally develops
after many weeks of continuous PN; it is most often multifactorial
in origin. Cyclic PN—in which PN is infused for only 12 h of the
day—may prevent or reduce the severity of this complication.
PN in the Intensive Care Unit  Current guidelines recommend
starting EN soon after a critically ill patient has been resuscitated
and stabilized and enteral access to an adequately functioning gas­
trointestinal tract has been established. If the protein goal has not
been achieved after 7–10 days, PN may be added, especially if the
protein-catabolic state has not yet abated. Soy-based lipid emulsions
should be avoided during the first week of PN; alternative lipid
emulsions may prove to be safe and beneficial.
SPECIAL CLINICAL SITUATIONS
The Critical Illness–Nutrition Paradox  There is now good clinical
trial evidence that confirms what was long indicated by a combi­
nation of physiologic reasoning, biologic plausibility, and formal
clinical observation—namely, that personalized nutritional inter­
ventions improve the clinical outcomes of starving, non–critically
ill patients. The case for SNS would seem to be much stronger for
patients with ADM, which induces rapid muscle loss and maintains
or increases energy expenditure in people rendered incapable of
eating voluntarily. Yet well-designed large-enrollment clinical trials
have repeatedly failed to demonstrate short-term clinical benefits
from nutritional interventions in critical illness. In fact, there is
good evidence that, unlike in noncritical illness, early energy provi­
sion at a rate near or above the rate of energy expenditure not only
fails to improve clinical outcomes but may be deleterious. The fail­
ure of formal clinical trials to show benefits from SNS in critical ill­
ness has several possible explanations: severe prolonged starvation
is so obviously harmful that it is unethical to intentionally include
such a treatment arm in a randomized clinical trial; critical illness
is extremely heterogeneous, and not every critically ill patient is, or
remains, severely protein-catabolic for long; and owing to generous
admission criteria and thanks to the high quality of modern inten­
sive care, many patients admitted to intensive care units rapidly
improve and are discharged in a handful of days, while others are
so mortally ill when admitted to the intensive care unit that their
outcome is virtually predetermined. For these reasons, clinical
trials that enroll and report the outcomes of all patients admitted
to an intensive care unit could well fail to demonstrate an average
benefit from SNS. More than that, in current practice, standard EN
regimens prescribed for most critically ill patients commonly fail
to deliver more than one-half the currently recommended amount
of protein, and the low protein-to-energy ratio of most standard
EN and PN products can make it impossible to provide critically
ill patients sufficiently generous amounts of protein or amino acids
without subjecting them to harmful energy overfeeding. We await
the publication of large-enrollment clinical trials in which critical
illness is diagnosed, not by where the patient is being treated, but
using anatomic and pathophysiologic criteria and in which current
standard care is compared with treatment arms that provide per­
sonalized SNS appropriate to patients’ anatomy and pathophysiol­
ogy. For now, along with other experts, we continue to recommend
EN and PN for critically ill patients. Attempts to match energy
provision to energy expenditure should be avoided as long as sys­
temic inflammation remains severe. The dose of protein or amino
acids should be guided by physiologic reasoning and a personalized
evaluation of each patient’s anatomic and metabolic condition.
We recommend bearing in mind that people who survive critical
illness become non–critically ill people for whom the benefit of
appropriate nutrition, together with other components of medical
and surgical care, is not in question. Accordingly, we recommend a
long-view approach to the design of nutritional support regimens
for critically ill patients that aims both to increase their likelihood of
short-term survival and anticipates their post–critical illness phase,
which will be improved by measures that mitigate muscle loss, avoid
toxic energy overfeeding, and prevent or correct micronutrient
deficiencies.
Iron and PN  Iron deficiency is common in hospitalized patients.
It has many causes: preexisting deficiency, insufficient food, macro-
or microscopic gastrointestinal blood loss, and repeated blood
sampling. Even though it is common, the diagnosis is often missed
because the anemia of systemic inflammation is even more com­
mon, and it confounds interpretation of the usual indicator of iron
deficiency, serum ferritin. Serum ferritin and fasting serum trans­
ferrin saturation should be part of the patient’s initial nutritional
evaluation. Current evidence suggests that a fasting transferrin
saturation <20% is the most reliable indicator of functional iron
deficiency, even in the absence of anemia. In principle, all micro­
nutrient deficiency states should be prevented and corrected. Inhospital iron deficiency causes and prevents recovery from anemia
and increases postoperative blood transfusion requirements. Sub­
clinical iron deficiency could contribute to cognitive and immune
dysfunction.
CHAPTER 346
Enteral and Parenteral Nutrition
Iron is not added to PN mixtures. Iron dextran is incompatible
with lipid emulsions, and although it appears to be chemically com­
patible with aqueous solutions of amino acids and glucose, there
is realistic concern that interactions between iron molecules and
certain vitamins and amino acids in PN solutions could catalyze
the formation of free radicals that degrade vitamins and exert subtle
adverse systemic effects.
Intravenous iron is usually necessary to treat in-hospital iron
deficiency, but it should be delayed if there is known or suspected
bacterial infection, and possibly not during the intense phase of
ADM. A substantial rise in the serum iron concentration could
increase susceptibility to gram-negative (and possibly other micro­
bial) infections, as well as catalyze the formation of free radicals
that increase the intensity of the catabolic response to major tissue
injury.
Zinc  One liter of secretory diarrhea contains ~12 mg of zinc.
Patients with intestinal fistulas or high-volume chronic diarrhea
require this amount of zinc in addition to their daily requirement of
15 mg to avoid zinc deficiency. Zinc may be provided parenterally
or enterally. Because of its low bioavailability, 12 mg of parenteral
zinc is equivalent to 30 mg of oral zinc.
Old Age  In addition to their other frailties, elderly people com­
monly suffer from age-related muscle loss (sarcopenia) com­
pounded by disuse muscle atrophy. These factors place them at
high risk of the consequences of starvation disease and make them
candidates for earlier SNS.
Inactivity  Physical activity and adequate nutrition interact inti­
mately. Reduced physical activity reduces appetite, and physical
rehabilitation and its associated emotional benefits restore opti­
mism and appetite. Full nutrient provision will maintain or normal­
ize many physiologic functions in bedridden patients, but without
exercise, they will not increase their muscle mass.
PART 10
Disorders of the Gastrointestinal System
Renal Failure  Protein provision should not be reduced in patients
with renal failure unless renal replacement therapy is unavailable.
Renal replacement therapy removes large amounts of amino acids,
vitamins, and trace elements from the circulation, so protein and
micronutrient provision should be increased to compensate for
these losses.
Liver Failure  Patients with severe hepatic disease are relatively
intolerant to starvation and commonly have CDM when admitted
to hospital, so they are prime candidates for SNS. Their SNS should
be generous both in energy and protein, despite an increased risk of
hepatic encephalopathy. The risk of encephalopathy can be reduced
by meticulous attention to fluid balance, acid-base balance, and
electrolyte status and by spreading protein provision over the day
to accommodate the liver’s reduced capacity to clear amino acid–
derived ammonia.
Perioperative SNS  Patients with SRM or CDM awaiting elective
major surgery will benefit from 7–10 days of preoperative SNS.
Optimized voluntary nutrition is always preferred, but when a
patient has been admitted to hospital in a semi-urgent condition,
EN or PN is usually required to quickly meet the patient’s nutri­
tional goal. Preoperative SNS improves immunity and reduces
postoperative complications, but it will not increase serum albu­
min concentrations, and it should not be provided for >7–10 days
with the expectation that it will do so. More prolonged preopera­
tive EN or PN may confer slight additional nutritional benefits,
but they are counterbalanced by their risks and the consequences
of prolonged hospitalization and delayed surgery. Surgery should
not be delayed for starving patients whose muscle mass is normal
or only mildly depleted and who are not experiencing systemic
inflammation, for they are well able to withstand the trauma of
even major uncomplicated surgery. The need for urgent surgery
often precludes otherwise indicated preoperative SNS. Early post­
operative PN is usually indicated for these patients, for they are
at increased risk of postoperative complications and are unlikely
to consume much food over the next many days. Patients with
only mild muscle loss, no systemic inflammation, and no postop­
erative complications do not require postoperative PN unless (1)
adequate feeding by mouth has not been achieved by day 5–7 after
surgery or (2) there are indications that voluntary feeding will be
further delayed. There is evidence that perioperative immuneenhancing EN reduces morbidity in patients undergoing major
elective gastrointestinal surgery.
Cancer  SNS plays a crucial role in cancer therapy. Many malig­
nant neoplasms (especially those that involve the gastrointestinal
tract or induce systemic inflammation) and their cytotoxic thera­
pies create the conditions for starvation and commonly lead to
SRM or CDM. The prevention or treatment of these starvation
diseases will improve patients’ quality of life and their tolerance
to anticancer therapy. EN and PN are generally not prescribed
to patients with advanced cancer for which effective anticancer
therapy is unavailable, because the side effects and complications
of instrumental SNS are not counterbalanced by an improved dis­
ease trajectory. In some cases, the disease progresses so slowly that
the patient will die of starvation long before they succumb to their
primary disease. EN or PN is obviously appropriate in such cases.
Advanced Dementia  Optimized voluntary nutrition is the key
approach in this situation, and it can be used to deal with problems
such as disability and dysphagia in patients who get pleasure from
eating. There is no evidence that EN or PN improves quality or
length of life in patients with advanced dementia who show little
or no interest in food, and the side effects and complications are
unpleasant and sometimes dangerous.
REFEEDING SYNDROME
The refeeding syndrome can occur in patients with successfully
adapted SRM during the first week of nutritional repletion when
carbohydrate and sodium are introduced too rapidly. Carbohy­
drate stimulates insulin secretion, which, owing to its antinatri­
uretic effect, expands the ECF volume, especially when excessive
sodium is provided. Refeeding edema can be minimized by severely
limiting sodium provision and increasing carbohydrate provision
slowly. Carbohydrate refeeding may stimulate intracellular glucose6-phosphate and glycogen synthesis sufficiently to seriously lower
serum phosphate concentrations, impairing skeletal and cardiac
muscle function, and can trigger hemolysis. Abrupt carbohydrate
provision increases the downregulated metabolic rate of patients
with adapted SRM and stimulates N retention, new cell synthe­
sis, and cellular rehydration. Because phosphorus, potassium, and
magnesium deficiencies occur and are dangerous during refeeding,
their serum concentrations should be measured frequently and
appropriate supplements provided. Left heart failure may occur
in predisposed patients; it has three causes: (1) an abrupt increase
of intravascular volume due to the administration of fluids and of
glucose, which stimulates insulin-mediated renal sodium retention;
(2) increased cardiac demand on an atrophic left ventricle created
by an insulin-mediated increase of resting energy expenditure; and
(3) myocardial deficiencies of potassium, phosphorus, or magne­
sium. Cardiac arrhythmias may occur. Acute thiamine deficiency
encephalopathy is a devastating preventable complication of refeed­
ing, even with simple glucose infusions.
■
■FURTHER READING
Berger MM et al: ESPEN micronutrient guideline. Clin Nutr 41:1357,
2022.
Berger MM, Pichard C: When is parenteral nutrition indicated?
J Intensive Med 2:22, 2022.
Compher C et al: Guidelines for the provision of nutrition support
therapy in the adult critically ill patient: The American Society for
Parenteral and Enteral Nutrition. JPEN J Parenter Enteral Nutr 46:12,
2022.

18 - SECTION 3 Liver and Biliary Tract Disease
SECTION 3 Liver and Biliary Tract Disease
Jensen GL et al: GLIM criteria for the diagnosis of malnutrition: A con­
sensus report from the global clinical nutrition community. JPEN J
Parenter Enteral Nutr 43:32, 2019.
Singer P et al: ESPEN practical and partially revised guideline: Clinical
nutrition in the intensive care unit. Clin Nutr 42:1671, 2023.
Wischmeyer PE et al: Personalized nutrition therapy in critical care:
10 expert recommendations. Crit Care 27:261, 2023.
Section 3	 Liver and Biliary Tract Disease
Marc G. Ghany, Jay H. Hoofnagle
Approach to the Patient
with Liver Disease
A diagnosis of liver disease usually can be made accurately by careful
elicitation of the patient’s history, physical examination, and applica­
tion of a few laboratory tests. In some circumstances, radiologic exami­
nations are helpful or, indeed, diagnostic. Liver biopsy is considered
the criterion standard in evaluation of liver disease but is now needed
less for diagnosis than for grading (activity) and staging (fibrosis) of
disease. Noninvasive means of assessing fibrosis stage have become
increasingly helpful and may allow for avoidance of biopsy in an
increasing proportion of patients. This chapter provides an introduc­
tion to diagnosis and management of liver disease, briefly reviewing
the structure and function of the liver; the major clinical manifestations
of liver disease; and the use of clinical history, physical examination,
laboratory tests, imaging studies, and liver biopsy.
LIVER STRUCTURE AND FUNCTION
The liver is the largest organ of the body, weighing 1–1.5 kg and rep­
resenting 1.5–2.5% of the lean body mass. The size and shape of the
liver vary and generally match the general body shape—long and lean
or squat and square. This organ is located in the right upper quadrant
of the abdomen under the right lower rib cage against the diaphragm
and projects for a variable extent into the left upper quadrant. It is held
in place by ligamentous attachments to the diaphragm, peritoneum,
great vessels, and upper gastrointestinal organs. The liver receives a
dual blood supply; ~20% of the blood flow is oxygen-rich blood from
the hepatic artery, and 80% is nutrient-rich blood from the portal vein
arising from the stomach, intestines, pancreas, and spleen.
The majority of cells in the liver are hepatocytes, which constitute
two-thirds of the organ’s mass. The remaining cell types are Kupffer
cells (members of the reticuloendothelial system), stellate (Ito or fatstoring) cells, endothelial and blood vessel cells, bile ductular cells, and
cells of supporting structures. Viewed by light microscopy, the liver
appears to be organized in lobules, with portal areas at the periphery
and central veins in the center of each lobule. However, from a func­
tional point of view, the liver is organized into acini, with both hepatic
arterial and portal venous blood entering the acinus from the portal
areas (zone 1) and then flowing through the sinusoids to the terminal
hepatic veins (zone 3); the intervening hepatocytes constitute zone 2.
The advantage of viewing the acinus as the physiologic unit of the liver
is that this perspective helps to explain the morphologic patterns and
zonality of many vascular and biliary diseases not explained by the
lobular arrangement.
Portal areas of the liver consist of small veins, arteries, bile ducts, and
lymphatics organized in a loose stroma of supporting matrix and small
amounts of collagen. Blood flowing into the portal areas is distributed
through the sinusoids, passing from zone 1 to zone 3 of the acinus and
draining into the terminal hepatic veins (“central veins”). Secreted bile
flows in the opposite direction—that is, in a countercurrent pattern
from zone 3 to zone 1. The sinusoids are lined by unique endothelial
cells that have prominent fenestrae of variable sizes, allowing the free
flow of plasma but not of cellular elements. The plasma is thus in direct
contact with hepatocytes in the subendothelial space of Disse.
Hepatocytes have distinct polarity. The basolateral side of the hepa­
tocyte lines the space of Disse and is richly lined with microvilli; it
exhibits endocytotic and pinocytotic activity, with passive and active
uptake of nutrients, proteins, and other molecules. The apical pole of
the hepatocyte forms the canalicular membranes through which bile
components are secreted. The canaliculi of hepatocytes form a fine
network, which fuses into the bile ductular elements near the portal
areas. Kupffer cells usually lie within the sinusoidal vascular space and
represent the largest group of fixed macrophages in the body. The stel­
late cells are located in the space of Disse but are not usually prominent
unless activated, whereupon they produce collagen and matrix. Red
blood cells stay in the sinusoidal space as blood flows through the lob­
ules, but white blood cells can migrate through or around endothelial
cells into the space of Disse and from there to portal areas, where they
can return to the circulation through lymphatics.
Hepatocytes perform numerous and vital roles in maintaining
homeostasis and health. These functions include the synthesis of most
essential serum proteins (albumin, carrier proteins, coagulation fac­
tors, many hormonal and growth factors), the production of bile and its
carriers (bile acids, cholesterol, lecithin, phospholipids), the regulation
of nutrients (glucose, glycogen, lipids, cholesterol, amino acids), and
the metabolism and conjugation of lipophilic compounds (bilirubin,
anions, cations, drugs) for excretion in the bile or urine. Measurement
of these activities to assess liver function is complicated by the multi­
plicity and variability of these functions. Commonly used liver “function”
tests are measurements of serum bilirubin, serum albumin, and pro­
thrombin time. The serum bilirubin level is a measure of hepatic con­
jugation and excretion; the serum albumin level and prothrombin time
are measures of protein synthesis. Abnormalities of bilirubin, albumin,
and prothrombin time are typical of hepatic dysfunction. Frank liver
failure is incompatible with life, and the functions of the liver are too
complex and diverse to be subserved by a mechanical pump; a dialysis
membrane; or a concoction of infused hormones, proteins, and growth
factors.
CHAPTER 347
Approach to the Patient with Liver Disease
LIVER DISEASES
While there are many causes of liver disease (Table 347-1), these disor­
ders generally present clinically in a few distinct patterns and are usu­
ally classified as hepatocellular, cholestatic (obstructive), or mixed. In
hepatocellular diseases (such as viral hepatitis and alcoholic liver dis­
ease), features of liver injury, inflammation, and necrosis predominate.
In cholestatic diseases, such as gallstone or malignant obstruction,
primary biliary cholangitis (previously referred to as primary biliary
cirrhosis), and some drug-induced liver diseases, features of inhibition
of bile flow predominate. In a mixed pattern, features of both hepato­
cellular and cholestatic injury are present (such as in cholestatic forms
of viral hepatitis and many drug-induced liver diseases). The pattern
of onset and prominence of symptoms can rapidly suggest a diagnosis,
particularly if major risk factors are considered, such as the age and sex
of the patient and a history of exposure or risk behaviors.
Typical presenting symptoms of liver disease include jaundice,
fatigue, itching, right-upper-quadrant pain, nausea, poor appetite,
abdominal distention, and intestinal bleeding. At present, however,
many patients are diagnosed with liver disease who have no symptoms
and who have been found to have abnormalities in biochemical liver
tests as a part of a routine physical examination or screening for blood
donation or for insurance or employment. The wide availability of bat­
teries of liver tests makes it relatively simple to demonstrate the pres­
ence of liver injury as well as to rule it out in someone in whom liver
disease is suspected.
Evaluation of patients with liver disease should be directed at deter­
mining (1) the etiologic diagnosis, (2) disease severity (grading), and
(3) disease stage (staging). Diagnosis should focus on the pattern of

19 - 347 Approach to the Patient with Liver Disease
347 Approach to the Patient with Liver Disease
Jensen GL et al: GLIM criteria for the diagnosis of malnutrition: A con­
sensus report from the global clinical nutrition community. JPEN J
Parenter Enteral Nutr 43:32, 2019.
Singer P et al: ESPEN practical and partially revised guideline: Clinical
nutrition in the intensive care unit. Clin Nutr 42:1671, 2023.
Wischmeyer PE et al: Personalized nutrition therapy in critical care:
10 expert recommendations. Crit Care 27:261, 2023.
Section 3	 Liver and Biliary Tract Disease
Marc G. Ghany, Jay H. Hoofnagle
Approach to the Patient
with Liver Disease
A diagnosis of liver disease usually can be made accurately by careful
elicitation of the patient’s history, physical examination, and applica­
tion of a few laboratory tests. In some circumstances, radiologic exami­
nations are helpful or, indeed, diagnostic. Liver biopsy is considered
the criterion standard in evaluation of liver disease but is now needed
less for diagnosis than for grading (activity) and staging (fibrosis) of
disease. Noninvasive means of assessing fibrosis stage have become
increasingly helpful and may allow for avoidance of biopsy in an
increasing proportion of patients. This chapter provides an introduc­
tion to diagnosis and management of liver disease, briefly reviewing
the structure and function of the liver; the major clinical manifestations
of liver disease; and the use of clinical history, physical examination,
laboratory tests, imaging studies, and liver biopsy.
LIVER STRUCTURE AND FUNCTION
The liver is the largest organ of the body, weighing 1–1.5 kg and rep­
resenting 1.5–2.5% of the lean body mass. The size and shape of the
liver vary and generally match the general body shape—long and lean
or squat and square. This organ is located in the right upper quadrant
of the abdomen under the right lower rib cage against the diaphragm
and projects for a variable extent into the left upper quadrant. It is held
in place by ligamentous attachments to the diaphragm, peritoneum,
great vessels, and upper gastrointestinal organs. The liver receives a
dual blood supply; ~20% of the blood flow is oxygen-rich blood from
the hepatic artery, and 80% is nutrient-rich blood from the portal vein
arising from the stomach, intestines, pancreas, and spleen.
The majority of cells in the liver are hepatocytes, which constitute
two-thirds of the organ’s mass. The remaining cell types are Kupffer
cells (members of the reticuloendothelial system), stellate (Ito or fatstoring) cells, endothelial and blood vessel cells, bile ductular cells, and
cells of supporting structures. Viewed by light microscopy, the liver
appears to be organized in lobules, with portal areas at the periphery
and central veins in the center of each lobule. However, from a func­
tional point of view, the liver is organized into acini, with both hepatic
arterial and portal venous blood entering the acinus from the portal
areas (zone 1) and then flowing through the sinusoids to the terminal
hepatic veins (zone 3); the intervening hepatocytes constitute zone 2.
The advantage of viewing the acinus as the physiologic unit of the liver
is that this perspective helps to explain the morphologic patterns and
zonality of many vascular and biliary diseases not explained by the
lobular arrangement.
Portal areas of the liver consist of small veins, arteries, bile ducts, and
lymphatics organized in a loose stroma of supporting matrix and small
amounts of collagen. Blood flowing into the portal areas is distributed
through the sinusoids, passing from zone 1 to zone 3 of the acinus and
draining into the terminal hepatic veins (“central veins”). Secreted bile
flows in the opposite direction—that is, in a countercurrent pattern
from zone 3 to zone 1. The sinusoids are lined by unique endothelial
cells that have prominent fenestrae of variable sizes, allowing the free
flow of plasma but not of cellular elements. The plasma is thus in direct
contact with hepatocytes in the subendothelial space of Disse.
Hepatocytes have distinct polarity. The basolateral side of the hepa­
tocyte lines the space of Disse and is richly lined with microvilli; it
exhibits endocytotic and pinocytotic activity, with passive and active
uptake of nutrients, proteins, and other molecules. The apical pole of
the hepatocyte forms the canalicular membranes through which bile
components are secreted. The canaliculi of hepatocytes form a fine
network, which fuses into the bile ductular elements near the portal
areas. Kupffer cells usually lie within the sinusoidal vascular space and
represent the largest group of fixed macrophages in the body. The stel­
late cells are located in the space of Disse but are not usually prominent
unless activated, whereupon they produce collagen and matrix. Red
blood cells stay in the sinusoidal space as blood flows through the lob­
ules, but white blood cells can migrate through or around endothelial
cells into the space of Disse and from there to portal areas, where they
can return to the circulation through lymphatics.
Hepatocytes perform numerous and vital roles in maintaining
homeostasis and health. These functions include the synthesis of most
essential serum proteins (albumin, carrier proteins, coagulation fac­
tors, many hormonal and growth factors), the production of bile and its
carriers (bile acids, cholesterol, lecithin, phospholipids), the regulation
of nutrients (glucose, glycogen, lipids, cholesterol, amino acids), and
the metabolism and conjugation of lipophilic compounds (bilirubin,
anions, cations, drugs) for excretion in the bile or urine. Measurement
of these activities to assess liver function is complicated by the multi­
plicity and variability of these functions. Commonly used liver “function”
tests are measurements of serum bilirubin, serum albumin, and pro­
thrombin time. The serum bilirubin level is a measure of hepatic con­
jugation and excretion; the serum albumin level and prothrombin time
are measures of protein synthesis. Abnormalities of bilirubin, albumin,
and prothrombin time are typical of hepatic dysfunction. Frank liver
failure is incompatible with life, and the functions of the liver are too
complex and diverse to be subserved by a mechanical pump; a dialysis
membrane; or a concoction of infused hormones, proteins, and growth
factors.
CHAPTER 347
Approach to the Patient with Liver Disease
LIVER DISEASES
While there are many causes of liver disease (Table 347-1), these disor­
ders generally present clinically in a few distinct patterns and are usu­
ally classified as hepatocellular, cholestatic (obstructive), or mixed. In
hepatocellular diseases (such as viral hepatitis and alcoholic liver dis­
ease), features of liver injury, inflammation, and necrosis predominate.
In cholestatic diseases, such as gallstone or malignant obstruction,
primary biliary cholangitis (previously referred to as primary biliary
cirrhosis), and some drug-induced liver diseases, features of inhibition
of bile flow predominate. In a mixed pattern, features of both hepato­
cellular and cholestatic injury are present (such as in cholestatic forms
of viral hepatitis and many drug-induced liver diseases). The pattern
of onset and prominence of symptoms can rapidly suggest a diagnosis,
particularly if major risk factors are considered, such as the age and sex
of the patient and a history of exposure or risk behaviors.
Typical presenting symptoms of liver disease include jaundice,
fatigue, itching, right-upper-quadrant pain, nausea, poor appetite,
abdominal distention, and intestinal bleeding. At present, however,
many patients are diagnosed with liver disease who have no symptoms
and who have been found to have abnormalities in biochemical liver
tests as a part of a routine physical examination or screening for blood
donation or for insurance or employment. The wide availability of bat­
teries of liver tests makes it relatively simple to demonstrate the pres­
ence of liver injury as well as to rule it out in someone in whom liver
disease is suspected.
Evaluation of patients with liver disease should be directed at deter­
mining (1) the etiologic diagnosis, (2) disease severity (grading), and
(3) disease stage (staging). Diagnosis should focus on the pattern of
TABLE 347-1  Liver Diseases
Inherited hyperbilirubinemia
  Gilbert syndrome
  Crigler-Najjar syndrome, types I
Liver involvement in systemic diseases
  Sarcoidosis
  Amyloidosis
  Glycogen storage diseases
  Celiac disease
  Tuberculosis
  Mycobacterium avium-intracellulare
and II
  Dubin-Johnson syndrome
  Rotor syndrome
Viral hepatitis
  Hepatitis A
  Hepatitis B
  Hepatitis C
  Hepatitis D
  Hepatitis E
  Others (Epstein-Barr virus
infection
Cholestatic syndromes
  Benign postoperative cholestasis
  Jaundice of sepsis
  Total parenteral nutrition–induced
jaundice
  Cholestasis of pregnancy
  Cholangitis and cholecystitis
  Extrahepatic biliary obstruction
[mononucleosis] herpesvirus,
cytomegalovirus, adenovirus
hepatitis)
  Cryptogenic hepatitis
Immune and autoimmune liver
diseases
  Primary biliary cholangitis
  Autoimmune hepatitis
  Sclerosing cholangitis
  Overlap syndromes
  Graft-versus-host disease
  Allograft rejection
Genetic liver diseases
  a1 Antitrypsin deficiency
  Hemochromatosis
  Wilson disease
  Benign recurrent intrahepatic
(stone, stricture, cancer)
  Biliary atresia
  Caroli disease
  Cryptosporidiosis
Drug-induced liver disease
  Hepatocellular patterns (isoniazid,
acetaminophen)
  Cholestatic patterns
(methyltestosterone)
  Mixed patterns (sulfonamides,
phenytoin)
  Micro- and macrovesicular steatosis
PART 10
Disorders of the Gastrointestinal System
(methotrexate, fialuridine)
Vascular injury
  Sinusoidal obstruction syndrome
  Budd-Chiari syndrome
  Ischemic hepatitis
  Passive congestion
  Portal vein thrombosis
  Nodular regenerative hyperplasia
Mass lesions
  Hepatocellular carcinoma
  Cholangiocarcinoma
  Adenoma
  Focal nodular hyperplasia
  Metastatic tumors
  Abscess
  Cysts
  Hemangioma
cholestasis
  Progressive familial intrahepatic
cholestasis, types I–III
  Others (galactosemia, tyrosinemia,
cystic fibrosis, Niemann-Pickdisease, Gaucher’s disease)
Alcohol-related liver disease
  Acute fatty liver
  Acute alcoholic hepatitis
  Laënnec cirrhosis
Nonalcoholic fatty livera
  Steatosis
  Steatohepatitis
Acute fatty liver of pregnancy
aAlso known as metabolic dysfunction–associated steatotic liver disease (MASLD).
disease presentation (hepatocellular, cholestatic, or mixed injury) as
well as on the specific etiologic diagnosis. Grading refers to assessment
of the severity or activity of disease—active or inactive as well as mild,
moderate, or severe. Staging refers to estimation of the point in the
course of the natural history of the disease, whether early or late; or
precirrhotic, cirrhotic, or end-stage. This chapter introduces general,
salient concepts in the evaluation of patients with liver disease that help
lead to the diagnoses discussed in subsequent chapters.
■
■CLINICAL HISTORY
The clinical history should focus on the symptoms of liver disease—
their nature, patterns of onset, and progression—and on potential risk
factors for liver disease. The manifestations of liver disease include
constitutional symptoms such as fatigue, weakness, nausea, poor appe­
tite, and malaise and the more liver-specific symptoms of jaundice,
dark urine, light stools, itching, abdominal pain, and bloating. Symp­
toms can also suggest the presence of cirrhosis, end-stage liver disease,
or complications of cirrhosis such as portal hypertension. Generally,
the constellation of symptoms and their patterns of onset, rather than
a specific symptom, point to an etiology.
Fatigue is the most common and most characteristic symptom of
liver disease. It is variously described as lethargy, weakness, listlessness,
malaise, increased need for sleep, lack of stamina, and poor energy.
The fatigue of liver disease typically arises after activity or exercise and
is rarely present or severe after adequate rest; that is, it is “afternoon”
rather than “morning” fatigue. Fatigue in liver disease is often intermit­
tent and variable in severity from hour to hour and day to day. In some
patients, it may not be clear whether fatigue is due to the liver disease
or due to other problems such as stress, anxiety, sleep disturbance, or
a concurrent illness.
Nausea occurs with more severe liver disease and may accompany
fatigue or be provoked by smelling food odors or eating fatty foods.
Vomiting can occur but is rarely persistent or prominent. Poor appetite
with weight loss occurs frequently in acute liver disease but is rare in
chronic disease except when cirrhosis is present and advanced. Diarrhea is
uncommon in liver disease except with severe jaundice, in which a lack
of bile acids reaching the intestine can lead to steatorrhea.
Right-upper-quadrant discomfort or ache (“liver pain”) occurs in
many liver diseases and is usually marked by tenderness over the liver
area. The pain arises from stretching or irritation of Glisson’s capsule,
which surrounds the liver and is rich in nerve endings. Severe pain is
most typical of gallbladder disease, liver abscess, and severe sinusoidal
obstruction syndrome (previously known as venoocclusive disease) but
is also an occasional accompaniment of acute hepatitis.
Itching occurs with acute liver disease, appearing early in obstruc­
tive jaundice (from biliary obstruction) or drug-induced cholestasis
and somewhat later in hepatocellular disease (acute hepatitis). Itching
also occurs in chronic liver diseases—typically the cholestatic forms
such as primary biliary cholangitis and sclerosing cholangitis, in which
it is often the presenting symptom, preceding the onset of jaundice.
However, itching can occur in any liver disease, particularly once cir­
rhosis develops.
Jaundice is the hallmark symptom of liver disease and perhaps the
most reliable marker of severity. Patients usually report darkening of
the urine before they notice scleral icterus. Jaundice is rarely detectable
with a bilirubin level <43 μmol/L (2.5 mg/dL). With severe cholestasis,
there will also be lightening of the color of the stools and steatorrhea.
Jaundice without dark urine usually indicates indirect (unconjugated)
hyperbilirubinemia and is typical of hemolytic anemia and the genetic
disorders of bilirubin conjugation, the common and benign form being
Gilbert syndrome and the rare and severe form being Crigler-Najjar
syndrome. Gilbert syndrome affects up to 5% of the general popula­
tion; the jaundice in this condition is more noticeable after fasting and
with stress.
Major risk factors for liver disease that should be sought in the clini­
cal history include details of alcohol use, medication use (including
herbal compounds, birth control pills, and over-the-counter medica­
tions), personal habits, sexual activity, travel, exposure to jaundiced or
other high-risk persons, injection drug use, recent surgery, remote or
recent transfusion of blood or blood products, occupation, accidental
exposure to blood or needlestick, and familial history of liver disease.
For assessing the risk of viral hepatitis, a careful history of sexual
activity is of particular importance and should include the number
of lifetime sexual partners and, for men, a history of having sex with
men. Sexual exposure is a common mode of spread of hepatitis B and
D but is uncommon for hepatitis C. A family history of hepatitis, liver
disease, and liver cancer is also important. Maternal-infant transmis­
sion occurs with both hepatitis B and C. Vertical spread of hepatitis B
can now be prevented by passive and active immunization of the infant
at birth. Additionally, antiviral therapy administered in the second or
third trimester of pregnancy is now recommended for mothers with
levels of hepatitis B virus DNA >200,000 IU/mL. Vertical spread of
hepatitis C is uncommon, but there are no reliable means of preven­
tion. Transmission is more common among HIV-co-infected mothers
and is also linked to prolonged and difficult labor and delivery, early
rupture of membranes, internal fetal monitoring, and a high maternal
viral load. A history of injection drug use, even in the remote past,
is of great importance in assessing the risk for hepatitis B, C, and D.
Injection drug use is the single most common risk factor for hepatitis
C. Transfusion with blood or blood products is no longer an important
risk factor for acute viral hepatitis due to screening of blood products.
However, blood transfusions received before the introduction of sensi­
tive enzyme immunoassays for antibody to hepatitis C virus in 1992
is an important risk factor for chronic hepatitis C. Blood transfusion
before 1986, when screening for antibody to hepatitis B core antigen
was introduced, is also a risk factor for hepatitis B. Travel to a develop­
ing area of the world, exposure to persons with jaundice, and exposure
to young children in day-care centers are risk factors for hepatitis A.
Tattooing and body piercing (for hepatitis B, C, and D) and eating
shellfish (for hepatitis A) are frequently mentioned but are actually
types of exposure that rarely lead to infection.
Hepatitis E is one of the more common causes of jaundice in Asia
and Africa but is uncommon in developed nations. In endemic areas,
transmission is usually through exposure to fecally contaminated
water. Non-travel-related (autochthonous) cases of hepatitis E have
been described in developed countries, including the United States.
These cases appear to be due to strains of hepatitis E virus that are
endemic in swine and some wild animals (genotypes 3 and 4). While
occasional cases are associated with eating raw or undercooked pork
or game (deer and wild boars), most cases of hepatitis E occur without
known exposure, predominantly in elderly men without typical risk
factors for viral hepatitis. Hepatitis E infection can become chronic in
immunosuppressed individuals (such as transplant recipients, patients
receiving chemotherapy, or patients with HIV infection), in whom it
presents with abnormal serum enzymes in the absence of markers of
hepatitis B or C.
A history of alcohol intake is important in assessing the cause of
liver disease and in planning management and recommendations. In
the United States, for example, at least 70% of adults drink alcohol
to some degree, but significant alcohol intake is less common; in
population-based surveys, only 5% of individuals have more than two
drinks per day, the average drink representing 11–15 g of alcohol. Alco­
hol consumption associated with an increased rate of alcoholic liver
disease is probably more than two drinks (22–30 g) per day in women
and three drinks (33–45 g) in men. Most patients with alcoholic cir­
rhosis have a much higher daily intake and have drunk excessively for
≥10 years before onset of liver disease. In assessing alcohol intake, the
history should also focus on whether alcohol abuse or dependence is
present. Alcoholism is usually defined by the behavioral patterns and
consequences of alcohol intake, not by the amount. Abuse is defined
by a repetitive pattern of drinking alcohol that has adverse effects on
social, family, occupational, or health status. Dependence is defined by
alcohol-seeking behavior, despite its adverse effects. Many alcoholics
demonstrate both dependence and abuse, and dependence is consid­
ered the more serious and advanced form of alcoholism. A clinically
helpful approach to diagnosis of alcohol dependence and abuse is the
use of the CAGE questionnaire (Table 347-2), which is recommended
for all medical history-taking.
Family history can be helpful in assessing liver disease. Familial
causes of liver disease include Wilson disease; hemochromatosis and
α1 antitrypsin deficiency; and the less common inherited pediatric
liver diseases—that is, familial intrahepatic cholestasis, benign recur­
rent intrahepatic cholestasis, and Alagille syndrome. Onset of severe
TABLE 347-2  CAGE Questionsa
ACRONYM
QUESTION
C
Have you ever felt you ought to cut down on your drinking?
A
Have people annoyed you by criticizing your drinking?
G
Have you ever felt guilty or bad about your drinking?
E
Have you ever had a drink first thing in the morning to steady
your nerves or get rid of a hangover (eye-opener)?
aOne “yes” response should raise suspicion of an alcohol use problem, and more
than one is a strong indication of abuse or dependence.
liver disease in childhood or adolescence in conjunction with a family
history of liver disease or neuropsychiatric disturbance should lead to
investigation for Wilson disease. A family history of cirrhosis, diabetes,
or endocrine failure and the appearance of liver disease in adulthood
suggest hemochromatosis and should prompt investigation of iron
status. Abnormal iron studies in adult patients warrant genotyping of
the HFE gene for the C282Y and H63D mutations typical of genetic
hemochromatosis. In children and adolescents with iron overload,
other non-HFE causes of hemochromatosis should be sought. A family
history of emphysema should lead to investigation of α1 antitrypsin
levels and, if low, for protease inhibitor (Pi) genotype.
■
■PHYSICAL EXAMINATION
The physical examination rarely uncovers evidence of liver dysfunction
in a patient without symptoms or abnormal laboratory findings, nor
are most signs of liver disease specific to one diagnosis. Thus, the phys­
ical examination complements rather than replaces the need for other
diagnostic approaches. In many patients, the physical examination is
normal unless the disease is acute or severe and advanced. Neverthe­
less, the physical examination is important in that it can yield the first
evidence of hepatic failure, portal hypertension, and liver decompen­
sation. In addition, the physical examination can reveal signs—related
either to risk factors or to associated diseases or findings—that point
to a specific diagnosis.
Typical physical findings in liver disease are icterus, hepatomegaly,
hepatic tenderness, splenomegaly, spider angiomata, palmar erythema,
and skin excoriations. Signs of advanced disease include muscle wast­
ing, ascites, edema, dilated abdominal veins, hepatic fetor, asterixis,
mental confusion, stupor, and coma. In male patients with cirrhosis,
particularly that related to alcohol use, signs of hyperestrogenemia
such as gynecomastia, testicular atrophy, and loss of male-pattern hair
distribution may be found.
CHAPTER 347
Icterus is best appreciated when the sclera is inspected under natu­
ral light. In fair-skinned individuals, a yellow tinge to the skin may be
obvious. In dark-skinned individuals, examination of the mucous
membranes below the tongue can demonstrate jaundice. Jaundice is
rarely detectable if the serum bilirubin level is <43 μmol/L (2.5 mg/dL)
but may remain detectable below this level during recovery from jaun­
dice (because of protein and tissue binding of conjugated bilirubin).
Approach to the Patient with Liver Disease
Spider angiomata and palmar erythema occur in both acute and
chronic liver disease; these manifestations may be especially prominent
in persons with cirrhosis but can develop in normal individuals and are
frequently found during pregnancy. Spider angiomata are superficial,
tortuous arterioles, and—unlike simple telangiectasias—typically fill
from the center outward. Spider angiomata occur only on the arms,
face, and upper torso; they can be pulsatile and may be difficult to
detect in dark-skinned individuals.
Hepatomegaly is not a very reliable sign of liver disease because of
variability in the liver’s size and shape and the physical impediments to
assessment of liver size by percussion and palpation. Marked hepato­
megaly is typical of cirrhosis, sinusoidal obstruction syndrome, infil­
trative disorders such as amyloidosis, metastatic or primary cancers of
the liver, and alcoholic hepatitis. Careful assessment of the liver edge
may also reveal unusual firmness, irregularity of the surface, or frank
nodules. Perhaps the most reliable physical finding in the liver exami­
nation is hepatic tenderness. Discomfort when the liver is touched or
pressed upon should be carefully sought with percussive comparison of
the right and left upper quadrants.
Splenomegaly, which occurs in many medical conditions, can be
a subtle but significant physical finding in chronic liver disease and
suggests underlying cirrhosis. The availability of ultrasound (US)
methods for assessment of the spleen allows confirmation of the physi­
cal finding.
Signs of advanced liver disease include muscle wasting and weight
loss as well as hepatomegaly, bruising, ascites, and edema. Ascites
is best appreciated by attempts to detect shifting dullness by careful
percussion. US examination will confirm the finding of ascites in
equivocal cases. Peripheral edema can occur with or without ascites. In
patients with advanced liver disease, other factors frequently contribute
to edema formation, including hypoalbuminemia, venous insuffi­
ciency, heart failure, and medications.
Hepatic failure is defined as the occurrence of signs or symptoms of
hepatic encephalopathy in a person with severe acute or chronic liver
disease. The first signs of hepatic encephalopathy can be subtle and
nonspecific—change in sleep patterns, change in personality, irritabil­
ity, and mental dullness. Thereafter, confusion, disorientation, stupor,
and eventually coma supervene. In acute liver failure, excitability and
mania may be present. Physical findings include asterixis and flapping
tremors of the body and tongue. Fetor hepaticus refers to the slightly
sweet, ammoniacal odor that can develop in patients with liver failure,
particularly if there is portal-venous shunting of blood around the liver.
Other causes of coma and disorientation should be excluded, mainly
electrolyte imbalances, sedative use, and renal or respiratory failure.
The appearance of hepatic encephalopathy during acute hepatitis is the
major criterion for diagnosis of acute liver failure and indicates a poor
prognosis. In chronic liver disease, encephalopathy is usually triggered
by a medical complication such as gastrointestinal bleeding, overdiure­
sis, uremia, dehydration, electrolyte imbalance, infection, constipation,
or use of narcotic analgesics.
A helpful measure of hepatic encephalopathy is a careful mental
status examination and use of the trail-making test, which consists of
a series of 25 numbered circles that the patient is asked to connect as
rapidly as possible using a pencil. The normal range for the connectthe-dot test is 15–30 s; it is considerably longer in patients with early
hepatic encephalopathy. Other tests include drawing of abstract objects
or comparison of a signature to previous examples. More sophisticated
testing—for example, with electroencephalography and visual evoked
potentials—can detect mild forms of encephalopathy but are rarely
clinically useful.
PART 10
Disorders of the Gastrointestinal System
Other signs of advanced liver disease include umbilical hernia from
ascites, hydrothorax, prominent veins over the abdomen, and caput
medusa, a condition that consists of collateral veins radiating from the
umbilicus and results from recanulation of the umbilical vein. Wid­
ened pulse pressure and signs of a hyperdynamic circulation can occur
in patients with cirrhosis as a result of fluid and sodium retention,
increased cardiac output, and reduced peripheral resistance. Patients
with long-standing cirrhosis and portal hypertension are prone to
develop the hepatopulmonary syndrome, which is defined by the triad
of liver disease, hypoxemia, and pulmonary arteriovenous shunting.
The hepatopulmonary syndrome is characterized by platypnea and
orthodeoxia: shortness of breath and oxygen desaturation that occur
paradoxically upon the assumption of an upright position. Measure­
ment of oxygen saturation by pulse oximetry is a reliable screening test
for hepatopulmonary syndrome.
Several skin disorders and changes are common in liver disease.
Hyperpigmentation is typical of advanced chronic cholestatic diseases
such as primary biliary cholangitis and sclerosing cholangitis. In these
same conditions, xanthelasma and tendon xanthomata occur as a result
of retention and high serum levels of lipids and cholesterol. Slate-gray
pigmentation of the skin is also seen with hemochromatosis if iron
levels are high for a prolonged period. Mucocutaneous vasculitis with
palpable purpura, especially on the lower extremities, is typical of
cryoglobulinemia of chronic hepatitis C but can also occur in chronic
hepatitis B.
Some physical signs point to specific liver diseases. Kayser-Fleischer
rings occur in Wilson disease and consist of a golden-brown copper
pigment deposited in Descemet’s membrane at the periphery of the
cornea; they are best seen by slit-lamp examination. Dupuytren con­
tracture and parotid enlargement are suggestive of alcohol use disorder
and alcohol-related liver disease. In metastatic liver disease or primary
hepatocellular carcinoma (HCC), signs of cachexia and wasting as well
as firm hepatomegaly and a hepatic bruit may be prominent.
■
■DIAGNOSIS OF LIVER DISEASE
The key diagnostic tests of major causes of acute and chronic liver dis­
ease are outlined in Table 347-3, and an algorithm for evaluation of the
patient with suspected liver disease is shown in Fig. 347-1. Specifics of
diagnosis are discussed in later chapters. The most common causes of
TABLE 347-3  Important Diagnostic Tests in Common Liver Diseases
DISEASE
DIAGNOSTIC TEST
Hepatitis A
Anti-HAV IgM
Hepatitis B
  Acute
HBsAg and anti-HBc IgM
  Chronic
HBsAg and HBeAg and/or HBV DNA
Hepatitis C
Anti-HCV and HCV RNA
Hepatitis D (delta)
HBsAg and anti-HDV
Hepatitis E
Anti-HEV IgM and HEV RNA
Autoimmune hepatitis
ANA or SMA, elevated IgG levels, and
compatible histology
Primary biliary cholangitis
Mitochondrial antibody, elevated IgM levels,
and compatible histology
Primary sclerosing cholangitis
P-ANCA, cholangiography
Drug-induced liver disease
History of drug ingestion
Alcohol-related liver disease
History of excessive alcohol intake and
compatible histology
Nonalcoholic steatohepatitisa
Ultrasound or CT evidence of fatty liver and
compatible histology
a1 Antitrypsin disease
Reduced α1 antitrypsin levels, phenotype PiZZ
or PiSZ
Wilson disease
Decreased serum ceruloplasmin and increased
urinary copper; increased hepatic copper level
Hemochromatosis
Elevated iron saturation and serum ferritin;
genetic testing for HFE gene mutations
Hepatocellular cancer
Elevated α-fetoprotein level (to >500 ng/mL);
ultrasound or CT image of mass
aAlso known as metabolic dysfunction–associated steatohepatitis (MASH).
Abbreviations: ANA, antinuclear antibody; anti-HBc, antibody to hepatitis B core
(antigen); CT, computed tomography; HAV, HBV, HCV, HDV, HEV, hepatitis A, B, C, D,
E virus; HBeAg, hepatitis B e antigen; HBsAg, hepatitis B surface antigen; P-ANCA,
peripheral antineutrophil cytoplasmic antibody; SMA, smooth-muscle antibody.
acute liver disease are viral hepatitis (particularly hepatitis A, B, and C),
drug-induced liver injury, cholangitis, and alcohol-related liver disease.
Liver biopsy usually is not needed for the diagnosis and management
of acute liver disease, exceptions being situations where the diagnosis
remains unclear despite thorough clinical and laboratory investigation.
Liver biopsy can be helpful in diagnosing drug-induced liver disease
and acute alcoholic hepatitis.
The most common causes of chronic liver disease, in general order
of frequency, are chronic hepatitis C, alcoholic liver disease, nonal­
coholic steatohepatitis (also called metabolic dysfunction–associated
steatohepatitis), chronic hepatitis B, autoimmune hepatitis, sclerosing
cholangitis, primary biliary cholangitis, hemochromatosis, and Wilson
disease. Hepatitis E virus is a rare cause of chronic hepatitis, with cases
occurring mostly in persons who are immunosuppressed or immuno­
deficient. Strict diagnostic criteria have not been developed for most
liver diseases, but liver biopsy plays an important role in the diagnosis
of autoimmune hepatitis, primary biliary cholangitis, nonalcoholic
and alcoholic steatohepatitis, and Wilson disease (with a quantitative
hepatic copper level in the last instance).
Laboratory Testing 
Diagnosis of liver disease is greatly aided by
the availability of reliable and sensitive tests of liver injury and func­
tion. A typical battery of blood tests used for initial assessment of
liver disease includes measurement of levels of serum alanine (ALT)
and aspartate (AST) aminotransferases, alkaline phosphatase (AlkP),
direct and total serum bilirubin and albumin, and prothrombin time.
The pattern of abnormalities generally points to hepatocellular versus
cholestatic liver disease and helps determine whether the disease is
acute or chronic and whether cirrhosis and hepatic failure are present.
Based on these results, further testing over time may be necessary.
Other laboratory tests may be helpful, such as γ-glutamyl transpepti­
dase (γGT) to define whether AlkP elevations are due to liver disease;
hepatitis serology to define the type of viral hepatitis; and autoimmune
markers to diagnose primary biliary cholangitis (antimitochondrial
Suspected liver disease
Abnormal liver tests
Acute
<6 months
Chronic
6 months
Hepatitic: ⇑⇑ALT
Mixed: ↑ALT,
↑AlkP
Hepatitic: ⇑⇑ALT
Mixed: ↑ALT,
↑AlkP
Cholestatic:
⇑⇑AlkP,
⇑⇑gGT,
↑ALT
Diagnostic
evaluation
IgM Anti-HAV
HBsAg
IgM Anti-HBc
Anti-HCV
ANA, SMA
Monospot,
heterophile
Ceruloplasmin
Alcohol history
Drug history
Diagnostic
evaluation
HBsAg
Anti-HCV
Fe saturation,
ferritin
Ceruloplasmin
α1AT
ANA, SMA
Ultrasound
Alcohol history
Diagnostic
evaluation
AMA
Drug history
Ultrasound/MRI
MRCP/ERCP
Liver biopsy in acute liver disease:
Reserved for patients in whom the diagnosis
remains unclear despite medical evaluation
Liver biopsy in chronic liver disease:
Often valuable for diagnosis as well as
staging and grading liver disease
FIGURE 347-1  Algorithm for evaluation of abnormal liver tests. For patients with suspected
liver disease, an appropriate approach to evaluation is initial routine liver testing—for
example, measurement of serum bilirubin, albumin, alanine aminotransferase (ALT), aspartate
aminotransferase (AST), and alkaline phosphatase (AlkP). These results (sometimes complemented
by testing of γ-glutamyl transpeptidase [γGT]) will establish whether the pattern of abnormalities is
hepatic, cholestatic, or mixed. In addition, the duration of symptoms or abnormalities will indicate
whether the disease is acute or chronic. If the disease is acute and if history, laboratory tests, and
imaging studies do not reveal a diagnosis, liver biopsy is appropriate to help establish the diagnosis.
If the disease is chronic, liver biopsy can be helpful not only for diagnosis but also for grading of
the activity and staging the progression of disease. This approach is generally applicable to
patients without immune deficiency. In patients with HIV infection or recipients of bone marrow or
solid organ transplants, the diagnostic evaluation should also include evaluation for opportunistic
infections (e.g., with adenovirus, cytomegalovirus, Coccidioides, hepatitis E virus) as well as for
vascular and immunologic conditions (venoocclusive disease, graft-versus-host disease). α1AT,
α1 antitrypsin; AMA, antimitochondrial antibody; ANA, antinuclear antibody; anti-HBc, antibody to
hepatitis B core (antigen); ERCP, endoscopic retrograde cholangiopancreatography; HAV, hepatitis
A virus; HBsAg, hepatitis B surface antigen; HCV, hepatitis C virus; MRCP, magnetic resonance
cholangiopancreatography; P-ANCA, peripheral antineutrophil cytoplasmic antibody; SMA, smoothmuscle antibody.
TABLE 347-4  Diagnostic Tests to Assess Liver Fat
IMAGING MODALITY
ADVANTAGES
DISADVANTAGES
CLINICAL UTILITY
Ultrasound
No radiation
Widely available
Transient elastography
with controlled attenuation
parameter
No radiation
Point-of-care assessment of liver fat
Provides semiquantitative assessment
of fat severity
Computed tomography
Rapid assessment
Non–operator dependent
Quantitative assessment of fat severity
Magnetic resonance imaging–
proton density fat fraction
Direct assessment of liver fat
Highly sensitive and specific
antibody), sclerosing cholangitis (peripheral antineutro­
phil cytoplasmic antibody), and autoimmune hepatitis
(antinuclear, smooth-muscle, and liver-kidney micro­
somal antibody). A simple delineation of laboratory
abnormalities and common liver diseases is given in
Table 347-3.
The use and interpretation of liver function tests
are summarized in Chap. 348.
Diagnostic Imaging 
Great advances have been
made in hepatobiliary imaging, although no method is
adequately accurate in demonstrating underlying cir­
rhosis in its early stages. Of the many modalities avail­
able for imaging the liver, US, computed tomography
(CT), and magnetic resonance imaging (MRI) are the
most commonly employed and are complementary to
one another. In general, US and CT are highly sensitive
for detecting biliary duct dilation and are the first-line
options for investigating cases of suspected obstructive
jaundice. All three modalities can detect a fatty liver,
which appears bright on imaging studies. Modifications
of CT and MRI can be used to quantify liver fat, and this
information may ultimately be valuable in monitoring
response to therapy in patients with fatty liver disease.
Advantages, disadvantages, and clinical utility of each
modality are presented in Table 347-4. Magnetic reso­
nance cholangiopancreatography (MRCP) and endo­
scopic retrograde cholangiopancreatography (ERCP) are
the procedures of choice for visualization of the biliary
tree. MRCP offers several advantages over ERCP: there is
no need for contrast media or ionizing radiation, images
can be acquired faster, the procedure is less operator
dependent, and it carries no risk of pancreatitis. MRCP
is superior to US and CT for detecting choledocholithia­
sis but is less specific. MRCP is useful in the diagnosis of
bile duct obstruction and congenital biliary abnormali­
ties, but ERCP is considered more valuable in evaluating
ampullary lesions and primary sclerosing cholangitis.
ERCP permits biopsy, direct visualization of the ampulla
and common bile duct, and intraductal ultrasonography
and brushings for cytologic evaluation of malignancy.
It also provides several therapeutic options in patients
with obstructive jaundice, such as sphincterotomy, stone
extraction, and placement of nasobiliary catheters and
biliary stents.
Cholestatic:
⇑⇑AlkP,
⇑⇑gGT,
↑ALT
Diagnostic
evaluation
Drug history
AMA
P-ANCA
Ultrasound
MRCP/ERCP
CHAPTER 347
Approach to the Patient with Liver Disease
Doppler US and MRI are used to assess hepatic
vasculature and hemodynamics and to monitor surgi­
cally or radiologically placed vascular shunts, including
Operator dependent
Imprecise qualitative assessment of fat severity,
particularly mild steatosis
Initial screening test for suspected
liver fat
Requires special software
No reliable cutoff for diagnosis of liver fat
Imprecise qualitative assessment of fat severity
Alternate screening test for suspected
liver fat if available
Requires radiation
Quantification of fat requires specific protocols
Imprecise quantitative assessment of fat
severity, particularly mild steatosis
Not recommended for clinical
assessment of liver fat due to need for
radiation exposure and low sensitivity
for mild fat
Relatively limited accessibility
Test of choice for quantitative
assessment of liver fat if available
transjugular intrahepatic portosystemic shunts. Multidetector or spiral
CT and MRI with contrast enhancement are the procedures of choice
for the identification and evaluation of hepatic masses, the staging of
liver tumors, and preoperative assessment. With regard to mass lesions,
the sensitivity of hepatic imaging continues to increase; unfortunately,
specificity remains a problem, and often two and sometimes three
studies are needed before a diagnosis can be reached. An emerging
imaging modality for the investigation of hepatic lesions is contrastenhanced US. This procedure permits enhancement of liver lesions
in a similar fashion as contrast-enhanced, cross-sectional CT or MRI.
Major advantages are real-time assessment of liver perfusion through­
out the vascular phases without risk of nephrotoxicity and radiation
exposure. Other advantages are its widespread availability and lower
cost. Limitations include body habitus of the patient and skill of the
operator.
US is the recommended modality for HCC screening. Contrastenhanced US, CT, and MRI are appropriate for further investigation
of lesions detected on screening US. The American College of Radi­
ologists has developed a Liver Imaging Reporting and Data System
(LI-RADS) to standardize the reporting and data collection of CT,
MRI, and contrast-enhanced US imaging for HCC. This system allows
for more consistent reporting and reduces imaging interpretation vari­
ability and errors.
Recently, several US-based elastographic techniques have been
developed and approved for the measurement of hepatic stiffness,
providing an indirect assessment of fibrosis and cirrhosis. The most
commonly used approaches in clinical practice include transient
elastography, acoustic radiation force impulse imaging, shear-wave
elasticity imaging, and supersonic shear imaging. These techniques
can eliminate the need for liver biopsy if the only indication for the test
is the assessment of disease stage. Magnetic resonance elastography is
more sensitive than US elastography but is also more expensive and
requires advanced scheduling and special equipment. Finally, interven­
tional radiologic techniques allow for the biopsy of solitary lesions, the
radiofrequency ablation and chemoembolization of cancerous lesions,
the insertion of drains into hepatic abscesses, the measurement of
portal pressure, and the creation of vascular shunts in patients with
portal hypertension. Which modality to use depends on factors such as
availability, cost, and experience of the radiologist with each technique.
PART 10
Disorders of the Gastrointestinal System
Liver Biopsy 
Liver biopsy remains the gold standard in the evalu­
ation of patients with liver disease, particularly chronic liver disease.
Liver biopsy is necessary for diagnosis in selected instances but is
more often useful for assessment of the severity (grade) and stage of
liver damage, prediction of prognosis, and monitoring of the response
to treatment. The size of the liver biopsy sample is an important
determinant of reliability; a length of 1.5–2 cm with 10 portal tracts
is necessary for accurate assessment of fibrosis. Because liver biopsy
is an invasive procedure and not without complications, it should be
used only when it will contribute materially to decisions about man­
agement and therapy. In the future, noninvasive means of assessing
disease activity (batteries of blood tests) and fibrosis (elastography and
fibrosis markers) may replace liver biopsy for the staging and grading
of disease.
■
■GRADING AND STAGING OF LIVER DISEASE
Grading refers to an assessment of the severity or activity of liver dis­
ease, whether acute or chronic; active or inactive; and mild, moderate,
or severe. Liver biopsy is the most accurate means of assessing severity,
particularly in chronic liver disease. Serum aminotransferase levels
serve as convenient and noninvasive markers for disease activity but do
not always reliably reflect disease severity. Thus, normal serum amino­
transferase levels in patients with hepatitis B surface antigen in serum
may indicate the inactive carrier state or may reflect mild chronic hepa­
titis B or hepatitis B with fluctuating disease activity. Serum testing for
hepatitis B e antigen and hepatitis B virus DNA can help sort out these
different patterns, but these markers can also fluctuate and change
over time. Similarly, in chronic hepatitis C, serum aminotransferase
levels can be normal despite moderate disease activity. Finally, in both
alcoholic and nonalcoholic steatohepatitis, aminotransferase levels are
quite unreliable in reflecting severity. In these conditions, liver biopsy
is helpful in guiding management and identifying appropriate therapy,
particularly if treatment is difficult, prolonged, and expensive. Of the
several well-verified numerical scales for histologic grading of chronic
liver disease, the most commonly used are the METAVIR and the his­
tology activity index.
Liver biopsy is also the most accurate means of assessing stage of
disease as early or advanced and as precirrhotic or cirrhotic. Staging of
disease pertains largely to chronic liver diseases in which progression
to cirrhosis and end-stage disease can occur but may require years or
decades. Clinical features, biochemical tests, and hepatic imaging stud­
ies are helpful in assessing stage but generally become abnormal only
in the middle to late stages of cirrhosis. Noninvasive tests that suggest
advanced fibrosis include mild elevations of bilirubin, prolongation
of prothrombin time, slight decreases in serum albumin, and mild
thrombocytopenia (which is often the first indication of worsen­
ing fibrosis). Combinations of blood test results that include clinical
features, routine laboratory tests, and special laboratory tests such as
serum proteins or small molecules that are affected by or involved with
fibrogenesis have been used to create models for predicting advanced
liver disease, but these models are not reliable enough to use on a
regular basis or for repeated measures and only separate advanced
from early disease (Table 347-5). Recently, elastography and nonin­
vasive breath tests using 13C-labeled compounds have been proposed
as a means of detecting early stages of fibrosis and liver dysfunction,
but their reliability and reproducibility remain to be proven. A major
limitation of noninvasive markers is that they can be affected by disease
activity. Even elastography is limited in this regard, in that it measures
liver stiffness, not fibrosis per se, and can be affected by inflammation,
edema, hepatocyte necrosis, and intrasinusoidal cellularity (inflamma­
tory, malignant, or sickled cells). Thus, at present, mild to moderate
stages of hepatic fibrosis are detectable only by liver biopsy. In the
assessment of stage, the degree of fibrosis is usually used as the quanti­
tative measure. The amount of fibrosis is generally staged on a scale of
0 to 4+ (METAVIR scale) or 0 to 6+ (Ishak scale). The importance of
staging relates primarily to prognosis, recommendation of therapy, and
optimal management to prevent complications of chronic liver disease.
Patients with cirrhosis are candidates for screening and surveillance for
esophageal varices and HCC. Patients without advanced fibrosis need
not undergo screening.
Once cirrhosis develops, other scoring systems are employed to
assess compensated versus decompensated disease and prognosis. The
first staging system used for this purpose was the modified Child-Pugh
TABLE 347-5  Selected Noninvasive Methods of Assessing Hepatic
Fibrosis and Cirrhosis
ADVANCED
FIBROSIS
CIRRHOSIS
METHOD
PARAMETERS
APRI
AST, platelet count
1
1.5 (1–2)
ELF
Age, hyaluronic acid, MMP-3, TIMP-1
7.7
9.3
FIB-4
Age, AST, ALT, platelet count
1.45
3.25
Fibro testa
Haptoglobin, α2-macroglobulin,
apolipoprotein A1, γGT, total bilirubin
0.45
0.63
TE
Measures speed of a shear wave
generated by vibration through liver
tissue
7.3 kPa
15 kPa
(9–26.5 kPa)
ARFI
Measures speed of shear wave
generated by acoustic radiation force
through liver tissue
1.3 m/s
1.87 m/s
aPatented models.
Note: The cut points presented in the table were mostly derived from patients with
chronic hepatitis C. The cut points for the noninvasive models and tests presented
in the table vary among different liver diseases and among patients with the same
disease among different populations.
Abbreviations: ALT, alanine aminotransferase; APRI, AST-to-platelet ratio; ARFI,
acoustic radiation force imaging; AST, aspartate aminotransferase; ELF, enhanced
liver fibrosis panel; γGT, γ-glutamyl transpeptidase; MMP-3, metalloproteinase-3;
TIMP-1, tissue inhibitor of metalloproteinase-1; TE, transient elastography.

20 - 348 Evaluation of Liver Function
348 Evaluation of Liver Function
TABLE 347-6  Child-Pugh Classification of Cirrhosis
POINTS TOWARD TOTAL SCORE
FACTOR
UNITS
Serum bilirubin
μmol/L
<34
34–51
51
mg/dL
<2.0
2.0–3.0
3.0
Serum albumin
g/L
35
30–35
<30
g/dL
3.5
3.0–3.5
<3.0
Prothrombin time
Seconds
prolonged
<4
4–6
6
INRa
<1.7
1.7–2.3
2.3
Ascites
None
Easily
controlled
Poorly
controlled
Hepatic
encephalopathy
None
Minimal
Advanced
aInternational normalized ratio.
Note: The Child-Pugh score is calculated by adding the scores for the five factors
and can range from 5 to 15. The resulting Child-Pugh class can be A (a score of
5–6), B (7–9), or C (≥10). Decompensation indicates cirrhosis, with a Child-Pugh
score of ≥7 (class B). This level has been the accepted criterion for listing a patient
for liver transplantation.
classification, with a scoring system of 5–15: scores of 5 and 6 represent
Child-Pugh class A (consistent with “compensated cirrhosis”), scores
of 7–9 represent class B, and scores of 10–15 represent class C (Table
347-6). This scoring system was initially devised to stratify patients
with cirrhosis into risk groups before portal decompressive surgery.
The Child-Pugh score is a reasonably reliable predictor of survival in
many liver diseases and predicts the likelihood of major complications
of cirrhosis, such as bleeding from varices and spontaneous bacterial
peritonitis. This classification scheme was used to assess prognosis in
cirrhosis and to provide standard criteria for listing a patient as a can­
didate for liver transplantation (Child-Pugh class B). The Child-Pugh
system has been replaced by the Model for End-Stage Liver Disease
(MELD) system for the latter purpose. The MELD score is a prospec­
tively derived system designed to predict the short-term mortality of
patients awaiting orthotopic liver transplant and following a transjugu­
lar intrahepatic portosystemic shunt (TIPS). This score is calculated
using three readily available objective variables: the prothrombin time
expressed as the international normalized ratio (INR), the serum bili­
rubin level, and the serum creatinine concentration. The ability of the
MELD score to predict outcome after liver transplantation is regularly
monitored and was modified to increase its accuracy and improve allo­
cation of donated livers. These modifications include serum sodium
concentration as a factor in the model and a reweighting of the MELD
components. A separate scoring system, the Pediatric End-Stage Liver
Disease (PELD) score, is used for children (<12 years old). Transient
elastography has also been used to stage cirrhosis and has been shown
to be useful in predicting complications such as variceal hemorrhage,
ascites development, and liver-related death.
The MELD system provides a more objective means of assess­
ing disease severity and has less center-to-center variation than the
Child-Pugh score as well as a wider range of values. The MELD and
PELD systems are currently used to establish priority listing for liver
transplantation in the United States. Convenient MELD and PELD cal­
culators are available via the Internet (https://optn.transplant.hrsa.gov).
■
■NONSPECIFIC ISSUES IN THE MANAGEMENT OF
PATIENTS WITH LIVER DISEASE
Specifics on the management of different forms of acute or chronic
liver disease are supplied in subsequent chapters, but certain issues
are applicable to any patient with liver disease. These issues include
advice regarding alcohol use, medication use, vaccination, and sur­
veillance for certain liver diseases and complications of liver disease.
Alcohol should be used sparingly, if at all, by patients with liver dis­
ease. Abstinence from alcohol should be encouraged for all patients
with alcohol-related liver disease, patients with cirrhosis, and patients
receiving interferon-based therapy for hepatitis B or D and during
antiviral therapy of hepatitis C. With regard to vaccinations, all patients
with liver disease should receive hepatitis A and hepatitis B vaccine, if
not previously vaccinated, and screened for hepatitis C if not already
done. Adherence to the recommendations of the Centers for Disease
Control and Prevention (CDC) on other adult vaccinations should
also be encouraged. Patients with liver disease should exercise caution
in using any medications other than those that are most necessary.
Drug-induced hepatotoxicity can mimic many forms of liver disease
and can cause exacerbations of chronic hepatitis and cirrhosis; drugs
should be suspected in any situation in which the cause of exacerba­
tion is unknown. Finally, consideration should be given to surveillance
for complications of chronic liver disease such as variceal hemorrhage
and HCC. Cirrhosis warrants upper endoscopy to assess the presence
of varices, and the patient should receive chronic therapy with beta
blockers or should be offered endoscopic obliteration if large varices
are found. Moreover, cirrhosis warrants screening and long-term sur­
veillance for development of HCC. While the optimal regimen for such
surveillance has not been established, an appropriate approach is US of
the liver at 6- to 12-month intervals.
■
■FURTHER READING
Jeng W-J et al: Hepatitis B. Lancet 401:1039, 2023.
Manns MP et al: Breakthroughs in hepatitis C research: From discovery
to cure. Gastroenterol Hepatol 19:533, 2022.
Powell EE et al: Non-alcoholic fatty liver disease. Lancet 397:2212,
2021.
Tapper EB, Lok AS: Use of liver imaging and biopsy in clinical practice.
N Engl J Med 377:756, 2017.
CHAPTER 348
Evaluation of Liver Function
Emily D. Bethea, Daniel S. Pratt
Evaluation of Liver
Function
There are a number of tests that can be used to evaluate liver function.
These tests include biochemical tests, radiologic tests, and pathologic tests.
Serum biochemical tests, also commonly referred to as “liver
function tests,” can be used to (1) detect the presence of liver disease,
(2) distinguish among different types of liver disorders, (3) gauge the
extent of liver damage, and (4) follow the response to treatment. How­
ever, serum biochemical tests have shortcomings. They lack sensitivity
and specificity; they can be normal in patients with serious liver disease
and abnormal in patients with diseases that do not affect the liver. Liver
tests rarely suggest a specific diagnosis; rather, they suggest a general
category of liver disease, such as hepatocellular or cholestatic, which
then further directs the evaluation. The liver carries out thousands
of biochemical functions, most of which cannot be easily measured
by blood tests. Laboratory tests measure only a limited number of
these functions. In fact, many tests, such as the aminotransferases and
alkaline phosphatase, do not measure liver function at all. Rather, they
detect liver cell damage or interference with bile flow. Thus, no one
biochemical test enables the clinician to accurately assess the liver’s
total functional capacity.
To increase the sensitivity and the specificity of biochemical tests
in the detection of liver disease, it is best to use them as a battery.
Tests usually employed in clinical practice include the bilirubin, ami­
notransferases, alkaline phosphatase, albumin, and prothrombin time
tests. When more than one of these tests provide abnormal findings
or the findings are persistently abnormal on serial determinations, the
probability of liver disease is high. When all test results are normal, the
probability of missing occult liver disease is low.
Serum Bilirubin 
(See also Chap. 52) Bilirubin, a breakdown
product of the porphyrin ring of heme-containing proteins, is found
in the blood in two fractions—conjugated and unconjugated. The
unconjugated fraction, also termed the indirect fraction, is insoluble in
water and is bound to albumin in the blood. The conjugated (direct)
bilirubin fraction is water-soluble and can therefore be excreted by the
kidney. Normal values of total serum bilirubin are reported between
1 and 1.5 mg/dL with 95% of a normal population falling between
0.2 and 0.9 mg/dL. If the direct-acting fraction is <15% of the total,
the bilirubin can be considered to all be indirect. The most frequently
reported upper limit of normal for conjugated bilirubin is 0.3 mg/dL.
Elevation of the unconjugated fraction of bilirubin is rarely due to
liver disease. An isolated elevation of unconjugated bilirubin is seen
primarily in hemolytic disorders and in a number of genetic conditions
such as Crigler-Najjar and Gilbert’s syndromes (Chap. 52). Isolated
unconjugated hyperbilirubinemia (bilirubin elevated but <15% direct)
should prompt a workup for hemolysis (Fig. 348-1). In the absence of
hemolysis, an isolated, unconjugated hyperbilirubinemia in an other­
wise healthy patient can be attributed to Gilbert’s syndrome, and no
further evaluation is required.
In contrast, conjugated hyperbilirubinemia almost always implies
liver or biliary tract disease. The rate-limiting step in bilirubin
metabolism is not conjugation of bilirubin, but rather the transport
of conjugated bilirubin into the bile canaliculi. Thus, elevation of the
conjugated fraction may be seen in any type of liver disease including
fulminant liver failure. In most liver diseases, both conjugated and
unconjugated fractions of the bilirubin tend to be elevated. Except in
PART 10
Disorders of the Gastrointestinal System
Liver Tests
Isolated elevation
of the bilirubin
Fractionate
bilirubin
Ducts not
dilated
Dilated ducts
15% Direct
Check AMA
<15% Direct
AMA positive
AMA
negative
Evaluation for
hemolysis
Dubin-Johnson or
Rotor syndrome
W/U
positive
W/U negative
Hepatocellular
pattern
(see Table 348-1)
Hemolysis
Gilbert’s
syndrome
Review drug list
Hepatitis C antibody
Hepatitis B surface Ag
Iron, TIBC, ferritin
ANA, SPEP
Ceruloplasmin
(if patient <40)
Ultrasound to look
for fatty liver
W/U negative
R/O Celiac disease
Consider other
nonhepatic cause
W/U negative
Consider liver biopsy
FIGURE 348-1  Algorithm for the evaluation of chronically abnormal liver tests. Ag, antigen; AMA, antimitochondrial antibody; ANA, antinuclear antibody; Bx, biopsy; CT,
computed tomography; ERCP, endoscopic retrograde cholangiopancreatography; GGT, γ-glutamyl transpeptidase; MRCP, magnetic resonance cholangiopancreatography;
R/O, rule out; SPEP, serum protein electrophoresis; TIBC, total iron-binding capacity; W/U, workup.
the presence of an isolated hyperbilirubinemia, fractionation of the
bilirubin is rarely helpful in determining the cause of jaundice.
Although the degree of elevation of the serum bilirubin has not been
critically assessed as a prognostic marker, it is important in a number of
conditions. In viral hepatitis, the higher the serum bilirubin, the greater
is the hepatocellular damage. Total serum bilirubin correlates with
poor outcomes in alcoholic hepatitis. It is also a critical component
of the Model for End-Stage Liver Disease (MELD) score, a tool used
to estimate survival of patients with end-stage liver disease, prioritize
patients awaiting liver transplantation, and assess operative risk of
patients with cirrhosis. An elevated total serum bilirubin in patients
with drug-induced liver disease indicates more severe injury. Total
bilirubin has also been shown to be of prognostic value in primary
biliary cholangitis.
Unconjugated bilirubin always binds to albumin in the serum and
is not filtered by the kidney. Therefore, any bilirubin found in the
urine is conjugated bilirubin; the presence of bilirubinuria implies
the presence of liver disease or obstructive jaundice. A urine dipstick
test can theoretically give the same information as fractionation of
the serum bilirubin. This test is almost 100% accurate. Phenothi­
azines may give a false-positive reading with the Ictotest tablet. In
patients recovering from jaundice, the urine bilirubin clears prior to
the serum bilirubin.
Serum Enzymes 
The liver contains thousands of enzymes, some of
which are also present in the serum in very low concentrations. These
enzymes have no known function in the serum and behave like other
Cholestatic pattern
(see Table 348-1)
Isolated elevation
of the alkaline
phosphatase
Review drugs
Ultrasound
CT/MRCP/ERCP
ERCP/Liver Bx
Liver Bx
Fractionate the alkaline
phosphatase or check
GGT or 5' nucleotidase
to assess origin of
alkaline phosphatase
Alkaline phos.
of liver origin
Alkaline phos.
of bone origin
Bone Eval
Ultrasound
Review drug list
Check AMA
Dilated ducts
Ducts not dilated
and/or AMA positive
MRCP
Liver biopsy
serum proteins. They are distributed in the plasma and in interstitial
fluid and have characteristic half-lives, which are usually measured
in days. Very little is known about the catabolism of serum enzymes,
although they are probably cleared by cells in the reticuloendothe­
lial system. The elevation of a given enzyme activity in the serum is
thought to primarily reflect its increased rate of entrance into serum
from damaged liver cells.
Serum enzyme tests can be grouped into two categories: (1) enzymes
whose elevation in serum reflects damage to hepatocytes and (2) enzymes
whose elevation in serum reflects cholestasis.
ENZYMES THAT REFLECT DAMAGE TO HEPATOCYTES  The amino­
transferases (transaminases) are sensitive indicators of liver cell injury
and are most helpful in recognizing acute hepatocellular diseases
such as hepatitis. They include aspartate aminotransferase (AST) and
alanine aminotransferase (ALT). AST is found in the liver, cardiac
muscle, skeletal muscle, kidneys, brain, pancreas, lungs, leukocytes,
and erythrocytes in decreasing order of concentration. ALT is found
primarily in the liver and is therefore a more specific indicator of liver
injury. The aminotransferases are normally present in the serum in low
concentrations. These enzymes are released into the blood in greater
amounts when there is damage to the liver cell membrane, resulting in
increased permeability. Liver cell necrosis is not required for the release
of the aminotransferases, and there is a poor correlation between the
degree of liver cell damage and the level of the aminotransferases. Thus,
the absolute elevation of the aminotransferases is of no prognostic sig­
nificance in acute hepatocellular disorders.
The normal range for aminotransferases varies widely among labo­
ratories, but generally ranges from 10 to 40 IU/L. The interlaboratory
variation in normal range is due to technical reasons; no reference stan­
dards exist to establish upper limits of normal for ALT and AST. Some
have recommended revisions of normal limits of the aminotransferases
to adjust for sex and body mass index, but others have noted the poten­
tial costs and unclear benefits of implementing this change.
Any type of liver cell injury can cause modest elevations in the
serum aminotransferases. Levels of up to 300 IU/L are nonspecific and
may be found in any type of liver disorder. Minimal ALT elevations in
asymptomatic blood donors rarely indicate severe liver disease; studies
have shown that fatty liver disease is the most likely explanation. Strik­
ing elevations—that is, aminotransferases >1000 IU/L—occur almost
exclusively in disorders associated with extensive hepatocellular injury
such as (1) viral hepatitis, (2) ischemic liver injury (prolonged hypoten­
sion or acute heart failure), or (3) toxin- or drug-induced liver injury.
The pattern of the aminotransferase elevation can be helpful diag­
nostically. In most acute hepatocellular disorders, the ALT is higher
than or equal to the AST. Whereas the AST:ALT ratio is typically <1
in patients with chronic viral hepatitis and nonalcoholic fatty liver
disease, a number of groups have noted that as cirrhosis develops, this
ratio rises to >1. An AST:ALT ratio >2:1 is suggestive, whereas a ratio
3:1 is highly suggestive, of alcohol-related liver disease. The AST in
alcohol-related liver disease is rarely >300 IU/L, and the ALT is often
normal. A low level of ALT in the serum is due to an alcohol-induced
deficiency of pyridoxal phosphate.
The aminotransferases are usually not greatly elevated in obstruc­
tive jaundice. One notable exception occurs during the acute phase of
biliary obstruction caused by the passage of a gallstone into the com­
mon bile duct. In this setting, the aminotransferases can briefly be in
the 1000–2000 IU/L range. However, aminotransferase levels decrease
quickly, and the biochemical tests rapidly evolve into those typical of
cholestasis.
ENZYMES THAT REFLECT CHOLESTASIS  The activities of three
enzymes—alkaline phosphatase, 5′-nucleotidase, and γ-glutamyl trans­
peptidase (GGT)—are usually elevated in cholestasis. Alkaline phos­
phatase and 5′-nucleotidase are found in or near the bile canalicular
membrane of hepatocytes, whereas GGT is located in the endoplasmic
reticulum and in bile duct epithelial cells. Reflecting its more diffuse
localization in the liver, GGT elevation in serum is less specific for
cholestasis than are elevations of alkaline phosphatase or 5′-nucleo­
tidase. Some have advocated the use of GGT to identify patients with
occult alcohol use. Its lack of specificity makes its use in this setting
questionable.
The normal serum alkaline phosphatase consists of many distinct
isoenzymes found in the liver, bone, placenta, and, less commonly, the
small intestine. Patients over age 60 can have a mildly elevated alkaline
phosphatase (1–1.5 times normal), whereas individuals with blood
types O and B can have an elevation of the serum alkaline phosphatase
after eating a fatty meal due to the influx of intestinal alkaline phos­
phatase into the blood. It is also elevated in children and adolescents
undergoing rapid bone growth because of bone alkaline phosphatase
and late in normal pregnancies due to the influx of placental alkaline
phosphatase.
Elevation of liver-derived alkaline phosphatase is not totally specific
for cholestasis, and a less than threefold elevation can be seen in almost
any type of liver disease. Alkaline phosphatase elevations greater than
four times normal occur primarily in patients with cholestatic liver
disorders, infiltrative liver diseases such as cancer and amyloidosis,
and bone conditions characterized by rapid bone turnover (e.g., Paget’s
disease). In bone diseases, the elevation is due to increased amounts of
the bone isoenzymes. In liver diseases, the elevation is almost always
due to increased amounts of the liver isoenzyme.
If an elevated serum alkaline phosphatase is the only abnormal
finding in an apparently healthy person or if the degree of elevation is
higher than expected in the clinical setting, identification of the source
of elevated isoenzymes is helpful (Fig. 348-1). This problem can be
approached in two ways. First, and most precise, is the fractionation of
the alkaline phosphatase by electrophoresis. The second, best substan­
tiated, and most available approach involves the measurement of serum
5′-nucleotidase or GGT.
CHAPTER 348
In the absence of jaundice or elevated aminotransferases, an elevated
alkaline phosphatase of liver origin often, but not always, suggests early
cholestasis and, less often, hepatic infiltration by tumor or granulo­
mata. Other conditions that cause isolated elevations of the alkaline
phosphatase include primary biliary cholangitis, primary or secondary
sclerosing cholangitis, Hodgkin’s disease, diabetes, hyperthyroidism,
congestive heart failure, and amyloidosis.
Evaluation of Liver Function
The level of serum alkaline phosphatase elevation is not helpful
in distinguishing between intrahepatic and extrahepatic cholestasis.
There is essentially no difference among the values found in obstruc­
tive jaundice due to cancer, common duct stone, sclerosing cholangitis,
or bile duct stricture. Values are similarly increased in patients with
intrahepatic cholestasis due to drug-induced hepatitis, primary biliary
cholangitis, sepsis, rejection of transplanted livers, and, rarely, alcoholinduced steatohepatitis. Values are also greatly elevated in hepatobili­
ary disorders seen in patients with AIDS (e.g., AIDS cholangiopathy
due to cytomegalovirus or cryptosporidial infection and tuberculosis
with hepatic involvement).
■
■TESTS THAT MEASURE BIOSYNTHETIC
FUNCTION OF THE LIVER
Serum Albumin 
Serum albumin is synthesized exclusively by
hepatocytes. Serum albumin has a long half-life: 18–20 days, with ~4%
degraded per day. Because of this slow turnover, the serum albumin is
not a good indicator of acute or mild hepatic dysfunction; only minimal
changes in the serum albumin are seen in acute liver conditions such
as viral hepatitis, drug-related hepatotoxicity, and obstructive jaundice.
In hepatitis, albumin levels <3 g/dL should raise the possibility of
chronic liver disease. Hypoalbuminemia is more common in chronic
liver disorders such as cirrhosis and usually reflects severe liver damage
and decreased albumin synthesis. However, hypoalbuminemia is not
specific for liver disease and may occur in protein malnutrition of any
cause, as well as protein-losing enteropathies, nephrotic syndrome, and
chronic infections that are associated with prolonged increases in levels
of cytokines that inhibit albumin synthesis, such as serum interleukin 1
and/or tumor necrosis factor. Serum albumin should not be measured
to screen patients in whom there is no suspicion of liver disease. A
general medical clinic study of consecutive patients in whom no indi­
cations were present for albumin measurement showed that although
12% of patients had abnormal test results, the finding was of clinical
importance in only 0.4%.
Serum Globulins 
  Serum globulins are a group of proteins made
up of γ globulins (immunoglobulins) produced by B lymphocytes and
α and β globulins produced primarily in hepatocytes. γ Globulins are
increased in chronic liver disease, such as chronic hepatitis and cirrho­
sis. In cirrhosis, the increased serum γ globulin concentration is due
to the increased synthesis of antibodies, some of which are directed
against intestinal bacteria. This occurs because the cirrhotic liver fails
to clear bacterial antigens that normally reach the liver through the
hepatic circulation.
Increases in the concentration of specific isotypes of γ globulins
are often helpful in the recognition of certain chronic liver diseases.
Diffuse polyclonal increases in IgG levels are common in autoim­
mune hepatitis; increases >100% should alert the clinician to this
possibility. Increases in the IgM levels are common in primary biliary
cholangitis, whereas increases in the IgA levels occur in alcoholic
liver disease.
■
■COAGULATION FACTORS
With the exception of factor VIII, which is produced by vascular endo­
thelial cells, the blood clotting factors are made exclusively in hepato­
cytes. Their serum half-lives are much shorter than albumin, ranging
from 6 h for factor VII to 5 days for fibrinogen. Because of their rapid
turnover, measurement of the clotting factors is the single best acute
measure of hepatic synthetic function and helpful in both diagnosis
and assessing the prognosis of acute parenchymal liver disease. Useful
for this purpose is the serum prothrombin time, which collectively mea­
sures factors II, V, VII, and X. Biosynthesis of factors II, VII, IX, and
X depends on vitamin K. The international normalized ratio (INR) is
used to express the degree of anticoagulation on warfarin therapy. The
INR standardizes prothrombin time measurement according to the
characteristics of the thromboplastin reagent used in a particular lab,
which is expressed as an International Sensitivity Index (ISI); the ISI is
then used in calculating the INR.
PART 10
Disorders of the Gastrointestinal System
The prothrombin time may be elevated in hepatitis and cirrhosis as
well as in disorders that lead to vitamin K deficiency such as obstruc­
tive jaundice or fat malabsorption of any kind. Marked prolongation
of the prothrombin time, >5 s above control and not corrected by
parenteral vitamin K administration, is a poor prognostic sign in
acute viral hepatitis and other acute and chronic liver diseases. The
INR, along with the total serum bilirubin, creatinine, albumin, and
sodium, are components of the MELD 3.0 score, which is used as a
measure of hepatic decompensation and to allocate organs for liver
transplantation.
■
■OTHER DIAGNOSTIC TESTS
Although tests may direct the physician to a category of liver disease,
additional biochemical testing, radiologic testing, and procedures are
often necessary to make the proper diagnosis, as shown in Fig. 348-1.
The most commonly used ancillary tests are reviewed here, as are the
noninvasive tests available for assessing hepatic fibrosis.
Ammonia 
Ammonia is produced in the body during normal
protein metabolism and by intestinal bacteria, primarily those in the
colon. The liver plays a role in the detoxification of ammonia by con­
verting it to urea, which is excreted by the kidneys. Striated muscle also
plays a role in detoxification of ammonia, where it is combined with
glutamic acid to form glutamine. Patients with advanced liver disease
typically have significant muscle wasting, which likely contributes
to hyperammonemia. Some physicians use the blood ammonia for
detecting encephalopathy or for monitoring hepatic synthetic function,
although its use for either of these indications has problems. There
is very poor correlation between either the presence or the severity
of acute encephalopathy and elevation of blood ammonia; it can be
occasionally useful for identifying occult liver disease in patients with
mental status changes. There is also a poor correlation of the blood
serum ammonia and hepatic function. The ammonia can be elevated
in patients with severe portal hypertension and portal blood shunting
around the liver even in the presence of normal or near-normal hepatic
function. Elevated arterial ammonia levels have been shown to corre­
late with outcome in fulminant hepatic failure.
Liver Biopsy 
Percutaneous biopsy of the liver is a safe procedure
that is easily performed with local anesthesia and ultrasound guid­
ance. Liver biopsy is of proven value in the following situations: (1)
hepatocellular disease of uncertain cause, (2) prolonged hepatitis with
the possibility of autoimmune hepatitis, (3) unexplained hepatomegaly,
(4) unexplained splenomegaly, (5) hepatic lesions uncharacterized
by radiologic imaging, (6) fever of unknown origin, and (7) staging
of malignant lymphoma. Liver biopsy is most accurate in disorders
causing diffuse changes throughout the liver and is subject to sam­
pling error in focal disorders. Liver biopsy should not be the initial
procedure in the diagnosis of cholestasis. The biliary tree should first
be assessed for signs of obstruction. Contraindications to performing
a percutaneous liver biopsy include significant ascites and prolonged
INR. Under these circumstances, the biopsy can be performed via the
transjugular approach.
Noninvasive Tests to Detect Hepatic Fibrosis 
Although liver
biopsy is the standard for the assessment of hepatic fibrosis, noninva­
sive measures of hepatic fibrosis have gained favor. These measures
include multiparameter tests aimed at detecting and staging the degree
of hepatic fibrosis and imaging techniques. FibroTest (marketed as
FibroSure in the United States) is the best evaluated of the multipa­
rameter blood tests. The test incorporates haptoglobin, bilirubin, GGT,
apolipoprotein A-I, and α2-macroglobulin and has been found to have
high positive and negative predictive values for diagnosing advanced
fibrosis in patients with chronic hepatitis C, chronic hepatitis B, alco­
holic liver disease, or metabolic dysfunction–associated steatotic liver
disease (MASLD) and patients taking methotrexate for psoriasis. The
Enhanced Liver Fibrosis (ELF) test is a multiparameter blood test that
uses markers of fibrogenesis and cytolysis to estimate hepatic fibrosis:
type III procollagen peptide, hyaluronic acid, and tissue inhibitor of
metalloproteinase-1. It has been validated in patients with MASLD and
is gaining favor in cholestatic liver disease. The FIB4 index is a liver
fibrosis biomarker that is easily calculated using age and three routine
laboratory parameters: ALT, AST, and platelet count. While origi­
nally proposed to help assess hepatic fibrosis in patients with human
immunodeficiency virus and hepatitis C virus coinfection, it has been
validated for hepatitis C virus, hepatitis B virus, and MASLD. Transient
elastography (TE), marketed as FibroScan, and magnetic resonance
elastography (MRE) both have gained U.S. Food and Drug Administra­
tion approval for use in the management of patients with liver disease.
TE uses ultrasound waves to measure hepatic stiffness noninvasively.
TE has been shown to be accurate for identifying advanced fibrosis in
patients with chronic hepatitis C, primary biliary cholangitis, hemo­
chromatosis, nonalcoholic fatty liver disease, and recurrent chronic
hepatitis after liver transplantation. MRE has been found to be superior
to TE for staging liver fibrosis in patients with a variety of chronic liver
diseases but requires access to a magnetic resonance imaging scanner
and is more expensive.
Ultrasonography 
Ultrasonography is the first diagnostic test to
use in patients whose liver tests suggest cholestasis, to look for the pres­
ence of a dilated intrahepatic or extrahepatic biliary tree or to identify
gallstones. In addition, it shows space-occupying lesions within the
liver, enables the clinician to distinguish between cystic and solid
masses, and helps direct percutaneous biopsies. Ultrasound with Dop­
pler imaging can detect the patency of the portal vein, hepatic artery,
and hepatic veins and determine the direction of blood flow. This
is the first test ordered in patients suspected of having Budd-Chiari
syndrome.
■
■USE OF LIVER TESTS
As previously noted, the best way to increase the sensitivity and
specificity of laboratory tests in the detection of liver disease is to
employ a battery of tests that includes the aminotransferases, alkaline
phosphatase, bilirubin, albumin, and prothrombin time along with the

21 - 349 The Hyperbilirubinemias
349 The Hyperbilirubinemias
TABLE 348-1  Liver Test Patterns in Hepatobiliary Disorders
TYPE OF DISORDER
BILIRUBIN
AMINOTRANSFERASES
ALKALINE PHOSPHATASE
ALBUMIN
PROTHROMBIN TIME
Hemolysis/Gilbert’s
syndrome
Normal to 86 μmol/L (5 mg/dL)
85% due to indirect fractions
No bilirubinuria
Normal
Normal
Normal
Normal
Acute hepatocellular
necrosis (viral,
ischemic, and drug-
or toxin-induced
hepatitis)
Both fractions may be
elevated
Peak usually follows
aminotransferases
Bilirubinuria
Elevated, often >500 IU,
ALT > AST
Chronic
hepatocellular
disorders
Both fractions may be
elevated
Bilirubinuria
Elevated, but usually
<300 IU
Alcoholic hepatitis,
cirrhosis
Both fractions may be
elevated
Bilirubinuria
AST:ALT >2 suggests
alcoholic hepatitis or
cirrhosis
Intra- and
extrahepatic
cholestasis
(obstructive
jaundice)
Both fractions may be
elevated
Bilirubinuria
Normal to moderate
elevation
Rarely >500 IU
Infiltrative diseases
(tumor, granulomata)
Usually normal
Normal to slight elevation
Elevated, often >4× normal elevation
Fractionate, or confirm liver origin
with 5′-nucleotidase or γ-glutamyl
transpeptidase
Abbreviations: ALT, alanine aminotransferase; AST, aspartate aminotransferase.
judicious use of the other tests described in this chapter. Table 348-1
shows how patterns of liver tests can lead the clinician to a category
of disease that will direct further evaluation. However, it is important
to remember that no single set of liver tests will necessarily provide a
diagnosis. It is often necessary to repeat these tests on several occasions
over days to weeks for a diagnostic pattern to emerge. Figure 348-1 is
an algorithm for the evaluation of chronically abnormal liver tests.
■
■GLOBAL CONSIDERATIONS
The tests and principles presented in this chapter are applicable world­
wide. The causes of liver test abnormalities vary according to region. In
developing nations, infectious diseases are more commonly the etiol­
ogy of abnormal serum liver tests than in developed nations.
■
■FURTHER READING
Berzigotti A et al: EASL Clinical Practice Guidelines on non-invasive
tests for evaluation of liver disease severity and prognosis: 2021
update. J Hepatology 75:659, 2021.
Kaplan M: Alkaline phosphatase. Gastroenterology 62:452, 1972.
Kim WR et al: MELD 3.0: The model for end-stage liver disease
updated for the modern era. Gastroenterology 161:1887, 2021.
Prati D et al: Updated definitions of healthy ranges for serum alanine
aminotransferase levels. Ann Intern Med 137:1, 2002.
Allan W. Wolkoff
The Hyperbilirubinemias
■
■BILIRUBIN METABOLISM
The details of bilirubin metabolism are presented in Chap. 52.
However, the hyperbilirubinemias are best understood in terms of
perturbations of specific aspects of bilirubin metabolism and transport,
and these will be briefly reviewed here as depicted in Fig. 349-1.
Bilirubin is the end product of heme degradation. Some 70–90% of
bilirubin is derived from degradation of the hemoglobin of senescent
Normal to <3× normal elevation
Normal
Usually normal. If >5×
above control and not
corrected by parenteral
vitamin K, suggests poor
prognosis
Normal to <3× normal elevation
Often decreased
Often prolonged
Fails to correct with
parenteral vitamin K
Normal to <3× normal elevation
Often decreased
Often prolonged
Fails to correct with
parenteral vitamin K
Elevated, often >4× normal elevation
Normal, unless
chronic
Normal
If prolonged, will
correct with parenteral
vitamin K
Normal
Normal
CHAPTER 349
red blood cells. Bilirubin produced in the periphery is transported to
the liver within the plasma, where, due to its insolubility in aqueous
solutions, it is tightly bound to albumin. Under normal circumstances,
bilirubin is removed from the circulation rapidly and efficiently by
hepatocytes. Transfer of bilirubin from blood to bile involves four dis­
tinct but interrelated steps (Fig. 349-1).
The Hyperbilirubinemias
Hepatocellular uptake: Uptake of bilirubin by the hepatocyte has
carrier-mediated kinetics. Although a number of candidate bilirubin
transporters have been proposed, the identity of the actual trans­
porter remains elusive.
Intracellular binding: Within the hepatocyte, bilirubin is kept in
solution by binding as a nonsubstrate ligand to several of the gluta­
thione-S-transferases, formerly called ligandins.
Conjugation: Bilirubin is conjugated with one or two glucuronic
acid moieties by a specific UDP-glucuronosyltransferase to form
OATP1B1
OATP1B3
ALB
UCB
BMG
BDG
UGT1A1
BMG
UGT1A1
MRP3
GST:UCB
UCB
BMG
BDG
MRP2
MRP2
UCB
GST
BDG
BT
ALB:UCB
Space
of
Disse
Sinusoid
FIGURE 349-1  Hepatocellular bilirubin transport. Albumin-bound bilirubin in
sinusoidal blood passes through endothelial cell fenestrae to reach the hepatocyte
surface, entering the cell by both facilitated and simple diffusional processes.
Within the cell, it is bound to glutathione-S-transferases and conjugated by
bilirubin-UDP-glucuronosyltransferase (UGT1A1) to mono- and diglucuronides,
which are actively transported across the canalicular membrane into the bile. In
addition to this direct excretion of bilirubin glucuronides, a portion are transported
into the portal circulation by MRP3 and subjected to reuptake into the hepatocyte
by OATP1B1 and OATP1B3. ALB, albumin; BDG, bilirubin diglucuronide; BMG,
bilirubin monoglucuronide; BT, proposed bilirubin transporter; GST, glutathioneS-transferase; MRP2 and MRP3, multidrug resistance–associated proteins 2 and
3; OATP1B1 and OATP1B3, organic anion transport proteins 1B1 and 1B3; UCB,
unconjugated bilirubin; UGT1A1, bilirubin-UDP-glucuronosyltransferase.
5′
3′
500 kb
Variable (Substrate Specific) First Exons
A13 A12 A11 A10
A9
A8
A7
A6
A5
A4
A3
A2
A1
A(TA)6TAA
TATA Box
FIGURE 349-2  Structural organization of the human UGT1 gene complex. This large complex on chromosome 2 contains
at least 13 substrate-specific first exons (A1, A2, etc.). Since four of these are pseudogenes, nine UGT1 isoforms with
differing substrate specificities are expressed. Each exon 1 has its own promoter and encodes the amino-terminal
substrate-specific ∼286 amino acids of the various UGT1-encoded isoforms, and common exons 2–5 encode the 245
carboxyl-terminal amino acids common to all of the isoforms. mRNAs for specific isoforms are assembled by splicing a
particular first exon such as the bilirubin-specific exon A1 to exons 2 to 5. The resulting message encodes a complete
enzyme, in this particular case, bilirubin-UDP-glucuronosyltransferase (UGT1A1). Mutations in a first exon affect only a
single isoform. Those in exons 2–5 affect all enzymes encoded by the UGT1 complex.
bilirubin mono- and diglucuronide, respectively. Conjugation dis­
rupts the internal hydrogen bonding that limits aqueous solubility
of bilirubin, and the resulting glucuronide conjugates are highly
soluble in water. Conjugation is obligatory for excretion of bilirubin
across the bile canalicular membrane into bile. The UDP-glucuro­
nosyltransferases have been classified into gene families based on
the degree of homology among the mRNAs for the various iso­
forms. Those that conjugate bilirubin and certain other substrates
have been designated the UGT1 family. These are expressed from
a single gene complex by alternative promoter usage. This gene
complex contains multiple substrate-specific first exons, desig­
nated A1, A2, etc. (Fig. 349-2), each with its own promoter and
each encoding the amino-terminal half of a specific isoform. In
addition, there are four common exons (exons 2–5) that encode the
shared carboxyl-terminal half of all of the UGT1 isoforms. The vari­
ous first exons encode the specific aglycone substrate binding sites
for each isoform, while the shared exons encode the binding site
for the sugar donor, UDP-glucuronic acid, and the transmembrane
domain. Exon A1 and the four common exons, collectively desig­
nated as the UGT1A1 gene (Fig. 349-2), encode the physiologically
critical enzyme bilirubin-UDP-glucuronosyltransferase (UGT1A1).
A functional corollary of the organization of the UGT1 gene is that
a mutation in one of the first exons will affect only a single enzyme
isoform. By contrast, a mutation in exons 2–5 will alter all isoforms
encoded by the UGT1 gene complex.
4.	Biliary excretion: It has been thought until recently that biliru­
PART 10
Disorders of the Gastrointestinal System
bin mono- and diglucuronides are excreted directly across the
canalicular plasma membrane into the bile canaliculus by an
ATP-dependent transport process mediated by a canalicular
membrane protein called multidrug resistance–associated protein 2
(MRP2, ABCC2). Mutations of MRP2 result in the Dubin-Johnson
syndrome (see below). However, studies in patients with Rotor
syndrome (see below) indicate that after formation, a portion of
the glucuronides is transported into the portal circulation by a
sinusoidal membrane protein called multidrug resistance–associated
protein 3 (MRP3, ABCC3) and is subjected to reuptake into the
hepatocyte by the sinusoidal membrane uptake transporters
organic anion transport protein 1B1 (OATP1B1, SLCO1B1) and
OATP1B3 (SLCO1B3).
■
■EXTRAHEPATIC ASPECTS OF BILIRUBIN
DISPOSITION
Bilirubin in the Gut 
Following secretion into bile, conjugated bili­
rubin reaches the duodenum and passes down the gastrointestinal tract
without reabsorption by the intestinal mucosa. An appreciable fraction
is converted by bacterial metabolism in the gut to the water-soluble col­
orless compound urobilinogen. Urobilinogen undergoes enterohepatic
cycling. Urobilinogen not taken up by
the liver reaches the systemic circula­
tion, from which some is cleared by the
kidneys. Unconjugated bilirubin ordi­
narily does not reach the gut except
in neonates or, by ill-defined alterna­
tive pathways, in the presence of severe
unconjugated hyperbilirubinemia (e.g.,
Crigler-Najjar syndrome, type I [CN-I]).
Unconjugated bilirubin that reaches the
gut is partly reabsorbed, amplifying any
underlying hyperbilirubinemia.
Common Exons
~245 AA
~286 AA
Renal Excretion of Bilirubin Con­
jugates 
Unconjugated bilirubin is not
excreted in urine, as it is too tightly
bound to albumin for effective glomeru­
lar filtration and there is no tubular
mechanism for its renal secretion. In
contrast, the bilirubin conjugates are
readily filtered at the glomerulus and
can appear in urine in disorders characterized by increased bilirubin
conjugates in the circulation. It should be kept in mind that the kidney
can serve as an “overflow valve” for conjugated bilirubin. Consequently,
the level of jaundice in individuals with conjugated hyperbilirubinemia
can be amplified in the presence of renal failure.
DISORDERS OF BILIRUBIN METABOLISM
LEADING TO UNCONJUGATED
HYPERBILIRUBINEMIA
■
■INCREASED BILIRUBIN PRODUCTION
Hemolysis 
Increased destruction of erythrocytes leads to increased
bilirubin turnover and unconjugated hyperbilirubinemia; the hyperbil­
irubinemia is usually modest in the presence of normal liver function.
In particular, the bone marrow is only capable of a sustained eightfold
increase in erythrocyte production in response to a hemolytic stress.
Therefore, hemolysis alone cannot result in a sustained hyperbilirubi­
nemia of more than ∼68 μmol/L (4 mg/dL). Higher values imply con­
comitant hepatic dysfunction. When hemolysis is the only abnormality
in an otherwise healthy individual, the result is a purely unconjugated
hyperbilirubinemia, with the direct-reacting fraction as measured in a
typical clinical laboratory being ≤15% of the total serum bilirubin. In
the presence of systemic disease, which may include a degree of hepatic
dysfunction, hemolysis may produce a component of conjugated
hyperbilirubinemia in addition to an elevated unconjugated bilirubin
concentration. Prolonged hemolysis may lead to the precipitation of
bilirubin salts within the gallbladder or biliary tree, resulting in the for­
mation of gallstones in which bilirubin, rather than cholesterol, is the
major component. Such pigment stones may lead to acute or chronic
cholecystitis, biliary obstruction, or any other biliary tract consequence
of calculous disease.
Ineffective Erythropoiesis 
During erythroid maturation, small
amounts of hemoglobin may be lost at the time of nuclear extrusion,
and a fraction of developing erythroid cells is destroyed within the
marrow. These processes normally account for a small proportion of
bilirubin that is produced. In various disorders, including thalassemia
major, megaloblastic anemias due to folate or vitamin B12 deficiency,
congenital erythropoietic porphyria, lead poisoning, and various
congenital and acquired dyserythropoietic anemias, the fraction of
total bilirubin production derived from ineffective erythropoiesis is
increased, reaching as much as 70% of the total. This may be sufficient
to produce modest degrees of unconjugated hyperbilirubinemia.
Miscellaneous 
Degradation of the hemoglobin of extravascu­
lar collections of erythrocytes, such as those seen in massive tissue
infarctions or large hematomas, may lead transiently to unconjugated
hyperbilirubinemia.
■
■DECREASED HEPATIC BILIRUBIN CLEARANCE
Decreased Hepatic Uptake 
Decreased hepatic bilirubin uptake
is believed to contribute to the unconjugated hyperbilirubinemia of
Gilbert syndrome (GS), although the molecular basis for this find­
ing remains unclear (see below). Several drugs, including flavaspidic
acid, novobiocin, and rifampin, as well as various cholecystographic
contrast agents, have been reported to inhibit bilirubin uptake. The
resulting unconjugated hyperbilirubinemia resolves with cessation of
the medication.
Impaired Conjugation 
• 
PHYSIOLOGIC NEONATAL JAUNDICE 
Bilirubin produced by the fetus is cleared by the placenta and elimi­
nated by the maternal liver. Immediately after birth, the neonatal
liver must assume responsibility for bilirubin clearance and excretion.
However, many hepatic physiologic processes are incompletely devel­
oped at birth. Levels of UGT1A1 are low, and alternative excretory
pathways allow passage of unconjugated bilirubin into the gut. Since
the intestinal flora that convert bilirubin to urobilinogen are also
undeveloped, an enterohepatic circulation of unconjugated bilirubin
ensues. As a consequence, most neonates develop mild unconjugated
hyperbilirubinemia between days 2 and 5 after birth. Peak levels are
typically <85–170 μmol/L (5–10 mg/dL) and decline to normal adult
concentrations within 2 weeks, as mechanisms required for bilirubin
disposition mature. Prematurity, often associated with more pro­
found immaturity of hepatic function and hemolysis, can result in
higher levels of unconjugated hyperbilirubinemia. A rapidly rising
unconjugated bilirubin concentration, or absolute levels >340 μmol/L
(20 mg/dL), puts the infant at risk for bilirubin encephalopathy, or
kernicterus. Under these circumstances, bilirubin crosses an imma­
ture blood-brain barrier and precipitates in the basal ganglia and
other areas of the brain. The consequences range from appreciable
neurologic deficits to death. Treatment options include phototherapy,
which converts bilirubin into water-soluble photoisomers that are
excreted directly into bile, and exchange transfusion. The canalicular
mechanisms responsible for bilirubin excretion are also immature at
birth, and their maturation may lag behind that of UGT1A1; this can
lead to transient conjugated neonatal hyperbilirubinemia, especially in
infants with hemolysis.
ACQUIRED CONJUGATION DEFECTS  A modest reduction in bilirubin
conjugating capacity may be observed in advanced hepatitis or cirrho­
sis. However, in this setting, conjugation is better preserved than other
aspects of bilirubin disposition, such as canalicular excretion. Various
drugs, including pregnanediol, novobiocin, chloramphenicol, genta­
micin, and atazanavir, may produce unconjugated hyperbilirubinemia
by inhibiting UGT1A1 activity. Bilirubin conjugation may be inhibited
by certain fatty acids that are present in breast milk, but not serum,
TABLE 349-1  Principal Differential Characteristics of Gilbert and Crigler-Najjar Syndromes
CRIGLER-NAJJAR SYNDROME
FEATURE
GILBERT SYNDROME
TYPE I
TYPE II
Total serum bilirubin, μmol/L (mg/dL)
310–755 (usually >345) (18–45
[usually >20])
Routine liver tests
Response to phenobarbital
Kernicterus
Hepatic histology
Normal
None
Usual
Normal
Bile characteristics
  Color
  Bilirubin fractions
 
Pale or colorless
90% unconjugated
Bilirubin UDP-glucuronosyltransferase
activity
Inheritance (all autosomal)
Typically absent; traces in some patients
Recessive
of mothers whose infants have excessive neonatal hyperbilirubinemia
(breast milk jaundice). Alternatively, there may be increased entero­
hepatic circulation of bilirubin in these infants. The pathogenesis of
breast milk jaundice appears to differ from that of transient familial
neonatal hyperbilirubinemia (Lucey-Driscoll syndrome), in which
there may be a UGT1A1 inhibitor in maternal serum.
■
■HEREDITARY DEFECTS IN BILIRUBIN
CONJUGATION
Three familial disorders characterized by differing degrees of uncon­
jugated hyperbilirubinemia have long been recognized. The defining
clinical features of each are described below (Table 349-1). While these
disorders have been recognized for decades to reflect differing degrees
of deficiency in the ability to conjugate bilirubin, recent advances in
the molecular biology of the UGT1 gene complex have elucidated their
interrelationships and clarified previously puzzling features.
Crigler-Najjar Syndrome, Type I 
CN-I is characterized by
striking unconjugated hyperbilirubinemia of ∼340–765 μmol/L (20–
45 mg/dL) that appears in the neonatal period and persists for life.
Other conventional hepatic biochemical tests such as serum amino­
transferases and alkaline phosphatase are normal, and there is no
evidence of hemolysis. Hepatic histology is also essentially normal
except for the occasional presence of bile plugs within canaliculi.
Bilirubin glucuronides are virtually absent from the bile, and there is
no detectable constitutive expression of UGT1A1 activity in hepatic
tissue. Neither UGT1A1 activity nor the serum bilirubin concentra­
tion responds to administration of phenobarbital or other enzyme
inducers. Unconjugated bilirubin accumulates in plasma, from which
it is eliminated very slowly by alternative pathways that include direct
passage into the bile and small intestine, possibly via bilirubin pho­
toisomers. This accounts for the small amount of urobilinogen found
in feces. No bilirubin is found in the urine. First described in 1952,
the disorder is rare (estimated prevalence, 0.6–1.0 per million). Many
patients are from geographically or socially isolated communities in
which consanguinity is common, and pedigree analyses show an auto­
somal recessive pattern of inheritance. The majority of patients (type
IA) exhibit defects in the glucuronide conjugation of a spectrum of
substrates in addition to bilirubin, including various drugs and other
xenobiotics. These individuals have mutations in one of the com­
mon exons (2–5) of the UGT1 gene (Fig. 349-2). In a smaller subset
(type IB), the defect is limited largely to bilirubin conjugation, and
the causative mutation is in the bilirubin-specific exon A1. Estrogen
glucuronidation is mediated by UGT1A1 and is defective in all CN-I
patients. More than 30 different genetic lesions of UGT1A1 respon­
sible for CN-I have been identified, including deletions, insertions,
alterations in intron splice donor and acceptor sites, exon skipping,
CHAPTER 349
The Hyperbilirubinemias
100–430 (usually ≤345) (6–25
[usually ≤20])
Typically ≤70 μmol/L (≤4 mg/dL) in
absence of fasting or hemolysis
Normal
Decreases bilirubin by >25%
Rare
Normal
Normal
Decreases bilirubin to normal
No
Usually normal; increased lipofuscin
pigment in some
 
Pigmented
Largest fraction (mean: 57%)
monoconjugates
 
Normal dark color
Mainly diconjugates but
monoconjugates increased (mean: 23%)
Markedly reduced: 0–10% of normal
 
Predominantly recessive
Reduced: typically 10–33% of normal
Promoter mutation: recessive
Missense mutations: 7 of 8 dominant;
1 reportedly recessive
and point mutations that introduce premature stop codons or alter
critical amino acids. Their common feature is that they all encode
proteins with absent or, at most, traces of bilirubin-UDP-glucuronos­
yltransferase enzymatic activity.
Prior to the use of phototherapy, most patients with CN-I died of
bilirubin encephalopathy (kernicterus) in infancy or early childhood.
A few lived as long as early adult life without overt neurologic damage,
although more subtle testing usually indicated mild but progressive
brain damage, now termed bilirubin-induced neurologic dysfunction
(BIND). In the absence of liver transplantation, death eventually super­
vened from late-onset bilirubin encephalopathy, which often followed
a nonspecific febrile illness. Although isolated hepatocyte transplanta­
tion has been used in a small number of cases of CN-I, early liver trans­
plantation (Chap. 356) remains the best hope to prevent brain injury
and death at present. It is anticipated that gene replacement therapy
may be an option in the future.
Crigler-Najjar Syndrome, Type II (CN-II) 
This condition was
recognized as a distinct entity in 1962 and is characterized by marked
unconjugated hyperbilirubinemia in the absence of abnormalities of
other conventional hepatic biochemical tests, hepatic histology, or
hemolysis. It differs from CN-I in several specific ways (Table 349-1):
(1) although there is considerable overlap, average bilirubin concentra­
tions are lower in CN-II; (2) accordingly, CN-II is only infrequently
associated with kernicterus; (3) bile is deeply colored, and bilirubin
glucuronides are present, with a striking, characteristic increase in the
proportion of monoglucuronides; (4) UGT1A1 in liver is usually pres­
ent at reduced levels (typically ≤10% of normal); and (5) while typically
detected in infancy, hyperbilirubinemia was not recognized in some
cases until later in life and, in one instance, at age 34. As with CN-I,
most CN-II cases exhibit abnormalities in the conjugation of other
compounds, such as salicylamide and menthol, but in some instances,
the defect appears limited to bilirubin. Reduction of serum bilirubin
concentrations by >25% in response to enzyme inducers such as
phenobarbital distinguishes CN-II from CN-I, although this response
may not be elicited in early infancy and often is not accompanied by
measurable UGT1A1 induction. Bilirubin concentrations during phe­
nobarbital administration do not return to normal but are typically in
the range of 51–86 μmol/L (3–5 mg/dL). Although the incidence of
kernicterus in CN-II is low, instances have occurred, not only in infants
but also in adolescents and adults, often in the setting of an intercur­
rent illness, fasting, or another factor that temporarily raises the serum
bilirubin concentration above baseline and reduces serum albumin
levels. For this reason, phenobarbital therapy is widely recommended,
a single bedtime dose often sufficing to maintain clinically safe serum
bilirubin concentrations.
PART 10
Disorders of the Gastrointestinal System
Over 100 different mutations in the UGT1 gene have been identi­
fied as causing CN-I or CN-II. It was found that missense mutations
are more common in CN-II patients, as would be expected in this
less severe phenotype. Their common feature is that they encode for
a bilirubin-UDP-glucuronosyltransferase with markedly reduced, but
detectable, enzymatic activity. The spectrum of residual enzyme activ­
ity explains the spectrum of phenotypic severity of the resulting hyper­
bilirubinemia. Molecular analysis has established that a large majority
of CN-II patients are either homozygotes or compound heterozygotes
for CN-II mutations and that individuals carrying one mutated and
one entirely normal allele have normal bilirubin concentrations.
Gilbert Syndrome 
This syndrome is characterized by mild uncon­
jugated hyperbilirubinemia, normal values for standard hepatic bio­
chemical tests, and normal hepatic histology other than a modest
increase of lipofuscin pigment in some patients. Serum bilirubin
concentrations are most often <51 μmol/L (<3 mg/dL), although both
higher and lower values are frequent. The clinical spectrum of hyper­
bilirubinemia fades into that of CN-II at serum bilirubin concentra­
tions of 86–136 μmol/L (5–8 mg/dL). At the other end of the scale,
the distinction between mild cases of GS and a normal state is often
blurred. Bilirubin concentrations may fluctuate substantially in any
given individual, and at least 25% of patients will exhibit temporarily
normal values during prolonged follow-up. More elevated values are
associated with stress, fatigue, alcohol use, reduced caloric intake,
and intercurrent illness, while increased caloric intake or administra­
tion of enzyme-inducing agents produces lower bilirubin levels. GS
is most often diagnosed at or shortly after puberty or in adult life
during routine examinations that include multichannel biochemical
analyses. UGT1A1 activity is typically reduced to 10–35% of normal,
and bile pigments exhibit a characteristic increase in bilirubin mono­
glucuronides. Studies of radiobilirubin kinetics indicate that hepatic
bilirubin clearance is reduced to an average of one-third of normal.
Administration of phenobarbital normalizes both the serum biliru­
bin concentration and hepatic bilirubin clearance; however, failure
of UGT1A1 activity to improve in many such instances suggests the
possible coexistence of an additional defect. Compartmental analysis
of bilirubin kinetic data suggests that GS patients may have a defect in
bilirubin uptake as well as in conjugation, although this has not been
shown directly. Defects in the hepatic uptake of other organic anions
that at least partially share an uptake mechanism with bilirubin, such
as sulfobromophthalein and indocyanine green (ICG), are observed
in a minority of patients. The metabolism and transport of bile acids
that do not utilize the bilirubin uptake mechanism are normal. The
magnitude of changes in the serum bilirubin concentration induced
by provocation tests such as 48 h of fasting or the IV administration
of nicotinic acid has been reported to be of help in separating GS
patients from normal individuals. Other studies dispute this assertion.
Moreover, on theoretical grounds, the results of such studies should
provide no more information than simple measurements of the base­
line serum bilirubin concentration. Family studies indicate that GS
and hereditary hemolytic anemias such as hereditary spherocytosis,
glucose-6-phosphate dehydrogenase deficiency, and β-thalassemia trait
sort independently. Reports of hemolysis in up to 50% of GS patients
are believed to reflect better case finding, since patients with both GS
and hemolysis have higher bilirubin concentrations and are more likely
to be jaundiced than patients with either defect alone.
GS is common, with many series placing its prevalence as high
as 8%. Males predominate over females by reported ratios ranging
from 1.5:1 to >7:1. However, these ratios may have a large artifactual
component since normal males have higher mean bilirubin levels than
normal females, but the diagnosis of GS is often based on comparison
to normal ranges established in men. The high prevalence of GS in the
general population may explain the reported frequency of mild uncon­
jugated hyperbilirubinemia in liver transplant recipients. The disposi­
tion of most xenobiotics metabolized by glucuronidation appears to
be normal in GS, as is oxidative drug metabolism in the majority of
reported studies. The principal exception is the metabolism of the anti­
tumor agent irinotecan (CPT-11), whose active metabolite (SN-38) is
glucuronidated specifically by bilirubin-UDP-glucuronosyltransferase.
Administration of CPT-11 to patients with GS has resulted in several
toxicities, including intractable diarrhea and myelosuppression. Some
reports also suggest abnormal disposition of menthol, estradiol benzo­
ate, acetaminophen, tolbutamide, and rifamycin SV. Although some
of these studies have been disputed, and there have been no reports of
clinical complications from use of these agents in GS, prudence should
be exercised in prescribing them or any agents metabolized primarily
by glucuronidation in this condition. It should also be noted that the
HIV protease inhibitors indinavir and atazanavir (Chap. 208) can
inhibit UGT1A1, resulting in hyperbilirubinemia that is most pro­
nounced in patients with preexisting GS.
Most older pedigree studies of GS were consistent with autosomal
dominant inheritance with variable expressivity. However, studies of
the UGT1 gene in GS have indicated a variety of molecular genetic
bases for the phenotypic picture and several different patterns of
inheritance. Studies in Europe and the United States found that nearly
all patients had normal coding regions for UGT1A1 but were homo­
zygous for the insertion of an extra TA (i.e., A[TA]7TAA rather than
A[TA]6TAA) in the promoter region of the first exon. This appeared
to be necessary, but not sufficient, for clinically expressed GS, since
15% of normal controls were also homozygous for this variant. While
normal by standard criteria, these individuals had somewhat higher
bilirubin concentrations than the rest of the controls studied. Hetero­
zygotes for this abnormality had bilirubin concentrations identical to
those homozygous for the normal A[TA]6TAA allele. The prevalence
of the A[TA]7TAA allele in a general Western population is 30%, in
which case 9% would be homozygotes. This is slightly higher than the
prevalence of GS based on purely phenotypic parameters. It was sug­
gested that additional variables, such as mild hemolysis or a defect in
bilirubin uptake, might be among the factors enhancing phenotypic
expression of the defect.
Phenotypic expression of GS due solely to the A[TA]7TAA promoter
abnormality is inherited as an autosomal recessive trait. A number of
CN-II kindreds have been identified in whom there is also an allele
containing a normal coding region but the A[TA]7TAA promoter
abnormality. CN-II heterozygotes, who have the A[TA]6TAA promoter,
are phenotypically normal, whereas those with the A[TA]7TAA pro­
moter express the phenotypic picture of GS. GS in such kindreds may
also result from homozygosity for the A[TA]7TAA promoter abnormal­
ity. Seven different missense mutations in the UGT1 gene that report­
edly cause GS with dominant inheritance have been found in Japanese
individuals. Another Japanese patient with mild unconjugated hyper­
bilirubinemia was homozygous for a missense mutation in exon 5. GS
in her family appeared to be recessive.
DISORDERS OF BILIRUBIN METABOLISM
LEADING TO MIXED OR PREDOMINANTLY
CONJUGATED HYPERBILIRUBINEMIA
In hyperbilirubinemia due to acquired liver disease (e.g., acute hepa­
titis, common bile duct stone), there are usually elevations in the
serum concentrations of both conjugated and unconjugated bilirubin.
Although biliary tract obstruction or hepatocellular cholestatic injury
may present on occasion with a predominantly conjugated hyper­
bilirubinemia, it is generally not possible to differentiate intrahepatic
from extrahepatic causes of jaundice based on the serum levels or
relative proportions of unconjugated and conjugated bilirubin. The
major reason for determining the amounts of conjugated and uncon­
jugated bilirubin in the serum is for the initial differentiation of
hepatic parenchymal and obstructive disorders (mixed conjugated and
unconjugated hyperbilirubinemia) from the inheritable and hemo­
lytic disorders discussed above that are associated with unconjugated
hyperbilirubinemia.
■
■FAMILIAL DEFECTS IN HEPATIC
EXCRETORY FUNCTION
Dubin-Johnson Syndrome (DJS) 
This benign, relatively rare
disorder is characterized by low-grade, predominantly conjugated
hyperbilirubinemia (Table 349-2). Total bilirubin concentrations
are typically between 34 and 85 μmol/L (2 and 5 mg/dL) but on
occasion can be in the normal range or as high as 340–430 μmol/L
TABLE 349-2  Principal Differential Characteristics of Inheritable Disorders of Bile Canalicular Function
 
DJS
ROTOR
PFIC1
BRIC1
PFIC2
BRIC2
PFIC3
Gene
Protein
Cholestasis
ABCCA
MRP2
No
SLCO1B1/SLCO1B3
OATP1B1/1B3
No
ATP8B1
FIC1
Yes
Serum GGT
Serum bile acids
Normal
Normal
Normal
Normal
Normal
↑↑
Clinical features
Mild conjugated
hyperbilirubinemia;
otherwise, normal
liver function; dark
pigment in liver;
characteristic
pattern of urinary
coproporphyrins
Mild conjugated
hyperbilirubinemia;
otherwise, normal
liver function; liver
without abnormal
pigmentation
Severe cholestasis
beginning in
childhood
Abbreviations: BRIC, benign recurrent intrahepatic cholestasis; BSEP, bile salt excretory protein; DJS, Dubin-Johnson syndrome; GGT, γ-glutamyl transferase; MRP2,
multidrug resistance–associated protein 2; OATP1A/1B, organic anion transport proteins 1B1 and 1B3; PFIC, progressive familial intrahepatic cholestasis; ↑↑, increased.
(20–25 mg/dL) and can fluctuate widely in any given patient. The
degree of hyperbilirubinemia may be increased by intercurrent illness,
oral contraceptive use, and pregnancy. Because the hyperbilirubinemia
is due to a predominant rise in conjugated bilirubin, bilirubinuria is
characteristically present. Aside from elevated serum bilirubin levels,
other routine laboratory tests are normal. Physical examination is usu­
ally normal except for jaundice, although an occasional patient may
have hepatosplenomegaly.
Patients with DJS are usually asymptomatic, although some may
have vague constitutional symptoms. These latter patients have usu­
ally undergone extensive diagnostic examinations for unexplained
jaundice and have high levels of anxiety. In women, the condition
may be subclinical until the patient becomes pregnant or receives oral
contraceptives, at which time chemical hyperbilirubinemia becomes
frank jaundice. Even in these situations, other routine liver function
tests, including serum alkaline phosphatase and transaminase activi­
ties, are normal.
A cardinal feature of DJS is the accumulation of dark, coarsely gran­
ular pigment in the lysosomes of centrilobular hepatocytes. As a result,
the liver may be grossly black in appearance. This pigment is thought
to be derived from epinephrine metabolites that are not excreted nor­
mally. The pigment may disappear during bouts of viral hepatitis, only
to reaccumulate slowly after recovery.
Biliary excretion of a number of anionic compounds is compro­
mised in DJS. These include various cholecystographic agents, as
well as sulfobromophthalein (Bromsulphalein [BSP]), a synthetic
dye formerly used in a test of liver function. In this test, the rate of
disappearance of BSP from plasma was determined following bolus
IV administration. BSP is conjugated with glutathione in the hepato­
cyte; the resulting conjugate is normally excreted rapidly into the bile
canaliculus. Patients with DJS exhibit characteristic rises in plasma
concentrations at 90 min after injection, due to reflux of conjugated
BSP into the circulation from the hepatocyte. Dyes such as ICG that are
taken up by hepatocytes but are not further metabolized prior to bili­
ary excretion do not show this reflux phenomenon. Continuous BSP
infusion studies suggest a reduction in the time to maximum plasma
concentration (tmax) for biliary excretion. Bile acid disposition, includ­
ing hepatocellular uptake and biliary excretion, is normal in DJS. These
patients have normal serum and biliary bile acid concentrations and do
not have pruritus.
CHAPTER 349
The Hyperbilirubinemias
By analogy with findings in several mutant rat strains, the selective
defect in biliary excretion of bilirubin conjugates and certain other
classes of organic compounds, but not of bile acids, that characterizes
DJS in humans was found to reflect defective expression of MRP2
(ABCC2), an ATP-dependent canalicular membrane transporter. Sev­
eral different mutations in the ABCC2 gene produce the Dubin-Johnson
phenotype, which has an autosomal recessive pattern of inheritance.
Although MRP2 is undoubtedly important in the biliary excretion of
ATP8B1
FIC1
Episodic
ABCB11
BSEP
Yes
ABCB11
BSEP
Episodic
ABCB4
MDR3
Yes
Normal
↑↑ during
episodes
Normal
↑↑
Normal
↑↑ during
episodes
↑↑
↑↑
Recurrent
episodes of
cholestasis
beginning at any
age
Severe cholestasis
beginning in
childhood
Recurrent
episodes of
cholestasis
beginning at any
age
Severe cholestasis
beginning in
childhood;
decreased
phospholipids
in bile
conjugated bilirubin, the fact that this pigment is still excreted in the
absence of MRP2 suggests that other, as yet uncharacterized, transport
proteins may serve in a secondary role in this process.
Patients with DJS also have a diagnostic abnormality in urinary
coproporphyrin excretion. There are two naturally occurring copro­
porphyrin isomers, I and III. Normally, ∼75% of the coproporphyrin
in urine is isomer III. In urine from DJS patients, total coproporphy­
rin content is normal, but ≥80% is isomer I. Heterozygotes for the
syndrome show an intermediate pattern. The molecular basis for this
phenomenon remains unclear.
Rotor Syndrome (RS) 
This benign, autosomal recessive disor­
der is clinically similar to DJS (Table 349-2), although it is seen even
less frequently. A major phenotypic difference is that the liver in
patients with RS has no increased pigmentation and appears totally
normal. The only abnormality in routine laboratory tests is an eleva­
tion of total serum bilirubin, due to a predominant rise in conjugated
bilirubin. This is accompanied by bilirubinuria. Several additional
features differentiate RS from DJS. In RS, the gallbladder is usually
visualized on oral cholecystography, in contrast to the nonvisualiza­
tion that is typical of DJS. The pattern of urinary coproporphyrin
excretion also differs. The pattern in RS resembles that of many
acquired disorders of hepatobiliary function, in which copropor­
phyrin I, the major coproporphyrin isomer in bile, refluxes from the
hepatocyte back into the circulation and is excreted in urine. Thus,
total urinary coproporphyrin excretion is substantially increased in
RS, in contrast to the normal levels seen in DJS. Although the fraction
of coproporphyrin I in urine is elevated, it is usually <70% of the total,
compared with ≥80% in DJS. The disorders also can be distinguished
by their patterns of BSP excretion. Although clearance of BSP from
plasma is delayed in RS, there is no reflux of conjugated BSP back into
the circulation as seen in DJS. Kinetic analysis of plasma BSP infusion
studies suggests the presence of a defect in intrahepatocellular storage
of this compound. This has never been demonstrated directly. Recent
studies indicate that the molecular basis of RS results from simulta­
neous deficiency of the hepatocyte plasma membrane transporters
OATP1B1 (SLCO1B1) and OATP1B3 (SLCO1B3). This results in
reduced reuptake by these transporters of conjugated bilirubin that
has been pumped out of the hepatocyte into the portal circulation by
MRP3 (ABCC3) (Fig. 349-1). It should be noted that these transport­
ers normally mediate uptake into the liver of a variety of drugs such
as statins, and the possibility of drug toxicity resulting from their
deficiency in RS must be considered.
PART 10
Disorders of the Gastrointestinal System
Benign Recurrent Intrahepatic Cholestasis (BRIC) 
This
rare disorder is characterized by recurrent attacks of pruritus and
jaundice. The typical episode begins with mild malaise and elevations
in serum aminotransferase levels, followed rapidly by rises in alkaline
phosphatase and conjugated bilirubin and onset of jaundice and itch­
ing. The first one or two episodes may be misdiagnosed as acute viral
hepatitis. The cholestatic episodes, which may begin in childhood or
adulthood, can vary in duration from several weeks to months, fol­
lowed by a complete clinical and biochemical resolution. Intervals
between attacks may vary from several months to years. Between
episodes, physical examination is normal, as are serum levels of bile
acids, bilirubin, transaminases, and alkaline phosphatase. The disorder
is familial and has an autosomal recessive pattern of inheritance. BRIC
is considered a benign disorder in that it does not lead to cirrhosis or
end-stage liver disease. However, the episodes of jaundice and pruritus
can be prolonged and debilitating, and some patients have undergone
liver transplantation to relieve the intractable and disabling symptoms.
Treatment during the cholestatic episodes is symptomatic; there is no
specific treatment to prevent or shorten the occurrence of episodes.
A gene termed FIC1 was recently identified and found to be mutated
in patients with BRIC. Curiously, this gene is expressed strongly in
the small intestine but only weakly in the liver. The protein encoded
by FIC1 shows little similarity to those that have been shown to play
a role in bile canalicular excretion of various compounds. Rather, it
appears to be a member of a P-type ATPase family that transports
aminophospholipids from the outer to the inner leaflet of a variety of
cell membranes. Its relationship to the pathobiology of this disorder
remains unclear. A second phenotypically identical form of BRIC,
termed BRIC type 2, has been described resulting from mutations in
the bile salt excretory protein (BSEP), the protein that is defective in
progressive familial intrahepatic cholestasis (PFIC) type 2 (Table 349-2).
How some mutations in this protein result in the episodic BRIC phe­
notype is unknown.
Progressive Familial Intrahepatic Cholestasis 
This name
is applied to three phenotypically related syndromes (Table 349-2).
PFIC type 1 (Byler’s disease) presents in early infancy as cholestasis
that may be initially episodic. However, in contrast to BRIC, Byler’s
disease progresses to malnutrition, growth retardation, and end-stage
liver disease during childhood. This disorder is also a consequence of
an FIC1 mutation. The functional relationship of the FIC1 protein to
the pathogenesis of cholestasis in these disorders is unknown. Two
other types of PFIC (types 2 and 3) have been described. PFIC type
2 is associated with a mutation in the protein originally named sis­
ter of P-glycoprotein, now known as bile salt excretory protein (BSEP,
ABCB11), which is the major bile canalicular exporter of bile acids.
As noted above, some mutations of this protein are associated with
BRIC type 2, rather than the PFIC type 2 phenotype. PFIC type 3 has
been associated with a mutation of MDR3 (ABCB4), a protein that is
essential for normal hepatocellular excretion of phospholipids across
the bile canaliculus. Although all three types of PFIC have similar
clinical phenotypes, only type 3 is associated with high serum levels
of γ-glutamyl transferase (GGT) activity. In contrast, activity of this
enzyme is normal or only mildly elevated in symptomatic BRIC and
PFIC types 1 and 2. Interestingly, mutations in FIC1 or BSEP are not
found in approximately one-third of patients with clinical PFIC and
normal GGT. Recent studies have shown that patients with mutations
in NR1H4, the gene encoding the farnesoid X receptor (FXR), a nuclear
hormone receptor activated by bile acids, have a syndrome identical
to PFIC2 with absent expression of BSEP. Mutations in tight junction
protein 2 (TJP2) as well as ubiquitin-specific protease 53 (USP53), a
protein that interacts with TJP2, have also been associated with severe
cholestasis with normal GGT levels, likely due to disruption of tight
junctions at the bile canaliculus.
■
■FURTHER READING
Bull LN, Thompson RJ: Progressive familial intrahepatic cholestasis.
Clin Liver Dis 22:657, 2018.
Canu G et al: Gilbert and Crigler Najjar syndromes: An update of the
UDP-glucuronosyltransferase 1A1 (UGT1A1) gene mutation data­
base. Blood Cells Mol Dis 50:273, 2013.
Hansen TW: Biology of bilirubin photoisomers. Clin Perinatol 43:277,
2016.
Hassan S, Hertel P: Overview of progressive familial intrahepatic
cholestasis. Clin Liver Dis 26:371, 2022.
Kavallar AM et al: Management and outcomes after liver transplantation
for progressive familial intrahepatic cholestasis: A systematic review and
meta-analysis. Hepatology Communications 7:e0286, 2023.
Lamola AA: A pharmacologic view of phototherapy. Clin Perinatol
43:259, 2016.
Memon N et al: Inherited disorders of bilirubin clearance. Pediatr Res
79:378, 2016.
Soroka CJ, Boyer JL: Biosynthesis and trafficking of the bile salt
export pump, BSEP: Therapeutic implications of BSEP mutations.
Mol Aspects Med 37:3, 2014.
van de Steeg E et al: Complete OATP1B1 and OATP1B3 deficiency
causes human Rotor syndrome by interrupting conjugated bilirubin
reuptake into the liver. J Clin Invest 122:519, 2012.
van Wessel DBE et al: Genotype correlates with the natural history of
severe bile salt export pump deficiency. J Hepatol 73:84, 2020.
Wolkoff AW: Organic anion uptake by hepatocytes. Compr Physiol
4:1715, 2014.

22 - 350 Acute Viral Hepatitis
350 Acute Viral Hepatitis
Esperance A. K. Schaefer,
Raymond T. Chung, Jules L. Dienstag
Acute Viral Hepatitis
Acute viral hepatitis is a systemic infection affecting the liver predomi­
nantly. Almost all cases of acute viral hepatitis are caused by one of five
viral agents: hepatitis A virus (HAV), hepatitis B virus (HBV), hepatitis
C virus (HCV), the HBV-associated delta agent or hepatitis D virus
(HDV), and hepatitis E virus (HEV). All these human hepatitis viruses
are RNA viruses, except for hepatitis B, which is a DNA virus but
replicates like a retrovirus. Although these agents can be distinguished
by their molecular and antigenic properties, all types of viral hepatitis
produce clinically similar acute illnesses. These range from asymp­
tomatic and inapparent to fulminant and fatal acute infections, which
can be observed in all types. On the other hand, the bloodborne types
(HBV, HCV, HDV) may also manifest a spectrum of chronic disease,
from subclinical, persistent infections to rapidly progressive chronic
liver disease with cirrhosis and even hepatocellular carcinoma. Rarely,
infections with other viruses (e.g., cytomegalovirus, Epstein-Barr virus,
other herpes viruses, SARS-CoV-2) can be associated with mild or
severe, even fulminant, hepatitis, more often in immunocompromised
hosts but also in otherwise healthy persons. In 2021 and 2022, after
the peak years of the COVID-19 pandemic, multiple reports appeared
globally of severe acute, often fulminant, hepatitis in previously
healthy children. While these instances of severe acute hepatitis remain
unexplained, links have been reported of infection with adenovirus
type 41 and adeno-associated virus 2, ordinarily not causes of liver
injury. A reduction in circulating respiratory virus infections during
the isolation of the COVID-19 pandemic has been postulated to have
accounted for the increased susceptibility to and severity of the acute
hepatitis associated with these nonhepatotropic viruses.
■
■VIROLOGY AND ETIOLOGY
Hepatitis A 
HAV is a nonenveloped 27-nm, heat-, acid-, and etherresistant, single-stranded, positive-sense RNA virus in the Hepatovirus
genus of the picornavirus family (Fig. 350-1). Quasi-enveloped virus
particles encased in host plasma membrane–derived membranous ves­
icles circulate in the bloodstream. The virion contains four structural
capsid polypeptides, designated VP1–VP4, as well as six nonstructural
proteins, which are cleaved posttranslationally from the polyprotein
product of a 7500-nucleotide genome. Inactivation of viral activity can
be achieved by boiling for 1 min, by contact with formaldehyde and
chlorine, or by ultraviolet irradiation. Despite nucleotide sequence
variation of up to 20% among isolates of HAV and despite the recog­
nition of six genotypes (three of which affect humans), all strains of
this virus are immunologically indistinguishable and belong to one
FIGURE 350-1  Electron micrographs of hepatitis A virus particles and serum from a patient with hepatitis B. Left:
27-nm hepatitis A virus particles purified from stool of a patient with acute hepatitis A and aggregated by antibody to
hepatitis A virus. Right: Concentrated serum from a patient with hepatitis B, demonstrating the 42-nm virions, tubular
forms, and spherical 22-nm particles of hepatitis B surface antigen. 132,000×. (Hepatitis D resembles 42-nm virions of
hepatitis B but is smaller, 35–37 nm; hepatitis E resembles hepatitis A virus but is slightly larger, 32–34 nm; hepatitis C
has been visualized as a 55-nm particle.)
serotype. Human HAV can infect and cause hepatitis in chimpanzees,
tamarins (marmosets), and several monkey species. HAV-like hepa­
toviruses have also been identified in small mammals, including bats
and rodents.
Hepatitis A has an incubation period of ~3–4 weeks. Its replication
is limited to the liver, but the virus is present in the liver, bile, stools,
and blood during the late incubation period and acute preicteric/
presymptomatic phase of illness. Despite slightly longer persistence
of virus in the liver, fecal shedding, viremia, and infectivity diminish
rapidly once jaundice becomes apparent. Detection of HAV RNA by
sensitive reverse transcription polymerase chain reaction assays has
been reported to persist at low levels in stool, the liver, and serum for
up to several months after acute illness; however, this does not cor­
relate with persistent infectivity, probably because of the presence of
neutralizing antibody. HAV can be cultivated reproducibly in vitro and
in primate models.
Antibodies to HAV (anti-HAV) can be detected during acute illness
when serum aminotransferase activity is elevated and fecal HAV shed­
ding is still occurring. This early antibody response is predominantly
of the IgM class and persists for several (~3) months, rarely for 6–12
months. During convalescence, however, anti-HAV of the IgG class
becomes the predominant antibody (Fig. 350-2). Therefore, the diag­
nosis of hepatitis A is made during acute illness by demonstrating antiHAV of the IgM class. After acute illness, anti-HAV of the IgG class
remains detectable indefinitely, and patients with serum anti-HAV
are immune to reinfection. Neutralizing antibody activity parallels the
appearance of anti-HAV, and the IgG anti-HAV present in immune
globulin accounts for the protection it affords against HAV infection.
CHAPTER 350
Hepatitis B 
HBV is a DNA virus with a remarkably compact
genomic structure; despite its small, circular, 3200-bp size, HBV DNA
codes for four sets of viral products with a complex, multiparticle
structure. HBV achieves its genomic economy by relying on an effi­
cient strategy of encoding proteins from four overlapping genes: S, C,
P, and X (Fig. 350-3), as detailed below. Once thought to be unique
among viruses, HBV is now recognized as one of a family of animal
viruses, hepadnaviruses (hepatotropic DNA viruses), and is classi­
fied as hepadnavirus type 1. Similar viruses infect certain species of
woodchucks, ground and tree squirrels, and Pekin ducks, to mention
the most carefully characterized; genetic evidence of ancient HBV-like
virus forbears has been found in fossils of ancient birds, and an HBV-like
virus has been identified in contemporary fish. Studies of ancient HBV
genomes date an association between HBV and human beings back
as long as 21,000 years ago; primate HBV-like viruses date back mil­
lions of years, suggesting that HBV predated the emergence of modern
humans. Like HBV, all have the same distinctive three morphologic
forms, have counterparts to the envelope and nucleocapsid virus anti­
gens of HBV, replicate in the liver but exist in extrahepatic sites, contain
their own endogenous DNA polymerase, have partially double-strand
and partially single-strand genomes, are associated with acute and
Acute Viral Hepatitis
chronic hepatitis and hepatocellular car­
cinoma, and rely on a replicative strategy
unique among DNA viruses but typi­
cal of retroviruses. Entry of HBV into
hepatocytes is mediated by binding to
the sodium taurocholate cotransporting
polypeptide (NTCP) receptor. Instead
of DNA replication directly from a DNA
template, hepadnaviruses rely on reverse
transcription (effected by the DNA poly­
merase) of minus-strand DNA from a
“pregenomic” RNA intermediate. Then,
plus-strand DNA is transcribed from
the minus-strand DNA template by the
DNA-dependent DNA polymerase and
converted in the hepatocyte nucleus to
a covalently closed circular DNA, which
serves as a template for messenger RNA
and pregenomic RNA. Viral proteins
Jaundice
IgG Anti-HAV
IgM Anti-HAV
ALT
Fecal HAV
Weeks after exposure
FIGURE 350-2  Scheme of typical clinical and laboratory features of hepatitis A
virus (HAV). ALT, alanine aminotransferase.
are translated by the messenger RNA, and the proteins and genome
are packaged into virions and secreted from the hepatocyte. Hepatitis
B virus has been difficult to cultivate in vitro from clinical materials;
therefore, a model that recapitulates the entire HBV life cycle has been
particularly elusive. In recent decades, however, advances in molecular
virology have permitted the comprehensive study of HBV replication
and its viral genes and proteins.
VIRAL PROTEINS AND PARTICLES  Three particulate forms of HBV
appear in the circulation (Table 350-1), the complete virion and two
incomplete or subviral particles. Of these, the most numerous are
the 22-nm particles, which appear as spherical or long filamentous
forms; these are antigenically indistinguishable from the outer surface
or envelope protein of HBV and are thought to represent excess viral
envelope protein. Outnumbered in serum by a factor of 100 or 1000
to 1 compared with the spheres and tubules are large, 42-nm, doubleshelled spherical particles, which represent the intact hepatitis B virion
(Fig. 350-1). The envelope protein expressed on the outer surface of the
virion and on the smaller spherical and tubular structures is referred
to as hepatitis B surface antigen (HBsAg). The concentration of HBsAg
and virus particles in the blood may reach 500 μg/mL and 10 trillion
particles per milliliter, respectively; HBsAg assays in common clinical
use detect all forms of HBsAg (virion and subviral particles). The enve­
lope protein, HBsAg, is the product of the S gene of HBV.
PART 10
Disorders of the Gastrointestinal System
HBV isolates fall into 10 different genotypes (A–J) and multiple sub­
types, which are determined by unique gene sequences of the envelope
Pre-S2
Pre-S1
S
P
C
Pre-C
X
FIGURE 350-3  Compact genomic structure of hepatitis B virus (HBV). This structure,
with overlapping genes, permits HBV to code for multiple proteins. The S gene
codes for the “major” envelope protein, HBsAg. Pre-S1 and pre-S2, upstream of
S, combine with S to code for two larger proteins, “middle” protein, the product of
pre-S2 + S, and “large” protein, the product of pre-S1 + pre-S2 + S. The largest gene,
P, codes for DNA polymerase. The C gene codes for two nucleocapsid proteins,
HBeAg, a soluble, secreted protein (initiation from the pre-C region of the gene),
and HBcAg, the intracellular core protein (initiation after pre-C). The X gene codes
for HBxAg, which can transactivate the transcription of cellular and viral genes; its
clinical relevance is not known, but it may contribute to carcinogenesis by binding
to p53.
HBsAg protein. Envelope HBsAg subdeterminants include a common
group-reactive antigen, a, shared by all HBsAg isolates and one of
several subtype-specific antigens—d or y, w or r. Geographic distribu­
tion of genotypes and subtypes varies; genotypes A (corresponding
to subtype adw) and D (ayw) predominate in the United States and
Europe, whereas genotypes B (adw) and C (adr) predominate in Asia;
however, these geographic distinctions have been blunted by recentdecade migration across continents. Clinical course and outcome
are independent of subtype, but genotype B appears to be associated
with less rapidly progressive liver disease and cirrhosis and a lower
likelihood, or delayed appearance, of hepatocellular carcinoma than
genotype C, D, or F. In a large Japanese cohort, after acute infection,
patients with genotype A were more likely to have persistent infection
(23.4% with genotype A vs 8.6% with non-A genotypes). Also, geno­
type may influence response to treatment with interferon and other
antiviral agents; for example, higher rates of HBeAg and HBsAg loss
have been observed in patients with genotype A. An additional impor­
tant consequence of genotype is the propensity for “precore” mutations
to emerge (see below).
Upstream of the S gene are the pre-S genes (Fig. 350-3), which code
for pre-S gene products, including receptors on the HBV surface for
polymerized human serum albumin and for hepatocyte membrane
proteins. The pre-S region actually consists of both pre-S1 and pre-S2.
Depending on where translation is initiated, three potential HBsAg
gene products are synthesized. The protein product of the S gene is
HBsAg (major protein), the product of the S region plus the adjacent
pre-S2 region is the middle protein, and the product of the pre-S1 plus
pre-S2 plus S regions is the large protein. Compared with the smaller
spherical and tubular particles of HBV, complete 42-nm virions are
enriched in the large protein. Both pre-S proteins and their respective
antibodies can be detected during HBV infection, and the period of
pre-S antigenemia appears to coincide with other markers of virus rep­
lication, as detailed below; however, pre-S proteins have little clinical
relevance and are not included in routine serologic testing repertoires.
The intact 42-nm virion contains a 27-nm nucleocapsid core par­
ticle. Nucleocapsid proteins are coded for by the C gene. The antigen
expressed on the surface of the nucleocapsid core is hepatitis B core
antigen (HBcAg), and its corresponding antibody is anti-HBc. A third
HBV antigen is HBeAg, a soluble, nonparticulate, nucleocapsid protein
that is immunologically distinct from intact HBcAg but is a product
of the same C gene. The C gene has two initiation codons: a precore
and a core region (Fig. 350-3). If translation is initiated at the precore
region, the protein product is HBeAg, which has a signal peptide that
binds it to the smooth endoplasmic reticulum, the secretory apparatus
of the cell, leading to its secretion into the circulation. If translation
begins at the core region, HBcAg is the protein product; it has no signal
peptide and is not secreted, but it assembles into nucleocapsid particles,
which bind to and incorporate RNA and which, ultimately, contain
HBV DNA. Also packaged within the nucleocapsid core is a DNA
polymerase, which directs replication and repair of HBV DNA. When
packaging within viral proteins is complete, synthesis of the incomplete
plus strand stops; this accounts for the single-strand gap and for differ­
ences in the size of the gap. HBcAg particles remain in the hepatocyte,
where they are readily detectable by immunohistochemical staining
and are exported after encapsidation by an envelope of HBsAg. There­
fore, naked core particles do not circulate in the serum. The secreted
nucleocapsid protein, HBeAg, provides a convenient, readily detect­
able, qualitative marker of HBV replication and relative infectivity.
HBsAg-positive serum containing HBeAg is more likely to be highly
infectious and to be associated with the presence of hepatitis B virions
(and detectable HBV DNA, see below) than HBeAg-negative or antiHBe-positive serum. For example, HBsAg-positive mothers who are
HBeAg positive almost invariably (>90%) transmit hepatitis B infec­
tion to their offspring, whereas HBsAg-positive mothers with anti-HBe
rarely (10–15%) infect their offspring.
Early during the course of acute hepatitis B, HBeAg appears tran­
siently; its disappearance may be a harbinger of clinical improvement
and resolution of infection. Persistence of HBeAg in serum beyond the
first 3 months of acute infection may be predictive of the development
TABLE 350-1  Nomenclature and Features of Hepatitis Viruses
HEPATITIS
TYPE
VIRUS
PARTICLE, nm
MORPHOLOGY
GENOMEa
CLASSIFICATION
ANTIGEN(S)
ANTIBODIES
REMARKS
HAV
Icosahedral
nonenveloped
7.5-kb RNA,
linear, ss, +
Hepatovirus
HAV
Anti-HAV
Early fecal shedding
Diagnosis: IgM anti-HAV
Previous infection: IgG anti-HAV
HBV
Double-shelled
virion (surface and
core) spherical 
 
3.2-kb DNA,
circular,
ss/ds
Hepadnavirus
HBsAg
HBcAg
HBeAg 
 
HBcAg
HBeAg
 
Nucleocapsid core
 
 
 
Spherical and
filamentous;
represents excess
virus coat material
HCV
Enveloped
9.4-kb RNA,
linear, ss, +
Hepacivirus
HCV core
antigen
HDV
35–37
Enveloped hybrid
particle with
HBsAg coat and
HDV core
1.7-kb RNA,
circular,
ss, –
Resembles viroids
and plant satellite
viruses (genus
Deltavirus)
HEV
32–34
Nonenveloped
icosahedral
7.6-kb RNA,
linear, ss, +
Orthohepevirus
HEV antigen
Anti-HEV
Agent of enterically transmitted hepatitis;
rare in the United States; occurs in Asia,
Mediterranean countries, Central America
Diagnosis: IgM/IgG anti-HEV (assays not
routinely available); virus in stool, bile,
hepatocyte cytoplasm
ass, single-strand; ss/ds, partially single-strand, partially double-strand; −, minus-strand; +, plus-strand.
Note: See text for abbreviations.
of chronic infection, and the presence of HBeAg during chronic hepa­
titis B tends to be associated with ongoing viral replication, infectivity,
and inflammatory liver injury (except during the early decades after
perinatally acquired HBV infection; see below).
The third and largest of the HBV genes, the P gene (Fig. 350-3),
codes for HBV DNA polymerase; as noted above, this enzyme has
both DNA-dependent DNA polymerase and RNA-dependent reverse
transcriptase activities. The fourth gene, X, codes for a small, non­
particulate protein, hepatitis B x antigen (HBxAg), which is capable of
transactivating the transcription of both viral and cellular genes (Fig.
350-3). This function underpins its role in promoting both viral rep­
lication and HBV-related hepatocellular carcinoma. In the cytoplasm,
HBxAg effects calcium release (possibly from mitochondria), which
activates signal-transduction pathways that lead to stimulation of HBV
reverse transcription and HBV DNA replication. Such transactivation
may enhance the replication of HBV; the transactivating activity can
enhance the transcription and replication of other viruses besides HBV,
such as HIV. The clinical relevance of HBxAg is limited, however, and
testing for it is not part of routine clinical practice.
SEROLOGIC AND VIROLOGIC MARKERS  After a person is infected
with HBV, the first virologic marker detectable in serum within
1–12 weeks, usually between 8 and 12 weeks, is HBsAg (Fig. 350-4).
Circulating HBsAg precedes elevations of serum aminotransferase
activity and clinical symptoms by 2–6 weeks and remains detectable
during the entire icteric or symptomatic phase of acute hepatitis B and
beyond. In typical self-limited cases, HBsAg becomes undetectable 1–2
months after the onset of jaundice and rarely persists beyond 6 months.
After HBsAg disappears, antibody to HBsAg (anti-HBs) becomes
Anti-HBs
Anti-HBc
Anti-HBe 
 
Anti-HBc
Anti-HBe
 
Bloodborne virus; carrier state
Acute diagnosis: HBsAg, IgM anti-HBc
Chronic diagnosis: IgG anti-HBc, HBsAg
Markers of replication: HBeAg, HBV DNA
Liver, lymphocytes, other organs
Nucleocapsid contains DNA and DNA
polymerase; present in hepatocyte nucleus;
HBcAg does not circulate; HBeAg (soluble,
nonparticulate) and HBV DNA circulate—
correlate with infectivity and complete virions
HBsAg detectable in >95% of patients with
acute hepatitis B; found in serum, body fluids,
hepatocyte cytoplasm; anti-HBs appears
following infection—protective antibody
 
 
HBsAg
Anti-HBs
Anti-HCV
Bloodborne agent, formerly labeled non-A,
non-B hepatitis
Acute diagnosis: anti-HCV, HCV RNA
Chronic diagnosis: anti-HCV, HCV RNA;
cytoplasmic location in hepatocytes
HBsAg
HDAg
Anti-HBs
Anti-HDV
Defective RNA virus, requires helper function
of HBV (hepadnaviruses); HDV antigen (HDAg)
present in hepatocyte nucleus
Diagnosis: anti-HDV, HDV RNA; HBV/HDV
co-infection—IgM anti-HBc and anti-HDV; HDV
superinfection—IgG anti-HBc and anti-HDV
CHAPTER 350
Acute Viral Hepatitis
detectable in serum and remains detectable indefinitely thereafter.
Because HBcAg is intracellular and, when in the serum, sequestered
within an HBsAg coat, naked core particles do not circulate in serum,
and therefore, assays for HBcAg are not included in routine clinical
testing. By contrast, anti-HBc is readily demonstrable in serum, begin­
ning within the first 1–2 weeks after the appearance of HBsAg and
preceding detectable levels of anti-HBs by weeks to months. Because
variability exists in the time of appearance of anti-HBs after HBV
Jaundice
ALT
HBeAg
Anti-HBe
IgG Anti-HBc
HBsAg
IgM Anti-HBc
Anti-HBs
Weeks after exposure
FIGURE 350-4  Scheme of typical clinical and laboratory features of acute hepatitis
B. ALT, alanine aminotransferase.
infection, occasionally a gap of several weeks or longer may separate
the disappearance of HBsAg and the appearance of anti-HBs. Dur­
ing this “gap” or “window” period, IgM anti-HBc may represent the
only serologic evidence of current or recent HBV infection, and blood
containing anti-HBc in the absence of HBsAg and anti-HBs had been
implicated in the past in transfusion-associated hepatitis B. In part
because the sensitivity of immunoassays for HBsAg and anti-HBs
has increased, however, this window period is rarely encountered. In
some persons, years after HBV infection, anti-HBc may persist in the
circulation longer than anti-HBs. Therefore, isolated anti-HBc does not
necessarily indicate active virus replication; most instances of isolated
anti-HBc represent hepatitis B infection in the remote past. Rarely,
however, isolated anti-HBc represents low-level hepatitis B viremia,
with HBsAg below the detection threshold, and occasionally, isolated
anti-HBc represents a cross-reacting or false-positive immunologic
specificity.
Recent and remote HBV infections can be distinguished by deter­
mination of the immunoglobulin class of anti-HBc. Anti-HBc of the
IgM class (IgM anti-HBc) predominates during the first 6 months after
acute infection, whereas IgG anti-HBc is the predominant class of antiHBc beyond 6 months. Therefore, patients with current or recent acute
hepatitis B, including those in the anti-HBc window, have IgM antiHBc in their serum. In patients who have recovered from hepatitis B in
the remote past as well as those with chronic HBV infection, anti-HBc
is predominantly of the IgG class. Infrequently, in ≤1–5% of patients
with acute HBV infection, levels of HBsAg are too low to be detected;
in such cases, the presence of IgM anti-HBc establishes the diagnosis
of acute hepatitis B. When isolated anti-HBc occurs in the rare patient
with chronic hepatitis B whose HBsAg level is below the sensitivity
threshold of contemporary immunoassays (a low-level carrier), antiHBc is of the IgG class. Generally, in persons who have recovered from
hepatitis B, anti-HBs and anti-HBc persist indefinitely.
PART 10
Disorders of the Gastrointestinal System
The temporal association between the appearance of anti-HBs and
resolution of HBV infection as well as the observation that persons
with anti-HBs in serum are protected against reinfection with HBV
suggests that anti-HBs is the protective antibody. Therefore, strategies
for prevention of HBV infection are based on providing susceptible
persons with circulating anti-HBs (see below). Occasionally, in ~5–10%
of patients with chronic hepatitis B, low-level, low-affinity anti-HBs can be
detected. Although an unusual serologic pattern in chronic hepatitis B,
coexisting HBsAg and anti-HBs positivity does not signal imminent
clearance of hepatitis B. Recently, the presence of coexisting HBsAg
and anti-HBs in 4.2% of 6534 Chinese patients with chronic hepatitis B
was reported (in HBeAg-negative but not HBeAg-reactive patients)
to be associated with a higher risk of advanced fibrosis and cirrhosis.
The fact that the coexistent HBsAg–anti-HBs subset was significantly
older suggests that the link with advanced disease was predominantly
the result of longer-duration disease. These patients with HBsAg and
such nonneutralizing anti-HBs should be categorized as having chronic
HBV infection and, because of the reported association with advanced
fibrosis, evaluated for a higher risk of advanced disease.
The other readily detectable serologic marker of HBV infection,
HBeAg, appears concurrently with or shortly after HBsAg. Its appear­
ance coincides temporally with high levels of virus replication and
reflects the presence of circulating intact virions and detectable HBV
DNA (with the notable exception of patients with precore mutations
who cannot synthesize HBeAg—see “Molecular Variants”). Pre-S1 and
pre-S2 proteins are also expressed during periods of peak replication,
but assays for these gene products are not routinely available. In selflimited HBV infections, HBeAg becomes undetectable shortly after
peak elevations in aminotransferase activity, before the disappearance
of HBsAg, and anti-HBe then becomes detectable, coinciding with
a period of relatively lower infectivity (Fig. 350-4). Because HBeAg
appears transiently (and HBV DNA, see below, is always present) dur­
ing typical cases of acute infection, testing for HBeAg and HBV DNA
is of limited clinical utility; in contrast, testing for the presence of these
markers is of substantial importance in patients with chronic infection.
Departing from the pattern typical of acute HBV infections, in
chronic HBV infection, HBsAg remains detectable beyond 6 months,
ALT
HBeAg
Anti-HBe
HBV   DNA
HBsAg
Anti-HBc
IgM   anti-HBc
Months after exposure
1 2 3 4 5
FIGURE 350-5  Scheme of typical laboratory features of wild-type chronic
hepatitis B. HBeAg and hepatitis B virus (HBV) DNA can be detected in serum
during the relatively replicative phase of chronic infection, which is associated
with infectivity and liver injury. Seroconversion from the replicative phase to the
relatively nonreplicative phase occurs at a rate of ~10% per year and is heralded by
an acute hepatitis–like elevation of alanine aminotransferase (ALT) activity; during
the nonreplicative phase, infectivity and liver injury are limited. In HBeAg-negative
chronic hepatitis B associated with mutations in the precore region of the HBV
genome, replicative chronic hepatitis B occurs in the absence of HBeAg.
anti-HBc is primarily of the IgG class, and anti-HBs is either undetect­
able or detectable at low levels (see “Laboratory Features”) (Fig. 350-5).
During early chronic HBV infection, HBV DNA can be detected both
in serum and in hepatocyte nuclei, where it is present in free or epi­
somal form. This relatively highly replicative stage of HBV infection is
the time of maximal infectivity and liver injury; HBeAg is a qualitative
marker and HBV DNA a quantitative marker of this replicative phase,
during which all three forms of HBV circulate, including intact virions.
Over time, the relatively replicative phase of chronic HBV infection
gives way to a relatively nonreplicative phase. This occurs at a rate of
~10% per year and is accompanied by seroconversion from HBeAg to
anti-HBe. In many cases, this seroconversion coincides with a tran­
sient, usually mild, acute hepatitis-like elevation in aminotransferase
activity, believed to reflect cell-mediated immune clearance of virusinfected hepatocytes. In this relatively nonreplicative or low-replicative
phase of chronic infection, when HBV DNA is demonstrable in hepa­
tocyte nuclei, it tends to be integrated into the host genome. In this
phase, only spherical and tubular forms of HBV, not intact virions,
circulate, and liver injury tends to subside. In this phase, clinical mea­
sures of liver injury (specifically alanine aminotransferase [ALT]) are
normal. Most such patients would be characterized as inactive HBV
carriers. In reality, the designations replicative and nonreplicative are
only relative; even in the so-called nonreplicative phase, circulating
HBV DNA can be detected at levels of approximately ≤103 virions/mL
with highly sensitive amplification probes such as the polymerase chain
reaction (PCR); below this replication threshold, liver injury and infec­
tivity of HBV are limited to negligible. Still, the distinctions are patho­
physiologically and clinically meaningful. While true inactive carriers
rarely progress to an immune active state, in 5% of cases, relatively
nonreplicative HBeAg-negative chronic HBV infection converts back
to replicative infection with re-expression of HBeAg. Such spontaneous
reactivations may also be accompanied by expression of IgM anti-HBc,
as well as by exacerbations of liver injury. Because high-titer IgM antiHBc can reappear during acute exacerbations of chronic hepatitis B,
relying on IgM anti-HBc versus IgG anti-HBc to distinguish between
acute and chronic hepatitis B infection, respectively, may not always
be reliable; in such cases, patient history and additional follow-up
monitoring over time are invaluable in helping to distinguish de novo
acute hepatitis B infection from acute exacerbation of chronic hepatitis
B infection.
MOLECULAR VARIANTS  Variation occurs throughout the HBV
genome, and clinical isolates of HBV that do not express typical viral
proteins have been attributed to mutations in individual or even multi­
ple gene locations. For example, variants have been described that lack
nucleocapsid proteins (commonly), envelope proteins (very rarely),
or both. Two categories of naturally occurring HBV variants have
attracted the most attention: precore mutations and escape mutations.
One precore mutation was identified initially in Mediterranean
countries among patients with severe chronic HBV infection and
detectable HBV DNA, but with anti-HBe instead of HBeAg. These
patients were found to be infected with an HBV mutant that contained
an alteration in the precore region, rendering the virus incapable of
encoding HBeAg. Although several potential mutation sites exist in
the pre-C region, the region of the C gene necessary for the expression
of HBeAg (see “Virology and Etiology”), the most commonly encoun­
tered in such patients is a single base substitution, from G to A in the
second to last codon of the pre-C gene at nucleotide 1896. This sub­
stitution results in the replacement of the TGG tryptophan codon by a
stop codon (TAG), which prevents the translation of HBeAg. Another
mutation, in the core-promoter region, prevents transcription of the
coding region for HBeAg and yields an HBeAg-negative phenotype.
Patients with such mutations in the precore region and who are unable
to secrete HBeAg may have severe liver disease that progresses more
rapidly to cirrhosis, or alternatively, they are identified clinically later
in the course of the natural history of chronic hepatitis B, when the
disease is more advanced. Both “wild-type” HBV and precore-mutant
HBV can coexist in the same patient, or mutant HBV may arise late
during wild-type HBV infection. Clusters of fulminant hepatitis B
have been observed in Japan and Israel among patients with precore
mutant mutations, although, typically, in Europe and North America,
most cases of fulminant hepatitis B occur in patients with wild-type
virus. HBeAg-negative chronic hepatitis with mutations in the precore
region is now the most frequently encountered form of hepatitis B in
Mediterranean countries and in Europe. In the United States, where
HBV genotype A (less prone to G1896A mutation) is prevalent, pre­
core-mutant HBV had been much less common; however, as a result of
immigration from Asia and Europe, the proportion of HBeAg-negative
hepatitis B–infected persons has increased in the United States, now
accounting for ~30–40% of patients with chronic hepatitis B. Char­
acteristic of such HBeAg-negative chronic hepatitis B are lower levels
of HBV DNA (usually ≤105 IU/mL) and one of several patterns of
aminotransferase activity—persistent elevations, periodic fluctuations
above the normal range, and periodic fluctuations between the normal
and elevated range.
The second important category of HBV mutants consists of escape
mutants, in which a single amino acid substitution, from glycine to
arginine, occurs at position 145 of the immunodominant a determi­
nant common to all HBsAg subtypes. This HBsAg alteration leads to
a critical conformational change that results in a loss of neutralizing
activity by anti-HBs. This specific HBV/a mutant has been observed
in two situations, active and passive immunization, in which humoral
immunologic pressure may favor evolutionary change (“escape”) in
the virus—in a small number of hepatitis B vaccine recipients who
acquired HBV infection despite the prior appearance of neutralizing
anti-HBs and in HBV-infected liver transplant recipients treated with
a high-potency human monoclonal anti-HBs preparation. Although
such mutants have not been recognized frequently, their existence
raises a concern that may complicate vaccination strategies and sero­
logic diagnosis.
Different types of mutations emerge during antiviral therapy of
chronic hepatitis B with nucleoside analogues; such YMDD and
similar mutations in the polymerase motif of HBV are described in
Chap. 352.
EXTRAHEPATIC SITES  Hepatitis B antigens and HBV DNA have been
identified in extrahepatic sites, including the lymph nodes, bone mar­
row, circulating lymphocytes, spleen, and pancreas. Although the virus
does not appear to be associated with tissue injury in any of these extra­
hepatic sites, its presence in these “remote” reservoirs has been invoked
(but is not necessary) to explain the recurrence of HBV infection after
orthotopic liver transplantation. The clinical relevance of such extra­
hepatic HBV is limited.
Hepatitis D 
The delta hepatitis agent, or HDV, the only member
of the genus Deltavirus, is a defective RNA virus that co-infects with
and requires the helper function of HBV (or other hepadnaviruses)
for its replication and expression. Slightly smaller than HBV, HDV
is a formalin-sensitive, 35- to 37-nm virus with a hybrid structure.
Its nucleocapsid expresses HDV antigen (HDAg), which bears no
antigenic homology with any of the HBV antigens, and contains the
virus genome. The HDV core is “encapsidated” by an outer envelope
of HBsAg, indistinguishable from that of HBV except in its relative
compositions of major, middle, and large HBsAg component proteins.
The genome is a small, 1700-nucleotide, circular, single-strand RNA
of negative polarity that is nonhomologous with HBV DNA (except
for a small area of the polymerase gene) but that has features and
the rolling circle model of replication common to genomes of plant
satellite viruses or viroids. HDV RNA contains many areas of internal
complementarity; therefore, it can fold on itself by internal base pairing
to form an unusual, very stable, rod-like structure that contains a very
stable, self-cleaving and self-ligating ribozyme. HDV RNA requires
host RNA polymerase II for its replication in the hepatocyte nucleus
via RNA-directed RNA synthesis by transcription of genomic RNA
to a complementary antigenomic (plus strand) RNA; the antigenomic
RNA, in turn, serves as a template for subsequent genomic RNA syn­
thesis effected by host RNA polymerase I. HDV RNA has only one
open reading frame, and HDAg, a product of the antigenomic strand,
is the only known HDV protein; HDAg exists in two forms: a small,
195-amino-acid species, which plays a role in facilitating HDV RNA
replication, and a large, 214-amino-acid species, which appears to
suppress replication but is required for assembly of the antigen into
virions. HDV antigens have been shown to bind directly to RNA
polymerase II, resulting in stimulation of transcription. Viral assem­
bly requires farnesylation of the large HDAg for ribonucleoprotein
anchoring to HBsAg. Both HBV and HDV enter hepatocytes via the
NTCP receptor. Although complete hepatitis D virions and liver injury
require the cooperative helper function of HBV, intracellular replica­
tion of HDV RNA can occur without HBV. Genomic heterogeneity
among HDV isolates has been described. Although pathophysiologic
and clinical consequences of this genetic diversity have not been estab­
lished definitively, preliminarily, genotype 2 has been linked to milder
disease and genotype 3 to severe acute disease. The clinical spectrum of
hepatitis D is common to all eight genotypes identified, the predomi­
nant of which is genotype 1.
CHAPTER 350
Acute Viral Hepatitis
HDV can either infect a person simultaneously with HBV (co-infection)
or superinfect a person already infected with HBV (superinfection). In
instances of superinfection, when HDV infection is transmitted from
a donor with one HBsAg subtype to an HBsAg-positive recipient with
a different subtype, HDV assumes the HBsAg subtype of the recipient,
rather than the donor. Because HDV relies absolutely on HBV for its
replication, the duration of HDV infection is determined by the dura­
tion of (and cannot outlast) HBV infection. HDV replication tends
to suppress HBV replication; therefore, patients with hepatitis D tend
to have lower levels of HBV replication. HDV antigen is expressed
primarily in hepatocyte nuclei and is occasionally detectable in serum.
During acute HDV infection, anti-HDV of the IgM class predominates,
and 30–40 days may elapse after symptoms appear before anti-HDV
can be detected. In self-limited infection, anti-HDV is low-titer and
transient, rarely remaining detectable beyond the clearance of HBsAg
and HDV antigen. In chronic HDV infection, anti-HDV circulates in
high titer, and both IgM and IgG anti-HDV can be detected. HDV
antigen in the liver and HDV RNA in serum and liver can be detected
during HDV replication.
Hepatitis C 
Hepatitis C virus, which, before its identification, was
labeled “non-A, non-B hepatitis,” is a linear, single-strand, positivesense, 9600-nucleotide RNA virus, the genome of which is similar
in organization to that of flaviviruses and pestiviruses; HCV is the
only member of the genus Hepacivirus in the family Flaviviridae.
The HCV genome contains a single, large open reading frame (ORF)
(gene) that codes for a virus polyprotein of ~3000 amino acids, which
is cleaved after translation to yield 10 viral proteins. The 5′ end of
the genome consists of an untranslated
region (containing an internal ribo­
somal entry site [IRES]) adjacent to the
genes for three structural proteins, the
nucleocapsid core protein, C, and two
envelope glycoproteins, E1 and E2. The
5′ untranslated region and core gene are
highly conserved among genotypes, but
the envelope proteins are coded for by
the hypervariable region, which varies
from isolate to isolate and may allow
the virus to evade host immunologic
containment directed at accessible virusenvelope proteins. The 3′ end of the
genome also includes an untranslated
region and contains the genes for seven
nonstructural (NS) proteins: p7, NS2,
NS3, NS4A, NS4B, NS5A, and NS5B.
p7 is a membrane ion channel pro­
tein necessary for efficient assembly and
release of HCV. The NS2 cysteine protease cleaves NS3 from NS2, and
the NS3-4A serine protease cleaves all the downstream proteins from
the polyprotein. Important NS proteins involved in virus replication
include the NS3 helicase; NS3-4A serine protease; the multifunctional
membrane-associated phosphoprotein NS5A, an essential component
of the viral replication membranous web (along with NS4B); and the
NS5B RNA-dependent RNA polymerase (Fig. 350-6). Because HCV
does not replicate via a DNA intermediate, it does not integrate into
the host genome. Because HCV tends to circulate in relatively low titer,
103−107 virions/mL, visualization of the 50- to 80-nm virus particles
remains difficult. Still, the replication rate of HCV is very high, 1012
virions per day; its half-life is 2.7 h. Historically, the study of HCV had
been hampered by few adequate in vitro and animal models. In 2005,
however, complete replication of HCV and intact 55-nm virions was
described in cell culture systems. These and other subsequent models
were fundamental toward the development of targeted, effective antivi­
ral therapies for HCV infection. Still, the continued absence of robust
animal models poses a barrier to much-needed vaccine development.
Albeit a helpful animal model, the chimpanzee is cumbersome to study,
and access to chimpanzees for medical research has been curtailed
substantially. In addition, while HCV replication has been documented
in a xenograft immunodeficient mouse model containing explants
of human liver and in transgenic mouse and rat models, a tractable
immunocompetent murine model has remained elusive.
AA
Envelope
glycoproteins
Core
5'
3'
C
E1
E2
NS2
NS3
NS4B
NS5A
NS5B
Conserved
region
Hypervariable
region
FIGURE 350-6  Organization of the hepatitis C virus genome and its associated, 3000-amino-acid (AA) proteins. The
three structural genes at the 5′ end are the core region, C, which codes for the nucleocapsid, and the envelope
regions, E1 and E2, which code for envelope glycoproteins. The 5′ untranslated region and the C region are highly
conserved among isolates, whereas the envelope domain E2 contains the hypervariable region. At the 3′ end are seven
nonstructural (NS) regions—p7, a membrane protein adjacent to the structural proteins that appears to function as an
ion channel; NS2, which codes for a cysteine protease; NS3, which codes for a serine protease and an RNA helicase;
NS4 and NS4B; NS5A, a multifunctional membrane-associated phosphoprotein, an essential component of the viral
replication membranous web; and NS5B, which codes for an RNA-dependent RNA polymerase. After translation of the
entire polyprotein, individual proteins are cleaved by both host and viral proteases.
PART 10
Disorders of the Gastrointestinal System
In vitro study has also elucidated important cofactors for the
hepatitis C viral life cycle. For example, HCV entry into the hepatocyte
occurs via the non-liver-specific CD81 receptor and the liver-specific
tight junction protein claudin-1. A growing list of additional host
receptors to which HCV binds upon cell entry includes occludin, lowdensity lipoprotein receptors, glycosaminoglycans, scavenger receptor
B1, and epidermal growth factor receptor, among others. After viral
entry and uncoating, translation is initiated by the IRES on the endo­
plasmic reticulum membrane, and the HCV polyprotein is cleaved dur­
ing translation and posttranslationally by host cellular proteases as well
as HCV NS2-3 and NS3-4A proteases. Host cofactors involved in HCV
replication include cyclophilin A, which binds to NS5A and yields
conformational changes required for viral replication, and the liverspecific host microRNA miR-122. Relying on the same assembly and
secretion pathway as low-density and very-low-density lipoproteins,
HCV is a lipoviroparticle and masquerades as a lipoprotein, which may
limit its visibility to the adaptive immune system, explain its ability to
evade immune containment and clearance, and account for the lower
circulating LDL levels observed in patients with chronic hepatitis C.
At least six distinct major genotypes (and a minor genotype 7),
as well as >50 subtypes within genotypes, of HCV have been identi­
fied by nucleotide sequencing. Some HCV genotypes are distributed
worldwide, whereas others are more confined geographically (see
“Epidemiology and Global Features”). Genotype differences can affect
choice of direct-acting antiviral treatment and can affect clinical course
Helicase
Serine
protease
RNA-dependent
RNA polymerase
p7
NS4A
(e.g., hepatic steatosis and clinical progression are more likely in geno­
type 3). Genotypes differ from one another in sequence homology by
≥30%, and subtypes differ by ~20%. Because divergence of HCV iso­
lates within a genotype or subtype and within the same host may vary
insufficiently to define a distinct genotype, these intragenotypic differ­
ences are referred to as quasispecies and differ in sequence homology by
only a few percent. The genotypic and quasispecies diversity of HCV,
resulting from its high mutation rate, interferes with effective humoral
immunity. Neutralizing antibodies to HCV have been demonstrated,
but they tend to be short-lived, and HCV infection does not induce
lasting immunity against reinfection with different virus isolates or
even the same virus isolate. Indeed, the presence of detectable antiHCV does not indicate protection against re-infection. Thus, neither
heterologous nor homologous immunity appears to develop commonly
after acute HCV infection
Currently available, third-generation immunoassays, which incor­
porate proteins from the core, NS3, and NS5 regions, detect anti-HCV
antibodies during acute infection. The most sensitive indicator of
HCV infection is the presence of HCV RNA, which requires molecular
amplification, for example, by PCR (Fig. 350-7). To allow standardiza­
tion of the quantification of HCV RNA among laboratories and com­
mercial assays, HCV RNA is reported as international units (IUs) per
milliliter; quantitative assays with a broad dynamic range are available
that allow detection of HCV RNA with a sensitivity as low as 5 IU/mL.
HCV RNA can be detected within a few days of exposure to HCV—
well before the appearance of anti-HCV—and tends to persist for the
duration of HCV infection. Application of sensitive molecular probes
for HCV RNA has revealed the presence of replicative HCV in periph­
eral blood lymphocytes of infected persons; however, as is the case for
HBV in lymphocytes, the clinical relevance of HCV lymphocyte infec­
tion is not known.
Anti-HCV
HCV  RNA
ALT
0 1 2 3 4 5 6
Months after exposure
FIGURE 350-7  Scheme of typical laboratory features during acute hepatitis C
progressing to chronicity. Hepatitis C virus (HCV) RNA is the first detectable
event, preceding alanine aminotransferase (ALT) elevation and the appearance of
anti-HCV.
Hepatitis E 
Previously labeled epidemic or enterically transmitted
non-A, non-B hepatitis, HEV is an enterically transmitted virus that
causes clinically apparent hepatitis primarily in India, Asia, Africa,
and Central America. In those geographic areas, HEV is the most
common cause of acute hepatitis; one-third of the global population
appears to have been infected. This agent, with epidemiologic features
resembling those of hepatitis A, is a 27- to 34-nm, nonenveloped, heatstable, HAV-like virus with a 7200-nucleotide, single-strand, positivesense RNA genome. Like HAV, HEV also exists in a quasi-enveloped
form enclosed within host-cell-derived membranes. HEV has three
overlapping ORFs (genes), the largest of which, ORF1, encodes non­
structural proteins involved in virus replication (the viral replicase,
which includes a protease, polymerase, and helicase). A middle-sized
gene, ORF2, encodes the nucleocapsid protein, the major structural
protein, and the smallest, ORF3, encodes a small structural phospho­
protein involved in virus particle secretion. All HEV isolates appear
to belong to a single serotype, despite genomic heterogeneity of up to
25% and the existence of four species (A–D) and eight genotypes, only
four of which, all within species A, have been detected in humans.
Genotype associations are important in HEV infection. Genotypes 1
and 2 (common in developing countries) appear to be more virulent
anthrotropic variants. In contrast, genotypes 3 (the most common
in the United States and Europe) and 4 (seen in China), are endemic
in animal species (enzootic variants), represent a zoonotic reservoir
for human infections, and are associated with more attenuated or
subclinical infection in healthy hosts but with chronic infection in
immunocompromised hosts. Contributing to the perpetuation of this
virus are the animal reservoirs described above, most notably in swine
but also in camels, deer, rats, and rabbits, among others. No genomic
or antigenic homology, however, exists between HEV and HAV or
other picornaviruses; and HEV, although resembling caliciviruses, is
sufficiently distinct from any known agent to merit its own classifica­
tion as a unique genus, Orthohepevirus, within the family Hepeviridae
(which includes similar viruses infecting mammals, birds, and fish).
The virus has been detected in stool, bile, and liver and is excreted in
the stool during the late incubation period. Both IgM anti-HEV during
early acute infection and IgG anti-HEV predominating after the first
3 months can be detected (a serologic pattern similar to that of hepatitis
A, see above). The presence of HEV RNA in serum and stool accompa­
nies acute infection; viremia resolves as clinical-biochemical recovery
ensues, while HEV RNA in stool may outlast viremia by several weeks.
Currently, serologic/virologic testing for HEV infection—not approved
or licensed by the U.S. Food and Drug Administration (FDA)—can be
done in specialized laboratories (e.g., the Centers for Disease Control
and Prevention [CDC]) and some commercial laboratories.
■
■PATHOGENESIS
Under ordinary circumstances, none of the hepatitis viruses are known
to be directly cytopathic to hepatocytes. Evidence suggests that the
clinical manifestations and outcomes after acute liver injury associated
with viral hepatitis are determined by the immunologic responses of
the host. Among the viral hepatitides, the immunopathogenesis of
hepatitis B and C has been studied most extensively.
Hepatitis B 
For HBV, the existence of inactive hepatitis B carriers
or “immune tolerant” (see below) patients with normal liver histology
and function suggests that the virus is not directly cytopathic. The fact
that patients with defects in cellular immune competence are more
likely to remain chronically infected rather than to clear HBV supports
the role of cellular immune responses in the pathogenesis of hepatitis B–
related liver injury. The model that has the most experimental support
involves cytolytic T cells sensitized specifically to recognize host and
hepatitis B viral antigens on the liver cell surface. Nucleocapsid pro­
teins (HBcAg and possibly HBeAg), present on the cell membrane in
minute quantities, are the viral target antigens that, with host antigens,
invite cytolytic T cells to destroy HBV-infected hepatocytes. Differ­
ences in the robustness and broad polyclonality of CD8+ cytolytic
T-cell responsiveness; in the level of HBV-specific helper CD4+ T cells;
in attenuation, depletion, and exhaustion of virus-specific T cells; in
viral T-cell epitope escape mutations that allow the virus to evade T-cell
containment; and in the elaboration of antiviral cytokines by T cells
have been invoked to explain differences in outcomes between those
who recover after acute hepatitis and those who progress to chronic
hepatitis or between those with mild and those with severe (fulminant)
acute HBV infection. Ultimately, a robust CD8+ T-cell response is cen­
tral both to viral clearance and to liver injury.
Although a robust cytolytic T-cell response occurs and eliminates
virus-infected liver cells during acute hepatitis B, >90% of HBV DNA
has been found in experimentally infected chimpanzees to disappear
from the liver and blood before maximal T-cell infiltration of the
liver and before most of the biochemical and histologic evidence of
liver injury. This observation suggests that components of the innate
immune system and inflammatory cytokines, independent of cyto­
pathic antiviral mechanisms, participate in the early immune response
to HBV infection; this effect has been shown to represent elimination
of HBV replicative intermediates from the cytoplasm and covalently
closed circular viral DNA from the nucleus of infected hepatocytes. In
turn, the innate immune response to HBV infection is mediated largely
by natural killer (NK) cell cytotoxicity, activated by immunosuppres­
sive cytokines (e.g., interleukin [IL] 10 and transforming growth factor
[TGF] β), reduced signals from inhibitory receptor expression (e.g.,
major histocompatibility complex), or increased signals from activat­
ing receptor expression on infected hepatocytes. In addition, NK cells
reduce helper CD4+ cells, which results in reduced CD8+ cells and
exhaustion of the virus-specific T-cell response to HBV infection.
Adding to the evidence supporting the role of these immunologic
perturbations in the pathogenesis of HBV-associated liver injury are
the observations that many of these departures from normal immune
function are restored after successful antiviral therapy. Ultimately,
HBV-HLA–specific cytolytic T-cell responses of the adaptive immune
system are felt to be responsible for recovery from HBV infection.
CHAPTER 350
Debate continues over the relative importance of viral and host
factors in the pathogenesis of HBV-associated liver injury and its out­
come. As noted above, precore genetic mutants of HBV have been asso­
ciated with the more severe outcomes of HBV infection (severe chronic
and fulminant hepatitis), suggesting that, under certain circumstances,
relative pathogenicity is a property of the virus, not the host. The facts
that concomitant HDV and HBV infections are associated with more
severe liver injury than HBV infection alone and that cells transfected
in vitro with the gene for HDV antigen express HDV antigen and then
become necrotic in the absence of any immunologic influences are also
consistent with a viral effect on pathogenicity. Similarly, in patients
who undergo liver transplantation for end-stage chronic hepatitis B,
occasionally, rapidly progressive liver injury appears in the new liver.
This clinical presentation is associated with an unusual histologic pat­
tern in the new liver, fibrosing cholestatic hepatitis, which, ultrastruc­
turally, appears to represent a choking of the cell with overwhelming
quantities of HBsAg. This observation suggests that, under the influ­
ence of the potent immunosuppressive agents required to prevent
allograft rejection, HBV may have a direct cytopathic effect on liver
cells, independent of the immune system.
Acute Viral Hepatitis
Although the precise mechanism of liver injury in HBV infection
remains elusive, studies of nucleocapsid proteins have shed light on
the profound immunologic tolerance to HBV of babies born to mothers
with highly replicative (HBeAg-positive), chronic HBV infection.
In HBeAg-expressing transgenic mice, in utero exposure to HBeAg,
which is sufficiently small to traverse the placenta, induces T-cell tol­
erance to both nucleocapsid proteins. This, in turn, may explain why,
when infection occurs so early in life, immunologic clearance does not
occur, and protracted, lifelong infection ensues. An alternative expla­
nation proposed to explain why robust liver injury does not accompany
neonatal HBV infection but predisposes to chronic infection is defec­
tive priming of HBV-specific T cells during in utero exposure to HBV.
“IMMUNE TOLERANT” VERSUS “IMMUNE ACTIVE” CHRONIC
HEPATITIS B  An important distinction should be drawn between
HBV infection acquired at birth, common in endemic areas, such as
East Asia, and infection acquired in adulthood, common in the West.
Infection in the neonatal period is associated with the acquisition of
what appears to be a high level of immunologic tolerance to HBV and
absence of an acute hepatitis illness but the almost invariable establish­
ment of chronic, often lifelong infection. Neonatally acquired HBV
infection can culminate decades later in cirrhosis and hepatocellular
carcinoma (see “Complications and Sequelae”). In contrast, when HBV
infection is acquired during adolescence or early adulthood, the host
immune response to HBV-infected hepatocytes tends to be robust,
an acute hepatitis-like illness is the rule, and failure to recover is the
exception. After adulthood-acquired infection, chronicity is uncom­
mon, and the risk of hepatocellular carcinoma is very low. Based on
these observations, some authorities categorize chronic HBV infec­
tion into an “immune tolerant” phase, an “immune active” phase, and
an “inactive” phase. This somewhat simplistic formulation, however,
does not apply at all to typical adult infection with self-limited acute
hepatitis B. Moreover, even in early-life-acquired chronic infection,
this duality is limited by nuance and imprecision in its distinctions. For
example, even among those with neonatally acquired HBV infection, in
whom immunologic tolerance appears to be established, immunologic
responses to HBV infection have been demonstrated (albeit typically at
reduced levels), and intermittent bursts of hepatic necroinflammatory
activity punctuate the early decades of life during which liver injury
appears to be quiescent. Thus, while labeled by some as the “immune
tolerant” phase, it more accurately is a period of dissociation between
high-level HBV replication and a paucity of inflammatory liver injury.
In addition, even when clinically apparent liver injury and progressive
fibrosis emerge and predominate during later decades (the so-called
immune active phase), the level of immunologic tolerance to HBV
and the corresponding failure of immunologic containment and clear­
ance of HBV infection remain substantial. More accurately, in patients
with neonatally acquired HBV infection, a dynamic equilibrium exists
between tolerance and intolerance, the outcome of which determines
the clinical expression of chronic infection. Persons infected as neo­
nates tend to have a relatively higher level of immunologic tolerance
(high replication, low necroinflammatory activity) during the early
decades of life and a relatively lower level (but only rarely a loss) of
tolerance (and necroinflammatory activity reflecting the level of virus
replication) in the later decades of life. Adding to the confusion is an
indeterminate category of patients in a gray zone that borders both
relative immune activity and relative immune tolerance.
PART 10
Disorders of the Gastrointestinal System
Hepatitis C 
Cell-mediated immune responses and elaboration
by T cells of antiviral cytokines contribute to the multicellular innate
and adaptive immune responses involved in the containment of infec­
tion and pathogenesis of liver injury associated with hepatitis C. The
fact that HCV is so efficient in evading these immune mechanisms
is a testament to its highly evolved ability to disrupt host immune
responses at multiple levels. After exposure to HCV, the host cell
identifies viral product motifs (pattern recognition receptors) that dis­
tinguish the virus from “self,” resulting in the elaboration of interferons
and other cytokines that result in activation of innate and adaptive
immune responses. Intrahepatic human leukocyte antigen (HLA)
class 1–restricted cytolytic T cells directed at nucleocapsid, envelope,
and nonstructural viral protein antigens have been demonstrated in
patients with chronic hepatitis C; however, such virus-specific cytolytic
T-cell responses do not correlate adequately with the degree of liver
injury or with recovery. Yet a consensus has emerged supporting a role
in the pathogenesis of HCV-associated liver injury of virus-activated
CD4+ helper T cells that stimulate, via the cytokines they elaborate,
HCV-specific CD8+ cytotoxic T cells. These responses appear to be
more robust (higher in number, more diverse in viral antigen speci­
ficity, more functionally effective, and longer lasting) in those who
recover from HCV infection than in those who have chronic infection.
Contributing to chronic infection are a CD4+ proliferative defect that
results in rapid contraction of CD4+ responses, mutations in CD8+ T cell–
targeted viral epitopes that allow HCV to escape immune-mediated
clearance, and upregulation of inhibitory receptors on functionally
impaired, exhausted T cells. Although attention has focused on adap­
tive immunity, HCV proteins have been shown to interfere with innate
immunity by resulting in blocking of type 1 interferon responses and
inhibition of interferon signaling and effector molecules in the inter­
feron signaling cascade.
Several single nucleotide polymorphisms have been linked with
self-limited hepatitis C, the most convincing of which is the CC hap­
lotype of the IL28B gene, which codes for interferon λ3, a component
of innate immune antiviral defense. The link between non-CC IL28B
polymorphisms and failure to clear HCV infection has been explained
by a chromosome 19q13.13 frameshift variant upstream of IL28B, the
ΔG polymorphism of which creates an ORF in a novel interferon gene
(IFN-λ4) associated with impaired HCV clearance. Also shown to con­
tribute to limiting HCV infection are NK cells of the innate immune
system; in spontaneous clearance, NK cells show enhanced IFN-γ
elaboration and expression of activating receptors. Conversely, in per­
sistent infection, both peripheral cytotoxicity and intrahepatic NK cell
cytotoxicity are dysfunctional. In addition, an important intersection
between IL28B genotype and NK function can be inferred from the
fact that patients with hepatitis C and unfavorable (non-CC, associated
with reduced HCV clearance) IL28B alleles have been shown to have
depressed NK cell/innate immune function. Adding to the complexity
of the immune response, HCV core, NS4B, and NS5B have been shown
to suppress the immunoregulatory nuclear factor (NF)-κB pathway,
resulting in reduced antiapoptotic proteins and a resultant increased
vulnerability to tumor necrosis factor (TNF) α-mediated cell death.
Finally, the rapid mutation rate of the virus impairs the host immune
response; the emergence of substantial viral quasispecies diversity and
HCV sequence variation allows the virus to evade attempts by the
host to contain HCV infection by both humoral and cellular immune
mechanisms.
Hepatitis A and E 
Viral shedding in these acute hepatitides pre­
dates clinical evidence of liver injury, consistent with the absence of a
relationship between viral replication and target-organ injury. Instead,
as shown for hepatitis B and C, in hepatitis A and E, experimental evi­
dence supports a cytolytic CD8+ T-cell response as the instrument of
liver cell injury, in concert with or dwarfed by CD4+ helper T cells or
CD4+ interferon γ–secreting cells. Furthermore, in acute HAV infec­
tion, expansion of CD8+ T cells active against other, non-HAV, viruses
has been demonstrated, and this nonspecific CD8+ T-cell activation
correlated with liver injury. HEV, similar to the other hepatitides, has
also been shown to interfere with host antiviral defenses, such as inter­
feron signaling and effector function, and to downregulate interferonstimulated genes. The demonstration of an activated innate immune
response in patients with these hepatitides argues for a multitude of
immunologic mechanisms in the pathogenesis of the acute liver injury
resulting from HAV and HEV infection.
■
■EXTRAHEPATIC MANIFESTATIONS
Immune complex–mediated tissue damage appears to play a pathoge­
netic role in the extrahepatic manifestations of acute hepatitis B. The
occasional prodromal serum sickness–like syndrome observed in acute
hepatitis B appears to be related to the deposition in tissue blood vessel
walls of HBsAg–anti-HBs circulating immune complexes, leading to
activation of the complement system and depressed serum comple­
ment levels. Other extrahepatic manifestations of HBV infection may
include immune complex–mediated diseases, such as polyarteritis
nodosa (generalized vasculitis affecting small- and medium-sized arte­
rioles), glomerulonephritis with nephrotic syndrome, and cryoglobu­
linemia, which are very rare in acute hepatitis B but far more common
in chronic disease (Chaps. 326 and 375).
Immune complex disorders have been linked, albeit rarely, with both
hepatitis A and E. In hepatitis E, rare neurologic (including GuillainBarré syndrome), renal, pancreatic, and hematologic complications
have been postulated to result from both immunologic mechanisms
and/or direct extrahepatic-site infection with the virus.
■
■PATHOLOGY
Liver biopsy is rarely needed for the diagnosis or management of acute
viral hepatitis, for which clinical features and serologic testing remain
the cornerstone of diagnosis. The typical morphologic lesions of all
types of viral hepatitis are similar and consist of panlobular infiltration
with mononuclear cells, hepatic cell necrosis, hyperplasia of Kupffer
cells, and variable degrees of cholestasis. Hepatic cell regeneration is
present, as evidenced by numerous mitotic figures, multinucleated
cells, and “rosette” or “pseudoacinar” formation. The mononuclear
infiltration consists primarily of small lymphocytes, although plasma
cells and eosinophils occasionally are present. Liver cell damage con­
sists of hepatic cell degeneration and necrosis, cell dropout, ballooning
of cells, and acidophilic degeneration of hepatocytes (forming so-called
Councilman or apoptotic bodies). Large hepatocytes with a groundglass appearance of the cytoplasm may be seen in chronic but not in
acute HBV infection; these cells contain HBsAg and can be identified
histochemically with orcein or aldehyde fuchsin. In uncomplicated
viral hepatitis, the reticulin framework is preserved.
In hepatitis C, the histologic lesion is often remarkable for a relative
paucity of inflammation, a marked increase in activation of sinusoidal
lining cells, lymphoid aggregates, the presence of fat (more frequent
in genotype 3 and linked to increased fibrosis), and, occasionally,
bile duct lesions in which biliary epithelial cells appear to be piled
up without interruption of the basement membrane. Occasionally,
microvesicular steatosis occurs in hepatitis D. In hepatitis E, a common
histologic feature is marked cholestasis. A cholestatic variant of slowly
resolving acute hepatitis A has been described as well.
In the setting of severe acute liver injury, less-specific histologic
findings, including portal and lobular inflammation and confluent
necrosis, may be present. A more severe histologic lesion, bridging
hepatic necrosis, also termed subacute or confluent necrosis or inter­
face hepatitis, is observed occasionally in acute hepatitis. “Bridging”
between lobules results from large areas of hepatic cell dropout, with
collapse of the reticulin framework. Characteristically, the bridge con­
sists of condensed reticulum, inflammatory debris, and degenerating
liver cells that span adjacent portal areas, portal to central veins, or
central vein to central vein. This lesion had been thought to have prog­
nostic significance; in many of the originally described patients with
this lesion, a subacute course terminated in death within several weeks
to months, or severe chronic hepatitis and cirrhosis developed; how­
ever, the association between bridging necrosis and a poor prognosis in
patients with acute hepatitis has not been upheld. Therefore, although
demonstration of this lesion in patients with chronic hepatitis has
prognostic significance (Chap. 352), its demonstration during acute
hepatitis is less meaningful, and liver biopsies to identify this lesion are
no longer undertaken routinely in patients with acute hepatitis. In mas­
sive hepatic necrosis (fulminant hepatitis, “acute yellow atrophy”), the
striking feature at postmortem examination is the finding of a small,
shrunken, soft liver. Histologic examination reveals massive necrosis
and dropout of liver cells of most lobules with extensive collapse and
condensation of the reticulin framework. When histologic documenta­
tion is required in the management of fulminant or very severe hepa­
titis, a biopsy can be done by the angiographically guided transjugular
route, which permits the performance of this invasive procedure in the
presence of severe coagulopathy.
Immunohistochemical and electron-microscopic studies have local­
ized HBsAg to the cytoplasm and plasma membrane of infected liver
cells. In contrast, HBcAg predominates in the nucleus, but, occasion­
ally, scant amounts are also seen in the cytoplasm and on the cell mem­
brane. HDV antigen is localized to the hepatocyte nucleus, whereas
HAV and HCV antigens are localized to the cytoplasm. Hepatitis E
ORF-2 protein staining is distributed in both a cytoplasmic and nuclear
pattern.
■
■EPIDEMIOLOGY AND GLOBAL FEATURES
Before the availability of serologic tests for hepatitis viruses, all viral
hepatitis cases were labeled either as “infectious” or “serum” hepatitis.
Modes of transmission overlap, however, and a clear distinction among
the different types of viral hepatitis cannot be made solely based on clini­
cal or epidemiologic features (Table 350-2). The most accurate means
to distinguish the various types of viral hepatitis involves specific
serologic testing.
Hepatitis A 
This agent is transmitted almost exclusively by the
fecal-oral route. Person-to-person spread of HAV is enhanced by poor
personal hygiene and overcrowding; large outbreaks as well as spo­
radic cases have been traced to contaminated food, water, milk, frozen
raspberries and strawberries, green onions imported from Mexico, and
shellfish (e.g., scallops imported from the Philippines used to make
sushi, the culprit identified in a 2016 Hawaiian outbreak). Intrafamily
and intrainstitutional spreads are also common. Early epidemiologic
observations supported a predilection for hepatitis A to occur in late
fall and early winter. In temperate zones, epidemic waves have been
recorded every 5–20 years as new segments of nonimmune population
appeared; however, in developed countries, the incidence of hepatitis
A has been declining, presumably as a function of improved sanitation
and environmental hygiene, and these cyclic patterns are no longer
observed; still, episodic outbreaks among certain high-risk populations
continue to be reported (see below). No HAV carrier state has been
identified after acute hepatitis A; perpetuation of the virus in nature
depends presumably on nonepidemic, inapparent subclinical infection,
ingestion of contaminated food or water in, or imported from, endemic
areas, and/or contamination linked to environmental reservoirs.
In the general population, anti-HAV, a marker for previous HAV
infection, increases in prevalence as a function of increasing age and
of decreasing socioeconomic status. In the 1970s, serologic evidence
of prior hepatitis A infection occurred in ~40% of urban popula­
tions in the United States, most of whose members never recalled
having had a symptomatic case of hepatitis. In subsequent decades,
however, the prevalence of anti-HAV declined in the United States. In
developing countries, exposure, infection, and subsequent immunity
are almost universal in childhood. As the frequency of subclinical
childhood infections declines in developed countries, a susceptible
cohort of adults emerges. Hepatitis A tends to be more symptomatic
in adults; therefore, paradoxically, as the frequency of HAV infection
declines, the likelihood of clinically apparent, even severe, HAV ill­
nesses increases in the susceptible adult population. Travel to endemic
areas is a common source of infection for adults from nonendemic
areas. Important recognized epidemiologic foci of HAV infection
include childcare centers, neonatal intensive care units, promiscuous
men who have sex with men, injection drug users, and unvaccinated
close contacts of newly arrived international adopted children, most
of whom emanate from countries with intermediate-to-high hepatitis
A endemicity. Although hepatitis A is rarely bloodborne, several out­
breaks have been recognized in recipients of clotting-factor concen­
trates. In the United States, the introduction of hepatitis A vaccination
programs among children from high-incidence states has resulted in
a >70% reduction in the annual incidence of new HAV infections and
has shifted the burden of new infections from children to adults. In
the 2007–2012 U.S. Public Health Service National Health and Nutri­
tion Examination Survey (NHANES), the prevalence of anti-HAV in
the U.S. population aged ≥20 years had declined to 24.2% from the
29.5% measured in NHANES 1999–2006. From 2007 to 2016, the
overall anti-HAV prevalence rose, from 30% in 2007–2010 to 40% in
2011–2016 in U.S.-born persons over the age of 2; the largest increases
were observed in children (2–11 years) and teenagers (12–19 years),
with a slight increase among young adults (age 20–29). In all other age
cohorts among U.S.-born adults, however, the prevalence of anti-HAV
remained unchanged between 2007 and 2017; no more than 25% of
any age group had protective antibodies. The lowest age-specific preva­
lence of anti-HAV (16.1–17.6%) occurred in adults in the fourth and
fifth decades (aged 30–49 years). This is a subgroup of the population
who remain susceptible to acute hepatitis A acquired during travel to
endemic areas and from contaminated foods, especially those imported
from endemic countries. Recognized initially in San Diego, California,
in 2016, widespread person-to-person outbreaks, attributed to fecally
contaminated environments, of acute hepatitis A occurred primarily
among homeless persons and persons who were using injection drugs.
Ultimately, this outbreak extended to at least 32 states (highest number
of cases in Kentucky), and by March 2020, 31,950 cases were reported,
resulting in 19,548 hospitalizations (61% of cases) and 322 deaths (1%
of reported cases, 1.6% of hospitalized cases). By 2022, 315 hepatitis A
CHAPTER 350
Acute Viral Hepatitis
TABLE 350-2  Clinical and Epidemiologic Features of Viral Hepatitis
FEATURE
HAV
HBV
HCV
HDV
HEV
Incubation (days)
15–45, mean 30
30–180, mean 60–90
15–160, mean 50
30–180, mean 60–90
14–60, mean 40
Onset
Acute
Insidious or acute
Insidious or acute
Insidious or acute
Acute
Age preference
Children, young
adults
Young adults (sexual and
percutaneous), babies,
toddlers
Transmission
  Fecal-oral
  Percutaneous
  Perinatal
  Sexual
 
+++
Unusual
−
±
 
−
+++
+++
++
Clinical
  Severity
  Fulminant
  Progression to chronicity
  Carrier
  Cancer
  Prognosis
 
Mild
0.1%
None
None
None
Excellent
 
Occasionally severe
0.1–1%
Occasional (1–10%)
(90% of neonates)
0.1–30%f + (neonatal infection)
Worse with age, debility
Prophylaxis
Ig, inactivated
vaccine
HBIG, recombinant vaccine
None
HBV vaccine (none for HBV
carriers)
Therapy
None
Interferonh
Lamivudineh
Adefovirh
Pegylated interferoni
Entecaviri
PART 10
Disorders of the Gastrointestinal System
Telbivudinei
Tenofovir disoproxil fumaratei
Tenofovir alafenamidei
aPrimarily with HIV co-infection and high-level viremia in index case; more likely in persons with multiple sex partners or sexually transmitted diseases; risk ~5%. bUp
to 5% in acute HBV/HDV co-infection; up to 20% in HDV superinfection of chronic HBV infection. e10–20% in pregnant women. dIn acute HBV/HDV co-infection, the
frequency of chronicity is the same as that for HBV; in HDV superinfection, chronicity is invariable. eExcept as observed in immunosuppressed liver allograft recipients or
other immunosuppressed hosts. fVaries considerably throughout the world and in subpopulations within countries; see text. gCommon in Mediterranean countries; rare in
North America and western Europe. hNo longer recommended or not included in first-line therapy. iFirst-line agents. jAnecdotal reports and retrospective studies suggest
that pegylated interferon and/or ribavirin are effective in treating chronic hepatitis E, observed in immunocompromised persons; ribavirin monotherapy has been used
successfully in acute, severe hepatitis E.
Abbreviation: HBIG, hepatitis B immunoglobulin. See text for other abbreviations.
outbreak–related deaths were reported across 37 states. The increased
clinical severity, rate of hospitalization, and death in these outbreaks
can be attributed to their involving an older population (mean age
ranging from 36 to 42 years; median age of death was 55), born before
the introduction of universal childhood hepatitis A vaccination and
in whom clinical severity, as noted above, is higher than in children.
Moreover, the affected homeless and drug-using populations suffer
from multiple comorbidities (including HBV or HCV co-infection)
and disparities in access to health care. Addressing this multistate
outbreak has required a vigorous hepatitis A vaccination effort (still
falling short of adequate coverage) as well as environmental sanitation/
hygiene and education among these susceptible populations.
Hepatitis B 
Percutaneous inoculation has long been recognized as
a major route of hepatitis B transmission, but the outmoded designa­
tion “serum hepatitis” is an inaccurate label for the epidemiologic
spectrum of HBV infection. As detailed below, most of the hepatitis
transmitted by blood transfusion is not caused by HBV; moreover, in
approximately two-thirds of patients with acute type B hepatitis, no
history of an identifiable percutaneous exposure can be elicited. We
now recognize that many cases of hepatitis B result from less obvi­
ous modes of nonpercutaneous or covert percutaneous transmission.
HBsAg has been identified in almost every body fluid from infected
persons, and at least some of these body fluids—most notably semen
and saliva—are infectious, albeit less so than serum, when adminis­
tered percutaneously or nonpercutaneously to experimental animals.
Among the nonpercutaneous modes of HBV transmission, oral
Any age, but more common
in adults
Any age (similar to HBV)
Epidemic cases: young
adults (20–40 years);
sporadic cases: older
adults (>60)
 
−
+++
±a
 
−
+++
+
++
 
+++
−
−
−
±a
 
Moderate
0.1%
Common (85%)
1.5–3.2%
+
Moderate
 
Occasionally severe
5–20%b
 
Mild
1–2%c
Commond
Nonee
Variableg
None
None
Good
±
Acute, good; chronic, poor
Vaccine
Pegylated interferon
ribavirin,h telaprevir,h
boceprevir,h simeprevir,h
sofosbuvir, ledipasvir,
paritaprevir/ritonavir,h
ombitasvir,h dasabuvir,h
daclatasvir,h velpatasvir,
grazoprevir, elbasvir,
glecaprevir, pibrentasvir,
voxilaprevir
Pegylated interferon ±
Nonej
ingestion has been documented as a potential but inefficient route of
exposure. By contrast, the two nonpercutaneous routes considered to
have the greatest impact are intimate (especially sexual) contact and
perinatal transmission.
In sub-Saharan Africa, intimate contact among toddlers is con­
sidered instrumental in contributing to the maintenance of the high
frequency of hepatitis B in the population. Perinatal transmission
occurs primarily in infants born to mothers with chronic hepatitis B
or (rarely) mothers with acute hepatitis B during the third trimester of
pregnancy or during the early postpartum period. Perinatal transmis­
sion is uncommon in North America and western Europe but occurs
with great frequency and is the most important mode of HBV per­
petuation in East Asia and developing countries. Although the precise
mode of perinatal transmission is unknown, and although ~10% of
infections may be acquired in utero, epidemiologic evidence suggests
that most infections occur approximately at the time of delivery and are
not related to breast-feeding (which is not contraindicated in women
with hepatitis B). The likelihood of perinatal transmission of HBV
correlates with the presence of HBeAg and high-level viral replication;
90% of HBeAg-positive mothers but only 10–15% of anti-HBe-positive
mothers transmit HBV infection to their offspring. In most cases, acute
infection in the neonate is clinically asymptomatic, but the child is very
likely to remain chronically infected.
The over 290 million persons with chronic HBV infection in the
world constitute the main reservoir of hepatitis B in human beings.
Whereas serum HBsAg is infrequent (0.1–0.5%) in normal popula­
tions in the United States and western Europe, the global prevalence is
estimated at 3.5%, and a prevalence of up to 5–10% has been found in
East Asia, sub-Saharan Africa, and tropical countries. The prevalence
can be even higher in certain high-risk groups, including persons with
Down syndrome, lepromatous leprosy, leukemia, Hodgkin disease,
polyarteritis nodosa, and chronic renal disease on hemodialysis, as well
as in injection drug users.
Other groups with high rates of HBV infection include spouses of
acutely infected persons; sexually promiscuous persons (especially
promiscuous men who have sex with men); health care workers and
first responders exposed to blood; persons who require repeated
transfusions especially with pooled blood-product concentrates (e.g.,
hemophiliacs); residents and staff of custodial institutions for the
developmentally handicapped; prisoners; and, to a lesser extent, family
members of chronically infected patients. Because of highly sensitive
virologic screening (antigen, antibody, and nucleic acid testing) of
donor blood, the risk of acquiring HBV infection from a blood transfu­
sion is estimated to be 1 in 360,000.
Prevalence of infection, modes of transmission, and human behav­
ior conspire to mold geographically different epidemiologic patterns
of HBV infection. In East Asia and Africa, hepatitis B, a disease of the
newborn and young children, is perpetuated by a cycle of maternalneonatal spread. In North America and western Europe, hepatitis B
is primarily a disease of adolescence and early adulthood, the time of
life when intimate sexual contact and recreational and occupational
percutaneous exposures tend to occur. To some degree, however, this
dichotomy between high-prevalence and low-prevalence geographic
regions has been minimized by immigration from high-prevalence
to low-prevalence areas. For example, in the United States, NHANES
data from 2007 to 2012 revealed an overall prevalence of current HBV
infection (detectable HBsAg) of 0.3%; however, the prevalence in Asian
persons, 93% of whom were foreign-born, was tenfold higher, 3.1%,
representing 50% of the U.S. national disease burden. As a result of
adoption of safe behaviors in high-risk groups as well as screening and
vaccination programs, the incidence of newly reported HBV infections
fell by >80% in the United States during the 1990s and has remained
low in the 21st century (with a low of 0.6 cases per 100,000 popula­
tion in 2021). Paralleling that trend, the imbalance between cases in
U.S.-born and foreign-born persons widened; currently, imported cases
in non-U.S.-born persons outnumber domestic cases by manyfold; in
NHANES 2017–2020, the prevalence of HBV infection was 1.0% in
foreign-born versus 0.2% in the entire cohort; non-U.S.-born persons
accounted for 73.6% of all infections. The introduction of hepatitis B
vaccine in the early 1980s and adoption of universal childhood vacci­
nation policies in many countries resulted in a dramatic, ~90% decline
in the incidence of new HBV infections in those countries as well as
in the dire consequences of chronic infection, including hepatocel­
lular carcinoma. In the United States, as demonstrated in NHANES
2007–2012, following the 1991 implementation of universal childhood
vaccination, HBsAg seropositivity had declined in children aged 6–19
years to as low as 0.03%, an ~85% reduction.
UNIVERSAL SCREENING FOR HEPATITIS B  Populations and groups at
high risk for hepatitis B are listed in Table 350-3. Screening for HBV
infection used to be recommended in these high-risk populations;
however, because risk-group screening had not been shown to be
effective in reducing the prevalence of HBV infection in the popula­
tion and the majority of acute hepatitis B cases were found to occur in
persons who were not in these high-risk groups, universal screening of
all adults (≥18 years) for hepatitis B was recommended in 2023 by the
CDC. Because current antiviral therapy has been highly effective in
reducing the clinical progression and consequences of HBV infection,
such universal screening was found to be cost-effective in reducing the
morbidity and mortality associated with chronic hepatitis B.
Hepatitis D 
Infection with HDV has a worldwide distribution,
but two epidemiologic patterns exist. In Mediterranean countries
(northern Africa, southern Europe, the Middle East), HDV infection
is endemic among those with hepatitis B, and the disease is trans­
mitted predominantly by nonpercutaneous means, especially close
personal contact. In nonendemic areas, such as the United States
TABLE 350-3  Populations with a High Risk for HBV Infectiona
Persons born in countries/regions with a high (≥8%) and intermediate (≥2%)
prevalence of HBV infection including immigrants and adopted children and
including persons born in the United States who were not vaccinated as infants
and whose parents emigrated from areas of high HBV endemicity
Household and sexual contacts of, or needle sharing with, persons who have
hepatitis B
Babies born to HBsAg-positive mothers
Persons who have used injection drugs
Persons with multiple sexual contacts or a history of sexually transmitted
disease
Men who have sex with men
Persons incarcerated in a correctional facility or other detention settings
Persons with elevated alanine or aspartate aminotransferase levels
Persons with HCV or HIV infection
Hemodialysis patients
Health care and laboratory workers and first responders exposed to blood
aScreening for hepatitis B had been recommended in the past for persons in these
high-risk groups; however, in 2023, the Centers for Disease Control and Prevention
recommended universal screening of all adults (>18 years of age) for hepatitis
B. Other groups who should be tested for hepatitis B include pregnant women;
persons who are the source of blood or body fluids that would be an indication for
postexposure prophylaxis (e.g., needlestick, mucosal exposure sexual assault); and
persons who require immunosuppressive or cytotoxic therapy (including anti–tumor
necrosis factor α therapy for rheumatologic or inflammatory bowel disorders).
(where hepatitis D is rare among persons with chronic hepatitis B)
and northern Europe, HDV infection is confined to persons exposed
frequently to blood and blood products, primarily injection drug
users (especially in HIV-infected injection drug users) and hemo­
philiacs. In the United States, the prevalence of HDV infection in the
national population was 0.02% in NHANES 1999–2012 and 0.11%
in NHANES 2011–2016; however, among HBsAg-positive persons,
the prevalence of HDV infection is highest in injection drug users
(11–36%) and hemophiliacs (19%). In one study of persons who
inject drugs (PWID) in San Francisco, the overall prevalence of HDV
infection was 1.1% but as high as 34.6% in PWID who had chronic
HBV infection. HDV infection can be introduced into a population
through drug users or by migration of persons from endemic to non­
endemic areas. Thus, patterns of population migration and human
behavior facilitating percutaneous contact play important roles in the
introduction and amplification of HDV infection. Occasionally, the
migrating epidemiology of hepatitis D is expressed in explosive out­
breaks of severe hepatitis, such as those that have occurred in remote
South American villages (e.g., “Lábrea fever” in the Amazon basin)
as well as in urban centers in the United States. Ultimately, such out­
breaks of hepatitis D—either of co-infections with acute hepatitis B
or of superinfections in those already infected with HBV—may blur
the distinctions between endemic and nonendemic areas. On a global
scale, HDV infection declined at the end of the 1990s. Even in Italy,
an HDV-endemic area, public health measures introduced to control
HBV infection (e.g., mass hepatitis B vaccination) resulted during
the 1990s in a 1.5%/year reduction in the prevalence of HDV infec­
tion. Still, the frequency of HDV infection during the first decade of
the twenty-first century has not fallen below levels reached during
the 1990s; the reservoir has been sustained by survivors infected
during 1970–1980 and recent immigrants from still-endemic (e.g.,
eastern Europe and Central Asia) to less-endemic countries. The
current global prevalence of HDV infection has been estimated at
12 million people, albeit with significant regional variation. Of the
eight HDV genotypes, genotype 1 is distributed worldwide, while the
others are more geographically confined (e.g., genotypes 2 and 4 in
the Far East, 3 in South America, and 5–8 in Africa).
CHAPTER 350
Acute Viral Hepatitis
Hepatitis C 
Routine screening of blood donors for HBsAg and
the elimination of commercial blood sources in the early 1970s
reduced the frequency of, but did not eliminate, transfusion-associ­
ated hepatitis. During the 1970s, the likelihood of acquiring hepatitis
after transfusion of voluntarily donated, HBsAg-screened blood was
~10% per patient (up to 0.9% per unit transfused); 90–95% of these
cases were classified, based on serologic exclusion of hepatitis A and
B, as “non-A, non-B” hepatitis. For patients requiring transfusion of
pooled products, such as clotting factor concentrates, the risk was
even higher, up to 20–30%.
During the 1980s, voluntary self-exclusion of blood donors with
risk factors for AIDS and then the introduction of donor screening
for anti-HIV reduced further the likelihood of transfusion-associated
hepatitis to <5%. During the late 1980s and early 1990s, the introduc­
tion first of “surrogate” screening tests for non-A, non-B hepatitis
(ALT and anti-HBc, both shown to identify blood donors with a
higher likelihood of transmitting non-A, non-B hepatitis to recipi­
ents) and, subsequently, after the discovery of HCV, progressively
more sensitive immunoassays for anti-HCV and then the application
of automated PCR testing of donated blood for HCV RNA reduced
the risk of transfusion-associated hepatitis C even further, to almost
imperceptible levels ranging between 1 in 2.3 million transfusions to
1 in 4.7 million transfusions.
In addition to being transmitted by transfusion, hepatitis C can be
transmitted by other percutaneous routes, such as injection drug use.
This virus can be transmitted by occupational exposure to blood,
and the likelihood of infection is increased in hemodialysis units.
Although the frequency of transfusion-associated hepatitis C fell as
a result of blood-donor screening, the overall frequency of reported
hepatitis C cases did not change until the 1990s, when the overall fre­
quency of reported cases fell by 80%, in parallel with a reduction in the
number of new cases in injection drug users, the source of most of the
HCV reservoir. After the exclusion of anti-HCV-positive plasma units
from the donor pool, rare, sporadic instances occurred of hepatitis C
among recipients of immunoglobulin preparations for intravenous (but
not intramuscular) use.
PART 10
Disorders of the Gastrointestinal System
Serologic evidence for HCV infection occurs in 90% of patients
with a history of transfusion-associated hepatitis (almost all occurring
before 1992, when second-generation HCV screening tests were intro­
duced); hemophiliacs and others treated with clotting factors; injection
drug users; 60–70% of patients with sporadic “non-A, non-B” hepatitis
who lack identifiable risk factors; 0.5% of volunteer blood donors;
and, in the NHANES survey conducted in the United States between
1999 and 2002, 1.6% of the general population in the United States,
which translated into 4.1 million persons (3.2 million with viremia),
the majority of whom were unaware of their infections. Moreover,
such population surveys do not include higher-risk groups such as
incarcerated persons, homeless persons, and active injection drug
users, indicating that the actual prevalence is even higher (estimated
to add an additional 1 million with anti-HCV antibody and 0.8 million
with HCV RNA in a later cohort assessed in 2003–2010). Comparable
frequencies of HCV infection occur in most countries around the
world, with 71 million persons infected worldwide, but extraordinarily
high prevalences of HCV infection occur in certain countries such as
Egypt, where >20% of the population (as high as 50% in persons born
prior to 1960) in some cities is infected. The high frequency in Egypt
is attributable to contaminated equipment used for medical procedures
and unsafe injection practices in the 1950s to 1980s (during a campaign
to eradicate schistosomiasis with intravenous tartar emetic). Thanks to
a 2018–2019 Egyptian government program to screen its entire adult
population (79% participation among >60 million people) for hepatitis
C and treat infected persons (2.2 million, 4.6% of those screened; of the
83% with a documented outcome, 99% were cured; the cost to identify
and cure a person was $130) with generic versions of direct-acting
antiviral (DAA) therapy (Chap. 352), hepatitis C has been nearly
eliminated there.
In the United States, African Americans and Mexican Americans
have higher frequencies of HCV infection than whites. Data from
NHANES showed that between 1988 and 1994, 30- to 40-year-old
men had the highest prevalence of HCV infection; however, in the
NHANES survey conducted between 1999 and 2002, the peak age
decile had shifted to those aged 40–49 years; an increase in hepatitis
C–related mortality has paralleled this secular trend, increasing since
1995 predominantly in the 45- to 65-year age group. Thus, despite an
80% reduction in new reported HCV infections during the 1990s, the
prevalence of HCV infection in the population was sustained by an
aging cohort that had acquired their infections three to four decades
earlier, during the 1960s and 1970s, as a result predominantly of selfinoculation with recreational drugs. Retrospective phylogenetic map­
ping of >45,000 HCV genotype 1a isolates revealed that the hepatitis
C epidemic emerged in the United States between 1940 and 1965,
peaking in 1950 and aligning temporally with the post–World War II
expansion of medical procedures (including reuse of glass syringes).
Thus, HCV was amplified iatrogenically not only in Egypt but also
in the United States; in the United States, the seeds sewn by medical
procedures in the 1950s were reaped in the 1960s and 1970s among
transfusion recipients and injection drug users, even those whose drug
use was confined to brief adolescent experimentation.
In NHANES 2003–2010, the prevalence of HCV infection (HCV RNA
reactivity) in the United States had actually fallen to 1% (2.7 million
persons) from 1.3% (3.2 million) the decade before (NHANES 1999–
2002), attributable to deaths among the HCV-infected population. In
NHANES data from 2013–2016, the prevalence of current HCV infec­
tion (HCV RNA reactivity) had fallen lower, to 0.9%. Worth emphasiz­
ing, however, is that NHANES datasets account for only the domiciled
and noninstitutionalized U.S. civilian population. In the same time
frame, the prevalence among incarcerated and homeless persons was
estimated to be 10.7%.
As deaths resulting from HIV infection fell after 1999, age-adjusted
mortality associated with HCV infection surpassed that of HIV infec­
tion in 2007; >70% of HCV-associated deaths occurred in the “baby
boomer” cohort born between 1945 and 1965. By 2012, HCV mortality
had surpassed deaths from HIV, tuberculosis, hepatitis B, and 57 other
notifiable infectious diseases (i.e., all infectious diseases) reported to
the CDC. In NHANES 1999–2002, compared to the 1.6% prevalence
of HCV infection in the population at large, the prevalence in the
1945–1965 birth cohort was 3.2%, representing three-quarters of all
infected persons. Therefore, in 2012, the CDC and, in 2013, the U.S.
Preventive Services Task Force (USPSTF) recommended that all per­
sons born between 1945 and 1965 be screened for hepatitis C, without
ascertainment of risk, a recommendation shown to be cost-effective
and predicted to identify 800,000 infected persons. Because of the
availability of highly effective antiviral therapy, such screening would
have the potential to avert 200,000 cases of cirrhosis and 47,000 cases
of hepatocellular carcinoma and to prevent 120,000 hepatitis-related
deaths; with the availability of the new generation of DAAs (efficacy
95%, see Chap. 352), screening baby boomers and treating those with
hepatitis C have been predicted to reduce the HCV-associated disease
burden by 50–70% through 2050.
Still, persons with chronic hepatitis C identified by 1945–1965 birthcohort screening are older than 50, and by the time they are identi­
fied, >20% already have advanced liver disease. In 2020, based on (1)
the 95–99% efficacy of all-oral, well-tolerated, highly effective DAAs;
(2) the demonstration that the endpoint of DAA therapy (sustained
virologic response) was associated with a marked decrease in liver and
all-cause mortality, cirrhosis, and hepatocellular carcinoma (Chap.
352); (3) a reduction in the initially high cost of DAA therapy; (4) the
demonstration of higher cost-effectiveness of screening all adults rather
than birth-cohort screening; and (5) the shifting demographics of HCV
infection (see below), especially since 2010, toward a younger popula­
tion exposed through injection drug use, the American Association for
the Study of Liver Diseases and the Infectious Diseases Society of America
as well as the USPSTF and CDC expanded recommended hepatitis C
screening to all adolescents and adults aged 18–79 (and because of the
substantial increase in HCV infections among women of child-bearing
age [age 20–39], expanded such screening to pregnant women).
Hepatitis C accounts for 40% of chronic liver disease and, before
the introduction of high-efficacy DAA therapy, was the most fre­
quent indication for liver transplantation; hepatitis C is estimated to
account for 8000–10,000 deaths per year in the United States. The
distribution of HCV genotypes varies in different parts of the world.
Worldwide, genotype 1 is the most common. In the United States,
genotype 1 accounts for 70% of HCV infections, whereas genotypes 2
and 3 account for the remaining 30%; among African Americans, the
frequency of genotype 1 is even higher (i.e., 90%). Genotype 4 pre­
dominates in Egypt; genotype 5 is localized to South Africa, genotype
6 to Hong Kong, and genotype 7 to Central Africa.
As a bloodborne infection, HCV potentially can be transmitted
sexually and perinatally; however, both modes of transmission are
inefficient for hepatitis C. Although 10–15% of patients with acute
hepatitis C report having potential sexual sources of infection, most
studies have failed to identify sexual transmission of this agent. The
chances of sexual and perinatal transmission have been estimated to
be ~5% but have shown in a prospective study to be only 1% between
monogamous sexual partners, well below comparable rates for HIV
and HBV infections. An important exception to the low rate of peri­
natal transmission is the high risk—typically in excess of 10%—in
babies born to HIV-HCV coinfected mothers. Moreover, sexual
transmission appears to be confined to such subgroups as persons
with multiple sexual partners and sexually transmitted diseases; for
example, isolated clusters of sexually transmitted HCV infection
have been reported in HIV-infected men who have sex with men.
Breast-feeding does not increase the risk of HCV infection between
an infected mother and her infant. Infection of health workers is not
dramatically higher than among the general population; however,
health workers are more likely to acquire HCV infection through
accidental needle punctures, the efficiency of which is ~3%. Infec­
tion of household contacts is rare as well. Most asymptomatic blood
donors found to have anti-HCV and ~20–30% of persons with
reported cases of acute hepatitis C do not fall into a recognized risk
group; however, many such blood donors do recall long forgotten,
risk-associated behaviors when questioned carefully.
Besides persons born between 1945 and 1965, other groups with
an increased frequency of HCV infection are listed in Table 350-4. In
immunosuppressed individuals, levels of anti-HCV may be undetect­
able, and a diagnosis may require testing for HCV RNA. Although
new acute cases of hepatitis C are rare outside of the injection
drug–using community, newly diagnosed cases are common among
otherwise healthy persons who experimented briefly with injection
drugs, as noted above, four or five decades earlier. Such instances
usually remain unrecognized for years, until unearthed by laboratory
screening for routine medical examinations, insurance applications,
and attempted blood donation. Although, overall, the annual inci­
dence of new HCV infections has continued to fall, the rate of new
infections has been increasing since 2002, has accelerated since 2010
(tripling from 0.3/100,000 to 1.2/100,000 between 2009 and 2018),
and has been amplified by the recent epidemic of opioid use in a new
cohort of young injection drug users aged 20–39 years (accounting
for a 3.8-fold increase in cases between 2010 and 2017 and for more
than two-thirds of all acute cases), who, unlike older cohorts, had not
TABLE 350-4  Populations with a High Risk for HCV Infectiona
All adults aged 18–79 should be screened, a recommendation that supplants the
earlier focus on persons born between 1945 and 1965a
Persons who have ever used injection drugs
Persons with HIV infection
Hemophiliacs treated with clotting factor concentrates prior to 1987
Persons who have ever undergone long-term hemodialysis
Persons with unexplained elevations of aminotransferase levels
Transfusion or transplantation recipients prior to July 1992
Recipients of blood or organs from a donor found to be positive for hepatitis C
Children born to women with hepatitis C
Health care, public safety, and emergency medical personnel following needle
injury or mucosal exposure to HCV-contaminated blood
Sexual partners of persons with hepatitis C infection
Pregnant women 
aScreening for hepatitis C had been recommended in the past for persons in these
high-risk groups; however, in 2020, the Centers for Disease Control and Prevention
recommended universal screening of all adults (>18 years of age) for hepatitis C.
learned to take precautions to prevent bloodborne infections. Reflect­
ing this emerging development, the prevalence of current HCV infec­
tion (HCV RNA reactivity) in the United States rose from 0.65% (1.7
million persons) in a 2010–2014 NHANES analysis to 0.9% (2.04
million persons) in a 2013–2016 NHANES analysis. Moreover, based
on an estimate of populations excluded from this NHANES analysis,
the prevalence would be even higher, 0.93% (2.27 million persons).
This late temporal trend was attributed to the increase of acute cases
in injections drug users, driven by increases in states most affected
by the opioid/injection drug use epidemic. Also, in parallel with this
trend, the prevalence of HCV infection in women aged 15–44 years
(of child-bearing age) doubled between 2016 and 2014; accordingly,
screening of pregnant women for HCV infection is now recom­
mended as well.
Hepatitis E 
This type of hepatitis, identified in India, Asia, Africa,
the Middle East, and Central America (endemic areas), resembles
hepatitis A in its primarily enteric mode of spread. The commonly
recognized cases occur after contamination of water supplies such
as after monsoon flooding, but sporadic, isolated cases occur. An
epidemiologic feature that distinguishes HEV from other enteric
agents is the rarity of secondary person-to-person spread from
infected persons to their close contacts. Large waterborne outbreaks
in endemic areas are linked to genotypes 1 and 2, arise in popula­
tions that are immune to HAV, favor young adults, and account for
antibody prevalences of 30–80%. The worldwide annual incidence
of acute HEV infections has been estimated conservatively to be at
least 20 million (of which 3.3 million are symptomatic), rendering
HEV infection as the most common cause of acute viral hepatitis. In
nonendemic areas of the world, such as the United States, clinically
apparent acute hepatitis E is extremely rare; however, during the
1988–1994 NHANES III survey conducted by the U.S. Public Health
Service, the prevalence of anti-HEV was 21%, reflecting subclinical
infections, infection with genotypes 3 and 4, predominantly in older
males (>60 years). A repeat NHANES study in 2009–2010, however,
showed a substantial 70% two-decade reduction in anti-HEV to
only 6%, more consistent with the rarity of acute hepatitis E in the
United States than the previous NHANES result would suggest and
perhaps a reflection of a more specific anti-HEV assay used in the
second time period. A subsequent examination of the 2009–2016
NHANES dataset demonstrated a seroprevalence of HEV infection
of 6.1%. Moreover, supporting the concern for influence of assay dif­
ferences on population estimates, a re-interrogation of the NHANES
III cohort based on assays of the highest sensitivity identified varying
prevalence estimates ranging from 16–21%. Again, in the 2009–2016
cohort, older age was associated with anti-HEV seropositivity, as were
non-Hispanic Asian ethnicity and birth outside the United States.
CHAPTER 350
Acute Viral Hepatitis
In nonendemic areas, HEV accounts for only a small propor­
tion of cases of sporadic (labeled “autochthonous” or indigenous)
hepatitis; however, cases imported from endemic areas have been
found in the United States. Evidence supports a zoonotic reservoir
for HEV primarily in swine (but also in deer, camels, and rabbits),
which may account for the mostly subclinical infections primarily
of genotypes 3 and 4 in nonendemic areas. Globally, ~60% of swine
have evidence of prior HEV infection, and 13% have active HEV
infection. A previously unrecognized high distribution of HEV infec­
tion, linked to uncooked or undercooked pork-product ingestion, has
been discovered in western Europe (e.g., in Germany, an estimated
annual incidence of 300,000 cases and a 17% prevalence of anti-HEV
among adults; in France, a 22% prevalence of anti-HEV in healthy
blood donors).
■
■CLINICAL AND LABORATORY FEATURES
Symptoms and Signs 
Acute viral hepatitis occurs after an incuba­
tion period that varies according to the responsible agent. Generally,
incubation periods for hepatitis A range from 15 to 45 days (mean,
4 weeks), for hepatitis B and D from 30 to 180 days (mean, 8–12 weeks),
for hepatitis C from 15 to 160 days (mean, 7 weeks), and for hepatitis
E from 14 to 60 days (mean, 5–6 weeks). The prodromal symptoms of
acute viral hepatitis are systemic and quite variable. Constitutional
symptoms—which may include anorexia, nausea and vomiting, diar­
rhea, fatigue, malaise, arthralgias, myalgias, headache, and photo­
phobia—tend to precede the onset of jaundice (if jaundice occurs).
The nausea, vomiting, and anorexia are frequently associated with
alterations in olfaction and taste. A low-grade fever between 38° and
39°C (100°–102°F) is more often present in hepatitis A and E than in
hepatitis B or C, except when hepatitis B is heralded by a serum
sickness–like syndrome; rarely, a fever of 39.5°–40°C (103°–104°F)
may accompany the constitutional symptoms. Dark urine and claycolored stools may be noticed by the patient from 1–5 days before the
onset of clinical jaundice.
Many patients will not become jaundiced (anicteric hepatitis), and
the likelihood of jaundice varies by virus (e.g., substantially more com­
mon in acute hepatitis B than in acute hepatitis C). With the onset
of clinical jaundice, the constitutional prodromal symptoms usually
diminish, but in some patients, mild weight loss (2.5–5 kg) is common
and may continue during the entire icteric phase. The liver becomes
enlarged and tender and may be associated with right upper quadrant
pain and discomfort. Infrequently, patients present with a cholestatic
picture, suggesting extrahepatic biliary obstruction. Splenomegaly
and cervical adenopathy are present in 10–20% of patients with acute
hepatitis. Rarely, a few spider angiomas appear during the icteric phase
and disappear during convalescence. During the recovery phase, consti­
tutional symptoms disappear, but usually some liver enlargement and
abnormalities in liver biochemical tests are still evident. The duration
of the posticteric phase is variable, ranging from 2 to 12 weeks, and is
usually more prolonged in acute hepatitis B and C. Complete clinical
and biochemical recovery is to be expected 1–2 months after all cases
of hepatitis A and E and 3–4 months after the onset of jaundice in
three-quarters of uncomplicated, self-limited cases of hepatitis B and
C (among healthy adults, acute hepatitis B is self-limited in 95–99%,
whereas hepatitis C is self-limited in only ~15–20%). In the remainder,
biochemical recovery may be delayed.
PART 10
Disorders of the Gastrointestinal System
Infection with HDV can occur in the presence of acute (coinfec­
tion) or chronic (superinfection) HBV infection; the duration of HBV
infection determines the duration of HDV infection. When acute
HDV and HBV infections occur simultaneously (coinfection), clini­
cal and biochemical features may be indistinguishable from those of
HBV infection alone, and the clinical severity can range from mild
to fulminant disease. When acute HDV infection occurs in a patient
with preexisting chronic HBV infection (superinfection), the clinical
presentation is more commonly severe, and the HDV superinfection
mirrors the chronic HBV infection in becoming persistent. In such
cases, the HDV superinfection appears as a clinical exacerbation or an
episode resembling acute viral hepatitis in someone already chronically
infected with HBV. Superinfection with HDV in a patient with chronic
hepatitis B often leads to clinical deterioration (see below).
In addition to superinfections with other hepatitis agents, acute
hepatitis-like clinical events in persons with chronic hepatitis B may
accompany spontaneous HBeAg to anti-HBe seroconversion or spon­
taneous reactivation (i.e., reversion from relatively nonreplicative to
replicative infection). Occasionally, acute clinical exacerbations of
chronic hepatitis B may represent the emergence of a precore mutant
(see “Virology and Etiology”), and the subsequent course in such
patients may be characterized by periodic exacerbations. Reactivations
can occur as well in therapeutically immunosuppressed patients with
chronic HBV infection when cytotoxic/immunosuppressive drugs
are withdrawn; in these cases, restoration of immune competence
is thought to allow resumption of previously checked cell-mediated
immune cytolysis of HBV-infected hepatocytes. Cancer-directed thera­
pies can also result in HBV or HCV reactivation, with the greatest risk
from the B-cell (CD20)-depleting antibody rituximab in HBV. Immune
checkpoint inhibitors (ICIs) have been an important new class of anti­
tumor therapies, and the impact of ICI therapy on hepatitis caused by
chronic HBV or HCV infection is not yet well characterized, but acute
exacerbations of chronic viral hepatitis have been reported and may
be related to restoring function of “exhausted” T-cell populations. Of
interest, however, in pilot studies of ICIs given for chronic hepatitis B,
restoration of T-cell function has also been associated with functional
cure in a limited subset.
Laboratory Features 
The serum aminotransferases aspartate
aminotransferase (AST) and ALT (previously designated SGOT and
SGPT) increase to a variable degree during the prodromal phase of
acute viral hepatitis and precede the rise in bilirubin level (Figs. 350-2
and 350-4). The level of these enzymes, however, does not correlate
well with the degree of liver cell damage. Peak levels vary from ~400
to ~4000 IU or more; these levels are usually reached at the time the
patient is clinically icteric and diminish progressively during the recov­
ery phase of acute hepatitis. The diagnosis of anicteric hepatitis is based
on clinical features and on aminotransferase elevations.
Jaundice is usually visible in the sclera or skin when the serum
bilirubin value is >43 μmol/L (2.5 mg/dL). When jaundice appears, the
serum bilirubin typically rises to levels ranging from 85 to 340 μmol/L
(5–20 mg/dL). The serum bilirubin may continue to rise despite falling
serum aminotransferase levels. In most instances, the total bilirubin is
equally divided between the conjugated and unconjugated fractions.
Bilirubin levels >340 μmol/L (20 mg/dL) extending and persisting late
into the course of viral hepatitis are more likely to be associated with
severe disease. In certain patients with underlying hemolytic anemia,
however, such as glucose-6-phosphate dehydrogenase deficiency and
sickle cell anemia, a high serum bilirubin level is common, resulting from
superimposed hemolysis. In such patients, bilirubin levels >513 μmol/L
(30 mg/dL) have been observed and are not necessarily associated with
a poor prognosis.
Neutropenia and lymphopenia are transient and are followed by a
relative lymphocytosis. Atypical lymphocytes (varying between 2 and
20%) are common during the acute phase. Measurement of the pro­
thrombin time (PT) is important in patients with acute viral hepatitis,
because a prolonged value may reflect a severe hepatic synthetic defect,
signify extensive hepatocellular necrosis, and indicate a worse progno­
sis. Occasionally, a prolonged PT may occur with only mild increases in
the serum bilirubin and aminotransferase levels. Prolonged nausea and
vomiting, inadequate carbohydrate intake, and poor hepatic glycogen
reserves may contribute to hypoglycemia noted occasionally in patients
with severe viral hepatitis. Serum alkaline phosphatase may be normal
or only mildly elevated, whereas a fall in serum albumin is uncom­
mon in uncomplicated acute viral hepatitis. In some patients, mild
and transient steatorrhea has been noted, as well as slight microscopic
hematuria and minimal proteinuria.
A diffuse but mild elevation of the γ globulin fraction is common
during acute viral hepatitis. Serum IgG and IgM levels are elevated in
about one-third of patients during the acute phase of viral hepatitis,
but the serum IgM level is elevated more characteristically during
acute hepatitis A. During the acute phase of viral hepatitis, antibodies
to smooth muscle and other cell constituents may be present, and low
titers of rheumatoid factor, nuclear antibody, and heterophile antibody
can also be found occasionally. In hepatitis C and D, antibodies to LKM
may occur; however, the species of LKM antibodies in the two types of
hepatitis are different from each other as well as from the LKM anti­
body species characteristic of autoimmune hepatitis type 2 (Chap.
352). The autoantibodies in viral hepatitis are nonspecific and can
also be associated with other viral and systemic diseases. In contrast,
virus-specific antibodies, which appear during and after hepatitis virus
infection, are serologic markers of diagnostic importance.
As described above, serologic tests are available routinely with
which to establish a diagnosis of hepatitis A, B, D, and C. Tests for fecal
or serum HAV are not routinely available. Therefore, a diagnosis of
hepatitis A is based on detection of IgM anti-HAV during acute illness
(Fig. 350-2). Worth considering, false-positive IgM anti-HAV results
have been observed in rheumatologic diseases (especially those with
rheumatoid factor, which can mediate nonspecific binding in solidphase immunoassays) and, very rarely, in other acute viral infections,
such as with Epstein-Barr virus (EBV).
A diagnosis of HBV infection can usually be made by detection
of HBsAg in serum. Infrequently, levels of HBsAg are too low to be
detected during acute HBV infection, even with contemporary, highly
sensitive immunoassays. In such cases, the diagnosis can be established
by the presence of IgM anti-HBc.
The titer of HBsAg bears little relation to the severity of clinical
disease. Indeed, an inverse correlation exists between the serum con­
centration of HBsAg and the degree of liver cell damage. For example,
titers are highest in immunosuppressed patients, lower in patients with
chronic liver disease (but higher in mild chronic than in severe chronic
hepatitis), and very low in patients with acute fulminant hepatitis.
These observations suggest that in hepatitis B the degree of liver cell
damage and the clinical course are related to variations in the patient’s
immune response to HBV rather than to the amount of circulating
HBsAg. In immunocompetent persons, however, a correlation exists
between markers of HBV replication and liver injury (see below).
Another important serologic marker in patients with hepatitis B is
HBeAg. Its principal clinical usefulness is as an indicator of relative
infectivity. Because HBeAg is invariably present during early acute
hepatitis B, HBeAg testing is indicated primarily in chronic infection.
In patients with hepatitis B surface antigenemia of unknown dura­
tion (e.g., blood donors found to be HBsAg positive), testing for IgM
anti-HBc may be useful to distinguish between acute or recent infec­
tion (IgM anti-HBc positive) and chronic HBV infection (IgM antiHBc negative, IgG anti-HBc positive). A false-positive test for IgM
anti-HBc may be encountered in patients with high-titer rheumatoid
factor. Also, IgM anti-HBc may be reexpressed during acute reactiva­
tion of chronic hepatitis B.
Anti-HBs is rarely detectable in the presence of HBsAg in patients
with acute hepatitis B, but up to 10% of persons with chronic HBV
infection may harbor anti-HBs. Variability of the a determinant of
the S protein is thought to be central to this anti-HBs. The antibody is
directed not against the common group determinant, a, but against the
relatively closely related heterotypic subtype determinant (e.g., HBsAg
of subtype ad with anti-HBs of subtype y). Clinically, patients with both
HBsAg and anti-HBs tend to have higher aminotransferase levels, are
older, and are more likely to have advanced fibrosis or cirrhosis (pri­
marily in HBeAg-negative patients). This serologic pattern in patients
with chronic hepatitis B is not a harbinger of imminent HBsAg clear­
ance; however, in patients clearing HBsAg as acute hepatitis B resolves,
infrequently, transiently, both HBsAg and anti-HBs may be detected
simultaneously.
After immunization with hepatitis B vaccine, which consists of
HBsAg alone, anti-HBs is the only serologic marker to appear. The
commonly encountered serologic patterns of hepatitis B and their
interpretations are summarized in Table 350-5. Tests for the detec­
tion of HBV DNA in liver and serum are now available. Like HBeAg,
serum HBV DNA is an indicator of HBV replication, but tests for HBV
TABLE 350-5  Commonly Encountered Serologic Patterns of Hepatitis B Infection
HBsAg
ANTI-HBs
ANTI-HBc
HBeAg
ANTI-HBe
INTERPRETATION
+
−
IgM
+
−
Acute hepatitis B, high infectivitya
+
−
IgG
+
−
Chronic hepatitis B, high infectivity
+
−
IgG
−
+
1.  Late acute or chronic hepatitis B, low infectivity
2.  HBeAg-negative (“precore-mutant”) hepatitis B (chronic or, rarely,
+
+
+
+/−
+/−
1.  HBsAg of one subtype and heterotypic anti-HBs (common)
2.  Process of seroconversion from HBsAg to anti-HBs (rare)
−
−
IgM
+/−
+/−
1.  Acute hepatitis Ba
−
−
IgG
−
+/−
1.  Low-level hepatitis B carrier
2.  Hepatitis B in remote past
−
+
IgG
−
+/−
Recovery from hepatitis B
−
+
−
−
−
1.  Immunization with HBsAg (after vaccination)
2.  Hepatitis B in the remote past (?)
3.  False-positive
aIgM anti-HBc may reappear during acute reactivation of chronic hepatitis B.
Note: See text for abbreviations.
DNA are more sensitive and quantitative. First-generation hybridiza­
tion assays for HBV DNA had a sensitivity of 105−106 virions/mL, a
relative threshold below which infectivity and liver injury are limited
and HBeAg is usually undetectable. Currently, testing for HBV DNA
has shifted from insensitive hybridization assays to amplification assays
(e.g., the PCR-based assay, which can detect as few as 10 or 100 virions/
mL); among the commercially available PCR assays, the most useful are
those with the highest sensitivity (5–10 IU/mL) and the largest dynamic
range (100–109 IU/mL). With increased sensitivity, amplification assays
remain reactive well below the current 103–104 IU/mL threshold for
infectivity and liver injury. These markers are useful in following the
course of HBV replication in patients with chronic hepatitis B receiving
antiviral chemotherapy (Chap. 352). Except for the early decades of
life after perinatally acquired HBV infection (see above), in immuno­
competent adults with chronic hepatitis B, a general correlation exists
between the level of HBV replication, as reflected by the level of serum
HBV DNA, and the degree of liver injury. High-serum HBV DNA
levels, increased expression of viral antigens, and necroinflammatory
activity in the liver go hand in hand unless immunosuppression inter­
feres with cytolytic T-cell responses to virus-infected cells; reduction
of HBV replication with antiviral drugs tends to be accompanied
by an improvement in liver histology. Among patients with chronic
hepatitis B, high levels of HBV DNA increase the risk of cirrhosis,
hepatic decompensation, and hepatocellular carcinoma (see “Compli­
cations and Sequelae”).
A diagnosis of acute hepatitis C hinges upon the presence of antiHCV and/or HCV RNA with or without the presence of anti-HCV.
Compared with the other viral hepatitides, assays specific for IgM
anti-HCV, never found to be of clinical value, are lacking; available
highly sensitive assays detect IgG anti-HCV. While, in the past, with
early-generation immunoassays, up to 12 weeks could elapse before
anti-HCV became detectable during acute hepatitis C, with contempo­
rary, late-generation assays, anti-HCV is readily detectable during early
acute hepatitis C. Thus, when contemporary immunoassays are used,
anti-HCV can be detected in acute hepatitis C during the initial phase
of elevated aminotransferase activity and remains detectable both
after recovery from acute infection as well as during chronic infection.
As alluded to above, nonspecificity can confound immunoassays for
anti-HCV, especially in persons with a low prior probability of infec­
tion, such as volunteer blood donors, or in persons with circulating
rheumatoid factor, which can bind nonspecifically to assay reagents;
testing for HCV RNA can be used in such settings to distinguish
between true-positive and false-positive anti-HCV determinations.
While spontaneous recovery from acute HCV infection occurs no
more often than 15–20% of the time, persons who do recover will have
CHAPTER 350
Acute Viral Hepatitis
acute)
2.  Anti-HBc “window”
a persistent anti-HCV but are not protected against reinfection. In
high-risk groups, notably PWID, a pattern of positive anti-HCV and
HCV RNA may reflect acute reinfection rather than chronic infection.
In this setting, prior laboratory tests and clinical history are critical in
differentiating between chronic and repeat acute infections.
Assays for HCV RNA are the most sensitive tests for HCV infec­
tion and represent the “gold standard” in establishing a diagnosis of
hepatitis C. HCV RNA can be detected even before acute elevation of
aminotransferase activity and before the appearance of anti-HCV in
patients with acute hepatitis C. In addition, HCV RNA remains detect­
able indefinitely, although at varying levels, in patients with chronic
hepatitis C. In the very small minority of patients with hepatitis C
who lack anti-HCV, a diagnosis can be supported by detection of HCV
RNA. In the rare instance when all these tests are negative and the
patient has a well-characterized case of hepatitis after percutaneous
exposure to blood or blood products, a diagnosis of hepatitis caused by
an unidentified agent can be entertained.
Amplification techniques are required to detect HCV RNA. Cur­
rently, such target amplification (i.e., synthesis of multiple copies
of the viral genome) is achieved by PCR, in which the viral RNA is
reverse transcribed to complementary DNA and then amplified by
repeated cycles of DNA synthesis. Quantitative PCR assays provide
a measurement of relative “viral load”; current PCR assays have a
sensitivity of 10 (lower limit of detection) to 25 (lower limit of quan­
titation) IU/mL and a wide dynamic range (10–107 IU/mL). Determi­
nation of HCV RNA level is not a reliable marker of disease severity
or prognosis but is helpful in predicting relative responsiveness to
antiviral therapy. The same is true for determinations of HCV geno­
type (Chap. 352). Of course, HCV RNA monitoring during and after
antiviral therapy is the sine qua non for determining on-treatment
and durable responsiveness.
PART 10
Disorders of the Gastrointestinal System
A proportion of patients with hepatitis C have isolated anti-HBc
in their blood, a reflection of a common risk in certain populations
of exposure to multiple bloodborne hepatitis agents. The anti-HBc in
such cases is almost invariably of the IgG class and usually represents
HBV infection in the remote past (HBV DNA undetectable); it rarely
represents current HBV infection with low-level virus carriage. Detect­
able anti-HCV in the absence of HCV RNA signifies spontaneous or
therapeutically induced recovery from (“cured”) hepatitis C.
The presence of HDV infection can be identified by demonstrating
intrahepatic HDV antigen or, more practically, an anti-HDV (IgM or
IgG) seroconversion (a rise in titer of anti-HDV or de novo appearance
of anti-HDV). Serum HDV RNA or antigen can be detected early dur­
ing acute coinfection but is often transient and detectable only briefly,
if at all. Because anti-HDV is often undetectable once HBsAg disap­
pears, retrospective serodiagnosis of acute self-limited, simultaneous
HBV and HDV infection is difficult. Early diagnosis of acute infection
may be hampered by a delay of up to 30–40 days in the appearance of
IgG anti-HDV.
When a patient presents with acute hepatitis and has HBsAg and
anti-HDV in serum, determination of the class of anti-HBc is helpful
in establishing the relationship between infection with HBV and HDV.
TABLE 350-6  Simplified Diagnostic Approach in Patients Presenting with Acute Hepatitis
SEROLOGIC TESTS OF PATIENT’S SERUM
DIAGNOSTIC INTERPRETATION
HBsAg
IgM ANTI-HAV
IgM ANTI-HBc
ANTI-HCV HCV RNA
+
−
+
−
Acute hepatitis B
+
−
−
−
Chronic hepatitis B
+
+
−
−
Acute hepatitis A superimposed on chronic hepatitis B
+
+
+
−
Acute hepatitis A and B
−
+
−
−
Acute hepatitis A
−
+
+
−
Acute hepatitis A and B (HBsAg below detection threshold)
−
−
+
−
Acute hepatitis B (HBsAg below detection threshold)
−
−
−
Either +
Acute hepatitis C (vs. chronic HCV)
Note: See text for abbreviations.
Although IgM anti-HBc does not distinguish absolutely between acute
and chronic HBV infection, its presence is a reliable indicator of recent
infection and its absence a reliable indicator of infection in the remote
past. In simultaneous acute HBV and HDV infections, IgM anti-HBc
will be detectable, whereas in acute HDV infection superimposed on
chronic HBV infection, anti-HBc will be of the IgG class. Assays for
HDV RNA, available in specialized laboratories and yet to be standard­
ized, can be used to confirm HDV infection and to monitor treatment
during chronic infection.
The serologic/virologic course of events during acute hepatitis E is
entirely analogous to that of acute hepatitis A, with brief fecal shedding
of virus and viremia and an early IgM anti-HEV response that predom­
inates during approximately the first 3 months but is eclipsed thereafter
by long-lasting IgG anti-HEV. Diagnostic tests of varying reliability for
hepatitis E are commercially available outside the United States; in the
United States, although tests for HEV infection are not approved by the
FDA, reliable diagnostic serologic/virologic assays can be performed at
the CDC or other commercial or academic laboratories.
Liver biopsy is rarely necessary or indicated in acute viral hepatitis,
except when the diagnosis is questionable or when clinical evidence
suggests an alternate diagnosis of chronic hepatitis (e.g., autoimmune
hepatitis).
A diagnostic algorithm can be applied in the evaluation of cases of
acute viral hepatitis. A patient with acute hepatitis should undergo five
serologic tests: HBsAg, IgM anti-HAV, IgM anti-HBc, anti-HCV, and
HCV RNA (Table 350-6). The presence of HBsAg, with or without IgM
anti-HBc, represents HBV infection. If IgM anti-HBc is present, the
HBV infection is considered acute (or, less commonly, an acute reac­
tivation); if IgM anti-HBc is absent, the HBV infection is considered
chronic. A diagnosis of acute hepatitis B can be made in the absence of
HBsAg when IgM anti-HBc is detectable. A diagnosis of acute hepatitis
A is based on the presence of IgM anti-HAV. If IgM anti-HAV coexists
with HBsAg, a diagnosis of simultaneous HAV and HBV infections can
be made; if IgM anti-HBc (with or without HBsAg) is detectable, the
patient has simultaneous acute hepatitis A and B, and if IgM anti-HBc
is undetectable, the patient has acute hepatitis A superimposed on
chronic HBV infection. In a patient with an acute hepatitis illness, the
presence of anti-HCV supports a diagnosis of acute hepatitis C. In the
past, sensitivity of anti-HCV assays was limited during acute hepatitis,
leading to a delay in the appearance anti-HCV; however, contemporary
assays detect anti-HCV reliably during acute hepatitis. If acute hepatitis
C is suspected but anti-HCV is undetectable, testing for HCV RNA
helps to establish the diagnosis. The serologic diagnosis of acute hepa­
titis C or B may be challenging, for example, when an acute hepatitislike illness occurs in someone already infected chronically with HBV
or HCV. Still, when encountering a person with acute hepatitis, a series
of standard serologic assays for hepatitis A, B, and C is a reliable first
step. If these assays are all negative, testing for hepatitis E, especially
in the presence of risk factors, is available from specialty laboratories
or the CDC. Acute hepatitis D belongs in the differential diagnosis of
acute hepatitis, but negative serologic testing for hepatitis B excludes
hepatitis D infection. In the setting of serologically documented acute
hepatitis B, patients who are at high risk for hepatitis D are candidates
to be tested for anti-HDV and/or HDV RNA. Absence of all serologic
markers is consistent with a diagnosis of “non-A-E,” hepatitis (no other
proven human hepatitis viruses have been identified). As noted above,
an outbreak of severe acute liver injury and liver failure in children was
found to be associated with adenovirus infection, and elevations in
aminotransferase levels are observed as well in a variety of other acute
viral illnesses.
In patients with chronic hepatitis, initial testing should consist of
HBsAg and anti-HCV. Anti-HCV supports and HCV RNA testing
establishes the diagnosis of chronic hepatitis C. If a serologic diagno­
sis of chronic hepatitis B is made, testing for HBeAg and anti-HBe is
indicated to evaluate relative infectivity. Testing for HBV DNA in such
patients provides a more quantitative and sensitive measure of the
level of virus replication and therefore is very helpful during antiviral
therapy (Chap. 352). In patients with established chronic hepatitis B
and normal aminotransferase activity in the absence of HBeAg, serial
testing over time is often required to distinguish between the inactive
carrier state and HBeAg-negative chronic hepatitis B with fluctuating
virologic and necroinflammatory activity. In persons with hepatitis
B, testing for anti-HDV is useful in those with severe and fulminant
disease, with severe chronic disease, with chronic hepatitis B and acute
hepatitis-like exacerbations, with frequent percutaneous exposures (or
known HBV acquisition from percutaneous exposure), and from areas
where HDV infection is endemic.
■
■PROGNOSIS
Virtually all previously healthy patients with hepatitis A recover
completely with no clinical sequelae. Similarly, in acute hepatitis
B, 95–99% of previously healthy adults have a favorable course and
recover completely. Certain clinical and laboratory features, how­
ever, suggest a more complicated and protracted course. Patients of
advanced age and with serious underlying medical disorders may
have a prolonged course and are more likely to experience severe
hepatitis. Initial presenting features such as ascites, peripheral edema,
and symptoms of hepatic encephalopathy suggest a poorer prognosis.
In addition, a prolonged PT, low serum albumin level, hypoglycemia,
and very high serum bilirubin values suggest severe hepatocellular
disease. Patients with these clinical and laboratory features deserve
prompt hospital admission. The case-fatality rate in hepatitis A and B
is very low (~0.1%) but is increased by advanced age and underlying
debilitating disorders. Among patients ill enough to be hospitalized
for acute hepatitis B, the fatality rate is 1%. Hepatitis C is less severe
during the acute phase than hepatitis B and is more likely to be anic­
teric; fatalities are rare, but the precise case-fatality rate is not known.
Interestingly, while spontaneous resolution of HCV is the exception
rather than the rule, symptomatic infection (and icteric hepatitis in
particular) is associated with a higher likelihood of viral contain­
ment. In outbreaks of waterborne hepatitis E (genotypes 1 and 2) in
India and Asia, the case-fatality rate is 1–2% and up to 10–20% in
pregnant women. Contributing to fulminant hepatitis E in endemic
countries (but only very rarely or not at all in nonendemic coun­
tries) are instances of acute hepatitis E superimposed on underlying
chronic liver disease (“acute-on-chronic” liver disease). Patients with
simultaneous acute hepatitis B and D coinfection do not necessarily
experience a higher mortality rate than do patients with acute hepati­
tis B alone; however, in several outbreaks of acute simultaneous HBV
and HDV infection among injection drug users, the case-fatality
rate was ~5%. When HDV superinfection occurs in a person with
chronic hepatitis B, the likelihood of fulminant hepatitis and death is
increased substantially. Although the case-fatality rate for hepatitis D
is not known definitively, in outbreaks of severe HDV superinfection
in isolated populations with a high hepatitis B carrier rate (“Lábrea
fever”), a mortality rate >20% has been recorded.
■
■COMPLICATIONS AND SEQUELAE
A small proportion of patients with hepatitis A experience relaps­
ing hepatitis weeks to months after apparent recovery from acute
hepatitis. Relapses are characterized by recurrence of symptoms,
aminotransferase elevations, occasional jaundice, and fecal excretion
of HAV. Another unusual variant of acute hepatitis A is cholestatic
hepatitis, characterized by protracted cholestatic jaundice and pruritus.
Rarely, liver test abnormalities persist for many months, even up to 1
year. Even when these complications occur, hepatitis A remains selflimited and does not progress to chronic liver disease.
During the prodromal phase of acute hepatitis B, a serum sick­
ness–like syndrome characterized by arthralgia or arthritis, rash,
angioedema, and, rarely, hematuria and proteinuria may develop in
5–10% of patients. This syndrome occurs before the onset of clini­
cal jaundice, and these patients are often diagnosed erroneously as
having rheumatologic diseases. The diagnosis can be established by
measuring serum aminotransferase levels, which are almost invariably
elevated, and serum HBsAg. In patients with chronic hepatitis B or C,
immune complex and other extrahepatic manifestations may occur,
including metabolic disorders (including insulin resistance and other
manifestations of the metabolic syndrome), dermatologic disorders
(e.g., porphyria cutanea tarda and lichen planus in chronic hepatitis
C), lymphoproliferative and rheumatologic disorders, and, in addition
to hepatocellular carcinoma, nonliver malignancies (Chap. 352). In
addition, population studies of patients with chronic hepatitis C have
shown a convincing increase in cardiovascular and cerebrovascular
disease, renal disease, and mental health and cognitive disorders.
The most feared complication of acute viral hepatitis is fulminant
hepatitis (massive hepatic necrosis); fortunately, this is a rare event.
Fulminant hepatitis is seen primarily in hepatitis B, D, and E, but rare
fulminant cases of hepatitis A occur primarily in older adults and in
persons with underlying chronic liver disease, including, according
to some reports, chronic hepatitis B and C. Hepatitis B accounts for
50% of fulminant cases of viral hepatitis, a sizable proportion of
which are associated with HDV infection and another proportion with
underlying chronic hepatitis C. Fulminant hepatitis is hardly ever seen
in hepatitis C, but hepatitis E, as noted above, can be complicated by
fatal fulminant hepatitis in 1–2% of all cases and in up to 20% of cases
in pregnant women. Patients usually present with signs and symptoms
of encephalopathy that may evolve to deep coma. The liver is usually
small and the PT excessively prolonged. The combination of rapidly
shrinking liver size, rapidly rising bilirubin level, and marked prolon­
gation of the PT, even as aminotransferase levels fall, together with
clinical signs of confusion, disorientation, somnolence, ascites, and
edema, indicates that the patient has hepatic failure with encephalopa­
thy. Cerebral edema is common; brainstem compression, gastrointes­
tinal bleeding, sepsis, respiratory failure, cardiovascular collapse, and
renal failure are terminal events. The mortality rate can be exceedingly
high (>80% in patients with deep coma) without liver transplantation,
but patients who survive may have a complete biochemical and histo­
logic recovery (Chap. 356). Documenting the disappearance of HBsAg
after apparent clinical recovery from acute hepatitis B is particularly
important. Before laboratory methods were available to distinguish
between acute hepatitis and acute hepatitis–like exacerbations (spon­
taneous reactivations) of chronic hepatitis B, observations suggested
that ~10% of previously healthy patients remained HBsAg positive for
6 months after the onset of clinically apparent acute hepatitis B. Onehalf of these persons cleared the antigen from their circulations during
the next several years, but the other 5% remained chronically HBsAg
positive. More recent observations suggest that the true rate of chronic
infection after clinically apparent acute hepatitis B is as low as 1% in
normal, immunocompetent, young adults. Earlier, higher estimates
may have been confounded by inadvertent inclusion of acute exacerba­
tions in chronically infected patients; these patients, chronically HBsAg
positive before exacerbation, were unlikely to seroconvert to HBsAg
negative thereafter. Whether the rate of chronicity is 10% or 1%, such
patients have IgG anti-HBc in serum; anti-HBs is either undetected
or detected at low titer against the opposite subtype specificity of the
antigen (see “Laboratory Features”). These patients may (1) be inactive
carriers; (2) have low-grade, mild chronic hepatitis; or (3) have moder­
ate to severe chronic hepatitis with or without cirrhosis. The likelihood
of remaining chronically infected after acute HBV infection is espe­
cially high among neonates, persons with Down syndrome, chronically
CHAPTER 350
Acute Viral Hepatitis
hemodialyzed patients, and immunosuppressed patients, including
persons with HIV infection.
Chronic hepatitis is an important late complication of acute hepatitis B
occurring in a small proportion of patients with acute disease but more
common in those who present with chronic infection without having
experienced an acute illness, as occurs typically after neonatal infec­
tion or after infection in an immunosuppressed host (Chap. 352). The
following clinical and laboratory features suggest progression of acute
hepatitis to chronic hepatitis: (1) lack of complete resolution of clini­
cal symptoms of anorexia, weight loss, fatigue, and the persistence of
hepatomegaly; (2) the presence of bridging/interface or multilobular
hepatic necrosis on liver biopsy during protracted, severe acute viral
hepatitis; (3) failure of the serum aminotransferase, bilirubin, and
globulin levels to return to normal within 6–12 months after the acute
illness; and (4) the persistence of HBeAg for >3 months or HBsAg for
6 months after acute hepatitis.
Although acute hepatitis D infection does not increase the likeli­
hood of chronicity of simultaneous acute hepatitis B, hepatitis D has
the potential for contributing to the severity of chronic hepatitis B.
Hepatitis D superinfection can transform inactive or mild chronic
hepatitis B into severe, progressive chronic hepatitis and cirrhosis;
it also can accelerate the course of chronic hepatitis B and acceler­
ate the risk of hepatocellular carcinoma. Some HDV superinfections
in patients with chronic hepatitis B lead to fulminant hepatitis. As
defined in longitudinal studies over three decades, the annual rate of
cirrhosis in patients with chronic hepatitis D is 4%. Although HDV
and HBV infections are associated with severe liver disease, mild
hepatitis and even inactive carriage have been identified in some
patients, and the disease may become indolent beyond the early years
of infection.
PART 10
Disorders of the Gastrointestinal System
After acute HCV infection, the likelihood of remaining chronically
infected approaches 85–90%. Although many patients with chronic
hepatitis C have no symptoms, cirrhosis may develop in as many as
20% within 10–20 years of acute illness; in prior series of cases reported
by referral centers, cirrhosis has been reported in as many as 50%
of patients with chronic hepatitis C. Among cirrhotic patients with
chronic hepatitis C, the annual risk of hepatic decompensation is ~4%;
the annual risk of hepatocellular carcinoma in cirrhotic patients with
chronic hepatitis C has been reported in a range between 1 and 4%
and, in a recent definitive report, was 3.36%. With the advent of highly
effective DAA therapy during the second decade of the 21st century,
a decline in HCV-related cirrhosis and hepatocellular carcinoma has
been observed. Progression of chronic hepatitis C may be influenced
by advanced age of acquisition, long duration of infection, immuno­
suppression, coexisting excessive alcohol use, concomitant hepatic
steatosis, other hepatitis virus infection, or HIV co-infection. In fact,
instances of severe and rapidly progressive chronic hepatitis B and C
are recognized with increasing frequency in patients with HIV infec­
tion (Chap. 208). In contrast, neither HAV nor HEV causes chronic
liver disease in immunocompetent hosts; however, cases of chronic
hepatitis E (including cirrhosis and end-stage liver disease and even
hepatocellular carcinoma) have been observed in immunosuppressed
organ-transplant recipients, persons receiving cytotoxic chemotherapy,
and persons with HIV infection. Among patients with chronic hepatitis
(e.g., caused by hepatitis B or C, alcohol, etc.) in endemic countries,
hepatitis E has been reported as the cause of acute-on-chronic liver fail­
ure; however, in most experiences among patients from nonendemic
countries, HEV has not been found to contribute commonly to hepatic
decompensation in patients with chronic hepatitis.
Persons with chronic hepatitis B, particularly those infected in
infancy or early childhood and especially those with HBeAg and/or
high-level HBV DNA, have an enhanced risk of hepatocellular car­
cinoma. The risks of cirrhosis and hepatocellular carcinoma increase
with the level of HBV replication. The annual rate of hepatocellular
carcinoma in patients with chronic hepatitis D and cirrhosis is ~3%.
Rare complications of acute viral hepatitis include pancreati­
tis, myocarditis, atypical pneumonia, aplastic anemia, transverse
myelitis, peripheral neuropathy, and Guillain-Barré syndrome. In
children, hepatitis B may present rarely with anicteric hepatitis,
a nonpruritic papular rash of the face, buttocks, and limbs, and
lymphadenopathy (papular acrodermatitis of childhood or GianottiCrosti syndrome).
Rarely, autoimmune hepatitis (Chap. 352) can be triggered by a
bout of otherwise self-limited acute hepatitis, as reported after acute
hepatitis A, B, and C.
■
■DIFFERENTIAL DIAGNOSIS
Viral diseases such as infectious mononucleosis (EBV); those due
to cytomegalovirus, herpes simplex virus, adenovirus, and cox­
sackieviruses; and toxoplasmosis may share certain clinical features
with viral hepatitis and cause elevations in serum aminotransferase
and, less commonly, in serum bilirubin levels. Tests such as the
differential heterophile and serologic tests for these agents may be
helpful in the differential diagnosis if HBsAg, anti-HBc, IgM antiHAV, and anti-HCV (and anti-HEV) determinations are negative.
Aminotransferase elevations can accompany almost any systemic
viral infection, including the coronavirus SARS-CoV-2 (~10% of all
cases and up to half of severe cases); other rare causes of liver injury
confused with viral hepatitis are infections with Leptospira, Candida,
Brucella, Mycobacteria, and Pneumocystis. A complete drug history
is particularly important, because many drugs and certain anesthetic
agents can produce a picture of either acute hepatitis or cholestasis
(Chap. 351). Equally important is a history of unexplained “repeated
episodes” of acute hepatitis. This history should alert the physician
to the possibility that the underlying disorder is chronic hepatitis,
for example, autoimmune hepatitis (Chap. 352). Alcohol-associated
hepatitis (AH), which can cause an acute icteric hepatitis, must
also be considered, but usually the serum aminotransferase levels
are not as markedly elevated (generally not much above 200 IU/
mL, almost always below 400 IU/mL) in AH, and a heavy drink­
ing history must be present (at least 40 g/d in women and 60 g/d
in men for at least 6 months). The finding on liver biopsy of fatty
infiltration, a neutrophilic inflammatory reaction, and “alcoholic
hyaline” would be consistent with alcohol-induced rather than viral
liver injury. Because acute hepatitis may present with right upper
quadrant abdominal pain, nausea and vomiting, fever, and icterus,
it is often confused with acute cholecystitis, common duct stone,
or ascending cholangitis. Patients with acute viral hepatitis may
tolerate surgery poorly, and surgical risk is particularly high in the
elderly and those with underlying chronic liver disease; therefore,
excluding biliary disease clinically and with imaging is important.
In confusing cases, a percutaneous liver biopsy may be helpful. Viral
hepatitis in the elderly is often misdiagnosed as obstructive jaundice
resulting from a common duct stone or carcinoma of the pancreas.
Another clinical constellation that may mimic acute hepatitis is right
ventricular failure with passive hepatic congestion or hypoperfusion
syndromes, such as those associated with shock, severe hypoten­
sion, and severe left ventricular failure. Also included in this general
category is any disorder that interferes with venous return to the
heart, such as right atrial myxoma, constrictive pericarditis, hepatic
vein occlusion (Budd-Chiari syndrome), or venoocclusive disease.
Clinical features are usually sufficient to distinguish among these
vascular disorders and viral hepatitis. Acute fatty liver of pregnancy,
cholestasis of pregnancy, eclampsia, and the HELLP (hemolysis,
elevated liver tests, and low platelets) syndrome can be confused with
viral hepatitis during pregnancy. Very rarely, malignancies meta­
static to the liver can mimic acute or even fulminant viral hepatitis.
Occasionally, genetic or metabolic liver disorders (e.g., Wilson’s dis­
ease, α1 antitrypsin deficiency) and nonalcoholic fatty liver disease
(NAFLD; now labeled metabolic dysfunction–associated steatotic
liver disease [MASLD]; what used to be called nonalcoholic steato­
hepatitis [NASH] is now labeled metabolic dysfunction–associated
steatohepatitis [MASH]) are confused with acute viral hepatitis
(Chap. 354). Among patients with biochemical evidence for severe
liver injury, i.e., aminotransferase levels of ≥1000 IU/L, the most
common causes are ischemic liver injury, drug-induced liver injury
(especially caused by acetaminophen), acute viral hepatitis, and pan­
creaticobiliary disorders.
TREATMENT
Acute Viral Hepatitis
Most persons with acute hepatitis (especially hepatitis A, B, and E)
recover spontaneously and do not require specific antiviral therapy.
In hepatitis B, among previously healthy adults who present with
clinically apparent acute hepatitis, recovery occurs in ~99%; there­
fore, antiviral therapy is not likely to improve the rate of recovery
and is not required. In rare instances of severe acute hepatitis B,
treatment with a nucleoside analogue at oral doses used to treat
chronic hepatitis B (Chap. 352) is included in treatment guide­
lines by the American Association for the Study of Liver Diseases
(AASLD) and the European Association for the Study of the Liver
(EASL). Indications include fulminant liver failure (along with such
therapy, referral to a liver transplantation center is advised) and a
prolonged course (>4 weeks) of acute hepatitis B. Although clinical
trials have not been done to establish the efficacy or duration of this
approach, most authorities would recommend institution of antivi­
ral therapy with a nucleoside analogue (entecavir or tenofovir, the
most potent and least resistance-prone agents) for severe, but not
mild-moderate, acute hepatitis B. Treatment should continue until
3 months after HBsAg seroconversion or 6 months after HBeAg
seroconversion.
In typical cases of acute hepatitis C, recovery is rare (~15–20% in
most experiences), and progression to chronic hepatitis is the rule.
Patients with jaundice, those with HCV genotype 1, women, and
those with earlier age of infection, lower level of HCV RNA, and
HBV co-infection, are more likely to recover from acute hepatitis C,
as are persons who have genetic markers associated with spontane­
ous recovery (favorable IL28B CC haplotype).
Because spontaneous recovery can occur and because most cases
of acute hepatitis C are not clinically severe or rapidly progres­
sive, during the era of interferon-based therapy, delaying antiviral
therapy of acute hepatitis C for 3–6 months was recommended;
however, in the current era of highly effective (95–100%) oral DAA
therapy, waiting for potential spontaneous recovery is no longer
advised. Instead, early treatment with a standard, full 8- to 12-week
course of one of the first-line drug combinations approved for
treatment of chronic hepatitis C (Chap. 352) is recommended for
treatment of patients with acute hepatitis C.
Because of the vast reservoir of acute HCV infections acquired
four to five decades ago in the 1945–1965 birth cohort, most newly
recognized HCV infections are chronic. Opportunities to identify
and treat patients with acute hepatitis C occur in two population
subsets: (1) in health care workers who sustain hepatitis C–contam­
inated needle sticks (occupational accidents), monitoring for ALT
elevations and the presence of HCV RNA identify acute hepatitis C
in ~3%, and this group should be treated; (2) in persons who use
injection drugs, the risk of acute hepatitis C has been on the rise
during the previous two decades, and the epidemic of opioid use
has contributed to an amplification of HCV infection among drug
users. Such persons are candidates for antiviral therapy, and efforts
to combine antiviral therapy with drug rehabilitation therapy have
been very successful.
Notwithstanding these specific therapeutic considerations, in
most cases of typical acute viral hepatitis, specific treatment gen­
erally is not necessary. Although hospitalization may be required
for clinically severe illness, most patients do not require hospi­
tal care. Forced and prolonged bed rest is not essential for full
recovery, but many patients will feel better with restricted physi­
cal activity. A high-calorie diet is desirable, and because many
patients may experience nausea late in the day, the major caloric
intake is best tolerated in the morning. Intravenous feeding is nec­
essary in the acute stage if the patient has persistent vomiting and
cannot maintain oral intake. Drugs capable of producing adverse
reactions such as cholestasis and drugs metabolized by the liver
should be avoided. If severe pruritus is present, the use of the bile
salt–sequestering resin cholestyramine is helpful. Glucocorticoid
therapy has no value in acute viral hepatitis, even in severe cases,
and may be deleterious, even increasing the risk of chronicity
(e.g., of acute hepatitis B).
Physical isolation of patients with hepatitis to a single room and
bathroom is rarely necessary except in the case of fecal inconti­
nence for hepatitis A and E or uncontrolled, voluminous bleeding
for hepatitis B (with or without concomitant hepatitis D) and C.
Because most patients hospitalized with hepatitis A excrete little, if
any, HAV, the likelihood of HAV transmission from these patients
during their hospitalization is low. Therefore, burdensome enteric
precautions are no longer recommended. Although gloves should be
worn when the bed pans or fecal material of patients with hepati­
tis A are handled, these precautions do not represent a departure
from sensible procedure and contemporary universal precautions
for all hospitalized patients. For patients with hepatitis B and C,
emphasis should be placed on blood precautions (i.e., avoiding
direct, ungloved hand contact with blood and other body fluids).
Enteric precautions are unnecessary. The importance of simple
hygienic precautions such as hand washing cannot be overempha­
sized. Universal precautions that have been adopted for all patients
apply to patients with viral hepatitis. Hospitalized patients may
be discharged following substantial symptomatic improvement,
a significant downward trend in the serum aminotransferase and
bilirubin values, and a return to normal of the PT. Mild aminotrans­
ferase elevations should not be considered contraindications to the
gradual resumption of normal activity.
CHAPTER 350
In fulminant hepatitis, the goal of therapy is to support the patient
by maintenance of fluid balance, nutrition, support of circulation and
respiration, control of bleeding, correction of hypoglycemia, early
identification and treatment of infection, and treatment of other com­
plications of the comatose state in anticipation of liver regeneration and
repair. Glucocorticoid therapy has been shown in controlled trials to
be ineffective. Likewise, exchange transfusion, plasmapheresis, human
cross-circulation, porcine liver cross-perfusion, hemoperfusion, and
extracorporeal liver-assist devices have not been proven to enhance
survival. The cornerstone of management is meticulous intensive
care with vigilant monitoring for infection (and low threshold for
initiation of antibiotics and antifungals), careful mental status/cerebral
edema monitoring, and management of bleeding complications. Liver
transplantation is resorted to with increasing frequency, with excellent
results, in patients with fulminant hepatitis (Chap. 356). Fulminant
hepatitis C is very rare; however, in fulminant hepatitis B, oral antiviral
therapy has been used successfully, although the benefit may be limited
once late-stage acute liver failure is present. In clinically severe acute
hepatitis E or acute-on-chronic liver failure, successful therapy with
ribavirin (600 mg twice daily, 15 mg/kg) has been reported; however,
the precise dose and duration of therapy have not been established, and
conflicting reports have appeared documenting the futility of ribavirin
in this setting. Unfortunately, when fulminant hepatitis E occurs in
pregnant women (as it does in up to 20% of pregnant women with
acute hepatitis E), ribavirin, which is teratogenic, is contraindicated. In
cases of hepatitis E in organ-transplant recipients, reduction in overall
immunosuppressive drug doses and switching from tacrolimus to
cyclosporine A have been shown to be effective, often without antiviral
therapy, in achieving eradication of HEV. If a change in immunosup­
pression is inadequate, ribavirin treatment for 3 months has been
observed to achieve a sustained virologic response in 78% of treated
patients; again however, the optimal dose, duration, and efficacy of
ribavirin therapy remain to be determined.
Acute Viral Hepatitis
■
■PROPHYLAXIS
Because application of therapy for acute viral hepatitis is limited and
because chronic viral hepatitis requires prolonged and costly courses
of antiviral therapy (Chap. 352), emphasis is placed on preven­
tion through immunization. The prophylactic approach differs for
each of the types of viral hepatitis. In the past, immunoprophylaxis
relied exclusively on passive immunization with antibody-containing
globulin preparations purified by cold ethanol fractionation from the
plasma of hundreds of normal donors. Currently, for hepatitis A, B,
and E, active immunization with vaccines is the preferable approach
to prevention.
Hepatitis A 
Both passive immunization with immunoglobulin
(IG) and active immunization with killed vaccines are available. Pas­
sive immunization, which can be considered in the setting of recent
exposure, is 80–90% effective at preventing infection and provides
12–20 weeks of protection; however, its utility is limited by availability
and cost (and the competing appeal of active immunization with vac­
cine). For postexposure prophylaxis of intimate contacts (household,
sexual, institutional) of persons with hepatitis A, the administration of
0.1 mL/kg is recommended as early after exposure as possible but must
be within 14 days. Prophylaxis is not necessary for those who have
already received hepatitis A vaccine, for casual contacts (office, factory,
school, or hospital), for most elderly persons, who are very likely to be
immune, or for those known to have anti-HAV in their serum. By the
time most common-source outbreaks of hepatitis A are recognized, it
is usually too late in the incubation period for IG to be effective; how­
ever, prophylaxis may have limited the frequency of secondary cases.
For travelers to tropical countries, developing countries, and other
areas outside standard tourist routes, IG prophylaxis had been recom­
mended before a vaccine became available.
Recommendations for postexposure prophylaxis and for preexpo­
sure prophylaxis for international travel were updated in 2020. Cur­
rently, hepatitis A vaccine, not IG, is recommended for all persons
aged ≥12 months for postexposure prophylaxis and for preexposure
prophylaxis prior to international travel to HAV-endemic areas. Even
though hepatitis A vaccine is indicated for children ≥12 months of
age, when infants aged 6–11 months travel internationally to areas
with a risk of HAV infection, they should receive the vaccine for
preexposure prophylaxis; however, this travel-related dose should
not be counted toward the universal childhood two-dose hepatitis A
vaccine recommendation, which begins at age 12 months. For post­
exposure prophylaxis of persons with contraindications to hepatitis
A vaccination and infants aged <12 months, the use of IG (0.1 mL/
kg) should be retained. In addition, for postexposure prophylaxis in
immunocompromised adults and persons with chronic liver disease,
both hepatitis A vaccination and IG administration (0.1 mL/kg), at
different IM sites, are recommended. Finally, for infants aged <6
months and for persons with contraindications to hepatitis A vac­
cination, preexposure prophylaxis for travel consists of IG at doses
of 0.1 mg/kg for travel durations up to 1 month, 0.2 mg/kg for travel
up to 2 months, and repeat 0.2 mg/kg every 2 months thereafter for
the remainder of travel. Thus, except for these limited considerations,
hepatitis A vaccine has supplanted IG in almost all cases for both
postexposure prophylaxis and preexposure prophylaxis for travel.
In general, practical use of IG at this point tends to be confined to
outbreaks, in which early containment is critical. Unlike IG prophy­
laxis, the protection afforded by active immunization with vaccine is
durable and simpler to administer.
PART 10
Disorders of the Gastrointestinal System
Formalin-inactivated vaccines made from strains of HAV attenu­
ated in tissue culture have been shown to be safe, immunogenic, and
effective in preventing hepatitis A. Hepatitis A vaccines are approved
for use in persons who are at least 1 year old and appear to provide
adequate protection beginning 4 weeks after a primary inoculation. As
noted above, for travel to an endemic area, hepatitis A vaccine is the
preferred approach to preexposure immunoprophylaxis and provides
long-lasting protection (protective levels of anti-HAV should last at
least 20 years after vaccination). Shortly after its introduction, hepatitis
A vaccine was recommended for children living in communities with
a high incidence of HAV infection; in 1999, this recommendation was
extended to include all children living in states, counties, and com­
munities with high rates of HAV infection. As of 2006, the Advisory
Committee on Immunization Practices of the U.S. Public Health Ser­
vice recommended routine hepatitis A vaccination of all children. Other
groups considered being at increased risk for HAV infection and who
are candidates for hepatitis A vaccination include military personnel,
populations with cyclic outbreaks of hepatitis A (e.g., Alaskan natives),
employees of day-care centers and persons working in facilities for
the developmentally delayed, primate handlers, laboratory workers
exposed to hepatitis A or fecal specimens, and patients with chronic
liver disease (including persons with aminotransferase elevations
≥2 times the upper limit of normal). Because of an increased risk of
fulminant hepatitis A—observed in some experiences but not con­
firmed in others—among patients with chronic hepatitis C, patients
with chronic hepatitis C are candidates for hepatitis A vaccination, as
are persons with chronic hepatitis B, persons with HIV infection, and
the expanding population of persons with nonalcoholic liver disease
(MASLD). Other populations whose recognized risk of hepatitis A is
increased should be vaccinated, including men who have sex with men,
persons who use injection or noninjection drugs, persons experienc­
ing homelessness, persons traveling from the United States to coun­
tries with high or intermediate hepatitis A endemicity, postexposure
prophylaxis for contacts of persons with hepatitis A, and household
members and other close contacts of (or anyone who anticipates close
personal contact with) adopted children arriving from countries with
high and moderate hepatitis A endemicity. Hepatitis A vaccine is now
recommended as well for pregnant women at risk of infection or severe
outcomes from infection during pregnancy. Recommendations for
dose and frequency differ for the two approved vaccine preparations in
the United States and the combination vaccines that include hepatitis
A (Table 350-7); all injections are IM. Hepatitis A vaccine has been
reported to be effective in preventing secondary household and daycare center–associated cases of acute hepatitis A. In the United States,
reported mortality resulting from hepatitis A declined in parallel with
hepatitis A vaccine–associated reductions in the annual incidence of
new infections.
Hepatitis B 
Until 1982, prevention of hepatitis B was based on
passive immunoprophylaxis either with standard IG, containing
modest levels of anti-HBs, or hepatitis B immunoglobulin (HBIG),
containing high-titer anti-HBs. The efficacy of standard IG has
never been established and remains questionable; even the efficacy
of HBIG, demonstrated in several clinical trials, has been challenged,
and its contribution appears to be in reducing the frequency of
clinical illness, not in preventing infection. The first vaccine for active
immunization, introduced in 1982, was prepared from purified,
noninfectious, 22-nm spherical HBsAg particles derived from the
plasma of healthy HBsAg carriers. In 1987, the plasma-derived vac­
cine was supplanted by a genetically engineered vaccine derived from
recombinant yeast. The latter vaccine consists of HBsAg particles that
are nonglycosylated but are otherwise indistinguishable from natural
HBsAg; two recombinant vaccines were licensed for use in the United
States in the 1980s (Recombivax-HB 1986; Engerix-B 1989), a third
(Heplisav-B) was licensed in 2017, and a fourth recombinant vaccine
(PreHevbrio, VBI Vaccines) was licensed in 2022. In the United States,
TABLE 350-7  Hepatitis A Vaccination Schedules
AGE, YEARS
NO. OF DOSES
DOSE
SCHEDULE, MONTHS
HAVRIX (GlaxoSmithKline)a
1–18
≥19
720 ELUb (0.5 mL)
1440 ELU (1 mL)
0, 6–12
0, 6–12
VAQTA (Merck)
1–18
≥19
25 units (0.5 mL)
50 units (1 mL)
0, 6–18
0, 6–18
aA combination of this hepatitis A vaccine and hepatitis B vaccine, TWINRIX, is
licensed for simultaneous protection against both of these viruses among adults
(age ≥18 years). Each 1-mL dose contains 720 ELU of hepatitis A vaccine and 20
μg of hepatitis B vaccine. These doses are recommended at months 0, 1, and 6.
bEnzyme-linked immunoassay units. cCombination hepatitis A and typhoid vaccines,
Hepatyrix (GlaxoSmithKline) and Viatim (Sanofi Pasteur), are available, targeted
primarily for travelers to endemic areas. Please consult product insert for doses
and schedules
TABLE 350-8  Preexposure Hepatitis B Vaccinations
NAME
VACCINE TYPE
AGE GROUP
NO. OF DOSES
SCHEDULE, MONTH
NOTES
Engerix-B
(GlaxoSmithKline)
Single antigen
recombinant
From birth, not on
dialysis
0, 1, 6
0.5-mL dose for 0–19 years
1-mL dose for ≥20 years
Contraindicated with severe yeast allergy
 
 
Adults on dialysis
0, 1, 2, 6
2-mL dose for adults
Recombivax HB
(Merck)
Single antigen
recombinant
From birth, not on
dialysis
0, 1, 6
0.5-mL dose for 0–19 years
1-mL dose for ≥20 years
Contraindicated with severe yeast allergy
Recombivax HBV
dialysis formulation
(Merck)
Single antigen
recombinant
Adults on dialysis
0, 1, 6
1-mL dose for adults 
Heplisav-B
(Dynavax)
Single antigen adjuvanted
recombinant
≥18 years
0, 1
0.5-mL dose
Higher levels of protective antibodies, particularly
in adults ≥40 years
Not for use in severe yeast allergy
PreHevbrio (VBI)*
Three antigen
recombinant
≥18 years
0, 1, 6
1-mL dose
Higher rates of seroprotection in adults ≥45 years
Not for use in pregnant women
Twinrix
(GlaxoSmithKline)
Combined single antigen
recombinant hepatitis A
and hepatitis B
≥18 years
0, 1, 6
Note: accelerated 4-dose
regimen, days 0, 7, and
21–30, 12 months
*In November 2024, distribution of this vaccine in the United States was discontinued by the manufacturer, because of bankruptcy and termination of operations.
universal birth vaccination against HBV infection has been the standard
of care since 2002, with more recent recommendations suggesting fur­
ther that the birth dose be administered within 12 hours of delivery.
For unvaccinated persons, current recommendations can be divided
into those for preexposure and postexposure prophylaxis.
For preexposure prophylaxis against hepatitis B in persons with a
high risk of exposure, three IM (deltoid, not gluteal) injections of hepa­
titis B vaccine are recommended at 0, 1, and 6 months (other, optional
schedules are summarized in Table 350-8). In 2022, the Advisory Com­
mittee on Immunization Practices (ACIP) recommended universal
vaccination for all unvaccinated persons aged 19–59 years. Among
persons 60 years and older, vaccination is recommended for those at
high risk as well as any interested person. Groups considered at high
risk include (1) those with sexual exposure (sex workers, those with
HBsAg partners, patients at sexually transmitted disease clinics), (2)
those with frequent blood exposure (health care workers, first respond­
ers, persons who use injection drugs, hemophiliacs, dialysis patients,
persons with end-stage renal disease), (3) those who frequently travel
to, or first-generation immigrants from, moderate- to high-endemicity
areas (>2% prevalence of HBV infection), (4) those with chronic liver
disease (HCV infection, steatotic liver disease, autoimmune hepatitis,
alcohol-associated liver disease, or aminotransferases >2 times the
upper limit of normal), and (5) several other groups (e.g., incarcerated
persons and those with HIV infection and diabetes mellitus). Pregnancy
is not a contraindication to vaccination details of the use of Heplisav-B,
a two-injection course a month apart, appear in Table 350-8. In areas of
low HBV endemicity such as the United States, despite the availability of
safe and effective hepatitis B vaccines, a strategy of vaccinating persons
in high-risk groups was not effective. The incidence of new hepatitis B
cases continued to increase in the United States after the introduction of
vaccines; <10% of all targeted persons in high-risk groups were actually
vaccinated, and ~30% of persons with sporadic acute hepatitis B did not
fall into any high-risk group category. Therefore, to have an impact on
the frequency of HBV infection in an area of low endemicity such as the
United States, universal hepatitis B vaccination is now recommended. In
HBV-hyperendemic areas (e.g., Asia), universal vaccination of children
has resulted in a marked (~70–90%) 40-year decline in complications
of hepatitis B, including liver-related mortality and hepatocellular
carcinoma.
1-mL dose
Severe allergy to yeast or any hepatitis A vaccine
CHAPTER 350
The original two available aluminum-adjuvanted recombinant
hepatitis B vaccines are comparable, one containing 10 μg of HBsAg
(Recombivax-HB) and the other containing 20 μg of HBsAg (Engerix-B),
and recommended doses for each injection vary for the two prepara­
tions (Table 350-8). Combinations of hepatitis B vaccine with other
childhood vaccines are available as well (Table 350-8).
Acute Viral Hepatitis
More recently, two new recombinant hepatitis B vaccines have been
approved for use. One is a recombinant vaccine with a novel adjuvant
that activates toll-like 9 receptors (Heplisav-B) and is approved for
adults aged 18 or older (although it should be avoided in pregnancy).
In a series of prospective trials, compared to three Engerix-B injec­
tions, two IM doses a month apart yielded higher proportions with
protective levels of anti-HBs (≥10 mIU/mL): 95% of adults aged 18–55
or 18–70 (vs 81% for Engerix-B), 90% of older adults aged 40–70 (vs
71% for Engerix-B), and 90% of adults aged 18–70 with type 2 diabetes
(vs 65% for Engerix-B). This two-injection regimen may be useful for
revaccination of persons who failed to respond to the original vaccines.
The second novel recombinant vaccine (PreHevbrio, VBI Vaccines),
which contains all three hepatitis B surface antigen proteins, S, pre-S1,
and pre-S2, has been shown in clinical trials (three IM doses at 0, 1, and
6 months) to achieve higher proportions with protective anti-HBs and
higher antibody levels than Engerix-B (which contains S antigen only).
In addition, PreHevbrio was shown in clinical trials to be associated
with higher rates of seroprotection, e.g., in persons ≥45 years (89%
compared to 73.1% with an old standard vaccine). This vaccine was
approved by the FDA on December 1, 2021, for adults age ≥18 years
and recommended for use by the ACIP of the CDC in March 2022. It is
the only vaccine not contraindicated in the presence of a yeast allergy.*
For unvaccinated persons sustaining an exposure to HBV, postex­
posure prophylaxis with a combination of HBIG (for rapid achieve­
ment of high-titer circulating anti-HBs) and hepatitis B vaccine
(for achievement of long-lasting immunity as well as its apparent
efficacy in attenuating clinical illness after exposure) is recom­
mended. For perinatal exposure of infants born to HBsAg-positive
*In November 2024, distribution of this vaccine in the United States was
discontinued by the manufacturer, because of bankruptcy and termination of
operations.
mothers, a single dose of HBIG, 0.5 mL, should be administered IM
in the thigh immediately after birth, followed by a complete course
of three injections of recombinant hepatitis B vaccines approved
for children (see doses above) to be started within the first 12 h of
life. For those experiencing a direct percutaneous inoculation or
transmucosal exposure to HBsAg-positive blood or body fluids (e.g.,
accidental needle stick, other mucosal penetration, or ingestion),
a single IM dose of HBIG, 0.06 mL/kg, administered as soon after
exposure as possible, is followed by a complete course of hepatitis
B vaccine to begin within the first week. For pregnant mothers with
high-level HBV DNA (>2 × 105 IU/mL), adding antiviral nucleoside
analogues (e.g., pregnancy class B tenofovir, see Chap. 352) during
the third trimester of pregnancy reduces perinatal transmission
even further. For persons exposed by sexual contact to a patient
with acute hepatitis B, a single IM dose of HBIG, 0.06 mL/kg,
should be given within 14 days of exposure, to be followed by a com­
plete course of hepatitis B vaccine. When both HBIG and hepatitis B
vaccine are recommended, they may be given at the same time but at
separate sites. Testing adults for anti-HBs after a course of vaccine is
advisable to document the acquisition of immunity (and can guide
management in nonresponders, e.g., whether HBIG and/or repeat
vaccination with a more immunogenic vaccine preparation is war­
ranted), but because hepatitis B vaccine immunogenicity is nearly
universal in infants, postvaccination anti-HBs testing of children is
not recommended. Similarly, testing for the presence of anti-HBs
in exposed adults can help guide management (e.g., HBIG, repeat
vaccination).
The precise duration of protection afforded by hepatitis B vaccine
is unknown; however, ~80–90% of immunocompetent adult vaccinees
retain protective levels of anti-HBs for at least 5 years, and 60–80% for
10 years, and protective antibody has been documented to last for at
least two decades after vaccination in infancy. Thereafter and even after
anti-HBs becomes undetectable, long-term protection (via an anam­
nestic immune response) persists against clinical hepatitis B, hepatitis
B surface antigenemia, and chronic HBV infection. Currently, booster
immunizations are not recommended routinely, except in immunosup­
pressed persons who have lost detectable anti-HBs or immunocompe­
tent persons who sustain percutaneous HBsAg-positive inoculations
after losing detectable antibody. Specifically, for hemodialysis patients,
annual anti-HBs testing is recommended after vaccination; booster
doses are recommended when anti-HBs levels fall to <10 mIU/mL. As
noted above, for persons at risk of both hepatitis A and B, a combined
vaccine is available containing 720 enzyme-linked immunoassay units
(ELUs) of inactivated HAV and 20 μg of recombinant HBsAg (at 0, 1,
and 6 months).
PART 10
Disorders of the Gastrointestinal System
Hepatitis D 
Infection with hepatitis D can be prevented by vac­
cinating susceptible persons with hepatitis B vaccine. No product is
available for immunoprophylaxis to prevent HDV superinfection in
persons with chronic HBV infection; for these patients, avoidance of
percutaneous exposures and limitation of intimate contact with per­
sons who have HDV infection are recommended.
Hepatitis E 
For prevention of hepatitis E, IG derived from HEVendemic populations does not appear to be effective. Two safe and
effective three-dose (0, 1, and 6 months), recombinant genotype 1
capsid protein vaccines have been shown in randomized, placebocontrolled trials to be highly protective (including against other
genotypes) against symptomatic acute hepatitis E. A Chinese vaccine,
Hecolin, achieved 100% 12-month efficacy and was licensed in China
in 2011; its long-lasting protection (87% efficacy) was documented for
up to 4.5 years. A second vaccine developed by GlaxoSmithKline and
the U.S. Army achieved a 12-month 96% efficacy. The second vaccine
was never developed commercially. The Chinese vaccine is available in
China but is not FDA approved or available in the United States.
Hepatitis C 
IG is ineffective in preventing hepatitis C and is
no longer recommended for postexposure prophylaxis in cases of
perinatal, needle stick, or sexual exposure. Although prototype
vaccines that induce antibodies to HCV envelope proteins have been
developed, currently, hepatitis C vaccination is not feasible practi­
cally. Genotype and quasispecies viral heterogeneity, as well as rapid
evasion of neutralizing antibodies by this rapidly mutating virus,
conspire to render HCV a difficult target for immunoprophylaxis
with a vaccine. Prevention of transfusion-associated hepatitis C has
been accomplished successfully by implementation of highly sensitive
virologic screening tests, and the current estimated risk is less than
1 in 10 million.
In the absence of active or passive immunization, prevention of
hepatitis C includes behavior changes and precautions to limit expo­
sures to infected persons. Recommendations designed to identify
patients with clinically inapparent hepatitis as candidates for medical
management have as a secondary benefit the identification of persons
whose contacts could be at risk of becoming infected. A so-called
look-back program has been recommended to identify persons who
were transfused before 1992 with blood from a donor found sub­
sequently to have hepatitis C. In addition, anti-HCV testing, once
recommended for persons born between 1945 and 1965, has now
been expanded to include all persons 18 years or older, indepen­
dent of risk factors. For stable, monogamous sexual partners, sexual
transmission of hepatitis C is unlikely, and sexual barrier precautions
are not recommended. For persons with multiple sexual partners or
with sexually transmitted diseases, the risk of sexual transmission of
hepatitis C is increased, and barrier precautions (latex condoms) are
recommended. A person with hepatitis C should avoid sharing such
items as razors, toothbrushes, and nail clippers with sexual partners
and family members. No special precautions are recommended for
babies born to mothers with hepatitis C, and breast-feeding does not
have to be restricted.
A more recent, ambitious goal toward HCV prevention steers focus
toward its global eradication by 2030, which would require higher
rates of and investment in global surveillance, immediate reflex treat­
ment in persons who test positive for HCV infection, and undelayed,
streamlined initiation of treatment (including, and perhaps most
importantly, in people who inject drugs to prevent secondary spread)
(Table 350-4).
■
■FURTHER READING
Asselah T, Rizzetto M: Hepatitis D virus infection. N Engl J Med
389:58, 2023.
Conners EE, et al: Screening and testing for hepatitis B virus infection:
CDC recommendations—United States, 2023. MMWR Morb Mortal
Wkly Rep 72:1, 2023.
Doshani M et al: Recommendations of the Advisory Committee on
Immunization Practices for use of hepatitis A vaccine for persons
experiencing homelessness. MMWR Morb Mortal Wkly Rep 68:153,
2019.
European Association for the Study of the Liver: EASL clinical
practice guidelines on hepatitis E virus infection. J Hepatol 68:1256,
2018.
Freedman M et al: Advisory Committee on Immunization Practices.
Recommended adult immunization schedule, United States, 2020.
Ann Intern Med 172:337, 2020.
Goldberg D et al: Changes in the prevalence of hepatitis C virus
infection, nonalcoholic steatohepatitis, and alcoholic liver disease
among patients with cirrhosis and liver failure on the waitlist for liver
transplantation. Gastroenterology 152:1090, 2017.
Hofmeister MG et al: Estimating prevalence of hepatitis C virus
infection in the United States, 2013-2016. Hepatology 69:1020,
Jeng W-J et al: Hepatitis B. Lancet 401:1039, 2023.
Kelgeri C et al: Clinical spectrum of children with acute hepatitis of
unknown cause. N Engl J Med 387:611, 2022.
Koh C et al: Pathogenesis of and new therapies for hepatitis D. Gastro­
enterology 156:461, 2019.
Le MH et al: Chronic hepatitis B prevalence among foreign-born and
U.S.-born adults in the United States, 1999-2016. Hepatology 71:431,
2020.

23 - 351 Toxic and Drug-Induced Hepatitis
351 Toxic and Drug-Induced Hepatitis
Lemon SM et al: Type A viral hepatitis: A summary and update on
the molecular virology, epidemiology, pathogenesis, and prevention.
J Hepatol 68:167, 2018.
Lewis KC et al: Estimated prevalence and awareness of hepatitis C
virus infection among US adults: National Health and Nutrition
Examination Survey, January 2017–March 2020. Clin Infect Dis
77:1413, 2023.
Nelson NP et al: Prevention of hepatitis A virus infection in the United
States; recommendations of the Advisory Committee on Immunization
Practices, 2020. MMWR Morb Mortal Wkly Rep 69:1, 2020.
Roberts H et al: Prevalence of chronic hepatitis B virus (HBV) infection
in U.S. households: National Health and Nutrition Examination Survey
(NHANES), 1988–2012. Hepatology 63:388, 2016.
Schillie S et al: CDC recommendations for hepatitis C screening
among adults—United States, 2020. MMWR Recommend Rep
69:1, 2020.
Schweitzer A et al: Estimations of worldwide prevalence of chronic
hepatitis B virus infection: A systematic review of data published
between 1965 and 2013. Lancet 386:1546, 2015.
Stockdale AJ et al: The global prevalence of hepatitis D virus infec­
tion: Systematic review and meta-analysis. J Hepatol 73:523, 2020.
Sureau C et al: The hepatitis delta virus: replication and pathogenesis.
J Hepatol 64:S102, 2016.
Van Damme P et al: Hepatitis A virus infection. Nat Rev Dis Primers
9:51, 2023.
Waked I et al: Screening and treatment program to eliminate hepatitis
C in Egypt. N Engl J Med 382:1166, 2020.
Weng MK et al: Universal hepatitis B vaccination in adults aged
19–59 years: Updated recommendations of the Advisory Committee on
Immunization Practices—United States, 2022. MMWR Morb Mortal
Wkly Rep 71:477, 2022.
William M. Lee, Jules L. Dienstag
Toxic and Drug-Induced
Hepatitis
Liver injury is a possible consequence of ingestion of any xenobiotic,
including industrial toxins, pharmacologic agents, and complementary
and alternative medications (CAMs). Among patients with acute liver
failure, drug-induced liver injury (DILI) is the most common cause,
and evidence for hepatotoxicity detected during clinical trials for drug
development is the most common reason for failure of compounds to
reach approval status. DILI requires careful history-taking to identify
unrecognized exposure to chemicals used in work or at home, drugs
taken by prescription or bought over the counter, and herbal or dietary
supplement medicines. Hepatotoxic drugs can injure the hepatocyte
directly, for example, via a free-radical or metabolic intermediate that
causes peroxidation of membrane lipids and that results in liver cell
injury. Alternatively, a drug or its metabolite may activate compo­
nents of the innate or adaptive immune system, stimulate apoptotic
pathways, or initiate damage to bile excretory pathways (Fig. 351-1).
Interference with bile canalicular pumps can allow endogenous bile
acids, which can injure the liver, to accumulate. Such secondary injury,
in turn, may lead to necrosis of hepatocytes; injure bile ducts, produc­
ing cholestasis; or block pathways of lipid movement, inhibit protein
synthesis, or impair mitochondrial oxidation of fatty acids, resulting in
lactic acidosis and intracellular triglyceride accumulation (expressed
histologically as microvesicular steatosis). In other instances, drug
metabolites sensitize hepatocytes to toxic cytokines. The differences
observed between susceptible and nonsusceptible drug recipients may
be attributable to human leukocyte antigen (HLA) haplotypes that
determine binding of drug-related haptens on the cell surface as well
as to polymorphisms in elaboration of competing, protective cytokines,
as has been suggested for acetaminophen hepatotoxicity (see below).
Immune mechanisms may include cytotoxic lymphocytes or antibodymediated cellular cytotoxicity. In addition, a role has been shown for
activation of nuclear transporters, such as the constitutive androstane
receptor (CAR) or, more recently, the pregnane X receptor (PXR), in
the induction of drug hepatotoxicity.
■
■DRUG METABOLISM
Most drugs, which are water-insoluble, undergo a series of metabolic
steps, culminating in a water-soluble form appropriate for renal or
biliary excretion. This process begins with oxidation or methylation
mediated initially by the microsomal mixed function oxygenases,
cytochrome P450 (phase I reaction), followed by glucuronidation or
sulfation (phase II reaction) or inactivation by glutathione. Most drug
hepatotoxicity is the result of formation of a phase I toxic metabolite,
but glutathione depletion, precluding inactivation of harmful com­
pounds by glutathione S-transferase, can contribute as well by ensuring
that the toxic compound is not abrogated.
■
■LIVER INJURY CAUSED BY DRUGS
In general, two major types of chemical hepatotoxicity have been rec­
ognized: (1) direct toxic and (2) idiosyncratic. As shown in Table 351-1,
direct toxic hepatitis occurs with predictable regularity in individuals
exposed to the offending agent and is dose-dependent. The latent
period between exposure and liver injury is usually short (often several
hours), although clinical manifestations may be delayed for 24–48 h.
Agents producing toxic hepatitis are generally systemic poisons or are
converted in the liver to toxic metabolites. The direct hepatotoxins
result in morphologic abnormalities that are reasonably characteristic
and reproducible for each toxin. Examples of rare toxins currently
include carbon tetrachloride and trichloroethylene that characteristi­
cally produce a centrilobular zonal necrosis. The hepatotoxic octapep­
tides of Amanita phalloides usually produce massive hepatic necrosis;
the lethal dose of the toxin is ~10 mg, the amount found in a single
deathcap mushroom. Acetaminophen, the prime example of a direct
toxin, is discussed below.
CHAPTER 351
Toxic and Drug-Induced Hepatitis
In idiosyncratic drug reactions, the occurrence of liver injury is
infrequent (1 in 103–105 patients) and unpredictable; the response
is not as clearly dose-dependent as is injury associated with direct
hepatotoxins, and liver injury may occur at any time after exposure to
the drug but typically between 5 and 90 days following its initiation.
Although regarded as not dose-related in the fashion of direct toxins,
most agents causing idiosyncratic toxicity are given at relatively high
daily doses, typically exceeding 100 mg, suggesting a role for dose—
drugs with low potency must be given in higher doses that engender
greater chances for “off-target” effects. Likewise, drugs given in mil­
ligram amounts are of high potency and rarely cause liver or other
off-target effects. Adding to the difficulty of predicting or identifying
idiosyncratic drug hepatotoxicity is the occurrence of mild, transient,
nonprogressive serum aminotransferase elevations that resolve with
continued drug use. Such “adaptation,” the mechanism of which is
unknown, is well recognized for drugs such as isoniazid (INH), valpro­
ate, phenytoin, and HMG-CoA reductase inhibitors (statins). Extrahe­
patic manifestations of hypersensitivity, such as rash, arthralgias, fever,
leukocytosis, and eosinophilia, occur in a small fraction of patients
with idiosyncratic hepatotoxic drug reactions but are characteristic for
certain drugs (phenytoin, trimethoprim-sulfamethoxazole) and not
others. Both primary immunologic injury and direct hepatotoxicity
related to idiosyncratic differences in generation of toxic metabolites
have been invoked to explain idiosyncratic drug reactions. The most
current data implicate the adaptive immune system responding to the
formation of immune stimulatory compounds resulting from phase I
metabolic activation of the offending drug. Differences in host sus­
ceptibility may result from varying kinetics of toxic metabolite gen­
eration and genetic polymorphisms in downstream drug-metabolizing
Six Mechanisms of Liver Injury
A
Membrane
Hepatocyte
PART 10
Disorders of the Gastrointestinal System
F
Vesicle
Triglycerides
Free fatty
acid
D
E
Cell death
Other
caspases
Inhibition of
β-oxidation, respiration,
or both
Caspase
Caspase
Mitochondrion
Lactate
A. Rupture of cell membrane.
B. Injury of bile canaliculus (disruption of transport pumps).
C. P-450-drug covalent binding (drug adducts).
FIGURE 351-1  Potential mechanisms of drug-induced liver injury. The normal hepatocyte may be affected adversely by drugs through (A) disruption of intracellular calcium
homeostasis that leads to the disassembly of actin fibrils at the surface of the hepatocyte, resulting in blebbing of the cell membrane, rupture, and cell lysis; (B) disruption
of actin filaments next to the canaliculus (the specialized portion of the cell responsible for bile excretion), leading to loss of villous processes and interruption of transport
pumps such as multidrug resistance–associated protein 3 (MRP3), which, in turn, prevents the excretion of bilirubin and other organic compounds; (C) covalent binding of
the heme-containing cytochrome P450 enzyme to the drug, thus creating nonfunctioning adducts; (D) migration of these enzyme-drug adducts to the cell surface in vesicles
to serve as target immunogens for cytolytic attack by T cells, stimulating an immune response involving cytolytic T cells and cytokines; (E) activation of apoptotic pathways
by tumor necrosis factor α (TNF-α) receptor or Fas (DD denotes death domain), triggering the cascade of intercellular caspases, resulting in programmed cell death;
or (F) inhibition of mitochondrial function by a dual effect on both β-oxidation and the respiratory-chain enzymes, leading to failure of free fatty acid metabolism, a lack
of aerobic respiration, and accumulation of lactate and reactive oxygen species (which may disrupt mitochondrial DNA). Toxic metabolites excreted in bile may damage
bile-duct epithelium (not shown). CTLs, cytolytic T lymphocytes. (From WM Lee: Drug-induced hepatotoxicity. N Engl J Med 349:474, 2003. Copyright © 2003, Massachusetts
Medical Society. Reprinted with permission from Massachusetts Medical Society.)
B
Transport
pumps (MRP3)
Canaliculus
P-450
Heme Drug
C
Endoplasmic
reticulum
Enzyme-drug
adduct
Other
caspases
Cytokines
Caspase
DD
DD
DD
DD
Cytolytic
T cell
TNF-α receptor,
Fas
D. Drug adducts targeted by CTLs/cytokines.
E. Activation of apoptotic pathway by TNFα/Fas.
F. Inhibition of mitochondrial function.
TABLE 351-1  Some Features of Toxic and Drug-Induced Hepatic Injury
DIRECT TOXIC EFFECTa
IDIOSYNCRATICa
OTHERa
CARBON
TETRACHLORIDE
ACETAMINOPHEN
AMOXICILLINCLAVULANATE
ISONIAZID
CIPROFLOXACIN
FEATURES
Predictable and doserelated toxicity
+
+
Latent period
Short
Short
Delayed onset
Variable
May be short
Variable
Arthralgia, fever, rash,
eosinophilia
Liver morphology
Necrosis, fatty
infiltration
Centrilobular
necrosis
Mixed hepatocellular/
cholestatic
aThe drugs listed are typical examples.
pathways or cytokine activation; in addition, certain HLA haplotypes
have been associated with hepatotoxicity of certain drugs such as
amoxicillin-clavulanate and flucloxacillin. Occasionally, however, the
clinical features of an allergic reaction (e.g., prominent tissue eosino­
philia, autoantibodies) are difficult to ignore and suggest activation of
IgE pathways. A few instances of drug hepatotoxicity are observed to
be associated with autoantibodies, including a class of antibodies to
liver-kidney microsomes, anti-LKM2, directed against a cytochrome
P450 enzyme. Four agents that specifically have a phenotype of auto­
immune hepatitis with a high likelihood of positive antinuclear anti­
bodies (ANAs) include nitrofurantoin, minocycline, hydralazine, and
α-methyldopa.
Idiosyncratic reactions lead to a morphologic pattern that is more
variable than those produced by direct toxins; a single agent is often
capable of causing a variety of lesions, although certain patterns tend
to predominate. Depending on the agent involved, idiosyncratic
hepatitis may result in a clinical and morphologic picture indistin­
guishable from that of viral hepatitis (e.g., INH or ciprofloxacin).
So-called hepatocellular injury is the most common form, featuring
spotty necrosis in the liver lobule with a predominantly lymphocytic
infiltrate resembling that observed in acute hepatitis A, B, or C.
Drug-induced cholestasis ranges from mild to increasingly severe:
(1) bland cholestasis with limited hepatocellular injury (e.g., estro­
gens, 17,α-substituted androgens); (2) inflammatory cholestasis (e.g.,
amoxicillin-clavulanic acid [the most frequently implicated antibiotic
among cases of DILI], oxacillin, erythromycin estolate); (3) sclerosing
cholangitis (e.g., after intrahepatic infusion of the chemotherapeutic
agent floxuridine for hepatic metastases from a primary colonic car­
cinoma); and (4) disappearance of bile ducts, “ductopenic” cholestasis
or vanishing bile duct syndrome, similar to that observed in chronic
rejection (Chap. 357) following liver transplantation (e.g., carbam­
azepine, levofloxacin). More recently, a picture resembling primary
sclerosing cholangitis has been observed secondary to chemotherapy
and after long-term ketamine use. Cholestasis may result from bind­
ing of drugs to canalicular membrane transporters, accumulation of
toxic bile acids resulting from canalicular pump failure, or genetic
defects in canalicular transporter proteins. Clinically, the distinction
between a hepatocellular and a cholestatic reaction is indicated by
the R value, the ratio of alanine aminotransferase (ALT) to alkaline
phosphatase values, both expressed as multiples of the upper limit of
normal. An R value of >5.0 is associated with hepatocellular injury,
R <2.0 with cholestatic injury, and R between 2.0 and 5.0 with mixed
hepatocellular-cholestatic injury.
Morphologic alterations may also include hepatic granulomas (e.g.,
sulfonamides) or macrovesicular or microvesicular steatosis or steato­
hepatitis. Severe hepatotoxicity associated with steatohepatitis, most
likely a result of mitochondrial toxicity, was recognized with certain
antiretroviral therapies, although most of these drugs have been with­
drawn (Chap. 208). Another potential target for idiosyncratic drug
hepatotoxicity is sinusoidal lining cells; when these are injured, such as
by high-dose chemotherapeutic agents (e.g., cyclophosphamide, mel­
phalan, busulfan) administered prior to bone marrow transplantation,
venoocclusive disease can result. Nodular regenerative hyperplasia,
a subtle form of portal hypertension, may also result from vascular
ESTROGENS/
ANDROGENIC STEROIDS
Hepatocellular injury
resembling viral
hepatitis
Hepatocellular injury
resembling viral
hepatitis
Cholestasis without portal
inflammation
injury to portal or hepatic venous endothelium following systemic
chemotherapy, such as with oxaliplatin, as part of adjuvant treatment
for colon cancer.
Not all adverse hepatic drug reactions can be classified as either toxic
or idiosyncratic. For example, oral contraceptives, which combine estro­
genic and progestational compounds, may result in impairment of liver
tests and, occasionally, jaundice; however, they do not produce necrosis
or fatty change, manifestations of hypersensitivity are generally absent,
and susceptibility to the development of oral contraceptive–induced
cholestasis appears to be genetically determined. Such estrogen-induced
cholestasis is more common in women with cholestasis of pregnancy,
a disorder linked to genetic defects in multidrug resistance–associated
canalicular transporter proteins.
CHAPTER 351
Any idiosyncratic reaction that occurs in <1:10,000 recipients will
go unrecognized in most clinical trials, which involve at most several
thousand subjects. The U.S. Food and Drug Administration (FDA)
and pharmaceutical companies have learned to look for even subtle
indications of serious toxicity and monitor regularly the number of
trial subjects in whom any aminotransferase elevations develop, as
a possible surrogate for more serious toxicity. Even more valid as a
predictor of severe hepatotoxicity is the occurrence of jaundice in
patients enrolled in a clinical drug trial, so-called “Hy’s Law,” named
after Dr. Hyman Zimmerman, one of the pioneers of the field of
drug hepatotoxicity. He recognized that, if jaundice occurred dur­
ing a phase 3 trial, more serious liver injury was likely, with a 10:1
ratio between cases of jaundice and liver failure (i.e., 10 patients with
jaundice would result in 1 patient with acute liver failure). Thus, the
finding of such Hy’s Law (jaundiced) cases during drug development
often portends failure of approval, particularly if any of the subjects
sustains a bad outcome. Troglitazone, a peroxisome proliferator–activated
receptor γ agonist, was the first in its class of thiazolidinedione
insulin-sensitizing agents. Although in retrospect, Hy’s Law cases of
jaundice had occurred during phase 3 trials, no instances of liver fail­
ure were recognized until well after the drug was introduced, empha­
sizing the importance of postmarketing surveillance in identifying
toxic drugs and in leading to their withdrawal from use. Fortunately,
such hepatotoxicity is not characteristic of the second-generation
thiazolidinediones rosiglitazone and pioglitazone; in clinical trials,
the frequency of aminotransferase elevations in patients treated with
these medications did not differ from that in placebo recipients, and
isolated reports of liver injury among recipients are extremely rare.
Since troglitazone was withdrawn from the market in 2001, no fully
approved drugs have had to be withdrawn from the market by the
FDA. Several agents have received black box warnings indicating that
caution is needed; overall, the industry and FDA in concert have been
able to avert severe toxicity in approved agents over the past 25 years.
Toxic and Drug-Induced Hepatitis
Proving that an episode of liver injury is caused by a drug (causality) is
difficult in many cases. DILI is nearly always a presumptive diagnosis,
and many other disorders produce a similar clinicopathologic picture.
Thus, causality may be difficult to establish and requires several sepa­
rate supportive assessment variables to lead to a high level of certainty,
including temporal association (time of onset, time to resolution), clin­
ical-biochemical features, type of injury (hepatocellular vs cholestatic),
extrahepatic features, likelihood that a given agent is to blame based on
its past record, and exclusion of other potential causes. Scoring systems
such as the Roussel-Uclaf Causality Assessment Method (RUCAM)
yield residual uncertainty and have not been adopted widely. Cur­
rently, the U.S. Drug-Induced Liver Injury Network (DILIN) relies on a
structured expert opinion process requiring detailed data on each case
and a comprehensive review by three experts who arrive at a consensus
on a five-degree scale of likelihood (definite, highly likely, probable,
possible, unlikely); however, this approach is not practical for routine
clinical application. A new Revised Electronic Causality Assessment
Method (RECAM) has now been developed to improve on the other
two methods but, to date, is not yet in widespread use. The RECAM
website can be accessed via the link dilirecam.com.
Generally, drug hepatotoxicity is not more frequent in persons with
underlying chronic liver disease, although the severity of the outcome
may be amplified. Reported exceptions include hepatotoxicity of aspi­
rin, methotrexate, INH (only in certain experiences), antiretroviral
therapy for HIV infection, and certain drugs such as conditioning regi­
mens for bone marrow transplantation in the presence of hepatitis C.
TREATMENT
Toxic and Drug-Induced Hepatic Disease
Treatment is largely supportive, except in acetaminophen hepato­
toxicity (for which N-acetylcysteine is effective, see below). Acute
liver failure develops in 10% of patients with DILI; spontaneous
recovery, once that threshold is reached, occurs in <30%, and
liver transplantation is performed in >40% of those who reach the
level of severity of acute liver failure (coagulopathy and hepatic
encephalopathy) (Chap. 356). Withdrawal of the suspected agent
is indicated at the first sign of an adverse reaction or when ami­
notransferase levels reach five times the upper limit of normal.
A number of studies have suggested that lethal outcomes follow
continued use of an agent in the face of symptoms and signs of liver
injury. In the case of the direct toxins, liver involvement should not
divert attention from renal or other organ involvement, which may
also threaten survival. Agents used occasionally but of question­
able value include glucocorticoids for DILI with allergic features,
silibinin for mushroom poisoning, and ursodeoxycholic acid for
cholestatic drug hepatotoxicity. While these medications have been
shown to be effective and to have reasonable safety profiles, they
are of uncertain value. A double-blind, randomized controlled trial
of the use of N-acetylcysteine for nonacetaminophen acute liver
failure, including cases of DILI, demonstrated benefit, particularly
for patients with early-stage hepatic encephalopathy; however, the
drug has not been approved by the FDA for this indication, pending
“further studies,” which, however, are unlikely to be undertaken.
PART 10
Disorders of the Gastrointestinal System
In Table 351-2, several classes of chemical agents are listed together
with examples of the pattern of liver injury they produce. Certain drugs
appear to be responsible for the development of chronic as well as acute
hepatic injury. For example, nitrofurantoin, minocycline, hydralazine,
and methyldopa have been associated with moderate to severe chronic
hepatitis with autoimmune features; corticosteroids may be used and
can virtually always be discontinued after 6 months of therapy. Metho­
trexate, tamoxifen, and amiodarone have been implicated in the devel­
opment of cirrhosis. Portal hypertension in the absence of cirrhosis,
termed nodular regenerative hyperplasia, may result from alterations in
hepatic architecture produced by excessive intake of vitamin A or fol­
lowing chemotherapy with oxaliplatin. Oral contraceptives have been
implicated in the development of focal nodular hyperplasia or hepatic
adenoma (both benign lesions) and, rarely, hepatocellular carcinoma
and hepatic vein occlusion (Budd-Chiari syndrome). Another unusual
lesion, peliosis hepatis (blood cysts of the liver), has been observed in
some patients treated with anabolic or contraceptive steroids. The exis­
tence of these hepatic disorders expands the spectrum of liver injury
induced by chemical agents and emphasizes the need for a thorough
drug history in all patients with liver dysfunction. The comprehensive,
authoritative LiverTox website, which contains up-to-date information
on DILI, is available as a valuable reference through the National Insti­
tutes of Health and the National Library of Medicine (livertox.nih.gov).
The following are patterns of adverse hepatic reactions for some
prototypic agents.
■
■ACETAMINOPHEN HEPATOTOXICITY
(DIRECT TOXIN)
Acetaminophen represents the most prevalent cause of acute liver failure
in the Western world; up to 72% of patients with acetaminophen hepa­
totoxicity in Scandinavia—somewhat lower frequencies in the United
Kingdom and the United States—progress to encephalopathy and
coagulopathy. Acetaminophen causes dose-related centrilobular hepatic
necrosis after single-time-point ingestions, as intentional self-harm, or
over extended periods, as unintentional overdoses, when multiple drug
preparations or inappropriate drug amounts are used daily for several
days, for example, for relief of pain or fever. In these instances, 8 g/d,
twice the daily recommended maximum dose, over several days can
readily lead to liver failure. Use of opioid-acetaminophen combinations
appears to be particularly harmful, because habituation to the opioid
may occur with a gradual increase in opioid-acetaminophen combina­
tion dosing over days or weeks. A single dose of 10–15 g, occasionally
less, may produce clinical evidence of liver injury. Fatal fulminant
disease is usually (although not invariably) associated with ingestion of
≥25 g. Blood levels of acetaminophen correlate with severity of hepatic
injury (levels >300 μg/mL 4 h after ingestion are predictive of the devel­
opment of severe damage; levels <150 μg/mL suggest that hepatic injury
is highly unlikely). Nausea, vomiting, diarrhea, abdominal pain, and
shock are early manifestations occurring 4–12 h after ingestion. Then
24–48 h later, when these features are abating, hepatic injury becomes
apparent. Maximal abnormalities and hepatic failure are evident 3–5 days
after ingestion, and aminotransferase levels exceeding 10,000 IU/L are
not uncommon (i.e., levels far exceeding those in patients with viral
hepatitis). Renal failure and myocardial injury may be present. Whether
or not a clear history of overdose can be elicited, clinical suspicion of
acetaminophen hepatotoxicity should be raised by the presence of the
extremely high aminotransferase levels in association with low bilirubin
levels that are characteristic of this hyperacute injury. This biochemical
signature should trigger further questioning of the subject if possible;
however, outright denial (or denial of high doses) or altered mentation
may confound diagnostic efforts. In this setting, a presumptive diagno­
sis is reasonable, and the proven antidote, N-acetylcysteine, is both safe
and will be effective if given early (within 12 h) but is also used even
when injury has evolved.
Acetaminophen is metabolized predominantly by a phase II reaction
to innocuous sulfate and glucuronide metabolites; however, a small
proportion is metabolized by a phase I reaction to a hepatotoxic metab­
olite formed from the parent compound by cytochrome P450 CYP2E1.
This metabolite, N-acetyl-p-benzoquinone-imine (NAPQI), is detoxi­
fied by binding to “hepatoprotective” glutathione to become harmless,
water-soluble mercapturic acid, which undergoes renal excretion.
When excessive amounts of NAPQI are formed, or when glutathione
levels are low, glutathione levels are depleted and overwhelmed, per­
mitting covalent binding to nucleophilic hepatocyte macromolecules
forming acetaminophen-protein “adducts.” These adducts, which can
be measured in serum by high-performance liquid chromatography,
hold promise as diagnostic markers of acetaminophen hepatotoxicity,
and a point-of-care assay for acetaminophen-Cys adducts is under
development. The binding of acetaminophen to hepatocyte macromol­
ecules is believed to lead to hepatocyte necrosis; the precise sequence
and mechanism are unknown. Hepatic injury may be potentiated by
prior administration of alcohol, phenobarbital, INH, or other drugs;
by conditions that stimulate the mixed-function oxidase system; or
by conditions such as starvation (including inability to maintain oral
intake during severe febrile illnesses) that reduce hepatic glutathione
levels. Alcohol induces cytochrome P450 CYP2E1; consequently,
increased levels of the toxic metabolite NAPQI may be produced in
chronic alcoholics after acetaminophen ingestion, but the role of alco­
hol in potentiating acute acetaminophen injury is still debated. Alcohol
TABLE 351-2  Principal Alterations of Hepatic Morphology Produced by Some Commonly Used Drugs and Chemicalsa
PRINCIPAL MORPHOLOGIC
CHANGE
CLASS OF AGENT
EXAMPLE
Cholestasis
Anabolic steroid
Antibiotic
Anticonvulsant
Antidepressant
Anti-inflammatory
Antiplatelet
Antihypertensive
Antithyroid
Calcium channel blocker
Immunosuppressive
Lipid-lowering
Oncotherapeutic
Oral contraceptive
Oral hypoglycemic
Tranquilizer
Fatty liver
Antiarrhythmic
Antibiotic
Anticonvulsant
Antiviral
Oncotherapeutic
Hepatitis
Anesthetic
Antiandrogen
Antibiotic 
Anticonvulsant
Antidepressant
 
Antifungal
Antihypertensive
Anti-inflammatory
Antipsychotic
Antiviral
Calcium channel blocker
Cholinesterase inhibitor
Diuretic
Laxative
Norepinephrine reuptake inhibitor
Oral hypoglycemic 
Immune checkpoint inhibitor
Mixed hepatitis/cholestatic
Antibiotic
Antibacterial
Antifungal
Antihistamine
Immunosuppressive
Lipid-lowering
Toxic (necrosis)
Analgesic
Hydrocarbon
Metal
Mushroom
Solvent
Granulomas
Antiarrhythmic
Antibiotic
Anticonvulsant
Anti-inflammatory
Xanthine oxidase inhibitor
Vascular injury
Chemotherapeutic
Oxaliplatin, melphalan
aSeveral agents cause more than one type of liver lesion and appear under more than one category. bRarely associated with primary biliary cholangitis-like lesion.
cOccasionally associated with chronic hepatitis or bridging hepatic necrosis or cirrhosis. dAssociated with an autoimmune hepatitis-like syndrome. eWithdrawn from use
because of severe hepatotoxicity.
Methyl testosterone, many other body-building supplements
Erythromycin estolate, nitrofurantoin, rifampin, amoxicillin-clavulanic acid, oxacillin
Carbamazepine
Duloxetine, mirtazapine, tricyclic antidepressants
Sulindac
Clopidogrel
Irbesartan, fosinopril
Methimazole
Nifedipine, verapamil
Cyclosporine
Ezetimibe
Anabolic steroids, busulfan, tamoxifen, irinotecan, cytarabine, temozolomide
Norethynodrel with mestranol
Chlorpropamide
Chlorpromazineb
Amiodarone
Tetracycline (high-dose, IV)
Valproic acid
Dideoxynucleosides (e.g., zidovudine), protease inhibitors (e.g., indinavir, ritonavir)
Asparaginase, methotrexate, tamoxifen
Halothane, fluothane
Flutamide
Isoniazid,c rifampicin, nitrofurantoin, telithromycin, minocycline,d pyrazinamide, trovafloxacine
CHAPTER 351
Phenytoin, carbamazepine, valproic acid, phenobarbital
Iproniazid, amitriptyline, trazodone, venlafaxine, fluoxetine, paroxetine, duloxetine, sertraline,
nefazodonee
Ketoconazole, fluconazole, itraconazole
Methyldopa,c captopril, enalapril, lisinopril, losartan
Ibuprofen, indomethacin, diclofenac, sulindac, bromfenac
Risperidone
Zidovudine, didanosine, stavudine, nevirapine, ritonavir, indinavir, tipranavir, zalcitabine
Nifedipine, verapamil, diltiazem
Tacrine
Chlorothiazide
Oxyphenisatinc,e
Toxic and Drug-Induced Hepatitis
Atomoxetine
Troglitazone,e acarbose
Ipilimumab, pembrolizumab, nivolumab
Amoxicillin-clavulanic acid, trimethoprim-sulfamethoxazole
Clindamycin
Terbinafine
Cyproheptadine
Azathioprine
Nicotinic acid, lovastatin, ezetimibe
Acetaminophen
Carbon tetrachloride
Yellow phosphorus
Amanita phalloides
Dimethylformamide
Quinidine, diltiazem
Sulfonamides
Carbamazepine
Phenylbutazone
Allopurinol
also suppresses hepatic glutathione production. Therefore, in chronic
alcoholics, the toxic dose of acetaminophen may be as low as 2 g, and
alcoholic patients should be warned specifically about the dangers of
even standard doses of this commonly used drug. In a 2006 study, ami­
notransferase elevations were identified in 31–44% of normal subjects
treated for 14 days with the maximal recommended dose of acetamino­
phen, 4 g daily (administered alone or as part of an acetaminophenopioid combination); because these changes were transient and never
associated with bilirubin elevation, the clinical relevance of these find­
ings remains to be determined. Although underlying hepatitis C virus
(HCV) infection was found to be associated with an increased risk of
acute liver injury in patients hospitalized for acetaminophen overdose,
generally, in patients with nonalcoholic liver disease, acetaminophen
taken in recommended doses is well tolerated. Acetaminophen use in
cirrhotic patients has not been associated with hepatic decompensa­
tion. On the other hand, because of the link between acetaminophen
use and liver injury and because of the limited safety margin between
safe and toxic doses, the FDA has recommended that the daily dose of
acetaminophen be reduced from 4 g to 3 g (even lower for persons with
chronic alcohol use), that all acetaminophen-containing products be
labeled prominently as containing acetaminophen, and that the poten­
tial for liver injury be prominent in the packaging of acetaminophen
and acetaminophen-containing products. Within opioid combina­
tion products, the limit for the acetaminophen component has been
lowered to 325 mg per tablet. Adoption of this limit appears to have
had a salutary effect in decreasing the number of hospital admissions
and instances of liver failure associated with acetaminophen-opioid
combinations.
PART 10
Disorders of the Gastrointestinal System
TREATMENT
Acetaminophen Overdosage
Treatment includes gastric lavage, supportive measures, and oral
administration of activated charcoal or cholestyramine to prevent
absorption of residual drug. Neither charcoal nor cholestyramine
appears to be effective if given >30 min after acetaminophen inges­
tion; if they are used, the stomach lavage should be done before
other agents are administered orally. The chances of possible, prob­
able, and high-risk hepatotoxicity can be derived from a nomogram
plot, readily available in emergency departments, as a function of
measuring acetaminophen plasma levels 4–8 h after ingestion. In
patients with high acetaminophen blood levels (>200 μg/mL mea­
sured at 4 h or >100 μg/mL at 8 h after ingestion), the administra­
tion of N-acetylcysteine reduces markedly the severity of hepatic
necrosis. This agent provides sulfhydryl donor groups to replete
glutathione, which is required to render harmless toxic metabolites
that would otherwise bind covalently via sulfhydryl linkages to
cell proteins, resulting in the formation of drug metabolite-protein
adducts. Therapy should be begun within 8 h of ingestion but
may be at least partially effective when given as late as 24–36 h
after overdose. Routine use of N-acetylcysteine has substantially
reduced the occurrence of fatal acetaminophen hepatotoxicity.
N-acetylcysteine may be given orally but is more commonly used
as an IV solution, with a loading dose of 140 mg/kg over 1 h, fol­
lowed by 70 mg/kg every 4 h for 15–20 doses. Whenever a patient
with potential acetaminophen hepatotoxicity is encountered, a
local poison control center should be contacted. Treatment can be
stopped when plasma acetaminophen levels indicate that the risk
of liver damage is low. If signs of hepatic failure (e.g., progressive
jaundice, coagulopathy, confusion) occur despite N-acetylcysteine
therapy for acetaminophen hepatotoxicity, liver transplantation
may be the only option. Early arterial blood lactate levels among
such patients with acute liver failure may distinguish patients highly
likely to require liver transplantation (lactate levels >3.5 mmol/L)
from those likely to survive without liver replacement. Acute renal
injury occurs in nearly 75% of patients with severe acetaminophen
injury but is virtually always self-limited.
Survivors of acute acetaminophen overdose rarely, if ever, have
ongoing liver injury or sequela but may be subject to repeat overdosing.
■
■ISONIAZID HEPATOTOXICITY
(IDIOSYNCRATIC REACTION)
INH remains central to most antituberculous prophylactic and
therapeutic regimens, despite its long-standing recognition as a
hepatotoxin. In 10% of patients treated with INH, elevated serum
aminotransferase levels develop during the first few weeks of ther­
apy; however, these elevations in most cases are self-limited, are mild
(values for ALT <200 IU/L), and resolve despite continued drug use.
This adaptive response allows continuation of the agent if symptoms
and progressive enzyme elevations do not follow the initial eleva­
tions. Acute hepatocellular DILI secondary to INH is evident with
a variable latency period up to 6 months and is more frequent in
alcoholics and patients taking certain other medications, such as
barbiturates, rifampin, and pyrazinamide. If the clinical threshold of
encephalopathy is reached, severe hepatic injury is likely to be fatal
or to require liver transplantation. Liver biopsy reveals morpho­
logic changes similar to those of viral hepatitis or bridging hepatic
necrosis. Substantial liver injury appears to be age-related, increas­
ing substantially after age 35; the highest frequency is in patients
over age 50, and the lowest is in patients under the age of 20. Even
for patients >50 years of age monitored carefully during therapy,
hepatotoxicity occurs in only ~2%, well below the risk estimate
derived from earlier experiences. Fever, rash, eosinophilia, and other
manifestations of drug allergy are distinctly unusual. Antibodies to
INH have been detected in INH recipients, but a link to causality of
liver injury remains unclear. A clinical picture resembling chronic
hepatitis has been observed in a few patients. Many public health
programs that require INH prophylaxis for a positive tuberculin skin
test or blood test (Quantiferon or T-Spot) include monthly monitor­
ing of aminotransferase levels, although this practice has been called
into question. Even more effective in limiting serious outcomes may
be encouraging patients to be alert for symptoms such as nausea,
fatigue, or jaundice, because most fatalities occur in the setting of
continued INH use despite clinically apparent illness. The incidence
of severe INH toxicity may be declining as a result of less frequent
use and/or better management.
■
■SODIUM VALPROATE HEPATOTOXICITY
(TOXIC AND IDIOSYNCRATIC REACTION)
Sodium valproate, an anticonvulsant useful in the treatment of petit
mal and other seizure disorders, has been associated with the devel­
opment of severe hepatic toxicity and, rarely, fatalities, predominantly
in children but also in adults. Among children listed as candidates
for liver transplantation, valproate is the most common antiepileptic
drug implicated. Asymptomatic elevations of serum aminotransferase
levels have been recognized in as many as 45% of treated patients.
These “adaptive” changes, however, appear to have no clinical impor­
tance, because major hepatotoxicity is not seen in the majority of
patients despite continuation of drug therapy. In the rare patients
in whom jaundice, encephalopathy, and evidence of hepatic failure
are found, examination of liver tissue reveals microvesicular fat and
bridging hepatic necrosis, predominantly in the centrilobular zone.
Bile duct injury may also be apparent. Most likely, sodium valproate
is not directly hepatotoxic, but its metabolite, 4-pentenoic acid, may
be responsible for hepatic injury. Valproate hepatotoxicity is more
common in persons with mitochondrial enzyme deficiencies and may
be ameliorated by IV administration of carnitine, which valproate
therapy can deplete. Valproate toxicity has been linked to HLA hap­
lotypes (DR4 and B∗1502) and to mutations in mitochondrial DNA
polymerase gamma 1.
■
■NITROFURANTOIN HEPATOTOXICITY
(IDIOSYNCRATIC REACTION)
This commonly used antibiotic for urinary tract infections may cause
an acute hepatitis leading to fatal outcome or, more frequently, chronic
hepatitis of varying severity but indistinguishable from autoimmune
hepatitis. These two scenarios may reflect the frequent use and reuse of
the drug for treatment of recurrent cystitis in women. Although most
toxic agents manifest injury within 6 months of first ingestion, nitro­
furantoin may have a longer latency period, in part perhaps because of
its intermittent, recurrent use. Autoantibodies to nuclear components,
smooth muscle, and mitochondria are seen and may subside after reso­
lution of injury; however, glucocorticoid or other immunosuppressive
medication may be necessary to resolve the autoimmune injury, and
cirrhosis may be seen in cases that are not recognized quickly. Intersti­
tial pulmonary fibrosis presenting as chronic cough and dyspnea may
be present and resolve slowly with medication withdrawal. Histologic
findings are identical to those of autoimmune hepatitis. A similar dis­
ease pattern can be observed with minocycline, which is used repeat­
edly for the treatment of acne in teenagers, as well as with hydralazine
and α-methyldopa.
■
■AMOXICILLIN-CLAVULANATE HEPATOTOXICITY
(IDIOSYNCRATIC MIXED REACTION)
Currently, the most common agent implicated as causing DILI in the
United States and in Europe is amoxicillin-clavulanate (most frequent
brand name: Augmentin). This medication causes a very specific syn­
drome of mixed or primarily cholestatic injury. Because hepatotoxic­
ity may follow amoxicillin-clavulanate therapy after a relatively long
latency period, the liver injury may begin to manifest after the drug
has been withdrawn. The high prevalence of hepatotoxicity reflects
in part the very frequent use of this drug for respiratory tract infec­
tions, including community-acquired pneumonia. The mechanism of
hepatotoxicity is unclear, but the liver injury is thought to be caused
by amoxicillin toxicity that is potentiated in some way by clavulanate,
which itself appears not to be toxic. Symptoms include nausea, anorexia,
fatigue, and jaundice—which may be prolonged—with pruritus. Rash
is quite uncommon. On occasion, amoxicillin-clavulanate, like other
cholestatic hepatotoxic drugs, causes permanent injury to small bile
ducts, leading to the so-called “vanishing bile duct syndrome.” In van­
ishing bile duct syndrome, initially, liver injury is minimal except for
severe cholestasis; however, over time, histologic evidence of bile duct
abnormalities is replaced by a paucity and eventual absence of discern­
ible ducts on subsequent biopsies.
■
■AMIODARONE HEPATOTOXICITY
(TOXIC AND IDIOSYNCRATIC REACTION)
Therapy with this potent antiarrhythmic drug is accompanied in
15–50% of patients by modest elevations of serum aminotransferase
levels that may remain stable or diminish despite continuation of the
drug. Such abnormalities may appear days to many months after begin­
ning therapy. A proportion of those with elevated aminotransferase
levels have detectable hepatomegaly, and clinically important liver dis­
ease develops in <5% of patients. Features that represent a direct effect
of the drug on the liver and that are common to the majority of longterm recipients are ultrastructural phospholipidosis, unaccompanied
by clinical liver disease, and interference with hepatic mixed-function
oxidase metabolism of other drugs. The cationic amphiphilic drug and
its major metabolite desethylamiodarone accumulate in hepatocyte
lysosomes and mitochondria and in bile duct epithelium. The relatively
common elevations in aminotransferase levels are also considered a
predictable, dose-dependent, direct hepatotoxic effect. On the other
hand, in the rare patient with clinically apparent, symptomatic liver
disease, liver injury resembling that seen in alcoholic liver disease is
observed. The so-called pseudoalcoholic liver injury can range from
steatosis, to alcoholic hepatitis–like neutrophilic infiltration and Mal­
lory’s hyaline, to cirrhosis. Electron-microscopic demonstration of
phospholipid-laden lysosomal lamellar bodies can help to distinguish
amiodarone hepatotoxicity from typical alcoholic hepatitis. This cat­
egory of liver injury appears to be a metabolic idiosyncrasy that allows
hepatotoxic metabolites to be generated. Rarely, an acute idiosyncratic
hepatocellular injury resembling viral hepatitis or cholestatic hepatitis
occurs. Hepatic granulomas have occasionally been observed. Because
amiodarone has a long half-life, liver injury may persist for months
after the drug is stopped.
■
■ANABOLIC STEROIDS (CHOLESTATIC REACTION)
The most common form of liver injury caused by CAMs is the profound
cholestasis associated with anabolic steroids used by body builders.
Unregulated agents sold in gyms and health food stores as diet supple­
ments, which are taken by athletes to improve their performance, may
contain anabolic steroids. In a young male, jaundice that is accompanied
by a cholestatic, rather than a hepatitic, laboratory profile almost invari­
ably will turn out to be caused by the use of one of a variety of androgen
congeners. Such agents have the potential to injure bile transport pumps
and to cause intense cholestasis; the time to onset is variable, and reso­
lution, which is the rule, may require many weeks to months. Initially,
anorexia, nausea, and malaise may occur, followed by pruritus in some
but not all patients. Serum aminotransferase levels are usually <100 IU/L,
and serum alkaline phosphatase levels are generally moderately elevated
with bilirubin levels frequently exceeding 342 μmol/L (20 mg/dL). Exami­
nation of liver tissue reveals cholestasis without substantial inflammation
or necrosis. Anabolic steroids have also been used by prescription to
treat bone marrow failure. In this setting, hepatic centrizonal sinusoidal
dilatation and peliosis hepatis have been reported in rare patients, as
have hepatic adenomas and hepatocellular carcinoma. Recently, a large
series of cases with a uniform phenotype has been described. Unfor­
tunately, no genomic signature has become evident despite the unique
features of the injury. No permanent sequelae are evident besides pro­
longed jaundice, lasting frequently 10 weeks or more.
■
■TRIMETHOPRIM-SULFAMETHOXAZOLE
HEPATOTOXICITY (IDIOSYNCRATIC REACTION)
This antibiotic combination is used routinely for urinary tract infec­
tions in immunocompetent persons and for prophylaxis against and
therapy of Pneumocystis jirovecii pneumonia in immunosuppressed
persons (transplant recipients, patients with AIDS). With its increas­
ing use, its occasional hepatotoxicity is being recognized with grow­
ing frequency. Its likelihood is unpredictable, but when it occurs,
trimethoprim-sulfamethoxazole hepatotoxicity follows a relatively
uniform latency period of several weeks and is often accompanied by
eosinophilia, rash, and other features of a hypersensitivity reaction,
including the drug reaction with eosinophilia and systemic symptoms
(DRESS) syndrome. Biochemically and histologically, acute hepatocel­
lular necrosis predominates, but cholestatic features are quite frequent.
Occasionally, cholestasis without necrosis occurs, and very rarely, a
severe cholangiolytic pattern of liver injury is observed. In most cases,
liver injury is self-limited, but rare fatalities have been recorded. The
hepatotoxicity is attributable to the sulfamethoxazole component of the
drug and is similar in features to that seen with other sulfonamides;
tissue eosinophilia and granulomas may be seen. The risk of trime­
thoprim-sulfamethoxazole hepatotoxicity is increased in persons with
HIV infection. In a recent study, unique HLA associations in European
Americans and in African Americans have been identified.
CHAPTER 351
Toxic and Drug-Induced Hepatitis
■
■HMG-COA REDUCTASE INHIBITORS (STATINS)
(IDIOSYNCRATIC MIXED HEPATOCELLULAR AND
CHOLESTATIC REACTION)
Between 1 and 2% of patients taking lovastatin, simvastatin, pravas­
tatin, fluvastatin, or one of the newer statin drugs for the treatment of
hypercholesterolemia experience asymptomatic, reversible elevations
(greater than threefold) of aminotransferase activity. Acute hepatitislike histologic changes, centrilobular necrosis, and centrilobular cho­
lestasis have been described in a very small number of cases. In a
larger proportion, minor aminotransferase elevations appear during
the first several weeks of therapy. Careful laboratory monitoring can
distinguish between patients with minor, transitory changes, who
may continue therapy, and those with more profound and sustained
abnormalities, who should discontinue therapy. Because clinically
meaningful aminotransferase elevations are so rare after statin use and
do not differ in meta-analyses from the frequency of such laboratory
abnormalities in placebo recipients, a panel of liver experts recom­
mended to the National Lipid Association’s Safety Task Force that liver
test monitoring was not necessary in patients treated with statins and
that statin therapy need not be discontinued in patients found to have
asymptomatic isolated aminotransferase elevations during therapy.
Statin hepatotoxicity is not increased in patients with chronic hepatitis
C, hepatic steatosis, or other underlying liver diseases, and statins can
be used safely in these patients.
ALTERNATIVE AND COMPLEMENTARY
MEDICINES (IDIOSYNCRATIC
HEPATITIS, STEATOSIS)
Herbal medications that are of scientifically unproven efficacy and that
lack prospective safety oversight by regulatory agencies account cur­
rently for >20% of DILI in the United States. Besides anabolic steroids,
the most common category of dietary or herbal products is weight
loss agents. Included among the herbal remedies associated with toxic
hepatitis are Jin Bu Huan, xiao-chai-hu-tang, germander, chaparral,
senna, mistletoe, skullcap, gentian, comfrey (containing pyrrolizidine
alkaloids), ma huang, bee pollen, valerian root, pennyroyal oil, kava,
celandine, Impila (Callilepis laureola), LipoKinetix, Hydroxycut, Oxy­
Elite Pro, Herbalife, herbal nutritional supplements, and herbal teas
containing Camellia sinensis (green tea extract). Well characterized are
the acute hepatitis-like histologic lesions following Jin Bu Huan use:
focal hepatocellular necrosis, mixed mononuclear portal tract infiltra­
tion, coagulative necrosis, apoptotic hepatocyte degeneration, tissue
eosinophilia, and microvesicular steatosis. Megadoses of vitamin A can
injure the liver, as can pyrrolizidine alkaloids, which often contaminate
Chinese herbal preparations and can cause a venoocclusive injury lead­
ing to sinusoidal hepatic vein obstruction. Because some alternative
medicines induce toxicity via active metabolites, alcohol and drugs
that stimulate cytochrome P450 enzymes may enhance the toxicity of
some of these products. Conversely, some alternative medicines also
stimulate cytochrome P450 and may result in or amplify the toxic­
ity of recognized drug hepatotoxins. In many instances, herbal and
dietary supplements actually contain chemicals rather than only leaves,
roots, and bark. Antirheumatic “herbs” have been found to contain
a nonsteroidal anti-inflammatory drug (NSAID) such as diclofenac,
for example. Given the widespread use of such poorly defined herbal
preparations, hepatotoxicity is likely to be encountered with increasing
frequency; therefore, a drug history in patients with acute and chronic
liver disease should include use of “alternative medicines” and other
nonprescription preparations sold in so-called health food stores.
PART 10
Disorders of the Gastrointestinal System
■
■CHECKPOINT INHIBITOR AND OTHER
IMMUNOTHERAPIES FOR CANCER
The introduction of a new class of immunotherapeutic agents for
melanoma and other cancers has ushered in a new kind of hepato­
toxicity, that associated with activation of the immune response. The
three classes of immune-active molecules are cytotoxic T-lymphocyteassociated antigen-4 (CTLA-4), programmed cell death receptor 1
(PD-1), and programmed cell death receptor ligand 1 (PD-L1). Within
weeks of beginning treatment with any one of several agents, including
ipilimumab (CTLA-4), pembrolizumab (PD-1), or nivolumab (PD-1),
an active hepatitis evolves that is associated with positive ANAs and
appears to respond to glucocorticoid therapy. Liver histology may share
some of the features of, but does not resemble the chronic changes
observed in, autoimmune hepatitis. Instead, histologic findings are
compatible with a nonspecific acute hepatic injury, assumed to result
from the release of host modulation of anti-self-immune responses and
mediated predominantly by CD8+ lymphocytes. Immune-mediated
injury to thyroid, muscle, and colon may be seen as well. Few deaths
have been reported related to these immunotherapies; while these
novel agents may need to be halted temporarily, in many cases, they
can be restarted (and are tolerated better on retreatment) if patients are
showing a favorable antitumor response.
■
■HIGHLY ACTIVE ANTIRETROVIRAL THERAPY
FOR HIV INFECTION (MITOCHONDRIAL TOXIC,
IDIOSYNCRATIC, STEATOSIS; HEPATOCELLULAR,
CHOLESTATIC, AND MIXED)
The recognition of drug hepatotoxicity in persons with HIV infec­
tion is complicated in this population by the many alternative causes
of liver injury (chronic viral hepatitis, fatty infiltration, infiltrative
disorders, mycobacterial infection, etc.), but drug hepatotoxicity
associated with highly active antiretroviral therapy (HAART) was a
common type of liver injury in HIV-infected persons in the early days
of HIV therapy; however, it is less frequent now (Chap. 208). Impli­
cated most frequently are combinations including the nucleoside
analogue reverse transcriptase inhibitors zidovudine, didanosine,
and, to a lesser extent, stavudine; the protease inhibitors ritonavir
and indinavir (and amprenavir when used together with ritonavir),
as well as tipranavir; and the nonnucleoside reverse transcriptase
inhibitors nevirapine and, to a lesser extent, efavirenz. Distinguish­
ing the impact of HAART hepatotoxicity in patients with HIV and
hepatitis virus co-infection is made challenging by the following: (1)
both chronic hepatitis B and hepatitis C can affect the natural history
of HIV infection and the response to HAART, and (2) HAART can
have an impact on chronic viral hepatitis. For example, immunologic
reconstitution with HAART can result in immunologically mediated
liver-cell injury in patients with chronic hepatitis B co-infection if
treatment with an antiviral agent for hepatitis B (e.g., nucleoside ana­
logues such as tenofovir) is withdrawn. Infection with HIV, especially
with low CD4+ T-cell counts, has been reported to increase the rate
of hepatic fibrosis associated with chronic hepatitis C, and HAART
therapy can increase levels of serum aminotransferases and HCV
RNA in patients with hepatitis C co-infection. Didanosine or stavu­
dine should not be used with ribavirin in patients with HIV/HCV
co-infection because of an increased risk of severe mitochondrial
toxicity and lactic acidosis.
■
■FURTHER READING
Ahmad J et al: Sclerosing cholangitis-like changes on magnetic resonance
cholangiography in patients with drug-induced liver injury. Clin
Gastroenterol Hepatol 17:789, 2019.
Björnsson ES, Hoofnagle JL: Categorization of drugs implicated in
causing liver injury: Critical assessment based upon published case
reports. Hepatology 63:590, 2016.
Chalasani NP et al: ACG clinical guideline: Diagnosis and manage­
ment of idiosyncratic drug-induced liver injury. Am J Gastroenterol
116:878, 2021.
Chalasani N et al: Drug Induced Liver Injury Network. Clinical features,
outcomes, and HLA risk factors associated with nitrofurantoininduced liver injury. J Hepatol 78:293, 2023.
Cirulli ET et al: A missense variant in PTPN22 is a risk factor for
drug-induced liver injury. Gastroenterology 156:1707, 2019.
de Boer YS et al: Features of autoimmune hepatitis in patients
with drug-induced liver injury. Clin Gastroenterol Hepatol 15:103,
European Association for the Study of the Liver: EASL clinical
practice guidelines: Drug-induced liver injury. J Hepatol 70:1222,
2019.
Fontana RJ et al: AASLD practice guidance on drug, herbal, and
dietary supplement-induced liver injury. Hepatology 77:1036, 2023.
Fontana RJ et al: The evolving profile of idiosyncratic drug-induced
liver injury. Clin Gastroenterol Hepatol 21:2088, 2023.
Hayashi PH et al: RECAM: A revised electronic version of RUCAM
for diagnosis of drug induced liver injury. Hepatology 76:18,
Hoofnagle JH et al: HLA-B*35:01 and green tea induced liver injury.
Hepatology 73:2484, 2021.
Orandi BJ et al: Association of FDA mandate limiting acetaminophen
(paracetamol) in prescription combination opioid products and
subsequent hospitalizations and acute liver failure. JAMA 329:735,
Peeraphatdit TB et al: Hepatotoxicity from immune checkpoint
inhibitors: A systematic review and management recommendation.
Hepatology 72:315, 2020.
Stolz A et al: Severe and protracted cholestasis in 44 young men
taking bodybuilding supplements: Assessment of genetic, clinical and
chemical risk factors. Aliment Pharmacol Ther 49:1195, 2019.

24 - 352 Chronic Hepatitis
352 Chronic Hepatitis
Esperance A. K. Schaefer,
Raymond T. Chung, Jules L. Dienstag
Chronic Hepatitis
Chronic hepatitis represents a series of liver disorders of varying causes
and severity in which hepatic inflammation and necrosis continue
for at least 6 months. Milder forms are nonprogressive or only slowly
progressive, while more severe forms may be associated with scarring
and architectural reorganization, which, when advanced, lead ulti­
mately to cirrhosis. Several broad categories of chronic hepatitis have
been recognized. These include chronic viral hepatitis, drug-or toxininduced chronic hepatitis (Chap. 351), alcohol-associated liver
disease (Chap. 353), metabolic dysfunction-related chronic hepatitis
(Chap 354), and autoimmune chronic hepatitis. In many cases, clinical
and laboratory features are insufficient to allow assignment into one of
these distinct categories; these “idiopathic” cases are also believed to
represent autoimmune chronic hepatitis. Moreover, because of the high
prevalence of steatotic liver diseases (metabolic dysfunction-associated
steatotic liver disease and alcohol-associated liver disease), patients
with more than one, overlapping type of hepatitis are encountered not
infrequently (e.g., viral and steatotic liver injury, alcohol-related and
nonalcohol-related steatotic liver injury). Finally, clinical and labora­
tory features of chronic hepatitis are observed occasionally in patients
with other hereditary/metabolic disorders, such as Wilson disease
(copper overload) and α1 antitrypsin deficiency (Chaps. 355 and 427).
Although all types of chronic hepatitis share certain clinical, laboratory,
and histopathologic features, chronic viral and chronic autoimmune
hepatitis are sufficiently distinct to merit separate discussions. For dis­
cussion of acute hepatitis, see Chap. 350.
CLASSIFICATION OF CHRONIC HEPATITIS
Common to all forms of chronic hepatitis are histopathologic distinc­
tions based on localization and extent of liver injury. These vary from
the milder forms, previously labeled chronic persistent hepatitis and
chronic lobular hepatitis, to the more severe form, formerly called
chronic active hepatitis. When first defined, these designations were
believed to have prognostic implications, which were not corroborated
by subsequent observations. Categorization of chronic hepatitis based
primarily on histopathologic features has been replaced by a more
informative classification based on a combination of clinical, serologic,
and histologic variables. Classification of chronic hepatitis is based on
(1) its cause; (2) its histologic activity, or grade; and (3) its degree of
progression based on level of fibrosis, or stage. Thus, neither clinical
features alone nor histologic features—requiring liver biopsy or non­
invasive markers of fibrosis—alone are sufficient to characterize and
distinguish among the several categories of chronic hepatitis.
■
■CLASSIFICATION BY CAUSE
Clinical and serologic features allow the establishment of a diagnosis
of chronic viral hepatitis, caused by hepatitis B, hepatitis B plus D, or
hepatitis C; autoimmune hepatitis, including several subcategories,
I and II, based on serologic distinctions; drug-associated chronic hepatitis;
and a category of unknown cause, or cryptogenic chronic hepatitis
(Table 352-1). These are addressed in more detail below.
■
■CLASSIFICATION BY GRADE
Grade, a histologic assessment of necroinflammatory activity, is based
on examination of the liver biopsy. An assessment of important histo­
logic features includes the degree of periportal necrosis and the disrup­
tion of the limiting plate of periportal hepatocytes by inflammatory
cells (so-called piecemeal necrosis or interface hepatitis); the degree
of confluent necrosis that links or forms bridges between vascular
structures—between portal tract and portal tract or even more impor­
tant bridges between portal tract and central vein—referred to as bridging
necrosis; the degree of hepatocyte degeneration and focal necrosis
TABLE 352-1  Clinical and Laboratory Features of Chronic Hepatitis
TYPE OF HEPATITIS
DIAGNOSTIC TEST(S)
THERAPY
Chronic hepatitis B
HBsAg, IgG anti-HBc,
HBeAg, HBV DNA
IFN-α, PEG IFN-α
Oral agents:
First-line: entecavir, tenofovir 
Chronic hepatitis C
Anti-HCV, HCV RNA
Sofosbuvir-ledipasvir, glecaprevirpibrentasvir, sofosbuvirvelpatasvir, elbasvir-grazoprevir,
sofosbuvir-velpatasvir-voxilaprevira
Chronic hepatitis D
Anti-HDV, HDV RNA,
HBsAg, IgG anti-HBc
IFN-α, PEG IFN-αc, bulevirtideb
Autoimmune
hepatitis
ANA (homogeneous),
anti-SMA, anti-SLA,
anti-LKM1 (±)
Hyperglobulinemia
First line: prednisone, azathioprine
Second line: mycophenolate
mofetil, tacrolimus, rituximab
See www.hcvguidelines.org.
aDetailed treatment recommendations can be found at www.hcvguidelines.org.
bCurrently approved by European Medicines Agency, under review at the U.S. Food
and Drug Administration.
Abbreviations: HBc, hepatitis B core; HBeAg, hepatitis B e antigen; HBsAg, hepatitis
B surface antigen; HBV, hepatitis B virus; HCV, hepatitis C virus; HDV, hepatitis D
virus; IFN-α, interferon α; IgG, immunoglobulin G; LKM, liver-kidney microsome;
PEG IFN-α, pegylated interferon α; SLA, soluble liver antigen; SMA, smooth muscle
antibody.
within the lobule; and the degree of portal inflammation. Several scor­
ing systems for viral hepatitis that take these histologic features into
account have been devised, and the most popular are the histologic
activity index (HAI), used commonly in the United States, and the
METAVIR score, used in Europe (Table 352-2). Based on the presence
and degree of these features of histologic activity, chronic hepatitis can be
graded as mild, moderate, or severe.
CHAPTER 352
■
■CLASSIFICATION BY STAGE
The stage of chronic hepatitis, which reflects the level of progression
of the disease, is based on the degree of hepatic fibrosis. When fibro­
sis is so extensive that fibrous septa surround parenchymal nodules
and alter the normal architecture of the liver lobule, the histologic
lesion is defined as cirrhosis. Staging is based on the degree of fibrosis
as categorized on a numerical scale 0−6 (HAI) or 0−4 (METAVIR)
(Table 352-2). Several noninvasive approaches have been introduced
to provide approximations of hepatic histologic stage, including serum
biomarkers of fibrosis; fibrosis scores such as FIB-4, a validated algo­
rithm based on such routine lab tests as aspartate and alanine amino­
transferase (AST and ALT) levels and platelet counts (PLT) (age [years]
× AST divided by PLT ×); the Enhanced Liver Fibrosis (ELF) score, a
proprietary score (applied more frequently in steatotic liver disease
than viral liver disease) based on a combination of three nonroutine
blood tests (hyaluronic acid, procollagen III N-terminal peptide, and
tissue inhibitor of metalloproteinase 1); and imaging determinations of
liver elasticity/stiffness.
Chronic Hepatitis
CHRONIC VIRAL HEPATITIS
Both of the enterically transmitted forms of viral hepatitis, hepatitis A
and E, are self-limited and do not cause chronic hepatitis (rare reports
notwithstanding in which acute hepatitis A serves as a trigger for the
onset of autoimmune hepatitis in genetically susceptible patients or
in which hepatitis E [Chap. 350] can cause chronic liver disease in
immunosuppressed hosts, for example, after liver transplantation). In
contrast, the entire clinicopathologic spectrum of chronic hepatitis
occurs in patients with chronic viral hepatitis B and C as well as in
patients with chronic hepatitis D superimposed on chronic hepatitis B.
■
■CHRONIC HEPATITIS B
The likelihood of chronicity after acute hepatitis B varies as a function
of age. Infection at birth is associated with clinically silent acute infec­
tion but a 90% chance of chronic infection, whereas infection in young
adulthood in immunocompetent persons is typically associated with
clinically apparent acute hepatitis but a risk of chronicity of only ~1%.
Most cases of chronic hepatitis B among adults, however, are recognized
TABLE 352-2  Histologic Grading and Staging of Chronic Hepatitis
 
HISTOLOGIC ACTIVITY INDEX (HAI)a
METAVIRb
HISTOLOGIC FEATURE
SEVERITY
SCORE
SEVERITY
SCORE
Necroinflammatory Activity (grade)
Periportal necrosis,
including piecemeal
necrosis and/or bridging
necrosis (BN)
None
Mild
Mild/moderate
Moderate
Severe
None
Mild
Moderate
Severe
Bridging
necrosis
Intralobular necrosis
Confluent
—None
—Focal
—Zone 3 some
—Zone 3 most
—Zone 3 + BN few
—Zone 3 + BN multiple
—Panacinar/multiacinar
None or mild
Moderate
Severe
Focal
—None
—≤1 focus/10× field
—2–4 foci/10× field
—5–10 foci/10× field
—>10 foci/10× field
Portal inflammation
None
Mild
Moderate
Moderate/marked
Marked
Total
0–18
 
 
 
 
 
A0–A3c
PART 10
Disorders of the Gastrointestinal System
Fibrosis (stage)
None
Portal fibrosis—some
Portal fibrosis—most
Bridging fibrosis—few
Bridging fibrosis—many
Incomplete cirrhosis
Cirrhosis
 
 
 
 
 
 
 
Total
F0
F1
F1
F2
F3
F4
F4
aIshak K, Baptista A, Bianchi L, et al: Histologic grading and staging of chronic hepatitis. J Hepatol
22:696, 1995. bBedossa P, Poynard T, French METAVIR Cooperative Study Group: An algorithm for
grading activity in chronic hepatitis C. Hepatology 24:289, 1996. cNecroinflammatory grade: A0 = none;
A1 = mild; A2 = moderate; A3 = severe.
in patients who never had a recognized episode of clinically apparent
acute viral hepatitis. The degree of liver injury (grade) in patients with
chronic hepatitis B is variable, ranging from none in inactive carriers
to mild to moderate to severe. Among adults with chronic hepatitis B,
histologic features are of prognostic importance. In one long-term study
of 379 patients with untreated chronic hepatitis B in the United States,
investigators found a 5-year survival rate of 97% for patients with mild
chronic hepatitis, 86% for patients with moderate to severe chronic hepa­
titis, and only 55% for patients with chronic hepatitis and cirrhosis. The
15-year survival in these cohorts was 77%, 66%, and 40%, respectively.
On the other hand, more recent observations do not allow us to be so
sanguine about the prognosis in patients with mild chronic hepatitis;
among such patients followed for 1−13 years, progression to more severe
chronic hepatitis and cirrhosis has been observed in more than a quarter
of cases. Co-infection with HIV has long been appreciated to accelerate
the progression of hepatitis B; early studies documented a nearly 20-fold
increase in liver-related mortality between persons with hepatitis B virus
(HBV)-HIV co-infection compared to those with HBV infection alone.
Similarly, coexisting steatohepatitis has also been found to accelerate
disease progression; in one study, patients with HBV infection and
steatohepatitis were found to have both higher-grade inflammation and
more advanced-stage fibrosis.
More important to consider than histology alone in
patients with chronic hepatitis B is the degree of HBV
replication. As reviewed in Chap. 350, chronic HBV
infection can occur in the presence or absence of serum
hepatitis B e antigen (HBeAg), and generally, for both
HBeAg-reactive and HBeAg-negative chronic hepati­
tis B, the level of HBV DNA correlates with the level
of liver injury and risk of progression. In otherwise
healthy adults with HBeAg-positive chronic hepatitis B,
two phases have been recognized based on the relative
level of HBV replication. The relatively replicative phase
is characterized by the presence in the serum of HBeAg
and HBV DNA levels well in excess of 103−104 IU/mL,
sometimes exceeding 109 IU/mL; by the presence in
the liver of detectable intrahepatocyte nucleocapsid
antigens (primarily hepatitis B core antigen [HBcAg]);
by high infectivity; and by accompanying liver injury
(see below for an important exception during the early
decades of life). In contrast, the relatively nonreplicative
phase is characterized by the absence of the conventional
serum marker of HBV replication (HBeAg), the appear­
ance of anti-HBe, levels of HBV DNA below a threshold
of ~103 IU/mL, the absence of intrahepatocytic HBcAg,
limited infectivity, and minimal liver injury. Patients in
the relatively replicative phase tend to have more severe
chronic hepatitis and more rapid progression, whereas
those in the relatively nonreplicative phase tend to have
minimal or mild chronic hepatitis or to be inactive hep­
atitis B carriers. Thus, the transition between HBeAgpositive chronic hepatitis B to HBeAg-negative disease
is of clinical importance, signaling a slowing of disease
progression. In the past, the likelihood of spontaneous
HBeAg loss was estimated to be ~10% per year; how­
ever, a 2020 systematic review of 26 studies involving
7550 persons with HBeAg-positive chronic hepatitis B
revealed a spontaneous HBeAg clearance rate of 6.46
per 100 person-years. Such spontaneous HBeAg loss
was found to occur more frequently with older age and
in European populations but less frequently in patients
with genotype C.
Yes
No
Distinctions in HBV replication and in histologic cat­
egory, however, do not always coincide. In patients with
HBeAg-positive chronic HBV infection, especially when
acquired at birth or in early childhood, as recognized
commonly in Asian countries, a dichotomy is common
between very high levels of HBV replication during the
early decades of life (when the level of apparent host
immunologic tolerance of HBV is relatively high) and negligible
levels of liver injury; during this phase of chronic hepatitis B (labeled
by some as “HBeAg-positive chronic infection” or “immune tolerant,”
see Chap. 350, “Complications and Sequelae”), the level of viral replica­
tion does not correlate with liver injury or late complications. Yet despite
the relatively immediate, apparently benign nature of liver disease for
many decades in this population, in the middle decades, activation of
liver injury emerges as what appears to be the relative tolerance of the
host to HBV declines, and these patients with childhood-acquired HBV
infection are ultimately at increased risk later in life for cirrhosis, hepa­
tocellular carcinoma (HCC) (Chap. 87), and liver-related death; the link
between high-level HBV replication and these late liver complications
has been demonstrated convincingly in, and confined mostly to, persons
in their middle decades, especially age ≥40. A discussion of the patho­
genesis of liver injury in patients with chronic hepatitis B appears in
Chap. 350.
HBeAg-negative chronic hepatitis B (i.e., chronic HBV infection with
active virus replication and readily detectable HBV DNA but without
HBeAg [anti-HBe-reactive]) is more common than HBeAg-positive
chronic hepatitis B in Mediterranean and European countries and
in Asia. Compared to patients with HBeAg-reactive chronic hepa­
titis B, patients with HBeAg-negative chronic hepatitis B have HBV
DNA levels several orders of magnitude lower (usually no more than
105−106 IU/mL) than those observed in the HBeAg-reactive subset.
Most such HBeAg-negative cases represent precore or core-promoter
mutations acquired late in the natural history of mostly early-lifeonset disease; these mutations prevent translation of HBeAg from
the precore component of the HBV genome (precore mutants) or are
characterized by downregulated transcription of precore mRNA (corepromoter mutants; Chap. 350). Although their levels of HBV DNA
tend to be lower than among patients with HBeAg-reactive chronic
hepatitis B, patients with HBeAg-negative chronic hepatitis B can
have progressive liver injury (complicated by cirrhosis and HCC) and
experience episodic reactivation of liver disease reflected in fluctuat­
ing levels of aminotransferase activity (“flares”). The biochemical and
histologic activity of HBeAg-negative disease tends to correlate closely
with levels of HBV replication. Worth reiterating, the level of HBV
replication is the most important risk factor for the ultimate devel­
opment of cirrhosis and HCC in both HBeAg-reactive and HBeAgnegative patients. Although levels of HBV DNA are lower and more
readily suppressed by therapy to undetectable levels in HBeAg-negative
(compared to HBeAg-reactive) chronic hepatitis B, achieving sustained
responses that permit discontinuation of antiviral therapy is less likely
in HBeAg-negative patients (see below). Inactive carriers are patients
with circulating hepatitis B surface antigen (HBsAg), normal serum
aminotransferase levels, minimal or no histologic evidence of liver
injury, undetectable HBeAg, and levels of HBV DNA that are either
undetectable or present at a threshold of ≤103 IU/mL. This serologic
profile occurs not only in inactive carriers but also in patients with
HBeAg-negative chronic hepatitis B during periods of relative inactiv­
ity; distinguishing between the two requires sequential biochemical
and virologic monitoring over many months.
The spectrum of clinical features of chronic hepatitis B is broad,
ranging from asymptomatic infection to debilitating disease or even
end-stage, fatal hepatic failure. As noted above, the onset of the dis­
ease tends to be insidious in most patients, apart from the very few in
whom chronic disease follows failure of resolution of clinically appar­
ent acute hepatitis B. The clinical and laboratory features associated
with progression from acute to chronic hepatitis B are discussed in
Chap. 350.
Fatigue is a common symptom, and persistent or intermittent
jaundice is a common feature in severe or advanced cases. Intermittent
deepening of jaundice and recurrence of malaise and anorexia, as well
as worsening fatigue, are reminiscent of acute hepatitis; such exacer­
bations may occur spontaneously, often coinciding with evidence of
virologic reactivation; may lead to progressive liver injury; and, when
superimposed on well-established cirrhosis, may cause hepatic decom­
pensation. Complications of cirrhosis occur in end-stage chronic
hepatitis and include ascites, edema, bleeding gastroesophageal varices,
hepatic encephalopathy, coagulopathy, and hypersplenism. Occasion­
ally, these complications bring the patient to initial clinical attention.
Extrahepatic complications of chronic hepatitis B, similar to those seen
during the prodromal phase of acute hepatitis B, are associated with
tissue deposition of circulating hepatitis B antigen–antibody immune
complexes. These include arthralgias and arthritis, which are common,
and the rarer purpuric cutaneous lesions (leukocytoclastic vasculi­
tis), immune-complex glomerulonephritis, and generalized vasculitis
(polyarteritis nodosa) (Chap. 375).
Laboratory features of chronic hepatitis B do not always distinguish
adequately between histologically mild and severe hepatitis. Amino­
transferase elevations tend to be modest for chronic hepatitis B but
may fluctuate in the range of 100 to 1000 units. As is true for acute viral
hepatitis B, ALT tends to be more elevated than AST; however, once
cirrhosis is established, AST tends to exceed ALT. Levels of alkaline
phosphatase activity tend to be normal or only marginally elevated. In
severe cases, moderate elevations in serum bilirubin (51.3−171 μmol/L
[3−10 mg/dL]) occur. Hypoalbuminemia and prolongation of the
prothrombin time occur in severe or end-stage cases. Hyperglobulin­
emia and detectable circulating autoantibodies are distinctly absent in
chronic hepatitis B (in contrast to autoimmune hepatitis). Viral markers
of chronic HBV infection are discussed in Chap. 350.
TREATMENT
Chronic Hepatitis B
Although progression to cirrhosis is more likely in severe than in
mild or moderate chronic hepatitis B, all forms of chronic hepatitis B
can be progressive, and progression occurs primarily in patients
with active HBV replication. Moreover, in populations of patients
with chronic hepatitis B who are at risk for HCC (Chap. 87), the
risk is highest for those with continued, high-level HBV replica­
tion and lower for persons in whom initially high-level HBV DNA
falls spontaneously over time. Therefore, management of chronic
hepatitis B is directed at suppressing the level of virus replication.
Although clinical trials tend to focus on clinical endpoints achieved
over 1−2 years (e.g., suppression of HBV DNA to undetectable
levels, loss of HBeAg/HBsAg, improvement in histology, normal­
ization of ALT), these short-term gains translate into reductions
in the risk of clinical progression, hepatic decompensation, HCC,
liver transplantation, and death; regression of cirrhosis and of
esophageal varices has been documented to follow long-term phar­
macologic suppression of HBV replication. In addition, restoration
of impaired HBV-specific T-cell function has been shown following
successful suppression of HBV replication with antiviral therapy.
To date, eight drugs have been approved for treatment of chronic
hepatitis B: injectable interferon (IFN) α and pegylated interferon
(long-acting IFN bound to polyethylene glycol, PEG [PEG IFN])
and the oral agents lamivudine, adefovir dipivoxil, entecavir, telbi­
vudine, tenofovir disoproxil fumarate (TDF), and tenofovir alafen­
amide (TAF).
CHAPTER 352
Antiviral therapy for hepatitis B has evolved rapidly since the
mid-1990s, as has the sensitivity of tests for HBV DNA. When IFN
and the first oral antiviral lamivudine were evaluated in clinical
trials, HBV DNA was measured by insensitive hybridization assays
with detection thresholds of 105−106 virions/mL; when subsequent
treatments were studied in clinical trials, HBV DNA was measured
by sensitive amplification assays (polymerase chain reaction [PCR])
with detection thresholds of 101−103 viral copies/mL or IU/mL.
Recognition of these distinctions is helpful when comparing results
of clinical trials that established the efficacy of these therapies
(reviewed below). Of the eight approved treatments, PEG IFN,
entecavir, and the two tenofovir preparations (TDF and TAF) are
recommended as first-line agents, and generally, the oral agents are
favored over injectable PEG IFN.
Chronic Hepatitis
HISTORICAL, EARLY-GENERATION ANTIVIRAL AGENTS
Interferon  IFN-α was the first approved therapy (1992) for
chronic hepatitis B. Although it is no longer used to treat hepatitis B,
standard IFN is important historically, having provided important
lessons about antiviral therapy in general. For immunocompetent
adults with HBeAg-reactive chronic hepatitis B (who tend to have
high-level HBV DNA [>105−106 virions/mL] and histologic evi­
dence of chronic hepatitis on liver biopsy), a 16-week course of
subcutaneous IFN, 5 million units daily or 10 million units thrice
weekly, resulted in a loss of HBeAg and hybridization-detectable
HBV DNA (i.e., a reduction to levels below 105−106 virions/mL) in
~30% of patients, with a concomitant improvement in liver histol­
ogy. Seroconversion from HBeAg to anti-HBe occurred in ~20%,
and, in early trials, ~8% lost HBsAg. Successful IFN therapy and
seroconversion were often accompanied by an acute hepatitis-like
elevation in aminotransferase activity, postulated to result from
enhanced cytolytic T-cell clearance of HBV-infected hepatocytes.
Relapse after successful therapy was rare (1 or 2%). Responsive­
ness to IFN was higher in patients with low-level HBV DNA and
substantial ALT elevations. Therapy with IFN was not effective in
immunosuppressed persons, persons with neonatal acquisition of
infection and minimal-to-mild ALT elevations, or patients with
decompensated chronic hepatitis B (in whom such therapy was
actually detrimental, sometimes precipitating decompensation,
often associated with severe adverse effects). After HBeAg loss
during IFN therapy, 80% experienced eventual loss of HBsAg and
ALT normalization over the ensuing decade. In addition, improved
long-term and complication-free survival as well as a reduction
in the frequency of HCC were documented among IFN respond­
ers, supporting the conclusion that successful antiviral therapy
improves the natural history of chronic hepatitis B.
Brief-duration IFN therapy in patients with HBeAg-negative
chronic hepatitis B was disappointing, suppressing HBV replication
transiently during therapy but resulting only rarely in sustained
antiviral responses.
Complications of IFN therapy include systemic “flu-like” symp­
toms; marrow suppression; emotional lability (irritability, depres­
sion, anxiety); autoimmune reactions (especially autoimmune
thyroiditis); and miscellaneous side effects such as alopecia, rashes,
diarrhea, and numbness and tingling of the extremities. With the
TABLE 352-3  Comparison of Pegylated Interferon (PEG IFN), Lamivudine, Adefovir, Entecavir, Telbivudine, and Tenofovir Therapy for
Chronic Hepatitis Ba
FEATURE
PEG IFNb
LAMIVUDINE
ADEFOVIR
ENTECAVIR
TELBIVUDINE
TENOFOVIR (TDF)
TENOFOVIR (TAF)
Route of administration
Subcutaneous
injection (180 μg/
week)
Oral (100 mg/d)
Oral (10 mg/d)
Oral (0.5 mg/d)
Oral (600 mg/d)
Oral (300 mg/d)
Oral 25 mg/d)
Status
First-line
No longer
preferred
No longer
preferred
Duration of therapyc
48–52 weeks
≥52 weeks
≥48 weeks
≥48 weeks
≥52 weeks
≥48 weeks
48 weeks
Tolerability
Poorly tolerated
Well tolerated
Well tolerated;
creatinine
monitoring
recommended
PART 10
Disorders of the Gastrointestinal System
HBeAg seroconversion
  1 yr Rx 
  >1 yr Rx
 
18–20% 
NA
 
16–21% 
up to 50% at 5 yrs
 
12% 
43% at 3 yrsd
Log10 HBV DNA reduction
(mean copies/mL)
  HBeAg-reactive
  HBeAg-negative
  
 
4.5
4.1
  
 
5.5
4.4–4.7
  
 
Median 3.5–5
Median 3.5–3.9
HBV DNA PCR negative
(at then current PCR
sensitivitya) at end of yr 1
  HBeAg-reactive
  HBeAg-negative
 
  
 
10–25%
63%
 
  
 
36–44%
60–73%
 
 
  
13–21%
48–77%
ALT normalization at end
of yr 1
  HBeAg-reactive
  HBeAg-negative
 
 
39%
34–38%
 
 
41–75%
62–79%
 
 
48–61%
48–77%
HBsAg loss, yr 1
  >yr 1
3–4%
12% 5 yr after 1 yr
of Rx
≤1%
No data
0%
5% at yr 5
Histologic improvement
(≥2 point reduction in
HAI) at yr 1
  HBeAg-reactive 
  HBeAg-negative
 
 
 
38% 6 months after
48% 6 months after
 
 
 
49–62% 
61–66%
 
 
 
53–68% 
64%
Viral resistance
None
15–30% at 1 yr
70% at 5 yrs
None at 1 yr
29% at 5 yrs
Pregnancy category
C
Cf
C
C
B
B
B
aGenerally, these comparisons are based on data on each drug tested individually versus placebo in registration clinical trials; with rare exception, these comparisons are
not based on head-to-head testing of these drugs. In addition, the sensitivity of HBV DNA assays increased in sensitivity over the two decades between the introduction
of the earliest and latest of these approved drugs. Therefore, relative advantages and disadvantages should be interpreted cautiously. bAlthough standard interferon
α administered daily or three times a week is approved as therapy for chronic hepatitis B, it has been supplanted by PEG IFN, which is administered once a week and
is more effective. Standard interferon has no advantages over PEG IFN. cDuration of therapy in clinical efficacy trials; use in clinical practice may vary. dBecause of a
computer-generated randomization error that resulted in misallocation of drug versus placebo during the second year of clinical trial treatment, the frequency of HBeAg
seroconversion beyond the first year is an estimate (Kaplan-Meier analysis) based on the small subset in whom adefovir was administered correctly. e7% during a year of
therapy (43% at year 4) in lamivudine-resistant patients. fDespite its category C designation, lamivudine has an extensive pregnancy safety record in women with HIV/AIDS.
Abbreviations: ALT, alanine aminotransferase; HAI, histologic activity index; HBeAg, hepatitis B e antigen; HBsAg, hepatitis B surface antigen; HBV, hepatitis B virus; NA, not
applicable; PEG IFN, pegylated interferon; PCR, polymerase chain reaction; Rx, therapy; yr, year; TAF, tenofovir alafenamide; TDF, tenofovir disoproxil fumarate.
possible exception of autoimmune thyroiditis, all these side effects
are reversible upon dose lowering or cessation of therapy.
Although no longer competitive with the newer generation of
antivirals, IFN did represent the first successful antiviral approach,
set a standard against which to measure efficacy of subsequent
drugs, and demonstrated the benefit of antiviral therapy on the
natural history of chronic hepatitis B. Standard IFN has been sup­
planted by long-acting PEG IFN (see below).
Lamivudine  The first of the nucleoside analogues to be approved
(in 1998) for hepatitis B, the dideoxynucleoside lamivudine inhibits
reverse transcriptase activity of both HIV and HBV and is an effec­
tive agent for chronic hepatitis B; however, it is now superseded by
newer, more potent, less resistance-prone agents. For a summary
of its virologic, serologic, biochemical, and histologic efficacy, as
well as its resistance profile, please refer to Table 352-3. In clinical
First-line
No longer
preferred,
withdrawn
First-line
First-line
Well tolerated
Well tolerated
Well tolerated;
creatinine
monitoring
recommended
Well tolerated
 
21% 
31% at 2 yrs
44% at 6 yrs
 
22% 
30% at 2 yrs
 
21% 
40% at 5 yrs
 
10% (14% HBeAg loss)
18% at yr 2 (HBeAg
loss 22%)
  
 
6.9
5.0
  
 
6.4
5.2
  
 
6.2
4.6
Not reported in
clinical trials, likely
same as TDF
 
 
  
67% (91% at 4 yrs)
90%
 
 
  
60%
88%
 
 
  
76%
93%
 
 
  
64%
94%
 
 
68%
78%
 
 
77%
74%
 
 
68%
76%
 
 
72%
83%
2%
6% at yr 6
<1%
No data
3%
10% at yr 5
1%
1%
 
 
 
72% 
70%
 
 
 
65% 
67%
 
 
 
74% 
72%
Not included in
clinical trials
≤1% at 1 yre
Up to 5% at yr 1
Up to 22% at yr 2
0% at yr 1
0% through yr 8
0% at yr 1
0% through yr 2
1.2% at 6 yrse
trials, lamivudine therapy at daily doses of 100 mg for 48−52 weeks
suppressed HBV DNA, as measured by sensitive PCR amplifica­
tion assays, by a median of ~5.5 log10 copies/mL in HBeAg-positive
chronic hepatitis B and ~4.5 log10 copies/mL in HBeAg-negative
chronic hepatitis B (baseline HBV DNA levels are lower in HBeAgnegative than in HBeAg-positive chronic hepatitis B) and to
undetectable levels in ~40% and ~70%, respectively. Lamivudine
was shown to improve histology, retard hepatic fibrosis, prevent
progression to cirrhosis, and, in patients with advanced fibrosis,
reduce decompensation. Lamivudine was also effective in patients
unlikely to respond to IFN (e.g., those with high-level HBV DNA)
or in prior IFN nonresponders.
In HBeAg-positive chronic hepatitis B, lamivudine-associated
HBeAg seroconversion occurred in ~20%, as was true for IFN
therapy. Also similar to IFN therapy, patients with near-normal
ALT activity tended not to experience HBeAg responses (despite
suppression of HBV DNA), while patients with ALT levels ≥5× the
upper limit of normal could expect 1-year HBeAg seroconversion
rates of 50−60%. Generally, HBeAg seroconversions were confined
to patients who achieved suppression of HBV DNA to <104 copies/
mL (equivalent to ~103 IU/mL). Lamivudine-associated HBeAg
responses were accompanied by a delayed posttreatment HBsAg
seroconversion rate comparable to that seen after IFN. Among
Western patients who experienced HBeAg responses during a yearlong course of therapy and in whom the response was sustained for
4−6 months after cessation of therapy, the response was durable
thereafter in the vast majority (>80%); therefore, the achievement of
an HBeAg response represented a viable stopping point in therapy for
HBeAg-positive patients. Reduced durability, however, was reported
in Asian patients; therefore, to support the durability of HBeAg
responses, a period of consolidation therapy of ≥6 months in
Western patients and ≥1 year in Asian patients was recommended
after HBeAg seroconversion (see treatment guidelines below; a
full 12-month consolidation period is recommended currently for
treatment extension after oral agent–induced HBeAg seroconver­
sion). Close posttreatment monitoring was recommended to iden­
tify HBV reactivation promptly and to resume therapy. If HBeAg
was unaffected by lamivudine therapy, lamivudine was continued
until an HBeAg response occurred, but long-term therapy was
required to suppress HBV replication and, in turn, limit liver injury;
HBeAg seroconversions increased to a level of 50% after 5 years of
therapy. After a cumulative course of 3 years of lamivudine therapy,
necroinflammatory activity was reduced in the majority of patients,
and even cirrhosis was shown to regress to precirrhotic stages in as
many as three-quarters of patients.
Losses of HBsAg were few during the first year of lamivudine
therapy, and this observation had been cited as an advantage of IFNbased therapy over lamivudine therapy; however, in head-to-head
comparisons between standard IFN and lamivudine monotherapy,
HBsAg losses were rare in both groups. Trials in which lamivudine
and IFN were administered in combination failed to show a benefit
of combination therapy over lamivudine monotherapy for either
treatment-naïve patients or prior IFN nonresponders.
Patients with HBeAg-negative chronic hepatitis B (i.e., in those
with precore or core-promoter HBV mutations and who lack
HBeAg) cannot achieve an HBeAg response to nucleoside ana­
logue therapy—a stopping point in HBeAg-reactive patients; almost
invariably, when therapy was discontinued, reactivation was the
rule. Therefore, these patients required long-term lamivudine
therapy.
Clinical and laboratory side effects of lamivudine were negligible
and indistinguishable from those observed in placebo recipients;
however, lamivudine doses were reduced in patients with reduced
creatinine clearance. During lamivudine therapy, transient ALT
elevations, resembling those seen during IFN therapy and dur­
ing spontaneous HBeAg-to-anti-HBe seroconversions, occurred
in one-fourth of patients. These ALT elevations may result from
restored cytolytic T-cell activation permitted by suppression of
HBV replication. Similar ALT elevations, however, occurred at an
identical frequency in placebo recipients; however, ALT elevations
associated temporally with HBeAg seroconversion in clinical trials
were confined to lamivudine-treated patients. When therapy was
stopped after a year, two- to threefold ALT elevations occurred
in 20−30% of lamivudine-treated patients, representing renewed
liver-cell injury as HBV replication returned. Although these post­
treatment flares were almost always transient and mild, rare severe
exacerbations, especially in cirrhotic patients, were observed, man­
dating close and careful clinical and virologic monitoring after
discontinuation of treatment. Many authorities cautioned against
discontinuing therapy in patients with cirrhosis, in whom posttreat­
ment flares could precipitate decompensation.
Long-term monotherapy with lamivudine was associated with
methionine-to-valine
(M204V)
or
methionine-to-isoleucine
(M204I) mutations, primarily at amino acid 204 in the tyrosinemethionine-aspartate-aspartate (YMDD) motif of the C domain of
HBV DNA polymerase, analogous to mutations that occur in HIVinfected patients treated with this drug. During a year of therapy,
YMDD mutations occurred in 15−30% of patients; the frequency
increased with each year of therapy, reaching 70% at year 5. Ulti­
mately, patients with YMDD mutants experienced degradation of
clinical, biochemical, and histologic responses; therefore, if treat­
ment was begun with lamivudine monotherapy, the emergence of
lamivudine resistance, reflected clinically by a breakthrough from
suppressed levels of HBV DNA and ALT, was managed by adding
another antiviral to which YMDD variants are sensitive (e.g., adefo­
vir, tenofovir; see below).
Currently, lamivudine has been eclipsed by more potent antivi­
rals that have superior resistance profiles (see below); it is no longer
recommended as first-line therapy. Still, as the first successful oral
antiviral agent for use in hepatitis B, lamivudine provided proof of
principle that polymerase inhibitors can achieve virologic, sero­
logic, biochemical, and histologic benefits, including retardation
and reversal of fibrosis and even of cirrhosis.
CHAPTER 352
Chronic Hepatitis
Cirrhosis 
In addition, lamivudine was shown to be effective
in the treatment of patients with decompensated hepatitis B (for
whom IFN is contraindicated), in some of whom decompensa­
tion can be reversed. Moreover, among patients with cirrhosis or
advanced fibrosis, lamivudine was shown to be effective in reducing
the risk of progression to hepatic decompensation and, based on
subsequent population studies, the risk of HCC. In the half decade
following the introduction in the United States of lamivudine
therapy for hepatitis B, referral of patients with HBV-associated
end-stage liver disease for liver transplantation fell by ~30%, sup­
porting further the beneficial impact of oral antiviral therapy on the
natural history of chronic hepatitis B.
HIV Co-infection 
Because lamivudine monotherapy in persons
with HIV infection can result universally in the rapid emergence of
YMDD variants, testing for HIV infection was recommended for
all patients with chronic hepatitis B prior to lamivudine therapy; if
HIV infection was identified, lamivudine monotherapy at the HBV
daily dose of 100 mg was contraindicated. These patients require
treatment for both HIV and HBV with an HIV drug regimen that
includes or is supplemented by at least two drugs active against
HBV; antiretroviral therapy (ART) often contains two drugs with
antiviral activity against HBV (e.g., tenofovir and emtricitabine),
but if lamivudine was part of the regimen, the 300-mg daily dose
was required (Chap. 208). The safety of lamivudine during preg­
nancy has not been established; however, the drug is not terato­
genic in rodents and has been used safely in pregnant women with
HIV infection and with HBV infection. As shown for subsequent
nucleoside analogues, administration of lamivudine during the last
months of pregnancy to mothers with high-level hepatitis B viremia
reduced the likelihood of perinatal transmission of hepatitis B.
Adefovir Dipivoxil  At an oral daily dose of 10 mg, the acyclic
nucleotide analogue adefovir dipivoxil, the prodrug of adefovir
(approved for hepatitis B in 2002), reduces HBV DNA by ~3.5−4 log10
copies/mL; i.e., it is less potent than lamivudine or any of the newer
antiviral agents. For a summary of its virologic, serologic, biochem­
ical, and histologic efficacy, as well as its resistance profile, please
refer to Table 352-3. Like IFN and lamivudine, adefovir dipivoxil
is more likely to achieve an HBeAg response in patients with high
baseline ALT; HBeAg responses to it are highly durable and can be
relied upon as a treatment stopping point, after a period of consoli­
dation therapy; and biochemical, serologic, and virologic outcomes
improve over time with continued therapy.
In HBeAg-negative chronic hepatitis B, as was true for lamivudine,
because HBeAg responses—a potential stopping point—cannot
be achieved, reactivation is the rule when adefovir therapy is dis­
continued, and indefinite, long-term therapy is required. Reported
attempts to stop adefovir after 5 years were followed by a period
of maintained suppression of HBV DNA and ALT; however, most
such patients had persistent hepatitis B viremia, and most HBeAgnegative patients were treated indefinitely unless HBsAg loss, albeit
very rare, was achieved.
Adefovir was less prone to resistance than lamivudine, and thus
the primary contribution of adefovir, its effectiveness in lamivudineresistant, YMDD-mutant HBV, led to its adoption for lamivudineresistant hepatitis B. When lamivudine resistance occurred, adding
adefovir (i.e., maintaining lamivudine to preempt the emergence of
adefovir resistance) was superior to switching to adefovir.
An advantage of adefovir was its relatively favorable resistance
profile; however, it was not as potent as the other approved oral
agents, it did not suppress HBV DNA as rapidly or as uniformly
as the others, it was the least likely of all agents to result in HBeAg
seroconversion, and 20−50% of patients failed to suppress HBV
DNA by 2 log10 (“primary nonresponders”). For these reasons,
adefovir, which has been supplanted in both treatment-naïve and
lamivudine-resistant patients by the more potent, less resistanceprone tenofovir (see below), is no longer recommended as first-line
therapy.
PART 10
Disorders of the Gastrointestinal System
Telbivudine  Telbivudine, a cytosine analogue (approved in 2006),
is similar in efficacy to entecavir (see below) but slightly less potent
in suppressing HBV DNA (median 6.4 log10 reduction in HBeAgreactive disease and a 5.2 log10 reduction in HBeAg-negative
disease). In its registration trial, telbivudine at an oral daily dose
of 600 mg suppressed HBV DNA to <300 copies/mL in 60% of
HBeAg-positive and 88% of HBeAg-negative patients, reduced ALT
to normal in 77% of HBeAg-positive and 74% of HBeAg-negative
patients, and improved histology in 65% of HBeAg-positive and
67% of HBeAg-negative patients. Although resistance to telbivu­
dine (M204I, not M204V, mutations) was less frequent than resis­
tance to lamivudine at the end of 1 year, resistance mutations after
2 years of treatment occurred in up to 22%. Generally well tolerated,
telbivudine was associated with a low frequency of asymptom­
atic creatine kinase elevations and with a very low frequency of
myopathy and peripheral neuropathy; frequency of administration
had to be reduced for patients with impaired creatinine clearance.
Its excellent potency notwithstanding, the inferior resistance and
safety profile of telbivudine limited its appeal; telbivudine is neither
recommended as first-line therapy nor used widely.
CURRENT FIRST-LINE TREATMENT OPTIONS
PEGYLATED IFN  After long-acting PEG IFN was shown to be
effective in the treatment of hepatitis C (see below), this more con­
venient IFN preparation was evaluated in the treatment of chronic
hepatitis B. Once-a-week PEG IFN is more effective than the more
frequently administered, standard IFN, and several large-scale trials
of PEG IFN versus oral lamivudine were conducted in patients with
chronic hepatitis B.
In HBeAg-reactive chronic hepatitis B, two large-scale studies
were done. In one study, PEG IFN-α2b (100 μg weekly for
32 weeks, then 50 μg weekly for another 20 weeks for a total of
52 weeks) was evaluated against a comparison arm of combina­
tion PEG IFN with oral lamivudine in 307 subjects. The other
study involved PEG IFN-α2a (180 μg weekly for 48 weeks) in 814
primarily Asian patients, three-fourths of whom had ALT ≥2×
the upper limit of normal, with comparison arms of lamivudine
monotherapy and combination PEG IFN plus lamivudine. At the
end of therapy (48−52 weeks) in the PEG IFN monotherapy arms,
HBeAg loss occurred in ~30%, HBeAg seroconversion in 22−27%,
undetectable HBV DNA (<400 copies/mL by PCR) in 10−25%,
and normal ALT in 34−39%, and a mean reduction in HBV DNA
of 2 log10 copies/mL (PEG IFN-α2b) to 4.5 log10 copies/mL (PEG
IFN-α2a) was seen. Six months after completing PEG IFN mono­
therapy in these trials, HBeAg losses were present in ~35%, HBeAg
seroconversion in ~30%, undetectable HBV DNA in 7−14%, and
normal ALT in 32−41%, and the mean reduction in HBV DNA
was 2−2.4 log10 copies/mL. Although the combination of PEG IFN
and lamivudine was superior at the end of therapy in one or more
serologic, virologic, or biochemical outcomes, neither the combina­
tion arm (in both studies) nor the lamivudine monotherapy arm
(in the PEG IFN-α2a trial) demonstrated any benefit compared to
the PEG IFN monotherapy arms 6 months after therapy. Moreover,
HBsAg seroconversion occurred in 3−7% of PEG IFN recipients
(with or without lamivudine); some of these seroconversions were
identified by the end of therapy, but many were identified during
the posttreatment follow-up period. The likelihood of HBeAg loss
in PEG IFN–treated HBeAg-reactive patients was associated with
HBV genotype A > B > C > D (shown for PEG IFN-α2b but not for
PEG IFN-α2a). PEG IFN-α2a was approved in the United States for
hepatitis B in 2005; PEG IFN-α2b, which is not approved for hepa­
titis B in the United States, is used in other countries.
Based on these results, some authorities concluded that PEG IFN
monotherapy should be the first-line therapy of choice in HBeAgreactive chronic hepatitis B; however, this conclusion has been
challenged. Although a finite, 1-year course of PEG IFN results
in a higher rate of sustained response (6 months after treatment)
than is achieved with oral nucleoside/nucleotide analogue therapy,
the comparison is confounded by the fact that oral agents are not
discontinued at the end of 1 year. Instead, taken orally and free of
side effects, therapy with oral agents is extended indefinitely or until
after the occurrence of an HBeAg response. The rate of HBeAg
responses after 2 years of oral-agent nucleoside analogue therapy
is at least as high as, if not higher than, that achieved with PEG
IFN after 1 year; favoring oral agents is the absence of injections,
difficult-to-tolerate side effects, and laboratory monitoring as well
as lower direct and indirect medical care costs and inconvenience.
HBsAg loss with PEG IFN therapy occurs in such a small propor­
tion of patients that subjecting everyone to PEG IFN for the mar­
ginal gain of HBsAg responses during or immediately after therapy
in such a very small minority is questionable. Moreover, with the
newer, more potent nucleoside analogues, the frequency of HBsAg
loss during the first year of therapy nearly equals that of PEG IFN.
Of course, resistance is not an issue during PEG IFN therapy, but
the risk of resistance is much lower with new agents (≤1% up to
3−8 years in previously treatment-naïve, entecavir-treated patients
and 0% in tenofovir-treated patients; see below). Finally, the level of
HBV DNA inhibition that can be achieved with the newer agents,
and even with lamivudine, exceeds that achieved with PEG IFN, in
some cases by several orders of magnitude.
In HBeAg-negative chronic hepatitis B, the role of PEG IFN is
uncertain. One trial of PEG IFN-α2a (180 μg weekly for 48 weeks vs
comparison arms of lamivudine monotherapy and of combination
therapy) in 564 patients showed that PEG IFN monotherapy resulted
at the end of therapy in suppression of HBV DNA by a mean of 4.1
log10 copies/mL, undetectable HBV DNA (<400 copies/mL by PCR)
in 63%, normal ALT in 38%, and loss of HBsAg in 4%. Although
lamivudine monotherapy and combination lamivudine−PEG IFN
therapy were both superior to PEG IFN at the end of therapy, no
advantage of lamivudine monotherapy or combination therapy was
apparent over PEG IFN monotherapy 6 months after therapy—
suppression of HBV DNA by a mean of 2.3 log10 copies/mL, unde­
tectable HBV DNA in 19%, and normal ALT in 59%. In patients
involved in this trial followed for up to 5 years, among the twothirds followed who had been treated initially with PEG IFN, 17%
maintained HBV DNA suppression to <400 copies/mL, but ALT
remained normal in only 22%; HBsAg loss increased gradually to
12%. As was the case for standard IFN therapy in HBeAg-negative
patients, only a small proportion maintained responsiveness after
completion of PEG IFN therapy, raising questions about the rela­
tive value of a finite period of PEG IFN versus a longer course with
a potent, low-resistance oral nucleoside analogue in these patients.
Moreover, the value of PEG IFN for HBeAg-negative chronic
hepatitis B has not been confirmed. In the only other controlled
clinical trial of PEG IFN for HBeAg-negative chronic hepatitis B,
the hepatitis C regimen of PEG IFN plus ribavirin was compared to
PEG IFN monotherapy. In this trial, HBV DNA suppression (<400
copies/mL) occurred in only 7.5% of the two groups combined, and
no study subject lost HBsAg.
Reductions in quantitative HBsAg levels have been shown to
correlate with and to be predictive of responsiveness to PEG IFN
in chronic hepatitis B. If HBsAg levels fail to fall within the first
12–24 weeks or to reach <20,000 IU/mL by week 24, PEG IFN
therapy is unlikely to be effective and should be discontinued.
(Similar observations of HBsAg levels in oral agent–treated patients
are of interest but of limited clinical relevance, given the very high
likelihood of virologic responses during such therapy.) While PEG
IFN remains one of the recommended first-line agents for hepatitis
B, subsequent-generation, injection-free, very-well-tolerated, highbarrier-to-resistance, oral agents are used much more widely.
Entecavir  Entecavir, an oral cyclopentyl guanosine analogue
polymerase inhibitor (approved in 2005), is a highly potent HBV
antiviral and is just as well tolerated as lamivudine. In HBeAg-positive
patients, one 709-subject clinical trial compared oral entecavir,
0.5 mg daily, with lamivudine, 100 mg daily. At 48 weeks, entecavir
was superior to lamivudine in suppression of HBV DNA (mean
6.9 vs 5.5 log10 copies/mL), percentage with undetectable HBV
DNA (<300 copies/mL by PCR; 67 vs 36%), histologic improve­
ment (≥2-point improvement in necroinflammatory HAI score;
72 vs 62%), and normal ALT (68 vs 60%). The two treatments were
indistinguishable in percentage with HBeAg loss (22 vs 20%) and
seroconversion (21 vs 18%). Among patients treated with entecavir
for 96 weeks, HBV DNA was undetectable cumulatively in 80% (vs
39% for lamivudine), and HBeAg seroconversions had occurred in
31% (vs 26% for lamivudine). After 3–6 years of entecavir, HBeAg
seroconversions were observed in 39–44% and HBsAg loss in 5–6%.
Similarly, in a 638-subject clinical trial among HBeAg-negative
patients, at week 48, oral entecavir, 0.5 mg daily, was superior to
lamivudine, 100 mg daily, in suppression of HBV DNA (mean 5.0
vs 4.5 log10 copies/mL) and in percentage with undetectable HBV
DNA (90 vs 72%), histologic improvement (70 vs 61%), and normal
ALT (78 vs 71%). No resistance mutations were encountered in pre­
viously treatment-naïve, entecavir-treated patients during 96 weeks
of therapy, and in a cohort of subjects treated for up to 6 years, resis­
tance emerged in only 1.2%. Entecavir-induced HBeAg seroconver­
sions are as durable as those achieved with other antivirals. Its high
barrier to resistance coupled with its high potency renders entecavir
a first-line drug for patients with chronic hepatitis B.
Entecavir is also effective against lamivudine-resistant HBV
infection. In a trial of 286 lamivudine-resistant patients, enteca­
vir, at a higher daily dose of 1 mg, was superior to lamivudine,
as measured at week 48, in achieving suppression of HBV DNA
(mean 5.1 vs 0.48 log10 copies/mL), undetectable HBV DNA (72
vs 19%), normal ALT (61 vs 15%), HBeAg loss (10 vs 3%), and
HBeAg seroconversion (8 vs 3%). In this population of lamivudineexperienced patients, however, entecavir resistance emerged in 7%
at 48 weeks. Although entecavir resistance requires both a YMDD
mutation and a second mutation at one of several other sites (e.g.,
T184A, S202G/I, or M250V), resistance to entecavir in lamivudineresistant chronic hepatitis B was reported to increase progressively
to 43% at 4 years and 57% at 6 years; therefore, entecavir is not as
attractive a choice (and is not recommended, despite its approval
for this indication) as adefovir was or as tenofovir is for patients
with lamivudine-resistant hepatitis B.
In clinical trials, entecavir had an excellent safety profile. In
addition, on-treatment and posttreatment ALT flares are relatively
uncommon and relatively mild in entecavir-treated patients. Doses
should be reduced for patients with reduced creatinine clearance.
Entecavir does have low-level antiviral activity against HIV and,
therefore, cannot be used as monotherapy to treat HBV infection in
HIV/HBV co-infected persons.
Tenofovir  Tenofovir disoproxil fumarate (TDF), an acyclic nucle­
otide analogue and potent antiretroviral agent used to treat HIV
infection (approved for hepatitis B in 2008), is similar to adefovir
but more potent in suppressing HBV DNA and inducing HBeAg
responses; it is highly active against both wild-type and lamivudineresistant HBV and active in patients whose response to adefovir
is slow and/or limited. At an oral once-daily dose of 300 mg for
48 weeks, tenofovir suppressed HBV DNA by 6.2 log10 (to undetect­
able levels [<400 copies/mL] in 76%) in HBeAg-positive patients
and by 4.6 log10 (to undetectable levels in 93%) in HBeAg-negative
patients; reduced ALT to normal in 68% of HBeAg-positive and
76% of HBeAg-negative patients; and improved histology in 74% of
HBeAg-positive and 72% of HBeAg-negative patients. In HBeAgpositive patients, HBeAg seroconversions occurred in 21% by the
end of year 1, 27% by year 2, 34% by year 3, and 40% by year 5 of
tenofovir treatment; HBsAg loss occurred in 3% by the end of
year 1, 6% at year 2, and 10% by year 5. After 5 years of tenofovir
therapy, 87% of patients experienced histologic improvement,
including reduction in fibrosis score (51%) and regression of cir­
rhosis (74%). The 5-year safety (negligible renal toxicity, in 1%, and
mild reduction in bone density, in ~0.5%) and resistance profiles
(none recorded through 8 years) of tenofovir are very favorable as
well; therefore, tenofovir has supplanted adefovir both as first-line
therapy for chronic hepatitis B and as rescue therapy for lamivu­
dine-resistant chronic hepatitis B. Studies of tenofovir and entecavir
reviewed in 2015 showed no difference in long-term risks of renal
and bone toxicity; however, among patients treated with tenofovir,
instances of acute renal failure and of low blood phosphate levels
have been reported. Thus, in patients receiving tenofovir, monitor­
ing bone density is not recommended, but periodic (at least annual)
monitoring for renal injury is recommended (serum creatinine
and phosphate, urine glucose and protein). Frequency of tenofovir
administration should be reduced for patients with impaired cre­
atinine clearance.
CHAPTER 352
Chronic Hepatitis
Tenofovir alafenamide (TAF), a second-generation tenofovir
(approved for hepatitis B in 2016), is a prodrug of tenofovir
that requires activation to tenofovir in hepatocytes. This targeted
delivery to hepatocytes allows a lower dose to suffice and reduces
systemic exposure by 90%, thereby minimizing TDF-associated
proximal tubular renal injury, its associated phosphate wasting, and
the potential consequent loss of bone mineral density. The dose of
TAF is 25 mg, which is equivalent in antiviral potency to 300 mg
of TDF; both formulations have the same high barrier to resis­
tance, and clinical resistance has not been encountered. Random­
ized, controlled, double-blind, phase 3 noninferiority trials, one in
HBeAg-positive patients and the other in HBeAg-negative patients,
provided the safety and efficacy data to support TAF approval.
In 873 HBeAg-positive patients treated for 48 weeks, TAF versus
TDF achieved (1) HBV DNA reductions to <29 IU/mL in 64%
versus 67%; (2) ALT normalization in 72% versus 67% (an unex­
plained TAF biochemical advantage confirmed in other trials); (3)
HBeAg loss in 14% versus 12%; (4) HBeAg seroconversion in 10%
versus 8%; and (5) a negligible loss of HBsAg in 1% versus 0.3%.
Compared to TDF, TAF was associated with reduced impairment of
renal function (median reduction in estimated glomerular filtration
rate of –0.6 mL/min for TAF vs –5.4 mL/min for TDF) and of bone
density (in hip measurements, mean reduction of –0.10% for TAF
vs –1.72% for TDF; adjusted difference, 1.62%).
In the parallel trial among 426 HBeAg-negative patients treated
for 48 weeks, reductions in HBV DNA to <29 IU/mL occurred in
94% versus 93% of individuals treated with TAF versus TDF, respec­
tively; normalization of ALT occurred in 83% versus 75%, but no
HBsAg loss occurred in either group. Similar TAF advantages in
maintaining renal function and bone density were reported: reduc­
tion in median estimated glomerular filtration rate (–1.8 mL/min
for TAF vs –4.8 mL/min for TDF) and in median bone density (in
hip measurements, mean reduction of –0.29% for TAF vs –2.16%
for TDF; adjusted percentage difference, 1.87%).
At week 96, TAF and TDF HBV DNA and ALT reductions
(including the TAF advantage observed at 48 weeks) were main­
tained. In the original TDF group, when TDF was switched to TAF
after week 96, all differences observed during the first 96 weeks (in
normalization of ALT and reductions in renal function and bone
density) had resolved at week 120. Resistance did not emerge to
either TAF or TDF throughout the trial.
Based on these trial outcomes, TAF joined the list of recom­
mended first-line antiviral agents for chronic hepatitis B. This drug
is recommended over TDF by the American Association for the
Study of Liver Diseases (AASLD) and the European Association
for the Study of the Liver (EASL) for patients with reduced renal
function (creatinine clearance <50 mL/min), reduced bone density,
and risk factors for renal injury (including, according to EASL
guidelines, decompensated cirrhosis, creatinine clearance <60 mL/
min, poorly controlled hypertension or diabetes, proteinuria, active
glomerulonephritis, concomitant nephrotoxic medications, or
solid-organ transplantation); the EASL recommendation extends to
persons >60 years, who are at increased risk of TDF nephrotoxicity.
In patients with creatinine clearances <15 mL/min, neither TDF
nor TAF is recommended.
PART 10
Disorders of the Gastrointestinal System
A comparison of antiviral therapies for chronic hepatitis B
appears in Table 352-3; their relative potencies in suppressing
HBV DNA are shown in Fig. 352-1.
Combination Therapies  Adding a nucleos(t)ide inhibitor to PEG
IFN does not appear to improve the frequency of HBeAg response
or other treatment endpoints. Although the combination of lamivu­
dine and PEG IFN suppresses HBV DNA more profoundly during
therapy than does monotherapy with either drug alone (and is
much less likely to be associated with lamivudine resistance), this
combination used for a year is no better than a year of PEG IFN
–1
–2
Log10 HBV DNA
–3
–3.5
–4
–4.5
–5
–5.5
–6
–6.2
–6.4
–6.9
–7
ADV
PEG IFN
LAM
TDF
TBV
ETV
FIGURE 352-1  Relative potency of antiviral drugs for hepatitis B, as reflected by
median log10 hepatitis B virus (HBV) DNA reduction in HBeAg-positive chronic
hepatitis B. These data are from individual reports of large, randomized controlled
registration trials that were the basis for approval of the drugs. In most instances,
these data do not represent direct comparisons among the drugs, because study
populations were different, baseline patient variables were not always uniform,
and the sensitivity and dynamic range of the HBV DNA assays used in the trials
varied. ADV, adefovir dipivoxil; ETV, entecavir; LAM, lamivudine; PEG IFN, pegylated
interferon α2a; TBV, telbivudine; TDF, tenofovir disoproxil fumarate. Because
of potency and a high barrier to resistance, ETV and tenofovir (either TDF or the
second-generation tenofovir alafenamide) are recommended as first-line therapy.
While PEG IFN remains a first-line agent, the oral agents developed earlier, LAM,
ADV, and TBV, are no longer preferred agents.
in achieving sustained responses. To date, combinations of oral
nucleoside/nucleotide agents have not achieved an enhancement
in virologic, serologic, or biochemical efficacy over that achieved
by the more potent of the combined drugs given individually. In a
2-year trial of combination entecavir and tenofovir versus entecavir
monotherapy, for a small subgroup of patients with very high HBV
DNA levels (≥108 IU/mL), a reduction in HBV DNA to <50 IU/mL
was higher in the combination group (79 vs 62%); however, no dif­
ferences in HBeAg responses or any other endpoint were observed
between the combination-therapy and monotherapy groups, even
in the high-HBV DNA subgroup.
In the setting of resistant virus, combinations of nucleos(t)ide
inhibitors can be effective. For resistance to lamivudine or adefo­
vir, adding a second, non-cross-resistant agent was previously the
chosen approach. Whereas, initially, in clinical studies of adefovir
as rescue therapy for lamivudine resistance, adding adefovir to
lamivudine (combination therapy) was considered a better strategy
than replacing lamivudine with adefovir monotherapy (to minimize
ALT flares and to avoid adefovir resistance), newer antiviral thera­
pies have altered the paradigm. According to current treatment
recommendations of the AASLD and the EASL, switching from the
resistant drug to the new drug is preferred. Because the current
generation of antivirals is so potent and has such a high barrier to
resistance, monotherapy with the rescue drug (e.g., tenofovir for
lamivudine resistance) is as effective (as demonstrated in observa­
tional reports for up to 5 years) in maintaining viral suppression
without the emergence of resistance as combination therapy with
the resistant drug and the rescue drug. Generally, in patients treated
initially with entecavir and tenofovir preparations, antiviral drug
resistance is no longer encountered. For currently rare patients
who already have acquired multidrug resistance (to both nucleo­
side analogues [lamivudine, entecavir, telbivudine] and nucleotide
analogues [adefovir, tenofovir]), treatment with a combination of
entecavir and tenofovir has been shown to be highly effective in
suppressing HBV DNA and overcoming drug resistance.
In addition to the eight approved antiviral drugs for hepatitis B,
emtricitabine, a fluorinated cytosine analogue very similar to lami­
vudine in structure, efficacy, and resistance profile, offers no advan­
tage over lamivudine. A combination of emtricitabine and tenofovir
is approved for the treatment of HIV infection and is an appealing
combination therapy for hepatitis B, especially for lamivudineresistant disease; however, neither emtricitabine nor the combina­
tion is approved for hepatitis B. Worth considering, however, in
patients with chronic hepatitis B who are at high risk of human
immunodeficiency virus (HIV) exposure, the combination of TDF
and emtricitabine (FTC) may have a unique but complicated role.
Although clinical trial data are limited, combined TDF/FTC is used
often as preexposure prophylaxis (PrEP) to prevent HIV infec­
tion, with a 94% reduction in new HIV infection. Thus, in patients
with chronic hepatitis B who require treatment for their hepatitis
B but who also desire PrEP, TDF/FTC may be an appealing treat­
ment option. Adherence is essential, however, because virologic
and clinical reactivation of hepatitis B is common if TDF/FTC is
discontinued.
Novel Antivirals and Strategies  The current generation of welltolerated oral antivirals has been very successful in the management
of chronic hepatitis B and in altering its natural history favorably;
however, most patients require long-duration, usually indefi­
nite, therapy. Therefore, the quest continues for novel antiviral
approaches that can achieve “cure” but are finite in duration. The
cure of HBV infection is defined as either “functional cure,” with
loss of HBsAg and undetectable DNA (that are durable for at least
6 months after treatment), with or without the emergence of antiHBs, or as “complete (sterilizing) cure,” in which covalently closed
circular DNA (cccDNA) is eradicated as well. Even an approach to
achieving “functional cure” with finite-duration therapy, which is
elusive in patients treated with nucleoside analogues, would be wel­
come. Currently, innovative approaches being investigated focus on
viral-targeting strategies or immunomodulatory strategies. Current
virologic approaches in clinical testing include HBV entry blockers,
therapies to block viral RNA translation, and HBV capsid assembly
modulators. Bulevirtide, an entry inhibitor blocking the receptor
for HBV on cell membranes, sodium taurocholate co-transporting
polypeptide (NTCP), that decreases new cccDNA production, has
been approved in the European Union for treatment of HBV/HDV
co-infection (see HDV, below) but has not met with success in treat­
ing HBV mono-infection. Several therapies targeting viral RNA
are under investigation. These include RNA interference (RNAi)
therapies and antisense oligonucleotides (ASO). Experimental
RNAi therapies tend to reduce HBsAg by ~2 log10 IU/mL during
therapy, but ultimately, HBsAg reappears over time after therapy.
An early signal of efficacy was reported for the ASO bepirovirsen,
which achieved functional cure in ~10% of patients treated with
six monthly injections (as monotherapy or in combination with
nucleoside analogues); however, the durability of this response was
degraded over ensuing follow-up, yielding a very small residual
rate of functional cure. Drugs to inhibit capsid assembly have the
potential to modulate cccDNA synthesis and contribute to func­
tional cure; however, while this class of drugs reduces HBV DNA
by ~2–3 log10 IU/mL, HBsAg levels remain unaffected. Moreover, at
least one experimental drug in each of these categories of virologic
inhibitors had to be abandoned for toxicity. Immunomodulators
being studied have included toll-like receptor agonists (TLR7,
TLR8), immune checkpoint inhibitors (programmed cell death 1
[PD-1] and programmed cell death ligand 1 [PD-L1] blockade),
T-cell vaccines, and broadly neutralizing antibody approaches.
While some immunomodulatory strategies have shown promise,
such as one small phase 2 clinical trial in which treatment with the
TLR8 agonist selgantolimod resulted in HBsAg loss in a small num­
ber, most immunomodulatory approaches have been abandoned
for lack of efficacy or for toxicity (e.g., upregulation of RNA sen­
sor retinoic acid-inducible gene I [RIG-I]). Even after more than a
decade of early clinical trials, none of these treatments has yet been
TABLE 352-4  Recommendations for Treatment of Chronic Hepatitis Ba
HBeAg STATUS
CLINICAL
HBV DNA (IU/mL)
ALT
RECOMMENDATION
HBeAg-reactive
b
2 × 104
≤2 × ULNc
   
 
   
 
   
 
Chronic hepatitis
Cirrhosis:
Cirrhosis compensated 
2 × 104d 
2 × ULNd 
2 × 103
< or > ULN
ULN
< or > ULN
< or > ULN
<2 × 103
Cirrhosis
decompensated
Detectable
Undetectable
HBeAg-negative
b
≤2 × 103
≤ULN
1 to >2 × ULNd
Chronic hepatitis
   
 
2 × 103
   
 
  
 
Chronic hepatitis
Cirrhosis:
Cirrhosis compensated 
2 × 103 
2 × ULNd 
2 × 103
< or > ULN
ULN
< or > ULN
< or > ULN
<2 × 103
Cirrhosis
decompensated
Detectable
Undetectable
aBased on practice guidelines of the American Association for the Study of Liver Diseases (AASLD). Except as indicated in footnotes, these guidelines are similar to those
issued by the European Association for the Study of the Liver (EASL). bLiver disease tends to be mild or inactive clinically; most such patients do not undergo liver biopsy.
cAccording to the EASL guidelines, treat if HBV DNA is >2 × 103 IU/mL and ALT >ULN. dOne of the potent oral drugs with a high barrier to resistance (entecavir or tenofovir)
or PEG IFN can be used as first-line therapy (see text for further detail). eBecause HBeAg seroconversion is not an option, the goal of therapy is to suppress HBV DNA and
maintain a normal ALT. PEG IFN is administered by subcutaneous injection weekly for a year; caution is warranted in relying on a 6-month posttreatment interval to define
a sustained response, because the majority of such responses are lost thereafter. Oral agents, entecavir or tenofovir, are administered daily, usually indefinitely or until,
as very rarely occurs, virologic and biochemical responses are accompanied by HBsAg seroconversion. fFor older patients and those with advanced fibrosis, consider
lowering the HBV DNA threshold to >2 × 103 IU/mL.
Abbreviations: ALT, alanine aminotransferase; HBeAg, hepatitis B e antigen; HBsAg, hepatitis B surface antigen; HBV, hepatitis B virus; PEG IFN, pegylated interferon;
ULN, upper limit of normal.
shown to “cure” hepatitis B, and none is likely to be competitive,
unless it can be shown to go beyond current antivirals in achieving
recovery (sustained HBsAg loss) from HBV infection. Whether
combinations of immune modulators and direct antiviral agents
will improve prospects for curative therapies remains to be seen.
TREATMENT RECOMMENDATIONS
Several learned societies and groups of expert physicians have
issued treatment recommendations for patients with chronic hepa­
titis B; the most authoritative and updated are those of the AASLD
and the EASL. Although the recommendations differ slightly, a con­
sensus has emerged on most of the important points (Table 352-4).
No treatment is recommended or available for inactive “nonrepli­
cative” hepatitis B carriers (undetectable HBeAg with normal ALT
and HBV DNA ≤103 IU/mL documented serially over time). Nor is
treatment recommended currently for HBeAg-positive chronic hep­
atitis B with normal ALT (labeled by some as reflecting a relatively
“immune-tolerant” phase but more accurately a period of dissocia­
tion between high-level HBV replication and a paradoxical paucity
of inflammatory liver injury; see Chap. 350, “Complications and
Sequelae”); however, such patients merit careful monitoring with
HBV DNA and ALT at least every 6 months. Conversely, treatment
is universally recommended for any patient with compensated
cirrhosis and detectable HBV DNA; in decompensated cirrhosis,
treatment should be carefully considered. Treatment considerations
otherwise are based on the presence or absence of HBeAg, HBV
DNA levels and presence or absence of active necroinflammatory
activity (as based on ALT or liver biopsy).
CHAPTER 352
HBeAg Positive  In patients with detectable HBeAg and HBV
DNA levels >2 × 104 IU/mL, treatment is recommended by the
AASLD for those with ALT levels >2× the upper limit of normal.
(The EASL recommends treatment in HBeAg-positive patients
for HBV DNA levels >2 × 103 IU/mL and ALT above the upper
limit of normal.) For HBeAg-positive patients with ALT ≤2× the
upper limit of normal, in whom sustained responses are not likely
Chronic Hepatitis
No treatment; monitor, except in patients >40, with family history of cirrhosis or
hepatocellular carcinoma, with extrahepatic manifestations, with a history of
previous treatment, and/or with liver biopsy (or noninvasive fibrosis determination)
evidence for moderate to severe inflammation or fibrosis
Treatd 
Treatd with oral agents, not PEG IFN
Treatd with oral agents, not PEG IFN; refer for liver transplantation
Observe; refer for liver transplantation
Inactive carrier; treatment not necessary
No treatment; monitor, except in patients >40, with family history of cirrhosis or
hepatocellular carcinoma, with extrahepatic manifestations, with a history of
previous treatment, and/or with liver biopsy (or noninvasive fibrosis determination)
evidence for moderate to severe inflammation or fibrosis
Treate,f 
Treatd with oral agents, not PEG IFN
Treatment suggestedf
Treatd with oral agentsg, not PEG IFN; refer for liver transplantation
Observe; refer for liver transplantation
and who would require multiyear therapy, antiviral therapy is not
recommended currently. As noted above, these untreated patients
(as well as patients with HBV DNA between 2 × 103 and 2 ×
104 IU/mL) merit continued monitoring with HBV DNA and
ALT every 3–6 months. Even in these groups (in the “gray zone”),
however, therapy would be considered for patients >40 years of
age, with extrahepatic manifestations of HBV infection, or with a
family history of cirrhosis or HCC; patients whose liver biopsy or
noninvasive testing shows moderate to severe necroinflammatory
activity or fibrosis; or patients with a history of previous treatment.
In HBeAg-positive patients, treatment may be discontinued after
HBeAg loss. In a 2016 systematic review of 1716 patients involved
in 25 clinical trials, responses after oral-agent therapy were found
to be durable. Among patients with HBeAg-reactive chronic hepa­
titis B, the pooled rates of durable HBeAg seroconversions main­
tained after cessation of nucleoside/nucleotide analogue therapy
(including all the oral agents) were 92 and 88% at posttreatment
months 12 and 24, respectively, unaffected by the duration of postHBeAg-response consolidation therapy (>6 months in all studies
evaluated); the pooled rate of durable biochemical remission after
therapy in this population was 76%. Thus, per current AASLD
recommendations, antiviral treatment with oral agents can be
stopped after HBeAg loss/seroconversion in noncirrhotics, and the
suggested period of consolidation therapy is 12 months with close
monitoring for recurrent viremia (monthly × 6, then every 3 months
for the rest of a year) after cessation of therapy.
HBeAg Negative  For patients with HBeAg-negative chronic
hepatitis B, antiviral therapy is recommended if ALT is >2× the
upper limit of normal (above the upper limit of normal according
to EASL) and HBV DNA is >2 × 103 IU/mL. If HBV DNA is >2 ×
103 IU/mL and ALT is 1 to >2× the upper limit of normal, the same
considerations apply as for HBeAg-positive patients with border­
line ALT levels (the “gray zone”) for patients >40 years of age, with
extrahepatic manifestations of HBV infection, or with a family his­
tory of cirrhosis or HCC; patients whose liver biopsy or noninvasive
testing shows moderate to severe necroinflammatory activity or
fibrosis; or patients with a history of previous treatment. This group
also merits continued careful monitoring with HBV DNA and ALT
every 3–6 months. For patients with HBeAg-negative chronic hepa­
titis, the current recommendation with oral agents is for indefinite
therapy; stopping therapy in this group can be considered after
HBsAg loss, which remains exceedingly rare.
PART 10
Disorders of the Gastrointestinal System
The potential for stopping antiviral therapy in noncirrhotic
HBeAg-negative patients after protracted (≥2–5 years) antiviral
therapy has been the subject of several studies After such pro­
longed courses of entecavir or tenofovir, in one small, 57-patient
study (DARING-B), 18-month virologic relapse rates (HBV DNA
2000 IU/mL) were high (in 72%); however, 25% of study subjects
underwent HBsAg loss after 18 months of follow-up. In a similar
study (FINITE), virologic relapse rates were high, but 62% did not
meet criteria for retreatment, and 19% lost HBsAg. In contrast, in
a study among Asian patients, only ~30% had sustained responses
for which resumption of therapy was not introduced, and HBsAg
responses were negligible. In the only randomized, controlled trial
of stopping therapy versus continuing therapy in HBeAg-negative
patients after prolonged antiviral therapy (Toronto STOP study),
only 33% had sustained responses after cessation of therapy, and
HBsAg loss occurred with equal, small frequencies in both the
stop-treatment group (4%) and the continue-treatment group (5%).
Most recently, in a large 1552-subject, international cohort study
(RETRACT B), treatment cessation in HBeAg-negative patients
without cirrhosis was followed by virologic relapse rates of 68.9%
at 12 months and 83.4% at 48 months. Relapse requiring retreat­
ment occurred in 29.8% of subjects at 12 months and 54.7% at
48 months. Similar to prior studies, HBsAg loss was observed in
3.2% of subjects at 12 months and increased to 13% at 48 months.
Predictors of HBsAg loss were identified, which may offer guidance
to determine which patients may benefit from stopping nucleos(t)
ide analogue therapy. In this study, the most important predictor
of HBsAg loss was quantitative HBsAg level at the time therapy
was discontinued (at posttreatment year 4, in Caucasian patients
with HBsAg <1000 IU/mL, 41% had lost HBsAg, and in Asian
patients with HBsAg <100 IU/mL, 33% had lost HBsAg). At pres­
ent, however, quantitative HBsAg determination is not routinely
available in clinical care. Generally, then, although HBsAg loss can
be achieved in a small fraction, and although a subgroup may not
require reintroduction of therapy in the short run, enthusiasm for
this approach is limited, and for most HBeAg-negative patients,
recommendations support indefinite treatment, unless they experi­
ence HBsAg loss.
For patients with compensated cirrhosis, because antiviral ther­
apy has been shown to retard clinical progression, treatment is
recommended regardless of HBeAg status and ALT as long as HBV
DNA is detectable. In addition, stopping therapy in patients with
cirrhosis comes with the risk of decompensation. For patients with
decompensated cirrhosis, treatment is recommended regardless of
serologic and biochemical status, as long as HBV DNA is detect­
able. In addition, patients with decompensated cirrhosis should be
evaluated as candidates for liver transplantation. Cirrhotics should
be treated indefinitely (see considerations for stopping antiviral
therapy in noncirrhotics, above).
Choice of Treatment Agent  Among the eight available drugs for
hepatitis B, PEG IFN has supplanted standard IFN, entecavir has
supplanted lamivudine, and tenofovir has supplanted adefovir
(Table 352-5). PEG IFN, entecavir, or tenofovir (TDF or TAF)
is recommended as first-line therapy (Table 352-3). PEG IFN
requires finite-duration therapy, achieves the highest rate of HBeAg
TABLE 352-5  Pegylated Interferon Versus Oral Nucleoside Analogues
for the Treatment of Chronic Hepatitis B
 
PEG IFN
NUCLEOSIDE ANALOGUES
Administration
Weekly injection
Daily, orally
Tolerability
Poorly tolerated,
intensive
monitoring
Well tolerated, limited
monitoring
Duration of therapy
Finite, 48 weeks
≥1 year, indefinite in most
patients
Maximum mean HBV DNA
suppression
4.5 log10
6.9 log10
Effective in high-level HBV
DNA (≥109 IU/mL)
No
Yes
HBeAg seroconversion
  During 1 year of therapy
  During >1 year of therapy
 
~30%
Not applicable
 
~20%
30% (year 2) to up to 50% (year 5)
HBeAg-negative
posttreatment HBV DNA
suppression
17% at 5 years
7% at 4 years (lamivudine)
HBsAg loss
  During 1 year of therapy
  During >1 year of therapy
    After 1 year of therapy–
HBeAg-negative
 
3–4%
Not applicable
12% at 5 years
 
0–3%
3–8% at 5 years of therapy
3.5% at 5 years
Antiviral resistance
None
Lamivudine: ~30% year 1,
~70% year 5
Adefovir: 0% year 1, ~30% year 5
Telbivudine: up to 4% year 1,
22% year 2
Entecavir: ≤1.2% through year 6
Tenofovir: 0% through year 8
Use in cirrhosis,
transplantation,
immunosuppressed
No
Yes
Cost, 1 year of therapy
++++
to ++
Abbreviations: HBV, hepatitis B virus; HBeAg, hepatitis B e antigen; HBsAg,
hepatitis B surface antigen; PEG IFN, pegylated interferon.
responses after a year of therapy, and does not support viral muta­
tions, but it requires subcutaneous injections and is associated with
inconvenience, more intensive clinical and laboratory monitor­
ing, and intolerability. Oral agents do not require injections or
cumbersome laboratory monitoring, are very well tolerated, lead
to improved histology in 50−90% of patients, suppress HBV DNA
more profoundly than PEG IFN, and are effective even in patients
who fail to respond to IFN-based therapy. Although oral agents
are less likely to result in HBeAg responses during the first year
of therapy, as compared to PEG IFN, treatment with oral agents is
usually extended beyond the first year and, by the end of the second
year (of the current-generation, potent, high-barrier-to-resistance
agents entecavir and tenofovir), yields HBeAg responses (and even
HBsAg responses) comparable in frequency to those achieved after
1 year of PEG IFN (and without the associated side effects). Choice
of agent should be based on clinical features, any history of prior
treatment, and patient and physician preference; availability, rela­
tive cost, and insurance coverage often also play a role. As noted
above, some physicians prefer to begin with PEG IFN, while most
physicians and patients prefer oral agents as first-line therapy.
Between the two first-line oral agents, entecavir and tenofovir, data
to support the choice of one compared to the other are limited.
Several observational studies have suggested that TDF is superior
to entecavir in reducing the risk of HCC. Such studies, however,
sophisticated statistical analyses notwithstanding, are subject to
confounding influences that could favor TDF (e.g., physician pref­
erence for TDF in milder disease and for entecavir in more severe
disease, including in older patients and those with impaired renal
function). In addition, while several studies confirm a differential
effect of TDF on long-term HCC risk, many others do not, and, in
most cases, adjusting for such factors as age, sex, country, albumin,
platelets, AFP, cirrhosis, and diabetes, eliminated the benefit of TDF
over entecavir. Therefore, currently, the preponderance of data are
insufficient to support this benefit of TDF over entecavir. Between
TDF and TAF, TAF is preferred in patients with creatinine clearance
<50 mL/min, reduced bone density, and risk factors for renal injury
(e.g., decompensated cirrhosis, creatinine clearance <60 mL/min,
poorly controlled hypertension or diabetes, proteinuria, active glo­
merulonephritis, concomitant nephrotoxic medications, or solidorgan transplantation). As noted above, the EASL recommendation
for TAF extends to persons >60 years, who are at increased risk of
TDF nephrotoxicity.
POPULATIONS WITH UNIQUE TREATMENT
CONSIDERATIONS
Pregnancy  Substantial experience with lamivudine during preg­
nancy (see above) has identified no teratogenicity; although widely
used during pregnancy, lamivudine remains classified as pregnancy
category C. Although IFNs do not appear to cause congenital
anomalies, these have antiproliferative properties and should be
avoided during pregnancy. Adefovir during pregnancy has not
been associated with birth defects; however, the risk of spontaneous
abortion may be increased, and adefovir is categorized as pregnancy
category C. Data on the safety of entecavir during pregnancy have
not been published (pregnancy category C). Sufficient data in
animals and limited data in humans suggest that telbivudine and
tenofovir (both pregnancy category B) can be used safely during
pregnancy; however, telbivudine is not an acceptable first-line drug.
In general, then, except for lamivudine and tenofovir, and until
additional data become available, the other antivirals for hepatitis B
should be avoided or used with extreme caution during pregnancy.
Tenofovir is the current drug of choice in pregnancy.
Children  For children aged 2 to <18 years old with HBeAgpositive hepatitis B (most children will be HBeAg-positive; children
with HBeAg-negative chronic hepatitis B have not been well rep­
resented in treatment studies), treatment is recommended if HBV
DNA is detectable and ALT levels are elevated, but not if ALT levels
are normal. Each of the available first-line drugs is approved for dif­
ferent childhood age groups (PEG IFN α2a age ≥5 years [approved
for hepatitis C, not B, but can be used in hepatitis B]; entecavir and
tenofovir age ≥2 years). Package inserts should be consulted for
childhood doses.
HIV Co-infection  Patients with HBV-HIV co-infection can have
progressive HBV-associated liver disease and, occasionally, a severe
exacerbation of hepatitis B resulting from immunologic recon­
stitution following antiretroviral therapy. Tenofovir and the com­
bination of tenofovir and emtricitabine in one pill are approved
therapies for HIV and represent excellent choices for treating
HBV infection in HBV-HIV co-infected patients. Generally, even
for HBV-HIV co-infected patients who do not yet meet treatment
criteria for HIV infection, treating for both HBV and HIV is recom­
mended. In HIV-HBV co-infection, TAF, because of its better safety
profile, is preferable to TDF.
Cirrhosis  For patients with compensated or decompensated cir­
rhosis, PEG IFN should not be used; the emergence of such lifethreatening side effects as sepsis can result in further deterioration
and loss of antiviral effectiveness. Therefore, in this patient subset,
therapy with a very favorable resistance profile (e.g., entecavir or
tenofovir) is the preferred treatment. For patients with end-stage
chronic hepatitis B who undergo liver transplantation, reinfec­
tion of the new liver is almost universal in the absence of antiviral
therapy. The majority of patients become high-level viremic carriers
with minimal liver injury. Before the availability of antiviral therapy,
an unpredictable proportion experienced severe hepatitis B-related
liver injury, sometimes a fulminant-like hepatitis and sometimes
a rapid recapitulation of the original severe chronic hepatitis B
(Chap. 350). Currently, however, prevention of recurrent hepati­
tis B after liver transplantation has been achieved definitively by
combining short-term (5–7 days) intravenous hepatitis B immune
globulin (HBIG) with lifelong low-resistance oral entecavir or TDF
or TAF (Chap. 356); in some patients, especially those with a low
risk for recurrence, the newer, more potent, and less resistanceprone oral agents may be used instead of HBIG for posttransplanta­
tion therapy. For patients at high risk for recurrence and progressive
disease (e.g., patients with HDV-HBV or HIV-HBV co-infections)
as well as for nonadherent patients, lifelong combination HBIG-oral
agent therapy should be considered. For patients receiving livers
from anti-HBc-positive donors, lifelong oral-agent therapy is rec­
ommended (without HBIG).
CHAPTER 352
Chronic Hepatitis
Immunosuppression  Patients with chronic hepatitis B who
undergo cytotoxic chemotherapy for treatment of malignancies as
well as patients treated with immunosuppressive, anticytokine, or
anti–tumor necrosis factor (TNF) therapies experience enhanced
HBV replication and viral expression on hepatocyte membranes
during chemotherapy coupled with suppression of cellular immu­
nity. When chemotherapy is withdrawn, such patients are at risk for
reactivation of hepatitis B, often severe and occasionally fatal. Such
rebound reactivation represents restoration of cytolytic T-cell func­
tion against a target organ enriched in HBV expression. Preemptive
treatment with the first of the oral HBV antivirals, lamivudine,
prior to the initiation of chemotherapy was shown to reduce the
risk of such reactivation substantially; treating after reactivation
has occurred is less effective. The newer, more potent oral antiviral
agents, entecavir and tenofovir, which are even more effective in
preventing hepatitis B reactivation and with a lower risk of antiviral
drug resistance, are preferred.
The risk of HBV reactivation and liver injury varies from highest
(e.g., B cell–depleting agents, anthracycline derivatives, moder­
ate-/high-dose corticosteroids for ≥4 weeks) to moderate (e.g.,
TNF-α inhibitors, cytokine or integrin inhibitors, tyrosine kinase
inhibitors, low-dose corticosteroids for ≥4 weeks) to lowest (e.g.,
immunosuppressive agents like methotrexate and azathioprine,
intraarticular corticosteroids, any dose of corticosteroids for
≤1 week). Most authorities (e.g., Centers for Disease Control and
Prevention, AASLD, American Gastroenterological Association,
EASL) recommend HBsAg and anti-HBc (± anti-HBs) screen­
ing of all patients prior to initiation of chemotherapy or other
long-term immunosuppressive regimens. Preemptive antiviral pro­
phylaxis is recommended for any patient with detectable HBsAg,
given the high risk of reactivation (which can occur in ~25% of
cases but potentially in up to 68%). While, generally, reactivation
in HBsAg-negative but anti-HBc-reactive patients is rare, in this
population, careful monitoring is recommended, except in the set­
ting of B cell–depleting therapies, which are associated with a high
risk for reactivation. Thus, antiviral prophylaxis is recommended
in HBsAg-negative/anti-HBc-positive patients undergoing B-cell
depletion therapy or stem cell transplantation.
The optimal duration of antiviral therapy after completion of
chemotherapy is not known, but a suggested approach is 6 months
(12 months for B cell–depleting agents) for inactive hepatitis B car­
riers and longer-duration therapy in patients with baseline HBV
DNA levels >2 × 103 IU/mL, until standard clinical endpoints are
met (Table 352-4).
■
■CHRONIC HEPATITIS D (DELTA HEPATITIS)
Chronic hepatitis D virus (HDV) infection is most commonly observed
in patients who were exposed to HDV while already harboring chronic
HBV infection (so-called superinfection). Chronic HDV infection may
also follow acute co-infection with HBV and HDV but at a rate no higher
than the rate of chronicity of acute hepatitis B. That is, although HDV
co-infection can increase the severity of acute hepatitis B, HDV does
not increase the likelihood of progression to chronic hepatitis B. When,
however, HDV superinfection occurs in a person who is already chron­
ically infected with HBV, long-term HDV infection is the rule, and a
worsening of the liver disease is the expected consequence. Except for
severity, chronic hepatitis B plus D has similar clinical and laboratory
features to those seen in chronic hepatitis B alone. Relatively severe
and progressive chronic hepatitis, with or without cirrhosis, is the rule,
and mild chronic hepatitis is the exception. Occasionally, however,
mild hepatitis or even, rarely, inactive carriage occurs in patients with
chronic hepatitis B plus D, and the disease may become indolent after
several years of infection. The clinical and laboratory features of
chronic HDV infection are summarized in Chap. 350.
PART 10
Disorders of the Gastrointestinal System
TREATMENT
Chronic Hepatitis D
Management is not well defined, and the host cellular RNA poly­
merase upon which HDV replication depends cannot be targeted by
conventional antiviral agents. Treatment with type I interferons is
the only approved treatment for chronic HDV infection, although,
per expert opinion, treating underlying HBV infection is suggested
as well if the HBV DNA level is >2 × 103 IU/mL. Preliminary
experimental trials of IFN-α suggested that conventional doses and
durations of therapy lower levels of HDV RNA and aminotransfer­
ase activity only transiently during treatment but have no impact
on the natural history of the disease. In contrast, high-dose IFN-α
(9 million units three times a week) for 12 months was reported to
be associated with a sustained loss of HDV replication and clinical
improvement in up to 50% of patients. Furthermore, a substantial
improvement in 12-year survival after liver transplantation was
observed in patients with hepatitis D who received high-dose IFN-α
(86 vs 31% with standard-dose IFN-α). A suggested approach to
therapy has been high-dose, long-term IFN for at least a year and,
in responders, extension of therapy until HDV RNA and HBsAg
clearance; however, extension of therapy to a second year provided
no advantage, and sustained responses after completion of therapy
have been rare. Compared to standard IFN-α, PEG IFN has been
shown to be more effective in chronic hepatitis D but still of limited
therapeutic value; after 48 weeks of therapy, durable undetectable
HDV RNA for 24 posttreatment weeks has been reported in a
quarter to just over a half of patients. Disappointingly, loss of viro­
logic responses (reappearance of HDV RNA) was observed during
long-term (median 4.5 years) monitoring in a majority of 24-weekposttreatment responders, with durable HDV RNA suppression to
undetectable in only 12%. None of the nucleoside analogue anti­
viral agents for hepatitis B are effective in hepatitis D, and adding
ribavirin or oral nucleoside agents (lamivudine and, more recently,
adefovir and tenofovir) to PEG IFN was not found to be more effec­
tive than PEG IFN monotherapy. While recommended, 12 months
of PEG IFN therapy is far from satisfactory. The lackluster response
to type I interferon treatment with or without nucleoside agents
highlights the need for novel therapies to target HDV infection.
The approach farthest along in clinical testing is the entry inhibi­
tor bulevirtide, a lipopeptide that blocks the NTCP receptor, pre­
venting hepatocyte entry of both HBV and HDV. In phase 2 studies
of subcutaneous bulevirtide alone or combined with PEG IFN for
varying treatment durations, a significant reduction occurred in
HDV RNA in >71% of subjects, with a more pronounced effect
in combination therapy. Similarly, a phase 2 study of bulevirtide
combined with tenofovir (vs tenofovir monotherapy) for 24 weeks
demonstrated significant reductions in HDV RNA at the end
of treatment (50–77% vs 4% with tenofovir alone), but rebound
of viremia was universal at follow-up week 48. Most recently, in
an ongoing phase 3 clinical trial, ALT normalized in 51–56% of
patients receiving bulevirtide after 48 weeks of treatment, and
45–48% of patients had a ≥2 log10 reduction in or undetectable
HDV RNA; this treatment effect was maintained though 96 weeks
of treatment. In 2020, bulevirtide was authorized conditionally by
the European Medicines Agency for the treatment of HDV with or
without PEG IFN or nucleoside analogues. Review by the U.S. Food
and Drug Administration (FDA) is anticipated.
Preliminary trials have also been performed with lonafarnib, a
prenylation inhibitor, which is combined with ritonavir to improve
bioavailability. Prenylation, the posttranslational covalent addition
of the prenyl lipid farnesyl to large HDV antigen, is required for
this HDV protein to interact and form secreted viral particles with
HBsAg. Phase 2 studies of lonafarnib/ritonavir have demonstrated
significant, on-treatment reductions in HDV RNA, and in one
study, the addition of IFNα increased the likelihood of achieving
a viral response to 89%. Additional experimental approaches to
the treatment of hepatitis D include nucleic acid polymer therapy,
often combined with PEG IFNα. Nucleic acid polymer therapy is
thought to inhibit HBsAg release and has been administered alone
or with PEG IFN; to date, these studies have been done in one
Eastern European site. While promising reductions in HDV RNA
and HBsAg have been reported, these studies have been plagued
by adverse effects, and confirmatory studies have not materialized.
In patients with end-stage liver disease secondary to chronic
hepatitis D, liver transplantation has been effective. If hepatitis D
recurs in the new liver without the expression of hepatitis B (an
unusual serologic profile in immunocompetent persons but com­
mon in transplant patients), liver injury is limited. In fact, the
outcome of transplantation for chronic hepatitis D is superior to
that for chronic hepatitis B; in such patients, combination HBIG
and nucleoside analogue therapy for hepatitis B is indicated
(Chap. 356).
■
■CHRONIC HEPATITIS C
Regardless of the epidemiologic mode of acquisition of hepatitis C
virus (HCV) infection, chronic hepatitis follows acute hepatitis C in
50−70% of cases; chronic infection is common even in those with a
return to normal in aminotransferase levels after acute hepatitis C,
adding up to an 85% likelihood of chronic HCV infection after acute
hepatitis C. Few clues had emerged to explain host differences associ­
ated with chronic infection until recently, when variation in a single
nucleotide polymorphism (SNP) on chromosome 19, IL28B (which
codes for IFN-λ3, now renamed IFNL3), was identified that distin­
guished between responders and nonresponders to IFN-based antiviral
therapy (see below). The same variants correlated with spontaneous
resolution after acute infection: 53% in genotype C/C, 30% in geno­
type C/T, but only 23% in genotype T/T. The association with HCV
clearance after acute infection is even stronger when IL28B (IFNL3)
haplotype is combined with haplotype G/G of an SNP near human
leukocyte antigen (HLA) class II DBQ1*03:01.
In patients with chronic hepatitis C followed for 20 years, progres­
sion to cirrhosis occurs in ~20−25%. Such is the case even for patients
with relatively clinically mild chronic hepatitis, including those without
symptoms, with only modest elevations of aminotransferase activity, and
with mild chronic hepatitis on liver biopsy. Even in cohorts of well-
compensated patients with chronic hepatitis C referred for clinical
research trials (no complications of chronic liver disease and with nor­
mal hepatic synthetic function), the prevalence of cirrhosis may be as
high as 50%. Most cases of hepatitis C are identified initially in asymp­
tomatic patients who have no history of acute hepatitis C (e.g., those dis­
covered while attempting to donate blood, while undergoing lab testing
as part of an application for life insurance, or as a result of routine labo­
ratory tests). The source of HCV infection in many of these cases is not
defined, although a long-forgotten percutaneous exposure (e.g., injection
drug use) in the remote past can be elicited in a substantial proportion
and probably accounts for most infections; most of these infections were
acquired in the 1960s and 1970s among persons in the 1945–1965 birth
cohort (Chap. 350), coming to clinical attention decades later.
Approximately one-third of patients with chronic hepatitis C have
normal or near-normal aminotransferase activity; although one-third
to one-half of these patients have chronic hepatitis on liver biopsy, the
grade of liver injury and stage of fibrosis tend to be mild in the vast
majority. In some cases, more severe liver injury has been reported—
even, rarely, cirrhosis, most likely the result of previous histologic activ­
ity. Among patients with persistent normal aminotransferase activity
sustained over ≥5−10 years, histologic progression has been shown to
be rare; however, approximately one-fourth of patients with normal
aminotransferase activity experience subsequent aminotransferase
elevations, and histologic injury can be progressive once abnormal
biochemical activity resumes. Therefore, continued clinical monitor­
ing and antiviral therapy are indicated, even for patients with normal
aminotransferase activity.
Despite this substantial rate of progression of chronic hepatitis C
and even though liver failure can result from end-stage chronic hepati­
tis C, the long-term prognosis over 1–2 decades for chronic hepatitis C
in most patients is relatively benign. Mortality over 10−20 years among
patients with transfusion-associated chronic hepatitis C has been
shown not to differ from mortality in a matched population of trans­
fused patients in whom hepatitis C did not develop. Although death
in the hepatitis group is more likely to result from liver failure and
although hepatic decompensation may occur in ~15% of such patients
over the course of a decade, the majority (almost 60%) of patients
remain asymptomatic and well compensated, with no clinical sequelae
of chronic liver disease. Overall, chronic hepatitis C tends to be very
slowly and insidiously progressive, if at all, in most patients, whereas
in approximately one-fourth of cases, chronic hepatitis C will progress
eventually to end-stage cirrhosis. In fact, because HCV infection is so
prevalent, and because a proportion of patients progress inexorably to
end-stage liver disease, hepatitis C was the most frequent indication
for liver transplantation (Chap. 356) in the era prior to the availability
of direct-acting antiviral (DAA) therapy (see below). In the United
States, hepatitis C accounts for up to 40% of all chronic liver disease;
as of 2007, mortality caused by hepatitis C surpassed that associated
with HIV/AIDS, and as of 2012, reported deaths caused by hepatitis C
surpassed those associated with all other notifiable infectious diseases
(HIV, tuberculosis, hepatitis B, and 57 other infectious diseases). More­
over, because the prevalence of HCV infection is so much higher in the
“baby boomer” cohort born between 1945 and 1965, three-quarters of
the mortality associated with hepatitis C occurs in this age cohort.
In a 2010 modeling study based on the prevalence of HCV infec­
tion and its known natural history, liver disease mortality associated
with hepatitis C was projected to peak in the year 2022; however, the
observed HCV-related mortality in the United States peaked in 2013
and has fallen since, paralleling the introduction in 2013 of highly
effective DAA agents (see below). The declining death rates related to
HCV infection following the wide use of DAAs have been observed
across most of the United States.
Progression of liver disease in patients with chronic hepatitis C has
been reported to be more likely in patients with older age, longer dura­
tion of infection, advanced histologic stage and grade, more complex
HCV quasispecies diversity, increased hepatic iron, concomitant other
liver disorders (alcoholic liver disease, chronic hepatitis B, hemochro­
matosis, α1 antitrypsin deficiency, and steatohepatitis), HIV infection,
and obesity. Among these variables, however, duration of infection
appears to be one of the most important, and some of the others prob­
ably reflect disease duration to some extent (e.g., quasispecies diversity,
hepatic iron accumulation). No other epidemiologic or clinical features
of chronic hepatitis C (e.g., severity of acute hepatitis, level of amino­
transferase activity, level of HCV RNA, presence or absence of jaundice
during acute hepatitis) are predictive of eventual outcome. Despite
the relatively benign nature of chronic hepatitis C over time in many
patients, cirrhosis following chronic hepatitis C has been associated
with the late development, after several decades, of HCC (Chap. 87);
the annual rate of HCC in cirrhotic patients with hepatitis C is 1–4%,
occurring primarily in patients who have had HCV infection for
30 years or more.
Referral bias may account for the more severe outcomes described in
cohorts of patients reported from tertiary care centers (20-year progres­
sion of ≥20%) versus the more benign outcomes in cohorts of patients
monitored from initial blood-product-associated acute hepatitis or
identified in community settings (20-year progression of only 4−7%).
Still unexplained, however, are the wide ranges in reported progression
to cirrhosis, from 2% over 17 years (eventually 19% over 36 years) in
a population of Irish women with hepatitis C infection acquired from
contaminated anti-D immune globulin to 30% over ≤11 years in recipi­
ents of contaminated intravenous immune globulin.
CHAPTER 352
Perhaps the best prognostic indicator in chronic hepatitis C is liver
histology; the rate of hepatic fibrosis may be slow, moderate, or rapid.
Patients with mild necrosis and inflammation as well as those with
limited fibrosis have an excellent prognosis and limited progression to
cirrhosis. In contrast, among patients with moderate to severe necro­
inflammatory activity or fibrosis, including septal or bridging fibrosis,
progression to cirrhosis is highly likely over the course of 10−20 years.
Among patients with compensated cirrhosis associated with hepatitis
C, the 10-year survival rate is close to 80%; mortality occurs at a rate
of 2−6% per year; decompensation at a rate of 4−5% per year; and, as
noted above, HCC at a rate of 1−4% per year. Estimates of the natural
history of chronic hepatitis C have been made, based on data available
on the prevalence of HCV infection in the U.S. population and on the
rate of disease progression. Weighted primarily by the concentration of
chronic hepatitis C in the baby boomer generation, the peak prevalence
was estimated to have occurred in 2015. The calculated frequency of
cirrhosis in U.S. patients with hepatitis C was 5% in 1990 and 25% in
2010 and was projected to be 37% in 2020. A discussion of the patho­
genesis of liver injury in patients with chronic hepatitis C appears
in Chap. 350.
Chronic Hepatitis
Clinical features of chronic hepatitis C are similar to those described
above for chronic hepatitis B. Generally, fatigue is the most common
symptom; jaundice is rare. Immune complex–mediated extrahepatic
complications of chronic hepatitis C are less common than in chronic
hepatitis B (despite the fact that assays for immune complexes are
often positive in patients with chronic hepatitis C), with the exception
of essential mixed cryoglobulinemia (Chap. 350), which is linked to
cutaneous vasculitis and membranoproliferative glomerulonephritis
as well as lymphoproliferative disorders such as B-cell lymphoma
and unexplained monoclonal gammopathy. In addition, chronic
hepatitis C has been associated with extrahepatic complications unre­
lated to immune-complex injury. These include Sjögren’s syndrome,
lichen planus, porphyria cutanea tarda, renal injury, type 2 diabetes
mellitus, and the metabolic syndrome (including insulin resistance and
steatohepatitis). In addition, a link has been observed between HCV
infection and cardiovascular/cerebrovascular disease, rheumatologic/
immunologic disorders, mental health and cognitive disorders, and
nonliver malignancies.
Laboratory features of chronic hepatitis C are similar to those in
patients with chronic hepatitis B, but aminotransferase levels tend to
fluctuate more (the characteristic episodic pattern of aminotransferase
activity) and to be lower, especially in patients with long-standing dis­
ease. An interesting and occasionally confusing finding in patients with
chronic hepatitis C is the presence of autoantibodies. Rarely, patients
with autoimmune hepatitis (see below) and hyperglobulinemia have
false-positive immunoassays for anti-HCV. On the other hand, some
patients with serologically confirmable chronic hepatitis C have circu­
lating anti-LKM. These antibodies are anti-LKM1, as seen in patients
with autoimmune hepatitis type 2 (see below), and are directed against
a 33-amino-acid sequence of cytochrome P450 IID6.
Histopathologic features of chronic hepatitis C, especially those
that distinguish hepatitis C from hepatitis B, are described in
Chap. 350.
TREATMENT
Chronic Hepatitis C
Therapy for chronic hepatitis C has evolved substantially in the
30 years since IFN-α was introduced for this indication in 1991.
The therapeutic armamentarium grew to include PEG IFN with
ribavirin and, then, in 2011, the introduction of the first protease
inhibitors, telaprevir and boceprevir, used in combination with
PEG IFN and ribavirin in patients with HCV genotype 1. The field
of antiviral therapy for hepatitis C was transformed beginning in
2013, with the approval of the first nucleoside analogue polymerase
inhibitor sofosbuvir. Although several of these combination regi­
mens have been supplanted by better, later-generation drugs, as of
2020, five all-oral, highly effective (>95%), low-resistance, pangeno­
typic, well-tolerated, short-duration (primarily 8–12 weeks) combi­
nation regimens of DAA drugs are recommended. The remarkable
historical evolution of antiviral therapy for hepatitis C is instructive.
THE INTERFERON ERA (1991–2011)
IFN-based therapy has been supplanted by DAA agents introduced
in the second decade of the twenty-first century; however, many
important lessons about antiviral therapy for chronic hepatitis
C were learned from the experience with IFN-based treatment,
and many of the limitations of—and disparities in responsiveness
to—IFN-based therapy have been overcome by current-generation
DAA treatments. Mechanistically, HCV proteins inhibit several
steps of the JAK-STAT signal transduction pathway, and by acti­
vation of JAK-STAT signaling, exogenous IFN culminates in and
restores intracellular expression of IFN-stimulated genes and their
protein products that have antiviral properties.
PART 10
Disorders of the Gastrointestinal System
When first approved, subcutaneous IFN-α three times a week
for 6 months achieved a sustained virologic response (SVR;
defined then as a reduction of HCV RNA to PCR-undetectable
levels ≥24 weeks after completion of therapy) in <10%. Doubling
Structural proteins
Nonstructural proteins
E1
E2
NS2
p7
NS3
C
4A
NS5A
NS5B
Target
NS3-4A protease inhibitors NS5A inhibitors
NS5B inhibitors
Function
Block N3-4A cleavage
Inhibit NS5A protein,
essential for
replication
Suffix
-previr
-asvir
-buvir
Drugs
Glecaprevir, voxilaprevir
Velpatasvir, ledipasvir,
pibrentasvir
Sofosbuvir
FIGURE 352-2  Hepatitis C virus (HCV) viral genome and drug targets. The HCV viral genome encodes three structural proteins (c-nucleocapsid and e1-2 envelope proteins)
and seven nonstructural proteins. Current, first-line direct-acting antivirals have three viral targets: (1) the NS3/4A viral protease; (2) the NS5A protein, which is a critical
component of the viral replication apparatus; and (3) NS5B RNA-dependent RNA polymerase.
the duration of therapy increased the SVR rate to ~20%, and addi­
tion to the regimen of daily ribavirin (ineffective when used alone),
an oral guanosine nucleoside, increased the SVR rate further to
40% by reducing the likelihood of virologic relapse after comple­
tion of treatment. Although its mechanism of action remains
poorly understood, ribavirin retains a limited role in supporting
DAA agents in several subgroups of otherwise refractory patients
(see below).
Beginning in 2011, for the treatment of hepatitis C, standard
IFNs were supplanted by PEG IFNs, which have substantially longer
half-lives, permitting administration once (rather than three times)
a week. Treatment with the combination of PEG IFN and ribavirin
increased SVR rates to 55% overall—to >40% in genotypes 1 and
4, requiring 48 weeks of therapy, and to >80% in genotypes 2 and
3, requiring only 24 weeks of therapy—and resulted in histologic
improvement in approximately three-fourths of patients. After
initiation of IFN treatment, ALT levels fell precipitously, and up to
90% of virologic responses were achieved within the first 12 weeks
of therapy. Failure to achieve an early virologic response (EVR), a
≥2 log10 reduction in HCV RNA by week 12, predicted failure to
experience a subsequent SVR. Similarly, patients in whom HCV
RNA became undetectable within 4 weeks (i.e., who achieve a rapid
virologic response [RVR]) had a very high likelihood of achieving
an SVR (Fig. 352-2).
Most relapses occurred within the first 12 weeks after treatment,
and absence of HCV RNA 12 weeks after completion of therapy has
become the current standard for SVR (SVR12); relapses are very rare
6 months to a year after SVR and almost unheard of after 2 years.
Of documented durability decades after successful therapy, an SVR
to antiviral therapy for chronic hepatitis C is tantamount to a cure
and is followed by marked improvements in liver disease outcomes
(see below).
Patient variables that correlated with IFN-based SVRs included
favorable genotype (genotypes 2 and 3 as opposed to genotypes 1
and 4); low baseline HCV RNA level (<800,000 IU/mL); histologi­
cally mild hepatitis and minimal fibrosis, especially absence of cir­
rhosis; immunocompetence; age <40; female gender; and absence
of comorbid metabolic disease. High levels of HCV RNA, more
histologically advanced liver disease, and high HCV quasispecies
diversity all went hand in hand with advanced duration of infection
and reduced IFN responsiveness. Ironically, patients whose disease
was least likely to progress were the ones most likely to respond to
IFN and vice versa.
As described above in the discussion of spontaneous recovery
from acute hepatitis C, IFN gene variants discovered in genomewide association studies were shown to have a substantial impact on
IFN responsiveness of patients with genotype 1 to antiviral therapy.
In studies of patients treated with PEG IFN and ribavirin, variants
RNA-dependent RNA
polymerase
Phosphoprotein
NS3/4a viral protease
Block the RNAdependent RNA
polymerase
of the IL28B (now renamed IFNL3) SNP that code for IFN-λ3 (a
type III IFN, the receptors for which are more discretely distributed
than IFN-α receptors and more concentrated in hepatocytes) cor­
related significantly with responsiveness. Homozygotes for the C
allele at this locus (C/C) achieved SVRs of ~80%, heterozygotes
(C/T) SVRs of ~35%, and homozygotes for the T allele (T/T) SVRs
of ~25%.
Side effects of IFN therapy are described in the section on
treatment of chronic hepatitis B (see above). Besides ribavirinassociated nasal and chest congestion, pruritus, and precipitation of
gout, the most pronounced ribavirin side effect is hemolysis, often
requiring dose reduction or addition of erythropoietin therapy (not
shown, however, to increase the likelihood of an SVR); therefore,
close monitoring of blood counts is crucial, and ribavirin should
be avoided in patients with anemia, hemoglobinopathies, coronary
artery disease or cerebrovascular disease, or renal insufficiency (the
drug is excreted renally) and in pregnancy (the drug is teratogenic,
mandating contraception during, and for several months after, ther­
apy in women of child-bearing age [because of their antiprolifera­
tive properties, IFNs also are contraindicated during pregnancy]).
For most of the decade prior to 2011, when protease inhibitors
were introduced for HCV genotype 1 (see below), the standard of
care was a combination of PEG IFN plus ribavirin (unless riba­
virin was contraindicated) for all HCV genotypes. Even after the
introduction of protease inhibitors for genotypes 1 and 4, however,
PEG IFN–ribavirin remained the standard of care for patients with
genotypes 2 and 3 until late 2013. Responsiveness to IFN-ribavirin–
based therapy was diminished in immunocompromised patients
and in patients with HIV-HCV co-infection and contraindicated in
patients with decompensated liver disease or end-stage renal disease.
The cumbersome nature of IFN-ribavirin–based therapy (injections,
complicated laboratory monitoring, side effects and poor tolerability,
modest efficacy, variables and patient subsets associated with poor
responsiveness, tailored therapy, futility rules, etc.) was supplanted
eventually (in 2016) by DAAs for all genotypes (see below). Most of
the variables associated with poor responsiveness to IFN-based ther­
apy became irrelevant, and difficult-to-treat patient subpopulations
began to experience responses to DAAs that were indistinguishable
from responses in standard patients (see below).
Persons with chronic HCV infection suffer increased liverrelated mortality, all-cause mortality, and multiple extrahepatic
disorders (see above). On the other hand, successful antiviral
therapy of chronic hepatitis C resulting in an SVR was shown to
improve survival (and to reduce the need for liver transplantation);
to lower the risk of liver failure, liver-related death, and all-cause
mortality; to slow the progression of chronic hepatitis C; to reverse
fibrosis and even cirrhosis; and to improve such HCV-associated
extrahepatic disorders as type 2 diabetes and renal disease. Whereas
the 10- and 20-year survival in the absence of an SVR is reduced in
cirrhotic patients with chronic hepatitis C, survival at these inter­
vals after an SVR has been found to be indistinguishable from that
of the general population. In cirrhotic patients (and in those with
advanced fibrosis), although successful treatment reduces mortality
and liver failure (three- to fourfold 10-year reduction) and reduces
the need for liver transplantation and the likelihood of HCC (14fold 10-year reduction), the risk of liver-related death and HCC
persists, albeit at a much reduced level, necessitating continued
clinical monitoring and cancer surveillance after SVR in patients
with advanced fibrosis and cirrhosis. On the other hand, in the
absence of an SVR, IFN-based therapy does not reduce the risk of
HCC. Fortunately, PEG IFN–ribavirin nonresponders can now be
retreated with DAAs and experience SVR rates comparable to those
in treatment-naïve persons (see below).
FIRST-GENERATION PROTEASE INHIBITORS (2011–2013)
The HCV RNA genome encodes a single polyprotein, which is
cleaved during and after translation by host and viral-encoded
proteases (Fig. 352-2). One protease involved in the cleavage of the
viral polyprotein is an NS3/4A viral protein that has serine protease
activity. Telaprevir and boceprevir were first-generation serine pro­
tease inhibitors that target the genotype 1 NS3/4A. In 2011, telapre­
vir and boceprevir used in combination with PEG IFN and ribavirin
were approved by the FDA as the first oral DAA agents for the
treatment of hepatitis C genotype 1 (not other genotypes) in adults
with stable liver disease. Although now replaced by more effective,
all-oral regimens, these first-in-class agents represented a break­
through in the treatment of chronic hepatitis C and established
milestones against which subsequent therapies could be measured.
Because selection of resistant variants developed rapidly during
monotherapy with telaprevir and boceprevir, these drugs had to
be used in combination with PEG IFN and ribavirin. Ribavirin,
in particular, appeared to reduce relapse rates significantly in
protease-inhibitor-based regimens, such that those who could not
take or were intolerant to ribavirin were unlikely to benefit from
the addition of these agents. Telaprevir and boceprevir regimens
consisted of periods of triple therapy (protease inhibitor plus PEG
IFN plus ribavirin) and periods of dual therapy (PEG IFN plus riba­
virin). Cumbersome periods of “run-in” therapy or late-treatment
consolidation therapy with PEG IFN-ribavirin, reliance on early
virologic responses to determine duration of therapy (“responseguided therapy”), and futility rules (for stopping therapy based on
poor early responsiveness) rendered first-generation DAA treat­
ment regimens complicated and unwieldy.
For patients with HCV genotype 1, protease inhibitors improved
the frequency of RVRs and SVRs significantly as compared to PEG
IFN plus ribavirin alone. In treatment-naïve patients, telaprevirbased SVRs were achieved in up to 79% of patients who received
12 weeks of triple therapy followed by 12–36 weeks of dual therapy,
with higher rates of response (83–92%) observed in those with
undetectable RNA at weeks 4 and 12. In studies with boceprevir
in treatment-naïve patients, SVRs occurred in 59–66% of patients,
and among those with undetectable HCV RNA at 8 weeks, the SVR
rate increased to 86–88%. Adding to the complexity of treatment
with these protease inhibitors were absolute stopping rules for
futility, that is, absence of HCV RNA reductions at critical treat­
ment milestones, which were shown to be invariably predictive of
nonresponse.
CHAPTER 352
Chronic Hepatitis
The first-generation protease inhibitors were successful as well
among patients in whom prior PEG IFN plus ribavirin had failed,
even in “null responders,” who had never experienced even a 2 log10
reduction in HCV RNA while on treatment. In this hardest-to-treat
group, telaprevir-based therapy achieved an SVR in 29–33% and
boceprevir achieved an SVR in 30–40%.
In a substantial proportion of protease inhibitor nonresponders,
resistance-associated substitutions (RASs; previously referred to
as resistance-associated variants [RAVs]) could be identified, but
these variants were not archived, and wild-type HCV reemerged in
almost all cases within 1.5–2 years. SVRs to these protease inhibitors
were highest in prior relapsers and treatment-naïve patients (with
SVR rates of >75%) and lowest in cirrhotic prior null responders,
for whom no benefit accrued over PEG IFN–ribavirin treatment.
Both protease inhibitors had substantial toxicities. Telaprevir was
associated with a severe, generalized (trunk and extremities), often
confluent, maculopapular, pruritic rash in ~6% of treated patients
(that required careful dermatologic monitoring in all patients and
systemic corticosteroid therapy in the most severely affected).
Other common side effects included pruritus, rectal burning, nau­
sea, diarrhea, fatigue, dysgeusia (altered or unpleasant taste), and
anemia, which required close monitoring, could be relatively refrac­
tory, and occasionally required transfusion and even hospitalization
(especially in cirrhotic prior nonresponders). Anemia occurred in
half of boceprevir-treated patients. The other most common side
effects of boceprevir were fatigue, nausea, headache, dysgeusia, dry
mouth, vomiting, and diarrhea.
Both drugs came with an inconveniently high pill burden and
had to be administered every 8 h with food (telaprevir with a 20-g
fat meal). Use of protease inhibitors was further complicated by
numerous drug-drug interactions. As telaprevir and boceprevir are
both eliminated by and inhibit CYP3A4, these agents could not
be administered with other medications that induce CYP3A4 or
are dependent on CYP3A4 for elimination. Care had to be taken
to examine for any potential interactions between these protease
inhibitors and other medications the patient was taking, and a
convenient website became available to check for such drug-drug
interactions (www.hep-druginteractions.org). This website remains
updated and of important clinical utility for checking drug-drug
interactions for current HCV DAA regimens.
Despite the improvement in SVRs with protease-inhibitor-based
regimens for genotype 1 compared to PEG IFN–ribavirin (e.g., in
treatment-naïve patients 66–79% vs 38–44%), triple-drug protease
inhibitor therapy was hampered by amplified intolerability, the
complexity of response-guided regimens and futility stopping rules,
the inconvenience of thrice-daily dosing with meals and a high
pill burden, the need for PEG IFN injections and ribavirin with all
their intolerability, and multiple drug-drug interactions. Moreover,
side effects appeared to be more severe and burdensome once these
drugs entered practice, especially in cirrhotic nonresponders, in
whom studies reported from Europe showed serious adverse events
in up to 45% and deaths in up to 3%. All these issues, as well as
rapidly accelerating progress on next-generation and all-oral DAA
TABLE 352-6  Indications and Recommendations for Antiviral Therapy of Chronic Hepatitis Ca
Standard Indications for Therapy
All patients with chronic HCV infection (detectable
HCV RNA, with or without elevated ALT) except for
those with short life expectancies owing to comorbid
conditions.
Any stage of fibrosis; pretreatment biopsy is no longer
embraced and has been supplanted by noninvasive
measures of fibrosis, e.g., imaging to determine liver
elasticity.
Retreatment Recommended
Relapsers, partial responders, or nonresponders after
a previous course of interferon-based therapy or prior
direct-acting antiviral therapy
Antiviral Therapy Not Recommended
Pregnancy: No sizable clinical studies of directacting antivirals during pregnancy are available.
Ribavirin is contraindicated during pregnancy;
therefore, any regimen including ribavirin should not
be used. Sofosbuvir and sofosbuvir plus ledipasvir
are classified as pregnancy category B, but the other
direct-acting antivirals do not have a pregnancy
classification. Therefore, these therapies are not
indicated routinely in pregnancy and should be used,
with caution, only if the benefit of treatment outweighs
the potential for fetal risk.
Therapeutic Regimens (based on AASLD-IDSA
recommendations, www.hcvguidelines.org)b  
Treatment-Naïve
Previously Treated with DAA
Simplified treatment regimenb
All genotypes
Sofosbuvir-velpatasvir × 12 weeks
Glecaprevir-pibrentasvir × 8 weeks  
PART 10
Disorders of the Gastrointestinal System
Cirrhosis – compensatedc
Genotypes 1–6
Glecaprevir-pibrentasvir × 8 weeks 
Genotypes 1, 2, 4, 5, or 6
Sofosbuvir-velpatasvir × 12 weeks
Cirrhosis – decompensated
Ribavirin-eligible
Genotypes 1–6
Sofosbuvir-velpatasvir with weight-based ribavirind ×
12 weeks 
Genotypes 1,4, 5, or 6
Ledipasvir-sofosbuvir with ribavirine × 12 weeks 
Ribavirin-ineligible
Genotypes 1–6
Sofosbuvir-velpatasvir × 24 weeks 
Genotypes 1, 4, 5, or 6
Ledipasvir-sofosbuvir × 24 weeks 
FEATURES ASSOCIATED WITH REDUCED RESPONSIVENESS TO DIRECT-ACTING ANTIVIRAL
COMBINATION THERAPY
Genotype and subtype (genotype 1a less responsive than genotype 1b for several drugs)
Treatment experience
Advanced fibrosis (bridging fibrosis, cirrhosis)
Reduced adherence
aRapidly evolving new recommendations continue to be issued; for up-to-date treatment recommendations, please see www.hcvguidelines.org. bExcludes: cirrhosis,
prior treatment, pregnancy, hepatitis B surface antigen positive, hepatocellular carcinoma (HCC), or liver transplant. No adjustment required for end-stage renal
disease. cChild-Pugh class A cirrhosis only; excludes: prior treatment, pregnancy, hepatitis B surface antigen positive, end-stage renal disease, HCC, or liver transplant.
After treatment, ongoing surveillance for HCC is required. dFor Child-Pugh class C, initial low dose of ribavirin 600 mg is recommended with increase as tolerated.
eInitial dose of ribavirin 600 mg with increase as tolerated. eNot recommended for prior NS3/4 protease inhibitor failure or genotype 3. fIn compensated cirrhosis, the
addition of weight-based ribavirin is recommended. gConsider extending treatment to 24 weeks if high-risk features (e.g., genotype 3, cirrhosis, or failure following
sofosbuvir-glecaprevir-pibrentasvir).
Abbreviations: ALT, alanine aminotransferase; HCV, hepatitis C virus.
therapy (see below), conspired to temper enthusiasm for these new
antivirals; after a brief stint as recommended therapy (2011–2013),
these drugs became obsolete and are no longer recommended or
available.
DIRECT-ACTING ANTIVIRAL COMBINATIONS OF SECONDGENERATION PROTEASE INHIBITORS, FIRST-GENERATION
POLYMERASE INHIBITORS, AND FIRST-GENERATION NS5A
INHIBITORS (2014–2015)
Since late 2013, the number of new antiviral agents for hepatitis C has
expanded substantially, and currently, PEG IFN–based treatments
have been supplanted by five remaining therapeutic regimens, which
are all oral, IFN-free, highly efficacious (>95% SVR), and well toler­
ated, with high barriers to resistance, simple dosing, low pill burdens,
treatment durations as brief as 8–12 weeks, and pangenotypic efficacy
(Table 352-6). These drugs are distributed among three classes of
DAAs: NS3/4 protease inhibitors (which cleave the single HCV poly­
protein into constituent structural and nonstructural proteins [drug
name ending in “-previr”]), NS5B nucleoside and nonnucleoside
polymerase inhibitors (which interfere with the RNA-dependent RNA
polymerase [a replicase] involved in synthesis of viral RNA [drug
name ending in “-buvir”]), and NS5A inhibitors (which interfere
Sofosbuvir failure
All genotypes
Sofosbuvir-velpatasvir-voxilaprevir × 12 weeks
Genotypes 1, 2, 4, 5, or 6 (alternative)
Glecaprevir-pibrentasvir × 16 weekse
Glecaprevir-pibrentasvir failure
All genotypes
Glecaprevir-pibrentasvir plus sofosbuvir and weightbased ribavirin × 16 weeks
Sofosbuvir-velpatasvir-voxilaprevir × 12 weeksf 
Multiple direct-acting antiviral failures
All genotypes
Glecaprevir-pibrentasvir plus sofosbuvir and weightbased ribavirin × 16 weeksg 
Sofosbuvir-velpatasvir-voxilaprevir plus weight-based
ribavirin × 24 weeks 
Cirrhosis – decompensated
All genotypes
Sofosbuvir-velpatasvir plus weight-based ribavirin ×
24 weeks 
Genotypes 1, 4, 5, or 6
Ledipasvir-sofosbuvir plus ribavirine × 24 weeks
DECOMPENSATED CIRRHOSIS
Any protease-inhibitor-containing regimen is
CONTRAINDICATED in decompensated cirrhosis
(glecaprevir, grazoprevir, voxilaprevir).
with a membrane-associated phosphoprotein essential to the HCV
RNA replication complex [drug name ending in “-asvir”]).
The first of the new DAA agents (approved in November 2013)
was simeprevir, a second-generation protease inhibitor for geno­
type 1, followed shortly thereafter (December 2013) by sofosbuvir,
a pangenotypic nucleoside polymerase inhibitor. For genotype 1,
both of these agents had to be combined with PEG IFN and
ribavirin; for genotypes 2 and 3, sofosbuvir was administered with
ribavirin, without PEG IFN; however, these treatment regimens
have been supplanted by combinations of all-oral, IFN-free DAAs,
and ribavirin is rarely needed and retained only for very limited
indications.
Simeprevir  When simeprevir was used with PEG IFN, its efficacy
(genotype 1b > 1a) was similar to that of first-generation prote­
ase inhibitors but required only once-a-day dosing without the
complexity of response-guided therapy. Similar to first-generation
protease inhibitors, however, simeprevir was hampered by many
drug-drug interactions, restriction to genotype 1 HCV, side effects,
and limited efficacy in patients with a highly prevalent HCV NS3
polymorphism. Little about simeprevir supported its adoption in
combination with PEG IFN and ribavirin. On the other hand, the
combination of simeprevir (150 mg) along with sofosbuvir (400 mg)
for 12 weeks was found to be effective in treatment-naïve (97%
SVR12) or treatment-experienced (95% SVR12) patients without cir­
rhosis and in treatment-naïve (88% SVR12) or treatment-refractory
(79% SVR12) patients with cirrhosis. While this all-oral, simeprevirsofosbuvir regimen was a landmark in the treatment of chronic
hepatitis C, simeprevir was eclipsed by more effective DAAs and is
no longer in clinical use.
Sofosbuvir  Sofosbuvir, which inhibits the NS5B RNA-dependent
RNA polymerase, was the first non–protease inhibitor DAA to
be approved and had a paradigm-shifting excellent profile—high
potency, high barrier to resistance, pangenotypic activity, very
well tolerated with limited adverse effects (most commonly mild
fatigue, insomnia, headache, and nausea), once-daily oral admin­
istration, and relative freedom from major drug-drug interactions.
Sofosbuvir has efficacy in all genotypes (1–6); in treatment-naïve
subjects and prior nonresponders to PEG IFN–based and proteaseinhibitor-based therapy; with PEG IFN–ribavirin or in IFN-free
regimens; in combination with ribavirin or with NS5A inhibitors
(see below); and for treatment periods as brief as 8–12 weeks. Cur­
rently, sofosbuvir is used in combination with one of two NS5A
inhibitors and is a component of three of the five currently recom­
mended DAA regimens (Table 352-6).
Paritaprevir-Ritonavir, Ombitasvir, and Dasabuvir  The combina­
tion of ritonavir (100 mg)–boosted paritaprevir (150 mg), a prote­
ase inhibitor; ombitasvir (25 mg), an NS5A inhibitor; and dasabuvir
(250 mg), a nonnucleoside NS5B polymerase inhibitor, with or
without weight-based ribavirin (total of five drugs), was approved
in December 2014 for genotypes 1 and 4. In clinical trials, this com­
bination achieved SVR12 rates of 87–100% in treatment-naïve and
treatment-experienced patients with genotype 1; without ribavirin,
this combination in genotype 1a was ~7% less responsive than in
genotype 1b. Therefore, in treatment-naïve patients with genotype
1a, this combination was administered with ribavirin for 12 weeks
in the absence of cirrhosis (95–97% SVR12) or for 24 weeks in the
presence of compensated cirrhosis (94% SVR12), whereas in patients
with genotype 1b, the combination did not require ribavirin, and
the duration of therapy was 12 weeks for both noncirrhotic and cir­
rhotic patients (99–100% SVR12). In July 2016, the FDA approved a
long-acting formulation of dasabuvir, allowing once-a-day instead
of twice-a-day treatment; for genotype 1a, twice-daily ribavirin
dosing remained.
This combination was well tolerated with generally mild side
effects. Hyperbilirubinemia (primarily unconjugated) and eleva­
tions in ALT levels could occur but resolved during or shortly
after treatment. Because of occasional hyperbilirubinemia and
potential hepatotoxicity (FDA warning letter issued October 2015
regarding hepatic failure/decompensation reported in treated cir­
rhotic patients), this combination (and all subsequently introduced
protease-inhibitor-containing combinations) was contraindicated
in patients with decompensated cirrhosis, and treated cirrhotic
patients had be monitored closely for decompensation.
Compared to superior regimens that came later (e.g., sofosbuvirledipasvir, see below), this regimen had the disadvantage of requir­
ing twice-a-day ribavirin therapy for genotype 1a and of being
contraindicated in decompensated cirrhosis; however, it had the
advantage of offering a 12-week, ribavirin-free regimen for prior
null responders with cirrhosis and providing an option for patients
with renal failure. Based on regimen simplicity and superior­
ity, subsequent-generation, ribavirin-free combination DAAs have
supplanted paritaprevir-ritonavir, ombitasvir, and dasabuvir; this
regimen is no longer recommended by the AASLD; however, it is
retained in EASL recommendations as an alternative regimen for
genotype 1b only.
Sofosbuvir and Daclatasvir  Daclatasvir, an NS5A inhibitor, along
with the polymerase inhibitor sofosbuvir, was approved by the FDA
in July 2015 for genotype 3 and in February 2016 for genotype 1.
At the time of its approval for genotype 3, daclatasvir filled a need
inadequately met by other available combination DAAs; however,
eventually, recommendation of this combination regimen was
extended to genotypes 1–4 in the United States and to all genotypes
(1–6) in Europe. Daclatasvir, a 60-mg tablet, and sofosbuvir, a sepa­
rate 400-mg tablet, were taken once a day for 12–24 weeks.
In clinical trials among treatment-naïve or treatment-experienced
patients, SVR12 rates for 12 weeks of daclatasvir plus sofosbuvir
were 98% for genotype 1 (comparable results in genotypes 1a and
1b), 92% for genotype 2, and 89% for genotype 3. In patients with
cirrhosis, SVR12 was achieved in 93% with Child-Pugh class A and B
but in only 56% with class C decompensated cirrhosis. For patients
with genotype 3 and cirrhosis, the combination was effective in
treatment-naïve patients (94% SVR12) but less so in prior nonre­
sponders (69% SVR12).
CHAPTER 352
Chronic Hepatitis
Like other sofosbuvir–NS5A inhibitor combinations, daclatasvir
plus sofosbuvir was well tolerated (mild fatigue, headache, nausea,
or diarrhea in 5–14%) but could cause severe bradycardia when
administered with amiodarone (contraindicated). Responsiveness
to daclatasvir-containing drug combination therapy was reduced
in cirrhotic patients with genotype 1a and in both cirrhotic and
noncirrhotic patients with genotype 3 who had baseline daclatasvirassociated NS5A RASs.
As new combination DAAs were introduced, however, dacla­
tasvir-sofosbuvir was less competitive and no longer filled a niche.
This regimen has been supplanted by better, later-generation com­
bination DAAs and is no longer recommended.
CURRENT FIRST-LINE COMBINATIONS OF
DIRECT-ACTING ANTIVIRALS
Sofosbuvir-Ledipasvir  The DAA combination that has had an
early and dominant role in the treatment of hepatitis C is sofosbuvir
(400 mg) plus the NS5A inhibitor ledipasvir (90 mg) in a once-aday, fixed-dose, single pill, approved in October 2014 for genotype
1 and in November 2015 for genotypes 4, 5, and 6. Phase 3 trials
were conducted in treatment-naïve noncirrhotic patients, in treat­
ment-naïve cirrhotic and noncirrhotic patients, and in treatmentexperienced cirrhotic and noncirrhotic patients treated for 8, 12,
or 24 weeks, both with and without ribavirin. In treatment-naïve
noncirrhotics, an SVR12 was achieved in 97–99% of subjects, and
no benefit was observed by extending therapy from 12 to 24 weeks
or by adding ribavirin. Moreover, for treatment-naïve, noncirrhotic
patients with baseline HCV RNA <6 × 106 IU/mL, a treatment dura­
tion of 8 weeks was as effective as one of 12 weeks (94–95% SVR12),
which may be a consideration for a proportion of patients. In cir­
rhotic patients, SVR12 was achieved in 97–100% of treatment-naïve
subjects (no advantage of extending therapy from 12 to 24 weeks or
of adding ribavirin); however, for cirrhotic prior nonresponders to
IFN-based therapy, 12 weeks of therapy was inferior (86% SVR12)
to 24 weeks of therapy (100% SVR12). This combination, which is
equally effective in patients with HIV-HCV co-infection and in
African-American patients, has been shown to be highly effective
in patients with decompensated cirrhosis and in patients with hepa­
titis C after liver transplantation and after kidney transplantation.
Initially, sofosbuvir-ledipasvir was not recommended in patients
with advanced renal failure; however, subsequently, the safety and
efficacy of sofosbuvir-ledipasvir in patients with advanced renal
failure were established, and the combination was approved for this
indication (November 2019). All sofosbuvir-containing regimens
can be associated with severe bradycardia in patients taking the
antiarrhythmic agent amiodarone, especially along with beta block­
ers; sofosbuvir-containing combinations are contraindicated with
amiodarone. Drug-drug interactions are few, but P-glycoprotein
inducers, such as St. John’s wort and rifampin, and proton pump
gastric acid inhibitors, such as omeprazole, may reduce sofosbuvirledipasvir concentrations. Generally, responsiveness to sofosbuvirledipasvir is not reduced in patients with baseline RASs to these
agents, with the exception of treatment-experienced patients who
have baseline NS5A RASs (see Table 352-6).
Elbasvir-Grazoprevir  Elbasvir (50 mg), an NS5A inhibitor, com­
bined in a single, fixed-dose pill with grazoprevir (100 mg), an
NS3/4 protease inhibitor, was approved in January 2016 as a oncea-day (with or without food) treatment for genotypes 1 and 4. In
clinical trials, a 12-week course was effective in treatment-naïve
and treatment-experienced patients without cirrhosis or with
compensated cirrhosis. In treatment-naïve patients, this combina­
tion yielded an SVR12 in 92% of patients with genotype 1a, 99%
with genotype 1b, and 100% with genotype 4 (very small numbers,
however); 10 patients with genotype 6 were included, but only 80%
achieved SVR12. Cirrhotic and noncirrhotic patients had compa­
rable rates of SVR12, 97% and 94%, respectively. For this drug com­
bination, however, ~11% of patients with genotype 1a harbor NS5A
polymorphisms, which may include RASs, at baseline. If present,
these NS5A RASs reduce efficacy of elbasvir-grazoprevir (unlike
baseline RASs to most of the other combination DAA regimens
described above and below) from 99% to 58% in treatment-naïve
patients. Therefore, all patients with genotype 1a required base­
line RAS testing prior to starting this regimen; when these RASs
were present, treatment extension to 16 weeks and the addition of
weight-based ribavirin were documented to bring SVR12 rates up to
expected levels of close to 100%. In treatment-experienced patients,
both extending treatment to 16 weeks and adding ribavirin were
studied; however, generally, in the absence of baseline NS5A RASs,
SVR12 rates were not increased over those without ribavirin for
12 weeks (94–97%). Among nonresponders to prior protease inhib­
itor therapy, even in the absence of baseline NS5A RASs, ribavirin
had to be added to a 12-week regimen; in the presence of baseline
NS5A RASs, treatment was extended to 16 weeks and ribavirin
added. For genotype 1b, NS5A RASs are not an issue, and the only
subgroup requiring modification of a 12-week course of therapy
were prior nonresponders to protease inhibitor regimens, for whom
ribavirin was added. For genotype 4, the recommended regimen
for all prior nonresponders (whether to PEG IFN–ribavirin or
protease inhibitor regimens) was 16 weeks of elbasvir-grazoprevir
plus ribavirin. Now that simpler, improved combination regimens
are available, for patients with NS5A RASs, extending the duration
of elbasvir-grazoprevir and adding ribavirin have been abandoned
(Table 352-6); however, elbasvir-grazoprevir is one of the currently
recommended DAA combinations (Table 352-6).
PART 10
Disorders of the Gastrointestinal System
This combination is just as effective in patients with HIVHCV co-infection and in patients with advanced renal failure
(including those requiring hemodialysis), but like all proteaseinhibitor-including DAA combinations, it is contraindicated in
decompensated cirrhosis. In this vein, like other protease inhibitor
regimens, elbasvir-grazoprevir can be associated with aminotrans­
ferase elevations and potential hepatotoxicity; because these drugs
are excreted by the liver, plasma drug concentrations may become
elevated substantially in the presence of impaired hepatic function.
Therefore, all treated patients should have ALT screening periodi­
cally during therapy, and the drug should be stopped for elevations
exceeding 10-fold or for elevations of conjugated bilirubin, alkaline
phosphatase, or prothrombin time.
Elbasvir-grazoprevir is well tolerated, with only low levels of
mild adverse effects (fatigue, headache, or nausea in 5–11%) seen
just as frequently in placebo recipients. Both elbasvir and grazopre­
vir are substrates for CYP3A and are subject to multiple potential
drug-drug interactions. Therefore, this combination should not
be used with potent CYP3A inducers; conversely, CYP3A and
OATP1B1 inhibitors can lead to untoward elevations of plasma
elbasvir-grazoprevir concentrations. Checking for potential drugdrug interactions is advisable prior to initiating therapy (www
.hep-druginteractions.org).
Compared to other available regimens for genotypes 1 and 4,
elbasvir-grazoprevir has the disadvantage/inconvenience of requir­
ing baseline NS5A RAS testing but the advantages of a comparable
regimen for cirrhotic and noncirrhotic patients, for treatment-naïve
and treatment-experienced patients, and for patients with normal
renal function and with renal failure. As noted above, its limitations
and complexities notwithstanding, elbasvir-grazoprevir remains
one of the recommended DAA combinations (albeit not one of the
“simplified” treatment algorithms, see below).
Sofosbuvir-Velpatasvir  The combination in a single, fixed-dose
pill of velpatasvir (100 mg), a highly potent, pangenotypic NS5A
inhibitor, and the polymerase inhibitor sofosbuvir (400 mg) was
approved in June 2016 for genotypes 1–6 in treatment-naïve
and treatment-experienced noncirrhotic and cirrhotic patients. In
August 2017, approval was extended to include patients with HCVHIV co-infection. Ribavirin is not required, including in patients
with genotypes 2 and 3, except in patients with decompensated
cirrhosis.
In a series of clinical trials, this combination for 12 weeks in
the absence of ribavirin was shown to yield a 99% SVR12 (range
97–100%) in genotypes 1, 2, 4, 5, and 6 and 95% in genotype 3.
Baseline NS5A RASs had no impact on responsiveness.
Prior to the availability of this drug combination, patients with
genotype 3, especially those with cirrhosis and prior null response
to other therapies, proved to be the most refractory subset of
patients. In treatment-naïve patients with genotype 3, 12 weeks
of sofosbuvir-velpatasvir (95% SVR12) was superior to 24 weeks of
sofosbuvir plus ribavirin (80% SVR12). In patients with genotype
3, the combination of sofosbuvir-velpatasvir for 12 weeks was
comparable in noncirrhotic (97% SVR12) and cirrhotic patients
(91% SVR12) and in treatment-naïve (97% SVR12) and treatmentexperienced (90% SVR12) patients and was superior in all these
categories to 24 weeks of sofosbuvir plus ribavirin (87, 66, 86, and
63%, respectively). In cirrhotic null responder patients, most avail­
able IFN-free regimens for genotype 3 (including daclatasvir plus
sofosbuvir, which had been approved specifically for this genotype)
achieved SVR12 rates in the range of ~60–75%, while the combina­
tion of PEG IFN, ribavirin, and sofosbuvir could boost SVR12 to the
mid-80% range. For treatment-experienced patients with genotype
3, sofosbuvir-velpatasvir in noncirrhotic and cirrhotic patients had
similarly high efficacy (91 and 89% SVR12, respectively); this was
the highest recorded SVR12 for genotype 3 cirrhotic null respond­
ers treated with IFN-free DAA regimens. Finally, in patients with
genotypes 1–4 and 6 and with decompensated, class B cirrhosis
(55% treatment-experienced), sofosbuvir-velpatasvir plus ribavirin
for 12 weeks yielded an SVR12 in 94%; this result was better than
sofosbuvir-velpatasvir without ribavirin for 12 weeks (83% SVR12)
or 24 weeks (86% SVR12).
Like other all-oral DAAs, sofosbuvir-velpatasvir was very well
tolerated; in noncirrhotic and compensated cirrhotic patients, mild
headache and fatigue were seen in >10% (this occurred in a compara­
ble proportion of placebo recipients), and in patients with decompen­
sated cirrhosis, mild fatigue, headache, nausea, insomnia, diarrhea,
and anemia (ribavirin was part of the regimen) were seen in >10%.
Like other sofosbuvir-containing regimens, sofosbuvir-velpatasvir
should not be administered along with amiodarone (potential serious
bradycardia); in addition, P-glycoprotein inducers and moderate-topotent CYP3A inducers can reduce plasma levels of sofosbuvir and/
or velpatasvir. Checking for drug-drug interactions prior to therapy
is advisable (www.hep-druginteractions.org). Baseline RASs do not
influence responsiveness to this combination. Sofosbuvir-velpatasvir
is one of the currently recommended DAA combinations for hepatitis
C (Table 352-6). Because it is so simple and broadly effective across
patient subgroups, sofosbuvir-velpatasvir is one of the two combination
DAA regimens recommended by the AASLD and EASL as a preferred,
simplified treatment algorithm (Table 352-6).
Sofosbuvir-Velpatasvir-Voxilaprevir  Approved in July 2017, the
pangenotypic, high-barrier-to-resistance protease inhibitor voxila­
previr (100 mg) added to the polymerase inhibitor–NS5A inhibitor
combination of sofosbuvir-velpatasvir yields a very well tolerated
triple-drug combination with ~97% SVR12 across all HCV geno­
types and patient subgroups. These include the small percentage of
patients with genotype 1 and genotype 3 refractory to previously
approved DAA combinations as well as noncirrhotic/cirrhotic,
treatment-naïve/treatment-experienced groups, including those
who had or who had not received prior NS5A treatment. Efficacy
was independent of the number of prior DAA drug classes received,
and no effects of baseline NS5A RASs were noted.
The potential for abbreviated (8-week) treatment with this triple
combination was explored in a clinical trial involving treatmentnaïve patients; however, the shortened duration was inferior to a full
12-week course. The side effect profile for sofosbuvir-velpatasvirvoxilaprevir was similar to that in the placebo arm of clinical trial
patients and included mild and uncommon headache, fatigue,
nausea, and diarrhea.
Because other DAA regimens are so effective in most patients
with chronic hepatitis C, recommendations for sofosbuvir-velpatasvirvoxilaprevir are limited to a small subset of otherwise refractory
patients: for treatment-naïve cirrhotic patients with genotype 3
and baseline NS5A velpatasvir RAS Y93H, for treatment-naïve
(according to AASLD, not EASL) or IFN-ribavirin–experienced
noncirrhotic or cirrhotic patients with genotype 3 (Table 352-6),
and for patients with or without compensated cirrhosis and
prior, failed NS5A inhibitor–containing therapy (consult www.
hcvguidelines.org).
This triple-drug combination, like all sofosbuvir-containing
combinations, is contraindicated in patients taking amiodarone
and, like all protease inhibitor–containing combinations, in patients
with decompensated cirrhosis. Concomitant omeprazole, 20 mg,
can be taken with this sofosbuvir-containing regimen. Prior to initi­
ating therapy, checking for drug-drug interactions is recommended.
Glecaprevir-Pibrentasvir  A regimen of 8 weeks of this single-pill,
fixed-dose combination of the protease inhibitor glecaprevir (300 mg)
and NS5A inhibitor pibrentasvir (120 mg), two pangenotypic, highpotency DAAs with high barriers to resistance (approved in August
2017), achieves SVR12 in close to 100% of treatment-naïve patients
with all genotypes, with or without cirrhosis: SVR12 of ~99% for
genotypes 1, 2, and 4–6 and of 95–98% for genotype 3. Extended
treatment for 12 weeks did not increase efficacy. In trials among
treatment-experienced patients, treatment with 12 weeks of this
DAA combination was just as effective as 16 weeks for all geno­
types except genotype 3; however, with increasing numbers of prior
treatment courses, SVR12 rates fell—100% for patients treated with
a protease inhibitor only, 88% for patients treated with an NS5A
inhibitor only, and 79% for patients treated previously with both a
protease inhibitor and an NS5A inhibitor. Similarly, baseline RASs
reduced SVR12 rates—from 100% without RASs (or with RASs
limited to those reflecting protease inhibitor resistance) to 89% for
baseline NS5A RASs.
For retreatment of patients with prior glecaprevir-pibrentasvir
failure, 16 weeks of glecaprevir-pibrentasvir plus sofosbuvir are
recommended (alternatively, sofosbuvir-velpatasvir-voxilaprevir
for 12 weeks [plus ribavirin in cirrhotic patients]). Glecaprevirpibrentasvir for 16 weeks is recommended as well after failure to
respond to the triple-drug combination of sofosbuvir-velpatasvirvoxilaprevir (see below). For retreatment of patients with
sofosbuvir-velpatasvir-voxilaprevir failure, 16 weeks of glecap­
revir-pibrentasvir plus ribavirin is recommended, as is a repeat
course of sofosbuvir-velpatasvir-voxilaprevir plus ribavirin for
24 weeks. For navigating the complexities of treating DAArefractory patients, checking up-to-date guidelines (www.hcvguid­
lines.org) and referral to a specialized center are advisable.
As is the case for any DAA combination containing a protease
inhibitor, glecaprevir-pibrentasvir is contraindicated in decom­
pensated cirrhosis; it has been shown to achieve an SVR12 in 98%
of patients with stage 4 or 5 renal disease (in treatment-naïve or
experienced, cirrhotic or noncirrhotic patients) and is a preferred
treatment for patients with severe renal impairment. This DAA
combination should be taken with food, and drug-drug interactions
should be considered prior to initiating treatment. Because it is so
simple and broadly effective across patient subgroups (8 weeks for all
noncirrhotic treatment-naïve patients [including with HIV co-infection];
12 weeks for all treatment-experienced cirrhotics [including with
HIV co-infection] and treatment-naïve cirrhotics with genotype 3
[except treatment-experienced cirrhotic or noncirrhotic genotype 3
{16 weeks}]), glecaprevir-pibrentasvir is one of the two combination
DAA regimens recommended by the AASLD and EASL as a preferred,
simplified treatment algorithm (Table 352-6).
Emerging data on the impact of DAAs on the natural history
of chronic hepatitis C indicated that, as was documented for IFNbased therapy, successful DAA therapy is associated with a gradual
reduction in fibrosis progression and a regression of advanced
fibrosis (cirrhosis), improvement in survival among patients with
decompensated cirrhosis, a reduction in HCC, and a decline in the
number of patients with hepatitis C being referred for liver trans­
plantation. Indeed, early and broad implementation of DAAs in the
treatment of chronic hepatitis C has been instrumental in enabling
the decline of HCV-related deaths since 2013. Modeling estimates
suggest that application of DAA therapies has the potential to result
in a 50–70% reduction in the hepatitis C–associated disease burden
by 2050.
CHAPTER 352
Chronic Hepatitis
TREATMENT RECOMMENDATIONS
Because the pace of new drug development and approval has been
so rapid, the AASLD and the Infectious Diseases Society of America
(IDSA) have been providing a consensus of updated treatment
recommendations for patients with hepatitis C; these recommenda­
tions, which continue to be revised regularly based on new data, are
available online at www.hcvguidelines.org and should be consulted
before initiating therapy. The EASL issues similar (but not identical)
treatment recommendations annually for hepatitis C (www.easl.eu),
most recently in November 2020.
Prior to therapy, all patients should undergo pretreatment evalu­
ation, which includes HCV RNA, HCV genotype, renal function,
hepatitis B status (testing should include HBsAg and anti-HBc),
HIV status, and fibrosis stage. Despite the presence of pangenotypic
regimens, HCV genotype should still be determined, because the
genotype may contribute to decisions about which treatment regi­
mens are indicated and can guide insurance coverage (Table 352-6).
Fibrosis assessment is an especially important element of the
pretreatment evaluation: the absence or presence of cirrhosis or
advanced fibrosis determines the treatment options from which to
select, including the antiviral agents to be used, the duration of ther­
apy, and the now rare need for ribavirin (Table 352-6). In the past, a
pretreatment liver biopsy was relied upon to assess histologic grade
and stage as well as to identify such histologic factors as steatosis,
which can influence adversely responsiveness to therapy. As therapy
has improved for patients with a broad range of histologic severity,
and as noninvasive measures of the stage of fibrosis (e.g., assess­
ment of liver stiffness by imaging, FIB-4 score [see above]) have
advanced in accuracy and popularity, noninvasive approaches have
supplanted histology in almost most cases. As noted above, if cir­
rhosis or advanced fibrosis is present prior to therapy, the risk
of HCC, although reduced substantially by successful therapy, is
not eliminated, and twice-yearly posttreatment imaging for HCC
surveillance remains indicated even after an SVR. In patients with
low-level fibrosis at baseline, achievement of SVR allows cessation
of such surveillance.
Several reports have appeared describing hepatitis B reactivation,
often severe, during and after DAA therapy in patients co-infected
with HCV and HBV who were not being treated for their HBV
infections. Therefore, screening for HBV infection is recommended
prior to initiating DAA therapy for hepatitis C (which should
have been done to determine HBV immunity status as a prelude
to recommended hepatitis B vaccination in patients with chronic
hepatitis C), and therapy for HBV infection (for those meeting
HBV treatment criteria, see above) should be initiated prior to or
simultaneously with HCV therapy.
Monitoring of serum HCV RNA levels before, during, and after
treatment is crucial in assessing the response to therapy; moreover,
the baseline level may contribute to determining the duration of
therapy (e.g., in noncirrhotic patients with genotype 1 and HCV RNA
<6 × 106 IU/mL, 8 [instead of the usual 12] weeks of sofosbuvir-ledi­
pasvir may be a consideration). The goal of treatment is to eradicate
HCV RNA during therapy and to document that the virus remains
undetectable for at least 12 weeks after completion of therapy (SVR12).
Most recent guidelines support a simplified treatment algorithm
for patients with HCV infection, leveraging the well tolerated and
highly effective pangenotypic regimens. Patients who are eligible
for this simplified algorithm must not have any of the following
clinical characteristics: prior treatment failure, cirrhosis, presence
of HBsAg, pregnancy, HCC, or history of liver transplantation. In
the absence of any of these features, the patient may be treated with
either sofosbuvir/velpatasvir for 12 weeks or glecaprevir/pibrentas­
vir for 8 weeks.
PART 10
Disorders of the Gastrointestinal System
INDICATIONS FOR ANTIVIRAL THERAPY
Patients with chronic hepatitis C who have detectable HCV RNA
in serum, whether or not aminotransferase levels are increased,
and chronic hepatitis of any grade and stage are candidates for
antiviral therapy with DAA agents. The only exception would be
patients with short life expectancies, for whom treating hepatitis C
would have no influence on longevity. Certainly, for patients with
advanced liver disease, early treatment merits a high priority.
Although patients with persistently normal aminotransferase activ­
ity tend to progress histologically very slowly or not at all, they
respond to antiviral therapy just as well as do patients with elevated
aminotransferase levels; therefore, such patients are candidates for
antiviral therapy. As noted above, antiviral therapy has been shown
to improve survival and complication-free survival and to slow
progression of and to reverse fibrosis.
Patients who have relapsed after, or failed to respond to, a course
of IFN-based (currently no longer encountered commonly) or
DAA agent–based therapy are candidates for retreatment with a
DAA therapy regimen (Table 352-6). For patients who have failed
to respond to a DAA combination, options include increasing the
duration of therapy with the failed regimen, adding ribavirin, or
changing the drug class (e.g., after failed protease and polymerase
inhibitors, switching to an NS5A-containing combination). In the
presence of cirrhosis or a need for urgent retreatment, patients
who have failed protease inhibitor plus polymerase inhibitor com­
bination therapy or who have failed an NS5A combination are
candidates for RAS testing and tailored therapy based on such
resistance testing. If reliable RAS testing is not available, adding
ribavirin or extending the duration of therapy are options. For prior
nonresponders to IFN-based therapy, NS5A inhibitor–containing
regimens are highly effective; however, reduced responsiveness
can be encountered, especially in cirrhotic patients. The simplest
current approach for patients who have not responded to prior
DAA therapy containing sofosbuvir is to re-treat with sofosbuvirvelpatasvir-voxilaprevir for 12 weeks or glecaprevir-pibrentasvir for
16 weeks. Additional details for treatment of such patient subgroups
can be found at www.hcvguidelines.org; in the small proportion
of patients in whom initial DAA courses do not achieve SVR12,
referral to an experienced treatment center is advisable. Worth
reiterating, protease inhibitors are contraindicated for patients
with decompensated cirrhosis, and sofosbuvir-containing regimens
are not recommended for patients taking amiodarone (especially
with beta blockers) for treatment of cardiac arrhythmias. While
sofosbuvir-containing DAA combinations were not recommended
initially for patients with advanced renal failure, subsequent studies
demonstrated safety and efficacy in this subgroup, and sofosbuvircontaining DAA combinations are now approved for advanced
renal failure.
Persons with acute hepatitis C are also candidates for antiviral
therapy (Chap. 350) with the same pangenotypic combination
DAA agents and the same duration of treatment approved for
chronic hepatitis C. For example, in a trial of a shortened, 6-week
course of sofosbuvir-velpatasvir for acute hepatitis C, a 6-week
course was less effective than a full 12-week course (81.7 vs 90.5%,
respectively). Delaying the initiation of therapy in acute hepatitis
C to allow for spontaneous recovery is no longer recommended.
While acute hepatitis C should be treated, postexposure prophylaxis
after HCV exposure is not recommended currently
In patients with biochemically and histologically mild chronic
hepatitis C, the rate of progression is slow; however, such patients
respond just as well to antiviral therapy as those with elevated
aminotransferase levels and more histologically severe hepatitis.
Because of the high cost of DAA treatments, in the past, a higher
priority was assigned to patients with advanced fibrosis/cirrhosis;
however, this controversial approach was relied upon by some
medical insurers and pharmacy benefit management organizations
to withhold therapy from patients with low-level fibrosis. Unfor­
tunately, delaying therapy until fibrosis becomes advanced misses
the opportunity to prevent all the dire consequences of chronic
hepatitis C (liver failure, death/transplantation, HCC), which can
be reduced, but not eliminated completely, once advanced fibrosis
is established. Therefore, therapy for patients with mild disease is
justified as well as cost-effective.
TREATMENT OF UNIQUE POPULATIONS
Decompensated Cirrhosis  Patients with compensated cirrhosis
can respond to therapy, and their likelihood of a sustained response
with DAAs is comparable to that in the absence of cirrhosis.
Patients with decompensated cirrhosis, who were not candidates
for IFN-based antiviral therapy, respond well to non–proteaseinhibitor-containing regimens, and specifically sofosbuvir com­
bined with an NS5A inhibitor, velpatasvir or ledipasvir, with or
without ribavirin. As noted above, protease-inhibitor-containing
combinations have been associated with potential hepatotoxicity
and hepatic decompensation and are contraindicated in this patient
subset. Treatment decisions are informed by whether the patient is
eligible for ribavirin: in ribavirin-eligible patients, ribavirin is added
to a 12-week course of sofosbuvir-NS5A therapy; however, in cases
of ribavirin ineligibility, the duration of therapy should be extended
to 24 weeks. In cases of prior failure to respond to sofosbuvirNS5A therapy, the sofosbuvir-NS5A regimen should be repeated
but supplemented with ribavirin and extended to 24 weeks (www.
hcvguidelines.org). Also, patients with decompensated cirrhosis
should be referred to a liver transplantation center.
DAAs are highly effective not only for patients with end-stage
liver disease awaiting liver transplantation but also for patients with
recurrent hepatitis C after liver transplantation. Ideally, patients
should be treated prior to liver transplantation. Given the effec­
tiveness of treatment immediately after transplantation (either for
donor or recipient HCV infection), however, the utility of treat­
ment may be limited if the patient has a high Model for EndStage Liver Disease (MELD) score and may soon receive a donor
liver for transplantation. Therefore, advocacy has been expressed
for postponing DAA therapy in patients with high-MELD-score
(≥18–20), HCV-associated, end-stage liver disease until after liver
transplantation; for patients with MELD scores <18–20, pretrans­
plantation DAA therapy is advised. Still, the decision whether to
treat pretransplantation or posttransplantation should be individu­
alized thoughtfully for each patient, based on such factors as MELD
score, time anticipated prior to availability of a donor organ, rela­
tive clinical stability, and comorbidities (Chap. 356). Because DAA
therapy is so effective, most transplantation centers are accepting
organs from HCV-infected donors, transplanting them into
HCV-uninfected recipients, and treating recipients with sofosbuvirvelpatasvir for 12 weeks or glecaprevir-pibrentasvir for 8 weeks after
transplantation—with excellent results.
Extrahepatic Manifestations  The cutaneous and renal vasculitis
of HCV-associated essential mixed cryoglobulinemia (Chap. 350)
may respond to antiviral therapy, but sustained responses were rare
after discontinuation of therapy in the IFN era, and prolonged,
potentially indefinite, therapy was recommended. Now that more
effective DAAs are available, a 12-week course of sofosbuvir-based
combination therapy has been shown to yield an SVR12 rate exceed­
ing 80% in cryoglobulinemic vasculitis. Anecdotal reports suggest
that IFN-based antiviral therapy may be effective in porphyria
cutanea tarda or lichen planus associated with hepatitis C; whether
the more appealing DAAs are effective in these groups remains to
be documented.
HIV Co-infection  In patients with HCV/HIV co-infection, hepa­
titis C is more progressive and severe than in HCV mono-infected
patients. Although patients with HCV/HIV co-infection responded
less well to IFN-based antiviral therapy for hepatitis C, they respond
as well as patients with HCV infection alone to DAA combination
regimens and, generally, should be treated with the same DAA
combinations and for the same duration as those for patients (in
the absence or presence of cirrhosis) with HCV mono-infection.
For patients with HCV/HIV co-infection, an abbreviated, 8-week
course of sofosbuvir-ledipasvir for low-level HCV RNA is not
recommended, and a full 12 weeks should be given; similarly, for
patients with genotype 4, a 12-week course of glecaprevir-pibren­
tasvir is recommended instead of an 8-week course for treatmentnaïve or -experienced patients with or without cirrhosis (Table 352-6).
In HCV/HIV-infected patients, ribavirin can potentiate the toxicity
of didanosine (e.g., lactic acidosis) and the lipoatrophy of stavu­
dine, and zidovudine can exacerbate ribavirin-associated hemolytic
anemia; therefore, these drug combinations should be avoided.
In HCV/HIV co-infected persons, the list of potential drug-drug
interactions is extensive and should be consulted carefully before
beginning DAA treatment (www.hcvguidelines.org).
Substance Use Disorder  Patients with a history of injection drug
use and alcoholism can be treated successfully for chronic hepati­
tis C, preferably in conjunction with drug and alcohol treatment
programs. Moreover, because persons who use injection drugs
(PWID), as a source of transmission to others, account dispro­
portionately for perpetuating the spread of HCV infection in the
population, the impact of treating people who are actively using
injection drugs is amplified by reducing such transmission. In a
randomized, controlled clinical trial, treatment with sofosbuvir and
velpatasvir for 12 weeks was highly successful in PWID, with SVR12
rates of 74–76%. Based on this study, and others, active injection
drug use is no longer considered a contraindication to treatment,
and indeed, mechanisms to facilitate care of PWID are being sought
to address the U.S. and global burden of HCV infection and move
toward the World Health Organization goal of HCV elimination by
2030 (see below).
Renal Impairment  The approved oral DAA combinations are
effective in patients with mild-modest renal impairment and require
no dose adjustments. For patients with severe renal impairment
(creatinine clearances <30 mL/min), including those undergoing
hemodialysis, recommended combinations are 12 weeks of elbas­
vir-grazoprevir for genotypes 1 and 4 or 12 weeks of glecaprevirpibrentasvir for all genotypes. Both in severe renal impairment and
after kidney transplantation, levels of SVR12 in patients treated with
these oral DAA combinations have approached 100%. Initially, in
patients with severe renal impairment, sofosbuvir-containing com­
binations were not recommended. Subsequently, however, based on
efficacy and safety in a series of clinical trials, sofosbuvir-containing
regimens were approved by the FDA in November 2019 for patients
with severe renal impairment.
Pregnancy  No large clinical studies of the use of DAAs during
pregnancy are available. One small study demonstrated the safety
and efficacy of sofosbuvir and ledipasvir given for 12 weeks at
week 23–24 of gestation. Ribavirin is contraindicated during preg­
nancy; therefore, any regimen including ribavirin should not be
used. Sofosbuvir; sofosbuvir-ledipasvir; and paritaprevir-ritonavir,
ombitasvir, and dasabuvir are classified as pregnancy category B;
the other DAAs do not have a pregnancy classification. Therefore,
these therapies are not indicated routinely in pregnancy and should
be used, with caution, only if the benefit of treatment is compelling
and justified compared to the potential for fetal risk. Currently,
screening all pregnant women for HCV infection is recommended.
Breast feeding is not contraindicated in women with HCV infection
(unless the mother has a break in the integrity of the nipples or is
co-infected with HIV).
CHOOSING AMONG AVAILABLE TREATMENT OPTIONS
Choosing among the number of all-oral DAA combinations
approved since 2013 was daunting to treating clinicians. Cur­
rently, however, the number of recommended DAA combina­
tions has narrowed to a very manageable few. The most popular
of the regimens have been fixed-dose, single-pill, pangenotypic
combinations. For simplicity, two “one-size-fits-all” pangenotypic
regimens—sofosbuvir-velpatasvir and glecaprevir-pibrentasvir—
can be used, for 8–12 weeks in almost all treatment-naïve, noncir­
rhotic and cirrhotic patients, including those with advanced renal
failure and HCV-HIV co-infection. Applicability of the triple-drug
combination sofosbuvir-velpatasvir-voxilaprevir is quite limited in
treatment-naïve patients, reserved primarily for cirrhotic patients
with genotype 3. As noted above, protease-inhibitor-containing
DAA regimens (elbasvir-grazoprevir, glecaprevir-pibrentasvir, and
sofosbuvir-velpatasvir-voxilaprevir) are contraindicated in decom­
pensated cirrhosis.
CHAPTER 352
Chronic Hepatitis
APPROACHES TO STEAMLINING DAA TREATMENT
The World Health Organization has set an elimination goal of a
90% reduction in the global prevalence of HCV infection by 2030,
and the Viral Hepatitis National Strategic Plan includes a goal of
an 80% reduction in the U.S. prevalence of HCV infection by 2030;
however, at the current pace of identification and treatment, this
ambitious goal is unlikely to be met. To address this shortfall, efforts
have been devoted to improving the frequency of screening and
diagnosis and to streamlining the process of enrolling persons with
HCV infection into treatment programs. Now that DAA regimens
have been simplified so dramatically, in-person specialist care is
usually not needed (except in complex patients such as those with
advanced fibrosis, comorbid conditions, and prior nonresponse to
DAA treatment). Programs with an impressive record of success
in streamlining the treatment of persons with hepatitis C include
those based on telehealth, one-stop screening and treatment initia­
tion, minimal monitoring (dispensing the full course of pills at the
beginning of therapy, relying on a limited number of in-person
or even all-remote patient contacts during therapy, and involving
no HCV RNA or liver test monitoring during treatment and not
until 12 weeks after completion of treatment), and collaborative
participation on the treatment team of clinicians experienced in
managing substance use disorder. The expansion of such programs
is anticipated eagerly.
AUTOIMMUNE HEPATITIS
■
■DEFINITION
Autoimmune hepatitis (AIH) is a chronic disorder characterized by
continuing hepatocellular necrosis and inflammation, usually with
fibrosis, which can progress to cirrhosis and liver failure. The presenta­
tion of AIH varies, from asymptomatic liver enzyme abnormalities to
severe or even, in its most severe form, marked hepatocellular drop­
out, multilobular collapse, and clinical features of fulminant hepatitis
with high mortality. Based on contemporary estimates of the natural
history of AIH, the 10-year survival is 80−98% for treated and 67%
for untreated patients. The prominence of extrahepatic features of
autoimmunity and seroimmunologic abnormalities in this disorder
supports an autoimmune process in its pathogenesis; this concept is
reflected in the prior labels lupoid and plasma cell hepatitis. Similar to
other autoimmune diseases, AIH has a strong female predominance,
with women accounting for ~75% of all cases. The age distribution is
bimodal, with one peak in the early decades (age <30) and the second
in middle to late years (often >60). Autoantibodies and other typical
features of autoimmunity, however, do not occur in all cases; among
the broader categories of “idiopathic” or cryptogenic chronic hepati­
tis, many, perhaps the majority, are probably autoimmune in origin.
Cases in which hepatotropic viruses, metabolic/genetic derangements
(including nonalcoholic fatty liver disease), and hepatotoxic drugs have
been excluded represent a spectrum of heterogeneous liver disorders of
unknown cause, a proportion of which are most likely AIH.
■
■IMMUNOPATHOGENESIS
The weight of evidence suggests that the progressive liver injury
in patients with AIH is the result of a cell-mediated immunologic
attack directed against liver cells in the setting of a loss of, or failed,
immunologic tolerance for self-liver antigens. In all likelihood, pre­
disposition to autoimmunity is inherited, whereas the liver specific­
ity of this injury is triggered by environmental (e.g., chemical, drug
[e.g., minocycline], or viral) factors. For example, patients have been
described in whom apparently self-limited cases of acute hepatitis
A, B, or C led to AIH, presumably because of genetic susceptibility
or predisposition. More recently, the emergence of AIH has been
reported following—presumably triggered by—COVID-19 and after
COVID-19 vaccination, postulated mechanistically to result from the
vigorous immune response and loss of immune tolerance in suscep­
tible persons.
PART 10
Disorders of the Gastrointestinal System
Evidence to support an autoimmune pathogenesis in this type of
hepatitis includes the following: (1) in the liver, the histopathologic
lesions are composed predominantly of cytotoxic T cells and plasma
cells; (2) circulating autoantibodies (nuclear, smooth muscle, and
others, see below), rheumatoid factor, and hyperglobulinemia are
common; (3) other autoimmune disorders—such as autoimmune
thyroiditis, rheumatoid arthritis, autoimmune hemolytic anemia,
ulcerative colitis, membranoproliferative glomerulonephritis, juvenile
diabetes mellitus, vitiligo, celiac disease, and Sjögren’s syndrome—
occur with increased frequency in patients who have AIH and in their
relatives; (4) histocompatibility haplotypes associated with autoim­
mune diseases, such as HLA-B1, B8, DR3, and DR4 as well as extended
haplotype DRB10301 and DRB10401 alleles, are common in patients
with AIH; and (5) this type of chronic hepatitis is responsive to gluco­
corticoid/immunosuppressive therapy, effective in a variety of autoim­
mune disorders.
Loss of immune tolerance as well as cytotoxic cellular immune
responses appear to be important in the pathogenesis of AIH. In vitro
studies have suggested that in patients with this disorder, CD4+ T
lymphocytes are capable of becoming sensitized to hepatocyte mem­
brane proteins and of destroying liver cells. Molecular mimicry by
cross-reacting antigens that contain epitopes similar to liver antigens
is postulated to activate these T cells, which infiltrate, and result in
injury to, the liver. Abnormalities of immunoregulatory control over
cytotoxic lymphocytes (impaired regulatory CD4+CD25+ T-cell influ­
ences) may play a role as well. Studies of genetic predisposition to AIH
demonstrate that certain haplotypes are associated with the disorder,
as enumerated above, as are polymorphisms in cytotoxic T lymphocyte
antigens (CTLA-4) and tumor necrosis factor α (TNFA*2). The precise
triggering factors, genetic influences, and cytotoxic and immunoregu­
latory mechanisms involved in this type of liver injury remain incom­
pletely defined.
Intriguing clues into the pathogenesis of AIH come from the obser­
vation that circulating autoantibodies are prevalent in patients with
this disorder. Among the autoantibodies described in these patients
are antibodies to nuclei (so-called antinuclear antibodies [ANAs],
primarily in a homogeneous pattern) and smooth muscle (so-called
anti-smooth-muscle antibodies [ASMAs], directed at actin, vimen­
tin, and skeletin), antibodies to F-actin and anti-LKM (see below),
antibodies to soluble liver antigen (directed against a uracil-guanineadenine transfer RNA suppressor protein), antibodies to α-actinin, and
antibodies to the liver-specific asialoglycoprotein receptor (or “hepatic
lectin”) and other hepatocyte membrane proteins. Although some of
these provide helpful diagnostic markers, their involvement in the
pathogenesis of AIH has not been established.
Humoral immune mechanisms have been shown to play a role
in the extrahepatic manifestations of autoimmune and idiopathic
hepatitis. Arthralgias, arthritis, cutaneous vasculitis, and glomerulo­
nephritis occurring in patients with AIH appear to be mediated by the
deposition of circulating immune complexes in affected tissue vessels,
followed by complement activation, inflammation, and tissue injury.
While specific viral antigen-antibody complexes can be identified in
acute and chronic viral hepatitis, the nature of the immune complexes
in AIH has not been defined.
■
■CLINICAL FEATURES
Many of the clinical features of AIH are similar to those described
for chronic viral hepatitis. The onset of disease may be insidious or
abrupt; the disease may present initially like, and be confused with,
acute viral hepatitis. The most common symptom, occurring in ~85%
of patients, is fatigue. In severe disease, which accounts for 20–25%
of cases, jaundice may be present as well. In approximately a quarter of
patients, the diagnosis is made in the absence of symptoms, based
on abnormal liver laboratory tests. A subset of patients with AIH has
distinct features. Such patients are predominantly young to middleaged women with marked hyperglobulinemia and high-titer circulat­
ing ANAs. This is the group with positive lupus erythematosus (LE)
preparations (initially labeled lupoid hepatitis) in whom other autoim­
mune features are common. Fatigue, malaise, anorexia, amenorrhea,
acne, arthralgias, and jaundice are common. Occasionally, arthritis,
maculopapular eruptions (including cutaneous vasculitis), erythema
nodosum, colitis, pleuritis, pericarditis, anemia, azotemia, and sicca
syndrome (keratoconjunctivitis, xerostomia) occur. In some patients,
complications of cirrhosis, such as ascites and edema (associated with
portal hypertension and hypoalbuminemia), encephalopathy, hyper­
splenism, coagulopathy, or variceal bleeding may bring the patient to
initial medical attention.
The course of AIH may be variable. In patients with mild dis­
ease or limited histologic lesions (e.g., interface hepatitis, labeled
in the past as piecemeal necrosis [inflammation and erosion of the
limiting place of periportal hepatocytes] without bridging necrosis
and fibrosis), progression to cirrhosis may be limited, but, even in
this subset, treatment and careful monitoring remain warranted,
because, if untreated, up to half can progress to cirrhosis over 15 years.
Also, asymptomatic patients, if left untreated, have a lower 10-year
overall survival. In North America, cirrhosis at presentation is
more common in African Americans than in whites. In those with
severe symptomatic AIH (aminotransferase levels >10 times normal,
marked hyperglobulinemia, “aggressive” histologic lesions—bridging
necrosis or multilobular collapse, cirrhosis), the 6-month mortality
without therapy may be as high as 40%. Such severe disease accounts
for only 20% of cases; the natural history of milder disease is variable,
often accentuated by spontaneous remissions and exacerbations. In
a 10-year (2006–2016) national Dutch study, mortality in patients
with AIH was higher than that of the general population only in
patients with cirrhosis; for patients without cirrhosis, survival was
comparable to that of the general population. Especially poor prog­
nostic signs include the presence histologically of multilobular col­
lapse at the time of initial presentation and failure of serum bilirubin
to improve after 2 weeks of therapy. Death may result from hepatic
failure, hepatic coma, other complications of cirrhosis (e.g., variceal
hemorrhage), and intercurrent infection. In patients with established
cirrhosis, HCC may be a late complication (Chap. 87) but occurs less
frequently than in cirrhosis associated with viral hepatitis.
Laboratory features of AIH are similar to those seen in chronic viral
hepatitis. Liver biochemical tests are invariably abnormal but may not
correlate with the clinical severity or histopathologic features in indi­
vidual cases. Many patients with AIH have normal serum bilirubin,
alkaline phosphatase, and globulin levels with only minimal amino­
transferase elevations. Most commonly, serum AST and ALT levels
are increased and fluctuate in the range of 100−1000 units. In severe
cases, the serum bilirubin level is moderately elevated (51−171 μmol/L
[3−10 mg/dL]). Hypoalbuminemia occurs in patients with very active
or advanced disease. Serum alkaline phosphatase levels may be moder­
ately elevated or near normal. In a small proportion of patients, marked
elevations of alkaline phosphatase activity occur; in such patients,
clinical and laboratory features overlap with those of primary biliary
cholangitis (Chap. 355). The prothrombin time is often prolonged,
particularly late in the disease or during active phases.
Polyclonal hypergammaglobulinemia (>2.5 g/dL) is common in
AIH, as is the presence of rheumatoid factor. As noted above, circulat­
ing autoantibodies are also prevalent, most characteristically ANAs in a
homogeneous staining pattern. Smooth-muscle antibodies are detected
in 63% of cases and, when combined with a positive ANA, have high
specificity. Of important note, low-titer ANA or ASMA is observed
often in other liver diseases, including metabolic dysfunction–associated
steatotic liver disease (in one series, 20% of patients had a positive ANA
and 14% had a positive ASMA) and viral hepatitis. Studies of autoan­
tibodies in AIH have led to the recognition of two categories of AIH.
Type I AIH is the classic syndrome prevalent in North America and
northern Europe occurring in young women, associated with marked
hyperglobulinemia, lupoid features, circulating ANAs, and HLA-DR3
or HLA-DR4 (especially B8-DRB1*03). Also associated with type I AIH
are autoantibodies against F-actin and atypical perinuclear antineutro­
philic cytoplasmic antibodies (pANCA). Included in the spectrum of
type I AIH is a subset of patients who lack ANA and anti-LKM1 but
who have circulating antibodies to soluble liver antigen (SLA). Most
of these patients are women and have clinical features similar to, or
perhaps more severe than, those of other patients with type I AIH. The
presence of SLA autoantibodies has been reported to be associated with
a worse prognosis.
Type II AIH, often seen in children, more common in Mediterranean
populations, and linked to HLA-DRB1 and HLA-DQB1 haplotypes, is
associated not with ANA but with anti-LKM. Actually, anti-LKMs rep­
resent a heterogeneous group of antibodies. In type II AIH, the antibody
is anti-LKM1, directed against cytochrome P450 2D6. This is the same
anti-LKM seen in some patients with chronic hepatitis C. Anti-LKM2 is
seen in drug-induced hepatitis, and anti-LKM3 (directed against uridine
diphosphate glucuronyltransferases) is seen in patients with chronic
hepatitis D but has also been observed in type II AIH. Another autoan­
tibody observed in type II AIH is directed against liver cytosol formimi­
notransferase cyclodeaminase (anti-liver cytosol 1 [LC-1]).
Liver biopsy is imperative to make the diagnosis of AIH. Abnormali­
ties are similar to those described for chronic viral hepatitis, but classic
features of AIH include inflammation involving not only the portal
tracts but extending beyond the portal tracts (interface hepatitis), the
presence of plasma cells in the inflammatory infiltrate, and central vein
perivenulitis. Necroinflammatory activity characterizes the lobular
parenchyma, and evidence of hepatocellular regeneration is reflected
by “rosette” formation, the occurrence of thickened liver cell plates, and
regenerative “pseudolobules.” Septal fibrosis, bridging fibrosis, and cir­
rhosis are frequent. In patients with early AIH presenting as an acutehepatitis-like illness, lobular and centrilobular (as opposed to the more
common periportal) necrosis has been reported. Bile duct injury and
granulomas are uncommon; however, a subgroup of patients with AIH
has histologic, biochemical, and serologic features overlapping those of
primary biliary cholangitis (Chap. 355).
■
■DIAGNOSTIC CRITERIA
An international group has suggested a set of criteria for establishing
a diagnosis of AIH. Exclusion of liver disease caused by genetic dis­
orders, viral hepatitis, drug hepatotoxicity, and alcohol is linked with
such inclusive diagnostic criteria as hyperglobulinemia, autoantibod­
ies, and characteristic histologic features. This international group has
also suggested a comprehensive diagnostic scoring system that, rarely
required for typical cases, may be helpful when typical features are not
present. Factors that weigh in favor of the diagnosis include female
gender; predominant aminotransferase elevation; presence and level
of globulin elevation; presence of nuclear, smooth-muscle, LKM1, and
other autoantibodies; concurrent other autoimmune diseases; charac­
teristic histologic features (interface hepatitis, plasma cells, rosettes);
HLA-DR3 or DR4 markers; and response to treatment (see below). A
more simplified, more specific scoring system relies on four variables:
presence of autoantibodies (ANA, ASMA, anti-SLA or LKM), elevated
serum IgG level, typical or compatible histologic features, and absence
of viral hepatitis markers. Elevated aminotransferase levels are also
required for the diagnosis of AIH. Weighing against the diagnosis are
predominant alkaline phosphatase elevation, mitochondrial antibod­
ies, markers of viral hepatitis, history of hepatotoxic drugs or excessive
alcohol, histologic evidence of bile duct injury, or such atypical histo­
logic features as fatty infiltration, iron overload, and viral inclusions.
■
■DIFFERENTIAL DIAGNOSIS
Early during the course of chronic hepatitis, AIH may resemble typical
acute viral hepatitis or drug-induced liver injury (Chap. 350). Exclu­
sion of acute viral hepatitis with serologic testing is critical. In adoles­
cence, Wilson disease (Chaps. 355 and 427) may present with features
of chronic hepatitis long before neurologic manifestations become
apparent and before the formation of Kayser-Fleischer rings (copper
deposition in Descemet’s membrane in the periphery of the cornea).
In this age group, serum ceruloplasmin and serum and urinary copper
determinations plus measurement of liver copper levels establish the
correct diagnosis. Cryptogenic cirrhosis and primary biliary cholan­
gitis (Chap. 355) share clinical features with AIH, and both alcoholassociated hepatitis (Chap. 353) and metabolic dysfunction–associated
steatohepatitis (MASH; formerly nonalcoholic steatohepatitis [NASH],
Chap. 354) may present with many features common to AIH; his­
toric, biochemical, serologic, and histologic assessments are usually
sufficient to allow these entities to be distinguished from AIH. The
distinction between autoimmune and chronic viral hepatitis may not
be straightforward, especially when viral antibodies occur in patients
with autoimmune disease or when autoantibodies occur in patients
with viral disease. Furthermore, the presence of extrahepatic features
such as arthritis, cutaneous vasculitis, or pleuritis—not to mention
the presence of circulating autoantibodies—may cause confusion with
rheumatologic disorders such as rheumatoid arthritis and systemic LE.
The existence of clinical and biochemical features of progressive necro­
inflammatory liver disease distinguishes chronic hepatitis from these
other disorders, which are not associated with severe liver disease.
Rarely, hepatic venous outflow obstruction (Budd-Chiari syndrome)
may present with features suggestive of AIH, but painful hepatomegaly,
ascites, and vascular imaging provide distinguishing diagnostic clues.
Other diagnostic considerations would include celiac disease and isch­
emic liver disease, which would be readily distinguishable by clinical
and laboratory features from AIH.
CHAPTER 352
Chronic Hepatitis
In patients treated with immune checkpoint inhibitors for malig­
nancy, the liver may be one of the autoimmune targets of therapy; the
syndrome of immune checkpoint inhibitor (ICI) hepatitis resembles
AIH in clinical features (without the female preponderance, how­
ever) and response to glucocorticoid-based treatment; however,
patients with ICI are less likely to have ANA, which, if present, is
low-titer and unlikely to have plasma cell inflammatory infiltrates
histologically. Similar to de novo AIH, ICI hepatitis encompasses a
spectrum of severity, and some cases may be refractory to standard
therapeutics. Finally, occasionally, features of AIH overlap with
features of autoimmune biliary disorders such as primary biliary
cholangitis, primary sclerosing cholangitis (Chaps. 355 and 357),
or, even more rarely, mitochondrial antibody-negative autoimmune
cholangitis. Such overlap syndromes are difficult to categorize, and
often response to therapy may be the distinguishing factor that estab­
lishes the diagnosis.
TREATMENT
Autoimmune Hepatitis
The mainstay of management in AIH is glucocorticoid therapy,
which is highly effective for the induction of remission. Several
controlled clinical trials have documented that such therapy leads
to symptomatic, clinical, biochemical, and histologic improvement
as well as increased survival. A therapeutic response to glucocorti­
coids can be expected in up to 80% of patients, and recent consensus
opinion defines response as a >50% reduction in ALT by 4 weeks
of therapy; however, a dramatic reduction within days is uncom­
mon after the initiation of corticosteroids. Although some advocate
the use of prednisolone (the hepatic metabolite of prednisone),
prednisone is just as effective and is favored by most authorities. A
popular regimen in the United States relies on an initiation dose of
60 mg/d. This high dose is tapered successively over the course of
a month down to a goal of 10–20 mg/d. In the European literature,
successful treatment for AIH has been achieved with a prednisone
starting dose of 20 mg. For patients with a mild presentation,
budesonide, while inferior to prednisone in AIH, is an alternative
first-line corticosteroid with limited systemic effects related to its
extensive hepatic first-pass metabolism that achieves similar rates
of biochemical response in patients with aminotransferase levels
<2× the upper limit of normal.
PART 10
Disorders of the Gastrointestinal System
After an early response to corticosteroids is observed, a steroidsparing agent should be initiated to maintain remission. Azathio­
prine, considered first-line treatment for nonicteric or severe AIH,
should be introduced 7–14 days after corticosteroids are started
and a response is observed. Testing for thiopurine S-methyltrans­
ferase activity is recommended during this period prior to initiation
of azathioprine. Dosing of azathioprine may be between 50 and 150 mg,
and some experts titrate the dose upward after starting at 50 mg.
Once on maintenance azathioprine (typically 1–2 mg/kg per day),
prednisone is tapered further to a maintenance dose of 5–10 mg. If
the patient achieves full biochemical remission, an attempt to taper
prednisone completely is reasonable.
The advantage of this combination approach is a reduction, over
the span of an 18-month course of therapy, in serious, life-threat­
ening complications of steroid therapy (e.g., cushingoid features,
hypertension, diabetes, osteoporosis) from 66% down to <20%.
Improvement of fatigue, anorexia, malaise, and jaundice tends to
occur within days to several weeks; biochemical improvement occurs
over the course of several weeks to months, with a fall in serum bili­
rubin and globulin levels and an increase in serum albumin. Serum
aminotransferase levels usually drop promptly, but improvements in
AST and ALT alone do not appear to be reliable markers of recovery
in individual patients; histologic improvement, characterized by a
decrease in mononuclear infiltration and in hepatocellular necrosis,
may be delayed for 6−24 months. Still, if interpreted cautiously, ami­
notransferase levels, and immunoglobulin levels in some, are valuable
indicators of relative disease activity, and although recommended,
many authorities do not advocate for serial liver biopsies to assess
therapeutic success or to guide decisions to alter or stop therapy.
Therapy should continue for at least 24 months after biochemical
remission. After tapering and cessation of therapy, the likelihood
of relapse is at least 50%, and in one multicenter study, the rate of
relapse was 81% after 3 years. In fact, most (95% of) patients will
require treatment at maintenance doses indefinitely. Patients who
relapse after completing one course of therapy almost always relapse
after subsequent courses; because each relapse increases the risk of
cirrhosis, death, and need for liver transplantation, ideally, patients
who relapse once should be treated indefinitely.
If azathioprine is not tolerated because of adverse effects (such
as marrow suppression, fatigue, or gastrointestinal symptoms), both
6-mercaptopurine (6MP) and mycophenolate mofetil (MMF) are
considered second-line alternative agents; if azathioprine is ineffec­
tive, MMF (1.5–2.0 g/d in divided doses) is the next drug of choice.
While MMF has yet to be compared to azathioprine in a random­
ized controlled clinical trial, in a recent prospective study, MMF
was at least as effective, if not more so, than azathioprine in achiev­
ing biochemical remission and may be better tolerated. A downside
of MMF that merits consideration is the risk of teratogenicity.
In cases of AIH refractory to standard treatment (azathioprine,
6-MP, or MMF, with or without prednisone) and in whom coex­
isting other causes for liver disease have been excluded, several
additional treatments are options. Tacrolimus has been shown to be
effective. Small studies of tacrolimus have shown a decline in ami­
notransferase levels in up to 80% after 3 months of therapy; how­
ever, tacrolimus requires careful trough monitoring and comes with
a risk of nephrotoxicity. Treatment with monoclonal anti-CD20
(rituximab), described initially in a small pediatric case series,
is another promising option that has been used with increasing
frequency in refractory AIH. In one cohort study, rituximab given
to patients with refractory AIH (1000-mg IV infusion given twice,
2 weeks apart) achieved remission rates of 95% at 1 year and 71%
at 2 years and permitted substantial corticosteroid dose reduction.
When medical therapy fails, including in severe acute AIH
refractory to high-dose corticosteroids or when AIH progresses
to cirrhosis and end-stage liver disease, liver transplantation is the
only recourse (Chap. 356); in patients with severe AIH, failure of
the bilirubin to improve after 2 weeks of therapy should prompt
early consideration of the patient for liver transplantation. Recur­
rence of AIH in the new liver occurs rarely in most experiences but
in as many as 35−40% of cases in others; nonetheless, 5-year patient
and graft survival exceed 80%.
Like all patients with chronic liver disease, patients with AIH
should be vaccinated against hepatitis A and B, ideally before
immunosuppressive therapy is begun, if practical. Similarly, age-
and gender-appropriate vaccinations, such as human papillomavi­
rus, varicella-zoster virus, pneumococcal, and COVID-19, should
be administered prior to immunosuppressive therapy. Patients with
AIH and cirrhosis should be screened for HCC with ultrasound at
6-month intervals and may merit screening for varices with upper
gastrointestinal endoscopy, based on the severity of liver disease
(Chap. 355).
■
■FURTHER READING
AASLD/IDSA HCV Guidance Panel: Hepatitis C guidance 2019
update: American Association for the Study of Liver Diseases–
Infectious Diseases Society of America recommendations for test­
ing, managing, and treating hepatitis C virus infection. Hepatology
71:686, 2020. Updated regularly and available at http://www.hcvguide­
lines.org. Accessed April 20, 2020.
Asselah T, Rizzetto M: Hepatitis D virus infection. N Engl J Med
389:58, 2023.
Bhattacharya D  et al: Hepatitis C guidance 2023 update: American
Association for the Study of Liver Diseases-Infectious Diseases Soci­
ety Recommendations for testing, managing, and treating hepatitis C
virus infection. Clin Infect Dis 2023.
Buti M et al: Tenofovir alafenamide versus tenofovir disproxil fuma­
rate for the treatment of HBeAg-negative chronic hepatitis B virus
infection: A randomized, double-blind, phase 3 non-inferiority trial.
Lance Gastroenterol Hepatol 1:196, 2017.
Carrat F et al: Clinical outcomes in patients with chronic hepatitis C
after direct-acting antiviral treatments: A prospective cohort study.
Lancet 393:1453, 2019.
Chan HLY et al: Tenofovir alafenamide versus tenofovir disproxil
fumarate for the treatment of HBeAg-positive chronic hepatitis B

25 - 353 Alcohol-Associated Liver Disease
353 Alcohol-Associated Liver Disease
virus infection: A randomized, double-blind, phase 3 non-inferiority
trial. Lancet Gastroenterol Hepatol 1:185, 2017.
Dalekos GN  et al: Long-term results of mycophenolate mofeteil vs.
azathioprine use in individuals in autoimmune hepatitis. J Hep Rep
4:100601, 2022.
Dougan M et al: AGA clinical practice update on diagnosis and man­
agement of immune checkpoint inhibitor colitis and hepatitis: Expert
review. Gastroenterology 160:1384, 2021.
Dusheiko G  et al: New approaches to chronic hepatitis B. N Engl J
Med 388:55, 2023.
European Association for the Study of the Liver: EASL 2017
clinical practice guidelines on the management of hepatitis B virus
infection. J Hepatol 67:370, 2017.
European Association for the Study of the Liver: EASL recom­
mendations on treatment of hepatitis C: Final update of the series.
J Hepatol 73:1170, 2020.
Forns X et al: Glecaprevir plus pibrentasvir for chronic hepatitis C
virus genotype 1, 2, 4, 5, or 6 infection in adults with compensated
cirrhosis (EXPEDITION-1): A single-arm, open-label, multicentre
phase 3 trial. Lancet Infect Dis 17:1062, 2017.
Heller T et al: Hepatitis D: Looking back, looking forward, seeing the
reward and the promise. Clin Gastroenterol Hepatol 21:2051, 2023.
Hirode G et al: Off-therapy response after nucleos(t)ide analogue
withdrawal in patients with chronic hepatitis B: An international,
multicenter, multiethnic cohort (RETRACT-B Study). Gastroenterol­
ogy 162:757, 2022.
Jacobson IM et al: American Gastroenterological Association Institute
clinical practice update-expert review: Care of patients who have
achieved a sustained virologic response after antiviral therapy for
chronic hepatitis C infection. Gastroenterology 152:1578, 2017.
Jeng W-J et al: Hepatitis B. Lancet 401:1039, 2023.
Kwo PY et al: Glecaprevir and pibrentasvir yield high response rates in
patients with HCV genotype 1-6 without cirrhosis. J Hepatol 67:263, 2017.
Lampertico P  et al: Hepatitis D virus infection: Pathophysiology,
epidemiology and treatment. Report from the first international delta
cure meeting 2022. JHEP Rep 5:100818, 2023.
Liem KS et al: Limited sustained response after stopping nucleos(t)ide
analogues in patients with chronic hepatitis B: Results from a ran­
domized controlled trial (Toronto STOP study). Gut 68:2206, 2019.
Litwin AH et al: Patient-centred models of hepatitis C treatment for
people who inject drugs: a multicentre, pragmatic randomized trial.
Lancet Gastroenterol Hepatol 7:1112, 2022.
Lok ASG et al: Antiviral therapy for chronic hepatitis B viral infection
in adults: A systematic review and meta-analysis. Hepatology 63:284,
2016.
Mack CL et al: Diagnosis and management of autoimmune hepatitis
in adults and children: 2019 practice guidance and guidelines from
the American Association for the Study of Liver Diseases. Hepatology
72:671, 2020.
Martinello M  et al: Hepatitis C. Lancet 402:1085, 2023.
Mohareb AM  et al: Clearance of hepatitis B e antigen in untreated
chronic hepatitis B virus infection: A systematic review and metaanalysis. J Infect Dis 11:761, 2022.
Negro F, Lok AS: Hepatitis D: A review. JAMA 330:2376, 2023.
Papatheodoridis GV et al: DARING-B: Discontinuation of effective
entecavir or tenofovir disoporxil fumarate long-term therapy before
HBsAg loss in non-cirrhotic HBeAg-negative chronic hepatitis B.
Antivir Ther 23:677, 2018.
Papatheodoridis GV et al: Eight-year survival in chronic hepatitis
B patients under long-term entecavir or tenofovir is similar to the
general population. J Hepatol 68:1129, 2018.
Pawlotsky J-M et al: From non-A, non-B hepatitis to hepatitis C virus
cure. J Hepatol 62:S87, 2015.
Perrillo RP et al: American Gastroenterological Association Institute
technical review on prevention and treatment of hepatitis B virus
reactivation during immunosuppressive drug therapy. Gastroenterol­
ogy 148:221, 2015.
Reddy KR et al: American Gastroenterological Association Insti­
tute guideline on the prevention and treatment of hepatitis B virus
reactivation during immunosuppressive drug therapy. Gastroenterol­
ogy 148:215, 2015.
Rizzetto M et al: The changing context of hepatitis D. J Hepatol
74:1200, 2021.
Romée JALM et al: An open-label randomised-controlled trial of aza­
thioprine vs. mycophenolate mofetil for the induction of remission in
treatment-naive autoimmune hepatitis. J Hepatol 80:576, 2024.
Rossi C et al: Sustained virologic response from interferon-based
hepatitis C regimens is associated with reduced risk of extrahepatic
manifestations. J Hepatol 71:1116, 2019.
Singal AG et al: AGA clinical practice update on interaction between
oral direct-acting antivirals for chronic hepatitis C infection and
hepatocellular carcinoma: Expert review. Gastroenterology 156:2149,
2019.
Sgamato C  et al: Autoimmune liver disease and SARS-CoV-2. World
J Gastroenterol 29:1838, 2023.
Terrault N et al: Update on prevention, diagnosis, and treatment of
chronic hepatitis B: AASLD 2018 hepatitis B guidance. Hepatology
67:1560, 2018.
Thomas DL: Global elimination of chronic hepatitis. N Engl J Med
380:2041, 2019.
Van Den Brand FF et al: Increased mortality among patients with
vs without cirrhosis and autoimmune hepatitis. Clin Gastroenterol
Hepatol 17:940, 2019.
Wedemeyer H  et al: A phase 3, randomized trial of bulevirtide in
chronic hepatitis D. N Engl J Med 389:22, 2023.
Yardeni D et al: Current best practice in hepatitis B management
and understanding long-term prospects for cure. Gastroenterology
164:42, 2023.
CHAPTER 353
Alcohol-Associated Liver Disease
Bernd Schnabl
Alcohol-Associated
Liver Disease
Alcohol-associated liver diseases (ALD) comprise a spectrum of dis­
eases associated with chronic alcohol consumption ranging from alcoholassociated fatty liver disease and steatohepatitis to more advanced
liver disease including fibrosis and cirrhosis. Acute alcohol-associated
hepatitis is an acute-on-chronic form of ALD that is associated with
liver failure and high mortality.
■
■EPIDEMIOLOGY
Approximately 7% of adults in the United States meet criteria for
unhealthy drinking, defined as ≥2 drinks per day in women and
≥3 drinks per day in men, or partake in heavy episodic drinking,
defined as ≥4 drinks for women and ≥5 drinks for men on a single
occasion (1 drink equals ~14 g of ethanol, which is 1 beer, 4 oz of wine,
or 1 oz of 80% spirits). The COVID-19 pandemic increased excessive
alcohol intake, resulting in a rise in morbidity and mortality linked to
ALD. Prevalence of ALD correlates with the amount of alcohol con­
sumption in different regions. Prevalence of alcohol-associated fatty
liver disease is 4.7% of the general population in the United States, and
1.5% has stage 2 or greater fibrosis. Liver cirrhosis is the eleventh lead­
ing cause of mortality worldwide, causing 1.16 million deaths annually;
48% of cases of cirrhosis can be attributed to alcohol. Among patients
with alcohol misuse, 18% had fibrosis, 26% had cirrhosis, and 7% had
acute alcohol-associated hepatitis without underlying cirrhosis. In the
European population, the annual incidence rate for acute alcoholassociated hepatitis is between 24 and 27 per million persons in women
and between 46 and 65 per million persons in men.
■
■PATHOGENESIS
Alcohol in the form of ethanol is rapidly absorbed in the upper gas­
trointestinal tract and predominantly metabolized in the liver. Ethanol
reaches the liver through the portal vein, and the majority of ethanol
is oxidized via alcohol dehydrogenase 1 (ADH1) into acetaldehyde
in hepatocytes. Chronic alcohol consumption induces the expression
of a second ethanol-metabolizing enzyme, cytochrome P450 family 2
subfamily E member 1 (CYP2E1), which also converts ethanol into
acetaldehyde. In addition to the direct cellular toxic effects of acetal­
dehyde, metabolism of ethanol into acetaldehyde causes the generation
of reactive oxygen species (ROS), resulting in further injury of hepa­
tocytes via lipid peroxidation and DNA damage. Acetaldehyde is
then oxidized into acetate via acetaldehyde dehydrogenase (ALDH).
Inherited deficiency of ALDH2 is common in Asian countries and
leads to acetaldehyde accumulation after alcohol consumption. These
individuals develop nausea and cutaneous flushing. Several mecha­
nisms contribute to the development of hepatic steatosis related to
alcohol consumption. Acetate is converted into acetyl-coenzyme A
(CoA), which contributes to fatty acid and triglyceride synthesis.
Alcohol, in part through epigenetic changes, increases the expression
of genes involved in lipogenesis, while genes involved in fatty acid
transport and oxidation are suppressed. Alcohol also increases the ratio
of reduced nicotinamide adenine dinucleotide (NAD)/oxidized NAD
(NADH/NAD+) in hepatocytes, which further reduces mitochondrial
β-oxidation. Alcohol can increase fatty acid mobilization in adipose
tissue and the intestine, which will lead to hepatic accumulation of fatty
acids and increased hepatic steatosis. Overall, the net effect of these
processes contributes to accumulation of lipids in the liver.
■
■RISK FACTORS FOR PROGRESSION OF ALD
Daily alcohol consumption or heavy drinking results in hepatic steato­
sis, but only 10–20% of such individuals will develop progressive liver
disease and cirrhosis. Therefore, other cofactors such as behavioral,
environmental, and genetic factors play important roles in progression
of ALD (Table 353-1). There is a dose-dependent increase, with regard
to the amount of alcohol consumed, in the likelihood of developing
liver cirrhosis. Women develop ALD at a lower daily alcohol intake.
Cigarette smoking is an independent risk factor for alcohol-associated
cirrhosis. The drinking pattern, in particular binge drinking and
excessive alcohol drinking outside meals, increases the risk of develop­
ing progressive ALD. Obesity and metabolic dysfunction–associated
steatotic liver disease (MASLD) are frequent cofactors contributing to
progression of ALD. A distinct subset of patients within ALD, termed
as metabolic ALD (MetALD), now describes patients with MASLD
and increased alcohol consumption. Other chronic liver diseases
such as viral hepatitis and hemochromatosis can have synergistic
effects on ALD. Twin studies demonstrated a genetic predisposition
to alcohol-associated liver cirrhosis that is independent from the
genetic predisposition to alcohol use disorder. Gene polymorphisms
conferring increased risk of alcohol-associated liver cirrhosis have been
found in three genes, patatin-like phospholipase domain-containing 3
(PNPLA3), membrane bound O-acyltransferase domain-containing 7
(MBOAT7), and transmembrane 6 superfamily member 2 (TM6SF2),
although the molecular mechanism is not well understood. Genetic
variants in hydroxysteroid 17-beta dehydrogenase 13 (HSD17B13) and
Fas-associated factor family member 2 (FAF2) are associated with a
PART 10
Disorders of the Gastrointestinal System
TABLE 353-1  Factors for Progression of Alcohol-Associated
Liver Disease
• Alcohol amount and duration
• Drinking pattern (drinking without meal, binge drinking)
• Genetic factors, especially PNPLA3 polymorphism
• Female sex
• Smoking
• Obesity and metabolic dysfunction–associated steatotic liver disease
(MASLD)
• Chronic liver diseases such as viral hepatitis and hemochromatosis
• Intestinal microbiota
TABLE 353-2  Symptoms and Signs Associated with Alcohol-Associated
Cirrhosis and Alcohol-Associated Hepatitis
• Tiredness
• Malnutrition and sarcopenia
• Abdomen: abdominal discomfort, hepatomegaly, splenomegaly, caput
medusae, ascites with weight gain, abdominal pain, and shortness of breath
• Skin: spider angioma, palmar erythema, jaundice, ecchymoses
• Eyes: icteric sclerae
• Hands: Dupuytren contracture
• Extremities: edema
• Face: rhinophyma
• Reproductive system: gynecomastia, gonadal atrophy, loss of libido,
amenorrhea
• Neurologic:
• Peripheral neuropathy
• Alcohol withdrawal: tachycardia, agitation, tremor, seizures, delirium
• Hepatic encephalopathy: asterixis (flapping tremor), forgetfulness,
inversion of sleep/wake pattern, altered consciousness, confusion,
lethargy, coma
• Wernicke-Korsakoff syndrome
reduced risk of developing ALD. A subset of patients with alcohol use
disorder develop changes in the gut microbiome and increased intes­
tinal permeability. This leads to the passage of microbial components,
like bacterial lipopolysaccharide (LPS), through the portal vein into the
liver resulting in hepatic inflammation, hepatocyte death, and activa­
tion of fibrotic pathways. Ongoing fibrosis resulting from sustained
alcohol consumption progresses to cirrhosis accompanied by portal
hypertension. Impaired liver regeneration might contribute to the
acute onset of alcohol-associated-hepatitis in patients with underlying
cirrhosis (Chap. 355).
■
■CLINICAL FEATURES
The development of alcohol-associated steatosis, steatohepatitis, and
cirrhosis is most often clinically silent. Symptoms arise once the patient
with alcohol-associated liver cirrhosis decompensates or develops
alcohol-associated hepatitis (Table 353-2). Patients with alcoholassociated hepatitis have been drinking heavily (>40 g/d for women and
50–60 g/d for men) for >6 months with <60 days of abstinence before
onset of symptoms. They present with rapid onset of jaundice (serum
bilirubin >3 mg/dL), often accompanied by fever, malaise, tender hepa­
tomegaly, and clinical signs of hepatic decompensation, such as ascites,
bacterial infection, variceal bleeding, and hepatic encephalopathy.
Infections occur in 12–26% of patients with severe alcohol-associated
hepatitis at the time of admission. Alcohol-associated hepatitis is often
accompanied by systemic inflammatory response syndrome (SIRS) and
acute kidney injury (AKI) secondary to hepatorenal syndrome.
■
■LABORATORY FINDINGS
Patients with simple hepatic steatosis can present with normal liver
function tests. Steatohepatitis is characterized by elevated levels of
aspartate aminotransferase (AST) and γ-glutamyl transferase (GGT).
Characteristic laboratory parameters for ALD include a ratio of AST
to alanine aminotransferase (ALT) of >1, and serum AST is rarely
300 IU/L. Serum bilirubin and international normalized ratio (INR)
are typically normal. Elevated bilirubin and INR and low serum albu­
min and platelet count are common laboratory finings in patients with
cirrhosis. Patients with alcohol-associated hepatitis have AST and ALT
elevations that do not exceed 400 IU/L, with AST/ALT ratio of >1.5 and
serum bilirubin >3 mg/dL.
■
■DIAGNOSIS
The Alcohol Use Disorders Inventory Test (AUDIT) is a validated tool
for identifying patients with alcohol use disorder (Chap. 464). Diagno­
sis of ALD requires exclusion of other liver diseases in heavy drinkers.
Alcohol-associated steatosis can be diagnosed by simple ultrasound,
magnetic resonance imaging (MRI), or computed tomography (CT).
Noninvasive quantification of hepatic fat can be achieved with the
ultrasound technique of controlled attenuation parameter (CAP) or
with magnetic resonance proton density fat fraction (MR-PDFF). Liver
biopsy is rarely indicated for diagnosing alcohol-associated hepatic ste­
atosis or steatohepatitis. Liver biopsy typically shows hepatocytes with
large lipid droplets (macrovesicular steatosis) around pericentral veins
(zone 3). Morphologic features of alcohol-associated steatohepatitis
include hepatocyte injury and ballooning with Mallory-Denk bodies,
necrosis, and lobular inflammation with mononuclear and neutro­
philic granulocytes.
Progression of alcohol-associated steatohepatitis to fibrosis can be
diagnosed using liver stiffness measurement by techniques such as
transient elastography (e.g., FibroScan). Liver stiffness <6 kPa indi­
cates normal liver, whereas cutoffs for each stage of alcohol-associated
liver fibrosis have been validated (>8 kPa indicates ≥F3 advanced
fibrosis; >12.5 kPa indicates F4 cirrhosis). Fibrosis-4 (Fib-4) score is
a serum marker test based on age, AST, ALT, and platelets and can
be used to exclude advanced fibrosis or cirrhosis at a threshold value
<3.25. Histology shows initially perivenular fibrosis with subsequent
extension of collagen fibers into hepatic lobules, described as septal
fibrosis. Patients with cirrhosis show liver nodularity on imaging with
ultrasound, MRI, or CT scan. Radiologic signs of portal hypertension
include ascites, splenomegaly, and portal-systemic collateral vessels.
Prognosis and risk of mortality are assessed using Child-Pugh-Turcotte
(CPT) or Model for End-Stage Liver Disease (MELD; or sodiumMELD) scores (Chap. 355).
In patients presenting with features suggestive of alcohol-associated
hepatitis, imaging is obtained to exclude biliary obstruction and hepa­
tocellular carcinoma (HCC). In addition, other causes of liver disease
such as viral hepatitis, Wilson’s disease, and severe autoimmune liver
disease should be ruled out. Histology shows macrovesicular steatosis,
hepatocyte ballooning with Mallory-Denk bodies, megamitochondria,
neutrophil infiltration, ductular reaction, bilirubinostasis, and chicken
wire fibrosis. The majority of patients with alcohol-associated hepatitis
have underlying cirrhosis (80%) (Chap. 355), and 10–20% of patients
with a clinical diagnosis of alcohol-associated hepatitis will have other
liver diseases on biopsy. Therefore, in the presence of potential con­
founding factors, including possible ischemic hepatitis (in the setting
of, e.g., hypotension, massive gastrointestinal bleeding, recent cocaine
use, septic shock), drug-induced liver injury (DILI), autoimmune liver
disease, uncertain alcohol use assessment, or atypical laboratory tests
(AST <50 IU/L or >400 IU/L, AST/ALT ratio <1.5), a transjugular liver
biopsy is recommended to confirm the diagnosis of alcohol-associated
hepatitis. Infections need to be assessed routinely with chest x-ray and
blood, urine, and ascites cultures in patients presenting with alcoholassociated hepatitis.
TREATMENT
Alcohol-Associated Liver Disease (Fig. 353-1)
To date, the most effective therapy to reduce the progression of and
reverse ALD is prolonged alcohol abstinence. In particular, alcoholassociated hepatic steatosis and steatohepatitis are reversible with
cessation of alcohol consumption. Thus, treatment of the underlying
alcohol use disorder is an integral part for therapy of ALD. There
are currently no approved drugs for treatment of alcohol-associated
steatosis and steatohepatitis with or without fibrosis.
Patients with alcohol-associated cirrhosis and ongoing alcohol
consumption are at risk for decompensation and development of
hepatic encephalopathy, ascites, variceal bleeding, hepatorenal
syndrome, and HCC (Chap. 355). Patients with cirrhosis should
undergo an upper gastrointestinal endoscopy to screen for varices.
HCC screening is recommended using liver ultrasonography and
serum α-fetoprotein (AFP) every 6 months in patients with cir­
rhosis. Management of complications of cirrhosis such as variceal
bleeding, ascites, hepatic encephalopathy, and HCC does not
differ from patients with cirrhosis due to a different etiology
(Chap. 355). Liver transplantation for patients with alcoholassociated decompensated cirrhosis or HCC is a definitive
Alcohol-Associated
Hepatitis (AH)
Clinical diagnosis with
laboratory findings
Confounding
diagnostic
factors
Moderate AH
MDF <32
or MELD ≤20
Severe AH
MDF ≥32
or MELD >20
TJ liver biopsy
Alcohol
abstinence
Nutritional
support
Oral prednisolone 40 mg/d
(unable to take oral medications:
methylprednisolone 32 mg/d IV)
Contraindications
for corticosteroids
7 days
Lille score <0.45
Lille score ≥0.45
Continue prednisolone
for 28 days total
Stop prednisolone
If eligible, early liver transplantation (LT)
If not eligible for LT: Supportive/palliative care
CHAPTER 353
FIGURE 353-1  Treatment algorithm for alcohol-associated hepatitis. In patients
with a clinical diagnosis of alcohol-associated hepatitis, confounding factors (see
text) need to be ruled out, if necessary, by transjugular (TJ) liver biopsy. Patients
with severe alcohol-associated hepatitis (AH), defined as Maddrey discriminant
function (MDF) ≥32 or Model for End-Stage Liver Disease (MELD) score >20,
without contraindications for glucocorticoids (see text) are candidates for such
treatment. Nonresponders or patients with contraindications for treatment should
be considered for early liver transplantation (LT) or supportive or palliative care, as
clinically appropriate.
Alcohol-Associated Liver Disease
therapy and is currently the leading indication for liver trans­
plantation in the United States. Liver transplantation evaluation
should be taken into consideration for patients with end-stage
liver disease (Chap. 356).
In patients diagnosed with alcohol-associated hepatitis (Fig. 353-1),
short-term mortality can be predicted using the Maddrey discrimi­
nant function (MDF; calculated as 4.6 × [the prolongation of the
prothrombin time above control {seconds}] + serum bilirubin [mg/dL]),
MELD score (Chap. 355), or age-bilirubin-INR-creatinine (ABIC)
score. Patients with MDF <32 or MELD ≤20 are defined as hav­
ing moderate alcohol-associated hepatitis. Currently, patients with
moderate alcohol-associated hepatitis are treated under a multidis­
ciplinary team including an alcohol use disorder specialist, dietitian
for nutritional supplementation for patients with markedly reduced
intake, and hepatologist for managing liver disease complications.
Enteral nutrition with a goal of 35–40 kcal/kg and supplementation
of micronutrients (in particular zinc) and vitamin supplementation
(in particular vitamin B1 and K) are recommended for patients
with alcohol-associated hepatitis. Intravenous albumin is preferred
for volume expansion. Patients with moderate alcohol-associated
hepatitis have a 20% risk of 1-year mortality. MDF ≥32 or MELD
20 identifies patients with severe alcohol-associated hepatitis and
high short-term mortality of ~30% at 3 months, who will have a
survival benefit with glucocorticoid treatment. Contraindications
for glucocorticoid treatment include uncontrolled infections or
sepsis, AKI and hepatorenal syndrome, uncontrolled upper gastro­
intestinal bleeding, concomitant diseases (including viral hepatitis,
HCC, pancreatitis, DILI, active tuberculosis, and HIV), multior­
gan failure, and shock. Glucocorticoids can be used once infec­
tion, sepsis, and gastrointestinal bleeding are adequately controlled.

27 - 355 Cirrhosis and Its Complications
355 Cirrhosis and Its Complications
Cusi K et al: American Association of Clinical Endocrinology clinical
practice guideline for the diagnosis and management of nonalcoholic
fatty liver disease in primary care and endocrinology clinical settings:
Co-sponsored by the American Association for the Study of Liver
Diseases (AASLD). Endocr Pract 28:528, 2022.
Diehl AM, Day CSC: Cause, pathogenesis, and treatment of nonalco­
holic steatohepatitis. N Engl J Med 377:2063, 2017.
European Association for the Study of the Liver (EASL) et al:
EASL-EASD-EASO Clinical Practice Guidelines for the management
of non-alcoholic fatty liver disease. J Hepatol 64:1388, 2016.
Harrison SA et al: A phase 3, randomized, controlled trial of resmeti­
rom in NASH with liver fibrosis. N Engl J Med 390:497, 2024.
Kanwal F et al: Clinical care pathway for the risk stratification and
management of patients with nonalcoholic fatty liver disease. Gastro­
enterology 161:1657, 2021.
Rinella ME et al: AASLD practice guidance on the clinical assessment
and management of nonalcoholic fatty liver disease. Hepatology
77:1797, 2023.
Vos MB et al: NASPGHAN clinical practice guideline for the diagnosis
and treatment of nonalcoholic fatty liver disease in children: Recom­
mendations from the Expert Committee on NAFLD (ECON) and the
North American Society of Pediatric Gastroenterology, Hepatology
and Nutrition (NASPGHAN). J Pediatr Gastroenterol Nutr 64:319,
2017.
Younossi ZM et al: The global epidemiology of nonalcoholic fatty
liver disease (NAFLD) and nonalcoholic steatohepatitis (NASH): A
systematic review. Hepatology 77:1335, 2023.
PART 10
Disorders of the Gastrointestinal System
Alex S. Befeler, Bruce R. Bacon
Cirrhosis and Its
Complications
Cirrhosis is a condition that is defined histopathologically and has a
variety of clinical manifestations and complications, some of which can
be life-threatening. In the past, it has been thought that cirrhosis was
never reversible; however, it has become apparent that when the under­
lying insult that has caused the cirrhosis has been removed, there can
be reversal of fibrosis. This is most apparent with the successful treat­
ment of chronic hepatitis C; however, reversal of fibrosis is also seen in
patients with hemochromatosis after iron removal and in patients with
alcohol-associated liver disease who have discontinued alcohol use.
Regardless of the cause of cirrhosis, the pathologic features consist
of the development of fibrosis to the point that there is architectural
distortion with the formation of regenerative nodules. This results in a
decrease in hepatocellular mass, and thus function, and an alteration of
blood flow. The induction of fibrosis occurs with activation of hepatic
stellate cells, resulting in the formation of increased amounts of col­
lagen and other components of the extracellular matrix.
Clinical features of cirrhosis are the result of pathologic changes and
mirror the severity of the liver disease. Most hepatic pathologists pro­
vide an assessment of grading and staging when evaluating liver biopsy
samples. These schemes vary between disease states and have been
developed for most conditions, including chronic viral hepatitis, non­
alcoholic fatty liver disease, and primary biliary cholangitis. Advanced
fibrosis usually includes bridging fibrosis with nodularity designated as
stage 3 and cirrhosis designated as stage 4. Patients who have cirrhosis
have varying degrees of liver function, and clinicians need to differen­
tiate between those who have stable, compensated cirrhosis and those
who have decompensated cirrhosis. Patients who have developed
ascites, hepatic encephalopathy, or variceal bleeding are classified as
TABLE 355-1  Causes of Cirrhosis
Alcohol
Cardiac cirrhosis
Chronic viral hepatitis
Inherited metabolic liver disease
  Hepatitis B
  Hemochromatosis
  Hepatitis C
  Wilson’s disease
Autoimmune hepatitis
  α1 Antitrypsin deficiency
Metabolic dysfunction-associated
steatohepatitis
  Cystic fibrosis
Biliary cirrhosis
Cryptogenic cirrhosis
  Primary biliary cholangitis
 
  Primary sclerosing cholangitis
  Autoimmune cholangiopathy
 
decompensated. They should be considered for liver transplantation,
particularly if the decompensations are poorly controlled. Many of the
complications of cirrhosis will require specific therapy. Portal hypertension
is a significant complicating feature of decompensated cirrhosis and is
responsible for the development of ascites and bleeding from esopha­
gogastric varices. Loss of hepatocellular function results in jaundice,
coagulation disorders, and hypoalbuminemia and contributes to the
causes of portosystemic encephalopathy. The complications of cirrho­
sis are basically the same regardless of the etiology. Nonetheless, it is
useful to classify patients by the cause of their liver disease (Table 355-1);
patients can be divided into broad groups, including those with alcoholassociated cirrhosis, cirrhosis due to chronic viral hepatitis, biliary
cirrhosis, metabolic dysfunction–associated steatotic liver disease (the
new consensus term for nonalcoholic fatty liver disease), and other, less
common causes, such as cardiac cirrhosis, cryptogenic cirrhosis, and
other miscellaneous causes.
ALCOHOL-ASSOCIATED CIRRHOSIS
Excessive chronic alcohol use can cause several different types of
chronic liver disease, including alcohol-associated fatty liver, alcoholic
hepatitis, and alcohol-associated cirrhosis. Furthermore, use of exces­
sive alcohol can contribute to liver damage in patients with other
liver diseases, such as hepatitis C, hemochromatosis, and metabolic
dysfunction–associated steatotic liver disease. Chronic alcohol use can
produce fibrosis in the absence of accompanying inflammation and/
or necrosis. Fibrosis can be centrilobular, pericellular, or periportal.
When fibrosis reaches a certain degree, there is disruption of the nor­
mal liver architecture and replacement of liver cells by regenerative
nodules. In alcohol-associated cirrhosis, the nodules are usually <3 mm
in diameter; this form of cirrhosis is referred to as micronodular. With
cessation of alcohol use, larger nodules may form, resulting in a mixed
micronodular and macronodular cirrhosis.
Pathogenesis 
Alcohol is the most commonly used drug in the
United States, and >70% of adults drink alcohol each year. Twenty
percent have had a binge within the past month, and >7% of adults
regularly consume more than four or five drinks five or more times
a month. Unfortunately, >14 million adults in the United States meet
the diagnostic criteria for alcohol use disorder. In the United States,
chronic liver disease is the tenth most common cause of death in
adults, and alcohol-associated cirrhosis accounts for ~48% of deaths
due to cirrhosis.
Ethanol is mainly absorbed by the small intestine and, to a lesser
degree, through the stomach. Gastric alcohol dehydrogenase (ADH)
initiates alcohol metabolism. Three enzyme systems account for
metabolism of alcohol in the liver. These include cytosolic ADH, the
microsomal ethanol oxidizing system (MEOS) utilizing the induc­
ible cytochrome P450 CYP2E1, and peroxisomal catalase. Normally
the majority of ethanol oxidation occurs via ADH to form acetaldehyde,
which is a highly reactive molecule that may have multiple effects. The
MEOS pathway in chronic alcohol use causes induction of CYP2E1,
which leads to generation of reactive oxygen species and produces
more acetaldehyde. Ultimately, acetaldehyde is metabolized to acetate
by aldehyde dehydrogenase (ALDH). Intake of ethanol increases intra­
cellular accumulation of triglycerides by increasing fatty acid uptake
and by reducing fatty acid oxidation and lipoprotein secretion. Protein
synthesis, glycosylation, and secretion are impaired. Oxidative damage
to hepatocyte membranes occurs due to the formation of reactive oxy­
gen species; acetaldehyde is a highly reactive molecule that combines
with proteins and nucleic acids to form acetaldehyde adducts. These
adducts may interfere with specific enzyme activities, including micro­
tubular formation and hepatic protein trafficking. With acetaldehydemediated hepatocyte damage, certain reactive oxygen species can result
in Kupffer cell activation. As a result, profibrogenic cytokines are pro­
duced that initiate and perpetuate stellate cell activation, with the resul­
tant production of excess collagen and extracellular matrix. Connective
tissue appears in both periportal and pericentral zones and eventually
connects portal triads with central veins forming regenerative nodules.
Hepatocyte loss occurs, and with increased collagen production and
deposition, together with continuing hepatocyte destruction, the liver
contracts and shrinks in size. This process generally takes from years
to decades to occur and requires repeated insults.
Clinical Features 
The diagnosis of alcohol-associated liver disease
requires an accurate history regarding both amount (>2 drinks per day
in women and >3 drinks per day in men) and duration (>12 months) of
alcohol consumption. Patients with alcohol-associated liver disease can
present with nonspecific symptoms such as vague right upper quad­
rant abdominal pain, fever, nausea and vomiting, diarrhea, anorexia,
and malaise. Alternatively, they may present with complications of
chronic liver disease, including ascites, edema, upper gastrointestinal
(GI) hemorrhage, jaundice, or encephalopathy. Many cases present
incidentally at the time of autopsy or elective surgery. Other patients
may be identified during an evaluation of routine laboratory studies
that are found to be abnormal. On physical examination, the liver and
spleen may be enlarged, with the liver edge being firm and nodular.
Other frequent findings include scleral icterus, palmar erythema
(Fig. 355-1), spider angiomas (Fig. 355-2), parotid gland enlargement,
digital clubbing, muscle wasting, edema, and ascites. Men may have
decreased body hair, gynecomastia, and testicular atrophy, which may
be a consequence of hormonal abnormalities or a direct toxic effect
of alcohol on the testes. In women with advanced disease, menstrual
irregularities usually occur including amenorrhea. These changes are
often reversible following cessation of alcohol ingestion.
Laboratory tests may be completely normal in patients with early
compensated alcohol-associated cirrhosis. Alternatively, in advanced
liver disease, many abnormalities usually are present. Patients may be
anemic from chronic GI blood loss, nutritional deficiencies, or hyper­
splenism or as a direct suppressive effect of alcohol on the bone marrow.
A unique form of hemolytic anemia (with spur cells and acanthocytes)
called Zieve’s syndrome can occur in patients with severe alcoholic
hepatitis. Platelet counts are often reduced early in the disease, reflec­
tive of portal hypertension with hypersplenism. Serum total bilirubin
FIGURE 355-1  Palmar erythema. This figure shows palmar erythema in a patient
with alcohol-associated cirrhosis. The erythema is peripheral over the palm with
central pallor.
FIGURE 355-2  Spider angioma. This figure shows a spider angioma in a patient with
hepatitis C cirrhosis. With release of central compression, the arteriole fills from the
center and spreads out peripherally.
can be normal or elevated with advanced disease. Prothrombin times
are often prolonged and usually do not respond to administration of
parenteral vitamin K. Serum sodium levels are usually normal unless
patients have ascites and then can be depressed, largely due to ingestion
of excess free water. Serum alanine and aspartate aminotransferases
(ALT, AST) are typically elevated but <400 IU/mL, with AST levels being
higher than ALT levels, usually by a 2:1 ratio, particularly in patients
who continue to drink.
CHAPTER 355
Diagnosis 
Patients who have any of the above-mentioned clinical
features, physical examination findings, or laboratory studies in the
setting of chronic alcohol consumption should be considered to have
alcohol-associated liver disease. Furthermore, other forms of chronic
liver disease (e.g., chronic viral hepatitis or metabolic or autoimmune
liver diseases) must be considered or ruled out, or if present, an esti­
mate of relative causality along with the alcohol use should be deter­
mined. Liver biopsy can be helpful to confirm a diagnosis but generally
is not performed unless there is a suspicion of an alternative diagnosis.
Cirrhosis and Its Complications
In patients who have had complications of cirrhosis and who con­
tinue to drink, there is a <50% 5-year survival. In contrast, in patients
who remain abstinent, the prognosis is significantly improved, particu­
larly when they have resolution of liver complications; however, some
individuals who remain abstinent do not improve and liver transplan­
tation is a viable option.
TREATMENT
Alcohol-Associated Cirrhosis and
Alcoholic Hepatitis
Abstinence is the cornerstone of therapy for patients with alcoholassociated liver disease. New clinically available biomarkers of
recent alcohol consumption such as phosphatidylethanol (Peth)
can be helpful in evaluating abstinence. In addition, patients require
good nutrition and long-term medical supervision to manage
underlying complications that may develop. Complications such
as the development of ascites and edema, variceal hemorrhage, or
portosystemic encephalopathy all require specific management and
treatment. Liver transplantation can be an effective long-term treat­
ment in those who have been deemed a low enough risk for alcohol
relapse and do not respond to other treatments.
Glucocorticoids are occasionally used in patients with severe
alcoholic hepatitis in the absence of infection. Short-term sur­
vival has been shown to be improved in certain studies and
meta-analysis, although 6-month survival is more dependent on
abstinence. Treatment is restricted to patients with a discriminant
function (DF) value of >32. The DF is calculated as the serum
total bilirubin plus the difference in the patient’s prothrombin time
compared to upper limit of control (in seconds) multiplied by 4.6.
Failure to improve total bilirubin after 7 days predicts treatment
failure, and glucocorticoids can be stopped; otherwise, they are
continued for 28 days.
There is modest evidence that intravenous N-acetylcysteine plus
glucocorticoids may have a short-term survival benefit in alcoholic
hepatitis if the DF is >32. Other therapies including oral pentoxifyl­
line, inhibitors of tumor necrosis factor (TNF) α, anabolic steroids,
propylthiouracil, antioxidants, colchicine, and penicillamine have
not shown clear-cut benefits and are not recommended. A variety
of nutritional therapies have been tried, both parenteral and enteral
feedings; however, there is no clear evidence of improved survival.
There is evidence that persons who consume >21.5 kcal/kg body
weight per day have better survival, so achieving better caloric
intake is recommended. Finally, in highly selected patients with
good social support structure who fail other treatments for alco­
holic hepatitis, early liver transplant can be an effective treatment.
Cessation of alcohol use is key. Recent experience with medica­
tions that reduce craving for alcohol, such as acamprosate calcium
and baclofen, have been favorable. Patients may take other neces­
sary medications even in the presence of cirrhosis. Acetaminophen
use is often discouraged in patients with liver disease; however, if
no more than 2 g of acetaminophen per day are consumed, there
generally are no problems unless there is active alcohol use.
■
■CIRRHOSIS DUE TO CHRONIC VIRAL
HEPATITIS B OR C
Of patients exposed to the hepatitis C virus (HCV), ~80% develop
chronic hepatitis C, and of those, ~20–30% will develop cirrhosis over
20–30 years. Many of these patients have had concomitant alcohol
use, and the true incidence of cirrhosis due to hepatitis C alone is
unknown. It is expected that an even higher percentage will go on to
develop cirrhosis over longer periods of time. In the United States,
~5–6 million people have been exposed to HCV, and ~4–5 million are
chronically viremic. Worldwide, ~170 million individuals have hepa­
titis C, with some areas of the world (e.g., Egypt) having up to 15%
of the population infected. HCV is a noncytopathic virus, and liver
damage is probably immune-mediated. Progression of liver disease
due to chronic hepatitis C is characterized by portal-based fibrosis with
bridging fibrosis and nodularity developing, ultimately culminating in
the development of cirrhosis. In cirrhosis due to chronic hepatitis C,
the liver is small and shrunken with characteristic features of a mixed
micro- and macronodular cirrhosis seen on liver biopsy. In addition, an
inflammatory infiltrate is found in portal areas with interface hepatitis
and occasionally some lobular hepatocellular injury and inflammation.
In patients with HCV genotype 3, steatosis is often present.
PART 10
Disorders of the Gastrointestinal System
Similar findings are seen in patients with cirrhosis due to chronic
hepatitis B. Of adult patients exposed to hepatitis B, ~5% develop
chronic hepatitis B, and ~20% of those patients will go on to develop
cirrhosis. Special stains for hepatitis B core (HBc) and hepatitis B
surface (HBs) antigen will be positive, and ground-glass hepatocytes
signify HBs antigen (HBsAg) may be present. In the United States,
there are ~2 million carriers of hepatitis B, whereas in other parts of
the world where hepatitis B virus (HBV) is endemic (i.e., Asia, Southeast
Asia, sub-Saharan Africa), up to 15% of the population may be
infected, having acquired the infection vertically at the time of birth.
Thus, >300–400 million individuals are thought to have hepatitis B
worldwide. Approximately 25% of these individuals may ultimately
develop cirrhosis.
Clinical Features and Diagnosis 
Patients with cirrhosis due to
either chronic hepatitis C or B can present with the usual symptoms
and signs of chronic liver disease. Fatigue, malaise, vague right upper
quadrant pain, and laboratory abnormalities are frequent presenting
features. Diagnosis requires a thorough laboratory evaluation, includ­
ing quantitative HCV RNA testing and analysis for HCV genotype, or
hepatitis B serologies to include HBsAg, HBeAg (hepatitis B e antigen),
anti-HBe, and quantitative HBV DNA levels.
TREATMENT
Cirrhosis due to Chronic Viral Hepatitis B or C
Management of complications of cirrhosis revolves around spe­
cific therapy for treatment of whatever complications occur (e.g.,
esophageal variceal hemorrhage, development of ascites and edema,
or encephalopathy). In patients with chronic hepatitis B, numerous
studies have shown beneficial effects of antiviral therapy, which
is effective at viral suppression, as evidenced by reducing amino­
transferase levels and HBV DNA levels and improving histology by
reducing inflammation and fibrosis. Several clinical trials and case
series have demonstrated that patients with decompensated liver
disease can become compensated with the use of antiviral therapy
directed against hepatitis B. Currently available therapy includes
lamivudine, adefovir, telbivudine, entecavir, and tenofovir, with the
latter two strongly preferred because of reduced risk of viral resis­
tance. Interferon α can also be used for treating hepatitis B, but it
should not be used in cirrhotics (see Chap. 352).
Treatment of patients with cirrhosis due to hepatitis C with
direct-acting antiviral protocols is highly successful (>95% cure
rate), well tolerated, and usually of short duration (8–12 weeks)
(see Chap. 352).
CIRRHOSIS FROM AUTOIMMUNE
HEPATITIS AND METABOLIC
DYSFUNCTION–ASSOCIATED
STEATOTIC LIVER DISEASE
Other causes of posthepatitic cirrhosis include autoimmune hepatitis
(AIH) and metabolic dysfunction–associated steatohepatitis (MASH),
which was previously called nonalcoholic steatohepatitis. Many patients
with AIH present with cirrhosis that is already established. Typically,
these patients will not benefit from immunosuppressive therapy with
glucocorticoids or azathioprine because the AIH is “burned out.” In
this situation, liver biopsy does not show a significant inflammatory
infiltrate. Diagnosis in this setting requires positive autoimmune
markers such as antinuclear antibody (ANA) or anti-smooth-muscle
antibody (ASMA). When patients with AIH present with cirrhosis and
active inflammation accompanied by elevated liver enzymes, there can
be considerable benefit from the use of immunosuppressive therapy.
Patients with MASH are increasingly being found to have pro­
gressed to cirrhosis. With the epidemic of obesity that continues in
Western countries, more and more patients are identified with meta­
bolic dysfunction–associated steatotic liver disease (Chap. 354). Of
these, a significant subset has MASH and can progress to increased
fibrosis and cirrhosis. Over the past several years, it has been increas­
ingly recognized that many patients who were thought to have crypto­
genic cirrhosis in fact have MASH. As their cirrhosis progresses, they
become catabolic and then lose the telltale signs of steatosis seen on
biopsy. Management of complications of cirrhosis due to either AIH or
MASH is similar to that for other forms of cirrhosis.
■
■BILIARY CIRRHOSIS
Biliary cirrhosis has pathologic features that are different from either
alcohol-associated cirrhosis or posthepatitic cirrhosis, yet the manifes­
tations of end-stage liver disease are the same. Cholestatic liver disease
may result from necroinflammatory lesions, congenital or metabolic
processes, or external bile duct compression. Thus, two broad catego­
ries reflect the anatomic sites of abnormal bile retention: intrahepatic
and extrahepatic. The distinction is important for obvious therapeutic
reasons. Extrahepatic obstruction may benefit from surgical or endo­
scopic biliary tract decompression, whereas intrahepatic cholestatic
processes will not improve with such interventions and require a dif­
ferent approach.
The major causes of chronic cholestatic syndromes are primary
biliary cholangitis (PBC), autoimmune cholangitis (AIC), primary
sclerosing cholangitis (PSC), and idiopathic adulthood ductopenia.
These syndromes are usually clinically distinguished from each other
by antibody testing, cholangiographic findings, and clinical presenta­
tion. However, they all share the histopathologic features of chronic
cholestasis, including cholate stasis; copper deposition; xanthomatous
transformation of hepatocytes; and irregular, so-called biliary fibro­
sis. In addition, there may be chronic portal inflammation, interface
activity, and chronic lobular inflammation. Ductopenia is a result of
progressive disease as patients develop cirrhosis.
■
■PRIMARY BILIARY CHOLANGITIS
PBC is seen in about 100–200 individuals per million, with a strong
female preponderance and a median age of ~50 years at the time of
diagnosis. The cause of PBC is unknown; it is characterized by portal
inflammation and necrosis of cholangiocytes in small- and mediumsized bile ducts. Cholestatic features prevail, and biliary cirrhosis is
characterized by an elevated bilirubin level and progressive liver fail­
ure. Liver transplantation is the treatment of choice for patients with
decompensated cirrhosis due to PBC.
Antimitochondrial antibodies (AMAs) are present in ~95% of
patients with PBC. These autoantibodies recognize lipoic acid on the
inner mitochondrial membrane proteins that are enzymes of the pyru­
vate dehydrogenase complex (PDC), the branched-chain 2-oxoacid
dehydrogenase complex, and the 2-oxogluterate dehydrogenase com­
plex. These autoantibodies are not pathogenic, but rather are useful
markers for making a diagnosis.
Pathology 
Histopathologic analysis identifies four distinct stages of
the disease as it progresses. The earliest lesion is termed chronic non­
suppurative destructive cholangitis and is a necrotizing inflammatory
process of the portal tracts. Medium and small bile ducts are infiltrated
with lymphocytes and undergo duct destruction. Mild fibrosis and
sometimes bile stasis can occur. With progression, the inflammatory
infiltrate becomes less prominent, but the number of bile ducts is
reduced and there is proliferation of smaller bile ductules. Increased
fibrosis ensues with the expansion of periportal fibrosis to bridging fibro­
sis. Finally, cirrhosis, which may be micronodular or macronodular,
develops.
Clinical Features 
Currently, most patients with PBC are middleaged women diagnosed well before the end-stage manifestations of
the disease are present, and as such, most patients are asymptomatic.
When symptoms are present, they most prominently include a sig­
nificant degree of fatigue out of proportion to either the severity of
the liver disease or the age of the patient. Pruritus is seen in ~50% of
patients at the time of diagnosis, and it can be debilitating. It might be
intermittent and usually is most bothersome in the evening. In some
patients, pruritus can develop toward the end of pregnancy and can
be mistaken for cholestasis of pregnancy. Pruritus that presents prior
to the development of jaundice indicates severe disease and a poor
prognosis.
Physical examination can show jaundice and other complications
of chronic liver disease including hepatomegaly, splenomegaly, ascites,
and edema. Other features that are unique to PBC include hyperpig­
mentation, xanthelasma, and xanthomata, which are related to altered
cholesterol metabolism. Hyperpigmentation is evident on the trunk
and the arms and is seen in areas of exfoliation and lichenification
associated with progressive scratching related to the pruritus. Bone
pain resulting from osteopenia or osteoporosis is occasionally seen at
diagnosis.
Laboratory Findings 
Laboratory findings in PBC show choles­
tatic liver enzyme abnormalities with an elevation in γ-glutamyl trans­
peptidase and alkaline phosphatase (ALP) along with mild elevations
in aminotransferases (ALT and AST). Immunoglobulins, particularly
IgM, are typically increased. Hyperbilirubinemia usually is seen once
cirrhosis has developed. Thrombocytopenia, leukopenia, and anemia
may be seen in patients with portal hypertension and hypersplenism.
Liver biopsy shows characteristic features as described above and
should be evident to any experienced hepatopathologist. Up to 10% of
patients with characteristic PBC will have features of AIH (moderate
to severe interphase hepatitis on biopsy, elevated ALT >5× the upper
limit of normal, and elevated IgG levels) as well and are defined as
having “overlap” syndrome. These patients are usually treated as PBC
patients and progress to cirrhosis with the same frequency as typical
PBC patients. Some patients require immunosuppressive medications
as well.
Diagnosis 
PBC should be considered in patients with chronic cho­
lestatic liver enzyme abnormalities. AMA testing may be negative in as
many as 5–10% of patients with PBC. These patients usually are posi­
tive for other PBC-specific autoantibodies including sp100 or gp210,
although these tests are not universally available. Liver biopsy is most
important in this setting of AMA-negative PBC. In patients who are
AMA negative with cholestatic liver enzymes, PSC should be ruled out
by way of cholangiography.
TREATMENT
Primary Biliary Cholangitis
Treatment of the typical manifestations of cirrhosis is no different
for PBC than for other forms of cirrhosis. Ursodeoxycholic acid
(UDCA) has been shown to improve both biochemical and histo­
logic features of the disease, thus slowing but not reversing or cur­
ing the disease. Improvement is greatest when therapy is initiated
early; the likelihood of significant improvement with UDCA is low
in patients with PBC who present with manifestations of cirrhosis.
UDCA is given in doses of 13–15 mg/kg per d; the medication
is usually well tolerated, although some patients have worsening
pruritus with initiation of therapy. A small proportion of patients
may have diarrhea or headache as a side effect of the drug. About
30–40% of patients with PBC do not have a satisfactory response
to UDCA; more than half of these patients will have signifi­
cant improvement with obeticholic acid, elafibranor or seladelpar
though these medication should be avoided in the setting of cir­
rhosis with portal hypertension or decompensation. Patients with
PBC require long-term follow-up by a physician experienced with
the disease. Certain patients may need to be considered for liver
transplantation should their liver disease decompensate.
CHAPTER 355
Cirrhosis and Its Complications
The main symptoms of PBC are fatigue and pruritus. Several
therapies have been tried for treatment of fatigue, but none of them
has been successful; frequent naps should be encouraged. Pruri­
tus is treated with antihistamines, narcotic receptor antagonists
(naltrexone), selective serotonin reuptake inhibitors, and rifampin.
Cholestyramine, a bile salt–sequestering agent, has been helpful in
some patients but is somewhat tedious and difficult to take. Plasma­
pheresis has been used rarely in patients with severe intractable pru­
ritus. There is an increased incidence of osteopenia and osteoporosis
in patients with cholestatic liver disease, and bone density testing
should be performed. Oral calcium and vitamin D are also recom­
mended. Treatment with a bisphosphonate should be instituted
when bone disease is identified. Screening for fat-soluble vitamin
deficiency (A, D, E, K) should done if total bilirubin is >2 mg/dL.
■
■PRIMARY SCLEROSING CHOLANGITIS
As in PBC, the cause of PSC remains unknown. PSC is a chronic cho­
lestatic syndrome that is characterized by diffuse inflammation and
fibrosis involving the entire biliary tree, resulting in chronic cholestasis.
This pathologic process ultimately results in obliteration of both the
intra- and extrahepatic biliary tree, leading to biliary cirrhosis, portal
hypertension, and liver failure. The cause of PSC remains unknown
despite extensive investigation into various mechanisms related to bac­
terial and viral infections, toxins, genetic predisposition, and immuno­
logic mechanisms, all of which have been postulated to contribute to
the pathogenesis and progression of this syndrome.
Liver biopsy changes in PSC are not pathognomonic, and estab­
lishing the diagnosis of PSC must involve imaging of the biliary tree.
Pathologic changes occurring in PSC show bile duct proliferation as
well as ductopenia and fibrous cholangitis (pericholangitis). Periductal
fibrosis is occasionally seen on biopsy specimens and can be quite
helpful in making the diagnosis. As the disease progresses, biliary cir­
rhosis is the end-stage manifestation of PSC.
Clinical Features 
The usual clinical features of PSC are those found
in cholestatic liver disease, with fatigue, pruritus, steatorrhea, deficien­
cies of fat-soluble vitamins, and the associated consequences. As in PBC,
the fatigue is profound and nonspecific. Pruritus can often be debilitating
and is related to the cholestasis. The severity of pruritus does not cor­
relate with the severity of the disease. Metabolic bone disease, as seen in
PBC, can occur with PSC and should be treated (see above).
Laboratory Findings 
Patients with PSC typically are identified
during an evaluation of abnormal liver enzymes. Most patients have at
least a twofold increase in ALP and may have elevated aminotransfer­
ases as well. Albumin levels may be decreased, and prothrombin times
can be prolonged at the time of diagnosis. Some degree of correction
of a prolonged prothrombin time may occur with parenteral vitamin K.
A small subset of patients has aminotransferase elevations >5× the
upper limit of normal and may have features of AIH on biopsy indicat­
ing an overlap syndrome between PSC and AIH. Autoantibodies are
frequently positive in patients with the overlap syndrome but are typi­
cally negative in patients who only have PSC. One autoantibody, the
perinuclear antineutrophil cytoplasmic antibody (pANCA), is positive
in ~65% of patients with PSC. Sixty to eighty percent of patients with
PSC have inflammatory bowel disease, predominately ulcerative colitis
(UC); thus, a colonoscopy is recommended at diagnosis.
Diagnosis 
The definitive diagnosis of PSC requires cholangio­
graphic imaging. Magnetic resonance imaging (MRI) with magnetic
resonance cholangiopancreatography (MRCP) is the imaging tech­
nique of choice for initial evaluation. Endoscopic retrograde cholan­
giopancreatography (ERCP) can be considered if the MRCP provided
suboptimal images or if there is newly elevated total bilirubin or MRCP
evidence of a dominant stricture. Typical cholangiographic findings in
PSC are multifocal stricturing and beading involving both the intrahe­
patic and extrahepatic biliary tree. These strictures are typically short
and with intervening segments of normal or slightly dilated bile ducts
that are distributed diffusely, producing the classic beaded appearance.
The gallbladder and cystic duct can be involved in up to 15% of cases.
Gradually, biliary cirrhosis develops, and patients will progress to
decompensated liver disease with manifestations of ascites, esophageal
variceal hemorrhage, and encephalopathy.
PART 10
Disorders of the Gastrointestinal System
TREATMENT
Primary Sclerosing Cholangitis
There is no specific proven treatment for PSC. Some clinicians
use UDCA at “PBC dosages” of 13–15 mg/kg per d with anecdotal
improvement, although no study has shown convincing evidence of
clinical benefit. A study of high-dose (28–30 mg/kg per d) UDCA
found it to be harmful. Endoscopic dilatation of dominant strictures
can be helpful, but the ultimate treatment is liver transplantation
when decompensated cirrhosis develops. Episodes of cholangitis
should be treated with antibiotics and can be an indication for liver
transplantation. A dreaded complication of PSC is the development
of cholangiocarcinoma, which is a relative contraindication to liver
transplantation.
■
■CARDIAC CIRRHOSIS
Definition 
Patients with long-standing right-sided congestive heart
failure may develop chronic liver injury from congestive hepatopa­
thy sometimes resulting in cardiac cirrhosis. This is an increasingly
uncommon, if not rare, cause of chronic liver disease given the
advances made in the care of patients with heart failure, particularly
valvular heart disease, but there has been an increase in patients with
congenital heart disease particularly after the Fontan operation.
Etiology and Pathology 
In the case of long-term right-sided
heart failure, there is an elevated venous pressure transmitted via the
inferior vena cava and hepatic veins to the sinusoids of the liver, which
become dilated and engorged with blood. The liver becomes enlarged
and swollen, and with long-term passive congestion and relative isch­
emia due to poor circulation, centrilobular hepatocytes can become
necrotic, leading to pericentral fibrosis. This fibrotic pattern can extend
to the periphery of the lobule outward until a unique pattern of fibrosis
causing cirrhosis can occur.
Clinical Features 
Patients typically have signs of congestive heart
failure and will manifest an enlarged firm liver on physical examina­
tion. ALP levels are characteristically elevated, and aminotransferases
may be normal or slightly increased, with AST usually higher than
ALT. It is unlikely that patients will develop variceal hemorrhage or
encephalopathy.
Diagnosis 
The diagnosis is usually made in someone with clear-cut
cardiac disease who has an elevated ALP and an enlarged liver. Liver
biopsy shows a pattern of fibrosis that can be recognized by an experi­
enced hepatopathologist. Differentiation from Budd-Chiari syndrome
(BCS) can be made by seeing extravasation of red blood cells in BCS,
but not in cardiac hepatopathy. Venoocclusive disease, now termed
sinusoidal obstructive syndrome (SOS), can also affect hepatic outflow
and has characteristic features on liver biopsy. SOS can be seen under
the circumstances of conditioning for bone marrow transplant with
radiation and chemotherapy; it can also be seen with the ingestion of
certain herbal teas as well as pyrrolizidine alkaloids. This is typically
seen in Caribbean countries and rarely in the United States. Treatment
is based on management of the underlying cardiac disease.
OTHER TYPES OF CIRRHOSIS
There are several other less common causes of chronic liver disease that
can progress to cirrhosis. These include inherited metabolic liver dis­
eases such as hemochromatosis, Wilson’s disease, α1 antitrypsin (α1AT)
deficiency, and cystic fibrosis. For these disorders, the manifestations
of cirrhosis are similar, with some minor variations, to those seen in
other patients with other causes of cirrhosis.
Hemochromatosis is an inherited disorder of iron metabolism that
results in a progressive increase in hepatic iron deposition, which, over
time, can lead to a portal-based fibrosis progressing to cirrhosis, liver
failure, and hepatocellular cancer. While the frequency of hemochro­
matosis is relatively common, with genetic susceptibility occurring in
1 in 250 individuals, the frequency of end-stage manifestations due to
the disease is relatively low, and <5% of those patients who are geno­
typically susceptible will go on to develop severe liver disease from
hemochromatosis. Diagnosis is made with serum iron studies showing
an elevated transferrin saturation and an elevated ferritin level, along
with abnormalities identified by HFE mutation analysis. Treatment is
straightforward, with regular therapeutic phlebotomy.
Wilson’s disease is an inherited disorder of copper homeostasis with
failure to excrete excess amounts of copper, leading to an accumulation
in the liver. This disorder is relatively uncommon, affecting 1 in 30,000
individuals. Wilson’s disease typically affects adolescents and young
adults. Prompt diagnosis before end-stage manifestations become
irreversible can lead to significant clinical improvement. Diagnosis
requires determination of ceruloplasmin levels, which are low; 24-h
urine copper levels, which are elevated; typical physical examination
findings, including Kayser-Fleischer corneal rings; and character­
istic liver biopsy findings. Treatment consists of copper-chelating
medications.
α1AT deficiency results from an inherited disorder that causes abnor­
mal folding of the α1AT protein, resulting in failure of secretion of that
protein from the liver. It is unknown how the retained protein leads to
liver disease. Patients with α1AT deficiency at greatest risk for develop­
ing chronic liver disease have the ZZ phenotype, but only ~10–20% of
such individuals will develop chronic liver disease. Diagnosis is made
by determining α1AT levels and phenotype. Characteristic periodic
acid–Schiff (PAS)–positive, diastase-resistant globules are seen on liver
biopsy. The only effective treatment is liver transplantation, which is
curative.
TABLE 355-2  Complications of Cirrhosis
Portal hypertension
Coagulopathy
  Gastroesophageal varices
  Factor deficiency
  Portal hypertensive gastropathy
  Fibrinolysis
  Splenomegaly, hypersplenism
  Thrombocytopenia
  Ascites
Bone disease
   Spontaneous bacterial peritonitis
  Osteopenia
Acute kidney injury-hepatorenal syndrome
  Osteoporosis
Chronic kidney disease-hepatorenal syndrome
  Osteomalacia
 
Hematologic abnormalities
Hepatic encephalopathy
  Anemia
Hepatopulmonary syndrome
  Hemolysis
Portopulmonary hypertension
  Thrombocytopenia
Malnutrition
  Neutropenia
Cystic fibrosis is an uncommon inherited disorder affecting whites
of northern European descent. A biliary-type cirrhosis can occur, and
some patients derive benefit from the chronic use of UDCA.
MAJOR COMPLICATIONS OF CIRRHOSIS
These include gastroesophageal variceal hemorrhage, splenomegaly,
ascites, hepatic encephalopathy, spontaneous bacterial peritonitis
(SBP), hepatorenal syndrome (HRS), and hepatocellular carcinoma
(Table 355-2). There are also more rare complications in the pulmo­
nary system including hepatopulmonary syndrome and portopulmo­
nary hypertension.
■
■PORTAL HYPERTENSION
Portal hypertension is defined as the elevation of the hepatic venous
pressure gradient (HVPG) to >5 mmHg. The portal venous system
normally drains blood from most of the GI tract including the stom­
ach, small and large intestines, spleen, pancreas, and gallbladder.
Portal hypertension is caused by a combination of two simultaneously
occurring hemodynamic processes: (1) increased intrahepatic resis­
tance to the passage of blood flow through the liver due to cirrhosis,
regenerative nodules, and microthrombi, and (2) increased splanchnic
blood flow secondary to vasodilation within the splanchnic vascular
bed. In more advanced stages, there is also activation of neurohumoral
responses and vasoconstrictive systems resulting in sodium and water
retention, increased blood volume, and hyperdynamic circulatory
system producing more portal hypertension. There is usually an initial
stage of compensated cirrhosis with HVPG between 5 and 10 mmHg
that can be asymptomatic and last for ≥10 years, but when clinically
significant portal hypertension (CSPH) develops (defined as a HVPG
≥10 mmHg), there is substantial risk of decompensation with variceal
bleeding, ascites, or hepatic encephalopathy. With decompensation,
median mortality is <2 years.
The causes of portal hypertension are usually subcategorized as
prehepatic, intrahepatic, and posthepatic (Table 355-3). Prehepatic
causes of portal hypertension are those affecting the portal venous
system before it enters the liver; they include portal vein thrombosis
and splenic vein thrombosis. Posthepatic causes encompass those
affecting the hepatic veins and venous drainage to the heart; they
include BCS and chronic right-sided cardiac congestion. Intrahepatic
causes account for >95% of cases of portal hypertension and are rep­
resented by the major forms of cirrhosis. Intrahepatic causes of portal
hypertension can be further subdivided into presinusoidal, sinusoidal,
and postsinusoidal causes. Postsinusoidal causes include venoocclusive
disease, whereas presinusoidal causes include congenital hepatic fibro­
sis and schistosomiasis. Sinusoidal causes are related to cirrhosis from
various causes.
Cirrhosis is overwhelmingly the most common cause of portal
hypertension in the United States. Portal vein thrombosis may contrib­
ute to portal hypertension and is most often associated with cirrhosis
but may be idiopathic or can occur in association with infection, pan­
creatitis, or abdominal trauma. Coagulation disorders that can lead
TABLE 355-3  Classification of Portal Hypertension
Prehepatic
  Portal vein thrombosis
  Splenic vein thrombosis
  Massive splenomegaly (Banti’s syndrome)
Hepatic
  Presinusoidal
    Schistosomiasis
    Congenital hepatic fibrosis
  Sinusoidal
    Cirrhosis—many causes
    Alcoholic hepatitis
  Postsinusoidal
    Hepatic sinusoidal obstruction (venoocclusive syndrome)
Posthepatic
  Budd-Chiari syndrome
  Inferior vena caval webs
  Cardiac causes
    Restrictive cardiomyopathy
    Constrictive pericarditis
    Severe congestive heart failure
to the development of portal vein thrombosis include polycythemia
vera; essential thrombocytosis; deficiencies in protein C, protein S,
antithrombin III, and factor V Leiden; and abnormalities in the generegulating prothrombin production. Some patients may have a sub­
clinical myeloproliferative disorder.
CHAPTER 355
Clinical Features 
The three primary complications of portal
hypertension are gastroesophageal varices with hemorrhage, ascites,
and hypersplenism. Thus, patients may present with upper GI bleed­
ing, which, on endoscopy, is found to be due to esophageal or gastric
varices; with the development of ascites along with peripheral edema;
or with an enlarged spleen with associated reduction in platelets and
white blood cells on routine laboratory testing.
Cirrhosis and Its Complications
ESOPHAGEAL VARICES  Variceal hemorrhage is an immediate lifethreatening problem with a 20–30% mortality rate associated with
each episode of bleeding. Over the past decade, it has become common
practice to screen known cirrhotics with endoscopy to look for esopha­
geal varices. Such screening studies have shown that approximately
one-third of patients with histologically confirmed cirrhosis have vari­
ces. Approximately 5–15% of cirrhotics per year develop varices, and
it is estimated that the majority of patients with cirrhosis will develop
varices over their lifetimes. Furthermore, it is anticipated that roughly
one-third of patients with varices will develop bleeding. Several fac­
tors predict the risk of bleeding, including the severity of cirrhosis
(Child-Pugh class, Model for End-Stage Liver Disease [MELD] score);
the height of wedged-hepatic vein pressure; the size of the varix; the
location of the varix; and certain endoscopic stigmata, including red
wale signs, hematocystic spots, diffuse erythema, bluish color, cherry
red spots, or white-nipple spots. Patients with tense ascites are also at
increased risk for bleeding from varices.
Diagnosis 
In patients with cirrhosis who are being followed chroni­
cally, the development of portal hypertension is usually revealed by the
presence of thrombocytopenia; the appearance of an enlarged spleen;
or the development of ascites, encephalopathy, and/or esophageal
varices with or without bleeding. In previously undiagnosed patients,
any of these features should prompt further evaluation to determine
the presence of portal hypertension and liver disease. Varices should
be identified by endoscopy. Contrasted-enhanced abdominal imaging,
either by computed tomography (CT) or MRI, can be helpful in dem­
onstrating a nodular liver and in finding changes of portal hyperten­
sion with intraabdominal collateral circulation. Rarely, the HVPG is
measured by interventional radiology. Patients with a gradient
12 mmHg are at risk for variceal hemorrhage.
TREATMENT
Variceal Hemorrhage
Treatment for esophageal varices as a complication of portal hyper­
tension is divided into two main categories: (1) primary prophylaxis
and (2) prevention of rebleeding once there has been an initial vari­
ceal hemorrhage. Primary prophylaxis requires routine surveillance
by endoscopy. Upper endoscopies are recommended at diagnosis of
compensated cirrhosis and then every 2 years if the liver disease is
active or every 3 years if inactive (alcohol cessation, viral hepatitis
eradication). In the absence of thrombocytopenia and with a liver
stiffness by transient elastography <20 kPa, high-risk varices are
rare and thus screening is not needed. Endoscopy is also recom­
mended at the time of hepatic decompensation. Once varices that
are at increased risk for bleeding are identified, usually defined as
medium or large varices or small varices with high-risk stigmata or
in decompensated cirrhosis, primary prophylaxis can be achieved
either through traditional nonselective beta blockade (NSBB) (pro­
pranolol or nadolol) titrated with a goal heart rate of 55–60 beats/
min with systolic blood pressure >90 mmHg or by variceal band
ligation. Carvedilol is becoming the NSBB of choice. Given its addi­
tional anti-α1-adrenergic vasodilating properties, it more effectively
lowers portal pressure, lacks the need for heart rate goals, and has
emerging data that suggest it can prevent hepatic decompensation
and improve survival in persons with CSPH.
PART 10
Disorders of the Gastrointestinal System
Endoscopic variceal ligation (EVL) has been compared to tradi­
tional NSBB and carvedilol for primary prophylaxis against variceal
bleeding, and EVL appears to have equivalent efficacy at preventing
bleeding. NSBBs, especially carvedilol, are generally recommended
as first-line treatment for primary prophylaxis of bleeding if toler­
ated, given their additional benefits. Once primary prophylaxis
has been initiated, repeat endoscopy for surveillance of varices is
unnecessary.
The approach to patients once they have had a variceal bleed is
first to treat the acute bleed, which can be life-threatening, and then
to prevent further bleeding. Treatment of acute bleeding requires
both fluid and red blood cell replacement to stabilize hemodynam­
ics. A randomized trial of restricted transfusion starting when
hemoglobin is <7 g/dL with a goal hemoglobin of 7–9 g/dL, com­
pared to a more liberal strategy, resulted in reduced early rebleeding
and mortality. This strategy is recommended, although adjustments
should be made based on cardiac risks and hemodynamics. Correct­
ing an elevated prothrombin time with fresh frozen plasma is not
recommended unless there is evidence of coagulopathy (bleeding at
other sites such as IV lines). The use of vasoconstricting agents, usu­
ally somatostatin or octreotide, has been shown to improve initial
bleeding control and reduce transfusion requirements and all-cause
mortality. Prophylactic antibiotics, usually with ceftriaxone, started
prior to endoscopy result in reduced infections, recurrent bleeding,
and mortality. Balloon tamponade (Sengstaken-Blakemore tube or
Minnesota tube) or placement of self-expandable metal stents can
be used in patients who need stabilization prior to endoscopic
therapy or as a bridge to transjugular intrahepatic portosystemic
shunt (TIPS) after endoscopic failure. Control of bleeding can be
achieved in the vast majority of cases; however, bleeding recurs in
the majority of patients if definitive endoscopic therapy has not been
instituted. Upper endoscopy is used as first-line treatment to diag­
nose the cause of the bleeding and to control bleeding acutely with
EVL. When esophageal varices extend into the proximal stomach or
the bleeding varices are entirely within the stomach, band ligation
is often unsuccessful. In these situations, consideration for a TIPS
should be made. This technique creates a portosystemic shunt by
a percutaneous approach using an expandable metal stent, which
is advanced under angiographic guidance to the hepatic veins and
then through the substance of the liver to create a direct portocaval
Recurrent acute bleeding
Endoscopic therapy
+
Pharmacologic therapy
Control of bleeding
Decompensated cirrhosis
Child’s class B or C
Compensated cirrhosis
Child’s class A
Transplant evaluation
TIPS
Endoscopic therapy or
beta blockers
Consider liver
transplantation
evaluation
Consider TIPS
Liver transplantation
FIGURE 355-3  Management of recurrent variceal hemorrhage. This algorithm
describes an approach to management of patients who have recurrent bleeding
from esophageal varices. Initial therapy is generally with endoscopic therapy often
supplemented by pharmacologic therapy. With control of bleeding, a decision
needs to be made as to whether patients should go on to transjugular intrahepatic
portosystemic shunt (TIPS; if they are Child’s class A) or if they should have TIPS
and be considered for transplant (if they are Child’s class B or C).
shunt. Encephalopathy can occur in as many as 20% of patients
after TIPS and is particularly problematic in elderly patients and in
patients with preexisting encephalopathy. TIPS is usually reserved
for individuals who fail or are unable to receive endoscopic therapy,
although there is emerging evidence that patients who are highly
selected to be at high risk for rebleeding may also benefit. TIPS can
sometimes be used as a bridge to transplantation, and all patients
requiring TIPS should be considered for transplant evaluation. Some
gastric varices are associated with a splenorenal shunt and can be
effectively treated with a balloon-occluded retrograde transvenous
obliteration (BRTO) of varices sometimes in combination with a
TIPS. Prevention of further bleeding is usually accomplished with
repeated variceal band ligation until varices are obliterated in com­
bination with NSBB. If recurrent variceal bleeding occurs, then TIPS
should be performed for long-term prevention of bleeding. Once a
TIPS has been performed, there is no need for further endoscopies
for variceal surveillance; however, the TIPS should be periodically
monitored with Doppler ultrasound for stenosis (Fig. 355-3).
■
■PORTAL HYPERTENSIVE GASTROPATHY
Portal hypertensive gastropathy can cause both acute clinical GI bleed­
ing and chronic bleeding resulting in iron-deficiency anemia. It is
associated with all causes of portal hypertension and is diagnosed by
characteristic endoscopy findings showing a snakeskin-like mosaic
pattern of gastric mucosa often with central red or brown spots. When
there is bleeding, treatment is with NSBB and iron repletion. Refrac­
tory bleeding may respond to TIPS.
■
■SPLENOMEGALY AND HYPERSPLENISM
Congestive splenomegaly with hypersplenism is common in patients
with portal hypertension and is usually the first indication of portal
hypertension. Clinical features include the presence of an enlarged
spleen on physical examination and the development of thrombocy­
topenia and leukopenia in patients who have cirrhosis. Some patients
will have significant left-sided and left upper quadrant abdominal pain
related to an enlarged spleen. Splenomegaly itself usually requires no
specific treatment.
■
■ASCITES
Definition 
Ascites is the accumulation of fluid within the peri­
toneal cavity. Overwhelmingly, the most common cause of ascites is
portal hypertension related to cirrhosis; however, clinicians should
remember that malignant, infectious, and cardiac causes of ascites can
be present as well, and careful differentiation of these other causes is
obviously important for patient care.
Pathogenesis 
The presence of portal hypertension contributes to
the development of ascites in patients who have cirrhosis (Fig. 355-4).
There is an increase in intrahepatic resistance, causing increased portal
pressure, but there is also vasodilation of the splanchnic arterial system,
which, in turn, results in an increase in portal venous inflow. Both
abnormalities result in increased production of splanchnic lymph.
Vasodilating factors such as nitric oxide are responsible for the vasodi­
latory effect. There is activation of the renin-angiotensin-aldosterone
system with the development of hyperaldosteronism and activation
of the sympathetic nervous system as a consequence of a homeostatic
response caused by underfilling of the arterial circulation secondary to
arterial vasodilation in the splanchnic vascular bed. The renal effects
of increased aldosterone and activation of the sympathetic nervous sys­
tem lead to sodium retention causing fluid accumulation and expan­
sion of the extracellular fluid volume, resulting in peripheral edema
and ascites. Because the retained fluid is constantly leaking out of the
intravascular compartment into the peritoneal cavity, the sensation of
vascular filling is not achieved, and the process continues. Hypoalbu­
minemia from decreased synthetic function in a cirrhotic liver results
in reduced plasma oncotic pressure and contributes to the loss of fluid
from the vascular compartment into the peritoneal cavity.
Clinical Features 
Patients typically note an increase in abdominal
girth that is often accompanied by peripheral edema. The development
of ascites is often insidious, and it is surprising that some patients wait
so long and become so distended before seeking medical attention.
Patients usually have at least 1–2 L of fluid in the abdomen before they
are aware that there is an increase. If ascitic fluid is massive, respiratory
function can be compromised, causing dyspnea. Hepatic hydrothorax
may also contribute to respiratory symptoms. Patients with massive
ascites are often malnourished and have muscle wasting and excessive
fatigue and weakness.
Diagnosis 
Diagnosis of ascites is by physical examination and is
often aided by abdominal imaging. Patients will have bulging flanks
Cirrhosis
Portal hypertension
Splanchnic vasodilation
↑ Splanchnic pressure
Arterial underfilling
Lymph formation
Activation of
vasoconstrictors and
antinatriuretic factors*
Formation of ascites
Sodium retention
Plasma volume
expansion
FIGURE 355-4  Development of ascites in cirrhosis. This flow diagram illustrates the
importance of portal hypertension with splanchnic vasodilation in the development
of ascites. *Antinatriuretic factors include the renin-angiotensin-aldosterone
system and the sympathetic nervous system.
and may have a fluid wave or the presence of shifting dullness. This is
determined by taking patients from a supine position to lying on either
their left or right side and noting the movement of the dullness to per­
cussion. Subtle amounts of ascites can be detected by ultrasound or CT
scanning. Hepatic hydrothorax is more common on the right side and
implicates a rent in the diaphragm with free flow of ascitic fluid into
the thoracic cavity.
When patients present with ascites for the first time, it is recom­
mended that a diagnostic paracentesis be performed to characterize
the fluid. This should include the determination of total protein and
albumin content, blood cell counts with differential, and cultures. In
the appropriate setting, amylase may be measured and cytology per­
formed. In patients with cirrhosis, the protein concentration of the
ascitic fluid is low, usually <2.5 g/dL. The serum ascites-to-albumin
gradient (SAAG), calculated by subtracting the fluid albumin level
from the serum albumin level, has replaced the description of exuda­
tive or transudative fluid. When the SAAG is >1.1 g/dL, the cause of
the ascites is most likely due to portal hypertension; this is usually
in the setting of cirrhosis. Cardiac ascites can be identified by SAAG
1.1 g/dL and ascites protein >2.5g/dL. When the SAAG is <1.1 g/dL,
infectious or malignant causes of ascites should be considered. When
ascitic fluid protein is very low, <1.5 g/dL, patients are at increased risk
for developing SBP. A high level of red blood cells in the ascitic fluid
usually signifies a traumatic tap but can also rarely occur with hepato­
cellular cancer or a ruptured omental varix. When the absolute level of
polymorphonuclear leukocytes is >250/μL, infection is likely.
TREATMENT
Ascites
CHAPTER 355
Patients with small amounts of ascites can usually be managed
with dietary sodium restriction alone. Most average diets in the
United States contain 6–8 g of sodium per day, and if patients eat at
restaurants or fast-food outlets, the amount of sodium in their diet
can exceed this amount. Thus, it is often extremely difficult to get
patients to change their dietary habits to ingest 2 g of sodium per
day, equivalent to slightly more than three-quarters of a teaspoon
of salt, which is the recommended amount. Sodium educational
pamphlets are helpful. Often, a simple recommendation is to eat
fresh or frozen foods, avoiding canned or processed foods. When
a moderate amount of ascites is present, diuretic therapy is usually
necessary. Traditionally, spironolactone at 100 mg/d as a single dose
is started, and furosemide may be added at 40 mg/d, particularly in
patients who have peripheral edema. Failure of the diuretics sug­
gests that patients may not be compliant with a low-sodium diet.
If compliance is confirmed and ascitic fluid is not being mobilized,
there should be incremental increases in spironolactone to a maxi­
mum of 400 mg/d and furosemide to 160 mg/d. If a large amount
of ascites is still present on diuretics in patients who are compliant
with a low-sodium diet, then they are defined as having refractory
ascites, and alternative treatment modalities including repeated
large-volume paracentesis (LVP) or a TIPS procedure should be
considered (Fig. 355-5). After LVP of ≥5 L, IV 25% albumin at
a dose of 6–8 g/L of removed ascites should be given to prevent
circulatory dysfunction. Multiple studies have shown that TIPS,
although effective at managing the ascites, does not improve sur­
vival. Unfortunately, TIPS is often associated with an increased
frequency of hepatic encephalopathy and must be considered care­
fully on a case-by-case basis. The prognosis for patients with cir­
rhosis with ascites is poor, and some studies have shown that <50%
of patients survive 2 years after the onset of ascites. Thus, there
should be consideration for liver transplantation in patients with
ascites. Patients with cirrhosis and ascites are at increased risk for
renal failure from certain medications including nonsteroidal antiinflammatory drugs and aminoglycosides; therefore, these medica­
tions should generally be avoided. Angiotensin-converting enzyme
inhibitors and angiotensin receptor blockers should be used cau­
tiously with close monitoring of blood pressure and renal function.
Cirrhosis and Its Complications
Highly symptomatic ascites
Large-volume paracentesis (LVP) + albumin
Dietary sodium restriction + diuretics
Ascites reaccumulation
Consider TIPS
Continue LVP with
albumin as needed
Consider liver
transplantation
FIGURE 355-5  Treatment of refractory ascites. In patients who develop azotemia
in the course of receiving diuretics in the management of their ascites, some will
require repeated large-volume paracentesis (LVP), some may be considered for
transjugular intrahepatic portosystemic shunt (TIPS), and some would be good
candidates for liver transplantation. These decisions are all individualized.
■
■SPONTANEOUS BACTERIAL PERITONITIS
SBP is a common and severe complication of ascites characterized by
spontaneous infection of the ascitic fluid without an intraabdominal
source. In hospitalized patients with cirrhosis and ascites, SBP can
occur in up to 30% of individuals and can have a 25% in-hospital
mortality rate. Bacterial translocation is the presumed mechanism for
development of SBP, with gut flora traversing the intestine into mes­
enteric lymph nodes, leading to bacteremia and seeding of the ascitic
fluid. The most common organisms are Escherichia coli and other gut
bacteria; however, gram-positive bacteria, including Streptococcus
viridans, Staphylococcus aureus, and Enterococcus spp., can also be
found. If more than two organisms are identified, secondary bacte­
rial peritonitis due to a perforated viscus should be considered. The
diagnosis of SBP is made when the fluid sample has an absolute neu­
trophil count >250/μL. Bedside cultures should be obtained by direct
injection of ascitic fluid into blood culture bottles. Patients with ascites
may present with fever, altered mental status, elevated white blood cell
count, abdominal pain or discomfort, and acute kidney injury, or they
may present without any of these features. Therefore, it is necessary
to have a high degree of clinical suspicion, and peritoneal taps are
recommended for cirrhosis patients hospitalized with ascites and cir­
rhosis complications or signs of infection. Treatment is commonly with
intravenous third-generation cephalosporin for 5 days. In addition,
intravenous albumin (1.5 g/kg body weight on day 1 and 1.0 g/kg on
day 3) has been shown to reduce the risk of renal failure and to improve
survival. In patients with variceal hemorrhage, the frequency of SBP is
significantly increased, and prophylaxis against SBP is recommended
at presentation with upper GI bleeding. Furthermore, in patients who
have had an episode of SBP and recovered, quinolone antibiotic pro­
phylaxis should be given to prevent recurrent SBP.
PART 10
Disorders of the Gastrointestinal System
■
■HEPATORENAL SYNDROME
HRS is a form of functional renal failure without renal pathology that
occurs in ~10% of patients with advanced cirrhosis or acute liver failure.
There are marked disturbances in the arterial renal circulation in patients
with HRS; these include an increase in vascular resistance accompanied
by a reduction in systemic vascular resistance. The reason for renal vaso­
constriction is most likely multifactorial and is poorly understood. The
diagnosis is made usually in the presence of a large amount of ascites in
patients who have a stepwise progressive increase in creatinine. Acute
kidney injury (AKI) is defined by a 0.3-mg/dL rise in creatinine over
48 h or a 50% rise in creatinine from baseline. HRS-AKI, the new term
for type 1 HRS, is characterized by a progressive rapid impairment in
renal function. Type 2 HRS, now termed HRS–chronic kidney disease
(CKD), is characterized by a reduction in glomerular filtration rate with
an elevation of serum creatinine level, but it is stable and is associated
with a better outcome than that of type 1 HRS.
HRS-AKI requires exclusion of other causes of acute renal failure,
most notably volume depletion. Diuretics should be stopped, and
infusion of albumin 1 g/kg per d for 48 h without significant improve­
ments is required. Other causes of AKI, including intrinsic (acute
tubular necrosis, acute interstitial nephritis, and glomerulonephritis)
and obstructive kidney disease, should be excluded. Treatment is
with vasoconstrictors such as terlipressin or, if not available, low-dose
norepinephrine, which requires intensive care unit monitoring. Mido­
drine, an α-agonist, along with octreotide and intravenous albumin
are also commonly used in the United States but are now third line. If
treatment fails, then renal replacement therapy can be initiated. The
best therapy for HRS is liver transplantation; recovery of renal function
is typical in this setting. In patients with either AKI-HRS or CKD-HRS,
the prognosis is poor unless transplant can be achieved.
■
■HEPATIC ENCEPHALOPATHY
Hepatic encephalopathy is a serious complication of chronic liver
disease and is broadly defined as an alteration in mental status and
cognitive function occurring as a consequence of liver failure. In severe
acute liver injury, the development of encephalopathy is a requirement
for a diagnosis of acute liver failure and can be seen in association
with life-threatening brain edema, which is not a feature in chronic
liver disease. Hepatic encephalopathy is much more commonly seen
in patients with chronic liver disease. Gut-derived neurotoxins that are
not removed by the liver because of vascular shunting and decreased
hepatic mass reach the brain and cause the symptoms known as hepatic
encephalopathy. Ammonia levels are typically elevated, but the correla­
tion between severity of liver disease and height of ammonia levels is
often poor, and most hepatologists do not rely on ammonia levels to
make a diagnosis or follow clinical progress. Other compounds and
metabolites that may contribute to the development of encephalopathy
include certain false neurotransmitters and mercaptans.
Clinical Features 
In acute liver failure, changes in mental status can
occur rapidly. Brain edema can be seen in these patients, with severe
encephalopathy associated with swelling of the gray matter. Cerebral
herniation is a feared complication of brain edema in acute liver failure,
and treatment to decrease edema is with hypertonic saline or mannitol.
In patients with cirrhosis, encephalopathy is often found as a result
of precipitating events such as volume depletion, gastrointestinal
bleeding, hyponatremia, infection, or constipation. Patients may be
confused or exhibit a change in personality. They may be quite violent
and difficult to manage; alternatively, patients may be very sleepy and
difficult to rouse. If patients have ascites, this should be tapped to rule
out infection. Evidence of GI bleeding should be sought, and patients
should be appropriately hydrated. Electrolytes should be measured,
and abnormalities corrected. In patients presenting with encepha­
lopathy, asterixis is often present. Asterixis can be elicited by having
patients extend their arms and bend their wrists back. Patients who are
encephalopathic have a “liver flap”—that is, a sudden forward move­
ment of the wrist. This requires patients to be able to cooperate with
the examiner. Alternative causes for altered mental status should also
be considered.
The diagnosis of hepatic encephalopathy is clinical and requires
an experienced clinician to recognize and put together all the various
features. Often when patients have encephalopathy for the first time,
they (and/or their caregivers) are unaware of what is transpiring, but
once they have been through the experience, they can identify when
this is developing in subsequent situations and can often self-medicate
to prevent the development or worsening of encephalopathy.
TREATMENT
Hepatic Encephalopathy
Treatment is multifactorial and includes management of the abovementioned precipitating factors. Sometimes hydration and correc­
tion of electrolyte imbalance are all that is necessary. In the past,
restriction of dietary protein was used; however, the negative impact
of that maneuver on overall nutrition is thought to outweigh the
benefit, and it is thus strongly discouraged. The mainstay of treat­
ment for encephalopathy is to use lactulose, a nonabsorbable disac­
charide, which results in colonic acidification. Catharsis ensues,
contributing to the elimination of nitrogenous products in the gut
that are responsible for the development of encephalopathy. The
goal of lactulose therapy is to promote two to three soft stools per
day. Patients are asked to titrate their amount of ingested lactulose
to achieve the desired effect. Lactulose is usually continued after the
initial episode of encephalopathy. Poorly absorbed antibiotics are
often used as adjunctive therapies for patients who have a difficult
time with lactulose or in those with recurrent episodes. Rifaximin
has replaced neomycin and metronidazole (because of their signifi­
cant toxicity). The dose is 550 mg twice daily and is very effective
in preventing recurrent encephalopathy. In patients with recurrent
symptoms despite treatment, closure of large portosystemic shunts
can be helpful. Zinc supplementation is sometimes helpful and is
relatively harmless. The development of encephalopathy in patients
with chronic liver disease is a poor prognostic sign, but this compli­
cation can be managed in the vast majority of patients.
■
■ACUTE-ON-CHRONIC LIVER FAILURE
Acute-on-chronic liver failure (ACLF) is a recently recognized clinical
syndrome that is characterized by acutely decompensating cirrhosis
with associated failure of one or more organ systems, including liver,
kidneys, brain, lung, circulatory system, and coagulation. It occurs in
the setting of chronic liver disease almost always with cirrhosis and is a
major cause of mortality in persons with cirrhosis. ACLF is precipitated
by either direct injury to the liver (most commonly from alcoholic
hepatitis and less often from new or flaring viral hepatitis, autoim­
mune hepatitis, or drug-induced liver injury) or systemic effects (most
commonly bacterial or fungal infection and less often GI bleeding or
the postoperative state), resulting in a marked systemic inflammatory
response followed by organ failure and is analogous to sepsis syndrome.
Mortality at 28 days ranges from 20 to 70% and increases with the
number of organ failures. Clinicians should search for and treat pre­
cipitating causes of ACLF, determine if intensive care unit (ICU) care
is needed, and consider immediate referral for liver transplantation.
Complications of cirrhosis should be treated as described elsewhere
in this chapter. If after 3–7 days of ICU support, there continue to be
four or more organ failures and liver transplantation is not an option,
consideration for a transition to palliative care is recommended.
■
■LIVER-LUNG SYNDROMES
Hepatopulmonary syndrome (HPS) is characterized by arterial hypox­
emia in a patient with cirrhosis without significant lung disease. The
liver disease causes intrapulmonary vascular dilations, resulting in
blood shunting past alveoli and significant ventilation-perfusion mis­
match. Clinical symptoms include dyspnea and platypnea. HPS is com­
mon, occurring in 4–32% of patients with cirrhosis; however, it is often
mild. Diagnosis involves demonstrating hypoxemia, without evidence
of significant lung disease, and shunt on bubble echocardiography.
Treatment is with oxygen supplementation and liver transplantation.
Portopulmonary hypertension (PPHT) is pulmonary hyperten­
sion in a patient with portal hypertension. The portal hypertension
results in the production of vasoconstrictor substances that affect the
pulmonary artery. Many patients are asymptomatic, especially early
in the disease; however, they later can develop dyspnea on exertion
and fatigue. PPHT is rare, occurring in ~5% of patients with advanced
cirrhosis. Diagnosis includes initial identification on echocardiogram
and confirmation on right heart catheterization showing elevated mean
pulmonary artery pressure, elevated pulmonary vascular resistance,
and normal pulmonary capillary wedge pressure. Prognosis is poor,
although liver transplantation after effective reduction in pulmonary
artery pressure with vasodilatory medications can be effective.
■
■MALNUTRITION IN CIRRHOSIS
Because the liver is principally involved in the regulation of protein
and energy metabolism in the body, patients with decompensated
cirrhosis often develop malnutrition, insufficient intake of nutrients,
which manifests clinically as sarcopenia, a loss in muscle mass and
function resulting in frailty, decreased physical reserve, and increased
susceptibility to health stressors. In the setting of cirrhosis and subse­
quent hepatic decompensation, patients become more catabolic with
increased energy needs but often have decreased caloric intake due
to anorexia, early satiety, ascites, restrictive diets (low sodium, potas­
sium, and fluid) and hepatic encephalopathy. Obesity and edema may
mask underlying sarcopenia. Lack of physical activity can further
exacerbate the sarcopenia and resulting frailty. Sarcopenia and frailty
have been associated with increased resource utilization and risk of
death and poorer quality of life. Guidelines recommend (1) educa­
tion for patients and families both with asymptomatic compensated
and symptomatic decompensated cirrhosis on appropriate nutritional
intake (both calories and protein) and the benefits of regular exercise;
(2) periodic clinical assessments for malnutrition, sarcopenia, and
frailty with increasing frequency as the patient decompensates; and (3)
early referrals to registered dietician and physical therapy to develop a
personalized therapeutic plan. Generally, patients with cirrhosis should
consume 35 kcal/kg per d based on actual weight with estimated dis­
counting for edema and ascites and 1.2–1.5 g/kg per d of protein based
upon ideal body weight. Caloric intake goals should be reduced for
patients with body mass index over 35 kg/m2. Multiple small meals are
recommended throughout waking hours to minimize muscle wasting
during the fasted state. Relaxation in sodium restriction should be
considered in those who are not meeting caloric intake goals. Optimi­
zation of care for hepatic encephalopathy and ascites is also beneficial.
General exercise recommendations include 150–300 min of aerobic
exercise a week and 2 days of muscle-strengthening exercises per week,
though these recommendations should be tailored, particularly in
decompensated cirrhosis.
CHAPTER 355
■
■ABNORMALITIES IN COAGULATION
Coagulation disorders in cirrhosis are poorly understood, and typical
clinical measures of coagulation, such as the prothrombin time and
international normalized ratio, are not reliable measures of clotting
ability. There is decreased synthesis of both pro- and anticoagulant fac­
tors and thus some rebalancing in coagulation; however, the coagula­
tion cascade can easily tip toward thrombosis or bleeding. In addition,
patients may have thrombocytopenia from hypersplenism due to portal
hypertension and some platelet dysfunction, which is counterbalanced
with increased von Willebrand factor. Adequate thrombin formation
can occur with platelet levels from cirrhosis patients >50,000–60,000/L.
Synthesis of vitamin K–dependent clotting factors II, VII, IX, and X
is diminished in patients with chronic cholestatic syndromes because
absorption of vitamin K requires good bile flow. Intravenous or intra­
muscular vitamin K can quickly correct this abnormality. Overall, the
status of coagulation in a cirrhotic patient needs to be judged clinically
rather than relying on current laboratory tests. Standard prophylaxis
for deep venous thrombosis when hospitalized is generally recom­
mended. Routine correction of international normalized ratio before
procedures or with variceal bleeding is generally not required.
Cirrhosis and Its Complications
■
■BONE DISEASE IN CIRRHOSIS
Osteoporosis is common in patients with chronic cholestatic liver
disease because of malabsorption of vitamin D and decreased calcium
ingestion. The rate of bone resorption exceeds that of new bone forma­
tion in patients with cirrhosis, resulting in bone loss. Dual-energy x-ray
absorptiometry (DEXA) is a useful method for determining osteopo­
rosis or osteopenia. When a DEXA scan shows osteoporosis, treatment
with bisphosphonates is effective.
■
■HEMATOLOGIC ABNORMALITIES IN CIRRHOSIS
Numerous hematologic manifestations of cirrhosis are present, includ­
ing anemia from a variety of causes including hypersplenism, hemoly­
sis, iron deficiency, and perhaps folate deficiency from malnutrition.
Macrocytosis is a common abnormality in red blood cell morphology
seen in patients with chronic liver disease, and neutropenia may be a
result of hypersplenism.

28 - 356 Liver Transplantation
356 Liver Transplantation
■
■FURTHER READING
Bhattacharya D et al: Hepatitis C guidance update 2023 update:
AASLD-IDSA recommendations for testing, managing, and treating
hepatitis C virus infection. Clin Infect Dis Ciad319:1, 2023.
Biggins SW et al: Diagnosis, evaluation and management of ascites,
spontaneous bacterial peritonitis and hepatorenal syndrome: 2021
practice guidance by the American Association for the Study of Liver
Diseases. Hepatology 74:1014, 2021.
EASL: Clinical practice guidelines on acute-on-chronic liver failure.
J Hepatol 79:461, 2023.
Kaplan D et al: AASLD practice guidance of risk stratification and
management of portal hypertension and varices in cirrhosis. Hepatol­
ogy 79:1180, 2024.
Lai JC et al: Malnutrition, frailty, and sarcopenia in patients with cir­
rhosis: 2021 practice guidance by the American Association for the
Study of Liver Diseases. Hepatology 74:1611, 2021.
Terrault NA et al: Update on prevention, diagnosis and treatment of
chronic hepatitis B: AASLD 2018 hepatitis B guidance. Hepatology
67:1560, 2018.
Vilstrup H et al: Hepatic encephalopathy in chronic liver disease:
2014 Practice Guideline by the American Association for the Study
of Liver Diseases and the European Association for the Study of the
Liver. Hepatology 60:715, 2014.
PART 10
Disorders of the Gastrointestinal System
Emily D. Bethea, Raymond T. Chung,
Jules L. Dienstag
Liver Transplantation
Liver transplantation—the replacement of the native, diseased liver by
a normal organ (allograft)—has matured from an experimental pro­
cedure reserved for desperately ill patients to an accepted, lifesaving
operation applied more optimally in the natural history of end-stage
liver disease. The preferred and technically most advanced approach
is orthotopic transplantation, in which the native organ is removed and
the donor organ is inserted in the same anatomic location. Pioneered
in the 1960s by Thomas Starzl at the University of Colorado and,
later, at the University of Pittsburgh and by Roy Calne in Cambridge,
England, liver transplantation is now performed routinely worldwide.
Success measured as 1-year survival has improved from ~30% in the
1970s to >90% today. These improved prospects for prolonged survival
resulted from refinements in operative technique, improvements in
organ procurement and preservation, advances in immunosuppres­
sive therapy, and, perhaps most influentially, more enlightened patient
selection and timing. Despite the perioperative morbidity and mortal­
ity, the technical and management challenges of the procedure, and
its costs, liver transplantation has become the approach of choice for
selected patients whose chronic or acute liver disease is progressive,
life-threatening, and unresponsive to medical therapy. Based on the
current level of success, the number of liver transplants has continued
to grow each year; in 2022, 9528 patients received liver allografts in the
United States—8925 deceased donor (94%) and 603 living donor (6%).
Still, the demand for new livers continues to outpace availability; as of
September 2023, 10,285 patients in the United States were on a waiting
list for a donor liver.
INDICATIONS
Potential candidates for liver transplantation are children and adults
who, in the absence of contraindications (see below), suffer from
severe, irreversible liver disease for which alternative medical or surgi­
cal treatments have been exhausted or are unavailable. Timing of the
operation is of critical importance. Indeed, improved timing and better
patient selection are felt to have contributed more to the increased
success of liver transplantation in the 1980s and beyond than all the
impressive technical and immunologic advances combined. Although
the disease should be advanced, and although opportunities for
spontaneous or medically induced stabilization or recovery should be
allowed, the procedure should be done sufficiently early to give the
surgical procedure a fair chance for success. Ideally, transplantation
should be considered in patients with end-stage liver disease who are
experiencing or have experienced a life-threatening complication of
hepatic decompensation or whose quality of life has deteriorated to
unacceptable levels. Although patients with well-compensated cirrho­
sis can survive for many years, many patients with quasi-stable chronic
liver disease have much more advanced disease than may be apparent.
As discussed below, the better the status of the patient prior to trans­
plantation, the higher will be its anticipated success rate. The decision
about when to transplant is complex and requires the combined judg­
ment of an experienced team of hepatologists, transplant surgeons,
anesthesiologists, and specialists in support services, not to mention
the well-informed consent of the patient and the patient’s family.
■
■TRANSPLANTATION IN CHILDREN
Indications for transplantation in children are listed in Table 356-1.
The most common is biliary atresia. Inherited or genetic disorders of
metabolism associated with liver failure constitute another major indi­
cation for transplantation in children and adolescents. In Crigler-Najjar
disease type I and in certain hereditary disorders of the urea cycle and
of amino acid or lactate-pyruvate metabolism, transplantation may be
the only way to prevent impending deterioration of central nervous
system function, despite the fact that the native liver is structurally nor­
mal. Combined heart and liver transplantation has yielded dramatic
improvement in cardiac function and in cholesterol levels in children
with homozygous familial hypercholesterolemia; combined liver and
TABLE 356-1  Indications for Liver Transplantation
CHILDREN
ADULTS
Biliary atresia
Primary biliary cholangitis
Neonatal hepatitis
Primary sclerosing cholangitis
Congenital hepatic fibrosis
Caroli’s diseasea
Alagille’s syndromeb
Secondary biliary cirrhosis
Byler’s diseasec
Autoimmune hepatitis
Inherited disorders of metabolism
Hemochromatosis-associated cirrhosis
Wilson’s disease
α1 Antitrypsin deficiency
  Tyrosinemia
Metabolic dysfunction–associated
steatohepatitis (MASH)d
  Glycogen storage diseases
Alcohol-associated cirrhosis
  Lysosomal storage diseases
Severe alcohol-associated hepatitis
  Protoporphyria
Cryptogenic cirrhosis
Crigler-Najjar disease type I
Chronic viral hepatitis with cirrhosis
Familial hypercholesterolemia
Hepatic venous outflow obstruction
(Budd-Chiari syndrome)
Primary hyperoxaluria type I
Hemophilia
Acute liver failure (ALF)
Hepatocellular carcinoma
 
 
 
 
 
 
Select cases for the following indications:
  Hepatic adenomas
  Familial amyloidosis
  Hepatic epithelioid
hemangioendothelioma (HEHE)
  Erythropoietic protoporphyria (EPP)
  Metastatic neuroendocrine tumors
 
  Polycystic liver disease
aMultiple cystic dilatations of the intrahepatic biliary tree. bArteriohepatic dysplasia,
with paucity of bile ducts, and congenital malformations, including pulmonary
stenosis. cIntrahepatic cholestasis, progressive liver failure, and mental and growth
retardation. dFormerly nonalcoholic steatohepatitis (NASH).
Liver Transplantation
CHAPTER 356
kidney transplantation has been successful in patients with primary
hyperoxaluria type I. In hemophiliacs with transfusion-associated
hepatitis and liver failure, liver transplantation has been associated with
recovery of normal factor VIII synthesis.
■
■TRANSPLANTATION IN ADULTS
Liver transplantation is indicated for end-stage cirrhosis of all causes
(Table 356-1). In sclerosing cholangitis and Caroli’s disease (multiple
cystic dilatations of the intrahepatic biliary tree), recurrent infections
and sepsis associated with inflammatory and fibrotic obstruction
of the biliary tree may be an indication for transplantation. Because
prior biliary surgery complicates and is a relative contraindication
for liver transplantation, surgical diversion of the biliary tree has
been all but abandoned for patients with sclerosing cholangitis.
Currently, the most common indication for liver transplantation is
steatotic liver disease (SLD), which encompasses alcohol-associated
liver disease (ALD) and metabolic dysfunction–associated steatotic
liver disease (MASLD), formerly known as nonalcoholic fatty liver
disease (NAFLD). Patients with alcohol-associated cirrhosis can be
considered as candidates for transplantation if they meet strict criteria
for abstinence and reform; however, these criteria still do not prevent
return to alcohol use in up to a quarter of cases. In highly selected
cases in a limited but growing number of centers, transplantation for
severe acute alcohol-associated hepatitis has been performed with suc­
cess; however, because patients with acute alcohol-associated hepatitis
are still actively using alcohol and because continued alcohol use
remains a concern, acute alcohol-associated hepatitis is not a routine
indication for liver transplantation. Patients with chronic hepatitis C
have early allograft and patient survival comparable to those of other
subsets of patients after transplantation; however, they have histori­
cally experienced a higher frequency of allograft failure beyond 5 years due
to universal reinfection and progressive fibrosis in cases of recurrent
hepatitis C is insidiously progressive, with allograft cirrhosis and
failure occurring at a higher frequency beyond 5 years. Fortunately,
with the introduction of highly effective direct-acting antiviral (DAA)
agents targeting hepatitis C virus (HCV), allograft outcomes have
improved substantially. In patients with chronic hepatitis B, in the
absence of measures to prevent recurrent hepatitis B, survival after
transplantation is reduced by ~10–20%; however, prophylactic use of
hepatitis B immune globulin (HBIg) during and after transplantation
increases the success of transplantation to a level comparable to that
seen in patients with nonviral causes of liver decompensation. Specific
oral antiviral drugs (e.g., entecavir, tenofovir disoproxil fumarate,
tenofovir alafenamide) (Chap. 352) can be used both for prophylaxis
against and for treatment of recurrent hepatitis B, facilitating further
the management of patients undergoing liver transplantation for endstage hepatitis B; most transplantation centers rely on antiviral drugs
with or without HBIg to manage patients with hepatitis B. Issues of
disease recurrence are discussed in more detail below. In patients who
undergo transplantation for hepatic vein thrombosis (Budd-Chiari syn­
drome), postoperative anticoagulation is essential, as is the treatment
of any myeloproliferative disorder. Patients with acute liver failure are
candidates for liver transplantation, provided a donor organ can be
located quickly, before life-threatening complications—including cerebral
edema—set in. Patients with nonmetastatic primary hepatobiliary
tumors—primary hepatocellular carcinoma (HCC), cholangiocarci­
noma, hepatoblastoma, angiosarcoma, epithelioid hemangioendothe­
lioma, and multiple or massive hepatic adenomata—have undergone
liver transplantation; however, for some hepatobiliary malignancies,
overall survival is significantly lower than that for other categories
of liver disease. Most transplantation centers have reported 5-year
recurrence-free survival rates in patients with unresectable HCC for
single tumors <5 cm in diameter or for three or fewer lesions all <3
cm comparable to those seen in patients undergoing transplantation
for nonmalignant indications. Consequently, liver transplantation is
currently restricted to patients whose hepatic malignancies meet these
criteria. Expanded criteria for patients with HCC continue to be eval­
uated in exploratory clinical trials. Because the likelihood of recurrent
cholangiocarcinoma is very high, only highly selected patients with
limited disease are being evaluated for transplantation after intensive
chemotherapy and radiation.
CONTRAINDICATIONS
Absolute contraindications for transplantation include life-threatening
systemic diseases, uncontrolled extrahepatic bacterial or fungal infec­
tions, preexisting advanced cardiovascular or pulmonary disease, mul­
tiple uncorrectable life-threatening congenital anomalies, metastatic
malignancy, and active drug or alcohol use disorders (Table 356-2).
Because carefully selected patients in their sixties and even seventies
have undergone transplantation successfully, advanced age per se is
no longer considered an absolute contraindication; however, in older
patients, a more thorough preoperative evaluation should be under­
taken to exclude ischemic cardiac disease and other comorbid condi­
tions. Still, advanced age (>70 years) should be considered a relative
contraindication—that is, a factor to be considered with other relative
contraindications. Other relative contraindications include extensive
portal vein thrombosis, preexisting renal disease not associated with
liver disease (which may prompt consideration of combined liver and
kidney transplantation), intrahepatic or biliary sepsis, severe hypox­
emia (PO2 <50 mmHg) resulting from right-to-left intrapulmonary
shunts, portopulmonary hypertension with high mean pulmonary
artery pressures (>35 mmHg), uncontrolled psychiatric conditions,
and lack of sufficient social supports. Any one of these relative contra­
indications may be insufficient in and of itself to preclude transplan­
tation. For example, the problem of portal vein thrombosis may be
overcome in certain cases by constructing a graft from the donor liver
portal vein to the recipient’s superior mesenteric vein. Now that com­
bination antiretroviral therapy has dramatically improved the survival
of persons with HIV infection (Chap. 208) and because end-stage liver
disease caused by chronic hepatitis C and B has emerged as a serious
source of morbidity and mortality in the HIV-infected population,
liver transplantation has now been performed successfully in selected
HIV-positive persons who have excellent control of HIV infection.
Selected patients with CD4+ T-cell counts >100/μL and with pharma­
cologic suppression of HIV viremia have undergone transplantation
for end-stage liver disease. HIV-infected persons who have received
liver allografts for end-stage liver disease resulting from chronic hepa­
titis B have experienced survival rates comparable to those of HIVnegative persons undergoing transplantation for the same indication.
In contrast, until recently, recurrent HCV infection in the allograft
has limited long-term success in persons with HCV-related end-stage
TABLE 356-2  Contraindications to Liver Transplantation
ABSOLUTE
RELATIVE
Uncontrolled extrahepatobiliary
infection
Advanced agea
Active, untreated sepsis
Prior extensive hepatobiliary surgery
Life-limiting congenital anomalies
Extensive portal vein thrombosis
Cholangiocarcinoma (except those
tumors that fit into protocols)
Renal failure not attributable to
liver disease (consider dual organ
transplantation)
Advanced cardiopulmonary disease
Extrahepatobiliary malignancy
 
Severe obesity
 
Severe malnutrition/wasting
 
Medical noncompliance
AIDS
HIV seropositivity with failure to control
HIV viremia or CD4 <100/μL
Life-threatening systemic diseases
Severe hypoxemia secondary to
right-to-left intrapulmonary shunts
(PO2 <50 mmHg)
 
Severe pulmonary hypertension (mean
pulmonary artery pressure >35 mmHg)
Active substance use disorder
Uncontrolled psychiatric disorder
aFor patients 70 years and older, comprehensive assessment to exclude concurrent
comorbidities, in particular cardiovascular compromise, is indicated.
liver disease. Again, the availability of DAA agents targeting HCV (see
below and Chap. 352) has improved allograft outcomes substantially.
TECHNICAL CONSIDERATIONS
■
■DECEASED-DONOR SELECTION
Deceased-donor livers for transplantation are procured primarily from
victims of head trauma, termed donation after brain death (DBD). Car­
diovascular and respiratory functions in these donors are maintained
artificially until the liver can be removed. Organs from brain-dead
donors up to age 60 are acceptable if the following criteria are met:
hemodynamic stability, adequate oxygenation, absence of bacterial or
fungal infection, absence of abdominal trauma, and absence of hepatic
dysfunction. Historically serologic exclusion of infections with hepatitis
B virus (HBV), HCV, and HIV was required, but organs from donors
infected with HBV, HCV, and HIV are now used in select cases when
matched appropriately with the recipient after thorough informed
consent and antiviral medication planning. Transplantation of organs
procured from deceased donors who have succumbed to cardiac death,
termed donation after circulatory death (DCD), can be performed
successfully under selected circumstances, when ischemic time is
minimized and liver histology preserved. Encouraging improvements in
normothermic ex vivo liver perfusion techniques may make broader use
of these organs possible. Compatibility in ABO blood group and organ
size between donor and recipient are important considerations in donor
selection; however, ABO-incompatible, split-liver, or reduced-donororgan allografts can be performed in emergencies or marked donor
scarcity. Tissue typing for human leukocyte antigen (HLA) matching is
not required, and preformed cytotoxic HLA antibodies do not preclude
liver transplantation. Following perfusion with cold electrolyte solution,
the donor liver is removed and packed in ice. The use of University of
Wisconsin (UW) solution, rich in lactobionate and raffinose, has per­
mitted the extension of cold ischemic time up to 20 h; however, 12 h
may be a more reasonable limit. Improved techniques for harvesting
multiple organs from the same donor have increased the availability of
donor livers, but the availability of donor livers is far outstripped by the
demand. Currently in the United States, all donor livers are distributed
through a nationwide organ-sharing network (United Network for
Organ Sharing [UNOS]) designed to allocate available organs based
on regional considerations and recipient acuity. Recipients who have
the highest disease severity generally have the highest priority, but
allocation strategies that balance highest urgency against best outcomes
continue to evolve to distribute deceased-donor organs most effectively.
Allocation based on the Child-Turcotte-Pugh (CTP) score, which uses
five clinical variables (encephalopathy stage, ascites, bilirubin, albumin,
and prothrombin time) and waiting time, has been replaced by alloca­
tion based on urgency alone, calculated using the Model for End-Stage
Liver Disease (MELD) score. The MELD score is based on a mathemati­
cal model that includes serum bilirubin, albumin, sodium, creatinine,
and prothrombin time expressed as international normalized ratio
(INR); gender assigned at birth is included in the MELD calculation
for females (Table 356-3). The MELD scale is continuous with scores
ranging between 6 and 40; higher values correlate with increased ill­
ness severity. On the lower end of this scale, the value of proceeding to
transplantation is limited, because liver recipients with MELD scores
<15 experience higher posttransplantation mortality rates than similarly
classified patients who remain on the waiting list. The MELD score
has undergone modifications over time, including the development of
MELD-Na in 2016 and MELD-3.0 in 2023 to improve model accuracy
in predicting short-term survival and prioritizing allocation. Neither
waiting time (except as a tie breaker between two potential recipients
with the same MELD scores) nor posttransplantation outcome is taken
into account, but use of the MELD score has been shown to reduce
waiting list mortality, to reduce waiting time prior to transplantation,
to be the best predictor of pretransplantation mortality, to satisfy the
prevailing view that medical need should be the decisive determinant,
and to eliminate both the subjectivity inherent in the CTP scoring sys­
tem (presence and degree of ascites and hepatic encephalopathy) and
the differences in waiting times among different regions of the country.
PART 10
Disorders of the Gastrointestinal System
TABLE 356-3  United Network for Organ Sharing (UNOS) Liver
Transplantation Waiting List Criteria
Status 1
Acute liver failure (including primary graft
nonfunction and hepatic artery thrombosis)a
The Model for End-Stage Liver Disease (MELD) score, on a continuous scale,
determines allocation of the remainder of donor organsb.
Candidates who are at least 18 years old at the time of registration receive an
initial MELD score equal to:
1.33 (if female) + [4.56 × loge(bilirubin)] + [0.82 × (137 – sodium)] – [0.24 × (137 –
sodium) × loge(bilirubin)] + [9.09 × loge(INR)] + [11.14 × loge(creatinine)] + [1.85 ×
(3.5 – albumin)] – [1.83 × (3.5 – albumin) × loge(creatinine)] + 6
Candidates who are currently at least 12 years old and were less than 18 years
old at the time of registration receive a MELD score equal to:
[4.56 × loge(bilirubin)] + [0.82 × (137 – sodium)] – [0.24 × 137 – sodium) ×
loge(bilirubin)] + [9.09 × loge(INR)] + 11.14 × loge(creatinine)] +
[1.85 × (3.5 – albumin)] – [1.83 × (3.5 – albumin)] – [1.83 × (3.5 – albumin) ×
loge(creatinine)] + 7.33
The PELD calculator is for candidates under the age of 12. 
Online calculators to determine MELD scores are available, such as the
following: https://optn.transplant.hrsa.gov/resources/allocation-calculators/
meld-calculator/
aFor children <18 years of age, status 1 includes acute or chronic liver failure plus
hospitalization in an intensive care unit or inborn errors of metabolism. Status 1
is retained for those persons with acute liver failure and supersedes the MELD
score. MELD scale is continuous, with 34 levels ranging between 6 and 40 (scores
above 40 are categorized as 40). bIn certain cases, the natural MELD score may
not represent the severity of illness or need for transplantation; in such cases, an
exception MELD score may be granted if select criteria are met upon review by the
National Liver Review Board (NLRB): https://optn.transplant.hrsa.gov/professionals/
by-topic/guidance/liver-review-board-guidance/.
The highest priority (status 1) continues to be reserved for patients
with acute liver failure (ALF) or complications following transplanta­
tion such as primary graft nonfunction and hepatic artery thrombosis.
Because candidates for liver transplantation who have HCC may not
be sufficiently decompensated to compete for donor organs based on
urgency criteria alone, and because protracted waiting for deceaseddonor organs often results in tumor growth beyond acceptable limits
for transplantation, such patients are assigned disease-specific MELD
exception points. In addition to HCC, other disease-specific MELD
exceptions are evaluated on a regular basis and updated by the National
Liver Review Board (NLRB) (Table 356-3).
■
■LIVING-DONOR TRANSPLANTATION
Occasionally, especially for liver transplantation in children, one
deceased-donor organ can be split between two recipients (one adult
and one child). A more viable alternative, transplantation of the right
lobe of the liver from a healthy adult donor into an adult recipient,
has gained increased popularity. Living-donor transplantation of the
left lobe (left lateral segment), introduced in the early 1990s to allevi­
ate the extreme shortage of donor organs for small children, accounts
currently for approximately one-third of all liver transplantation pro­
cedures in children. Driven by the shortage of deceased-donor organs,
living-donor transplantation involving the more sizable right lobe is
being considered with increasing frequency in adults; however, livingdonor liver transplantation cannot be expected to solve the donor
organ shortage; 603 such procedures were done in 2023, representing
only ~6% of all liver transplant operations done in the United States.
Living-donor transplantation can reduce waiting time and cold
ischemia time; is done under elective, rather than emergency, circum­
stances; and is lifesaving in recipients who cannot afford to wait for
a deceased donor. The downside, of course, is the risk to the healthy
donor (a mean of 10 weeks of medical disability; biliary complications
in ~5%; postoperative complications such as wound infection, smallbowel obstruction, and incisional hernias in 9–19%; and death in
0.2–0.4%) as well as the increased frequency of biliary (15–32%) and
vascular (10%) complications in the recipient. Potential donors must
participate voluntarily without coercion, and transplantation teams
should go to great lengths to exclude subtle coercive or inappropriate
psychological factors as well as outline carefully to both donor and
recipient the potential benefits and risks of the procedure. Donors for
the procedure should be 18–65 years old; have a compatible blood type
with the recipient; have no serious chronic medical problems or history
of major abdominal surgery; and pass an exhaustive series of clinical,
biochemical, and serologic evaluations to unearth disqualifying medi­
cal disorders. The recipient should meet the same UNOS criteria for
liver transplantation as recipients of a deceased donor allograft.
■
■SURGICAL TECHNIQUE
Removal of the recipient’s native liver is technically difficult, particu­
larly in the presence of portal hypertension with its associated collat­
eral circulation and extensive varices and especially in the presence
of scarring from previous abdominal operations. The combination of
portal hypertension and coagulopathy (elevated prothrombin time and
thrombocytopenia) may translate into large blood-product transfusion
requirements. After the portal vein and infrahepatic and suprahepatic
inferior vena cava are dissected, the hepatic artery and common bile
duct are dissected. Then the native liver is removed and the donor
organ inserted. During the anhepatic phase, coagulopathy, hypogly­
cemia, hypocalcemia, and hypothermia are encountered and must be
managed by the anesthesiology team. Caval, portal vein, hepatic artery,
and bile duct anastomoses are performed in succession, the last by endto-end suturing of the donor and recipient common bile ducts (Fig.
356-1) or by choledochojejunostomy to a Roux-en-Y loop if the recipi­
ent common bile duct cannot be used for reconstruction (e.g., in scle­
rosing cholangitis). A typical transplant operation lasts 8 h, with a range
of 6–18 h. Because of excessive bleeding, large volumes of blood, blood
products, and volume expanders may be required during surgery;
however, blood requirements have fallen sharply with improvements in
surgical technique, blood-salvage interventions, and experience.
As noted above, emerging alternatives to orthotopic liver transplan­
tation include split-liver grafts, in which one donor organ is divided
and inserted into two recipients, and living-donor procedures, in which
part of the left (for children), the left (for children or small adults), or
the right (for adults) lobe of the liver is harvested from a living donor
for transplantation into the recipient. In the adult procedure, once the
right lobe is removed from the donor, the donor right hepatic vein
is anastomosed to the recipient right hepatic vein remnant, followed
by donor-to-recipient anastomoses of the portal vein and then the
hepatic artery. Finally, the biliary anastomosis is performed, ductto-duct if practical or via Roux-en-Y anastomosis. Heterotopic liver
Suprahepatic
vena cava
Donor
liver
Hepatic artery
Portal vein
Common bile duct
Infrahepatic vena cava
FIGURE 356-1  The anastomoses in orthotopic liver transplantation. The
anastomoses are performed in the following sequence: (1) suprahepatic and
infrahepatic vena cava, (2) portal vein, (3) hepatic artery, and (4) common bile ductto-duct anastomosis. (Reproduced with permission from S Vilarinho, RP Lifton: Liver
transplantation: From inception to clinical practice. 150:1096, 2012.)
transplantation, in which the donor liver is inserted without removal
of the native liver, has met with very limited success and acceptance,
except in a very small number of centers. Areas of research with the
potential to overcome the shortage of donor organs include hepatocyte
transplantation and xenotransplantation with genetically modified
organs of nonhuman origin (e.g., swine).
POSTOPERATIVE COURSE AND
MANAGEMENT
■
■IMMUNOSUPPRESSIVE THERAPY
The introduction in 1980 of cyclosporine as an immunosuppressive
agent contributed substantially to the improvement in survival after
liver transplantation. Cyclosporine, a calcineurin inhibitor, blocks
early activation of T cells and is specific for T-cell functions that result
from the interaction of the T cell with its receptor and that involve the
calcium-dependent signal transduction pathway. As a result, the activ­
ity of cyclosporine leads to inhibition of lymphokine gene activation,
blocking interleukins 2, 3, and 4, tumor necrosis factor α, and other
lymphokines. Cyclosporine also inhibits B-cell functions. This process
occurs without affecting rapidly dividing cells in the bone marrow,
which may account for the reduced frequency of posttransplantation
systemic infections. The most common and important side effect of
cyclosporine therapy is nephrotoxicity. Cyclosporine causes dosedependent renal tubular injury and direct renal artery vasospasm.
Following renal function is therefore important in monitoring cyclo­
sporine therapy and is perhaps even a more reliable indicator than
blood levels of the drug. Nephrotoxicity is reversible and can be man­
aged by dose reduction. Other adverse effects of cyclosporine therapy
include hypertension, hyperkalemia, tremor, hirsutism, glucose intol­
erance, and gingival hyperplasia.
CHAPTER 356
Tacrolimus, a macrolide lactone antibiotic isolated from a Japanese soil
fungus, Streptomyces tsukubaensis, has the same mechanism of action
as cyclosporine but is 10–100 times more potent. Initially applied as
“rescue” therapy for patients in whom rejection occurred despite the
use of cyclosporine, tacrolimus was shown to be associated with a
reduced frequency of acute, refractory, and chronic rejection. Although
patient and graft survival are the same with these two drugs, the advan­
tage of tacrolimus in minimizing episodes of rejection, reducing the
need for additional glucocorticoid doses, and lowering the likelihood
of bacterial and cytomegalovirus (CMV) infection has simplified the
management of patients undergoing liver transplantation. In addi­
tion, the oral absorption of tacrolimus is more predictable than that of
cyclosporine. As a result, tacrolimus has now supplanted cyclosporine
for primary immunosuppression, and most centers rely on oral rather
than IV administration from the outset.
Liver Transplantation
Although more potent than cyclosporine, tacrolimus is also more
toxic and more likely to be discontinued for adverse events. The toxic­
ity of tacrolimus is similar to that of cyclosporine; nephrotoxicity and
neurotoxicity are the most commonly encountered adverse effects,
and neurotoxicity (tremor, seizures, hallucinations, psychoses, coma)
is more likely and more severe in tacrolimus-treated patients. Both
drugs can cause diabetes mellitus, but tacrolimus does not cause hirsut­
ism or gingival hyperplasia. Because of overlapping toxicity between
cyclosporine and tacrolimus, especially nephrotoxicity, and because
tacrolimus reduces cyclosporine clearance, these two drugs should not
be used together. Because 99% of tacrolimus is metabolized by the liver,
hepatic dysfunction reduces its clearance; in primary graft nonfunction
(when, for technical reasons or because of ischemic damage prior to
its insertion, the allograft is defective and does not function normally
from the outset), tacrolimus doses have to be reduced substantially,
especially in children. Both cyclosporine and tacrolimus are metabo­
lized by the cytochrome P450 IIIA system, and therefore, drugs that
induce cytochrome P450 (e.g., phenytoin, phenobarbital, carbamaze­
pine, rifampin) reduce available levels of cyclosporine and tacrolimus,
and drugs that inhibit cytochrome P450 (e.g., erythromycin, flucon­
azole, ketoconazole, clotrimazole, itraconazole, verapamil, diltiazem,
danazol, metoclopramide, the HIV protease inhibitor ritonavir, and
the HCV protease inhibitors glecaprevir [cyclosporine only] and
grazoprevir) increase cyclosporine and tacrolimus blood levels. Indeed,
itraconazole is used occasionally to help boost tacrolimus levels. Like
azathioprine, cyclosporine and tacrolimus appear to be associated with
a risk of lymphoproliferative malignancies (see below), which may
occur earlier after cyclosporine or tacrolimus than after azathioprine
therapy. Because of these side effects, combinations of cyclosporine
or tacrolimus with prednisone and an antimetabolite (azathioprine or
mycophenolic acid, see below)—all at reduced doses—are preferable
regimens for immunosuppressive therapy.
Mycophenolic acid, a nonnucleoside purine metabolism inhibitor
derived as a fermentation product from several Penicillium species,
is another immunosuppressive drug being used for patients undergo­
ing liver transplantation. Mycophenolate has been shown to be better
than azathioprine, when used with other standard immunosuppressive
drugs, in preventing rejection after renal transplantation and has been
adopted widely as well for use in liver transplantation. The most com­
mon adverse effects of mycophenolate are bone marrow suppression
and gastrointestinal complaints.
In patients with pretransplantation renal dysfunction or renal dete­
rioration that occurs intraoperatively or immediately postoperatively,
tacrolimus or cyclosporine therapy may not be practical; under these
circumstances, induction or maintenance of immunosuppression
with antithymocyte globulin (ATG; thymoglobulin) or monoclonal
antibodies to T cells (basiliximab, daclizumab) may be appropriate.
Therapy with some of these agents (ATG) has also been effective in
reversing acute rejection in the posttransplantation period and is the
standard treatment for acute rejection that fails to respond to methyl­
prednisolone boluses. Available data support the use of thymoglobulin
induction to delay calcineurin inhibitor use and its attendant nephro­
toxicity. IV infusions of thymoglobulin may be complicated by fever
and chills, which can be ameliorated by premedication with antipyret­
ics and a low dose of glucocorticoids.
PART 10
Disorders of the Gastrointestinal System
Sirolimus, an inhibitor of the mammalian target of rapamycin
(mTOR), blocks later events in T-cell activation and is another agent
approved for use in solid-organ transplantation. Sirolimus, however,
is not used early in the posttransplantation course, because it impairs
wound healing and is associated with an increased frequency of hepatic
artery thrombosis in the first 30 days after transplantation. In patients
with calcineurin inhibitor–related nephrotoxicity, conversion to siroli­
mus has been demonstrated to be effective in preventing rejection with
accompanying improvements in renal function. Because of its profound
antiproliferative effects, sirolimus has also been suggested to be a use­
ful immunosuppressive agent in patients with a prior or current his­
tory of malignancy, such as HCC. Side effects include hyperlipidemia,
peripheral edema, oral ulcers, and interstitial pneumonitis. Everolimus
is a hydroxyethyl derivative of sirolimus that, when used in conjunction
with low-dose tacrolimus, also provides successful protection against
acute rejection, with decreased renal impairment compared to that
associated with standard tacrolimus dosing. Everolimus and sirolimus
share a similar adverse event profile. The most important principle of
immunosuppression is that the ideal approach strikes a balance between
immunosuppression and immunologic competence. In general, given
sufficient immunosuppression, acute liver allograft rejection is nearly
always reversible. On one hand, incompletely treated acute rejection
predisposes to the development of chronic rejection, which can threaten
graft survival. On the other hand, if the cumulative dose of immunosup­
pressive therapy is too large, the patient may succumb to opportunistic
infection. Data show a benefit to minimizing the use of glucocorticoids,
a mainstay of immunosuppressive regimens, as cases allow. In select
conditions, such as hepatitis C, early steroid withdrawal and initiation of
DAAs can avert recurrent allograft hepatitis successfully. Patients who
undergo liver transplantation for autoimmune diseases such as autoim­
mune hepatitis are less likely to achieve freedom from glucocorticoids,
although, in many cases, immunosuppression can be narrowed to a
dual-agent regimen of a calcineurin inhibitor and an antimetabolite.
■
■POSTOPERATIVE COMPLICATIONS
Complications of liver transplantation can be divided into nonhepatic
and hepatic categories (Tables 356-4 and 356-5). In addition, both
TABLE 356-4  Nonhepatic Complications of Liver Transplantation
CATEGORY
COMPLICATION
Cardiovascular
instability 
 
Arrhythmias
Congestive heart failure
Cardiomyopathy
Pulmonary compromise
Pneumonia
 
Pulmonary capillary vascular permeability
 
Fluid overload
Renal dysfunction
Prerenal azotemia
 
Hypoperfusion injury (acute tubular necrosis)
 
Drug nephrotoxicity
 
↓ Renal blood flow secondary to ↑ intraabdominal
pressure
Hematologic
Anemia secondary to gastrointestinal and/or
intraabdominal bleeding
 
Hemolytic anemia, aplastic anemia
 
Thrombocytopenia
Infection
Bacterial: early, common postoperative infections
 
Fungal/parasitic: late, opportunistic infections
 
Viral: late, opportunistic infections, recurrent hepatitis
Neuropsychiatric
Seizures
 
Metabolic encephalopathy
 
Depression
 
Difficult psychosocial adjustment
Diseases of donor
Infectious
 
Malignant
Malignancy
B-cell lymphoma (posttransplantation
lymphoproliferative disorders)
 
De novo neoplasms (particularly squamous cell skin
carcinoma)
immediate postoperative and late complications are encountered. As
a rule, patients who undergo liver transplantation have been chroni­
cally ill for protracted periods and may be malnourished and wasted.
The impact of such chronic illness and the multisystem failure that
TABLE 356-5  Hepatic Complications of Liver Transplantation
Hepatic Dysfunction Common after Major Surgery
Prehepatic
Pigment load
 
Hemolysis
 
Blood collections (hematomas, abdominal
collections)
Intrahepatic
 
  Early
Hepatotoxic drugs and anesthesia
 
Hypoperfusion (hypotension, shock, sepsis)
 
Benign postoperative cholestasis
  Late
Transfusion-associated hepatitis
 
Exacerbation of primary hepatic disease
  Posthepatic
Biliary obstruction
 
↓ Renal clearance of conjugated bilirubin (renal
dysfunction)
Hepatic Dysfunction Unique to Liver Transplantation
Primary graft nonfunction
 
Vascular compromise
Portal vein obstruction
 
Hepatic artery thrombosis
 
Anastomotic leak with intraabdominal bleeding
Bile duct disorder
Stenosis, obstruction, leak
Rejection
 
Recurrent primary hepatic
disease
accompanies liver failure continue to require attention in the post­
operative period. Because of the massive fluid losses and fluid shifts
that occur during the operation, patients may remain fluid overloaded
during the immediate postoperative period, straining cardiovascular
reserve; this effect can be amplified in the face of transient renal dys­
function and pulmonary capillary vascular permeability. Continuous
monitoring of cardiovascular and pulmonary function, measures to
maintain the integrity of the intravascular compartment and to treat
extravascular volume overload, and scrupulous attention to potential
sources and sites of infection are of paramount importance. Cardiovas­
cular instability may also result from the electrolyte imbalance that may
accompany reperfusion of the donor liver as well as from restoration of
systemic vascular resistance following implantation. Pulmonary func­
tion may be compromised further by paralysis of the right hemidia­
phragm associated with phrenic nerve injury. The hyperdynamic state
with increased cardiac output that is characteristic of patients with liver
failure reverses rapidly after successful liver transplantation.
Other immediate management issues include renal dysfunction.
Prerenal azotemia, acute kidney injury associated with hypoperfusion
(acute tubular necrosis), and renal toxicity caused by antibiotics, tacro­
limus, or cyclosporine are encountered frequently in the postoperative
period, sometimes necessitating dialysis. Hemolytic-uremic syndrome
can be associated with cyclosporine and tacrolimus. Occasionally,
postoperative intraperitoneal bleeding may be sufficient to increase
intraabdominal pressure, which, in turn, may reduce renal blood flow;
this effect is rapidly reversible when abdominal distention is relieved
by exploratory laparotomy to identify and ligate the bleeding site and
to remove intraperitoneal clot.
Anemia may also result from acute upper gastrointestinal bleed­
ing or from transient hemolytic anemia, which may be autoimmune,
especially when blood group O livers are transplanted into blood group
A or B recipients. This autoimmune hemolytic anemia is mediated by
donor intrahepatic lymphocytes that recognize red blood cell A or B
antigens on recipient erythrocytes. Transient in nature, this process
resolves once the donor liver is repopulated by recipient bone mar­
row–derived lymphocytes; the hemolysis can be treated by transfusing
blood group O red blood cells and/or by administering higher doses of
glucocorticoids. Transient thrombocytopenia is also encountered com­
monly. Aplastic anemia, a late occurrence, is rare but has been reported
in almost 30% of patients who underwent liver transplantation for
acute, severe hepatitis of unknown cause.
Bacterial, fungal, or viral infections are common and may be lifethreatening postoperatively. Early after transplant surgery, common
postoperative infections predominate—pneumonia, wound infections,
infected intraabdominal collections, urinary tract infections, and IV
line infections—rather than opportunistic infections; these infections
may involve the biliary tree and liver as well. Beyond the first post­
operative month, the toll of immunosuppression becomes evident,
and opportunistic infections—CMV, herpes viruses, fungal infections
(Aspergillus, Candida, cryptococcal disease), mycobacterial infections,
parasitic infections (Pneumocystis, Toxoplasma), and bacterial infec­
tions (Nocardia, Legionella, Listeria)—predominate. Rarely, early infec­
tions represent those transmitted with the donor liver, either infections
present in the donor or infections acquired during procurement
processing. De novo viral hepatitis infections acquired from the donor
organ or, almost unheard of now, from transfused blood products
occur after typical incubation periods for these agents (well beyond
the first month). Obviously, infections in an immunosuppressed host
demand early recognition and prompt management; prophylactic
antibiotic therapy is administered routinely in the immediate postop­
erative period. Use of sulfamethoxazole with trimethoprim reduces the
incidence of postoperative Pneumocystis jirovecii pneumonia. Antiviral
prophylaxis for CMV with ganciclovir should be administered in
patients at high risk (e.g., when a CMV-seropositive donor organ is
implanted into a CMV-seronegative recipient).
Neuropsychiatric complications include seizures (commonly associ­
ated with cyclosporine and tacrolimus toxicity), metabolic encephalop­
athy, depression, and difficult psychosocial adjustment. Rarely, diseases
are transmitted by the allograft from the donor to the recipient. In
addition to viral and bacterial infections, malignancies of donor ori­
gin have occurred. Posttransplantation lymphoproliferative disorders,
especially B-cell lymphoma, are a recognized complication associated
with immunosuppressive drugs such as azathioprine, tacrolimus, and
cyclosporine (see above). Epstein-Barr virus has been shown to play
a contributory role in some of these tumors, which may regress when
immunosuppressive therapy is reduced. De novo neoplasms appear at
increased frequency after liver transplantation, particularly squamous
cell carcinomas of the skin. Routine screening should be performed.
Long-term complications after liver transplantation attributable
primarily to immunosuppressive medications include diabetes mel­
litus and osteoporosis (associated with glucocorticoids and calcineu­
rin inhibitors) as well as hypertension, hyperlipidemia, and chronic
renal insufficiency (associated with cyclosporine and tacrolimus).
Monitoring and treating these disorders are routine components of
posttransplantation care; in some cases, they respond to changes in
immunosuppressive regimen, while in others, specific treatment of the
disorder is introduced. Data from a large U.S. database showed that the
prevalence of renal failure was 18% at year 5 and 25% at year 10 after
liver transplantation. Similarly, the high frequency of diabetes, hyper­
tension, hyperlipidemia, obesity, and the metabolic syndrome renders
patients susceptible to cardiovascular disease after liver transplanta­
tion; although hepatic complications account for most of the mortality
after liver transplantation, renal failure and cardiovascular disease are
the other leading causes of late mortality after liver transplantation.
■
■HEPATIC COMPLICATIONS
Hepatic dysfunction after liver transplantation is similar to the hepatic
complications encountered after major abdominal and cardiotho­
racic surgery; however, in addition, hepatic complications include
primary graft failure, vascular compromise, failure or stricture of the
biliary anastomoses, and rejection. As in nontransplantation surgery,
postoperative jaundice may result from prehepatic, intrahepatic,
and posthepatic sources. Prehepatic sources represent the massive
hemoglobin pigment load from transfusions, hemolysis, hematomas,
ecchymoses, and other collections of blood. Early intrahepatic liver
injury includes effects of hepatotoxic drugs and anesthesia; hypoperfu­
sion injury associated with hypotension, sepsis, and shock; and benign
postoperative cholestasis. Late intrahepatic sources of liver injury
include exacerbation of primary disease. Posthepatic sources of hepatic
dysfunction include biliary obstruction and reduced renal clearance
of conjugated bilirubin. Hepatic complications unique to liver trans­
plantation include primary graft failure associated with ischemic injury
to the organ during harvesting; vascular compromise associated with
thrombosis or stenosis of the portal vein or hepatic artery anastomoses;
vascular anastomotic leak; stenosis, obstruction, or leakage of the anas­
tomosed common bile duct; recurrence of primary hepatic disorder
(see below); and rejection.
CHAPTER 356
Liver Transplantation
■
■ALLOGRAFT REJECTION
Despite the use of immunosuppressive drugs, rejection of the trans­
planted liver still occurs in a proportion of patients, beginning 1–2 weeks
after surgery. Clinical signs suggesting rejection may include fever,
right upper quadrant pain, and reduced bile pigment and volume.
Leukocytosis may occur, but the most reliable indicators are increases
in serum bilirubin and aminotransferase levels. Because these tests lack
specificity, and because patients can be asymptomatic, distinguishing
among rejection, biliary obstruction, primary graft nonfunction, vas­
cular compromise, viral hepatitis, CMV infection, drug hepatotoxic­
ity, and recurrent primary disease may prove difficult. Radiographic
visualization of the biliary tree and/or percutaneous liver biopsy often
help to establish the correct diagnosis. Morphologic features of acute
rejection include a mixed portal cellular infiltrate, bile duct injury, and/
or endothelial inflammation (“endothelialitis”); some of these findings
are reminiscent of graft-versus-host disease, primary biliary cholangi­
tis, or recurrent allograft hepatitis C. As soon as allograft rejection is
suspected, treatment consists of increased immunosuppression, most
commonly IV methylprednisolone in repeated boluses; if this fails
to abort rejection, consideration is given to thymoglobulin. Caution
should be exercised when managing acute rejection with pulse gluco­
corticoids in patients with HCV infection, because of the high risk of
triggering recurrent allograft hepatitis C; however, the availability of
DAAs for HCV can obviate this concern effectively.
Chronic rejection is a relatively rare outcome that can follow repeated
bouts of acute rejection or that occurs unrelated to preceding rejection
episodes. Morphologically, chronic rejection is characterized by pro­
gressive cholestasis, focal parenchymal necrosis, mononuclear infiltra­
tion, vascular lesions (intimal fibrosis, subintimal foam cells, fibrinoid
necrosis), and fibrosis. This process may be reflected as ductopenia—
the vanishing bile duct syndrome, which is more common in patients
undergoing liver transplantation for autoimmune liver disease. Revers­
ibility of chronic rejection is limited; in patients with therapy-resistant
chronic rejection, retransplantation has yielded encouraging results.
OUTCOME
■
■SURVIVAL
The survival rate for patients undergoing liver transplantation has
improved steadily since 1983. One-year survival rates have increased
from ~70% in the early 1980s to 85–90% from 2003 to the present time.
Currently, the 5-year survival rate exceeds 60%. An important observa­
tion is the relationship between clinical status before transplantation
and outcome. For patients who undergo liver transplantation when
their level of compensation is high (e.g., still working or only partially
disabled), a 1-year survival rate of >85% is common. For those whose
level of decompensation mandates continuous in-hospital care prior
to transplantation, the 1-year survival rate is ~70%, whereas for those
who are so decompensated that they require life support in an intensive
care unit, the 1-year survival rate is ~50%. Since the adoption by UNOS
in 2002 of the MELD system for organ allocation, posttransplantation
survival has been found to be affected adversely for candidates with
MELD scores >25, considered high disease severity. Thus, irrespec­
tive of allocation scheme, high disease severity before transplantation
corresponds to diminished posttransplantation survival. Another
important distinction in survival has been drawn between high- and
low-risk patient categories. For patients who do not fit any “high-risk”
designations, 1- and 5-year survival rates of 85% and 80%, respectively,
have been recorded. In contrast, among patients in high-risk categories—
cancer, fulminant hepatitis, age >65, concurrent renal failure, respira­
tor dependence, portal vein thrombosis, and history of a portacaval
shunt or multiple right upper quadrant operations—survival statistics
fall into the range of 60% at 1 year and 35% at 5 years. Survival after
retransplantation for primary graft nonfunction is ~50%. Causes of
failure of liver transplantation vary with time. Failures within the first
3 months result primarily from technical complications, postoperative
infections, and hemorrhage. Transplant failures after the first 3 months
are more likely to result from infection, rejection, or recurrent disease
(such as malignancy, viral hepatitis, or return to alcohol use).
PART 10
Disorders of the Gastrointestinal System
■
■RECURRENCE OF PRIMARY DISEASE
Features of autoimmune hepatitis, primary sclerosing cholangitis, and
primary biliary cholangitis overlap with those of rejection or post­
transplantation bile duct injury. Autoimmune hepatitis and sclerosing
cholangitis can recur after liver transplantation. Similarly, reports of
recurrent primary biliary cholangitis after liver transplantation have
appeared; however, the histologic features of primary biliary cholan­
gitis and chronic rejection are virtually indistinguishable and occur as
frequently in patients with primary biliary cholangitis as in patients
undergoing transplantation for other reasons. The presence of a florid
inflammatory bile duct lesion is highly suggestive of the recurrence
of primary biliary cholangitis, but even this lesion can be observed in
acute rejection. Hereditary disorders such as Wilson’s disease and α1
antitrypsin deficiency have not recurred after liver transplantation;
however, recurrence of disordered iron metabolism has been observed
in some patients with hemochromatosis. Hepatic vein thrombosis
(Budd-Chiari syndrome) may recur; this can be minimized by treat­
ing underlying myeloproliferative disorders and by anticoagulation.
Because cholangiocarcinoma recurs almost invariably, few centers now
offer transplantation to such patients; however, a few highly selected
patients with operatively confirmed stage I or II cholangiocarci­
noma who undergo liver transplantation combined with neoadjuvant
chemoradiation may experience excellent outcomes. In patients with
intrahepatic HCC who meet criteria for transplantation, 1- and 5-year
survivals are similar to those observed in patients undergoing liver
transplantation for nonmalignant disease. Finally, metabolic disorders
such as MASLD are seen de novo and recur frequently, especially if the
underlying metabolic predisposition is not altered.
Hepatitis A can recur after transplantation for fulminant hepatitis A,
but such acute reinfection has no serious clinical sequelae. In fulminant
hepatitis B, recurrence is not the rule; however, in the absence of any
prophylactic measures, hepatitis B usually recurs after transplantation
for end-stage chronic hepatitis B. Before the introduction of prophylac­
tic antiviral therapy, immunosuppressive therapy sufficient to prevent
allograft rejection led inevitably to marked increases in hepatitis B
viremia, regardless of pretransplantation levels. Overall graft and patient
survival were poor, and some patients experienced a rapid recapitulation
of severe injury—severe chronic hepatitis or even fulminant hepatitis—
after transplantation. Also recognized in the era before availability of
antiviral regimens was fibrosing cholestatic hepatitis, rapidly progressive
liver injury associated with marked hyperbilirubinemia, substantial pro­
longation of the prothrombin time (both out of proportion to relatively
modest elevations of aminotransferase activity), and rapidly progressive
liver failure. This lesion has been suggested to represent a “choking off”
of the hepatocyte by an overwhelming density of HBV proteins. Com­
plications such as sepsis and pancreatitis were also observed more fre­
quently in patients undergoing liver transplantation for hepatitis B prior
to the introduction of antiviral therapy. The introduction of long-term
prophylaxis with HBIg revolutionized liver transplantation for chronic
hepatitis B. Preoperative hepatitis B vaccination, preoperative or postop­
erative interferon (IFN) therapy, or short-term (≤2 months) HBIg pro­
phylaxis has not been shown to be effective, but a retrospective analysis
of data from several hundred European patients followed for 3 years after
transplantation has shown that long-term (≥6 months) prophylaxis with
HBIg is associated with a lowering of the risk of HBV reinfection from
~75 to 35% and a reduction in mortality from ~50 to 20%.
As a result of long-term HBIg use following liver transplantation
for chronic hepatitis B, similar improvements in outcome have been
observed in the United States, with 1-year survival rates between 75 and
90%. Currently, with HBIg prophylaxis, the outcome of liver transplan­
tation for chronic hepatitis B is indistinguishable from that for chronic
liver disease unassociated with chronic hepatitis B; essentially, medical
concerns regarding liver transplantation for chronic hepatitis B have
been eliminated. Passive immunoprophylaxis with HBIg is begun dur­
ing the anhepatic stage of surgery, repeated daily in the postoperative
days for high-risk patients (active viremia at the time of transplanta­
tion and/or co-infection with hepatitis D or HIV), and then continued
with infusions that are given either at regular intervals of 4–6 weeks
or, alternatively, when antibody to hepatitis B surface (anti-HBs) levels
fall below a threshold of 100 mIU/mL. The current approach in most
centers is to continue HBIg indefinitely in select high-risk cases, which
can add ~$20,000 per year to the cost of care, and to transition to
maintenance with antiviral alone in low-risk-hepatitis B liver allograft
recipients. Still, “breakthrough” HBV infection occurs occasionally.
Further improving the outcome of liver transplantation for chronic
hepatitis B is the current availability of such antiviral drugs as entecavir,
tenofovir disoproxil fumarate, and tenofovir alafenamide (Chap. 352).
When these drugs are administered to patients with decompensated
liver disease, a proportion improves sufficiently to postpone imminent
liver transplantation. In addition, antiviral therapy can be used to
prevent recurrence of HBV infection when administered prior to trans­
plantation; to treat hepatitis B that recurs after transplantation, includ­
ing in patients who break through HBIg prophylaxis; and to reverse
the course of otherwise fatal fibrosing cholestatic hepatitis. Clinical
trials have shown that entecavir or tenofovir antiviral therapy reduces
the level of HBV replication substantially, sometimes even resulting in
clearance of hepatitis B surface antigen (HBsAg); reduces alanine ami­
notransferase (ALT) levels; and improves histologic features of necrosis
and inflammation. Currently, most liver transplantation centers com­
bine HBIg plus one of the high-barrier-to-resistance oral nucleoside
(entecavir) or nucleotide analogues (tenofovir). In low-risk patients
with no detectable hepatitis B viremia at the time of transplantation,
results from a number of clinical trials have suggested that antiviral
prophylaxis can suffice, without HBIg or with a finite duration of
HBIg, to prevent recurrent HBV infection of the allograft. In patients
documented at the time of liver transplantation to have undetectable
HBV DNA in serum and covalently closed circular DNA in the liver
(i.e., with low risk for recurrence of HBV infection), results of a clinical
trial suggested that, after receipt of 5 years of combined therapy, both
HBIg and oral-agent therapy can be withdrawn sequentially (over two
6-month periods) with a success rate, as monitored over a median of
6 years after withdrawal, of 90% and an anti-HBs seroconversion rate
of 60% (despite transient reappearance of HBV DNA and/or HBsAg in
some of these patients).
Antiviral prophylactic approaches applied to patients undergoing
liver transplantation for chronic hepatitis B are being used as well
for patients without hepatitis B who receive organs from donors with
antibody to hepatitis B core antigen (anti-HBc) but who do not have
detectable HBsAg. Patients who undergo liver transplantation for
chronic hepatitis B plus D are less likely to experience recurrent liver
injury than patients undergoing liver transplantation for hepatitis B
alone; still, such co-infected patients would also be offered standard
posttransplantation prophylactic antiviral therapy for hepatitis B.
Until recently, the most common indication for liver transplantation
was end-stage liver disease resulting from chronic hepatitis C (Fig. 356-2).
For patients undergoing liver transplantation for hepatitis C, because
of an aggressive natural history of recurrent allograft hepatitis C, graft
and patient survival were diminished substantially compared to other
indications for transplantation.
The approval over the last decade of several DAA agents and of
IFN-free DAA regimens against HCV has had a major impact on the
Percent
A
Year
FIGURE 356-2  Trends in liver transplantation. A. Candidates waiting for liver transplantation by indication at any time in the given year. Candidates listed at more than one
center are counted once per listing. Active and inactive patients are included. B. All liver allograft recipients by indication, including adult and pediatric, retransplantation,
and multiorgan recipients. HCC, hepatocellular carcinoma; HCV, hepatitis C virus; MASH, metabolic dysfunction–associated steatohepatitis. (From the Scientific Registry of
Transplant Recipients (SRTR.org).)
management and outcome of both pretransplantation and posttrans­
plantation HCV infection. Such therapeutic approaches (1) permit
the clearance of viremia in a substantial proportion of decompensated
cirrhotics, thereby preventing recurrent allograft infection and even
improving the clinical status of most of these patients, delaying or obvi­
ating the need for liver replacement; and (2) achieve sustained virologic
responses in a much higher proportion of persons with allograft HCV
infection, because of improvements in antiviral treatment efficacy and
tolerability. Ideally, patients should be treated prior to liver transplanta­
tion. This approach has already reduced the numbers of patients with
HCV infection referred for liver transplantation and led to delisting of
others. A concern, however, is that eradication of HCV infection will
reduce the MELD score and lower the priority for a donor organ in
some patients who still require transplantation because of continued
hepatic decompensation and profound reduction in quality of life (a
situation referred to as MELD purgatory). In addition, elimination
of HCV infection prior to transplantation, historically, would have
disqualified such patients as recipients of donor livers from persons
with HCV infection, contracting the potential donor pool and limit­
ing accessibility to donor organs and timely transplantation. With
the advent of highly effective DAAs, however, allografts from HCVinfected donors are being transplanted into both HCV-infected and
HCV-uninfected recipients, who can be treated posttransplantation
with DAAs. Thus, earlier concerns about the potential impact of pre­
transplantation DAA treatment on limiting the donor pool are no lon­
ger an issue. The approach to HCV treatment should be individualized
thoughtfully for each patient, based on such factors as MELD score,
time anticipated prior to availability of a donor organ, relative clinical
stability, and comorbidities.
CHAPTER 356
DAA combinations that are being used successfully against allograft
HCV include ledipasvir plus sofosbuvir; velpatasvir plus sofosbuvir;
and grazoprevir plus pibrentasvir. (For updated guidelines, see www.
hcvguidelines.org.) In patients with recurrent HCV infection after
Liver Transplantation
Acute liver failure
HCV
Alcoholic liver disease
Cholestatic disease
HCC
MASH
Other/unknown
Transplants
PART 10
Disorders of the Gastrointestinal System
Year
B
FIGURE 356-2  (Continued)
liver transplantation, each of these regimens has yielded response
rates approaching those seen in compensated nontransplant patient
populations.
Historically, small numbers of allograft recipients have succumbed
to early HCV-associated liver injury, and a syndrome reminiscent of
fibrosing cholestatic hepatitis (see above) has been observed rarely.
Currently, however, the routine use of DAA regimens early after trans­
plantation, before the onset of these variant presentations, has already
had a profound impact on the frequency of severe recurrent allograft
hepatitis C.
Patients who undergo liver transplantation for end-stage alcoholassociated cirrhosis are at risk of a return to drinking again after
transplantation, a potential source of recurrent alcohol-associated liver
injury. Currently, alcohol-associated liver disease is the most common
indication for liver transplantation, accounting for 40% of all liver
transplantation procedures, and most transplantation centers screen
candidates carefully for predictors of continued abstinence. Return to
alcohol use is more likely in patients whose sobriety prior to transplan­
tation was <6 months. For abstinent patients with alcohol-associated
cirrhosis, liver transplantation can be undertaken successfully, with
outcomes comparable to those for other categories of patients with
chronic liver disease, when coordinated by a team approach that
includes substance use counseling.
The syndemic of hazardous alcohol consumption, opioid use, and
obesity continues to influence important changes in liver disease epi­
demiology, health disparities, indications for transplant (Fig. 356-2),
and disease recurrence.
■
■POSTTRANSPLANTATION QUALITY OF LIFE
Full rehabilitation is achieved in most patients who survive the early
postoperative months and escape chronic rejection or unmanageable
infection. Psychosocial maladjustment interferes with medical com­
pliance in a small number of patients, but most manage to adhere to
immunosuppressive regimens, which must be continued indefinitely.
Acute liver failure
HCV
Alcoholic liver disease
Cholestatic disease
HCC
MASH
Other/unknown
In one study, 85% of patients who survived their transplant operations
returned to gainful activities. In fact, some women have conceived
and carried pregnancies to term after transplantation without demon­
strable injury to their infants.
■
■FURTHER READING
Bethea  E et al: Immediate administration of antiviral therapy after
transplantation of hepatitis C-infected livers into uninfected recipients:
Implications for therapeutic planning. Am J Transplant 20:1619, 2020.
Bhattacharya  D et al: Hepatitis C Guidance 2023 Update: American
Association for the Study of Liver Diseases-Infectious Diseases Soci­
ety of America recommendations for testing, managing, and treating
hepatitis C virus infection. Clin Infect Dis 2023. (Updated regularly,
available at http://www.hcvguidelines.org. Accessed September 23, 2023.)
Brustia  R et al: Guidelines for perioperative care for liver transplanta­
tion: Enhanced Recovery After Surgery (ERAS) recommendations.
Transplantation 106:552, 2022.
European Association for the Study of the Liver: EASL Clini­
cal Practice Guidelines on acute-on-chronic liver failure. J Hepatol
79:461, 2023.
Goldberg D et al: Changes in the prevalence of hepatitis C virus
infection, nonalcoholic steatohepatitis, and alcoholic liver disease
among patients with cirrhosis or liver failure on the waitlist for liver
transplantation. Gastroenterology 152:1090, 2017.
Kim  WR et al: MELD 3.0: The Model for End-State Liver Disease
updated for the modern era. Gastroenterology 161:1887, 2021.
Louvet A et al: Early liver transplantation for severe alcohol-related
hepatitis not responding to medical treatment: A prospective con­
trolled study. Lancet Gastroenterol Hepatol 7:416, 2022.
Lucey MR et al: Liver transplantation. N Engl J Med 389:1888, 2023.
Nephew  LD, Serper M: Racial, gender, and socioeconomic disparities
in liver transplantation. Liver Transpl 27:900, 2021.
Terrault NA et al: Liver transplantation 2023: Status report, current
and future challenges. Clin Gastroeterol Hepatol 21:2150, 2023.

29 - 357 Diseases of the Gallbladder and Bile Ducts
357 Diseases of the Gallbladder and Bile Ducts
Norton J. Greenberger*, Gustav Paumgartner*,
Daniel S. Pratt
Diseases of the
Gallbladder and Bile Ducts
PHYSIOLOGY OF BILE PRODUCTION
AND FLOW
■
■BILE SECRETION AND COMPOSITION
Bile formed in hepatocytes is secreted into a complex network of
canaliculi lined by hepatocytes that flow into the canals of Hering
lined partially by hepatocytes and partially by cholangiocytes, leading
to small bile ductules and larger bile ducts lined by cholangiocytes that
run with lymphatics and branches of the portal vein and hepatic artery
in portal tracts situated between hepatic lobules. These interlobular
bile ducts coalesce to form larger septal bile ducts that join to form the
right and left hepatic ducts, which in turn, unite to form the common
hepatic duct. The common hepatic duct is joined by the cystic duct of
the gallbladder to form the common bile duct (CBD), which enters the
duodenum (often after joining the main pancreatic duct) through the
ampulla of Vater.
Hepatic bile is an isotonic fluid with an electrolyte composition
resembling blood plasma. The electrolyte composition of gallbladder
bile differs from that of hepatic bile because most of the inorganic
anions, chloride, and bicarbonate have been removed by reabsorption
across the gallbladder epithelium. As a result of water reabsorption,
total solute concentration of bile increases from 3–4 g/dL in hepatic
bile to 10–15 g/dL in gallbladder bile.
Major solute components of bile by moles percent include bile acids
(80%), phospholipids (lecithins, cephalins, and sphingomyelin) (16%),
and unesterified cholesterol (4.0%). In the lithogenic state, the choles­
terol value can be as high as 8–10%. Other constituents include conju­
gated bilirubin; proteins (all immunoglobulins, albumin, metabolites
of hormones, and other proteins metabolized in the liver); electrolytes;
mucus; heavy metals; and, often, drugs and their metabolites.
The total daily basal secretion of hepatic bile is ~500–600 mL. Many
substances taken up or synthesized by the hepatocyte are secreted into
the bile canaliculi. The canalicular membrane forms microvilli and is
associated with microfilaments of actin, microtubules, and other con­
tractile elements. Prior to their secretion into the bile, many substances
are taken up into the hepatocyte, while others, such as phospholipids,
a portion of primary bile acids, and some cholesterol, are synthesized
de novo in the hepatocyte. Three mechanisms are important in regu­
lating bile flow: (1) active transport of bile acids from hepatocytes into
the bile canaliculi, (2) active transport of other organic anions, and
(3) cholangiocellular secretion. The last is a secretin-mediated and
cyclic AMP–dependent mechanism that results in the secretion of a
bicarbonate-rich fluid into the bile ducts.
Active vectorial trans-hepatocellular movement of bile acids from
the portal blood into the bile canaliculi is driven by a set of trans­
port systems at the basolateral (sinusoidal) and the canalicular apical
plasma membrane domains of the hepatocyte. Two sinusoidal bile salt
uptake systems have been cloned in humans, the Na+/taurocholate
cotransporter (NTCP, SLC10A1) and the organic anion–transporting
proteins (OATP1B1/1B3), which also transport a large variety of non–
bile salt organic anions. Several ATP-dependent canalicular transport
systems, “export pumps” (ATP-binding cassette transport proteins,
also known as ABC transporters), have been identified, the most
important of which are the bile salt export pump (BSEP, ABCB11); the
anionic conjugate export pump (MRP2, ABCC2), which mediates the
canalicular excretion of various amphiphilic conjugates formed by
phase II conjugation (e.g., bilirubin mono- and diglucuronides and
*Deceased.
drugs); the multidrug export pump (MDR1, ABCB1) for hydropho­
bic cationic compounds; and the phospholipid export pump (MDR3,
ABCB4). Two hemitransporters, ABCG5/G8, functioning as a couple,
constitute the canalicular cholesterol and phytosterol transporter. F1C1
(ATP8B1) is an aminophospholipid transferase (“flippase”) essential
for maintaining the lipid asymmetry of the canalicular membrane. The
canalicular membrane also contains ATP-independent transport sys­
tems such as the Cl/HCO3 anion exchanger isoform 2 (AE2, SLC4A2)
for canalicular bicarbonate secretion. For most of these transporters,
genetic defects have been identified that are associated with various
forms of cholestasis or defects of biliary excretion. F1C1 (ATP8B1) is
defective in progressive familial intrahepatic cholestasis type 1 (PFIC1)
and benign recurrent intrahepatic cholestasis type 1 (BRIC1) and
results in ablation of all other ATP-dependent transporter functions.
BSEP (ABCB11) is defective in PFIC2 and BRIC2. Mutations of MRP2
(ABCC2) cause the Dubin-Johnson syndrome, an inherited form
of conjugated hyperbilirubinemia (Chap. 349). A defective MDR3
(ABCB4) results in PFIC3. ABCG5/G8, the canalicular half transport­
ers for cholesterol and other neutral sterols, are defective in sitosterol­
emia. The cystic fibrosis transmembrane regulator (CFTR, ABCC7),
located on bile duct epithelial cells but not on canalicular membranes,
is defective in cystic fibrosis, which is associated with impaired cholan­
giocellular pH regulation during ductular bile formation and chronic
cholestatic liver disease, occasionally resulting in biliary cirrhosis.
■
■THE BILE ACIDS
The primary bile acids, cholic acid and chenodeoxycholic acid (CDCA),
are synthesized in hepatocytes from cholesterol, conjugated with gly­
cine or taurine, and secreted into the bile canaliculus. Secondary bile
acids, including deoxycholate and lithocholate, are formed in the colon
as bacterial metabolites of the primary bile acids. However, lithocholic
acid is much less efficiently absorbed from the colon than deoxycho­
lic acid. Another secondary bile acid, found in low concentration, is
ursodeoxycholic acid (UDCA), a stereoisomer of CDCA. In healthy
subjects, the ratio of glycine to taurine conjugates in bile is ~3:1.
CHAPTER 357
Diseases of the Gallbladder and Bile Ducts
Bile acids are detergent-like molecules that in aqueous solutions
and above a critical concentration of ~2 mM form molecular aggre­
gates called micelles. Cholesterol alone is sparingly soluble in aqueous
environments, and its solubility in bile depends on both the total lipid
concentration and the relative molar percentages of bile acids and leci­
thin. Normal ratios of these constituents favor the formation of solubi­
lizing mixed micelles, while abnormal ratios promote the precipitation
of cholesterol crystals in bile via an intermediate liquid crystal phase.
In addition to facilitating the biliary excretion of cholesterol, bile
acids facilitate the normal intestinal absorption of dietary fats, mainly
cholesterol, and fat-soluble vitamins, via a micellar transport mecha­
nism (Chap. 336). Bile acids also serve as a major physiologic driving
force for hepatic bile flow and aid in water and electrolyte transport in
the small bowel and colon.
Bile acids also function as hormones binding to nuclear (farnesoid X
receptor [FXR]) and G protein–coupled (TGR5) receptors that regulate
bile acid metabolism and their enterohepatic circulation.
■
■ENTEROHEPATIC CIRCULATION
Bile acids are efficiently conserved under normal conditions. Uncon­
jugated, and to a lesser degree also conjugated, bile acids are absorbed
by passive diffusion along the entire gut. Quantitatively much more
important for bile salt recirculation, however, is the active transport
mechanism for conjugated bile acids in the distal ileum (Chap. 336).
The reabsorbed bile acids enter the portal bloodstream and are taken
up rapidly by hepatocytes, reconjugated, and resecreted into bile
(enterohepatic circulation).
The normal bile acid pool size is ~2–4 g. During digestion of a meal,
the bile acid pool undergoes at least one or more enterohepatic cycles,
depending on the size and composition of the meal. Normally, the bile
acid pool circulates ~5–10 times daily. Intestinal reabsorption of the
pool is ~95% efficient; therefore, daily fecal loss of bile acids is in the
range of 0.2–0.4 g. In the steady state, this fecal loss is compensated by
an equal daily synthesis of bile acids by the liver, and thus, the size
of the bile acid pool is maintained. Bile acids in the intestine stimulate
the release of fibroblast growth factor 19 (FGF19), which suppresses
the hepatic synthesis of bile acids from cholesterol by inhibiting the
rate-limiting enzyme cytochrome P450 7A1 (CYP7A1). FGF19 also
promotes gallbladder relaxation. While the loss of bile salts in stool is
usually matched by increased hepatic synthesis, the maximum rate of
synthesis is ~5 g/d, which may be insufficient to replete the bile acid
pool size when there is pronounced impairment of intestinal bile salt
reabsorption.
The expression of ABC transporters in the enterohepatic circulation
and of the rate-limiting enzymes of bile acid and cholesterol synthesis
are regulated in a coordinated fashion by nuclear receptors, which are
ligand-activated transcription factors. The hepatic BSEP (ABCB11)
is upregulated by the FXR that also represses bile acid synthesis. The
expression of the cholesterol transporter, ABCG5/G8, is upregulated by
the liver X receptor (LXR), which is an oxysterol sensor.
■
■GALLBLADDER AND SPHINCTERIC FUNCTIONS
In the fasting state, the sphincter of Oddi (SOD) offers a high-pressure
zone of resistance to bile flow from the CBD into the duodenum. Its
tonic contraction serves to (1) prevent reflux of duodenal contents into
the pancreatic and bile ducts and (2) promote filling of the gallblad­
der. The major factor controlling the evacuation of the gallbladder is
the peptide hormone cholecystokinin (CCK), which is released from
the duodenal mucosa in response to the ingestion of fats and amino
acids. CCK produces (1) powerful contraction of the gallbladder, (2)
decreased resistance of the SOD, and (3) enhanced flow of biliary con­
tents into the duodenum.
Hepatic bile is “concentrated” within the gallbladder by energydependent transmucosal absorption of water and electrolytes. Almost
the entire bile acid pool may be sequestered in the gallbladder follow­
ing an overnight fast for delivery into the duodenum with the first meal
of the day. The normal capacity of the gallbladder is ~30 mL.
PART 10
Disorders of the Gastrointestinal System
DISEASES OF THE GALLBLADDER
■
■CONGENITAL ANOMALIES
Anomalies of the biliary tract are not uncommon and include abnor­
malities in number, size, and shape (e.g., agenesis of the gallbladder,
duplications, rudimentary or oversized “giant” gallbladders, and diver­
ticula). Phrygian cap is a clinically innocuous entity in which a partial
or complete septum (or fold) separates the fundus from the body.
Anomalies of position or suspension are not uncommon and include
left-sided gallbladder, intrahepatic gallbladder, retrodisplacement of
the gallbladder, and “floating” gallbladder. The latter condition predis­
poses to acute torsion, volvulus, or herniation of the gallbladder.
■
■GALLSTONES
Epidemiology and Pathogenesis 
Gallstones are quite prevalent
in most Western countries. Gallstone formation increases after age 50.
In the United States, the prevalence is highest in Native Americans fol­
lowed by Hispanics, non-Hispanic whites, and black Americans. The
prevalence is higher in women than men across all ages.
Gallstones form because of abnormal bile composition. They are
divided into two major types: cholesterol stones and pigment stones.
Cholesterol stones account for >90% of all gallstones in Western indus­
trialized countries. Cholesterol gallstones usually contain >50% choles­
terol monohydrate plus an admixture of calcium salts, bile pigments,
proteins, and fatty acids. Pigment stones are composed primarily of
calcium bilirubinate; they contain <20% cholesterol and are classified
into “black” and “brown” types, the latter forming secondary to chronic
biliary infection.
CHOLESTEROL STONES AND BILIARY SLUDGE  Cholesterol is essen­
tially water-insoluble and requires aqueous dispersion into either
micelles or vesicles, both of which require the presence of a second
lipid to solubilize the cholesterol. Cholesterol and phospholipids are
secreted into bile as unilamellar bilayered vesicles, which are con­
verted into mixed micelles consisting of bile acids, phospholipids, and
ABCG5/G8
CYP7A1
MDR3 (ABCB4)
I.
Normal cholesterol
Normal bile acids
Lecithin
Cholesterol
Normal bile acids
Normal lecithin
Normal cholesterol
Bile acids
Normal lecithin
II. Supersaturation
Inhibit nucleation
Apolipoproteins
Lecithin vesicles
Promote nucleation
Mucous glycoproteins
Heat-labile proteins
III. Nucleation
IV. Microstone
Gallstone
FIGURE 357-1  Scheme showing pathogenesis of cholesterol gallstone
formation. Conditions or factors that increase the ratio of cholesterol to bile
acids and phospholipids (lecithin) favor gallstone formation. ABCB4, ATP-binding
cassette transporter; ABCG5/8, ATP-binding cassette (ABC) transporter G5/G8;
CYP7A1, cytochrome P450 7A1; MDR3, multidrug resistance protein 3, also called
phospholipid export pump.
cholesterol by the action of bile acids. If there is an excess of cholesterol
in relation to phospholipids and bile acids, unstable, cholesterol-rich
vesicles remain, which aggregate into large multilamellar vesicles from
which cholesterol crystals precipitate (Fig. 357-1).
There are several important mechanisms in the formation of
lithogenic (stone-forming) bile. The most important is increased
biliary secretion of cholesterol. This may occur in association with
obesity, the metabolic syndrome, high-caloric and cholesterol-rich
diets, or drugs (e.g., clofibrate) and may result from increased activity
of hydroxymethylglutaryl-coenzyme A (HMG-CoA) reductase, the
rate-limiting enzyme of hepatic cholesterol synthesis, and increased
hepatic uptake of cholesterol from blood. In patients with gallstones,
dietary cholesterol increases biliary cholesterol secretion. This does not
occur in nongallstone patients on high-cholesterol diets. In addition to
environmental factors such as high-caloric and cholesterol-rich diets,
genetic factors play an important role in gallstone disease. A large
study of symptomatic gallstones in Swedish twins provided strong
evidence for a role of genetic factors in gallstone pathogenesis. Genetic
factors accounted for 25%, shared environmental factors for 13%, and
individual environmental factors for 62% of the phenotypic variation
among monozygotic twins. A single nucleotide polymorphism of the
gene encoding the hepatic cholesterol transporter ABCG5/G8 has
been found in 21% of patients with gallstones, but only in 9% of the
general population. It is thought to cause a gain of function of the
cholesterol transporter and to contribute to cholesterol hypersecretion.
A high prevalence of gallstones is found among first-degree relatives of
gallstone carriers and in certain ethnic populations such as American
Indians, Chilean Indians, and Chilean Hispanics. A common genetic
trait has been identified for some of these populations by mitochon­
drial DNA analysis. In some patients, impaired hepatic conversion of
cholesterol to bile acids may also occur, resulting in an increase of the
lithogenic cholesterol/bile acid ratio. Although most cholesterol stones
have a polygenic basis, there are rare monogenic (Mendelian) causes.
Mutations in the CYP7A1 gene have been described that result in a
deficiency of the enzyme cholesterol 7-hydroxylase, which catalyzes
the initial step in cholesterol catabolism and bile acid synthesis. The
homozygous state is associated with hypercholesterolemia and gall­
stones. Because the phenotype is expressed in the heterozygote state,
mutations in the CYP7A1 gene may contribute to the susceptibility to
cholesterol gallstone disease in the population. Mutations in the MDR3
(ABCB4) gene, which encodes the phospholipid export pump in the
canalicular membrane of the hepatocyte, may cause defective phospho­
lipid secretion into bile, resulting in cholesterol supersaturation of bile
and formation of cholesterol gallstones in the gallbladder and in the
bile ducts. Thus, an excess of biliary cholesterol in relation to bile acids
and phospholipids is primarily due to hypersecretion of cholesterol,
but hyposecretion of bile acids or phospholipids may contribute. An
additional disturbance of bile acid metabolism that is likely to con­
tribute to supersaturation of bile with cholesterol is enhanced conver­
sion of cholic acid to deoxycholic acid, with replacement of the cholic
acid pool by an expanded deoxycholic acid pool. It may result from
enhanced dehydroxylation of cholic acid and increased absorption of
newly formed deoxycholic acid. An increased deoxycholate secretion is
associated with hypersecretion of cholesterol into bile. A recent study
suggested a role for the gut microbiota as an environmental factor con­
tributing to gallstone formation through an influence on cholesterol
and bile acid metabolism.
While supersaturation of bile with cholesterol is an important pre­
requisite for gallstone formation, it is generally not sufficient by itself
to produce cholesterol precipitation in vivo. Most individuals with
supersaturated bile do not develop stones because the time required
for cholesterol crystals to nucleate and grow is longer than the time bile
remains in the gallbladder.
An important mechanism is nucleation of cholesterol monohydrate
crystals, which is greatly accelerated in human lithogenic bile. Accel­
erated nucleation of cholesterol monohydrate in bile may be due to
either an excess of pronucleating factors or a deficiency of antinucleating
factors. Mucin and certain nonmucin glycoproteins, principally immu­
noglobulins, appear to be pronucleating factors, while apolipoproteins
A-I and A-II and other glycoproteins appear to be antinucleating
factors. Pigment particles may possibly play a role as nucleating fac­
tors. In a genome-wide analysis of serum bilirubin levels, the uridine
diphosphate-glucuronyltransferase 1A1 (UGT1A1) Gilbert’s syndrome
gene variant was associated with the presence of gallstone disease. Because
most gallstones associated with the UGT1A1 variant were cholesterol
stones, this finding points to the role of pigment particles in the patho­
genesis of gallbladder stones. Cholesterol monohydrate crystal nucle­
ation and crystal growth probably occur within the mucin gel layer.
Vesicle fusion leads to liquid crystals, which, in turn, nucleate into solid
cholesterol monohydrate crystals. Continued growth of the crystals
occurs by direct nucleation of cholesterol molecules from supersatu­
rated unilamellar or multilamellar biliary vesicles.
A third important mechanism in cholesterol gallstone formation is
gallbladder hypomotility. If the gallbladder emptied all supersaturated
or crystal-containing bile completely, stones would not be able to grow.
A high percentage of patients with gallstones exhibits abnormalities
of gallbladder emptying. Ultrasonographic studies show that gallstone
patients display an increased gallbladder volume during fasting and
after a test meal (residual volume) and that fractional emptying after
gallbladder stimulation is decreased. The incidence of gallstones is
increased in conditions associated with infrequent or impaired gall­
bladder emptying such as fasting, parenteral nutrition, or pregnancy
and in patients using drugs that inhibit gallbladder motility.
Biliary sludge is a thick, mucous material that, upon microscopic
examination, reveals lecithin-cholesterol liquid crystals, cholesterol
monohydrate crystals, calcium bilirubinate, and mucin gels. Biliary
sludge typically forms a crescent-like layer in the most dependent
portion of the gallbladder and is recognized by characteristic echoes
on ultrasonography (see below). The presence of biliary sludge implies
two abnormalities: (1) the normal balance between gallbladder mucin
secretion and elimination has become deranged, and (2) nucleation
of biliary solutes has occurred. That biliary sludge may be a precursor
form of gallstone disease is evident from several observations. In one
study, 96 patients with gallbladder sludge were followed prospectively
by serial ultrasound studies. In 18%, biliary sludge disappeared and did
not recur for at least 2 years. In 60%, biliary sludge disappeared and
reappeared; in 14%, gallstones (8% asymptomatic, 6% symptomatic)
developed; and in 6%, severe biliary pain with or without acute pan­
creatitis occurred. In 12 patients, cholecystectomies were performed, 6
for gallstone-associated biliary pain and 3 in symptomatic patients with
sludge but without gallstones who had prior attacks of pancreatitis; the
latter did not recur after cholecystectomy. It should be emphasized that
biliary sludge can develop with disorders that cause gallbladder hypo­
motility; that is, surgery, burns, total parenteral nutrition, pregnancy,
and oral contraceptives—all of which are associated with gallstone
formation. However, the presence of biliary sludge implies supersatura­
tion of bile with either cholesterol or calcium bilirubinate.
Two other conditions are associated with cholesterol-stone or bili­
ary-sludge formation: pregnancy and rapid weight reduction through a
very-low-calorie diet. There appear to be two key changes during preg­
nancy that contribute to a “cholelithogenic state”: (1) a marked increase
in cholesterol saturation of bile during the third trimester and (2) slug­
gish gallbladder contraction in response to a standard meal, resulting
in impaired gallbladder emptying. That these changes are related to
pregnancy per se is supported by several studies that show reversal of
these abnormalities quite rapidly after delivery. During pregnancy, gall­
bladder sludge develops in 20–30% of women and gallstones in 5–12%.
Although biliary sludge is a common finding during pregnancy, it is
usually asymptomatic and often resolves spontaneously after delivery.
Gallstones, which are less common than sludge and frequently associ­
ated with biliary colic, may also disappear after delivery because of
spontaneous dissolution related to bile becoming unsaturated with
cholesterol postpartum.
CHAPTER 357
Diseases of the Gallbladder and Bile Ducts
Approximately 10–20% of persons with rapid weight reduction
achieved through very-low-calorie dieting develop gallstones. In a
study involving 600 patients who completed a 3-month, 520-kcal/d
diet, UDCA in a dosage of 600 mg/d proved highly effective in pre­
venting gallstone formation; gallstones developed in only 3% of UDCA
recipients, compared to 28% of placebo-treated patients. In obese
patients treated by gastric banding, 500 mg/d of UDCA reduced the
risk of gallstone formation from 30 to 8% within a follow-up of
6 months.
To summarize, cholesterol gallstone disease occurs because of sev­
eral defects, which include (1) bile supersaturation with cholesterol,
(2) nucleation of cholesterol monohydrate with subsequent crystal
retention and stone growth, and (3) abnormal gallbladder motor
function with delayed emptying and stasis. Other important factors
known to predispose to cholesterol-stone formation are summarized
in Table 357-1.
PIGMENT STONES  Black pigment stones are composed of either
pure calcium bilirubinate or polymer-like complexes with calcium
and mucin glycoproteins. They are more common in patients who
have chronic hemolytic states (with increased conjugated bilirubin
in bile); cirrhosis, especially related to alcohol; Gilbert’s syndrome;
or cystic fibrosis. Gallbladder stones in patients with ileal diseases,
ileal resection, or ileal bypass generally are also black pigment stones.
Enterohepatic recycling of bilirubin in ileal disease states contributes
to their pathogenesis. Brown pigment stones are composed of calcium
salts of unconjugated bilirubin with varying amounts of cholesterol
and protein. They are caused by the presence of increased amounts
of unconjugated, insoluble bilirubin in bile that precipitates to form
stones. Deconjugation of an excess of soluble bilirubin mono- and
diglucuronides may be mediated by endogenous β-glucuronidase but
may also occur by spontaneous hydrolysis. Sometimes, the enzyme is
also produced when bile is chronically infected by bacteria, and such
TABLE 357-1  Predisposing Factors for Cholesterol and Pigment
Gallstone Formation
Cholesterol Stones
Demographic/genetic factors: Prevalence highest in North American Indians,
Chilean Indians, and Chilean Hispanics, greater in Northern Europe and North
America than in Asia, lowest in Japan; familial disposition; hereditary aspects
Obesity, metabolic syndrome: Normal bile acid pool and secretion but
increased biliary secretion of cholesterol
Rapid weight loss: Mobilization of tissue cholesterol leads to increased
biliary cholesterol secretion while enterohepatic circulation of bile acids is
decreased
Female sex hormones
a.	 Estrogens stimulate hepatic lipoprotein receptors, increase uptake of
dietary cholesterol, and increase biliary cholesterol secretion
b.	 Natural estrogens, other estrogens, and oral contraceptives lead to
decreased bile salt secretion and decreased conversion of cholesterol to
cholesteryl esters
Pregnancy: Impaired gallbladder emptying caused by progesterone
combined with the influence of estrogens, which increase biliary cholesterol
secretion
Increasing age: Increased biliary secretion of cholesterol, decreased size of
bile acid pool, decreased secretion of bile salts
Gallbladder hypomotility leading to stasis and formation of sludge
a.	 Total parenteral nutrition
b.	 Fasting
c.	 Pregnancy
d.	 Drugs such as octreotide
Clofibrate therapy: Increased biliary secretion of cholesterol
Decreased bile acid secretion
PART 10
Disorders of the Gastrointestinal System
a.	 Genetic defect of the CYP7A1 gene
Decreased phospholipid secretion: Genetic defect of the MDR3 gene
Miscellaneous
a.	 High-calorie, high-fat diet
b.	 Spinal cord injury
Pigment Stones
Demographic/genetic factors: Asia, rural setting (presumed due to increased
prevalence of parasitic biliary infections; the incidence has been dropping
with time)
Chronic hemolysis (example: sickle cell disease)
Alcohol related liver cirrhosis
Ineffective erythropoiesis (example: pernicious anemia)
Cystic fibrosis
Chronic biliary tract infection, parasite infections
Increasing age
Ileal disease, ileal resection or bypass
stones are brown. Pigment stone formation is frequent in Asia and is
often associated with parasitic infections in the gallbladder and biliary
tree (Table 357-1).
Diagnosis 
Procedures of potential use in the diagnosis of choleli­
thiasis and other diseases of the gallbladder are detailed in Table 357-2.
Ultrasonography of the gallbladder is very accurate in the identifica­
tion of cholelithiasis and has replaced oral cholecystography (OCG)
(Fig. 357-2A). Stones as small as 1.5 mm in diameter may be con­
fidently identified provided that firm criteria are used (e.g., acoustic
“shadowing” of opacities that are within the gallbladder lumen and that
change with the patient’s position [by gravity]). In major medical cen­
ters, the false-negative and false-positive rates for ultrasound in gall­
stone patients are ~2–4%. Biliary sludge is material of low echogenic
activity that typically forms a layer in the most dependent position of
the gallbladder. This layer shifts with postural changes but fails to pro­
duce acoustic shadowing; these two characteristics distinguish sludges
from gallstones. Ultrasound can also be used to assess the emptying
function of the gallbladder.
The plain abdominal film may detect gallstones containing suf­
ficient calcium to be radiopaque (10–15% of cholesterol and ~50% of
TABLE 357-2  Diagnostic Evaluation of the Gallbladder
DIAGNOSTIC
ADVANTAGES
DIAGNOSTIC
LIMITATIONS
COMMENT
Ultrasound
Rapid
Accurate identification of
gallstones (>95%)
Simultaneous scanning
of GB, liver, bile ducts,
pancreas
“Real-time” scanning
allows assessment of GB
volume, contractility
Not limited by jaundice,
pregnancy
May detect very small
stones
Bowel gas
Massive obesity
Ascites
Procedure of choice for
detection of stones
Plain Abdominal X-Ray
Low cost
Readily available
Relatively low yield
Contraindicated in
pregnancy
Pathognomonic findings
in: calcified gallstones,
limey bile, porcelain GB,
emphysematous
cholecystitis,
gallstone ileus
Cholescintigraphy (HIDA, DISIDA, etc.)
Accurate identification of
cystic duct obstruction
Simultaneous
assessment of bile ducts
Contraindicated in
pregnancy
Serum bilirubin
103–205 μmol/L
(6–12 mg/dL)
Cholecystogram of low
resolution
Indicated for
confirmation of
suspected acute
cholecystitis; less
sensitive and less
specific in chronic
cholecystitis; useful
in the diagnosis
of acalculous
cholecystopathy,
especially if given
with CCK to assess GB
emptying
Abbreviations: CCK, cholecystokinin; DISIDA, diisopropyl iminodiacetic acid; GB,
gallbladder; HIDA, hydroxyl iminodiacetic acid.
pigment stones). Plain radiography may also be of use in the diagnosis
of emphysematous cholecystitis, porcelain gallbladder, limey bile, and
gallstone ileus.
OCG was historically a useful procedure for the diagnosis of gall­
stones but has been replaced by ultrasound and is now regarded as
obsolete. It may be used to assess the patency of the cystic duct and
gallbladder emptying function. Further, OCG can also delineate the
size and number of gallstones and determine whether they are calci­
fied, useful information if medical dissolution is being considered.
Radiopharmaceuticals such as 99mTc-labeled N-substituted iminodi­
acetic acids (HIDA and DISIDA) are rapidly extracted from the blood
and are excreted into the biliary tree in high concentration even in
the presence of mild to moderate serum bilirubin elevations. Failure
to image the gallbladder in the presence of biliary ductal visualization
may indicate cystic duct obstruction, acute or chronic cholecystitis,
or surgical absence of the organ. Such scans have application in the
diagnosis of acute cholecystitis and may play a role in the detection of
a postcholecystectomy bile leak.
Symptoms of Gallstone Disease 
Gallstones usually produce
symptoms by causing inflammation or obstruction following their
migration into the cystic duct or CBD. The most specific and charac­
teristic symptom of gallstone disease is biliary colic that is a constant
and often long-lasting pain (see below). Obstruction of the cystic
duct or CBD by a stone produces increased intraluminal pressure and
distention of the viscus that cannot be relieved by repetitive biliary
contractions. The resultant visceral pain is characteristically a severe,
chd
gb
cbd
shadow
pd
A
B
C
D
FIGURE 357-2  Examples of ultrasound and radiologic studies of the biliary tract. A. An ultrasound study showing a distended gallbladder (GB) containing a single large
stone (arrow), which casts an acoustic shadow. B. Endoscopic retrograde cholangiopancreatogram (ERCP) showing normal biliary tract anatomy. In addition to the
endoscope and large vertical gallbladder filled with contrast dye, the common hepatic duct (CHD), common bile duct (CBD), and pancreatic duct (PD) are shown. The arrow
points to the ampulla of Vater. C. Endoscopic retrograde cholangiogram (ERC) showing choledocholithiasis. The biliary tract is dilated and contains multiple radiolucent
calculi. D. ERCP showing sclerosing cholangitis. The CBD shows areas that are strictured and narrowed.
steady ache or fullness in the epigastrium or right upper quadrant
(RUQ) of the abdomen with frequent radiation to the interscapular
area, right scapula, or shoulder.
Biliary colic begins quite suddenly and may persist with severe
intensity for 30 min to 5 h, subsiding gradually or rapidly. It is steady
rather than intermittent, as would be suggested by the word colic,
which must be regarded as a misnomer, although it is in widespread
use. An episode of biliary pain persisting beyond 5 h should raise the
suspicion of acute cholecystitis (see below). Nausea and vomiting fre­
quently accompany episodes of biliary pain. An elevated level of serum
bilirubin and/or alkaline phosphatase suggests a common duct stone.
Fever or chills (rigors) with biliary pain usually imply a complication,
that is, cholecystitis, pancreatitis, or cholangitis. Complaints of shortlasting, vague epigastric fullness, dyspepsia, eructation, or flatulence,
especially following a fatty meal, should not be confused with biliary
pain. Such symptoms are frequently elicited from patients with or
without gallstone disease but are not specific for biliary calculi. Bili­
ary colic may be precipitated by eating a fatty meal, by consumption
of a large meal following a period of prolonged fasting, or by eating a
normal meal; it is frequently nocturnal, occurring within a few hours
of retiring.
Natural History 
Gallstone disease discovered in an asymptomatic
patient or in a patient whose symptoms are not referable to choleli­
thiasis is a common clinical problem. Sixty to 80% of persons with
asymptomatic gallstones remain asymptomatic over follow-up periods
of up to 25 years. The probability of developing symptoms within
5 years after diagnosis is 2–4% per year and decreases in the years
thereafter to 1–2%. The yearly incidence of complications is about
0.1–0.3%. Patients remaining asymptomatic for 15 years were found to
be unlikely to develop symptoms during further follow-up, and most
patients who did develop complications from their gallstones experi­
enced prior warning symptoms. Similar conclusions apply to diabetic
patients with silent gallstones. Decision analysis has suggested that (1)
the cumulative risk of death due to gallstone disease while on expect­
ant management is small, and (2) prophylactic cholecystectomy is not
warranted.
Complications requiring cholecystectomy are much more common
in gallstone patients who have developed symptoms of biliary pain.
Patients found to have gallstones at a young age are more likely to
develop symptoms from cholelithiasis than are patients >60 years at the
time of initial diagnosis. Patients with diabetes mellitus and gallstones
may be somewhat more susceptible to septic complications, but the
magnitude of risk of septic biliary complications in diabetic patients is
incompletely defined.
CHAPTER 357
TREATMENT
Gallstones
SURGICAL THERAPY
In asymptomatic gallstone patients, the risk of developing symp­
toms or complications requiring surgery is quite small (see above).
Thus, a recommendation for cholecystectomy in a patient with
gallstones should probably be based on assessment of three factors:
(1) the presence of symptoms that are frequent enough or severe
enough to interfere with the patient’s general routine; (2) the pres­
ence of a prior complication of gallstone disease, that is, history
of acute cholecystitis, pancreatitis, gallstone fistula, etc.; or (3) the
presence of an underlying condition predisposing the patient to
increased risk of gallstone complications (e.g., a previous attack
of acute cholecystitis regardless of current symptomatic status).
Patients with very large gallstones (>3 cm in diameter) and patients
harboring gallstones in a congenitally anomalous gallbladder might
also be considered for prophylactic cholecystectomy. Although
young age is a worrisome factor in asymptomatic gallstone patients,
few authorities would now recommend routine cholecystectomy in
young patients with silent stones. Laparoscopic cholecystectomy
is a minimal-access approach for the removal of the gallbladder
together with its stones. Its advantages include a markedly short­
ened hospital stay, minimal disability, and decreased cost, and it
is the procedure of choice for most patients referred for elective
cholecystectomy.
Diseases of the Gallbladder and Bile Ducts
From several studies involving >4000 patients undergoing lapa­
roscopic cholecystectomy, the following key points emerge: (1) com­
plications develop in ~4% of patients, (2) conversion to laparotomy
occurs in 5%, (3) the death rate is remarkably low (i.e., <0.1%), and
(4) the rate of bile duct injuries is low (i.e., 0.2–0.6%) and compa­
rable with open cholecystectomy. These data indicate why laparo­
scopic cholecystectomy has become the “gold standard” for treating
symptomatic cholelithiasis.
MEDICAL THERAPY—GALLSTONE DISSOLUTION
In carefully selected patients with a functioning gallbladder and
with radiolucent stones <10 mm in diameter, complete dissolution
can be achieved in ~50% of patients within 6 months to 2 years. For
good results within a reasonable time period, this therapy should
be limited to radiolucent stones <5 mm in diameter. The dose of
UDCA should be 10–15 mg/kg per d. Stones >10 mm in size rarely
dissolve. Pigment stones are not responsive to UDCA therapy. Prob­
ably ≤10% of patients with symptomatic cholelithiasis are candidates
for such treatment. However, in addition to the vexing problem of
recurrent stones (30–50% over 3–5 years of follow-up), there is also
the factor of taking a drug for up to 2 years and perhaps indefinitely.
The advantages and success of laparoscopic cholecystectomy have
largely reduced the role of gallstone dissolution to patients who
wish to avoid or are not candidates for elective cholecystectomy.
However, patients with cholesterol gallstone disease who develop
recurrent choledocholithiasis after cholecystectomy should be on
long-term treatment with UDCA.
■
■ACUTE AND CHRONIC CHOLECYSTITIS
Acute Cholecystitis 
Acute inflammation of the gallbladder wall
usually follows obstruction of the cystic duct by a stone. Inflammatory
response can be evoked by three factors: (1) mechanical inflamma­
tion produced by increased intraluminal pressure and distention with
resulting ischemia of the gallbladder mucosa and wall, (2) chemical
inflammation caused by the release of lysolecithin (due to the action
of phospholipase on lecithin in bile) and other local tissue factors, and
(3) bacterial inflammation, which may play a role in 50–85% of patients
with acute cholecystitis. The organisms most frequently isolated by
culture of gallbladder bile in these patients include Escherichia coli,
Klebsiella spp., Streptococcus spp., and Clostridium spp.
PART 10
Disorders of the Gastrointestinal System
Acute cholecystitis often begins as an attack of biliary pain that pro­
gressively worsens. Approximately 60–70% of patients report having
experienced prior attacks that resolved spontaneously. As the episode
progresses, however, the pain of acute cholecystitis becomes more
generalized in the right upper abdomen. As with biliary colic, the pain
of cholecystitis may radiate to the interscapular area, right scapula, or
shoulder. Peritoneal signs of inflammation such as increased pain with
jarring or on deep respiration may be apparent. The patient is anorectic
and often nauseated. Vomiting is relatively common and may produce
symptoms and signs of vascular and extracellular volume depletion.
Jaundice is unusual early in the course of acute cholecystitis but may
occur when edematous inflammatory changes involve the bile ducts
and surrounding lymph nodes.
A low-grade fever is characteristically present, but shaking chills or
rigors are not uncommon. The RUQ of the abdomen is almost invari­
ably tender to palpation. An enlarged, tense gallbladder is palpable in
25–50% of patients. Deep inspiration or cough during subcostal palpa­
tion of the RUQ usually produces increased pain and inspiratory arrest
(Murphy’s sign). Localized rebound tenderness in the RUQ is com­
mon, as are abdominal distention and hypoactive bowel sounds from
paralytic ileus, but generalized peritoneal signs and abdominal rigidity
are usually lacking, in the absence of perforation.
The diagnosis of acute cholecystitis is usually made on the basis of
a characteristic history and physical examination. The triad of sudden
onset of RUQ tenderness, fever, and leukocytosis is highly suggestive.
Typically, leukocytosis in the range of 10,000–15,000 cells per microli­
ter with a left shift on differential count is found. The serum bilirubin
is mildly elevated (<85.5 μmol/L [5 mg/dL]) in fewer than half of
patients, whereas about one-fourth have modest elevations in serum
aminotransferases (usually less than a fivefold elevation). Ultrasound
will demonstrate calculi in 90–95% of cases and is useful for detection
of signs of gallbladder inflammation including thickening of the wall,
pericholecystic fluid, and dilatation of the bile duct. The radionuclide
(e.g., HIDA) biliary scan may be confirmatory if bile duct imaging is
seen without visualization of the gallbladder.
Approximately 75% of patients treated medically have remission
of acute symptoms within 2–7 days following hospitalization. In
25%, however, a complication of acute cholecystitis will occur despite
conservative treatment (see below). In this setting, prompt surgical
intervention is required. Of the 75% of patients with acute cholecystitis
who undergo remission of symptoms, ~25% will experience a recur­
rence of cholecystitis within 1 year, and 60% will have at least one
recurrent bout within 6 years. In view of the natural history of the
disease, acute cholecystitis is best treated by early surgery whenever
possible. Mirizzi’s syndrome is a rare complication in which a gallstone
becomes impacted in the cystic duct or neck of the gallbladder causing
compression of the CBD, resulting in CBD obstruction and jaundice.
Ultrasound shows gallstone(s) lying outside the hepatic duct. Endo­
scopic retrograde cholangiopancreatography (ERCP) (Fig. 357-2B),
percutaneous transhepatic cholangiography (PTC), or magnetic reso­
nance cholangiopancreatography (MRCP) will usually demonstrate
the characteristic extrinsic compression of the CBD. Surgery consists
of removing the cystic duct, diseased gallbladder, and impacted stone.
The preoperative diagnosis of Mirizzi’s syndrome is important to avoid
CBD injury.
ACALCULOUS CHOLECYSTITIS  In 5–14% of patients with acute chole­
cystitis, calculi obstructing the cystic duct are not found at surgery. In
50% of such cases, an underlying explanation for acalculous inflam­
mation is not found. An increased risk for the development of acalcu­
lous cholecystitis is especially associated with prolonged fasting, with
serious trauma or burns, in the postpartum period following prolonged
labor, and with orthopedic and other nonbiliary major surgical opera­
tions in the postoperative period. It may possibly complicate periods
of prolonged parenteral hyperalimentation. For some of these cases,
biliary sludge in the cystic duct may be responsible. Other precipitating
factors include vasculitis, obstructing adenocarcinoma of the gallblad­
der, diabetes mellitus, torsion of the gallbladder, “unusual” bacterial
infections of the gallbladder (e.g., Leptospira, Streptococcus, Salmonella,
or Vibrio cholerae), and parasitic infestation of the gallbladder. Acalcu­
lous cholecystitis may also be seen with a variety of other systemic dis­
ease processes (e.g., sarcoidosis, cardiovascular disease, tuberculosis,
syphilis, actinomycosis).
Although the clinical manifestations of acalculous cholecystitis are
indistinguishable from those of calculous cholecystitis, the setting of
acute gallbladder inflammation complicating severe underlying ill­
ness is characteristic of acalculous disease. Ultrasound or computed
tomography (CT) examinations demonstrating a large, tense, static
gallbladder without stones and with evidence of poor emptying over
a prolonged period may be diagnostically useful in some cases. The
complication rate for acalculous cholecystitis exceeds that for calculous
cholecystitis. Successful management of acute acalculous cholecystitis
appears to depend primarily on early diagnosis and surgical interven­
tion, with meticulous attention to postoperative care.
ACALCULOUS CHOLECYSTOPATHY  Disordered motility of the gall­
bladder can produce recurrent biliary pain in patients without gall­
stones. Infusion of an octapeptide of CCK can be used to measure the
gallbladder ejection fraction during cholescintigraphy. The surgical
findings have included abnormalities such as chronic cholecystitis,
gallbladder muscle hypertrophy, and/or a markedly narrowed cystic
duct. Some of these patients may well have had antecedent gallbladder
disease. The following criteria can be used to identify patients with
acalculous cholecystopathy: (1) recurrent episodes of typical RUQ
pain characteristic of biliary tract pain, (2) abnormal CCK cholescin­
tigraphy demonstrating a gallbladder ejection fraction of <40%, and
(3) infusion of CCK reproducing the patient’s pain. An additional clue
would be the identification of a large gallbladder on ultrasound exami­
nation. Importantly, it should be noted that SOD dysfunction can also
give rise to recurrent RUQ pain and CCK-scintigraphic abnormalities.
EMPHYSEMATOUS CHOLECYSTITIS  So-called emphysematous cho­
lecystitis is thought to begin with acute cholecystitis (calculous or
acalculous) followed by ischemia or gangrene of the gallbladder wall
and infection by gas-producing organisms. Bacteria most frequently
cultured in this setting include anaerobes, such as Clostridium welchii
or C. perfringens, and aerobes, such as E. coli. This condition occurs
most frequently in elderly men and in patients with diabetes mellitus.
The clinical manifestations are essentially indistinguishable from those
of nongaseous cholecystitis. The diagnosis is usually made on plain
abdominal film by finding gas within the gallbladder lumen, dissecting
within the gallbladder wall to form a gaseous ring, or in the perichole­
cystic tissues. The morbidity and mortality rates with emphysematous
cholecystitis are considerable. Prompt surgical intervention coupled
with appropriate antibiotics is mandatory.
Chronic Cholecystitis 
Chronic inflammation of the gallbladder
wall is almost always associated with the presence of gallstones and is
thought to result from repeated bouts of subacute or acute cholecys­
titis or from persistent mechanical irritation of the gallbladder wall
by gallstones. The presence of bacteria in the bile occurs in >25% of
patients with chronic cholecystitis. The presence of infected bile in a
patient with chronic cholecystitis undergoing elective cholecystectomy
probably adds little to the operative risk. Chronic cholecystitis may be
asymptomatic for years, which may progress to symptomatic gallblad­
der disease or to acute cholecystitis or may present with complications
(see below).
Complications of Cholecystitis 
• 
EMPYEMA AND HYDROPS 
Empyema of the gallbladder usually results from progression of acute
cholecystitis with persistent cystic duct obstruction to superinfection
of the stagnant bile with a pus-forming bacterial organism. The clini­
cal picture resembles that of cholangitis with high fever; severe RUQ
pain; marked leukocytosis; and often, prostration. Empyema of the
gallbladder carries a high risk of gram-negative sepsis and/or perfora­
tion. Emergency surgical intervention with proper antibiotic coverage
is required as soon as the diagnosis is suspected.
Hydrops or mucocele of the gallbladder may also result from pro­
longed obstruction of the cystic duct, usually by a large solitary calcu­
lus. In this instance, the obstructed gallbladder lumen is progressively
distended, over a period of time, by mucus (mucocele) or by a clear
transudate (hydrops) produced by mucosal epithelial cells. A visible,
easily palpable, nontender mass sometimes extending from the RUQ
into the right iliac fossa may be found on physical examination. The
patient with hydrops of the gallbladder frequently remains asymptom­
atic, although chronic RUQ pain may also occur. Cholecystectomy is
indicated, because empyema, perforation, or gangrene may complicate
the condition.
GANGRENE AND PERFORATION  Gangrene of the gallbladder results
from ischemia of the wall and patchy or complete tissue necrosis.
Underlying conditions often include marked distention of the gall­
bladder, vasculitis, diabetes mellitus, empyema, or torsion resulting
in arterial occlusion. Gangrene usually predisposes to perforation of
the gallbladder, but perforation may also occur in chronic cholecys­
titis without premonitory warning symptoms. Localized perforations
are usually contained by the omentum or by adhesions produced by
recurrent inflammation of the gallbladder. Bacterial superinfection of
the walled-off gallbladder contents results in abscess formation. Most
patients are best treated with cholecystectomy, but some seriously ill
patients may be managed with cholecystostomy and drainage of the
abscess. Free perforation is less common but is associated with a mor­
tality rate of ~30%. Such patients may experience a sudden transient
relief of RUQ pain as the distended gallbladder decompresses; this is
followed by signs of generalized peritonitis.
FISTULA FORMATION AND GALLSTONE ILEUS  Fistula formation into
an adjacent organ adherent to the gallbladder wall may result from
inflammation and adhesion formation. Fistulas into the duodenum are
most common, followed in frequency by those involving the hepatic
flexure of the colon, stomach or jejunum, abdominal wall, and renal
pelvis. Clinically “silent” biliary-enteric fistulas occurring as a compli­
cation of acute cholecystitis have been found in up to 5% of patients
undergoing cholecystectomy. Asymptomatic cholecystoenteric fistulas
may sometimes be diagnosed by finding gas in the biliary tree on plain
abdominal films. Barium contrast studies or endoscopy of the upper
gastrointestinal tract or colon may demonstrate the fistula. Treatment
in the symptomatic patient usually consists of cholecystectomy, CBD
exploration, and closure of the fistulous tract.
Gallstone ileus refers to mechanical intestinal obstruction result­
ing from the passage of a large gallstone into the bowel lumen. The
stone customarily enters the duodenum through a cholecystoenteric
fistula at that level. The site of obstruction by the impacted gallstone
is usually at the ileocecal valve, provided that the more proximal
small bowel is of normal caliber. Most patients do not give a his­
tory of either prior biliary tract symptoms or complaints suggestive
of acute cholecystitis or fistula formation. Large stones, >2.5 cm in
diameter, are thought to predispose to fistula formation by gradual
erosion through the gallbladder fundus. Diagnostic confirmation
may occasionally be found on the plain abdominal film (e.g., smallintestinal obstruction with gas in the biliary tree [pneumobilia] and
a calcified, ectopic gallstone) or following an upper gastrointestinal
series (cholecystoduodenal fistula with small-bowel obstruction at
the ileocecal valve). Laparotomy with stone extraction (or propulsion
into the colon) remains the procedure of choice to relieve obstruc­
tion. Evacuation of large stones within the gallbladder should also
be performed. In general, the gallbladder and its attachment to the
intestines should be left alone.
LIMEY (MILK OF CALCIUM) BILE AND PORCELAIN GALLBLADDER 
Calcium salts in the lumen of the gallbladder in sufficient concentra­
tion may produce calcium precipitation and diffuse, hazy opacifica­
tion of bile or a layering effect on plain abdominal roentgenography.
This so-called limey bile, or milk of calcium bile, is usually clinically
innocuous, but cholecystectomy is often performed, especially when it
occurs in a hydropic gallbladder. In the entity called porcelain gallblad­
der, calcium salt deposition within the wall of a chronically inflamed
gallbladder may be detected on the plain abdominal film. In the past,
cholecystectomy was advised in all patients with porcelain gallbladder
because there was felt to be a high incidence of carcinoma of the gall­
bladder associated with this condition, an association challenged by a
number of studies. Two patterns of gallbladder wall calcification have
now been appreciated: complete intramural calcification and selective
mucosal calcification. The incidence of cancer in those with selective
intramural calcification is higher than those with complete mucosal
wall calcification, but the risk is very small. As such, the need for chole­
cystectomy for porcelain gallbladder is not absolute; close surveillance
in these patients is also acceptable.
CHAPTER 357
Diseases of the Gallbladder and Bile Ducts
TREATMENT
Acute Cholecystitis
MEDICAL THERAPY
Although surgical intervention remains the mainstay of therapy
for acute cholecystitis and its complications, a period of in-hospital
stabilization may be required before cholecystectomy. Oral intake
is eliminated, nasogastric suction may be indicated, extracellular
volume depletion and electrolyte abnormalities are repaired, and
analgesia is provided. Intravenous antibiotic therapy is indicated
in patients with severe acute cholecystitis, even though bacterial
superinfection of bile may not have occurred in the early stages of
the inflammatory process. Antibiotic therapy is guided by the most
common organisms likely to be present including E. coli, Klebsiella,
Enterococcus, Enterobacter, and Streptococcus. Effective antibiot­
ics include piperacillin plus tazobactam, imipenem, meropenem,
ceftriaxone plus metronidazole, and levofloxacin plus metronida­
zole (Chap. 166). Postoperative complications of wound infection,
abscess formation, and sepsis are reduced in antibiotic-treated
patients.
SURGICAL THERAPY
The optimal timing of surgical intervention in patients with acute
cholecystitis depends on stabilization of the patient. The clear
trend is toward earlier surgery, and this is due in part to require­
ments for shorter hospital stays. Urgent (emergency) cholecystec­
tomy or percutaneous cholecystostomy is probably appropriate in
most patients in whom a complication of acute cholecystitis such
as empyema, emphysematous cholecystitis, or perforation is sus­
pected or confirmed. Patients with uncomplicated acute cholecys­
titis should undergo early elective laparoscopic cholecystectomy,
ideally within 48–72 h after diagnosis. The complication rate is
not increased in patients undergoing early as opposed to delayed
(>6 weeks after diagnosis) cholecystectomy. Delayed surgical
intervention is probably best reserved for (1) patients in whom
the overall medical condition imposes an unacceptable risk for
early surgery and (2) patients in whom the diagnosis of acute cho­
lecystitis is in doubt. Thus, early cholecystectomy (within 72 h) is
the treatment of choice for most patients with acute cholecystitis.
Mortality figures for emergency cholecystectomy in most centers
range from 1 to 3%, whereas the mortality risk for early elective
cholecystectomy is ~0.5% in patients under age 60. Of course, the
operative risks increase with age-related diseases of other organ
systems and with the presence of long- or short-term complica­
tions of gallbladder disease. Seriously ill or debilitated patients
with cholecystitis may be managed with percutaneous drainage (a
cholecystostomy tube), transpapillary drainage (an endoscopically
placed transpapillary drainage catheter via the cystic duct), or
transmural drainage (an endoscopically placed covered, lumenapposing stent). Elective cholecystectomy may then be done at a
later date.
Postcholecystectomy Complications 
Early complications fol­
lowing cholecystectomy include atelectasis and other pulmonary
disorders, abscess formation (often subphrenic), external or internal
hemorrhage, biliary-enteric fistula, and bile leaks. Jaundice may indi­
cate absorption of bile from an intraabdominal collection following a
biliary leak or mechanical obstruction of the CBD by retained calculi,
intraductal blood clots, or extrinsic compression.
Overall, cholecystectomy is a very successful operation that pro­
vides total or near-total relief of preoperative symptoms in 75–90% of
patients. The most common cause of persistent postcholecystectomy
symptoms is an overlooked symptomatic nonbiliary disorder (e.g.,
reflux esophagitis, peptic ulceration, pancreatitis, or—most often—
irritable bowel syndrome). In a small percentage of patients, however,
a disorder of the extrahepatic bile ducts may result in persistent symp­
tomatology. These so-called postcholecystectomy syndromes may be
due to (1) biliary strictures, (2) retained biliary calculi, (3) cystic duct
stump syndrome, (4) stenosis or dyskinesia of the SOD, or (5) bile
salt–induced diarrhea or gastritis.
PART 10
Disorders of the Gastrointestinal System
CYSTIC DUCT STUMP SYNDROME  In the absence of cholangiographi­
cally demonstrable retained stones, symptoms resembling biliary pain
or cholecystitis in the postcholecystectomy patient have frequently
been attributed to disease in a long (>1 cm) cystic duct remnant (cystic
duct stump syndrome). Careful analysis, however, reveals that post­
cholecystectomy complaints are attributable to other causes in almost
all patients in whom the symptom complex was originally thought to
result from the existence of a long cystic duct stump. Accordingly, con­
siderable care should be taken to investigate the possible role of other
factors in the production of postcholecystectomy symptoms before
attributing them to cystic duct stump syndrome.
SOD STENOSIS, SOD DYSKINESIA, AND BILIARY DYSKINESIA  Symp­
toms of biliary colic accompanied by signs of recurrent, intermittent
biliary obstruction may be produced by acalculous cholecystopathy,
SOD stenosis, or SOD dyskinesia. SOD stenosis is thought to result
from acute or chronic inflammation of the papilla of Vater or from
glandular hyperplasia of the papillary segment. Five criteria have been
used to define SOD stenosis: (1) upper abdominal pain, usually RUQ
or epigastric; (2) abnormal liver tests; (3) dilatation of the CBD upon
MRCP or ERCP examination; (4) delayed (>45 min) drainage of con­
trast material from the duct; and (5) increased basal pressure of the
SOD. Treatment consists of endoscopic or surgical sphincteroplasty to
ensure wide patency of the distal portions of both the bile and pancre­
atic ducts. The greater the number of the preceding criteria present, the
greater is the likelihood that a patient does have a degree of SOD suf­
ficient to justify correction. The factors usually considered as indica­
tions for sphincterotomy include (1) prolonged duration of symptoms,
(2) lack of response to symptomatic treatment, (3) presence of severe
disability, and (4) the patient’s choice of sphincterotomy over surgery
(given a clear understanding on the patient’s part of the risks involved
in both procedures).
Biliary SOD disorders are characterized by three criteria: (1) biliary
pain, (2) absence of bile duct stones or other abnormalities, and (3)
elevated liver enzymes or a dilated CBD, but not both. In this setting,
either hepatobiliary scintigraphy or SOD manometry can support the
diagnosis. Importantly, the presence of both elevated liver enzymes and
a dilated CBD should raise the question of obstruction. Proposed mech­
anisms to account for SOD dysfunction include spasm of the sphincter,
denervation sensitivity resulting in hypertonicity, and abnormalities
in the sequencing or frequency rates of the sphincteric-contraction
waves. When thorough evaluation has failed to demonstrate another
cause for the pain and when cholangiographic and manometric criteria
suggest a diagnosis of SOD dyskinesia, medical treatment with nitrites
or anticholinergics to attempt pharmacologic relaxation of SOD has
been proposed but not evaluated in detailed studies. Endoscopic biliary
sphincterotomy (EBS) or surgical sphincterotomy may be indicated in
patients who fail to respond to a 2- to 3-month trial of medical therapy,
especially if SOD pressures are elevated. Approximately 45% of such
patients have long-term pain relief after EBS. EBS has become the
procedure of choice for removing bile duct stones and for other biliary
and pancreatic problems.
BILE SALT–INDUCED DIARRHEA AND GASTRITIS  Postcholecystectomy
patients may develop symptoms of dyspepsia, which have been attrib­
uted to duodenogastric reflux of bile. However, firm data linking these
symptoms to bile gastritis after surgical removal of the gallbladder are
lacking. Cholecystectomy induces persistent changes in gut transit, and
these changes effect a noticeable modification of bowel habits. Chole­
cystectomy shortens gut transit time by accelerating passage of the fecal
bolus through the colon with marked acceleration in the right colon,
thus causing an increase in colonic bile acid output and a shift in bile
acid composition toward the more diarrheagenic secondary bile acids,
that is, deoxycholic acid. Diarrhea that is severe enough, that is, three
or more watery movements per day, can be classified as postcholecys­
tectomy diarrhea, and this occurs in 5–10% of patients undergoing
elective cholecystectomy. Treatment with bile acid–sequestering agents
such as cholestyramine or colestipol is often effective in ameliorating
troublesome diarrhea.
■
■THE HYPERPLASTIC CHOLECYSTOSES
The term hyperplastic cholecystoses is used to denote a group of dis­
orders of the gallbladder characterized by excessive proliferation of
normal tissue components.
Adenomyomatosis is characterized by a benign proliferation of
gallbladder surface epithelium with glandlike formations, extramural
sinuses, transverse strictures, and/or fundal nodule (“adenoma” or
“adenomyoma”) formation.
Cholesterolosis is characterized by abnormal deposition of lipid,
especially cholesteryl esters, within macrophages in the lamina propria
of the gallbladder wall. In its diffuse form (“strawberry gallbladder”),
the gallbladder mucosa is brick red and speckled with bright yellow
flecks of lipid. The localized form shows solitary or multiple “choles­
terol polyps” studding the gallbladder wall. Cholesterol stones of the
gallbladder are found in nearly half the cases. Cholecystectomy is only
indicated in adenomyomatosis or cholesterolosis when biliary symp­
toms are present.
The prevalence of gallbladder polyps in the adult population is
3–6% with a marked male predominance. Types of gallbladder polyps
include cholesterol polyps, adenomyomas, inflammatory polyps, and
adenomas (rare). No significant changes have been found over a 5-year
period in asymptomatic patients with gallbladder polyps <6 mm and
few changes in polyps 7–9 mm. Currently, cholecystectomy is recom­
mended in symptomatic patients as well as in asymptomatic patients
50 years whose polyps are >10 mm or associated with gallstones or
polyp growth on serial ultrasonography. A recent consensus conference
of the Society of Radiologists in Ultrasound recommends that patients
with polyps between 10 and 14 mm and low-risk radiologic features
could be monitored instead of undergoing cholecystectomy.
DISEASES OF THE BILE DUCTS
■
■CONGENITAL ANOMALIES
Biliary Atresia and Hypoplasia 
Atretic and hypoplastic lesions
of the extrahepatic and large intrahepatic bile ducts are the most com­
mon biliary anomalies of clinical relevance encountered in infancy.
The clinical picture is one of severe obstructive jaundice during the
first month of life, with pale stools. When biliary atresia is suspected
on the basis of clinical, laboratory, and imaging findings, the diagnosis
is confirmed by surgical exploration and operative cholangiography.
Approximately 10% of cases of biliary atresia are treatable with Roux-en-Y
choledochojejunostomy, with the Kasai procedure (hepatic portoenter­
ostomy) being attempted in the remainder in an effort to restore some
bile flow. Most patients, even those having successful biliary-enteric
anastomoses, eventually develop chronic cholangitis, extensive hepatic
fibrosis, and portal hypertension.
Choledochal Cysts 
Cystic dilatation may involve the free portion
of the CBD, that is, choledochal cyst, or may present as diverticulum
formation in the intraduodenal segment. In the latter situation, chronic
reflux of pancreatic juice into the biliary tree can produce inflamma­
tion and stenosis of the extrahepatic bile ducts, leading to cholangitis
or biliary obstruction. Because the process may be gradual, ~50% of
patients present with onset of symptoms after age 10. The diagnosis
may be made by ultrasound, abdominal CT, MRCP, or cholangiog­
raphy. Only one-third of patients show the classic triad of abdominal
pain, jaundice, and an abdominal mass. Ultrasonographic detection
of a cyst separate from the gallbladder should suggest the diagnosis
of choledochal cyst, which can be confirmed by demonstrating the
entrance of extrahepatic bile ducts into the cyst. Surgical treatment
involves excision of the “cyst” and biliary-enteric anastomosis. Patients
with choledochal cysts are at increased risk for cholangiocarcinoma.
Congenital Biliary Ectasia 
Dilatation of intrahepatic bile ducts
may involve either the major intrahepatic radicles (Caroli’s disease), the
inter- and intralobular ducts (congenital hepatic fibrosis), or both. In
Caroli’s disease, clinical manifestations include recurrent cholangitis,
abscess formation in and around the affected ducts, and, often, brown
pigment gallstone formation within portions of ectatic intrahepatic
biliary radicles. Ultrasound, MRCP, and CT are of great diagnostic
value in demonstrating cystic dilatation of the intrahepatic bile ducts.
Treatment with ongoing antibiotic therapy is usually undertaken in
an effort to limit the frequency and severity of recurrent bouts of
cholangitis. Progression to secondary biliary cirrhosis with portal
hypertension, extrahepatic biliary obstruction, cholangiocarcinoma, or
recurrent episodes of sepsis with hepatic abscess formation is common.
■
■CHOLEDOCHOLITHIASIS
Pathophysiology and Clinical Manifestations 
Passage of gall­
stones into the CBD occurs in ~10–15% of patients with cholelithiasis.
The incidence of common duct stones increases with increasing age
of the patient, so up to 25% of elderly patients may have calculi in
the common duct at the time of cholecystectomy. Undetected duct
stones are left behind in ~1–5% of cholecystectomy patients. The
overwhelming majority of bile duct stones are cholesterol stones
formed in the gallbladder, which then migrate into the extrahepatic
biliary tree through the cystic duct. Primary calculi arising de novo
in the ducts are usually brown pigment stones developing in patients
with (1) hepatobiliary parasitism or chronic, recurrent cholangitis;
(2) congenital anomalies of the bile ducts (especially Caroli’s disease);
(3) dilated, sclerosed, or strictured ducts; or (4) an MDR3 (ABCB4)
gene defect leading to impaired biliary phospholipids secretion (low
phospholipid–associated cholesterol cholelithiasis). Common duct
stones may remain asymptomatic for years, may pass spontaneously
into the duodenum, or (most often) may present with biliary colic or
a complication.
Complications 
• 
CHOLANGITIS  Cholangitis may be acute or
chronic, and symptoms result from inflammation, which usually is
caused by at least partial obstruction to the flow of bile. Bacteria are
present on bile culture in ~75% of patients with acute cholangitis early
in the symptomatic course. The characteristic presentation of acute
cholangitis involves biliary pain, jaundice, and spiking fevers with
chills (Charcot’s triad). Blood cultures are frequently positive, and
leukocytosis is typical. Nonsuppurative acute cholangitis is most com­
mon and may respond relatively rapidly to supportive measures and to
treatment with antibiotics. In suppurative acute cholangitis, however,
the presence of pus under pressure in a completely obstructed ductal
system leads to symptoms of severe toxicity—mental confusion, bac­
teremia, and septic shock. Response to antibiotics alone in this setting
is relatively poor, multiple hepatic abscesses are often present, and the
mortality rate approaches 100% unless prompt endoscopic or surgical
relief of the obstruction and drainage of infected bile are carried out.
Endoscopic management of bacterial cholangitis is as effective as surgi­
cal intervention. ERCP with endoscopic sphincterotomy is safe and the
preferred initial procedure for both establishing a definitive diagnosis
and providing effective therapy.
OBSTRUCTIVE JAUNDICE  Gradual obstruction of the CBD over a
period of weeks or months usually leads to initial manifestations of
jaundice or pruritus without associated symptoms of biliary colic or
cholangitis. Painless jaundice may occur in patients with choledocho­
lithiasis but is much more characteristic of biliary obstruction second­
ary to malignancy of the head of the pancreas, bile ducts, or ampulla
of Vater.
In patients whose obstruction is secondary to choledocholithiasis,
associated chronic calculous cholecystitis is very common, and the
gallbladder in this setting may be unable to distend. The absence of
a palpable gallbladder in most patients with biliary obstruction from
duct stones is the basis for Courvoisier’s law, that is, that the presence
of a palpably enlarged gallbladder suggests that the biliary obstruction
is secondary to an underlying malignancy rather than to calculous
disease. Biliary obstruction causes progressive dilatation of the intra­
hepatic bile ducts as intrabiliary pressures rise. Hepatic bile flow is
suppressed, and reabsorption and regurgitation of conjugated bilirubin
into the bloodstream lead to jaundice accompanied by dark urine (bili­
rubinuria) and light-colored (acholic) stools.
CHAPTER 357
Diseases of the Gallbladder and Bile Ducts
CBD stones should be suspected in any patient with cholecystitis
whose serum bilirubin level is >85.5 μmol/L (5 mg/dL). The maximum
bilirubin level is seldom >256.5 μmol/L (15.0 mg/dL) in patients with
choledocholithiasis unless concomitant hepatic or renal disease or
another factor leading to marked hyperbilirubinemia exists. Serum
bilirubin levels ≥342.0 μmol/L (20 mg/dL) should suggest the pos­
sibility of neoplastic obstruction. The serum alkaline phosphatase
level is almost always elevated in biliary obstruction. A rise in alkaline
phosphatase often precedes clinical jaundice and may be the only
abnormality in routine liver function tests. There may be a two- to
tenfold elevation of serum aminotransferases, especially in association
with acute obstruction. Following relief of the obstructing process,
serum aminotransferase elevations usually return rapidly to normal,
while the serum bilirubin level may take 1–2 weeks to return to normal.
The alkaline phosphatase level usually falls slowly, lagging behind the
decrease in serum bilirubin.
PANCREATITIS  The most common associated entity discovered in
patients with nonalcoholic acute pancreatitis is biliary tract disease.
Biochemical evidence of pancreatic inflammation complicates acute
cholecystitis in 15% of cases and choledocholithiasis in >30%, and the
common factor appears to be the passage of gallstones through the
common duct. Coexisting pancreatitis should be suspected in patients
with symptoms of cholecystitis who develop (1) back pain or pain to
the left of the abdominal midline, (2) prolonged vomiting with para­
lytic ileus, or (3) a pleural effusion, especially on the left side. Surgical
treatment of gallstone disease is usually associated with resolution of
the pancreatitis.
SECONDARY BILIARY CIRRHOSIS  Secondary biliary cirrhosis may
complicate prolonged or intermittent duct obstruction with or with­
out recurrent cholangitis. Although this complication may be seen
in patients with choledocholithiasis, it is more common in cases of
prolonged obstruction from stricture. Once established, secondary bili­
ary cirrhosis may be progressive even after correction of the obstruct­
ing process, and increasingly severe hepatic cirrhosis may lead to portal
hypertension, hepatic failure and death. Prolonged biliary obstruction
may also be associated with clinically relevant deficiencies of the fatsoluble vitamins A, D, E, and K.
Diagnosis and Treatment 
The diagnosis of choledocholithiasis
is made by cholangiography (Table 357-3), either preoperatively by
endoscopic retrograde cholangiogram (ERC) (Fig. 357-2C) or MRCP
or intraoperatively at the time of cholecystectomy. As many as 15% of
patients undergoing cholecystectomy will prove to have CBD stones.
When CBD stones are suspected prior to laparoscopic cholecystec­
tomy, preoperative ERCP with endoscopic papillotomy and stone
extraction is the preferred approach. It not only provides stone clear­
ance but also defines the anatomy of the biliary tree in relationship to
the cystic duct. CBD stones should be suspected in gallstone patients
who have any of the following risk factors: (1) a history of jaundice
TABLE 357-3  Diagnostic Evaluation of the Bile Ducts
DIAGNOSTIC ADVANTAGES
DIAGNOSTIC LIMITATIONS
CONTRAINDICATIONS
COMPLICATIONS
COMMENT
Ultrasound
Rapid
Simultaneous scanning of GB, liver, bile
ducts, pancreas
Accurate identification of dilated bile ducts
Not limited by jaundice, pregnancy
Guidance for fine-needle biopsy
Bowel gas
Massive obesity
Ascites
Barium
Partial bile duct obstruction
Poor visualization of distal CBD
PART 10
Disorders of the Gastrointestinal System
Computed Tomography
Simultaneous scanning of GB, liver, bile
ducts, pancreas
Accurate identification of dilated bile ducts,
masses
Not limited by jaundice, gas, obesity, ascites
High-resolution image
Guidance for fine-needle biopsy
Extreme cachexia
Movement artifact
Ileus
Partial bile duct obstruction
Magnetic Resonance Cholangiopancreatography
Noninvasive modality for visualizing
pancreatic and biliary ducts
Has excellent sensitivity for bile duct
dilatation, biliary stricture, and intraductal
abnormalities
Can identify pancreatic duct dilatation or
stricture, pancreatic duct stenosis, and
pancreas divisum
Cannot offer therapeutic
intervention
High cost
Endoscopic Retrograde Cholangiopancreatography
Simultaneous pancreatography
Best visualization of distal biliary tract
Bile or pancreatic cytology
Endoscopic sphincterotomy and stone
removal
Biliary manometry
Gastroduodenal obstruction
Roux-en-Y biliary-enteric
anastomosis
Percutaneous Transhepatic Cholangiogram
Best when bile ducts dilated
Best visualization of proximal biliary tract
May be used to obtain bile cytology/culture
Allows for percutaneous transhepatic
drainage
Nondilated or sclerosed ducts
Pregnancy
Uncorrectable
coagulopathy
Massive ascites
Hepatic abscess
Endoscopic Ultrasound
Most sensitive method to detect ampullary
stones and exclude pathology in the head of
the pancreas
Abbreviations: CBD, common bile duct; ERCP, endoscopic retrograde cholangiopancreatography; GB, gallbladder; US, hepatobiliary ultrasound.
or pancreatitis, (2) abnormal tests of liver function, and (3) ultraso­
nographic or MRCP evidence of a dilated CBD or stones in the duct.
Alternatively, if intraoperative cholangiography reveals retained stones,
postoperative ERCP can be carried out. The need for preoperative
ERCP is expected to decrease further as laparoscopic techniques for
bile duct exploration improve.
The widespread use of laparoscopic cholecystectomy and ERCP
has decreased the incidence of complicated biliary tract disease and
the need for choledocholithotomy and T-tube drainage of the bile
ducts. EBS followed by spontaneous passage or stone extraction is the
treatment of choice in the management of patients with common duct
stones, especially in elderly or poor-risk patients.
■
■TRAUMA, STRICTURES, AND HEMOBILIA
Most benign strictures of the extrahepatic bile ducts result from surgi­
cal trauma and occur in about 1 in 500 cholecystectomies. Strictures
may present with bile leak or abscess formation in the immediate post­
operative period or with biliary obstruction or cholangitis as long as
None
None
Initial procedure of choice in
investigating possible biliary tract
obstruction
Pregnancy
Reaction to iodinated
contrast, if used
Indicated for evaluation of hepatic
or pancreatic masses or for
assessing for complications related
to gallstones (pancreatitis)
Procedure of choice in investigating
possible biliary obstruction if
diagnostic limitations limit US
Claustrophobia
Certain metals (iron)
None
First choice to assess for
choledocholithiasis given
comparable sensitivity and
specificity to ERCP
Pregnancy
Acute pancreatitis
Severe
cardiopulmonary
disease
Pancreatitis
Cholangitis, sepsis
Infected pancreatic
pseudocyst
Perforation (rare)
Hypoxemia, aspiration
Cholangiogram of choice if there
is believed to be a need for
intervention:
Diagnosed or high clinical
probability of choledocholithiasis
Biliary stricture requiring sampling
and stenting
Need for sphincterotomy such as
sphincter of Oddi dysfunction
Bleeding
Hemobilia
Bile peritonitis
Bacteremia, sepsis
Indicated for the drainage of
obstructed and infected ducts when
ERCP is contraindicated or failed
Excellent for detecting
choledocholithiasis
2 years or more following the inciting trauma. The diagnosis is estab­
lished by percutaneous or endoscopic cholangiography. Endoscopic
brushing of biliary strictures may be helpful in establishing the nature
of the lesion and is more accurate than bile cytology alone. When posi­
tive exfoliative cytology is obtained, the diagnosis of a neoplastic stric­
ture is established. This procedure is especially important in patients
with primary sclerosing cholangitis (PSC) who are predisposed to the
development of cholangiocarcinomas. Successful operative correction
of non-PSC bile duct strictures by a skillful surgeon with duct-to-bowel
anastomosis is usually possible, although mortality rates from surgical
complications, recurrent cholangitis, or secondary biliary cirrhosis are
high.
Hemobilia may follow traumatic or operative injury to the liver or
bile ducts, intraductal rupture of a hepatic abscess or aneurysm of the
hepatic artery, biliary or hepatic tumor hemorrhage, or mechanical
complications of choledocholithiasis or hepatobiliary parasitism. Diag­
nostic procedures such as liver biopsy, PTC, and transhepatic biliary
drainage catheter placement may also be complicated by hemobilia.
Patients often present with a classic triad of biliary pain, obstructive
jaundice, and melena or occult blood in the stools. The diagnosis is
sometimes made by cholangiographic evidence of blood clot in the
biliary tree, but selective angiographic verification may be required.
Although minor episodes of hemobilia may resolve without interven­
tion, arteriography and transcatheter embolization or surgical ligation
of the bleeding vessel may be required.
■
■EXTRINSIC COMPRESSION OF THE BILE DUCTS
Partial or complete biliary obstruction may be produced by extrinsic
compression of the ducts. The most common cause of this form of
obstructive jaundice is carcinoma of the head of the pancreas. Biliary
obstruction may also occur as a complication of either acute or chronic
pancreatitis or involvement of lymph nodes in the porta hepatis by
lymphoma or metastatic carcinoma. The latter should be distinguished
from cholestasis resulting from massive replacement of the liver by
tumor.
■
■HEPATOBILIARY PARASITISM
Infestation of the biliary tract by adult helminths or their ova may pro­
duce a chronic, recurrent pyogenic cholangitis with or without multiple
hepatic abscesses, ductal stones, or biliary obstruction. This condition
is relatively rare but does occur in inhabitants of southern China and
elsewhere in Southeast Asia. The organisms most commonly involved
are trematodes or flukes, including Clonorchis sinensis, Opisthorchis
viverrini or Opisthorchis felineus, and Fasciola hepatica. The biliary
tract also may be involved by intraductal migration of adult Ascaris
lumbricoides from the duodenum or by intrabiliary rupture of hydatid
cysts of the liver produced by Echinococcus spp. The diagnosis is made
by cholangiography and the presence of characteristic ova on stool
examination. When obstruction is present, the treatment of choice is
laparotomy under antibiotic coverage, with common duct exploration
and a biliary drainage procedure.
■
■SCLEROSING CHOLANGITIS
PSC is characterized by a progressive, inflammatory, sclerosing, and
obliterative process affecting the extrahepatic and/or the intrahepatic
bile ducts. PSC is strongly associated with inflammatory bowel disease,
especially ulcerative colitis.
Immunoglobulin G4 (IgG4)–associated cholangitis is a biliary
disease of unknown etiology that presents with biochemical and chol­
angiographic features indistinguishable from PSC, is often associated
with autoimmune pancreatitis and other fibrosing conditions and is
characterized by elevated serum IgG4 and infiltration of IgG4-positive
plasma cells in bile ducts and liver tissue. All patients diagnosed with
sclerosing cholangitis should have a serum IgG4 level checked to rule
out IgG4 disease as a cause of secondary sclerosing cholangitis, particu­
larly if they do not have inflammatory bowel disease. Glucocorticoids
are the initial treatment of choice. Relapse is common after steroid
withdrawal, especially with proximal strictures. Steroid-sparing agents
such as azathioprine or rituximab may be needed after relapse or for
inadequate response (Chap. 359).
Patients with PSC often present with signs and symptoms of chronic
or intermittent biliary obstruction: RUQ abdominal pain, pruritus, jaun­
dice, or acute cholangitis. Late in the course, complete biliary obstruction,
secondary biliary cirrhosis, hepatic failure, and/or portal hypertension
with bleeding varices may occur. The diagnosis is established by finding
multifocal, diffusely distributed strictures with intervening segments of
normal or dilated ducts, producing a beaded appearance on cholangi­
ography (Fig. 357-2D). The cholangiographic technique of choice in
suspected cases is MRCP. When a diagnosis of sclerosing cholangitis has
been established, causes of secondary sclerosing should be considered
before reaching a diagnosis of PSC. Patients with PSC should undergo
testing for associated diseases, especially inflammatory bowel disease, if
that diagnosis has not already been established.
Small duct PSC is defined by the presence of chronic cholestasis
and hepatic histology consistent with PSC in a patient with IBD, but
with normal findings on cholangiography. Small duct PSC is found in
~5% of patients with PSC and is associated with a significantly better
long-term prognosis. However, such patients may progress to classic
PSC and/or end-stage liver disease with consequent necessity of liver
transplantation.
In patients with AIDS, cholangiopancreatography may demon­
strate a broad range of biliary tract changes as well as pancreatic duct
obstruction and occasionally pancreatitis (Chap. 208). Further, biliary
tract lesions in AIDS include infection and cholangiopancreatographic
changes of sclerosing cholangitis. Changes noted include (1) diffuse
involvement of intrahepatic bile ducts alone, (2) involvement of both
intra- and extrahepatic bile ducts, (3) ampullary stenosis, (4) stricture
of the intrapancreatic portion of the CBD, and (5) pancreatic duct
involvement. Associated infectious organisms include Cryptosporid­
ium, Mycobacterium avium-intracellulare, cytomegalovirus, Microspo­
ridia, and Isospora. ERCP sphincterotomy can provide significant pain
reduction in patients with AIDS-associated papillary stenosis.
CHAPTER 357
TREATMENT
Primary Sclerosing Cholangitis
Diseases of the Gallbladder and Bile Ducts
There is no proven medical therapy for PSC. Therapy to treat pruritus
associated with PSC includes cholestyramine, rifampin, fenofibrate/
bezafibrate, sertraline, and naltrexone. Antibiotics are useful when
bacterial cholangitis complicates the clinical picture. Vitamin D
and calcium supplementation may be used as initial therapy to
help prevent the loss of bone mass frequently seen in patients with
chronic cholestasis. In cases where high-grade biliary obstruction
(dominant strictures) has occurred, balloon dilatation is preferred
over stenting due to the higher complication rate associated with
stenting including pancreatitis and cholangitis. Only rarely is sur­
gical intervention indicated. PSC is a progressive disease with a
median survival of 12–18 years following the diagnosis, regardless
of therapy. Four variables (age, serum bilirubin level, histologic
stage, and splenomegaly) predict survival in patients with PSC and
serve as the basis for a risk score. PSC is a common indication for
liver transplantation.
■
■FURTHER READING
Baron TH et al: Interventional approaches to gallbladder disease. N Engl
J Med 373:357, 2015.
Hu H et al: Gut microbiota promotes cholesterol gallstone formation
by modulating bile acid composition and biliary cholesterol secre­
tion. Nature Comm 13:252, 2022.
Kamaya A et al: Management of incidentally detected gallbladder
polyps: Society of Radiologists in Ultrasound consensus conference
recommendations. Radiology 305:227, 2022.
Lindor K et al: American College of Gastroenterology (ACG) guide­
lines: Primary sclerosing cholangitis. Hepatology 51:660, 2010.
Ryl JK et al: Clinical features of acute acalculous cholecystitis. J Clin
Gastroenterol 36:166, 2003.
Strasberg S: Clinical practice. Acute calculous cholecystitis. N Engl J
Med 358:2804, 2008.

30 - SECTION 4 Disorders of the Pancreas
SECTION 4 Disorders of the Pancreas
Section 4	 Disorders of the Pancreas
Somashekar G. Krishna,
Darwin L. Conwell, Phil A. Hart
Approach to the Patient
with Pancreatic Disease
■
■GENERAL CONSIDERATIONS
Globally, pancreatic disorders, including acute and chronic pancreati­
tis, pancreatic cysts, and pancreatic cancer, are challenging to manage
and associated with a high burden on health care resources. While the
relationships between these diseases are multifaceted, there is ongoing
scientific progress and a growing understanding in this field. Acute
pancreatitis is one of the most common reasons for hospitalizations in
gastroenterology, and there is increasing evidence of sequelae includ­
ing diabetes, exocrine pancreas insufficiency, and chronic pancreatitis.
In elderly patients, acute pancreatitis may serve as an early symptom
of pancreatic cancer. Chronic pancreatitis, an irreversible disease of the
pancreas, associated with poor quality of life due to abdominal pain
and associated exocrine insufficiency, is also an established risk factor
for pancreatic cancer. Pancreatic cysts, mostly incidental, are increas­
ingly detected on cross-sectional abdominal imaging studies. Although
only a small proportion of pancreatic cysts can progress to pancreatic
cancer, the diagnostic uncertainty can introduce unwanted anxiety to
patients and treating physicians. Meanwhile, with persistently high
mortality rates, the incidence of pancreatic adenocarcinoma is increas­
ing and is the seventh leading cause of cancer-related death in the
industrialized world and the third most common in the United States.
PART 10
Disorders of the Gastrointestinal System
As emphasized in Chap. 359, the etiologies and clinical manifesta­
tions of pancreatitis are quite varied. Although it is well-appreciated
that acute pancreatitis is frequently secondary to biliary tract disease or
alcohol abuse, it can also be caused by medications, genetic mutations,
and trauma. In ~30% of patients with acute pancreatitis and 25–40% of
patients with chronic pancreatitis, the etiology is initially unexplained.
The global pooled incidence of acute pancreatitis is ~33.7 cases
(95% confidence interval [CI], 23.3–48.8) with 1.16 deaths (95% CI,
0.85–1.6) per 100,000 person-years. The global pooled incidence of
chronic pancreatitis is ~9.6 cases (95% CI, 7.9–11.8) with 0.09 attrib­
utable deaths (95% CI, 0.02–0.5) per 100,000 person-years. In the
A
B
FIGURE 358-1  A. Side-branch intraductal papillary mucinous neoplasm (magnetic resonance imaging [MRI] with magnetic resonance cholangiopancreatography [MRCP]).
T2-weighted MRCP image demonstrates a dominant, lobulated, hyperintense cystic structure (arrow) within the posterior body of the pancreas. The pancreatic duct
upstream from the cyst is dilated and irregular. Endoscopic ultrasound and fine-needle aspiration of cyst fluid were consistent with a mucinous cyst. Surgical histopathology
revealed an infiltrating moderately differentiated adenocarcinoma, 0.3 cm, arising in a background of an intraductal papillary mucinous neoplasm (IPMN). B. Mucinous
cystic neoplasm (computed tomography [CT] scan). In the tail of the pancreas, there is a well-circumscribed hypodense cyst (arrow) without any nodular enhancing
components. Endoscopic ultrasound and fine-needle aspiration of cyst fluid were suggestive of a mucinous cyst. Surgical histopathology revealed a mucinous cystic
neoplasm (3.4 cm) with low-grade dysplasia. The stroma of the cyst demonstrated diffuse positivity for progesterone receptor and focal positivity for CD10 (ovarian stroma),
confirming the diagnosis. C. Serous cystadenoma (MRI). A lobulated microcystic cyst (arrow) is observed in the tail of the pancreas. Neither a communication with the main
pancreatic duct nor intracystic soft tissue enhancing nodular components were observed. However, the cyst continued to increase in size, and a distal pancreatectomy
was performed. Histopathology revealed a serous microcystic adenoma. (Courtesy of Dr. Z.K. Shah, The Ohio State University Wexner Medical Center; with permission.)
United States, the number of patients admitted to the hospital with
acute pancreatitis is increasing, with estimated rates of almost 300,000
annually, whereas the number of patients hospitalized for chronic pan­
creatitis is decreasing, with recent estimates of ~13,000 admissions per
year. Chronic pancreatitis has an annual prevalence of 42–73 cases per
100,000 adults in the United States, although higher prevalence rates
(0.04–5%) have been noted among adults at autopsy. Together, acute
and chronic pancreatic disease costs an estimated $3 billion annually
in health care expenditures. During the COVID-19 pandemic, it was
noted that the infection was associated with elevated pancreas enzyme
serum levels and presumed acute pancreatitis, though causal relation­
ships have not been definitively established.
The diagnosis of acute pancreatitis is generally defined based on a
combination of laboratory, imaging, and clinical symptoms. The diag­
nosis of chronic pancreatitis, especially in mild disease, is hampered by
the relative inaccessibility of the pancreas to direct examination and the
nonspecificity of the associated abdominal pain. Many patients with
chronic pancreatitis do not have elevated blood amylase or lipase levels.
Some patients with chronic pancreatitis develop signs and symptoms
of exocrine pancreatic insufficiency (EPI), and thus, objective evidence
for pancreatic disease can be demonstrated. However, there is a large
reservoir of pancreatic exocrine function. Maldigestion of fat and
protein becomes evident only when more than 90% of the pancreas
is functionally damaged or obstructed. Noninvasive, indirect tests of
pancreatic exocrine function (e.g., fecal elastase) are much more likely
to give abnormal results in patients with obvious advanced pancreatic
disease (i.e., pancreatic calcification, steatorrhea, or diabetes mellitus)
than in patients with occult disease. Invasive, direct tests of pancreatic
secretory function (e.g., secretin stimulation test) are the most sensitive
and specific tests to detect early chronic pancreatic disease when imag­
ing is equivocal or normal.
The increasing utilization of cross-sectional imaging modalities
with their improved resolution has contributed to a high prevalence
(2–5% with computed tomography [CT] scans, 20–30% with magnetic
resonance imaging [MRI]) of incidentally detected pancreatic cysts.
The most common cyst type encountered is an intraductal papillary
mucinous neoplasm (IPMN), which is classified as a precancerous
mucinous cyst. In the absence of high-risk features, radiographic
surveillance is typically recommended (Fig. 358-1). Mucinous cystic
neoplasms (MCNs) are relatively less common and occur almost exclu­
sively in women. Among the neoplastic cysts, serous cystadenomas
have a negligible risk of progression to malignancy. Other infrequent
neoplastic cysts include neuroendocrine tumors and solid pseudo­
papillary neoplasms. The most commonly encountered benign cyst
is a pseudocyst, which can occur in patients with a history of acute
C

31 - 358 Approach to the Patient with Pancreatic Disease
358 Approach to the Patient with Pancreatic Disease
Section 4	 Disorders of the Pancreas
Somashekar G. Krishna,
Darwin L. Conwell, Phil A. Hart
Approach to the Patient
with Pancreatic Disease
■
■GENERAL CONSIDERATIONS
Globally, pancreatic disorders, including acute and chronic pancreati­
tis, pancreatic cysts, and pancreatic cancer, are challenging to manage
and associated with a high burden on health care resources. While the
relationships between these diseases are multifaceted, there is ongoing
scientific progress and a growing understanding in this field. Acute
pancreatitis is one of the most common reasons for hospitalizations in
gastroenterology, and there is increasing evidence of sequelae includ­
ing diabetes, exocrine pancreas insufficiency, and chronic pancreatitis.
In elderly patients, acute pancreatitis may serve as an early symptom
of pancreatic cancer. Chronic pancreatitis, an irreversible disease of the
pancreas, associated with poor quality of life due to abdominal pain
and associated exocrine insufficiency, is also an established risk factor
for pancreatic cancer. Pancreatic cysts, mostly incidental, are increas­
ingly detected on cross-sectional abdominal imaging studies. Although
only a small proportion of pancreatic cysts can progress to pancreatic
cancer, the diagnostic uncertainty can introduce unwanted anxiety to
patients and treating physicians. Meanwhile, with persistently high
mortality rates, the incidence of pancreatic adenocarcinoma is increas­
ing and is the seventh leading cause of cancer-related death in the
industrialized world and the third most common in the United States.
PART 10
Disorders of the Gastrointestinal System
As emphasized in Chap. 359, the etiologies and clinical manifesta­
tions of pancreatitis are quite varied. Although it is well-appreciated
that acute pancreatitis is frequently secondary to biliary tract disease or
alcohol abuse, it can also be caused by medications, genetic mutations,
and trauma. In ~30% of patients with acute pancreatitis and 25–40% of
patients with chronic pancreatitis, the etiology is initially unexplained.
The global pooled incidence of acute pancreatitis is ~33.7 cases
(95% confidence interval [CI], 23.3–48.8) with 1.16 deaths (95% CI,
0.85–1.6) per 100,000 person-years. The global pooled incidence of
chronic pancreatitis is ~9.6 cases (95% CI, 7.9–11.8) with 0.09 attrib­
utable deaths (95% CI, 0.02–0.5) per 100,000 person-years. In the
A
B
FIGURE 358-1  A. Side-branch intraductal papillary mucinous neoplasm (magnetic resonance imaging [MRI] with magnetic resonance cholangiopancreatography [MRCP]).
T2-weighted MRCP image demonstrates a dominant, lobulated, hyperintense cystic structure (arrow) within the posterior body of the pancreas. The pancreatic duct
upstream from the cyst is dilated and irregular. Endoscopic ultrasound and fine-needle aspiration of cyst fluid were consistent with a mucinous cyst. Surgical histopathology
revealed an infiltrating moderately differentiated adenocarcinoma, 0.3 cm, arising in a background of an intraductal papillary mucinous neoplasm (IPMN). B. Mucinous
cystic neoplasm (computed tomography [CT] scan). In the tail of the pancreas, there is a well-circumscribed hypodense cyst (arrow) without any nodular enhancing
components. Endoscopic ultrasound and fine-needle aspiration of cyst fluid were suggestive of a mucinous cyst. Surgical histopathology revealed a mucinous cystic
neoplasm (3.4 cm) with low-grade dysplasia. The stroma of the cyst demonstrated diffuse positivity for progesterone receptor and focal positivity for CD10 (ovarian stroma),
confirming the diagnosis. C. Serous cystadenoma (MRI). A lobulated microcystic cyst (arrow) is observed in the tail of the pancreas. Neither a communication with the main
pancreatic duct nor intracystic soft tissue enhancing nodular components were observed. However, the cyst continued to increase in size, and a distal pancreatectomy
was performed. Histopathology revealed a serous microcystic adenoma. (Courtesy of Dr. Z.K. Shah, The Ohio State University Wexner Medical Center; with permission.)
United States, the number of patients admitted to the hospital with
acute pancreatitis is increasing, with estimated rates of almost 300,000
annually, whereas the number of patients hospitalized for chronic pan­
creatitis is decreasing, with recent estimates of ~13,000 admissions per
year. Chronic pancreatitis has an annual prevalence of 42–73 cases per
100,000 adults in the United States, although higher prevalence rates
(0.04–5%) have been noted among adults at autopsy. Together, acute
and chronic pancreatic disease costs an estimated $3 billion annually
in health care expenditures. During the COVID-19 pandemic, it was
noted that the infection was associated with elevated pancreas enzyme
serum levels and presumed acute pancreatitis, though causal relation­
ships have not been definitively established.
The diagnosis of acute pancreatitis is generally defined based on a
combination of laboratory, imaging, and clinical symptoms. The diag­
nosis of chronic pancreatitis, especially in mild disease, is hampered by
the relative inaccessibility of the pancreas to direct examination and the
nonspecificity of the associated abdominal pain. Many patients with
chronic pancreatitis do not have elevated blood amylase or lipase levels.
Some patients with chronic pancreatitis develop signs and symptoms
of exocrine pancreatic insufficiency (EPI), and thus, objective evidence
for pancreatic disease can be demonstrated. However, there is a large
reservoir of pancreatic exocrine function. Maldigestion of fat and
protein becomes evident only when more than 90% of the pancreas
is functionally damaged or obstructed. Noninvasive, indirect tests of
pancreatic exocrine function (e.g., fecal elastase) are much more likely
to give abnormal results in patients with obvious advanced pancreatic
disease (i.e., pancreatic calcification, steatorrhea, or diabetes mellitus)
than in patients with occult disease. Invasive, direct tests of pancreatic
secretory function (e.g., secretin stimulation test) are the most sensitive
and specific tests to detect early chronic pancreatic disease when imag­
ing is equivocal or normal.
The increasing utilization of cross-sectional imaging modalities
with their improved resolution has contributed to a high prevalence
(2–5% with computed tomography [CT] scans, 20–30% with magnetic
resonance imaging [MRI]) of incidentally detected pancreatic cysts.
The most common cyst type encountered is an intraductal papillary
mucinous neoplasm (IPMN), which is classified as a precancerous
mucinous cyst. In the absence of high-risk features, radiographic
surveillance is typically recommended (Fig. 358-1). Mucinous cystic
neoplasms (MCNs) are relatively less common and occur almost exclu­
sively in women. Among the neoplastic cysts, serous cystadenomas
have a negligible risk of progression to malignancy. Other infrequent
neoplastic cysts include neuroendocrine tumors and solid pseudo­
papillary neoplasms. The most commonly encountered benign cyst
is a pseudocyst, which can occur in patients with a history of acute
C
or chronic pancreatitis. The challenges with accurately predicting the
risk of malignant transformation of precancerous pancreatic cysts has
contributed to the growing number of patients on imaging surveillance
protocols placing a burden on health care systems in industrialized
nations.
TABLE 358-1  Tests Useful in the Diagnosis of Acute and Chronic Pancreatitis and Pancreatic Neoplasms
TEST
PRINCIPLE
COMMENT
Pancreatic Enzymes in Body Fluids
Serum lipase
Pancreatic inflammation leads to increased serum
enzyme levels
Amylase
 
 
Serum
Pancreatic inflammation leads to increased serum
enzyme levels
Urine
Renal clearance of amylase is increased in acute
pancreatitis
Ascitic fluid
Disruption of gland or main pancreatic duct leads to
increased amylase concentration
Pleural fluid
Exudative pleural effusion with pancreatitis
False positives occur with carcinoma of the lung and esophageal perforation
Studies Pertaining to Pancreatic Structure
Radiologic and radionuclide tests
Plain film of the abdomen or
Can demonstrate large calcifications in chronic
pancreatitis
upper gastrointestinal x-rays
Ultrasonography (US)
Can provide limited information on edema,
inflammation, calcification, pseudocysts, and mass
lesions
Computed tomography (CT)
Permits detailed visualization of pancreas and
surrounding structures, pancreatic fluid collection,
pseudocyst; assessment of necrosis or interstitial
disease
scan
Magnetic resonance
Permits noninvasive detailed evaluation of the
pancreatic parenchyma, biliary and pancreatic
ducts, adjacent soft tissues, and vascular
structures.
imaging (MRI) and
cholangiopancreatography
(MRCP)
Endoscopic ultrasonography
High-frequency transducer used with EUS produces
very-high-resolution images permitting focused
evaluation of pancreatic parenchyma and biliary
and pancreatic ducts, and FNA/B provides targeted
tissue acquisition
(EUS) and fine-needle
aspiration/biopsy (FNA/B)
Endoscopic retrograde
Cannulation of pancreatic and/or common bile duct
permits visualization of pancreaticobiliary ductal
system
cholangiopancreatography
(ERCP)
Tests of Exocrine Pancreatic Function
Direct stimulation of the pancreas with analysis of duodenal contents
Secretin test
Secretin leads to increased output of pancreatic
juice and HCO3
–; pancreatic secretory response is
related to the functional mass of pancreatic tissue;
involves duodenal intubation and fluoroscopic
placement of gastroduodenal tube
Endoscopic pancreatic
Secretin-stimulated collection of pancreatic juice
performed during upper endoscopy; replaces need
for tube placement in the duodenum
function test (ePFT)
EUS-ePFT
Combines endosonographic evaluation of the
pancreas and endoscopic collection of pancreatic
juice
Secretin-stimulated MRCP
Combines imaging evaluation of the pancreas and
a semiquantitative estimation of pancreatic juice
output in the duodenum
Measurement of intraluminal digestion products
Stool fat determination
Lack of lipolytic enzymes brings about impaired
fat digestion; quantitative 72-h stool collection
and estimation are more reliable than qualitative
analysis of a random stool sample
Measurement of pancreatic enzymes in feces
Fecal elastase
Pancreatic secretion of proteolytic enzymes; not
degraded in intestine
■
■TESTS USEFUL IN THE DIAGNOSIS OF
PANCREATIC DISEASE
Several tests are of value in the evaluation of pancreatic disease. Examples
of specific tests and their usefulness in the diagnosis of acute and chronic
pancreatitis are summarized in Table 358-1 and Fig. 358-2. At some
Enzyme measurement of choice for the diagnosis of acute pancreatitis;
increased specificity if the level is more than three times the upper limit of
normal (3× ULN)
Simple; increased specificity if the level is >3× ULN; may be falsely normal in
patients with hypertriglyceridemic pancreatitis
Infrequently used
Can help establish source of ascites; false positives occur with intestinal
obstruction and perforated ulcer; can also measure lipase
Infrequently used
Simple, noninvasive; sequential studies quite feasible; useful in diagnosis
of gallstones; pancreas visualization limited by interference from overlying
bowel gas
CHAPTER 358
Useful in the diagnosis of pancreatic calcification, dilated pancreatic ducts,
and pancreatic tumors; may not be able to distinguish between inflammatory
and neoplastic mass lesions; multiphasic CT scans are the preferred
imaging modality for staging pancreatic cancer; IV contrast is needed for
characterization of most features
Approach to the Patient with Pancreatic Disease
Has mostly replaced ERCP for diagnostic assessment of the pancreatic duct;
more sensitive than CT scan for detection of mild pancreatitis, necrosis,
choledocholithiasis, pancreatic ductal abnormalities, and cystic neoplasms;
no exposure to ionizing radiation
Can be used to assess gallstones, choledocholithiasis, chronic pancreatitis,
pancreatic masses, and cystic neoplasms; FNA/B facilitates diagnostic and
therapeutic management of pancreatic diseases
Primarily a therapeutic procedure; invasive with risks for iatrogenic
complications
Sensitive to detect occult disease; poorly defined normal enzyme response;
large secretory reserve capacity of the pancreas; rarely performed
Sensitive to detect occult disease; high negative predictive value for chronic
pancreatitis; requires sedation
Single endoscopic evaluation of pancreatic structure and function
Improved visualization of pancreatic ductal anatomy; functional evaluation is
less accurate than ePFT; noninvasive
Reliable reference standard for defining severity of fat malabsorption;
does not distinguish between pancreatic and nonpancreatic cause of
malabsorption
Diagnostic accuracy is highest when the pretest probability is high and the
value is <100 μg/g; false positives will occur in patients with nonformed stools
• Clinical signs and symptoms suggestive of chronic pancreatic disease: abdominal pain,
nausea, weight loss, steatorrhea, malabsorption, history of alcohol abuse, recurrent
pancreatitis, fatty-food intolerance
• Perform history, physical examination, review of laboratory studies; consider fecal elastase
measurement
Step 1
• Contrast-enhanced CT scan
• CP diagnostic criteria: calcifications in combination with atrophy and/or dilated duct
• Diagnostic criteria met; no further imaging needed
• Inconclusive or nondiagnostic results; continue to step 2
Step 2
• MRI and MRCP, with or without secretin enhancement (sMRCP)
• CP diagnostic criteria: Cambridge class III,a dilated duct, atrophy of gland, filling
defects in duct suggestive of stones
• Diagnostic criteria met; no further imaging needed
• Inconclusive or nondiagnostic results; continue to step 3
Step 3
• EUS with quantification of parenchymal and ductal criteria
• CP diagnostic criteria: ≥5 EUS CP criteria
• Diagnostic criteria met; no further imaging needed
• Inconclusive or nondiagnostic results; continue to step 4
Step 4
• Pancreas function test (with secretin)—endoscopic (ePFT) collection method
preferred; consider combining ePFT with EUS
• CP diagnostic criteria: peak [bicarbonate] <80 mEq/L
• Diagnostic criteria met; no further imaging needed
• Inconclusive or nondiagnostic results require monitoring of signs and symptoms and repeat
testing in 6 months–1 year
FIGURE 358-2  A stepwise diagnostic approach to the patient with suspected chronic pancreatitis (CP). Endoscopic ultrasonography (EUS) and magnetic resonance
imaging (MRI) with secretin-stimulated magnetic resonance cholangiopancreatography (sMRCP/MRCP) are appropriate diagnostic alternatives to endoscopic retrograde
cholangiopancreatography (ERCP). CT, computed tomography; ePFT, endoscopic pancreas function test. aCambridge classification of pancreatic duct findings: class 0: normal—
visualization of complete normal ductal anatomy; class I: equivocal—normal main duct, 1–3 abnormal side branches; class II: mild—normal main duct, >3 abnormal
side branches; class III—dilated and irregular main duct, >3 abnormal side branches, small (<10 mm) cysts; class IV—irregular main duct, intraductal calculi, strictures,
obstruction with dilation, or large (>10 mm) cysts.
PART 10
Disorders of the Gastrointestinal System
institutions, pancreatic function tests are available and performed if the
diagnosis of chronic pancreatitis remains a possibility after noninvasive
tests (i.e., ultrasound, CT scan, MRI with magnetic resonance cholangio­
pancreatography [MRCP]) or invasive tests (i.e., endoscopic retrograde
cholangiopancreatography [ERCP], endoscopic ultrasound [EUS]) have
given normal or inconclusive results. In this regard, tests using direct
stimulation of the pancreas with secretin are the most sensitive.
Pancreatic Enzymes in Body Fluids 
The serum amylase and
lipase levels are widely used as screening tests for acute pancreatitis
in the patient with acute abdominal pain or back pain. Lipase is more
specific for the pancreas, and values greater than three times the upper
limit of normal (3× ULN) in combination with epigastric pain strongly
suggest the diagnosis of acute pancreatitis. In acute pancreatitis, the
serum amylase and lipase are usually elevated within 24 h of onset
and remain so for 3–7 days. Levels usually return to normal within 7 days
unless there is pancreatic ductal disruption, ductal obstruction, or
pseudocyst formation. Approximately 85% of patients with acute pan­
creatitis have threefold or greater elevated serum lipase and amylase
levels. The values may be normal if (1) there is a delay (2–5 days) before
blood samples are obtained, (2) the underlying disorder is chronic
pancreatitis rather than acute pancreatitis, or (3) hypertriglyceridemia
is present. Patients with hypertriglyceridemia and acute pancreatitis
have been found to have spuriously low levels of amylase and perhaps
lipase activity. In the absence of objective evidence of pancreatitis by
abdominal ultrasound, contrast-enhanced CT scan, MRI with MRCP,
or EUS, mild to moderate elevations of amylase and/or lipase are not
helpful in making a diagnosis of chronic pancreatitis.
It should be noted that the serum amylase can be elevated in other
conditions (Table 358-2), in part because the enzyme is found in many
organs. In addition to the pancreas and salivary glands, small quantities
of amylase are found in the tissues of the fallopian tubes, lung, thyroid,
and tonsils and can be produced by various tumors (carcinomas of the
lung, esophagus, breast, and ovary). Isoamylase determinations do not
accurately distinguish elevated blood amylase levels from pancreatic
or nonpancreatic sources. In patients with unexplained hyperamylas­
emia, the measurement of macroamylase can avoid numerous tests in
patients with this rare disorder.
Elevation of ascitic fluid amylase occurs in acute pancreatitis as well
as in (1) ascites due to disruption of the main pancreatic duct or a leak­
ing pseudocyst and (2) other abdominal disorders that simulate pan­
creatitis (e.g., intestinal obstruction, intestinal infarction, or perforated
peptic ulcer). Elevation of pleural fluid amylase can occur in acute
pancreatitis, chronic pancreatitis, carcinoma of the lung, and esopha­
geal perforation. Serum lipase is the single best enzyme to measure
for the diagnosis of acute pancreatitis. It is important to acknowledge
that levels are often mildly elevated in the setting of renal disease, so
determining whether a patient with renal failure and abdominal pain
has pancreatitis remains a challenging clinical problem. One study
found that serum amylase levels were elevated in patients with renal
dysfunction only when creatinine clearance was <0.8 mL/s (<50 mL/min).
In such patients, the serum amylase level was invariably <500 IU/L in
the absence of objective evidence of acute pancreatitis. In that study,
serum lipase and trypsin levels paralleled serum amylase values. With
these limitations in mind, the recommended screening test for acute
pancreatitis in renal disease is serum lipase, but a high index of clinical
suspicion is needed based on symptoms. Elevations in serum lipase
3× ULN due to nonpancreatic etiology can be observed in hepato­
biliary or gastrointestinal malignancies, septicemia, liver cirrhosis,
systemic lupus erythematosus, severe head injury, chronic alcoholism,
diabetes mellitus, and post-ERCP without any associated evidence of
pancreatitis.
COVID-19 infection has been associated with asymptomatic eleva­
tions of both amylase and lipase. In a systematic review of 21 studies
involving 36,496 patients during the COVID-19 pandemic, the pooled
prevalence of hyperlipasemia (>3 × ULN) and hyperamylas­
emia (>3 × ULN) was 5.6% (95% CI, 2.8–9.3%) and 4.0% (95% CI,
0.9–8.7%), respectively. Importantly, the overall prevalence of acute
pancreatitis in this cohort was 1.7%. Abnormal levels of amylase and
lipase were more closely associated with the severity of COVID-19
than a diagnosis of acute pancreatitis.
Studies Pertaining to Pancreatic Structure 
• 
RADIOLOGIC
TESTS  Plain films of the abdomen rarely provide useful informa­
tion related to pancreatic disease and have been superseded by more
detailed imaging studies (ultrasound, EUS, CT, and MRI with MRCP).
TABLE 358-2  Causes of Hyperamylasemia
Pancreatic Disease
I.	 Pancreatitis
A.	 Acute
B.	 Chronic: ductal obstruction
C.	 Complications of pancreatitis
Pancreatic pseudocyst
Ascites caused by pancreatic duct disruption
Pancreatic necrosis
II.	 Pancreatic trauma
III.	 Pancreatic adenocarcinoma
Nonpancreatic Disorders
I.	 Renal insufficiency
II.	 Salivary gland lesions
A.	 Mumps
B.	 Calculus
C.	 Irradiation sialadenitis
D.	 Maxillofacial surgery
III.	 “Tumor” hyperamylasemia
A.	 Carcinoma of the lung, esophagus, breast, or ovary
IV.	 Macroamylasemia
V.	 Burns
VI.	 Diabetes mellitus, particularly when ketoacidosis is present
VII.	 Pregnancy
VIII.	 Renal transplantation
IX.	 COVID-19 infection
X.	 Cerebral trauma
XI.	 Drugs: opiates
Other Abdominal Disorders
I.	 Biliary tract disease: cholecystitis, choledocholithiasis
II.	 Intraabdominal disease
A.	 Perforated or penetrating peptic ulcer
B.	 Intestinal obstruction or inflammation
C.	 Ruptured ectopic pregnancy
D.	 Peritonitis
E.	 Aortic aneurysm
F.	 Postoperative hyperamylasemia
Ultrasonography (US) can provide important information in the
initial emergency ward evaluation of patients with acute pancreatitis,
chronic pancreatitis, pseudocysts, and pancreatic adenocarcinoma.
Sonographic changes can indicate the presence of edema, inflamma­
tion, and calcification (not obvious on plain films of the abdomen),
as well as gallstones, biliary dilation, pseudocysts, and mass lesions. In
acute pancreatitis, the pancreas is characteristically enlarged. In pan­
creatic pseudocyst, the usual appearance is primarily that of a smooth,
round fluid collection. Pancreatic adenocarcinoma distorts the usual
landmarks, and mass lesions >3.0 cm are usually detected as localized,
solid lesions. US is often the initial investigation for patients with sus­
pected pancreatic disease. However, obesity and excess intestinal bowel
gas can interfere with pancreatic imaging, limiting its sensitivity.
CT with intravenous contrast is the best imaging study for the
assessment of complications of acute and chronic pancreatitis. It is
especially useful in the detection of pancreatic and peripancreatic
acute fluid collections, fluid-containing lesions such as pseudocysts,
walled-off necrosis (see Chap. 359, Figs. 359-1, 359-2, and 359-4),
and pancreatic neoplasms. Acute pancreatitis is characterized by (1)
enlargement of the pancreas, (2) distortion of the pancreatic contour
with peripancreatic stranding of adjacent fat tissue, and/or (3) the
presence of pancreatic fluid that has a different attenuation coeffi­
cient than normal pancreas. When possible, CT scans should ideally
be performed with oral and intravenous contrast to detect areas of
pancreatic necrosis. The major benefit of CT scan in acute pancreatitis
is the diagnosis of pancreatic necrosis in patients not responding to
conservative management within 72 h. It may take 48–72 h to develop
perfusion defects indicative of pancreatic necrosis. Therefore, if acute
pancreatitis is confirmed with serology and physical examination find­
ings, CT scan in the first 3 days is not recommended to minimize risk
of contrast-induced nephropathy and unnecessary health care costs.
Improved imaging technology and increased resolution are facilitated
by multiphasic CT scans using multidetector technology (MDCT)
in which a pancreas protocol consisting of dual-phase scanning with
intravenous contrast is utilized for the detection and staging of pan­
creatic cancers. While the sensitivity of MDCT for detecting smaller
(≤2 cm) lesions is lower, the reported overall sensitivity for pancreatic
cancers is 76–97%. The contraindications to using intravenous contrast
include renal failure (serum creatinine >2 mg/dL) and a history of
severe allergic reaction to iodinated contrast agents. In situations where
EUS is not available, CT-guided percutaneous aspiration or biopsy of a
pancreatic mass can be performed. Prior to the major advance of EUSguided fine-needle aspiration (FNA), CT-guided biopsy was utilized in
the preceding decades and is regarded as a safe procedure.
MRI and MRCP provide excellent imaging of the bile duct, pan­
creatic duct, and pancreas parenchyma in both acute pancreatitis and
chronic pancreatitis. MRI is more sensitive than transabdominal US
and CT scans and comparable to EUS for the detection of choledo­
cholithiasis. Similar to CT, MRI can evaluate for the severity of acute
pancreatitis. Moreover, T2-weighted MRI of fluid collections can dif­
ferentiate necrotic debris from fluid in suspected walled-off necrosis,
and T1 imaging can diagnose hemorrhage in suspected pseudoaneu­
rysm rupture. In chronic pancreatitis, secretin-enhanced MRCP is a
method to enhance the evaluation of major and minor ductal changes.
While imaging is comparable to CT for evaluating pancreatic mass
lesions, MRI with MRCP is the preferred imaging modality for evalu­
ating pancreatic cystic lesions. Nephrogenic systemic fibrosis has been
described in patients with chronic renal failure following exposure to
the gadolinium contrast, but incidence rates are extraordinarily low
with contemporary contrast agents.
CHAPTER 358
Approach to the Patient with Pancreatic Disease
EUS produces high-resolution images of the bile duct, pancreatic
parenchyma, and pancreatic duct with a transducer fixed to an endo­
scope that can be directed onto the surface of the pancreas through the
stomach or duodenum. EUS is not beneficial for the evaluation of pan­
creas during acute pancreatitis. It is preferable to perform EUS after the
resolution of acute pancreatitis (~4 weeks) to detect any predisposing
factors, including malignancy, choledocholithiasis, pancreatic divisum,
or ampullary lesions. EUS can be combined with ERCP in a single ses­
sion and is increasingly preferred for the diagnosis and management
of choledocholithiasis in acute pancreatitis and pancreatic neoplasm
with biliary obstruction. EUS has been studied as a diagnostic modal­
ity for chronic pancreatitis. Criteria for abnormalities on EUS in severe
chronic pancreatic disease have been developed. There is general
agreement that the presence of five or more of the nine criteria listed
in Table 358-3 is highly predictive of chronic pancreatitis in the correct
clinical context. The sensitivity of EUS (81%; 95% CI, 70–89%) to diag­
nose chronic pancreatitis is comparable to that of MRI/MRCP (78%;
95% CI, 69–85%) and better than CT (75%; 95% CI, 66–83%); however,
nonspecific changes are commonly seen in the pancreas that may be
attributable to cigarette smoking, diabetes, or normal aging. EUS also
facilitates the delivery of nerve-blocking agents via fine-needle injec­
tion in patients suffering from pancreatic pain from chronic pancre­
atitis (celiac plexus block) or cancer (celiac plexus neurolysis). When
TABLE 358-3  Endoscopic Ultrasonographic Criteria for Chronic
Pancreatitis (Total Criteria = 9)
DUCTAL
PARENCHYMAL
Stones
Echogenic strands
Hyperechoic main duct margins
Echogenic foci
Main duct irregularity
Lobular contour
Main duct dilatation
Cysts
Visible side branches

32 - 359 Acute and Chronic Pancreatitis
359 Acute and Chronic Pancreatitis
clinically suspected, EUS imaging is more sensitive than MDCT for
the detection of pancreatic malignancy and permits fine-needle aspira­
tion/biopsy (FNA/B). Currently, EUS-guided FNA/B is the diagnostic
modality of choice for the acquisition of diagnostic tissue and cyst fluid
in patients with pancreatic masses and cystic lesions, respectively.
Although a pancreatogram during ERCP is the most specific and
sensitive test for evaluating the ductal anatomy, EUS and MRI/MRCP
have largely replaced ERCP in the diagnostic evaluation of pancreatic
disease to avoid the risk of complications. Therefore, ERCP is primar­
ily of therapeutic value after CT, EUS, or MRI/MRCP has detected
abnormalities requiring endoscopic treatment. ERCP is the most sensi­
tive modality for the detection of bile duct stones. In the management
of acute biliary pancreatitis, ERCP should not be unduly delayed in
patients with high clinical suspicion of biliary obstruction. In chronic
pancreatitis, ERCP abnormalities in the main pancreatic duct and side
branches have been outlined by the Cambridge classification (Fig. 358-2).
The presence of ductal stenosis and irregularity can make it difficult
to distinguish chronic pancreatitis from pancreatic adenocarcinoma. It
is important to be aware that ERCP changes interpreted as indicating
chronic pancreatitis actually may be due to the effects of aging on the
pancreatic duct, sequelae of a recent attack of acute pancreatitis, or
changes secondary to placement of pancreatic duct stent. Pancreatic
adenocarcinoma is characterized by stenosis or obstruction of either
the pancreatic duct or the common bile duct; both ductal systems are
often abnormal (double-duct sign). When indicated, ERCP permits
acquisition of diagnostic tissue as in biopsy of ampullary lesions or
biliary brushings for distal bile duct strictures. Elevated serum amy­
lase levels after ERCP have been reported in the majority of patients,
and clinical pancreatitis has been reported in 5–10% of patients.
Historically, pancreatic duct stents were commonly placed to prevent
post-ERCP pancreatitis. Recent data, however, have revealed that the
periprocedural administration of rectal indomethacin can decrease
the incidence of post-ERCP pancreatitis. In a randomized multicenter
study, it was demonstrated that the combination of rectal indometha­
cin with prophylactic pancreatic duct stents was more effective than
indomethacin alone in preventing post-ERCP pancreatitis in high-risk
patients.
PART 10
Disorders of the Gastrointestinal System
■
■TESTS OF EXOCRINE PANCREATIC FUNCTION
Pancreatic function tests (Table 358-1) can be divided into the
following:
Direct stimulation of the pancreas by IV administration of secretin
followed by collection and measurement of duodenal contents: The
secretin test, used to detect diffuse pancreatic disease, is based on
the physiologic principle that the pancreatic secretory response is
directly related to the functional mass of pancreatic tissue. In the
standard assay, secretin is given IV in a dose of 0.2 μg/kg of synthetic
human secretin as a bolus. Normal values for the standard secretin
test are (1) volume output >2 mL/kg per h, (2) bicarbonate (HCO3
–)
concentration >80 mmol/L, and (3) HCO3
– output >10 mmol/L in 1 h.
The most reproducible measurement, giving the highest level of
discrimination between normal subjects and patients with chronic
pancreas dysfunction, appears to be the maximal bicarbonate con­
centration. A cutoff point <80 mmol/L is considered abnormal and
suggestive of reduced secretory function that is most commonly
observed in early chronic pancreatitis.
There may be a dissociation between the results of the secretin test
and other tests of absorptive function. For example, patients with
chronic pancreatitis often have abnormally low outputs of HCO3
–
after secretin but have normal fecal fat excretion. The secretin test
directly measures the secretory capacity of ductular epithelium,
whereas fecal fat excretion indirectly reflects intraluminal lipolytic
activity. Steatorrhea does not occur until intraluminal levels of
lipase are markedly reduced, underscoring the fact that only small
amounts of enzymes are necessary for intraluminal digestive activi­
ties. It must be emphasized that an abnormal secretin test result
suggests that pancreatic ductal secretory function is abnormal. This
is an early abnormality in chronic pancreatitis but should not be
considered diagnostic and must be interpreted within the proper
clinical context.
3.	 Measurement of fecal pancreatic enzymes such as elastase: Measure­
ment of intraluminal digestion products (i.e., undigested muscle
fibers, stool fat, and fecal nitrogen) is discussed in Chap. 336. The
amount of human elastase in stool reflects the pancreatic output of
this proteolytic enzyme. Decreased fecal elastase-1 (FE-1) activity in
stool is a test to detect severe EPI in patients with chronic pancre­
atitis and cystic fibrosis. FE-1 levels >200 μg/g are normal, levels of
100–200 μg/g are considered mild-moderate EPI, and levels <100 μg/g
are severe EPI. Although the test is simple and noninvasive, it can
yield false-positive results if stools are not formed and should not
generally be used for the evaluation of a patient with diarrhea. Falsepositive results have also been observed in diabetes and irritable
bowel syndrome.
Tests useful in the diagnosis of EPI and the differential diagnosis
of malabsorption are also discussed in Chaps. 336 and 359.
■
■FURTHER READING
Conwell DL et al: American Pancreatic Association practice guidelines
in chronic pancreatitis: Evidence-based report on diagnostic guidelines.
Pancreas 43:1143, 2014.
Hart PA et al: Endoscopic pancreas fluid collection: Methods and
relevance for clinical care and translational science. Am J Gastroenterol
111:1258, 2016.
Petrov MS, Yadav D: Global epidemiology and holistic prevention of
pancreatitis. Nat Rev Gastroenterol Hepatol 16:175, 2019.
Singh  VK et al: Diagnosis and management of chronic pancreatitis:
A review. JAMA 322:2422, 2019.
Phil A. Hart, Darwin L. Conwell,
Somashekar G. Krishna
Acute and Chronic
Pancreatitis
BIOCHEMISTRY AND PHYSIOLOGY OF
PANCREATIC EXOCRINE SECRETION
■
■GENERAL CONSIDERATIONS
The pancreas secretes 1500–3000 mL of isosmotic alkaline (pH >8)
fluid per day containing ~20 enzymes. Pancreatic secretions provide
the enzymes and bicarbonate needed to perform the major digestive
activity of the gastrointestinal tract and produce an optimal pH for the
function of these enzymes.
■
■REGULATION OF PANCREATIC SECRETION
Secretions from the exocrine pancreas are highly regulated by neuro­
hormonal systems in a phasic manner (cephalic, gastric, and intestinal
phases). Gastric acid is the stimulus for the release of secretin from the
duodenal mucosa (S cells), which stimulates the secretion of water and
electrolytes from pancreatic ductal cells. Release of cholecystokinin
(CCK) from the duodenal and proximal jejunal mucosa (Ito cells)
is largely triggered by long-chain fatty acids, essential amino acids
(tryptophan, phenylalanine, valine, methionine), and gastric acid
itself. CCK evokes an enzyme-rich secretion from acinar cells in the
pancreas. The parasympathetic nervous system (via the vagus nerve)
exerts significant control over pancreatic secretion, particularly during
the cephalic phase. Secretion evoked by secretin and CCK depends on
the permissive roles of vagal afferent and efferent pathways. This is
particularly true for enzyme secretion, whereas water and bicarbonate
secretions are heavily dependent on the hormonal effects of secretin
and to a lesser extent CCK. Also, vagal stimulation affects the release
of vasoactive intestinal peptide (VIP), a secretin agonist. Pancreatic
exocrine secretion is also influenced by inhibitory neuropeptides
including somatostatin, pancreatic polypeptide, peptide YY, neuro­
peptide Y, enkephalin, pancreastatin, calcitonin gene–related peptides,
glucagon, and galanin. Pancreatic polypeptide and peptide YY may act
primarily on nerves outside the pancreas, while somatostatin acts at
multiple sites.
■
■WATER AND ELECTROLYTE SECRETION
Bicarbonate is the ion of primary physiologic importance in pancreatic
secretions. The ductal cells secrete bicarbonate predominantly derived
from plasma (93%) more so than from intracellular metabolism (7%).
Bicarbonate enters the duct lumen through the sodium bicarbonate
cotransporter with depolarization caused by chloride efflux through
the cystic fibrosis transmembrane conductance regulator (CFTR).
Secretin and VIP bind at the basolateral surface and cause an increase
in secondary messenger intracellular cyclic AMP and act on the apical
surface of the ductal cells opening the CFTR to promote secretion.
CCK, acting as a neuromodulator, markedly potentiates the stimula­
tory effects of secretin. Acetylcholine also plays an important role in
ductal cell secretion. Intraluminal bicarbonate secreted from the ductal
cells helps neutralize gastric acid, increases the solubility of fatty acids
and bile acids, maintains an optimal pH for pancreatic and brush bor­
der enzymes, and prevents intestinal mucosal damage.
■
■ENZYME SECRETION
The acinar cell is highly compartmentalized for the production and
secretion of pancreatic enzymes. Proteins synthesized by the rough
endoplasmic reticulum are processed in the Golgi apparatus and then
targeted to the appropriate site: zymogen granules, lysosomes, or
other cell compartments. The zymogen granules migrate to the apical
region of the acinar cell awaiting the appropriate neural or hormonal
stimulatory response. The pancreas secretes amylolytic, lipolytic, and
proteolytic enzymes into the duct lumen. Amylolytic enzymes, such
as amylase, hydrolyze starch to oligosaccharides and the disaccharide
maltose. The lipolytic enzymes include lipase, phospholipase A2, and
cholesterol esterase. Bile salts inhibit lipase in isolation, but colipase,
another constituent of pancreatic secretion, binds to lipase and pre­
vents this inhibition. Bile salts activate phospholipase A and choles­
terol esterase. Proteolytic enzymes include endopeptidases (trypsin,
chymotrypsin), which act on internal peptide bonds of proteins and
polypeptides; exopeptidases (carboxypeptidases, aminopeptidases),
which act on the free carboxyl- and amino-terminal ends of pep­
tides, respectively; and elastase. The proteolytic enzymes are secreted
as inactive zymogen precursors. Ribonucleases (deoxyribonucleases,
ribonuclease) are also secreted. Enterokinase, an enzyme found in
the duodenal mucosa (“brush border”), cleaves the lysine-isoleucine
bond of trypsinogen to form trypsin. Trypsin then activates the other
proteolytic zymogens and phospholipase A2 in a cascade. The nervous
system initiates pancreatic enzyme secretion. The neurologic stimula­
tion is cholinergic, involving extrinsic innervation by the vagus nerve
and subsequent innervation by intrapancreatic cholinergic nerves. The
stimulatory neurotransmitters are acetylcholine and gastrin-releasing
peptides. These neurotransmitters activate calcium-dependent second­
ary messenger systems, resulting in the release of zymogens into the
pancreas duct. VIP is present in intrapancreatic nerves and potentiates
the effect of acetylcholine. In contrast to other species, there are no
CCK receptors on acinar cells in humans. CCK in physiologic con­
centrations contributes to pancreatic secretion by stimulating afferent
vagal and intrapancreatic nerves.
■
■AUTOPROTECTION OF THE PANCREAS
Autodigestion of the pancreas is prevented by (1) the packaging of pan­
creatic proteases in the precursor (proenzyme) form, (2) intracellular
calcium homeostasis (low intracellular calcium in the cytosol of the
acinar cell promotes the destruction of spontaneously activated tryp­
sin), (3) acid-base balance, and (4) the synthesis of protective protease
inhibitors serine protease inhibitor, Kazal type 1 (SPINK1) which can
bind and inactivate ~20% of intracellular trypsin activity. Chymotryp­
sin C can also lyse and inactivate trypsin. These protease inhibitors are
found in acinar cells, pancreatic secretions, and the α1- and α2-globulin
fractions of plasma. Loss of any of these four protective mechanisms
leads to premature enzyme activation, autodigestion, and ultimately
acute pancreatitis.
■
■ENTEROPANCREATIC AXIS AND
FEEDBACK INHIBITION
Pancreatic enzyme secretion is controlled, at least in part, by a negative
feedback mechanism induced by the presence of active serine prote­
ases in the duodenum and nutrients in the distal small intestine. For
example, perfusion of the duodenal lumen with phenylalanine (stimu­
lates early digestion) causes a prompt increase in plasma CCK levels
as well as increased secretion of chymotrypsin and other pancreatic
enzymes. However, simultaneous perfusion with trypsin (stimulates
late digestion) blunts both responses. Conversely, perfusion of the duo­
denal lumen with protease inhibitors leads to enzyme hypersecretion.
Available evidence supports the concept that the duodenum contains a
peptide called CCK-releasing factor (CCK-RF) that is involved in stim­
ulating CCK release. It appears that serine proteases inhibit pancreatic
secretion by inactivating a CCK-releasing peptide in the lumen of the
small intestine. Thus, the integrative result of both bicarbonate and
enzyme secretion depends on a feedback process for both bicarbon­
ate and pancreatic enzymes. Acidification of the duodenum releases
secretin, which stimulates vagal and other neural pathways to activate
pancreatic duct cells, which secrete bicarbonate. This bicarbonate then
neutralizes the duodenal acid, and the feedback loop is completed.
Dietary proteins bind proteases, thereby leading to an increase in free
CCK-RF. CCK is then released into the blood in physiologic concentra­
tions, acting primarily through the neural pathways (vagal-vagal). This
leads to acetylcholine-mediated pancreatic enzyme secretion. Proteases
continue to be secreted from the pancreas until the protein within the
duodenum is digested. At this point, pancreatic protease secretion is
reduced to basic levels, thus completing this step in the feedback pro­
cess. Additional hormonal feedback inhibition of pancreatic enzyme
secretion occurs via peptide YY and glucagon-like peptide-1 following
lipid or carbohydrate exposure to the ileum.
CHAPTER 359
Acute and Chronic Pancreatitis
ACUTE PANCREATITIS
■
■GENERAL CONSIDERATIONS
Recent U.S. estimates indicate that acute pancreatitis is one of the most
common principal gastrointestinal diagnoses, accounting for almost
300,000 hospitalizations annually. The annual incidence ranges from
15–45/100,000 persons, depending on the distribution of etiologies
(e.g., alcohol, gallstones, metabolic factors, drugs; Table 359-1) and
country of study. The median length of hospital stay is 4 days, with a
median hospital cost of ~$6000 and a mortality of ~1%. The estimated
annual cost approaches $3 billion. Hospitalization rates increase with
age and are higher among African Americans. Globally, there is a
steady increase in the incidence of acute pancreatitis, most promi­
nently in North America and Europe. Thus, the incidence of acute
pancreatitis continues to rise and is associated with substantial health
care costs.
■
■ETIOLOGY AND PATHOGENESIS
There are many causes of acute pancreatitis (Table 359-1), and the
mechanisms by which each of these conditions triggers pancreatic
inflammation have not been fully elucidated. Gallstones and alcohol
account for 80–90% of identified cases of acute pancreatitis in the
United States. Gallstones are the leading cause of acute pancreatitis
in most series (30–60%). The risk of acute pancreatitis in patients
with at least one gallstone <5 mm in diameter is fourfold greater than
patients with larger stones. Alcohol is the second most common cause,
responsible for 15–30% of cases in the United States. The incidence
of pancreatitis in individuals with heavy alcohol use is surprisingly
low (5/100,000), indicating that in addition to the amount of alcohol
TABLE 359-1  Causes of Acute Pancreatitis
Common Causes
Gallstones (including microlithiasis)
Heavy alcohol use
Severe hypertriglyceridemia
Endoscopic retrograde cholangiopancreatography (ERCP), especially after
therapeutic intervention
Idiopathic
Uncommon Causes
Drugs (azathioprine, 6-mercaptopurine, sulfonamides, estrogens, tetracycline,
valproic acid, 5-aminosalicylic acid [5-ASA], dipeptidyl peptidase-4 [DPP4]
inhibitors)
Connective tissue disorders and thrombotic thrombocytopenic purpura (TTP)
Pancreatic cancer
Hypercalcemia
Periampullary diverticulum
Pancreas divisuma
Hereditary pancreatitis
Cystic fibrosis
Renal failure
Infections (mumps, coxsackievirus, cytomegalovirus, echovirus, parasites)
Autoimmune (e.g., type 1 and type 2)
Trauma (especially blunt abdominal trauma)
Postoperative (abdominal and nonabdominal operations)
Causes to Consider in Patients With Recurrent Bouts of Acute
Pancreatitis Without an Obvious Etiology
PART 10
Disorders of the Gastrointestinal System
Occult disease of the biliary tree or pancreatic ducts, especially microlithiasis/
biliary sludge
Alcohol abuse
Metabolic: Hypertriglyceridemia, hypercalcemia
Anatomic: Pancreas divisuma
Pancreatic cancer
Intraductal papillary mucinous neoplasm (IPMN)
Hereditary pancreatitis
Cystic fibrosis
Idiopathic
aPancreas divisum is not believed to cause acute pancreatitis in isolation of another
disease precipitant.
ingested, other factors affect a person’s susceptibility to pancreatic
injury, such as cigarette smoking and genetic predisposition. Acute
pancreatitis occurs in 5–10% of patients following endoscopic retro­
grade cholangiopancreatography (ERCP); however, this risk can be
decreased with proper patient selection and the use of a prophylactic
pancreatic duct stent and/or rectal nonsteroidal anti-inflammatory
drugs (NSAIDs; indomethacin). Risk factors for post-ERCP pan­
creatitis include minor papilla sphincterotomy, suspected functional
sphincter dysfunction (previously termed sphincter of Oddi dysfunc­
tion), prior history of post-ERCP pancreatitis, age <60 years, more
than two contrast injections into the pancreatic duct, and endoscopist
experience.
Hypertriglyceridemia is the cause of acute pancreatitis in 1–4%
of cases; serum triglyceride levels are usually >1000 mg/dL. Most
patients with hypertriglyceridemic pancreatitis have undiagnosed or
uncontrolled diabetes mellitus. An additional subset has an underlying
derangement in lipid metabolism, probably unrelated to pancreati­
tis. Such patients are prone to recurrent episodes of pancreatitis and
increased severity. Any factor (e.g., alcohol or medications, such as
oral contraceptives) that causes an abrupt increase in serum triglyc­
erides can potentially precipitate a bout of acute pancreatitis. Patients
with a deficiency of apolipoprotein CII have an increased incidence of
pancreatitis; apolipoprotein CII activates lipoprotein lipase, which is
important in clearing chylomicrons from the bloodstream. Although
frequently entertained, <2% of cases of acute pancreatitis are drug
related. Drugs cause pancreatitis either by a hypersensitivity reaction
or by the generation of a toxic metabolite, although in some cases,
it is not clear which of these mechanisms is operative (Table 359-1).
Increased attention has been directed toward a potential increased risk
for acute pancreatitis in patients taking glucagon-like peptide-1 recep­
tor agonists (GLP-1RA) and dipeptidyl peptidase-4 (DPP4) inhibitors.
The most recent metanalysis of cardiovascular outcome trials does
support an increased risk with DPP4 inhibitors, but not with GLP-1RA
medications. When used for treating hyperglycemia, it is important to
be conscious that >10% of individuals with diabetes will have elevated
amylase or lipase levels irrespective of medication use.
Pathologically, acute pancreatitis ranges from interstitial pancreatitis
(pancreas blood supply maintained), which is generally self-limited, to
necrotizing pancreatitis (pancreas blood supply interrupted). Autodi­
gestion is a currently accepted pathogenic theory resulting when pro­
teolytic enzymes (e.g., trypsinogen, chymotrypsinogen, proelastase)
and lipolytic enzymes (such as phospholipase A2) are activated in the
pancreas acinar cell compartment rather than the intestinal lumen.
Several factors (e.g., endotoxins, exotoxins, viral infections, ischemia,
oxidative stress, lysosomal calcium, direct trauma) are believed to
facilitate premature activation of trypsin, which can activate other
enzymes. Spontaneous activation of trypsin also can occur, resulting
in autodigestion.
■
■ACTIVATION OF PANCREATIC ENZYMES IN THE
PATHOGENESIS OF ACUTE PANCREATITIS
Several studies have suggested that pancreatitis is a disease that evolves
in three phases. The initial phase is characterized by intrapancreatic
digestive enzyme activation and acinar cell injury. Trypsin activation
appears to be mediated by lysosomal hydrolases such as cathepsin
B that become colocalized with digestive enzymes in intracellular
organelles; it is currently believed that acinar cell injury is the conse­
quence of trypsin activation. The second phase of pancreatitis involves
the activation, chemoattraction, and sequestration of leukocytes and
macrophages in the pancreas, resulting in an enhanced intrapancre­
atic inflammatory reaction. Neutrophil depletion induced by prior
administration of an antineutrophil serum has been shown to reduce
the severity of experimentally induced pancreatitis. There is also evi­
dence to support the concept that neutrophils can activate trypsinogen.
Thus, intrapancreatic acinar cell activation of trypsinogen could be
a two-step process (i.e., an early neutrophil-independent and a later
neutrophil-dependent phase). The third phase of pancreatitis is due
to the effects of activated proteolytic enzymes and cytokines, released
by the inflamed pancreas, on distant organs. Activated proteolytic
enzymes, especially trypsin, not only digest pancreatic and peripan­
creatic tissues but also activate other enzymes such as elastase and
phospholipase A2. The active enzymes and cytokines then digest cel­
lular membranes and cause proteolysis, edema, interstitial hemorrhage,
vascular damage, coagulation necrosis, fat necrosis, and cellular necro­
sis in the parenchyma. Cellular injury and death result in the liberation
of bradykinin peptides, vasoactive substances, and histamine that can
produce vasodilation, increased vascular permeability, and edema with
profound effects on other organs. The systemic inflammatory response
syndrome (SIRS) and acute respiratory distress syndrome (ARDS), as
well as multiorgan failure, may occur because of this cascade producing
local and distant effects.
Several genetic factors can increase the susceptibility and/or modify
the severity of pancreatic injury in acute pancreatitis, recurrent acute
pancreatitis, and chronic pancreatitis. The major genetic susceptibility
factors center on the control of trypsin activity within the pancreatic
acinar cell, in part because they were identified as candidate genes
linked to intrapancreatic trypsin control. Six genetic variants have
been identified as being associated with susceptibility to pancreatitis.
The genes that have been identified include (1) cationic trypsinogen
gene (PRSS1), (2) SPINK1, (3) the cystic fibrosis transmembrane
conductance regulator gene (CFTR), (4) the chymotrypsin C gene
(CTRC), (5) the calcium-sensing receptor (CASR), and (6) claudin-2
(CLDN2). Among these variants, only PRSS1 mutations are sufficient
to precipitate acute pancreatitis in the absence of other risk factors,
whereas the other variants are disease modifiers. Investigations for
other genetic variants are in progress, so it is expected this list will be
expanded in the future.
APPROACH TO THE PATIENT
Abdominal Pain
Abdominal pain is the major symptom of acute pancreatitis. Pain
may vary from mild discomfort to severe, constant, and incapacitat­
ing distress. Characteristically, the pain, which is steady and boring
in character, is located in the epigastrium region and may radiate to
the back, chest, flanks, and lower abdomen. Nausea, vomiting, and
abdominal distention due to gastric and intestinal hypomotility are
also frequent complaints.
Physical examination frequently reveals a distressed and anxious
patient. Low-grade fever, tachycardia, and hypotension are com­
mon. Shock is not unusual and may result from (1) hypovolemia
secondary to exudation of blood and plasma proteins into the
retroperitoneal space; (2) increased formation and release of kinin
peptides, which cause vasodilation and increased vascular perme­
ability; and (3) systemic effects of proteolytic and lipolytic enzymes
released into the circulation. Jaundice occurs infrequently; when
present, it may be a consequence of extrinsic compression due
to peripancreatic edema, a pancreatic head mass, or intraductal
obstruction from a common bile duct stone or sludge. Erythema­
tous skin nodules (“pancreatic panniculitis”) due to subcutaneous
fat necrosis are possible, but rarely occur. There are pulmonary
findings in 10–20% of patients, including basilar rales, atelectasis,
and pleural effusion, the latter most frequently left-sided. Abdomi­
nal tenderness and muscle rigidity are present to a variable degree,
but compared with the intense pain, these signs may be less impres­
sive. Bowel sounds are usually diminished or absent. An enlarged
pancreas from an acute fluid collection, walled-off necrosis, or
a pseudocyst may be palpable in the upper abdomen later in the
course of the disease (i.e., 4–6 weeks). A faint blue discoloration
around the umbilicus (Cullen’s sign) may occur as the result of
hemoperitoneum, and a blue-red-purple or green-brown discol­
oration of the flanks (Turner’s sign) reflects tissue breakdown of
hemoglobin from severe necrotizing pancreatitis with hemorrhage;
both findings are rare but reflect an increased clinical severity.
■
■LABORATORY DATA
Serum amylase and lipase values greater than three times the upper
limit of normal are strongly supportive of the diagnosis if alternate
etiologies, including gut perforation, ischemia, and infarction, are
excluded. However, it should be noted that there is no correlation
between the severity of pancreatitis and the degree of serum lipase and
amylase elevations or serial trends. After 3–7 days, even with continu­
ing evidence of pancreatitis, total serum amylase values tend to return
to normal. However, pancreatic lipase levels may remain elevated for
7–14 days. It should be recognized that amylase elevations in serum
and urine occur in many conditions other than pancreatitis (see Chap.
358, Table 358-2). Importantly, patients with acidemia (arterial pH
≤7.32) may have spurious elevations in serum amylase. This finding
explains why patients with diabetic ketoacidosis may have marked ele­
vations in serum amylase without any other evidence of acute pancre­
atitis. On the other hand, serum amylase levels can be spuriously low in
severe hypertriglyceridemia. Serum lipase activity increases in parallel
with amylase activity and is more specific than amylase, making it the
preferred test. A serum lipase measurement can be instrumental in dif­
ferentiating a pancreatic or nonpancreatic cause for hyperamylasemia.
Leukocytosis (15,000–20,000 leukocytes/μL) occurs frequently.
Patients with more severe disease may show hemoconcentration with
hematocrit values >44% and prerenal azotemia (elevated blood urea
nitrogen [BUN]) resulting from loss of plasma into the retroperito­
neal space and peritoneal cavity. In fact, one study reported that a
combination of admission BUN and a change in BUN at 48 h was a
strong predictor of mortality (area under the curve [AUC] 0.91).
Hemoconcentration may be the harbinger of more severe disease,
whereas azotemia is a significant risk factor for mortality. Hypergly­
cemia is common and is due to multiple factors, including decreased
insulin release, increased glucagon release, and increased output of
adrenal glucocorticoids and catecholamines. Hypocalcemia occurs in
~25% of patients, and its pathogenesis is incompletely understood.
Although earlier studies suggested that the response of the parathy­
roid gland to a decrease in serum calcium is impaired, subsequent
observations have failed to confirm this phenomenon. Intraperitoneal
saponification of calcium by fatty acids in areas of fat necrosis occurs
occasionally, with large amounts (up to 6.0 g) dissolved or suspended in
ascitic fluid. Such “soap formation” may also be significant in patients
with pancreatitis, mild hypocalcemia, and little or no obvious ascites.
Hyperbilirubinemia (serum bilirubin >4.0 mg/dL) occurs in ~10% of
patients. However, jaundice is usually transient, and serum bilirubin
levels return to normal in 4–7 days. Serum alkaline phosphatase and
transaminase levels may also be transiently elevated and parallel serum
bilirubin values. Elevations of alanine aminotransferase (ALT) >3× the
upper limit of normal are strongly associated with a gallstone etiology
in patients with acute pancreatitis. Approximately 5–10% of patients
have hypoxemia (arterial PO2 ≤60 mmHg), which may herald the onset
of ARDS. Finally, the electrocardiogram is occasionally abnormal in
acute pancreatitis with ST-segment and T-wave abnormalities simulat­
ing myocardial ischemia.
An abdominal ultrasound is recommended as the initial diagnostic
imaging modality and is most useful for evaluating gallstones and com­
mon bile duct dilation.
CHAPTER 359
The Revised Atlanta Criteria have clearly outlined the morpho­
logic features of acute pancreatitis on computed tomography (CT)
scan as follows: (1) interstitial pancreatitis, (2) necrotizing pancreati­
tis, (3) acute pancreatic fluid collection, (4) pancreatic pseudocyst,
(5) acute necrotic collection (ANC), and (6) walled-off necrosis
(WON) (Table 359-2 and Fig. 359-1). Radiologic studies useful in
the diagnosis of acute pancreatitis are discussed in Chap. 358 and
listed in Table 358-1.
Acute and Chronic Pancreatitis
■
■DIAGNOSIS
Any severe acute pain in the abdomen or back should raise the pos­
sibility of acute pancreatitis. The diagnosis is established by two of the
following three criteria: (1) typical abdominal pain in the epigastrium
that may radiate to the back, (2) threefold or greater elevation in serum
lipase and/or amylase, and (3) confirmatory findings of acute pancre­
atitis on cross-sectional abdominal imaging. Although not required
for diagnosis, markers of severity may include hemoconcentration
(hematocrit >44%), admission azotemia, SIRS, and signs of organ fail­
ure (Table 359-3).
The differential diagnosis should include the following disorders: (1)
perforated viscus, especially peptic ulcer; (2) acute cholecystitis and
biliary colic; (3) acute intestinal obstruction; (4) mesenteric vascular
occlusion; (5) renal colic; (6) inferior myocardial infarction; (7) dissect­
ing aortic aneurysm; (8) connective tissue disorders with vasculitis; (9)
pneumonia; and (10) diabetic ketoacidosis. It may be difficult to differ­
entiate acute cholecystitis from acute pancreatitis, because an elevated
serum amylase may be found in both disorders. Pain of biliary tract
origin is more right sided or epigastric than periumbilical or left upper
quadrant and can be more severe; ileus is usually absent. Ultrasound is
helpful in establishing the diagnosis of cholelithiasis and cholecystitis.
Intestinal obstruction due to mechanical factors can be differentiated
from pancreatitis by the history of crescendo-decrescendo pain, find­
ings on abdominal examination, and CT of the abdomen showing
changes characteristic of anatomic obstruction. Acute mesenteric vas­
cular occlusion is usually suspected in elderly debilitated patients with
leukocytosis, abdominal distention, and bloody diarrhea, confirmed by
CT or magnetic resonance angiography. Vasculitides secondary to sys­
temic lupus erythematosus and polyarteritis nodosa may be confused
with pancreatitis, especially because pancreatitis may develop as a com­
plication of these diseases. Diabetic ketoacidosis is often accompanied
TABLE 359-2  Revised Atlanta Definitions of Morphologic Features of Acute Pancreatitis
 
DEFINITION
COMPUTED TOMOGRAPHY FEATURES
Types of Acute Pancreatitis
Interstitial
pancreatitis
Acute inflammation of the pancreatic parenchyma and
peripancreatic tissues, but without recognizable tissue
necrosis
Necrotizing
pancreatitis
Inflammation associated with pancreatic parenchymal
and/or peripancreatic necrosis
Morphologic Features
Acute pancreatic
fluid collection
Peripancreatic fluid associated with interstitial edematous
pancreatitis with no associated peripancreatic necrosis.
This term applies only to areas of peripancreatic fluid
seen within the first 4 weeks after onset of interstitial
edematous pancreatitis and without the features of a
pseudocyst.
Pancreatic
pseudocyst
An encapsulated collection of fluid with a well-defined
wall usually outside the pancreas with minimal or no
necrosis. This usually occurs >4 weeks after onset of
interstitial edematous pancreatitis.
Acute necrotic
collection (ANC)
A collection containing variable amounts of both fluid and
necrosis associated with necrotizing pancreatitis; the
necrosis can involve the pancreatic parenchyma and/or
the peripancreatic tissues.
Walled-off necrosis
(WON)
A mature, encapsulated collection of pancreatic and/or
peripancreatic necrosis that has developed a well-defined
inflammatory wall. WON usually occurs >4 weeks after
onset of acute necrotizing pancreatitis.
PART 10
Disorders of the Gastrointestinal System
Source: Data from P Banks et al: Gut 62:102, 2013.
by abdominal pain and elevated total serum amylase levels, thus closely
mimicking acute pancreatitis; however, the serum lipase level is often
not elevated in diabetic ketoacidosis, and pancreas imaging is normal.
■
■CLINICAL COURSE, DEFINITIONS, AND
CLASSIFICATIONS
The Revised Atlanta Criteria define (1) phases of acute pancreatitis,
(2) severity of acute pancreatitis, and (3) radiographic definitions, as
outlined below.
Phases of Acute Pancreatitis 
Two clinical phases of acute pan­
creatitis have been defined, early (<2 weeks) and late (>2 weeks), which
primarily describe the hospital course of the disease. In the early phase
of acute pancreatitis, which lasts 1–2 weeks, severity is defined by clini­
cal parameters rather than morphologic findings. Most patients exhibit
SIRS, and if this persists, patients are predisposed to organ failure.
Three organ systems should be assessed to define organ failure: respi­
ratory, cardiovascular, and renal. Organ failure is defined as a score
of 2 or more for one of these three organ systems using the modified
Marshall scoring system. Persistent organ failure (>48 h) is the most
important clinical finding regarding the severity of the acute pancreati­
tis episode. CT imaging is usually not needed or recommended during
the first 48 h of admission in acute pancreatitis.
The late phase is characterized by a protracted course of illness and
may require imaging to evaluate for local complications. The critical
clinical parameter of severity, as in the early phase, is persistent organ
failure. These patients may require supportive measures such as renal
dialysis, ventilator support, or need for supplemental nutrition. The
radiographic feature of greatest importance to recognize in this phase is
the development of necrotizing pancreatitis on CT imaging. Necrosis is
associated with prolonged hospitalization and, if infected, may require
intervention (percutaneous, endoscopic, or, rarely, surgical).
Severity of Acute Pancreatitis 
Three classes of severity have been
defined: mild, moderately severe, and severe. Mild acute pancreatitis is
Pancreatic parenchyma enhancement by IV contrast agent and without
peripancreatic necrosis
Lack of pancreatic parenchymal enhancement by IV contrast agent and/or
presence of findings of peripancreatic necrosis (see below—ANC and WON)
Occurs in the setting of interstitial pancreatitis
Homogeneous collection with fluid density
Confined by normal peripancreatic fascial planes
No definable wall encapsulating the collection
Adjacent to pancreas (no intrapancreatic extension)
Well circumscribed, usually round or oval
Homogeneous fluid density
No solid component
Well-defined wall; that is, completely encapsulated
Maturation usually requires >4 weeks after onset of acute interstitial pancreatitis
Occurs in the setting of acute necrotizing pancreatitis
Heterogeneous and nonliquid density of varying degrees in different locations
(some appear homogeneous early in their course)
No definable wall encapsulating the collection
Location—intrapancreatic and/or extrapancreatic
Heterogeneous with liquid and nonliquid density with varying degrees of loculations
(some may appear homogeneous)
Well-defined wall; that is, completely encapsulated
Location—intrapancreatic and/or extrapancreatic
Maturation usually requires >4 weeks after onset of acute necrotizing pancreatitis
without local complications or organ failure. Most patients with inter­
stitial acute pancreatitis have mild pancreatitis. In mild acute pancre­
atitis, the disease is self-limited, usually within 3–7 days after onset.
Oral intake can be resumed if the patient is hungry, has normal bowel
function, and is without nausea and vomiting. Typically, a clear or full
liquid diet has been recommended for the initial meal; however, a lowfat solid diet is advised early in mild acute pancreatitis.
Moderately severe acute pancreatitis is characterized by transient
organ failure (i.e., it resolves in <48 h) or local or systemic complica­
tions in the absence of persistent organ failure. These patients may or
may not have necrosis but may develop a local complication such as
a fluid collection, which often requires a prolonged hospitalization
1 week. As with mild acute pancreatitis, the mortality rate for these
patients remains low.
Severe acute pancreatitis is characterized by persistent organ failure
(>48 h), involving one or more organs. A CT scan or magnetic reso­
nance imaging (MRI) should be obtained to assess for necrosis and/
or complications. If a local complication is encountered, management
is dictated by clinical symptoms, evidence of infection, the maturity
of fluid collection, and clinical stability of the patient. Prophylactic
antibiotics are no longer recommended for severe acute pancreatitis.
Imaging in Acute Pancreatitis 
Two types of pancreatitis are
recognized on imaging as interstitial or necrotizing based on pancreatic
perfusion. CT imaging with IV contrast is best evaluated 3–5 days into
hospitalization if patients are not responding to supportive care to
assess for local complications such as necrosis. Recent studies report
the overutilization of CT imaging within 72 h for acute pancreatitis,
including those with a mild severity of disease. The Revised Atlanta Cri­
teria outline the terminology for local complications and fluid collec­
tions along with a CT imaging template to guide reporting of findings.
Local morphologic features are summarized in Table 359-2. Interstitial
pancreatitis occurs in 90–95% of admissions for acute pancreatitis and
is characterized by diffuse gland enlargement, homogenous contrast
enhancement, and mild inflammatory changes or peripancreatic
A
B
C
D
FIGURE 359-1  Evolution of changes of acute necrotizing pancreatitis on computed tomography (CT). A. CT scan of the abdomen without IV contrast performed on admission
for a patient with acute gallstone pancreatitis, showing mild peripancreatic stranding. B. Contrast-enhanced CT scan of the abdomen performed on the same patient
1 week after admission shows extensive intrapancreatic necrosis, evidenced by the lack of contrast enhancement in the pancreatic body with minimal enhancement noted
at the distal most aspect of the pancreatic tail. C. Contrast-enhanced CT scan of the abdomen performed on the same patient 2 weeks after admission demonstrates a
semiorganized, heterogeneous fluid collection, referred to as an acute necrotic collection. On this image, a small area of viable pancreatic parenchyma is seen at the tail
of the pancreas. D. Contrast-enhanced CT scan of the abdomen performed on the same patient 5 weeks after admission demonstrates a well-encapsulated fluid collection,
essentially replacing the pancreas, referred to as walled-off necrosis.
stranding. Symptoms generally resolve with a week of hospitalization.
Necrotizing pancreatitis occurs in 5–10% of acute pancreatitis admis­
sions and may not evolve until after several days of hospitalization. It
is characterized by lack of pancreatic parenchymal enhancement by
intravenous contrast agent and/or presence of peripancreatic necrosis.
The natural history of pancreatic and peripancreatic necrosis is vari­
able because it may remain solid or liquefy, remain sterile or become
infected, and persist or resolve over time. Importantly, those with only
extrapancreatic necrosis have a more favorable prognosis than patients
with pancreatic necrosis (with or without extrapancreatic necrosis). CT
identification of local complications, particularly necrosis, is critical
in patients who are not responding to therapy because patients with
necrosis are at the greatest risk of mortality (Figs. 359-1 and 359-2). The
median prevalence of organ failure is >50% in necrotizing pancreatitis
and is slightly higher in infected versus sterile necrosis. With singleorgan system failure, the mortality is 3–10%, but increases to nearly
50% with multiorgan failure.
■
■ACUTE PANCREATITIS MANAGEMENT
The management of patients with acute pancreatitis from the time
of diagnosis in the emergency ward to hospital discharge is briefly
reviewed, highlighting salient features based on severity and complica­
tions. It is important to recognize that 85–90% of cases of acute pancre­
atitis are self-limited and subside spontaneously, usually within 3–7 days
after onset, and do not exhibit organ failure or local complications.
The management of acute pancreatitis begins in the emergency
ward. After a diagnosis has been confirmed, early and aggressive
fluid resuscitation is critical. Additionally, intravenous analgesics are
administered, severity is assessed, and a search for etiologies that may
CHAPTER 359
Acute and Chronic Pancreatitis
impact acute care is begun. Patients who do not respond to aggressive
fluid resuscitation in the emergency ward should be considered for
admission to a step-down or intensive care unit for further intensive
monitoring and management, including hemodynamic monitoring
and management of organ failure, if present.
Fluid Resuscitation and Monitoring Response to Therapy 
The
most important treatment intervention for acute pancreatitis is early,
intravenous fluid resuscitation to prevent systemic complications from
the secondary systemic inflammatory response. The patient is initially
made NPO to minimize nutrient-induced stimulation of the pancreas
and is given intravenous narcotic analgesics to control abdominal pain
and supplemental oxygen (as needed).
Lactated Ringer’s solution has been shown to decrease systemic
inflammation (lower C-reactive protein levels from admission) and
may be a better crystalloid than normal saline. Aggressive intravenous
fluids have historically been promoted including an initial bolus of
approximately 15–20 mL/kg (1050–1400 mL), followed by 2–3 mL/kg
per h (200–250 mL/h), to maintain urine output >0.5 mL/kg per hour.
However, a recent randomized controlled trial demonstrated that this
hydration strategy is associated with an increased risk of fluid overload
(based on signs or symptoms) compared to a less aggressive strategy
(10 mL/kg bolus, followed by 1.5 mL/kg per h) without an improve­
ment in clinical outcomes. Irrespective of the initial fluid management
strategy, serial bedside evaluations are required every 6–8 h to assess
vital signs, oxygen saturation, and change in physical examination
to optimize fluid resuscitation. A targeted resuscitation strategy with
measurement of hematocrit and BUN every 8–12 h is recommended
to ensure the adequacy of fluid resuscitation and monitor response
TABLE 359-3  Severity Assessment of Acute Pancreatitis
Risk Factors for Increased Severity
• Age >60 years
• Obesity, BMI >30 kg/m2
• Comorbid disease (based on Charlson comorbidity index)
Markers of Severity at Admission or Within 24 h
• SIRS—defined by presence of 2 or more criteria:
• Core temperature <36° or >38°C
• Heart rate >90 beats/min
• Respirations >20/min or PCO2 <32 mmHg
• White blood cell count >12,000/μL, <4000/μL, or 10% bands
• APACHE II (≥8 at 24 h)
• Hemoconcentration (hematocrit >44%)
• Admission BUN (>20 mg/dL)
• BISAP score (≥3 present)
• (B) BUN >25 mg/dL
• (I) Impaired mental status
• (S) SIRS: ≥2 of 4 present
• (A) Age >60 years
• (P) Pleural effusion
• Organ failure (Modified Marshall score) (≥1 present):
• Cardiovascular: systolic BP <90 mmHg, heart rate >130 beats/min
• Pulmonary: PaO2 <60 mmHg
• Renal: serum creatinine >2.0 mg/dL
Markers of Severity during Hospitalization
• Elevated C-reactive protein (CRP) (variable thresholds >100 mg/L reported)
• Persistent organ failure (≥48 h)
• Pancreatic or extrapancreatic necrosis
PART 10
Disorders of the Gastrointestinal System
Abbreviations: APACHE II, Acute Physiology and Chronic Health Evaluation II;
BISAP, Bedside Index of Severity in Acute Pancreatitis; BMI, body mass index; BP,
blood pressure; BUN, blood urea nitrogen; SIRS, systemic inflammatory response
syndrome.
to therapy, noting that a less aggressive resuscitation strategy may be
sufficient in milder forms of pancreatitis. Importantly, a rising BUN
during hospitalization is associated not only with inadequate hydration
but also higher in-hospital mortality.
A decrease in hematocrit and BUN during the first 12–24 h is strong
evidence that an adequate volume of fluids is being administered.
Adjustments in fluid resuscitation may be required in patients with
cardiac, pulmonary, or renal disease.
Assessment of Severity and Hospital Triage 
Severity of acute
pancreatitis should be determined in the emergency ward to assist in
A
B
C
FIGURE 359-2  Imaging features of a pancreaticopleural fistula secondary to acute pancreatitis. A. Pancreaticopleural fistula: pancreatic duct leak on endoscopic
retrograde cholangiopancreatography. Pancreatic duct leak (arrow) demonstrated at the time of retrograde pancreatogram in a patient with exacerbation of alcoholinduced acute pancreatitis. B. Pancreaticopleural fistula: computed tomography (CT) scan. Contrast-enhanced CT scan (coronal view) with arrows showing fistula tract
from pancreatic duct disruption in the pancreatic pleural fistula. C. Pancreaticopleural fistula: chest x-ray. Large pleural effusion in the left hemithorax from a disrupted
pancreatic duct. Analysis of pleural fluid revealed elevated amylase concentration. (Courtesy of Dr. K.J. Mortele, Brigham and Women’s Hospital; with permission.)
patient triage to a regular hospital ward, step-down unit, or intensive
care unit. The Bedside Index of Severity in Acute Pancreatitis (BISAP)
incorporates five clinical and laboratory parameters obtained within
the first 24 h of hospitalization (Table 359-3)—BUN >25 mg/dL,
impaired mental status (Glasgow coma scale score <15), SIRS, age
60 years, and pleural effusion on radiography—that can be useful in
assessing severity. The presence of three or more of these factors was
associated with substantially increased risk for in-hospital mortal­
ity among patients with acute pancreatitis. In addition, an elevated
hematocrit and admission BUN are also associated with more severe
acute pancreatitis. Incorporating these indices with the overall patient
response to initial fluid resuscitation in the emergency ward can be
useful in triaging patients to the appropriate hospital acute care setting.
In general, patients with lower BISAP scores, hematocrit, and
admission BUN levels tend to respond to initial management and can
be safely triaged to a regular hospital ward for ongoing care. If SIRS is
not present at 24 h, the patient is unlikely to develop organ failure or
necrosis. Conversely, patients with persistent SIRS at 24 h or underly­
ing comorbid illnesses (e.g., chronic obstructive pulmonary disease,
congestive heart failure) should be considered for a step-down unit
setting if available. Patients with higher BISAP scores and elevations
in hematocrit and admission BUN who do not respond to initial fluid
resuscitation and exhibit evidence of respiratory failure, hypotension,
or organ failure should be considered for direct admission to an inten­
sive care unit.
Special Considerations Based on Etiology 
A careful history,
review of medications, selected laboratory studies (liver profile, serum
triglycerides, serum calcium), and abdominal ultrasound are recom­
mended in the emergency ward to assess for etiologies that may impact
acute management. An abdominal ultrasound is the initial imaging
modality of choice and will evaluate the gallbladder, common bile duct,
and pancreatic head.
GALLSTONE PANCREATITIS  Patients with evidence of ascending chol­
angitis (including sepsis with evidence of biliary obstruction on labo­
ratory or imaging studies) should undergo ERCP within 24–48 h of
admission. Patients with gallstone pancreatitis are at increased risk of
recurrence, and consideration should be given to performing a chole­
cystectomy during the same admission in mild acute pancreatitis. An
alternative for patients who are not surgical candidates would be to
perform an endoscopic biliary sphincterotomy before discharge.
HYPERTRIGLYCERIDEMIA  Serum triglycerides >1000 mg/dL are asso­
ciated with acute pancreatitis. Initial therapy should focus on the treat­
ment of hyperglycemia with intravenous insulin, which often corrects
hypertriglyceridemia. In contrast to the standard nutritional manage­
ment of acute pancreatitis, patients should remain fasting for the initial
24–36 h to promote the resolution of severe triglyceride elevations.
Adjunct therapies may also include heparin or plasmapheresis, but
there is no compelling evidence these measures improve clinical
outcomes. Outpatient therapies include control of diabetes if present,
administration of lipid-lowering agents, weight loss, and avoidance of
drugs that elevate lipid levels.
Other potential etiologies that may impact acute hospital care
include hypercalcemia and post-ERCP pancreatitis. Treatment of hyper­
parathyroidism or malignancy is effective at reducing serum calcium.
Pancreatic duct stenting and/or rectal indomethacin administration are
effective at decreasing the risk of pancreatitis after ERCP. Drugs that
cause pancreatitis should be discontinued. Multiple drugs have been
implicated, but only about 30 have been rechallenged (Class 1A) and
found to be causative.
Nutritional Therapy 
A low-fat solid diet can be administered to
subjects with mild acute pancreatitis once they are able to eat. Enteral
nutrition should be considered 2–3 days after admission in subjects
with more severe pancreatitis instead of total parenteral nutrition
(TPN). Enteral feeding maintains gut barrier integrity, limits bacterial
translocation, is less expensive, and has fewer complications than TPN.
While there may be physiologic advantages of jejunal feeding, gastric
feeding is safe, and the benefits of jejunal over gastric feeding remain
under investigation.
Management of Local Complications (Table 359-4) 
Patients
exhibiting signs of clinical deterioration despite aggressive fluid resus­
citation and hemodynamic monitoring should be assessed for local
complications, including necrosis, pseudocyst formation, pancreas
duct disruption, peripancreatic vascular complications, and extrapan­
creatic infections. A multidisciplinary team approach is recommended,
including gastroenterology, surgery, interventional radiology, and
intensive care specialists; consideration should be made for transfer
to a tertiary pancreas center of excellence if these services are not
available.
NECROSIS  The management of necrosis requires a multidisciplinary
team approach. Percutaneous fine-needle aspiration of necrosis with
Gram stain and culture was previously performed to evaluate for
infected pancreatic necrosis in those with sustained leukocytosis, fever,
or organ failure. However, the current use of this technique varies
depending on institutional preference, with many abandoning this
diagnostic test to avoid potentially contaminating an otherwise sterile
collection, particularly when culture results will not lead to a clinical
decision to de-escalate antimicrobial therapy. Even though there is
currently no role for prophylactic antibiotics in necrotizing pancreatitis,
empiric antibiotics should be considered in those with clinical decom­
pensation. Prophylactic antibiotics do not lead to improved survival
and may promote the development of opportunistic fungal infections.
Repeated CT or MRI imaging should also be considered with any
change in clinical course to monitor for complications (e.g., thrombo­
ses, hemorrhage, abdominal compartment syndrome).
In general, sterile necrosis is most often managed conservatively
unless complications arise. Once a diagnosis of infected necrosis is
established and an organism identified, targeted antibiotics should be
instituted. Pancreatic drainage and/or debridement (necrosectomy)
should be considered for definitive management of infected necrosis,
but management decisions are ultimately influenced by the clinical
response since almost two-thirds of patients respond to antibiotic
treatment with or without drainage. Symptomatic local complications
as outlined in the Revised Atlanta Criteria typically require definitive
therapy.
A step-up approach (percutaneous or endoscopic transgastric/
transduodenal drainage followed, if necessary, by endoscopic or, rarely,
surgical necrosectomy) is the current treatment paradigm. Random­
ized trials have reported advantages to an initial endoscopic approach
compared to an initial surgical necrosectomy approach in patients
requiring intervention for symptomatic WON. Taken together, a more
conservative approach to the management of infected pancreatic
necrosis has evolved under the close supervision of a multidisci­
plinary team. If conservative therapy can be safely implemented, it is
TABLE 359-4  Complications of Acute Pancreatitis
Local
Pancreatic/peripancreatic fluid collections (Table 359-2):
  Acute necrotic collection (sterile or infected)
  Walled-off necrosis (sterile or infected)
  Pancreatic pseudocyst
Disruption of main pancreatic duct or secondary branches
Pancreatic ascites
Chylous ascites (secondary to disruption of lymphatic ducts)
Involvement of contiguous organs by necrotizing pancreatitis (e.g., colon
perforation)
Splanchnic thromboses (splenic vein, superior mesenteric vein, and/or portal
vein)
Gastric outlet obstruction
Biliary obstruction (jaundice)
Systemic
Pulmonary
  Pleural effusion
  Atelectasis
  Mediastinal fluid
  Pneumonitis
  Acute respiratory distress syndrome
Cardiovascular
  Hypotension
  Hypovolemia
  Nonspecific ST-T changes in electrocardiogram simulating myocardial
CHAPTER 359
infarction
  Pericardial effusion
Hematologic
  Disseminated intravascular coagulation
Gastrointestinal hemorrhage
  Peptic ulcer disease
  Erosive gastritis
  Hemorrhagic pancreatic necrosis with erosion into major blood vessels
  Variceal hemorrhage secondary to splanchnic thrombosis
Renal
  Oliguria (<300 mL/d)
  Azotemia
  Renal artery and/or renal vein thrombosis
  Acute tubular necrosis
Metabolic
  Hyperglycemia
  Hypertriglyceridemia
  Hypocalcemia
  Encephalopathy
  Sudden blindness (Purtscher’s retinopathy)
Central nervous system
  Psychosis
  Fat emboli
Fat necrosis
  Subcutaneous tissues (erythematous nodules)
  Bone
  Miscellaneous (mediastinum, pleura, nervous system)
Acute and Chronic Pancreatitis
recommended to do so for 4–6 weeks to allow the pancreatic collec­
tions to either resolve or evolve to develop a more organized boundary
(i.e., to “wall off”) since interventions are generally safer and more
technically straightforward.
PSEUDOCYST  The incidence of pseudocyst is low, and most acute
collections resolve over time. Less than 10% of patients have persis­
tent fluid collections after 4 weeks that would meet the definition of
a pseudocyst. Only symptomatic collections require intervention with
endoscopic or surgical drainage.
PANCREATIC DUCT DISRUPTION  Pancreatic duct disruption may
present with symptoms of increasing abdominal pain or shortness of
breath in the setting of an enlarging fluid collection resulting in pan­
creatic ascites (ascitic fluid has high amylase level). Diagnosis can be
confirmed on magnetic resonance cholangiopancreatography (MRCP)
or ERCP. Placement of a bridging pancreatic stent for at least 6 weeks
is >90% effective at resolving the leak with or without parenteral nutri­
tion and octreotide. Nonbridging stents are less effective (25–50%) but
may be useful in combination with parenteral nutrition and octreotide
prior to surgical evaluation.
PERIVASCULAR COMPLICATIONS  Perivascular complications may
include splenic vein thrombosis with gastric varices and pseudoaneu­
rysms, as well as portal and superior mesenteric vein thromboses. Gastric
varices rarely bleed but can be life-threatening. Similarly, life-threaten­
ing bleeding from a ruptured pseudoaneurysm can be diagnosed and
treated with mesenteric angiography and embolization.
EXTRAPANCREATIC INFECTIONS  Hospital-acquired infections occur
in up to 20% of patients with acute pancreatitis. Patients should be con­
tinually monitored for the development of pneumonia, urinary tract
infection, and line infection. Continued culturing of urine, monitoring
of chest x-rays, and routine changing of intravenous lines are important
during hospitalization.
Follow-Up Care 
Hospitalizations for moderately severe and severe
acute pancreatitis can be prolonged and last weeks to months and often
involve periods of intensive care unit admission and outpatient reha­
bilitation or subacute nursing care. Follow-up evaluation should assess
for development of diabetes, exocrine pancreatic insufficiency, recur­
rent cholangitis, or infected fluid collections. As mentioned previously,
cholecystectomy should be performed during the initial hospitalization
for acute gallstone pancreatitis with mild clinical severity. For patients
with necrotizing gallstone pancreatitis, the timing of cholecystectomy
needs to be individualized.
PART 10
Disorders of the Gastrointestinal System
■
■RECURRENT ACUTE PANCREATITIS
Approximately 25% of patients who have had an attack of acute pancre­
atitis will experience a recurrence. The two most common etiologic fac­
tors are alcohol and cholelithiasis. In patients with recurrent pancreatitis
without an obvious cause, the differential diagnosis should consider
occult biliary tract disease, including microlithiasis, hypertriglyceri­
demia, pancreatic cancer, and hereditary pancreatitis (Table 359-1). In
one small series, up to two-thirds of patients with recurrent acute pan­
creatitis without an initially apparent cause had occult gallstone disease
attributed to microlithiasis. Genetic defects as in hereditary pancreatitis
and cystic fibrosis mutations can result in recurrent pancreatitis. Other
diseases of the biliary tree and pancreatic ducts that can cause acute
pancreatitis include choledochocele; ampullary tumors; pancreas divi­
sum; and pancreatic duct stones, stricture, and tumors. Approximately
2–4% of patients with pancreatic cancer present with acute pancreatitis.
■
■PANCREATITIS IN PATIENTS WITH AIDS
The incidence of acute pancreatitis is theoretically increased in
patients with AIDS for two reasons: (1) the high incidence of infec­
tions involving the pancreas such as infections with cytomegalovirus,
Cryptosporidium, and the Mycobacterium avium complex; and (2)
the frequent use of medications such as pentamidine, trimethoprimsulfamethoxazole, and protease inhibitors. The incidence has been
markedly reduced due to advances in therapy, including the disuse of
didanosine (Chap. 208).
CHRONIC PANCREATITIS AND EXOCRINE
PANCREATIC INSUFFICIENCY
■
■PATHOPHYSIOLOGY
Chronic pancreatitis is a disease process characterized by irreversible
damage to the pancreas, in contrast to the reversible changes noted in
acute pancreatitis (Table 359-4). The events that initiate and then per­
petuate the inflammatory process in the pancreas are becoming more
clearly understood. Irrespective of the mechanism of injury, it is apparent
that stellate cell activation leads to cytokine expression and production
of extracellular matrix proteins that contribute to acute and chronic
inflammation and collagen deposition in the pancreas. This condition
is defined by the presence of histologic abnormalities, including chronic
inflammation, fibrosis, and progressive destruction (atrophy) of both
exocrine and endocrine tissue. A number of etiologies have been associ­
ated with chronic pancreatitis resulting in the cardinal manifestations
of the disease such as abdominal pain, steatorrhea, weight loss, diabetes
mellitus, and, less commonly, pancreatic cancer (Table 359-5).
Even in individuals in whom alcohol is believed to be the primary
cause of chronic pancreatitis, other factors are likely required for the
development and progression of disease, which explains why not all
heavy consumers of alcohol develop pancreatic disease. Importantly,
there is a strong association between smoking and chronic pancreatitis.
Cigarette smoke leads to an increased susceptibility to pancreatic auto­
digestion and predisposes to dysregulation of duct cell CFTR function.
Smoking is an independent, dose-dependent risk factor for chronic
pancreatitis and recurrent acute pancreatitis. Both continued alcohol
and smoking exposure are associated with disease progression, includ­
ing pancreatic fibrosis and calcifications.
Characterization of pancreatic stellate cells (PSCs) has added insight
into the underlying cellular responses behind development of chronic
pancreatitis. Specifically, PSCs are believed to play a role in maintaining
normal pancreatic architecture that shifts toward fibrogenesis in those
who develop chronic pancreatitis. It is believed that alcohol or additional
stimuli lead to matrix metalloproteinase–mediated destruction of nor­
mal collagen in pancreatic parenchyma, which later allows for pancreatic
TABLE 359-5  Classification of Chronic Pancreatitis: The TIGAR-O
System
Toxic-metabolic
Alcoholic
Tobacco smoking
Hypercalcemia
Hyperlipidemia (hypertriglyceridemia)
Chronic renal failure
Idiopathic
Early onset
Late onset
Tropical
Genetic
Cationic trypsinogen (PRSS1)
Cystic fibrosis transmembrane conductance regulator gene (CFTR)a
Calcium-sensing receptor (CASR)a
Chymotrypsin C gene (CTRC)a
Pancreatic secretory trypsin inhibitor gene (SPINK1)a
Autoimmune
Type 1 autoimmune pancreatitis (associated with IgG4-related disease)
Type 2 autoimmune pancreatitis (idiopathic duct-centric chronic pancreatitis)
Recurrent and severe acute pancreatitis
Postnecrotic (severe acute pancreatitis)
Recurrent acute pancreatitis
Vascular diseases/ischemia
Radiation induced
Obstructive
Pancreas divisuma
Duct obstruction (e.g., tumor)
Preampullary duodenal wall cysts
Posttraumatic pancreatic duct strictures
aThese conditions are believed to be disease modifiers that require additional
factors to cause chronic pancreatitis.
Abbreviations: TIGAR-O, toxic-metabolic, idiopathic, genetic, autoimmune, recurrent
and severe acute pancreatitis, obstructive.
remodeling. Proinflammatory cytokines, tumor necrosis factor α (TNF-α),
interleukin 1 (IL-1), and interleukin 6 (IL-6), as well as oxidant com­
plexes, can induce PSC activity with subsequent new collagen synthesis.
In addition to being stimulated by cytokines, oxidants, or growth factors,
PSCs also possess transforming growth factor β (TGF-β)–mediated selfactivating autocrine pathways that may explain disease progression in
chronic pancreatitis even after removal of noxious stimuli.
■
■ETIOLOGIC CONSIDERATIONS
Among adults in the United States, heavy alcohol use is the most com­
mon cause of clinically apparent chronic pancreatitis. As many as 25% of
adults in the United States with chronic pancreatitis have the idiopathic
form, including a subset of patients who do not develop clinical manifes­
tations until later in life (referred to as idiopathic late-onset chronic pan­
creatitis). Recent investigations have indicated that up to 15% of patients
with chronic pancreatitis previously classified as having idiopathic pan­
creatitis may have an underlying genetic predisposition (Table 359-5).
The prototypical genetic defect was identified in the cationic tryp­
sinogen gene (PRSS1) by studying several large families with chronic
pancreatitis. Additional pathogenic and nonpathogenic mutations
have been identified in this gene. The defect prevents the destruction
of prematurely activated trypsin and allows it to be resistant to the
intracellular protective effect of trypsin inhibition. It is hypothesized
that this continual activation of digestive enzymes within the gland
leads to acute injury and, finally, chronic pancreatitis. Since the initial
discovery of the PRSS1 mutation defect, other genetic disease modifiers
have been identified (Table 359-5).
The CFTR gene functions as a cyclic AMP–regulated chloride
channel. In patients with cystic fibrosis, the high concentration of
macromolecules can block the pancreatic ducts. It must be appreciated,
however, that there is a great deal of heterogeneity in relationship to
the CFTR gene defect. More than 1700 putative mutations of the CFTR
gene have been identified. The large number and different classes of
CFTR mutations have hampered attempts to elucidate the relationship
between the genotype and pancreatic manifestations. The ability to
detect CFTR mutations has led to the recognition that the clinical spec­
trum of the disease is broader than initially thought. Studies have clari­
fied the association between mutations of the CFTR gene and another
monosymptomatic form of cystic fibrosis (i.e., chronic pancreatitis). It
is estimated that in patients with idiopathic pancreatitis, the frequency
of a single CFTR mutation is 11 times the expected frequency and the
frequency of two mutant alleles is 80 times the expected frequency. In
these studies, patients were adults when the diagnosis of pancreatitis
was made; none had any clinical evidence of pulmonary disease, and
sweat test results were not diagnostic of cystic fibrosis. The prevalence
of such mutations is unclear, and further studies are needed. In addi­
tion, the therapeutic and prognostic implication of these findings with
respect to managing pancreatitis remains to be determined. CFTR
mutations are common in the general population, so it is unclear
whether the CFTR mutation alone can lead to pancreatitis as an auto­
somal recessive disease. A study evaluated 39 patients with idiopathic
chronic pancreatitis to assess the risk associated with these mutations.
Patients with two CFTR mutations (compound heterozygotes) dem­
onstrated CFTR function at a level between that seen in typical cystic
fibrosis and cystic fibrosis carriers and had a 40-fold increased risk
of pancreatitis. The presence of a separate genetic mutation (N34S
SPINK1) increased the risk 20-fold. A combination of two CFTR muta­
tions and an N34S SPINK1 mutation increased the risk of pancreatitis
900-fold. Knowledge of the genetic defects and downstream alterations
in protein expression has led to the development of novel therapies in
children with cystic fibrosis that potentiate the CFTR channel, result­
ing in improvement in lung function, quality of life, and weight gain.
Some studies have shown that use of CFTR modulators may reduce the
frequency of acute pancreatitis in heterozygous carriers. Table 359-5
lists other recognized causes of chronic pancreatitis.
■
■AUTOIMMUNE PANCREATITIS (TABLE 359-6)
Autoimmune pancreatitis (AIP) refers to a form of chronic pancre­
atitis with distinct histopathology and several unique differences in
TABLE 359-6  Comparison of the Autoimmune Pancreatitis (AIP)
Subtypes
 
TYPE 1 AIP
TYPE 2 AIP
Age at diagnosis, mean
Seventh decade
Fifth decade
Male sex
75%
50%
Serum IgG4 elevation
~66%
~25%
Other organ involvement
50%
Noa
Histologic findings:
 
 
  Lymphoplasmacytic
  Infiltration
++
++
  Periductal inflammation
++
++
  Storiform fibrosis
++
+
  Obliterative phlebitis
++
+
  Granulocytic epithelial
–
+++
lesion (GEL)
  IgG4 tissue staining
Abundant
(≥10 cells/hpf)
Scant
(<10 cells/hpf)
Response to steroids
~100%
~100%
Risk for relapse
Moderate to high
(20–60%)
Low (<10%)
Associated with IgG4-RD
Yes
No
aInflammatory bowel disease is seen in ~10–20% of patients with idiopathic ductcentric chronic pancreatitis but may also occur in type 1 AIP.
Abbreviations: hpf, high-power field; IgG4-RD, IgG4-related disease.
Source: Reproduced with permission from PA Hart: Reviews in basic and clinical
gastroenterology and hepatology. Gastroenterology 149:39, 2015.
CHAPTER 359
the clinical phenotype. Currently, two subtypes of AIP are recognized,
type 1 AIP and idiopathic duct-centric chronic pancreatitis (IDCP,
also referred to as type 2 AIP). Type 1 AIP is identified as the pancre­
atic manifestation of a multiorgan syndrome currently referred to as
IgG4-related disease (Chap. 380). The characteristic histopathologic
findings of type 1 AIP include lymphoplasmacytic infiltrate, storiform
fibrosis, and abundant IgG4 cells. IDCP is histologically defined by
the presence of granulocytic infiltration of the duct wall (termed a
granulocytic epithelial lesion [GEL]) but without IgG4-positive cells.
Type 1 AIP is often associated with involvement of other organs in
the setting of IgG4-related disease, including bilateral submandibular
gland enlargement, characteristic renal lesions, retroperitoneal fibrosis,
and stricturing of the extrapancreatic biliary tree. In contrast, IDCP is a
pancreas-specific disorder that is associated with inflammatory bowel
disease in ~10% of patients. AIP is not a common cause of idiopathic
recurrent acute pancreatitis.
Acute and Chronic Pancreatitis
Jaundice, weight loss, and new-onset diabetes are the most common
presenting symptoms. Elevated serum IgG4 levels are supportive of the
diagnosis (elevated in two-thirds of patients with type 1 AIP) but have
a low positive predictive value when used in isolation of other clinical
findings. CT imaging demonstrates abnormalities in the majority of
patients with either diffuse or focal enlargement during active disease,
unless the gland is atrophic due to previous disease (Fig. 359-3). The
presence of an inflammatory rim, termed a capsule sign, is highly
specific (but not sensitive) for AIP. ERCP or MRCP reveals strictures
in the bile duct in more than one-third of patients with AIP, including
some patients with isolated intrahepatic bile duct strictures (type 1
AIP only), which can mimic primary sclerosing cholangitis, and is
referred to as IgG4-related sclerosing cholangitis (previously termed
IgG4-associated cholangitis).
The Mayo Clinic HISORt criteria provide a helpful mnemonic to
remember the key diagnostic features of this disease, including (1)
histology; (2) imaging; (3) serology (elevated serum IgG4 levels); (4)
other organ involvement; and (5) response to glucocorticoid therapy.
These diagnostic criteria have been harmonized with those from other
countries to develop the International Consensus Diagnostic Criteria
for AIP, which are the most comprehensive criteria. Glucocorticoids
have shown efficacy in alleviating symptoms, decreasing the size of
PART 10
Disorders of the Gastrointestinal System
A
B
FIGURE 359-3  Imaging features of the pancreatic parenchyma in a patient with type 1 autoimmune pancreatitis on computed tomography (CT). A. Contrast-enhanced CT
scan of the abdomen demonstrates diffuse pancreatic enlargement and a hypoechoic rim (capsule sign) in a patient who presented with jaundice. The serum IgG4 level
was elevated to 942 mg/dL (reference range 4–86 mg/dL), so the patient was diagnosed with definitive type 1 autoimmune pancreatitis. B. Contrast-enhanced CT scan of the
abdomen following a treatment course with high-dose steroids demonstrates return to normal size of the pancreas, reappearance of normal lobulations along the margin,
and absence of the hypoechoic rim.
the pancreas, and reversing histopathologic features in patients with
AIP. Patients typically respond dramatically to glucocorticoid therapy
within a 2- to 4-week period. Prednisone is usually administered at
an initial dose of 40 mg/d for 4 weeks followed by a taper of the daily
dosage by 5 mg per week based on monitoring of clinical parameters.
Relief of symptoms, liver biochemistries, and abnormal imaging of the
pancreas and bile ducts are followed to assess for treatment response. A
poor response to glucocorticoids should raise suspicion of an alternate
diagnosis, such as pancreatic cancer. A prior multicenter international
study examined >1000 patients with AIP. Clinical remission was
achieved in 99% of type 1 AIP and 92% of type 2 AIP patients with
steroids. However, disease relapse occurred in 31 and 9% of patients
with type 1 and type 2 AIP, respectively. Patients with multiple relapses
have been managed with immunomodulators (e.g., azathioprine,
6-mercaptopurine, or mycophenolate mofetil) with variable success.
Management with B-cell depletion therapy (e.g., rituximab) is likely
the most effective treatment option for patients with recurrent relapses
with other therapies under investigation. The appearance of interval
cancers following a diagnosis of AIP is uncommon.
Clinical Features of Chronic Pancreatitis 
Patients with chronic
pancreatitis primarily seek medical attention due to abdominal pain or
symptoms of maldigestion. The abdominal pain may be variable in loca­
tion, severity, and frequency. The pain can be constant or intermittent
with pain-free intervals. Eating may exacerbate the pain, leading to a fear
of eating with consequent weight loss. The spectrum of abdominal pain
ranges from mild to quite severe, with narcotic dependence as a frequent
consequence. There is often a disparity between the reported severity
of abdominal pain and the physical findings, which primarily consist of
nonfocal abdominal tenderness. Patients with chronic abdominal pain
may or may not experience symptoms of maldigestion, such as chronic
diarrhea, steatorrhea, and/or weight loss. Fat-soluble vitamin deficien­
cies are increasingly recognized. Importantly, there is an exceedingly
high prevalence of metabolic bone disease in chronic pancreatitis, with
~65% of patients having either osteopenia or osteoporosis. Patients with
chronic pancreatitis have impaired quality of life and develop significant
morbidity, requiring frequent health encounters.
The diagnosis of early or mild chronic pancreatitis can be challeng­
ing because there is no accurate biomarker for the disease. In contrast
to acute pancreatitis, the serum amylase and lipase levels are usually not
strikingly elevated in chronic pancreatitis. Rather, low serum pancreatic
enzyme levels are moderately specific for a diagnosis of chronic pancre­
atitis but have poor sensitivity. Elevation of serum bilirubin and alkaline
phosphatase may indicate cholestasis secondary to common bile duct
stricture caused by chronic inflammation or fibrosis. The cumulative
prevalence of exocrine pancreatic insufficiency is >80%. The presence
of overt steatorrhea in a patient with chronic pancreatitis is highly sug­
gestive of this complication. However, in those with milder symptoms,
additional testing, such as a random fecal elastase-1 level (on a formed
stool specimen) may be needed to confirm the diagnosis of exocrine
pancreatic insufficiency. The radiographic evaluation of a patient with
suspected chronic pancreatitis usually proceeds from a noninvasive to
more invasive approach. Abdominal CT imaging (Fig. 359-4) is the
initial modality of choice, followed by MRI, endoscopic ultrasound, and
pancreas function testing. In addition to excluding a pseudocyst and
pancreatic cancer, CT imaging may show calcifications, dilated pancre­
atic or biliary ducts, or an atrophic pancreas. Although abdominal CT
scanning and MRI greatly aid in the diagnosis of pancreatic disease, the
diagnostic test with the best sensitivity is the direct pancreatic function
test using secretin. The secretin test becomes abnormal when ≥60% of
the pancreatic exocrine function has been lost and usually correlates
well with the onset of chronic abdominal pain. The role of endoscopic
ultrasonography (EUS) in diagnosing early chronic pancreatitis is still
evolving. A total of nine endosonographic features have been described
in chronic pancreatitis. The presence of five or more features is consid­
ered diagnostic of chronic pancreatitis. EUS is not a specific enough
test for detecting early chronic pancreatitis alone (Chap. 358) and may
show positive features in patients with diabetes, patients with a his­
tory of cigarette smoking, or even in normal aging individuals. Recent
data suggest that EUS can be combined with endoscopic pancreatic
function testing (EUS-ePFT) during a single endoscopy to screen for
chronic pancreatitis in patients with chronic abdominal pain. Diffuse
calcifications noted on plain film of the abdomen usually indicate sig­
nificant damage to the pancreas and are pathognomonic for chronic
pancreatitis. While patterns of calcification are not specific for an etiol­
ogy, patients with tropical pancreatitis (often associated with a SPINK1
mutation) characteristically have bulky calcifications.
Complications of Chronic Pancreatitis 
There are a number of
disease-related complications from chronic pancreatitis in addition to the
aforementioned abdominal pain and exocrine pancreatic insufficiency
(Table 359-7). The lifetime prevalence of chronic pancreatitis–related
diabetes exceeds 80%. Although most patients develop hyperglycemia
due to insulin deficiency caused by loss of islet cells, diabetic ketoaci­
dosis and diabetic coma are uncommon. Likewise, end-organ damage
(retinopathy, neuropathy, nephropathy) is also uncommon. Nondia­
betic retinopathy may be due to vitamin A and/or zinc deficiency.
Osteoporosis and osteopenia are increasingly recognized in chronic
pancreatitis and likely related to a combination of shared risk factors
(e.g., alcohol use, cigarette smoking), vitamin D deficiency, and detri­
mental effects on the bone from chronic inflammation. Gastrointesti­
nal bleeding may occur from peptic ulceration, gastritis, a pseudocyst
A
B
C
D
FIGURE 359-4  Distribution of imaging features of chronic pancreatitis on computed tomography (CT). Distinct features of chronic pancreatitis are seen on selected
images from contrast-enhanced CT scans of the abdomen from four unique patients, including the following. A. Numerous punctate calcifications involving the pancreatic
parenchyma in the head and body. B. A moderate-sized calculus visualized in the pancreatic duct with associated ductal dilation. C. Significant pancreatic duct dilation
and adjacent parenchymal atrophy secondary to a pancreatic duct stricture (which is not well seen on this scan). D. A large unilocular, encapsulated cyst in the tail of the
pancreas consistent with a pseudocyst from prior pancreatitis. Note adjacent pancreatic parenchymal atrophy.
eroding into the duodenum, arterial bleeding into the pancreatic duct
(hemosuccus pancreaticus), or ruptured varices secondary to splenic
vein thrombosis. Jaundice, cholestasis, and biliary cirrhosis may occur
from the chronic inflammatory reaction around the intrapancreatic
portion of the common bile duct. Twenty years after the diagnosis of
chronic calcific pancreatitis, the cumulative risk of pancreatic cancer
is 4%. Patients with hereditary PRSS1 or tropical pancreatitis have
an increased risk for pancreatic cancer compared to other forms of
chronic pancreatitis.
TREATMENT
Chronic Pancreatitis
There are currently no therapies to reverse or delay the disease pro­
gression of chronic pancreatitis, so management is primarily focused
on screening for and management of disease-related complications.
STEATORRHEA
The treatment of steatorrhea with pancreatic enzyme replacement
therapy is conceptually straightforward, yet complete correction of
steatorrhea is uncommon. Enzyme therapy usually brings diarrhea
under control and restores absorption of fat to an acceptable level
TABLE 359-7  Complications of Chronic Pancreatitis
Chronic abdominal pain
Exocrine pancreatic insufficiency
Diabetes mellitus
Splanchnic venous thrombosis
Metabolic bone disease (osteoporosis)
Biliary stricture and/or biliary cirrhosis
Pancreatic duct stricture
Pseudocyst
Pancreatic cancer
Malnutrition, micronutrient deficiencies
CHAPTER 359
Acute and Chronic Pancreatitis
and affects weight gain. Thus, pancreatic enzyme replacement is the
cornerstone of therapy. In treating steatorrhea, it is important to use
a potent pancreatic formulation that will deliver sufficient lipase into
the duodenum to correct maldigestion and decrease steatorrhea. For
adult patients with exocrine pancreatic insufficiency, it is generally
recommended to start at a dosage of 25,000–50,000 units of lipase
taken during each meal; however, the dose may need to be increased
up to 100,000 units of lipase depending on the response in symp­
toms, nutritional parameters, and/or pancreas function test results
(although rarely used for this purpose). Additionally, some may
require acid suppression with proton pump inhibitors to optimize the
response to pancreatic enzymes. Monitoring nutritional parameters
such as fat-soluble vitamins, zinc levels, body weight, and periodic
bone mineral density measurement should be considered.
ABDOMINAL PAIN
The management of pain in patients with chronic pancreatitis is
challenging due to the complex mechanisms of pancreatitis-related
pain. Recent meta-analyses have shown no consistent benefit of
enzyme therapy at reducing pain in chronic pancreatitis. Pain relief
experienced by patients treated with pancreatic enzymes may be
due to improvements in dyspepsia from maldigestion. One shortterm randomized trial showed that pregabalin could decrease pain
in chronic pancreatitis and lower pain medication requirement.
Other studies using antioxidants have yielded mixed results.
Endoscopic treatment of chronic pancreatitis pain may involve
sphincterotomy, pancreatic duct stenting, stone extraction, and
drainage of a pancreatic pseudocyst. Therapy directed to the pan­
creatic duct would seem to be most appropriate in the setting of a
dominant stricture, especially if there is an obstructing intraductal
stone. The use of endoscopic stenting for patients with chronic pain
has not been investigated in controlled trials. It is now appreci­
ated that significant complications can occur from stenting (e.g.,
stent migration, stent occlusion, and stent-induced pancreatic duct
strictures). Recent guidelines recommend considering celiac plexus
block for treatment of pain in chronic pancreatitis, but recommen­
dations were conditional with very low quality of evidence. Celiac
plexus block has not been rigorously studied for chronic pancreati­
tis and does not provide durable pain relief. It can provide relief in
some selected patients, but the a priori identification of those who
will respond is difficult. In patients with pancreatic duct dilation,
ductal decompression with surgical therapy has been the therapy of
choice. Among such patients, 80% seem to obtain immediate relief;
however, at the end of 3 years, one-half of the patients have recur­
rence of pain. Two randomized prospective trials comparing endo­
scopic to surgical therapy for chronic pancreatitis demonstrated
that surgical therapy was superior to endoscopy at decreasing pain
and improving quality of life in selected patients with dilated ducts
and abdominal pain. This would suggest that chronic pancreatitis
patients with dilated ducts and pain should be considered for surgi­
cal intervention. A recent randomized controlled trial suggests that
early surgical intervention may provide superior pain relief in the
short term compared to a conservative, nonoperative approach.
The role of preoperative stenting prior to surgery as a predictor of
response has yet to be proven.
Total pancreatectomy with or without autologous islet cell trans­
plantation has been used in highly selected patients with chronic
pancreatitis and abdominal pain refractory to conventional therapy.
However, some patients will continue to have pain postoperatively,
illustrating the complex nature of pain in this patient population.
Patients who benefit most from total pancreatectomy have a shorter
duration of symptoms and lower pain medication requirements.
The role of this procedure remains to be fully defined but may be an
option in lieu of ductal decompression surgery or partial pancreatic
resection in patients with intractable, painful, small-duct disease or
hereditary pancreatitis.
PART 10
Disorders of the Gastrointestinal System
■
■HEREDITARY PANCREATITIS
Hereditary pancreatitis (PRSS1) is a rare form of pancreatitis with early
age of onset that is typically associated with familial aggregation of
cases. A genome-wide search using genetic linkage analysis identified
the hereditary pancreatitis gene on chromosome 7. Mutations in ion
codons 29 (exon 2) and 122 (exon 3) of the cationic trypsinogen gene
(PRSS1) cause an autosomal dominant form of pancreatitis. The codon
122 mutations lead to a substitution of the corresponding arginine
with another amino acid, usually histidine. This substitution, when it
occurs, eliminates a fail-safe trypsin self-destruction site necessary to
eliminate trypsin that is prematurely activated within the acinar cell.
These patients have recurring episodes of acute pancreatitis. Patients
frequently develop pancreatic calcification, diabetes mellitus, and ste­
atorrhea; in addition, they have an increased incidence of pancreatic
cancer with a cumulative incidence of ~10%. A previous natural history
study of hereditary pancreatitis in >200 patients from France reported
that abdominal pain started in childhood at age 10 years, steatorrhea
developed at age 29 years, diabetes at age 38 years, and pancreatic can­
cer at age 55 years. Abdominal complaints in relatives of patients with
hereditary pancreatitis should raise the question of pancreatic disease.
■
■PANCREATIC ENDOCRINE TUMORS
Pancreatic endocrine tumors are discussed in Chap. 89.
OTHER CONDITIONS
■
■ANNULAR PANCREAS
When the ventral pancreatic anlage fails to migrate correctly to make
contact with the dorsal anlage, the result may be a ring of pancreatic
tissue encircling the duodenum. Such an annular pancreas may cause
intestinal obstruction in the neonate or adult. Symptoms of postpran­
dial fullness, epigastric pain, nausea, and vomiting may be present for
years before the diagnosis is entertained. The radiographic findings
are symmetric dilation of the proximal duodenum with bulging of the
recesses on either side of the annular band, effacement but not destruc­
tion of the duodenal mucosa, accentuation of the findings in the right
anterior oblique position, and lack of change on repeated examina­
tions. The differential diagnosis should include duodenal webs, tumors
of the pancreas or duodenum, duodenal ulcer, regional enteritis, and
adhesions. Patients with annular pancreas have an increased incidence
of pancreatitis and peptic ulcer. Because of these and other potential
complications, the treatment for refractory symptoms is surgical.
Retrocolic duodenojejunostomy is the procedure of choice, although
some surgeons advocate Billroth II gastrectomy, gastroenterostomy,
and vagotomy.
■
■PANCREAS DIVISUM
Pancreas divisum is present in 7–10% of the population and occurs
when the embryologic ventral and dorsal pancreatic anlagen fail to
fuse, so that pancreatic drainage is accomplished mainly through
the accessory minor papilla. Pancreas divisum is the most common
congenital anatomic variant of the human pancreas. Current evidence
indicates that this anomaly does not predispose to the development of
pancreatitis in the majority of patients. However, the combination of
pancreas divisum and a small accessory orifice can result in dorsal duct
obstruction. The challenge is to identify this subset of patients with
dorsal duct pathology. Cannulation of the dorsal duct by ERCP is tech­
nically challenging and associated with a very high risk of post-ERCP
pancreatitis, so patients with pancreatitis and pancreas divisum should
likely be treated with conservative measures. In many of these patients,
pancreatitis is idiopathic and unrelated to the pancreas divisum. Endo­
scopic or surgical intervention is indicated if pancreatitis recurs and no
other cause can be found. It should be stressed that the ERCP/MRCP
appearance of pancreas divisum (i.e., a small-caliber ventral duct with
an arborizing pattern) may be mistaken as representing an obstructed
main pancreatic duct secondary to a mass lesion.
■
■MACROAMYLASEMIA
In macroamylasemia, amylase circulates in the blood in a polymer
form too large to be easily excreted by the kidney. Patients with this
condition demonstrate an elevated serum amylase value and a low uri­
nary amylase value. The presence of macroamylase can be documented
by chromatography of the serum. The prevalence of macroamylasemia
is 1.5% of the nonalcoholic general adult hospital population. Usu­
ally, macroamylasemia is an incidental finding and is not related to
disease of the pancreas or other organs. Macrolipasemia has now been
documented in patients with cirrhosis or non-Hodgkin’s lymphoma.
In these patients, the pancreas appeared normal on ultrasound and CT
examination. Lipase was shown to be complexed with immunoglobulin
A. Thus, the possibility of both macroamylasemia and macrolipasemia
should be considered in patients with elevated blood levels of these
enzymes.
Acknowledgment
This chapter represents a revised version of chapters by Drs. Norton J.
Greenberger (deceased), Phillip P. Toskes (deceased), Peter A. Banks, and
Bechien Wu that were in previous editions of Harrison’s.
■
■FURTHER READING
Crockett SD et al: American Gastroenterological Association Insti­
tute guideline on initial management of acute pancreatitis. Gastroen­
terology 154:1096, 2018.
De-Maderia E et al: Aggressive or moderate fluid resuscitation in
acute pancreatitis. N Engl J Med 387:989, 2022.
Forsmark CE et al: Acute pancreatitis. N Engl J Med 375:1972, 2016.
Gardner TB et al: ACG clinical guideline: Chronic pancreatitis. Am J
Gastroenterol 115:322, 2020.
Hart PA et al: Recent advances in autoimmune pancreatitis. Gastro­
enterology 149:39, 2015.
Petrov MS, Yadav D: Global epidemiology and holistic prevention of
pancreatitis. Nat Rev Gastroenterol Hepatol 16:175, 2019.
Vege SS, Chari ST: Chronic pancreatitis. N Engl J Med 386:869, 2022.