15.24.2 Nonalcoholic fatty liver disease 3147

15.24.2 Nonalcoholic fatty liver disease 3147

15.24.2  Nonalcoholic fatty liver disease 3147 related to the stage of disease, as assessed by standard chronic liver disease scores such as the MELD score or the Child–​Pugh score. The prognosis is improved by sustained abstinence. Liver transplantation Liver transplantation should be considered for those patients whose clinical condition remains poor despite sustained abstin- ence. Although there is no requirement for a set period of abstin- ence before considering liver transplantation, many patients will improve clinically up to 6  months after stopping drinking, and such improvement might render referral for transplant unneces- sary. Those patients who are assessed for transplant require a rigorous psychiatric evaluation. Despite this, around 10% of pa- tients transplanted for ALD will return to problem drinking. In the long term, recidivism is associated with an increased mortality and may also adversely affect the attitude of the public to this form of treatment in ALD. Liver transplantation for alcoholic hepatitis has been suggested for a highly selected group of patients who have not responded to corticosteroids (‘Lille nonresponders’). This is a controversial area, with concerns raised about the appropriate use of donor organs and reservations about the specificity of Lille nonresponse to predict mortality. Such nonresponders may still have a 50% survival without transplantation. FURTHER READING European Association for the Study of Liver (2012). EASL clinical prac- tical guidelines: management of alcoholic liver disease. J Hepatol,
57, 399–​420. Forrest EH, et al. (2005). Analysis of factors related to mortality in al- coholic hepatitis and the derivation and validation of the Glasgow alcoholic hepatitis score. Gut, 54, 1174–​9. Hodgson R, et al. (2002). The FAST alcohol screening test. Alcohol Alcohol, 37, 61–​6. Lafferty H, Stanley AJ, Forrest EH (2013). The management of al- coholic hepatitis:  a prospective comparison of scoring systems. Aliment Pharmacol Ther, 38, 603–​10. Louvet A, et al. (2007). The Lille model: a new tool for therapeutic strategy in patients with severe alcoholic hepatitis treated with ster- oids. Hepatology, 45, 1348–​54. Lucey MR, Mathurin P, Morgan TR (2009). Alcoholic hepatitis.
N Engl J Med, 360, 2758–​69. Mathurin P, et al. (2011). Early liver transplantation for severe alco- holic hepatitis. N Engl J Med, 365, 1790–​800. Mathurin P, et al. (2013). Prednisolone with vs without pentoxifylline and survival of patients with severe alcoholic hepatitis: a random- ized clinical trial. JAMA, 310, 1033–​41. National Institute for Health and Care Excellence (NICE) (2010). Alcohol-​use disorders: diagnosis and clinical management of alcohol-​ related physical complications. Clinical guideline. NICE, London. Nguyen-​Khac E, et al. (2011). Glucocorticoids plus N-​acetylcysteine in severe alcoholic hepatitis. N Engl J Med, 365, 1781–​9. Seth D, et  al. (2011). Pathogenesis of alcohol-​induced liver dis- ease: classical concepts and recent advances. J Gastroenterol Hepatol, 26, 1089–​105. Thursz MR, et  al. (2015). Steroids or pentoxifylline for alcoholic hepatitis. N Engl J Med, 372, 1619–​28. 15.24.2  Nonalcoholic fatty
liver disease Quentin M. Anstee and Christopher P. Day ESSENTIALS Nonalcoholic fatty liver disease (NAFLD) is the most common liver disorder in the developed world, affecting 20 to 30% of Western adults. NAFLD occurs with a range of severity from simple steatosis (nonalcoholic fatty liver, NAFL) through nonalcoholic steatohepatitis (NASH) to fibrosis and, ultimately, cirrhosis. The condition is a mani- festation of the metabolic syndrome and so is strongly associated with obesity, insulin resistance, and dyslipidaemia. Dietary and gen- etic factors appear to determine susceptibility to the disease and its progression. In most patients, the condition is discovered incidentally when ab- normal values of serum liver-​related liver tests are reported. The diag- nosis is usually one of exclusion. Liver biopsy is not always required but, in the absence of well-​validated noninvasive biomarkers, remains the only way to detect steatohepatitis and accurately stage fibrosis of intermediate severity. However, biopsy is not practical in all cases and so a staged approach to patient assessment and risk stratification is ad- vised. Treatment is directed at components of the metabolic syndrome to reduce cardiovascular disease and liver disease progression: weight loss through diet and exercise has been shown to ameliorate NAFLD. A range of novel pharmacological drug treatments are under evaluation. Introduction Excessive hepatic fat accumulation is a common radiological and histological finding. In the absence of excessive alcohol consump- tion this is termed nonalcoholic fatty liver disease (NAFLD), which describes a broad spectrum of histologically defined progressive liver disease (Fig. 15.24.2.1). NAFLD ranges from hepatic stea- tosis (nonalcoholic fatty liver, NAFL) where hepatocellular trigly- ceride accumulation is >5%, through inflammatory nonalcoholic steatohepatitis (NASH), to fibrosis or cirrhosis, and in some cases hepatocellular carcinoma. Critically, most patients with NAFLD do not progress beyond steatosis, but a substantial minority do progress to cirrhosis and end-​stage liver disease and so experience associated morbidity and mortality. The presence of NAFLD is also an inde- pendent risk factor for development of cardiovascular disease and stroke. The key challenge in the management of NAFLD is to iden- tify the patients at greatest risk of disease progression so that treat- ment may be targeted at those that will benefit most. Aetiology Hepatic steatosis and steatohepatitis may be attributable to a di- verse range of acquired exposures as well as numerous rare mono- genic inherited disorders (summarized in Table 15.24.2.1). The

section 15  Gastroenterological disorders 3148 Steatohepatitis (NASH) Cirrhosis Steatosis (NAFL) NAFLD Fat infiltration >5% with or without mild inflammation Steatosis + necroinflammation (ballooning, Mallory bodies, megamitochondria) Increasing fibrosis, eventually leading to cirrhosis Fig. 15.24.2.1  NAFLD—​a spectrum of liver disease. Table 15.24.2.1  Aetiology of steatosis and steatohepatitis Aetiological factor Predominant pattern of histological steatosis Acquired metabolic and/​or nutritional disorders Metabolic syndrome (obesity, insulin resistance/​ type 2 diabetes mellitus) Macrovesicular Starvation and cachexia Macrovesicular Protein malnutrition (Kwashiorkor, anorexia nervosa) Macrovesicular Dietary choline deficiency Macrovesicular Total parenteral nutrition Macrovesicular Drugs Amiodarone Macrovesicular Aspirin Microvesicular Chloroquine Macrovesicular Corticosteroids Macrovesicular Methotrexate Macrovesicular NSAIDs (naproxen, ibuprofen, ketoprofen) Microvesicular Nucleoside analogues used in HAART (zidovudine, didanosine, zalcitabine, fialuridine, etc.) Microvesicular Oestrogens Macrovesicular Tamoxifen Macrovesicular Tetracycline Microvesicular Valproic acid Microvesicular Toxins Amanita phalloides mushroom poisoning Bacillus cereus emetic toxin Microvesicular Carbon tetrachloride Ethanol Macrovesicular Petrochemicals Toxic shock syndrome Microvesicular Aetiological factor Predominant pattern of histological steatosis Genetic factors Abetalipoproteinaemia Macrovesicular Alpers’ disease Microvesicular Cholesterol ester storage disease Microvesicular Familial combined hyperlipidaemia Familial hypobetalipoproteinaemia Glycogen storage disease Inherited defects in fatty acid β-​oxidation Microvesicular Lecithin–​cholesterol acyltransferase deficiency Microvesicular Lipodystrophy Macrovesicular Lysosomal acid lipase deficiency (Wolman’s disease) Microvesicular Ornithine transcarbamylase deficiency Microvesicular Wilson’s disease Macrovesicular Infections Chronic hepatitis C (genotype 3) Macrovesicular Bacterial overgrowth following jejunoileal bypass Macrovesicular Others Acute fatty liver of pregnancy Microvesicular Coeliac disease HELLP syndrome Microvesicular Reye’s syndrome Microvesicular Metals HAART, highly active antiretroviral therapy; NSAIDs, nonsteroidal anti-​inflammatory drugs.

15.24.2  Nonalcoholic fatty liver disease 3149 most common aetiological factors underlying the development of NAFLD are central obesity, insulin resistance/​type 2 diabetes mellitus, hypertension, and dyslipidaemia, a group of chronic condi- tions indicative of increased cardiovascular risk that together com- prise the ‘metabolic syndrome’. NAFLD is also associated with conditions including polycystic ovary syndrome, obstructive sleep apnoea, and small-​bowel bac- terial overgrowth. Due to sedentary lifestyles and the increasing consumption of diets enriched in fats and carbohydrates, the meta- bolic syndrome is now endemic in many developed countries and so the incidence of NAFLD has risen rapidly to become the leading cause of chronic liver disease worldwide. Epidemiology The true worldwide prevalence of NAFLD is not known as estimates vary between the populations studied due to different ethnicities, dietary patterns, and the sensitivity of the modality used to detect disease. Overall, however, NAFLD is estimated to affect approxi- mately 20 to 30% of the population in Western countries and 5 to 18% in Asia, with about 1 in 10 NAFLD cases exhibiting features of NASH. Studies from the United States of America indicate that the frequency of steatosis varies significantly with ethnicity (45% in Hispanics, 33% in white people, and 24% in black people) and sex (42% white males vs 24% white females). Prevalence increases dramatically when populations with known metabolic syndrome risk factors are selected. Illustrating this, 91% of obese patients (body mass index (BMI) ≥30 kg/​m2), 67% of overweight (BMI 25–​30 kg/​m2), and 25% in normal weight individuals in an un- selected European population sample had NAFLD, and 40 to 70% of patients with type 2 diabetes mellitus also have NAFLD. Of concern, while the prevalence of most liver diseases is stable, the prevalence of NAFLD is increasing, placing a greater burden on healthcare resources. There is an important paradox: most individuals with features of the metabolic syndrome develop steatosis and so NAFLD is highly prevalent in the general population, but only a subgroup progress to advanced liver disease and experience liver-​related morbidity (Fig. 15.24.2.2). During a median 12.6-​year follow-​up period in a cohort of 619 NAFLD patients, an overall 33.2% risk of death or liver transplantation was observed, with liver-​related mortality being the third most common cause of death, after cardiovascular disease and extrahepatic malignancy. Challenging the dogma that ‘simple’ steatosis is a benign con- dition with no clinical sequelae, serial biopsy studies indicate that patients with either steatosis or NASH may exhibit progres- sive fibrosis, with approximately 40% showing increased fibrosis, 40% stable disease, and 20% disease regression over a 6-​ to 7-​year period. The presence and severity of hepatic fibrosis on liver bi- opsy (or as assessed by noninvasive testing) appears to be the single most important determinant of long-​term prognosis, with advanced fibrosis/​cirrhosis (fibrosis stages F3–​4) being predictive of liver-​related events, transplantation, and death in patients with NAFLD. Compared with cases without histological evidence of fi- brosis, the presence of early [F1–​2] fibrosis conferred an 11.2-​fold risk. This increased to an 85.8-​fold risk in patients with advanced [F3–​4] fibrosis/​cirrhosis. The average age of NASH patients is 40 to 50 years and for NASH cirrhosis is 50 to 60 years, but the emerging epidemic of childhood obesity means that NAFLD is present in increasing numbers of younger patients and so the age that patients develop significant liver disease is likely to fall. Pathogenesis/​pathology It is generally accepted that the initiating events in NAFLD are dependent on the development of obesity and insulin resistance. Together these lead to increased fatty acid content in the liver due to the combination of de novo lipogenesis and fatty acid import from Early fibrosis F1 F2 F3 F4 Advanced fibrosis & Cirrhosis Steatosis Steatosis + Lobular Inflammation NASH +/–Portal Inflammation Steatotic/Steatohepatitic phase Fibrotic phase Rates of Fibrosis Progression vary between individuals Environmental Factors Diet, Microbiome, Xenobiotics Genetic & Epigenetic Factors Genetic variants in PNPLA3, TM6SF2, MBOAT7, HSD17B13 DNA methylation Risk of Death or Transplantation Fig. 15.24.2.2  Natural history of NAFLD.

section 15  Gastroenterological disorders 3150 dietary sources and adipose tissue stores. This places hepatocytes under considerable metabolic pressure, promotes lipotoxicity, in- creases oxidative stress secondary to free radical production during β-​ and ω-​fatty acid oxidation, and induces endoplasmic reticulum stress. Hepatocellular triglyceride accumulation (i.e. steatosis) is the histologically visible evidence of these metabolic stressors but is it- self unlikely to be directly harmful to the liver, being an adaptive response through which potentially lipotoxic fatty acids are parti- tioned into relatively inert intracellular stores. Ultimately, these in- sults combine with the additive effects of endotoxin-​initiated Kupffer cell cytokine release and immune-​mediated hepatocellular injury to induce cellular damage and activate cell death pathways, marking the transition to steatohepatitis. If these processes persist, stellate cell activation, collagen deposition, and hepatic fibrosis occur. Interpatient variation in disease progression and prognosis is thought to be determined by the combined effects of environ- mental (dietary and intestinal flora) exposures acting on a suscep- tible polygenic background. Studies implicate single nucleotide polymorphisms in a number of genes including PNPLA3, TM6SF2, MBOAT7 and HSD17B13 as factors that modify development of NAFLD and subsequent progression of fibrosis. Clinical features NAFLD is frequently asymptomatic although may be associated with nonspecific symptoms such as fatigue and mild right upper quadrant abdominal discomfort due to steatotic hepatomegaly distending the liver capsule. It is most commonly identified as an incidental radiological finding or a mild biochemical abnormality noted during routine blood tests taken for another indication. Alternatively, patients with progressive NAFLD may present late in the course of disease with complications of cirrhosis and portal hypertension such as variceal haemorrhage, or hepatocellular car- cinoma (HCC). HCC is a frequent complication of NAFLD (cu- mulative incidence 2.4–​12.8%) that may occur in both precirrhotic and postcirrhotic patients. Recognized independent risk factors for NAFLD progres- sion and advanced fibrosis include age greater than 45  years, presence of diabetes (or severity of insulin resistance), obesity (BMI >30  kg/​m2), and hypertension (Table 15.24.2.2). These factors are clinically useful as they assist with identification of ‘high-​risk’ patient groups. There are no specific physical signs to establish a diagnosis of NAFLD. Hepatomegally may be noted, but abdominal adiposity may hamper effective examination. In the absence of advanced disease, where the classical stigmata associated with the presence of chronic liver disease such as jaundice, spider naevi, and ascites may be apparent, clinical examination is frequently unremarkable. In light of the strong association between NAFLD and the meta- bolic syndrome, height and weight should be recorded, hip/​waist circumference measured, and evidence of end-​organ damage due to insulin resistance and hypertension sought. Differential diagnosis Key to the diagnosis of NAFLD is recognition that an individual possesses features of metabolic syndrome (central obesity, insulin resistance/​type 2 diabetes mellitus, dyslipidaemia, and hyperten- sion) and therefore is at risk of NAFLD. The more features of the metabolic syndrome that are present, the higher the probability that a patient has underlying NAFLD and the more likely they are to ex- hibit steatohepatitis and progressive liver disease. There are many causes for steatosis and steatohepatitis and hence a potentially wide differential diagnosis (Table 15.24.2.1). In practice, the principal differential is between metabolic syndrome-​ related NAFLD and alcoholic liver disease. Discriminating these is reliant upon a detailed history and seeking corroboration from family members (where available) to ensure that any history of concealed excessive alcohol consumption is excluded. An arbi- trary threshold for ethanol consumption of less than 20 g/​day for women and less than 30 g/​day for men is adopted to sustain a diag- nosis of NAFLD. Metabolic syndrome-​related NAFLD sensitizes the liver to the effects of other injurious processes such as alcohol consump- tion. As obesity and the wider metabolic syndrome become en- demic within the population, the distinction between NAFLD and alcohol-​related liver disease can easily become blurred with these apparently mutually exclusive conditions coexisting in some indi- viduals. To avoid this diagnostic oxymoron, the condition may be described as ‘dual-​aetiology fatty liver disease’. Table 15.24.2.2  Common risk factors for NAFLD Risk factor Effect Age Higher risk of NAFLD and advanced fibrosis aged >45 Metabolic syndrome (obesity, insulin resistance/​type 2 diabetes mellitus, dyslipidaemia, and hypertension) 70–​90% of metabolic syndrome patients have NAFLD. The more features an individual possesses, the greater the likelihood of NASH and severity of fibrosis Sex Males >females Ethnicity High risk in Hispanics, white individuals intermediate, lower risk in black individuals Dietary factors Diets high in cholesterol, saturated fats, and fructose but low in carbohydrate increase risk Caffeine may be protective Obstructive sleep apnoea Increased risk of hepatic fibrosis Genetic factors Variants in Patatin-​like phospholipase domain-​containing 3 (PNPLA3) and transmembrane 6 superfamily member 2 (TM6SF2) genes are associated with greater steatosis, NASH, and more advanced liver fibrosis. PNPLA3 mutations also increase the risk of NAFLD-​associated hepatocellular carcinoma

15.24.2  Nonalcoholic fatty liver disease 3151 A detailed drug history including prescribed medication, over-​ the-​counter medication, and traditional/​herbal remedies should be taken. Clinical investigation Once excessive alcohol consumption has been excluded and other confounding environmental factors discounted, investi- gation of suspected NAFLD should be directed first towards ex- clusion of other liver diseases and then confirming the presence of NAFLD: • Exclusion of other liver diseases:  laboratory testing should confirm absence of chronic viral hepatitis (hepatitis B virus sur- face antigen and hepatitis C virus serology), autoimmune liver disease (antinuclear antibodies, antimitochondrial antibodies, smooth muscle antibodies, liver–​kidney microsomal type 1 anti- bodies, immunoglobulins) and other treatable metabolic dis- eases (haemochromatosis, Wilson’s disease, coeliac disease, α1 antitrypsin deficiency). • Confirmation of the presence of NAFLD:  may be obtained radiologically or histologically. • Assessment of underlying liver damage: discriminating steatosis from NASH and determining the extent of hepatic fibrosis pre- sent (i.e. grade and stage of disease) is important to assist with individual risk stratification and prognostication. This has trad- itionally necessitated liver biopsy for histological assessment, but techniques for noninvasive testing to detect significant liver fi- brosis are becoming more robust. Biochemical tests The diagnosis, staging, and long-​term follow-​up of NAFLD is made more complicated as there is no single blood test that is diagnostic for NAFLD or which can be used to accurately determine disease severity and treatment response. Aminotransferases Elevations of serum alanine transaminase (ALT) and aspartate transaminase (AST) are modest and usually less than twice the upper limit of normal. Although many NAFLD cases are first de- tected due to the incidental finding of a mild transaminitis, these routine biochemical tests are insensitive with approximately 80% of patients having normal-​range ALT levels. Indeed, ALT levels fall as hepatic fibrosis progresses and so the characteristic AST/ALT ratio of less than 1 commonly seen in NAFLD reverses (AST/ALT

1. as fibrosis progresses toward cirrhosis, meaning that fibrosing steatohepatitis and/​or advanced liver disease may be present even in those with normal-​range ALT levels. Other biochemical markers Other laboratory abnormalities that may be present include non­ specific elevations of γ-​glutamyl transferase, low-​titre antinuclear antibodies, and/​or antismooth muscle antibody in 20 to 30% of patients, and elevated ferritin levels despite normal transferrin sat- uration. These can lead to diagnostic uncertainty as they may be incorrectly ascribed to high alcohol consumption, autoimmune hepatitis, and haemochromatosis respectively. Raised ferritin levels and the presence of increased IgA levels are indicative but not diag- nostic of more advanced fibrosis. Radiological studies Steatosis can readily be detected by various imaging modalities. Ultrasonography is widely available and cost-​effective and is there- fore most often used. A steatotic liver appears ‘bright’ due to in- creased echogenicity, but this finding is subjective and provides only a qualitative assessment of hepatic fat content. Sensitivity of ultrasonography is poor when less than 33% of hepatocytes are steatotic and so significant steatosis can be missed. Alternatives in- clude CT, MRI-derived Proton Density Fat Fraction (MRI-PDFF), or magnetic resonance spectroscopy, which offer greater sensitivity for detecting lesser degrees of steatosis but are resource inten- sive and not widely used in routine practice. Currently no widely available clinical imaging modality can distinguish steatosis from steatohepatitis or accurately quantify intermediate stages of fibrosis short of cirrhosis. Assessing the severity of the underlying liver disease In the absence of routinely available ‘standard’ tests to con- firm the diagnosis, assess the degree of inflammation, and accurately determine the extent of liver fibrosis to assist prog- nostication, liver biopsy is commonly used and remains the ‘gold standard’ investigation. However, it is invasive, carries a modest but appreciable risk of complications, and is unsuitable for widespread use outside the secondary care setting. Several noninvasive techniques, both commercial and noncommercial, have been proposed that may be used to predict the presence of NAFLD or the stage of fibrosis. The performance of selected tests that are commonly used are summarized in Table 15.24.2.3. Of these, simple scores like the NAFLD Fibrosis Score and FIB-​4 Score, or specialist panels such as the Enhanced Liver Fibrosis (ELF) Test, offer the ability to rule out significant underlying liver disease (fibrosis stages F3–​4) with a high degree of confi- dence and so are widely used to assist risk stratification, enabling invasive testing to be reserved for those where it may offer clin- ically relevant additional information. All three tests have also been shown to be predictive of long-​term disease outcome and mortality. Techniques that measure liver stiffness as a surrogate for fi- brosis severity, such as ultrasonography-​based transient elasto­ graphy (Fibroscan), are also widely adopted, although adiposity can interfere with accuracy in some patients. Promising alter- natives that are emerging include acoustic radiation force im- pulse imaging and magnetic resonance elastography. In general, noninvasive techniques have good negative predictive values for advanced fibrosis but more modest specificity and so lower positive predictive values. A suggested algorithm for the assess- ment and risk stratification of patients with NAFLD is provided in Fig. 15.24.2.3. Histology NAFLD ranges from hepatic steatosis, through inflammatory steatohepatitis, to fibrosis or cirrhosis (Fig. 15.24.2.1). The histo- logical classification of NAFLD, which allows grading of disease

section 15  Gastroenterological disorders 3152 activity and staging of fibrosis progression, is based on the assess- ment of three domains: • Degree of steatosis: this usually has a mainly centrilobular, acinar zone 3, distribution. Steatosis levels fall as cirrhosis develops and so NAFLD is underdiagnosed in the setting of advanced liver disease. It is thought to be the underlying cause of 30 to 75% of cases of ‘cryptogenic cirrhosis’ where no clear aetiology is apparent. • Grade of steatohepatitis (i.e. hepatocellular injury and inflam- mation): specific features of hepatocellular injury and inflamma- tion include hepatocyte ballooning degeneration with or without acidophil bodies or spotty necrosis and a mild, mixed inflam- matory infiltrate. These may be accompanied by Mallory–​Denk bodies. • Stage of fibrosis, scored from F0 (none) to F4 (cirrhosis): perisinusoidal fibrosis is a characteristic feature of NASH. To improve consistency between histopathologists, either of two well-​validated and widely used semiquantitative histological scoring systems may be used. These are the NIDDK NAFLD Activity Score (NAS) and the FLIP Steatosis/​Activity/​Fibrosis Score (SAF) (Table 15.24.2.4). Both are robust, but the SAF score may have some advantages as it specifically reports degree of steatosis and grade of activity separately, while these are combined as a single value in the NAS. Management Management of patients with NAFLD depends largely on the stage of disease and is reliant on careful risk stratification. There are at pre- sent no licensed prescription medications that are of proven efficacy for the treatment of NAFLD. Despite this, there are well-​defined evidenced-​based therapeutic interventions that reduce morbidity and mortality. Weight reduction and lifestyle modification All patients with NAFLD require advice about lifestyle modification aimed at weight loss and increased physical activity. A sustained loss of 7 to 9% of body weight reduces steatosis, hepatocellular in- jury, and inflammation. Weight loss of >10% may lead to regres- sion of fibrosis, if sustained. To achieve this, a calorie-​restricted diet (600 kcal/​day less than required to maintain body weight) should be recommended, targeting a weight loss of 0.5 kg per week, and patients should be advised to increase physical activity, performing at least 30 min of moderate exercise five times per week. Selected patients may benefit from pharmacological weight-​reduction treat- ments (e.g. orlistat) or bariatric surgery. Management of cardiovascular risk and the metabolic syndrome The presence of constituents of the metabolic syndrome should be sought and aggressively treated to reduce cardiovascular risk. The use of statins is not contraindicated in patients with NAFLD. Liver-​directed pharmacotherapy for high-​risk patients Although no pharmacological therapies are licensed for the treat- ment of NAFLD, there is evidence to support the use of specific classes of drugs to treat diabetes and hypertension that may offer some additional liver specific benefits. There are some preliminary evidence that angiotensin-​converting-​ enzyme inhibitors and angiotensin II receptor blockers may slow progression of liver fibrosis. Metformin does not appear to be an effective treatment for NAFLD but may reduce development of NAFLD-​associated hepatocellular carcinoma, with a 7% per annum reduction in hepatocellular car- cinoma risk reported in one retrospective study. Pioglitazone im- proves histological NASH, with a recent meta-​analysis suggesting that it may have some additional antifibrotic effects. Weight gain is a well-​recognized adverse effect of this treatment and long-​term use Table 15.24.2.3  Noninvasive tests for NAFLD and NAFLD fibrosis Test Formula Sensitivity and specificity Fatty Liver Index (FLI) predicts NAFLD FLI = (e 0.953 × loge (triglycerides) + 0.139 × BMI + 0.718 × loge (γGT) + 0.053 × waist circumference − 15.745)/​(1 + e 0.953 × loge (triglycerides) + 0.139 × BMI + 0.718 × loge (γGT) + 0.053 × waist circumference − 15.745) × 100 High risk (FLI >60) Intermediate (FLI 30–​60) Low risk (FLI <30) Se 0.87, Sp 0.86 NAFLD Fibrosis Score (NFS) predicts advanced fibrosis [F3–​4] NFS = −1.675 + 0.037 × age (years) + 0.094 × BMI (kg/​m2) + 1.13 × IFG or diabetes (yes = 1, no = 0) + 0.99 × AST/​ALT ratio − 0.013 × platelet (× 109/​L) − 0.66 × albumin (g/​dL) High risk (NFS >0.676): Se 0.51, Sp 0.98, PPV 0.90, NPV 0.85 Indeterminate (NFS −1.455 to 0.676). Low risk (NFS <−1.455): Se 0.82, Sp 0.77, PPV 0.56, NPV 0.93 FIB-​4 score predicts advanced fibrosis [F3–​4] FIB-​4 = age (years) × AST (IU/​L)/​platelet count (×109/​L) × √
ALT (IU/​L) High risk (FIB-​4 >2.67): Se 0.33, Sp 0.98, PPV 0.80, NPV 0.83 Indeterminate (1.30–​2.67) Low risk (FIB-​4 <1.30): Se 0.74, Sp 0.71, PPV 0.43, NPV 0.90 Enhanced Liver Fibrosis (ELF) test predicts advanced fibrosis [F3–​4] ELF = −7.412 + (ln(HA) × 0.681) + (ln(PIIINP) × 0.775) + (ln(TIMP1) × 0.494) High risk (ELF > 0.3576): Se 0.80, Sp 0.90, PPV 0.71, NPV 0.94 Low risk (ELF < 0.3576) Fibroscan predicts advanced fibrosis [F3–​4] Ultrasonography-​based transient elastography measures the velocity of an elastic shear wave propagating through the liver to assess liver stiffness. Three probes S, M, and XL are used according to body habitus. Greater adiposity may limit test accuracy High risk (M >9.6 kPa; XL >9.3 kPa): M/​XL Se 0.75/​0.57,
Sp 0.92/​0.90, PPV 0.72/​0.71, NPV 0.93/​0.84 Indeterminate (M 7.9–​9.6; XL 7.2–​9.3) Low risk (M <7.9 kPa; XL <7.2 kPa): M/​XL Se 0.91/​0.78,
Sp 0.75/​0.78, PPV 0.52/​0.60, NPV 0.97/​0.89 γGT, γ-​glutamyl transferase; NPV, negative predictive value; PPV, positive predictive value; Se, sensitivity; Sp, specificity. Simple scores such as NFS and FIB-​4 are based on the results of routinely available blood tests and anthropometrics. In contrast, commercial tests such as the ELF Test rely on measurement of nonstandard indices (hyaluronic acid (HA), amino-​ terminal propeptide of type III procollagen (PIIINP), and tissue inhibitor of metalloproteinase 1 (TIMP-​1)) and Fibroscan requires specialist equipment.

Calculate FIB-4 Score Age <65: FIB-4 <1.3 Age >65: FIB-4 <2.0 1.3/2.0 to 2.67 More than 2.67 Low risk High risk Transient elastography (fibroscan)
M probe <7.9k Pa M probe 7.9–9.6 kPa XL probe 7.2–9.3 kPa XL probe <7.2k Pa

M probe >9.6 kPa
XL probe >9.3 kPa
Advanced [F3–4] Fibrosis Unlikely Advanced [F3–4] Fibrosis Likely Indeterminate Manage in primary care, lifestyle advice, address CVD risks, Recalculate NFS 3–5 years. Refer to secondary care Fibrosis F0–1 Fibrosis F2–3 Cirrhosis F4 Lifestyle advice, address CVD risks, NAFLD directed therapy Lifestyle advice, address CVD risks, NAFLD-directed therapy, HCC & variceal surveillance Suspected NAFLD Features of the metabolic syndrome, radiological evidence of steatosis and/or abnormal liver biochemistry, raised FLI, alternative diagnoses including high alcohol consumption excluded Recalculate NAFLD fibrosis score in 3–5 years or if patient develops type 2 diabetes Indeterminate Liver Biopsy Fig. 15.24.2.3  NAFLD risk stratification and management algorithm. Several different risk stratification algorithms have been proposed. This example represents a pragmatic approach to risk stratification that sequentially applies tests with high negative predictive value so that liver biopsy is reserved for a highly selected group of patients at high risk of advanced fibrosis/​cirrhosis. The NAFLD Fibrosis Score may be substituted for the FIB-4 Score, or a validated commercial fibrosis panel such as the ELF Test may replace Fibroscan as the second-​stage screen for fibrosis. Age adjusted thresholds for FIB-4 Score are shown. These maintain the high NPV of the test, which otherwise falls in patients aged >65 years. CVD, cardiovascular disease; FLI, Fatty Liver Index; HCC, hepatocellular carcinoma; NFS, NAFLD Fibrosis Score; NPV/​PPV, negative/​positive predictive value. F0–​4 refer to histological stage of fibrosis.

section 15  Gastroenterological disorders 3154 of thiazolidinediones in patients with diabetes has been associated with congestive heart failure, bladder cancer, and bone fractures. Preliminary data suggests that the glucagon-​like peptide-​1 agonist Liraglutide may be beneficial in slowing NAFLD progression, which effect may be largely through weight loss. Vitamin E (800 IU/​day) has been shown to lower ALT levels and to ameliorate histological steatohepatitis in pre-cirrhotic, nondiabetic patients in the PIVENS and TONIC trials, but a clear benefit on liver fibrosis was not demonstrated. Concern regarding an association with increased all-​cause mortality and prostate cancer mean that its use remains controversial, but it is considered by many to be a first-​ line therapy for selected patients with histologically confirmed NASH without cirrhosis or type 2 diabetes. A Cochrane analysis in 2017 of pharmacological interventions for NAFLD found most of the evidence to be of very low quality and concluded that there was no proven benefit for the majority of agents tested up to that point. However, there is substantial drug develop- ment activity in this field with a number of pharmacological agents currently in clinical trials such as FXR agonists, PPARa/d/g agonists and CCR2/5 antagonists. It is therefore likely that additional, liver-​ directed therapies will become available within the next few years. Prognosis The presence of NAFLD appears to be associated with greater all-​cause mortality, estimated to be 34 to 69% higher than the age and sex-​ matched general population. In most cases it has an indolent course, with patients ultimately succumbing to nonhepatic causes of death (principally cardiovascular disease), but some patients exhibit a more aggressive and rapidly progressive form of disease that leads to cirrhosis. FURTHER READING Natural history Angulo P, et al. (2015). Liver fibrosis, but no other histologic features, is associated with long-​term outcomes of patients with nonalcoholic fatty liver disease. Gastroenterology, 149, 389–​97.e10. Anstee QM, Targher G, Day CP (2013). Progression of NAFLD to diabetes mellitus, cardiovascular disease or cirrhosis. Nat Rev Gastroenterol Hepatol, 10, 330–​44. Ekstedt M, et al. (2015). Fibrosis stage is the strongest predictor for disease-​specific mortality in NAFLD after up to 33 years of follow-​ up. Hepatology, 61, 1547–​54. McPherson S, et al. (2015). Evidence of NAFLD progression from stea- tosis to fibrosing-​steatohepatitis using paired biopsies: Implications for prognosis and clinical management. J Hepatol, 62, 1148–​55. Musso G, et al. (2011). Meta-​analysis: natural history of non-​alcoholic fatty liver disease (NAFLD) and diagnostic accuracy of non-​invasive tests for liver disease severity. Ann Med, 43, 617–​49. Diagnosis and risk-​stratification Angulo P, et al. (2007). The NAFLD fibrosis score: a noninvasive system that identifies liver fibrosis in patients with NAFLD. Hepatology, 45, 846–​54. Anstee Q, Day C (2015). The genetics of non-​alcoholic fatty liver
disease:  spotlight on PNPLA3  & TM6SF2. Semin Liver Dis, 35, 270–​90. Burt AD, Lackner C, Tiniakos DG (2015). Diagnosis and assessment of NAFLD: definitions and histopathological classification. Semin Liver Dis, 35, 207–​20. Dyson JK, McPherson S, Anstee QM (2013). Non-​alcoholic fatty liver disease:  non-​invasive investigation and risk stratification. J Clin Pathol, 66, 1033–​45. Table 15.24.2.4  Validated histological scores commonly used for reporting NAFLD liver biopsies The FLIP Steatosis/​Activity/​Fibrosis Score (SAF) NIDDK NAFLD Activity Score (NAS) Histological feature Category Definition Histological feature Category Definition Steatosis 0 1 2 3 <5% 5–​33% 34–​66%

66% Steatosis 0 1 2 3 <5% 5–​33% 34–​66% 66% (S) Steatosis score 0–​3 PLUS Hepatocyte ballooning 0 1 2 None Clusters of hepatocytes with rounded shape and pale cytoplasm Same as grade 1 with enlarged hepatocytes (>2× normal size) Hepatocyte ballooning 0 1 2 None Few Many PLUS PLUS Inflammation 0 1 2 None < 2 foci per 20× field 2 foci per 20× field Inflammation 0 1 2 3 None 1–​2 foci per ×20 field 2–​4 foci per ×20 field 4 foci per ×20 field (A) Total = activity score 0–​4 (NAS) Total = NAFLD activity score 0–​8 Fibrosis 0 1a 1b 1c 2 3 4 No fibrosis Zone 3 mild perisinusoidal Fibrosis Zone 3 moderate Perisinusoidal fibrosis Periportal/​portal fibrosis only Zone 3 plus portal/​periportal fibrosis Bridging fibrosis Cirrhosis Fibrosis 0 1a 1b 1c 2 3 4 No fibrosis Zone 3 mild perisinusoidal Fibrosis Zone 3 moderate Perisinusoidal fibrosis Periportal/​portal fibrosis only Zone 3 plus portal/​periportal fibrosis Bridging fibrosis Cirrhosis (F) Fibrosis score 0–​4 Fibrosis score 0–​4