# 79 - 192 Rickettsial Diseases

### 192 Rickettsial Diseases

■
■FURTHER READING
Arvikar SL, Steere AC: Diagnosis and treatment of Lyme arthritis. 

Infect Dis Clin North Am 36:563, 2022.
Aucott JN: Posttreatment Lyme disease syndrome. Infect Dis Clin 
North Am 29:309, 2015.
Branda JA, Steere AC: Laboratory diagnosis of Lyme borreliosis. 
Clin Micro Rev 34:e00018, 2021.
Branda JA et al: Two-tiered antibody testing for Lyme disease with 
use of 2 enzyme immunoassays, a whole-cell sonicate enzyme immu­
noassay followed by a VlsE C6 peptide enzyme immunoassay. Clin 
Infect Dis 53:541, 2011.
Klempner MS et al: Two controlled trials of antibiotic treatment in 
patients with persistent symptoms and a history of Lyme disease. 
N Engl J Med 345:85, 2001.
Lantos PM et al: Clinical practice guidelines by the Infectious Diseases 
Society of America (IDSA), American Academy of Neurology, and 
the American College of Rheumatology (ACR): 2020 guidelines for 
the prevention, diagnosis, and treatment of Lyme disease. Clin Infect 
Dis 72:1, 2021.
Lemieux JE et al: Whole genome sequencing of human Borrelia burg­
dorferi isolates reveals linked blocks of accessory genome elements 
located on plasmids and associated with human dissemination. PLoS 
Pathog 17:e1011243, 2023.
Li X et al: Burden and viability of Borrelia burgdorferi in skin or joints 
of patients with erythema migrans or Lyme arthritis. Arthritis Rheum 
63:2238, 2011.
Lochhead RB et al: Lyme arthritis, linking infection, inflammation, 
and autoimmunity. Nat Rev Rheumatol 17:449, 2021.
Oschmann P et al: Stages and syndromes of neuroborreliosis. J Neurol 
PART 5
Infectious Diseases
245:262, 1998.
Steere AC: Lyme disease. N Engl J Med 345:115, 2001.
Steere AC: Posttreatment Lyme disease syndromes: Distinct pathogenesis 
caused by maladaptive host responses. J Clin Invest 130:2148, 2020.
Steere AC et al: Prospective study of serologic tests for Lyme disease. 
Clin Infect Dis 47:188, 2008.
Steere AC et al: Lyme borreliosis. Nat Rev Dis Primers 2:16090, 2016.
Ursinus J et al: Prevalence of persistent symptoms after treatment for 
Lyme borreliosis: A prospective observational cohort study. Lancet 
Reg Health Eur 6:100142, 2021. Erratum in: Lancet Reg Health Eur 
27:100622, 2023.
Section 10	 Diseases Caused by Rickettsiae, 
Mycoplasmas, and Chlamydiae
David H. Walker, J. Stephen Dumler, 

Lucas S. Blanton, Chantal P. Rovers

Rickettsial Diseases
Rickettsiae are a heterogeneous group of small, obligately intracel­
lular, gram-negative coccobacilli and short bacilli, most of which are 
transmitted by a tick, mite, flea, or louse vector. Except in the case of 
louse-borne typhus, humans are incidental hosts. Among rickettsiae, 
Coxiella burnetii, Rickettsia prowazekii, and Rickettsia typhi have the 
well-documented ability to survive for an extended period outside the 
reservoir or vector and to be extremely infectious: inhalation of a single 
Coxiella microorganism can cause pneumonia. High-level infectivity 
and severe illness after inhalation make R. prowazekii, R. rickettsii, R. 
typhi, R. conorii, and C. burnetii bioterrorism threats (Chap. S4).
Clinical infections with rickettsiae can be classified according to (1) 
the taxonomy and diverse microbial characteristics of the agents, which 

belong to seven genera (Rickettsia, Orientia, Ehrlichia, Anaplasma, Neor­
ickettsia, “Candidatus Neoehrlichia,” and Coxiella); (2) epidemiology; or 
(3) clinical manifestations. The clinical manifestations of all the acute 
presentations are similar during the first 5 days: fever, headache, and 
myalgias with or without nausea, vomiting, and cough. As the course 
progresses, clinical manifestations—including a macular, maculopapu­
lar, or vesicular rash; eschar; pneumonitis; and meningoencephalitis—
vary from one disease to another. Given the many etiologic agents with 
varied mechanisms of transmission, geographic distributions, and asso­
ciated disease manifestations, the consideration of rickettsial diseases as 
a single entity poses complex challenges (Table 192-1).
Establishing the etiologic diagnosis of rickettsioses is very diffi­
cult during the acute stage of illness, and definitive diagnosis usually 
requires the examination of serum samples during the acute and con­
valescent phases of illness. Heightened clinical suspicion is based on 
epidemiologic data, history of exposure to vectors or reservoir animals, 
travel to endemic locations, clinical manifestations (sometimes includ­
ing rash or eschar), and characteristic laboratory findings (including 
thrombocytopenia, normal or low white blood cell [WBC] counts, 
elevated hepatic enzyme levels, and hyponatremia). Such suspicion 
should prompt empirical treatment. Doxycycline is the empirical drug 
of choice for most of these infections. Only one agent, C. burnetii, 
has been documented to cause chronic illness. One other species, R. 
prowazekii, causes recrudescent illness (Brill-Zinsser disease) when 
latent infection is reactivated years after resolution of the acute illness.
Rickettsial infections dominated by fever may resolve without fur­
ther clinical evolution. However, after nonspecific early manifestations, 
the illnesses can also evolve along one or more of several principal 
clinical lines: (1) development of a macular or maculopapular rash; 
(2) development of an eschar at the site of tick or mite feeding, which 
can occur during the incubation period; (3) development of a vesicular 
rash (often in rickettsialpox, R. parkeri infection, and African tick-bite 
fever); (4) development of pneumonitis with chest radiographic opaci­
ties and/or rales (Q fever and severe cases of Rocky Mountain spotted 
fever [RMSF], Mediterranean spotted fever [MSF], louse-borne typhus, 
human monocytotropic ehrlichiosis [HME], human granulocytotropic 
anaplasmosis [HGA], scrub typhus, and murine typhus); (5) develop­
ment of meningoencephalitis (louse-borne typhus and severe cases of 
RMSF, scrub typhus, HME, murine typhus, MSF, and [rarely] Q fever); 
and (6) progressive hypotension and multiorgan failure as seen with 
sepsis or toxic shock syndromes (RMSF, MSF, louse-borne typhus, 
murine typhus, scrub typhus, HME, HGA, and neoehrlichiosis).
Epidemiologic clues to the transmission of a particular pathogen 
include (1) environmental exposure to ticks, fleas, or mites during the 
season of activity of the vector species for the disease in the appropri­
ate geographic region (spotted fever and typhus rickettsioses, scrub 
typhus, ehrlichiosis, anaplasmosis); (2) travel to or residence in an 
endemic geographic region during the incubation period (Table 192-1); 
(3) exposure to parturient ruminants, cats, and dogs (Q fever); (4) 
exposure to flying squirrels (R. prowazekii infection); and (5) history 
of previous louse-borne typhus (recrudescent typhus).
Clinical laboratory findings such as thrombocytopenia (particularly 
in spotted fever and typhus rickettsioses, ehrlichiosis, anaplasmosis, 
and scrub typhus), normal or low WBC counts, mild to moderate 
serum elevations of hepatic aminotransferases, and hyponatremia sug­
gest some common pathophysiologic mechanisms.
Application of these clinical, epidemiologic, and laboratory prin­
ciples requires consideration of a rickettsial diagnosis and knowledge 
of the individual diseases.
TICK-, MITE-, LOUSE-, AND FLEA-BORNE 
RICKETTSIOSES
These diseases, caused by organisms of the genera Rickettsia and Ori­
entia in the family Rickettsiaceae, result from endothelial cell infection 
and increased vascular permeability. Pathogenic rickettsial species are 
very closely related, have small genomes (as a result of reductive evolu­
tion, which eliminated many genes for biosynthesis of intracellularly 
available molecules), and are traditionally separated into typhus and 
spotted fever groups on the basis of lipopolysaccharide antigens. Some

TABLE 192-1  Features of Selected Rickettsial Infections
DISEASE
ORGANISM
TRANSMISSION
Rocky Mountain 
spotted fever (RMSF)
Rickettsia 
rickettsii
Tick bite: Dermacentor 
andersoni, D. variabilis, 

D. similis
Amblyomma cajennense, 

A sculptum, A. patinoi, 

A. mixtum, A. aureolatum
Rhipicephalus sanguineus
Mexico, South 
America, United 
States
Mediterranean 
spotted fever (MSF)
R. conorii
Tick bite: R. sanguineus, 

R. pumilio
African tick-bite 
fever
R. africae
Tick bite: A. hebraeum, 

A. variegatum
Maculatum disease
R. parkeri
Tick bite: A. maculatum, 

A. triste, A. tigrinum, A. ovale
Pacific Coast tick 
fever
Rickettsia ssp 
californica 
Tick bite: D. occidentalis
United States
3–9
5–14

+++
Rickettsialpox
R. akari
Mite bite: Liponyssoides 
sanguineus
Tick-borne 
lymphadenopathy
R. slovaca, 

R. raoultii
Tick bite: D. marginatus, 

D. reticularis
Flea-borne spotted 
fever
R. felis
Flea (mechanism 
undetermined): 
Ctenocephalides felis
Epidemic typhus
R. prowazekii
Louse feces: Pediculus 
humanus humanus, fleas 
and lice of flying squirrels, 
or recrudescence
Murine typhus
R. typhi
Flea feces: Xenopsylla 
cheopis, C. felis, others
Human 
monocytotropic 
ehrlichiosis (HME)
Ehrlichia 
chaffeensis
Tick bite: A. americanum, 
D. variabilis
Ewingii ehrlichiosis
E. ewingii
Tick bite: A. americanum
United States
1–21
4–21

None
 
Unnamed 
ehrlichiosis
E. muris ssp. 
eauclairensis
Tick bite: Ixodes scapularis
United States
Unknown
3–14

None
 
Human 
granulocytotropic 
anaplasmosis (HGA)
Anaplasma 
phagocytophilum
Tick bite: I. scapularis, 
I. ricinus, I. pacificus, 
I. persulcatus, 
Haemaphysalis concinna
Unnamed disease
A. capra
I. persulcatus
Northeastern 
China, France
Neoehrlichiosis
“Candidatus 
Neoehrlichia 
mikurensis”
Tick bite: I. ricinus, 
I. persulcatus, 
Haemaphysalis concinna
Scrub typhus
Orientia 
tsutsugamushi
Mite bite: Leptotrombidium 
deliense, others
O. chuto
Mite bite
Africa and 
Western Asia
Candidatus O. 
chiloensis
Mite bite
Chile
Q fever
Coxiella burnetii
Inhalation of aerosols 
of infected parturition 
material (goats, sheep, 
cattle, cats, others), 
ingestion of infected milk 
or milk products
a++++, severe; +++, marked; ++, moderate; +, present in a small proportion of cases; —, not a noted feature.

GEOGRAPHIC 
RANGE
INCUBATION 
PERIOD, DAYS
DURATION, 
DAYS
RASH, 
%
ESCHAR, 
%
LYMPHADENOPATHYa
United States
2–14
10–20

<1
+
Central/

South America
Southern 
Europe, Africa, 
Middle East, 
central Asia
5–7
7–14

+
Sub-Saharan 
Africa, West 
Indies
4–10
4–19

+++
United States, 
South America
2–10
6–16

++
United States, 
Ukraine, Turkey, 
Mexico, Croatia
10–17
3–11

+++
Europe
7–9
17–180

++++
CHAPTER 192
Worldwide
8–16
8–16

—
Worldwide
7–14
10–18

None
—
Rickettsial Diseases
Worldwide
8–16
9–18

None
—
United States
1–21
3–21

None
++
United States, 
Europe, Asia
4–8
3–14
Rare
None
—
Unknown
11–21

+
Europe, China
≥8
11–75

None
 
Asia, Australia, 
Pacific and 
Indian Ocean 
islands
9–18
6–21

+++
Worldwide 
except New 
Zealand, 
Antarctica
3–30
5–57
4–18%
None
—

diseases and their agents (e.g., R. africae, R. parkeri, and R. sibirica) 
are too similar to require separate descriptions. Indeed, the similarities 
among MSF (R. conorii [all strains] and R. massiliae), North Asian tick 
typhus (R. sibirica), Japanese spotted fever (R. japonica), and Flinders 
Island spotted fever (R. honei) far outweigh their minor variations. 
The Rickettsiaceae that cause life-threatening infections are, in order 
of decreasing case–fatality rate, R. rickettsii (RMSF); R. prowazekii 
(louse-borne typhus); Orientia tsutsugamushi (scrub typhus); R. cono­
rii (MSF); R. typhi (murine typhus); and, in rare cases, other spotted 
fever–group (SFG) organisms. Some agents (e.g., R. parkeri, R. africae, 
Rickettsia 364D, R. akari, R. slovaca, R. honei, R. felis, R. massiliae, 

R. helvetica, R. heilongjiangensis, R. aeschlimannii, and R. monacensis) 
have never been documented to cause a fatal illness. The most preva­
lent SFG rickettsia in the United States, R. amblyommatis, has been 
circumstantially associated with asymptomatic seroconversion in most 
persons and with self-limited illness in others.

■
■ROCKY MOUNTAIN SPOTTED FEVER
Epidemiology 
RMSF occurs in 47 states (with the highest preva­
lence in the south-central and southeastern states) as well as in Canada, 
Mexico, and Central and South America. The infection is transmit­
ted by Dermacentor variabilis, the American dog tick, in the eastern 
two-thirds of the United States; by D. andersoni, the Rocky Mountain 
wood tick, in the western United States; by D. similis in California; 
by Rhipicephalus sanguineus, the brown dog tick, in Mexico, Arizona, 
and probably Colombia; and by Amblyomma sculptum, A. mixtum, A. 
patinoi, A. cajennense, A. tonelliae, and A. aureolatum in Central and/
or South America. Maintained partially by transovarian transmission 
from one generation of ticks to the next, R. rickettsii can be acquired 
by uninfected ticks through the ingestion of a blood meal from rick­
ettsemic small mammals or by co-feeding adjacent to an infected tick.
PART 5
Infectious Diseases
Humans become infected during tick season (in the Northern 
Hemisphere, from April to September), although some cases occur in 
winter. The mortality rate was 20–25% in the preantibiotic era and has 
been reported at ~3–5% in the postantibiotic era, principally because 
of delayed diagnosis and treatment. Recent reporting of a relatively low 
mortality rate (0.4%) for spotted fever rickettsiosis is likely an artifact 
related to the abundance of less pathogenic SFG rickettsial species 
likely causing subclinical or undiagnosed infection with cross-reactive 
anti-SFG antibodies and to a relatively low proportion of diagnostically 
confirmed cases. Indeed, the reported case–fatality rates in confirmed 
cases in the United States and in parts of Arizona, where R. rickettsii is 
the sole infecting SFG species, are 9% and 10%, respectively. The case–
fatality rate is highest among children (<10 years of age) and in the later 
decades of life (>70 years). For unknown reasons, the case–fatality rate 
of RMSF in Mexico and Brazil approaches 50%.
Pathogenesis 
R. rickettsii organisms are inoculated into the dermis 
along with secretions of the tick’s salivary glands after ≥6 h of feeding. 
The rickettsiae spread lymphohematogenously throughout the body 
and infect numerous foci of contiguous endothelial cells. The dosedependent incubation period is ~1 week (range, 2–14 days). Occlusive 
thrombosis and ischemic necrosis are not the fundamental pathologic 
bases for tissue and organ injury. Instead, increased vascular perme­
ability, with resulting edema, hypovolemia, and ischemia, is responsible. 
Consumption of platelets results in thrombocytopenia in 32–52% of 
patients, but disseminated intravascular coagulation (DIC) with hypofi­
brinogenemia is rare. Activation of platelets, generation of thrombin, and 
activation of the fibrinolytic system all appear to be homeostatic physi­
ologic responses to endothelial injury by nonocclusive hemostatic plugs.
Clinical Manifestations 
Early in the illness, when medical atten­
tion usually is first sought, RMSF is difficult to distinguish from many 
self-limiting viral illnesses. Fever, headache, malaise, myalgia, nausea, 
vomiting, and anorexia are the most common symptoms during the first 
3 days. The patient becomes progressively more ill as vascular infec­
tion and injury advance. In one large series, only one-third of patients 
were diagnosed with presumptive RMSF early in the clinical course and 
treated appropriately as outpatients. In the tertiary-care setting, RMSF is 

all too often recognized only when late severe manifestations, develop­
ing at the end of the first week or during the second week of illness in 
patients without appropriate treatment, prompt return to a physician or 
hospital and admission to an intensive care unit.
The progressive nature of the infection is clearly manifested in the skin. 
Rash is evident in only 14% of patients on the first day of illness and in 
only 49% during the first 3 days. Erythematous macules (1–5 mm) appear 
first on the wrists and ankles and then on the remainder of the extremities 
and the trunk. Later, more severe vascular damage results in frank hemor­
rhage at the center of the maculopapule, producing a petechia that does 
not disappear upon compression (Fig. 192-1). This sequence of events is 
sometimes delayed or aborted by effective treatment. However, the rash is 
a variable manifestation, appearing on day 6 or later in 20% of cases and 
not appearing at all in ~10% of cases. Petechiae occur in 41–59% of cases, 
appearing on or after day 6 in 74% of cases that manifest a rash. Involve­
ment of the palms and soles, often considered diagnostically important, 
usually develops relatively late in the course (after day 5 in 43% of cases) 
and does not develop at all in 18–64% of cases.
Hypovolemia leads to prerenal azotemia and (in 17% of cases) hypo­
tension. Infection of the pulmonary microcirculation leads to noncar­
diogenic pulmonary edema; 12% of patients have acute respiratory 
distress syndrome, and 8% require mechanical ventilation. Cardiac 
involvement manifests as dysrhythmia in 7–16% of cases.
Besides respiratory failure, central nervous system (CNS) involve­
ment is the other important determinant of the outcome of RMSF. 
Encephalitis, presenting as confusion or lethargy, is apparent in 
26–28% of cases. Progressively severe encephalitis manifests as stupor 
or delirium in 21–26% of cases, ataxia in 18%, coma in 10%, and sei­
zures in 8%. Numerous focal neurologic deficits have been reported. 
A
B
FIGURE 192-1  A. Petechial lesions of Rocky Mountain spotted fever on the lower 
legs and soles of a young, previously healthy patient. B. Close-up of lesions from the 
same patient. (Photos courtesy of Dr. Lindsey Baden; with permission.)

Meningoencephalitis results in cerebrospinal fluid (CSF) pleocytosis 
in 34–38% of cases; usually there are 10–100 cells/μL and a mono­
nuclear predominance, but occasionally there are >100 cells/μL and 
a polymorphonuclear predominance. The CSF protein concentration 
is increased in 30–35% of cases, but the CSF glucose concentration is 
usually normal.
Acute kidney injury, often reversible with rehydration, is caused by 
acute tubular necrosis in severe cases with shock. Hepatic injury with 
increased serum aminotransferase concentrations (38% of cases) is due 
to multifocal death of individual hepatocytes without hepatic failure. 
Jaundice is recognized in 9% of cases and an elevated serum bilirubin 
concentration in 18–30%.
Life-threatening bleeding is rare. Anemia develops in 30% of cases 
and is severe enough to require transfusions in 11%. Blood is detected 
in the stool or vomitus of 10% of patients, and death has followed mas­
sive upper-gastrointestinal hemorrhage.
Other characteristic clinical laboratory findings include increased 
plasma levels of proteins of the acute-phase response (C-reactive pro­
tein, procalcitonin, fibrinogen, ferritin, and others), hypoalbuminemia, 
and hyponatremia (in 56% of cases) due to the appropriate secretion of 
antidiuretic hormone in response to the hypovolemic state. Myositis 
occurs occasionally, with marked elevations in serum creatine kinase 
levels and multifocal rhabdomyonecrosis. Ocular involvement includes 
conjunctivitis in 30% of cases and retinal vein engorgement, flame 
hemorrhages, arterial occlusion, and papilledema with normal CSF 
pressure in some instances.
Severe RMSF can present as sepsis or septic shock. In untreated fatal 
cases, death occurs 8–15 days after onset. A rare presentation, fulmi­
nant RMSF, is fatal within 5 days after onset. This fulminant presenta­
tion is seen most often in male black patients with glucose-6-phosphate 
dehydrogenase (G6PD) deficiency and may be related to an undefined 
effect of hemolysis on the rickettsial infection. Although survivors 
of RMSF usually return to their previous state of health, permanent 
sequelae, including neurologic deficits and gangrene necessitating 
amputation of extremities, may follow severe illness.
Diagnosis 
The diagnosis of RMSF during the acute stage is more 
difficult than is generally appreciated. The most important epidemio­
logic factor is a history of exposure to a potentially tick-infested envi­
ronment within the 14 days preceding disease onset during a season 
of possible tick activity. However, only 60% of patients actually recall 
being bitten by a tick during the incubation period.
The differential diagnosis for early clinical manifestations of RMSF 
(fever, headache, and myalgia without a rash) includes influenza, 
enteroviral infection, infectious mononucleosis, viral hepatitis, lep­
tospirosis, typhoid fever, gram-negative or gram-positive bacterial 
sepsis, HME, HGA, murine typhus, sylvatic flying-squirrel typhus, 
and rickettsialpox. Enterocolitis may be suggested by nausea, vomit­
ing, and abdominal pain; prominence of abdominal tenderness has 
resulted in exploratory laparotomy. CNS involvement can masquerade 
as bacterial or viral meningoencephalitis. Cough, pulmonary signs, and 
chest radiographic opacities can lead to a diagnostic consideration of 
bronchitis or pneumonia.
At presentation during the first 3 days of illness, only 3% of patients 
exhibit the classic triad of fever, rash, and history of tick exposure. When 
a rash appears, a diagnosis of RMSF should be considered. However, 
many illnesses considered in the differential diagnosis also can be associ­
ated with a rash, including rubeola, rubella, meningococcemia, dissemi­
nated gonococcal infection, secondary syphilis, toxic shock syndrome, 
drug hypersensitivity, immune thrombocytopenic purpura, thrombotic 
thrombocytopenic purpura, Kawasaki syndrome, and immune complex 
vasculitis. Conversely, any person in an endemic area with a provisional 
diagnosis of one of the above illnesses could have RMSF. Thus, if a viral 
infection is suspected during RMSF season in an endemic area, it should 
always be kept in mind that RMSF can mimic viral infection early in the 
course; if the illness worsens over the next couple of days after initial 
presentation, the patient should return for reevaluation.
The most common serologic test for confirmation of the diagnosis is 
the indirect immunofluorescence assay. Not until 7–10 days after onset 

is a reactive titer of ≥64 first detectable. The sensitivity and specific­
ity of the indirect immunofluorescence IgG assay are 89–100% and 
99–100%, respectively. Detection of IgM is no more sensitive in early 
illness and is subject to nonspecific cross-reactivity. It is important 
to understand that serologic tests for RMSF are usually negative at 
the time of presentation for medical care and that treatment should 
not be delayed while a positive serologic result is awaited. Because of 
significant seropositivity in the expanding geographic distribution of 
Amblyomma americanum ticks, a positive serologic titer in the first 
5–6 days of illness represents preexisting antibody unrelated to the 
current illness.

The only diagnostic test that has proven useful during the acute ill­
ness is immunohistologic examination of a cutaneous biopsy sample 
from a rash lesion for R. rickettsii. Examination of a 3-mm punch 
biopsy from such a lesion is 70% sensitive and 100% specific, and poly­
merase chain reaction (PCR) on a rash biopsy would likely yield even 
higher sensitivity. PCR amplification for detection of R. rickettsii DNA 
in peripheral blood is not adequately sensitive. Although rickettsiae 
are present in large quantities in heavily infected foci of endothelial 
cells, there are relatively low quantities in the circulation. Cultiva­
tion of rickettsiae in cell culture is feasible but is seldom undertaken 
because of technical difficulty and biohazard concerns. The recent 
dramatic increase in the reported incidence of RMSF correlates with 
the use of single-titer SFG cross-reactive enzyme immunoassay serol­
ogy. Few cases are specifically determined to be caused by R. rickettsii. 
Currently, many febrile persons who do not have RMSF present with 
cross-reactive antibodies, possibly because of previous exposure to the 
highly prevalent SFG rickettsia R. amblyommatis.
CHAPTER 192
TREATMENT
Rocky Mountain Spotted Fever
The drug of choice for the treatment of both children and adults 
with RMSF is doxycycline. Because of the severity of RMSF, imme­
diate empirical administration of doxycycline should be strongly 
considered for any patient with a consistent clinical presentation 
in the appropriate epidemiologic setting. Doxycycline is adminis­
tered orally (or, with coma or vomiting, intravenously) at 100 mg 
twice daily. For children with suspected RMSF, up to five courses 
of doxycycline may be administered with minimal risk of den­
tal staining. In patients with allergy to doxycycline, desensitiza­
tion should be considered. Once considered an alternative during 
pregnancy, chloramphenicol is not readily available in the United 
States. Although available in much of the world, it is less effective 
than doxycycline. Fortunately, there is little evidence to support 
the occurrence of tetracycline-associated adverse events in moth­
ers (hepatotoxicity) and fetuses (staining of deciduous teeth and 
teratogenicity) who receive doxycycline. The antirickettsial drug 
should be administered until the patient is afebrile and improving 
clinically—usually 3–5 days after defervescence. β-Lactam antibiot­
ics, erythromycin, and aminoglycosides have no role in the treat­
ment of RMSF, and sulfa-containing drugs are associated with more 
adverse outcomes than no treatment at all. There is little clinical 
experience with fluoroquinolones, clarithromycin, and azithromy­
cin, which are not recommended. The most seriously ill patients 
are managed in intensive care units, with careful administration 
of fluids to achieve optimal tissue perfusion without precipitating 
noncardiogenic pulmonary edema. In some severely ill patients, 
hypoxemia requires intubation and mechanical ventilation; oliguric 
or anuric acute renal failure requires renal replacement therapy; 
seizures necessitate the use of antiseizure medication; anemia or 
severe hemorrhage necessitates transfusions of packed red blood 
cells; or bleeding with severe thrombocytopenia requires platelet 
transfusions.
Rickettsial Diseases
Prevention 
Avoidance of tick bites is the only available preventive 
approach. Use of protective clothing and tick repellents, inspection of 
the body once or twice a day, and removal of ticks before they inoculate

rickettsiae reduce the risk of infection. Prophylactic doxycycline treat­
ment of tick bites has no proven role in preventing RMSF.

■
■MEDITERRANEAN SPOTTED FEVER 
(BOUTONNEUSE FEVER), AFRICAN TICK-BITE 
FEVER, AND OTHER TICK-BORNE SPOTTED FEVERS
Epidemiology and Clinical Manifestations 
R. conorii is preva­
lent in southern Europe, Africa, and southwestern and south-central 
Asia. The disease is characterized by high fever, rash, and—in most 
geographic locales—an inoculation eschar (tâche noire) that appears 
before the onset of fever at the site of the tick bite. A severe form of the 
disease (mortality rate, 50%) occurs in patients with diabetes, alcohol­
ism, or heart failure.
African tick-bite fever, caused by R. africae, occurs in rural areas of 
sub-Saharan Africa and in the Caribbean islands and is transmitted by 
Amblyomma hebraeum and A. variegatum ticks. The average incuba­
tion period is 4–10 days. The mild illness consists of headache, fever, 
eschar, and regional lymphadenopathy. Amblyomma ticks, a high por­
tion of which are infected with R. africae, often feed in groups, with the 
consequent development of multiple eschars. Rash may be vesicular, 
sparse, or absent altogether. Because of tourism in sub-Saharan Africa, 
African tick-bite fever is the rickettsiosis most frequently imported 
into Europe and North America. Maculatum disease, a similar disease 
caused by the closely related species R. parkeri, is transmitted by A. 
maculatum and found in a low percentage of A. americanum ticks in 
the United States. It is also transmitted by A. triste in South America 
and Arizona as well as A. tigrinum and A. ovale in South America.
R. japonica causes Japanese spotted fever, which also occurs in 
Korea, Thailand, Laos, and China. Similar diseases in northern Asia are 
caused by R. sibirica and R. heilongjiangensis. Queensland tick typhus 
due to R. australis is transmitted by Ixodes holocyclus ticks. Flinders 
Island spotted fever, found on the island for which it is named as well 
as in Tasmania, mainland Australia, and Asia, is caused by R. honei. In 
Europe, patients infected with R. slovaca by a wintertime Dermacentor 
tick bite usually manifest an afebrile illness with an eschar (usually on 
the scalp) and painful regional lymphadenopathy; R. raoultii causes a 
similar syndrome in Europe and Asia.
PART 5
Infectious Diseases
Diagnosis 
Diagnosis of these tick-borne spotted fevers is based 
on clinical and epidemiologic findings and is confirmed by serology, 
immunohistochemical demonstration of rickettsiae in skin biopsy 
specimens, cell-culture isolation of rickettsiae, or PCR of skin biopsy, 
eschar biopsy or swab, or blood samples. Serologic diagnosis detects 
antibodies to antigens shared among SFG rickettsiae, hindering iden­
tification of the etiologic species. In an endemic area, a possible diag­
nosis of rickettsial spotted fevers should be considered when patients 
present with fever, rash, and/or a skin lesion consisting of a black 
necrotic lesion or a crust surrounded by erythema.
TREATMENT
Tick-Borne Spotted Fevers
As with RMSF, severe cases should be treated with doxycycline 
(100 mg bid orally) continued for 3–5 days after defervescence. 
Alternative agents for milder disease include doxycycline (100 mg 
bid orally for 1–5 days), chloramphenicol (500 mg qid orally 
for 7–10 days), ciprofloxacin (750 mg bid orally for 7 days), and 
josamycin (3 g/d orally for 5 days) where available. Data on the 
efficacy of treatment of mildly ill children with clarithromycin or 
azithromycin should not be extrapolated to adults or to patients 
with moderate or severe illness.
■
■RICKETTSIALPOX
R. akari infects mice and their mites (Liponyssoides sanguineus), which 
maintain the organisms by transovarial transmission.
Epidemiology 
Rickettsialpox is recognized principally in New 
York City, but cases have also been reported in other urban and rural 

FIGURE 192-2  Eschar at the site of the mite bite in a patient with rickettsialpox. 
(Reprinted from A Krusell et al: Emerg Infect Dis 8:727, 2002. Photo obtained by 
Dr. Kenneth Kaye.)
locations in the United States and in Ukraine, Croatia, Mexico, and 
Turkey. Investigation of eschars suspected of representing bioterror­
ism-associated cutaneous anthrax revealed that rickettsialpox occurs 
more frequently than previously realized.
Clinical Manifestations 
A papule forms at the site of the mite’s 
feeding, develops a central vesicle, and becomes a 1- to 2.5-cm painless 
black crusted eschar surrounded by an erythematous halo (Fig. 192-2). 
Enlargement of the regional lymph nodes draining the eschar suggests 
initial lymphogenous spread. After an incubation period of 10–17 days, 
during which the eschar and regional lymphadenopathy frequently go 
unnoticed, disease onset is marked by malaise, chills, fever, headache, and 
myalgia. A macular rash appears 2–6 days after onset and usually evolves 
sequentially into papules, vesicles, and crusts that heal without scarring 
(Fig. 192-3); in some cases, the rash remains macular or maculopapular. 
Some patients develop nausea, vomiting, abdominal pain, cough, con­
junctivitis, or photophobia. Without treatment, fever lasts 6–10 days.
Diagnosis and Treatment 
Clinical, epidemiologic, and conva­
lescent serologic data establish the diagnosis of a SFG rickettsiosis 
that is seldom pursued further. Doxycycline is the drug of choice for 
treatment.
■
■FLEA-BORNE SPOTTED FEVER
Rickettsia felis is suspected to cause an emerging rickettsiosis world­
wide. Maintained transovarially in the geographically widespread cat 
flea, Ctenocephalides felis, the infection has been described as moder­
ately severe, with fever, rash, and headache as well as CNS, gastroin­
testinal, and pulmonary symptoms on the basis of PCR, which often 
detects organisms in healthy persons. Patient isolates are lacking.
■
■EPIDEMIC (LOUSE-BORNE) TYPHUS
Epidemiology 
The human body louse (Pediculus humanus humanus) 
lives in clothing under poor hygienic conditions and usually in impov­
erished cold areas. Lice acquire R. prowazekii when they ingest blood 
from a rickettsemic patient. The rickettsiae multiply in the louse’s mid­
gut epithelial cells and are shed in its feces. The infected louse leaves a 
febrile person and deposits infected feces on its subsequent host during 
its blood meal; the patient autoinoculates the organisms by scratching. 
The louse is killed by the rickettsiae and does not pass R. prowazekii 
to its offspring.
Epidemic typhus haunts regions afflicted by wars and disasters. 
An outbreak involved 100,000 people in refugee camps in Burundi in 
1997. A small focus was documented in Russia in 1998, sporadic cases 
were reported from Algeria, and frequent outbreaks occurred in Peru, 
Rwanda and Burundi. Eastern flying squirrels (Glaucomys volans) and 
their lice and fleas maintain R. prowazekii in a zoonotic cycle and transmit 
infection to humans.

A
B
FIGURE 192-3  A. Papulovesicular lesions on the trunk of the patient with 
rickettsialpox shown in Fig. 192-2. B. Close-up of lesions from the same patient. 
(Reprinted from A Krusell et al: Emerg Infect Dis 8:727, 2002. Photos obtained 
by Dr. Kenneth Kaye.)
Brill-Zinsser disease is a recrudescent illness occurring years after 
acute epidemic typhus, probably as a result of waning immunity. 

R. prowazekii remains latent for years; its reactivation results in spo­
radic cases of disease in louse-free populations or in epidemics in 
louse-infested populations. Recrudescence has been documented after 
flying squirrel–associated typhus.
Rickettsiae are potential agents of bioterrorism (Chap. S4). Infec­
tions with R. prowazekii and R. rickettsii have high case–fatality ratios. 
These organisms cause difficult-to-diagnose diseases and are highly 
infectious when inhaled as aerosols. Organisms resistant to tetracycline 
or chloramphenicol have been developed in the laboratory.
Clinical Manifestations 
After an incubation period of ~1–2 
weeks, the onset of illness is abrupt, with prostration, severe headache, 
and fever rising rapidly to 38.8°–40.0°C (102°–104°F). Cough is promi­
nent, developing in 70% of patients. Myalgias are usually severe. A rash 
begins on the upper trunk, usually on the fifth day, and then becomes 
generalized, involving the entire body except the face, palms, and soles. 
Initially, this rash is macular; without treatment, it becomes maculo­
papular, petechial, and confluent. The rash often goes undetected on 
black skin; 60% of African patients have spotless epidemic typhus. Pho­
tophobia, with considerable conjunctival injection, is common. The 
tongue may be dry, brown, and furred. Confusion and coma associated 
with meningoencephalitis are common. Skin necrosis and gangrene of 
the digits as well as interstitial pneumonia may occur in severe cases. 
Untreated disease is fatal in 7–40% of cases, with outcome depending 
primarily on the condition of the host. Patients with untreated infec­
tions develop renal insufficiency and multiorgan involvement in which 

neurologic manifestations are frequently prominent. Overall, 12% of 
patients with epidemic typhus have neurologic involvement. Infection 
associated with North American flying squirrels is a milder illness; 
whether this milder disease is due to host factors (e.g., better health 
status) or attenuated virulence is unknown.

Diagnosis and Treatment 
Epidemic typhus is sometimes misdiag­
nosed as typhoid fever in tropical countries (Chap. 171). This means 
even serologic studies are often unavailable in settings of louse-borne 
typhus. Epidemics can be recognized by the serologic or immunohisto­
chemical diagnosis of a single case or by detection of R. prowazekii in a 
louse found on a patient. Doxycycline (100 mg bid) is administered orally 
or—if the patient is comatose or vomiting—intravenously and continued 
until 3–5 days after defervescence. Under epidemic conditions, a single 
200-mg oral dose can be tried but fails in some cases. Pregnant patients 
should be evaluated individually and treated with chloramphenicol early 
in pregnancy or with doxycycline late in pregnancy.
Prevention 
Prevention of epidemic typhus involves control of 
body lice. Clothes should regularly be changed and laundered in hot 
water, and insecticides can be used every 6 weeks to control the louse 
population.
■
■ENDEMIC MURINE TYPHUS
Epidemiology 
R. typhi is maintained in mammalian host–flea 
cycles, with rats (Rattus rattus and R. norvegicus) and the Oriental rat 
flea (Xenopsylla cheopis) as the classic zoonotic niche. Fleas acquire R. 
typhi from rickettsemic rats and carry the organism throughout their 
life span. Nonimmune rats and humans are infected when rickettsialaden flea feces contaminate pruritic bite lesions; less frequently, the 
flea bite transmits the organisms. Transmission can also occur via inha­
lation of aerosolized rickettsiae from flea feces. Infected rats appear 
healthy, although they are rickettsemic for ~2 weeks.
CHAPTER 192
In the United States, murine typhus occurs mainly in Texas, where 
the disease is expanding northward, and in southern California, where 
it appears to affect people experiencing homelessness disproportion­
ally. The incidence is steadily increasing in both of these states, where 
the classic rat–flea cycle is absent and an opossum–cat flea (C. felis) 
cycle is highly suspected. Globally, endemic typhus occurs mainly 
in warm (often coastal) areas throughout the tropics and subtropics, 
where it is highly prevalent though often unrecognized. The incidence 
peaks from April through July in Texas and during the warm months 
of summer and early fall in other geographic locations. Patients seldom 
recall exposure to fleas, although exposure to animals such as cats, 
opossums, and rats is reported in nearly 40% of cases.
Rickettsial Diseases
Clinical Manifestations 
The incubation period of experimental 
murine typhus averages 11 days (range, 8–16 days). Headache, myalgia, 
arthralgia, nausea, and malaise develop 1–3 days before onset of chills 
and fever. Patients often experience nausea and vomiting.
The duration of untreated illness averages 12 days (range, 9–18 
days). Rash occurs in approximately half of all patients. It is present in 
only 13% of patients at presentation for medical care (usually ~4 days 
after onset of fever), appearing an average of 2 days later in half of the 
remaining patients. The initial macular rash is often faint and detected 
by careful inspection of the axilla or the inner surface of the arm. Sub­
sequently, the rash becomes maculopapular, involving the trunk more 
often than the extremities; it is seldom petechial and rarely involves 
the face, palms, or soles. A rash is detected in only 20% of patients with 
darkly pigmented skin.
Pulmonary involvement is frequently prominent; 35% of patients 
have a hacking, nonproductive cough, and 23% of patients who 
undergo chest radiography have pulmonary densities due to intersti­
tial pneumonia, pulmonary edema, and pleural effusions. Bibasilar 
rales are the most common pulmonary sign. Less common clinical 
manifestations include abdominal pain, confusion, stupor, seizures, 
ataxia, coma, and jaundice. Clinical laboratory studies frequently 
reveal anemia and leukopenia early in the course, leukocytosis late 
in the course, thrombocytopenia, hyponatremia, hypoalbuminemia, 
increased serum levels of hepatic aminotransferases, and prerenal

azotemia. Complications can include respiratory failure, hematemesis, 
cerebral hemorrhage, and hemolysis. Severe illness necessitates the 
admission of 10% of hospitalized patients to an intensive care unit. 
Greater severity is generally associated with old age, underlying dis­
ease, and treatment with a sulfonamide; the case–fatality rate is 0.4%.

Diagnosis and Treatment 
Serologic studies of acute- and con­
valescent-phase serum samples can provide a diagnosis, and an 
immunohistochemical method for identification of typhus groupspecific antigens in biopsy samples has been developed. Cultivation 
is used infrequently and is not widely available. PCR of the blood is 
not adequately sensitive. When endemic typhus is suspected, patients 
should be treated empirically with doxycycline (100 mg twice daily by 
mouth for 7 days). Where available, chloramphenicol is an alternative. 
Ciprofloxacin and azithromycin are active in vitro, but clinically, they 
are less effective than doxycycline.
■
■SCRUB TYPHUS
Epidemiology 
O. tsutsugamushi differs substantially from Rickett­
sia species both genetically and in cell-wall composition (i.e., it lacks 
lipopolysaccharide). O. tsutsugamushi is maintained by transovarial 
transmission in trombiculid mites. After hatching, infected larval mites 
(chiggers, the only stage that feeds on a host) inoculate organisms into 
the skin. Infected chiggers are particularly likely to be found in areas 
of heavy scrub vegetation during the wet season, when mites lay eggs.
Scrub typhus is endemic and reemerging in eastern and southern 
Asia, northern Australia, and islands of the western Pacific and Indian 
Oceans. Infections are prevalent in these regions; in some areas, >3% 
of the population is infected or reinfected each month. Immunity to 
the homologous strain wanes over 1–3 years, and the organisms exhibit 
remarkable antigenic diversity with loss of cross-protective immunity 
in as short a period as 1 month. Emerging cases in Chile, serologic 
evidence of infection in other parts of South and Central America and 
Africa, and the discovery of Orientia chuto infection in southwestern 
Asia challenge the classic epidemiology of scrub typhus.
PART 5
Infectious Diseases
Clinical Manifestations 
Illness varies from mild and self-limiting 
to fatal. After an incubation period of 6–21 days, onset is characterized 
by fever, headache, myalgia, cough, and gastrointestinal symptoms. 
Some patients recover spontaneously after a few days. The classic case 
description includes an eschar where the chigger has fed, regional 
lymphadenopathy, and a maculopapular rash—signs that are seldom 
seen in indigenous patients. In fact, <50% of returning travelers develop 
an eschar, and <40% develop a rash (on day 4–6 of illness). Severe cases 
typically manifest with encephalitis and interstitial pneumonia due to 
vascular injury. The case–fatality rate for untreated classic cases is 6% 
but would probably be lower if all mild cases were diagnosed.
Diagnosis and Treatment 
Serologic assays (indirect fluorescent 
antibody, indirect immunoperoxidase, enzyme immunoassays, and 
rapid diagnostic assays using lateral flow chromatographic tests) are 
the mainstays of laboratory diagnosis. PCR amplification of Orientia 
genes from eschars is effective, but less so for blood. Patients are treated 
with oral doxycycline (100 mg twice daily for 7–15 days), azithromycin 
(500 mg for 3 days), or chloramphenicol (500 mg four times daily for 
7–15 days). In severe scrub typhus, the combination of doxycycline and 
azithromycin is beneficial compared to monotherapy of either agent.
Some cases of scrub typhus in Thailand are poorly responsive to doxy­
cycline or chloramphenicol but respond to azithromycin and rifampin.
EHRLICHIOSES AND ANAPLASMOSES
Ehrlichioses are acute febrile infections caused by members of the family 
Anaplasmataceae, which is made up of obligately intracellular organ­
isms of more than five genera, including those with established human 
pathogens: Ehrlichia, Anaplasma, Wolbachia, “Candidatus Neoehrlichia,” 
and Neorickettsia. The bacteria reside in vertebrate reservoirs and target 
vacuoles of hematopoietic and, for some species, endothelial cells 
(Fig. 192-4). Four Ehrlichia species, two Anaplasma species, and one 
Neoehrlichia species are transmitted by ticks to humans and cause infec­
tion that can be severe and prevalent. E. chaffeensis, the agent of HME, 

FIGURE 192-4  Peripheral-blood smear from a patient with human granulocytotropic 
anaplasmosis. A neutrophil contains two morulae (vacuoles filled with Anaplasma 
phagocytophilum). (Photo courtesy of Dr. J. Stephen Dumler.)
and E. muris subsp. eauclairensis infect predominantly mononuclear 
phagocytes; E. ewingii and A. phagocytophilum infect neutrophils. Infec­
tions with “Candidatus Neoehrlichia mikurensis,” A. capra, and A. bovis 
are less well characterized but have been reported to grow in endothe­
lium, erythrocytes, and monocytes, respectively.
Ehrlichia, “Candidatus Neoehrlichia,” and Anaplasma are main­
tained by horizontal tick–mammal–tick transmission, and humans 
are only inadvertently infected. Wolbachiae are associated with human 
filariasis, since they are important for filarial viability and pathogenic­
ity; antibiotic treatment targeting wolbachiae is a strategy for filariasis 
control. Neorickettsiae parasitize flukes (trematodes) that in turn 
parasitize aquatic snails, fish, and insects. Only a single pathogen of 
human neorickettsiosis has been described: N. sennetsu causes sen­
netsu fever, an infectious mononucleosis–like illness first identified in 
1953 in association with the ingestion of raw fish containing N. sen­
netsu–infected flukes.
■
■HUMAN MONOCYTOTROPIC EHRLICHIOSIS
Epidemiology 
More than 25,480 cases of E. chaffeensis infec­
tion had been reported to the U.S. Centers for Disease Control and 
Prevention (CDC) as of November 2023. However, active prospective 
surveillance documented an incidence as high as 414 cases per 100,000 
population in some U.S. regions. Most E. chaffeensis infections are 
identified in the south-central, southeastern, and mid-Atlantic states, 
but cases have also been recognized in California, New York, New Eng­
land, and midwestern states. All stages of the Lone Star tick (A. ameri­
canum), which is expanding its geographic range, feed on white-tailed 
deer—a major reservoir. Dogs and coyotes also serve as reservoirs 
and often lack clinical signs. Tick bites and exposures are frequently 
reported by patients in rural areas, and 64% of infections occur in May 
through July. Transfusion-related infections are documented, as are 
infections in patients with organ transplants. The incidence of HME is 
highest in patients 60–84 years of age; however, 9% of infections occur 
in children ≤19 years of age, and these include severe and fatal infec­
tions. Of patients with HME, 59% are male.
E. chaffeensis has been detected in South and Central America, 
Africa, and Asia.
Clinical Manifestations 
E. chaffeensis disseminates hematog­
enously from the dermal blood pool created by the feeding tick. After a 
median incubation period of 8 days, illness develops. Clinical manifes­
tations are undifferentiated and include fever (97% of cases), headache 
(70%), myalgia (68%), and malaise (77%). Less frequently observed 
are nausea, vomiting, and diarrhea (28–57%); cough (30%); rash (29% 
overall, 6% at presentation); and confusion (20%). HME can be severe:

77% of patients with confirmed cases are hospitalized, and 2% die. Lifethreatening complications include renal failure, meningoencephalitis, 
acute respiratory distress syndrome, a DIC-like syndrome, pneumonia, 
septic shock, cardiac failure, hepatitis, hemorrhage, hemophagocytic 
lymphohistiocytosis and—in immunocompromised patients—
overwhelming ehrlichial infection; patients with diabetes, cancer, 
organ transplantation, asplenia, hepatitis C, or HIV infection have a 2.3 
relative risk for death. Laboratory findings are valuable in the differen­
tial diagnosis of HME; 66% of patients have leukopenia (initially lym­
phopenia, later neutropenia), 86% have thrombocytopenia, and 89% 
have elevated serum levels of hepatic aminotransferases. Despite low 
blood cell counts, the bone marrow is hypercellular, and noncaseating 
granulomas can be present. Vasculitis is not a component of HME.
Diagnosis 
HME can be fatal. If not given empirical doxycycline 
treatment, 39% and 40% of patients with HME require admission to 
an intensive care unit and mechanical ventilation, respectively; these 
measures are not necessary in patients receiving prompt empirical 
treatment. In addition, hospital stay and illness duration are length­
ened in untreated patients by 8 and 12 days, respectively. The diagnosis 
is suggested by fever, known tick exposure in the preceding 3 weeks, 
thrombocytopenia and/or leukopenia, and increased serum amino­
transferase activities. Morulae are demonstrated in <10% of peripheralblood smears. HME can be confirmed during active infection by PCR 
amplification of E. chaffeensis nucleic acids in blood obtained before 
the start of doxycycline therapy. Retrospective serodiagnosis requires 
a consistent clinical picture and a fourfold increase in E. chaffeensis 
antibody titer to ≥128 in paired serum samples obtained ~3 weeks 
apart. Separate specific diagnostic tests are necessary for HME and 
HGA (see below).
■
■EWINGII EHRLICHIOSIS AND EHRLICHIA MURIS 
EAUCLAIRENSIS INFECTIONS
Ehrlichia ewingii resembles E. chaffeensis in its tick vector (A. americanum) 
and vertebrate reservoirs (white-tailed deer and dogs). E. muris eauclai­
rensis causes human infections after Ixodes scapularis tick exposure in 
Wisconsin and Minnesota. E. ewingii and E. muris illnesses are similar to 
but less severe than HME. Many cases occur in immunocompromised 
patients. Human infections with E. canis have been documented as 
subclinical ehrlichemia. No specific serologic diagnostic tests for these 
other ehrlichiae are readily available, and E. chaffeensis serologic tests 
can be positive when the infecting agent is actually a different species 
of Ehrlichia.
■
■“CANDIDATUS NEOEHRLICHIA MIKURENSIS” 
INFECTION
“Candidatus Neoehrlichia mikurensis,” a bacterium in a phylogenetic 
clade between Ehrlichia and Anaplasma, was originally identified in 
Ixodes ricinus ticks from the Netherlands and in mice and Ixodes ovatus 
ticks from Japan. By means of broad-range 16S rRNA gene amplifica­
tion and sequence analysis, this organism was identified as the cause of 
severe and sometimes prolonged febrile illnesses in European immu­
nocompromised patients with tick bites or exposures and in Chinese 
patients developing a mild febrile illness after being bitten by Ixodes 
persulcatus and Haemaphysalis concinna ticks. The clinical presenta­
tion is similar to those of HME and HGA except for an often prolonged 
course and a predilection for venous thrombosis. Specific diagnostic 
methods have been developed but are not widely available.
TREATMENT
Ehrlichioses
Doxycycline is effective for HME as well as other ehrlichioses; the 
use of this drug in “Candidatus N. mikurensis” infection is associated 
with disease resolution. Therapy with doxycycline (100 mg given PO 
or IV twice daily) or tetracycline (250–500 mg given PO every 6 h) 
lowers hospitalization rates and shortens fever duration. E. chaffeen­
sis is not susceptible to chloramphenicol in vitro, and the use of this 
drug is controversial. While a few reports document E. chaffeensis 

persistence in humans, this finding is rare; most infections are cured 
by short courses of doxycycline continuing for 3–5 days after defer­
vescence. Although poorly studied for this indication, rifampin may 
be suitable when doxycycline is contraindicated.

Prevention 
HME, E. ewingii ehrlichiosis, E. muris ehrlichiosis, and 
“Candidatus N. mikurensis” infection can be prevented by the avoid­
ance of ticks in endemic areas. The use of protective clothing and tick 
repellents, careful postexposure tick searches, and prompt removal of 
attached ticks probably diminish infection risk.
■
■HUMAN GRANULOCYTOTROPIC ANAPLASMOSIS
Epidemiology 
As of November 2023, 64,225 cases of HGA had 
been reported to the CDC, most in the upper Midwest and north­
eastern United States. The global geographic distribution is similar to 
that of Lyme disease because of the shared Ixodes tick vectors. Natural 
reservoirs for A. phagocytophilum are small mammals, such as whitefooted mice and squirrels, and large animals such as white-tailed deer 
in the United States and red deer in Europe. HGA incidence peaks in 
May through July, but the disease can occur throughout the year with 
exposure to Ixodes ticks. HGA often affects males (59%), and the inci­
dence of infection in persons older than 60 years is 7 times higher than 
in those younger than 15 years.
Clinical Manifestations 
Seroprevalence rates are high in endemic 
regions; thus, it seems likely that most individuals develop subclinical 
infections. The incubation period for HGA is 4–8 days, after which the 
disease manifests as fever (75–100% of cases), myalgia (73%), headache 
(82%), and malaise (97%). A minority of patients develop nausea, vom­
iting, or diarrhea (22–40%); cough (27%); or confusion (17%). A rash 
in HGA (5%) most often reflects co-infection with Borrelia, resulting 
in erythema migrans. Most patients develop thrombocytopenia (80%) 
and/or leukopenia (63%) with increased serum hepatic aminotransfer­
ase levels (80%).
CHAPTER 192
Rickettsial Diseases
Life-threatening complications occur most often in the elderly and 
include renal failure, acute respiratory distress syndrome, a toxic 
shock–like syndrome, pneumonia, a DIC- or sepsis-like syndrome and 
hemophagocytic lymphohistiocytosis. Meningoencephalitis is rare in 
documented cases of HGA. Other documented neurologic sequelae 
include brachial plexopathy, cranial nerve involvement, and demyelinat­
ing polyneuropathy. Infection of patients with a preexisting immuno­
compromising condition (diabetes, immunosuppressive medications, 
asplenia, arthritis) is associated with a 3.0 relative risk for life-threatening 
complications. Of patients with HGA, 31% are hospitalized, and 7% 
require intensive care. The case–fatality rate is 0.3%, but the relative risk 
for death is 16 if infection occurs with an immunosuppressive condi­
tion. Neither vasculitis nor granulomas are components of HGA. While 
patients can be co-infected with Borrelia burgdorferi and Babesia microti 
(transmitted by the same tick vector[s]), there is little evidence that these 
infections increase the severity or persistence of HGA. HGA transmit­
ted by transfusion (including the transfusion of leukoreduced blood or 
platelets) has now been reported in at least nine cases, including a fatality.
Diagnosis 
HGA should be included in the differential diagnosis of 
influenza-like illnesses during seasons with Ixodes tick activity (May 
through December), especially in the context of a known tick bite 
or exposure. Concurrent thrombocytopenia, leukopenia, or elevated 
serum levels of alanine or aspartate aminotransferase further increase 
the likelihood of HGA. Many HGA patients develop Lyme disease 
antibodies in the absence of clinical findings consistent with that diag­
nosis. Thus, HGA should be considered in the differential diagnosis 
of atypical severe Lyme disease presentations. Peripheral-blood film 
examination for neutrophil morulae can yield a diagnosis in 20–75% 
of infections. PCR testing of blood from patients with active disease 
before doxycycline therapy is sensitive and specific. Serodiagnosis is 
retrospective, requiring a fourfold increase in A. phagocytophilum anti­
body titer (to ≥128) in paired serum samples obtained 1 month apart. 
Since seroprevalence is high in some regions, a single acute-phase titer 
should not be used for diagnosis.

Anaplasma capra and A. bovis Infection 
Human infection by A. 
capra, first isolated from goat blood, was identified in 28 patients from 
northeastern China. Patients presented with fever, headache, malaise, 
dizziness, myalgias, and chills, but these manifestations were less severe 
than in HGA. Hospitalization was recorded for 18% of patients, and 
14% had underlying disorders, including hyperglycemia, hypertension, 
coronary heart disease, diabetes, and cancer. Five patients had severe 
manifestations, including one with encephalitic signs and A. capra 
DNA present in CSF. A. capra is found most often in Haemaphysalis 
longicornis ticks, and at least seven additional species can carry it over 
a distribution across Asia and eastern Europe. All patients responded 
to doxycycline treatment and survived.

Anaplasma bovis was recently identified in four human patients in 
a surveillance study of suspected Lyme borreliosis, HGA, HME, and 
babesiosis in the United States and in two patients with fever, myalgias, 
and rash in China. All six patients were between the ages of 35 and 
71 years; four were male, and two were female. The apparent tick vec­
tors are Dermacentor variabilis in the United States and Haemaphysalis 
longicornis in China. Both Chinese patients recovered after antimicro­
bial treatment.
TREATMENT
Human Granulocytotropic Anaplasmosis
No prospective studies of therapy for HGA have been conducted. 
However, doxycycline (100 mg PO twice daily) is effective. Rifampin 
therapy is associated with improvement of HGA in pregnant women 
and children. Most treated patients defervesce within 24–48 h.
PART 5
Infectious Diseases
Prevention 
HGA prevention requires tick avoidance. Transmission 
can be documented as few as 4 h after a tick bite.
Q FEVER
The agent of Q fever is C. burnetii, a small pleomorphic coccobacillus 
with a gram-negative cell wall, that was first isolated in 1935 and called 
a rickettsia due to its presence in ticks, intracellular replication, small 
size, and staining characteristics, but it is now known to be genetically 
quite distinct from Rickettsiaceae and to have a number of unique 
features. It survives in harsh environments, escapes intracellular killing 
in macrophages by inhibiting the final step in phagosome maturation, 
and has adapted to the acidic phagolysosome.
Epidemiology 
Q fever is a zoonosis: transmission of C. burnetii 
to humans typically occurs by inhalation after it has been shed by 
animals. The primary sources of human infection are infected sheep, 
goats, and cattle. At parturition, when large amounts of C. burnetii are 
present in the fetus, placenta, membranes, and fluids, the bacterium 
readily contaminates the environment. Smaller amounts can be shed in 
milk for weeks to months after parturition, but also in urine and feces. 
Once shed, C. burnetii can remain viable in manure, hay, soil, etc., for 
many years, after which it can be aerosolized and inhaled, even after 
traveling miles from the source by wind. A variety of other vertebrate 
animals can be hosts of C. burnetii, including birds, cats, dogs, rabbits, 
skunks, raccoons, deer, bears, sloths, kangaroos, and marine animals. 
C. burnetii has also been found in several tick species, which could 
be important for maintenance of the agent in veterinary populations, 
but the majority of human Q fever cases are associated with aerosol 
transmission from infected livestock. Infections in animals are usu­
ally asymptomatic, but abortions and stillbirth have been observed in 
pregnant goats and sheep. Because it is easily dispersed as an aerosol 
and because of the extremely low infectious dose required for human 
infection (probably between 1 and 10 viable bacteria), C. burnetii is a 
potential agent of bioterrorism (Chap. S4), with a high infectivity rate 
and pneumonia as the major manifestation.
Persons most at risk for Q fever include abattoir workers, veterinar­
ians, farmers, and other individuals who have contact with infected 
animals (particularly newborn animals), but outbreaks often affect 
people without any livestock contact. In Canada and the Netherlands, 
65% and 72%, respectively, of persons living and/or working on dairy 

cattle farms were seropositive, and in the United States, 22% of veteri­
narians were seropositive, compared to ~3% of the population overall. 
Seroprevalence data confirm that Q fever is underreported in many 
countries. An outbreak of Q fever associated with ingestion of raw 
milk confirmed the oral route of transmission, although this route is 
uncommon. In rare instances, person-to-person transmission follows 
childbirth in an infected woman, surgery, autopsy, blood transfusion, 
or organ transplantation of an infected individual. Multiple outbreaks 
involving laboratory staff have been reported in the past. Some evi­
dence suggests that C. burnetii can be sexually transmitted among 
humans. Some unusual modes of C. burnetii transmission to humans 
include treatment with live fetal sheep cells and percutaneous infection 
after crushing an infected tick between the fingers.
Infections due to C. burnetii occur in most geographic locations 
except New Zealand and Antarctica. Several factors influence the epi­
demiology: environmental conditions such as high concentrations of 
animals, high animal pregnancy rates, dry weather, and the strength 
and direction of winds. In addition to differences between strains of C. 
burnetii, the variability in human susceptibility to C. burnetii can influ­
ence transmission and development of disease. Some people become 
sick after exposure, whereas others have only mild symptoms that are 
not sufficient to lead them to seek medical assistance, and ~60% have 
asymptomatic seroconversion. Q fever continues to be endemic in 
Australia and France. In Cayenne, French Guiana, Q fever is hyper­
endemic: 40% of all community-acquired pneumonias are caused by 
C. burnetii. The largest known outbreak occurred between 2007 and 
2010 in the Netherlands. Over 4000 cases were reported, and over 
40,000 people were infected. The outbreak was due to a combination 
of high-density goat farming in areas with large urban populations and 
environmental factors. Farms where spread did not occur had high 
vegetation density and lower groundwater concentrations.
Young age seems to be protective against disease caused by C. 
burnetii. In a large outbreak in Switzerland, symptomatic infection 
occurred five times more often among persons >15 years of age than 
among younger individuals. In many outbreaks, men are affected more 
commonly than women.
Clinical Manifestations 
• 
ACUTE Q FEVER  The incubation 
period is 3–30 days. The primary manifestations of acute Q fever dif­
fer geographically. During the Dutch outbreak, but also in Canada 
and Croatia, pneumonia is the more common presentation. In some 
countries where Q fever is endemic, such as France and Israel, hepatitis 
is more common. These differences could reflect the route of infec­
tion (i.e., ingestion of contaminated milk for hepatitis and inhalation 
of contaminated aerosols for pneumonia) or strain differences. In the 
Dutch outbreak, sequelae of infection in pregnant women were rare; 
this was not the case among pregnant women elsewhere. Pericarditis, 
myocarditis, acalculous cholecystitis, pancreatitis, lymphadenitis, spon­
taneous rupture of the spleen, transient hypoplastic anemia, hemolytic 
anemia, hemophagocytic lymphohistiocytosis, meningitis, encephalitis, 
optic neuritis, and erythema nodosum are less common manifestations.
The symptoms of acute Q fever are nonspecific; common among 
them are fever, extreme fatigue, photophobia, and severe headache that 
is frequently retro-orbital. Other symptoms include chills, sweats, nau­
sea, vomiting, and diarrhea. Cough develops in about half of patients 
with Q fever pneumonia. A nonspecific rash may be evident in 4–18% 
of patients. The WBC count is usually normal. Thrombocytopenia 
occurs in ~25% of patients, and reactive thrombocytosis frequently 
develops during recovery. Biochemical markers of autoimmunity, such 
as anticytoplasmic antibodies (ANCA), antinuclear antibodies (ANA), 
anti–smooth muscle antibodies, or antiphospholipid antibodies, are 
often present in acute Q fever. Chest radiography can show opacities 
similar to those seen in pneumonia caused by other pathogens. Case–
fatality rate of acute Q fever is ~1–2%.
Acute Q fever occasionally complicates pregnancy. In one series, it 
resulted in premature birth in 35% of cases and in abortion or neo­
natal death in 43%. Neonatal death and lower infant birth weight are 
reported up to three times more often among women seropositive for 
C. burnetii in some areas but not others.

Rickettsial Diseases

CHAPTER 192
TABLE 192-2  Diagnostic Criteria for Chronic Q Fever as Defined by the 
Dutch Q Fever Consensus Group
PROVEN CHRONIC 
Q FEVER
PROBABLE CHRONIC 
Q FEVER
POSSIBLE CHRONIC 
Q FEVER
1.  Positive Coxiella 
burnetii PCR in 
blood or tissuea
or
2.  IFA ≥1:800 or 1:1024 
for C. burnetii phase 
I IgG
and
Definite endocarditis 
according to the 
modified Duke criteria
or
Proven large vessel 
or prosthetic infection 
by imaging studies 
(18FDG-PET, CT, MRI, 
or AUS)
IFA ≥1:800 or 1:1024 for C. 
burnetii phase I IgG
AND AT LEAST ONE OF THE 
FOLLOWING:
Valvulopathy not meeting the 
major criteria of the modified 
Duke criteria
Known aneurysm and/or 
vascular or cardiac valve 
prosthesis without signs of 
infection by means of TEE/TTE, 
18FDG-PET, CT, MRI, or AUS
Suspected osteomyelitis or 
hepatitis as manifestation of 
chronic Q fever
Pregnancy
Symptoms and signs of 
chronic infection, such as 
fever, weight loss, night 
sweats, hepatosplenomegaly, 
and persistently raised ESR 
and CRP
Granulomatous tissue 
inflammation, proven by 
histologic examination
Immunocompromised state
IFA ≥1:800 or 1:1024 for 
C. burnetii phase I IgG 
without manifestations 
meeting the criteria 
for proven or probable 
chronic Q fever
aIn absence of acute infection.
Abbreviations: AUS, abdominal ultrasound; CRP, C-reactive protein; CT, computed 
tomography; 18FDG-PET, fluorodeoxyglucose positron emission tomography; 
ESR, erythrocyte sedimentation rate; IFA, indirect fluorescent antibody assay; 
MRI, magnetic resonance imaging; PCR, polymerase chain reaction; TEE, 
transesophageal echocardiography; TTE, transthoracic echocardiography.
Q FEVER FATIGUE SYNDROME  Prolonged fatigue follows acute Q fever 
in up to 20% of cases and can be accompanied by a constellation of 
symptoms, including headaches, sweats, arthralgia, and myalgias. Sev­
eral hypotheses regarding the etiology exist, including host and genetic 
factors, cytokine dysregulation, and a partly biopsychological etiology 
with C. burnetii acting as a trigger for fatigue development. A random­
ized controlled trial including 155 patients with strictly diagnosed 
Q fever fatigue syndrome showed that long-term treatment with doxy­
cycline did not reduce fatigue severity compared to placebo, so antibi­
otics should not be prescribed for these patients. Cognitive behavioral 
therapy aimed at fatigue-related cognitions and behaviors thought to 
perpetuate symptoms was effective in reducing fatigue severity in the 
short term. The beneficial effect of this treatment, however, was not 
maintained after 1 year.
CHRONIC Q FEVER  Following symptomatic or asymptomatic primary 
infection, 1−5% of all patients develop chronic Q fever. Chronic Q 
fever endocarditis, infected aneurysms, or infected vascular prosthe­
ses are most frequently observed. The primary infection often has 
not been recognized or was asymptomatic, and the median duration 
between primary infection and manifestation of chronic infection is 
13 months, with the largest observed interval between acute infection 
and diagnosis of chronic Q fever being >9 years. Risk factors for the 
development of chronic Q fever include male sex, older age, valvu­
lopathy or prior valve surgery, aneurysms, vascular prostheses, renal 
insufficiency, and immunocompromised state. Diagnosing chronic Q 
fever is difficult, as patients are sometimes asymptomatic until com­
plications occur and often present with nonspecific symptoms, such 
as fever, night sweats, weight loss, fatigue, and malaise. Fever can be 
absent and is frequently low grade. C-reactive protein is often low or 
even normal. Chronic Q fever endocarditis differs from endocarditis 
caused by other bacteria, manifesting as endothelium-covered nodules 
on the valves, aortic root abscess, or new or rapidly increasing valve 
insufficiency. A high index of suspicion is necessary for timely diag­
nosis. Patients with chronic Q fever are often ill for >1 year before the 
diagnosis is made. The disease should be suspected in all patients with 
culture-negative endocarditis. In addition, all patients with valvular 
heart disease, an aneurysm or vascular prosthesis and unexplained 
weight loss, fever, elevated inflammatory markers, stroke, unexpected 
aneurysm growth, and/or progressive heart failure should be tested for 
C. burnetii infection. Other manifestations of chronic Q fever include 
lymphadenitis and bone infection including vertebral osteomyelitis 
and prosthetic joint infection. Of 313 patients with proven chronic 
Q fever in the Netherlands, 65% developed complications. The most 
frequently observed complications were acute aneurysms, heart failure, 
and noncardiac abscesses. One in six patients with vascular chronic Q 
fever develops arterial fistula, including aortoenteric fistula, aortocaval 
fistula, aortobronchial fistula, and arteriocutaneous fistula. PCR posi­
tivity at any time during the disease, presence of prosthetic material, 
and older age were associated with complications. Q fever–related 
mortality was 25% in patients diagnosed with proven chronic Q fever 
after the Dutch outbreak. Chronic Q fever–related mortality was high­
est in patients with both endocarditis and vascular infection (33%), 
followed by patients with vascular infection only (25%), and was lowest 
in endocarditis patients (12%).
Diagnosis 
Culture of C. burnetii from buffy-coat blood samples or 
tissue specimens is possible but requires a biosafety level 3 laboratory 
and is not used in clinical practice. PCR detects C. burnetii DNA in 
blood and tissue specimens, including paraffin-embedded samples. 
The detection of antibodies to C. burnetii is the most commonly used 
method for the diagnosis of Q fever. Available serologic assays are 
indirect fluorescent antibody (IFA) assay, enzyme-linked immuno­
sorbent assay (ELISA), and complement fixation test (CFT), with IFA 
being the gold standard. IFA tests are useful for the detection of and 
discrimination between acute and chronic infection and have excellent 
sensitivity and specificity. The diagnosis of acute Q fever is dependent 
on seroconversion, defined as a fourfold increase in IgG titer for phase 
II antigens between acute- and convalescent-phase samples. In the first 
1−2 weeks of illness, PCR on blood or serum can also be positive. 
A high phase I IgG titer (e.g., >512) is suggestive of chronic Q fever, but 
alone, it is not enough for a definite diagnosis. A positive PCR for C. 
burnetii in blood or tissue in the absence of an acute infection confirms 
the diagnosis, but PCR on blood is negative in the majority of patients 
and tissue samples are often very difficult to obtain. The diagnosis of 
chronic Q fever should be based on a combination of clinical, labora­
tory, and imaging criteria. There has been debate on the optimal set of 
criteria, but the Dutch literature-based consensus guideline (Table 192-2) 
appears to be the most sensitive and is easy to use.
Valvular vegetations are detected in only 12% of patients with Q 
fever endocarditis by transthoracic echocardiography, but the rate of 
detection is higher (21–50%) with transesophageal echocardiography. 
Fluorodeoxyglucose positron emission tomography combined with 
computed tomography (FDG-PET/CT) can detect not only prosthetic 
valvular infection but also intravascular infection elsewhere, osteo­
myelitis, and lymphadenitis. In native valve endocarditis, specificity is 
very high but sensitivity is low, so a normal FDG-PET/CT scan cannot 
exclude native valve endocarditis. A study including 273 FDG-PET/
CT scans performed at diagnosis in patients suspected of chronic Q 
fever showed that, even after serology, PCR, and often ultrasound or 
CT had been performed, FDG-PET/CT led to a change in diagnosis or 
treatment in 20% of patients. Adding FDG uptake in a heart valve as a 
major criterion to the Duke criteria led to a 1.9-fold increase of definite 
endocarditis diagnoses. Of 218 scans performed during follow-up, 57% 
resulted in treatment adjustment. In case of suspected chronic Q fever, 
FDG-PET/CT should be considered.
TREATMENT
Q Fever
ANTIBIOTICS
In vitro, C. burnetii is susceptible to doxycycline, quinolones, 
trimethoprim-sulfamethoxazole (TMP-SMX), macrolides, and