Coxiella

TAXONOMY Back to top

Coxiella burnetii is a small gram-negative rod that grows within a parasitophorous vacuole

located in the cytoplasm of a host eucaryotic cell (invertebrate or vertebrate animal). It is a

member of the γ subgroup of theProteobacteria microbial phylum. The closest related

bacterium is in the genus Legionella. Bergey’s Manual of Systematic Microbiology (2005)

classifies it under the order Legionellales, family Coxiellaceae (7). The only other member of

the genus Coxiella is the proposed “C. cheraxi” bacterium, a pathogen of the Australian

freshwater crayfish Cherax quadricarinatus (14).

DESCRIPTION OF THE AGENT Back to top

C. burnetii consists of different morphological forms, depending on the stage of its life cycle.

Large-cell variants (LCV; 0.4 to 1.5 μm by 0.2 to 0.5 μm) are metabolically active and divide

by binary fission inside the parasitophorous vacuole of the host eukaryotic cell. Small-cell

variants (SCV; 0.5 μm by 0.2 μm) are electron dense and form when conditions are no

longer conducive to active growth (13) (Fig. 1). SCV are filterable (0.22 μm), and this

quiescent form of the cell differentially expresses certain proteins compared to LCV

(26,64, 66, 79). SCV are functionally spores, although chemically different from the grampositive

bacterial spore, as it lacks diaminopimelic acid. They act as the survival and

transmissible form of the bacterium when it is extracellular and in the environment. SCV can

survive in animal-contaminated environments (e.g., soil, hay, etc.) for many years, probably

decades. However, they are not as heat stable as normal bacterial spores and can be

inactivated at 63°C for 40 minutes (56). The cell wall, while gram-negative, stains poorly by

Gram stain and better by Gimenez stain.

The complete genome sequences of Nine Mile and other strains of C. burnetii have

demonstrated a circular genome with approximately 2 million base pairs, including many

insertion sequences, and a single plasmid is found in most strains (65). The presence of

many pseudogenes implies a process of ongoing gene degradation presumably associated

with an evolutionarily recent adaptation to an intracellular lifestyle. Unlike

the Rickettsiaceae, C. burnetii does not transport ATP across its cell membrane and has

almost full biosynthetic capabilities. It is metabolically active in the extracellular phase,

utilizes glucose and glutamate at low pH, and has recently been grown in cell-free medium

(48). Using microarray and whole-genome sequence analysis, variability in open reading

frames and transposon-mediated genomic plasticity have been demonstrated (4).

The virulent form of C. burnetii is referred to as “phase I” because it is first isolated from

humans with Q fever, infected vertebrate animals (especially cattle, sheep, and goats), and

ticks. When these isolates are grown in the laboratory in tissue culture or embryonated eggs,

the population of bacteria gradually change to a second form (“phase II”) and become

avirulent. This population change can involve loss of genetic material (29), such that the

microbe cannot synthesize a full-length polysaccharide, lacking a terminal sugar chain as

part of its cell wall lipopolysaccharide (LPS) (78). However, in some strains, loss of virulence

does not seem to involve genomic changes (15), presumably due to changes in the

expression of genes coding for virulence determinants. The in vitro change from virulent

phase I to avirulent phase II is analogous to the “smooth”-to- “rough” transition that occurs

in bacteria of the Enterobacteriaceae group. In phase I cells, the terminal glycan chain of the

LPS contains three unique sugars, L-virenose, dihydrohydroxystreptose, and

galactosaminuronyl-α-(1, 6)- glucosamine, that are not present in phase II LPS (1). Phase I

LPS appears to be a key virulence determinant of C. burnetii.

EPIDEMIOLOGY AND TRANSMISSION Back to top

Coxiella burnetii is associated with vertebrate animals, especially cattle, sheep, and goats. At

parturition, when large concentrations of C. burnetii are present in the placenta, fetus, and

associated membranes and fluids, the microbe readily contaminates the animal’s

environment. It can remain viable in soil, hay, etc., for many years (possibly decades),

presumably in its “spore-like” form (63, 85). Milk from infected cows and other lactating

animals may contain C. burnetii (30), which is destroyed by the temperature reached during

pasteurization (39). Goats are a significant source of infection (19).

A number of other vertebrate animals can be hosts for C. burnetii, especially native animals,

for example, native rats, wombats, bandicoots, and kangaroos in Australia. Infections have

been described for pets, including cats and dogs, with human outbreaks in North America

linked to parturition by these animals (8, 32,53). Birds are also hosts (70), and various other

domestic and wildlife species have been reported as potential host species, including mice,

horses, rabbits, and buffalo (43).

Numerous tick species either harbor or transmit C. burnetii and may be important for

maintenance of the agent in veterinary populations, but tick transmission is not considered a

major route of transmission to humans. Although there are a few reports of probable human

infection by tick bite, the great bulk of human infections are by aerosol transmission from an

infected animal focus, such as a herd of parturient goats. Infection in animals appears to be

almost always asymptomatic, with recrudescence only during parturition usually followed by

recovery of the maternal animal. Australian studies showed the importance of the bandicoot

(Isoodon macrourus) (a marsupial) and its tick Haemaphysalis humerosa (17) and the

kangaroo and its tick Amblyomma triguttatum (54) in sylvatic cycles of C. burnetii in

Australia.

Derrick described Q fever in a large number of Australian patients, including some probably

infected by tick bite, although most were infected by inhalation of dust associated with cattle

(16). Now, Q fever is thought to be endemic in every country except New Zealand (28). In

the United States, 22% of veterinarians are seropositive compared to approximately 3% of

the population overall (2, 86). Travelers and military personnel can also be infected (12, 25).

Transmission by aerosol and wind is well recognized (73). Nosocomial transmission,

presumably by coughing with aerosol production, has rarely been reported (49), as has

infection of surgical and obstetric staff during a Caesarean section on an infected patient

(60). Sexual transmission has also been noted, as presumably C. burnetii was present in

semen, from a chronic focus in the prostate gland or testis (45).

The epidemiology of human Q fever is a combination of the worldwide ubiquitous distribution

of C. burnetii; the extremely low infectious dose required for human infection (probably

between 1 and 10 viable C. burnetii cells) (72); the environmental conditions favoring

transmission, such as high concentrations of infected animals, high pregnancy rates,

appropriate environmental conditions (transmission is greater under dry conditions), and the

strength and direction of prevailing winds (73); and the inherent variability in human

susceptibility to C. burnetii. While some people are exposed and become sick, others are

exposed and seroconvert asymptomatically or have only mild symptoms, not sufficient to

seek medical assistance. The proportion of persons that become ill after natural exposure

may be as low as 50% (18, 76).

CLINICAL SIGNIFICANCE Back to top

Q fever can present in many forms: (i) as an acute undifferentiated febrile illness, (ii) as a

chronic infection (usually involving the cardiovascular system), or (iii) as a postinfectious

chronic fatigue syndrome (only recently recognized). Q fever can be latent and recrudesce

during periods of relative immunosuppression, such as late pregnancy, causing fetal

infection. Infection can also result in asymptomatic seroconversion and complete clearance

of the microbe, with the patient being unaware of infection or being only mildly ill.

Other features of Q fever include a higher incidence of symptomatic disease in men than

women and a higher incidence in middle-age men than those of other age groups. Although

it is claimed that occupational and exposure differences explain the gender differences, it is

probably not the complete picture. For example, female hormones appear to be protective in

mice (59).

Pathogenesis

Because C. burnetii is an intracellular pathogen, growing in a membrane-bound vacuole in

the cytoplasm of a host cell, its survival in the host animal (or patient) depends on its ability

to survive and grow intracellularly in the host cell (82). This in turn depends on its ability to

keep the host cell alive, and this requires microbe-directed immunomodulation of the host

(84). Normally a host cell would deal with an invasion by an intracellular microbe by

activating the host-cell self-destructing apoptotic process. Indeed, this is what happens in

phase II (avirulent) C. burnetii infection. The microbe is rapidly internalized (involving

receptor α Vβ3 integrin and complement receptor 3) (10), and it grows until the host’s cellmediated

immune response (involving gamma interferon and other molecules) induces

apoptosis and the infected host cell is destroyed, along with the microbe. Certain Toll-like

receptors are involved in the initial microbe-host cell interaction (88).

However, in the case of phase I (virulent) C. burnetii, a different sequence of events occurs,

leading to the survival and growth of the microbe (36, 81). The type 1 LPS fails to activate

dendritic cells (67), and the initial contact between the host cell membrane and the

bacterium bypasses complement receptor 3 (44). This leads to a different sequence of

intracellular events. Genes of a type IV secretion system (similar to the Dot/Icm system

of Legionella pneumophila) are expressed, leading to secretion of bacterial proteins into the

cytoplasm of the host cell (89). These proteins divert the normal intracellular autophagy

pathway, which results in the formation of a parasitophorous vacuole. It gradually enlarges

by incorporating recycled endoplasmic reticulum membrane into its own vesicular

membrane, allowing the phase I C. burnetii to grow. The inhibition of apoptosis is mediated

by host kinases (80), which allow the microbe to survive by preserving the infected host cell.

The metabolically active C. burnetii LCV continues to grow in the absence of any effective

host immune response despite the presence of circulating antibodies and cell-mediated

immunity mediated by T lymphocytes. Eventually the conditions inside the parasitophorous

vacuole become unsuitable for ongoing logarithmic bacterial growth, presumably due to

nutrient depletion, and C. burnetii converts to the SCV, the nonreplicating survival form.

Even then, the host cell may not be destroyed, leading to a state of chronicity or latent

infection. Host responses keep the microbe in check, but should immunosuppression

develop, such as during pregnancy, the bacterium starts growing again, causing a Q fever

relapse. This is well recognized during the third trimester of pregnancy. In patients with

chronic Q fever, there is impaired maturation of phagolysosomes, permitting ongoing

survival of C. burnetii (23).

Acute Q Fever

Q fever is a difficult disease to diagnose, as there are no pathognomic symptoms or signs

that give health care providers a clue to the etiology. Many doctors rarely consider Q fever in

the differential diagnosis of an acute febrile illness unless a link with animal contact is

established from the patient’s history. In fact, many patients without any significant animal

contact or tick bite develop Q fever, due to its dispersal by wind (73). Living downwind of a

herd of parturient animals, an animal-holding yard, or an abattoir is a risk for Q fever.

Presenting features include fever, headache, myalgia, elevated liver transaminases, and

interstitial pneumonia. It is claimed that the disease may differ in its presentation in different

countries, especially with respect to pneumonia.

The great diversity in acute symptoms is probably due to (i) differences in strains of C.

burnetii; (ii) differences in human immune response genes and how they process and

eliminate C. burnetii (27); (iii) route of infection (patients infected by the respiratory route

[the most common route] are likely to develop pneumonia, and patients infected by other

routes [e.g. tick bite, oral, sexually transmitted, and needle stick accident] can manifest the

illness differently); and (iv) infecting dose, since the dose of C. burnetii (phase I) needed to

infect a person is between 1 and 10 bacteria. An increased dose leads to a reduced

incubation period (72). Infecting dose is likely to influence the symptoms and clinical course

of the illness. Those with higher infecting doses are more likely to have severe symptoms.

Reviews of Q fever from Australia (16, 55, 68) and elsewhere (41, 43,57) show the diversity

of symptoms in this disease.

Occasionally, acute Q fever can be fulminant (46). However, most patients survive acute Q

fever and defervesce in about 10 to 14 days, at which time they develop either sterilizing or

nonsterilizing immunity. It is the latter patients who can go on to develop chronic Q fever.

Chronic Q Fever

As a result of the early dissemination of C. burnetii, many organ systems are exposed and

can become chronically infected. The cardiovascular system is particularly susceptible. Most

cases of chronic Q fever involve endocarditis (5), including infection of congenitally abnormal

(e.g., bicuspid) or previously damaged aortic and mitral valves, aneurysms, and vascular

grafts. Pericarditis, myocarditis, and splenic rupture have been reported. Other systems that

are often involved in chronic Q fever include the gastrointestinal tract, with chronic

granulomatous hepatitis (Fig. 2), acalculous cholecystitis, and diarrhea; central nervous

system, characterized by meningitis and meningoencephalitis; and bones and tendons, with

infections. Rare features include lymphadenopathy, migratory thrombophlebitis, rash, and

prostatitis. Many of these pathogenic features appear to involve autoimmunity, and the

presence of autoantibodies is a feature of chronic Q fever (9, 87).

Pregnancy and Q Fever

Q fever in pregnancy is an underrecognized problem rarely mentioned as a cause of

congenital infection, yet it is clearly a problem in many countries where Q fever is significant.

As pregnancy develops, latent, viable C. burnetii starts growing in the placenta and the

fetus, leading to infection and fetal death (71). While not recognized as one of the classical

“TORCH” agents of congenital infection, it should be included under “O” (for “other”).

Q Fever in Children

Children appear to be less susceptible to symptomatic Q fever than adults, as do young mice

compared to older mice (34). Nevertheless, infections, mainly febrile or influenza like, have

been reported (37, 47, 75).

Post-Q Fever Fatigue Syndrome

While a postinfectious fatigue syndrome, lasting weeks to months after an infectious disease

is well recognized, post-Q fever fatigue syndrome is not yet universally acknowledged. First

described in 1996 (3, 40) in the United Kingdom and Australia, post-Q fever fatigue

syndrome is defined as fatigue persisting more than 12 months after the onset of acute Q

fever.

COLLECTION, TRANSPORT, AND STORAGE OF

SPECIMENS Back to top

Specimens for the diagnosis of Q fever in humans include blood and tissue, the latter most

commonly from heart valves following valve replacement surgery. The sample collected will

depend on the diagnostic test(s) available. Whole blood may be used for isolation and nucleic

acid detection methods, and serum or plasma for serologic methods. Tissue samples are

most commonly used for isolation, PCR, and immunohistochemistry (IHC). Whole blood may

be collected in sodium citrate or EDTA tubes. Samples to be used for isolation should be

collected aseptically and shipped promptly, while maintaining refrigeration. If storage of

specimens prior to culture is necessary, samples should be kept frozen before and during

shipment (at least −20°C;−80°C is preferable). Blood or tissue samples to be tested by PCR

or IHC should be frozen (−20°C) prior to and during shipment, although tissues fixed by the

diagnostic laboratory for IHC may be shipped at room temperature.

For the diagnosis of a suspected acute infection by molecular methods or isolation, whole

blood should be collected during the acute phase, preferably prior to antibiotic therapy. For

serologic diagnosis of an acute infection, a serum sample should be collected during the

acute phase and a second sample at 3 or 4 weeks after onset. While the same specimens

may be used for the diagnosis of chronic infection as for acute infection, the timing of

collection is not as critical. Blood and tissue samples may be persistently positive by culture

and/or PCR in chronic infections, and serum antibody levels are typically elevated (phase I

and phase II immunoglobulin G [IgG] titers of >1,000) and sustained relative to acute

infections, which have generally lower peak titers that decrease postinfection.

C. burnetii can be isolated from blood or tissue samples, but since it is an obligate

intracellular bacterium, this must be done in cell culture or embryonated chicken eggs or by

animal inoculation. Isolation must currently be performed in specialized high-containment

biosafety level 3 (BSL-3) facilities, as the agent is highly infectious and classified as a select

agent and a CDC category B bioterrorism agent (86a). If an isolate is propagated by a

diagnostic laboratory, U.S. federal regulations require that it be transported to a registered

select agent laboratory or destroyed, within 7 days (National Select Agent Registry, Centers

for Disease Control and Prevention; www.selectagents.gov/cdForm.html). Diagnostic

samples can be evaluated by PCR and serologic methods in BSL-2 facilities with the use of

appropriate personal protective equipment.

DIRECT EXAMINATION Back to top

Microscopy and Antigen Detection

IHC is an excellent method for the detection of C. burnetii antigens in tissue samples (Fig.

3), particularly cardiac valve tissues that are colonized during chronic Q fever. Organisms in

heart valve tissues have also been demonstrated by direct immunofluorescent methods or

visualized by electron microscopy (Fig. 1). However, these methods are rarely used for the

diagnosis of acute Q fever, as the appropriate tissue samples are not often collected and will

vary among patients and because other methods (PCR and serology) that are simpler to

perform on blood samples are available.

Nucleic Acid Detection

PCR can be a useful diagnostic tool for acute and chronic Q fever infections (62). It is

important that the sample is collected during the early period of an acute infection while the

patient is bacteremic (optimally within 4 weeks of onset of symptoms). A recent study

showed that PCR can be more sensitive than serology during the first 2 weeks after the onset

of symptoms and provide an earlier diagnosis than with serology alone (22). Whole blood is

most commonly used for the analysis of acute infections, although enrichment of the white

cell fraction (buffy coat) may increase sensitivity. Serum may be used if whole blood is not

available, although it is less likely to be positive due to the lack of infected cells. For chronic

Q fever with endocarditis, the valve tissue is typically positive, while blood may be positive or

negative. It has been reported that PCR is generally positive in chronically infected patients

with phase I IgG antibody titers between 1:800 and 1:6,400, but PCR is often negative in

those with higher titers. A number of PCR assays are described that amplify the multicopy

IS1111 insertion sequence (31, 50), and these generally provide increased sensitivity

compared to assays that amplify single-copy genes (com1, 16S, 23S, etc.). However, the

potential for false positives with PCR makes it imperative that these results be interpreted

relative to other diagnostic assays, such as serology and other clinical data.

ISOLATION PROCEDURES Back to top

Isolation of C. burnetii from human blood or tissue must be performed in a BSL-3

containment facility due to the low infectious dose of the agent and potential for generating

aerosols. Isolation can be accomplished in tissue culture cells or embryonated chicken eggs

or by inoculation into animals such as mice or guinea pigs; the last requires BSL-3

containment facilities. Any infected tissue sample can be used for isolation, and PCR assays

are quite useful for screening tissue samples prior to isolation to determine those potentially

containing organisms. The organism can be stable in tissue samples for months before

isolation attempts. Animal inoculation is the most sensitive method for isolation. A mouse

can be injected intraperitoneally with up to 0.5 ml of inoculum, and the spleen harvested at

10 to 14 days postinjection. The spleen is homogenized and injected into another mouse,

inoculated onto uninfected tissue culture cells, or used to infect embryonated eggs for further

propagation. Isolation by animal inoculation is particularly useful for tissue or environmental

samples that could be contaminated with organisms other than C. burnetii, as the animal

serves to amplify Coxiella while killing other agents. Aseptically collected tissues that likely

contain only C. burnetii can be inoculated directly onto tissue culture cells or embryonated

chicken eggs. A shell vial method that works well is described for isolation in human

embryonic lung fibroblast tissue culture cells (61) and can be used with many commonly

available cell lines (e.g., Vero, RK13, THP1, and A549) susceptible to infection byC. burnetii.

Egg inoculations work particularly well for propagating large amounts of C. burnetii for

antigen preparation.

IDENTIFICATION Back to top

Only reference laboratories are likely to isolate C. burnetii in pure culture and be required to

confirm its identification by classical means. In recent years, identification is more often

accomplished by amplifying and sequencing key genes (e.g. com1 and IS1111a). A routine

diagnostic laboratory that inadvertently isolates C. burnetii in tissue culture could identify it

by direct fluorescent-antibody assay (Fig. 4), although specific antisera are not commercially

available.

TYPING SYSTEMS Back to top

Typing systems, based on genetic differences between isolates, are not standardized and are

still undergoing development. While no typing scheme is universally accepted, it is clear that

there is considerable strain/isolate variability within the species C. burnetii.

SEROLOGIC TESTS Back to top

The detection of antibodies to C. burnetii is the most commonly used and effective method

for the diagnosis of Q fever. The primary serologic assays in use today are the indirect

immunofluorescent antibody (IFA) assay, the complement fixation (CF) test, and the

enzyme-linked immunosorbent assay (ELISA), with the IFA assay being the gold standard

and most commonly used method. CF methods generally lack sensitivity and are used less

commonly today, while ELISAs are growing in use and availability. The diagnosis of infection

by any serologic assay may be complicated by the facts that C. burnetii has a worldwide

distribution in nature and many humans have been exposed and may be seropositive. All

serologic methods make use of C. burnetiiantigen grown in either tissue culture cells or

embryonated chicken eggs. Serologic methods also take advantage of the antigenic

differences between naturally occurring virulent phase I C. burnetii and attenuated phase II

isolates (69). Phase I strains contain intact LPS antigens, while phase II strains lack

complete LPS antigens. The antibodies produced during natural human infection respond in a

unique time sequence to the phase I and phase II forms, with the acute response directed

primarily to phase II antigens while the response in chronic infections is a mixture of phase I

and phase II antibodies.

IFA Test

IFA tests detect IgG, IgM, and IgA antibodies for both phase I and phase II antigens and

allow for the detection of, and discrimination between, acute and chronic infection (51). The

IFA test has excellent specificity and sensitivity for the diagnosis of Q fever if the appropriate

samples are available. The diagnosis of acute Q fever is dependent on seroconversion,

defined as a fourfold increase in the IgG titer for phase II between acute and convalescent

samples. Alternately, a single serum sample with a phase II IgG titer of ≥200 and a phase II

IgM titer of ≥50 has been used to diagnose acute Q fever (74). Chronic infections typically

present with phase I IgG titers of >800 (43). However, the choice of cutoff titer for acute or

chronic disease should be determined in each laboratory, as methods for antigen

preparations, assays, and interpretation can vary. Ideally, reference laboratories should

provide controls with known positive and negative samples that could be used for inlaboratory

assay validation and cutoff determination. FDA-approved, commercial C.

burnetii IFA tests are available for the diagnosis of acute Q fever (IgG and IgM assays for

phase II antigens) and provide the best source for laboratories not equipped to prepare their

own antigens (Focus Diagnostics, Cypress, CA). The diagnosis of chronic Q fever is best

performed at reference laboratories that have BSL-3 facilities for the propagation and

storage of phase I organisms since C. burnetii is classified as a U.S. category B bioterrorism

agent.

ELISA

ELISAs are reported to be as sensitive and specific as IFA tests (33, 52, 77, 83). FDAapproved,

commercial ELISAs that detect IgG or IgM antibodies to phase II antigens are

available (Inverness, Brisbane, Australia) and are useful for detection of IgM antibodies (20),

although not currently marketed in the United States. A number of reference laboratories

and research institutions have also developed in-house ELISAs. In general, ELISAs are

qualitative and have not been evaluated as thoroughly as IFA tests. Commercial and inhouse

ELISAs can be automated to require minimal interpretation and are particularly useful

for the detection of IgG antibodies to phase II antigens when employed in seroprevalence

studies. However, their lack of standardization and quantification make them less than ideal

for the diagnosis of acute Q fever, and the lack of phase I antigens in commercial assays

limits their use for the diagnosis of chronic Q fever.

CF Test

Whereas CF is still commonly used as a diagnostic assay for veterinary testing, the assay is

rarely used for human diagnostics due to advances in IFA and ELISA methods and the

demonstration that the CF test has lower sensitivity, is more time consuming, and detects

seroconversion at a later date than the other assays (21). False negatives with the CF assay

have also been described for chronic infections with high titers due to a prozone effect.

ANTIMICROBIAL SUSCEPTIBILITIES, TREATMENT, AND

PREVENTION Back to top

In vitro antimicrobial susceptibility testing is not routinely performed. C. burnetii is an

intracellular pathogen, and correlations between in vitro MICs of antibiotics in an infected

tissue culture system and antibiotic activity in infected patients are uncertain.

The recommended treatment for acute Q fever is doxycycline, although strains with partial

doxycycline resistance have been reported. Erythromycin and azithromycin are not

recommended, as many strains are resistant (MIC of >8 μg/ml) (35, 58). Neither

ciprofloxacin nor chloramphenicol are recommended, as they do not inhibit in vitro growth

(6, 61). Rifampin (rifampicin) may be used, although it is not recommended in the United

States. Treatment in pregnancy with co-trimoxazole is recommended, as adverse outcomes

are well recognized (11).

Chronic Q fever, especially Q fever endocarditis, requires long-term antibiotics and often

valve replacement. Doxycycline and hydroxychloroquine are recommended for a minimum of

1 year and possibly longer. Hydroxychloroquine increases the pH of the phagolysosome from

4.8 to 5.7 and renders doxycycline bactericidal rather than bacteriostatic (42).

A human Q fever vaccine (QVAX) is commercially available only in Australia (24, 38).

EVALUATION, INTERPRETATION, AND REPORTING OF

RESULTS Back to top

Serological and nucleic acid amplification (PCR) reports for a patient evaluated for Q fever

can be difficult to understand. Interpretation should always be provided and should include

(i) the significance of the antibody titer, particularly the difference between antibodies to

phase I and phase II C. burnetii and the different antibody classes detected (IgM, IgG, and

IgA); (ii) whether the serologic result supports recent Q fever (phase II IgM high), past Q

fever (phase II IgG high), or chronic Q fever (phase I IgG and IgA high); (iii) if there is only

one serum sample available for testing, the importance of receiving a second serum sample

to detect a fourfold increase in antibody titer (for example, a raised phase II IgM only in one

serum sample may represent very early acute Q fever or a false-positive result); (iv) in

locales where vaccination occurs or where the post-Q fever fatigue syndrome is considered, a

comment that no antibody pattern can currently define either condition should be provided;

and (v) an indication that detection of C. burnetii DNA in blood usually indicates acute, very

early Q fever, often before seroconversion. Tissue specimens (e.g., heart valve, liver, and

bone marrow) are most often positive in chronic Q fever, although peripheral blood

leukocytes from such patients can be positive or negative.