Q Fever in humans, domestic animals and wildlife

Epidemiology

Although C. burnetii have a near global distribution, the epidemiology of Q fever is highly variable geographically, which may reflect differences between strains, host biology and ecology, and environmental conditions.⁴ Coxiella burnetii have an unusually wide range of potential hosts and animal reservoirs, involving several taxa of vertebrate and invertebrate species, as well as amoebae.^1,4,5 Their ability to adapt to and persist in so many host types and under varied environmental conditions has probably contributed to their widespread geographical distribution. Domestic livestock are the primary reservoir for human infections and humans are usually considered incidental hosts, with human-to-human infection exceedingly rare.^1,4,5,7 Although most human cases are sporadic, with outbreaks typically small and limited, with a common point source, large-scale outbreaks can occur.^7,8

Shedding of C. burnetii has been demonstrated in the birth products, milk, urine, faeces, semen and vaginal secretions of infected animals.^1,4,5,9,10 Inhalation of contaminated aerosols is the primary mode of transmission to humans, most commonly after parturition or abortion in infected domestic animals.^1,8 Other sources include the slaughter and dressing of infected carcasses, wool shearing or aerosolisation of contaminated faecal material.^11,12 Air samples have been found to contain C. burnetii for several weeks or months after parturition in infected herds^10,13 and aerosols can be carried by wind over several kilometres.^14,15 Other possible routes of transmission that have been explored experimentally include arthropod vectors, ingestion, sexual contact and vertical transmission to offspring, but the epidemiological significance of these routes is likely to be low or negligible.^16,17 Because livestock are a common reservoir, the groups at risk of acquiring Q fever include abattoir workers, veterinary personnel, farm workers, livestock handlers and wool shearing staff,^1,18 whereas large outbreaks have been reported in communities living in relative proximity to high-density livestock operations.^7,8,15,19 Cases of Q fever have also been reported in those assisting cats during parturition²⁰ and in Australia, there have been increasing numbers of human cases associated with direct or indirect exposure to wildlife, rather than the traditional risk factors, pointing towards the likelihood of wildlife reservoirs of infection.²¹

Indirect methods

Serology is the easiest way to detect exposure to C. burnetii in unvaccinated individuals or animal populations, as antibodies tend to persist for extended periods after infection. Serological assays also have the benefit of being cheap and reasonably easy to perform. The main disadvantage of serology is that a positive result merely indicates historical exposure to the pathogen, which may or may not reflect the current disease status. Serial testing to demonstrate a rising antibody titre is necessary to differentiate a recent infection from a historical one, which can be logistically challenging in animals, particularly in wildlife.

The earliest techniques used to identify antibodies against C. burnetii were complement fixation tests (CFTs) and agglutination (AG) tests. These were commonly used for screening animal sera, regardless of species, but are insensitive, particularly in birds.^6,25,26 As a result, they have largely been superseded by more sensitive tests, such as enzyme-linked immunosorbent assays (ELISAs) and indirect immunofluorescence assays (IFAs).⁶ IFAs are the preferred (reference) test for detecting antibodies against C. burnetii in humans. Although they are more laborious to perform than ELISAs and require a skilled technician for accurate interpretation, IFAs have frequently been shown to have higher diagnostic accuracy in human studies.^27,28 Conversely, the World Organisation for Animal Health recommends the ELISA for veterinary use,⁶ although findings and recommendations about their diagnostic accuracy vary considerably between kits and host species in different studies.^29–31 Although there are several commercially available ELISA kits available for use in domestic ruminants, few have been validated for use in other species.

Direct methods

Tests to detect the pathogen directly are the most reliable method to demonstrate an active (current) infection. However, the sensitivity of these tests may be limited if the pathogen is localised in specific tissues or if the duration of bacteraemia or infection is short, necessitating careful selection and timing of sample collection.

Although bacterial culture and isolation could be considered the most definitive test for diagnosis of infection, culture of C. burnetii is difficult and laborious. Its high infectivity restricts culture and isolation to facilities with appropriate levels of biological containment. Culture has historically relied on inoculation of cell cultures, embryonated chicken eggs or laboratory animals, but more recent advances have facilitated in vitro growth in axenic medium.³²

Visualisation of bacteria in situ may be possible in smears from clinical samples or in histopathological sections. Although C. burnetii are Gram-negative organisms, they stain poorly with Gram stain.³³ However, they can be well visualised in clinical samples and cultures using Gimenez, Macchiavello or Giemsa stains, providing the bacterial load is sufficient (>10⁵ bacteria mL^–1).^6,33 Immunohistochemistry (IHC) or, less commonly, fluorescence in situ hybridisation (FISH) techniques are more specific methods for confirming the presence of C. burnetii in clinical samples, although the stains and controls required may be less readily available.^34,35

Polymerase chain reaction (PCR) assays are the direct method most commonly used to demonstrate active infection with C. burnetii, because of their greater sensitivity.⁶ In addition, PCR assays can have a high throughput, a quick turnaround time and have a relatively greater ease of use than other methods. A wide range of protocols have been described, targeting a variety of plasmid and chromosomal genes, with varying levels of sensitivity and specificity. Common targets include the transposase-like insertion sequence (IS1111), the isocitrate dehydrogenase gene (icd), the superoxide dismutase gene (sodB), the outer membrane protein 1 gene (com1), the outer membrane protein A gene (OmpA), heat shock operon genes (htpA and htpB) and the 16S ribosomal RNA gene (16S rRNA).^4,36,37 In addition, use of real-time PCR assays enables quantification of the bacterial load in a sample, although the high genetic diversity between isolates may complicate quantification for some targets. For instance, the copy numbers of IS1111, which is frequently the target used in screening tests because it yields assays with a higher sensitivity, has been shown to vary from zero to over 100 between different isolates, making quantification based on this target unreliable.^37,38 Single copy targets (such as com1) are therefore more useful for quantification, although the sensitivity of these tests may be significantly reduced. However, care is needed because of the potential for cross-reactivity with Coxiella-like tick endosymbionts, many of which have similar genes to C. burnetii and are still poorly characterised.¹⁶

Table 1 provides details of the sensitivity and specificity of each of the diagnostic tests described above. Fig. 1 shows the change in IgA, IgG and IgM responses and quantitation of C. burnetii by real time PCR, as a function of the number of days since onset of symptoms or clinical signs in in infected humans and animals.

Interpretation of PCR and serology test results using the guidelines listed above may assist in providing an estimate of the date of onset of clinical signs or exposure, where this is unknown by the human patients or animal owners. This may allow a time window of infection to be determined and assist with the identification of a likely source. In turn, this can assist in implementation of control measures to manage or eliminate spread infection.