Abstract
Practical relevance:
There has been increasing identification of vector-borne pathogens in cats presented to veterinary clinics around the world for evaluation of fever and the associated secondary effects, such as signs of depression and loss of appetite.
Aim:
The aim of this article is to summarize the clinically relevant information concerning fever in cats that is associated with pathogens known or suspected to be vectored by fleas, with an emphasis on presenting clinical abnormalities and optimal diagnostic, treatment and prevention strategies. Fever in cats that is associated with pathogens vectored by ticks or sandflies is discussed in Part 2 of this article series.
Keywords
Introduction
The two major differentials for elevated body temperature >39.2°C (>102.5°F) in cats are hyperthermia and fever (pyrexia). 1 Hyperthermia can result from increased muscle activity, increased environmental temperature, stress or increased metabolic rate (eg, hyperthyroidism). With fever, the thermoregulatory set point in the hypothalamus is increased, secondary to the release of pyrogens, resulting in increased body temperature from physiologic mechanisms inducing endogenous heat production or heat conservation. Fever develops when leukocytes, particularly mononuclear cells and neutrophils, are activated to release pyrogens such as interleukin. 1
Leukocytes are generally stimulated by contact with bacterial, viral, fungal and parasitic agents, as well as by neoplasia, tissue necrosis (eg, extensive trauma, pancreatitis), and primary immune-mediated diseases such as immune-mediated hemolytic anemia, immune-mediated thrombocytopenia and systemic lupus erythematosus (see the algorithm for initial evaluation of cats with elevated body temperature on page 32). A variety of soluble factors such as interleukin 1 and tumor necrosis factor, which are released by the activated cells, enter the central nervous system and change the thermoregulatory set point. 2 The thermoregulatory set point may also be altered by intracranial disease including trauma and neoplasia, or drugs such as tetracycline. Shivering and vasoconstriction are two of the most important physiologic responses to a thermoregulatory set point change, and result in generation and conservation of heat, respectively.
The differential list for fever in cats is long. In a recent review of 106 referral feline cases with a body temperature >39.2°C, infectious causes of fever were the most common category. 3 There are a number of infectious agents that are known causes of (or are considered reasonable differential diagnoses for) fever in cats that are vectored by several arthropods.
The following discussion provides veterinarians with an update on the flea-borne agents that could be associated with fever in cats. The particular focus is common clinical and laboratory findings, optimal diagnostic tests, treatments and strategies for prevention (Table 1). Here, as well as within the accompanying update on the tick- and sandfly-borne disease agents, 4 year-round flea and tick control in endemic areas is emphasized, as it is better to prevent these infections than have to treat clinically ill cats.
Concurrent findings and diagnostic plan suggestions for feline flea-borne pathogens associated with fever
Results of direct tests (cytology ± staining techniques, culture, PCR assays) confirm infection when they are positive. However, for some agents, such as Bartonella species and the hemoplasmas, there is a carrier phase in many healthy cats and so positive test results do not confirm disease induced by the agent. Similarly, most positive antibody test results merely indicate past or current infection, but do not confirm current infection or disease
Flea-Borne Agents Associated with Fever
Bartonella species
A number of Bartonella species, including Bartonella henselae, Bartonella clarridgeiae, Bartonella koehlerae and Bartonella bovis, have been cultured, or their DNA has been amplified, from client-owned cats with fever.5–7 Fleas, in particular Ctenocephalides felis, are the known vectors for B henselae, B clarridgeiae and B koehlerae. 8 Bartonella quintana is associated with fever in humans, has been detected in cats and is transmitted among humans by lice. However, C felis has also been shown to acquire and excrete B quintana. 9 Evidence of exposure to Bartonella species has been found in cats of many countries, particularly in regions with high humidity and fleas.10–22
Fever following experimental inoculation with B henselae has been documented in a number of studies including one where 3/6 cats developed B henselae infection with associated fever that lasted at least 2 days after exposure to infected C felis. 23 The fever resolved in these three cats after enrofloxacin was administered. Several other studies have associated Bartonella species exposure or infection with fever in naturally exposed cats.5–7 However, most cats infected with a Bartonella species will remain clinically normal.
Whether fever will occur during Bartonella species infection is likely influenced by both host and organism factors in a complex interaction. Lymphadenopathy, endocarditis, myocarditis, hyperglobulinemia, osteomyelitis and uveitis are other well-documented manifestations of bartonellosis in cats. 24 The association with hyperglobulinemia and fever is notable because feline infectious peritonitis is also common in febrile cats with hyperglobulinemia. 3
As B henselae, B clarridgeiae and B koehlerae are transmitted by fleas, bacteremia and antibody-positive rates can be very high in flea-endemic areas. For example, serum antibodies were detected in 93% of cats housed in an animal shelter in North Carolina, USA. 25 The majority of these cats were thought to be normal, which emphasizes that fever from bartonellosis cannot be documented by test results alone. In one study of pair-matched cats with or without fever, serum Bartonella antibodies detected by ELISA or Western blot immunoassay were not correlated with the presence of fever. 6 In addition, serum antibody test results are negative in 3–15% of bacteremic cats, particularly in the acute phase of bacteremia. 23 Thus, if a cat with fever is to be evaluated for Bartonella species infection, the combination of blood culture and/or PCR assay on blood, and serologic testing will detect the greatest number of cats that are currently, or were previously, infected (Table 1; www.dlab.colostate.edu; www.galaxydx.com). 26 Laboratories performing solely serology, culture or PCR assays are likely misdiagnosing some cases of fever associated with bartonellosis. Febrile cats that are seronegative and negative for Bartonella species in blood by culture or PCR are unlikely to have this organism as the cause of fever.
If fever or other clinical signs from bartonellosis are suspected in a cat, administration of doxycycline or a fluoroquinolone is generally effective.5,7,23 The American Association of Feline Practitioners Panel Report recommends doxycycline at 10 mg/kg PO q24h for 7 days as the initial therapeutic trial. 26 If a positive response is achieved, treatment should be continued for 2 weeks past resolution of clinical signs or for a minimum of 28 days. 26 If the response achieved by day 7 is poor or doxycycline is not tolerated and bartonellosis is still considered a valid differential diagnosis, fluoroquinolones are an appropriate second choice. In experimental or field studies, administration of enrofloxacin or orbifloxacin has led to rapid resolution of fever in cats with presumed bartonellosis.7,23 Azithromycin is now considered contraindicated for treatment of bartonellosis in cats because of rapid selection for resistance. 27 The new veterinary fluoroquinolone, pradofloxacin, is the least likely to cause resistant strains of B henselae and so may be the preferred quinolone for the treatment of this pathogen. 27 However, some chronic cases of feline bartonellosis may require the administration of two drugs owing to the presence of coinfections, and quinolones are not effective for the treatment of potential pathogens such as Anaplasma species, Borrelia burgdorferi or Ehrlichia species. 28
B henselae is common in fleas and survives at least 9 days in flea frass, and so flea control is imperative to attempt to lessen infection of other cats, dogs or people.12–14,29–31 There have been experimental studies showing that the use of imidacloprid-containing compounds can block transmission of B henselae among cats by C felis.23,31 The flea and tick collar containing imidacloprid was well tolerated by cats owned by veterinary students. 32 Whether other flea control products are effective has not been studied. While several Bartonella species have been grown or amplified by PCR from ticks, the role ticks play in the transmission of Bartonella species among cats is unclear. 33 However, the presence of multiple tick-borne diseases in cats, as described in Part 2, 4 supports the use of acaricides in this species.
Coxiella burnetii
Coxiella burnetii is a rickettsial organism commonly carried by cats and many other mammals,21,34,35 and can be zoonotic to people, usually by airborne transmission.36,37 While long considered a tick-borne disease, C burnetii has also been detected in fleas in Cyprus. 38 In addition, the importance of tick transmission has been questioned. 39
While C burnetii is associated with fever in some people, additional studies will be required to determine the importance of this agent as a cause of vector-borne fever in cats.
Hemoplasmas
Hemolytic anemia, with or without fever, can occur following infection with pathogenic hemoplasmas, especially Mycoplasma haemofelis, but sometimes also Candidatus Mycoplasma haemominutum’ and ‘Candidatus Mycoplasma turicensis’.3,40–43 Like Bartonella species, the DNA of the hemoplasmas has been amplified from the blood of cats as well as fleas collected from cats in many regions of the world.13–17,42,44–46
Based on multiple studies of experimentally infected cats, M haemofelis is usually the most pathogenic hemoplasma species.47,48 Dual infection with hemoplasmas may potentiate pathogenesis of disease. In one study, cats with chronic ‘Candidatus M haemominutum’ infection had more severe and longer duration anemia when they were experimentally infected with M haemofelis compared with cats infected with M haemofelis alone. 47 Clinical signs of disease depend on the degree of anemia, the stage of infection and the immune status of infected cats. In one unpublished study, one of the authors (ML) detected an association between M haemofelis and fever in cats without anemia.
Direct transmission may occur with the hemoplasmas. Studies have found some of the agents in saliva and have also documented infection transmission following subcutaneous inoculation of hemoplasma-containing blood. Hence this pathogen should be on the differential list for cats with a history of fighting and fever, particularly if the fever does not respond to beta-lactam antibiotics.49,50
Diagnosis of hemoplasmosis can be based on demonstration of the organism on the surface of erythrocytes on examination of a thin blood film (Figure 1) or, far more reliably, on PCR assay results (Table 1). Organism numbers fluctuate and so blood film examination can be very insensitive for diagnosis (0–37.5%).47,51–53 Specificity is also likely to be an issue with blood film examination, especially when performed by someone inexperienced in clinical pathology. When properly designed and executed, PCR is far more sensitive and specific than cytology, and species-specific assays are now routinely used. 54 However, healthy cats can also be positive for haemoplasma DNA in blood and so PCR assay results do not always correlate with clinical illness, as is the case with the interpretation of many PCR assays.41,55,56

Blood smear cytology for hemoplasma infection. Hemoplasmas are very difficult to find on cytology unless extremely high numbers of organisms are present in the blood at the time of sampling. Cytology of this blood smear (Wright’s stain, x 100) shows a number of erythrocytes with small purple-staining Mycoplasma haemofelis organisms epicellularly (arrows in inset). Large shift platelets are also visible
Doxycycline, often administered as a flavored suspension (to avoid esophagitis and esophageal strictures) at 10 mg/kg PO q24h or 5 mg/kg PO q12h, is generally effective for the treatment of clinical feline hemoplasmosis. The duration of recommended therapy varies, but is typically 2–4 weeks.57,58 In cats intolerant of doxycycline, fluoroquinolones have also been effective. Administration of marbofloxacin or orbifloxacin gave similar results to doxycycline in two studies.7,59 Pradofloxacin has shown promise in eliminating M haemofelis infection in some experimentally inoculated cats. 60 A recent study reported that to facilitate clearance of M haemofelis, when this is required, doxycycline treatment is administered for 28 days followed by monitoring of copy numbers in the blood by quantitative PCR. 61 If the cat remains PCR positive and clearance is needed, treatment should be switched to a fluoro-quinolone (marbofloxacin was used in the study) for 14 days, as this was associated with apparent clearance of infection. Owners should be informed that recurrences may occur with treatment, but are unusual. 43
Historically, enrofloxacin was also usually effective for the treatment of hemoplasmosis; however, there are safety concerns that need to be considered (see box). Azithromycin was not effective for the treatment of hemoplasmosis in one study. 47
While the hemoplasmas appear to be of low risk to people, there have been reports of infection with similar organisms and one described infection in a human with a hemoplasma species that could have originated in a cat.62,63
Feline hemoplasma DNA has been amplified from fleas and ticks.13,14,29,33,40,64,65 However, this does not equate with vector-mediated transmission, as the presence of hemoplasma DNA could merely reflect their hematophagous activity on infected hosts. Studies on the cat flea, C felis, showed only very transient M haemofelis infection transmission in cats experimentally infected via the hematophagous activity of fleas, and clinical and hematological signs of M haemofelis infection were not induced in the recipient cat. 66 However, the clustered geographical distribution of infection in some studies strongly supports the role of an arthropod vector in hemoplasma transmission and so hemoplasmas are often considered as vector-borne infections.22,67 Recently, Aedes aegypti were shown to ingest M haemofelis or ‘Candidatus M haemominutum’ but transmission to naive cats was not documented, suggesting this mosquito is not a biological vector. 68 Flea control and avoiding fights with other cats should lessen the risk of acquiring hemoplasmosis.
Rickettsia species
Rickettsia species are obligate intracellular Gram-negative bacteria that are divided into the spotted fever group and the typhus group. Rickettsia felis has been found in fleas that infest cats in almost all countries that have been studied.12,13,21,29,30,69,70 R felis DNA has been amplified from C felis, Ctenocephalides canis and Pulex irritans; these fleas have a worldwide distribution. C felis is a biological vector for R felis; this pathogen can be transmitted transovarially and transtadially within the flea. Rickettsia typhi DNA has been amplified from fleas in Spain and the USA.69,71–73 Some C felis fleas have been positive for both R felis and R typhi, suggesting cats could be exposed to both concurrently.69,72 In addition, DNA most consistent with Rickettsia asembonensis or ‘Candidatus Rickettsia senegalensis’ has been amplified from fleas collected from cats. 73
An attempt to associate fever with exposure to R felis was made in a small group of cats in the USA and, while seropositive cats were detected, an association with fever was not made. 74 To date, only small numbers of cats have been studied to attempt to make disease associations with R typhi, 69 and clinical illness has not been identified to the authors’ knowledge. Since flea-borne Rickettsia species are associated with fever in other species, it is possible that fever can also occur in cats, as has been suggested by others; 75 however, further data are needed to determine the significance of disease associations.
Currently, routine testing for cats is not available unless PCR panels using primers capable of amplifying R felis, R asembonensis, ‘Candidatus R senegalensis’ and R typhi are used.
Because clinical illness in cats has not been documented with the flea-associated Rickettsia species, optimal treatment is unknown. However, based on results in other species, doxycycline or a fluoroquinolone would be logical choices. As both R felis and R typhi are associated with disease in humans, it is imperative to provide flea control to cats all year round to potentially lessen the risk of exposure. 76
Yersinia pestis
Yersinia pestis is the cause of the plague in people and, when infection occurs, fever is common in both cats and people.77–80 However, transmission of this pathogen among cats or people by C felis is considered rare because of various flea-associated factors that include rapid bacterial clearance from the midgut of this flea species. 81 While it is possible that rodent fleas could transmit Y pestis to cats, ingestion of bacteremic rodents is likely the most common route of transmission. Y pestis should thus be considered on the differential list for fever of outdoor cats in endemic areas.
Bipolar staining rods are often seen on cytological evaluation of stained smears of lymph node aspirates. The diagnosis can be confirmed by culture (in laboratories with appropriate certification), fluorescent antibody staining or PCR assay; the latter techniques are available at many diagnostic laboratories, particularly in western USA (www.cdc.gov/plague/healthcare/veterinarians.html). In the USA, the state public health veterinarian should be contacted and sample submissions to the diagnostic laboratory clearly labeled as from a cat with possible plague so appropriate techniques can be used and precautions taken (nasphv.org).
Cats with Y pestis infection have responded to a number of antimicrobial drugs including aminoglycosides and doxycycline. Unpublished data of one of the authors (ML) recorded successful treatment of six cats with injectable enrofloxacin at 5 mg/kg SC q24h for 3–5 days (see box on page 35 for safety cautions when using enrofloxacin in cats) followed by doxycycline at 10 mg/kg PO q24h for 7–21 days. Cats should be provided with flea control all year round and hunting behavior should be minimized when possible, particularly for cats living in areas with proven cases of plague.
Key Points
Some flea-borne pathogens can be associated with fever in cats, and the history and concurrent clinical signs can trigger a diagnostic investigation and, ultimately, appropriate therapy.
It has been well documented that flea control can block the transmission of B henselae among cats.
Prevention of flea-borne infections is always preferred to treating clinically ill cats and so flea products should be used all year round.
Hands should be washed after handling cats with fleas.
Bites and scratches from cats should be avoided, particularly if fleas are present.
Footnotes
Acknowledgements
The authors thank Dr Susanne Siebert and Dr Annette Boegel for their support and encouragement as well as editorial comments.
Conflict of Interest
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The authors received no financial support for the research, authorship, and/or publication of this article.
Ethical Approval
This work did not involve the use of animals and therefore ethical approval was not required.
Informed Consent
This work did not involve the use of animals and therefore informed consent was not required. No animals or humans are identifiable within the publication, and therefore additional consent for publication was not required.
