FELINE
BARTONELLA INFECTIONS – Separating Fact from Fiction©2006
Alice
M. Wolf, DVM, DACVIM, DABVP
Chief
Consultant: Veterinary Information
Network
Adjunct
Professor, Department of Small Animal Veterinary Clinical Sciences
College
of Veterinary Medicine, Texas A&M University
College
Station, TX 77843-4474
Introduction
Bartonella
spp. cause cat scratch disease (CSD) and other clinical syndromes in human
beings, and are an important cause of endocarditis in dogs. On the other hand, there is scant documented
scientific evidence that Bartonella
infection causes overt clinical disease in naturally infected cats, in spite of
a high prevalence of bacteremia and seropositivity in areas of the United
States with warm temperatures and high humidity.
Microbiology
Bartonella
spp. are facultatively intracellular gram-negative rods that are related
closely to Brucella spp and the
rickettsiae. Their intra-erythrocytic
location precludes easy blood culture and a reliable response to antimicrobial
therapy.
Four species of Bartonella have been shown to infect pet cats. B.
henselae infection is most common, and is the most important cause of
CSD. B. clarridgeiae may be responsible for a small number of cases of
CSD. Rare infections of cats with B. koehlerae and B. bovis also have been reported.
Two main genotypes of Bartonella
henselae have been identified
worldwide – Houston and Marseille. A
third genotype, Berlin, has only been identified from one cat in Germany. Exotic cats have also been found to carry Bartonella sp. including: mountain lions, cheetahs, African lions,
Florida panthers, pumas, and bobcats.
Transmission
Bartonella
spp. are transmitted between cats by Ctenocephalides
felis – the cat flea. Fleas ingest
the organism during a blood meal from a bacteremic cat, and infect a naïve cat
through regurgitation of infected saliva during a subsequent blood meal. Ticks may transmit the organism rarely
between cats, and are the primary mode of transmission of Bartonella spp. between dogs.
The organisms are not transmitted between cats by fighting, grooming,
mating, or in-utero.
Human beings become infected with Bartonella spp. when flea feces from a
bacteremic cat are inoculated into a cat scratch. Although not confirmed, rarely, infection may possibly be
acquired directly through the bite of an infected flea.
Cats at Risk
Although Bartonella infections in cats have only been reported in the modern
veterinary literature since 1992, the organism has apparently been infecting
and adapting to cats for hundreds of years.
A recent paper reports the isolation of Bartonella antigen from dental pulp by PCR assay in 800 year-old
cat teeth from France. The prevalence of bacteremia and seropositivity in cats
in the United States is highest in regions that favor the reproduction and
persistence of fleas. Rates are highest
in the southeastern United States (up to 40%), and lowest in the northern tier
of states. The incidence in the EU,
UK, and other countries also mirrors this pattern with higher seroprevalence
rates in warm, moist locales, and much lower rates in colder climes.
Bacteremia is more likely to occur
in cats with fleas, free-roaming cats, young cats, and those from multiple-cat
populations.
Pathogenesis
After experimental inoculation into Bartonella-naïve cats, the organisms
infect erythrocytes and the initial bacteremia lasts 2 to 32 weeks. After this, infected cats undergo bouts of
cyclic bacteremia. Between bacteremic
phases, when blood cultures (and blood PCR tests) are negative, the organisms
may persist in endothelial cells, lymph nodes, or the central nervous system.
Following initial production of
anti-Bartonella IgM antibody, IgG
antibody is produced and remains detectable for months to years. There is no evidence that the height of the
IgG antibody titer correlates with the presence of bacteremia – this
observation is important when considering the diagnostic utility of antibody
titers.
Clinical Signs – Experimental Infections
Part of the confusion about the
clinical importance of Bartonella
infections in cats arises from the observation that clinical signs are more
likely to occur in cats after experimental inoculation that in naturally
infected cats. Inappropriate
extrapolation of data from experimental studies is one of many factors that
have led to an overdiagnosis of clinical bartonellosis in the general cat
population.
Following experimental inoculation,
some infected cats have developed an inflammatory lesion at the injection site,
mild generalized lymphadenopathy, splenomegaly, fever, lethargy, anorexia,
myalgia, behavioral or neurologic changes, and reproductive abnormalities. In a number of other studies, no clinical
signs were seen following infection.
This probably relates to variable pathogenicity among the strains of B. henselae used in these studies.
Clinical Signs – Natural Infections
With rare exception, Bartonella spp. cause prolonged
asymptomatic infections in naturally infected cats. Well-documented clinical signs arising directly from infection
are very unusual and mostly anecdotal.
Bartonella
spp. have been linked directly with endocarditis in one cat. Based on anecdotal reports, the organism may
be a rare cause of lymphoplasmacytic gingivostomatitis (LPG) and uveitis. Clinicians should remember that feline
calicivirus and plaque intolerance are very common causes of chronic LPG, and
that many affected cats coincidentally will be Bartonella-seropositive given the high prevalence of infection with
the organism.
A
similar situation arises with uveitis, which usually is caused by viral or
fungal infections. Because uveitis
often is accompanied by intra-ocular bleeding, the isolation of Bartonella organisms or antibody from
within the eye does not confirm infection.
The diagnosis of Bartonella-induced
uveitis is supported by the exclusion of all other more common causes, the
demonstration of higher antibody levels in the aqueous humor than in serum, and
a specific response to selective antimicrobial treatment.
Recent
studies by Dr. Mike Lappin and the infectious disease group at Colorado State
University have shown no statistical differences in Bartonella seropositivity
between cats with and without uveitis, oral cavity disease, and central nervous
system disease (ACVIM 2005).
Bartonella Infections in Human Beings
The CDC estimates that there are
24,000 cases of CSD/year in the U.S.
This is an incidence of 9.3/10,000 ambulatory patients/year. The seropositivity rate for Bartonella in humans is between 3.6% to
15%; with the latter value occurring in a survey of veterinary
professionals. Most persons inoculated
accidentally with infected fleas feces through a cat scratch probably show no
clinical signs or suffer from a vague, mild self-resolving, flu-like illness
that does not prompt a visit to the physician.
On the other hand, some immunocompetent people will develop typical CSD,
with or without systemic complications.
Persons with impaired immune systems are at risk for more severe
complications of infection.
Typical signs of CSD include the
development of a pustule (primary inoculation lesion) in the infected scratch
within 7 to 10 days of the injury.
Regional lymphadenitis, usually non-painful, occurs within 1 to 3 weeks
of the injury. Lymph node enlargement
may persist for weeks to months. Antimicrobial
treatment does not shorten the duration of disease reliably.
Atypical signs of CSD include
Parinaud’s oculoglandular syndrome (associated with a primary inoculation
lesion on the conjunctiva and regional lymphadenopathy following infection of
the conjunctiva with flea feces from a bacteremic cat), relapsing bacteremias
and fevers, encephalitis, endocarditis, hepatitis, pneumonia, and
osteomyelitis.
Angioproliferative lesions are more
common in immunocompromised persons and include cutaneous lesions of bacillary
angiomatosis, and cystic hepatic lesions of bacillary peliosis. Systemic complications of zoonotic Bartonella infections are more likely to
be severe in immunocompromised human patients.
Paradoxically, the response of these patients to antimicrobial treatment
is better than that of immunocompetent patients with typical CSD.
Diagnosis
Laboratory tests to detect or
exclude infection with Bartonella spp
in cats include the detection of anti-Bartonella
antibodies through immunofluorescent antibody (IFA) and Western Blot (WB)
tests, blood cultures, and the amplification of Bartonella DNA by polymerase chain reaction (PCR) tests. These tests are used to place cats into one
of the following categories:
1. Healthy
cats that are not infected with Bartonella
spp., and therefore are safe companion animals for immunocompromised
persons.
2. Healthy
cats that currently are infected with Bartonella
spp., or that have been infected previously with Bartonella spp.
3. Sick
cats (for example, cats with uveitis or stomatitis) with concurrent Bartonella infection that is not the
cause of their clinical illness.
4. Cats
with Bartonella-induced illness (an
unusual occurrence in clinical practice).
Because
of the high prevalence of seropositivity to Bartonella
in the general cat population and the low incidence of Bartonella-induced disease, the detection of serum antibodies has a
poor predictive value (42 to 46 per cent) for the confirmation of disease
caused by the organism. Similarly,
using the IFA test, there is a poor correlation between the height of the
antibody titer and the ability to detect bacteremia. Because titers in infected cats vary greatly over time, increases
in titers associated with vague clinical signs should be interpreted with
caution.
Conversely,
a negative IFA titer has a high negative predictive value (>90 per cent),
making it a useful screening test to exclude bacteremia in an asymptomatic or
symptomatic cat. A small number of cats
may be seronegative between cycles of bacteremia, and blood cultures and PCR
tests may be needed to confirm the status of these cats, especially if they are
being considered as companion animals for immunocompromised persons.
With
the possible exception of endocarditis, the clinical diseases attributed
anecdotally to infection with Bartonella spp.
usually are caused by more common infections.
Therefore, tests for these other diseases (for example, FeLV/FIV,
toxoplasmosis, cryptococcosis, histoplasmosis, and the aforementioned causes of
stomatitis) should be performed and interpreted before tests for Bartonella infection are ordered.
Based
on our present scientific knowledge of the epidemiology of Bartonella infections in cats and human beings, there are no valid
indications for the routine testing and subsequent treatment of healthy pet
cats that live with healthy owners.
This recommendation is also that of the Centers for Disease Control and
this information is available on their website: www.cdc.org
Treatment
Treatment
should be reserved for that small group of sick cats with apparent Bartonella-induced disease, based on
careful interpretation of serological and culture/PCR results and an exhaustive
exclusion of other more common diseases. At the present time, there is no
evidence that antimicrobial therapy eradicates Bartonella organisms completely from infected cats. Although the level of bacteremia may be
reduced temporarily, recurrence of bacteremia usually occurs due to the
intra-erythrocytic location of the organism.
Enrofloxacin and doxycycline have
been used to reduce the level of bacteremia.
Unfortunately, the recommended dose of enrofloxacin has a high risk of
inducing retinotoxicity, precluding its safe use in cats. Azithromycin (5-10 mg/kg PO q24h for 5d,
then q48h for 40d) has been recommended for anecdotal cases of stomatitis that
were suspected to be caused by Bartonella
infection. Unfortunately, because
bacteremia is cyclic, and because organisms are rarely cleared with antibiotic
therapy, there is no good endpoint on which to base apparent success of
treatment. PCR testing may be negative
at some point, and then return to positive weeks to months later. Antibodies will persist for years, often at
high levels even after organisms are gone.
Prevention
Prevention of CSD in human beings sharing
a house with cats depends primarily on scrupulous and effective flea
control. Even if the cat is bacteremic,
human infection from cat scratches will not occur unless the injuries are
contaminated with flea feces. Children,
who are at most risk for the development of CSD if infected, should be taught
to play gently with their pet cats, especially new kittens, to avoid scratches.
Cats being considered as companion
animals for immunocompromised persons should be selected from a flea-free
background. The cats should be screened
initially with an IFA test. If the test
is positive, it would be wise to consider the cat no further as a safe
companion. If the IFA test is negative,
blood cultures and PCR tests should be performed. If these latter tests are negative, the cat can be considered
safe, as long as it is kept indoors exclusively, not exposed to cats with
fleas, and is treated diligently with a year-round flea-control program.
References
- Pressler B: Bartonellosis. In
August JR (ed): Consultations in Feline Internal Medicine, vol 5. St. Louis, Elsevier, 2006, in press.
- Chomel BB, Boulouis HJ,
Breitschwerdt EB: Cat scratch disease and other zoonotic Bartonella infections. J Am Vet Med Assoc 224:1270-1279, 2004.
- Guptill L: The diagnosis and treatment
of Bartonella in dogs and cats.
Proc ACVIM Forum, Charlotte NC, 2003.
- Kumasaka K, et al: Survey of veterinary professionals for
antibodies to Bartonella henselae in Japan. Rinsho Byori, 49:906-910, 2001.
- Rolain JM, et al: Immunofluorescent detection of
intraerythrocytic Bartonella
henselae in naturally infected cats.
J Clin Microbiol
39:2978-2980, 2001
- Jacomo V, Kelly PJ, Raoult D: Natural history of Bartonella infections (an exception to Koch’s postulate). Clin Diag Lab Immunol 9:8-18, 2002
- Yamamoto K, et al: Experimental infection of domestic cats
with Bartonella koehlerae and
comparison of protein and DNA profiles with those of other Bartonella species infecting
cats. J Clin Microbiol 40:466-474, 2002
- La VD, et al: Molecular detection of Bartonella henselae DNA in the
dental pulp of 800-year-old French cats.
Clin Infec Dis
39:1391-1394, 2004.
