CAT SCRATCH DISEASE
• The combination of a compatible clinical syndrome of regional lymphadenopathy with a primary inoculation lesion and cat contact is highly suggestive of cat scratch disease.
• Results of Bartonella henselae serology are helpful in confirming the diagnosis; however, up to 20% of patients remain seronegative throughout their disease.
• Detection of B. henselae DNA by polymerase chain reaction from pus obtained by needle aspiration or lymph node biopsy is a highly sensitive diagnostic modality and may be particularly useful in seronegative patients.
• Typical cat scratch disease is a self-limited disease; therefore, systemic antimicrobials are not routinely indicated in immunocompetent patients with uncomplicated cat scratch disease.
• Indications for antimicrobial therapy include an immunocompromised host, endocarditis, retinitis, and other atypical syndromes.
• Treatment may also be considered for patients who have typical cat scratch disease and who have extensive or bulky lymphadenopathy and are highly symptomatic.
• Suppurative lymph nodes should be drained by large-bore needle aspiration.
Cat scratch disease is a worldwide zoonotic infection caused by Bartonella henselae, an intracellular, pleomorphic, gram-negative bacillus. Other Bartonella species have rarely been implicated. The disease typically manifests as benign regional lymphadenopathy, but atypical disease can involve almost any organ system and is associated with significant morbidity.
B. henselae infection is widespread among domestic cats and other felids worldwide, and serologic studies indicate a higher prevalence in warm, humid climates. Transmission among cats occurs via an arthropod vector, the cat flea Ctenocephalides felis. B. henselae bacteremia is detected at higher rates in feral than in domesticated cats and in kittens as compared to adult cats, explaining the higher infectivity of these animals. Cats infected with B. henselae are asymptomatic. Transmission to humans occurs via a scratch, bite, or lick; contamination of cat scratches by flea feces may enhance Bartonella transmission.
Cat scratch disease has been estimated to occur in the United States at a rate approaching 10 per 100,000 population. There is some seasonality, with incidence peaking between September and January. Although traditionally considered a disease of childhood, epidemiologic surveys have found a similar incidence of cat scratch disease in adults, and 6% of patients are aged 60 years or older. Cat contact is the most important risk factor: Almost all patients are cat owners or were otherwise exposed to cats, and about half can recall a recent bite or scratch, most commonly by a kitten.
Cat scratch disease is divided into two clinical syndromes. Typical cat scratch disease is a subacute, self-limited regional lymphadenopathy and constitutes 80% to 90% of cases. Atypical cat scratch disease encompasses Parinaud’s oculoglandular syndrome and other clinical entities with systemic extranodal involvement. Bacillary angiomatosis and bacillary peliosis are manifestations of B. henselae infection in immunocompromised persons and are not discussed here.
Typical Cat Scratch Disease
A primary skin lesion, usually a papule or pustule, appears at the site of inoculation 3 to 10 days after cat contact and persists for 1 to 3 weeks. Regional lymphadenopathy develops within 1 to 7 weeks and resolves spontaneously after 2 to 4 months. The primary lesion is still present in about two thirds of patients when they present for evaluation of lymphadenopathy. The most commonly involved lymph nodes are, in descending order of frequency, axillary and epitrochlear nodes, head and neck nodes, and femoral and inguinal nodes. One third of patients develop lymphadenopathy at multiple sites. Lymph nodes are often painful and red; suppuration occurs in 10% of nodes.
Mild constitutional symptoms, including low-grade fever and malaise, are noted in about half the cases. Rash, night sweats, anorexia, and weight loss are uncommon. Infection results in lifelong immunity.
Atypical Cat Scratch Disease
Parinaud’s oculoglandular syndrome, the most common form of atypical cat scratch disease, is a specific type of regional lymphadenopathy that occurs following conjunctival inoculation of B. henselae. The syndrome includes granulomatous conjunctivitis and preauricular lymphadenopathy. Endocarditis and encephalitis represent severe forms of B. henselae infection and are more common in elderly patients. Encephalitis manifests with various degrees of altered mental status, agitation, headache, and seizures. Cerebrospinal fluid lymphocytic pleocytosis occurs in only one third of cases, and brain-imaging studies usually fail to show any abnormalities.
Neuroretinitis manifests as sudden unilateral loss of visual acuity. The diagnosis is suspected on the basis of typical findings on fundoscopic examination: papilledema and macular exudates in a starlike configuration. However, these findings are not pathognomonic of cat scratch disease. Other infrequent manifestations are self-limited granulomatous hepatitis and splenitis and osteoarticular disease.
Prolonged fever of unknown origin has been described in children and adolescents.
A compatible clinical syndrome in a person with a positive history of cat exposure suggests cat scratch disease. The diagnosis is most commonly confirmed with specific serologic assays.
Immunofluorescence and enzyme immunoassays are comparably sensitive, although cross-reactivity can occur with other organisms such as Chlamydia species, Coxiella burnetii, and non-henselae Bartonella species. A single elevated titer of immunoglobulin (Ig)M or IgG in the acute phase or a fourfold increase of IgG in convalescent serum supports the diagnosis.
Biopsy and histopathologic examination of lymph nodes are usually only performed when malignancy is suspected. Necrotizing granulomas are typically observed. The pleomorphic gram-negative bacilli are best visualized with the Warthin-Starry stain; however, the sensitivity of this method is too low to be clinically useful. Similarly, culture lacks sensitivity and requires prolonged incubation, making it impractical for routine use.
Polymerase chain reaction assays for the detection of B. henselae in pus aspirated from suppurative lymph nodes or primary skin lesions is highly sensitive and specific, and it can be performed with a rapid turnaround time. However, these assays are not widely available in most clinical microbiology laboratories.
Cat scratch disease should be differentiated from other infectious and noninfectious causes of regional lymphadenopathy. Importantly, lymphoma and solid tumors with metastases to lymph nodes may be confused with cat scratch disease. Malignancy should be suspected in patients who have marked constitutional symptoms, who are seronegative, or whose lymphadenopathy fails to resolve spontaneously after more than 6 months. Unlike cat scratch disease, pyogenic lymphadenitis manifests with rapid progression, high-grade fever, and a septic-appearing patient. Mycobacterial infection, syphilis, bubonic plague, tularemia, histoplasmosis, and sporotrichosis should be considered based on the presence of specific risk factors and endemic exposures. Bartonella endocarditis must be differentiated from other causes of culture-negative endocarditis, specifically Q fever and brucellosis.
Because of the self-limiting course of typical cat scratch disease, patients usually only require observation and reassurance of its benign nature. The goals of therapy are alleviation of symptoms in patients with bulky lymphadenopathy and drainage of suppurating lymph nodes to prevent spontaneous formation of chronic sinus tracts. If drug treatment is initiated, azithromycin (Zithromax)1 is the agent of choice based on a small randomized placebo-controlled study that showed more-rapid resolution of lymphadenopathy in the azithromycin arm as determined by ultrasound. Suppurative nodes should be drained by large-bore needle aspiration. Incisional drainage is best avoided because it can promote sinus tract formation.
Atypical cat scratch disease syndromes usually require treatment, but there are scant data to support specific regimens, and treatment duration is not well defined. For patients with endocarditis, treatment with an aminoglycoside for at least 14 days is associated with a higher likelihood of recovery and survival (Table 1).
Antibiotic Treatment Regimens for Cat Scratch Disease
Abbreviation: NA = not available.
1 Not FDA approved for this indication.
* Treatment may be considered for patients with bulky or extensive lymphadenopathy and for immunocompromised patients.
Flea control in domestic cats can reduce the likelihood of Bartonella bacteremia and transmission to humans. Bites and scratches should be promptly rinsed. HIV-infected patients and other immunocompromised persons should be cautioned about the risks of cat exposure because they are susceptible to disseminated visceral B. henselae infection, as well as other zoonotic infections such as toxoplasmosis.
1. Ben-Ami R., Ephros M., Avidor B., et al. Cat-scratch disease in elderly patients. Clin Infect Dis. 2005;41:969–974.
2. Giladi M., Kletter Y., Avidor B., et al. Enzyme immunoassay for the diagnosis of cat-scratch disease defined by polymerase chain reaction. Clin Infect Dis. 2001;33:1852–1858.
3. Hansmann Y., DeMartino S., Piemont Y., et al. Diagnosis of cat scratch disease with detection of Bartonella henselae by PCR: A study of patients with lymph node enlargement. J Clin Microbiol. 2005;43:3800–3806.
4. Jackson L.A., Perkins B.A., Wenger J.D. Cat scratch disease in the United States: An analysis of three national databases. Am J Public Health. 1993;83:1707–1711.
5. Jacobs R.F., Schutze G.E. Bartonella henselae as a cause of prolonged fever and fever of unknown origin in children. Clin Infect Dis. 1998;26:80–84.
6. Raoult D., Fournier P.E., Vandenesch F., et al. Outcome and treatment of Bartonella endocarditis. Arch Intern Med. 2003;163:226–230.
7. Rolain J.M., Brouqui P., Koehler J.E., et al. Recommendations for treatment of human infections caused by Bartonella species. Antimicrob Agents Chemother. 2004;48:1921–1933.
8. Zangwill K.M., Hamilton D.H., Perkins B.A., et al. Cat scratch disease in Connecticut. Epidemiology, risk factors, and evaluation of a new diagnostic test. N Engl J Med. 1993;329:8– 13.
1 Not FDA approved for this indication.