The epidemiology of bacillary angiomatosis and bacillary peliosis.

OBJECTIVE To determine environmental risk factors for bacillary angiomatosis-bacillary peliosis (BAP), and to confirm infection with Rochalimaea species. DESIGN Case-control study. SETTING Community and university hospitals and clinics. PATIENTS Case patients (N = 48) had biopsy-confirmed BAP. Controls (N = 94) were matched to patients by institution and by human immunodeficiency virus (HIV) serological status. MAIN OUTCOME MEASURES Clinical information was obtained from medical records. Subjects were queried about environmental exposures. Univariate odds ratios (ORs) with 95% confidence intervals (CIs) were determined. Bivariate analyses were performed on variables associated with disease by univariate analysis. DNA from 22 available case-patient tissues and from 22 control tissues was amplified with the polymerase chain reaction (PCR) using primers designed to detect Rochalimaea species. RESULTS We identified five HIV-negative, immunocompetent case patients; one HIV-negative, immunodeficient case patient; and 42 HIV-positive case patients. There were no significant differences between case patients and controls by race, sex, age, or risk factors for HIV infection. Owning a cat (OR, 2.8; CI, 1.4 to 5.8) and history of a recent cat lick (OR, 1.95; CI, 1.0 to 3.8), cat scratch (OR, 3.7; CI, 1.7 to 8.0), or cat bite (OR, 3.9; CI, 1.8 to 8.9) were associated with disease in the univariate analysis. In bivariate analyses, only the variables representing traumatic contact with a cat (bite or scratch) remained associated with disease. No other environmental exposure was associated with disease. The PCR amplified a DNA fragment of the size expected for Rochalimaea species in all 22 case-patient tissue specimens. CONCLUSIONS These data suggest that BAP is a new zoonosis associated with both traumatic exposure to cats and infection with Rochalimaea species or a closely related organism.