A prospective, double-blind, placebo-controlled study was undertaken to determine the influence of prophylactic oxacillin on the frequency of infection in cat bite wounds. Adult patients with uninfected full-thickness wounds presenting within 24 hours of injury were considered. Emergency department management consisted of cleansing, irrigation, debridement, and closure as indicated; no topical antibiotics were applied. Patients were randomly assigned to receive oxacillin 500 mg qid for five days or identically appearing placebo. Home wound care was standardized and patients were observed at least every two days for a minimum of five days, or until wounds were sufficiently healed to allow discharge from the study. Clinical assessment of infection was confirmed microbiologically when possible. Twelve patients were admitted and 11 completed the study. Oxacillin (n = 5) and placebo (n = 6) groups were identical in sex, age, number of wounds per patient, wound location and type, delay to emergency department presentation, length of follow-up observation, medication compliance, and adequacy of home wound care. Four of six patients receiving placebo, but none of the five receiving oxacillin, developed a wound infection (P = .045). Material obtained from three of these four patients yielded Pasteurella multocida as the responsible organism. Prophylactic oxacillin was thus associated with a significant reduction in the frequency of infection following cat bites. We recommend such therapy in the care of these wounds.