OBJECTIVE: To assess the efficacy and safety of trimethoprim-sulfamethoxazole for the prevention of spontaneous bacterial peritonitis in patients with cirrhosis and ascites. DESIGN: A randomized controlled trial. SETTING: University-affiliated Veterans Affairs medical center. PATIENTS: 60 consecutive patients with cirrhosis and ascites. INTERVENTIONS: Consecutive patients were randomly assigned to receive either no prophylaxis or trimethoprim-sulfamethoxazole, one double-strength tablet daily, five times a week (Monday through Friday). Patient entry was stratified by serum bilirubin (> 51 mumol/L [> 3 mg/dL]), ascitic fluid protein (< 1 g/dL), and serum creatinine (> 177 mumol/L [> 2 mg/dL]) levels to ensure that high-risk patients would be similarly distributed in the two groups. The median duration of follow-up for the study patients was 90 days. MAIN OUTCOME MEASURES: Spontaneous bacterial peritonitis or spontaneous bacteremia as defined by objective criteria. RESULTS: Spontaneous bacterial peritonitis or spontaneous bacteremia developed in 27% (8 of 30) of patients who did not receive prophylaxis compared with 3% (1 of 30) of patients receiving trimethoprim-sulfamethoxazole (P = 0.025). Overall, infections developed in 9 of 30 patients (30%) not receiving prophylaxis and in 1 of 30 patients (3%) receiving trimethoprim-sulfamethoxazole (P = 0.012). Death occurred in 6 of 30 patients (20%) who did not receive prophylaxis and in 2 of 30 patients (7%) who received trimethoprim-sulfamethoxazole (P = 0.15). Side effects--particularly, hematologic toxicity--could not be attributed to trimethoprim-sulfamethoxazole in any patient. CONCLUSIONS: Trimethoprim-sulfamethoxazole was efficacious, safe, and cost-effective for the prevention of spontaneous bacterial peritonitis in patients with cirrhosis.