OBJECTIVE Treatment of intra-abdominal infections remains a challenge because of their polymicrobial nature and associated mortality risk. Broad-spectrum empiric coverage is usually required. This randomized study compared the efficacy and safety of intravenous piperacillin/tazobactam with those of intravenous imipenem/cilastatin in the treatment of 293 hospitalized patients with intra-abdominal infection. METHODS A total of 149 patients received piperacillin/tazobactam 4 g/500 mg every 8 hours, and 144 patients received imipenem/cilastatin 500 mg/500 mg every 6 hours. Efficacy was evaluated by clinical and bacteriological response. Safety was evaluated by analysis of adverse events and physical and laboratory examinations. RESULTS Clinical and bacteriological responses in both evaluable treatment groups were equivalent. The clinical success was 97% (108/111) for piperacillin/tazobactam and 97% (100/103) for imipenem/cilastatin. Bacteriological success was 97% (67/69) for piperacillin/tazobactam and 95% (61/64) for imipenem/cilastatin. The most common pathogens were Escherichia coli, Klebsiella pneumoniae, Enterobacter species and Pseudomonas aeruginosa. The frequencies of treatment-related adverse events were similar (16 with piperacillin/tazobactam and 19 with imipenem/cilastatin). CONCLUSIONS These results suggest that the safety and efficacy of piperacillin/tazobactam administered every 8 hours are equivalent to those of imipenem/cilastatin administered every 6 hours for the treatment of intra-abdominal infections.