The clinical safety and efficacy of imipenem/cilastatin in the treatment of intra-abdominal infections was compared with the combination of aztreonam and clindamycin in a randomized prospective trial. The severity of illness was determined by means of the Apache II score and a fixed outcome reporting scheme was used. One hundred and four patients were entered into the study, of whom 80 were evaluable. Forty-two patients were treated with imipenem/cilastatin (500 + 500 mg qds) and 38 with aztreonam (600 tds) and clindamycin (1000 mg tds). The study groups were comparable for age and sex. The imipenem/cilastatin group differed from the aztreonam and clindamycin group in having significantly more patients with the diagnosis of acute appendicitis (P < 0.01) and a significantly lower mean Apache score (P < 0.05). The predominate microorganisms isolated in both groups were Escherichia coli and Bacteroides fragilis. Treatment with imipenem/cilastatin proved successful in 71% and failed in 24%, and initial success only was seen in 5%. The numbers in the group treated with aztreonam and clindamycin were 64%, 29% and 7% respectively. Severity of illness, as measured by Apache II score, had no influence on the study outcome. Imipenem/cilastatin as well as the combination of aztreonam and clindamycin were effective in the treatment of abdominal infections and no major adverse reactions were seen.