The efficacy and safety of ceftizoxime and cefoxitin were compared in a randomized, double-blind study of therapy for lower extremity infections in patients with diabetes mellitus or peripheral vascular disease. Overall clinical responses were satisfactory in 82% (23/28) of patients treated with ceftizoxime and in 68% (17/25) of patients treated with cefoxitin. The difference was not statistically significant. Ceftizoxime had superior in vitro activity against Enterobacteriaceae, especially Enterobacter cloacae, whereas cefoxitin had better activity against the Bacteroides fragilis group. Relapses of infection were common in both groups during long-term follow-up; only about one third of patients in either group maintained satisfactory outcomes after one year. More than half of the patients in both groups responded to one or more courses of medical therapy and avoided major amputations for one year following entry into the study.