This multicenter, prospective, randomized trial of short-term antimicrobial prophylaxis in colorectal surgery was designed to compare the efficacy of a single dose of cefotetan vs. multiple doses of cefoxitin. Of the 403 evaluable patients, 206 received cefoxitin (1 gm intravenously at the beginning of surgery and after 3, 6, and 12 hours, group A) and 197 cefotetan (2 gm intravenously at the beginning of surgery, group B). The two groups of patients were similar in respect to age, sex, obesity, preoperative weight loss, diabetes, type of disease, type, and mean duration of surgery. The abdominal wound infection rate was 11.2 percent in group A and 9.1 percent in group B (difference not significant). The number of patients with infections at nonsurgical sites (respiratory and urinary tract, phlebitis, and septicemia) was significantly greater in group A than in group B (17 vs. 9.1 percent, P<.05). The mean postoperative hospital stay was similar in both the cefoxitin and cefotetan groups (15.1 vs. 15.9 days). Both regimens were inadequate in preventing infections in the presence of severe contamination of the operative field.