The impact on wound infection of the addition of perioperative cefamandole to a mechanical bowel preparation with oral antibiotics was studied in a prospective randomized series of patients undergoing elective colectomy for biopsy-proven carcinoma or adenomatous polyps. Seventy patients were randomized, all underwent mechanical bowel preparation and received oral neomycin and erythromycin base. Thirty-four patients also received a preoperative and four postoperative doses of cefamandole, while 36 patients were randomized to receive no parenteral antibiotics. The two groups were well stratified for age, sex, and risk factors. The Dukes stage was similar and the surgical procedures were equally distributed in the two groups. There were no wound infections in the 34 patients receiving cefamandole and only one wound infection (2.8 percent) in the 36 control patients. Therefore, the addition of perioperative intravenous cefamandole to a good mechanical bowel preparation with oral antibiotics was of no benefit in reducing wound infections following resection of colorectal malignancies in this select group of patients.