Two hundred twenty-three women were given a single, 1-g, intravenous dose of cefamandole or cefotaxime at elective abdominal hysterectomy in a multicenter, prospective, randomized, blind clinical trial of efficacy and safety. The demographic, surgical, efficacy and safety variables were statistically similar. Prior to discharge from the hospital, 12 women (5.3%) developed major postoperative pelvic infections that required parenteral antimicrobial therapy; no wound infections occurred. There was no correlation between a depressed antimicrobial development of significant postoperative infection. An expanded spectrum of antibacterial activity and a longer serum half-life did not improve clinical efficacy, and single-dose intravenous cephalosporin prophylaxis before abdominal hysterectomy was associated with a low incidence of pelvic infection.