In a randomized controlled trial single-dose peroperative intravenous mezlocillin, 5 g (group 1, n = 51), was compared with single-dose mezlocillin, 5 g, and metronidazole, 500 mg (group 2, n = 58), in 109 patients undergoing elective colorectal surgery. The two groups were similar in terms of age, sex, underlying pathology, obesity, prolonged operations, operative procedure and operator status. There was no significant difference in postoperative complications – five deaths 4.6% (three group 1), two septicaemias 2% (both group 2), five intra-abdominal abscesses 5% (three group 1) and 34 (31 %) wound infections. The incidence of wound infection was not related to the obesity of the patient, underlying pathology, duration of operation, operative procedure or operator status. Although mezlocillin was active against all anaerobes isolated during the study, analysis of the pattern of isolated organisms suggests that its spectrum of activity and its efficacy against aerobes is inadequate, rendering it inappropriate as a single-dose prophylactic agent in elective colorectal surgery even in combination with metronidazole.