Prophylactic peroperative intravenous metronidazole in elective colorectal surgery.

In a prospective double-blind randomised trial 83 patients undergoing elective colorectal surgery were given either preoperative intravenous metronidazole or intravenous normal saline. No other antimicrobials were given. Bowel preparation was the same for both groups. Deep post-operative wound sepsis occurred in 6 of 44 (13-6%) patients on metronidazole but in 20 of 39 (51-2%) control untreated patients. Anaerobes were responsible for all cases of deep sepsis in the metronidazole group (in 5 of the 6 cases aerobes were also isolated), and for 16 of the 20 cases of deep sepsis in the control group (all with aerobes). Superficial infection in the metronidazole group was caused by aerobes. Anastomotic leakage occurred in 5 of the 6 patients who developed deep sepsis on metronidazole and in 10 of 20 patients on placebo. Peroperative intravenous metronidazole dramatically reduced postoperative sepsis but failed to prevent infection in the presence of anastomotic breakdown.