OBJECTIVETo estimate the efficacy of preoperative administration of intravaginal metronidazole for the prevention of postcesarean endometritis.METHODSThis double-masked, placebo-controlled randomized trial included patients of at least 24 weeks' gestation undergoing cesarean deliveries for various indications. Patients were randomized to receive either 5 g of metronidazole gel intravaginally or matching placebo before the initiation of the cesarean. All patients underwent surgical cleansing of the abdomen, and most received prophylactic antibiotics after cord clamping. Patients with chorioamnionitis and/or suspected allergy to metronidazole were excluded. For a two-sided α of 0.05 and β of 0.20 (80% power), 120 subjects were required in each group. The main outcome variable was the incidence of postcesarean endometritis. Secondary outcome variables included presence of febrile morbidity, wound infection, days on antibiotics, and length of postpartum hospitalization. Neonatal outcomes included birth weight, Apgar scores less than 7 at 5 minutes, umbilical arterial pH less than 7.16, admission and length of stay in the neonatal intensive care unit, and length of hospital stay.RESULTSOf 112 patients receiving metronidazole, eight (7%) developed postcesarean endometritis, compared with 19 of 112 (17%) of those receiving placebo gel (relative risk 0.42, 95% confidence interval 0.19, 0.92). No significant differences were noted between treatment groups with respect to the other outcome variables.CONCLUSIONThe preoperative administration of 5 g of intravaginal metronidazole gel appears to reduce the incidence of postcesarean endometritis.