OBJECTIVE: To estimate the effect of a hospital-wide change in the timing of antimicrobial prophylaxis in cesarean deliveries on maternal and neonatal infections. METHODS: In November 2004, our institution instituted guidelines recommending that the administration of antimicrobial prophylaxis for cesarean delivery be administered before skin incision rather than after umbilical-cord clamping. We reviewed all cesarean deliveries from two time periods. Group 1 received antibiotics after umbilical-cord clamping (July 2002 to November 2004). Group 2 received antibiotics before skin incision (June 2005 to August 2007). Rates of maternal and neonatal infectious complications were compared between groups. RESULTS: There were 4,229 cesarean deliveries in group 1 and 4,781 cesarean deliveries in group 2. Compared with women receiving antimicrobial prophylaxis after umbilical-cord clamping, those administered antimicrobial prophylaxis before skin incision had lower rates of postpartum endometritis (2.2% compared with 3.9%) and wound infection (2.5% compared with 3.6%). After multivariable logistic regression, antimicrobial prophylaxis before skin incision remained associated with lower rates of endometritis (odds ratio [OR] 0.61, 95% confidence interval [CI] 0.47-0.79) and wound infection (OR 0.70, 95% CI 0.55-0.90). Antimicrobial prophylaxis before skin incision had no adverse effect on neonatal infection rates or on the evaluation of the neonate. CONCLUSION: Antimicrobial prophylaxis before skin incision, compared with after cord clamping, resulted in lower rates of maternal infections and had no effect on neonatal infections. Antimicrobial prophylaxis for cesarean delivery should occur before skin incision, consistent with basic tenets of surgical antimicrobial prophylaxis.