OBJECTIVE： To explore the effects of prophylactic use timing of antibiotics on postoperative infection of cesarean section patients. METHODS： 100 patients undergoing selective cesarean section were divided into observation group and control group according to random number table， with 50 cases in each group. Both groups received selective cesarean section. Control group was given Cefathiamidine for injection 2.0 g intravenously immediately after cutting umbilical cord of fetus， and then given another 2.0 g intravenously within 6 h after surgery. Observation group was given Cefathiamidine for injection 2.0 g intravenously 0.5-2 h before skin incision， and then given another 2.0 g intravenously 6 h after surgery. Surgery duration， intraoperative blood loss， prophylactic duration of antibiotics were compared between 2 groups as well as body temperature 6， 12， 24， 48 h after surgery， leukocyte and neutrophile granulocyte 48 h after surgery； the incidence of complication such as endometrial inflammation， incision infection， symptomatic urinary tract infection， puerperal disease were compared between 2 groups after surgery. RESULTS： There was no statistical significance in surgery duration and intraoperative blood loss as well as body temperature 6， 12， 24， 48 h after surgery， leukocyte and neutrophile granulocyte 48 h after surgery between 2 groups （P＞0.05）. The prophylactic duration of antibiotics in control group [（4.8±1.60）h] was shorter than observation group [（5.5±1.65）h]， with statistical significance （P＜0.05）. There was no statistical significance in the incidence of endometrial inflammation （0 vs. 2.00％）， incision infection （0 vs. 0）， urinary tract infection （2.00％ vs. 2.00％）， puerperal disease （4.00％ vs. 6.00％） between 2 groups （P＞0.05）. CONCLUSIONS： The use of antibiotics before skin incision is similar to that after cutting umbilical cord in preventive effects of postoperative infection for cesarean section patients. The use of antibiotics before skin incision can prolong the preventive duration of antibiotics.