e15576 Background: In Japan, antimicrobial prophylaxis (AMP) is typically administered for 3 to 4 days postoperatively in gastric cancer surgery. This far exceeds the recommended 24h or less laid out by the Centers for Disease Control (CDC) guidelines for the prevention of surgical-site infections, after a clean-contaminated operation. Methods: A multicenter randomized phase III trial was designed to evaluate the effect of postoperative AMP in gastric cancer surgery. Patients (pts) were required to have histologically proven gastric cancer which was curable by distal gastrectomy, be classifiable as ASA 1 or 2, and have adequate organ function. Pts were randomized to: (A) perioperative AMP (cefazolin 1g, at <30min before incision, every 3h intraoperative supplements) plus postoperative AMP (cefazolin 1g, twice daily for 2 postoperative days) or (B) perioperative AMP alone. Pts were stratified by institution and ASA. The primary endpoint was the incidence of surgical site infection (SSI). With 171 pts per arm, this study had 80% power to demonstrate non-inferiority with 5% margin of peri-AMP alone and 0.05 1-sided alpha. Results: 355 patients were recruited (A: 179, B: 176) in 7 centers between June 2005 and December 2007. The surgical-site infection rate was 9.0 percent (16 of 178) for peri-/post AMP and 4.5 percent (8 of 176) for peri-AMP alone, with no significant differences (Fisher's exact test: P=0.14, RR=1.98 [95%CI, 0.89–4.44]), but showing a significant non-inferiority (P<0.001). The remote site infection rate was 3.4 percent (6 of 178) for peri-/post AMP and 5.1 percent (9 of 175) for peri-AMP alone, with no significant differences (P=0.44, RR=0.66 [95%CI, 0.25- 1.70]). Conclusions: This multicenter randomized phase III trial confirms that postoperative AMP is unnecessary in patients undergoing distal gastrectomy for gastric cancer. No significant financial relationships to disclose.