Antibiotic irrigation during pancreatoduodenectomy to prevent infection and pancreatic fistula: A randomized controlled clinical trial.

Abstract

Surgical site infection affects 25% of patients undergoing pancreatoduodenectomy. This double-blind, randomized controlled trial tested the efficacy of intraperitoneal antibiotic irrigation in decreasing infection and pancreatic fistula after pancreatoduodenectomy. Patients undergoing pancreatoduodenectomy were randomized (1:1 ratio) to intraperitoneal antibiotic (polymyxin B, 500,000 units/L) irrigation or 0.9% NaCl irrigation. All patients received 1 dose of standard parenteral antibiotics within 1 hour of incision. The trial was powered to detect a 15% difference in any surgical site infection (primary endpoint) within 30 days after pancreatoduodenectomy. One hundred ninety patients undergoing pancreatoduodenectomy were randomized: 95 to antibiotic irrigation and 95 to saline irrigation. Groups were well matched regarding demographics, diagnosis, preoperative biliary stenting, bactibilia, texture of the pancreatic parenchyma, pancreatic and bile duct size, portal vein resection, and anastomotic technique. Overall, 30-day surgical site infection was observed in 24 (13%) patients: antibiotic irrigation in 10 (11%) versus saline in 14 (15%) (P = .62). Superficial (n = 9, 5%) and organ-space (n = 15, 8%) surgical site infection rates were 3% and 7% (antibiotic) and 6% and 8% (saline), respectively (P > .31). Clinically relevant postoperative pancreatic fistula occurred in 11 (12%) patients in the antibiotic arm and 10 (11%) in saline controls (P > .95). The addition of antibiotic solution to intraperitoneal irrigation does not decrease the incidence of postoperative infectious complications or pancreatic fistula after pancreatoduodenectomy.