Prevention of anastomotic leak in rectal cancer surgery with local antibiotic decontamination: a prospective, randomized, double-blind, placebo-controlled single center trial.

ABSTRACT

PURPOSE: Anastomotic leak and other infectious complications are septic complications of rectal cancer surgery caused by bacteria. Data from registry analysis show a beneficial effect of local antimicrobial administration on anastomotic leaks, but data are inconsistent in recent clinical trials. Therefore, our aim was to study the efficacy of topical antibiotic treatment on the incidence of anastomotic leaks in rectal cancer surgery. METHODS: A prospective, randomized, double-blind and placebo-controlled, single center trial was conducted. Patients received either placebo and amphotericin B or decontamination with polymyxin B, tobramycin, vancomycin, and amphotericin B four times per day starting the day before surgery until postoperative day 7. If a protective ileostomy was created, a catheter was placed transanally and the medication was administered locally to the anastomotic site. All patients received an intravenous perioperative antibiotic prophylaxis. RESULTS: The trial had to be stopped for ethical reasons after first interim analysis with 80 patients instead of the initially planned 280 patients. Of the 40 patients randomized to receive placebo, eight (20%) developed anastomotic leak compared to only 2 (5%) in the treatment group of 40 patients (decontamination) with significant difference in the χ2 test (p = 0.0425). Twenty percent of the placebo group and 12.5% in the treatment group developed infectious complications not associated with anastomotic leak (p = 0.5312). One patient (2.5%) in the placebo group died (p = 0.3141). CONCLUSION: Local decontamination with polymyxin, tobramycin, vancomycin, and amphotericin B is safe and effective in the prevention of anastomotic leak in rectal cancer surgery.