Selective decontamination to reduce gram-negative colonisation and infections after oesophageal resection.

181 patients undergoing resection of the oesophagus for carcinoma were randomised to receive selective decontamination (test group) or conventional perioperative antibiotic prophylaxis (controls). 114 patients were finally included in the study: 12 of 56 test patients had 18 infections, whereas 32 of 58 controls acquired 51 infections. Colonisation with aerobic gram-negative microorganisms, and the number of postoperative respiratory tract infections were significantly lower in the test patients. The postoperative therapeutic use of antibiotics was significantly lower in the test group. No endogenous infections were caused by gram-negative bacilli in the test group. Selective decontamination reduces colonisation with gram-negative bacilli and postoperative infections after resection of the oesophagus.