Selective decontamination of the digestive tract to prevent postoperative infection: A randomized placebo-controlled trial in liver transplant patients

OBJECTIVETo determine the efficacy of selective decontamination of the digestive tract (SDD) in patients undergoing elective transplantation of the liver. DESIGNRandomized, double-blind, placebo-controlled study. SETTINGTwo academic teaching hospitals. PATIENTSAdult patients undergoing elective liver transplantation26 patients receiving SDD and 29 patients receiving a placebo. INTERVENTIONSPatients undergoing SDD were administered 400 mg of norfloxacin once daily as soon as they were accepted for transplantation. Postoperative treatment for this group consisted of 2 mg of colistin, 1.8 mg of tobramycin, and 10 mg of amphotericin B, four times daily, combined with an oral paste containing a 2% solution of the same drugs until postoperative day 30. Prophylactic intravenous administration of antibiotics was not part of the SDD regimen in this study. Control patients were given a similar regimen with placebo drugs. MEASUREMENTSThe mean number of postoperative bacterial and fungal infections in the first 30 days after transplantation was the primary efficacy end point. Days on a ventilator, days spent in the intensive care unit, and medical costs were registered as secondary outcome variables. MAIN RESULTSOf the 26 patients undergoing SDD, 22 (84.5%) developed an infection in the postoperative study period; in the placebo group (n = 29), these numbers were not significantly different (25 patients, 86%). The mean number of postoperative infectious episodes per patient was also not significantly different1.77 (SDD) vs. 1.93 (placebo). Infections involving Gram-negative aerobic bacteria and Candida species were significantly less frequent in patients receiving SDD (p < .001 and p < .05). Total costs were higher in the group receiving SDD. CONCLUSIONSSelective decontamination of the digestive tract does not prevent infection in patients undergoing elective liver transplantation and increases the cost of their care. It does, however, affect the type of infection. Infections with Gram-negative bacilli and with Candida species are replaced by infections with Gram-positive cocci.