SELECTIVE DECONTAMINATION OF THE DIGESTIVE-TRACT IN MULTIPLE TRAUMA PATIENTS - IS THERE A ROLE - RESULTS OF A PROSPECTIVE, DOUBLE-BLIND, RANDOMIZED TRIAL

Objective: To evaluate the efficacy of the technique of selective decontamination of the digestive tract in preventing the development of secondary infection and its influence on morbidity and mortality rates in multiple trauma patients with chest injuries requiring intermittent positive-pressure ventilation. Design: Prospective, double-blind, randomized study. Setting: A multidisciplinary respiratory intensive care unit (ICU) in a 1,500-bed teaching hospital. Patients: Seventy-two patients (mean Injury Severity Score of 29.5) who were intubated for >48 hrs and remained in the ICU for >5 days. Interventions: Patients were randomized on admission to receive selective decontamination therapy or placebo. All patients received intravenous cefotaxime for 72 hrs and the treatment group received oral and enteral selective decontamination with amphotericin B, polymyxin E, and tobramycin (n = 39), while the placebo group received a placebo containing oral paste and enteral solution (n = 33). Measurements: Secondary infection was determined clinically and microbiologically and surveillance cultures were monitored for gastrointestinal colonization Results: The patient groups were fully comparable for age, severity of illness, and compromising factors, There was no difference in the number of patients infected (11 treatment group vs. 11 placebo), infections (17 vs. 16) and deaths (5 vs. 3); the duration of ICU (15.5 vs. 14.2 days) and hospital stays (26.3 vs. 25.5) were also similar. Microbiological surveillance cultures confirmed effective elimination of aerobic Gram-negative bacilli, and infections in the treatment group were largely due to Staphylococcus aureus and Staphylococcus epidermidis. Conclusion: We have been unable to show any benefit from the use of selective decontamination of the digestive tract in the prevention of secondary infections in multiple trauma patients.