Selective decontamination of the digestive tract in the intensive care unit: current status and future prospects.

OBJECTIVE: To evaluate the available data on selective decontamination of the digestive tract. This therapy aims to prevent infection in critically ill patients admitted to ICUs. Microbial carriage in the oropharynx, stomach, and gut; infection; mortality rate; and antibiotic resistance are the outcome events that are being reviewed. DATA SOURCES: Published indices, abstract booklets, and conference proceedings up to the end of 1990. STUDY SELECTION: Sixteen controlled trials of selective decontamination were identified. None of these trials was conducted as a randomized, placebo-controlled, double-blind trial. DATA EXTRACTION: Each of the three authors independently conducted a meta-analysis (thorough review) of the data from the 16 studies. DATA SYNTHESIS: Of the 16 trials, 15 show significant reduction of acquired infection among patients who received selective decontamination. Of the 14 studies that consider carriage of disease-causing microorganisms as an outcome event, 13 demonstrate a reduction in carriage of Gram-negative bacilli. Only ten studies consider mortality as an end-point, of which four demonstrate that the administration of selective decontamination is associated with decreased mortality rates in certain subgroups but not in the overall mortality rate. Of 11 studies that provide data about antibiotic resistance during selective decontamination, ten reported no increase in resistant microorganisms. Selective decontamination has been reported to be effective in controlling an outbreak of multi-resistant microorganisms.
CONCLUSIONS: There is a general consensus about the efficacy of selective decontamination in diminishing microbial carriage and acquired infection rates, although conclusions about benefits related to mortality rates vary. Differences in mortality rate are found in the selective decontamination studies of patients with curable diseases, including multitrauma and cardiovascular patients. More data on resistance, collected over a longer period of time, are needed. Practical problems of blinding and the major ecological effect of selective decontamination may explain the lack of a randomized, placebo-controlled, double-blind trial.