Better outcome for ICU patients after removal of bacteria from digestive tract

NB. Please note that if you are outside North America, the embargo for LANCET press material is 0001 hours UK Time 26 September 2003.

Selective use of SDD has occurred for two decades although its effect on infection with antibiotic-resistant bacteria and death is not clear. In a randomised unblinded trial, Evert de Jonge from Amsterdam's Academic Medical Center, Netherlands, and colleagues randomised 934 patients admitted to a surgical and medical ICU to receive either appropriate antibiotic treatment to facilitate SDD, or standard treatment (the control group).

There were fewer deaths in ICUs (15%) and fewer overall hospital deaths (24%) among SDD patients compared with those given standard treatment (23% and 31%, respectively). Infection with antibiotic-resistant bacteria in ICUs was also lower among SDD patients (16%) than patients given standard treatment (26%).

Evert de Jonge comments: "We conclude that SDD may improve survival of ICU patients and lowers the acquisition of resistant gram-negative aerobic bacteria. In ICUs that have low prevalence of vancomycin-resistant enterococcus and meticillin-resistant S aureus, we advocate the use of SDD in all patients expected to be on mechanical ventilation for at least 2 days or to be in the ICU for at least 3 days."

In an accompanying Commentary (p 1006), Jean-Louis Vincent from Free University of Brussels, Belgium, concludes: "So should SDD be applied routinely in all ICUs? To the question does SDD work, the answer now must definitely be yes-SDD reduces mortality. But, do the data apply to all environments? Despite the results in de Jonge's study from the Netherlands, the risks of bacterial resistance remain. Whether to use SDD or not will depend on the risk of resistant organisms in a given environment, and the population of patients. So, in ICUs in an area with a high incidence of vancomycin-resistant enterococci or meticillin- resistant S aureus, SDD may not be appropriate, and in general, surgical and trauma patients will benefit more than medical patients who enter the ICU already colonised. Whatever individual units decide, regular surveillance samples must be taken to monitor the long-term effects of this intervention."

Contact: Dr Evert de Jonge, c/o Edith Gerritsma or Arthur van Zuylen, Department of Public Relations, Academic Medical Center, PO Box 22660, 1100 DD Amsterdam, The Netherlands;
T) 31-20-566-2929;
F) 31-20-696-7899;
E) [email protected]

Professor Jean-Louis Vincent, Department of Intensive Care, Erasme Hospital, Free University of Brussels, 1070 Brussels, Belgium;
T) 32-2-555-3380;
F) 32-2-555-4555;
E) [email protected]