Selective digestive tract decontamination and vancomycin-resistant enterococcus isolation in the surgical intensive care unit.

Vancomycin-resistant Enterococcus (VRE) has emerged as a significant nosocomial pathogen in the surgical intensive care unit (SICU). We wished to test the hypothesis that the use of selective digestive tract decontamination (SDD) in the SICU affects the frequency of VRE isolation. A retrospective review of hospital records and the SICU database was performed using patients admitted to the SICU service for three or more days from January 1, 1996 to December 31, 1999 at our large tertiary-care teaching hospital. During this time use of SDD in selected patient populations decreased due to physician preference. Information gathered included length of SICU stay, presence of VRE infection or colonization, and use and duration of SDD protocol, vancomycin, and ceftazidime. There were 110 newly diagnosed VRE cases in the SICU during this time period. During the same time period 54 patients received SDD. Eight patients who received SDD had positive VRE cultures and seven had the initial positive culture after receiving SDD. Overall, 9.1% of eligible SICU patients received SDD, 18.5% of patients in the SICU for over 3 days had VRE, 7.3% of VRE patients received SDD, and 13.0% of the SICU patients who received SDD subsequently developed VRE. SDD use was not associated with VRE in univariate analysis. Logistic regression analysis showed higher odds ratios for SDD use in combination with vancomycin than for vancomycin use alone (OR=4.3 vs. 10.9). Odds ratios were over three times higher for SDD plus vancomycin plus ceftazidime use when compared to vancomycin plus ceftazidime use alone (OR=70.5 vs. 19.8). We conclude that administration of SDD alone did not correlate with increased VRE isolation, but that SDD use in conjunction with vancomycin and ceftazidime was associated with VRE isolation.