Screening for vancomycin resistant enterococcal (VRE) colonization/infection in critically III patients

We performed a 3 month prospective study of patients requiring >72 hour ICU admission to determine strategies for identification of VRE colonization or infection. Demographic data was collected and surveillance samples from the oropharynx and rectum, tracheal and gastric aspirates and urine specimens were cultured for the presence of VRE on admission to the ICU and then twice weekly until discharge; additional cultures were performed when clinically indicated. Of 160 patients enrolled, 31(19%) were colonized by VRE. Of these colonized patients, 12 (39%) developed infections, ten of which were incident in the ICU. Of the 12, 6 patients were colonized prior to infection, with a mean time to infection in the remaining patients of 9 days. Infections included 5 urinary, 3 intraabdominal, 8 bacteremias, 1 pleural and 4 deep wound infections. The ICU mortality was 25%, hospital mortality was 29%. Among the colonized patients, frequency of new colonization varied by surveillance site: rectal with 16/28(57%), oropharynx 7/14(50%), gastric 8/12(67%), sputum 7/8(87%), and urine 7/8(87%). The rectum was the first site of colonization in 26/28 cases (93%), while gastric was the site in 6/12 (50%), oropharynx 6/14(43%), and urine and sputum both 2/8(25%). Positive rectal surveillance cultures preceded 60% (6/10) infections incident in the ICU. In light of a 39% VRE infection rate, rectal surveillance is an effective strategy for identifying high risk critically ill patients.