Main Menu Browse by Day or Program Author Keywords

614 Characterizing and Containing VRE Acquisition Among Stem Cell Transplant and Medical Oncology Inpatients

Sunday, April 3, 2011
Trinity Ballroom (Hilton Anatole)
Robin T. Odom, MS , National Institutes of Health Clinical Center, Bethesda, MD
Ninet Sinaii, PhD , National Institutes of Health Clinical Center, Bethesda, MD
Philip L. Mussenden Jr., BS , Howard University College of Medicine, Washington, DC
Frida Stock, BS , National Institutes of Health Clinical Center, Bethesda, MD
MaryAnn Bordner, MS , National Institutes of Health Clinical Center, Bethesda, MD
Angela V. Michelin, MPH , National Institutes of Health Clinical Center, Bethesda, MD
Daniel P. Fedorko, PhD , National Institutes of Health Clinical Center, Bethesda, MD
Patrick R. Murray, PhD , National Institutes of Health Clinical Center, Bethesda, MD
David K. Henderson, MD , National Institutes of Health Clinical Center, Bethesda, MD
Tara N. Palmore, MD , National Institutes of Health Clinical Center, Bethesda, MD
Background: Vancomycin-resistant Enterococcus faecium (VRE) has become a scourge of hematopoietic stem cell transplant (HSCT) and medical oncology wards. VRE is transmitted person-to-person or selected by antibiotic use. Our hospital experienced a surge in VRE infection and colonization in a HSCT ward and a medical oncology ward in the spring of 2009.

Objective: To address an outbreak using active surveillance, pulsotyping of every isolate, and a case-control study to determine risk factors for acquisition of VRE.

Methods: From 7/09 through 6/10, rectal swabs sent on admission and then weekly from patients not known to be colonized in the affected wards were cultured for VRE on selective media. Starting in 7/10, swabs were tested by vanA/B PCR (Cepheid), and those with positive results were cultured. Strains were typed using pulsed-field gel electrophoresis. A strain was classified as “nosocomial” if its pulsotype matched, and “unique” if it did not match, that of another isolate. Infection control interventions included improved environmental cleaning, enhanced contact precautions, and frequent feedback to the wards. A case-control study was performed using cases that occurred between 7/09 and 4/10. Cases were defined as patients with no prior known VRE infection or colonization who grew VRE from surveillance or clinical cultures during the study. Controls had negative surveillance cultures, and were matched 3:1 by ward and date of sampling. Data were analyzed by conditional logistic regression analyses.

Results: VRE colonization or infection was detected in 76 patients from 7/09 to 10/10; 43 had unique strains and 26 had nosocomial strains; 7 patients had positive PCRs but negative cultures, thus no associated strain typing.  Case-control analysis of 50 cases between 7/09 and 4/10 highlighted HSCT, gut GVHD, and receipt of vancomycin as risk factors for VRE acquisition in a multivariable model (p<0.05, <0.05, <0.01, respectively). A patient’s odds of acquiring a nosocomial strain of VRE increased by 5% per day of exposure to VRE colonization pressure in the ward (OR 1.05; p<0.04).

Active surveillance stopped after 6 months on the oncology unit, as no new cases appeared after the 3rd month of the study. Active surveillance is ongoing in the HSCT unit; however, only 1 isolate between 5/10 and 10/10 has been identified as a nosocomial strain.

Conclusions: HSCT patients with gut GVHD are at particularly high risk of acquiring VRE. Active surveillance, enhanced environmental cleaning, and improved isolation adherence were temporally associated with control of nosocomial transmission. Even after transmission was curbed, patients continued to acquire VRE with unique strains, suggesting selection due to antimicrobial use, either in our institution or elsewhere. Our experience underscores the crucial role for antimicrobial stewardship in controlling VRE acquisition among hematology-oncology patients.