Risk Factors for Colonization or Infection with Vancomycin-Resistant Enterococcus on an Inpatient Hematology/Oncology Ward

Background: Vancomycin-resistant Enterococcus (VRE) continues to be a major cause of nosocomial infection with associated morbidity and mortality.  Infection control surveillance at a tertiary care center found an increase in incidence of clinically reported VRE infection in November 2009 among patients on an inpatient Hematology/Oncology ward.  As part of outbreak containment, we wished to determine risk factors predisposing this population to colonization and infection with VRE.

Objective: To determine the point-prevalence of VRE colonization on the affected ward and to identify risk factors for VRE colonization and infection.

Methods: Screening rectal swabs were obtained from 39 inpatients on the Hematology/Oncology ward over 4 weeks at a single tertiary care medical center  in April and May 2010 and cultured for VRE using selective media.  Medical records for these patients were reviewed for potential risk factors for colonization.  Patients with positive screening cultures were identified as being colonized.  Patients with negative cultures served as controls.  Additionally, patients with VRE-positive clinical cultures obtained on the ward over the previous six months were also included as cases for identification of risk factors for infection.  Univariate analysis was undertaken with Chi-square test for categorical and T-test for continuous variables.  Multivariate testing was done using logistic regression.

Results: 5 of 39 (12.8%) screening rectal cultures were positive for VRE.  An additional 4 cultures grew intermediately-resistant Enterococcus.  Eleven clinical specimens from the same ward over the preceeding 6 months were identified as VRE-positive.  16 cases and 34 controls were included.  The mean age of the study population was 60 (SD 14.5, range 20-89 yrs).  Significant (p < 0.05) major risk factors for colonization or infection in univariate analysis include exposure to intravenous vancomycin, fluoroquinolones, extended-spectrum penicillins, and fourth generation cephalosporins; total number of antibiotics received; antibiotic-days; length of stay; presence of a central venous catheter; a previous diagnosis of a hematologic malignancy; a history of receiving bone marrow transplantation; and renal failure. Protective factors on univariate analysis included statin use at time of admission.  When these were entered into multivariate stepwise logistic regression, only exposure to IV vancomycin remained significant, with OR 1.79 (95% CI 1.27-2.51).

Conclusions: Recent intravenous vancomycin exposure is a significant risk factor for colonization or infection with vancomycin-resistant Enterococcus. Outbreak investigation for VRE should include assessment of antibiotic use.