Trends in incidence of vancomycin-resistant Enterococcus colonization and bacteremia among allogeneic hematopoietic cell transplant recipients at a large cancer center

Abstract

Background: Vancomycin-resistant Enterococcus (VRE) are important hospital-acquired pathogens among hematopoietic cell transplant (HCT) recipients. We examined the incidence and outcomes of patients with VRE colonization and bacteremia (VREB) over a ten-year period at a center that routinely screens and uses barrier precautions for VRE. Methods: Adults receiving their first allogeneic HCT at our center between September 2007 and August 2016 were eligible for inclusion. Patients who were positive either by standardized pre-HCT stool/rectal screening or at any point two years prior to HCT were considered VRE colonized. Patients with acquired VRE were those with positive VRE cultures only post-HCT. Colonization and 100-day post-HCT VREB incidence rates were compared over time using linear regression. Cox proportional hazards models were constructed to assess the relationship between 100-day mortality and: a) pre-HCT colonization, and b) the number of days with sequential VREB cultures. Pre-transplant Assessment of Mortality (PAM) scores were calculated to allow adjustment for underlying disease severity. Results: Of 1,492 eligible HCT recipients, 203 (14%) were colonized with VRE pre-HCT; an additional 90 (6.0%) acquired VRE colonization post-HCT. Forty-two patients (2.8%) developed VREB, the majority among those colonized with VRE (32 [76%] vs. 10 [24%] non-colonized). The cumulative incidence of VREB for the cohort was 2.9 per 10,000 patient-days. Over the study period, there were no significant changes in VRE colonization or VREB (p-values>0.1). Those with multiple days with positive VRE blood cultures had higher mortality than those with one positive culture (HR 3.23; 95%CI: 0.88, 11.8). Patients with pre-HCT colonization had an increased risk of death compared to non-colonized patients in an unadjusted model (HR 2.1: 95%CI: 1.4, 3.2) and after adjustment by PAM score (HR = 2.2; 95%CI: 1.5, 3.3). Patients with higher PAM scores and VRE colonization pre-HCT had higher mortality than non-colonized patients with high PAM scores. Conclusion: Despite nearly 15% of patients with pre-HCT colonization in our cohort, VREB was an infrequent post-HCT complication. We identified a subgroup of patients at high risk of VREB who may be targeted for VRE-specific measures. Studies examining the impact of antimicrobial stewardship programs are needed to inform infection prevention interventions.