Unintended Consequences of Pre-transplant VRE Screening on Antimicrobial Stewardship Among Allogeneic Hematopoietic Cell Transplant Recipients

Abstract

BackgroundPre-transplant screening for VRE is commonly performed among patients prior to hematopoietic cell transplantation (HCT) although its role in prevention of horizontal transmission is controversial. The impact of pre-transplant VRE colonization on antimicrobial stewardship and use of VRE therapy is unknown. The purpose of this investigation is to determine whether pre-transplant VRE colonization affects the use of VRE therapy among patients during the first 100 days post-transplant.MethodsWe analyzed patients >18 years old within the first 100 days post- allogeneic HCT at a cancer center from September 1, 2007 to August 31, 2016 from a prospectively collected database. We performed retrospective chart review among patients who did and did not develop VRE bacteremia to obtain antimicrobial utilization data for agents with in vitro activity against VRE based on colonization status. Patients colonized post-HCT and those with positive surveillance blood cultures were excluded from analysis. Continuous variables were assessed using t-tests or Mann–Whitney U tests for normal or non-normal data respectively.ResultsOf 1402 allogeneic HCT patients, 203 (14%) were colonized pre-transplant. Among 22 (1.6%) patients who developed VRE bacteremia, 19 (86%) were colonized pre-transplant. Eight (42%) colonized patients received empiric VRE therapy within 24 hours of blood culture collection compared with 0 (0%) of non-colonized patients (P = 0.23). Among the 1371 patients who did not develop VRE bacteremia, 66 (5%) received VRE therapy. 32/179 (18%) of the colonized group received VRE therapy compared with 34/1192 (3%) non-colonized (P <.001). The median duration of VRE therapy was 3 days in the colonized group and 2 among non-colonized (P = 0.86). Indications for VRE therapy among colonized patients without bacteremia were empiric therapy (63%), therapy for VRE isolated from non-blood sites (19%), alternative therapy for non-VRE Gram-positive infection (13%), and unknown (6%).ConclusionIn our population, VRE-colonized patients received significantly more VRE therapy than non-colonized patients in the absence of VRE bacteremia. Among patients who developed bacteremia, less than half received timely initiation of empiric therapy.Disclosures S. Pergam, Merck: Consultant and Investigator, Consulting fee.