Vancomycin resistant enterococcus (VRE) bloodstream infections (BSI): a case-controlled analysis to assess implications on mortality in haematology patients at a single tertiary referral centre

Abstract

<h3>Aims</h3> To describe vancomycin resistant enterococcus (VRE) patient characteristics and compare mortality rates to case-matched controls. <h3>Methods</h3> Admissions and microbiology from haematology-oncology ward (1/6/2009-1/6/2010) were reviewed. Case-controlled analysis of VRE impact on mortality was undertaken. <h3>Results</h3> VRE isolated in 29 [11 bloodstream infections (BSI)]. Mean length of stay (LOS) was significantly longer in VRE than overall haematology (19.2±4.8 versus 10.8±1.4, <i>p</i>=0.01). Acute myeloid leukaemia (AML) / acute lymphoblastic leukaemia (ALL) patients accounted for 7 VRE (4 BSI). 33 VRE isolates in 29 separate patients: 16 swabs (mostly rectal), 11 BSI, 5 urine. 16/29 VRE patients died, median time-to-death from detection 30 days (0–236). No significant increase in mortality with VRE BSI compared to other modes: 10/16 (63%) swab, 5/11 (45%) BSI and 4/5 (80%) urine. 8/15 deaths due to progressive disease, 3/15 deaths due to overwhelming sepsis. Case-controlled analysis performed: age, case and stage of disease matched analysis comparing 14 VRE and 14 non-VRE. Median follow-up 272 and 364 days, respectively. Mortality rate 50% versus 29%. Majority of deaths due to progressive disease. <h3>Discussion</h3> VRE detection by screening and BSI are increasingly observed at our hospital. Risk factors from our analysis are LOS >20 days and AML/ALL. VRE detection is associated with increased frequency of death but progressive disease remains the most common cause of death. VRE BSI does not appear to confer a worse prognosis.