The Mortality of Vancomycin-Resistant Enterococci Bloodstream Infections (VRE BSI)

Abstract

4928 Background VRE are nosocomial pathogens with resistance to most commonly used antimicrobial agents. VRE BSI occurs in as few as 4% of patients colonised with VRE[i]. A review of factors contributing to the development of VRE BSI was performed to improve patient safety at St George Hospital. Alarmingly, there were 9 new cases of VRE BSI detected amongst haematology inpatients in early 2010, as compared with only 2 in the previous 6 months. In high risk populations, the rate of VRE BSI amongst patients colonised with VRE can be as high as 29%[ii]. VRE BSI is associated with an increased length of hospital stay from 10.5 to 46 days[iii][iv]and an estimated increased cost of $27,190 per patient[v]. Annual net savings of $100,000-150,000 can be achieved by hospitals detecting 6–9 cases of VRE BSI per year by utilising enhanced infection control strategies[vi]. Methods Admission details for haematology in-patients and their microbiology results between 1/6/2009-30/11/2010 were reviewed. A subsequent case-controlled analysis was performed matching for patient age, disease and disease stage. Interventions to reduce the rate of VRE transmission were introduced in July 2010: improved hand hygiene education, additional staffing allocations, additional cleaning services, antimicrobial stewardship, improved patient education, increased staff awareness, monthly census screening of all patients on the 4 East Oncology/Haematology ward for VRE and contact tracing measures. Results 471 patients were admitted a total of 943 times. VRE was isolated in 61 patients. Average length of stay was significantly longer in patients with VRE than for patients without VRE [16.3±3.0 vs 8.4±1.0 days, p=0.01]. 16 patients had VRE BSI. After a median follow-up of 9.8 months, eight of these patients have died (50% mortality), mostly due to progressive disease. 45 patients had non-BSI VRE and were followed up for a median 8.4 months: 24 died (53% mortality), also mostly due to progressive disease. Case-controlled Analysis An age, case and stage of disease matched analysis compared 14 patients with VRE and 14 with no evidence of VRE. Median follow-up from presentation with disease was 16.3 months and from VRE detection was 8 months. Mortality rate was 64% vs 29%. Seven out of nine deaths in the VRE cohort were due to progressive disease. Conclusions The clinical significance of VRE BSI over other modes of detection remains uncertain for individual patients, however high rates may reflect uncontrolled VRE transmission. Case controlled analysis demonstrates an associated additional mortality risk for VRE positive patients. Disclosures: No relevant conflicts of interest to declare.