Risk Factors and Outcomes for Vancomycin-ResistantEnterococcusBloodstream Infection in Children

Abstract

Background and Objective.Enterococcal bloodstream infections (BSIs) cause morbidity and mortality in children. This study aims to describe the epidemiological characteristics of enterococcal BSI, to determine the risk factors for vancomycin-resistantEnterococcus(VRE) BSI, and to compare outcomes of VRE BSI and vancomycin-susceptibleEnterococcus(VSE) BSI in this population.Methods.A retrospective cohort study at a 418-bed tertiary care children's hospital in Philadelphia, Pennsylvania, examined the epidemiological characteristics of children hospitalized with enterococcal BSI during the period from 2001 through 2006. A nested case-control study compared patients with VRE BSI with control patients with VSE BSI. Analysis included regression modeling to identify independent risk factors for VRE BSI.Results.We identified 339 patients with enterococcal BSI during the study period, including 39 patients with VRE infection. Fifty-three patients (16%) died before hospital discharge. Risk factors for VRE included long-term receipt of mechanical ventilation (adjusted odds ratio [OR], 5.40 [95% confidence interval {CI}, 1.28-6.48]), receipt of immunosuppressive medications during the preceding 30 days (adjusted OR, 2.88 [95% CI, 1.40-20.78]), use of vancomycin during the 2 weeks before onset of bacteremia (adjusted OR per day of vancomycin use, 1.25 [95% CI, 1.14-1.38]), and older age (adjusted OR, 1.08 [95% CI, 1.03-1.14]). VRE BSI was not associated with an increased length of stay after onset of bacteremia (0.77 days [95% CI, 0.55-1.07 days]). Mortality was higher for VRE BSI, but the difference was not statistically significant (adjusted OR, 1.94 [95% CI, 0.78-4.8]).Conclusion.Most enterococcal BSI in children was caused by VSE. Risk factors for VRE BSI included receipt of vancomycin, long-term receipt of mechanical ventilation, immunosuppression, and older age. Differences in length of stay and mortality were not detected.