Rising Rates of Carbapenem-Resistant Enterobacteriaceae in Community Hospitals: A Mixed-Methods Review of Epidemiology and Microbiology Practices in a Network of Community Hospitals in the Southeastern United States

Abstract

(See the commentary by Pfeiffer and Beldavs, on pages 984–986.)ObjectiveDescribe the epidemiology of carbapenem-resistant Enterobacteriaceae (CRE) and examine the effect of lower carbapenem breakpoints on CRE detection.DesignRetrospective cohort.SettingInpatient care at community hospitals.PatientsAll patients with CRE-positive cultures were included.MethodsCRE isolated from 25 community hospitals were prospectively entered into a centralized database from January 2008 through December 2012. Microbiology laboratory practices were assessed using questionnaires.ResultsA total of 305 CRE isolates were detected at 16 hospitals (64%). Patients with CRE had symptomatic infection in 180 cases (59%) and asymptomatic colonization in the remainder (125 cases; 41%). Klebsiella pneumoniae (277 isolates; 91%) was the most prevalent species. The majority of cases were healthcare associated (288 cases; 94%). The rate of CRE detection increased more than fivefold from 2008 (0.26 cases per 100,000 patient-days) to 2012 (1.4 cases per 100,000 patient-days; incidence rate ratio (IRR), 5.3 [95% confidence interval (CI), 1.22–22.7]; P = .01). Only 5 hospitals (20%) had adopted the 2010 Clinical and Laboratory Standards Institute (CLSI) carbapenem breakpoints. The 5 hospitals that adopted the lower carbapenem breakpoints were more likely to detect CRE after implementation of breakpoints than before (4.1 vs 0.5 cases per 100,000 patient-days; P < .001; IRR, 8.1 [95% CI, 2.7–24.6]). Hospitals that implemented the lower carbapenem breakpoints were more likely to detect CRE than were hospitals that did not (3.3 vs 1.1 cases per 100,000 patient-days; P = .01).ConclusionsThe rate of CRE detection increased fivefold in community hospitals in the southeastern United States from 2008 to 2012. Despite this, our estimates are likely underestimates of the true rate of CRE detection, given the low adoption of the carbapenem breakpoints recommended in the 2010 CLSI guidelines.