The effect of carbapenem-resistant versus carbapenem-susceptible Enterobacterales infections on patient outcomes at an academic medical center

Abstract

Abstract Objective: We sought to compare patient outcomes between carbapenem-resistant Enterobacterales (CRE) and carbapenem-susceptible Enterobacterales (CSE) infections at our academic medical center. Design: We conducted a retrospective cohort study of adult patients with a positive culture of E. coli, E. cloacae, K. aerogenes, K. oxytoca, and/or K. pneumoniae admitted at UK HealthCare (January 1, 2010–December 31, 2019). Based on the type of pathogen on the date of the first culture (index date), patients were included in the CRE (i.e., exposed) group, or the CSE (comparator) group. Exclusion criteria were age < 18 years old, pregnancy, endocarditis, osteomyelitis, necrotizing fasciitis, or cystic fibrosis. We evaluated the impact of CRE vs CSE on a composite outcome of 30-day of all-cause mortality or discharge to hospice using Kaplan–Meier survival curves and Cox proportional hazard regression with inverse probability of treatment weights (IPTW). Results: Of 17,839 hospitalized patients, 128 and 6,953 patients were included in the CRE and CSE groups, respectively. Baseline differences existed in sex-assigned-at-birth, admission source, time-to-index culture, and infection type/severity. Most CRE index cultures observed (76%) only exhibited resistance to ertapenem. IPTW-adjusted HR [95% CI] of composite outcome was 0.99 [0.65, 1.51] after 30 days. Follow-up analysis in patients with carbapenem-non-susceptible Enterobacteralesbloodstream infections on index yielded an HR of 1.38 [0.85, 2.24]. Conclusions: Risk of composite outcome was not estimated to differ between patients with CRE and CSE in the overall analysis. Although follow-up analysis identified an increased risk, we cannot statistically distinguish this from a null effect.