Increased mortality in hospital- compared to community-onset carbapenem-resistant enterobacterales infections

Abstract

Background: The CDC reported a 35% increase in hospital-onset (HO) carbapenem-resistant Enterobacterales (CRE) infections during the COVID-19 pandemic. We evaluated patient outcomes following HO and community-onset (CO) CRE bloodstream infections (BSI).

Methods: Patients prospectively enrolled in CRACKLE-2 from 56 hospitals in 10 countries between 30 April 2016 and 30 November 2019 with a CRE BSI were eligible. Infections were defined as CO or HO by CDC guidelines, and clinical characteristics and outcomes were compared. The primary outcome was desirability of outcome ranking (DOOR) 30 days after index culture. Difference in 30-day mortality was calculated with 95% CI.

Results: Among 891 patients with CRE BSI, 65% were HO (582/891). Compared to those with CO CRE, patients with HO CRE were younger [median 60 (Q1 42, Q3 70) years versus 65 (52, 74); P < 0.001], had fewer comorbidities [median Charlson comorbidity index 2 (1, 4) versus 3 (1, 5); P = 0.002] and were more acutely ill (Pitt bacteraemia score ≥4: 47% versus 32%; P < 0.001). The probability of a better DOOR outcome in a randomly selected patient with CO BSI compared to a patient with HO BSI was 60.6% (95% CI: 56.8%-64.3%). Mortality at 30-days was 12% higher in HO BSI (192/582; 33%) than CO BSI [66/309 (21%); P < 0.001].

Conclusion: We found a disproportionately greater impact on patient outcomes with HO compared to CO CRE BSIs; thus, the recently reported increases in HO CRE infections by CDC requires rigorous surveillance and infection prevention methods to prevent added mortality.

Trial registration: ClinicalTrials.gov NCT03646227.

Figure 1.

Kaplan–Meier curve of all-cause 30-day mortality by infection onset. Survival for 889 patients with CRE blood stream infection by infection onset type (CO versus hospital onset). Two subjects (one in each group) are excluded from the figure due to missing mortality outcome information. One subject in the HO group died on day 30 and is still considered at risk in the figure.