Infections caused by extended-spectrum β-lactamase-producing Enterobacterales after rectal colonization with ESBL-producing Escherichia coli or Klebsiella pneumoniae

Abstract

ObjectivesInfections as a result of extended-spectrum β-lactamase-producing Enterobacterales (ESBL-E) are considered infections with a high public health burden. In this study, we aimed to identify incidences of and risk factors for healthcare-associated infections (HAIs) after rectal colonization with ESBL-producing Escherichia coli (ESBL-EC) or Klebsiella pneumoniae (ESBL-KP). MethodsThis prospective cohort study was performed in 2014 and 2015. Patients colonized with ESBL-EC or ESBL-KP were monitored for subsequent HAI with ESBL-E and other pathogens. In the case of an ESBL-E infection, rectal and clinical isolates were compared using pulsed-field gel electrophoresis (PFGE), and whole-genome sequencing (WGS) for ESBL-KP isolates. Proportional hazard models were applied to identify risk factors for HAIs, and to analyse competing risks. ResultsAmong all patients admitted to the hospital during the study period, 13.6% were rectally screened for third-generation cephalosporin-resistant Enterobacterales (3GCREB). A total of 2386 rectal carriers of ESBL-EC and 585 of ESBL-KP were included in the study. Incidence density (ID) for HAI with ESBL-E was 2.74 per 1000 patient days at risk (95% confidence interval (CI) 2.16–3.43) among carriers of ESBL-EC, while it was 4.44 per 1000 patient days at risk (95% CI 3.17–6.04) among carriers of ESBL-KP. In contrast, ID for HAI with other pathogens was 4.36 per 1000 patient days at risk (95% CI 3.62–5.21) among carriers of ESBL-EC, and 5.00 per 1000 patient days at risk (95% CI 3.64–6.69) among carriers of ESBL-KP. Cox proportional hazard regression analyses identified colonization with ESBL-KP (HR = 1.58, 95% CI 1.068–2.325) compared with ESBL-EC as independent risk factor for HAI with ESBL-E. The results were consistent over all competing risk analyses. ConclusionsClinicians should be aware of the increased risk of ESBL-E infections among patients colonized with ESBL-KP compared with ESBL-EC that might be caused by underlying diseases, higher pathogenicity of ESBL-KP and other factors.