Estimating the number and incidence of carbapenemase-producing Enterobacterales infections in France in 2020: A capture-recapture study

Objectives

Even though France faces few severe infections due to carbapenem-producing Enterobacterales (CPE), inter-regional epidemic stages render their dissemination a cause for considerable concern. CPE reporting relies in France on three non-exhaustive monitoring systems (MS): an early-alert system, a nationwide passive surveillance system and the National Reference Centre. We aim to estimate the number and incidence of CPE-related infections in France in 2020 and to identify any overlap between the three systems to determine whether their continued use still serves a purpose.

Methods

Data on clinical CPE isolates in 2020 were extracted from the three MS databases. Screening samples were excluded. Datasets were manually merged, isolate by isolate, so as to identify in which system(s) each isolate was reported. A system-participant was defined as any declarant reporting at least one isolate in an MS. Using our matched dataset, we performed Bayesian model averaging for capture-recapture estimations.

Results

All in all, 1722 CPE isolates were reported through the monitoring systems in 2020. We estimated that the number of CPE infections was almost twice this number, corresponding to incidence of 0.031 CPE/1000 hospital-days [CI95% 0.015–0.057/1,000 hospital-days], with regional disparities taken into account. Among participating the laboratories, 86% were involved in only one of the systems. Among clinical CPE isolates, 56% were isolated from urine.

Conclusion

Regarding this rare infection, surveillance based only on passive surveillance from voluntary hospitals does not reflect actual epidemiology. We recommend maintaining the three monitoring systems and improving the participation of hospitals’ nationwide surveillance, the objective being to more accurately capture the real incidence of CPE infections.