Decrease of carbapenemase-producing Enterobacteriaceae incidence during the first year of the COVID-19 pandemic.

Dear Editors,

We read with interest the article by Lemenand et al. who showed a decrease in the proportion of extended-spectrum beta-lactamase among E. coli infections in primary care and nursing home during the first year of the coronavirus disease 2019 (COVID-19) pandemic in France. Several factors may have contributed to the decrease of antimicrobial resistance: the reinforcement of hygiene measures in the general population (hand hygiene, limitation of gatherings during lockdown periods), the decrease of antibiotic consumption in primary care, the limitation of the international travel and the increased awareness of healthcare workers’ (HCWs) to comply with infection control measures and hand hygiene.

On the other hand, in hospitals, some factors could have contributed to worsen the spread of antibiotic resistance during the COVID-19 pandemic: increase of bacterial infections, a higher antibiotic use in hospital, a more important proportion of patients requiring hospitalization in intensive care, overcrowding in hospitals and HCWs fatigue after months of COVID-19 surge.2, 3, 4, 5, 6

The aim of our observational study was to evaluate the impact of the COVID-19 pandemic, which began in January 2020 in France, on the evolution of carbapenemase-producing Enterobacteriaceae (CPE) incidence in the Assistance Publique–Hôpitaux de Paris (AP-HP). AP-HP, the largest public healthcare institution in France with 20,000 beds, is a network of 38 university-affiliated public hospitals, spread over Paris and its suburbs. Infection prevention and control teams (IPCT), coordinated by a multidisciplinary central IPCT, are in charge of prevention and surveillance of healthcare-associated infections in each hospital.

Since 2004, AP-HP has implemented a long-term programme for CPE surveillance and control which is based on a bundle of measures to prevent cross transmission, including pre-emptive isolation (contact precautions) and screening for CPE of every patient with a history of stay or hospitalization in a foreign country within the past year.

A case was defined as any patient infected or colonized with CPE and an episode as one index case, followed or not by secondary case(s). For each identification of CPE case, histories of recent hospitalization or stay abroad, the occurrence and number of secondary cases(s) were collected by the IPCT.

Data consisted of the quarterly number of CPE index cases from 2014 to 2020 in AP-HP hospitals. The time-series was decomposed into three components: trend, seasonal and random, from 2014 to 2019 (period 1), using a linear additive model. We calculated the expected number of index cases during the quarters in 2020 with 95% prediction intervals (PI) and compared the number of quarterly observed and forecasted CPE index cases. The statistical programming language 'R' was used.

From 2014 to 2019, 2131 CPE index cases were identified, 380 in 2020, with a history of previous stay abroad in respectively 1438 (67%) and 156 (41%) cases.

Fig. 1 shows the observed and fitted time-series of the quarterly CPE index cases per 100,000 hospitalization-days (HD) in period 1 and the predicted values with 95%PI for 2020. We observed a significant linear upward trend (p < 10−6) of +0.34 CPE cases/100,000 HD each quarter and a seasonal component with higher number in the third quarter, corresponding to summer periods (p < 10−3). By comparing the number of observed and forecasted CPE incidence in 2020, it emerged that the observed cases did not fit the projection. Indeed, the observed CPE incidence was lower than the predicted lower limit of the 95% PI forecast from the second quarter.

Fig. 1

Observed carbapenemase-producing Enterobacteriaceae index cases from 2014 to 2020 and 2020 forecasted cases with 95% prediction intervals.

Analysis of the CPE incident cases with a recent stay abroad showed a significant upward trend (p < 10−6) of +0.17 CPE cases/100,000 HD each quarter and significant seasonality (p < 10−5). By comparing the number of quarterly observed and forecasted CPE cases with a history of stay abroad, the observed CPE incidence was lower than the predicted lower limit of the 95% PI forecast from the second quarter of 2020 (Fig. 2 A). Analysis of the CPE cases without a history of stay abroad showed a similar upward trend (p < 10−6) of +0.17 CPE cases/100,000 HD each quarter but no seasonal component (Fig. 2B). Moreover, the number of observed CPE cases without a history of stay abroad remained in the 95% PI forecast during 2020.

Fig. 2

Observed carbapenemase-producing Enterobacteriaceae index cases with (2A) and without (2B) a known history of stay abroad from 2014 to 2020 and 2020 forecasted cases with 95% prediction intervals.

The mean number of secondary cases per episode remained stable with 399 for 2131 episodes in period 1 (0.2 cases/episode) and 84 for 380 episodes in 2020 (0.2 cases/episode) which does not argue for an increase of CPE intra-hospital transmission during the pandemic.

Finally, this study shows that the annual number of index CPE cases dramatically decreased in 2020 in the largest public health institution in France. This concerned the CPE index cases with a history of stay abroad while those without continued to increase following the same trend as in previous years. This is probably the result of governmental restrictions during the pandemic which have limited international travel, the most important source of CPE in countries as France with a low incidence. In contrast, the evolution of the CPE incidence without a recent stay abroad does not seem to have been affected during the pandemic: the reinforcement of hygiene measures in the general population and in hospitals did not bend the incidence curve, but the modified conditions of care in hospitals did not increase the cross transmission either.

This study has several limitations. Firstly, the completeness of the data is questionable because the IPCT had little time to do these reports in 2020, even if a catch-up was made retrospectively in 2021. In addition, a decrease in CPE screening in 2020 is possible due to the overload of work in the intensive care unit and a shortage of swabs that lasted a few weeks during the first COVID-19 wave. However, the fact that the decrease in CPE cases concerned only cases with a history of stay abroad tempers these limitations.

In conclusion, our study shows a decrease in the incidence of CPE cases in our institution during the COVID-19 pandemic. It suggests that this decrease is linked to a decrease in international exchanges. This underlines that, to be effective, the fight against antimicrobial resistance will have to be considered at an international level.

Availability of data and materials

The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Declaration of Competing Interest

The authors declare that they have no competing interests.

Funding

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

CRediT authorship contribution statement

Clarisse Duverger: Visualization, Formal analysis, Writing – original draft, Data curation. Catherine Monteil: Formal analysis, Data curation. Valérie Souyri: Visualization, Formal analysis, Data curation. Sandra Fournier: Visualization, Formal analysis, Writing – original draft, Data curation.