Viral epidemiology and SARS‐CoV‐2 co‐infections with other respiratory viruses during the first COVID‐19 wave in Paris, France

Abstract Objectives Our work assessed the prevalence of co‐infections in patients with SARS‐CoV‐2. Methods All patients hospitalized in a Parisian hospital during the first wave of COVID‐19 were tested by multiplex PCR if they presented ILI symptoms. Results A total of 806 patients (21%) were positive for SARS‐CoV‐2, 755 (20%) were positive for other respiratory viruses. Among the SARS‐CoV‐2‐positive patients, 49 (6%) had viral co‐infections. They presented similar age, symptoms, except for fever (P = .013) and headaches (P = .048), than single SARS‐CoV‐2 infections. Conclusions SARS‐CoV‐2‐infected patients presenting viral co‐infections had similar clinical characteristics and prognosis than patients solely infected with SARS‐CoV‐2.

all respiratory viruses with SARS-CoV-2, the number of SARS-CoV-2 viral co-infections, and the clinical features of such co-infections.

| ME THODS
All adult patients hospitalized in a COVID-19 first-line hospital in Paris, France, from January 25, 2020, to April 30, 2020, were included. All patients were tested by systematic mPCR testing if they presented ILI symptoms, according to the eCDC definition, and required hospitalization. The two mPCR assays used during the study period, the QIAstat-Dx SARS-CoV-2 respiratory panel, Qiagen, 8  CoV-2 assay (Roche Diagnostics). All these assays provided similar performance and limit of detection. 10,11 Demographic, clinical, and biological features were prospectively collected in the Emergency Department (ED) and retrospectively from the other units. Baseline characteristics within each group were summarized using appropriate descriptive statistics. The statistical analysis was performed using Stata15. The research was approved by the local ethic committee N° CER-2020-6 ( Table 1).

| RE SULTS
A total of 3768 patients were included during the study pe-

| D ISCUSS I ON
This work highlights that 6% of SARS-CoV-2-infected patient presented with viral co-infection at our adult ED. This proportion is higher than previously reported for SARS-CoV-2 6 but at a level similar to the other respiratory viruses. 12 This high prevalence of viral co-infections was observed, despite the limited circulation of other respiratory viruses due to lockdown, curfew, and being in the tail of the season of respiratory viruses. 7 Rhinoviruses, adenoviruses, and other coronaviruses were the most frequently detected viruses with SARS-CoV-2. Adenoviruses and rhinoviruses have already been reported, outside the scope of SARS-CoV-2, as being more frequently involved in viral co-infection, contrary to influenza viruses. 13 In our population, only 6 patients with SARS-CoV-2 were also infected with atypical bacteria. Co-infections can lead to viral interference, one virus limiting or suppressing the replication of the second virus, or to an enhancement of disease severity compared to mono-infection. 14  study is monocentric, and the SARS-CoV-2 epidemic flared in Ile-de-France when the incidence of most respiratory viruses was waning.
Prevalence of viral co-infections with SARS-CoV-2 might be higher in settings with an active circulation of respiratory viruses and/or once social distancing will be over. We also cannot rule out that some specific co-infections might have a deleterious impact, notably SARS-CoV-2/influenza, as only 4 were detected during our study period. Higher severity of SARS-CoV-2/influenza A H1N1pdm2009 has recently been described in golden Syrian hamsters when the two viruses were simultaneously inoculated. 15 We also did not retrieve data on the other pneumonia diagnosis related to pneumococcus or staphylococcus. Thus, although we found that SARS-CoV-2 viral coinfections were rare during the first epidemic wave and did not differ either by their clinical presentation or by their outcome from SARS-CoV-2 mono-infections, this reassuring finding must be confirmed in the upcoming months.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are available from the corresponding author upon reasonable request.

TR A N S PA R EN C Y D ECL A R ATI O N S
DB and BV have received funds for speaking at symposia organized on behalf of Qiagen and have also received funds for research from Qiagen.

PE E R R E V I E W
The peer review history for this article is available at https://publo ns.com/publo n/10.1111/irv.12853.