Shifts in Respiratory Virus Epidemiology on Reunion Island From 2017 to 2023: Impact of COVID-19 Pandemic and Non-Pharmaceutical Interventions

Abstract

The emergence of coronavirus disease 2019 (COVID-19), caused by the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), in late December 2019 in Wuhan, China, has had a profound impact worldwide [1, 2]. Reunion Island, situated in the southwest of the Indian Ocean and spanning 2500 km² as a French overseas department, is inhabited by around 885,700 people, primarily residing along the coastal areas. On this island, the first case of COVID-19 was reported on March 11, 2020, at the University Hospital Center (UHC) of La Reunion, in a group of travelers returning from a cruise [3]. Given the lack of available treatments to combat the infection during that period, the government implemented a comprehensive array of public health measures on Reunion Island, similar to those in metropolitan France, in an effort to control the spread of the virus. These measures, referred to as non-pharmaceutical interventions (NPIs), included a strict lockdown period and the closure of all schools and childcare centers on the island from March 17, 2020, to May 11, 2020. Subsequently, in August 2020, control measures were formally introduced on the island, encompassing social distancing and mandatory mask-wearing in public. Starting in July 2020, several SARS-CoV-2 screening campaigns were launched across all cities on the island, aiming to test all travelers entering and leaving the island, as well as all hospital admissions.

These NPIs were implemented both to reduce the transmission of the virus and to prevent the healthcare system from exceeding its capacity of intensive care beds [4]. Thanks to these measures, Reunion Island registered only 9037 cases and 42 deaths during the initial year of the COVID-19 pandemic, managing to remain relatively unscathed [5]. The vaccination campaign began on January 15, 2021, and in the subsequent months, a range of new NPIs, including lockdowns and curfews, were introduced. Nevertheless, the island encountered numerous successive waves of infection caused by different variants of SARS-CoV-2 (such as Beta, Delta, and Omicron variants) [6]. This led to over 82,796 reported cases in 2021 and a surge to more than 420,850 cases in 2022, according to regional public health authority statistics [6, 7]. The year 2023, on the other hand, was a year of low SARS-CoV-2 circulation on the island [8].

It has been reported that the combination of SARS-CoV-2 circulation and NPIs across the world has had a significant impact on the circulation of other respiratory viruses [9-14]. Notably, a global decline in influenza cases and a shift in the seasonal transmission pattern of respiratory syncytial virus (RSV) were evident worldwide. In mainland France, no influenza outbreaks were reported for the 2020–2021 season following the implementation of NPIs, while the RSV bronchiolitis outbreak in children was delayed by 3 months [15, 16].

Reunion Island represents an ideal observatory for monitoring the spread of SARS-CoV-2, its evolution in relation to NPIs, and its impact on the circulation of other respiratory viruses within a subtropical, closely-knit environment with the presence of dense urban areas. This French overseas region in the Southern Hemisphere benefits from a robust surveillance network in a geographic area with a critical lack of data. On this territory, the COVID-19 pandemic seems to have also impacted the circulation of respiratory viruses, as the epidemiology of severe community-acquired pneumonia has changed, with a significant decrease in the number of cases in 2020–2021 and no cases of pneumonia related to influenza viruses [17]. To gain a better understanding of how the COVID-19 pandemic has affected the circulation of influenza virus, RSV, and other respiratory viruses on Reunion Island, we compared the prevalence of these viruses before (2017–2019) and after the pandemic (2020–2023). This study relied on retrospective epidemiological data collected at Reunion UHC from tests conducted within the hospital environment and through the sentinel physician network, which monitors respiratory virus activity across the entire island.

Since March 2020, COVID-19 has dominated the landscape of respiratory viral infections worldwide. However, the burden of other respiratory viral diseases has become a matter of crucial concern as COVID-19 containment measures had been lifted [10-14]. It is now imperative to closely monitor the changes that have occurred in the prevalence of human viral respiratory diseases locally. In this comprehensive epidemiological study, we have made several observations that shed light on the dynamics of human respiratory viral infections after the emergence of the COVID-19 pandemic compared with the pre-COVID-19 period in Reunion Island. The retrospective data analyzed in this study were derived from samples processed by the virology laboratory of the UHC La Reunion, which may significantly represent the respiratory epidemic situation across the entire island, as compared with the number of consultations for ARI symptoms in the island's emergency departments and among a panel of general practitioners.

Among the major respiratory viruses, influenza viruses and RSV are recognized for exhibiting regular seasonal circulation on a global scale. On Reunion Island, influenza outbreaks typically occur during the austral winter season, spanning from June to November, with peak activity in October. In contrast, RSV epidemics occur during the austral summer season, with a peak observed in March. Epidemics caused by these viruses are typically characterized by the dominance of one subtype at a time.

Since the COVID-19 pandemic, the seasonality of these two viruses has been completely disrupted. In mainland France, the 2020–2021 winter season saw no influenza epidemic, while the RSV epidemic began 12 weeks later compared with previous seasons [20, 21].

On Reunion Island, there were no epidemics caused by these viruses in 2020 (Figures 2 and 3) as observed in many parts of the world [15, 22, 23]. These results suggest that the effects of the NPIs implemented to stop the circulation of SARS-CoV-2 may have also been very effective in stopping the circulation of the other respiratory viruses. In 2021, the circulation of influenza and RSVs resumed on our territory, but with a 10- to 12-week delay in the epidemic. A change in the seasonality of RSV was also observed in Germany, Australia, and South Africa, coinciding with the easing of the NPIs at the end of 2020 [13, 24, 25].

Since 2022, the influenza epidemic on Reunion Island has undergone significant changes, featuring two distinct outbreaks within the same year, each involving different subtypes.

Meanwhile, the 2021–2022 influenza epidemic in France appeared earlier and lasted 25 weeks [20]. Also, the RSV epidemic on Reunion Island began earlier and lasted longer than in the pre-COVID-19 years, a phenomenon similarly observed in mainland France from 2021–2022 [26].

On Reunion Island, in late 2021 and early 2022, the emergence of the Omicron variant coincided with a notable decline in cases of both influenza virus and RSV, although the number of tests performed remained stable during this period, suggesting a potential displacement effect where the SARS-CoV-2 wave superseded the other epidemics (Figures 2-4). As soon as the curfew was lifted in 2022, and the number of SARS-CoV-2 cases decreased, the epidemics of RSV and influenza virus resumed. While this finding may be incidental, a viral competition mechanism known as viral interference might have occurred between these respiratory viruses and SARS-CoV-2 [11, 27]. Evidence of such inhibitory interactions between respiratory viruses was described in France in 2009 during the influenza A(H1N1)pdm09 virus pandemic, where rhinoviruses were shown to delay the circulation of the pandemic influenza virus [28].

In addition to the significant change in the circulation periods, we also observed a notable modification in the proportion of positive cases caused by non-SARS-CoV-2 respiratory viruses. Regarding the influenza virus, even though multiple subtypes have circulated since late 2021, with numerous epidemic peaks, its prevalence and positivity rate have decreased compared with the pre-COVID-19 period on Reunion Island. Similar results were observed globally, with influenza virus activity remaining low during the 2021–2022 influenza season worldwide [29]. RSV, on the other hand, has been circulating significantly more in the territory since late 2021. This observation also applies to HRV, which lacks seasonality, as it became the predominant non-SARS-CoV-2 respiratory virus circulating in Reunion Island post-COVID-19, according to our results. The latter resumed its circulation immediately following the end of the 2020 lockdown and circulates throughout the year to a significant extent. Around the world, multiple studies have reported comparable results for HRV [11, 28].

One possible explanation is that NPIs might not have impacted the circulation of non-enveloped viruses, such as HRV, in the same way as they did with enveloped viruses like influenza A and SARS-CoV-2. Indeed, multiple studies indicate that the intrinsic characteristics of rhinoviruses and respiratory enteroviruses, including their lack of a viral envelope, resilience on surfaces, and the genomic diversity with multiple co-circulating strains, may facilitate rapid and sustained infections when NPIs were eased [30, 31]. Another possibility is that NPIs, which include measures like mask usage, improved hand hygiene, social distancing, and restricted travel, might not have significantly impacted infants and children. These groups were indeed the most affected by RSV and HRV locally when schools and daycare centers reopened in mid-2020. Finally, it is also possible that children and infants who were not immunized during the year 2020 due to lockdowns and NPIs were more susceptible to these infections in 2021–2022, thus contributing significantly to the spread of these viruses. It is indeed hypothesized that children who experience reduced early-life exposure to infectious agents may become more susceptible to diseases later in life [32, 33]. On a reassuring note, our findings from 2022 to 2023 suggest that on our island, the infection rates among different age groups for the three main viruses—influenza virus, RSV, and HRV—are returning to normal levels, as if the immune debt has been effectively erased.

Caution should be exercised when interpreting these results and assessing the direct role of COVID-19 and NPIs in the observed changes, as modifications in testing strategies, shifts in public behavior, and environmental or climatic factors may also have influenced the epidemiology of respiratory viruses following the COVID-19 pandemic on Reunion Island.

Furthermore, we chose to include all samples tested at the virology laboratory of the UHC of Reunion Island in this study, which may represent a limitation of our observational approach. These samples were primarily sourced from symptomatic patients within the hospital setting, with a smaller proportion collected by sentinel physicians or from private laboratories on the island. This sampling strategy could introduce bias into our results, as hospital patients may not be representative of the broader general population. Consequently, certain mildly symptomatic pathogens may be underrepresented in our analysis.

In conclusion, our findings suggest a notable impact of the COVID-19 pandemic and associated NPIs on the circulation of various respiratory viruses on Reunion Island. The influenza epidemic has occurred in several waves throughout the year, and the RSV epidemic now appears to start earlier and last longer. Additionally, our results underscore the increased circulation of viruses that were less prevalent than influenza in the pre-COVID period, such as PIVs, MPV, and HEV. This increase has occurred at the expense of influenza, the latter seeing a notable decline in circulation. Moreover, the infection rates among different age groups for the main respiratory viruses are returning to pre-COVID-19 levels. These alterations in the epidemiology of these viruses are probably multifactorial, influenced by a combination of factors, including the widespread implementation of several NPIs, reductions in travel, the immune debt, and possible SARS-CoV-2 viral interference. Modifications in testing priorities and surveillance systems due to the SARS-CoV-2 pandemic may have also influenced the statistics of respiratory viruses.

Describing these changes in our territory throughout the COVID-19 pandemic offers insights into the complex factors influencing the co-circulation of respiratory viruses in the community, which can inform the implementation of health measures for future epidemics or pandemics. This will allow health authorities to inform the medical community about these changes and adjust the influenza vaccination strategy and RSV immunization protocols for infants.

Nicolas M’nemosyme: writing–original draft, data curation, formal analysis, validation, investigation, visualization, software. Etienne Frumence: writing–original draft, writing–review and editing, formal analysis, validation, investigation. Laurent Souply: investigation, resources. Diana Heaugwane: investigation, resources. Nicolas Traversier: investigation, resources. Alizé Mercier: investigation, resources. Jamel Daoudi: investigation, resources. Jean-Sébastien Casalegno: writing–review and editing. Martine Valette: writing–review and editing. Marie-Pierre Moiton: investigation, resources. Rodolphe Manaquin: investigation, resources. Etienne Darieux: investigation, resources. Raphaëlle Sarton: investigation, resources. Anaïs Grimal: investigation, resources. Fabian Thouillot: investigation, resources. Xavier Deparis: investigation, resources. Bruno Lina: writing–review and editing. Marie-Christine Jaffar-Bandjee: writing–original draft, conceptualization, methodology, investigation, supervision, funding acquisition, project administration, resources.

Ethical approval was not required for this work as data were obtained within routine surveillance.

The authors declare no conflicts of interest.