Educational Setting and SARS-CoV-2 Transmission Among Children With Multisystem Inflammatory Syndrome: A French National Surveillance System

Background: Multisystem inflammatory syndrome in children (MIS-C) is the most severe form associated with SARS-CoV-2 infection in children. To reduce the spread of SARS-CoV-2 at the population level, educational setting closure have been implemented in many countries. However, the direct benefit of school closure on the MIS-C burden remains to be explored. We aimed to assess the role of educational settings in SARS-CoV-2 transmission among children with MIS-C. Methods: We conducted a French national prospective surveillance of MIS-C, coordinated by Public Health France, from April 2020 to March 2021. During this period, we included all children with MIS-C fulfilling the WHO definition who were reported to Public Health France. For each child, we traced the source of SARS-CoV-2 transmission. The main outcome was the proportion of children with MIS-C, with educational setting-related SARS-CoV-2 infection, during the period of school opening. Results: We included 142 children fulfilling WHO criteria for MIS-C: 104 (70%) cases occurred during school opening periods. In total, 62/104 children (60%, 95%CI [50; 69]) had been contaminated by a household contact and 5/104 in educational settings (5%, 95%CI [2; 11]). Among children with MIS-C occurring during school closure periods, the proportion of household transmission remained similar (66%, 25/38). Conclusion: Children with MIS-C were mainly infected by SARS-CoV-2 within their family environment, and the educational setting played a marginal role in this transmission. This suggests that mitigating school attendance may not reduce substantially the burden of MIS-C.

Background: Multisystem inflammatory syndrome in children (MIS-C) is the most severe form associated with SARS-CoV-2 infection in children. To reduce the spread of SARS-CoV-2 at the population level, educational setting closure have been implemented in many countries. However, the direct benefit of school closure on the MIS-C burden remains to be explored. We aimed to assess the role of educational settings in SARS-CoV-2 transmission among children with MIS-C.
Methods: We conducted a French national prospective surveillance of MIS-C, coordinated by Public Health France, from April 2020 to March 2021. During this period, we included all children with MIS-C fulfilling the WHO definition who were reported to Public Health France. For each child, we traced the source of SARS-CoV-2 transmission. The main outcome was the proportion of children with MIS-C, with educational settingrelated SARS-CoV-2 infection, during the period of school opening.
Conclusion: Children with MIS-C were mainly infected by SARS-CoV-2 within their family environment, and the educational setting played a marginal role in this transmission. This suggests that mitigating school attendance may not reduce substantially the burden of MIS-C.

INTRODUCTION
Multisystem inflammatory syndrome in children (MIS-C) is a life-threatening emerging disease associated with SARS-CoV-2 infection. Cardiovascular complications are common, often requiring inotropic support (1). Thus, MIS-C is by far the most severe form associated with SARS-CoV-2 infection in children and the leading source of burden related to SARS-CoV-2 in this age group (1).
Several non-pharmaceutical interventions implemented during the COVID-19 pandemic included school closure in more than 150 countries (2). After simulation studies (3), several countries including the United Kingdom, Germany and Italy closed educational institutions to mitigate the new wave of COVID-19 in winter 2020-21, and France again in April 2021. This intervention is likely to be discussed again to fight further waves of epidemics.
However, these decisions raised many issues, given their devastating consequences on child well-being and development (4). To help evaluate the public health benefit of these measures over the direct negative consequences, the direct benefit of school closure on the burden of MIS-C, the most severe form of SARS-CoV-2 infection in children, needs to be estimated.
Here we assessed the role of educational setting in SARS-CoV-2 transmission among children with MIS-C.

Study Design, Patients, and Settings
We used a national prospective surveillance of MIS-C from April 1, 2020to March 30, 2021. In April 2020, the reporting of all suspected MIS-C cases became mandatory in France, coordinated by the French National Public Health Agency, with a methodology previously published (5,6). Each suspected MIS-C case was reported by clinicians to the French National Public Health Agency, and was then classified as confirmed MIS-C or not, following World Health Organization (WHO) criteria (6). All children fulfilling WHO criteria for MIS-C up to March 30, 2021 and for whom an electronic case report was completed were included in this study.

Outcome Measure and Definitions
The primary outcome was the proportion of children with MIS-C with educational setting-related SARS-CoV-2 infection, during the period of school opening.
Secondary outcomes were the proportion of other sources of SARS-CoV-2 infection according to school opening or closure.
During the school opening periods of this study, the following non-pharmaceutical interventions were established by the French authorities for educational settings: the use of masks was mandatory for staff and children older than 11, hand washing had to be carried out at a minimum, on arrival in the establishment, before each meal, after going to the toilet, in the evening before returning home. Physical distancing between children in class was recommended (at least one meter) but not compulsory when it was materially impossible. Children and staff had to stay at home when sick or after close contact with someone infected by SARS-CoV-2 (8). Classrooms were ventilated every 3 h for at least 15 min. They were cleaned and disinfected regularly, as are the materials. The restriction of mixing between groups of pupils was not mandatory but each school had to reorganize children's activities to limit the regroupings and crossings between groups.

SARS-CoV-2 Contact Tracing
The source of SARS-CoV-2 infection, i.e., index case, was defined by a symptomatic contact and/or a confirmed contact (positive SARS-CoV-2 RT-PCR result from a nasopharyngeal swab) during the previous 6 weeks. To define a symptomatic contact, we considered the following symptoms based on the literature (9): cough, ageusia, anosmia, fever, asthenia, influenza-like symptoms, shortness of breath, sore throat, rhinitis, and digestive disorders.
The French National Public Health Agency was involved in contact tracing strategies to identify contacts for children with acute SARS-CoV-2 infection or MIS-C. Furthermore, in France, during school opening periods, recommendations were published to perform SARS-CoV-2 RT-PCR from a nasopharyngeal swab for children with symptoms evoking SARS-COV-2 infection or after contact with cases of SARS-CoV-2 infection (details in Supplementary 1) (10).

Analysis
We estimated the proportion of the different sources of SARS-CoV-2 infection with their 95% confidence intervals (CIs) by using the adjusted Wald method.

Ethical Statement
The study was approved by the INSERM Ethics Committee for evaluation (IRB00003888). A written information form validated by the ethics committee was given to all participants. Oral consent was obtained from study participants or parents.

RESULTS
Among the 172 pediatric patients reported to the French National Public Health Agency with WHO criteria for MIS-C and a completed case report form, 30 children were not included because the source of infection was not investigated or had missing data; 142 children were included (Figure 1).
(Supplementary Table 1 The sensitivity analysis considering short holiday periods as additional exclusion periods provided similar results [102 children included in the opening school period, of whom 5 (5%) were infected at school, and 60 (59%) by a household contact].
The characteristics of the 5 cases of educational setting-related MIS-C are detailed in Box 1. Four contacts involved classmates; the fifth case was a teacher. All education settings were involved (from kindergarten to high school).

DISCUSSION
Children with MIS-C were mainly infected by SARS-CoV-2 within their family environment. Educational settings played a marginal role in the transmission.
Several studies suggested that the circulation of SARS-CoV-2 within educational settings may be low (11)(12)(13), when appropriate infection control measures are used (14,15), as the use of mask, social distancing, and room ventilation (16). This suggestion agrees with many reports showing that children are less infected by SARS-CoV-2 than other age groups (17) and that infected children seem less infectious than adults, thus leading to reduced risk of a transmission chain within schools (11,12,18).
Our results could be influenced by an underestimation of SARS-CoV-2 circulation in educational settings. Indeed, symptoms of SARS-CoV-2 infection in children are not specific and are commonly shared by many other respiratory pathogens, which could lead to underdiagnosis of SARS-CoV-2 infection. However, French guidelines recommend performing SARS-CoV-2 RT-PCR with a nasopharyngeal swab for any child attending school with evocative symptoms (10). Furthermore, staff members with any evocative symptoms needed to be tested for SARS-CoV-2 before returning to school, which thus reduced without fully eliminating the risk of underestimating SARS-CoV-2 transmission at school.
The viral inoculum may also play a role in our findings. In numerous common infectious diseases, such as influenza or varicella, the severity of the disease seems linked to the viral inoculum (19). For SARS-CoV-2, the same phenomenon is suspected (20). Because intra-familial contacts are more intense and prolonged than within school, a high viral inoculum may lead to increased risk of severe SARS-CoV-2 infection. Ongoing studies may help elucidate the MIS-C pathophysiology and this specific mechanism.
Several limitations should be discussed. First, the diagnosis of MIS-C relies on non-specific criteria, and misdiagnosis cannot be ruled out. However, all children included in our study fulfilled the WHO criteria for MIS-C. The main clinical characteristics of MIS-C cases included in our study were similar to those in the literature, including median age, rate of cardiac dysfunction, PICU care, and hemodynamic support, enhancing the generalizability of our findings (1). Second, interpreting the minor role of educational setting in SARS-CoV-2 transmission among children with MIS-C should take into account mitigation measures implemented in schools. Third, the source of SARS-CoV-2 infection was not identified in 37/104 cases (36%), which means that patients could have contracted the infection in any setting (family, school, or via social contacts outside home). Several points should be considered. National guidelines to perform SARS-CoV-2 PCR for suspected cases were produced in March 2020. These guidelines regarding SARS-CoV-2 PCR indication did not change thereafter but were gradually applied due to initial lack of availability of PCR testing. The general population was initially not aware about BOX 1 | Characteristics of education setting-related SARS-CoV-2 transmission among children with multisystem in ammatory syndrome (MIS-C) (N = 5). the broad clinical spectrum of SARS-CoV-2 infections, which may have led to under-diagnosis in the first months of the study period. Furthermore, we cannot exclude that transmission from asymptomatic children occurred in educational settings, thus leading to MIS-C with undocumented sources of infection. No systematic test strategies have been implemented in school at a national scale. However, in our study, the proportion of MIS-C with undocumented source of infection remained similar and stable over time in "school closure" and "school opening" periods (32 and 36%), suggesting that educational settings may not play an important role among MIS-C cases with undocumented source of infection.
In conclusion, educational settings seem rarely involved in SARS-CoV-2 transmission among children with MIS-C. Thus, mitigating school attendance does not likely have a direct effect on the burden of MIS-C. By contrast, prolonged school closure have multiple adverse social, educational, health, and economic impacts (21). It might have negative effects on children physical and mental health, with an increase of anxiety and depressive symptoms (22), and could also worsen an already severe food security crisis. Among the consequences in terms of education, there is loss of learning, demotivation, dropping out of school, and increased disparities in academic results, due to unequal access to educational materials. Finally, economic difficulties, an increase in social inequalities and a predicted lower salary for current school children are described in the literature (23). These data have led the French Pediatric Society to publish recommendations for the reopening of schools in September 2020 (10). Likewise, an editorial published in Science in September 2020 (24) and an article in the New England Journal of Medicine in September 2020 (25) advocate giving maximum priority to the opening of schools with regard to the benefitrisk balance that this measure represents. In this context, our findings may contribute to better weight the expected benefit of school closure against the massive damage they gender, and may be considered by policy-makers when assessing the public health benefit of this intervention.

DATA AVAILABILITY STATEMENT
The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

ETHICS STATEMENT
The studies involving human participants were reviewed and approved by INSERM Ethics Committee for evaluation, Paris, France (IRB00003888). Written informed consent to participate in this study was provided by the participants' legal guardian/next of kin.