Figure 1

Figure 1. Weekly SARS-CoV-2 sentinel surveillance in primary schools, kindergartens, and nurseries, Germany, June‒November 2020.Timeline of Münchner Virenwächter study in context of pandemic activity in Munich, Germany. The 7-day incidence rates were...

This study was approved by the ethics committee of the Ludwig-Maximilians University under project no. 20-484. We intended to accomplish a timely detection of infected cases and offer an additional level of safety to participating institutions during regular operating mode. The study spanned over 2 phases (Figure 1): phase 1, June 15‒July 26, 2020; and phase 2, September 7‒November 1, 2020. Participating institutions were randomly selected, and written informed consent was obtained in the first week of each phase. To correct for underrepresentation of younger children (<5 years of age), we included an additional nursery/kindergarten into phase 2.

We tested oropharyngeal swab specimens for SARS-CoV-2 by using real-time reverse transcription PCR (rRT-PCR); weekly samples were obtained from randomly selected children (n = 20) and staff (n = 5) in each institution. Swab specimens were taken on-site by trained medical personnel, and results were timely reported. For rRT-PCR, we processed specimens by using the AmpliCube Coronavirus SARS-CoV-2 Panel (Mikrogen, https://www.mikrogen.de) on a CFX96 Touch rRT-PCR Detection System (Bio-Rad, https://www.bio-rad.com). We retested single gene results by using the Xpert Xpress SARS-CoV-2 Test (Cepheid, https://www.cepheid.com).

We performed SARS-CoV-2 IgG screening at 3 sequential time points for samples from consenting staff members by using the Liaison SARS-CoV-2 S1/S2 IgG System (DiaSorin, https://www.diasorin.com) (Figure 1). We confirmed active results by using the RecomLine SARS-CoV-2 IgG Lineblot (Mikrogen). Antibody screening was complemented by obtaining a throat swab specimen at the same time to exclude active infection. Institutions were asked to respond to a questionnaire assessing implementation of infection control measures for phases 1 and 2.

Figure 2

Figure 2. Weekly SARS-CoV-2 sentinel surveillance in primary schools, kindergartens, and nurseries, Germany, June‒November 2020. Distribution of weekly pediatric oropharyngeal swab samples for SARS-CoV-2 testing by real-time reverse transcription PCR. Color code...

We processed 3,169 oropharyngeal swab specimens during the 12-week testing period, 2,149 from children (median age 7 years, range 1‒11 years, male:female ratio 1.03) and 1,020 from staff (median age 41 years, range 17–76 years, male:female ratio 0.13). We also obtained 493 swab specimens from staff during weekly testing and 527 swab specimens to complement serologic testing. We also tested 527 blood samples from staff for SARS-CoV-2 IgG. We determined pediatric sample distribution per study week (Figure 2).

No SARS-CoV-2 infections were detected during phase 1 of the study. During phase 2, only week 12 yielded 2 positive samples from 1 primary school. All SARS-CoV-2 IgG test results were negative at timepoints 1 and 2; only 1 positive serologic result was detected at timepoint 3. We identified some changes for implemented infection control measures between study phases and for individual facets between schools and childcare facilities (Table). All children attending primary schools were wearing face masks on school premises, except when seated for classes. Regular ventilation was begun as a daily routine in all institutions, as per national infection prevention and control guidance (Robert Koch Institute, https://www.rki.de).

Designed during the first lockdown in Munich, our study was intended to determine a feasible SARS-CoV-2 sentinel program in primary schools and childcare facilities in anticipation of a second pandemic wave and increasing incidence rates. Although public health and political authorities were concerned that childcare institutions would be major drivers of the pandemic, our results suggest that this did not happen. This result was consistent with those of another report suggesting that it was unlikely that children are major drivers of the pandemic even if attending schools (11).

Our study was not powered to accurately illustrate changes in incidences during low-incidence periods because of small sample sizes. However, we detected 2 cases in a primary school, 1 child and 1 teacher, during a high, local, 7-day incidence rate of 50 cases/100,000 children 1–11 years of age and 150 cases/100,000 persons in the general population. Tracing of 36 close contacts (33 classmates and 3 private contacts) identified only 1 additional case in another asymptomatic child in the same class. Telephone interview‒based contact tracing showed that the teacher reported to have experienced unspecific symptoms of headache and malaise 6 days before testing. Thus, it seems reasonable to deduce that transmission occurred from staff to both children.

Several reports have assessed the role of children in the dynamics of SARS-CoV-2 transmission. A study conducted in day care centers in Germany that used buccal mucosal and anal swabs for SARS-CoV-2 detection concluded that day care centers are not relevant reservoirs in a low prevalence setting (12). However, this study used self-testing, lacked oropharyngeal swab specimens, and was conducted during a minimal local incidence rate. Our study covered both low and high 7-day incidence periods while obtaining oropharyngeal swab specimens from children 1‒11 years of age. Ismail et al. reported complementary data from the United Kingdom, which showed that staff members had an increased risk for SARS-CoV-2 infection compared with students in any educational setting and that most cases linked to outbreaks were in staff (13).

Secondary attack rate analysis of the cases in our study also suggests that infections were transmitted from staff to children. In addition, low prevalence for SARS-CoV-2 antibodies in staff over the 3-month study period suggests no relevant infection activity in neither work nor private setting. Another recent report highlighted the need for maintaining low infection rates in the community to keep schools open during the pandemic (14).

Our study was conducted before the emergence of SARS-CoV-2 variants, such as B.1.1.7. Thus, the effect of this variant on children could not be addressed in our study. However, recent data from the United Kingdom found no evidence of more severe disease in children during the second wave, suggesting that infection with the B.1.1.7 variant does not result in a greatly different clinical course than the original strain (15).

We conclude that asymptomatic children attending primary schools, kindergartens, and nurseries are not greatly contributing to pandemic distribution of SARS-CoV-2 while adhering to infection control measures described above, even during high local background incidence. Thus, these children are unlikely to initiate clusters or outbreaks in the community when these institutions continue to play their critical role for the physical and emotional well-being of children and their families.