Changes in Reverse Transcription Polymerase Chain Reaction–positive Severe Acute Respiratory Syndrome Coronavirus 2 Rates in Adults and Children According to the Epidemic Stages

Abstract

From March 2, 2020, to April 26, 2020, 52,588 reverse transcription polymerase chain reaction (RT-PCR) tests for severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) were performed in France, 6490 in children and 46,098 in adults. The rate of RT-PCR–positive SARS-CoV-2 tests for children (5.9%) was always less than that for adults (20.3%) but vary according to the epidemic stage. The risk ratio of RT-PCR–positive SARS-CoV-2 tests for adults compared with children was 3.5 (95% confidence interval: 3.2–3.9) for the whole study period.

France has been markedly affected by coronavirus disease 2019 (COVID-19), with more than 83,000 hospitalized cases at the epidemic peak.¹ The mortality and morbidity of this virus is highly variable among age groups.^2,3 In France and in other countries, the number of confirmed pediatric cases is relatively low, and they account for less than 1% of hospitalized cases and deaths.^3–6 In France, the strategy of closing schools and the lockdown started on March 17, 2020, but little information is available about the transmission between children and adults.^4,7 We aimed to describe the trends of reverse transcription polymerase chain reaction (RT-PCR)–positive severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) rates in children and adults when compared with the profile of the national epidemic curve of new COVID-19 cases in France.⁶

METHODS

With the Association Clinique et Thérapeutique Infantile du Val de Marne and Groupe de Pathologie Infectieuse Pédiatrique network research units, we conducted a French prospective multicenter study involving 45 hospitals, pediatric wards, emergency units and virology laboratories which performed RT-PCR analysis for SARS-CoV-2 from March 2, 2020, to April 26, 2020. The strategy of closing schools and the lockdown decided by the French government for the whole country started on March 17 and finished on May 11, 2020. Here, we collected aggregate data of RT-PCR–positive SARS-CoV-2 rates from each hospital: the proportion of children with positive results of children tested and the proportion of adults with positive results of adults tested. We defined different age groups: preschool age, <5 years; school age, 5 to <10 years; preadolescent, 10 to <15 years; adolescent (young/middle age adults), 15 to <60 years; and older adults, 60+ years. During the study period, the patients visiting emergency departments or who were hospitalized were selected for RT-PCR SARS-CoV-2 testing if they had severe disease (adults and children), and/or they were contact with a confirmed COVID-19 case and/or they were healthcare workers (adults) with symptoms. Twice a week, each clinical investigator from each participating ward was contacted to obtain information on SARS-CoV-2–positive tests.

Risk ratio and 95% confidence intervals (95% CIs) of RT-PCR–positive SARS-CoV-2 tests for adults compared with children were calculated using Stata SE v13.1 (Statacorp, College Station, TX).

RESULTS

During the study period, 52,588 RT-PCR tests for SARS-CoV-2 were performed, 6490 in children (12.3%) and 46,098 in adults (87.7%). The cumulative rate of positive tests for children was 5.9% (n = 382), 3.5-fold less than that for adults, 20.3% (n = 9346). For 59% of children (3845/6490), age data were available, and the cumulative positive tests were 5.4% for preschool age, 7.1% for school age and preadolescent and 7.6% for adolescent (P = 0.042). For 70% of adults (32,375/46,098), age data were available, and the cumulative positive tests were 18.1% for young/middle age adults and 23.5% for older adults (P < 0.001). Among the tested children, the rates of RT-PCR–positive SARS-CoV-2 were always lesser than that of adults whatever the age of children.

Figure 1 shows the trends of positive testing in children and adults in France and in the Paris area, one of the most affected regions, as well as the overall national trend in new COVID-19 cases.

FIGURE 1.: Evolution of RT-PCR–positive SARS-CoV-2 rates in France and in Paris area in children and adults when compared with new COVID-19 cases reported by the National Health Institute.⁶

In France, from the beginning of the epidemic until March 15, only 3.1% (n = 53) of 1690 pediatric samples were positive, 4.5-fold less than for adults (13.8%, n = 1124 of 8155 adult samples). At the peak of the national outbreak, on March 30, 9.7% (n = 85) of 877 pediatric samples were positive, 2.8-fold lower than for adults, 27.2% (n = 2054 of 7557 adult samples). A rapid decrease was observed during the following weeks, with the lowest rate reported the week of April 20, 3.4% (n = 33) of 960 pediatric samples were positive, 2.2-fold lower than for adults, 7.6% (n = 514 of 6791 adult samples).

In Paris area, the same trends were observed with marked differences between adults and children.

The risk ratio of RT-PCR–positive SARS-CoV-2 tests for adults compared with children was 3.5 (95% CI: 3.2–3.9) for the entire period of the study. Figure, Supplemental Digital Content 1, https://links.lww.com/INF/E86, shows the evolution of this risk ratio in France and in the Paris area week per week during the study period. For Paris area, at the beginning of the epidemic, the risk ratio was 7.1 (95% CI: 4.3–11.7), whereas it ranged from 3 to 4.9 during the following weeks.

DISCUSSION

Our results showed that, in a large cohort (n = 52,588), the risk ratio of RT-PCR–positive SARS-CoV-2 tests for adults compared with children was 3.5 for the entire period of the study. However, the dynamics of the curve for children followed that for adults and the National curve for new COVID-19.¹ The difference in rates between children and adults persisted during the surveillance period but varied according to the time in the epidemic, the rate of positivity and the region. In the Paris area, the most affected region, we observed a high spread of the disease in children, reaching more than 14.3% positive tests at the epidemic peak. At the beginning of the epidemic, the risk ratio was 7.1 (95% CI: 4.3–11.7), whereas it ranged from 3 to 4.9 during the following weeks.

As expected, the major impact of the lockdown was observed about 15 days after it began, with a rapid decrease in SARS-CoV-2–positive rate. Of note, the more the rate in adults increased, the less the difference between the 2 populations (adults and children), which supports that the main way of transmission was from adults to children.^4,5,7 Indeed, in this study, as in many others, the rate of RT-PCR–positive SARS-CoV-2 was significantly lower for children than for adults.^5,8 It is worthy to note that we observed higher rates of RT-PCR–positive SARS-CoV-2 for older children than for younger children (7.6% vs. 5.4%). However, the rates were always significantly lower in children than in adults. Even if the viral load detected by PCR is comparable for children and adults, this fact added to the low rates of secondary cases at schools argue simultaneously for a modest role of children in the dynamic of the COVID-19 pandemic and the reopening schools.^4,7,9,10

The slight increase of the risk ratio of RT-PCR–positive SARS-CoV-2 tests for adults compared with children just before the end of the epidemic is possibly explained by the occurrence of Kawasaki-like syndromes or hyperinflammatory shock in children 2–4 weeks after the peak epidemic.^11,12

Our study has several limitations. Our rates of RT-PCR–positive SARS-CoV-2 must take into account that RT-PCR practices were heterogeneous and could have evolved depending on the guidelines and the availability of the tests for adults and children. Furthermore, we were not able to describe precisely the reasons for why the tests were taken—disease requiring hospitalization versus contact of a confirmed COVID-19 case with symptoms. This stratification by reason for test could be important because children and adults were unevenly represented in these groups. However, during the study period, RT-PCR SARS-CoV-2 testing were performed if they had severe disease (adults and children), and/or they were contact with a confirmed COVID-19 case. Even if we cannot rule out that some cases could have escaped our surveillance in our centers and that we did not survey the whole country, our results are consistent with the dynamic of the COVID-19 pandemic in France.⁶ Finally, the viral load was not available in our study for children and adults.

We think that the surveillance of RT-PCR–positive tests in adults and children could be a simple and reliable tool to survey the epidemiology of SARS-CoV-2 infection, allowing to quickly detect any re-emergence of the disease.

ACKNOWLEDGMENTS

We thank all pediatricians and microbiologists who participated in the study: Belgaid A, Belivier E, Bueno B, Brehin C, Canarelli J, Castain L, Caurier B, Chevret L, Claris O, Cohen L, Colonna de Cinarca V, De Barbentane MC, De Pontual L, De Rougemont A, Dolfi-Fiette H, Dutron S, Faye A, Fernandez F, Flatres C, Foulongne V, Gajdos V, Garraffo A, Grosjean J, Guillotel E, Hau I, Jarlier V, Jeziorski E, Kuentz M, Labarthe F, Laisney N, Landraud L, Larrat S, Lemee V, Leruez M, Le Stradic C, Lina B, Loeile C, Mandelcwajg A, Marques C, Minodier P, Morand A, Morvan O, Ouldali N, Pantalone L, Peigne C, Pierre MH, Pinquier D, Raoult D, Rey A, Teissier R, Thach C, Varon E, Vignaud O.