Since the onset of coronavirus disease (COVID-19) pandemic, children have been less affected than adults in terms of severity1–3 and frequency, accounting for <2% of the cases.2–5 Unlike with other viral respiratory infections, children do not seem to be a major vector of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) transmission, with most pediatric cases described inside familial clusters6 and no documentation of child-to-child or child-to-adult transmission.7,8 The aim of this work was to describe the clinical presentation of the first 40 pediatric cases of COVID-19 in our city and the dynamics of their familial clusters.
Methods
From March 10 to April 10, 2020, all patients <16 years old with SARS-CoV-2 infection were identified by means of the Geneva University Hospital’s surveillance network (Switzerland). The network notifies the institution’s pediatric infectious diseases specialists about results of nasopharyngeal specimens tested for SARS-CoV-2 by reverse-transcription polymerase chain reaction. This study was approved by the Regional Ethics Committee. After informed oral parental consent and its documentation in the medical charts, chart reviews were used to retrieve clinical data, and parents were called for patients and household contacts (HHCs) follow-up. HHCs were considered suspect if they had fever or acute respiratory symptoms, as per the Swiss Federal Office for Public Health’s case definition9 (Supplemental Information).
Categorical data were compared using the χ2 test, with P values <.05 considered significant. Statistics were performed using SPSS version 23.0 (IBM SPSS Statistics, IBM Corporation).
Results
Among a total of 4310 patients with SARS-CoV-2, 40 were <16 years old (0.9%). One patient for which telephone follow-up was not possible was excluded because of the inability to evaluate clinical evolution and HHC symptoms. The median follow-up of the households was 18 days (interquartile range [IQR]: 14–28).
Clinical Presentation, Diagnosis, and Management
Demographics, clinical presentation, and diagnosis of the study children are detailed in Table 1. Of note, 29 (74%) patients were previously healthy; the most frequently reported comorbidities were asthma (10%), diabetes (8%), obesity (5%), premature birth (5%), and hypertension (3%). Seven patients (18%) were hospitalized to the ward, for a median duration of 3 days (IQR: 2–4); reasons for admission were surveillance for nonhypoxemic viral pneumonia (n = 2), fever without source (n = 2), apparent life-threatening event (n = 1), and sepsis-like event (n = 1); 1 paucisymptomatic child admitted because both parents had severe COVID-19 (n = 1). No patient required ICU admission or SARS-CoV-2–specific therapies. The others 32 patients were managed as outpatients. All patients had a complete resolution of symptoms by day 7 after diagnosis.
Demographics and Clinical Presentation of Study Patients
| n = 39 | |
|---|---|
| Demographics | |
| Median age, y (IQR) | 11.1 (5.7–14.5) |
| Female sex, No. (%) | 22 (56) |
| Clinical presentation | |
| Median time between symptom onset and diagnosis, d (IQR) | 2 (1–3) |
| Reported symptoms, No. (%) | |
| Cough | 32 (82) |
| Fever | 26 (67) |
| Nasal discharge | 25 (64) |
| Headache | 22 (56) |
| Sore throat | 14 (36) |
| Shortness of breath | 13 (33) |
| Myalgia | 13 (33) |
| Abdominal pain | 11 (28) |
| Anosmia | 8 (21) |
| Arthralgia | 7 (18) |
| Diarrhea | 7 (18) |
| Fatigue | 5 (13) |
| Rash | 5 (13) |
| Dysgueusia | 4 (10) |
| Nausea | 4 (10) |
| Vomiting | 3 (8) |
| Thoracic pain | 2 (5) |
| Conjunctivitis | 1 (3) |
| Diagnosis, No. (%) | |
| Upper respiratory tract infection | 27 (69) |
| Influenza-like illness | 2 (5) |
| Fever without source | 2 (5) |
| Pneumonia | 2 (5) |
| Obstructive bronchitis | 2 (5) |
| Sepsis-like event | 1 (3) |
| Croup | 1 (3) |
| ALTEs | 1 (3) |
| Asymptomatic | 1 (3) |
| n = 39 | |
|---|---|
| Demographics | |
| Median age, y (IQR) | 11.1 (5.7–14.5) |
| Female sex, No. (%) | 22 (56) |
| Clinical presentation | |
| Median time between symptom onset and diagnosis, d (IQR) | 2 (1–3) |
| Reported symptoms, No. (%) | |
| Cough | 32 (82) |
| Fever | 26 (67) |
| Nasal discharge | 25 (64) |
| Headache | 22 (56) |
| Sore throat | 14 (36) |
| Shortness of breath | 13 (33) |
| Myalgia | 13 (33) |
| Abdominal pain | 11 (28) |
| Anosmia | 8 (21) |
| Arthralgia | 7 (18) |
| Diarrhea | 7 (18) |
| Fatigue | 5 (13) |
| Rash | 5 (13) |
| Dysgueusia | 4 (10) |
| Nausea | 4 (10) |
| Vomiting | 3 (8) |
| Thoracic pain | 2 (5) |
| Conjunctivitis | 1 (3) |
| Diagnosis, No. (%) | |
| Upper respiratory tract infection | 27 (69) |
| Influenza-like illness | 2 (5) |
| Fever without source | 2 (5) |
| Pneumonia | 2 (5) |
| Obstructive bronchitis | 2 (5) |
| Sepsis-like event | 1 (3) |
| Croup | 1 (3) |
| ALTEs | 1 (3) |
| Asymptomatic | 1 (3) |
ALTE, apparent life-threatening event.
Familial Clusters
Familial cluster evaluation revealed a t number of 4 household members per family (IQR: 3–4). Among the 111 HHCs of study children, mothers predominated (n = 39), followed by fathers (n = 32), pediatric siblings (n = 23), adult siblings (n = 8), and grandparents (n = 7) (Fig 1). Adult HHCs were suspected or confirmed with COVID-19 before the study child in 79% (31/39) of cases. In only 8% (3/39) of households did the study child develop symptoms before any other HHC (Fig 1). Interestingly, 85% (75/88) of adult HHCs developed symptoms at some point, compared with 43% (10/23) of pediatric HHCs (P < .001). Also, 92% (36/39) of mothers developed symptoms, compared with 75% (24/32) of fathers (P = .04).
Description of individual household clusters with asymptomatic, suspected, and confirmed SARS-CoV-2 cases. Green, yellow, and red squares correspond to symptomatic HHCs who developed symptoms, respectively, before, simultaneously to, and after study patients. White squares correspond to asymptomatic HHCs. The “+” and “−” signs correspond to the results of SARS-CoV-2 nasopharyngeal reverse-transcription polymerase chain reaction (RT-PCR); patients without testing have an empty square. The absence of a family member inside a given household cluster is shown in gray. The study patient was the first to develop symptoms in cluster numbers 2, 3, and 13 only. Cycle threshold (CT) values correspond to the number of polymerase chain reaction cycles required to amplify the virus; hence the CT value is inversely proportional to the viral load.
Description of individual household clusters with asymptomatic, suspected, and confirmed SARS-CoV-2 cases. Green, yellow, and red squares correspond to symptomatic HHCs who developed symptoms, respectively, before, simultaneously to, and after study patients. White squares correspond to asymptomatic HHCs. The “+” and “−” signs correspond to the results of SARS-CoV-2 nasopharyngeal reverse-transcription polymerase chain reaction (RT-PCR); patients without testing have an empty square. The absence of a family member inside a given household cluster is shown in gray. The study patient was the first to develop symptoms in cluster numbers 2, 3, and 13 only. Cycle threshold (CT) values correspond to the number of polymerase chain reaction cycles required to amplify the virus; hence the CT value is inversely proportional to the viral load.
Discussion
Most children in our study had mild or atypical presentations: headache and nasal discharge were described in more than half of cases, and anosmia and abdominal symptoms were described in <20%, which is more frequent than previously described.2 Some of these symptoms might be underreported because younger children may not be able to describe them.
In 79% of households, ≥1 adult family member was suspected or confirmed for COVID-19 before symptom onset in the study child, confirming that children are infected mainly inside familial clusters.6 Surprisingly, in 33% of households, symptomatic HHCs tested negative despite belonging to a familial cluster with confirmed SARS-CoV-2 cases, suggesting an underreporting of cases. In only 8% of households did a child develop symptoms before any other HHC, which is in line with previous data in which it is shown that children are index cases in <10% of SARS-CoV-2 familial clusters10 ; however, with our study design, we cannot confirm that child-to-adult transmission occurred.
This study has some limitations. The study sample likely does not represent the total number of pediatric SARS-CoV-2 cases during this time period. Indeed, patients with milder or atypical presentation might not have sought medical attention. Moreover, the recall of symptom onset among HHCs might be inaccurate, although this seems for once less likely because of the confinement measures and anxiety in the community.
The results of this study are important because of the extensive HHC tracing and the almost absence of loss to follow-up. Extended diagnostic screening of suspected cases and thorough contact tracing are needed to better understand the dynamics of transmission within households.
Drs Posfay-Barbe, Wagner, and L’Huillier conceived and designed the study, designed the data collection instruments, conducted the initial analyses, drafted the initial manuscript, and reviewed and revised the manuscript. Drs Gauthey, Moussaoui, Loevy, and Diana critically reviewed the manuscript for important intellectual content and reviewed and revised the manuscript; and all the authors coordinated and supervised data collection and approved the final manuscript as submitted and agree to be accountable for all aspects of the work.
FUNDING: No external funding.
Comments
Unclear implications for schools and public health
To the Editor:
The ongoing COVID-19 pandemic has substantially impacted the lives of children, despite the lower diagnosed case burden and severity among children compared with adults. The role of children in transmission of SARS-CoV-2 has crucial implications for the control of the pandemic in both pediatric and adult populations, and must be balanced with the desire to support children’s intellectual, social, and emotional development in schools settings. I read with interest this report on the tracing of household contacts of pediatric patients diagnosed with SARS-CoV-2 infection (1). While it provides interesting observations regarding potential dynamics of household disease transmission between children and adults among the 39-case cohort, several crucial caveats should be made clear about the conclusions that can be drawn from this report.
First, the observation that children did not appear to be index cases within the household context is based on timing of symptoms rather than contact tracing outside of the household or timing of laboratory confirmation of infection. Incubation time of symptomatic COVID-19 remains poorly understood in children, thought the scant evidence that exists suggests that children may experience a longer incubation time in adults (2), potentially obviating the conclusion that transmission did not occur from child to adult in the majority of cases. Symptomatic disease may in fact be a misleading standard from a transmission standpoint. The relative rarity of diagnosed COVID-19 among children may not be evidence of rarity of infection, but rather of a significant rate of asymptomatic or paucisymptomatic disease, raising questions about children’s role in silent transmission.
Second, during a pandemic in which social-distancing measures have been put in place, including school-closure, the likelihood that children will represent index cases within a household would be expected to be decreased compared to a more typical situation in which children have considerable contact with others in a school setting. The observation that children did not appear to be the index cases in this study may therefore be a result of the atypical situation in which it was made.
Third, the biologic plausibility that infected children are substantially less likely to infect others remains unsupported. Since the publication of this work, data from a larger cohort has shown that household transmission from index case children does occur, and while there is a trend toward lower transmission from children in the first decade of life, the transmission rate is similar to the general population from index case children over 10 years old (3). Looking to evidence from other viral diseases spread by respiratory droplets, such as seasonal influenza, children and adolescents represent crucial demographics in community transmission (4,5).
The data presented by the authors raises important questions regarding the role of children in the transmission of SARS-CoV-2. However, it is not a burden of evidence sufficient for crucial public health recommendations - a limitation that deserves to be made explicit. Unfortunately, these extrapolations are being made, and their impact may be detrimental as parents, schools, and policy makers looks to the medical community for guidance in an uncertain time.
References:
1. Posfay-Barbe KM, Wagner N, Gauthey M, Moussaoui D, Loevy N, Diana A, et al. COVID-19 in Children and the Dynamics of Infection in Families. Pediatrics. 2020;e20201576.
2. Cai J, Xu J, Lin D, Yang Z, Xu L, Qu Z, et al. A Case Series of children with 2019 novel coronavirus infection: clinical and epidemiological features. Clin Infect Dis. 2020;ciaa198:1–5.
3. Park YJ, Choe YJ, Park O, Park SY, Kim Y-M, Kim J, et al. Contact Tracing during Coronavirus Disease Outbreak, South Korea, 2020. Emerg Infect Dis J. 2020;26(10). Available from: https://wwwnc.cdc.gov/eid/article/26/10/20-1315_article
4. Worby CJ, Chaves SS, Wallinga J, Lipsitch M, Finelli L, Goldstein E. On the relative role of different age groups in influenza epidemics. Epidemics. 2015;13:10–6. Available from: http://dx.doi.org/10.1016/j.epidem.2015.04.003
5. Ferguson NM, Cummings DAT, Fraser C, Cajka JC, Cooley PC, Burke DS. Strategies for mitigating an influenza pandemic. Nature. 2006;442(7101):448–52.
The importance of the epidemiological situation for safe school opening
Dear Editor,
Our data (ref 1) alongside others show that children are mostly infected within households by adults are to be interpreted cautiously given the fact that they were collected in an artificial setting where daycare and schools were closed. Even though children do not seem to be major drivers of transmission, there is no biological reason explaining why they could not be contagious under certain circumstances given the fact that their shedding pattern of live virus is similar to what has been shown in adults (ref 2 & 3).
The major determinant about when to safely reopen schools is the level of virus circulation in the community. Most European countries reopened schools and daycare when there was almost no circulation of SARS-CoV-2 in the community. Consequently, there was no significant increase in pediatric – and adult – COVID-19 cases.
Therefore, the current available evidence is not sufficient to conclude that it is safe to open schools in areas where there is still sustained viral circulation in the community. The epidemiological situation must be taken into account and data following school openings in Europe can certainly not be applied to areas with sustained SARS-CoV-2 circulation.
References
1.Posfay-Barbe KM, Wagner N, Gauthey M, et al. COVID-19 in Children and the Dynamics of Infection in Families. Pediatrics July 2020, e20201576; DOI: https://doi.org/10.1542/peds.2020-1576
2. L'Huillier AG, Torriani G, Pigny F, Kaiser L, Eckerle I. Culture-Competent SARS-CoV-2 in Nasopharynx of Symptomatic Neonates, Children, and Adolescents. Emerg Infect Dis 2020 Jun 30;26(10). doi: 10.3201/eid2610.202403
3. Wolfel R, Corman Vm, Guggemos W, et al. Virological assessment of hospitalized patients with COVID-2019. Nature 2020 May;581(7809):465-469. doi: 10.1038/s41586-020-2196-x.
Results are not generalizable to children attending school
To the Editor:
Dr. Posfay-Barbe and colleagues provide a thoughtful analysis of pediatric patients (n=39) who sought medical care and tested positive for SARS-CoV-2 in Geneva, Switzerland during March 10-April 10, 2020, and provide evidence that children were infrequently the index case within their households during this period.
While these data are being used to provide reassurance regarding school re-openings (ref 1), the risk of onward household transmission among children attending school cannot be extrapolated from these data. Public schools in Geneva closed on March 13, 2020 (ref 2) - three days into the one-month study period - and did not reopen until May 11 (ref 3). Therefore, the children in this study would be expected to be at lower risk of acquiring SARS-CoV-2 outside the household than children attending school.
We need data on the risks of acquiring and transmitting this virus among children who are attending school in a community with ongoing transmission. While informative, this article does not provide such data.
Sincerely,
Alison S. Rustagi, MD PhD
University of California-San Francisco
References
1. Lee B, Raszka WV. COVID-19 Transmission and Children: The Child Is Not to Blame. Pediatrics 2020;e2020004879; DOI:10.1542/peds.2020-004879.
2. Leybold-Johnson, I. and Boehler, P. Switzerland closes its schools to slow virus spread. swissinfo.ch. March 13, 2020.
3. Dahinten, J. and Kehnscherper, L. Switzerland Unveils Plan to Reopen Schools and Businesses. Bloomberg News. April 16, 2020.
RE:
The authors try to shed light on the urgent question whether children pose a lower risk in transmitting SARS-2-CoV infection into the family unit.
Of 4,310 SARS-2-CoV positive cases identified in the period March 10 - April 10 through the Geneva University Hospital Surveillance Network, 40 were younger than 16 years.
After analysing the timing of positive tests within households, the authors conclude “In 79% of households, at least one adult family member was suspected or confirmed for COVID-19 prior to symptom onset in the study child, confirming that children are infected mainly inside familial clusters."
Switzerland closed all schools from March 16 - six days into a 30 day data acquisition period - and has only recently begun to reopen certain school types, with a plan to have all students back at school from June. (1)
Children's largest exposure to others outside the family unit usually occurs at school. The sample for this paper was obtained during a period when children had significantly reduced exposure opportunities outside the family unit than they would have had when attending school. Accordingly, the authors' conclusion that children are "infected mainly within the family unit" can not be derived from this dataset and no policy decisions should be based upon it.
(1) https://www.coronavirus.bs.ch/schulen