The unpredictable emergency of SARS-CoV-2 has posed a substantial threat to public health. Implementing efforts on aggregating the existing data about epidemiology, clinical, laboratory, and imaging characteristics to have a better understanding of the virus, its patterns of spread and the spectrum of illness is of critical significance. To our knowledge, this is the first meta-analysis to study presentations in children with COVID-19. Through a comprehensive searching, a total of 14 articles with 361 laboratory-confirmed cases were included.
The general proportion of male to female of this analysis (55.8% vs 44.2%) is similar to the gender distribution in an initial investigation9(57.5% vs 42.5% in 731 confirmed cases) and in general population (55.9% vs 44.1%) (Table 3). All the results seem to show that male have a slightly higher incidence than female in COVID-19.
According to our results, children got infected with SARS-CoV-2 mainly through family clustering, quite the same as SARS-CoV7,27 and MERS-CoV8.While compared to adults, children are more likely to be asymptomatic or present milder symptoms, this reminds us that, whenever there’s a family member caught with this virus, it is necessary to conduct a virologic screening test on the child as soon as possible. Otherwise, the infected child may become a threat to other vulnerable populations (e.g. elderly people or people with severe underlying disease), resulting in further extension of ongoing pandemic, as was seen during influenza outbreak.
Fever and cough are the most common symptoms in COVID-19 children, in our study, the pooled incidence of fever is 54.4%, which is lower than that in adults35, SARS26 and influenza38. (Table 3). Clearly, children with COVID-19 rarely had obvious signs and symptoms of upper respiratory tract (pharyngeal congestion, rhinorrhea, sore throat, stuffy nose). Through a comprehensive review, it’s easy to draw the conclusion that SARS-CoV-2 leads to a less aggressive clinical course in children with more asymptomatic and fewer symptoms, compared to that in adults and the other two pathogens. (Table 3).
In terms of laboratory abnormalities, interestingly, the frequency of lymphocytosis is higher than lymphopenia (25.8% vs 9.7%), which is quite different from findings in COVID-19 adults35and SARS26.Besides, leucopenia was found in 17.3% of patients, nevertheless, a research including 80 virologic-confirmed children cited by Henry32 reported 46% of leukopenia. Theoretically, virus particles primarily spread through the respiratory mucosa, initially using the angiotensin-converting enzyme 2(ACE2) receptor(the cell-entry receptor for SARS-CoV-2) at ciliated bronchial epithelial cells and infect other cells, induce a cytokine storm in the body, generate a series of immune responses, and cause changes in peripheral white blood cells and immune cells such as lymphocytes33,34,35.Presumptions have been made that children may be protected against SARS-CoV-2 because this enzyme is less mature at a younger age, since the immune system undergoes substantial changes from birth to adulthood. In general, WBC and lymphocyte remained normal in the majority of pediatric patients, suggesting that the newly emerging virus, SARS-CoV-2, may have a marginal influence on the immune function of children.
As for radiologic aspects, our research found that a proportion of 32.3% of 308 virologic positive cases were in absence of CT abnormalities, and Ground glass opacity, also typical signs of severe acute respiratory syndrome (SARS)26, was shown in only 28.2% of pediatric patients. In addition, 14 cases in our research underwent chest X-ray, four of them (28.6%) subsequently showed abnormal opacities in the lung. This kind of low sensitivity hints us that routinely radiologic scans should not be over emphasized for screening or early identification of COVID-19 in children in consideration of substantial radiation exposure, especially when the child is lack of symptoms or running a mild clinical course. Therefore, more strict strategies and screening practices are required for the better management of pediatric cases.
It’s worth noting that underlying conditions have played a pivotal role in COVID-19. Four of 361 cases included were diagnosed as critical type of novel coronavirus infected pneumonia and required mechanical ventilation, three of them were undergoing hydronephrosis, leukemia, and intussusception respectively. A 10-month-old infant with intussusception had multi-organ failure and died 4 weeks after admission21.Compared to adults35, the spread of SARS-CoV-2 yield a much better prognosis in pediatric patients, similar to SARS7and Middle East Respiratory Syndrome (MERS)8. 75.1% cases were discharged, the discharge rate ought to be higher actually since many children were still in hospital before the submission of the papers. The reasons why children experience a milder COVID-19 disease remain elusive. One possible explanation is that the response of children to SARS-CoV-2 is fundamentally different from that of adults, as demonstrated in earlier reports30, the frequency of lymphopenia found in adults suggests that SARS-CoV-2 might act on lymphocytes, which is rare in children. Prior exposure to other respiratory virus may exert an influence, making children’s immune systems more resilient26. Another potential theory is related to differences in the expression of ACE 2 receptor which was thought to be the cell-entry receptor for SARS-CoV-233,36.Besides, some researchers proposed that the mild disease in children may be associated with trained immunity, which refers to the use of certain vaccines such as Bacille de Calmette Guerin (BCG). BCG has been proved to provide nonspecific protection of mice against influenza virus infection probably by the induction of trained immunity39. In addition, children suffered from fewer comorbidities than adults. Accordingly, further studies in fields of immunology, anatomy and virology are required to ravel out this puzzle.
With massive public health interventions implemented actively and effectively, the spread of SARS-CoV-2 seems to have been under control in several countries. On 18 April 2020, there were only 16 newly confirmed cases across mainland China, nine of which were imported39. At present, while some countries are considering enhance control measures, China is planning to lift restrictions, work resumes and school starts are on agenda. Nevertheless, concerns have been proposed that a second wave of cases might occur in light of the absence of herd immunity against COVID-19, escalating case importation or residual infected seeds and resumption of economic activities.40,41It’s plausible to suggest whether children have to get away from school again to mitigate the revival transmission. School closures can affect the spread of virus during a pandemic through reducing transmission and new cases, while long periods of social distancing interventions in school may put students in a disadvantaged situation. Recently, some scholars are questioning the benefits brought by closing schools. On the one hand, school closures are based on empirical evidence and assumptions from influenza outbreaks, it’s hard to say such measures are also effective in coronavirus outbreaks like SARS, MERS and especially COVID-19, for which transmission dynamics appear to be different.42 A systemic review42 concluded that school closures in SARS did not contribute to the control of the epidemic and its effectiveness in COVID-19 would be less than other social distancing interventions, with only 2-4% of death prevention. Meanwhile, less comprehensive and deliberate plan can result in a completely converse consequence. Jude Bayham and Eli P Fenichel43 estimated that school closures could lead to mortality rate increased by 0.35% and a greater number of deaths than they prevent when the health-care workforce declines by 15.0% due to unintended childcare obligations. (Table 4 shows alternative closure strategies in five countries). On the other hand, prolonged school dismissals can be detrimental to children’s physical and mental health.44,45 Out of school means a totally altered lifestyle—for example，fewer physical activities, less interaction with peer groups and longer screen time. Besides, many schools are offering online courses, but this is not available to all, especially to children from low socioeconomic households, and they may be further disadvantaged by nutrition shortfalls. Moreover, with home confinement, communities lockdown and economic recession deepens, family conflicts are rising, children are more likely to be exposed to domestic violence and abuse. Consequently, it is imperative for the policy makers to weigh the benefits of school closure against its costs carefully and deliberately and provide alternative strategies to minimize the adverse impacts of the COVID-19 on children’s well-being.
There are several limitations need to be acknowledged. Firstly, the number of included cases is relatively small, a few researches have reported larger simple size of pediatric patients, whereas sufficient data are unavailable. Secondly, most of articles in this meta-analysis are descriptive and retrospective with a low quality, they highlighted different aspects of the illness, consequently, high heterogeneity was inevitable. Thirdly, all of the studies included are derived from China, many reports from countries outside China like US, Italy, where the pandemic is progressing with a sharp increasing number at present, provide individual cases or only crude epidemiological data, which didn’t meet our inclusion criteria. In addition, we have intended to conduct a subgroup analysis based on age stratification and severity of the disease, while enough information was unavailable. Therefore, the findings of this meta-analysis still need to be updated by more relevant studies with more strict design and larger sample size.