Since December 2019, the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) which causes coronavirus disease 2019 (COVID-19) emerged as a virus causing substantial morbidity and mortality worldwide. There have been over 10 million confirmed cases and over 500,000 deaths worldwide [1
]. The transmission of COVID-19 has been shown to occur during pre-symptomatic or asymptomatic periods which contribute to widespread transmission of the infection [2
The reported number of cases worldwide implicates that elderly persons are especially vulnerable to the disease, contracting from children [3
]. In addition, most children with COVID-19 have been reported to present with mild symptoms [5
]. However, many studies suggested that children, despite mild or no symptoms, could be a source of viral transmission [6
]. It has been reported that the viral load in asymptomatic patients was similar to that in symptomatic patients, especially during the early phase of infection [8
There have been some systematic reviews and meta-analyses on COVID-19 aimed at characterizing SARS-CoV-2 infections in pediatric age groups [10
]. However, no study, to date, has focused on the difference between symptomatic and asymptomatic pediatric patients. Previous studies were mainly based on information from adult populations and limited data were available for children with COVID-19 [5
]. Therefore, we conducted a systematic review with the aim to investigate clinical, laboratory, and radiographic characteristics and differences among symptomatic and asymptomatic COVID-19 confirmed children.
The primary goal of this study was to systematically evaluate and characterize reported pediatric COVID-19 cases. An age threshold of 10 years was selected to distinguish between findings before and after the onset of puberty, which can vary with ethnicity and gender but generally begins as early as age 10. Analyses were stratified by clinical characteristics, laboratory findings, and radiological signs in children with COVID-19 based on this age threshold of 10 years and the presence or absence of symptoms. For asymptomatic patients, the continued absence of symptoms was also evaluated when reported at follow-up.
2.2. Search Strategy
The following databases were utilized: Embase, Medline, PubMed, and Web of Science, using the search terms: “COVID-19”, “SARS-CoV-2”, “novel coronavirus 2019”, “pediatrics”, “child*”, “infant*”, “neonate*”, and “adolescent*”. Articles were evaluated independently by four different researchers (D. Kim, H. Im, W. Rha, and E. Kim). The protocol followed the recommendations set forth by the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement [13
2.3. Eligible Criteria
We included peer-reviewed articles that reported confirmed SARS-CoV-2 infections by reverse transcription polymerase chain reaction. Eligible study designs included case reports and case series. Comments/letters and reviews were excluded. Articles not published in English were also excluded. Articles with duplicate cases were excluded.
2.4. Study Selection and Assessment of Risk of Bias
Initial results were reviewed by screening for duplicates, and then were screened by title and format for eligibility criteria. After full-text screening, a targeted search was conducted to generate additional articles focusing on asymptomatic children with COVID-19, using the same method used for the initial search.
2.5. Data Extraction and Statistical Analysis
Three independent investigators extracted data from each eligible study using a standardized data extraction form. Data were cross-checked for duplicate articles, duplicate cases, or discrepancies. Disagreements between reviewers regarding the data were resolved with reviews from three other independent investigators (E. Kim, S. Cha, and J. Yang). Statistical analyses were performed using SPSS version 26.0. Power was calculated using G*Power version 220.127.116.11. Percentages and means ± standard deviation were used to summarize categorical and continuous variables, respectively. The Kolmogorov–Smirnov test was used to test the normal distribution and the Mann–Whitney U test was used for non-normal distribution variables. The Chi-square test was performed to evaluate differences between age groups and symptomatic/asymptomatic groups, while the number in each cell was below 5, Fisher’s exact test was performed, and the student’s T test was used to evaluate differences in age.
With the rapid and continued propagation of SARS-CoV-2 infection, it is critical that its mode of transmission and the clinical characteristics of infection are clearly delineated. The pediatric population has recently been examined by numerous systematic reviews [5
], both to establish clinical characteristics of pediatric infection and to evaluate the extent that children may act as asymptomatic or mildly symptomatic carriers of the disease. Although it is suspected that mildly symptomatic pediatric cases can become rapid transmitters of SARS-CoV-2 during the incubation period [55
], we were unable to find studies identifying predisposing risk factors or clinical characteristics distinguishing symptomatic from asymptomatic patients. Furthermore, although many case studies were comprised of either infants or adolescents, no study, to the best of our knowledge, separated COVID-19 pediatric clinical features into younger and older age categories. Therefore, our study serves to (1) identify distinguishing features of symptomatic from asymptomatic infection and (2) characterize COVID-19 in young (<10 years) and older (10–17 years) pediatric populations. Forty-three articles published up to 4 September 2020 were aggregated. Twenty-two were conducted in China, five in Europe, two in the USA, and one each in Malaysia, the Republic of Korea, Vietnam, and Singapore. Twenty-three out of 24 asymptomatic patients remained asymptomatic throughout follow-up, decreasing the possibility that symptomatic patients experiencing an incubation period were falsely included in the asymptomatic group.
This systematic review reveals that symptomatic infection more frequently presents with abnormal laboratory characteristics indicative of infection. Low creatinine levels and high CRP, indicators of abnormal renal function and inflammatory response, respectively, were present at higher rates in symptomatic cases. Diagnostic markers more accurately determined the onset of symptoms in the population below 10 years of age. Within this group, abnormal laboratory characteristics, especially high lymphocyte levels, were associated with symptomatic infections, and abnormal radiological findings, low WBC count, low neutrophil count, and low creatinine levels neared statistical significance. This finding builds on works by Tian et al. [56
], and Shi et al. [57
] which linked serum creatinine and CRP to COVID-19 mortality, suggesting these markers also coincide with symptoms in pediatric patients. Interestingly, male sex was associated with asymptomatic infection. Previous studies in adults identified male sex as being associated with more severe infection, and a female predominance in asymptomatic infections [58
]. Dong et al. extensively examined a pediatric population and found that males and females were equally likely to be infected, but they did not specify the effect of sex on the manifestation of symptoms [61
]. Our data suggest that, unlike in adults, male sex in children may be associated with the absence of symptoms.
The characteristics of COVID-19 did not differ greatly between patients <10 years (including neonates and infants) and ≥10 years. Elevated serum creatinine levels were present at higher rates in older patients with symptoms. It is worth noting that typical GGOs were present at higher frequencies within asymptomatic older patients (83.3%) and were only present in 20.8% of younger asymptomatic patients. Liu et al. evaluated CT imaging as a diagnostic tool in children when nucleic acid tests were unavailable and found less lung involvement in children. However, these results indicate that chest CT may be more specific within an older pediatric population, which warrants further study.
Our data confirms previous findings that demonstrated milder infection in the pediatric population. No deaths were reported, and most cases recovered fully. Asymptomatic patients rarely received treatment, and symptomatic patients typically recovered well with antibiotics, antivirals, and oxygen therapy. For both asymptomatic and symptomatic populations, ICU admission occurred at similar rates. Only one ARDS was observed, which was consistent with previous research that complication rate of children was <2% as compared to 5% in adults [11
]. The most common symptoms, when present, were fever, cough, and vomiting. Cases of neonatal COVID-19 were reported in 15 cases across six studies [14
], and prognoses and symptoms were similar to other age groups, with six neonatal infections being asymptomatic.
There is still not enough information about the clinical features, laboratory, and radiologic findings of children, infants, and neonates COVID-19 infections. Fortunately, the pediatric population is deemed to have milder symptoms and better prognosis than adults. Thus, clinicians should be alert to asymptomatic SARS-CoV-2 infected children. Children cannot stay at home forever, and they need more definite preventive and treatment measures than just hand hygiene and supportive care. Therefore, continuous updates on the susceptibility and mechanisms on childhood COVID-19 infection are important.
Our study has several limitations. First, blood laboratory samples and CT were taken at different time points of infection for different patients. It is possible that as disease progressed from the contraction of illness to the onset of symptoms, laboratory characteristics and radiological features could have changed among cases. Second, included articles were not prospective, which would have helped in identifying risk factors predisposing certain children to symptomatic infection. Third, most articles originated from China, as studies from other countries with regards to children were rare. Furthermore, we found low numbers of cases within some categories of age and symptoms. For example, the subset of symptomatic patients ≥10 years with reported GGOs included only three cases, which could have interfered with statistical reliability. Fourth, multisystem inflammatory syndrome, a severe pediatric condition associated with COVID-19, was not specified during data extraction. Although the condition is rare, with a total of 570 reported to the CDC as of July 2020, it is possible that signs and symptoms of multisystem inflammatory syndrome in one or more samples included in our analysis could have altered our results [62
]. Fifth, although this study aggregates evidence sufficiently to achieve a power of 0.75–0.85 for comparisons of asymptomatic and symptomatic or younger and older children, some subgroup comparisons performed using Fisher’s exact test were unable to achieve this level of power, ranging from as low as 0.15 for GGO comparisons in asymptomatic vs. symptomatic children below 10 years to as high as 0.90 for GGO comparisons in above vs. below age 10 in asymptomatic children. Finally, we were unable to perform contact-tracing on our cases to evaluate the transmissibility of the symptomatic and asymptomatic cases. However, the strength of this systematic review is that it is the first, to our knowledge, to aggregate evidence of symptomatic and asymptomatic SARS-CoV-2 infection in children of varying ages.