Introduction

In early January 2020, a novel type of Coronavirus (CoV) was identified in the bronchoalveolar lavage sample of a subject affected by pneumonia of unknown origin [1]. The virus was provisionally named novel coronavirus (2019-nCoV) [2] to differentiate it from the Severe Acute Respiratory Syndrome Coronavirus (SARS-CoV) [3] and the Middle East Respiratory Syndrome Coronavirus (MERS-CoV) [4], responsible for two previous outbreaks, in 2002 and 2012, respectively [5]. Successively, the International Committee on Taxonomy of Viruses defined it as SARS-CoV-2 [6] and the associated disease has been called 2019 Coronavirus Disease (COVID-19). SARS-CoV-2 rapidly spread worldwide, forcing the World Health Organization (WHO) to declare the outbreak as a pandemic on 11 March [7, 8].

As of 1 May, 3,519,901 cases have been reported in 187 countries in all continents except for Antarctica, with 247,630 deaths [9]. Children seem to be less affected than adults, but data regarding epidemiologic characteristics and clinical features of COVID-19 in pediatric age are very poor and essentially based on limited case series [10, 11]. In a report of 72,314 cases from Chinese Center for Disease Control and Prevention (CDC), about 2% of all patients were aged < 19 years, but no specific clinical information was available [12].

Italy was among the first countries in the world to be hit by the COVID-19 outbreak, with 1.2% of all patients represented by children [9, 13, 14]. The estimated overall case-fatality rate in Italy resulted higher than in China (7.2% vs. 2.3%) [15], but no death in the pediatric age has been reported, confirming that the mortality remains low and no specific risk factor could be identified [16].

Neonatal SARS-CoV-2 infections are also extremely rare and, to date, there is no evidence of intrauterine infection caused by vertical transmission [17, 18]. As described in a case report and a case series, amniotic fluid, cord blood, neonatal throat swab, and colostrum samples collected from infected mothers were negative for COVID-19 [19, 20]. However, the question remains controversial, as IgM antibodies have been detected in newborns from mothers with COVID-19 [21], even though the probability of a false positivity should be taken in account. There is also growing evidence of cases of neonatal pneumonia which may be explained by SARS-CoV-2 infection [22, 23]. Currently in China, all newborns are separated from their infected mothers for at least 14 days [24], while the US CDC advise to consider a temporary separation between the infected mother and her infant on a case-by-case basis, using shared decision-making between the patient and the clinical team [25].

Despite the global interest and concern about COVID-19, clinical pattern is still unclear for the pediatric health community. The aim of our review is to provide a concise and systematic overview of the available evidence on clinical, laboratory, and radiological findings in children with SARS-CoV-2 infection.

Materials and methods

This study is in compliance with the Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) guidelines (Supplementary material S1) [26].

An electronic search was conducted on studies published from 1 January 2020 to 1 May 2020 in PubMed database. We used the search terms “2019 novel coronavirus OR COVID-19 OR SARS-CoV-2 AND child* OR pediatric* OR newborn OR infant” with no language restriction to include as much data as possible. However, since full texts available only in Chinese language could not be evaluated, we had to rely on English-language abstracts. For the purpose of this review, studies on children aged 0 to 18 years were included. Case reports, case series, and retrospective or observational studies were all considered eligible.

Articles were first screened by title and abstract: duplicates and those with no available English summary were excluded. Eligible full texts were then assessed for pediatric clinical, laboratory, and radiological data. Papers reporting information on both children and adults were included only if pediatric data could be retrieved. To identify missing studies, we also checked the reference list for each selected paper. Included studies were also assessed for methodological quality according to the Joanna Briggs Institute (JBI) Critical Appraisal Tools (Supplementary material S2) [27].

Three reviewers extracted data independently. A standardized table with the following information was used for data extraction: first author, date and journal of publication, study design (cohort study, case series, case report), sample size, age, mortality and morbidity rate, clinical features, laboratory and radiological results, and treatment information. All included studies were differentiated in tables for newborns and for children aged > 1 month.

During the data analysis process, clinical patterns were grouped according to the pediatric scoring for patients with COVID-19 (recommendations issued by the pediatric branch of the Chinese Medical Association [28]). In particular, cases described as “upper respiratory tract infection” (i.e., pharyngeal congestion, sore throat, and fever) with no abnormal radiographic and septic presentation were included in the “mild” symptomatic category. Children with radiological findings of “pneumonia” and no complications were categorized as “moderate.” Patients with mild or moderate clinical patterns, plus any manifestations suggesting disease progression (i.e., tachypnea, hypoxia, neurological deterioration, dehydration, myocardial injury, coagulation dysfunction, rhabdomyolysis), were considered “severe.” Critically ill children were those with a rapid disease progression in particular those who developed respiratory failure with need for mechanical ventilation (i.e., acute respiratory distress syndrome, persistent hypoxia), septic shock, or multiple organ failure (MOF) [28].

Laboratory data were presented as abnormally high or low according to the reference value reported by the paper, but normal ranges of main laboratory parameters were not always clearly defined. Similarly, radiological findings were categorized according to the available description.

A quantitative synthesis of the included studies was performed. For continuous variables, weighted mean (range) was calculated as appropriate, while categorical variables were expressed as percentages or frequencies in relation to the total or subtotal sample size, according to the number of missing data.

Results

We initially identified 243 papers on SARS-CoV-2 infection in children published from 1 January 2020 to 1 May 2020. After screening the title and abstract, 119 full articles were evaluated for eligibility. At the end of the selection process (Fig. 1), 62 [10, 11, 14, 16, 22, 23, 29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84] studies and three previously published reviews [85,86,87] with a total sample size of 7480 children (2428/4660 males, 52.1%) were included in the systematic review (Table 1). The weighted mean age of patients was 7.6 years, ranging from 0 to 18 years. Children included were mainly Italian (3293/7479; 44.1%), from the USA (2572/7479; 34.4%), and Chinese (1358/7479; 18.2%). The most extensive retrospective study [16] described clinical characteristics of 2143 children with confirmed (n = 713) or suspected (n = 1430) SARS-CoV-2 infection. For the purpose of this review, we considered only the confirmed cases. Overall, 2926/4709 (71%) were discharged after a weighted mean hospitalization of 11.2 days (range 2–27). The estimated mortality of confirmed cases of SARS-CoV-2 infection was 0.08%, with six reported deaths: a 10-month-old baby with intussusception who developed MOF [10], a 14-year-old boy from Hubei province [16], and a preterm newborn who died from complications of sepsis [65]. Other three children died according to the report of CDC [76], but review of these cases is still ongoing to confirm COVID-19 as the likely cause of death. Significant comorbidities were reported in 129/587 (22%) patients with known underlying disease status: 58/129 (45%) were asthmatic or had chronic lung disease [61, 62, 76, 78], 30/129 (23%) had a diagnosis of congenital heart disease [32, 61, 76, 77], 15/129 (12%) were on immunosuppressive treatment [30, 76, 77], and 8/129 (6%) children had hemato-oncological diseases [31, 61, 76]. Neurological conditions, prematurity, and metabolic disease were also reported [62, 76].

Fig. 1
figure 1

PRISMA flow diagram of the included studies on children with SARS-CoV-2 infection

Full size image
Table 1 Characteristics of the included studies and main outcome measures in children with documented SARS-CoV-2 infection
Full size table

Clinical features

Clinical findings were available in 1780 children aged 1 month to 18 years (49 studies) [10, 11, 16, 30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62, 64, 67, 69, 71, 72, 76,77,78,79,80,81, 84] (Table 2). Severe and critically ill children accounted for 2% (30/1475) and 0.6% (10/1475) of the total sample size, respectively (see also Table 5). The most commonly described symptoms in pediatric age were fever (51.6%), cough (47.3%), and sore throat (17.9%). Rarely children were also dyspneic (7.7%) and required oxygen supplementation for SpO2 below 92% (3.3%) [10, 31]. Extrarespiratory symptoms were mainly represented by diarrhea (9.7%), vomiting (7.2%), and fatigue (10.6%). A familial history of positive contact could be identified in 73.3% of the cases.

Table 2 Clinical features in children with documented SARS-CoV-2 infection
Full size table

Laboratory investigations

Table 3 summarizes the main laboratory investigations reported in 655 children (38 studies) [10, 11, 31,32,33,34, 36,37,38,39,40,41,42,43,44,45,46,47,48,49,50, 55, 58,59,60,61,62, 64, 67, 69, 71, 72, 75, 77, 79,80,81, 84]. The full blood cell count was unremarkable in most patients, with less than one fifth of them (17.1%) showing low white blood cell (WBC) and lympho- or neutropenia (13.3%). Elevated inflammatory indexes such as C-reactive protein (CRP) and procalcitonin (PCT) were shown by 31.1% of children. Creatine kinase (CPK) and liver enzymes were also altered, as shown by 14.5% and 12.4% of all patients, respectively.

Table 3 Lab investigations in children with documented SARS-CoV-2 infection
Full size table

Radiology findings

Of the 674 children who had radiological examinations (40 studies) [10, 11, 31,32,33,34, 36,37,38,39,40,41,42,43,44,45,46,47,48,49,50, 53, 55, 57,58,59,60,61,62, 64, 67, 69, 71, 72, 75, 77, 79,80,81, 84], up to 49.1% of them showed abnormalities, even if asymptomatic (17/113, 15%) (Table 4). Most patients underwent a chest CT scan (73.9%) that resulted normal in 198 out of 605 patients (32.7%), whereas typical ground-glass opacities (GGO), nonspecific unilateral and bilateral lesions were identified in 29.4%, 26.6%, and 23.2% of patients, respectively.

Table 4 Radiological findings in children with documented SARS-CoV-2 infection
Full size table

Treatment

Of all patients, about 2% were admitted in the pediatric intensive care unit (PICU) and required mechanical ventilation (MV) (Table 5) [10, 11, 31,32,33, 36,37,38, 40, 41, 43, 44, 49, 51, 53, 55, 57,58,59,60,61,62, 71, 72, 75,76,77,78,79,80, 84]. Most authors described the use of nebulized Interferon (IFN) (51.5%), and of other antiviral agents (38.4%) or antibiotics (23.3%). The use of intravenous immunoglobulin (IVIg) and corticosteroids (CCS) was less frequently described (5.2% and 5.8%, respectively).

Table 5 Treatments used in children with documented SARS-CoV-2 infection
Full size table

Neonatal cases

A few case reports [22, 49, 63, 68, 70, 73, 74, 82, 83] and case series [23, 65, 66] including a total of 25 newborns (17 males, 74%) with SARS-CoV-2 were identified (Table 6). Neonates were usually screened because of a history of primary maternal infection (84%). Similarly to older children, most of them were asymptomatic (20%) or had mild (48%) and moderate (20%) signs of clinical infection. However, a slightly higher proportion of them was severely ill (12%). Dyspnea was the most common reported sign in neonatal age (40%). Fever (32%) and feeding intolerance (24%) were also described. Blood tests showed high WBC (20%), CRP and/or PCT (12%), CPK (20%), and liver enzymes (16%).

Table 6 Clinical features and laboratory results in newborns and infants ≤ 3 months of age with documented SARS-CoV-2 infection
Full size table

Unlike older children, newborns underwent a chest X-ray in most cases (64%) (Table 7). Abnormal radiological findings could be recognized in less than half of them (48%), but specific lesions were not so frequently described: 4% had GGO, 20% unilateral patchy area, and 12% bilaterally. Poor information on treatment options was obtainable from the included reports, and most patients (48%) received only symptomatic therapies.

Table 7 Radiological findings and treatments used in newborns and infants ≤ 3 months of age with documented SARS-CoV-2 infection
Full size table

Discussion

The knowledge about epidemiological and clinical features of novel Coronavirus disease needs to be continuously updated. However, current data have been reported mainly from adult patients, while many aspects of the infection remain unclear in children. This systematic review fully assesses epidemiological trends and clinical characteristics of SARS-CoV-2 infections in pediatric age, giving an analytical summary of the available heterogeneous evidence.

Children are less likely to develop severe symptoms of COVID-19 than adults, with 95% of all cases ranging from asymptomatic to mild-moderate clinical patterns, as described by different case series [10, 11, 30, 33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62, 71, 72, 76, 77, 81]. Moreover, only 2% of patients were admitted to PICU or required MV [10, 31, 32, 75,76,77,78]. Overall, six deaths were reported (mortality rate 0.08%): most patients developed complications [10, 16], including a preterm newborn who died from sepsis [65]. In adulthood, over two thirds of those who died from COVID-19 had a comorbidity [88], while about 20% of children with underlying disease could be identified, and none of them showed worse clinical course of the infection in comparison to previously healthy patients [10, 30, 32, 61, 62]. The explanation for the lower mortality rate in children might rely not only on epidemiological reasons. Some authors hypothesized a molecular mechanism making children less susceptible to SARS-CoV-19 infection. The pathogenesis of human coronavirus mainly depends on the interactions between its transmembrane spike glycoprotein (S-protein) and specific cell receptors of angiotensin-converting enzyme II (ACE2) [89]. It has been shown that the expression of this enzyme starts to increase in later childhood [90], and this may protect children from the most aggressive pattern of the infection.

However, children might not be tested for SARS-CoV-2 as frequently as adults. Most countries choose to test and centralize only symptomatic patients. We may speculate that a small number of children have been tested because they are mostly asymptomatic and therefore just locked down at home. Moreover, the current gold standard for the diagnosis of SARS-CoV-2 infection is real time-polymerase chain reaction (RT-PCR) on respiratory tract specimen. The diagnostic accuracy of RT-PCR highly depends on the virus-specific diagnostic window (e.g., the viral load may decrease from the second week of the disease), and the analytical sensitivity of this assay is potentially plagued by false SARS-CoV-2 negativity attributable to the low viral loads, especially in asymptomatic or mild symptomatic patients that might transmit the disease as well [91]. There are very recent data from a German study indicating that viral loads in the very young (age group 0–6 years) do not significantly differ from those of adults. This means that even if children are less often symptomatic, they might be as infectious as adults [92].

The majority of pediatric patients infected with SARS-CoV-2 have been part of a family cluster outbreak or had a household positive contact (85% of all reported cases) [10, 32,33,34, 36, 58, 60,61,62]. Several other studies documented how COVID-19 spreads within families [93]. Preliminary findings suggested differences in the onset and duration of viral shedding between and within family clusters, with adult subjects remaining positive for a prolonged time [93]. Further investigation with the inclusion of the virus genome may be worthy.

Clinical features

Infected children usually show typical symptoms of acute respiratory infections including fever (49.6%) and cough (45.5%) [10, 11, 16, 30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62, 64, 67, 69, 71, 72, 76,77,78,79,80,81, 84]. In particular, some authors reported that up to one third of symptomatic children may have high fever [10, 77], but generally below 39 °C [38, 77]. Differently from adults, children are more likely to present with extra-respiratory symptoms [94]: diarrhea (9.4%) and vomiting (7.3%) are the most frequently reported ones. It has been showed that, when present, gastrointestinal symptoms usually anticipate the typical respiratory pattern [95]. Previous studies on SARS-CoV cases demonstrated the viral detection in gut biopsy specimens and stool of recovered patients, indicating a possible gastrointestinal tract tropism that may partially provide explanations for extra-respiratory symptoms and persistent viral shedding through fecal–oral route [96]. There is growing evidence that this mechanism of excretion may be typical also for SARS-CoV-2 [2, 37]. As described in a case series of ten infected children, SARS-CoV-2 remained detectable in rectal swabs after nasopharyngeal swabs turned negative [37]. However, the extra-pulmonary detection of viral RNA does not mean infectious virus is present, but two independent laboratories from China declared that they have successfully isolated live 2019-nCoV from the stool of patients (unpublished) [95]. Moreover, recent analysis revealed that ACE2 was expressed also in upper esophagus and absorptive enterocytes from ileum and colon [61].

According to evaluated studies, clinical presentation in newborns could be slightly different than in older children, with a higher proportion of them (12%) presenting with a severe pattern [65, 66]. Even if the vertical transmission for SARS-CoV-2 has not been demonstrated [19, 97], in 84% of neonatal cases the mother was infected [23, 63, 65, 66, 82, 83]. Moreover, nosocomial infection may also occur, and strict measures to reduce this risk should be always observed [24]. However, some authors hypothesized that 2019-nCoV infection and morbidity in newborns may be related to possible hypoxemia in the infected mother that increases the risk of perinatal adverse events such as birth asphyxia and premature birth [65, 98]. Evidence is still too limited.

Laboratory findings

Overall, no significant abnormalities were observed and this was consistent with the results of a previously published review including a total sample size of 66 children with confirmed SARS-CoV-2 infection [85]. In particular, full blood cell count was normal in most patients, with only two case series reporting high rates of lymphopenia (10/25, 40% [32] and 11/36, 30.1% [60], respectively). However, this finding seems to be in contrast with adult data, as low lymphocyte count has been noted in up to 80% of infected critically ill subjects [91, 99]. The limited number of severe COVID-19 infection may in part explain the lack of significant lymphopenia in children.

Our results suggested that inflammatory indexes may be abnormal in one third of children with SARS-CoV-2 infection (30.9%), while Henry et al. described only 10–13% of cases with high CRP and/or PCT [85]. This controversial finding could be explained by the high heterogeneity in defining a cut-off of abnormal values across all the included studies. However, in adults a PCT value of ≥ 0.5 ng/mL was observed to be associated with a near fivefold increase in risk of severe clinical course of COVID-19 infection [100].

Other significant reported laboratory investigations were high CPK values (13.6%) and liver enzymes (12.3%). These enzymes could be often altered during viral infections [101]. In particular, high CPK levels or aspartate aminotransferase activity correlated with more severe clinical patterns in adult patients with SARS-CoV-2 infection [102]. Abnormal transaminase levels may also express a sign of direct liver damage. Recently published data demonstrated enrichment of ACE2 expression in cholangiocytes (59.7% of cells) suggesting that SARS-CoV-2 might lead to direct damage of intrahepatic bile ducts [103].

Radiology findings

Half of the children who had radiological examinations showed abnormalities. The sensitivity of chest X-ray is supposed to be inferior to that of CT-scan: in adults, nonspecific patchy peripheral and peri-bronchial opacities may be shown in all lung zones, according to the severity of the infection [104]. As children usually develop milder patterns of the disease, chest X-ray may fail to identify typical lesions, and it is mainly adopted in newborns and younger infants [23, 65, 69]. We identified 15% of asymptomatic children with abnormal radiological findings across all studies; however, both clinical and radiological patterns were specifically described only in a few case series, therefore this proportion of patients may be even higher. CT scan is frequently performed in children with suspected or confirmed SARS-CoV-2 infection (up to 74% of all cases reported). The most frequently recognizable lesions are represented by GGO, with unilateral or bilateral distribution [33,34,35]. However, CT scan resulted completely normal in more than one third of asymptomatic patients who underwent this examination [10, 33,34,35,36, 44, 45, 55, 63]. The use of CT scan as a diagnostic screening tool in confirmed or suspected COVID-19 patients is supported by recent evidence showing that its sensitivity could be greater than that of real-time PCR in detecting the virus (98% vs. 71%, respectively) [105]. However, routine use of CT scan has several obvious implications, in particular in a pediatric setting where concern about unnecessary exposure to radiation source should be raised. Therefore, other possible diagnostic imaging tools may be used. Lung ultrasound (LUS) has been successfully adopted in adult subjects with SARS-CoV-2 infection [106]. A recently published Italian case series reported LUS findings in eight children with COVID-19 infection: subpleural consolidations and confluent B-lines were identified in all patients and results were confirmed by a chest radiography [107].

Treatment

Currently, a few registered drug trials for COVID-19 experimental treatment included children [108]. Symptomatic treatment alone was used in most cases, in particular in newborns [65].

The only therapeutic recommendation in pediatric age is to use nebulized IFN and oral antiviral agents (i.e., lopinavir/ritonavir), with CCS for complications (acute respiratory distress syndrome, encephalitis, hemophagocytic syndrome, or septic shock) and IVIg for severe cases [28]. However, none of these therapies have shown a clear benefit in the treatment of SARS-CoV-2, and neither the WHO nor the US CDC recommends any specific treatment in children [25, 109].

The main limitation of our review is the difficulty to retrieve the full text of some Chinese studies; thus, we had to rely on English-language summaries when data were easily retrievable or publications that referenced papers published in Chinese. Given the large amount of currently published papers on COVID-19, we have decided to base our research on a single but robust database. However, this strategy might implicate possible selection bias of the included studies. Moreover, our findings are essentially based on limited case series and case reports, therefore laboratory parameters of interest were not consistently reported and reference ranges were not always clearly defined. Similarly, radiological patterns were difficult to compare, as the description of the included cases was not standardized.

However, despite the scarce and extremely heterogeneous evidence on pediatric patients with COVID-19, the use of systematic databases allowed us to review at a glance and interpret the majority of published studies up to 1 May 2020.

Conclusion

SARS-CoV-2 affects children less commonly and severely in comparison to adults, with an estimated very low mortality rate, but there is growing evidence showing that they are as susceptible to become infected as adults. This could be due either to the fact that children are less frequently exposed to the main sources of transmission (in particular, nosocomial), and that they tend to show milder symptoms, therefore may be less often tested. In symptomatic patients, laboratory and radiology findings are mainly nonspecific, but could help identifying those who are severely ill. Larger epidemiological and clinical cohort studies are needed to better understand possible implications of COVID-19 infection in children.