SARS-CoV-2 infection in children
Children are less susceptible to SARS-COV-2 infection, have a lower seroprevalence and a less severe COVID-19 disease course than adults (Castagnoli 2020, Viner 2020, Merckx 2020, Zimmermann 2020, Parri 2020, Ludvigsson 2020). In this regard, COVID is strikingly different from other virus-induced respiratory diseases, which can be fatal in children (e.g. RSV in infants). The CoV-2 pandemic causes a large collateral damage to children because they are taken out of their normal social environment (kindergartens, schools etc.), and because of parents’ resistance to seek medical care despite need e.g. for vaccination (Bramer 2020) or even if their children are having an emergency (Lazzerini 2020).
Open the references of this chapter in a separate window.
Common coronaviruses in children: tropism, incubation period and spreading
The first International Corona Virus Conference was organized by Volker Termeulen in Würzburg/Germany in 1980. At the time only one human coronavirus, HCoV2229E, was known to be associated with the common cold (Weiss 2020). Commonly circulating human coronaviruses (COV) can be isolated from 4-8% of all children with acute respiratory tract infections, which tend to be mild, unless the child is immunocompromised (Ogimi 2019). Seven coronaviruses circulate among humans (Hufert 2020): α-Coronaviruses HCoV-229e (discovered in 1966), HCoV-NL63 (in 2004); β-Coronaviruses HCoV-OC43 (in 1967)-HKU1 (in 2005), -OC43; MERS-CoV (in 2012), SARS-CoV (in 2003) and SARS-CoV-2 that originally derive from bats (NL63, 229e, SARS-CoV), dromedary camels (229e, MERS-CoV), cattle (OC43) and pangolins (SARS-CoV-2) (Zimmermann 2020). There appear to be re-infections with the earlier described common COV despite the fact that most individuals seroconvert to human coronaviruses. In many children there are co-infections with other viruses such as Adeno-, Boca-, Rhino-, RSV-, Influenza- or Parainfluenza virus. There seems to be a cyclical pattern with seasonal outbreaks between December and May or March to November in the southern hemisphere.
Single-strand RNA coronaviruses are capable of mutation and recombination leading to novel coronaviruses that can spread from animals to humans. They have caused epidemics leading to significant case fatality rates (10% in SARS-CoV, Hong Kong 2002; more than 30% in MERS-CoV, Saudi Arabia 2012). Because of the high case fatality rate, both SARS-COV and MERS-COV have a low potential for long-term sustained community transmission. Accordingly, no human SARS-CoV infections have been reported since July 2003.
It is estimated that in SARS-CoV-2, one person infects 2-3 other persons. In clusters (e.g. nosocomial outbreaks) this number might be much higher. In both SARS-CoV and MERS-CoV, super-spreading events with one individual infecting up to 22 (SARS) or even 30 individuals (MERS) have been reported, especially in nosocomial outbreaks. In SARS-CoV a total of 41 children were reported with no deaths. Similarly, in MERS-CoV only 38 children were reported in two studies, with two deaths (Zimmermann 2020).
COVID Reference resources:
Home | Daily Top 10 | 7-Day Overview | Download | Epidemiology | Transmission | Prevention | Virology | Vaccines | Diagnosis | Clinical Manifestations | Treatment | Severe COVID-19 | Comorbidities / Special Populations | Pediatrics| Timeline | Preface
Epidemiology of COVID-19 in children
On April 6 the US CDC reported 2572 (1.7%) children under 18 years among 149,082 reported cases from 12 February to 2 April 2020. The availability of data was extremely limited (less than 10% available on symptoms, 13% on underlying conditions, 33% on whether children were hospitalized or not). Three deaths were reported to the CDC but no details were given. The median age was 11 and they were 57% males. 15 children were admitted to an ICU (≤ 2%). Children < 1 year accounted for the highest percentage (15-62%) of hospitalization (CDC 2020). The Chinese CDC report (Dong 2020) comprises 2143 pediatric patients from January 16 to February 8 2020. Only 731 children (34,1%) were laboratory confirmed cases. The median age was 7 years with 56,6% boys, less than 5% were classified as severe and less than 1% as critical.. The Korean Center for Disease Control and Prevention reported on 20 March that 6.3% of all COVID-19 cases were children under 19 years of age; again, the children had a mild form of the disease (Korean Center for Disease Control and Prevention. Press releases, https://www.cdc.go.kr). Italian data published on 18 March showed that only 1,2% of the 22,512 Italian cases with COVID-19 were children; no deaths were reported in this and in the Spanish cohort from Madrid (2 March to 16 March) (Livingstone 2020, Tagarro 2020). In Germany, 9657 children and adolescents with COVID-19 were reported up to May 4th, 2020; only 128 were admitted to 66 hospitals, only one child died (Armann 2020).
The European Surveillance System (TESSy) collects data from EU/EEA countries and the UK on laboratory-confirmed cases of COVID-19. Out of 576,024 laboratory confirmed COVID-19 cases 0,7% were 0-4 years, 0,6% 5-9 years, 0,9% 10-14 years (https://covid19-surveillance-report.ecdc.europa.eu). The multicentre cohort study (82 participating health-care institutions across 25 European countries), Paediatric Tuberculosis research Network (ptbnet) confirmed that COVID-19 is generally a mild disease in children. Of 582 children and adolescents (median age 5.0 years, 25% with pre-existing conditions) with PCR-confirmed SARS-CoV-2 infection, 363 (62%) were admitted to hospital and 48 (8%) required ICU admission. Significant risk factors for requiring ICU admission in multivariate analyses were being younger than 1 month (odds ratio 5.1), male sex (2.1) and pre-existing medical conditions (3.3). Four children died (Götzinger 2020).
Natural course and risk factors for complications
The incubation period is believed to be 3-7 days (range 1-14 days) (She 2020), the clinical onset 5-8 days after infection with the virus. Children often have an asymptomatic or less severe COVID-19 disease course than adults (Zimmermann 2020, Parri 2020). Among a total of 100 children with SARS-CoV-2 from Italy, 21% were asymptomatic, 58% had mild disease, 19% had moderate disease, 1% had severe disease, and 1% were in critical condition (Parri 2020).
Due to the paucity of data it is as yet unclear which group of children may be at a higher risk for development of complications, e.g. children with underlying chronic pulmonary or cardiac disease, severe neurologic deficits, immunosuppressed or critically ill children, etc. Analogous to influenza there might be genetic susceptibility in some children (see below, pathophysiology, Clohisey 2019). Interestingly, in a flash survey from 25 countries with 10,000 children with cancer at risk and 200 tested, only 9 were found to be CoV-2 positive. They were asymptomatic or had mild disease (Hrusak 2020). Even in the severely immunosuppressed and in children with significant cardiac and pulmonary comorbidities COVID-19 can be overcome (Dinkelbach 2020).
In The European Surveillance System (TESSy) deaths among children aged below 15 years are rare, 4 out of 44,695 (0.009%) were reported. The rate of hospitalization was higher in children under the age of five especially in infants compared to persons aged 5-29. However, it is believed that the threshold for admission is lower in young children. A severe course requiring admission to ICU seems not to be more likely in younger children. The likelihood of being hospitalised was higher when children had an underlying condition, and a severe course was rare (https://covid19-surveillance-report.ecdc.europa.eu).
In a cross-sectional study including 48 children with COVID-19 (median age 13 years; admitted to 46 North American pediatric ICUs between March 14 and April 3, 2020), forty patients (83%) had significant pre-existing co-morbidities and 18 (38%) required invasive ventilation. Targeted therapies were used in 28 patients (61%, mainly HCQ). Two patients (4%) died and 15 (31%) were still hospitalized, with 3 still requiring ventilatory support and 1 receiving ECMO (Shekerdemian 2020). In an observational retrospective cohort study that included 177 children and young adults with clinical symptoms and laboratory confirmed SARS-CoV-2 infection treated between March 15 and April 30, 2020 at the Children’s National Hospital in Washington, 44 were hospitalized and 9 were critically ill. Of these, 6/9 were adolescents and young adults > 15 years of age. Although asthma was the most prevalent underlying condition overall, it was not more common among patients with severe disease (DeBiasi 2020).
Although the natural course of COVID-19 is uneventful in most pediatric patients, a very small percentage can develop a potentially fatal severe hyperinflammatory state 2-4 weeks after acute infection with SARS-CoV-2 (Riphagen 2020). This hyperinflammatory state is termed as pediatric inflammatory multisystem syndrome temporarily associated with SARS-CoV-2 (PIMS-TS) (or synonym Multisystem Inflammatory Syndrome in Children (MIS-C). Of the 570 MIS-C cases reported to the CDC by July 2020, 10 patients had died (1.8% ) and 364 (63.9%) patients required treatment in an intensive care unit. Obesity was the most commonly reported underlying medical condition (Godfred-Cato 2020).
Pathophysiology and immunopathology
It is unclear why COVID-19 in children is associated with a less severe disease course.
The tissue expression pattern of the receptor for CoV-2 angiotensin converting enzyme (ACE2) and the transmembrane serine protease TMPRSS2 (essential for CoV-2 cell entry) as well as the tissue tropism of CoV-2 in childhood are unknown but age-dependent differences in ACE2 receptor expression may explain why outcomes differ in children versus adults (Bunyavanich 2020).
ACE2 is expressed on cells of the airways, the lungs, mucosal cells (lids, eyelids, nasal cavities), intestines and on immune cells (monocytes, lymphocytes, neutrophils) (Molloy 2020, reviewed in Brodin 2020). It needs to be clarified whether there is neurotropism (e.g. affecting the developing brain of newborns).
The main target of CoV-2 is the respiratory tract. As respiratory infections are extremely common in children it is to be expected that there are other viruses present in the respiratory tract of young children concomitantly with the coronavirus, which may limit its growth and the number of CoV-2 copies in the respiratory tract of children. Systematic viral load measurements in the respiratory tract of different viruses in children are underway. Key to the later immunopathologic stages of COVID-19 pneumonia is the macrophage activation syndrome (MAS)-like hyperinflammatory phase with a cytokine storm and acute respiratory distress syndrome (ARDS), usually within 10-12 days after symptom onset. In general, children are not less prone to develop ARDS during respiratory tract infections than adults. In the H1N1 flu pandemic in 2009, being under the age of 1 year was a significant risk factor for developing a severe form of the infection and ARDS (Bautista 2010). Why ARDS is less common in children compared to adults with COVID-19 is unclear. SARS-CoV-2 infection of cardiac tissue can be a major contributor to fatal myocarditis (Dolhnikoff 2020, Prieto 2020).
An explanation for the milder disease course in children could be age-related differences in innate or adaptive immune responses to CoV-2 between adults and children. In the innate immune response to any virus, Type I (IFN α, IFN β) and type III (IFN Ω) interferons are the most important cytokines. In 659 patients (1 month to 99 years old) with life-threatening COVID-19 pneumonia, inborn defects in the type 1 IFN signaling were found in 23 unrelated patients (Zhang 2020). Moreover, neutralizing auto-antibodies to type I/III IFN were found in 101/987 patients with life-threatening COVID-10 pneumonia (Bastard 2020). These findings show that inborn defects in the IFN I/II pathway or auto-antibodies to IFN I/III may predispose to life-threatening COVID-19. Based on influenza animal models it has been proposed that BCG vaccination (for tuberculosis prevention, done in the first week of life in some countries) may enhance non-specific innate immunity in children to infections like COVID-19 (so-called trained immunity) (Moorlag 2019). A search of the BCG World Atlas and correlation with data of COVID-19 cases and death per country found that countries without universal policies of BCG vaccination (Italy, the Netherlands, USA) have been more severely affected compared to countries with universal and long-standing BCG policies and that BCG vaccination also reduced the number of reported COVID-19 cases in a country (Miyasaka 2020, Hauer 2020). Recent data from a large population-based study did not show decreased infection rates in Israeli adults aged 35 to 41 years who were BCG-vaccinated in childhood as compared to non-BCG-vaccinated. Data on the effect of BCG vaccination on COVID-19 disease severity are unavailable (Hamiel 2020).
In the adaptive response to any virus, cytotoxic T cells play an important role in regulating responses to viral infections and control of viral replication. Children could benefit from the fact that the cytotoxic effector function of CD8 T cells in viral infection in children may be less detrimental compared to adults. Immune dysregulation with exhaustion of T cells has been reported in adults with COVID-19 infection. Regarding humoral immunity, IgG maternal antibodies are actively transferred to the child via placenta and/or IgA via breast milk. They may not include anti CoV-2 antibodies, if the mother is naïve to CoV-2 or infected late in pregnancy. In mothers with COVID-19 pneumonia serum and throat swabs of their newborns were negative for CoV-2 but virus-specific IgG antibodies were detected (Zeng H 2020). Thus, neonates may benefit from placental transmission of virus-specific antibodies from pre-exposed mothers. As shown in SARS CoV-1 it is likely that in SARS-CoV-2 a newly infected child will mount a significant humoral response with neutralizing IgM (within days) and IgG antibodies (within 1-3 weeks) to one of the immunodominant epitopes, e.g. the crown-like spike proteins giving the coronaviruses their name. Infections with non-SARS COV are very common in children (see above); however, to what extent previous infections with non-SARS coronaviruses may have led to protective cross-reactive antibodies is unclear.
Data regarding IgG and IgM seroprevalence and quality of the immune response in children are lacking. No human re-infections with CoV-2 have been demonstrated yet but overall it is not clear whether children mount a durable memory immune response to CoV-2. In summary, differences in the immune system such as more efficient innate and adaptive immunity to COV-2 (associated with better thymic function), cross-reactive immunity to common cold coronaviruses and differences in the ACE2 receptor expression as well as better overall health may be factors leading to a better COVID-19 outcome in children (Consiglio 2020).
Studies on the risk of acquiring SARS-CoV-2 infection in children in comparison to adults have shown contradicting results (Mehta 2020, Gudbjartsson 2020, Bi 2020). The exact role that children play in the transmission of SARS-CoV-2 is not yet fully understood. Population based studies performed so far indicate that children might not play a major factor in the spreading of COVID-19 (Gudbjartsson 2020).
Contraction of COVID-19 in a pregnant woman may have an impact on fetal outcome, namely fetal distress, potential preterm birth or respiratory distress if the mother gets very sick. Schwartz reviewed 5 publications from China and was able to identify 38 pregnant women with 39 offspring among whom 30 were tested for COVID-19 and all of them were negative (Schwartz 2020, Chen 2020). Among the 24 infants born to women with COVID-19, 15 (62.5%) had detectable IgG and 6 (25.0%) had detectable IgM; nucleic acid test results were all negative. Among 11 infants tested at birth, all had detectable IgG and 5 had detectable IgM. IgG titers with positive IgM declined more slowly than those without (Gao 2020). In the PRIORITY study (n = 263), adverse outcomes, including preterm birth, NICU admission, and respiratory disease, did not differ between infants born to mothers testing positive for SARS-CoV-2 (n = 184) and those born to mothers testing negative (n = 79), suggesting that infants born to mothers infected with SARS-CoV-2 generally do well in the first 6-8 weeks after birth (Flaherman 2020).
Transmission of COVID-19 appears unlikely to occur if correct hygiene precautions are undertaken. In 1481 deliveries at three hospitals in New York City, 116 (8%) mothers tested positive for SARS-CoV-2; 120 neonates were identified and none were positive for SARS-CoV-2 (Salvatore 2020).
In another study from New York, 101 newborns of SARS-CoV-2 infected mothers no transmission was observed despite sleeping in the same room and breastfeeding (Dumitriu 2020). Initially it was thought that CoV-2 is not vertically transmitted, but in a more recent analysis of 31 mothers with SARS-CoV-2, SARS-CoV-2 genome was detected in one umbilical cord blood, two at-term placentas, one vaginal mucosa and one breast-milk specimen. Three cases of vertical transmission of SARS-CoV-2 have been documented (Fenizia 2020).
In a UK national population-based cohort study on SARS-CoV-2 infected pregnant women, twelve (5%) of 265 infants subsequently tested positive for SARS-CoV-2 RNA, six of them within the first 12 hours after birth (Knight 2020). Postpartum acquisition appears to be the most common mode of infection; in a recent review only 4/1141 neonates born to SARS-CoV-2 infected mothers were thought to have a congenital infection (Dhir 2020).
Culture-competent SARS-CoV-2 has been grown from the nasopharynx of symptomatic neonates, children, and adolescents: 12 (52%) of 23 symptomatic SARS-CoV-2–infected children, the youngest being 7 days old. SARS-CoV-2 viral load and shedding patterns of culture-competent virus in the 12 symptomatic children resembled those in adults. Systematic measurements of SARS-CoV-2 viral load measurements in children are lacking. Therefore, transmission of SARS-CoV-2 from children is plausible (L’Huillier 2020). SARS-CoV-2 in children is transmitted through family contacts and mainly through respiratory droplets (Garazzino 2020). In a study from France, child-to-child and child-to-adult transmission seems to be uncommon (Danis 2019). Prolonged exposure to high concentrations of aerosols may facilitate transmission (She 2020).
SARS-CoV-2 may theoretically also be transmitted through the digestive tract. ACE2 is also found in upper esophageal and epithelial cells as well as intestinal epithelial cells in the ileum and colon (She 2020). SARS-CoV-2 RNA can be detected in the feces of patients (Holshue 2020). Cai revealed that viral RNA is detected from feces of children at a high rate (and can be excreted for as long as 2-4 weeks) (Cai 2020). However, direct evidence of a fecal-to-oral transmission has not yet been documented.
Onward transmission from children to others is low (Viner 2020, Merckx 2020). In a study from Milan, Itlay, in 83 children and 131 adults hospitalized and symptomatic in regard to COVID-19, adults were retrospectively more likely to be CoV-2 positive, asymptomatic carriers as compared to children (9% vs 1%) (Milani 2020).
Diagnosis and classification
Testing for the virus is only necessary in clinically suspect children. If the result is initially negative, repeat nasopharyngeal or throat swab testing of upper respiratory tract samples or testing of lower respiratory tract samples should be done. Sampling of the lower respiratory tract (induced sputum or bronchoalveolar lavage) is more sensitive (Han 2020). This is not always possible in critically ill patients and in young children.
Diagnosis is usually made by real-time polymerase chain reaction RT-PCR on respiratory secretions. For SARS-CoV, MERS-CoV and SARS-CoV-2, higher viral loads have been detected in samples from lower respiratory tract compared with upper respiratory tract.
In some patients, SARS-CoV-2 RNA is negative in respiratory samples while stool samples are still positive indicating that a viral gastrointestinal infection can last even after viral clearance in the respiratory tract. (Xiao 2020). Fecal testing may thus be of value in diagnosing COVID-19 in these patients.
As in other viral infections, a CoV-2 IgM and IgG seroconversion will appear in days (IgM) to 1-3 weeks (IgG) after infection and may or may not indicate protective immunity (still to be determined). Interestingly, asymptomatic seroconversion has been hypothesized in a very small series of health workers (mean age 40 years) exposed to a child with COVID-19 in a pediatric dialysis unit (Hains 2020).
Serology may be useful in patients with clinical symptoms highly suggestive of SARS-CoV-2 who are RNA negative, i.e in children with pediatric inflammatory multisystem syndrome temporarily associated with SARS-CoV-2 (PIMS-TS). If serology indicates protective immunity, this will be extremely important from a public health perspective, e.g. it will allow for strategic staffing in medical care and for the assessment of CoV-2 epidemiology (herd immunity).
|Table 1. COVID classification in children (Shen 2020)|
|1||Asymptomatic without any clinical symptoms|
|2||Mild fever, fatigue, myalgia and symptoms of acute respiratory tract infections|
|3||Moderate pneumonia, fever and cough, productive cough, wheezing but no hypoxemia|
|4||Severe fever, cough, tachypnea, oxygen saturation less than 92%, somnolence|
|5||Critical quick progress to acute respiratory distress syndrome (ARDS) or respiratory failure|
Laboratory and radiology findings
Laboratory and/or radiology studies in outpatient children who have mild disease are not indicated. Upon admission to the hospital the white blood cell count is usually normal. In a minority of children decreased lymphocyte counts have been documented. In contrast, adults (with hyperinflammation and cytokine release syndrome) often have an increase in neutrophils and lymphopenia. The inflammation parameters C reactive protein and procalcitonin can be slightly elevated or normal while there are elevated liver enzymes, creatine kinase CK-MB and D dimers in some patients. LDH appears to be elevated in severe cases and can be used to monitor severe disease.
A chest X-ray should only be done in children with moderate or more severe disease as CT scans mean a very high radiation exposure for the child and should only be done in complicated or high-risk cases. In the beginning of the pandemic in China, children all received CT scans even when they were asymptomatic and oligosymptomatic; surprisingly, they displayed very severe changes. On chest radiography there are bilateral patchy airspace consolidations and so-called ground-glass opacities. CT scans were more impressive than chest x-ray examinations. In 20 children with CT, 16 (80%) had some abnormalities (Xia 2020).
Symptoms and signs: Acute infection
Children and adolescents
In a clinical trial of 171 children from Wuhan, fever was reported in 41% (71 of 171), cough in over 50% (83 of 171), tachypnea in 28% (49 of 171). In 27 of the patients there were no symptoms at all (15,8%). At initial presentation very few children required oxygen supplementation (4 of 171, 2,3%). Other symptoms like diarrhea, fatigue, runny nose and vomiting were observed in less than 10% of the children (Lu 2020). In the cohort from Zhejiang as many as 10 out of 36 patients (28%) had no symptoms at all. None of the children had an oxygen saturation below 92% (Qiu 2020). In a Korean case series of children with COVID-19, 20 children (22%) were asymptomatic during the entire observation period. Among 71 symptomatic cases, only 6 (9%) were diagnosed at the time of symptom onset while 47 children (66%) had unrecognized symptoms before diagnosis and 18 (25%) developed symptoms after diagnosis. Fifty-one percent had “mild” disease, 22% “moderate” disease and 2% “severe” disease. No patient required intensive care (Han 2020). A larger UK series reports on 651 children and young people aged less than 19 years. Median age was 4.6 years, 35% (225/651) were under 12 months old. 18% (116/632) of children were admitted to critical care. Six patients died in hospital, all of whom had profound comorbidity (Swann 2020).
A recent comprehensive systematic review analysed 131 studies in 7780 pediatric COVID-19 patients across 26 different countries (Hoang 2020). In this review 19,3% of the patients were asymptomatic, the most common symptoms were fever (59%), cough (55,9%), rhinorrhea (20%) and myalgia/fatigue (18,7%). The need for intensive care treatment was low (3,3%).
In 52 hospitalized children from London, UK, renal dysfunction was frequent especially in those with pediatric inflammatory multisystem syndrome temporarily associated with SARS-CoV-2. 24 (46%) had elevated serum creatinine, and 15 (29%) met the diagnostic criteria for acute kidney injury (Stewart 2020).
In a case series of 4 children with PIMS-TS (see below) from London, UK, neurological symptoms were described (encephalopathy, headaches, brainstem and cerebellar signs, muscle weakness, reduced reflexes) with signal changes in the splenium of the corpus callosum on neuroimaging and required intensive care admission for the treatment of COVID-19 pediatric multisystem inflammatory syndrome (Abdel-Mannan 2020).
Neonates and infants
Zeng reports 33 newborns born to mothers with COVID-19 in Wuhan. Three of the 33 infants (9%) presented with early-onset SARS-CoV-2 infection. In 2 of the 3 neonates there were radiological signs of pneumonia. In one child disseminated intravascular coagulation was described but eventually all children had stable vital signs three weeks after the infection (Zeng L 2020). In a second cohort, 9 infants aged 1 month to 9 months were described without any severe complications (Wei 2020). Whether there are long-term complications of COVID-19 in these newborns and infants is unclear at this stage of the pandemic.
Pediatric inflammatory multisystem syndrome temporarily associated with SARS-CoV-2 (PIMS-TS) (or synonym Multisystem Inflammatory Syndrome in Children (MIS-C) or Kawasaki-like Disease
While most children with COVID-19 have a very mild disease, in April 2020 clinicians from the UK, France, Italy, Spain and the US reported on children with a severe inflammatory syndrome with Kawasaki-like features, some of whom had tested positive for CoV-2, while others not. Prior to this, Jones had described the case of a six-month-old baby girl with fever, rash and swelling characteristic of a rare pediatric inflammatory condition, Kawasaki disease (Jones 2020).
Eight patients from the UK and 10 patients from Bergamo in Italy with features of Kawasaki disease were published including one death in a 14-year-old boy in the UK during the SARS-CoV-2 epidemic (Riphagen 2020, Verdoni 2020). Some children presented with vasculitic skin rash (Schneider 2020). In Bergamo, the region with the highest infection rate in Italy, a 30-fold increased incidence of Kawasaki disease has been reported following the SARS-CoV-2 epidemic (Verdoni 2020). Of 21 children and adolescents from London, UK (19 with recent SARS-CoV-2 infection), 12 (57%) presented with Kawasaki disease shock syndrome, 16 (76%) with myocarditis, 17 (81%) required intensive care support. All had noticeable gastrointestinal symptoms and high levels of inflammatory markers, received intravenous immunoglobulin and 10 (48%) corticosteroids; the outcome was favourable in all (Toubiana 2020).
In the UK, 78 of the PIMS-TS cases reported 36 (46%) were invasively ventilated, 28 (36%) had evidence of coronary artery abnormalities, three children needed ECMO and two children died (Davies 2020).
In another study from the UK, 50% of the 58 “PIMS-TS” cases developed shock and required inotropic support or fluid resuscitation; 22% met diagnostic criteria for Kawasaki disease; and 14% had coronary artery dilatation or aneurysms (Whittaker 2020).
In a US MIS-C study on 186 patients 131 (70%) were positive for SARS-CoV-2 by RT-PCR or antibody testing. Detailed analysis of clinical manifestation revealed the gastrointestinal system (92%), cardiovascular (80%), hematologic (76%), mucocutaneous (74%), and respiratory involvement (70%). In total, 148 patients (80%) received intensive care, 37 (20%) received mechanical ventilation, and 4 (2%) died. Coronary-artery aneurysms were documented in 15 patients (8%), and Kawasaki disease–like features were documented in 74 (40%) (Feldstein 2020). In the largest cohort to date,
570 US MIS-C patients were reported as of July 29. A total of 203 (35.6%) of the patients had a typical MIS-C clinical course (shock, cardiac dysfunction, abdominal pain, and markedly elevated inflammatory markers) and almost all had positive SARS-CoV-2 test results (Class 1). The remaining 367 (64.4%) of MIS-C patients (Class 2 and 3) had manifestations that appeared to overlap with acute COVID-19 or had features of Kawasaki disease. 364/570 patients (63.9%) required care in an intensive care unit. Ten patients (1.8%) died. Approximately two thirds of the children had no pre-existing underlying medical conditions (Godfred-Cato 2020).
In summary, the pathophysiological overlap between COVID-19-associated inflammation and Kawasaki disease is not yet clear, their features are summarized in Table 2. The main pathophysiological differences appear to be an IL17A-driven inflammation in Kawasaki disease (KD) and a stronger endothel activation in coronary artery involvement in MIS-C. In both, MIS-C and KD autoantibodies may play an important role and MIS-C patients show distinct CD4 subset abnormalities. (Consiglio 2020).
|Table 2. Features of Kawasaki Disease and pediatric inflammatory multisystem syndrome temporarily associated with SARS-CoV-2|
|Kawasaki (Hedrich 2017, ECDC 2020) (previously called mucocutaneous lymph -node syndrome)||PIMS-TS (pediatric inflammatory multisystem syndrome temporarily associated with SARS-CoV-2 or MIS-C (multisystem inflammatory syndrome in children) (Verdoni 2020; Riphagen 2020, https://covid19-surveillance-report.ecdc.europa.eu/)
|Epidemiology||Incidence 5–19/100,000 annually < 5 years of age (EU, US), in north-east Asia higher; seasonal increase in winter/spring, geographic wave-like spread of illness during epidemics (Rowley 2018)||
|Age, sex||>90% < 5 years of age, more males||5-15 years of age, sex distribution unclear|
|Etiology||Unknown, hypothesis: infection with common pathogens, e.g. bacteria, fungi and viruses which cause immune-mediated damage (Dietz 2017) (Jordan-Villegas 2010, Kim 2012, Turnier 2015). Genetic factors (increased frequency in Asia and among family members of an index case)||
|fever ≥5 days, combined with at least 4 of the 5 following items
1.Bilateral bulbar conjunctival injection
2. Oral mucous membrane changes, including injected or fissured lips, injected pharynx, or strawberry tongue
3. Peripheral extremity changes, including erythema of palms or soles, edema of hands or feet (acute phase) or periungual desquamation (convalescent phase)
4. Polymorphous rash
5. Cervical lymphadenopathy
Children suspected of having KD who do not fulfill diagnostic criteria may have incomplete or atypical KD (Cimaz 2009)
|1. Persistent fever, inflammation (neutrophilia, elevated CRP and lymphopenia) and single or multi-organ dysfunction (shock, cardiac, respiratory, renal, gastrointestinal or neurological disorder) with other additional clinical, laboratory or imagining and ECG features. Children fulfilling full or partial criteria for Kawasaki Disease may be included
2. Exclusion of any other microbial cause, including bacterial sepsis, staphylococcal or streptococcal shock syndromes, infections associated with myocarditis such as enterovirus
3. SARS-CoV-2 PCR testing positive or negative (Royal College of Paediatrics and Child Health)
|CoV-2 status in most cases
|CoV-2 Ag (PCR); Abs (Elisa) negative||CoV-2 Ag (PCR) negative and Abs (Elisa) positive|
|Typical Lab||Marked Elevation of acute-phase reactants (eg, C-reactive protein [CRP] or erythrocyte sedimentation rate [ESR])
Thrombocytosis (generally after day 7 of illness
Leukocytosis, left-shift (increased immature neutrophils)
|Marked elevation of acute phase reactants CRP, ESR
Elevated myocarditis markers Troponin, pro-BNP
|Kawasaki disease shock syndrome (KSSS) (rare), features of macrophage activation syndrom, MAS (rare), coronary artery abnormalities, mitral regurgitation, prolonged myocardial dysfunction, disseminated intravascular coagulation (Kanegaye 2009)
Gastrointestinal complications (Ileitis, vomiting, abdominal pain) rare
|Shock (common), features of macrophage activation syndrome (common), myocardial involvement evidenced by markedly elevated cardiac enzymes (common), myocardial infarction, aneurysms, disseminated intravascular coagulation
Gastrointestinal complications (Ileitis, vomiting, abdominal pain) are very common
|Artery abnormalities (aneurysms of mid-sized arteries, giant coronary artery aneurysms CAAs)||Aneurysms|
|Management||High-dose intravenous immunoglobulin (IVIG) (2g/kg) first-line treatment; effective in reducing the risk of coronary artery disease when administered within 10 days of onset of fever. In addition, acetylsalicylic acid, glucocorticoids and anti-TNF monoclonal antibodies have been used
|So far, most patients published were treated with high dose IVIG, glucocorticoids, ASS (Verdoni 2020, Riphagen 2020, Ahmed 2020)
IVIG resistance requiring adjunctive steroid treatment is common (Verdoni 2020,)
Management on the pediatric intensive care unit is often necessary: progression to vasoplegic shock is common
Hemodynamic support, treatment with noradrenaline and milrinone, mechanical ventilation is often required (Riphagen 2020)
|Prognosis||Self-limited vasculitis lasting for an average of 12 days without therapy. Without timely treatment, CAAs, and in particular aneurysms, can occur in up to 25% of children||Overall prognosis not yet clear
More severe course than KD
Potentially fatal in individual cases
National guidelines and guidance documents have been published from different medical societies in China, North America, Italy, UK and Germany (https://rcpch.ac.uk; Venturini 2020, Chiotos 2020, Liu 2020; https://www.rcpch.ac.uk/key-topics/covid-19; https://dgpi.de/stellungnahme-medikamentoese-behandlung-von-kindern-mit-covid-19/)
Infection control in the medical setting
Early identification of COVID-19 and quarantine of contacts is imperative. In the in- and out-patient setting it is advised to separate children who have infectious diseases from healthy non-infectious children. Nosocomial outbreaks have played a role in the clustering of COVID-19. It is advised to admit children with COVID-19 to the hospital only if an experienced pediatrician feels it is medically necessary (e.g. tachypnea, dyspnea, oxygen levels below 92%). In the hospital the child with COVID-19 or suspicion of COVID-19 needs to be isolated in a single room or admitted to a COVID-19-only ward in which COVID-19-exposed medical personnel is protected by non-pharmacological interventions (wearing FFP-2 masks, gowns, etc.) and maintains distance and is cohorted themselves (e.g. no shifts on other wards). The presence of one parent is not negotiable in the care of the sick child both for emotional reasons as well as for help in the nursing of the child.
At present it is not recommended to separate healthy newborns from mothers with suspicion of COVID-19 (CDC-2 2020). Clearly, a preterm or newborn that has been exposed to CoV-2 needs to be closely monitored by the hospital and/or the primary care pediatrician. If there are signs of COVID (e.g. poor feeding, unstable temperature, tachy/dyspnea) it needs to be hospitalized and tested and lab examinations and chest x-ray to be done. Testing for CoV-2 is not useful before day 5 because of the incubation period. There needs to be strict hygiene as much as possible in this mother-child setting.
During peak phases of the COVID-19 pandemic, precautions in the outpatient and hospital setting include entrance control, strict hand and respiratory hygiene, daily cleaning and disinfection of the environment, and provision of protection (gloves, mask, goggles) for all medical staff when taking care of a COVID-19 or a suspected COVID-19 case (Wang 2020). In neonatal intensive care units (NICU), negative pressure rooms and filtering of exhaust would be ideal (Lu Q 2020). Respirators with closed circuit and filter systems should be used. Aerosol generating procedures, e.g. intubation, bronchoscopy, humidified inhalations/nebulization should be avoided as much as possible.
Infection control outside the medical setting
Some of the interventions to control the COVID pandemic have caused significant damage to children and adolescents. The description of their impact is beyond the scope of this artivcel and reviewed elsewhere.
Supportive treatment (respiratory support, bronchodilatation therapy, fever, superinfection, psychosocial support)
Having the child sitting in an upright position will be helpful for breathing. It might be useful to have physiotherapy. Insufflation of oxygen via nasal cannula will be important to children as it will increase lung ventilation and perfusion. In neonates, high flow nasal cannula (HFNC) has been utilized widely due to its superiority over other non-invasive respiratory support techniques.
The clinical use and safety of inhaling different substances is unclear in COVID-19. In other common obstructive and infectious childhood lung diseases, e.g. in bronchiolitis, the American Academy of Pediatrics is now recommending against the use of bronchodilators (Dunn 2020). Regarding the inhalation of steroids as part of maintenance therapy for asthma bronchiale there is no evidence to discontinue this treatment in children with COVID-19.
There is a large controversy over the extent of antipyretics usage in children. Still, in a child with COVID-19 who is clinically affected by high-degree fever, paracetamol or ibuprofen may be useful. There is no restriction despite initial WHO warnings of using ibuprofen, there is no evidence that the use of paracetamol or ibuprofen is harmful in COVID-19 in children (Day 2020).
The differentiation between CoV-2-induced viral pneumonia and bacterial superinfection is difficult unless there is clear evidence from culture results or typical radiological findings. Bacterial superinfection will be treated according the international and national guidelines (Mathur 2018).
The virus outbreak brings psychological stress to the parents and family as well as medical staff; therefore, social workers and psychologists should be involved when available.
Treatment of respiratory failure
The treatment of pediatric acute respiratory distress syndrome (pARDS) is reviewed elsewhere (Allareddy 2019). For neonates with pARDS high-dose pulmonary surfactant replacement, nitric oxide inhalation, and high-frequency oscillatory ventilation might be effective. In critically ill neonates, continuous renal replacement and extracorporeal membrane oxygenation need to be implemented if necessary.
COVID-19-specific drug treatment
As of yet there are no data from controlled clinical trials and thus there is currently no high-quality evidence available to support the use of any medication to treat COVID-19. The drugs listed below are repurposed drugs and there is limited or almost no pediatric experience. In the case of a severe or critically ill child with COVID the pediatrician has to make a decision whether to try a drug or not. If initiation of a drug treatment is decided, children should be included into clinical trials (https://www.clinicaltrialsregister.eu) if at all possible. However, there are very few, if any, studies open for recruitment in children.
When to treat with drugs
Under the lead of the German Society for Pediatric Infectiology (DGPI) an expert panel has proposed a consensus on when to start antiviral or immunomodulatory treatment in children (Table 3)).
A panel of pediatric infectious diseases physicians and pharmacists from North American institutions published an initial guidance on the use of antivirals for children with SARS. It is advised to limit antiviral therapy to children in whom the possibility for benefit outweighs the risk of toxicity and remdesvir is the preferred agent (Chiotos 2020).
Inhibitors of viral RNA synthesis
Remdesivir is available as 150 mg vials. Child dosing is
- < 40 kg: 5 mg/kg iv loading dose, then 2,5 mg/kg iv QD for 9 days
- ≥ 40 kg: 200 mg loading dose, then 100 mg QD for 9 days
Remdesivir is an adenosine nucleotide analogue with broad-spectrum antiviral activity against various RNA viruses. The compound undergoes a metabolic mechanism, activating nucleoside triphosphate metabolites for inhibiting viral RNA polymerases. Remdesivir has demonstrated in vitro and in vivo activity in animal models against MERS and SARS-CoV. Remdesivir showed good tolerability and a potential positive effect in regard to decrease of the viral load and mortality in Ebola in Congo in 2018 (Mulangu 2019). In Europe this drug has rarely been used in children so one should be extremely careful. It can be obtained through compassionate use programs (https://rdvcu.gilead.com).
|Table 3. Consensus on antiviral or immunomodulatory treatment in children|
|Disease severity in child||Intervention|
|Mild or moderate disease
pCAP, upper respiratory tract infection, no need for oxygen
No need for antiviral or immunomodulatory treatment
|More severe disease and risk groups*
pCAP, need for oxygen
Consider antiviral therapy
|Critically ill, admitted to ICU||Treat symptomatically
Consider antiviral therapy
Consider immunomodulatory treatment
|Secondary HLH (hemophagocytic lymphohistiocytosis)||Treat with immunomodulatory or immunosuppressive drugs|
* Congenital heart disease, immunosuppression, inborn/acquired immunodeficiencies, cystic fibrosis, chronic lung disease, chronic neurological/kidney/liver disease, diabetes/metabolic disease
Lopinavir/r (LPV/r, Kaletra®) is a co-formulation of lopinavir and ritonavir, in which ritonavir acts as a pharmacokinetic enhancer (booster). LPV/r is an HIV-1 protease inhibitor successfully used in HIV-infected children as part of highly active antiretroviral combination therapy (PENTA Group, 2015). In the SARS epidemics, LPV/r was recommended as a treatment. A recent study in adult COVID-19 patients did not show an effect regarding the primary end point in a controlled clinical trial. Despite the fact that there is long experience with LPV/r in HIV, it is not advised to use it in children with COVID-19 as it does not appear to be effective at all (see Treatment chapter, page 329
Inhibitors of viral entry
Hydroxychloroquine (HCQ, Quensyl®), Chloroquine (CQ, Resochin junior®, Resochin®) The experience among pediatricians with HCQ/CQ (except pediatriciancs working with malaria) is very limited. Authorities in the US are now warning about a widespread use of HCQ/CQ in COVID-19 (https://mailchi.mp/clintox/aact-acmt-aapcc-joint-statement). It is not advised to use HCQ or CQ in children with COVID as neither drug appears to be effective at all (see Treatment chapter, page 329).
Immunomodulatory drug treatment
The rational for immunomodulation in COVID-19 adult patients comes from a high expression of pro-inflammatory cytokines (Interleukin-1 (IL-1) and interleukin-6 (IL-6)), chemokines (“cytokine storm”) and the consumption of regulatory T cells resulting in damage of the lung tissue as reported in patients with a poor outcome. In children, the proinflammatory cytokines TNF and IL-6 do not appear to be central in CoV-2 induced hyperinflammation (Consiglio 2020). Blocking IL-1 or IL-6 can be successful in children with (auto) inflammatory disease (reviewed in Niehues 2019), but both interleukins are also key to the physiological immune response and severe side effects of immunomodulators have been reported. In adults with COVID-19, blocking interleukin-1/6 might be helpful (see the Treatment chapter). In the rare situation that the condition of the child deteriorates due to hyperinflammation and they are resistant to other therapies, anakinra may be an option as IL-1 seems to play a role in endothelial activation.
Steroids (e.g. prednisone, prednisolone) are available as oral solution, tablets or different vials for intravenous application. Dosage in children is 0,5 to 1 mg/kg iv or oral BID. Short term use of steroids has few adverse events. Administration of steroids will affect inflammation by inhibiting the transcription of some of the pro-inflammatory cytokines and various other effects. Initially, the use of corticosteroids in children and adults with CoV-induced ARDS was controversial (Lee 2004, Arabi 2018, Russell 2020). Only in severe and critically ill children the use of dexamethasone appears justified in children. In adults, the use of steroids in severe COVID-19 is clearly beneficial although the corticosteroid-induced decrease of antiviral immunity (e.g. to eliminate CoV-2 viruses) might be theoretically disadvantageous. Data supporting the use of steroids in children with CoV-induced ARDS are lacking. Only in severe and critically ill children might the use of dexamethasone appear justified.
Most patients with pediatric inflammatory multisystem syndrome associated with SARS-CoV-2 (PIMS-TS) published so far were treated with high dose IVIG and methylprednisolone (Verdoni 2020, Riphagen 2020). In these patients, features of macrophage activation syndrome and IVIG resistance were common, requiring adjunctive steroid treatment (Verdoni 2020). Clearly, any child severly affected by CoV-2 will need steroids at some stage.
Tocilizumab (Roactemra®) is available in 80/200/400 mg vials (20 mg/ml). Dosing is
- < 30 kg: 12 mg/kg iv QD, sometimes repeated after 8 hrs
- ≥ 30 kg: 8mg/kg iv QD iv (max. 800 mg)
Adverse events (deriving largely from long term use in chronic inflammatory diseases and use in combination with other immunomodulatory drugs): severe bacterial or opportunistic infections, immune dysregulation (anaphylactic reaction, fatal macrophage activation), psoriasis, vasculitis, pneumothorax, fatal pulmonary hypertension, heart failure, gastrointestinal bleeding, diverticulitis, gastrointestinal perforation (reviewed in Niehues 2019).
Anakinra (Kineret®) is available as 100 mg syringes (stored at 4-8° C). Dosing is 2-4 mg/kg sc QD daily as long as hyperinflammation persists. Thereafter, dose reduction by 10-30% per day. Higher dosage (> 4mg/kg-10mg/kg; max 400mg/d) may be necessary in patients with PIMS-TS. Adverse events (deriving largely from long-term use in chronic inflammatory diseases and use in combination with other immunomodulatory drugs): severe bacterial or opportunistic infections, fatal myocarditis, immune dysregulation, pneumonitis, colitis, hepatitis, endocrinopathies, nephritis, dermatitis, encephalitis, psoriasis, vitiligo, neutropenia (reviewed in Niehues 2019).
There are no systematic data on the use of convalescent plasma in children yet, but in a child with acute lymphoblastic leukemia and a young adult with a SCID (Severe Combined Immunodeficiency) phenotype and a high CoV-2 viral load, administration of convalescent plasma resulted in complete viral suppression (Shankar 2020, unpublished observation). Engineering monoclonal antibodies against the CoV spike proteins or against its receptor ACE2 or specific neutralizing antibodies against CoV-2 present in convalescent plasma may provide protection but are generally not available yet.
Interferon α has been inhaled by children with COVID-19 in the original cohorts but there are no data on its effect (Qiu 2020). Type I/III interferons (e.g. interferon α) are central to antiviral immunity. When coronaviruses (or other viruses) invade the host, viral nucleic acid activates interferon-regulating factors like IRF3 and IRF7 which promote the synthesis of type I interferons (IFNs).
PIMS / MIS-C
Based on the information published so far, most patients were treated with high dose intravenous Immunglobulin (see Table 2) and corticosteroids (Verdoni 2020). More data are needed to determine the optimal treatment strategies for patients with MIS-C.
 Recommendations regarding attendance to kindergartens and schools have been published (Cohen 2020).
Acknowledgements: Without the skillful help of Andrea Groth (Helios Klinikum Krefeld), the preparation of this manuscript would not have been possible. We thank cand. med. Lars Dinkelbach (Heinrich Heine Universität Düsseldorf) for critically reading the manuscript.
By Tim Niehues
& Jennifer Neubert