Neurologic Involvement in Children and Adolescents Hospitalized in the United States for COVID-19 or Multisystem Inflammatory Syndrome

Key Points

Question  What is the extent of neurologic involvement in US hospitalized children and adolescents with coronavirus disease 2019 (COVID-19)?

Findings  In this study of 1695 patients 21 years or younger hospitalized for acute COVID-19 or multisystem inflammatory syndrome, 365 (22%) had neurologic involvement. Forty-three patients (12%) developed COVID-19–related life-threatening neurologic disorders, 11 (26%) died, and 17 (40%) survived with new neurologic sequelae.

Meaning  In this study, COVID-19–related neurologic involvement was common in hospitalized children and adolescents and mostly transient.

Importance  Coronavirus disease 2019 (COVID-19) affects the nervous system in adult patients. The spectrum of neurologic involvement in children and adolescents is unclear.

Objective  To understand the range and severity of neurologic involvement among children and adolescents associated with COVID-19.

Setting, Design, and Participants  Case series of patients (age <21 years) hospitalized between March 15, 2020, and December 15, 2020, with positive severe acute respiratory syndrome coronavirus 2 test result (reverse transcriptase-polymerase chain reaction and/or antibody) at 61 US hospitals in the Overcoming COVID-19 public health registry, including 616 (36%) meeting criteria for multisystem inflammatory syndrome in children. Patients with neurologic involvement had acute neurologic signs, symptoms, or diseases on presentation or during hospitalization. Life-threatening involvement was adjudicated by experts based on clinical and/or neuroradiologic features.

Exposures  Severe acute respiratory syndrome coronavirus 2.

Main Outcomes and Measures  Type and severity of neurologic involvement, laboratory and imaging data, and outcomes (death or survival with new neurologic deficits) at hospital discharge.

Results  Of 1695 patients (909 [54%] male; median [interquartile range] age, 9.1 [2.4-15.3] years), 365 (22%) from 52 sites had documented neurologic involvement. Patients with neurologic involvement were more likely to have underlying neurologic disorders (81 of 365 [22%]) compared with those without (113 of 1330 [8%]), but a similar number were previously healthy (195 [53%] vs 723 [54%]) and met criteria for multisystem inflammatory syndrome in children (126 [35%] vs 490 [37%]). Among those with neurologic involvement, 322 (88%) had transient symptoms and survived, and 43 (12%) developed life-threatening conditions clinically adjudicated to be associated with COVID-19, including severe encephalopathy (n = 15; 5 with splenial lesions), stroke (n = 12), central nervous system infection/demyelination (n = 8), Guillain-Barré syndrome/variants (n = 4), and acute fulminant cerebral edema (n = 4). Compared with those without life-threatening conditions (n = 322), those with life-threatening neurologic conditions had higher neutrophil-to-lymphocyte ratios (median, 12.2 vs 4.4) and higher reported frequency of D-dimer greater than 3 μg/mL fibrinogen equivalent units (21 [49%] vs 72 [22%]). Of 43 patients who developed COVID-19–related life-threatening neurologic involvement, 17 survivors (40%) had new neurologic deficits at hospital discharge, and 11 patients (26%) died.

Conclusions and Relevance  In this study, many children and adolescents hospitalized for COVID-19 or multisystem inflammatory syndrome in children had neurologic involvement, mostly transient symptoms. A range of life-threatening and fatal neurologic conditions associated with COVID-19 infrequently occurred. Effects on long-term neurodevelopmental outcomes are unknown.

Coronaviruses primarily cause respiratory disease; however, severe acute respiratory syndrome coronavirus (SARS-CoV), Middle East respiratory syndrome coronavirus, and newly identified SARS-CoV-2 have been associated with a range of disorders of the peripheral and central nervous system (CNS).^1-9 Early reports from Wuhan, China, described a spectrum of neurologic conditions associated with SARS-CoV-2 infection in 36% of 214 adults hospitalized with coronavirus disease 2019 (COVID-19).¹⁰ Reported neurologic and psychiatric symptoms in adult patients with COVID-19 include anosmia/ageusia,^11,12 headaches,¹³ dizziness/ataxia,¹⁰ psychosis, dementia, depression, anxiety, and mania.¹⁴Quiz Ref ID Reported severe neurologic involvement in adult patients with COVID-19 includes acute encephalopathy or encephalitis,^15-18 acute necrotizing encephalopathy,^19,20 epilepsy/seizures,^21,22 acute transverse myelitis,^23-25 Guillain-Barré syndrome (GBS),^26,27 posterior reversible encephalopathy syndrome,²⁸ and acute ischemic or hemorrhagic stroke.^14,29-34

Although most children and adolescents are spared from severe COVID-19, there have been reports of life-threatening neurologic involvement in patients developing multisystem inflammatory syndrome in children (MIS-C), a relatively rare, hyperinflammatory, severe illness temporally associated with SARS-CoV-2 infection, presumably postinfectious.^35,36 Across case series published between March and August 14, 2020, between 6% and 58% of children and adolescents hospitalized with MIS-C developed central and/or peripheral nervous system involvement.^36-40 The frequency of neurologic involvement in children hospitalized with acute COVID-19 is unclear with 150 of 4190 patients reported across 9 international case series.^41-46 Using the Overcoming COVID-19 US public health surveillance registry of children and adolescents hospitalized with COVID-19–related complications,³⁶ we aimed to describe the type and severity of neurologic involvement and documented hospital outcomes.

Active surveillance was performed at 61 hospitals in 31 states in the Overcoming COVID-19 network to identify children and adolescents (age <21 years) with SARS-CoV-2–related illness hospitalized from March 15, 2020, to December 15, 2020. The study was approved by the central institutional review board at Boston Children’s Hospital and determined to meet the requirement of public health surveillance as defined in 45 CFR 46.102(I)(2) at Boston Children’s Hospital and the US Centers for Disease Control and Prevention under a waiver of consent.

Quiz Ref IDPatients were included if they were hospitalized for acute illness at a participating site, were younger than 21 years, had a positive SARS-CoV-2 test result (reverse transcriptase–polymerase chain reaction and/or antibody) and symptoms associated with acute COVID-19, or met US Centers for Disease Control and Prevention criteria for MIS-C (eTable 1 in Supplement 1).⁴⁷ Patients were excluded if they had asymptomatic SARS-CoV-2 infection or a non–COVID-19–related cause for hospitalization or death. Race and ethnicity were extracted from the patient’s medical record and included to evaluate risk of neurologic involvement.

Patients were stratified by the presence of neurologic involvement, defined as (1) suspected acute neurologic disease (eg, CNS infection/demyelination or stroke) on presentation or that developed during hospitalization (eMethods in Supplement 1) or (2) acute neurologic signs or symptoms on presentation.

Severity of neurologic involvement was adjudicated by neurology and critical care experts on the central study team (K.L.L., B.J.R., T.Y.P., and A.G.R.; eMethods in Supplement 1). Cases were classified as life-threatening based on clinical and/or neuroradiologic features associated with more severe outcomes and included the following diagnoses: acute CNS infection (aseptic meningitis, encephalitis by International Encephalitis Consortium definition,⁴⁸ and Brighton criteria⁴⁹), central demyelinating disorder (acute disseminated encephalomyelitis [ADEM]), acute ischemic or hemorrhagic stroke, GBS and variants, or severe encephalopathy with or without COVID-19–related neuroimaging abnormalities (eg, virus-associated necrotizing disseminated acute leukoencephalopathy⁵⁰ and/or cytotoxic splenial lesions^44,51,52). Cases with neurologic involvement that did not meet any of these criteria and had cerebrospinal fluid and/or neuroimaging results that were normal or not performed were categorized as non–life-threatening neurologic involvement.

Neurology and critical care experts (K.L.L., B.J.R., and A.G.R.) determined through case review and consensus whether life-threatening neurologic conditions were directly associated with COVID-19 or secondary to exacerbation of primary neurologic disease or complication of critical illness associated with COVID-19. Sites with abnormal neuroimaging studies sent deidentified brain magnetic resonance imaging (MRI) and computed tomography studies for central review. Images were reviewed by a pediatric neuroradiologist (T.Y.P.) and discussed with a pediatric neurologist (K.L.L.) reaching consensus opinion about whether the clinicoradiologic link was directly associated with COVID-19 or secondary to an alternate etiology (eg, extracorporeal membrane oxygenation [ECMO] or preexisting neurologic condition).

Outcomes were determined at hospital discharge. Neurologic deficits were defined as gross impairment in motor, cognitive, or speech and language functions. Psychiatric sequelae (eg, anxiety, depression, and/or suicidal ideation) were not included. New neurologic deficits were determined by medical record review at each site and adjudicated by the experts for all patients with and without neurologic involvement (eMethods in Supplement 1).

We report the frequency of clinical characteristics, underlying conditions, type of neurologic involvement on admission or during hospitalization, and hospital outcomes. Continuous variables were expressed as medians and interquartile range. Categorical variables were expressed as counts and percentages. Between-group differences were analyzed using a χ² test, Fisher exact test, or Kruskal-Wallis test where appropriate. Two-sided P values less than .05 were considered statistically significant. We did not impute missing data. We analyzed all data using R software, version 3.6.1 (R Project for Statistical Computing).

From March 15, 2020, to December 15, 2020, a total of 1784 hospitalized children and adolescents with COVID-19–related illness were reported to the registry. Of these, 89 patients were excluded on the basis of being 21 years or older (n = 27), epidemiologic link to SARS-CoV-2 without a positive test result (n = 54), and non–COVID-19–related cause for hospitalization or death (n = 8) (eFigure in Supplement 1). We describe 1695 patients (909 male [54%]; median [interquartile range] age, 9.1 [2.4-15.3] years) from 61 sites in 31 states (Table 1). Most patients were either Hispanic or Latino (638 of 1695 [38%]) or non-Hispanic Black (442 of 1695 [26%]).

There were 365 patients (22%) with neurologic involvement reported from 52 sites in 29 states. The characteristics of the patients with and without neurologic involvement are shown in Table 1. The frequencies of previously healthy patients (195 [53%] vs 723 [54%]) and patients meeting MIS-C criteria (126 [35%] vs 490 [37%]) were similar. Patients with neurologic involvement were more likely to have underlying neurologic disorders (81 [22%]) compared with those without (113 [8%]), including seizure disorders, neuromuscular disorders, and autism or developmental delay. Quiz Ref IDPresenting neurologic signs and symptoms differed by age with seizures or status epilepticus most common in children younger than 5 years and anosmia and/or ageusia most common in patients between ages 13 and 20 years (Figure 1A).

Most patients with and without neurologic involvement were discharged alive (351 [96%] and 1322 [99%], respectively). Children with neurologic involvement had a higher rate of survival with new neurologic deficits (20 of 365 [5%]) compared with those without COVID-19–associated neurologic involvement (2 of 1330 [0.2%]) (Table 1). Neurologic deficits in those without neurologic involvement included cognitive and motor impairments as a result of sequelae of critical illness and intensive care therapies.

Among 365 patients with neurologic involvement, 43 (12%) had life-threatening neurologic involvement associated with COVID-19 (Table 2). Among these, 34 of 43 (79%) had no major underlying conditions, 20 (47%) met criteria for MIS-C, and 3 (7%) had a preexisting neurologic disorder. Quiz Ref IDLife-threatening neurologic conditions included severe encephalopathy (n = 15; 5 with white-matter hyperintensities and splenial lesions), acute ischemic or hemorrhagic stroke (n = 12), acute CNS infection/ADEM (n = 8), acute fulminant cerebral edema (n = 4), and GBS (n = 4) (Table 2; eTable 2 in Supplement 1). Eight patients with stroke had underlying risk factors (5 experienced stroke during ECMO [eTable 3 in Supplement 1]; 2 were attributed to possible COVID-19–related exacerbation of an underlying primary neurologic disorder [eg, arteriovenous malformation rupture and ischemic stroke in a patient with history of moyamoya syndrome]; and a previously healthy patient presented with a new diagnosis of acute myelogenous leukemia). Four patients were previously healthy and did not have stroke risk factors (eTable 4 in Supplement 1). Five children with severe encephalopathy had brain MRI findings of diffuse white-matter hyperintensities on T2-weighted images and restricted diffusion in the periventricular white matter, deep white matter, and/or corpus callosum (60% with MIS-C; 3 of 5 with unfavorable neurologic outcomes). Representative CNS images from patients with life-threatening neurologic involvement associated with COVID-19 are shown in Figure 2 and Figure 3.

Compared with those with non–life-threatening neurologic involvement, children with life-threatening neurologic disease were more likely to undergo lumbar puncture (20 of 43 [47%] vs 72 of 322 [22%]), head computed tomography (23 of 43 [53%] vs 40 of 322 [12%]) or brain MRI (26 of 43 [60%] vs 28 of 322 [9%]). The cerebrospinal fluid results showed unremarkable findings in both groups (eTable 5 in Supplement 1). As shown in Figure 1B, patients with life-threatening neurologic conditions were more inflamed and coagulopathic than those with no or non–life-threatening neurologic involvement. Patients with life-threatening vs non–life-threatening neurologic involvement had higher neutrophil-to-lymphocyte ratios (median, 12.2 vs 4.4), and higher reported frequency of D-dimer >3 μg/mL fibrinogen equivalent units (21 [49%] vs 72 [22%]; to convert D-dimer to nanomoles per liter, multiply by 5.476; Figure 1; eTable 6 in Supplement 1).

In patients who developed life-threatening neurologic involvement, 11 (26%) died and 17 (40%) were discharged from hospital with new neurologic deficits (Table 2). Of survivors with new deficits, 16 (94%) were previously healthy, none had prior neurologic disorders, 7 (41%) met MIS-C criteria, and 14 (82%) required rehabilitative services on discharge (eTable 2 in Supplement 1).

Fourteen patients with COVID-19 neurologic involvement died in the hospital. Three deaths were associated with acute COVID-19 cardiorespiratory disease. Two patients with asthma had cardiac arrest on hospital presentation, and 1 previously healthy teenager with anosmia/ageusia died of multiorgan failure. These patients were excluded from further evaluation. The other 11 deaths were classified by expert consensus as either directly associated with COVID-19 neurologic involvement or with catastrophic neurologic events secondary to COVID-19–related critical illness (Table 2). These cases are briefly summarized below (eTable 2 in Supplement 1).

Three previously healthy children with acute fulminant cerebral edema died within 48 hours of hospital admission (eTable 7 in Supplement 1). One male infant with COVID-19 presented with fever, seizures, and gastrointestinal symptoms and within 24 hours of hospitalization developed status epilepticus and had a cardiac arrest, with subsequent imaging showing global cerebral edema. One elementary school–aged girl presented with fever and sore throat, then developed status epilepticus with subsequent imaging revealing cerebral edema with tonsillar herniation. One elementary school–aged boy met criteria for MIS-C 1 month after a positive SARS-CoV-2 respiratory test result. He developed status epilepticus shortly after hospital admission and imaging showed global cerebral edema and uncal herniation.

Four patients with stroke died. Three of these patients had strokes with malignant edema and examinations consistent with brain death on ECMO. The fourth died of multiple ischemic strokes owing to rapidly progressive large-vessel CNS vasculitis despite intensive immunotherapies.

Four patients who developed severe encephalopathy died. One immunocompromised adolescent with leukemia and acute COVID-19 pneumonia had diffuse T2 prolongation and reduced diffusivity in the bilateral periventricular white matter, with involvement of the splenium and genu of the corpus callosum, who also developed acute motor-sensory axonal neuropathy confirmed by electromyography/nerve conduction study. Two other patients who died required intubation for severe encephalopathy, complicated by cardiovascular collapse with cannulation for venoarterial ECMO and progression to brain death. One teenager with obesity who died had preexisting hypertension and diabetes and received venoarterial ECMO for cardiorespiratory failure. A brain MRI on decannulation obtained for prolonged encephalopathy showed multifocal areas of restricted diffusion and hemorrhage throughout the posterior white matter and brainstem.

In a large, multicenter case series of US children and adolescents hospitalized with acute COVID-19 or MIS-C, 22% of reported patients had neurologic involvement. Quiz Ref IDApproximately half of patients with and without neurologic involvement were previously healthy, a similar percentage had MIS-C, but more patients with neurologic involvement had underlying neurologic disorders (22% vs 8%). Neurologic involvement in most patients was transient and resolved by hospital discharge; however, 43 patients (12%) developed a range of life-threatening neurologic conditions associated with COVID-19, and 66% of these patients had unfavorable outcomes, including death or new neurologic disability at hospital discharge.

The range of neurologic symptoms associated with COVID-19 in children and adolescents was broad and varied by age including seizures/status epilepticus in the younger patients and reports of anosmia and/or ageusia, headache, and fatigue/weakness in older patients. Approximately 1 in 4 patients with neurologic involvement across age groups presented with altered awareness or confusion. The range of severe neurologic complications including peripheral nerve disorders (GBS and variants), focal CNS disease (ischemic stroke due to large vessel occlusion, cerebral venous sinus thrombosis, and focal cerebral arteriopathy), and diffuse CNS involvement (CNS infection, ADEM, severe encephalopathy with white matter and corpus callosum lesions, and acute fulminant cerebral edema) make it likely that multiple mechanisms underlie this wide spectrum of disease. These include putative mechanisms such as neuroinvasive or neurotropic (direct viral entry and/or neuronal infection via angiotensin-converting enzyme 2^53,54 and/or olfactory tract^55,56), neuroinflammatory (exaggerated cytokine/immune mediated response leading to blood brain barrier breakdown^57,58), postinfectious immune dysregulation,^59,60 and/or as secondary injury from complications of systemic inflammation or other non-CNS organ failure.⁶¹

We observed 4 cases of GBS that presented with classic neurologic signs, symptoms, and electrophysiologic features within 1 month following SARS-CoV-2 exposure, similar to reports in adults and children in association with COVID-19,^26,27 and 1 case of acute motor-sensory axonal neuropathy. Animal models and clinicopathological evidence support an autoimmune mechanism and potential molecular mimicry between antibodies against myelin and gangliosides in the nervous system and recent infectious agents, now including COVID-19 infection,^62,63 and suggests a potential role for antiganglioside antibodies in immunomodulatory therapies.

We report 12 cases of acute ischemic or hemorrhagic stroke, with 8 having underlying stroke risk factors. Five of these cases occurred while receiving ECMO, but were included because COVID-19 may have exacerbated an underlying pathophysiologic state (eg, hypercoagulability, hyperinflammation, increased risk of bleeding, and endothelial dysfunction) predisposing to stroke while receiving ECMO.^64,65 The 4 cases without stroke risk factors were directly associated with COVID-19. In the pediatric literature consisting of case reports and 2 international case series, ischemic stroke type has been reported in 18 children with COVID-19 with stroke mechanisms similar to those observed in our study.^44,66-69 Acute ischemic stroke in hospitalized adults with COVID-19 is not uncommon.^8,33,70

We also describe global cerebral involvement in 15 patients (8 with MIS-C) with severe encephalopathy, 8 patients with acute CNS infection (encephalitis, aseptic meningitis) or postinfectious, central demyelination (ADEM), and 4 patients with acute fulminant cerebral edema. We identified 5 previously healthy patients who presented with severe encephalopathy, focal neurologic deficits, and visual hallucinations (4 of 5 cases) and had diffuse abnormal T2 hyperintensities and reduced diffusivity involving the white matter and genu or splenium of the corpus callosum on MRI. These imaging features have been ascribed to COVID-19 in adults^50,71 and in children with MIS-C.^44,51,52 Cytotoxic lesions in the corpus callosum are thought to be associated with increased numbers of glutamate and cytokine receptors in the corpus callosum, particularly the splenium.^44,72,73 Similar to the range of outcomes in 1 small adult case series,⁵⁰ 3 patients had unfavorable outcomes (1 died and 2 were discharged with new deficits including cognitive impairment and painful neuropathy requiring gabapentin). Of those with acute CNS infections/ADEM, 7 patients in our study could be confirmed as having probable acute CNS infection using published case definitions.^48,49 Case reports and small case series also support a link between meningoencephalitis and COVID-19 in adults^74-76 and children.^44,77-79

There were 4 cases of previously healthy children who developed acute fulminant cerebral edema directly associated with COVID-19 or MIS-C, and 3 died. Acute fulminant cerebral edema has been previously reported in a child with COVID-19⁸⁰ and is a recognized phenotype with high mortality in adults^81,82 and children^44,83 associated with other viral causes.

Our study has several strengths. There was expert adjudication of cases with fatal and life-threatening neurological involvement and new neurologic deficits by pediatric neurology, pediatric critical care, and pediatric neuroradiology experts. The central study team also had personal communication with site clinicians contributing cases with fatal and life-threatening neurologic involvement or new neurologic deficits to confirm diagnoses and clinical course. Neuroimaging was associated with clinical information to document supportive imaging findings and confirm diagnoses. We also captured patients across most US states from a large number of pediatric centers.

The study has certain limitations. First, cases of COVID-19–related neurologic involvement were identified only at reporting hospitals and may not accurately reflect the true range and severity of COVID-19 neurologic involvement. Second, in patients with underlying neurologic diseases, neurologic presentations may be owing to COVID-19 neurologic effects or exacerbation of underlying neurologic conditions. Third, not all patients underwent neuroimaging (possibly owing to infection control concerns or critical illness-related instability) and image acquisition was not standardized, which could result in misclassification or an underestimation of neurologic involvement. Fourth, although standardized case report forms were used, we may not have captured certain variables completely, such as the indications for procedures (eg, lumbar puncture and imaging). Fifth, some neurologic symptoms (eg, anosmia or ageusia) may be underreported in very young patients. Sixth, nonstandardized diagnostic workups performed under routine clinical conditions may have missed non–COVID-19–related causes of life-threatening neurological conditions attributed to COVID-19. Seventh, standardized and validated assessments of neurologic outcomes at or after hospital discharge were not performed, likely underestimating the nature and extent of neurologic sequelae. Eighth, this is not a prospective cohort study but a case series, and caution is warranted in interpreting these data to identify risk factors for neurologic involvement.

In this study, neurologic involvement was common in children and adolescents with COVID-19–related hospitalization and is mostly transient. A spectrum of life-threatening neurologic involvement infrequently occurred and was associated with more extreme inflammation and severe sequelae. Future immunologic studies of cell-mediated and cytokine immune responses in young individuals may provide insight into the pathogenesis of neurologic disease in COVID-19 and MIS-C.⁸⁴ Patients with less severe neurologic involvement could have future sequelae. Long-term follow-up of pediatric patients with COVID-19–related neurologic involvement is needed to evaluate effects on cognition and development.

Corresponding Author: Adrienne G. Randolph, MD, Boston Children’s Hospital, 300 Longwood Ave, Bader 634, Boston, MA 02115 (adrienne.randolph@childrens.harvard.edu).

Accepted for Publication: February 12, 2021.

Published Online: March 5, 2021. doi:10.1001/jamaneurol.2021.0504

Author Contributions: Dr Randolph had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Drs LaRovere, Riggs, and Poussaint contributed equally. Drs Patel and Randolph contributed equally.

Concept and design: LaRovere, Walker, Hobbs, Feldstein, Doymaz, Spinella, Patel, Randolph.

Acquisition, analysis, or interpretation of data: LaRovere, Riggs, Poussaint, Young, Newhams, Maamari, Singh, Dapul, Hobbs, McLaughlin, Son, Maddux, Clouser, Rowan, McGuire, Fitzgerald, Gertz, Shein, Coronado Munoz, Thomas, Irby, Levy, Staat, Tenforde, Halasa, Giuliano, Hall, Kong, Carroll, Schuster, Loftis, Tarquinio, Babbitt, Nofziger, Kleinman, Keenaghan, Cvijanovich, Spinella, Hume, Wellnitz, Mack, Michelson, Flori, Patel, Randolph.

Drafting of the manuscript: LaRovere, Riggs, Poussaint, Hobbs, Kong, Spinella, Randolph.

Critical revision of the manuscript for important intellectual content: All authors.

Statistical analysis: LaRovere, Giuliano, Randolph.

Obtained funding: Randolph.

Administrative, technical, or material support: Poussaint, Newhams, Maamari, Singh, Hobbs, McLaughlin, Son, Rowan, Fitzgerald, Gertz, Shein, Coronado Munoz, Levy, Tenforde, Feldstein, Hall, Kong, Carroll, Schuster, Doymaz, Loftis, Babbitt, Nofziger, Cvijanovich, Spinella, Hume, Wellnitz, Michelson, Flori, Patel, Randolph.

Supervision: Newhams, Walker, Singh, Hobbs, McLaughlin, Thomas, Irby, Staat, Kong, Schuster, Loftis, Nofziger, Cvijanovich, Spinella, Hume, Patel, Randolph.

Conflict of Interest Disclosures: Dr Riggs reported grants from the US Centers for Disease Control and Prevention (CDC) funding through Boston Children's Hospital during the conduct of the study. Dr Newhams reported grants from the CDC during the conduct of the study. Dr Maamari reported other support from the CDC during the conduct of the study. Dr McLaughlin reported grants from Boston Children's Hospital and the CDC during the conduct of the study. Dr Maddux reported grants from the National Institutes of Health (NIH)/Eunice Kennedy Shriver National Institute of Child Health and Human Development during the conduct of the study. Dr Rowan reported grants from the CDC during the conduct of the study and from the NIH outside the submitted work. Dr McGuire reported grants from the CDC during the conduct of the study. Dr Fitzgerald reported grants from the CDC during the conduct of the study and from the NIH outside the submitted work. Dr Gertz reported grants from Boston Children’s Hospital as a passthrough for the CDC during the conduct of the study. Dr Shein reported grants from the CDC during the conduct of the study. Dr Coronado Munoz reported grants from the CDC during the conduct of the study. Dr Levy reported grants from the CDC during the conduct of the study and from the National Institute of Allergy and Infectious Diseases outside the submitted work. Dr Staat reported other support from Boston Children's Hospital during the conduct of the study. Dr Halasa reported grants from the CDC during the conduct of the study; grants from Sanofi and Quidel; and personal fees from Genentech outside the submitted work. Dr Hall reported grants from the CDC during the conduct of the study and personal fees from LaJolla Pharmaceuticals outside the submitted work. Dr Schuster reported other from the CDC during the conduct of the study; other support from Merck; and grants from the CDC outside the submitted work. Dr Doymaz reported grants from the CDC during the conduct of the study. Dr Tarquinio reported grants from the CDC during the conduct of the study. Dr Nofziger reported other from the CDC during the conduct of the study. Dr Kleinman reported grants from Boston Children's Hospital during the conduct of the study and grants from Health Services Research Administration and NICHD outside the submitted work. Dr Cvijanovich reported grants from the CDC during the conduct of the study and grants from Cincinnati Children’s Medical Center and Boston Children's Hospital outside the submitted work. Dr Hume reported grants from the CDC during the conduct of the study. Dr Wellnitz reported other support from the CDC and NIH during the conduct of the study. Dr Michelson reported grants from the CDC during the conduct of the study and grants National Palliative Care Research Center and the National Institutes of Health outside the submitted work. Dr Randolph reported grants from the CDC during the conduct of the study and other support from UpToDate outside the submitted work. Dr Poussaint reported receiving grants from the National Institutes of Health and royalties from Springer Publishing outside of the submitted work. No other disclosures were reported.

Funding/Support: This study was funded by the US Centers for Disease Control and Prevention under a contract to Boston Children’s Hospital.

Role of the Funder/Sponsor: The US Centers for Disease Control and Prevention designed and conducted the study; collected, managed, analyzed, and interpreted the data; prepared, reviewed, and approved the manuscript; had a role in the decision to submit the manuscript for publication and journal choice; and had the right to veto publication.

Disclaimer: The findings and conclusions in this report are those of the authors and do not necessarily represent the official position of the US Centers for Disease Control and Prevention.

Group Information: The Overcoming COVID-19 Investigators are listed in Supplement 2.

Additional Contributions: We appreciate and thank the many research coordinators at the Overcoming COVID-19 hospitals who assisted in data collection for this study. We thank the leadership of the Pediatric Acute Lung Injury and Sepsis Investigator’s (PALISI) Network for their ongoing support.