Coronavirus disease 2019 (COVID-19), caused by severe acute respiratory syndrome coronavirus 2, disproportionally affects adults (children <5% in most reports).1 Adult critical illness is characterized by acute hypoxemia, multiorgan failure, and high mortality.2,3 Reported risk factors for severe illness include age, cardiorespiratory comorbidities, obesity, and laboratory findings (lymphopenia and elevated D-dimer).2,4 Pediatric reports describe low infection rates and infrequent PICU admission.5,6 The largest PICU report consists of 48 North American children.7 It describes treatments and outcomes but not with adequate granularity to understand critical pediatric COVID-19. The Critical Coronavirus and Kids Epidemiology Study was designed to specifically investigate severe cases and provide detailed data. It involves >60 centers in nearly 20 countries from the Americas and Europe. In this report, we provide preliminary insights into our first 17 patients.
Methods
The Critical Coronavirus and Kids Epidemiology is a cohort study of children <19 years old with severe or critical COVID-19. The study period runs from April through December 2020. For this report, we included patients enrolled through April 23.
We defined critical COVID-19 as a positive severe acute respiratory syndrome coronavirus 2 test result and requiring ICU therapies (high-flow nasal cannula [HFNC], noninvasive ventilation [NIV], invasive mechanical ventilation [IMV], vasoactive support, continuous renal replacement therapy). Severe COVID-19 included those receiving mask or nasal oxygen exceeding the pediatric acute respiratory distress syndrome (ARDS) “at risk” threshold.8
Deidentified data were collected by using a modification of the International Severe Acute Respiratory and Emerging Infection Consortium form (https://isaric.tghn.org/COVID-19-CRF/). Local ethics approval was obtained with a waiver of need for consent.
Results
We enrolled 17 children from 10 PICUs in Chile, Colombia, Italy, Spain, and the United States. Detailed data are in the Supplemental Information. Most patients were male (65%), young (median 4 years; range 0.08–18 years), and without known COVID-19 exposure (14 of 17). Comorbidities (Table 1, Supplemental Table 3) were common (71%) but variable. Symptoms were heterogenous, with fever and cough being most frequent (Table 1, Supplemental Table 3). Most with gastrointestinal (GI) symptoms (4 of 6) were also diagnosed with myocarditis (Supplemental Table 4). All these were from Europe and without previous cardiovascular disease.
Demographics, Presenting Symptoms, and Selected Laboratory Findings
| Characteristic . | Result . |
|---|---|
| Days of symptoms preadmission, median (IQR) | 3.5 (2–5.8) |
| Days of symptoms before positive test, median (IQR) | 3.5 (2–6.8) |
| Comorbiditiesa | |
| None | 5 (29) |
| Respiratory | 1 (6) |
| Cardiac | 2 (12) |
| Cancer and/or immune | 2 (12) |
| Obesity | 2 (12) |
| Otherb | 8 (47) |
| Symptoms at admissiona | |
| Fever | 13 (76) |
| Cough | 9 (53) |
| Dyspnea | 6 (35) |
| Congestion | 6 (35) |
| GI | 6 (35) |
| Other | 5 (29) |
| Laboratory value on admission | |
| Leukocytosis, WBC count >11 000 per μL | 9 (53) |
| Elevated D-dimer >0.5 mg/μL | 7 (41) |
| Procalcitonin >2 ng/mL, at admission | 6 (35) |
| C-reactive protein >2 mg/L, at admission | 13 (76) |
| Laboratory value ever during hospitalization | |
| Leukocytosis | 12 (71) |
| Lymphopenia <1000 per μL | 8 (47) |
| Elevated D-dimer | 9 (53) |
| Ferritin >200 ng/mL | 7 (41) |
| Troponin I >1 ng/mL | 4 (25) |
| Characteristic . | Result . |
|---|---|
| Days of symptoms preadmission, median (IQR) | 3.5 (2–5.8) |
| Days of symptoms before positive test, median (IQR) | 3.5 (2–6.8) |
| Comorbiditiesa | |
| None | 5 (29) |
| Respiratory | 1 (6) |
| Cardiac | 2 (12) |
| Cancer and/or immune | 2 (12) |
| Obesity | 2 (12) |
| Otherb | 8 (47) |
| Symptoms at admissiona | |
| Fever | 13 (76) |
| Cough | 9 (53) |
| Dyspnea | 6 (35) |
| Congestion | 6 (35) |
| GI | 6 (35) |
| Other | 5 (29) |
| Laboratory value on admission | |
| Leukocytosis, WBC count >11 000 per μL | 9 (53) |
| Elevated D-dimer >0.5 mg/μL | 7 (41) |
| Procalcitonin >2 ng/mL, at admission | 6 (35) |
| C-reactive protein >2 mg/L, at admission | 13 (76) |
| Laboratory value ever during hospitalization | |
| Leukocytosis | 12 (71) |
| Lymphopenia <1000 per μL | 8 (47) |
| Elevated D-dimer | 9 (53) |
| Ferritin >200 ng/mL | 7 (41) |
| Troponin I >1 ng/mL | 4 (25) |
Results are presented as n (%) unless otherwise noted. IQR, interquartile range; WBC, white blood cell.
Total adds up to >100% because some had >1 comorbidity or symptom.
Includes chronic GI disorders (3), chronic neurologic disorders (2), prematurity (1), trisomy 21 (1), and tracheomalacia (1).
Patients had frequent laboratory testing (Table 1, Supplemental Table 5). Common findings included leukocytosis, lymphopenia, elevated inflammatory markers, D-dimer, and troponin I. Four had viral or bacterial respiratory coinfection.
Most subjects required respiratory support (Table 2, Supplemental Table 6), with nearly half requiring IMV. Five initially treated with HFNC needed no escalation; 2 were intubated. One initially treated with NIV was intubated. Pulmonary-specific adjuncts were uncommon. Most patients received antibiotics; fewer received antiviral agents (Table 2, Supplemental Table 6). Corticosteroids, hydroxychloroquine, and tocilizumab were each prescribed to nearly half. Intravenous immunoglobulin (IVIg) was prescribed exclusively for myocarditis.
ICU Therapies and Medications
| Treatment . | Result . |
|---|---|
| Respiratory supporta | |
| None | 3 (18) |
| HFNC | 7 (41) |
| NIV | 4 (24) |
| IMV | 8 (47) |
| Vasoactive infusion | 9 (53) |
| Respiratory adjunctsb | 1 (6) |
| Medications | |
| Antibiotics | 15 (88) |
| Remdesivir | 4 (24) |
| Lopinavir and/or ritonavir | 1 (6) |
| Corticosteroids | 9 (53) |
| Tocilizumab | 7 (41) |
| Hydroxychloroquine | 8 (47) |
| Diagnosis and/or complication | |
| Pneumonia | 13 (76) |
| ARDSc | 8; 2 mild, 1 moderate, 3 severe (47) |
| Myocarditis | 4 (24) |
| Cardiac arrest | 3 (18) |
| AKI | 3 (18) |
| Outcome | |
| Died | 1 (6) |
| MV duration, d, median (IQR) | 6 (4–11) |
| ICU LOS, d, median (IQR) | 5.5 (4.3–8.5) |
| Hospital LOS, d, median (IQR) | 13 (6.8–15) |
| Treatment . | Result . |
|---|---|
| Respiratory supporta | |
| None | 3 (18) |
| HFNC | 7 (41) |
| NIV | 4 (24) |
| IMV | 8 (47) |
| Vasoactive infusion | 9 (53) |
| Respiratory adjunctsb | 1 (6) |
| Medications | |
| Antibiotics | 15 (88) |
| Remdesivir | 4 (24) |
| Lopinavir and/or ritonavir | 1 (6) |
| Corticosteroids | 9 (53) |
| Tocilizumab | 7 (41) |
| Hydroxychloroquine | 8 (47) |
| Diagnosis and/or complication | |
| Pneumonia | 13 (76) |
| ARDSc | 8; 2 mild, 1 moderate, 3 severe (47) |
| Myocarditis | 4 (24) |
| Cardiac arrest | 3 (18) |
| AKI | 3 (18) |
| Outcome | |
| Died | 1 (6) |
| MV duration, d, median (IQR) | 6 (4–11) |
| ICU LOS, d, median (IQR) | 5.5 (4.3–8.5) |
| Hospital LOS, d, median (IQR) | 13 (6.8–15) |
Data are expressed as n (%) unless otherwise noted. AKI, acute kidney injury; IQR, interquartile range; LOS, length of stay; MV, mechanical ventilation.
Percentage adds up to >100 because some patients received >1 modality.
Includes inhaled nitric oxide, prone positioning, and neuromuscular blockade.
Two were supported with NIV, so we were unable to classify severity.
Pneumonia and ARDS were common diagnoses (Table 2, Supplemental Table 6). Vasoactive infusions were frequent, including 3 of 4 with myocarditis. Other organ support or complications were uncommon. Outcomes (minimum 3 weeks data) are shown in Table 2 and Supplemental Table 6. As of submission, 3 patients remained hospitalized, 1 remained in the ICU, and 1 died.
Discussion
Our description exclusively about critical pediatric COVID-19 reveals an uncommon (17 patients, 60 centers) but heterogeneous disease. Children frequently had GI rather than respiratory symptoms after a brief illness and recovered quickly despite significant support. We found regional variability of diagnoses (myocarditis in Europe), treatments (remdesivir in North America), and age.
Our findings parallel recent studies describing frequent comorbidities but a short PICU stay and low mortality, contrasting with adults.2,3,6,7 Compared to the North American series, our study was international, younger, included only severe disease, and revealed a wider range of common symptoms.7 We also provide critical COVID-19 laboratory findings.
We found that 3 children had peri-intubation arrest, markedly higher than expected.9 At least 1 resulted from unfamiliar protective equipment and intubation processes. Clinicians must consider the risks before intubating these children. Pediatric COVID-19 myocarditis has not been previously reported, although adult cases are described.10 It is unclear why myocarditis was only identified in Europe, but pediatric clinicians should consider cardiac involvement, particularly in those with the GI complaints common in our myocarditis patients.
This is a small case series and should be used to generate hypotheses for research rather than informing current treatment. Regional variations may limit our ability to identify outcome associations but do reveal regional differences. Finally, others use different definitions for COVID-19 severity, but their subjectivity could lead to patient misclassification.6 Our definitions are simple, objective, and reflect clinically relevant distinctions.
Conclusions
We provide early clinical and laboratory data about critical pediatric COVID-19, which suggest a variable disease but generally good outcomes compared with adults. Targets for research include the course of organ failure in pediatric critical COVID-19, laboratory findings for predicting illness course or complications, the inflammatory response and its role in pathophysiology, best treatments, and specific organ involvement, such as myocarditis.
Acknowledgments
We thank Martha I. Alvarez-Olmos from Fundación Cardioinfantil - Instituto de Cardiología (Bogotá, Colombia) and Agustín Cavagnaro from Complejo Asistencial Dr Sótero del Río, Santiago, Chile. We thank Laura Merson and the International Severe Acute Respiratory and Emerging Infection Consortium research team for developing and sharing their case report form.
Drs González-Dambrauskas and Vásquez-Hoyos designed the study, oversaw data collection and analysis, and participated in drafting and editing the manuscript; Dr Karsies designed the study, supervised data collection and analysis, conducted statistical analysis, and participated in drafting and editing the manuscript; Dr Shein designed the study, participated in data analysis and interpretation, and participated in drafting and editing the manuscript; Drs Camporesi, Díaz-Rubio, Piñeres-Olave, Fernández-Sarmiento, Gertz, Harwayne-Gidansky, Pietroboni, Urbano, Wegner, and Zemanate participated in creation of the study concept and data interpretation and were involved in data acquisition and drafting and editing the manuscript; and all authors had final approval of the manuscript, approved the final manuscript as submitted, and agree to be accountable for all aspects of the work.
FUNDING: No external funding.
References
Competing Interests
POTENTIAL CONFLICT OF INTEREST: The authors have indicated they have no potential conflicts of interest to disclose.
FINANCIAL DISCLOSURE: The authors have indicated they have no financial relationships relevant to this article to disclose.
Comments
iNOS Surge - Hypothesis on Which COVID-19's Complications Converge
Letter to the Editor:
I read with interest Dr. Sebastian Gonzalez-Dambrauskas’ article Pediatric critical care and COVID-19.
It is possible excessive iNOS (inducible nitric oxide synthase) and peroxynitrite (nitric oxide derived reactive nitrogen species) are vitally relevant.
Some viruses stimulate iNOS which increases nitric oxide and subsequent peroxynitrite. Evidence linking excessive iNOS/peroxynitrite to COVID-19’s complications include:
1. In a chimeric antigen receptor T cell therapy murine model, iNOS induction was implicated in cytokine storm while iNOS inhibition improved cytokine storm including survival from severe disease (1)
2. Peroxynitrite is linked to:
- Kawasaki Disease (murine model) - significantly increased iNOS/nitrotyrosine (peroxynitrite marker) staining was present at sites of coronary arteritis and aneurysm (2)
- Heart failure - peroxynitrite inhibits myofibrillar creatine kinase which is a critical component of cardiomyocyte contractility (3)
- Thromboembolism – peroxynitrite decreases blood clot retraction (via platelet mitochondrial inhibition) which may increase thromboembolic events (4)
COVID-19 may trigger excessive iNOS and downstream peroxynitrite which lead to cytokine storm and oxidative/nitrosative stress associated organ injuries. Current COVID-19 therapies (antivirals, convalescent plasma, interleukin inhibition, steroids) may not resolve excessive iNOS’/peroxynitrite’s harmful effects. As such, a critical gap in COVID-19 treatment remains.
Early intervention with N-acetylcysteine or methylene blue may fill this gap and represent an important adjunct to COVID-19 treatment. N-acetylcysteine inhibits iNOS induction (5) while methylene blue may decrease nitric oxide and peroynitrite production.
REFERENCES:
1. Giavridis T et al. CAR T cell-induced cytokine release syndrome is mediated by macrophages and abated by IL-1 blockade. Nat Med. 2018 June; 24(6): 731-738.
2. Adewuya O et al. Mechanism of vasculitis and aneurysms in Kawasaki disease: role of nitric oxide. Nitric oxide. 2003 Feb; 8(1): 15-25.
3. Mihm MJ and Bauer JA. Peroxynitrite-induced inhibition and nitration of cardiac myofibrillar creatine kinase. Biochimie. 2002 Oct; 84(10): 1013-1019.
4. Misztal T et al. Peroxynitrite may affect clot retraction in human blood through the inhibition of platelet mitochondrial energy production. Thromb Res. 2014 Mar; 133(3): 402-411.
5. Bergamini S et al. N-acetylcysteine inhibits in vivo nitric oxide production by inducible nitric oxide synthase. Nitric Oxide. 2001 Aug; 5(4): 349-360.