Risk factors for severe PCR-positive SARS-CoV-2 infection in hospitalised children

Abstract

Objective: To identify risk factors for severe disease in children hospitalised for SARS-CoV-2 infection.

Design: Multicentre retrospective cohort study.

Setting: 18 hospitals in Canada, Iran and Costa Rica from 1 February 2020 to 31 May 2021.

Patients: Children<18 years of age hospitalised for symptomatic PCR-positive SARS-CoV-2 infection, including PCR-positive multisystem inflammatory syndrome in children (MIS-C).

Main outcome measure: Severity on the WHO COVID-19 Clinical Progression Scale was used for ordinal logistic regression analyses.

Results: We identified 403 hospitalisations. Median age was 3.78 years (IQR 0.53-10.77). At least one comorbidity was present in 46.4% (187/403) and multiple comorbidities in 18.6% (75/403). Eighty-one children (20.1%) met WHO criteria for PCR-positive MIS-C. Progression to WHO clinical scale score ≥6 occurred in 25.3% (102/403). In multivariable ordinal logistic regression analyses adjusted for age, chest imaging findings, laboratory-confirmed bacterial and/or viral coinfection, and MIS-C diagnosis, presence of a single (adjusted OR (aOR) 1.90, 95% CI 1.13 to 3.20) or multiple chronic comorbidities (aOR 2.12, 95% CI 1.19 to 3.79), obesity (aOR 3.42, 95% CI 1.76 to 6.66) and chromosomal disorders (aOR 4.47, 95% CI 1.25 to 16.01) were independent risk factors for severity. Age was not an independent risk factor, but different age-specific comorbidities were associated with more severe disease in age-stratified adjusted analyses: cardiac (aOR 2.90, 95% CI 1.11 to 7.56) and non-asthma pulmonary disorders (aOR 3.07, 95% CI 1.26 to 7.49) in children<12 years old and obesity (aOR 3.69, 1.45-9.40) in adolescents≥12 years old. Among infants<1 year old, neurological (aOR 10.72, 95% CI 1.01 to 113.35) and cardiac disorders (aOR 10.13, 95% CI 1.69 to 60.54) were independent predictors of severe disease.

Conclusion: We identified risk factors for disease severity among children hospitalised for PCR-positive SARS-CoV-2 infection. Comorbidities predisposing children to more severe disease may vary by age. These findings can potentially guide vaccination programmes and treatment approaches in children.

Keywords: COVID-19; Epidemiology.

Figure 1

Flow chart of hospitalised children<18 years of age reported to the PICNIC study between 1 February 2020 and 31 May 2021 with a diagnosis of SARS-CoV-2 infection. MIS-C, multisystem inflammatory syndrome in children; PICNIC, Pediatric Investigators Collaborative Network on Infections in Canada.

Figure 2

Multivariable ordinal logistic regression models for factors associated with more severe PCR-positive SARS-CoV-2 infection on WHO COVID-19 Clinical Progression Scale. (A) Model evaluating individual comorbidities. (B) Model evaluating age and the number of comorbidities as risk factors for disease severity. aOR, adjusted OR.

Figure 3

Age as an exposure variable for factors associated with more severe PCR-positive SARS-CoV-2 infection on the WHO COVID-19 Clinical Progression Scale. Analyses according to ordinal logistic regression. (A) Unadjusted OR. (B) OR adjusted for the number of comorbidities, presence of any bacterial and/or viral coinfections, chest imaging results and multisystem inflammatory syndrome in children (MIS-C). Shaded areas represent the 95% CIs.