Pediatric SARS-CoV-2 long term outcomes study: chest radiographic and computed tomography findings at baseline

BACKGROUND: Radiologic findings associated with SARS-CoV-2 infection have been well described since its emergence. However, long-term clinical and radiologic complications of pediatric SARS-CoV-2 infection are less defined, including radiologic findings in children following recovery. OBJECTIVE: This study compares chest radiographic (CXR) and low-dose chest computed tomography (CT) findings in pediatric participants; a longitudinal cohort of children who have recovered from SARS-CoV-2 infection and uninfected controls. MATERIALS AND METHODS: Eight hundred forty-six children (700 laboratory confirmed SARS-CoV-2-infected children and 146 uninfected controls) completed a radiologic exam at the baseline visit, which occurred 9.8 months (mean) post-infection. CXR (n=485) and CT (n=362) images were evaluated by three radiologists for known radiologic manifestations of COVID-19 including various patterns of opacities (ground glass, reticular, consolidation), nodules, perihilar thickening, effusions, and/or cystic changes. Comparisons were made using generalized estimating equations with cohort (infected vs. uninfected) as the predictor. RESULTS: Overall, the incidence of radiologic abnormalities detected at baseline was 32.1% in the infected cohort and 24.0% in the uninfected cohorts. Infected participants were more likely to have CXR abnormalities than uninfected (21.2% vs. 12.4%; OR 2.44; 95% CI 1.19-5.02; P=0.016). This was primarily due to an increased incidence of perihilar peribronchial thickening (17.4% in infected vs. 10.1% in noninfected). There were no statistically significant differences between cohorts for CT abnormalities (46.2% vs. 42.1%; OR 1.24; 95% CI 0.71-2.16; P=0.46). Radiologic abnormalities were not statistically significantly associated with presence of pulmonary symptoms after recovery. CONCLUSIONS: At the baseline visit, infected participants were more likely than uninfected controls to have chest radiographic abnormalities, limited to perihilar peribronchial thickening. There was no significant difference in CT abnormalities between the two cohorts. There was no increased frequency of radiographic findings in children with underlying asthma, nor in children who experienced pulmonary symptoms following recovery at the baseline visit.