Background: COVID-19 infection continues to be a serious health concern in pediatric patients, associated with substantial morbidity and mortality. An important clinical question is whether laboratory hematologic parameters may act as a surrogate for imaging findings in pediatric COVID-19 pneumonia, in order to reduce exposure to ionizing radiation. The aim is to investigate the relationship between radiographic findings and hematological laboratory parameters in pediatric patients with COVID-19 infection by directly comparing the findings from both studies.
Methods: This was an institutional review board-approved retrospective study of 187 consecutive pediatric patients with microbiologically confirmed COVID-19 pneumonia and available initial chest radiographs at the time of diagnosis, obtained between March 2020 and December 2020. Two groups of patients were created: 1) 1st group (without radiographic evidence of COVID-19 pneumonia) and 2) second group (with radiographic evidence of COVID-19 pneumonia). Two experienced radiologists independently reviewed the initial chest radiographs for abnormalities. Patients’ medical records were reviewed for clinical presentation, hematological laboratory parameters, and patient outcomes. The relationship between initial chest radiographic findings and hematological laboratory parameters was evaluated between the two groups. Interobserver agreement was estimated with the Cohen k coefficient.
Results: The study population consisted of 187 chest radiographs from 187 individual pediatric patients (95 males and 92 females; mean age ± SD, 10.1 ± 6.0 years; range, 9 months – 18 years). Group 1 consisted of 103 chest radiographs (55.0%) from 103 individual pediatric patients (54 males and 49 females; mean age ± SD, 12.5 ± 5.2 years; range, 9 months – 18 years). Group 2 consisted of 84 chest radiographs (45.0%) from 84 individual pediatric patients (41 males and 43 females; mean age ± SD, 7.0 ± 5.6 years; range, 11 months – 18 years). In group 2, observed chest radiographic abnormalities included ground-glass opacity (GGO) in 75 patients (89.2%), GGO and consolidation in 6 (7.2%), peribronchial thickening in 1 patient (1.1%), consolidation in 1 patient (1.1%), and peribronchial thickening, GGO and consolidation in one patient (1.1%). In addition, group 2 patients (with radiographic abnormalities of COVID-19 pneumonia) had significantly elevated LDH (p = 0.001) compared to group 1 patients (without radiographic findings of COVID-19 pneumonia). All pediatric patients in both groups completely recovered from COVID-19 infection. Interobserver agreement for chest radiographic findings was nearly perfect between the two reviewers in both groups (k = 0.96, p = 0.001).
Conclusion: The presence of radiographic abnormalities on initial chest radiographs correlates well with hematologic laboratory abnormalities, specifically elevated LDH, in pediatric patients with COVID-19 pneumonia. Therefore, it is reasonable to conclude that hematologic laboratory parameters may act as a surrogate for chest radiography in pediatric COVID-19 pneumonia and that chest radiography may not be clinically indicated in pediatric patients with COVID-19 infection with normal LDH level, resulting in decreased exposure to ionizing radiation in this vulnerable population.