Abstract
PURPOSE: Investigate the clinical/hematological characteristics of children infected with the Omicron variant of severe acute respiratory syndrome-coronavirus-2 (SARS-CoV-2) and identify an effective indicator to distinguish coronavirus disease 2019 (COVID-19) severity in children. METHODS: A retrospective study was conducted through electronic medical records from pediatric patients. The demographic, clinical, and routine blood test (RBT) features of children diagnosed by real-time PCR for SARS-CoV-2 were collected. RESULTS: Data of 261 patients were analyzed. The most common abnormality shown by RBTs was increased monocyte count (68%). Children had "mild-moderate" or "severe" forms of COVID-19. Prevalence of abnormal neutrophil count (p = 0.048), eosinophil count (p = 0.006), mean corpuscular volume (p = 0.033), mean platelet volume (p = 0.006), platelet-large cell ratio (p = 0.043), and red blood cell distribution width-standard deviation (p = 0.031) were significantly different in the two types. A combination of the neutrophil: lymphocyte ratio (NLR) and eosinophil count for diagnosing severe COVID-19 presented the largest AUC (0.688, 95% CI = 0.599-0.777; p < 0.001), and the AUC increased with a decrease in age. CONCLUSIONS: Combination of the NLR and eosinophil count might be a promising indicator for identifying severe COVID-19 in children at infection onset.