ABSTRACT
Objective: It is thought that hyperinflammation has an important role in the pathogenesis of severe COVID-19 and tests that determine the de-gree of inflammation can be used to predict the severity of the disease. From this point of view, we aimed to determine the hematological parameters that can predict the severity of COVID-19 in pediatric patients. Material and Methods: Symptomatic and SARS-CoV-2-PCR test positive 105 children were included to study. Seventy-nine patients had mild, 26 had moderate to severe COVID-19 at admission. Data about their demo-graphic characteristics, clinical and laboratory findings were collected from their medical records. Correlations between the hematological parameters and disease severity of patients were investigated by using uni-variate and multivariate regression analyses. Predictive value of different diagnostic markers was studied. Results: Mean age was older (177 months vs. 70 months) and mean body mass index (BMI) was higher (18.8 kg/m2 vs. 25.0 kg/m2) in patients with severe COVID-19 than those with mild. Univariate analysis showed that mean leucocyte (WBC), lymphocyte, eosinophiles, and platelet counts were lower;mean platelet volume (MPV), neutrophil to lympho-cyte ratio (NLR), and derived neutrophil to lymphocyte ratio (dNLR) were higher in severe COVID-19 group (p< 0.05). Multivariant analysis showed low lymphocyte (OR 0.072) and WBC count (OR 0.085), high dNLR (OR 2.14) and MPV (OR 2.35) indexes were the most valuable parameters to predict disease severity, ROC curve analysis revealed lymphocyte count has superior predictive value (<1.55 /mm3 has 84.6% sensitivity, 70.9% specificity) than other CBC parameters have. Conclusion: Low lymphocyte and leukocyte count, high MPV and dNLR values have significant predictive value in predicting COVID-19 severity. In particular, lymphopenia appears to be a valuable parameter to identify patients at high risk for severe disease and initiate accurate treatment to prevent disease deterioration.