Diagnostic and predictive value of hematological parameters of COVID-19 patients: a retrospective study

Abstract

Link of Video Abstract: https://youtu.be/K2R3dfdvi20 Background: Simple, cost-effective, and practical laboratory indicators are required to diagnose and evaluate COVID-19 disease severity. This study assessed the diagnostic value and predictor of severity and outcome parameters of NLR, d-NLR, MLR, PLR, and ALC in COVID-19 patients. Methods: A retrospective study used medical record data from 100 COVID-19 patients from November 2020 to March 2021. SPSS version 22 was used to analyze the data. The severity of COVID-19 was predicted using a ROC curve. Kaplan-Meier analysis was employed to evaluate the ability of various inflammatory markers to predict COVID-19 prognosis. A multivariate analysis with logistic regression was conducted to assess the ability of an independent predictor of COVID-19 severity. A p-value of <0.05 was considered significant. Results: ALC values were lower in the severe-critical COVID-19 group, whereas NLR, d-NLR, MLR, and PLR values were higher. The NLR, d-NLR, and ALC parameters had sufficient accuracy, whereas the MLR and PLR parameters had low accuracy. NLR, d-NLR, MLR, PLR, and ALC had optimal cut-off values of 7.865, 4.82, 0.455, 235.000, and 0.895, respectively. The multivariate odds ratio for ALC was 7.348 (95% CI = 1.914-28.214; p = 0.004). Kaplan-Meier analysis revealed differences in survival time based on the optimal NLR, MLR, d-NLR, and ALC cut-offs obtained. Conclusion: NLR, d-NLR, MLR, PLR, and ALC are all potential predictors of COVID-19 severity and prognosis. ALC is a reliable predictor of COVID-19 severity.