The Role of Hematological Parameters in Predicting the Death of Hospitalized Patients With Covid-19 Publisher

Summary: Can blood tests predict COVID-19 severity? Findings suggest certain hematological parameters can identify patients at risk for ICU admission and worse outcomes. #COVID19 #HealthData

Mobinikhaledi M ; Mousavi ZS ; Falahati V ; Ghasemi A ; Alamsihashiani A ; Ghaffari K
Source: Advanced Biomedical Research Published:2023

Abstract

Background: The role of the hematologic indicators in the identification of severe or critical patients requires further investigation. In this study, we focused on predicting Covid-19 patients at risk of progression using blood parameters. Materials and Methods: We performed a retrospective study including 444 patients with confirmed Covid-19. Hematological parameters were evaluated. The logistic regression analysis was performed with step-wise method with dependent variables such as intensive care units admission, partial pressure of oxygen saturation, and mortality. Also, independent variables such as hematological parameters, age and sex to assess variables that are likely to predict patients at risk of progression. Results: Patients in intensive care units had significantly higher mean absolute neutrophil count than outpatients (P < 0.001). There was a statistically significant difference in the mean absolute lymphocyte count between dead and survived patients (P = 0.015). Multivariate analysis confirmed the positive association of the white blood cells (P < 0.001), absolute neutrophil count (P < 0.004), red cell distribution width (P < 0.001), and lactate dehydrogenase (P = 0.007) to be positively associated with the admission of Covid-19 patients in the intensive care units and the absolute monocyte count (P = 0.012, Odds ratios = 0.100, CI95% = 0.066-0.605) to be negatively associated with mortality. Conclusion: Based on the results of our study, it is recommended to use hematological data to make clinical decisions and evaluate the patient's prognosis. © 2024 Elsevier B.V., All rights reserved.