Validation of red cell distribution width as a COVID-19 severity screening tool

Aim: The aim of this study is the predictive validation of red cell distribution width (RDW) in COVID-19 patients. Method: In total, 331 COVID-19 patients were classified as 'severe' and 'nonsevere' groups based on the WHO standard criteria. The levels of RDW standard deviation (SD) were evaluated as both continuous and categorical variables. Multivariate statistical analyses were used. Results: RDW-SD <= 43 and <= 47 fl thresholds showed high specificity (90.1-91.4%) for diagnosing nonsevere illness and no risk of death. RDW-SD >47 indicated severe illness and a high mortality risk while 43<RDW-SD <= 47 indicated severe illness with low risk of death. Conclusion: RDW-SD levels may be a potent independent predictor of the infection severity and mortality probability in COVID-19 patients. Lay abstract: The red cell distribution width (RDW) test is part of a complete blood count, and calculates the difference in size from the largest to the smallest red blood cell. Our analysis of 331 COVID-19 patients showed that the RDW standard deviation index (RDW-SD) has the potential to differentiate nonsevere from severe patients as well as predict the death of COVID-19 patients. In this regard, RDW-SD <= 47 fl indicated nonsevere illness and low risk of death while RDW-SD >47 indicated severe illness and a high risk of death. Also, 43< RDW-SD <= 47 indicated severe illness with low risk of death.