Background
Red blood cell distribution width (RDW) has been assessed during COVID-19 patient hospitalization, however, further research should be done to evaluate RDW from routine community blood tests, before infection, as a risk factor for COVID-19 related hospitalization and mortality.
Patients and methods
RDW was measured as a predictor along with age, sex, chronic illnesses, and BMI in logistic regressions to predict hospitalization and mortality. Hospitalization and mortality odds ratios (ORs) were estimated with 95% confidence intervals (CI). RDW was evaluated separately as continuous and discrete (High RDW ≥ 14.5) variables.
Results
Four thousand one hundred and sixty-eight patients were included in this study, where 824 patients (19.8%) had a high RDW value ≥14.5% (High RDW: 64.7% were female, mean age 58 years [±22]
vs.
Normal RDW: 60.2% female, mean age 46 years [±19]). Eight hundred and twenty-nine patients had a hospitalization, where the median time between positive PCR and hospital entry was 5 [IQR 1–18] days. Models were analyzed with RDW (continuous) and adjusted for age, sex, comorbidities, and BMI suggested an OR of 1.242 [95% CI = 1.187–2.688] for hospitalization and an OR of 2.911 [95% CI = 1.928–4.395] for mortality (
p
< .001). RDW (discrete) with the same adjustments presented an OR of 2.232 [95% CI = 1.853–1.300] for hospitalization and an OR of 1.263 [95% CI = 1.166–1.368] for mortality (
p
< .001).
Conclusions
High RDW values obtained from community blood tests are associated with greater odds of hospitalization and mortality for patients with COVID-19.
KEY MESSAGES
RDW measures before SARS-CoV-2 infection is a predictive factor for hospitalization and mortality.
RDW threshold of 14.5% provides high sensitivity and specificity for COVID-19 related mortality, comparatively to other blood tests.
Patient records should be accessed by clinicians for prior RDW results, if available, followed by further monitoring.