Red Blood Cell Distribution Width (RDW) Predicts COVID-19 Severity: A Prospective, Observational Study from the Cincinnati SARS-CoV-2 Emergency Department Cohort

Brandon Michael Henry*, Justin Lee Benoit, Stefanie Benoit, Christina Pulvino, Brandon A. Berger, Maria Helena Santos de Olivera, Christopher A. Crutchfield, Giuseppe Lippi

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

62 Scopus citations

Abstract

Since previous evidence has demonstrated that red blood cell distribution width (RDW) may be a useful prognostic parameter in many critical illnesses and infectious diseases, we investigated the utility of RDW for monitoring patients with coronavirus disease 2019 (COVID-19). The study population consisted of 49 COVID-19 patients, including 16 (32.6%) with severe illness, 12 (24.5%) with severe acute kidney injury (AKI), and 8 (16.3%) requiring renal replacement therapy (RRT). The predictive value of blood tests, performed during emergency department evaluation, was then addressed. A progressive increase of RDW was observed with advancing COVID-19 severity. The area under the curve (AUC) of RDW was 0.73 for predicting severe illness, 0.80 for severe AKI, and 0.83 for RRT, respectively. In multivariate analysis, elevated RDW was associated with 9-fold and 16-fold increased odds of severe COVID-19 and AKI, respectively. The results of this study suggest that RDW should be part of routine laboratory assessment and monitoring of COVID-19.

Original languageEnglish (US)
Article number618
JournalDiagnostics
Volume10
Issue number9
DOIs
StatePublished - Sep 2020

Keywords

  • Acute kidney injury
  • Anisocytosis
  • Diagnostics
  • Hematology
  • Infections
  • Red blood cells

ASJC Scopus subject areas

  • Clinical Biochemistry

Fingerprint

Dive into the research topics of 'Red Blood Cell Distribution Width (RDW) Predicts COVID-19 Severity: A Prospective, Observational Study from the Cincinnati SARS-CoV-2 Emergency Department Cohort'. Together they form a unique fingerprint.

Cite this