The association between red cell distribution width and poor outcomes in hospitalized patients with influenza

Guy Topaz, Yona Kitay-Cohen, Lee Peled, Wesal Gharra, Keren Kaminer, Mayan Eitan, Lamis Mahamid, Lotan Shilo

Research output: Contribution to journalArticlepeer-review

Abstract

Purpose To examine an association between red blood cell distribution width (RDW) and the prognosis of influenza patients. Methods We conducted a retrospective analysis of patients hospitalized with influenza during 2012–2015 in the internal medicine wards of one medical center. RDW measurements during hospitalization were analyzed. Primary outcome was complicated hospitalization (defined as at least one of: length of stay ≥ 7 days, need for mechanical ventilation, septic shock, transfer to intensive-care, or 30-day mortality). Secondary outcome was 30-day mortality. Results 153 patients were included, mean age: 62.5 ± 1, 82 (54%) male; 84 (55%) had a high RDW value (> 14.5%) during hospitalization. Patients with high and low RDW (≤ 14.5%) had similar age and comorbidity profiles, but those with high RDW had lower hemoglobin and higher creatinine levels. Patients with high RDW had a higher rate of complicated hospitalization (32.5% vs. 10.3%, p < 0.01) and a trend for increased 30-day mortality. In a multivariate regression model, high RDW was a predictor of complicated hospitalization (OR 5.03, 95% CI 1.81–13.93, p < 0.01). Each 1-point increase in RDW was associated with a 29% increase in the risk for the primary outcome. Conclusion RDW > 14.5% was a predictor of severe hospital complications in patients with influenza.

Original languageEnglish
Pages (from-to)166-169
Number of pages4
JournalJournal of Critical Care
Volume41
DOIs
StatePublished - 1 Oct 2017

Keywords

  • Hospital complications
  • Influenza
  • Mortality
  • Red cell distribution width

Fingerprint

Dive into the research topics of 'The association between red cell distribution width and poor outcomes in hospitalized patients with influenza'. Together they form a unique fingerprint.

Cite this