1. Abstract 11936: Red Blood Cell Distribution Width Predicts Near-Term Mortality in Patients With Atherosclerotic Cardiovascular Disease.
- Author
-
Dikilitas, Ozan, He, Baosheng, and Kullo, Iftikhar
- Subjects
- *
ERYTHROCYTES , *CARDIOVASCULAR diseases , *PERIPHERAL vascular diseases , *ELECTRONIC health records , *MORTALITY - Abstract
Introduction: Red blood cell distribution width (RDW), a measure of heterogeneity in the size of circulating red blood cells, is predictive of adverse outcomes in a variety of diseases. We assessed the prognostic value of RDW in patients with atherosclerotic cardiovascular disease (ASCVD). Methods: We studied 7,849 patients (69±11 years, 65% male) referred to the noninvasive vascular laboratories at Mayo Clinic from January 2005 to March 2013 with follow up through March 2018. Baseline RDW, comorbidities, and mortality were obtained from the electronic health record (EHR). ASCVD phenotypes, namely, abdominal aortic aneurysm (AAA), cerebrovascular disease (CVD), coronary heart disease (CHD), and peripheral artery disease (PAD), were ascertained by validated electronic phenotyping algorithms that included ICD-9-CM and CPT codes, lab values, and medications. Extended multivariable Cox regression models that included a time-dependent coefficient were used to examine changes in the association between RDW and mortality over time. Results: The median RDW was 13.5 (1.3)%, and was highest in PAD patients 13.7 (1.6)%. During the mean follow up of 7.7 years, 1,900 patients (24.2%) died (Table). In Cox models stratified by ASCVD subtypes, with adjustment for age, sex and comorbidities, RDW was independently associated with mortality in all four ASCVD phenotypes. For a unit increase in RDW, impact on mortality was greatest within the first 3 years following the baseline measurement (p-value; AAA 0.02, PAD, CVD, and CHD <0.001), attenuated thereafter and was not significant beyond 5 years (Table). Conclusions: RDW is an independent predictor of near-term mortality across the spectrum of ASCVD and may be useful for prognostication. [ABSTRACT FROM AUTHOR]
- Published
- 2018
- Full Text
- View/download PDF