Background: Red cell distribution width (RDW) is a hematological parameter routinely obtained as part of the complete blood count. Recently, RDW has emerged as a potential independent predictor of clinical outcome in patients with established cardiovascular disease. However, little is known about the role of RDW as a prognosticator among persons with stroke, especially with regard to an incontrovertible endpoint like mortality. We assessed the association of RDW with stroke, and its effect on mortality among persons with stroke.
Methods: Data from the National Health and Nutrition Examination Survey (NHANES) a nationally representative sample of United States adults were analyzed. The study population consisted of 480 individuals aged > or =25 years with a baseline history of stroke followed-up from survey participation (1988-1994) through mortality assessment in 2000. Proportional hazard regression (Cox) was utilized to explore the independent relationship between RDW and mortality after adjusting for potential confounders.
Results: Among the cohort, 52.4% were female, 64% aged > or =65 years. Mean RDW was significantly higher among persons with stroke compared to individuals without a stroke (13.7% vs.13.2%,p<0.001). Baseline RDW was higher among persons with known stroke who later died vs. remained alive (13.9% vs.13.4%,p<0.001). After adjusting for confounders, those with elevated RDW (fourth vs. first quartile) were more likely to have experienced a stroke (OR 1.71, CI=1.20-2.45). Higher RDW level (fourth vs. first quartile) among those with known stroke independently predicted subsequent cardiovascular deaths (HR=2.38 and CI=1.41-4.01) and all-cause deaths (HR=2.0, CI=1.25-3.20).
Conclusions: Elevated RDW is associated with stroke occurrence and strongly predicts both cardiovascular and all-cause deaths in persons with known stroke.