Correlations of red blood cell distribution width with risk and severity of noncardiac ischemic stroke

Abstract

Objective To investigate the correlation between the red blood cell distribution width (RDW) with risk and and severity of noncardiac ischemic stroke . Methods The patients with acute noncardiac ischemic stroke were enrolled retrospetively and were used as a case group, and the non-stroke patients admitted to the Department of Neurology at the same period were served as a control group. The National Institutes of Health Stroke Scale (NIHSS) was used to assess the baseline neurological deficits. The NIHSS score <8 was defined as mild stroke and ≥8 was defined as moderate to severe stroke. The demographics, clinical data, and RDW were collected and compared. Multivariate logistic regression analysis was used to identify the independent risk factors for the risk of disease and severity of stroke in noncardiac ischemic stroke. Receiver operator characteristic (ROC) curve was used to analyze the predictive value of RDW for acute noncardiac ischemic stroke and its severity. Results A total of 1 173 patients with acute noncardiac ischemic stroke were enrolled. Over the same period, 996 non-stroke patients were hospitalized in the Department of Neurology. The RDW in the case group was significantly higher than that in the control group (13.1%±1.0% vs. 12.7%±0.7%; t=9.545, t<0.001). Multivariate logistic regression analysis showed that RDW (odds ratio [OR] 9.920, 95%confidence interval [CI] 6.286-15.655, P<0.001) was an independent risk factor for acute noncardiac ischemic stroke. ROC curve analysis showed that the RDW cutoff value for predicting acute noncardiac ischemic stroke was 12.55%, the area under the ROC curve was 0.611 (95% CI 0.588-0.635), the sensitivity was 73.3%, the specificity was 43.7%, the positive predictive value was 60.5%, the negative predictive value was 58.2%, and the accuracy was 59.7%. RDW in the moderate to severe stroke subgroup was significantly higher than that in the mild stroke subgroup (13.7%±1.6% vs. 12.9%±0.6%, t=-13.794, P<0.001). Multivariate logistic regression analysis showed that RDW (OR 3.835, 95% CI 3.002-4.899, P<0.001) was an independent risk factor for the severity of acute non-cardiac ischemic stroke. ROC curve analysis showed that the RDW cutoff value for predicting moderate to severe non-cardiac ischemic stroke was 13.45%, the area under the ROC curve was 0.713 (95% CI 0.674-0.752), the sensitivity was 47.7%, the specificity was 86.1%, the positive predictive value was 49.8%, the negative predictive value was 85.1%, and the accuracy was 77.6%. Conclusions RDW is an independent risk factor for acute non-cardiac ischemic stroke. Increased RDW is associated with the severity of acute non-cardiac ischemic stroke. Key words: Stroke; Brain Ischemia; Erythrocyte Indices; Risk Factors; Severity of Illness Index