Association Between Red Blood Cell Distribution width and Long-Term Mortality in Patients with Intracerebral Hemorrhage.

backgroundThe association between the red cell distribution width (RDW) and long-term mortality in patients with intracerebral hemorrhage (ICH) has not been clearly established.

methodsWe conducted a retrospective cohort study of patients with ICH admitted to two tertiary hospitals. The primary outcome was long-term mortality, and the effect of elevated RDW (RDW coefficient of variation [RDW-CV]; RDW standard deviation [RDW-SD]) on outcomes was assessed by using logistic regression analysis. Serum RDW levels was divided into four levels by quartiles (the lowest quartile [Q1]; the highest quartile [Q4]).

resultsThis study included 4223 patients with ICH. After adjustment for potential confounders, admission RDW-CV (Quartile 4 [Q4] vs. Quartile 1 [Q1], adjusted hazard ratio [HR] 1.61, 95% confidence interval [CI] 1.34-1.92) and median RDW-CV within the first month after admission (Q4 vs. Q1, adjusted HR 1.69, 95% CI 1.40-2.04) were both associated with 1-year mortality following ICH. Parallel results were found for RDW-SD. In the receiver operating characteristic analyses, both RDW-CV and RDW-SD outperformed some inflammatory biomarkers, such as albumin, hemoglobin, total cholesterol, platelet count, lymphocyte, and fibrinogen, in predicting long-term mortality following ICH. Additionally, compared with admission RDW, median RDW-CV and RDW-SD (areas under the curve [AUC] 0.668 and 0.652, respectively) was superior to predict long-term mortality, (P < 0.001). Furthermore, median RDW-CV level was a better predictor than RDW-SD (P = 0.03).

conclusionsIn patients with ICH, RDW independently predicted long-term mortality. Median RDW levels within the first month after admission were better predictors of long-term mortality compared with RDW levels on admission. Additionally, median RDW-CV showed superior predictive capacity than median RDW-SD for long-term mortality following ICH.