Objective: To determinate the prognostic value of red cell distribution width (RDW) and the relationships between RDW and clinical characteristics in patients with chronic heart failure (CHF).
Methods: A total of 16 681 in-hospital patients with chronic systolic HF and LVEF < 50% from 12 hospitals in Hubei province, China were enrolled. All patients were followed up with telephone call. Patients were divided into RDW ≤ 13.2% (n = 3981), 13.3% - 14.1% (n = 3996), 14.2% - 14.8% (n = 4319) and ≥ 14.9% (n = 4385) groups. Multivariate Cox regression analysis was performed to determine whether RDW is an independent risk factor of all-cause mortality in overall patients, patients with various etiologies. Multivariate Cox proportional hazard analysis was performed to determine the risk of all-cause mortality among various RDW groups.
Results: (1) Compared with RDW ≤ 13.2% group, adjusted hazard ratios (HRs) and 95% confidence intervals (CIs) of all-cause mortality for RDW 13.3% - 14.1%, 14.2% - 14.8% and ≥ 14.9% were 0.892 (95%CI 0.818 - 0.973, P = 0.01), 0.859 (95%CI 0.793 - 0.931, P < 0.01) and 1.034 (95%CI 0.961 - 1.111, P = 0.373) respectively. (2) Compared with MCV normal group, the adjusted HRs of MCV elevation and MCV decline groups were 1.351 (95%CI 1.063 - 1.718, P < 0.01) and 1.316 (95%CI 1.034 - 1.675, P < 0.01), respectively. (3) Compared to patients with rheumatic heart diseases, the adjusted HR for all-cause mortality in patients with coronary heart disease, dilated cardiomyopathy and hypertensive heart disease with RDW > 16% were 1.437 (95%CI 1.141 - 1.810, P < 0.01), 1.651 (95%CI 1.276 - 2.138, P < 0.01) and 1.276 (95%CI 1.004 - 1.621, P < 0.01), respectively. (4) The RDW is independently correlated with BMI (r = -0.345, P < 0.01), diastolic blood pressure (r = -0.321, P < 0.01), albumin (r = -0.411, P < 0.01), blood urine nitrogen (r = 0.476, P < 0.01), right ventricular end-diastolic diameter (r = 0.383, P < 0.01), LVEF (r = -0.463, P < 0.01) and heart rate (r = 0.379, P < 0.01).
Conclusions: There is a J shape relationship between all-cause mortality and RDW. The elevation or decline of MCV with increased RDW is linked with increased all-cause mortality in CHF patients.