Association between red blood cell distribution width and survival rate of maintenance dialysis patients

Objective To Study the association between red blood cell distribution width (RDW) and mortality in patients with maintenance hemodialysis (MHD) and continuous ambulatory peritoneal dialysis (CAPD). Methods A total of 331 adult patients diagnosed with end-stage renal diseases at the Department of Nephrology, the Second Affiliated Hospital of Kunming Medical University from January 1, 2015 to October 31, 2016, and undergoing dialysis for the first time in this hospital. All of dialysis patients were followed up to reference to other medical centers, kidney transplantation, death, loss of follow-up, or follow-up observation for 2 years, until the study termination date (October 31, 2016). Clinical data of these patients were collected, and the patients were divided into high RDW (RDW>14.3%) and low RDW (RDW ≤ 14.3%) groups according to the median RDW value. Two-year survival rate and risk factors for all-cause death in all the patients were studied. Survival rates and effects of different dialysis methods on survival and prognosis of the patients were compared between the high RDW group and the low RDW group, by using Kaplan-Meier analysis. Cox regression model was employed to analyze independent risk factors for all-cause death of the patients. Receiver operator characteristic (ROC) curve was employed to analyze predictive value of RDW for all-cause death of the patients with MHD and CAPD. Results The median survival time of 331 dialysis patients was 19 months, with the median RDW value of 14.3%. The median survival times for the high RDW group (161 patients) and the low RDW group (170 patients) were 19 months and 24 months, respectively. There were statistically significant differences between the two groups in white blood cell count, hematocrit, C-reactive protein, total cholesterol, folic acid, and complicated hypertension rate (all P<0.05). The Kaplan-Meier survival curve showed that the survival rate of patients with RDW ≤ 14.3% was significantly higher than that of patients with RDW>14.3% (P<0.05). The 2-year survival rates of all dialysis patients in both the high RDW group and the low RDW group were 67.9% and 81.1%, and the 2-year survival rates of hemodialysis patients in the two groups were 62.5% to 67.9%. The 2-year survival rates of peritoneal dialysis patients in the two groups were 77.4% and 91.4%. Among them, the differences of peritoneal dialysis patients and all dialysis patients between the two groups were statistically significant (P<0.05). Well-corrected multivariate Cox regression analysis showed that high RDW was an independent risk factor for all-cause death of all the dialysis patients (HR=1.311, 95% CI:1.106~1.555, P=0.002) and the CAPD patients(HR=3.230, 95% CI:1.598~6.527, P=0.001). The areas under the curve (AUC) for RDW to predict whether all the dialysis patients, hemodialysis ones and peritoneal dialysis ones died were 0.606, 0.560, and 0.726, respectively. Conclusions RDW is associated with the risk for all-cause death in all dialysis patients and CAPD patients, and It can be used as a valuable predictor of the risk for death in all dialysis patients and CAPD patients.