Red blood cell distribution width as a prognostic marker of mortality in patients on chronic dialysis: a single...

To determine if red cell distribution width (RDW) is associated with all-cause mortality in patients on chronic dialysis and to evaluate its prognostic value among validated prognostic biomarkers.

This is a single center, prospective longitudinal study. At the time of inclusion in January 2011, all patients were physically examined and a routine blood analysis was performed. A sera sample was preserved for determination of NT-pro-brain natriuretic peptide (NT-pro-BNP) and eosinophil cationic protein. Carotid intima media thickness (IMT) was also measured. Following one year, all-cause mortality was evaluated.

Of 100 patients, 25 patients died during the follow-up period of one-year. Patients who died had significantly higher median [range] RDW levels (16.7% [14.3-19.5] vs 15.5% [13.2-19.7], P<0.001. They had significantly higher Eastern Cooperative Oncology Group (ECOG) performance status (4 [2-4] vs 2 [1-4], Plt;0.001), increased intima-media thickness (IMT) (0.71 [0.47-1.25] vs 0.63 [0.31-1.55], P=0.011), increased NT-pro-BNP levels (8300 [1108-35000] vs 4837 [413-35000], P=0.043), and increased C-reactive protein (CRP) levels (11.6 [1.3-154.2] vs 4.9 [0.4-92.9], Plt;0.001). For each 1% point increase in RDW level as a continuous variable, one-year all cause mortality risk was increased by 54% in univariate Cox proportional hazard analysis. In the final model, when RDW was entered as a categorical variable, mortality risk was significantly increased (hazard ratio, 5.15, 95% confidence interval, 2.33 to 11.36) and patients with RDW levels above 15.75% had significantly shorter survival time (Log rank Plt;0.001) than others.

RDW could be an additive predictor for all-cause mortality in patients on chronic dialysis. Furthermore, RDW combined with sound clinical judgment improves identification of patients who are at increased risk compared to RDW alone.