The goal of this study was to independently validate the recent observations on the predictive role of red cell distribution width (RDW) for outcomes in chronic heart failure and to provide epidemiologic data on the biological correlates of RDW in heart failure (HF). Understanding the mechanism underlying this observation is unclear, largely hampered by the lack of epidemiologic studies demonstrating factors that are associated with anisocytosis in cardiovascular diseases.

One hundred ninety-five patients (145 men, 50 women) with systolic HF were enrolled and followed up for a median of 14.5 months. Primary end points were all-cause mortality and hospital readmission due to worsening HF symptoms. A total of 19 clinical chemistry, hematology, and biochemical variables were considered for analysis together with clinical parameters in Cox proportional hazards and multiple regression models.

Red cell distribution width was found to be an N-terminal pro–brain natriuretic peptide independent predictor of all-cause mortality (adjusted HR 1.61 per 1 SD increase) in our study. Multiple correlations between biomarkers of ineffective erythropoiesis (serum iron, ferritin, and soluble transferrin receptor levels), inflammation and acute-phase reaction (interleukin-6, soluble tumor necrosis factor (TNF) receptor I and soluble TNF receptor II, C-reactive protein, and prealbumin concentrations), undernutrition (total cholesterol and albumin levels), and renal function were observed. In the multiple regression model, the strongest relationship for RDW was obtained with soluble transferrin receptor, soluble TNF receptor I, soluble TNF receptor II, and total cholesterol.

Here we validate the strong, independent prediction of morbidity and mortality in HF by RDW. The described correlations between RDW and inflammation, ineffective erythropoiesis, undernutrition, and impaired renal function may facilitate the understanding why this marker is associated with adverse outcomes in HF.