Prognostic value of reticulocyte production ability in patients with chronic heart failure

Abstract Background Anaemia is frequently observed in patients with chronic heart failure (CHF), and many studies have elucidated the association between anaemia and mortality in patients with CHF. The quality of erythropoiesis is an intrinsic aspect of pathophysiology of anaemia, and the ability of reticulocyte production would be one of the pivotal factors in the quality of erythropoiesis. Purpose We sought to investigate whether the reticulocyte production ability is associated with clinical outcomes, and clarify the pathophysiology of reticulocyte production in patients with CHF. Methods We prospectively examined 1,328 symptomatic CHF patients (median age 71 [IQR 65-82] years, mean left ventricular ejection fraction 50 ± 16%, median NT-proBNP 743 [IQR 360–1710] pg/mL) in a multicentre registry with 27 Japanese sites between January 2020 and October 2023. The reticulocyte production ability was evaluated by calculating the reticulocyte production index (RPI) using reticulocyte counts and serum haematocrits. Patients were divided into four groups according to the presence or absence of WHO defined anaemia and the median value of RPI. The primary outcome was a composite of all-cause death and hospitalisation for worsening HF. Results During a median follow-up period of 551 (IQR 321–712) days, the primary outcome occurred in 219 (17%) patients. The group with anaemia and high RPI (≥0.978, the median) showed the highest incidence of the primary outcome among the groups (P < 0.001) (Figure 1). Multivariable regressions showed that higher RPI was independently associated with the higher risk of the primary outcome, even after adjustment for NT-proBNP, sodium, haemoglobin, serum iron, erythropoietin, and MAGGIC risk score (HR 1.41, 95% CI 1.09–1.82). Erythrocyte counts were significantly increased in patients with higher RPI compared to those with lower RPI in the groups without anaemia (P < 0.001)whereas there were no significant differences on erythrocyte counts between the groups with anaemia (P = 0.90) (Figure 2A). Patients with higher RPI had higher log erythropoietin levels compared to those with lower RPI in either anaemic or non-anaemic group (Figure 2B). In addition, serum iron levels and transferrin saturation did not significantly differ between the RPI groups in either anaemic or non-anaemic group (Figures 2C and D). Furthermore, in patients without anaemia, erythrocyte counts positively correlated with RPI (R = 0.27, P < 0.001) whereas there was very weak correlation between erythrocyte counts and RPI in those with anaemia (R = 0.07, P = 0.045). Conclusions Higher RPI was independently associated with worse clinical outcomes, suggesting RPI could be a surrogate marker for risk stratification in patients with CHF, especially in anaemic population. Higher reticulocyte production ability might reflect the impaired maturation of reticulocytes to erythrocytes independent of iron parameters.