Prognostic Value of the Red Cell Distribution Width-to-eGFR Ratio (RGR) Across Chronic Heart Failure Phenotypes: A Retrospective Observational Pilot Study

Abstract

Background/Objectives: This study aimed to investigate the prognostic value of the red cell distribution width-to-estimated glomerular filtration rate (RGR) ratio in patients hospitalized with chronic heart failure (CHF) and its potential interaction with NT-proBNP levels. By integrating anemia and renal dysfunction markers, the RGR may provide enhanced predictive insights regarding extended length of hospital stay (ELOS) > 7 days, in-hospital mortality, and 6-month all-cause mortality across specific CHF phenotypes. Methods: In this retrospective, single-center pilot observational study, 627 CHF admissions (January 2022–August 2024) were analyzed. Patients were classified according to the ESC guidelines into heart failure with reduced (HFrEF), mildly reduced (HFmrEF), or preserved ejection fraction (HFpEF). The RGR was calculated as red cell distribution width standard deviation (RDW-SD) divided by estimated glomerular filtration rate (eGFR). Predictive accuracy was evaluated using logistic regression, receiver operating characteristic (ROC) analyses, and stepwise Cox proportional hazard regression. Results: RGR was significantly higher in HFrEF than in HFpEF (p = 0.042) and predicted ELOS only in HFpEF (AUC = 0.619). In contrast, for in-hospital mortality, RGR achieved excellent discrimination in HFrEF (AUC = 0.945), outperforming RDW and NT-proBNP. In HFmrEF, RDW exhibited the highest predictive power (AUC = 0.826), whereas in HFpEF, NT-proBNP was the strongest predictor (AUC = 0.958), although RGR preserved good discrimination (AUC = 0.746). Across the entire cohort and HF phenotypes, RGR consistently emerged as a significant predictor in univariable analysis. In multivariable models, it improved the significance prognosis especially alongside NT-proBNP in the entire cohort and HFrEF. For 6-month all-cause mortality, RGR surpassed RDW in prediction in all HF phenotypes. Conclusions: The RGR independently predicts prolonged hospitalization, in-hospital, and 6-month mortality in CHF—often outperforming RDW and eGFR and being comparable to NT-proBNP, especially in HFrEF. These findings suggest that RGR may serve as a valuable risk stratification tool in CHF management.