Proposed Novel Heart Failure Biomarkers and Their Association with Length of Hospital Stay and Mortality: A Retrospective Observational Pilot Study

<b>Background/Objectives:</b> Chronic heart failure (CHF) remains a significant global health burden, with high morbidity, prolonged hospitalizations, and increased mortality. Traditional biomarkers such as NT-proBNP provide prognostic value; however, novel biomarker ratios may enhance risk stratification. This study evaluated the predictive utility of the NT-proBNP-to-albumin ratio (NTAR), red cell distribution width-to-eGFR ratio (RGR), and red cell distribution width-to-fibrinogen ratio (RFR) for hospital length of stay (LOS), extended hospitalization (ELOS), in-hospital mortality, and 6-month all-cause mortality. <b>Methods:</b> A retrospective observational pilot study was conducted on 382 CHF admissions (2022-2024) with comprehensive laboratory assessment. Biomarker performance was assessed through uni- and multivariate logistic regression, receiver operating characteristic curve, and Cox proportional hazards stepwise methods of analyses for refining predictive models. <b>Results:</b> NTAR and RGR emerged as significant predictors of hospitalization outcomes. NTAR demonstrated a moderate correlation with prolonged LOS (r = 0.45, <i>p</i> < 0.001) and was an independent predictor of ELOS (AUC = 0.697, OR = 2.438, <i>p</i> < 0.001), outperforming NT-proBNP. Additionally, NTAR significantly predicted in-hospital mortality (AUC = 0.768, OR = 4.461, <i>p</i> < 0.001) and 6-month all-cause mortality (AUC = 0.766, OR = 4.185, <i>p</i> < 0.001). RGR was the strongest predictor of in-hospital mortality (AUC = 0.785, HR = 2.18, <i>p</i> = 0.005), highlighting its role in renal dysfunction and erythropoietic alterations in CHF. The RFR observed prognostic value was minimal. <b>Conclusions:</b> In our study, NTAR and RGR offered valuable prognostic value underscoring the interplay of cardiac stress, nutritional status, and renal function in CHF prognosis. Further multicenter validation is warranted for these biomarkers.