Role of mitral regurgitation in a new predictive model for mortality and hospitalization risk in heart failure

Background Heart failure (HF) carries a high burden of in-hospital readmission and mortality, with 1-year rates approximating 35% and 20%, respectively. Several risk factors are known to be associated with poor prognosis, including age, biomarkers of renal impairment, congestion, and echocardiographic parameters of cardiac remodeling. Nevertheless, predictive models in the era of modern treatments are still lacking. Purpose The primary objective was to develop an algorithm to predict a composite endpoint of 12-month re-hospitalization or death after acute HF hospitalization. Methods A cohort of 231 patients admitted for acute HF (34% HFpEF; 66% HFrEF, mean age 78 years; 35.9% male) at our Institution, between January 2022 and December 2023, was considered. Demographic, biochemical, and echocardiographic data were retrospectively collected. Univariate and multivariate logistic regression analyses were used to identify variables associated with the 12-month composite endpoint of re-hospitalization or death. These variables were then incorporated into a logistic regression-based risk calculator to assess the primary endpoint. Results Age, creatinine, NT-proBNP, hemoglobin at admission and more than moderate mitral regurgitation were found to be significantly associated with re-hospitalization and mortality events at multivariate analysis. A logistic regression-based risk-score was developed to predict the likelihood of being alive and free from HF hospitalization at 12 months [Risk-formula = -6.1+(0.09*Age)-(0.2*Hb)+(0.02 *Creatinine)+(0.34*Mitral Regurgitation)+(0.00003*NT-proBNP)]. Youden's Index defined the optimal cut-off value (-0.02) to distinguish between high- and low-risk individuals (AUROC: 76.6%; 64% sensitivity and 79% specificity). Patients classified as "High-Risk" according to risk-score had significantly lower 1-year overall survival rates compared to "Low-Risk" patients (69.6% vs. 95.1%, p<0.01) and a shorter median overall survival (χ²: 28.1, Log-rank p<0.001, Fig. 1A). Similarly, evaluating the composite endpoint of mortality and HF readmission, "High-Risk" patients had significantly lower 1-year event-free survival rates than "Low-Risk" patients (42.9% vs. 82.9%, p<0.01) along with a shorter median event-free survival (χ²: 23.7, p<0.01, Fig. 1B). Conclusions We proposed a risk prediction score based on a modern HF population, including a large proportion of HFpEF subjects. In addition to the expected biomarkers of renal impairment, anemia, and congestion, significant mitral regurgitation emerged as the strongest prognostic factor, confirming the pathophysiological role of backward volume overload. Remarkably, the degree of systolic dysfunction was not prognostically relevant. The score, still pending of prospective validation in a larger and independent cohort, can contribute to early risk stratification and treatment optimization of HF patients in the modern era.