Long-Term Outcomes and Risk Stratification of Patients With Heart Failure With Recovered Ejection Fraction

This study aimed to understand the long-term outcomes of patients with heart failure with recovered ejection fraction, identify predictors of adverse events, and develop a risk stratification model. From an academic healthcare system, we retrospectively identified 133 patients (median age 66, 38% female, 30% ischemic etiology) who had an improvement in left ventricular ejection fraction (LVEF) from <40% to ≥53%. Significant predictors of all-cause mortality, hospitalization, and future reduction in LVEF were identified through Cox regression analysis. Kaplan-Meier survival was 70% at 5 years. Freedom from hospitalization was 58% at 1 year, and the risk of future LVEF reduction to <40% was 28% at 3 years. Diuretic dose and B-type natriuretic peptide (BNP) at the time of LVEF recovery were the strongest predictors of mortality and hospitalization in multivariate-adjusted analysis (BNP hazard ratio 1.13 per 100 pg/ml increase [p <0.01]; furosemide-equivalent dose hazard ratio 1.19 per 40 mg increase [p = 0.02]). An all-cause mortality Cox proportional hazard risk model incorporating New York Heart Association functional class, BNP and diuretic dose at the time of recovery showed excellent risk discrimination (c-statistic 0.79) and calibration. In conclusion, patients with heart failure with recovered ejection fraction have heterogenous clinical outcomes and are not "cured." A risk model using New York Heart Association functional class, BNP, and diuretic dose can accurately stratify mortality risk.