Clinical outcomes and latent risk in HFimpEF: insights from real-world data

Introduction

A subset of patients with heart failure and reduced ejection fraction (HFrEF) demonstrates significant functional and structural recovery under optimal medical therapy (GDMT), transitioning to heart failure with improved EF (HFimpEF).

Per ESC criteria, HFimpEF requires baseline LVEF ≤40%, follow-up LVEF >40% and an absolute rise ≥10 points. However, characteristics, EF trajectories, treatment patterns, and residual risk in HFimpEF remain incompletely defined.

Methods

We conducted a prospective study of 270 de novo HFrEF patients at a tertiary outpatient clinic. HFimpEF was defined per ESC criteria. Baseline clinical, laboratory, echocardiographic data, GDMT use and doses were collected. LVEF was measured at baseline, 3 months, 1 year, and last follow-up. The primary endpoint was a 3-year composite of all-cause mortality and HF hospitalization (HHF). Kaplan–Meier and Cox regression were used to access EF improvement prognostic impact.

Results

Overall, 112 patients (41%) recovered EF; 158 (59%) had persistent HFrEF. HFimpEF patients were slightly younger (64 vs 67 years, p=0.04) and more often female (37% vs 23%, p=0.02), while comorbidities, AF, HF etiology, NT-proBNP, renal function, and baseline LVEF were similar. HFimpEF patients had smaller LAVI (39 vs 46 mL/m², p<0.001) and more often reached maximum ARNI (42% vs 34%, p=0.02) and MRA dosing (90% vs 81%, p=0.04).

Despite identical baseline LVEF, patients with HFimpEF markedly improved EF at 3 month and kept improving slightly throughout follow-up, whereas patients with persistent HFrEF showed modest EF improvement at the 3-month mark, and only non-relevant improvement beyond 3-month - Figure 1. HFimpEF patients presented better prognosis, with almost no events when compared to persistent HFrEF patients: HFimpEF conferred a 79% lower risk of death/HHF (HR 0.21, 95% CI 0.08–0.55) – Figure 2.

Within HFimpEF, only 6 adverse events occurred (2 deaths, 4 HHF). Those with events showed higher frailty (CFS 4.8 vs 3.1), lower eGFR (44 vs 74 mL/min), and much higher NT-proBNP (24,289 vs 4,759 pg/mL), identifying these as potential markers of residual risk.

Conclusion

HFimpEF is common, confers a favorable prognosis and seems to be associated with higher GDMT doses. Although overall risk of events is low in this population, frailty, renal impairment and elevated NT-proBNP may identify HFimpEF patients needing closer follow-up and alternative therapies to further reduce residual risk.For image description, please refer to the figure legend and surrounding text.