Main Results Of The Empagliflozin Evaluation By Measuring Impact On Hemodynamics In Patients With Heart Failure Trial

Introduction Sodium glucose cotransporter 2 inhibitors (SGLT2i) prevent heart failure (HF) in patients with Type 2 Diabetes (T2D), and reduce CV death or worsening HF in HF and reduced ejection fraction (HFrEF), regardless of T2D. Mechanisms of HF benefits are unclear, and effects of SGLT2i on filling pressures are not known. EMBRACE-HF was designed to address this knowledge gap. Methods EMBRACE-HF was an investigator-initiated, multi-center, double-blind, randomized trial. From July 2017 to November 2019, 65 patients with HF (regardless of EF, with or without T2D) and previously implanted pulmonary artery (PA) pressure sensor (CardioMEMS) were randomized across 10 US centers to empagliflozin (empa) 10 mg daily or placebo (pbo) and treated for 12 weeks. Primary endpoint was change in PA diastolic pressure (PADP) from baseline to end of treatment (average PADP weeks 8-12). Results 93 patients were screened, and 65 were randomized (33 empa, 32 pbo). Mean age was 66 yrs, 63% were male, 52% had T2D, 54% NYHA class III/IV; LVEF 44%, NTproBNP 1665 pg/mL and PADP 20 mmHg. Empa significantly reduced PADP, with effects beginning at week 1, and amplified over time (Figure 1). Average PADP (week 8-12) was 1.5 mmHg lower (95% CI: 0.2, 2.8; p = 0.02); and at week 12, PADP was 1.7 mmHg lower (95% CI: 0.3, 3.2; p = 0.02) with empa vs pbo. Results were consistent for PA systolic and PA mean pressures. There was no difference in loop diuretic dose (measured in daily furosemide equivalents) at baseline (90 mg empa, 85 mg pbo; p = 0.51), week 12 (92 mg empa, 86 mg pbo; p = 0.24), or any interim time point. Majority of patients in both groups (76% empa, 81% placebo; p = 0.89) had no change in loop diuretic dose during the study. Conclusion EMBRACE-HF is the first randomized controlled trial to show a direct beneficial effect of SGLT2i on decongestion in patients with HF (reduced or preserved EF), with rapid reductions in PA pressures that were amplified over time and appeared to be independent of loop diuretic management.