Early in-hospital initiation of dapagliflozin versus empagliflozin in acute heart failure: impact on in-hospital survival and safety

Introduction

Acute heart failure (AHF) remains a major cause of morbidity and mortality worldwide, with persistently high in-hospital death rates despite advances in guideline-directed therapy (1).

Sodium–glucose cotransporter-2 (SGLT2) inhibitors have demonstrated rapid hemodynamic and renal benefits across the spectrum of left ventricular ejection fraction (2). Although both dapagliflozin and empagliflozin have proven efficacy in heart failure, direct comparative data, particularly when initiated early during cardiac ICU admission, remain limited (3).

Objective

To compare early in-hospital initiation of dapagliflozin versus empagliflozin in patients admitted for acute heart failure and evaluate their impact on in-hospital survival and safety.

Methods

We conducted a retrospective cohort study of 1,800 consecutive adults admitted to the cardiac ICU for acute heart failure (de novo or acute decompensation) between July 2023 and April 2025. Early initiation of SGLT2 inhibitors was defined as treatment within 48 hours of admission. After applying exclusion criteria, including dialysis, prior transplantation, eGFR <30 mL/min/1.73 m², cardiogenic shock, diabetic ketoacidosis, symptomatic hypotension, or early transfer, 1,546 patients were included. Patients were categorised into dapagliflozin (n=450), empagliflozin (n=398), or no early SGLT2 inhibitor groups. The primary endpoint was all-cause in-hospital mortality. Multivariable logistic regression assessed independent associations with mortality.

Results

Among 1,546 included patients (mean age 63.4 ± 15.1 years; 57.5% male; mean LVEF 44.4 ± 14.6%), early SGLT2 inhibitors were initiated in 848 patients. Overall, in-hospital survival did not differ significantly between dapagliflozin and empagliflozin (P=0.29). Subgroup analyses showed higher survival with dapagliflozin in patients with reduced LVEF (76% vs 61%; P=0.001), whereas empagliflozin was associated with higher survival in patients with right ventricular dysfunction (36.8% vs 11.8%; P=0.02) and chronic kidney disease (33.3% vs 25%; P=0.032). Hypotension was rare (0% empagliflozin vs 0.5% dapagliflozin; P<0.001), and rates of acute kidney injury were similar between groups. Low event rates were observed in patients receiving combined SGLT2 inhibitor and sacubitril/valsartan therapy.

Conclusion

Early initiation of dapagliflozin and empagliflozin was associated with comparable overall survival and favourable safety profiles. Subgroup findings suggest differential benefits according to ventricular function and renal status, supporting early SGLT2 inhibitor use in acute heart failure.