Impact of LVEF and de novo heart failure on outcomes of SGLT2 inhibitor therapy initiated during acute heart failure

Background

Acute heart failure (AHF) carries high short-term mortality and rehospitalizationdespite advances in therapy. Sodium–glucose cotransporter 2 inhibitors (SGLT2i) improveoutcomes in chronic heart failure (HF), but their efficacy when initiated during AHFhospitalization is less clear. Effects may differ by de novo versus chronic HF and left ventricularejection fraction (LVEF).

Purpose

To assess the efficacy of SGLT2i initiated during AHF hospitalization and examinewhether LVEF or de novo HF influence outcomes.

Methods

A systematic review and meta-analysis of RCTs was performed across five databases.Trials initiating dapagliflozin or empagliflozin during AHF hospitalization alongside standardtherapy were included. Random-effects meta-analyses estimated pooled risk ratios (RRs) for all-cause mortality and HF hospitalization (HFH), and mixed-effects meta-regression evaluated theinfluence of LVEF and de novo HF.

Results

Eight RCTs involving 3596 patients (1817 SGLT2i; 1779 controls) were included. Theproportion of de novo HF ranged from 29.9% to 53.3%, mean LVEF was consistently ≤40%,and over half of patients were NYHA class III/IV. Initiation of SGLT2i during hospitalizationsignificantly reduced all-cause mortality (RR 0.63[0.46–0.85] P=0.003). No significantreduction was observed for HFH (RR 0.86[0.69–1.07] P=0.78).

Meta-regression analyses showed that neither baseline LVEF nor de novo HF prevalencesignificantly influenced outcomes. For all-cause mortality, the RR at LVEF 20% was 0.56[0.21–1.48], with each 10% increase in LVEF associated with a nonsignificant 9% rise in RR (1.09[0.58–2.02] P=0.79). For HF hospitalization, the baseline RR at LVEF 20% was 0.99[0.48–2.03], with each 10% increase in LVEF linked to a nonsignificant 10% reduction in RR(0.90[0.57–1.44] P=0.67). Regarding de novo HF, the RR for all-cause mortality at 0%prevalence was 0.49[0.04–5.73], with each 10% increase in prevalence associated with anonsignificant 6% increase in RR (1.06[0.59–1.92] P=0.84). For HF hospitalization, thebaseline RR at 0% de novo HF was 2.62[0.28–24.3], and each 10% increase in prevalence waslinked to a nonsignificant 23% reduction in RR (0.77 [0.46–1.30] P=0.32).

Conclusion

Initiation of SGLT2i during hospitalization for AHF significantly reduces short-term all-cause mortality. The benefits appear consistent across patients with varying baselineLVEF and proportions of de novo heart failure, suggesting broad applicability in AHFmanagement.For image description, please refer to the figure legend and surrounding text.