Early versus delayed or no empagliflozin initiation after hospitalisation for heart failure in the United States: A target trial emulation

BACKGROUND: Sodium-glucose cotransporter-2 inhibitors (SGLT2is) are a foundational therapy for heart failure (HF), yet initiation is frequently delayed in clinical practice. Evidence comparing outcomes associated with early versus delayed or no initiation following hospitalisation for heart failure (HHF) remains limited. METHODS: This retrospective cohort study used the Optum de-identified Clinformatics Data Mart and included adults hospitalised for HF between September 2021 and December 2023. A target trial emulation using the clone-censor-weight approach compared patients according to the timing of empagliflozin initiation: during HHF or within 0-7 days post-discharge (early), 8-91 days post-discharge (intermediate), and 92-365 days post-discharge or no initiation within 1 year of discharge (delayed or no initiation). Outcomes assessed during the first year after discharge included all-cause mortality, worsening HF events, and all-cause hospitalisations. Health care resource utilisation (HCRU) was assessed using landmark analyses at Days 7 and 91 post-discharge. RESULTS: Among 99,738 eligible patients (mean age [standard deviation] 78.0 [9.2] years; 54.2% female), early and intermediate empagliflozin initiation were associated with lower 1-year post-discharge all-cause mortality than delayed or no initiation (risk ratio [RR] 0.73, 95% confidence interval [CI]: 0.69-0.78; and RR 0.93, 95% CI: 0.89-0.97, respectively). Further, 1-year absolute mortality risk was 24.7% with early, 31.1% with intermediate, and 33.6% with delayed or no initiation. Early initiation was also associated with lower risks of worsening HF events (RR 0.89, 95% CI: 0.84-0.94), HF rehospitalisations (RR 0.86, 95% CI: 0.81-0.92), emergency department (ED) visits (RR 0.87, 95% CI: 0.82-0.93), and all-cause hospitalisations (RR 0.91, 95% CI: 0.88-0.94). Associations for intermediate initiation were in the same direction but smaller in magnitude. In landmark analyses, early initiation was associated with lower acute care utilisation rates, including all-cause and HF-related hospitalisations (for 7 days analysis: incidence rate ratio [IRR] 0.80, 95% CI: 0.77-0.84; IRR 0.76, 95% CI: 0.70-0.83, respectively), and all-cause and HF-related ED visits (IRR 0.87, 95% CI 0.83-0.91; IRR 0.78, 95% CI 0.72-0.86, respectively), compared with no initiation. CONCLUSIONS: Early empagliflozin initiation was associated with lower risks of 1-year post-discharge all-cause mortality, worsening HF events, and all-cause hospitalisations compared with delayed or no initiation, and with lower acute care utilisation rates compared with no initiation.