Evidence Hub

Early empagliflozin initiation after hospitalization for heart failure in the United States: a target trial emulation

S. Greene, G. Tskhvarashvil, C. Carlsen, P. Tran, R. Hermans, S. Enshaeifar, A. Adam, M. Martinez Traba et al.

European Journal of Heart Failure, 2026

99,738 participants

Abstract

Sodium-glucose cotransporter-2 inhibitors (SGLT2i), such as empagliflozin, have been shown to reduce mortality and heart failure (HF) events across the entire spectrum of ejection fraction. Despite evidence from the EMPULSE trial and guideline recommendations for early initiation, real-world implementation varies substantially, with many patients experiencing delays in treatment initiation.

This study aimed to evaluate real-world 1-year clinical outcomes of early versus delayed empagliflozin initiation after HF hospitalisation.

This retrospective cohort study utilised Optum’s de-identified Clinformatics Data Mart database (Sep 2021–Dec 2023). Adults (≥18 years) hospitalised with HF in the U.S., with ≥1-year continuous enrolment and no prior SGLT2i use were included. Patients with type 1 diabetes, diabetic ketoacidosis, dialysis, LVAD, cardiac/renal transplant, or eGFR 91 days or no initiation. Study outcomes were all-cause mortality, HF events (HF rehospitalisation and emergency department visits with HF), and all-cause hospitalisation. Risk Ratios (RR) were calculated using a pooled logistic regression model with 95% confidence intervals (CI), using robust sandwich variance estimators. Initiation >91 days or no initiation served as a reference.

Among 99,738 patients hospitalised for HF (mean age 78 y; 54% female), 3,873 initiated empagliflozin pre-discharge or within 7 days post-discharge, 3,087 initiated between 8–91 days, and 92,778 had delayed or no initiation. Baseline…

How strong is the evidence?

Evidence Hub reads this as a cohort study, done in people. Its evidence level is 3 of 5 (Moderate · trial / cohort / secondary analysis). Trials without a large randomized design, cohorts and secondary analyses give moderate evidence. Reported size: 99,738 participants or records.

How the five levels of evidence work. The level is assigned automatically from the abstract and can be wrong, so check the paper.

Read the paper

Related studies

Go deeper in Evidence Hub

Evidence Hub adds a study data sheet (design, sample size, evidence level, key numbers), an AI summary and a chat grounded in the paper's text.

Open this study Sign up free