Empagliflozin after Acute Myocardial Infarction (EMPACT-MI)

Authors, Journal, Affiliations, Type, DOI

• Authors: Javed Butler, W. Schuyler Jones, Jacob A. Udell, Stefan D. Anker, Mark C. Petrie, Deepak L. Bhatt, Adrian F. Hernandez, et al.
• Journal: New England Journal of Medicine
• Affiliations: Baylor Scott and White Research Institute, Dallas; Duke University Medical Center; Duke Clinical Research Institute; Women's College Hospital Toronto; German Heart Center Charité Berlin; and 451 sites in 22 countries
• Type: International, event-driven, double-blind, randomised, placebo-controlled trial
• DOI: https://doi.org/10.1056/NEJMoa2314051
• Funding: Boehringer Ingelheim and Eli Lilly

Overview

EMPACT-MI randomised 6,522 patients hospitalised for acute MI at increased risk of HF (new LVEF <45% and/or in-hospital congestion with treatment) to empagliflozin 10 mg OD or placebo within 14 days of admission. The primary composite of HF hospitalisation or all-cause death was neutral (8.2% vs 9.1%; HR 0.90; 95% CI 0.76–1.06; P=0.21) over median 17.9 months. HF hospitalisation alone was significantly reduced (HR 0.77; 95% CI 0.60–0.98), and total HF hospitalisations were markedly lower in an exploratory analysis (rate ratio 0.67), while mortality was unchanged. The trial establishes that SGLT2 inhibitors do not prevent early de novo HF or death when initiated within 2 weeks of MI, even in patients enriched for HF risk — a critical boundary condition for the SGLT2i benefit zone, in contrast to their proven efficacy in established HF.

Keywords

Empagliflozin; SGLT2 inhibitor; acute myocardial infarction; heart failure prevention; HF hospitalisation; all-cause death; EMPACT-MI; post-MI; LVEF; congestion; primary endpoint neutral

Key Takeaways

Trial Design and Population

• Event-driven, double-blind RCT; 1:1 empagliflozin 10 mg OD vs placebo; randomised within 14 days of MI admission (median 5 days, IQR 3–8)
• Stratified by T2DM status and geographic region (North America, Latin America, Europe, Asia)
• Inclusion: Acute MI hospitalisation + EITHER new LVEF <45% OR in-hospital signs/symptoms of congestion requiring treatment + ≥1 HF enrichment factor
• Enrichment factors (≥1 required): age ≥65; LVEF <35%; prior MI, AF, or T2DM; eGFR <60; elevated NT-proBNP or BNP; elevated PA/RV pressure; 3-vessel CAD; PAD; no revascularisation for index MI
• Exclusions: prior HF diagnosis; current or planned SGLT2i use
• Population profile: 78.4% LVEF <45%; 57.0% had congestion treated; 74.8% STEMI; 89.3% revascularised; 31.9% T2DM; 50.0% aged ≥65; 70.5% had >1 enrichment factor
• 6,328 patients (97.0%) followed to end of trial for primary endpoint events; 99.2% with vital status data
• 6.7% started open-label SGLT2i during trial (6.2% empagliflozin arm; 7.2% placebo arm)

Primary Endpoint — Neutral

• HF hospitalisation or all-cause death (time to first event): 8.2% empagliflozin vs 9.1% placebo
• HR 0.90 (95% CI 0.76–1.06); P=0.21 — primary endpoint NOT met
• Incidence rates: 5.9 vs 6.6 events per 100 patient-years
• Death composed 52% of primary endpoint events and was similar in both arms — largely non-modifiable by SGLT2i in the acute post-MI setting

Primary Endpoint Components

• HF hospitalisation (first event): 3.6% vs 4.7%; HR 0.77 (95% CI 0.60–0.98) — statistically significant as exploratory component
• All-cause death: 5.2% vs 5.5%; HR 0.96 (95% CI 0.78–1.19) — NS
• CV death: 4.0% vs 4.0%; HR 1.03 (95% CI 0.81–1.31) — NS, identical rates

Key Secondary Endpoints (All Neutral — Hierarchical Procedure)

• Total HF hospitalisations or death from any cause: rate ratio 0.87 (95% CI 0.68–1.10)
• Total nonelective CV hospitalisations or death: rate ratio 0.92 (95% CI 0.78–1.07)
• Total nonelective hospitalisations or death: rate ratio 0.87 (95% CI 0.77–1.00)
• Total MI hospitalisations or death: rate ratio 1.06 (95% CI 0.83–1.35)

Exploratory Analysis — Total HF Hospitalisations

• Total HF hospitalisation count: 148 vs 207 events; rate ratio 0.67 (95% CI 0.51–0.89)
• This exploratory finding suggests a real anti-HF signal from empagliflozin, consistent in direction with established HF trials, but insufficient to reach the composite primary endpoint

Subgroup Analysis

• Primary endpoint results consistent across all prespecified subgroups: diabetes status, geographic region, eGFR strata, LVEF category, STEMI vs NSTEMI, atrial fibrillation, peripheral artery disease, revascularisation status
• No subgroup demonstrated a significant interaction with treatment effect

Safety

• Serious adverse events: 23.7% empagliflozin vs 24.7% placebo — similar
• Adverse events leading to permanent discontinuation: 3.8% in both groups
• Contrast-induced AKI: 0.2% vs 0.3% — no excess with empagliflozin
• Safety profile consistent with known empagliflozin experience; no unexpected signals in this early post-MI population

Why Did the Primary Endpoint Fail?

1. Non-modifiable mortality: Deaths (52% of events) occurred from causes not amenable to SGLT2i — stent thrombosis, recurrent MI, mechanical complications, scar arrhythmia in the first 30 days
2. Lower-than-expected event rates: 8.2–9.1% vs historically predicted higher rates — attributed to modern revascularisation (90%), optimal GDMT, and reduced HF hospitalisation during COVID-19 pandemic
3. Stunned/reversible myocardium: 74.8% STEMI with ~90% revascularisation; many patients had transient LVEF reduction that recovered after successful PCI — not the persistent substrate that drives HF benefit in chronic HF trials
4. Outpatient HF events excluded: Worsening HF requiring outpatient IV diuretics or oral diuretic intensification was not counted — if included (as in DAPA-HF), total HF event burden would have been higher and empagliflozin benefit more visible

Limitations

• End-point events assessed by site investigators (not centrally adjudicated), though prespecified definitions and blinded assessors were used
• Outpatient heart failure events not included in the primary endpoint
• Only focused safety data collected (serious AEs, discontinuation-related AEs, specific AEs of interest)
• Underrepresentation of women, older adults, and racial/ethnic minorities
• COVID-19 pandemic and regional wars reduced hospitalisation events, lowering observed event rates below projections
• Open-label SGLT2i initiation in 6.7% during trial period (slightly higher in placebo arm 7.2% vs 6.2%) may have diluted the difference

Key Concepts Mentioned

• concepts/SGLT2-Inhibitors-in-CKD — boundary condition: SGLT2i do not benefit primary de novo HF prevention post-MI despite proven benefit in established HF
• concepts/Cardiorenal-Syndrome — post-MI renal function preserved with empagliflozin (no excess AKI)

Key Entities Mentioned

• entities/Acute-Coronary-Syndrome — EMPACT-MI enrolled exclusively post-MI patients
• entities/Heart-Failure — HF hospitalisation was the key modifiable endpoint component; reduction with empagliflozin (HR 0.77) consistent with HF trials
• entities/HFrEF — post-MI LVEF <45% population; contrast with EMPEROR-Reduced (established HFrEF, HR 0.75)

Wiki Pages Updated

• Created wiki/sources/empagliflozin-empactmi-nejm-2024.md
• Updated wiki/entities/Acute-Coronary-Syndrome.md
• Updated wiki/concepts/SGLT2-Inhibitors-in-CKD.md
• Updated wiki/sourceindex.md
• Updated wiki/wikiindex.md