Empagliflozin Misses Primary Endpoint in Post-MI Patient Population, EMPACT-MI Trial

Published on: 

Results of EMPACT-MI suggest use of empagliflozin was not associated with a statistically significant reduction in a composite of death and heart failure hospitalization.

Use of empagliflozin did not lead to a significant reduction in the composite of hospitalizations for heart failure or all-cause mortality in patients at increased risk for heart failure after acute myocardial infarction, according to results of the EMPACT-MI trial.1

Presented at the American College of Cardiology 2024 (ACC.24) Annual Scientific Session, results of the trial, which come less than 6 months after results of DAPA-MI were presented at the American Heart Association 2023 Scientific Sessions, detail a nonsignificant trend toward benefit with use of empagliflozin among this patient population driven by reductions in heart failure hospitalizations.1,2

“We did not reach our primary endpoint—there was a 10% relative risk reduction in the primary endpoint that was not statistically significant, a P-value was 0.21," said lead investigator Javed Butler, MD, president of the Baylor Scott and White Research Institute in Dallas, distinguished professor of medicine at the University of Mississippi, in an ACC.24 episode of Don’t Miss a Beat. “Now, when we dissected the primary endpoint into its components, there was no signal of mortality benefit, but there was a 23% relative risk reduction [that was] statistically significant in failure hospitalization.”

In the 11 years since canagliflozin (Invokana) received approval for the treatment of adults with type 2 diabetes, the SGLT2 inhibitor class, driven primarily by empagliflozin and dapagliflozin, has been put to the test for a multitude of different cardiometabolic conditions. Like dominoes, the class proved its benefit for 1 condition after the next starting with heart failure with reduced ejection fraction, then preserved ejection fraction, and eventually chronic kidney disease. Leading into the AHA 2023, DAPA-MI offered a glimpse of the class’s safety and efficacy profile in a fourth patient population defined by presence of a recent myocardial infarction.1,2,3

In the trial, use of dapagliflozin was examined against placebo plus standard of care for a hierarchal composite endpoint of all-cause mortality, heart failure hospitalization, nonfatal myocardial infarction, atrial fibrillation/flutter, type 2 diabetes, change in NYHA class, and weight loss. Results of the trial suggested use was associated with a positive win ratio relative to placebo (win ratio, 1.34; 95% Confidence Interval [CI], 1.20 to 1.50; P < .001). However, as this effect appeared to be driven by items in the latter half of the composite endpoint, many in the community questioned the benefit of SGLT2 inhibition in this patient population and looked ahead to the EMPACT-MI trial for more clarity.2

Unlike the aforementioned DAPA-MI trial, the primary outcome of interest for the EMPACT-MI trial was a composite of first hospitalization for heart failure and all-cause mortality assessed in a time-to-first-event analysis. Another key difference in the trials was the enrollment of a patient population with greater risk in the EMPACT-MI trial, which included patients experiencing an acute myocardial infarction within 14 days before randomization with evidence of a newly developed left ventricular ejection fraction of less than 45% or signs or symptoms of congestion that resulted in treatment during the index hospitalization plus an additional clinical factor known to be associated with hospitalization for heart failure, including but not limited to type 2 diabetes. In contrast, DAPA-MI enrolled a patient population with acute myocardial infarction without a history of diabetes or chronic heart failure.1,2

An event-driven, double-blind, randomized, placebo-controlled trial, EMPACT-MI randomized 6522 patients in a 1:1 ratio to empagliflozin or placebo in addition to standard of care within 14 days of admission, with 3260 randomized to empagliflozin and 3262 randomized to placebo.1

This cohort had a mean age of 64 years, 75.1% were male, and 83.6% were White. The median follow-up of the study population was 17.9 months and the median time to randomization after acute myocardial infarction was 5 days. Investigators pointed out 31.9% of the cohort had diabetes, 74.3% of patients presented with a ST elevation acute myocardial infarction, 57.0% of patients had signs or symptoms of congestion requiring treatment, and 78.4% had newly depressed left ventricular ejection fraction less than 45%.1

Results of the primary outcome analysis suggested a first heart failure hospitalization or death occurred among 9.1% (n=267) of the placebo arm and 8.2% (n=298) of the empagliflozin arms. Investigators pointed out these correlate to incidence rates of 6.6 and 5.5 events, respectively, per 100 patient-years (Hazard Ratio [HR], 0.90; 95% Confidence Interval [CI], 0.76 to 1.06; P= .21).1

Further analysis of individual components of the primary outcome indicated a first hospitalization for heart failure occurred among 4.7% (n=153) of the placebo arm and 3.6% (n=118) of the empagliflozin arm (HR, 0.77; 95% CI, 0.60 to 0.98). When assessing all-cause mortality, such an event occurred among 5.5% (n=178) of the placebo arm and 5.2% (n=169) of the empagliflozin arm (HR, 0.96; 95% CI, 0.78 to 1.19).1

“To have a 25% to 30% reduction in heart failure hospitalizations is pretty clinically meaningful, but if you put it together with all-cause mortality, it was not a positive study for our primary endpoint,” said Butler, in a press release.4


  1. Butler J, Anker SD, Bhatt DL, et al. Empagliflozin After Acute Myocardial Infarction: Results Of The EMPACT-MI Trial. Presented at: American College of Cardiology (ACC.24) Annual Scientific Session. April 6 – 8, 2024. Atlanta, GA.
  2. James S, Erlinge D, Storey RF, et al. Dapagliflozin in Myocardial Infarction without Diabetes or Heart Failure. NEJM Evid. 2024;3(2):EVIDoa2300286. doi:10.1056/EVIDoa2300286
  3. Campbell P, Iapoce C. 10 years of SGLT2 inhibitors: A Decade of Redefining Cardiometabolic Care. HCP Live. March 29, 2023. Accessed April 6, 2024.
  4. American College of Cardiology. SGLT-2 Inhibitors Show Mixed Results After Heart Attack. April 6, 2024. Accessed April 6, 2024.