Semaglutide and Tirzepatide for Heart Failure With Preserved Ejection Fraction: The Trials, the FDA, and Who Benefits

Heart failure means the heart cannot move enough blood to meet the body's needs. Doctors sort it into types using the ejection fraction, which is the share of blood the main pumping chamber pushes out with each beat.

In heart failure with reduced ejection fraction (HFrEF), the heart muscle is weakened and pumps out too little blood with each contraction. In heart failure with preserved ejection fraction (HFpEF), the heart squeezes normally, but its walls are stiff and do not relax and fill properly between beats, so pressure builds up and fluid backs into the lungs and legs. People with HFpEF commonly feel short of breath, tire quickly during ordinary activity, and retain fluid. The trials described here enrolled people whose ejection fraction was at least 45% or 50%.

HFpEF is the most common type of heart failure, and obesity raises the risk of developing it. The American College of Cardiology's 2026 HFpEF pathway describes treating visceral fat, the fat packed around the internal organs, as part of HFpEF care. When the semaglutide trials began, no therapy had been approved to target obesity-related HFpEF, which is why trials of semaglutide (Wegovy, Ozempic) and tirzepatide (Zepbound, Mounjaro) in this group drew so much attention.

Novo Nordisk ran two one-year trials of semaglutide 2.4 mg, the Wegovy dose, given by injection once a week. STEP-HFpEF enrolled 529 people with HFpEF and a body mass index (BMI) of 30 or more who did not have diabetes, while STEP-HFpEF DM enrolled 616 people with the same heart condition who also had type 2 diabetes.

Both trials measured symptoms with the Kansas City Cardiomyopathy Questionnaire (KCCQ), a standard survey about shortness of breath, fatigue, swelling, and limits on daily activity. Its summary score runs from 0 to 100, and a higher score means fewer symptoms. In STEP-HFpEF, the score rose 16.6 points on semaglutide and 8.7 points on placebo, a difference of 7.8 points, while body weight fell 13.3% on semaglutide compared with 2.6% on placebo. People taking the drug also walked about 20 meters farther in a six-minute walking test than people on placebo.

The results in people with diabetes were similar but somewhat smaller: the KCCQ score rose 13.7 points on semaglutide and 6.4 on placebo, and weight fell 9.8% compared with 3.4%. In both trials, fewer people on semaglutide than on placebo reported serious adverse events.

A planned combined analysis of all 1,145 participants found that the symptom benefit held across age, sex, race, BMI, and ejection fraction. These trials were designed to measure symptoms and weight, however, and they were too small to show whether semaglutide prevents hospital stays or deaths.

Unlike the STEP trials, Eli Lilly's SUMMIT trial was designed to count heart failure events as a main outcome. It enrolled 731 people with HFpEF (an ejection fraction of 50% or more) and a BMI of 30 or more, with or without diabetes, who took tirzepatide at up to 15 mg a week or a placebo and were followed for a median of 104 weeks.

Cardiovascular death or a worsening heart failure event occurred in 9.9% of people on tirzepatide and 15.3% of people on placebo. That works out to a 38% lower relative risk (hazard ratio 0.62), or in absolute terms about 5 fewer people out of every 100 having one of these events over roughly two years. Worsening heart failure events on their own fell from 14.2% to 8.0%.

The benefit came from fewer heart failure events rather than fewer deaths, because cardiovascular deaths were rare in both groups (8 on tirzepatide and 5 on placebo), which is far too few to support any conclusion about survival. At one year, the KCCQ score rose 19.5 points on tirzepatide and 12.7 on placebo. More people stopped tirzepatide because of side effects, mostly digestive problems, at 6.3% compared with 1.4% on placebo.

Because the STEP-HFpEF trials were small, researchers in a Novo Nordisk-funded analysis pooled them with two larger trials. SELECT studied people with heart disease and overweight or obesity, and FLOW studied people with type 2 diabetes and kidney disease. Out of 22,282 people in the four trials, 3,743 had a history of HFpEF.

In that group, cardiovascular death or a heart failure event occurred in 5.4% of people on semaglutide and 7.5% on placebo (hazard ratio 0.69), and worsening heart failure events alone occurred in 2.8% compared with 4.7%. The analysis found no clear effect on cardiovascular death considered separately.

This was an analysis done after the trials ended, and in SELECT and FLOW, HFpEF was based on what study doctors wrote down rather than tests done for the study. The current Wegovy label shows that in SELECT as a whole, hospital stays or urgent visits for heart failure occurred in 1.4% of people on placebo and 1.1% on semaglutide, a difference that was not part of the trial's formal testing and could have been due to chance. The label states that the effect on heart failure "has not been established."

No randomized trial has compared the two drugs head to head in heart failure. A study published on October 29, 2026, in the American Diabetes Association journal Diabetes, Obesity, and CardioMetabolic CARE tried to estimate that comparison from insurance records. The team, led by John Ostrominski and Elisabetta Patorno at Brigham and Women's Hospital, used two U.S. commercial claims databases. They set up the analysis to mimic a trial, an approach called target trial emulation.

The study included 37,183 adults with HFpEF and type 2 diabetes who started tirzepatide (11,780) or semaglutide (25,387) between June 2022 and June 2025. The groups were balanced on measured traits using statistical weighting. Over a median of only 5.1 months, hospital stays for heart failure occurred at a rate of 19.7 per 1,000 person-years on tirzepatide and 25.5 on semaglutide (hazard ratio 0.80). Deaths from any cause were similar (hazard ratio 0.99).

This fits with tirzepatide producing more weight loss than semaglutide in other studies (our review of the head-to-head data). Still, a records study cannot rule out hidden differences between the people who were prescribed each drug, and five months is a short period for judging heart outcomes. The study also says nothing about people with HFpEF who do not have diabetes, who made up the whole STEP-HFpEF trial and part of SUMMIT.

Neither drug is approved by the FDA to treat heart failure. Both companies asked, and both withdrew their requests.

Novo Nordisk withdrew its Wegovy heart failure application in 2024 after talks with the FDA and said it would resubmit in 2025. The European Medicines Agency's advisory committee backed adding the STEP-HFpEF symptom results to the European Wegovy label in September 2024. The current U.S. Wegovy label, revised in June 2026, does not list heart failure as an indication.

Lilly withdrew its U.S. application for tirzepatide in HFpEF in May 2025. Lilly's chief scientific officer, Dan Skovronsky, said the FDA wanted "an additional confirmatory clinical trial." The current Zepbound label lists weight management and obstructive sleep apnea only.

In practice, many people with HFpEF and obesity already qualify for these drugs under their weight-loss approval. Insurance coverage for weight loss varies widely, and a heart failure diagnosis alone does not guarantee coverage.

Professional groups have moved faster than the FDA. In 2025, the American College of Cardiology (ACC) published a statement on treating obesity in people with heart failure. It called the early results for semaglutide and tirzepatide in HFpEF promising. It also listed open questions, including the safety of these drugs in HFrEF and whether the benefits come mainly from weight loss.

The ACC's 2026 HFpEF decision pathway, published online in July, goes further. According to HCPLive's summary, it recommends semaglutide or tirzepatide for people with HFpEF and a BMI of 30 or more. It names an SGLT2 inhibitor and finerenone as the base of drug treatment. The 2026 ADA and EASD diabetes consensus also says semaglutide or tirzepatide should be used to ease symptoms in people with diabetes, obesity, and HFpEF (our guide to that consensus).

The encouraging results in HFpEF do not carry over to HFrEF, where the heart muscle itself is weak. Two older trials of liraglutide, an earlier GLP-1 drug injected daily, raised concerns about this group.

The FIGHT trial enrolled 300 people who had recently been hospitalized with HFrEF, with a median ejection fraction of 25%. Liraglutide did not improve their clinical stability over 180 days, and heart failure readmissions were 41% on liraglutide compared with 34% on placebo, a gap that was not statistically significant. The LIVE trial, in 241 people with stable HFrEF, found no change in the heart's pumping function, while heart rate rose by 7 beats per minute and serious cardiac events were more common on liraglutide (10% compared with 3%).

These trials were small and tested an older drug, so they do not prove harm. They do explain why the ACC lists safety in HFrEF as an open question, and why people with a weakened heart muscle should not assume the HFpEF results apply to them. GLP-1 drugs also raise resting heart rate in general (our explainer).

If you have heart failure along with obesity, a GLP-1 drug may be worth discussing with the cardiologist who manages your heart treatment. Questions to bring to that appointment:

• What is my current ejection fraction, and does that mean I have HFpEF or HFrEF?
• Do I resemble the people studied in STEP-HFpEF or SUMMIT, including their BMI of 30 or more and their level of symptoms?
• Would semaglutide or tirzepatide make more sense for me, considering my other medical conditions and the absence of a head-to-head randomized trial?
• How should my diuretics (water pills) and blood pressure medicines be adjusted as my weight comes down?
• How will we watch for dehydration and kidney problems if I develop nausea, vomiting, or diarrhea?
• Will my insurance cover the drug under its weight-loss approval, given that it is not approved for heart failure?
• Which measures will we track to judge whether it is helping, such as symptoms, walking distance, weight, or hospital visits?

Do not start, stop, or change heart medicines on your own. Weight loss can lower blood pressure and change how much fluid your body holds, so the doses of other medicines may need adjusting. For readers over 65, our guide to GLP-1 drugs after 65 covers related issues.