Tirzepatide HFpEF imaging substudy finds smaller atria and right ventricle in SUMMIT
A cardiac MRI analysis of 94 SUMMIT patients found modest falls in atrial volumes and right ventricular size on tirzepatide against placebo. The lead investigator says the changes are small and visible only on average.
What's new
A SUMMIT cardiac MRI substudy presented at AHA 2026 found tirzepatide reduced left and right atrial volumes and right ventricular mass and volumes against placebo at 52 weeks.
Why it matters
It suggests a structural change behind SUMMIT's lower heart failure events, and the atrial changes tracked with better symptom scores.
What's next
The full results have yet to appear in a peer reviewed journal, and the analysis cannot separate weight loss from any direct drug effect.
Tirzepatide shrank both atria and the right ventricle in people with obesity related heart failure with preserved ejection fraction, a cardiac MRI substudy of the SUMMIT trial reports. The tirzepatide HFpEF imaging work, led by Christopher Kramer of the University of Virginia, was presented at the American Heart Association Scientific Sessions 2026 in Chicago. It covered 94 patients with usable scans at baseline and one year. The changes were small, and Kramer cautioned that they appear only when patients are averaged together.
What the SUMMIT imaging substudy measured
SUMMIT was a randomized, double blind Phase 3 trial that compared weekly tirzepatide with placebo in adults with heart failure, an ejection fraction of 50% or more, raised NT proBNP and a body mass index of at least 30. Doses were raised in steps every four weeks toward a set maximum. Of the trial's 731 patients, 175 had cardiac magnetic resonance imaging at the start and again at 52 weeks, according to HCPLive's report of the presentation. Only 94 had images good enough for analysis at both visits and a follow up scan taken while still on treatment or within two weeks of finishing. Of those, 42 had received tirzepatide and 52 had received placebo. The team measured chamber volumes and right ventricular mass in a full volumetric approach.
Kramer explained why the team used magnetic resonance rather than ultrasound. "Most studies previously with HFpEF therapies have been with echo, and echo is less sensitive for assessment of the right ventricle, left atrium, and right atrium," he told HCPLive. Cardiac MRI measures size and pumping more precisely, so a smaller group can show a measurable difference. That precision is the reason a study of fewer than 100 people can report changes of a few millilitres. The same SUMMIT imaging group earlier reported a fall of about 11 grams in left ventricular mass with tirzepatide, together with less fat around the heart, in an analysis that UVA Health summarized. That earlier analysis found no effect on left atrial volumes.
Tirzepatide HFpEF imaging results
The new analysis found that tirzepatide reduced the largest volume of the left atrium by 8 mL and of the right atrium by 10 mL compared with placebo. It also cut right ventricular mass by 5 grams, right ventricular end diastolic volume by 10 mL and end systolic volume by 7 mL. All five differences reached statistical significance, though several confidence intervals came close to zero. The right ventricular changes tracked with falls in left ventricular mass and in fat around the heart. The shrinking atria tracked with gains on the Kansas City Cardiomyopathy Questionnaire, a patient reported score of symptoms and limits. A correlation of this kind does not show that the smaller atria caused patients to feel better.
The shrinking atria tracked with gains on the Kansas City Cardiomyopathy Questionnaire, a patient reported score of symptoms and limits.
Kramer read the results as support both for the drug and for the imaging method. He did not suggest routine scans for patients. "I think that this points out the power of CMR in clinical trials, but I'm not advocating for doing yearly CMRs in patients with HFpEF," he told HCPLive. "The changes are subtle, small, and only seen when you average patients together." The left atrial result differs from the earlier SUMMIT imaging report, which found no change in that chamber, and the new work used a fuller volumetric method and a smaller analysed group. The full findings have not yet appeared in a peer reviewed journal, so the numbers come from the presentation as reported.
| Measure against placebo at 52 weeks | Change with tirzepatide | 95% CI |
|---|---|---|
| Left atrial maximal volume | 8 mL smaller | 2 to 15 mL |
| Right atrial maximal volume | 10 mL smaller | 1 to 18 mL |
| Right ventricular mass | 5 g lower | 1 to 10 g |
| Right ventricular end diastolic volume | 10 mL smaller | 2 to 19 mL |
| Right ventricular end systolic volume | 7 mL smaller | 0 to 13 mL |
How the findings fit the main SUMMIT trial
The parent SUMMIT trial, published in the New England Journal of Medicine in 2025, randomized 731 patients, 364 to tirzepatide and 367 to placebo. Over a median follow up of 104 weeks, the composite of cardiovascular death or a worsening heart failure event occurred in 36 patients on tirzepatide and 56 on placebo, a hazard ratio of 0.62. At 52 weeks the symptom score rose by 19.5 points on tirzepatide and 12.7 points on placebo, a difference of 6.9 points. Both primary endpoints favoured tirzepatide, and the imaging substudy drew its patients from this same randomized population. The trial is registered on ClinicalTrials.gov as NCT04847557.
The imaging data add a possible mechanism to those outcomes. In heart failure with preserved ejection fraction linked to obesity, the heart chambers can enlarge and stiffen, and extra fat around the heart may add to the load. Smaller atria and a lighter right ventricle are consistent with lower filling pressures, though the substudy did not measure pressure directly. A separate study of incretin drugs in heart failure patients with implanted sensors found that pulmonary artery pressure fell on semaglutide and tirzepatide. Together the data point in the same direction, but neither was designed to separate the effect of weight loss from any direct action of the drug on the heart.
Limits and what comes next
The substudy has clear limits. Only 94 of 731 trial patients contributed, about one in eight, and those with usable scans may differ from the rest. The groups were unequal in size, with 42 on tirzepatide and 52 on placebo. The analysis was not the trial's primary aim, so it should be read as supporting evidence rather than proof. Changes of 5 to 10 millilitres are small against normal chamber sizes and below what doctors would usually notice on a single scan. Kramer reports disclosures with Bristol Myers Squibb and Cytokinetics, HCPLive noted, and the SUMMIT trial was sponsored by Eli Lilly, which markets tirzepatide.
Tirzepatide is approved for type 2 diabetes, weight management and obstructive sleep apnea in the United States, and heart failure use depends on each regulator's label. The next step is publication of the full tirzepatide HFpEF imaging paper with its methods and complete results. Longer trials will also matter, including cardiovascular outcome studies of other incretin drugs, such as the survodutide cardiovascular safety trial reported this week. Readers can follow new readouts and what they do and do not show on the trials desk. Patients should not start, stop or change any heart failure treatment on the strength of imaging results from a substudy.
Conclusion
The tirzepatide HFpEF imaging substudy suggests the drug reverses some enlargement of the atria and right ventricle in obesity related heart failure, with the atrial changes linked to better symptom scores. The effects are small and come from 94 patients in a secondary analysis that has not yet been published in full. They add a structural signal to SUMMIT's outcome results without settling how much is due to weight loss alone.