The SURPASS-CVOT results showed that tirzepatide was not meaningfully worse than dulaglutide for major heart and stroke events. It did not establish that tirzepatide was better. Over about four years, 12.2% of people taking tirzepatide had a heart attack, stroke or heart-related death, compared with 13.1% taking dulaglutide. Everyone had type 2 diabetes and known artery disease.
What question did SURPASS-CVOT ask?
Could tirzepatide hold its own against a medicine with heart benefits already demonstrated in a trial?
Dulaglutide, sold as Trulicity, was an active comparator: a working medicine used for comparison. In REWIND, published in The Lancet in 2019, major heart or stroke events occurred in 12.0% on dulaglutide versus 13.4% on placebo. Placebo was a dummy injection, and follow-up lasted a median 5.4 years (Gerstein and colleagues).
So SURPASS-CVOT wasn't comparing a medicine with an empty shot. It tested whether tirzepatide preserved the benefit of an established treatment, and whether it could go further.
Who took part, and for how long?
Researchers assigned 13,299 adults by chance to weekly tirzepatide injections, up to 15 mg, or dulaglutide 1.5 mg. The study was double-blind: participants and study teams didn't know which medicine each person received.
Everyone was at least 40, had type 2 diabetes and had established artery disease in the heart, brain or legs. Their body mass index, a weight-to-height measure, was at least 25. HbA1c, a blood test reflecting average sugar over roughly three months, ranged from 7.0% to 10.5%.
The average participant was 64 and had lived with diabetes for about 15 years. Follow-up lasted a median four years: half were followed longer, half less.
People with a recent heart or stroke event, previous pancreatitis, meaning pancreas inflammation, or the most advanced heart failure were excluded. The trial registry, NCT04255433, gives the full eligibility rules.
The main analysis excluded 134 people enrolled in error who didn't meet the rules. That left 6,586 taking tirzepatide and 6,579 taking dulaglutide. These are the groups in the Nicholls and colleagues paper, New England Journal of Medicine, December 2025.
What did the trial measure?
The primary endpoint, the main result chosen before the study began, was the first heart-related death, heart attack or stroke. Together these are called MACE-3, a three-part count of major cardiovascular events. Each person counted once, at their first event.
An independent committee checked the events without knowing treatment assignments. Participants also continued usual care, and clinicians could adjust their other medicines.
SURPASS-CVOT results: the heart numbers
About 12 in 100 people on tirzepatide had a first major event, compared with 13 in 100 on dulaglutide.
| Main outcome over a median four years | Tirzepatide | Dulaglutide |
|---|---|---|
| People with a first heart-related death, heart attack or stroke | 801 of 6,586 | 862 of 6,579 |
| Percentage with an event | 12.2% | 13.1% |
The hazard ratio, which compares how quickly events accumulated, was 0.92. Its 95.3% confidence interval, the statistical range around that estimate, was 0.83 to 1.01. That spans 17% lower risk to 1% higher risk, including no difference.
Noninferiority means not meaningfully worse by a limit chosen beforehand. Here, the range's upper end had to fall below 1.05, ruling out a relative increase of 5% or more. It did.
Superiority means evidence that one treatment is better. That required the upper end to fall below 1.00. It didn't. The trial met its noninferiority test but missed its superiority test.
At 36 months, average weight fell 11.6% with tirzepatide versus 4.8% with dulaglutide. Better weight results did not establish better results for the main heart outcome.
What side effects showed up?
Stomach problems were the clearest tradeoff. About one in four people taking tirzepatide reported nausea. About as many reported diarrhea.
Selected rows from the paper's side-effect table follow. They count all treated participants, 6,647 in each group. Reported health problems aren't automatically caused by the medicine.
| Reported problem during treatment | Tirzepatide | Dulaglutide |
|---|---|---|
| Stomach or bowel problems | 42.5% | 35.9% |
| Nausea | 25.1% | 22.4% |
| Diarrhea | 24.8% | 19.1% |
| Vomiting | 11.6% | 9.7% |
| Stopped the study medicine because of a health problem | 13.2% | 10.1% |
| Serious health event | 31.8% | 31.9% |
| Pancreatitis | 0.6% | 0.6% |
Serious events include problems requiring hospital care. That row excludes the heart outcomes reviewed separately by the trial's committee. Similar serious-event percentages don't mean everyone tolerated treatment equally: more people stopped tirzepatide.
What did critics and later studies add?
A 2025 commentary by Gian Paolo Fadini in Diabetes Therapy cautioned against treating a reconstructed placebo comparison as a direct one. Such estimates borrow results from older trials and depend on comparable patients and background care.
The main trial's authors also noted limited representation: only 29% of participants were women, and more than 80% were White. Lilly funded the study. Its findings cannot describe every person who might receive tirzepatide.
As of September 30, 2026, the day this article was written, newer evidence includes an August 5, 2026 BMJ study. It compared 52,971 people taking tirzepatide or sitagliptin, another diabetes medicine, using insurance records. Estimated one-year risks were 2.9% versus 4.4% for its combined outcome of heart attack, stroke or death from any cause.
Treatment wasn't assigned by chance, and that outcome counted deaths beyond heart causes. It adds evidence from everyday care, but cannot replace the trial's direct comparison with dulaglutide.
What changed on the Mounjaro label?
As of September 30, 2026, the day this article was written, Mounjaro carries a heart-risk indication, a use written into its label. The FDA letter signed August 27, 2026 added reduction of heart-related death, nonfatal heart attack or nonfatal stroke in high-risk adults with type 2 diabetes.
The current DailyMed label reads Revised: 8/2026. Its table includes all randomized participants who received treatment, yielding slightly different percentages: 12.1% versus 13.0%. It also states that superiority was not established.
Tirzepatide heart benefits explains who that wording covers. For semaglutide's separate heart trials, SUSTAIN-6 results covers weekly injections and SOUL trial results covers a daily tablet.
What this means for a compounded prescription
SURPASS-CVOT tested Lilly's injections. It did not test Promise's formulation or establish identical outcomes for every pharmacy preparation.
Tirzepatide prescribed through Promise is dispensed by a licensed U.S. compounding pharmacy as a compounded medication, which is different from an FDA-approved product: the formulation offered here is not FDA-approved. The FDA explains that it does not verify compounded drugs' safety, effectiveness or quality before marketing.
A licensed provider may still prescribe a compounded formulation; that decision is between you and your doctor.
Its place in a visit
The useful questions are whether someone's health history resembles the trial group and which product is being considered. Existing diabetes and heart medicines, past illnesses and tolerability all shape that discussion. The prescriber sets the actual dose.
At Promise, a licensed provider reviews every request, and not everyone qualifies. The trial informs that judgment; it doesn't make it for one person.
