FDA Expands Mounjaro to Reduce Cardiovascular Risk: What SURPASS-CVOT Actually Showed

What changed on August 28, 2026

The FDA-approved Mounjaro prescribing information now states that tirzepatide is indicated to reduce the risk of major adverse cardiovascular events—cardiovascular death, nonfatal myocardial infarction, or nonfatal stroke—in adults with type 2 diabetes mellitus who are at high risk for these events. That is an expansion of an approved drug's uses, not the first approval of tirzepatide. Mounjaro's earlier indication was as an adjunct to diet and exercise to improve glycemic control in adults and pediatric patients age 10 years and older with type 2 diabetes.

The wording matters. The cardiovascular indication is for adults, even though the glycemic-control indication reaches patients age 10 and older. It is specific to type 2 diabetes and a defined MACE-3 outcome. It does not say that Mounjaro treats an acute heart attack, reverses atherosclerosis, prevents heart failure, or reduces cardiovascular risk in every person without diabetes.

Mounjaro and Zepbound contain the same active ingredient, tirzepatide, but they are separately labeled products with different approved indications. This action expanded Mounjaro's type 2 diabetes label; it should not be silently transferred to a different product or population. For background on the molecule, see our tirzepatide overview and guides to GLP-1 biology and GIP biology.

How SURPASS-CVOT was designed

SURPASS-CVOT (NCT04255433) was a Phase 3, event-driven, multicenter, international, randomized, double-blind, active-comparator trial. It randomized 13,299 people at 640 sites in 30 countries, 1:1, to subcutaneous tirzepatide once weekly or dulaglutide once weekly. Tirzepatide started at 2.5 mg and was escalated by 2.5 mg every four weeks toward 15 mg or the maximum tolerated dose; dulaglutide was given at 1.5 mg. Median follow-up was 210.1 weeks, approximately four years.

Participants had longstanding, high-risk disease: mean age was about 64 years, mean diabetes duration about 15 years, mean HbA1c 8.4%, and mean BMI about 33 kg/m². At baseline, 65% had coronary artery disease, 47% had a prior myocardial infarction, 19% a prior stroke, 25% peripheral arterial disease, and 20% a history of heart failure. Most were also receiving established preventive therapies: 86% statins and 83% antiplatelet therapy; 31% used an SGLT2 inhibitor.

This background care is a strength because the trial asked whether tirzepatide could preserve cardiovascular protection in contemporary practice. It also means the results describe adding one of two incretin therapies to broader care, not using either injection as a substitute for that care. Investigators could adjust glucose, lipid, blood-pressure, and cardiovascular medicines to local treatment targets during follow-up.

The main result—and the caveat behind “8% lower”

In the peer-reviewed modified intention-to-treat population, a first MACE-3 event occurred in 801 of 6,586 participants (12.2%) assigned tirzepatide and 862 of 6,579 (13.1%) assigned dulaglutide. The hazard ratio was 0.92, with a multiplicity-adjusted 95.3% confidence interval of 0.83 to 1.01. The noninferiority test was positive (P=0.003), while the superiority test was not (P=0.09).

Noninferiority trials set a margin defining how much worse an experimental treatment could be before it would no longer preserve an acceptable amount of the comparator's effect. SURPASS-CVOT used an upper confidence-bound margin of 1.05. The observed upper bound was 1.01, below 1.05, so the trial met that objective. Superiority required the upper confidence bound to fall below 1.00; it did not.

The FDA label's table uses all randomized and treated participants, whereas the journal's primary analysis excluded 134 randomized people later found not to meet inclusion criteria. That is why its MACE counts and denominators differ slightly from the publication. The label also warns that component outcomes and all-cause death were not controlled for family-wise type I error. They are clinically interesting, especially the all-cause mortality estimate, but they should not be promoted as independently confirmed superiority claims.

Why comparing with dulaglutide changes the interpretation

Dulaglutide (Trulicity) was not a neutral placebo. It is a once-weekly GLP-1 receptor agonist already indicated to reduce major cardiovascular events in adults with type 2 diabetes who have established cardiovascular disease or multiple cardiovascular risk factors. In the earlier placebo-controlled REWIND trial, dulaglutide reduced the primary MACE outcome, providing the evidence base that made it a meaningful active comparator.

That choice created a demanding clinical question: could tirzepatide deliver cardiovascular outcomes no worse than a medicine already known to help, while also treating hyperglycemia? SURPASS-CVOT answered yes within its prespecified margin. An active comparator can be more useful than placebo for choosing between therapies, and it avoids withholding an evidence-based incretin option for years.

But the design has limits. SURPASS-CVOT did not directly estimate tirzepatide versus placebo, and the effect of dulaglutide in REWIND cannot simply be mathematically transferred to a different trial population. It also tested dulaglutide 1.5 mg, the cardiovascular-outcomes dose used in REWIND, rather than every available GLP-1 drug or dose. The findings therefore do not establish that tirzepatide is superior to the GLP-1 class, semaglutide, SGLT2 inhibitors, or other cardiovascular therapies.

Who the new indication covers—and who was studied

The label uses the phrase “high risk for these events.” The pivotal efficacy evidence, however, came from secondary-prevention patients with established disease—such as coronary, cerebrovascular, or peripheral arterial disease—not a general low-risk diabetes population. Clinicians must use the current full prescribing information and an individual's documented history and risk profile rather than assuming that any single risk factor automatically reproduces the trial population.

The approval also does not mean everyone within the indication should start or switch therapy. Treatment selection can depend on prior cardiovascular events, glucose control, kidney function, concurrent medicines, tolerability, contraindications, cost and coverage, patient preferences, and competing therapeutic priorities. A cardiovascular label supports a claim at the population level; it does not predict an individual's outcome.

Safety and contraindications still matter

SURPASS-CVOT did not reveal a fundamentally new tolerability pattern. Overall adverse-event incidence appeared similar between groups in the publication, but gastrointestinal adverse events were more common with tirzepatide. The familiar common adverse reactions listed for Mounjaro are nausea, diarrhea, decreased appetite, vomiting, constipation, dyspepsia, and abdominal pain. These effects commonly emerge during dose escalation, but “common” should not be confused with harmless for every patient.

• Boxed warning: Tirzepatide causes thyroid C-cell tumors in rats; whether it causes medullary thyroid carcinoma in humans is unknown.
• Contraindications: Personal or family history of medullary thyroid carcinoma, MEN 2, or a known serious hypersensitivity to tirzepatide or its excipients.
• Important warnings: Acute pancreatitis; hypoglycemia when combined with insulin or an insulin secretagogue; serious hypersensitivity; acute kidney injury from volume depletion; severe gastrointestinal reactions; diabetic-retinopathy complications; acute gallbladder disease; and pulmonary aspiration during anesthesia or deep sedation.
• Gastroparesis: The label does not recommend Mounjaro in patients with severe gastroparesis.
• Oral medicines and contraception: Delayed gastric emptying can affect absorption of oral medications. The label advises users of oral hormonal contraceptives to use a non-oral method or add a barrier method for four weeks after starting and for four weeks after each dose increase.
• Pregnancy: Based on animal data, tirzepatide may cause fetal harm; pregnancy and pregnancy plans require clinician review.

What the trial cannot tell us

• No demonstrated superiority for MACE-3: The estimate favored tirzepatide, but the prespecified superiority threshold was not met.
• Not a primary-prevention trial: All participants had established cardiovascular disease. Extrapolation to people who only have risk factors is less direct.
• Representation was uneven: Participants were 71% male, 82% White, 2% Black, and 9% Asian in the FDA label. Average results may not precisely estimate effects in every subgroup.
• One active comparator and dose: The trial compared tirzepatide with dulaglutide 1.5 mg, not with placebo, semaglutide, every GLP-1 receptor agonist, or an SGLT2 inhibitor.
• Secondary findings need restraint: Individual endpoint and mortality estimates were not multiplicity-controlled in the label, so they should not be treated as separate definitive superiority findings.
• A composite can hide differences: MACE-3 combines cardiovascular death, myocardial infarction, and stroke. The component hazard ratios were not identical, and none alone proves the overall effect.
• Sponsor involvement: Eli Lilly funded the trial. Randomization, blinding, adjudicated outcomes, registration, and peer review strengthen credibility, but sponsorship remains relevant when assessing the evidence.

These limitations do not negate the result. They define it. The defensible conclusion is that, among adults with type 2 diabetes and established cardiovascular disease receiving contemporary care, tirzepatide preserved MACE-3 protection relative to dulaglutide within a strict noninferiority margin over about four years.

What the approval means in practice

The expanded label allows cardiovascular risk reduction to be part of the benefit-risk conversation when clinicians select glucose-lowering therapy for high-risk adults with type 2 diabetes. It also resolves an important uncertainty: before SURPASS-CVOT, tirzepatide had strong evidence for glycemic and weight effects, but dedicated cardiovascular-outcomes evidence was incomplete. The trial now shows that its MACE-3 performance was not unacceptably worse than a proven GLP-1 comparator.

The result may simplify therapy for some patients when glucose, weight, and cardiovascular goals overlap. It does not make tirzepatide a stand-alone cardiovascular prevention program. Blood-pressure control, lipid lowering, smoking cessation, physical activity, nutrition, kidney-protective therapy, and antithrombotic treatment when indicated remain separate decisions with their own evidence.

Nor does FDA approval settle comparative effectiveness for every choice. Dulaglutide remains an FDA-approved option with placebo-controlled cardiovascular evidence and a broader trial history that included many participants without established cardiovascular disease. Tirzepatide now has a label backed by a direct head-to-head noninferiority trial in established disease. Choosing between them requires more than comparing one headline percentage.