Tirzepatide and cardiovascular outcomes: what has been tested

TL;DR

• In a randomized trial of people with type 2 diabetes and established cardiovascular disease, tirzepatide was noninferior to dulaglutide for cardiovascular death, heart attack, or stroke.¹
• A post hoc analysis of the same trial found fewer broad cardiovascular and kidney events with tirzepatide, although the expanded endpoint makes this a supportive finding.²
• A meta-analysis of 22 randomized trials linked tirzepatide with lower odds of major cardiovascular events across adults with diabetes, overweight, or obesity.³
• A predicted-risk analysis modeled future cardiovascular risk, but it did not count heart attacks, strokes, or cardiovascular deaths over ten years.⁴

For Filipino readers, the key question is not whether a headline sounds positive.¹ It is what researchers actually measured, in whom, and against which comparator.¹ Tirzepatide has been tested through a major randomized cardiovascular trial, pooled trial analyses, health-record studies, and early mechanistic work.¹ Those designs answer different questions and carry different levels of certainty.¹

What counts as a cardiovascular outcome?

Researchers often combine cardiovascular death, heart attack, and stroke into one major-event endpoint.⁵ A composite can detect an overall difference even when individual events remain uncertain.⁵ In SURPASS-CVOT, participants had type 2 diabetes and established cardiovascular disease.¹ They were randomly assigned to tirzepatide or dulaglutide, an active GLP-1 comparator.¹ The main finding was noninferiority for the three-part composite.¹ In plain language, tirzepatide was not shown to be unacceptably worse than dulaglutide under the trial’s statistical boundary.¹ That is an important cardiovascular result, but it is not the same as proving superiority over every alternative or in every population.¹

This endpoint gives one overall answer. It does not guarantee that each part changed by the same amount.⁵

The tirzepatide entry below is included as a neutral catalog reference.

The clearest randomized test

SURPASS-CVOT enrolled 13,165 people with type 2 diabetes and pre-existing cardiovascular disease across 640 centers.² Random assignment helps balance measured and unmeasured differences between groups, giving this design stronger causal footing than a health-record comparison.¹

The reported main cardiovascular result established noninferiority for cardiovascular death, heart attack, or stroke.¹ That wording matters.¹ Noninferiority asks whether a treatment avoids being meaningfully worse than its comparator within a prespecified margin.¹ It does not automatically show that the treatment is better.¹ The trial population also matters: these were people with both diabetes and established cardiovascular disease.¹ The finding cannot simply be transferred to people without diabetes or without prior cardiovascular disease.¹

The active comparison also shapes the answer. This was not a test of tirzepatide against no treatment.¹ It asked how tirzepatide compared with dulaglutide in a group already at high cardiovascular risk.¹ That makes the result useful for a focused clinical research question. It also limits broad claims based on the trial. Readers should keep the comparator and enrolled population attached to every summary of the finding.¹

What the expanded analysis added

Researchers later performed a post hoc analysis of SURPASS-CVOT using a broader six-part cardiorenal endpoint.² It included all-cause death, heart attack, stroke, coronary revascularization, heart-failure hospitalization, and adverse kidney outcomes.² That endpoint occurred in 23.7% of tirzepatide participants and 27.4% of dulaglutide participants.² The reported hazard ratio was 0.84, with a 95% confidence interval from 0.79 to 0.90.²

Narrower five-part and four-part composites produced similar hazard ratios of 0.86.² These results support the broader pattern in this high-risk diabetes group.² Still, the analysis was post hoc.² Its expanded endpoint was not the trial’s original main cardiovascular question.² Gastrointestinal adverse events were also more common with tirzepatide than dulaglutide in this trial.²

The six-part measure blended heart, brain, kidney, and survival events.² That wider view can be useful, but it answers a different question from the original three-part endpoint. The event rates and hazard ratio should therefore be read with the full endpoint definition, not as a stand-alone claim that every included outcome improved.²

What pooled randomized trials found

A 2026 systematic review combined 22 randomized trials with 29,023 adults who had type 2 diabetes, overweight, or obesity.³ Each trial followed participants for at least 24 weeks.³ Across those trials, tirzepatide was linked with lower odds of major cardiovascular events.³ The pooled odds ratio was 0.87, with a 95% confidence interval from 0.79 to 0.94.³

The review also reported lower all-cause mortality, with an odds ratio of 0.84.³ Yet it found no link for individual major-event components or heart-failure hospitalization.⁵ That difference guards against a broad conclusion.⁵ A pooled composite can look favorable while separate outcomes remain too uncommon or unclear for firm answers.⁵ Combining trials can add statistical power. Even so, the result still depends on the populations, comparators, follow-up periods, and event definitions in those trials.³

Why comparison studies disagree

A separate meta-analysis compared tirzepatide with GLP-1 receptor agonists in adults with overweight or obesity.⁶ It included eight studies and 323,439 participants. Seven studies were observational, and only one was randomized.⁶ The pooled major-event hazard ratio was 0.85, with a confidence interval from 0.70 to 1.04.⁶ That result was not statistically significant. The heterogeneity statistic was high.⁶

One matched health-record study reached a more favorable result in 16,402 people with diabetes and atherosclerotic cardiovascular disease.⁷ Tirzepatide use was linked with lower one-year major-event risk than GLP-1 receptor agonist use, with a hazard ratio of 0.75.⁷ Matching can reduce measured differences. It cannot reproduce random assignment or remove every hidden difference between groups.⁷ Together, these studies suggest a signal, not a settled head-to-head ranking across the whole drug class.⁶

A useful check is simple: ask whether groups were randomized or reconstructed from past records, then compare baseline risk, outcome definitions, and follow-up. Even a large record set cannot become a randomized trial.⁶⁷

The semaglutide entry below is included as a neutral catalog reference.

What semaglutide comparisons can show

A target-trial emulation used health records from 217,920 adults starting semaglutide or tirzepatide.⁸ Researchers matched groups by diabetes status and followed outcomes for up to three years.⁸ Tirzepatide use was linked with lower risks across the heart and brain outcomes studied.⁸ The size of some links differed between participants with and without diabetes.⁸ The methods were designed to imitate a trial, but the evidence remains observational.⁸

Another record-based comparison was released as a preprint. It included two matched cohorts of 195,837 people with type 2 diabetes.⁹ It reported lower composite cardiovascular events and all-cause mortality with tirzepatide than semaglutide.⁹ The study was not randomized, and the report was a preprint.⁹ Its estimates therefore need caution and independent review.⁹ These comparisons can raise useful questions, but they do not replace direct randomized cardiovascular testing.⁸

What indirect studies can and cannot tell us

A Japanese post hoc analysis used trial measurements to calculate predicted 10-year cardiovascular risk.⁴ The participants had no diabetes or prior cardiovascular disease.⁴ The model estimated lower future risk with tirzepatide than placebo.⁴ This was a prediction based on a risk equation. Researchers did not watch the group for ten years and count heart attacks, strokes, or cardiovascular deaths.⁴

That gap is easy to miss. A modeled score can show how measured risk factors changed within a study. It cannot show whether the predicted events later happened. The analysis was also post hoc and included 186 trial completers, which narrows how far the estimate can be applied.⁴

Mechanistic research sits even earlier on the evidence path. One preprint studied injured arteries in mice and nitric-oxide responses in cultured human endothelial cells.¹⁰ These tests may help researchers form ideas about how a vascular effect could occur.¹⁰ They do not measure cardiovascular outcomes in people, and a preprint has a more provisional status.¹⁰ Such work can guide later research. It should not be presented as proof of fewer human events.

Bottom line

Tirzepatide has undergone a serious cardiovascular test, not merely studies of body weight or blood markers.¹ In people with type 2 diabetes and established cardiovascular disease, the randomized SURPASS-CVOT comparison showed noninferiority to dulaglutide for cardiovascular death, heart attack, or stroke.¹ Its post hoc expansion and a large randomized-trial meta-analysis add supportive signals.²³ However, head-to-head observational comparisons can retain hidden differences, and pooled results do not agree uniformly.⁶⁷ Predicted-risk and laboratory studies answer narrower questions than trials that count events in people.⁴¹⁰ The most accurate reading is encouraging but bounded by population, comparator, endpoint, and study design.¹

Read the main trial first, then use pooled and observational work as context.¹²³⁶ Risk scores, mouse work, and cell tests remain earlier steps in the research chain.⁴¹⁰

Frequently asked questions

Did tirzepatide prevent heart attacks and strokes in a randomized trial?

The main randomized cardiovascular trial tested a combined endpoint of cardiovascular death, heart attack, or stroke.¹ Tirzepatide met the criterion for noninferiority versus dulaglutide in people with type 2 diabetes and established cardiovascular disease.¹ That result does not mean every individual outcome was separately reduced.¹

Was tirzepatide better than dulaglutide?

The original main result established noninferiority, not a universal claim of superiority.¹ A later post hoc analysis found fewer events on a broader cardiovascular and kidney composite.² Because that analysis expanded the endpoint after the main trial analysis, it should be read as supportive evidence.²

Do meta-analyses settle the question?

Not completely.³ One meta-analysis of 22 randomized trials found lower odds of major cardiovascular events with tirzepatide.³ Another comparison, mostly built from observational studies, found no statistically significant major-event advantage over GLP-1 receptor agonists and reported substantial heterogeneity.⁶

What is known for people without diabetes?

A Japanese post hoc analysis included trial completers without diabetes or prior cardiovascular disease.⁴ It estimated lower 10-year cardiovascular risk from a prediction model.⁴ Because researchers modeled future risk instead of observing events for ten years, that study cannot establish that fewer heart attacks, strokes, or cardiovascular deaths occurred.⁴

Is tirzepatide proven better than semaglutide for cardiovascular outcomes?

Large health-record studies report favorable associations for tirzepatide over semaglutide.⁸ These studies used observational matching, which cannot remove every source of confounding.⁸ Another favorable comparison was reported as a preprint, giving that result a more provisional status.⁹

Do predicted-risk and laboratory studies count as outcome evidence?

A Japanese post hoc analysis estimated future risk instead of observing events.⁴ A preprint studied injured mouse arteries and human cells in a laboratory.¹⁰ Neither design directly proves fewer cardiovascular events in people.

This article is educational research context, not medical advice. Catalog links are neutral references and do not imply a treatment or human-use recommendation.

Sources

Footnotes

1. Cardiorenal Outcomes With Tirzepatide Compared With Dulaglutide in Patients With Diabetes and Cardiovascular Disease: A Post Hoc Analysis of the SURPASS-CVOT Randomized Clinical Trial.. Completed randomized, double-blind cardiovascular outcomes trial in people with type 2 diabetes and established cardiovascular disease. Noninferiority does not by itself establish superiority, and the result applies to the enrolled high-risk diabetes population. ↩ ↩² ↩³ ↩⁴ ↩⁵ ↩⁶ ↩⁷ ↩⁸ ↩⁹ ↩¹⁰ ↩¹¹ ↩¹² ↩¹³ ↩¹⁴ ↩¹⁵ ↩¹⁶ ↩¹⁷ ↩¹⁸ ↩¹⁹ ↩²⁰ ↩²¹ ↩²² ↩²³ ↩²⁴ ↩²⁵ ↩²⁶ ↩²⁷ ↩²⁸

2. Cardiorenal Outcomes With Tirzepatide Compared With Dulaglutide in Patients With Diabetes and Cardiovascular Disease: A Post Hoc Analysis of the SURPASS-CVOT Randomized Clinical Trial.. Post hoc analysis of 13,165 randomized participants with a median treatment duration of 46.9 months. The expanded endpoint was analyzed after the trial's main cardiovascular analysis and included kidney outcomes. ↩ ↩² ↩³ ↩⁴ ↩⁵ ↩⁶ ↩⁷ ↩⁸ ↩⁹ ↩¹⁰ ↩¹¹ ↩¹² ↩¹³ ↩¹⁴ ↩¹⁵ ↩¹⁶ ↩¹⁷

3. Tirzepatide, cardiovascular outcomes and mortality in obesity and diabetes: a systematic review and meta-analysis.. Systematic review of 29,023 adults with type 2 diabetes, overweight, or obesity and at least 24 weeks of follow-up. The pooled result spans different trial populations, comparators, and cardiovascular ascertainment. ↩ ↩² ↩³ ↩⁴ ↩⁵ ↩⁶ ↩⁷ ↩⁸ ↩⁹ ↩¹⁰ ↩¹¹

4. Post hoc analysis of SURMOUNT-J: tirzepatide versus placebo and predicted 10-year cardiovascular disease risk in Japanese adults with obesity disease.. Post hoc analysis of 186 trial completers without diabetes or prior cardiovascular disease using Framingham risk equations. The analysis predicted risk from a model and did not observe cardiovascular events over ten years. ↩ ↩² ↩³ ↩⁴ ↩⁵ ↩⁶ ↩⁷ ↩⁸ ↩⁹ ↩¹⁰ ↩¹¹ ↩¹²

5. Tirzepatide, cardiovascular outcomes and mortality in obesity and diabetes: a systematic review and meta-analysis.. Secondary outcome analysis within a 22-trial systematic review. A null association may reflect limited event counts or precision and does not prove no effect. ↩ ↩² ↩³ ↩⁴ ↩⁵ ↩⁶

6. Cardiovascular Outcomes With Tirzepatide Versus GLP-1 Receptor Agonists in Overweight or Obesity: A Systematic Review and Meta-Analysis.. Eight-study meta-analysis in adults with overweight or obesity; seven studies were observational and one was randomized. Substantial heterogeneity and the predominance of observational evidence limit causal interpretation. ↩ ↩² ↩³ ↩⁴ ↩⁵ ↩⁶ ↩⁷ ↩⁸ ↩⁹

7. Comparative Cardiovascular Outcomes of Tirzepatide and Glucagon-Like Peptide-1 Receptor Agonists in Patients With Type 2 Diabetes and Atherosclerotic Cardiovascular Disease.. TriNetX study of 16,402 matched people with type 2 diabetes and atherosclerotic cardiovascular disease. Retrospective matching cannot eliminate residual confounding or establish causality. ↩ ↩² ↩³ ↩⁴ ↩⁵

8. Cardiovascular and Cerebrovascular Outcomes Risk Reduction Associated With Semaglutide vs Tirzepatide: A Target Trial Emulation.. Observational TriNetX analysis of 217,920 adults, stratified and matched by diabetes status, with follow-up to three years. The design emulated a trial but remained observational and may retain unmeasured confounding. ↩ ↩² ↩³ ↩⁴ ↩⁵ ↩⁶ ↩⁷ ↩⁸

9. Cardiovascular Outcomes with Tirzepatide versus Semaglutide in Type 2 Diabetes. Preprint using two propensity-matched cohorts of 195,837 people with type 2 diabetes. The study was observational and reported as a preprint, so it was not equivalent to a peer-reviewed randomized comparison. ↩ ↩² ↩³ ↩⁴ ↩⁵

10. Tirzepatide Attenuates Wire Injury-Induced Arterial Remodeling in Non-Diabetic and Diabetic Mice: Comparison with Semaglutide. Preprint animal and laboratory study comparing tirzepatide with GLP-1 receptor agonists. Animal and cell findings cannot establish cardiovascular outcomes in humans. ↩ ↩² ↩³ ↩⁴ ↩⁵ ↩⁶