Semaglutide and heart risk: a 2026 review

A 2026 systematic review and meta-analysis in Biomedical Reports pooled 11 randomised controlled trials, 12 comparisons in total, covering 25,067 participants, and found that semaglutide was associated with a 32% lower odds of major adverse cardiovascular events (MACE) than the comparator, with a pooled odds ratio of 0.68^[1]. The headline that matters for readers: the benefit looked similar in people who had diabetes and in people who did not.

MACE is the standard cardiovascular trial endpoint. It is a composite of cardiovascular death, non-fatal heart attack (myocardial infarction) and non-fatal stroke. A lower number means fewer of those events. Semaglutide is a once-weekly GLP-1 receptor agonist (a drug that mimics a gut hormone called glucagon-like peptide-1) sold as Ozempic and Rybelsus for type-2 diabetes and as Wegovy for weight management.

What the meta-analysis found

Across the 12 comparisons, the pooled odds ratio for MACE was 0.68 (95% confidence interval 0.52 to 0.91)^[1]. An odds ratio below 1 favours semaglutide, so 0.68 translates to roughly a third fewer events in the pooled data. The authors also reported a 95% prediction interval of 0.44 to 1.04, which is wider than the confidence interval and crosses 1. That prediction interval is the honest part of the picture: it says a future trial in a new setting might land anywhere in that range, including no benefit.

The result held up under sensitivity testing. When the authors removed the STEP obesity trials, the odds ratio was 0.70. When they removed the heart-failure trials, it was 0.66^[1]. Dropping any single trial moved the estimate only between 0.63 and 0.76. The signal does not depend on one outlier study.

The 11 trials were published between 2016 and 2024 and spanned the major semaglutide programmes: SUSTAIN-6 and PIONEER-6 in type-2 diabetes, several STEP weight-management trials, the SELECT obesity-outcomes trial, and heart-failure trials^[1]. Pooling trials that recruited different populations is a strength and a weakness at once. It widens the evidence base, but it also mixes people whose baseline risk and reasons for taking the drug were not the same.

Diabetes or no diabetes: the benefit looked similar

This is the question the review was built to answer. In the subgroup with diabetes, the odds ratio was 0.72 (95% confidence interval 0.54 to 0.96). In the subgroup without diabetes, it was 0.67 (0.46 to 0.98)^[1]. The test for a difference between the two subgroups was not significant (P=0.62), which means the data do not show the cardiovascular benefit being driven by diabetes status one way or the other.

That fits the two large outcome trials underneath the pooled estimate. SUSTAIN-6 (Marso et al., 2016) studied 3,297 people with type-2 diabetes at high cardiovascular risk and reported a hazard ratio of 0.74 for the same composite endpoint^[2]. SELECT (Lincoff et al., 2023) studied 17,604 adults with overweight or obesity and established cardiovascular disease but no diabetes, and reported a hazard ratio of 0.80 (95% confidence interval 0.72 to 0.90, P<0.001)^[3]. Two different populations, two trials, the same direction of effect.

What the review does not say

A meta-analysis pools odds, it does not prove that the drug will help any one person. Most participants in these trials already had diabetes, cardiovascular disease, or both, so the pooled figure speaks to higher-risk groups, not to healthy adults taking semaglutide for cosmetic weight loss. The prediction interval crossing 1 is the same caution stated in statistical form. The review also reported the costs. The authors put the US incremental cost-effectiveness ratio at $180,000 to $260,000 per quality-adjusted life-year, above the willingness-to-pay thresholds usually applied in that setting.

The safety side was consistent with the rest of the semaglutide literature. The pooled data showed a higher risk of gastrointestinal disorders (relative risk 1.47) and of gallbladder events (relative risk 2.37), and discontinuation because of gut intolerance was more than twice as common on semaglutide^[1]. The authors also flagged a representation gap: 7 of the 11 trials enrolled at least 75% White participants, and none came from low-income countries.

Where this lands

Semaglutide already carries cardiovascular-outcome data in its regulatory file. The EMA authorises Wegovy as a prescription-only medicine for weight management, with the cardiovascular trial results recorded in the product information^[4]. This meta-analysis does not change the label. It pools the existing trials and confirms that the cardiovascular signal does not disappear when you separate people by diabetes status.

If you are weighing semaglutide for heart reasons, the decision belongs with a clinician who knows your history and your baseline risk. For how semaglutide is classified where you live, see our semaglutide page and the regulation hub.