Are GLP-1s Good for Your Heart?

For one specific group of people, the answer is a well-documented yes. A trial called SELECT put 17,604 adults on either semaglutide 2.4 mg or a placebo, followed them a little over three years, and counted 20% fewer heart attacks, strokes, and cardiovascular deaths in the treated group.¹ The FDA looked at that and added a cardiovascular indication to the label, which no weight management medication had ever carried before.²

Every single person in that trial had already survived a heart attack, a stroke, or serious artery disease in their legs. That detail tends to fall out of the headline version, and it changes almost everything about who the finding applies to.

Who Was Actually in This Trial

To get into SELECT you had to be at least 45, have a BMI of 27 or higher, and have a documented history of a heart attack, a stroke, or peripheral artery disease bad enough to cause pain when you walked. Anyone with diagnosed diabetes was turned away, along with anyone whose HbA1c came back at 6.5% or above.¹

The average person who made it in was 62 years old with a BMI of 33.3, and 76% of them had already had a heart attack.¹ This is what researchers call a secondary prevention population, meaning the event has already happened once and the whole question is whether you can stop the next one. Effects show up more clearly in a group like this, simply because there are more events available to prevent.

Also worth knowing: 88% of participants were on a statin the entire time.¹ So the 20% reduction happened on top of solid cardiac care that was already running. Nobody in this trial was relying on semaglutide alone to protect their heart, and you shouldn't read the result that way either.

The Numbers, Including the One That Missed

The main endpoint bundled three things together: cardiovascular death, nonfatal heart attack, and nonfatal stroke. That bundle happened to 6.5% of the semaglutide group and 8.0% of the placebo group, for a hazard ratio of 0.80.¹

Here's how each piece did on its own.

All figures from the SELECT trial.¹

Look at that last row for a second. Cardiovascular death on its own didn't reach statistical significance, with a p-value of 0.07 and a confidence interval that crosses 1.0.¹ You'll see plenty of articles round that up to "reduced cardiovascular death." The composite was significant and death from any cause was significant, so it's not as though the trial came up empty, but the isolated cardiac death number didn't clear the bar and there's no reason to pretend otherwise.

A couple of other numbers help explain what was going on underneath. Systolic blood pressure came down 3.8 mm Hg on semaglutide versus 0.5 on placebo, and by week 104 the treated group was down 9.4% of their body weight against 0.9% for placebo.¹

Side effects were part of the story too. About 16.6% of people on semaglutide quit because of an adverse event, roughly double the 8.2% who quit on placebo, and most of that was gastrointestinal.¹ Serious adverse events overall actually came in lower on semaglutide, 33.4% versus 36.4%.¹ More people quit than on placebo, and fewer of them had a serious event.

Tirzepatide's Data Is Different

People assume tirzepatide has the same cardiovascular résumé, and it doesn't, though that's less damning than it sounds once you see how its trial was built.

SURPASS-CVOT didn't test tirzepatide against a placebo. It tested tirzepatide against dulaglutide, an older GLP-1 that already had cardiovascular evidence behind it. Beating a placebo is one thing. Beating a drug that already works is a much steeper hill.

Across 13,165 people with type 2 diabetes and atherosclerotic cardiovascular disease, followed a median of 46.9 months, the primary composite hit 12.2% on tirzepatide and 13.1% on dulaglutide, a hazard ratio of 0.92.³ That cleared noninferiority at p = 0.003 and missed superiority at p = 0.09.³ Death from any cause landed at 8.6% versus 10.2%, though the investigators flagged that this endpoint wasn't adjusted for multiple comparisons, so it's suggestive rather than settled.³

Read plainly: tirzepatide held its own against a GLP-1 with a real cardiovascular track record, in people who had both diabetes and existing heart disease. It hasn't been shown to beat that comparator, and nobody has run it through a SELECT-style trial in people without diabetes. So if you find yourself comparing SELECT's 20% to SURPASS-CVOT's 8% as if that's a head-to-head, it isn't one. The trials used different comparators and enrolled different people.

If You Have Heart Failure, There's a Separate Trial

Heart failure with preserved ejection fraction, shortened to HFpEF, is the version where your heart squeezes normally but has stiffened up enough that it doesn't fill well between beats. It's tightly linked to carrying excess weight and it has been notoriously hard to treat, which is exactly why the SUMMIT trial got attention.

SUMMIT enrolled 731 adults with HFpEF and a BMI of 30 or above, then followed them a median of two years. Cardiovascular death or a worsening heart failure event happened to 9.9% of the tirzepatide group and 15.3% of the placebo group, a 38% relative reduction.⁴ Worsening heart failure events by themselves fell from 14.2% to 8.0%.⁴ People also walked an average of 18.3 meters farther on a six-minute walk test and scored 6.9 points better on a standard heart failure quality-of-life questionnaire.⁴

Semaglutide pointed the same direction. When researchers went back and looked at the SELECT participants who happened to have heart failure at baseline, the benefit held up in both the reduced and preserved ejection fraction groups.⁵ That's a subgroup analysis, so treat it as supporting evidence rather than proof of anything.

Is It the Drug or Is It the Weight Loss?

Nobody's untangled that, and designing a trial that could would be genuinely hard.

What SELECT does show is that several things moved at the same time. Weight dropped 9.4%. Systolic blood pressure came down about 3.3 mm Hg more than placebo. And progression toward diabetes was dramatically different, with 3.5% of the semaglutide group hitting an HbA1c of 6.5% or higher compared to 12.0% on placebo.¹

Any one of those, sustained over four years, would plausibly cut your cardiac risk. All of them together almost certainly do. Whether there's some additional direct effect on blood vessels or heart tissue beyond those pathways is still an open question, and the honest answer right now is that nobody knows.

If you're wondering whether that uncertainty makes the finding less useful, it doesn't. It just means the accurate way to say it is that research suggests these medications may support cardiovascular risk reduction in people who already have heart disease, mostly through the metabolic changes they cause.

What This Doesn't Mean

If you're 38, healthy, and you want to lose 25 pounds, SELECT isn't about you. Everyone in it was 45 or older with documented cardiovascular disease. There's no trial showing a GLP-1 prevents heart attacks in people who've never had one, and you shouldn't let anyone imply there is.

None of this turns a GLP-1 into a treatment for heart disease, either. These medications are approved for chronic weight management and type 2 diabetes, with that one added cardiovascular risk-reduction indication for one branded product in one defined population. That cardiovascular indication belongs to the branded FDA-approved product, and it does not extend to compounded semaglutide. They don't replace a statin or a blood pressure medication, and they don't replace the cardiologist you're already seeing. Every trial here ran on top of standard cardiac care.

This piece is about cardiovascular outcomes, not the full safety picture. This class also carries an FDA boxed warning for a risk of thyroid C-cell tumors, and the side effects worth planning for are laid out in our guide to managing GLP-1 side effects.

What We Actually Do With This in Lisle

Mostly it's a reason to look at more than the scale.

Every Defiant weight loss protocol opens with a provider consult and lab work, and the panel we build is where cardiometabolic markers like lipids, HbA1c, and hsCRP belong. We write the orders, and the draw itself happens through your primary care physician or Rythm Health, since we don't do blood draws in-clinic.

If you've got a cardiac diagnosis already, or you're on cardiovascular medications, say so at intake and keep your cardiologist in the loop. Coordination goes better when the person managing your heart and the person managing your weight each know what the other is doing.

You'll also get a Styku 3D body composition scan at baseline and monthly after that, which splits fat mass from lean mass in a way the bathroom scale never will. Our medically supervised weight loss program starts at $295 per month and covers medication, provider oversight, bi-weekly titration check-ins, monthly scans, and nurse access. We're in Lisle and we see patients from Naperville, Downers Grove, Wheaton, Oak Brook, and around the western suburbs.

Frequently Asked Questions

Last updated September 7, 2026.