Wegovy and Heart Attacks: The SELECT Trial, No Diabetes

You had a heart attack a few years back. Maybe a stent, maybe two. You carry more weight than your cardiologist would like, but your blood sugar is fine — no diabetes, never has been. You take a statin, an aspirin, maybe a blood-pressure pill. And somewhere along the way you heard that the weight-loss shot everyone's talking about might also protect your heart.

That question has a trial attached to it now. It's called SELECT, it ran for years, and it's the reason Wegovy can talk about heart attacks and strokes at all. Did it help? Yes. How much it helped, and for whom, is where the headlines and the data quietly part ways.

What follows is the real number, the real caveats, and what someone in exactly your situation should take from it.

The gap SELECT was built to fill

For most of the GLP-1 story, cardiovascular protection came bundled with diabetes. SUSTAIN-6, LEADER, REWIND — every major heart-outcomes trial in this drug class enrolled people with type 2 diabetes. The benefit was real, but it always carried an asterisk: did the heart protection come from the drug, or from better blood sugar?

That left a big group unanswered. Plenty of people carry extra weight and have heart disease without ever crossing into diabetes. Their cardiologists had statins and antiplatelets and blood-pressure meds — the standard toolkit — but no GLP-1 trial that spoke directly to them.

SELECT was the first large trial to put a GLP-1 in front of that exact group: obesity or overweight, established heart disease, and no diabetes. Take diabetes out of the equation entirely, and ask the clean question. If a drug developed for weight still moves hard cardiac outcomes in people whose sugar was never the problem, that tells you something the diabetes trials couldn't.

What SELECT actually tested

SELECT enrolled 17,604 patients. To get in, you had to be 45 or older, have preexisting cardiovascular disease, carry a body-mass index of 27 or higher, and have no history of diabetes. The researchers split them evenly — 8,803 to semaglutide, 8,801 to placebo — and ran the trial for a mean of 39.8 months, roughly 3.3 years.

The drug here matters. This was semaglutide at 2.4 mg once a week, given as a subcutaneous shot. That's the obesity dose, sold as Wegovy. Same molecule as Ozempic, the diabetes brand, but a higher dose studied in people who didn't have diabetes. Keep that distinction straight, because it's the whole point of the trial.

The primary endpoint was a composite that cardiologists know cold: death from cardiovascular causes, a non-fatal heart attack, or a non-fatal stroke. Three hard outcomes, bundled into one yardstick. In the trial literature it's called 3-point MACE, and it's the same measuring stick the diabetes CVOTs used.

One detail in that table is easy to skim past and shouldn't be: the comparator was placebo on top of standard heart care. Nobody traded their statin for a weekly shot. The semaglutide arm kept taking everything their cardiologist already had them on, and the shot got added to the pile. We'll come back to why that framing changes how you read the result.

The headline: a 20% lower risk

Here's the number that made the news. A primary endpoint event — a cardiovascular death, a non-fatal heart attack, or a non-fatal stroke — happened in 569 of the 8,803 people on semaglutide, which is 6.5%. In the placebo group it happened in 701 of 8,801, or 8.0%. The hazard ratio came out to 0.80, with a 95% confidence interval of 0.72 to 0.90, and the p-value landed below 0.001.

A hazard ratio of 0.80 is what "20% lower risk" means. The confidence interval doesn't cross 1.0, which is why the result is statistically solid rather than a fluke. The drug beat placebo on a hard, pre-specified cardiovascular endpoint, in people without diabetes, over three-plus years. That had never been shown before in this population.

SELECT's own conclusion, stated plainly: in patients with preexisting cardiovascular disease and overweight or obesity but without diabetes, weekly semaglutide 2.4 mg was superior to placebo in reducing cardiovascular death, non-fatal heart attack, or non-fatal stroke.

That sentence is the foundation Wegovy's cardiovascular story is built on. It's real. It's also where most summaries stop — and stopping there is how a genuinely modest benefit gets read as something much bigger than it is.

"20% lower" is not "20 fewer out of 100"

Slow down here. This is exactly where almost everyone's intuition goes sideways.

The 20% is a relative number. It describes how much the risk shrank compared to where it started. It is not a 20-percentage-point drop, and it does not mean one in five people avoided a heart attack. Look at the absolute figures again: the event rate went from 8.0% on placebo to 6.5% on semaglutide. That's a gap of about 1.5 percentage points, accumulated over roughly 3.3 years.

Both readings come from the same trial. The 20% and the 1.5 points are not in conflict — they're two honest ways of describing one result. The relative number tells you the drug works in a direction worth caring about. The absolute number tells you the size of the effect for any single person over that window.

Why does the framing matter so much? Because a 20% relative reduction sounds like the drug cuts your personal odds by a fifth no matter where you start — and it doesn't work like that. If your baseline risk is high, that same 20% buys you a bigger absolute drop. If your baseline risk is low, it buys you less. The relative number travels; the absolute benefit depends entirely on how much risk you were carrying to begin with. That's a conversation for you and your cardiologist, with your actual numbers in front of you — not a number you can read off a press release.

The cost side: who stopped, and why

A benefit only counts if people can stay on the drug, so the discontinuation data deserves equal billing with the headline.

In SELECT, adverse events that led people to permanently stop the trial medication happened in 1,461 patients on semaglutide — 16.6% — versus 718 on placebo, or 8.2%. The gap was statistically clear, and the reasons were mostly gastrointestinal: nausea, vomiting, diarrhea, the now-familiar GLP-1 starter pack.

That number cuts in two directions. Roughly twice as many people quit semaglutide for side effects as quit placebo. It's a genuine tolerability tradeoff, not a footnote — over three years, one in six people on the drug stopped because of how it made them feel.

None of that makes the drug dangerous, though. "Stopped because of nausea" and "was harmed" are different sentences. GI side effects in this class cluster early, usually during the dose-escalation weeks, and for many people they settle. The honest summary: the benefit is modest, the tolerability cost is non-trivial, and both of those are true at once — neither one cancels the other out.

What SELECT is not

This trial gets stretched to cover things it never studied. Four limits are worth nailing down.

It's secondary prevention, not primary. Every single person in SELECT already had cardiovascular disease — a prior heart attack, a prior stroke, or peripheral artery disease. The trial says nothing about whether semaglutide protects the heart of someone with obesity but no established heart disease. If you've never had a cardiac event, SELECT is not your trial. The benefit was measured in people already on the wrong side of that line.

It's an add-on, not a swap. The semaglutide group stayed on standard cardiovascular care — statins, antiplatelets, blood-pressure medication. The shot was layered on top of that care, and the 20% landed on top of it too. Nothing in SELECT suggests semaglutide replaces a statin or lets anyone stop their existing heart meds. That's not what the trial tested, and reading it that way is how people get hurt.

The endpoint was the three-part composite, and only that. SELECT measured cardiovascular death, non-fatal heart attack, and non-fatal stroke combined. It does not prove the drug "prevents all deaths" or that people on it simply "live longer" in some general sense. The composite moved. Claims beyond the composite aren't claims SELECT can back.

It studied the 2.4 mg obesity dose, in people without diabetes. This is Wegovy territory — the higher weight-management dose — not the diabetes story. The diabetes CVOTs are separate trials in separate populations. Don't blend them.

The safety basics worth knowing

None of SELECT's safety picture was a surprise to anyone who's followed this drug class, but it's worth having the facts in one place.

The most common adverse reactions with semaglutide are gastrointestinal — nausea, vomiting, diarrhea, abdominal pain, and constipation. They tend to show up early and ease over time for most people, which is exactly the pattern SELECT's discontinuation data reflected.

There's a boxed warning to know about, too. Semaglutide is contraindicated in anyone with a personal or family history of medullary thyroid carcinoma (MTC) or Multiple Endocrine Neoplasia syndrome type 2 (MEN 2), based on thyroid C-cell tumor signals seen in rodents. And acute pancreatitis has been reported with this class; if it's suspected, the drug should be stopped promptly. These aren't reasons to panic — they're reasons the decision belongs in a clinic, with someone who knows your history.

So what does this mean for you?

If you're the reader I described at the top — obesity or overweight, a heart event already behind you, no diabetes — SELECT is genuinely relevant to you in a way it wouldn't be for someone without established disease. You're the population it studied.

Here's the grounded takeaway:

Over about 3.3 years, adding semaglutide 2.4 mg to standard heart care lowered the combined risk of cardiovascular death, heart attack, and stroke by 20% in relative terms — roughly 1.5 fewer events per 100 people in absolute terms. It did that on top of statins and antiplatelets, not instead of them. And it came with a real tolerability cost: about one in six people stopped, mostly for GI reasons.

Whether that math favors you depends on numbers SELECT can't see: your baseline risk, how well your current regimen is controlling it, how your body tolerates GLP-1 side effects, and what the cost of the drug looks like through your coverage. That's not a calculation a blog can finish for you. It's the agenda for a visit. Talk to your cardiologist about whether the SELECT data changes anything in your particular case — that's the move here, not a prescription you can self-write from an article.

Questions worth bringing to your cardiologist

If this trial is on your mind, walking into the appointment with specifics gets you a better conversation than "should I be on the weight-loss shot?"

• I have established heart disease and obesity but no diabetes. Does the SELECT data change what you'd consider for me?
• I'm already on a statin, an antiplatelet, and blood-pressure medication. Is adding semaglutide additive for my cardiovascular risk, or redundant with what's already working?
• What's my actual baseline risk, and roughly how much absolute benefit would a 20% relative reduction translate to for someone with my numbers?
• I've heard GI side effects are common. If I tolerate it poorly during the dose-escalation weeks, what's the plan — slow the titration, or step away from the drug?
• Given my thyroid and pancreatitis history, is there any reason this class isn't a fit for me?
• If we do start it, how will we tell over the next year whether it's earning its place in my regimen?

None of those are gotcha questions. Any cardiologist will be glad you read the trial instead of the headline — it means the decision gets made with you, not just handed to you.

The bottom-line read

SELECT did something no trial had done before: it showed that a weight-management dose of semaglutide reduces hard cardiovascular events in people who have heart disease and obesity but not diabetes. That result is real, statistically clean, and the reason the US FDA recognized a cardiovascular risk-reduction use for the obesity-dose semaglutide in adults with established cardiovascular disease and obesity or overweight.

It also didn't do several things people assume it did. It didn't turn a 20% relative reduction into a 20-point drop in anyone's personal odds. It didn't show benefit for people without established heart disease. It didn't license anyone to drop their statin. And it didn't come for free — the side-effect discontinuation rate was about double placebo's.

Hold both halves at once and you've got the trial as it actually is: a meaningful, modest addition to secondary cardiovascular prevention, in a specific population, with a real tradeoff. If that population is you, the next step isn't a pharmacy — it's the conversation where your own numbers finally enter the picture.

This article draws on published clinical-trial and peer-reviewed research, including the SELECT trial in the New England Journal of Medicine, and is meant for information rather than medical advice. Any decision to start, stop, or change a medication belongs with your doctor, who can weigh it against your personal history.

Start managing your GLP-1 with Blueshot

AI coaching, injection scheduling, and weight tracking in one app