For decades, weight-loss drugs and cardiovascular safety had an uneasy, often outright damaging relationship. The regulatory and clinical memory of obesity pharmacotherapy was shaped by failure: fen-phen was withdrawn in 1997 after valvular heart disease; more recently and more relevantly, sibutramine — an appetite suppressant used for over a decade — was withdrawn from the market in 2010 after the SCOUT trial found it significantly increased cardiovascular events in patients with existing heart disease. That history left an entire generation of cardiologists and regulators primed to expect that weight-loss drugs, at best, did nothing for the heart, and at worst, actively harmed it. Every obesity drug since has had to clear a high bar of cardiovascular safety, typically framed as "non-inferiority" — proving a drug doesn't cause harm, not that it provides benefit.
The SELECT trial broke that pattern entirely. It was designed not merely to rule out harm, but to test whether semaglutide — already established for type 2 diabetes and, since 2021, for chronic weight management as Wegovy — could actively reduce cardiovascular events in people who did not have diabetes. That distinction matters enormously: prior GLP-1 cardiovascular outcome trials (LEADER, SUSTAIN-6) had already shown cardiovascular benefit, but only in diabetic populations, where a plausible dominant mechanism (glycemic control) was available to explain part of the effect. SELECT deliberately excluded diabetics to isolate the cardiovascular effect of semaglutide and weight loss in a non-diabetic population with obesity and existing heart disease — a population previously with no drug proven to reduce their cardiovascular risk through weight-targeted therapy.
SELECT randomized 17,604 participants 1:1 to once-weekly subcutaneous semaglutide 2.4mg (the Wegovy dose) or placebo, on top of standard-of-care cardiovascular risk management, and followed them for a median of 39.8 months — among the longest follow-up periods of any GLP-1 cardiovascular trial. The primary endpoint was time to first occurrence of 3-point MACE (major adverse cardiovascular events): cardiovascular death, non-fatal myocardial infarction, or non-fatal stroke.
The result was unambiguous: 6.5% of the semaglutide group experienced a primary endpoint event compared to 8.0% of the placebo group, a hazard ratio of 0.80 (95% CI 0.72–0.90, P<0.001) — a 20% relative risk reduction. This crossed not just the pre-specified non-inferiority margin but the superiority threshold, meaning the trial demonstrated semaglutide actively reduces cardiovascular events, not merely that it fails to increase them. This was the first time any anti-obesity medication had demonstrated a direct reduction in major cardiovascular events in a dedicated outcomes trial.
The composite 3-point MACE endpoint is built from three distinct event types, and their individual hazard ratios tell a more nuanced story than the headline number alone:
| Endpoint | Hazard Ratio | Statistical Significance | Interpretation |
|---|---|---|---|
| 3-point MACE (primary composite) | 0.80 (0.72–0.90) | Significant, P<0.001 | 20% relative risk reduction; the trial's primary, pre-specified superiority result |
| Cardiovascular death | 0.85 | Directional trend, not independently powered for significance | Favored semaglutide but the component alone was not statistically powered as a standalone endpoint |
| Non-fatal myocardial infarction | 0.72 | Significant | A meaningful, statistically robust reduction in heart attacks — one of the strongest individual components |
| Non-fatal stroke | 0.67 | Significant | The largest relative reduction among individual components — roughly a third fewer strokes |
| Heart failure hospitalization (secondary endpoint) | 0.82 | Favored semaglutide | Consistent with the broader pattern of cardiovascular benefit extending beyond the primary composite |
Participants on semaglutide lost an average of 9.4% of body weight over the trial versus 0.9% in the placebo group — a substantial and expected difference given semaglutide's well-established weight-loss efficacy. The obvious hypothesis is that the cardiovascular benefit is simply downstream of weight loss: less body mass, less strain on the cardiovascular system, improved blood pressure and lipids, therefore fewer events. This is the assumption that has generally justified the "if it helps you lose weight, the heart benefit will follow" framing in obesity medicine.
But the formal mediation analysis (Ryan et al. 2023) found that weight loss explains only an estimated 31–43% of the observed cardiovascular risk reduction. That leaves the majority of the benefit unaccounted for by weight change alone — strong evidence that semaglutide is doing something to the cardiovascular system directly, independent of how much weight a given patient loses. This is a genuinely important and somewhat surprising finding: it suggests semaglutide should not be thought of purely as a weight-loss drug that happens to produce downstream cardiovascular improvement, but as a drug with its own direct cardioprotective biology.
Several mechanisms have been proposed to explain the "unaccounted for" 57–69% of the cardiovascular benefit. Semaglutide produces measurable reductions in systemic inflammatory markers, including C-reactive protein (CRP) and IL-6, at magnitudes that appear to exceed what would be expected from weight loss alone — consistent with a direct anti-inflammatory action on vascular endothelium and atherosclerotic plaque biology. GLP-1 receptors are also expressed directly on cardiac tissue (cardiomyocytes and coronary vascular endothelium), providing a plausible substrate for direct cardiac signaling rather than purely indirect, weight-mediated effects. Additional proposed contributors include plaque-stabilizing effects on existing atherosclerotic lesions, reductions in VLDL cholesterol particle production, and a modest but consistent blood pressure reduction of approximately 3.4 mmHg systolic — smaller than what a dedicated antihypertensive would produce, but large enough over 39.8 months of follow-up to contribute meaningfully to event reduction across a population of this size.
Beyond the primary cardiovascular endpoints, secondary and exploratory analyses of the SELECT trial dataset found a 36% lower incidence of alcohol use disorder diagnoses in the semaglutide group — consistent with the broader, independently growing evidence base on GLP-1 drugs and reward-pathway modulation covered elsewhere on this site. The trial also found a 14% lower rate of COVID-19 hospitalization among semaglutide-treated participants, a finding that is biologically plausible given obesity's known status as an independent COVID-19 severity risk factor, but which should be interpreted cautiously as an exploratory, non-primary finding rather than a confirmed drug effect.
The gastrointestinal side effect profile in SELECT matched what's well established from semaglutide's other trials: nausea occurred in 44% of the semaglutide group versus 16% of placebo, consistent with the drug's known GI tolerability pattern, and the majority of these events were mild to moderate and occurred during dose escalation. Critically, given the historical shadow of prior obesity drug withdrawals, SELECT found no pancreatic or thyroid cancer safety signal over the trial's multi-year follow-up — an important reassurance given ongoing monitoring for these theoretical risks associated with GLP-1 receptor agonism (based on rodent thyroid C-cell tumor data that has not been replicated in human epidemiological data to date).
SELECT did not emerge in isolation — it built on a body of GLP-1 cardiovascular outcomes trials conducted primarily in diabetic populations. LEADER (liraglutide, Marso et al. 2016, NEJM) found a 13% MACE reduction in type 2 diabetics with high cardiovascular risk. SUSTAIN-6 (semaglutide, also 2016) found a larger 26% MACE reduction, though in a smaller, shorter trial not powered as a definitive outcomes study. AMPLITUDE-O (efpeglenatide, Bhatt et al. 2021) found a 27% reduction using a different GLP-1 molecule, reinforcing that the cardiovascular benefit is likely a class effect of GLP-1 receptor agonism rather than unique to any single drug. What made SELECT distinct and, in some ways, more scientifically important than any of these was the deliberate exclusion of diabetics — removing glycemic control as a competing explanation and definitively establishing that GLP-1 cardiovascular benefit is not simply a diabetes-management side effect.
| Trial | Population | Drug | MACE Reduction |
|---|---|---|---|
| LEADER (2016) | Type 2 diabetes, high CV risk | Liraglutide | 13% |
| SUSTAIN-6 (2016) | Type 2 diabetes, high CV risk | Semaglutide | 26% |
| AMPLITUDE-O (2021) | Type 2 diabetes, CV or renal risk | Efpeglenatide | 27% |
| SELECT (2023) | Obesity + established CVD, NO diabetes | Semaglutide | 20% |
The FDA responded to SELECT's results by expanding Wegovy's label in March 2024 to include reduction of cardiovascular risk — specifically, reducing the risk of cardiovascular death, heart attack, and stroke in adults with cardiovascular disease and either obesity or overweight. This was a significant regulatory milestone: it reframed Wegovy from a weight-management drug to a drug with an FDA-recognized cardiovascular indication, a distinction with major implications for insurance coverage determinations, since many payers had historically excluded weight-loss drugs from coverage while covering drugs with cardiovascular indications more readily.
Cost-effectiveness analyses following SELECT have estimated a number needed to treat (NNT) of approximately 67 patients over 3 years to prevent one MACE event — a figure that compares reasonably against established secondary-prevention therapies; for context, statin therapy for secondary cardiovascular prevention carries an NNT of roughly 77 over a similar timeframe. Given semaglutide's substantially higher cost relative to generic statins, this comparison has become central to ongoing payer and health-system debates about cost-effectiveness thresholds for GLP-1 drugs in cardiovascular risk reduction, even as the clinical evidence for benefit is now well established.
SELECT enrolled a specific population: adults with obesity or overweight AND established cardiovascular disease, without diabetes. The 20% MACE reduction should not be assumed to generalize identically to people taking semaglutide purely for weight loss without pre-existing cardiovascular disease, or to people using compounded, non-FDA-approved semaglutide formulations, which were not studied in this trial. Discuss your individual cardiovascular risk profile and whether a GLP-1 drug is appropriate with your physician.
If you have established cardiovascular disease and obesity but not diabetes, SELECT provides the strongest evidence to date that semaglutide 2.4mg (Wegovy) can reduce your risk of a major cardiovascular event — independent of, and beyond, whatever weight you personally lose on the drug. Home monitoring of blood pressure is a reasonable complement to track the modest BP reduction semaglutide produces over time, and omega-3 fatty acids (EPA+DHA) are commonly used alongside GLP-1 therapy as a complementary cardiovascular risk-reduction strategy, given the independent evidence base for high-dose omega-3 in secondary cardiovascular prevention (though the two have not been studied together in a dedicated combination trial).