Among 119,911 matched pairs of adults with established atherosclerotic cardiovascular disease, BMI of 30 or higher, and statin use, those who started a GLP-1 receptor agonist had markedly lower three-year all-cause mortality (4.0% vs 7.5%; HR 0.479), myocardial infarction (HR 0.762), and ischemic stroke (HR 0.671) than statin-only controls. Results held in a BMI 35-plus subgroup, four washout sensitivity analyses, and after excluding familial hypercholesterolemia. LDL differences between groups were negligible.
The direction fits the randomized evidence: SELECT showed roughly 20% fewer major cardiovascular events with semaglutide in people with obesity and prior cardiovascular disease. But a 52% reduction in death is implausibly large, far beyond anything seen in trials. That gap is a red flag for residual confounding, since people who obtain and persist on GLP-1 drugs tend to be healthier, better insured, and more engaged with care, and TriNetX electronic records capture lifestyle, adherence, and cause of death poorly. Immortal-time bias and differential loss to follow-up could also inflate the effect. The near-null LDL finding suggests any benefit runs through weight loss, inflammation, or blood pressure rather than cholesterol. This is a preprint awaiting peer review, and the estimates may change. Practically, adults with heart disease and obesity have a sound reason to ask their cardiologist about GLP-1 therapy, but they should anchor expectations to trial-sized benefits. Confirmatory in direction, unreliable in magnitude.