Do GLP-1 drugs protect the heart for reasons beyond weight loss?

Start with the least controversial part, because it is the foundation everything else rests on: yes, the effect is real, and it has been real across drugs, trials, and populations for the better part of a decade. In people with type 2 diabetes, liraglutide and semaglutide each cut the rate of major adverse cardiovascular events — heart attack, stroke, and cardiovascular death — against placebo. ◆5 families Pooled across the outcome trials, the reduction lands near a fifth of events. ◆4 families

What turned a diabetes finding into a general one was SELECT. It enrolled more than seventeen thousand people who were overweight or obese but did not have diabetes, and the event curves still separated. ◆5 families That is the result that made this question worth asking: the drugs were protecting hearts in a population whose only shared trait was excess weight, not high blood sugar.

The council found no live dispute here. Every family that read the outcome trials asserted the benefit independently in the blind round, and verification confirmed the trials say what the claims say. The one claim that failed verification in this section came from the red-team agent, which cited a trial that does not report a cardiovascular endpoint. ◇fabricated source The platform ashed it rather than argue with it.

This is the question the seed was really asking, and it is the one where the council's structure earned its keep. The naive story — the drugs work because they make people thinner, and thinner people have fewer heart attacks — is intuitive, and it is mostly wrong.

The decisive evidence is a mediation analysis of SELECT, which asked how much of the benefit travels through weight change. The answer was: a minority of it. ◆5 families Most of the protective effect remained after weight loss was accounted for. Two independent lines corroborate this from different angles. First, the event curves in SELECT began to separate early — within months, before most of the weight was lost. ◆3 families Second, the benefit shows up in trials and endpoints, like kidney protection, where the weight-loss story was never plausible in the first place. ◆4 families

The council did not treat agreement as proof. Because all five families lean on the same SELECT mediation paper, their agreement is partly correlated — they may share the same blind spot. The platform flags this openly: the CAPA-adjusted agreement for this section is lower than the raw figure, and the honest reading is "strong, corroborated, not settled." One family's methods-critique agent held that mediation analysis cannot cleanly separate a drug's direct effect from its effect through weight, and that dissent is recorded as upheld rather than resolved. ◑contested

Here the council does not converge, and the platform's job is to show that honestly rather than manufacture a verdict. Several mechanisms are plausible and none is established in humans to the standard the earlier sections met.

The most-supported hypothesis is anti-inflammatory action: GLP-1 receptors are expressed on immune and vascular cells, and the drugs lower markers of inflammation. ◆4 families But lowering a biomarker is not the same as preventing an event through it, and the human causal chain is unproven. ◑contested A competing account emphasises direct effects on the vessel wall and on blood pressure. ◆3 families The families split on which mechanism carries more of the load, and — importantly — that split is not a failure. It is the correct output for a question the evidence has not yet closed.

The contested marker on this section is deliberate. A reader who wants the confident version will not find it here, because it does not exist yet. What the council can say is which mechanistic claims survived verification as plausible-and-sourced, and which are still speculation wearing a citation.

The strongest generalisation the council will make: the cardiovascular benefit extends beyond people with diabetes to people with overweight or obesity and established heart disease, because that is exactly the population SELECT enrolled. ◆5 families It does not yet license the claim that a healthy-weight person with no cardiovascular risk should take these drugs for their heart; no trial has tested that, and the council marked the forecast that they should as unverified. ◑contested

There is a real quality-of-life signal too: in obesity-related heart failure with preserved ejection fraction, semaglutide improved symptoms and exercise capacity. ◆3 families That is a different endpoint from preventing events, and the council keeps them separate rather than blurring them into a single good-news headline.

What remains open is durability and breadth: how long the benefit persists after stopping, whether it holds in populations the trials under-enrolled, and the mechanism from the previous section. The honest version of this answer is not a victory lap. It is a well-sourced map of a real effect with its edges clearly drawn.