The review pooled seven randomised trials with prespecified, adjudicated stroke outcomes comparing GLP-1 receptor agonists with placebo or usual care in adults with type 2 diabetes at high cardiovascular risk, totalling 50,617 participants. Stroke risk was 17% lower (HR 0.83, 95% CI 0.75-0.92), and meta-regression associated the effect with trial-level systolic blood pressure differences but not with HbA1c or weight, though the mean 1.45 mmHg difference predicts only about a 4% reduction. Four trials in established neurodegenerative disease, 4,206 participants, reported no cognitive benefit, in a preprint.
Why it is interesting: It separates the stroke benefit of these drugs from the cardiometabolic changes usually invoked to explain it, and records no parallel signal for cognition.