Across 37 phase 3 trials of GLP-1 drugs and newer multi-hormone medicines, the trials that produced more weight loss also produced bigger drops in blood pressure, and differences in weight loss accounted for about three-quarters of the differences in systolic pressure between trials. The pooled analysis, published in EClinicalMedicine on 12 October 2026, covers 50,171 adults with overweight or obesity. For anyone taking one of these drugs, it suggests the blood-pressure effect largely travels with the weight that comes off, rather than arriving as a separate benefit.
What the analysis measured
The authors, based mainly at the University Health Network and Sunnybrook Health Sciences Centre in Toronto, searched MEDLINE, Embase and the Cochrane databases to December 2025 for phase 3 randomised trials designed to treat obesity. Thirty-seven qualified: 18 of semaglutide, seven of tirzepatide, six of liraglutide, two each of CagriSema and orforglipron, and one each of mazdutide and ecnoglutide. Median treatment lasted 68 weeks; 57% of participants had hypertension and 16% had type 2 diabetes. Thirty-six of the trials were rated at low risk of bias in every domain.
Every figure is placebo-adjusted: the drug group’s change minus the placebo group’s. Pooled, that came to a 10.1% weight reduction (95% confidence interval 9.0% to 11.3%) and a 4.9 mmHg fall in systolic pressure (4.3 to 5.5 mmHg). But the trials differed enormously in how much weight they produced (an I² of 98.4%, meaning almost all of the spread reflects real differences between trials rather than chance).
The central result is a meta-regression, which plotted each trial’s weight loss against its blood-pressure change. The fitted slope was 0.37 mmHg of systolic pressure per percentage point of weight lost (95% CI 0.28 to 0.45), and weight loss explained 76% of the between-trial variance (R² = 0.76). On that slope, a trial producing 5 points more weight loss than another would be expected to show about 1.9 mmHg more blood-pressure reduction (5 × 0.37); 15 points more, about 5.6 mmHg. The slope stayed at 0.37 after adjusting for follow-up length, baseline BMI, baseline systolic pressure, sex and diabetes. It rose to 0.52 (0.21 to 0.82) when the specific drug was added to the model, a much wider interval that leaves room for drug-specific effects in either direction.
How this fits what came before
That weight loss lowers blood pressure is not new. A 2003 meta-analysis in Hypertension, which the authors cite, pooled 25 lifestyle trials in 4,874 people and found about 1 mmHg of systolic reduction per kilogram lost. That figure is per kilogram rather than per percentage point, so the two are not directly comparable. The question this paper addresses is narrower: whether the newer drugs, which act on hormone receptors found well beyond the gut, lower blood pressure through some route other than weight. If they did, the trials would scatter well away from the regression line. Mostly, they did not.
That changes how to read trial reports. When results for a new agent highlight a fall in blood pressure, this analysis suggests the first question is how much weight the trial produced, because a drug that takes off more weight would be expected to show a larger blood-pressure effect for that reason alone. The same applies when comparing results for semaglutide and tirzepatide: differences in blood-pressure results across their trials largely track differences in weight loss.
The authors found no meaningful change in the relationship by baseline blood pressure, and the direction of effect was the same in every subgroup they examined. Blood pressure is still a surrogate, though. This analysis says nothing directly about heart attacks, strokes or kidney outcomes.
What is not yet known
The authors list several limits themselves. Blood pressure was not a primary endpoint in any of the 37 trials, so none was designed or powered to measure it, and readings were taken in the clinic rather than by home or 24-hour monitoring. The regression uses one average per trial, which leaves it open to ecological bias: it shows that trials with more weight loss had larger blood-pressure falls, not that each person’s pressure falls in step with their own weight. Changes to participants’ blood-pressure medication during the trials could not be adjusted for uniformly. Half the trials were of semaglutide, and agents still in phase 3, such as retatrutide and survodutide, were not included. The study had no funding; the lead author joined Novo Nordisk Canada after the work was done and states the company had no involvement.
Sources: Nardone M, Kugathasan L, Daunt L, et al. “Weight loss and blood pressure lowering with GLP-1 receptor agonists and multi-hormone receptor modulators in overweight and obesity: a systematic review and meta-regression of phase 3 clinical trials.” EClinicalMedicine 2026;101:104251, published 12 October 2026. Journal article (DOI); PubMed record, PMID 42903654.
