Hypertension and obesity travel together, so the people most likely to be prescribed a weight drug are disproportionately the people this review is about. After four updates and eight trials, its conclusion is that the evidence remains insufficient to say whether these drugs reduce death or cardiovascular events in that group [1].
The reason is not that the trials found nothing. It is that the two newest trials measured nothing the review needed. For both semaglutide and tirzepatide, the review records that no results were reported for all-cause mortality, cardiovascular morbidity, serious adverse events or adverse events. Liraglutide could not be entered at all, because no trial has published separate figures for hypertensive participants — a gap of exactly the kind this site described in who actually gets into trials.
The older drugs did report, and the reporting is mixed. Orlistat probably increases serious adverse events against placebo, at a risk ratio of 1.45 with a confidence interval from 1.10 to 1.91 across three trials and 1,476 participants, rated moderate certainty. Naltrexone with bupropion probably increases all adverse events at RR 1.69 (95% CI 1.58 to 1.80) in a single trial of 8,283 people, while showing no difference in serious ones. Phentermine with topiramate probably increases all adverse events at RR 1.13 (95% CI 1.08 to 1.20) — the tolerability ordering this site set out in the network meta-analysis of nineteen drugs.
Two facts about the evidence base frame the rest. Seven of the eight trials were funded by the pharmaceutical industry, which is ordinary and worth knowing. And the review itself had no dedicated funding, which is why its authors can end on a request rather than a conclusion: that completed trials enrolling both hypertensive and normotensive participants publish the hypertensive results separately.
For a reader with high blood pressure weighing one of these drugs, the practical position is that the weight and blood-pressure numbers exist while the hard-outcome numbers do not, in this population. That is a narrower gap than it sounds, because the cardiovascular outcome trials in broader populations are real and this site has covered them — including the surgery comparison on cardiovascular outcomes. It is still a gap, and it is four updates old.