There is one study, it covers 141 matched pairs, and it publishes no effect size for any outcome, so the answer is that nobody can tell you how much difference it makes. A left ventricular assist device is a mechanical pump implanted when a heart is failing badly enough to need one, often as a bridge to transplant. Obesity complicates the surgery, increases infection around the driveline that exits the skin, and at many centers blocks transplant listing outright. Among 527 people with obesity who received one, 231 had documented GLP-1 exposure before implantation, and 141 pairs were matched[1]. It sits among the growing set of results in populations trials never enroll, alongside the sickle cell cohort.
The reported direction is favorable across the board: fewer driveline infections at one year, and at three years lower mortality along with more transplant listings and more transplants performed. In a population this sick, all of those would be meaningful.
The exposure definition carries the heavier problem. Patients qualified by having a GLP-1 prescription before implantation, so the comparison is between people who arrived at major cardiac surgery already on one and people who did not. Being on a GLP-1 drug beforehand means being stable enough to have been started on it, engaged enough to have continued, and managed in advance rather than arriving in crisis. In a group where reaching elective implantation in good condition largely determines what follows, that is close to selecting on the outcome — the same structure flagged in the dialysis cohort.
One finding deserves separating from the rest because its mechanism is probably not cardiac at all. Transplant listing is a committee decision, and a BMI ceiling is a common explicit criterion. More listings among GLP-1 users may be recording that people crossed a threshold on a chart rather than that their hearts improved. That is a genuine and important benefit, because access to transplantation is the thing that changes survival here. It is still a different claim from the drug treating heart failure, which is examined on its own terms in the heart failure evidence and the rheumatoid arthritis cohort.
A second cohort in a comparably sick population shows what this one is missing. Among patients with end-stage kidney disease on dialysis, GLP-1 use was studied for hospitalization and mortality with published risk estimates attached [2], which is what lets a reader judge whether a difference is large or marginal. Both studies select people well enough to have been started on the drug, and only one of them lets you see how big the resulting gap was.
The authors describe their findings as observational and hypothesis-generating and ask for prospective evaluation, which is the right conclusion from 141 pairs with no published effect sizes. What it establishes is that the question deserves asking properly, in a population where weight is not a cosmetic concern but a gate to the treatment that would save them.