An Italian study found roughly double the combined outcome rate among patients with Type 2 diabetes who stopped GLP-1 receptor agonists compared with those who continued; none had established cardiovascular disease initially.
Published Nov. 9 in The Lancet Regional Health – Europe, the retrospective study analyzed data spanning 2004-24 from 301 Italian diabetes clinics. Its cohort included 80,302 people who had no established cardiovascular disease when they began treatment.
By their last recorded visit, 48,615 participants (60.5%) had stopped treatment, while 31,687 (39.5%) remained on it. For the matched comparison, researchers selected 28,713 people from each group.
The combined measure counted nonfatal heart attacks and strokes, along with deaths from any cause. The rate was 3.94 events per 100 person-years among those who stopped treatment and 1.96 among those who continued. The hazard ratio was 2.03 (95% confidence interval, 1.92-2.14).
Treatment duration also differed between the groups: an average of 1.21 years among those who stopped and 2.97 years among those who continued. When researchers looked at death from any cause separately, they reported a hazard ratio of 4.81 (95% confidence interval, 4.39-5.26) for the group that stopped, compared with those who continued. That measure covers deaths from any cause, not cardiovascular deaths alone.
The records did not explain why participants ended treatment, and the observational analysis cannot show whether stopping led to the higher event rate. Prof. Naveed Sattar, a professor of cardiometabolic medicine and honorary consultant at the University of Glasgow, said the excess could “readily reflect residual confounding.” People who stop treatment may differ from those who continue in ways that also affect their health, he said.
The results concern people with Type 2 diabetes who had no established cardiovascular disease when they started GLP-1 treatment. They do not establish outcomes for people using the medicines for obesity without diabetes or for people with existing cardiovascular disease. Researchers said further work is needed to determine whether discontinuation contributes to risk or reflects other health changes.
