GLP-1 medications have become increasingly popular in recent years for managing obesity.1 Another obesity management treatment that has been an option for patients for decades is bariatric surgery. Bariatric surgery works by restricting or bypassing the stomach to reduce the amount of food and caloric intake the patient has following surgery.2 We aimed to compare the benefits and risks of both treatment options.
We first analyzed weight change patterns in the year following initial treatment with either bariatric surgery or prescription of a weight-loss dosage of a GLP-1 medication. We matched patients based on sex, age, number of weight readings in the follow-up year, and BMI classification at the start of treatment. We found that patients who underwent bariatric surgery experienced significantly greater weight loss in the year following treatment and had lower rates of weight gain compared to those on one of the GLP-1 medications, as seen in Figure 1.
We further explored the proportion of patients who lost at least 15% of their starting body weight. Of patients who underwent bariatric surgery, 76% lost at least 15% of their starting body weight within one year, as seen in Figure 2. Among the patients prescribed GLP-1 medications, those who were prescribed tirzepatide experienced the greatest weight loss, with 46% losing at least 15% of their starting body weight within a year, while 30% of semaglutide patients and 8% of liraglutide patients lost at least 15% of their starting body weight.
We also examined the three-year prevalence of angina, transient ischemic attack (TIA), myocardial infarction (MI), ischemic stroke, and hemorrhagic stroke among 208,692 patients on any dose of a GLP-1 medication and 208,388 patients who had bariatric surgery. We matched on patient age, sex, and BMI classification.
We found that patients who were prescribed a GLP-1 medication had higher rates of angina and MI within three years, while the rates of ischemic stroke, TIA, and hemorrhagic stroke were higher but not statistically significant, as seen in Figure 3.
These data come from Cosmos, a dataset created in collaboration with a community of Epic health systems representing more than 277 million patient records from 1,500 hospitals and more than 36,000 clinics from all 50 states, Lebanon, and Saudi Arabia. This study was completed by two teams that worked independently, each composed of a clinician and research scientists. The two teams came to similar conclusions. Graphics by Brian Olson.
- Yeo YH, Rezaie A, Hsieh TYJ, et al. Shifting trends in the indication of glucagon-like peptide-1 receptor agonist prescriptions: A nationwide analysis. Ann Intern Med. 2024;177(9):1289-1291. doi:10.7326/m24-0019
- Faria GR. A brief history of bariatric surgery. Porto Biomed J. 2017;2(3):90-92. doi:10.1016/j.pbj.2017.01.008