Semaglutide is associated with lower rates of cardiovascular outcomes and death than positive airway pressure in patients with overweight/obesity and sleep apnea: a propensity score matching study.
B S Prado, D R Mazzotti, A Franci, L M Baracioli, K R Barbado, R Saretta, R H Furtado, R D Lopes, R Kalil-Filho, L F DragerAbstract
Background
The SELECT trial demonstrated that semaglutide reduces cardiovascular (CV) events in patients with overweight or obesity without diabetes. Obstructive Sleep Apnea (OSA) is very common in patients with obesity. Positive airway pressure (PAP) is the main treatment for OSA, but it usually does not promote weight loss. In addition, its CV benefit is uncertain. The association between semaglutide, PAP, and CV outcomes in patients with OSA remains unclear.
Purpose
To evaluate the association between semaglutide, PAP, and CV events in patients with OSA.
Methods
We conducted a retrospective cohort study using the TriNetX Global Health Research Network based on anonymized electronic health records (EHR). We identified adult patients with a body mass index ≥ 27 kg/m2 (as in the SELECT trial) and OSA (using a validated EHR algorithm). We excluded patients with diabetes, use of GLP-1 agonists (other than semaglutide), tirzepatide, orlistat, phentermine/topiramate, or bupropion/naltrexone, and those with prior bariatric surgery. Evidence of PAP initiation was based on PAP claims after OSA diagnosis. We performed three distinct pair analyses: Semaglutide-only vs. PAP-only; semaglutide-only vs. PAP+semaglutide, and PAP-only vs. PAP+semaglutide. We performed 1:1 propensity score matching on age, sex, ethnicity, body mass index (BMI), smoking, hypertension, dyslipidemia, previous myocardial infarction, stroke, heart failure, and chronic kidney disease. Our primary outcome was the combined events of myocardial infarction, stroke, and mortality. Secondary outcomes included individual components of the primary outcome.
Results
After propensity score matching, 45,700 patients were included for the comparison of semaglutide-only vs. PAP-only (mean age: 55±13 years, 51.9% males, mean BMI: 39±8 Kg/m2). After a median follow-up of 365 days, semaglutide-only was associated with fewer combined events than PAP-only (n=571 vs. n=2,574; HR=0.203; 95% CI 0.185-0.222). The figure shows the survival probability of the primary outcome in the two groups. Secondary outcome analysis showed consistent results. In comparisons of semaglutide-only vs. semaglutide+PAP (n=1,123 patients per group), the differences were attenuated but clearly favored semaglutide-only (n=65 vs. n=120; HR=0.527; 95% CI 0.390-0.713). When comparing PAP-only with PAP+semaglutide (n=1,127 patients per group), we observed twice as many combined clinical events in the PAP-only group (n=240 vs. n=120; HR=2.239; 95% CI 1.799-2.788).
Conclusions
In patients with overweight/obesity and OSA, with no diagnosis of diabetes, semaglutide is associated with lower rates of CV outcomes and mortality when compared with PAP or combination therapy. Randomized trials are warranted to confirm our findings.Figure