Impact of GLP-1 and dual agonists on the incidence of new cases of physician-reported sleep apnoea: A real-world propensity score-matched study

Abstract Background Obesity is the main risk factor for sleep apnoea. In the last years, few randomized studies have shown that glucagon-like peptide-1 (GLP-1) agonists and GLP-1/ glucose-dependent insulinotropic peptide (GIP) agonists (tirzepatide) were able to reduce the severity of sleep apnoea. Whether these important anti-obesity drugs may decrease the incidence of new cases of physician-reported sleep apnoea (PRSA) is unknown. Purpose To evaluate the impact of GLP-1 agonists (liraglutide, dulaglutide, semaglutide) and dual agonists (tirzepatide) as compared to no pharmacological/surgical weight loss treatment on the incidence of new cases of PRSA. Methods We conducted a retrospective cohort analysis using the TriNetX Global Health Research Network through anonymised electronic medical records. We identified all adult patients (18 years) with a history of obesity and/or type 2 diabetes mellitus without previous anti-obesity therapy (reference arm) and those new starters of GLP-1 agonists and tirzepatide. The index event (June 2022) was chosen to coincide with the launching of tirzepatide in the market. The follow-up time was 982 days (up to February 07, 2025). We excluded patients with previous diagnosis of PRSA, under previous use of GLP-1/dual agonists, orlistat, bupropion/naltrexone as well as patients with previous bariatric surgery. New starters were matched to the reference cohort using propensity score matching in a 1:1 ratio for age, sex, ethnicity and body mass index. We used multiple code recommendations for tracking previous or after the index event PRSA, according to the international Classification of Diseases 10th edition: G47.33, G47.3, G47.39 or G47.30. Results After propensity score matching, a total of 1,808,446 patients (904,223 in each group) were included in the reference arm versus GLP-1 and dual agonists’ comparison. Overall, GLP-1/dual agonists were associated with 60% lower incidence of new cases of PRSA (HR: 0.40; 95% CI: 0.38-0.42). Individually, all drugs were able to reach this outcome as compared to the reference arm: liraglutide (34% reduction; HR: 0.66, 95% CI: 0.52-0.83); dulaglutide (74% reduction; HR: 0.27, 95% CI: 0.23-0.31); semaglutide (46% reduction; HR:0.54, 95% CI: 0.50-0.57); tirzepatide (37.2% reduction; HR: 0.63, 95% CI: 0.54-0.73). Conclusion As compared to no pharmacological/surgical weight loss therapies, GLP-1 and dual agonists reduced the incidence of new cases of PRSA in patients with obesity and/or type 2 diabetes. Further research and focused randomised controlled trials are warranted to explore the broader clinical implications of these treatments as a preventive strategy for sleep apnoea incidence.