Abstract 14489: GLP-1 Receptor Agonists vs. SGLT2 Inhibitors and Risk of Atrial Fibrillation in Patients with Obstructive Sleep Apnea and Obesity: A Propensity-Matched Analysis
Abstract
Introduction: Obstructive sleep apnea (OSA) and obesity are established risk factors for atrial fibrillation (AF) through shared mechanisms including chronic intermittent hypoxia, sympathetic activation, and atrial remodeling. GLP-1 receptor agonists (GLP-1 RA) and SGLT2 inhibitors (SGLT2i) are increasingly used in this population, yet their comparative effects on AF and cardiovascular outcomes remain unclear. Hypothesis We hypothesized that GLP-1 RA use is associated with lower risk of incident AF compared to SGLT2i in patients with OSA and obesity, with concurrent benefits across secondary cardiovascular outcomes. Methods: Using the TriNetX Global Collaborative Network (171 healthcare organizations), we identified adults aged 18-85 with OSA and obesity who initiated GLP-1 RA (N=320,923) or SGLT2i (N=127,974) after January 2012, with mutual exclusion of the other drug class (Figure 1). Outcomes were analyzed at 1, 3, and 5 years. Propensity score matching (PSM) on 50 covariates (demographics, diagnoses, medications, lab values, CPAP use) yielded 72,778 matched pairs. Primary outcome was incident AF (ICD-10: I48.0-I49.2, I48.91). Secondary outcomes included pulmonary hypertension (PH), myocardial infarction (MI), heart failure (HF), and all-cause mortality. Results: After PSM, cohorts were well-balanced (mean age 62 years, 59% male, 72% T2DM). GLP-1 RA was associated with significantly lower risk of incident AF at 1 year (HR 0.682, 95% CI 0.637-0.731), 3 years (HR 0.726, 95% CI 0.691-0.764), and 5 years (HR 0.761, 95% CI 0.727-0.797; all p<0.001) (Figure 2). For secondary outcomes at 5 years, GLP-1 RA demonstrated markedly lower risk of PH (HR 0.603, 95% CI 0.573-0.635), MI (HR 0.696, 95% CI 0.660-0.734), HF (HR 0.798, 95% CI 0.764-0.833), and all-cause mortality (HR 0.870, 95% CI 0.839-0.902; all p<0.001). The protective effect of GLP-1 RA was most pronounced at 1 year and attenuated modestly over time, yet remained significant across all timepoints. Conclusion: In patients with OSA and obesity, GLP-1 RA use is associated with substantially lower risk of incident AF and major cardiovascular outcomes compared to SGLT2i across 1-, 3-, and 5-year follow-up. These findings support a preferential cardioprotective role for GLP-1 RA in this high-risk population and warrant further investigation through prospective trials.