AHA Scientific Sessions 2026 Abstract · Nov 2, 2026
Cost-Offset Analysis of GLP-1 Treatment for Atrial Fibrillation: The Liraglutide Effects in Atrial Fibrillation (LEAF) Study
By McCollister K, French M, Aguilar V et al.
Background: Atrial Fibrillation (AF) drives substantial healthcare cost among individuals with obesity and persistent AF, where recurrence and repeat ablation are common. Epicardial adipose tissue (EAT) is linked to AF burden, and risk factor modification (RFM) may improve outcomes and reduce costs. Liraglutide, a GLP-1 receptor agonist, reduces weight and EAT. The Liraglutide Effects in Atrial Fibrillation (LEAF) trial tested adding liraglutide to RFM before ablation; this analysis evaluates whether cost reductions offset liraglutide's incremental cost versus RFM alone. Research Questions. To assess 6- and 12-month healthcare costs of adjunctive liraglutide in overweight or obese patients with AF undergoing ablation, and whether cost reductions offset its incremental cost. Methods: This cost-offset analysis was conducted alongside the LEAF randomized trial in overweight/obese adults (BMI ≥27 kg/m ² ; 80% persistent AF) randomized to RFM or RFM plus liraglutide (RFM+L) for 3 months pre-ablation. Costs were estimated using national fee schedules and published cost studies. Per-patient totals at 6 and 12 months, along with cost offsets were computed with outlier-exclusion and log-transformed sensitivity analyses. A gamma model (log link) tested predictors of 12-month cost, and a prospective analysis compared repeat ablations from months 12-36. Results: Twenty-eight patients were assigned to RFM and 31 to RFM+L. Mean age was 61.8±10.3 vs 62.2±8.6 years; most were male and White, with a considerable proportion identifying as Hispanic. Body weight and BMI were slightly higher in RFM+L (109.6 vs 102.8 kg; 37.2 vs 34.2 kg/m ² ), though neither difference was significant. At 6 months, per-patient total costs were $36,023 (SD $12,544) for RFM and $33,614 (SD $11,957) for RFM+L (p=0.68); incremental liraglutide cost was $3,146. Over 12 months, totals were $44,055 (SD $16,138) and $42,002 (SD $19,263) (p=0.35), with no significant differences in medications (excluding GLP-1), diagnostics, procedures, or visits. After accounting for GLP-1 costs, the net cost offset for RFM+L was -$2,052 per patient (nonsignificant); excluding outliers (>3 SD) yielded -$4,775 (also nonsignificant). From months 12-36, repeat ablations were more frequent in RFM vs RFM+L (9 vs 2; Fisher p=0.018). One-year freedom from AF/atrial flutter was 81% (95% CI 62-91%) for RFM+L vs 54% (95% CI 34-70%) for RFM (log-rank p=0.007). In gamma regression, no covariate—including weight change, BMI, sex, age, or AF type—significantly predicted 12-month cost. Conclusions: Although adjunctive liraglutide did not significantly reduce overall healthcare costs, it reduced AF/atrial flutter recurrence and repeat ablations through 36 months. The -$2,052 net cost offset likely underestimates longer-term economic value of improved arrhythmia control. Larger studies of pharmacologic risk-factor modification are warranted.