Geographical region differences in the Cabana trials

Abstract

Background: Unexpected variations in treatment benefits by geographic region raise questions about the generalizability of findings from international clinical trials. Purpose: Investigate the outcome of catheter ablation vs drug therapy in CABANA patients with AF treated at North American (NA) compared with non‐NA sites, a pre‐specified comparison. Methods: CABANA randomized 2204 symptomatic patients with atrial fibrillation (AF) to ablation vs drug therapy at 126 centers in 10 countries. The primary endpoint was a composite of death, disabling stroke, serious bleeding, or cardiac arrest. Results: There were 1285 patients enrolled at 85 NA sites, and 919 at 45 non‐NA sites. At baseline, NA and non‐NA patients differed significantly with regard to age, sex, BMI, CCS, CHA2DS2‐VASc score, family history of AF, prior AF hospitalization, history of A Flutter, diabetes, NYHA class, and proportion with sleep apnea, heart disease, CAD, CHF, or EF >35%. Fouryear Kaplan‐Meier event rates showed considerable differences in treatment effects between NA and non‐NA patients (see Table). Ablation was associated with a 50% reduction in the primary endpoint in non‐NA (HR 0.50, 95% CI 0.24 to 1.03) vs a 5% reduction in NA (HR 0.95, 95% CI 0.69 to 1.30) enrollees despite higher event rates in NA patients. The trend of favorable outcome with ablation at non‐NA sites was maintained after adjustment for baseline covariates that differed between the non‐NA (HR 0.51, 95% CI 0.24 to 1.04) and NA populations (HR 0.93, 95% CI 0.68 to 1.20). Conclusion: We observed substantial geographic variation in clinical outcomes and treatment effects in the CABANA trial that suggests an important disconnection between overall event rates and treatment benefit from ablation in AF.