Abstract
BACKGROUND: Conventional follow-up after atrial fibrillation (AF) catheter ablation relies on physician-led interval monitoring and often fails to characterize paroxysmal symptoms. An increasing number of patients use smartwatch-based ECG devices for rhythm monitoring, but their structured integration into clinical workflows and the handling of the resultant data are not well described. OBJECTIVES: To describe the design, operationalization, data pipeline, and user engagement of a patient-led smartwatch ECG follow-up strategy after AF ablation within a randomized clinical trial. METHODS: A prospective, randomized controlled trial of adults undergoing first-time AF ablation was conducted. Participants were randomized to an Apple Watch-based protocol (daily and symptom-triggered ECGs) or standard follow-up. A prespecified audit of the smartwatch-derived rhythm classification was conducted. User engagement, symptom annotation, and downstream resource use were quantified. Primary clinical outcomes are reported in a companion Brief Report. RESULTS: Of the 168 enrolled participants (mean age 60.5 ± 9.9 years, 52 (31.0%) female, 84 (50.0%) persistent AF), Active-arm participants recorded a median of 170 (IQR 93-380) ECGs over 12 months and transmitted a median of 1.9% (0.0-8.3) for review. Symptom-annotated ECGs were more likely to show AF compared with unannotated ECGs (OR 16.1, 95% CI 13.0-19.9, P < 0.001) Watch-derived AF and sinus rhythm labels had positive predictive values of 0.96 and 0.95 respectively, although one-third of ECGs were unclassified. CONCLUSIONS: A structured, patient-led smartwatch ECG workflow can be embedded into routine post-ablation care with high engagement, modest staff workload, and accurate device-level rhythm classification. This implementation framework provides a practical template for integrating patient-generated wearable data into AF follow-up pathways and future digitally enabled trials.