Atrial fibrillation (AF) is the most common sustained arrhythmia in the adult population, with estimates that it costs the NHS in the UK over one billion pounds per year. There is growing interest in the ability to predict those individuals who are most likely to develop atrial fibrillation (AF) based on 12-lead electrocardiograms (ECGs) whilst in sinus rhythm, which has been fuelled by advancements in digital ECG analysis and machine learning.
Identifying such individuals may facilitate more aggressive approaches to AF-related stroke prevention, and it could potentially reduce the burden of symptoms and AF-cardiomyopathy-associated morbidity.
In this comprehensive narrative review, we assess the role of different 12-lead ECG parameters as predictive tools for future AF development and consider the direction of future research in the field of the ECG prediction of AF. View this paper