Introduction: Atrial fibrillation (AF) is the most prevalent cardiac arrhythmia, affecting approximately 3% of adults over 60, with prevalence increasing with age. AF is known to be associated with a five-fold increase in stroke risk and a 3.5-fold increase in mortality. This research investigates the atrial fibrillation burden (AFB), defined as the percentage of time spent in AF, as a potential predictor of heart failure development.
Methods: To test our hypothesis, we analyzed 69,663 Holter recordings from 47,729 individual patients. We trained a random forest classifier using AFB, age, sex, and body mass index to predict the development of heart failure over a five-year period.
Results: The model using AFB as a sole predictor achieved an AUROC of 0.64, which improved to 0.78 when incorporating additional demographic variables.
Conclusion: This study provides evidence that AFB serves as an independent risk factor for the development of heart failure, with predictive power that is enhanced when combined with traditional clinical variables.