Wearable Devices for Quantifying Atrial Fibrillation Burden: A Systematic Review and Bayesian Meta-Analysis

Background: Atrial fibrillation (AF) is the most common supraventricular arrhythmia and is associated with an impaired prognosis. Studies using implantable cardiac monitors suggest that this association is closely linked to AF burden, defined as the percentage of time spent in AF. Consequently, there is a growing need for affordable and comfortable alternative devices, such as wearables, capable of reliably monitoring AF burden in patients with AF. Methods: Major electronic databases were searched for studies comparing AF burden quantification using wearables and reference ECG monitoring methods. A Bayesian approach was adopted for the final analysis. Results: Six studies, including a total of 448 patients and 36,978 h of valid simultaneous recordings, were analyzed. Bayesian analysis revealed no statistically significant differences between wearables and reference methods in AF burden quantification. The mean error was 1% (95% CrIs: -4% to 7%). Similar findings were observed in the subgroup analysis of studies assessing only smartwatches. Between-study heterogeneity was low, and no evidence of publication bias was detected. Conclusion: Our analysis suggests that AF burden quantification using wearables is comparable to reference ECG monitoring methods. These findings support the potential role of wearables in clinical practice, particularly for research and prognostic purposes. However, more studies are needed to determine whether the observed statistical equivalence translates to clinical significance, thereby supporting the widespread use of wearables in the assessment of rhythm control therapeutic strategies.