Is a single-lead smartwatch ECG capable of detecting a wider spectrum of arrhythmias in clinical practice?

Abstract Introduction A commercially available smartwatch (SW), such as those equipped with photoplethysmography (PPG) and electrocardiogram (ECG) functions, is capable of detecting atrial fibrillation (AF) in a real-world setting. However, previous studies have reported a high occurrence of inconclusive results, ranging from 10% to 30%, potentially leading to patient uncertainty and unnecessary medical consultations. Expanding the clinical application of such devices beyond AF detection could enhance remote cardiac monitoring, facilitate the diagnosis of sporadic arrhythmias, and reduce the burden on healthcare systems. Objective This study assesses the potential of a commercially available SW to detect a wider spectrum of arrhythmias beyond AF. Methods This was a subanalysis within the EQUAL study, a prospective, multicenter, randomised control trial. The study population included patients aged 65 years and older with a CHA2DS2-VASc-score of ≥2 for men or ≥3 for women, with no prior history of AF. Patients in the intervention group were connected with a Dutch remote monitoring program by a SW with built-in PPG and single-lead ECG functions. Patients were instructed to record a 30-second ECG using the SW upon receiving an irregular rhythm notification or when experiencing symptoms. The study period ended after six months or upon AF diagnosis, whichever occurred first. All ECG recordings were evaluated by an independent eHealth team. Final data collection will be completed on July 17, 2025. This subanalysis includes all recorded single-lead ECG data. Results Currently, 6425 ECG recordings were collected from 218 patients. Of these ECG recordings, 935 (14.5%) were classified as inconclusive by the SW. The eHealth team successfully interpreted 911 (97.0%) of these inconclusive recordings, identifying multiple distinct arrhythmias beyond AF, including 4 (0.4%) AV nodal re-entry tachycardia, 15 (1.6%) atrial tachycardia, 612 (67.2%) premature ectopic beats in bigeminy or trigeminy patterns, and 10 (1.1%) supraventricular tachycardia, not otherwise specified. Additionally, 2 (0.2%) ECG recordings were classified as AF, 134 (14.7%) as sinus rhythm, 26 (2.9%) sinus bradycardia, 19 (2.1%) sinus tachycardia, and 89 (9.8%) as sinus rhythm with noise. Conclusion This study challenges the limits of a commercially available SW, revealing its potential as a powerful remote cardiac monitoring tool in clinical practice. When paired with a healthcare professional’s expertise, SW-based telemonitoring goes beyond AF, hinting at a future where wearables play a game-changing role in arrhythmia detection.