Abstract
BACKGROUND: Smartwatches, such as the Apple Watch (AW), are well-established tools for detecting atrial fibrillation (AF). We hypothesize that atrial flutter (AFL) is frequently misdiagnosed using traditional single-lead electrocardiogram configurations and that modified device positioning could substantially improve diagnostic accuracy. METHODS: Standard smartwatch lead-I (AW-I) recordings were obtained from 75 patients, including 25 with AFL, 25 with AF, and 25 with sinus rhythm. Additionally, modified lead-II (AW-II) recordings were collected for all AFL cases, resulting in a total of 100 tracings. Twenty blinded physicians from 4 different specialties independently analyzed all recordings. RESULTS: Physicians' ability to detect AFL using the AW-I lead was poor, with only 11.6% of cases correctly identified (P = 0.362). AFL was most often misdiagnosed as AF (55.6%), undetermined (21%), or sinus rhythm (11.8%). Diagnostic accuracy improved significantly with the AW-II lead, exceeding 80% among electrophysiologists, cardiologists, and primary care physicians (P = 0.001). Variable atrioventricular conduction was associated strongly with correct diagnosis exclusively through the AW-II lead (odds ratio 1.85, 95% confidence interval 1.14-3.0, P = 0.012). CONCLUSIONS: The standard lead-I configuration used by smartwatches is prone to misclassifying AF as AFL, particularly in the setting of variable conduction. A simple modification to lead-II positioning significantly enhances diagnostic accuracy. This adjustment may be especially valuable during post-pulmonary vein isolation surveillance and in broader clinical scenarios in which precise rhythm identification can influence therapeutic decision-making.