Participants
We conducted a prospective multicenter study that consecutively recruited patients suspected of CV in four referral-based university hospitals between November 2019 and April 2022. The inclusion criteria for this study were 1) recurrent attacks of spontaneous spinning or non-spinning vertigo that lasted less than 5 min, 2) symptoms onset over the age of 60 or at any age with known heart diseases, and 3) not better explained by another diagnosis. We excluded patients who were unable to use a smartphone, had accompanying auditory or neurological symptoms/signs, or cognitive dysfunction, and declined to participate in the study.
Procedures
Patients received simultaneous ambulatory ECG recordings using a 72-hour wearable ECG patch monitoring (Wellysis S-Patch Cardio) and a conventional 24-hour Holter monitoring. Heart rhythm analysis was performed by the cardiologists at each center, who were blinded to clinical information. Wellysis S-Patch Cardio is a lightweight (11 g) device containing a bioprocessor that continuously collects ECG data and transmits it in real time via a bluetooth connection to an Android smartphone application. The actual device is easily applied with only two standard ECG electrodes typically placed in the V2-V4 position. Patients kept the phone within 3 m of them throughout the test period. Arrhythmic events as causes of CV were defined as supraventricular tachycardia (> 4 beats), atrial fibrillation (> 4 beats), sinus pause > 3 s, atrioventricular block (Mobitz type II or third-degree atrioventricular block), ventricular tachycardia (> 4 beats), or polymorphic ventricular tachycardia/ventricular fibrillation. The primary end point was overall detection rate of arrhythmia events during the entire wearing period. The secondary end points included device preference and factors affecting CV detection.
Statistical Analysis
Based on the data from previously published studies1,8,9, we estimated that the diagnostic efficacy would be 10% in Holter monitoring group, and approximately 40% in patch monitoring group. By adopting 0.8 power to detect a significant difference and a drop rate of 5%, 50 patients were required.
All analyses were performed using SPSS (version 27.0, SPSS, Chicago, IL, USA). Differences in the detection rates of arrhythmic events between Holter and wearable patch monitoring were evaluated by the McNemar test. Levene's homogeneity of variance and two-sample independent t-test were performed for numeric variables. Between groups with and without documented arrhythmic events, continuous variables were compared using the t test or Mann-Whitney U tests, and nominal variables were compared using the χ2 or Fisher exact tests. A p-value less than 0.05 was considered statistically significant.
Standard protocol approvals, registrations, and patient consents
All experiments followed the tenets of the Declaration of Helsinki and were approved by the institutional review boards of the Pusan National University Hospital (IRB NO. 2106-030-104), Pusan National University Yangsan Hospital (IRB NO. 05-2021-100), Ulsan University Hospital (IRB NO. 2021-05-024), and Keimyung University Dongsan Hospital (IRB NO. 2021-05-035). Written informed consents were obtained after the nature and possible consequences of this study had been explained to the participants. This study complies with Standards for Reporting Diagnostic accuracy studies guidelines and are registered through International Clinical Trials Registry Platform (https://www.who.int/ictrp; Unique identifiers: KCT0009088).