Abstract
Background
Patch-electrocardiography (ECG) enables prolonged ECG monitoring beyond 24 hours. However, diagnostic yield between patch-ECG and Holter monitoring needs further validation. We aimed to compare diagnostic capabilities of 14-day patch-ECG and one day Holter monitoring to detect cardiac arrhythmias.
Methods
Patients with suspected cardiac arrhythmias but could not be diagnosed by 12-lead ECG were enrolled from two tertiary centers. Patch-ECG and Holter monitoring was attached simultaneously in enrolled patients. Primary endpoint was detection of major arrhythmias which was defined as atrial fibrillation (AF), atrial tachycardia, atrial flutter, paroxysmal supraventricular tachycardia (PSVT), ventricular tachycardia, 2nd or 3rd degree atrioventricular block, sick pause (> 2 seconds of pause), sick sinus syndrome, tachycardia-bradycardia syndrome, and Wolff-Parkinson-White syndrome.
Results
A total of 147 patients were analyzed. Major arrhythmias were detected in 75.5% and 48.3% in patch-ECG and Holter monitoring, respectively (P < 0.001). Detection rate between the first day of patch-ECG and Holter monitoring was identical. Detection rate for AF was significantly higher in patch-ECG (23.8% vs. 11.6%; P < 0.001). Substantial proportion of AF events were detected in the first day of monitoring (42.9%) but diagnosis rate increased steadily between day 2–14 of monitoring. Detection rate of supraventricular tachycardia (atrial tachycardia or PSVT), ventricular tachycardia, and brady-arrhythmias was higher in the patch-ECG. Four patients had to detach their patch-ECG due to skin side effects.
Conclusion
Patch-ECG has higher diagnostic capabilities compared to Holter monitoring for diagnosis of various cardiac arrhythmias.
Keywords: Arrhythmia, Wearable Device, Electrocardiography, Patch-ECG, Holter Monitor
Graphical Abstract
INTRODUCTION
Diagnosis of various types of arrhythmias is often limited by paroxysmal nature. Diagnostic yield of traditional 12-lead surface electrocardiography (ECG) and Holter monitoring is not satisfactory especially for paroxysmal arrhythmias such as paroxysmal atrial fibrillation (AF), paroxysmal supraventricular tachycardia (PSVT), sick sinus syndrome (SSS), and ventricular tachycardia.1 The confirmative diagnosis of most arrhythmia requires ECG documentation and paroxysmal nature of arrhythmia is the main obstacle for appropriate and timely diagnosis. Some arrhythmias, AF or SSS for example, might cause complications which may not be fully reversible such as ischemic stroke or head trauma with intracranial bleeding.2,3,4,5 Various wearable ECG devices such as watch-type ECG or patch-type ECG have been introduced to overcome the difficulties in diagnosis of paroxysmal arrhythmias.6,7,8 In the mSToPS trial, immediate monitoring with wearable patch-type ECG showed higher rate of AF diagnosis at four months of follow-up as compared with delayed monitoring group and steep rise of AF cumulative incidence was observed after the initiation of the patch ECG monitoring.9 Watch-type ECG also demonstrated 2.3-fold higher rate of arrhythmia diagnosis rate compared with traditional Holter monitoring.1
The utilization of patch-type ECG devices is rapidly increasing in the clinics. However, its diagnostic value as compared with Holter monitoring is not fully validated. We evaluated a patch-type wearable ECG device called Memopatch (patch-ECG) to efficiently diagnose paroxysmal arrhythmias (HUINNO Co., Ltd., Seoul, Korea). The recorded ECG signals by patch-ECG were initially analyzed by an artificial intelligence-based algorithm and only ECGs assumed to represent cardiac arrhythmia were presented to physicians who finally confirmed the diagnosis. The diagnostic yield of the patch-ECG, as compared with Holter monitoring, was evaluated in this prospective study.
METHODS
Participants
We performed this prospective study with patients visiting Korea University Medicine Anam Hospital, Seoul or Seoul National University Bundang Hospital, Seongnam, Republic of Korea due to symptoms suggestive of arrhythmias. Enrollment criteria was: 1) presence of arrhythmic symptoms such as palpitation, chest pain, syncope, stroke, or dizziness; 2) failure of 12-lead ECG to reveal any type of arrhythmia assumed to be the cause of symptoms; 3) age equal or older than 19 years and equal or younger than 80 years; and 4) no contraindications for application of patch-ECG. If definite diagnosis was made by 12-lead ECG, the patient was not enrolled. Exclusion criteria were as follows: 1) patients younger than 19 years or older than 80 years; 2) inability to record patch-ECG or Holter monitoring due to factors such as contact dermatitis; and 3) patient refusal. Prior diagnosis of arrhythmia was not an exclusion criterion if current symptoms could not be explained by it.
Primary and secondary endpoints
The aim of this study was to compare diagnostic yield of patch-ECG vs. Holter monitoring. Primary endpoint was diagnosis of arrhythmia including atrial premature contraction (APC), premature ventricular contraction (PVC), atrial tachycardia, atrial flutter, AF, PSVT, Wolff-Parkinson-White (WPW) syndrome, ventricular tachycardia, atrioventricular block, sinus node dysfunction, or SSS. Major arrhythmia was defined as AF, atrial tachycardia, atrial flutter, PSVT, ventricular tachycardia, 2nd or 3rd degree atrioventricular block, SSS (sinus pause more than 3 seconds that was accompanied by symptoms), tachycardia-bradycardia syndrome, and WPW syndrome. Diagnostic yield of major arrhythmia was compared between the patch-ECG and Holter monitoring.
Secondary outcome endpoint includes signal quality of patch-ECG and compliance of participants. Safety outcome endpoint was adverse effects related with patch attachment such as dermatitis.
Patch-ECG monitoring
The Memopatch (patch-ECG) was developed by HUINNO Co., Ltd., a medical device company in the Seoul, Republic of Korea. The sample image of the Memopatch used in this study is presented in Fig. 1. The Ministry of Food and Drug Safety of Korea approved the Memopatch as a medical device to record ECGs. The patch-ECG has 2 electrodes to record single-strip ECG. We recorded ECG for 2 weeks. Holter monitoring was performed simultaneously with patch-ECG. Since the Memopatch is consisted of 2 leads connected with wire, its size is small enough to be co-attached with Holter monitoring.8
Fig. 1. Patch ECG used in this study. Real model image and actual recorded ECG are presented. During AF status, fibrillatory waves were clearly identified. P-wave was well recorded during sinus rhythm.
ECG = electrocardiography, AF = atrial fibrillation.
The recorded ECGs by patch-ECG were initially screened by an artificial intelligence-based reading program developed by HUINNO Co., Ltd.1 The reported accuracy of the program is reported to be 97.1%.10 The initial diagnosis of an ECG recorded by either patch-ECG or Holter monitoring was made by physicians who prescribed the test. However, any abnormal ECGs recorded by either patch-ECG or Holter monitoring was confirmed by Choi IJ and Oh IY. If there was disagreement between the two physicians, it was settled by discussion or additional confirmation by Kim YG.
Statistical analysis
Student’s t-test was used to compare continuous variables. Categorical variables were compared with Fisher’s exact test or the χ2 test. Logistic regression analysis with a fixed effect for treatment group (patch-ECG vs. Holter monitoring) was used to calculate odds ratio. All tests were 2-tailed and P values of equal or less than 0.05 were considered statistically significant. All statistical analyses were performed using SAS version 9.4 (SAS Institute, Cary, NC, USA).
Ethics statement
The Institutional Review Board of Korea University Medicine Anam Hospital (approval number is 2021AN0247) and Seoul National University Bundang Hospital approved this study. We obtained written informed consent from all patients. The study protocol strictly adhered to the ethical guidelines of the 2013 Declaration of Helsinki and legal regulations of Republic of Korea.
RESULTS
Patients
From September 2021 to July 2022, a total of 169 patients were screened for eligibility to be enrolled in this study. After screening, 150 patients were enrolled. One patient failed to meet the eligibility criteria, and 18 patients refused to participate. Among 150 patients, 147 attached patch-ECG for more than 7 days. Two patients had technical issues (early depletion of the battery and program error) and one patient withdrew informed consent. Our main analysis was performed with this cohort. One hundred and forty-one patients completed the study protocol which was attachment of the patch-ECG for 14 days. Four patients had to detach the patch-ECG due to dermatitis, one patient arbitrary detached the patch-ECG for personal reasons, and one withdrew informed consent. The flow of the study is summarized in Fig. 2.
Fig. 2. Study flow of subjected patients.
Baseline demographics of the study participants are summarized in Table 1. Mean age of the cohort was 50.1 ± 15.0 years and 46.3% were male patients. The prevalence of hypertension, diabetes mellitus, and dyslipidemia was 29.9%, 7.5%, and 7.5%, respectively. One quarter of patients had a prior diagnosis of AF.
Table 1. Baseline demographics.
| Characteristics | Values (N = 147) | |
|---|---|---|
| Age, yr | 50.1 ± 15.0 | |
| Sex | ||
| Male | 68 (46.3) | |
| Female | 79 (53.7) | |
| Cardiovascular disorders | ||
| Atrial fibrillation | 38 (25.9) | |
| Atrial flutter (typical) | 3 (2.0) | |
| Atrial flutter (atypical) | 3 (2.0) | |
| Premature atrial contraction | 4 (2.75) | |
| Premature ventricular contraction | 8 (5.4) | |
| Paroxysmal supraventricular tachycardia | 4 (2.7) | |
| Complete atrioventricular block | 0 (0.0) | |
| Angina pectoris | 4 (2.7) | |
| Variant angina | 1 (0.7) | |
| Ventricular tachycardia | 1 (0.7) | |
| Hypertension | 44 (29.9) | |
| Diabetes mellitus | 11 (7.5) | |
| Dyslipidemia | 11 (7.5) | |
| Depression | 5 (3.4) | |
| Anxiety disorder | 1 (0.7) | |
| Panic disorder | 3 (2.0) | |
| Syncope | 8 (5.4) | |
| Ischemic stroke | 7 (4.8) | |
Values are presented as mean ± standard deviation or number (%).
Compliance of the patch-ECG
Among 147 patients who completed at least 7 days of monitoring and 95.2% of patients underwent more than 10 days of monitoring. Adverse events regarding dermatologic issues occurred in 63 patients (42.9%). Of these 63 patients with skin issues, 28 (44.4%) occurred within 5 days of patch-ECG attachment. Four patients failed to complete the study (patch-ECG attachment up to 14 days) due to the skin problems. Two, 1, and 1 patient detached their patch on the 10th, 11th, and 12th day, respectively. Chronological data of skin adverse events are depicted in Fig. 3. No adverse events other than dermatologic problems were reported throughout the study period.
Fig. 3. Skin side-effects by patch-ECG.
ECG = electrocardiography.
Detection of major arrhythmias
Detection rate of various arrhythmia is summarized in Table 2. In Holter monitoring, major arrhythmias (AF, atrial tachycardia, atrial flutter, PSVT, ventricular tachycardia, 2nd or 3rd degree atrioventricular block, sick pause of more than 2 seconds, SSS, tachycardia-bradycardia syndrome, and WPW syndrome) were detected in 71 (48.3%) patients which was identical with the first-day detection rate in patch-ECG. However, the detection rate of major arrhythmias was significantly higher in patch-ECG with full monitoring duration (111 patients; 75.5%; P < 0.001). Cumulative detection of major arrhythmia events is presented in Fig. 4A. Atrial fibrillation was detected in 35 (23.8%) and 17 (11.6%) in patch-ECG and Holter monitoring, respectively (P < 0.001). Among 35 patients diagnosed with AF by patch-ECG, 15 (42.9%) were detected in the first day by patch-ECG monitoring. The remaining 20 patients were detected during 2 to 14 days after attachment of patch-ECG. Time to first-detection of AF by patch-ECG and Holter monitoring is depicted in Fig. 4B.
Table 2. Detection of various arrhythmias by Holter monitoring and patch-ECG.
| Variables | Patch-ECG | Patch-ECGa | Holter | P valueb | P valuec | |
|---|---|---|---|---|---|---|
| All patients (N = 147) | ||||||
| All arrhythmia | 147 (100.0) | 143 (97.3) | 143 (97.3) | 0.125 | > 0.999 | |
| Major arrhythmia | 111 (75.5) | 71 (48.3) | 71 (48.3) | < 0.001 | > 0.999 | |
| Atrial fibrillation | 35 (23.8) | 18 (12.2) | 17 (11.6) | < 0.001 | 0.317 | |
| Atrial flutter | 3 (2.0) | 3 (2.0) | 4 (2.7) | 0.317 | 0.317 | |
| PSVT or atrial tachycardia (including non-sustained) | 97 (66.0) | 50 (34.0) | 47 (32.0) | < 0.001 | 0.083 | |
| Sinus pause (> 3 seconds) | 26 (17.7) | 19 (12.9) | 19 (12.9) | 0.008 | > 0.999 | |
| Sick sinus syndrome | 3 (2.0) | 1 (0.7) | 1 (0.7) | 0.157 | > 0.999 | |
| Tachycardia-bradycardia syndrome | 3 (2.0) | 2 (1.4) | 2 (1.4) | 0.317 | > 0.999 | |
| AV block (2nd or 3rd degree) | 12 (8.2) | 8 (5.4) | 8 (5.4) | 0.046 | > 0.999 | |
| Ventricular tachycardia (including non-sustained) | 9 (6.1) | 1 (0.7) | 1 (0.7) | 0.005 | > 0.999 | |
| Per-protocol analysis (n = 141) | ||||||
| All arrhythmia | 141 (100.0) | 137 (97.2) | 137 (97.2) | 0.125 | > 0.999 | |
| Major arrhythmia | 106 (75.2) | 68 (48.2) | 68 (48.2) | < 0.001 | > 0.999 | |
| Atrial fibrillation | 35 (24.8) | 18 (12.8) | 17 (12.1) | < 0.001 | 0.317 | |
| Atrial flutter | 3 (2.1) | 3 (2.1) | 4 (2.8) | 0.317 | 0.317 | |
| PSVT or atrial tachycardia (including non-sustained) | 93 (66.0) | 49 (34.8) | 46 (32.6) | < 0.001 | 0.083 | |
| Sinus pause (> 3 seconds) | 25 (17.7) | 18 (12.8) | 18 (12.8) | 0.008 | > 0.999 | |
| Sick sinus syndrome | 3 (2.1) | 1 (0.7) | 1 (0.7) | 0.157 | > 0.999 | |
| Tachycardia-bradycardia syndrome | 3 (2.1) | 2 (1.4) | 2 (1.4) | 0.317 | > 0.999 | |
| AV block (2nd or 3rd degree) | 10 (7.1) | 7 (5.0) | 7 (5.0) | 0.083 | > 0.999 | |
| Ventricular tachycardia (including non-sustained) | 8 (5.7) | 1 (0.7) | 1 (0.7) | 0.008 | > 0.999 | |
Values are presented as numbers (%).
ECG = electrocardiography, PSVT = paroxysmal supraventricular tachycardia, AV = atrioventricular block.
aSubjects: Same monitoring duration with Holter; bHolter vs. patch-ECG with full duration; cHolter vs. patch-ECG with same monitoring duration.
Fig. 4. Fourteen-day patch-ECG vs. Holter monitoring. Cumulative detection of major arrhythmia events (A) and atrial fibrillation (B) by 14-day patch ECG are depicted. Patch ECG showed comparable detection rate of AF compared with Holter monitoring in the first and second day of analysis.
The detection rate of supraventricular tachycardia (atrial tachycardia or PSVT) was significantly higher in the patch-ECG (66.0% vs. 32.0%, P < 0.001). Ventricular tachycardia was also detected more frequently in the patch-ECG (6.1% vs 0.7%, P = 0.005). Sinus node dysfunction was detected in 17.7% and 12.9% of patients in patch-ECG and Holter monitoring, respectively (P = 0.008). The 2nd or 3rd degree atrioventricular block was detected in 12 (8.2%) patients in patch-ECG and 8 (5.4%) in Holter monitoring.
DISCUSSION
The main findings of this study can be summarized as follows: 1) the patch-ECG can be superior to Holter monitoring in detecting major cardiac arrhythmias including AF; 2) detection rate was highest on the first day of patch-ECG monitoring; 3) however, we observed a steady increase in detection of major arrhythmias over the day 2 to 14 of patch-ECG monitoring; and 4) skin issues occurred in significant proportion of patients with some of them leading to early detachment of patch-ECG (Fig. 4).
The major strength of this study is simultaneous attachment of patch-ECG and Holter monitoring in same patients. We were able to directly compare detection yield of patch-ECG and Holter monitoring. The results of this study can assist clinicians to decide whether to apply patch-ECG rather than Holter monitoring despite potential skin side-effects and relatively higher medical cost.
We observed significantly higher diagnostic yield for various cardiac arrhythmias in patch-ECG as compared to traditional single day Holter monitoring. Detection capability for AF was significantly better in the patch-ECG than in the Holter monitoring (detection rate was 23.8% vs. 11.6%). Despite 42.9% of AF was detected on the first day of patch-ECG monitoring, substantial proportion of AF was diagnosed in day 2–14 of monitoring. By performing 147 patch-ECGs, we were able to detect additional 18 AF patients who were not diagnosed with AF by Holter monitoring. Since AF requires specific treatment such as anticoagulation or rhythm control therapy,11,12,13 performing patch-ECG in eligible patients can lead to significant change in patient management. In a prior study performed by Kim et al.,8 the 7-day patch-ECG did not show superiority over Holter monitoring to detect atrial fibrillation. However, our study enrolled more patients (147 vs. 58 patients) and applied prolonged patch-monitoring (14 vs. 7 days). Since 8 additional patients were diagnosed during day 8 to 14 monitoring, extended monitoring beyond 7 days can be beneficial to detect AF events. Our results are in accordance with the mSToPS trial. Steinhubl et al.9 reported that monitoring with a 14-day patch-ECG, compared with nonmonitored controls, had higher rates of AF diagnosis and greater initiation of anticoagulants at 1 year. Our study further intensified clinical usefulness of patch-ECG in AF diagnosis by directly comparing patch-ECG with Holter monitoring in same patients (paired-group analysis). In a trial comparing Holter monitoring, 12-lead ECG, trans-telephonic ECG monitoring, and implanted cardiac monitoring in post-AF catheter ablation patients, significant difference in detection rate of arrhythmia recurrence was observed with implanted cardiac monitoring having highest detection rate.14 Interestingly, Holter monitoring did not show superiority over 12-lead ECG or trans-telephonic monitoring suggesting that it might be a suboptimal diagnostic test to detect paroxysmal arrhythmias. In our study, the 14-day patch-ECG was superior to Holter monitoring to detect AF events suggesting that it can replace Holter monitoring in clinical practice.
We also demonstrated that other cardiac arrhythmias such as supraventricular tachycardia (atrial tachycardia or PSVT) or VT were detected more frequently by the patch-ECG. Patch-ECG demonstrated its capability for detection of bradycardias in this study. Some proportions of sinus node dysfunction or atrioventricular block are paroxysmal. Therefore, prolonged monitoring of ECG can be beneficial to detect such bradycardias. Furthermore, brady-arrhythmias are often difficult to detect with patient-triggered event recorders since dizziness or syncope can inhibit self-recording of their ECG.
The patch-ECG was able to diagnose additional 18 AF patients among 147 patients who underwent both patch-ECG and Holter monitoring. The prices of 14-day patch ECG and Holter monitoring are 221,000 and 61,292 Korean won (KRW) in December 2024, respectively. Therefore, a total of 23,477,076 ([221,000 – 61,292] × 147) KRW was required to diagnose 18 AF patients (1,304,282 KRW was spent to diagnose 1 additional AF patient). Since AF is associated with various cardiovascular adverse events including ischemic stroke and early rhythm control has demonstrated to reduce such events, 1,304,282 KRW to diagnose 1 additional AF patient seems highly cost-effective.11,15 Management of AF can also improve patient symptoms and quality of life.16 Furthermore, the cost-effectiveness of early rhythm control therapy in AF patients is also validated in recent study.17 In addition to AF, 2 sick sinus syndrome, 1 tachycardia-bradycardia syndrome, 4 2nd or 3rd degree atrioventricular block, and 8 ventricular tachycardia (including non-sustained) events were additionally diagnosed by patch-ECG. These diseases can lead to critical complications such as syncope, head trauma, or sudden cardiac death reinforcing cost-effectiveness of patch-ECG monitoring.
Patch-ECG is basically a non-invasive diagnostic exam. Although we observed 63 patients (42.9%) with skin side effects, all were temporary. We confirmed patient safety profile of patch-ECG, but skin side effects led to premature detachment in 4 patients. Most of the skin side effects occurred within 5 days after the patch attachment. Data acquired before the occurrence of skin side effects were analyzable and no patient required specific treatment for dermatitis. Furthermore, no other side effects that required any specific medical management were observed. Based on our observations, patch-ECG is a safe diagnostic tool despite potential skin irritation.
This study has several limitations. First, our cohort is exclusively consisted with East Asian patients. Therefore, extrapolation to other ethnic groups can be limited. Second, duration of patch-ECG attachment might not have been sufficient. Some arrhythmias occur very rarely, irrespective of its severity, and left undetected by patch-ECG used in this study. Comparison between patch-ECG Vs. implantable loop recorders can help to address such issues. Third, technical problems led to data loss in two patients. One patient had a problem with early battery depletion and the other had data loss due to program error. Fourth, the higher diagnostic yield of cardiac arrhythmia detection in patch-ECG, compared to 24-hour Holter monitoring, can be primarily due to the extended monitoring period of 14 days rather than better ECG quality as demonstrated in Fig. 4 (similar diagnostic yield in the first and second day). Fifth, since this study was performed in 2 tertiary referral hospitals in South Korea, the enrolled patients are highly likely to have cardiac arrhythmias and the difference in diagnostic yield between patch-ECG and Holter monitoring might have been overestimated. Sixth, we focused on diagnostic yield of patch-ECG Vs. Holter monitoring and not physician versus artificial intelligence-based reading program and therefore, cannot confirm the diagnostic accuracy of machine-based ECG interpretation.
Patch-ECG demonstrated superior diagnostic capabilities compared to Holter monitoring for diagnosis of cardiac arrhythmias. Atrial fibrillation was more frequently diagnosed with patch-ECG and diagnosis rate increased steadily between day 2 to 14 indicating clinical benefit of prolonged ECG monitoring beyond 24 hours. Skin side effects occurred in substantial proportion of patients but did not require any specific treatment other than patch detachment.
ACKNOWLEDGMENTS
We thank Junsang Park, Sunghoon Jung, Daein Kim, Jinkook Kim, and Junho An for their support in the study process and technical assistance.
Footnotes
Funding: This work was supported by the HUINNO Co., Ltd., Seoul, Republic of Korea. This work was supported by a National IT Industry Promotion Agency (NIPA) grant funded by the Korea government (MSIT) (Special Regulatory Treatment for Demonstration 2019-03 [ICT regulatory sandbox; I1902421] to Ho Sung Son) and in part by the Korea Medical Device Development Fund grant funded by the Korea government (the Ministry of Science and ICT, the Ministry of Trade, Industry and Energy, the Ministry of Health & Welfare, the Ministry of Food and Drug Safety [Project Number: RS-2021-D00001] to Jong-Il Choi), and Korea University research fund (K2407611 to Yun Gi Kim). The funders had no role in data collection, analysis, interpretation, trial design, patient recruitment, or any other aspects of this study.
Disclosure: The authors have no potential conflicts of interest to disclose.
- Conceptualization: Kim YG, Choi JI, Oh IY, Cho Y, Son HS, Kim YH.
- Data curation: Lee JH, Shim J, Son HS.
- Formal analysis: Kim YG, Choi YY, Lee JH, Shim J.
- Funding acquisition: Oh IY, Son HS.
- Investigation: Kim YG, Choi JI, Oh IY, Jeong JH, Lee HS, Choi YY, Lee JH, Cho Y, Shim J, Son HS.
- Methodology: Kim YG, Choi JI, Oh IY, Jeong JH, Lee HS, Choi YY, Lee JH, Cho Y, Shim J.
- Project administration: Cho Y, Son HS.
- Supervision: Kim YG, Choi JI, Oh IY, Son HS, Kim YH.
- Validation: Kim YG, Oh IY, Jeong JH, Lee HS, Choi YY, Shim J, Kim YH.
- Visualization: Kim YG, Jeong JH, Lee HS, Choi YY, Shim J.
- Writing - original draft: Kim YG, Choi JI.
- Writing - review & editing: Kim YG, Choi JI, Oh IY, Kim YH.
References
- 1.Kim YG, Choi JI, Kim HJ, Min K, Choi YY, Shim J, et al. A watch-type electrocardiography is a reliable tool for detecting paroxysmal cardiac arrhythmias. J Clin Med. 2022;11(12):3333. doi: 10.3390/jcm11123333. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Freedman B, Potpara TS, Lip GY. Stroke prevention in atrial fibrillation. Lancet. 2016;388(10046):806–817. doi: 10.1016/S0140-6736(16)31257-0. [DOI] [PubMed] [Google Scholar]
- 3.Anter E, Jessup M, Callans DJ. Atrial fibrillation and heart failure: treatment considerations for a dual epidemic. Circulation. 2009;119(18):2516–2525. doi: 10.1161/CIRCULATIONAHA.108.821306. [DOI] [PubMed] [Google Scholar]
- 4.Kamel H, Okin PM, Elkind MS, Iadecola C. Atrial fibrillation and mechanisms of stroke: time for a new model. Stroke. 2016;47(3):895–900. doi: 10.1161/STROKEAHA.115.012004. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Kim YG, Lee HS, Kim H, Kim M, Jeong JH, Choi YY, et al. Association of antiarrhythmic drug therapy with syncope and pacemaker implantation in patients with atrial fibrillation. J Am Coll Cardiol. 2024;83(11):1027–1038. doi: 10.1016/j.jacc.2024.01.013. [DOI] [PubMed] [Google Scholar]
- 6.Fung E, Järvelin MR, Doshi RN, Shinbane JS, Carlson SK, Grazette LP, et al. Electrocardiographic patch devices and contemporary wireless cardiac monitoring. Front Physiol. 2015;6:149. doi: 10.3389/fphys.2015.00149. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Marques Antunes M, Silva Cunha P, Lacerda Teixeira B, Alves S, Portugal G, Valente B, et al. Very-early detection of atrial fibrillation in patients after ablation evaluated by a home-based wearable ECG-patch. Europace. 2023;25(Supplement_1):euad122.634. doi: 10.1016/j.ijcha.2024.101369. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Kim JY, Oh IY, Lee H, Lee JH, Cho Y, Gil Y, et al. The efficacy of detecting arrhythmia is higher with 7-day continuous electrocardiographic patch monitoring than with 24-h Holter monitoring. J Arrhythm. 2023;39(3):422–429. doi: 10.1002/joa3.12865. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Steinhubl SR, Waalen J, Edwards AM, Ariniello LM, Mehta RR, Ebner GS, et al. Effect of a home-based wearable continuous ECG monitoring patch on detection of undiagnosed atrial fibrillation: the mSToPS randomized clinical trial. JAMA. 2018;320(2):146–155. doi: 10.1001/jama.2018.8102. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Park J, Kim J, Jung S, Gil Y, Choi JI, Son HS. ECG-signal multi-classification model based on squeeze-and-excitation residual neural networks. Appl Sci (Basel) 2020;10(18):6495. [Google Scholar]
- 11.Kirchhof P, Camm AJ, Goette A, Brandes A, Eckardt L, Elvan A, et al. Early rhythm-control therapy in patients with atrial fibrillation. N Engl J Med. 2020;383(14):1305–1316. doi: 10.1056/NEJMoa2019422. [DOI] [PubMed] [Google Scholar]
- 12.Patel MR, Mahaffey KW, Garg J, Pan G, Singer DE, Hacke W, et al. Rivaroxaban versus warfarin in nonvalvular atrial fibrillation. N Engl J Med. 2011;365(10):883–891. doi: 10.1056/NEJMoa1009638. [DOI] [PubMed] [Google Scholar]
- 13.Giugliano RP, Ruff CT, Braunwald E, Murphy SA, Wiviott SD, Halperin JL, et al. Edoxaban versus warfarin in patients with atrial fibrillation. N Engl J Med. 2013;369(22):2093–2104. doi: 10.1056/NEJMoa1310907. [DOI] [PubMed] [Google Scholar]
- 14.Balabanski T, Brugada J, Arbelo E, Laroche C, Maggioni A, Blomström-Lundqvist C, et al. Impact of monitoring on detection of arrhythmia recurrences in the ESC-EHRA EORP atrial fibrillation ablation long-term registry. Europace. 2019;21(12):1802–1808. doi: 10.1093/europace/euz216. [DOI] [PubMed] [Google Scholar]
- 15.ACTIVE Writing Group of the ACTIVE Investigators. Connolly S, Pogue J, Hart R, Pfeffer M, Hohnloser S, et al. Clopidogrel plus aspirin versus oral anticoagulation for atrial fibrillation in the Atrial fibrillation Clopidogrel Trial with Irbesartan for prevention of Vascular Events (ACTIVE W): a randomised controlled trial. Lancet. 2006;367(9526):1903–1912. doi: 10.1016/S0140-6736(06)68845-4. [DOI] [PubMed] [Google Scholar]
- 16.Kim YG, Shim J, Choi JI, Kim YH. Radiofrequency catheter ablation improves the quality of life measured with a short form-36 questionnaire in atrial fibrillation patients: a systematic review and meta-analysis. PLoS One. 2016;11(9):e0163755. doi: 10.1371/journal.pone.0163755. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Gottschalk S, Kany S, König HH, Crijns HJ, Vardas P, Camm AJ, et al. Cost-effectiveness of early rhythm control vs. usual care in atrial fibrillation care: an analysis based on data from the EAST-AFNET 4 trial. Europace. 2023;25(5):euad051. doi: 10.1093/europace/euad051. [DOI] [PMC free article] [PubMed] [Google Scholar]