Detection and Predictors of Arrhythmia in Patients with Chronic Noncardioembolic Ischemic Stroke on Wearable Electrocardiogram Device

Keywords

wearable device - noncardioembolic ischemic stroke - arrhythmia - atrial fibrillation - frequent premature atrial contractions - frequent premature ventricular contractions

Introduction

The causes of ischemic stroke are classified following the “Trial of Org 10172 in Acute Stroke Treatment” as large artery atherosclerosis, cardioembolism, small vessel occlusion, stroke of other determined etiology, and stroke of undetermined etiology.[1] As treatment differs depending on the stroke subtype, identifying the cause of ischemic stroke is important. Antiplatelet therapy is used for large artery atherosclerosis and small vessel occlusion, whereas anticoagulant therapy is used for cardioembolisms. Nonvalvular atrial fibrillation (AF) is the leading cause of cardioembolism, and the indication for anticoagulants is based on stroke risk score in patients with nonvalvular AF.[2] When a patient with ischemic stroke due to large artery atherosclerosis or small vessel occlusion has AF, secondary stroke prevention using antiplatelet therapy plus anticoagulation is needed. However, the risk of major bleeding complications increases when patients receive antiplatelet agents and anticoagulants.[3] Consensus regarding antithrombotic therapy for patients with noncardioembolic ischemic stroke complicated with AF is lacking.

The prevalence of AF among patients with acute ischemic stroke is reported to be 4.0% for small vessel occlusion and 7.3% for large artery atherosclerosis.[4] [5] The risk of arrhythmia may increase under mental stress, such as in acute ischemic stroke.[6] Therefore, whether anticoagulation is based on arrhythmia detected in the acute phase is questionable. The prevalence of AF in patients with a history of large artery atherosclerosis and small vessel occlusion at 12 months has been reported to be 11.7 and 12.6%, respectively, in studies using implantable cardiac recording monitors (ICMs).[7] The prevalence of AF in patients with chronic noncardioembolic ischemic stroke is higher than that in the general population (1.6–2.0%).[8] [9] The detection of chronic phase arrhythmia in patients with noncardioembolic ischemic stroke is important for lifelong secondary stroke prevention. Furthermore, premature atrial contractions (PACs) and premature ventricular contractions (PVCs) other than AF have been associated with ischemic stroke.[10] [11] [12] [13] Holter electrocardiogram (ECG) and ICM can detect asymptomatic arrhythmia. However, Holter ECG has a short examination time, and ICM as a screening test is unsuitable for clinical practice.

In recent years, wearable devices that can noninvasively monitor patient data have been developed.[14] [15] In the present study, we aimed to assess the efficacy of a stick-on wearable device in detecting AF, PAC, and PVC in patients with chronic noncardioembolic ischemic stroke and investigate the factors associated with arrhythmias.

Materials and Methods

Study Design

This prospective observational study involving 176 patients was approved by the Ethics Committee of our hospital (R01–175), and the registration period spanned from July 2020 to February 2022. The study took place in five hospitals in Japan.

Participants

The participants were patients with a history of noncardioembolic ischemic stroke (including large artery atherosclerosis and small vessel occlusion) who received antiplatelet therapy and could attend the outpatient clinic. Other types referred to as noncardioembolic ischemic stroke other than large artery atherosclerosis and small vessel occlusion include branch atheromatous disease, vascular dissection, and aortogenic ischemic stroke. Patients receiving anticoagulant therapy were excluded. The diagnosis of noncardioembolic ischemic stroke was defined as the absence of cardiac disease or an arrhythmia that could cause cardioembolic ischemic stroke and imaging studies consistent with noncardioembolic ischemic stroke. Written informed consent was obtained from all the participants.

Procedure

Equipment and Data Analysis

ECG monitoring was performed using a wearable device and a stick-on ECG sensor (Heartnote; JSR Corporation, Tokyo, Japan). This device was certified as a medical device by the Japanese Ministry of Health, Labor, and Welfare (medical device approval number 302ACBZX00015000) and has been used in previous clinical studies at other institutions.[14] [15] The device measures 30 mm × 100 mm × 5 mm and weighs 12 g. After registration in this study, the device was attached to the upper chest wall with adhesive tape, and ECGs were continuously recorded for a maximum of approximately 7 days. After recording, the device was returned to JSR Corporation for data analysis. All QRS complex, nose, and arrhythmia types were annotated on a long-term Holter ECG analysis viewer (NEY-HEA3000, Nexis Corporation, Fukuoka, Japan) using the Holter analysis program (JMDN36827012, Nexis Corporation), which has been approved by the Japanese Ministry of Health, Labor, and Welfare (medical device approval number 228AGBZX00099000). QRS complexes were classified according to the standard cycle length criteria for supraventricular ectopic heartbeats, grouped by QRS morphology, and labeled according to the arrhythmia type. Skilled technicians reviewed and edited these results, and skilled doctors confirmed the morphological classification.[14] [15]

Data Selection

The following data were collected: total heart rate, AF, PAC, and PVC. Frequent PAC was defined as PAC ≥ 100/d or ≥ 3 consecutive PACs ≥ 1/d, and frequent PVC was defined as PVCs ≥ 30/h or ≥ 2 consecutive PVCs ≥ 1/d, based on Lown's classification.[16] Age, sex, time from onset to the examination, stroke type (large artery atherosclerosis, small vessel occlusion, and others), height, weight, medications, and medical history of the patients were extracted from medical records. As the onset time was unknown in 11 of the 176 patients, the time from onset to examination was investigated in 165 patients.

In this study, we investigated the prevalence of arrhythmia and its associated risk factors. Additionally, we investigated the usefulness of a wearable device for detecting arrhythmia in patients with chronic noncardioembolic ischemic stroke.

Statistical Analysis

Continuous variables are presented as means ± standard deviation, and discrete data are presented as counts and percentages. The independence of continuous variables was evaluated using the t-test, and discrete data were analyzed using the chi-square test. Stepwise multivariate logistic regression analysis was used to determine the relationship between the predicted factors and frequent PACs or PVCs. SPSS software (version 27.0; IBM, Armonk, New York, United States) was used for all the analyses. Statistical significance was set at p < 0.05.

Results

This study included 176 patients––125 men and 51 women––with a mean age of 70.0 ± 9.8 years. The mean time from onset to the examination was 54.7 ± 59.1 months. Small vessel occlusion was the most common type of ischemic stroke, occurring in 92 (52.3%) patients, followed by large artery atherosclerosis in 64 (36.4%) patients. The most common antihypertensive drugs were calcium channel blockers (108 patients, 61.4%), angiotensin receptor blockers (74 patients, 42.0%), and diuretics (18 patients, 6.8%). The most common antiplatelet therapies were clopidogrel (65 patients, 36.9%), aspirin (62 patients, 35.2%), and cilostazol (62 patients, 35.2%). [Table 1] presents the characteristics of the 176 patients. The mean measurement time using the wearable device was 121.3 ± 45.3 hours. In 31 (17.6%) patients, measurements were recorded for < 3 days, whereas in 98 (55.7%) patients, the measurements were recorded for > 5 days. The mean total heart rate was 491,985.5 ± 227,918.7 beats. Arrhythmia occurred in 2 (1.1%) patients with AF, 69 (39.2%) patients with frequent PAC, and 36 (20.5%) patients with frequent PVCs. The rate of frequent PACs did not differ between 15 (48.4%) patients with measurements for < 3 days and 54 (37.2%) patients with measurements for > 3 days (p = 0.25), while the rate of frequent PVCs did not differ between 4 (12.9%) patients and 32 (22.1%) patients, respectively (p = 0.25).

Table 1 Characteristics of patients

	Patients
Number	176
Age, y	70.0 ± 9.8
Male, n (%)	125 (71.0)
Body mass index, kg/m2	24.0 ± 3.3
Time from onset to the examination, mo	52.2 ± 57.7
Present illness, n (%)
Hypertension	128 (72.7)
Diabetes mellitus	50 (28.4)
Hyperlipidemia	96 (54.5)
Type of cerebral infarction, n (%)
Small vessel occlusion	92 (52.3)
Large artery atherosclerosis	64 (36.4)
Other	21 (11.9)
Medication, n (%)
Antihypertensive drug
Calcium channel blocker	108 (61.4)
Angiotensin-converting enzyme inhibitor	4 (2.3)
Angiotensin receptor blocker	74 (42.0)
Diuretic	18 (10.2)
Beta-blocker	12 (6.8)
Other	4 (2.3)
Thyroid hormone therapy	2 (1.1)
Antithyroid therapy	0
Antiplatelet
Aspirin	62 (35.2)
Clopidogrel	65 (36.9)
Cilostazol	62 (35.2)
Measurement time using the wearable device, h	121.3 ± 45.3
Total heart rate, beat	491,985.5 ± 227,918.7
Atrial fibrillation, n (%)	2 (1.1)
Frequent premature atrial contractions, n (%)	69 (39.2)
≥ 100/d	49 (27.8)
3 consecutive contractions, ≥ 1/d	61 (34.7)
Frequent premature ventricular contractions, n (%)	36 (20.5)
≥ 30/h	14 (8.0)
2 consecutive contractions, ≥ 1/d	33 (18.8)

The device wearing time in the two patients with AF was 166.7 and 103.1 hours, and the total arrhythmia durations were 24 seconds and 2 hours 36 minutes 45 seconds, respectively. Patients with frequent PAC were significantly older and more likely to be female than those without frequent PAC (74.4 ± 7.44 years vs. 67.1 ± 10.1, p < 0.01, and 39.1% vs. 22.5%, p = 0.03). Furthermore, patients with frequent PACs were less likely to use clopidogrel and more likely to use cilostazol than those without frequent PACs (23.1% vs. 44.9%, p < 0.01; 47.8% vs. 27.1%, p < 0.01). Patients with frequent PVCs were more likely to be men (86.1% vs. 53.4%, p = 0.04). Furthermore, patients with frequent PVCs received more cilostazol than those without frequent PVCs (55.6% vs. 30.0%, p < 0.01) ([Table 2]).

Table 2 Characteristics of patients with frequent premature atrial contractions and premature ventricular contractions

	Frequent premature atrial contractions		p-Value	Frequent premature ventricular contractions		p-Value
	(+)	(–)		(+)	(–)
Number	69 (39.2)	107 (60.8)		36 (20.5)	140 (79.5)
Age, y	74.4 ± 7.44	67.1 ± 10.1	< 0.01	72.4 ± 10.9	69.4 ± 9.4	0.10
Male, n (%)	42 (60.9)	83 (77.5)	0.03	31 (86.1)	94 (53.4)	0.04
Body mass index, kg/m2	23.4 ± 3.1	24.3 ± 3.4	0.06	23.9 ± 2.8	24.0 ± 3.4	0.93
Time from onset to the examination, months	66.0 ± 65.9	50.7 ± 53.8	0.04	67.5 ± 68.9	53.8 ± 56.0	0.15
Present illness, n (%)
Hypertension	51 (73.9)	76 (71.0)	0.65	29 (80.6)	99 (70.7)	0.33
Diabetes mellitus	14 (20.2)	35 (32.7)	0.16	9 (25.0)	41 (29.3)	0.76
Hyperlipidemia	35 (50.7)	61 (57.0)	0.51	21 (58.3)	75 (53.6)	0.75
Type of cerebral infarction, n (%)
Small vessel occlusion	40 (58.0)	52 (48.6)	0.29	21 (58.3)	71 (50.7)	0.53
Large artery atherosclerosis	21 (30.4)	42 (39.3)	0.41	14 (38.9)	50 (35.7)	0.85
Other	8 (11.6)	13 (12.1)	1.00	2 (5.6)	19 (13.6)	0.03
Medication, n (%)
Antihypertensive drug
Calcium channel blocker	48 (69.6)	60 (56.1)	0.07	26 (72.2)	82 (58.6)	0.13
Angiotensin-converting enzyme inhibitor	2 (2.9)	2 (1.9)	0.66	1 (2.8)	3 (2.1)	0.82
Angiotensin receptor blocker	31 (44.9)	43 (40.2)	0.53	17 (47.2)	57 (40.7)	0.48
Diuretic	8 (11.6)	10 (9.3)	0.63	5 (13.9)	13 (9.3)	0.42
Beta-blocker	4 (5.8)	8 (7.5)	0.67	2 (5.6)	10 (7.1)	0.74
Other	2 (2.9)	2 (1.9)	0.66	0	4 (2.9)	0.31
Thyroid hormone therapy	1 (1.4)	1 (0.9)	0.75	1 (2.8)	1 (0.7)	0.30
Antithyroid therapy	0	0		0	0
Antiplatelet
Aspirin	25 (36.2)	37 (34.6)	0.97	8 (22.2)	54 (38.6)	0.09
Clopidogrel	16 (23.1)	48 (44.9)	< 0.01	11 (30.6)	54 (38.6)	0.43
Cilostazol	33 (47.8)	29 (27.1)	< 0.01	20 (55.6)	42 (30.0)	< 0.01

Multivariate logistic regression analyses performed for PAC and PCV revealed that the risk factors independently associated with frequent PACs were age (odds ratio [OR] 1.103, 95% confidence interval [CI] 1.055–1.153]; p < 0.001) and cilostazol use (OR 2.681, 95% CI 1.338–5.371; p = 0.005). The risk factors independently associated with frequent PVCs were age (OR 1.047, 95% CI 1.002–1.095; p = 0.043), male (OR 3.834, 95% CI 1.441–11.045; p = 0.013), and cilostazol use (OR 2.968, 95% CI 1.363–6.463; p = 0.006) ([Table 3]).

Table 3 Risk factors independently associated with frequent premature atrial contractions and frequent premature ventricular contractions

	Odds ratio	95% confidence interval	p-Value
Frequent premature atrial contractions
Age	1.103	1.055–1.153	< 0.001
Cilostazol	2.681	1.338–5.371	0.005
Frequent premature ventricular contractions
Age	1.047	1.002–1.095	0.043
Male	3.834	1.331–11.045	0.013
Cilostazol	2.968	1.363–6.463	0.006

Discussion

In this study, the prevalence of arrhythmias in patients with chronic noncardiogenic ischemic stroke was as follows: AF (1.1%), frequent PAC (39.2%), and frequent PVC (20.5%). ECG examination using wearable devices was possible for a mean of 5 days, and 56% of the patients were able to undergo examinations for more than 5 days.

The prevalence of AF in patients with noncardioembolic ischemic stroke is higher than that in the general population. In the general population, the prevalence of AF is 2.4% in men and 1.2 to 1.6% in women aged over 40 years.[8] [9] Moreover, AF is common in the elderly and those with clinical risk factors.[8] [9] A meta-analysis of studies on AF using Holter ECG reported prevalence rates of 2.4 and 2.2% in patients with small vessel occlusion and large artery atherosclerosis, respectively.[17] A study on AF using ICM reported prevalence rates of 12.6 and 11.7%, respectively.[7] Furthermore, elderly patients have a higher rate of cardioembolism and poorer functional prognosis than nonelderly patients.[18] Antithrombotic therapy is recommended for patients with noncardioembolic ischemic stroke complicated with AF. Therefore, evaluating AF in the chronic phase in patients with noncardioembolic ischemic stroke to select appropriate treatments for recurrence is important.

We found that patients with chronic noncardioembolic ischemic stroke had a 1.1% prevalence of AF and a frequency similar to that of the general population. Several factors may have influenced this result. AF may have been excluded during the examination at the initial ischemic stroke and detected between the time of onset and study period. This device can record measurements for 7 days, and measurements were only available for less than 3 days for 31 (17.5%) patients. Therefore, the measurement time may have been insufficient. Additionally, Asians have a lower prevalence of AF than other ethnic groups, and AF may be less common in patients with noncardioembolic ischemic stroke.[19]

In this study, the rate of frequent PAC was 39.2%, with PAC ≥ 100/d account for 27.8% and the rate of three consecutive PACs ≥ 1/d was 34.7%. The definition of PAC varies among reports; however, the prevalence of PAC ≥ 100/d in the general population has been reported to be 25.0%, and our result (27.8%) was comparable to that of a previous report.[20] Frequent PACs ≥ 100/d were associated with increased risks of AF, stroke, and death.[10] [11] [20] [21] Frequent PACs could accelerate the process of atrial remodeling and contribute to atrial dysfunction, progressing to atrial cardiomyopathy.[10] [22] In this study, the rate of frequent PVCs was 20.5%, with PVC ≥ 30/h was 8.0% and the rate of two consecutive PVC ≥ 1/d was 18.8%. The prevalence of PVC ≥ 30/h in this study was higher than that in the general population (2.1%).[13] Frequent PVC also increases the risk of heart failure, AF, cardiovascular events, and ischemic stroke.[12] [23] [24] Frequent PVC may be associated with the development of ventricular cardiomyopathy.[25] This process may lead to cardiac remodeling, such as apoptosis and fibrosis of myocardial cells, resulting in permanent structural changes of the ventricle.[12] In our study, frequent PVCs were more prevalent in patients with chronic noncardioembolic ischemic stroke than in the general population. Frequent PVCs can be a risk factor for cardiocerebrovascular events; therefore, evaluating arrhythmia is important.

This study showed no difference in the detection rate of frequent PACs and PVCs based on the time of measurement. The main aim of measurements was to detect AF. Both patients with AF were measured for > 3 days. The detection rate of AF is considered to be higher with affixed ECGs than with conventional Holter ECGs, and the detection rate can be increased with patient education and strict wearing practices to encourage long-term use.[14] [26]

In this study, frequent PACs were independently associated with age and cilostazol use, whereas frequent PVCs were independently associated with age, male sex, and cilostazol use. Previous reports have shown that frequent PACs in the general population increases with age and that frequent PVCs is independently associated with age and male sex, similar to the findings of our study.[13] [27] A new finding of this study is that frequent PACs and PVCs were significantly increased in patients who received cilostazol. Cilostazol increases cyclic adenosine monophosphate (cAMP) by selectively inhibiting phosphodiesterase 3, a cAMP-degrading enzyme.[28] An increase in cAMP induces an aggregation inhibitory effect in platelets and causes ventricular arrhythmia.[29] How frequently PACs and PVCs are clinically essential in patients with chronic noncardioembolic ischemic stroke is unclear. As frequent PACs and PVCs can lead to AF and ischemic stroke, caution may be required when detecting frequent PACs and PVCs in patients receiving cilostazol.

There is no consensus regarding the management of newly detected arrhythmias in patients with chronic noncardioembolic ischemic stroke; therefore, each case must be evaluated individually. If AF is detected, a change in anticoagulation therapy or percutaneous left ventricular closure may be considered.[2] [30] If arrhythmias such as PACs or PVCs are detected, a change in antiplatelet medication should be considered in patients on cilostazol. A specialist must be consulted to determine the appropriate indications for catheter ablation in patients with asymptomatic arrhythmia.

Long-term ECG measurement methods include Holter ECG and ICMs. Conventional Holter ECGs typically records measurements for 24 to 48 hours, but it can interfere with daily life, such as restricting bathing. On the other hand, ICMs can provide long-term measurements but require subcutaneous implantation, making them suitable only for specific indications such as syncope or unexplained stroke, and therefore not ideal for screening.[31] Wearable devices, such as wearable Holter ECGs and smartwatches, offer more convenience as they do not require multiple hospital visits for attachment and removal and allow for bathing. Wearable Holter ECGs can record measurements for longer periods than conventional Holter ECGs, resulting in increased sensitivity for detecting AF.[26] They also have the potential to classify various arrhythmias using deep learning, which could lead to further advancements.[32] In contrast, smartwatches measure pulse rate using photoplethysmography but have limitations in terms of accuracy, continuous monitoring, and detecting arrhythmias other than AF.[33] The device used in this study (Heartnote, JSR Corporation) is a flexible, cordless, integrated, waterproof, and lightweight stick-on ECG sensor. It allows for longer measurement periods than conventional Holter ECGs without the need for implantation like an ICM.[26] Moreover, it can detect not only AF but also frequent PACs and PVCs. Furthermore, in one of the two patients with AF, the total arrhythmia duration was 24 seconds. This device also has the advantage of detecting short-term AF, although ICM can only detect AF lasting > 30 seconds.[34] This makes it useful for stroke risk screening in daily clinical practice. The device needs to be worn multiple times to improve its arrhythmia detection capability.

This study had some limitations. This study included patients for whom measurements could only be performed for a short period. Thirty-one (17.6%) patients were assessed for < 3 days, and 99 (56.2%) patients were assessed for > 5 days. Therefore, the device wearing time may have been insufficient to detect arrhythmia. Most of the participants were men. The risk of arrhythmia is affected by sex; therefore, caution should be exercised when interpreting these results.[13] [27] Examinations at the time of ischemic stroke were not investigated, and some patients who underwent insufficient examinations may have been included in the study. Information that could affect autonomic activity and heart rhythm, such as the location and size of ischemic lesions and caffeine use, was neither obtained in previous studies nor in this study.

Conclusion

The frequency of AF in patients with chronic noncardioembolic ischemic stroke was 1.1%, which was not greater than that reported previously in the general population. The frequency of frequent PACs in our study was comparable to that in the general population, and the frequency of frequent PVCs was higher than that in the general population. The wearable device can easily detect various arrhythmias and be used to screen patients for stroke risk in daily medical practice.

• References

• 1 Adams Jr HP, Bendixen BH, Kappelle LJ. et al. Classification of subtype of acute ischemic stroke. Definitions for use in a multicenter clinical trial. TOAST. Trial of Org 10172 in Acute Stroke Treatment. Stroke 1993; 24 (01) 35-41
• 2 Hindricks G, Potpara T, Dagres N. et al; ESC Scientific Document Group. 2020 ESC Guidelines for the diagnosis and management of atrial fibrillation developed in collaboration with the European Association for Cardio-Thoracic Surgery (EACTS): the Task Force for the diagnosis and management of atrial fibrillation of the European Society of Cardiology (ESC) Developed with the special contribution of the European Heart Rhythm Association (EHRA) of the ESC. Eur Heart J 2021; 42 (05) 373-498
• 3 Toyoda K, Yasaka M, Iwade K. et al; Bleeding with Antithrombotic Therapy (BAT) Study Group. Dual antithrombotic therapy increases severe bleeding events in patients with stroke and cardiovascular disease: a prospective, multicenter, observational study. Stroke 2008; 39 (06) 1740-1745
• 4 Bansil S, Karim H. Detection of atrial fibrillation in patients with acute stroke. J Stroke Cerebrovasc Dis 2004; 13 (01) 12-15
• 5 Kimura K, Kazui S, Minematsu K, Yamaguchi T. Japan Multicenter Stroke Investigator's Collaboration. Analysis of 16,922 patients with acute ischemic stroke and transient ischemic attack in Japan. A hospital-based prospective registration study. Cerebrovasc Dis 2004; 18 (01) 47-56
• 6 Lampert R. Mental stress and ventricular arrhythmias. Curr Cardiol Rep 2016; 18 (12) 118
• 7 Bernstein RA, Kamel H, Granger CB. et al; STROKE-AF Investigators. Effect of long-term continuous cardiac monitoring vs usual care on detection of atrial fibrillation in patients with stroke attributed to large- or small-vessel disease: the STROKE-AF randomized clinical trial. JAMA 2021; 325 (21) 2169-2177
• 8 Iguchi Y, Kimura K, Aoki J. et al. Prevalence of atrial fibrillation in community-dwelling Japanese aged 40 years or older in Japan: analysis of 41,436 non-employee residents in Kurashiki-city. Circ J 2008; 72 (06) 909-913
• 9 Davis RC, Hobbs FD, Kenkre JE. et al. Prevalence of atrial fibrillation in the general population and in high-risk groups: the ECHOES study. Europace 2012; 14 (11) 1553-1559
• 10 Huang BT, Huang FY, Peng Y. et al. Relation of premature atrial complexes with stroke and death: systematic review and meta-analysis. Clin Cardiol 2017; 40 (11) 962-969
• 11 Acharya T, Tringali S, Bhullar M. et al. Frequent atrial premature complexes and their association with risk of atrial fibrillation. Am J Cardiol 2015; 116 (12) 1852-1857
• 12 Rujirachun P, Wattanachayakul P, Phichitnitikorn P, Charoenngam N, Kewcharoen J, Winijkul A. Association of premature ventricular complexes and risk of ischemic stroke: a systematic review and meta-analysis. Clin Cardiol 2021; 44 (02) 151-159
• 13 Ofoma U, He F, Shaffer ML, Naccarelli GV, Liao D. Premature cardiac contractions and risk of incident ischemic stroke. J Am Heart Assoc 2012; 1 (05) e002519
• 14 Hayano J, Yuda E. Enhanced detection of abnormalities in heart rate variability and dynamics by 7-day continuous ECG monitoring. Ann Noninvasive Electrocardiol 2022; 27 (01) e12897
• 15 Hayano J, Yuda E. Night-to-night variability of sleep apnea detected by cyclic variation of heart rate during long-term continuous ECG monitoring. Ann Noninvasive Electrocardiol 2022; 27 (02) e12901
• 16 Lown B, Wolf M. Approaches to sudden death from coronary heart disease. Circulation 1971; 44 (01) 130-142
• 17 Demeestere J, Fieuws S, Lansberg MG, Lemmens R. Detection of atrial fibrillation among patients with stroke due to large or small vessel disease: a meta-analysis. J Am Heart Assoc 2016; 5 (09) 5
• 18 Matsuzono K, Mashiko T, Ozawa T. et al. Characteristics of aged ischemic stroke patients indicative of cardioembolism. J Thromb Thrombolysis 2021; 51 (02) 522-529
• 19 Alonso A, Roetker NS, Soliman EZ, Chen LY, Greenland P, Heckbert SR. Prediction of atrial fibrillation in a racially diverse cohort: the Multi-Ethnic Study of Atherosclerosis (MESA). J Am Heart Assoc 2016; 5 (02) 5
• 20 Chong BH, Pong V, Lam KF. et al. Frequent premature atrial complexes predict new occurrence of atrial fibrillation and adverse cardiovascular events. Europace 2012; 14 (07) 942-947
• 21 Liu HY, Wu JY, Chung CP. et al. Premature atrial contractions and their association with stroke features and outcome. J Stroke Cerebrovasc Dis 2020; 29 (10) 105118
• 22 Hasdemir C, Simsek E, Yuksel A. Premature atrial contraction-induced cardiomyopathy. Europace 2013; 15 (12) 1790
• 23 Agarwal SK, Chao J, Peace F. et al. Premature ventricular complexes on screening electrocardiogram and risk of ischemic stroke. Stroke 2015; 46 (05) 1365-1367
• 24 Lin CY, Chang SL, Lin YJ. et al. Long-term outcome of multiform premature ventricular complexes in structurally normal heart. Int J Cardiol 2015; 180: 80-85
• 25 Huizar JF, Kaszala K, Potfay J. et al. Left ventricular systolic dysfunction induced by ventricular ectopy: a novel model for premature ventricular contraction-induced cardiomyopathy. Circ Arrhythm Electrophysiol 2011; 4 (04) 543-549
• 26 Barrett PM, Komatireddy R, Haaser S. et al. Comparison of 24-hour Holter monitoring with 14-day novel adhesive patch electrocardiographic monitoring. Am J Med 2014; 127 (01) 95.e11-95.e17
• 27 Conen D, Adam M, Roche F. et al. Premature atrial contractions in the general population: frequency and risk factors. Circulation 2012; 126 (19) 2302-2308
• 28 Lee TM, Lin SZ, Chang NC. Differential effect of phosphodiesterase-3 inhibitors on sympathetic hyperinnervation in healed rat infarcts. Circ J 2014; 78 (02) 366-376
• 29 Barta J, Sanganalmath SK, Kumamoto H, Takeda N, Edes I, Dhalla NS. Antiplatelet agents sarpogrelate and cilostazol affect experimentally-induced ventricular arrhythmias and mortality. Cardiovasc Toxicol 2008; 8 (03) 127-135
• 30 Osmancik P, Herman D, Neuzil P. et al; PRAGUE-17 Trial Investigators. 4-year outcomes after left atrial appendage closure versus nonwarfarin oral anticoagulation for atrial fibrillation. J Am Coll Cardiol 2022; 79 (01) 1-14
• 31 Nogami A, Kurita T, Abe H. et al; JCS/JHRS Joint Working Group. JCS/JHRS 2019 Guideline on non-pharmacotherapy of cardiac arrhythmias. Circ J 2021; 85 (07) 1104-1244
• 32 Hannun AY, Rajpurkar P, Haghpanahi M. et al. Cardiologist-level arrhythmia detection and classification in ambulatory electrocardiograms using a deep neural network. Nat Med 2019; 25 (01) 65-69
• 33 Perez MV, Mahaffey KW, Hedlin H. et al; Apple Heart Study Investigators. Large-scale assessment of a smartwatch to identify atrial fibrillation. N Engl J Med 2019; 381 (20) 1909-1917
• 34 Rogers JD, Sanders P, Piorkowski C. et al. In-office insertion of a miniaturized insertable cardiac monitor: Results from the Reveal LINQ In-Office 2 randomized study. Heart Rhythm 2017; 14 (02) 218-224