Role of Remote Monitoring in Detection of Atrial Arrhythmia, Stroke Reduction, and Use of Anticoagulation Therapy　― A Systematic Review and Meta-Analysis ―

Jia-Pei Jang; Hui-Ting Lin; Yu-Jen Chen; Ming-Hsiung Hsieh; Yu-Chen Huang

doi:10.1253/circj.CJ-20-0633

Abstract

Background: The effect of remote monitoring (RM) in atrial arrhythmia detection, stroke reduction, and anticoagulation therapy remains unknown, particularly for patients with implantable or wearable cardiac devices.

Methods and Results: We performed a systematic review and meta-analysis to evaluate the role of RM in atrial arrhythmia detection, stroke reduction and anticoagulation therapeutic intervention. Online databases were queried to include randomized controlled trials comparing detection of atrial arrhythmia and stroke risk between patients undergoing RM and those receiving in-office (IO) follow-up. Outcomes and complications of RM-guided anticoagulation therapy and conventional therapy in patients with atrial fibrillation were also reviewed. A total of 16 studies were included. Compared with patients receiving IO follow-up, patients undergoing RM had a significantly higher detection rate of atrial arrhythmia (risk ratio [RR], 1.363; 95% confidence interval [CI], 1.147–1.619), and a lower risk of stroke (RR, 0.539; 95% CI, 0.301–0.936). The higher rate of atrial arrhythmia was only noted in patients with wearable devices (RR, 4.070; 95% CI, 2.408–6.877), and the lower risk of stroke was only noted in patients with cardiovascular implantable electronic devices (CIED) (RR, 0.513; 95% CI, 0.265–0.996).

Conclusions: RM is effective for atrial arrhythmia detection in patients using wearable devices and for reducing the risk of stroke in patients with CIED.

Atrial fibrillation (AF) is the most common and clinically significant cardiac arrhythmia. The paroxysmal nature of AF may result in diagnostic delays because the ECG may appear normal between episodes. In addition, AF can be accompanied by a few symptoms or even be clinically silent and thereby remain undiagnosed.¹ However, AF is associated with a 5-fold risk of developing ischemic stroke and is estimated to be the cause of 15% of all ischemic strokes.² Early diagnosis of subclinical atrial arrhythmias might prompt early intervention and decrease morbidity or mortality.

People suffering from certain types of heart failure (HF), or an abnormal heart rate/rhythm, may be treated with cardiovascular implantable electronic devices (CIED). These patients are particularly susceptible to the occurrence of AF, which may trigger inappropriate shock, increase the risk of thromboembolic events such as stroke, and precipitate HF.³ Patients with CIEDs may benefit from continuous remote monitoring (RM) of arrhythmic episodes. This technology has the ability to maintain surveillance and rapidly bring to attention significant data, enabling clinically appropriate intervention. In 2015, a meta-analysis showed that the rate of atrial arrhythmia did not differ between RM and in-office (IO) follow-up groups of patients with implantable cardioverter-defibrillators (ICD).⁴ A couple of studies on RM of patients with permanent pacemakers (PM) were recently published, but the results regarding detection of atrial arrhythmia were controversial,⁵^,⁶ and no systematic review was performed.

Besides CIED, wearable devices such as self-applied ECGs⁷ and smartwatches⁸ enabling continuous heart rate monitoring have been developed. A randomized controlled trial (RCT) showed that screening with twice-weekly single-lead iECG with RM of patients aged ≥65 years at increased risk of stroke was significantly more likely to identify incident AF than routine care.⁹ Although multiple studies on wearable technology have been carried out recently,⁸^,¹⁰ there is insufficient RCT-based evidence to evaluate the overall effect of RM in patients with wearable devices.

Even if it were possible to detect AF earlier, there are very little data to inform anticoagulation strategies for AF detected by RM. Some studies tried to demonstrate that intermittent anticoagulation therapy based on daily rhythm monitoring is safe for patients with paroxysmal AF and might decrease the bleeding complications associated with anticoagulation therapy.¹¹^,¹² However, the risk or benefit of oral anticoagulation therapy based on RM-detected AF remains uncertain.

In the current study, we aimed to perform a systematic review and meta-analysis to evaluate the applicability of RM in atrial arrhythmia detection and its effect on stroke reduction in patients with CIED or wearable devices. We also aimed to provide an evidence-based solution regarding the risks and benefits of RM-tailored anticoagulation therapy.

Methods

This study was performed in agreement with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses Statement (http://www.prisma-statement.org/).

Data Sourcing and Search Strategy

We extracted studies indexed in PubMed, EMBASE, CINAHL, Airiti Library, and the Cochrane Library from the inception of these databases to 29 February 2020, using the following MeSH terms: “remote monitoring”, “telemonitoring” or “telemedicine” combined with “atrial arrhythmia”, “atrial fibrillation”, “stroke’’, “cerebrovascular accident”, “anticoagulation” or “anticoagulant.” Included articles were limited to human studies published in English. All articles were screened for cross-references.

Eligibility Criteria and Study Selection

We identified RCT-included patients with (1) CIED, including PM, ICD, or cardiac resynchronization therapy (CRT) or (2) wearable devices, through which RM could be performed. All identified articles were screened and selected according to the following inclusion criteria: (1) one study group received RM and another group received regular IO follow-up with atrial arrhythmia detection or incidence of stroke as reported outcomes; or (2) one group received RM-guided anticoagulation therapy and another group received IO follow-up and conventional anticoagulation therapy with incidence of stroke, mortality, and bleeding as reported outcomes. Duplicate publications, review articles, and conference reports were excluded.

Two authors (J.-P.J. and H.-T.L.) independently reviewed the titles and abstracts of the articles. A subsequent full-text review was performed when abstracts did not provide sufficient information. Any disagreements were discussed with a third reviewer (Y.-C.H.) and resolved by consensus.

Outcome Measures

The primary outcomes were comparisons of atrial arrhythmia detection and stroke incidence in patients with and without RM. Secondary outcomes were comparisons of the incidence of stroke, death, minor and major bleeding in patients with RM-guided anticoagulation therapy and those with conventional anticoagulation therapy.

Data Extraction

Two reviewers (J.-P.J. and H.-T.L.) independently extracted the following data for all studies meeting the inclusion criteria: authors, year of publication, study design, sample size, inclusion criteria, type of RM device, frequency of RM and IO group visits, and follow-up time; age, sex, left ventricular ejection fraction (LVEF), CHA₂DS₂-VAS score, definition of atrial arrhythmia, number of detected atrial arrhythmias, and number of stroke patients. All the extracted data was cross-checked to rule out discrepancies. The corresponding authors were contacted for additional information when necessary.

Assessment of the Risk of Bias

Study quality was determined by 2 of the authors (J.-P.J. and H.-T.L.) using the Cochrane Collaboration Risk-of-Bias Instrument (RoB2.0).

Data Analysis

We performed a pooled estimate of the clinical outcomes (detection of atrial arrhythmia and incidence of stroke) in patients in the RM group compared with those in the IO follow-up group. Subgroup analyses were performed according to the different devices for RM (CIED [PM or ICD/CRT] and wearable devices). Dichotomous data were analyzed using risk ratio (RR) with 95% confidence intervals (CI). We performed the I square test to test heterogeneity. A random effects model was used for all analyses. Publication bias was tested by application of the Begg-Mazumdar test. All analyses were performed using Comprehensive Meta-Analysis v.3 (Biostat, Inc., Englewood, NJ, USA).

Results

Search Results and Trial Characteristics

Of 1,809 studies screened, 16 RCTs involving 15,036 patients with CIED or wearable devices fulfilled the inclusion criteria (Figure 1). The characteristics and detailed data of the included studies are summarized in Tables 1–3. Of the 16 included studies, 12 reported the detection rate of atrial arrhythmia (7 for ICD/CRT,¹³^–¹⁹ 3 for PM,⁵^,⁶^,²⁰ and 2 for wearable devices⁷^,⁹), 7 reported the incidence of stroke (2 for ICD/CRT,¹⁹^,²¹ 3 for PM,⁵^,⁶^,²⁰ 1 for both ICD and PM,²² and 1 for wearable devices⁹), and 3 studies provided results on RM-guided anticoagulation therapy.¹¹^,¹²^,¹⁸

Figure 1.

Flow chart of the selection of eligible studies included in the systematic review and meta-analysis. CRT, cardiac resynchronization therapy; ICD, implantable cardioverter-defibrillator; RCT, randomized controlled trial.

Table 1. Summary of Included Studies Regarding Atrial Arrhythmia and Stroke

Study	Center	Sample size	Randomization	Inclusion criteria	Type of device	RM visit	Control visit	Follow-up duration
Al-Khatib et al, 2010¹³	Multicenter	151 (76/75)	1:1	Implantation ICD	ICD	1. RM system every 3 months 2. Transmitted data reviewed within 2 business days 3.Telephone contact at 6 months to collect data	Every 3 months	12 months
Varma et al, 2010¹⁹ (TRUST)	Multicenter	1,339 (908/431)	2:1	0–45 days after successful device implantation	ICD	1. Automatic upload once every 24 h 2. IO visit every 3 months	IO visit every 3 months	12 months
Crossley et al, 2011¹⁵ (CONNECT)	Multicenter	1,997 (1,014/983)	1:1	ICD or CRT-D	ICD or CRT-D	Remote device transmission	IO visits at 1, 3, 6, 9, 12, and 15 months post-implantation	15 months
Mabo et al, 2012²⁰ (CAMPAS)	Multicenter	494 (248/246)	1:1	At least 1 month after implantation of PM	PM	1. Automatic upload once every 24 h 2. If an abnormal warning was found, immediately passed to physician	Doctor decided the time of returning	18 months
Landolina et al, 2012¹⁷ (EVOLVO)	Multicenter	200 (99/101)	1:1	LVEF <35%	ICD	1. Remote follow-up performed at 4 and 12 months 2. IO visits at 8 and 16 months	IO visits at 4, 8, 12, and 16 months	16 months
Perl et al, 2013²² (SAVE-HM)	Single center	PM 115 (50/65)	1:1	3 months after implantation of device	PM or ICD	Automatic upload once daily	Follow-up once per year	–
Perl et al, 2013²² (SAVE-HM)	Single center	ICD 36 (18/18)	1:1	3 months after implantation of device	PM or ICD	Automatic upload once daily	Follow-up once per year	–
Boriani et al, 2013¹⁴ (MORE-CARE)	Multicenter	105 (57/48)	1:1	Systolic heart failure with NYHA class III/IV (LVEF <35%)	CRT-D	1. Remote follow-up at 4 and 12 months with activation of automatic alerts 2. IO visits at baseline and at 8 months	IO visits at baseline and every 4 months	12 months
Guedon-Moreau et al, 2013¹⁶ (E-COST)	Multicenter	433 (221/212)	1:1	Heart failure with NYHA class IV	ICD	1–3 months after ICD implantation, and at 15 and 27 months of follow-up	1–3 months after ICD implantation for first follow-up ,and at 9, 15, 21, and 27 months of follow-up	27 months
Martin et al, 2015¹⁸	Multicenter	2,718 (1,357/1,361)	1:1	CHADS₂ risk score ≥1 and ability to tolerate anticoagulation	ICD or CRT-D	1. Continuously monitored by remote technology that issued notifications when AT occurred	Standard follow-up	Median 701 days
Sardu et al, 2016²¹ (TELECART)	Multicenter	183 (89/94)	1:1	NYHA class II or III, LVEF <35%	ICD or CRT-D	1. Continuous automatic RM 2. IO visit at 10 days, then 3, 6, 9 and 12 months after discharge	IO visit at 10 days, then 3, 6, 9 and 12 months after discharge	12 months
Lima et al, 2016⁶	Single center	300 (150/150)	1:1	≥60 years old	PM	1. 24-h continuous automatic RM 2. IO visit at 1, 3 and 6 months, then every 6 months	IO visit at 1, 3 and 6 months, then every 6 months	24 months
Amara et al, 2017⁵ (SETAM)	Multicenter	595 (291/304)	1:1	CHA₂DS₂-VASc score ≥2	PM	1. 24-h continuous automatic RM 2. 1–3 months and at 12 months after device implantation	1–3 months and at 12 months after device implantation	12 months
Halcox et al, 2017⁹ (REHEARSE-AF)	Single center	1,001 (500/501)	1:1	1. ≥65 years old 2. CHADS-VASc score ≥2 3. Not taking anticoagulant drugs	AliveCor device (iECG)	Twice-weekly recording and transmission of a 30-s single-lead iECG trace to a secure server	At 12, 32 and 52 weeks	12 months
Steinhubl et al, 2018⁷	Multicenter	5,214 (1,738/3,476)	1:2	1. Undiagnosed AF 2. ≥75 years old or 3. Male ≥55 years old or female ≥65 years old with ≥1 comorbidities	Self-applied continuous ECG	All participants were asked to wear 2 different patches for a period of up to 2 weeks for each patch, each 3 months apart to evaluate the additional potential benefit of more than 2 weeks of monitoring		4 months

AF, atrial fibrillation; AT, atrial tachycardia; CRT, cardiac resynchronization therapy; ECG, electrocardiography; ICD, implantable cardioverter-defibrillators; IO, in-office; LVEF, left ventricular ejection fraction; NYHA, New York heart association; PM, pacemaker; RM, remote monitoring.

Table 2. Summary of Included Studies Regarding Anticoagulation Therapy

Study	Center	Sample size	Randomization	Inclusion criteria	Type of monitor	Intervention group visit (RM-guided anticoagulation therapy)	Control group visit (conventional anticoagulation therapy)	Follow-up duration
Martin et al, 2015¹⁸	Multicenter	2,718 (1,357/1,361)	1:1	CHADS₂ risk score ≥1 and ability to tolerate anticoagulation	ICD or CRT-D	1. Continuously monitored by remote technology that issued notifications when AT occurred 2. Start and stop anticoagulation based on RM of rhythm 3. Stroke symptom questionnaire every 3 months	1. Standard follow-up and anticoagulation based on clinical criteria as determined by treating physician 2. Stroke symptom questionnaire every 3 months	Median 701 days
Stavrakis et al, 2017¹¹ (iCARE-AF)	Single center	58 (29/29)	1:1	1. Patients with paroxysmal AF, within 6 months of randomization on 2 separate occasions, at least 1 day part. 2. ≥1 additional risk factor for stroke	iPhone-based rhythm monitoring device	1. Transmit daily 30-s ECG rhythm strip at approximately the same time of the day 2. When AF was detected based on rhythm monitoring, patients received anticoagulation	1. Received a NOAC, with the choice of the NOAC left to the discretion of the referring physician
Waks et al, 2018¹² (TACTIC-AF)	Multicenter	61 (48/16)	1:1	1. Nonpermanent AF 2. Current DOAC use 3. >18 years old, had a St. Jude Medical PM or ICD 4. CHADS₂ score <3	PM or ICD	1. Biweekly remote transmissions (1 manual and 1 automatic respectively), automatic alert-triggered transmissions for AT/AF burden above a set threshold 2. IO visit at 6 and 12 months	1. Anticoagulation initiated/discontinued based on standard of care/guidelines 2. IO visit at 6 and 12 months	12 months

CRT-D, cardiac resynchronization therapy with defibrillators; DOAC, direct oral anticoagulants; NOAC, non-vitamin K antagonist oral anticoagulant. Other abbreviations as in Table 1.

Table 3. Detailed Data of Included Studies

Study	Group		Age (years)	M/F	LVEF (%)	CHA₂DS₂- VAS score	Definition of atrial arrhythmias	Atrial arrhythmia (n)	Stroke (n)
Al-khatib et al, 2010¹³	RM		63.0 (54–70)	4/72	–	–	AF or atrial flutter	45	–
Al-khatib et al, 2010¹³	Control		63.0 (54–72)	2/73	–	–	AF or atrial flutter	26	–
Varma et al, 2010¹⁹ (TRUST)	RM		63.3±12.8	653/255	29.0±10.7	–	AF	82	3
Varma et al, 2010¹⁹ (TRUST)	Control		64.0±12.1	315/116	28.5±9.8	–	AF	33	5
Crossley et al, 2011¹⁵ (CONNECT)	RM		65.2±12.4	278/705	28.6±10.0	–	AT or AF	107	–
Crossley et al, 2011¹⁵ (CONNECT)	Control		64.9±11.9	300/714	29.2±10.3	–	AT or AF	105	–
Mabo et al, 2012²⁰ (CAMPAS)	RM		76.0±9.0	320/174	–	–	Atrial arrhythmia	4	2
Mabo et al, 2012²⁰ (CAMPAS)	Control				–	–	Atrial arrhythmia	10	8
Landolina et al, 2012¹⁷ (EVOLVO)	RM		66 (60–72)	81/18	31 (25–35)	–	AT or AF	13	–
Landolina et al, 2012¹⁷ (EVOLVO)	Control		69 (60–73)	76/25	30 (25–34)	–	AT or AF	17	–
Perl et al, 2013²² (SAVE-HM)	PM	RM	74.5±10.3	24/26	–	–	–	–	1
	PM	Control	74.3±8.6	42/23	–	–		–	1
	ICD	RM	62.1±8.4	14/4	–	–		–	0
	ICD	Control	63.3±12.8	16/2	–	–		–	0
Boriani et al, 2013¹⁴ (MORE-CARE)	RM		67.0±10.0	21/55	27±6.0	–	AT or AF	12	–
Boriani et al, 2013¹⁴ (MORE-CARE)	Control		69.0±9.0	18/54	27±6.0	–	AT or AF	7	–
Guedon-Moreau et al, 2013¹⁶ (E-COST)	RM		62.0±13.0	193/28	34.7±13.0	–	Supraventricular arrhythmia	5	–
Guedon-Moreau et al, 2013¹⁶ (E-COST)	Control		61.2±12.0	189/23	35.1±13.6	–	Supraventricular arrhythmia	1	–
Martin et al, 2015¹⁸	RM		64.7±10.8	347/1,010	29.9±10.8	2.0^a,b	AF or atrial flutter	34	–
Martin et al, 2015¹⁸	Control		64.2±11.5	368/993	29.4±11.3	2.0^a,b	AF or atrial flutter	30	–
Sardu et al, 2016²¹ (TELECART)	RM		71.8±8.5	64/25	–	–	AF	–	3
Sardu et al, 2016²¹ (TELECART)	Control		72.6±5.7	75/19	–	–	AF	–	4
Lima et al, 2016⁶	RM		75.6±7.9	68/82	57.8±12.4	1.8±0.9^b	AF	36	1
Lima et al, 2016⁶	Control		74.8±7.8	64/86	58.3±12.9	1.8±0.8^b	AF	29	0
Amara et al, 2017⁵ (SETAM)	RM		79.0±8.0	187/104	–	3.7±1.2	AT, AF, or atrial flutter	83	4
Amara et al, 2017⁵ (SETAM)	Control		79.0±8.0	186/118		3.7±1.2	AT, AF, or atrial flutter	66	7
Halcox et al, 2017⁹ (REHEARSE-AF)	RM		72.6±5.4	241/259	–	3.0±1.0	AF	19	6
Halcox et al, 2017⁹ (REHEARSE-AF)	Control		72.6±5.4	225/276	–	3.0±1.0	AF	5	10
Steinhubl et al, 2018⁷	RM		73.5±7.4	845/521	–	3.0 (2–4)	AF	53	–
Steinhubl et al, 2018⁷	Control		73.1±7.2	788/505	–	3.0 (2–4)	AF	12	–

All data presented as mean±standard deviation or number or median (25–75th percentile). ^aMedian; ^bCHADS₂ score. Abbreviations as in Table 1.

Risk-of-Bias Assessment

The results of the risk-of-bias assessment are summarized in Figure 2. All studies except 1 had low risk of bias. The overall risk of bias in the study was of some concern due to bias in the missing outcome data.

Figure 2.

Risk of bias in included studies.

RM-Guided Anticoagulation Therapy

In 2 of the 3 included studies, the focus was RM-guided anticoagulation therapy in paroxysmal AF patients with implantable or wearable devices.¹¹^,¹² The 3rd study included all patients with RM devices.¹⁸ Considering the heterogeneity of studies, we did not pool the studies focusing on RM-guided anticoagulation therapy. For patients with paroxysmal AF, 1 study showed gastrointestinal bleeding was more frequent with conventional anticoagulation therapy than with RM-guided anticoagulation therapy (16% vs. 0%; P=0.047).¹¹ However, another study reported 1 case of fatal bleeding in the RM-guided anticoagulation therapy group and no major bleeding in the control group.¹² In another study of anticoagulation therapy,¹⁸ patients were included without existing arrhythmia and patients were divided into RM-guided anticoagulation therapy and conventional anticoagulation therapy if atrial tachyarrhythmia was noted. Major bleeding (hazard ratio [HR], 1.39; 95% CI, 0.89–2.17) was similar in the 2 groups. The authors concluded that the strategy of early initiation and interruption of anticoagulation based on RM-detected atrial tachyarrhythmia did not prevent thromboembolism and bleeding.¹⁸

Results of Statistical Analysis

The results of the meta-analysis are shown in Table 4. The pooled analysis showed that the detection rate of atrial arrhythmia was significantly higher in the RM group than that in the IO follow-up group (RR, 1.363; 95% CI, 1.147–1.619). However, the subgroup analysis showed that a higher rate of detection of atrial arrhythmia was only reported in patients using wearable devices (RR, 4.070; 95% CI, 2.408–6.877) (Figure 3, Table 4). The pooled incidence of stroke was significantly lower in the RM group than in the IO follow-up group (RR, 0.539; 95% CI, 0.301–0.936). In particular, the incidence of stroke was significantly lower in the RM group than in the IO follow-up group among patients with CIED (RR, 0.513; 95% CI, 0.265–0.996) but not in those with wearable devices (Figure 4). However, the incidence of stroke was similar in the RM and IO groups when we further divided the patients with CIED into ICD/CRT and PM groups (Table 4). No publication bias was noted for any of the analyses.

Table 4. Results of the Meta-Analysis

	Study no.	Effect size	Effect estimate (95% CI)	P value	I² (%)
RM vs. IO
Detection of atrial arrhythmia
Overall	12	RR	1.363 (1.147–1.619)	<0.0001	67.2
CIED	10	RR	1.195 (0.996–1.433)	0.056	34.5
ICD/CRT	7	RR	1.205 (0.953–1.524)	0.120	38.3
PM	3	RR	1.141 (0.770–1.690)	0.512	48.7
Wearable device	2	RR	4.070 (2.408–6.877)	<0.0001	0
Incidence of stroke
Overall	7	RR	0.539 (0.301–0.936)	0.028	0
CIED	6	RR	0.513 (0.265–0.996)	0.048	0
ICD/CRT	2	RR	0.468 (0.168–1.302)	0.146	0
PM	4	RR	0.549 (0.230–1.308)	0.176	0
Wearable device	1	RR	0.539 (0.215–1.653)	0.320	0

CI, confidence interval; CIED, cardiovascular implantable electronic device. Other abbreviations as in Table 1.

Figure 3.

Forest plot showing that the detection rate of atrial arrhythmia were significantly higher in the RM group than in the IO follow-up group (odds ratio: 1.507, 95% confidence interval (CI): 1.197–1.899). IO, in-office; RM, remote monitoring.

Figure 4.

Forest plot showing that the incidence of stroke was significantly lower in the RM group than in the IO follow-up group (odds ratio: 0.530, 95% confidence interval (CI): 0.302–0.930). IO, in-office; RM, remote monitoring.

Discussion

This meta-analysis included not only patients with CIED but also those with wearable devices. Compared with the patients in the IO follow-up, patients receiving RM had a significantly higher rate of detection of atrial arrhythmia and a lower risk of stroke. The higher detection rate was only noted in patients with wearable devices and the lower risk of stroke was only found in patients with CIED.

In agreement with the meta-analysis published by Parthiban et al,⁴ there was no significant difference in the rate of detection of atrial arrhythmia in patients with RM of ICD devices compared with those receiving regular IO follow-up; similar results were noted in patients with RM of PMs. However, the detection rate of atrial arrhythmia was significantly higher in patients with RM of wearable devices. In the study by Steinhubl et al, new AF was identified after 4 months in 3.9% (53/1,366) of the RM group vs. 0.9% (12/1,293) in the group without RM (absolute difference, 3.0%; 95% CI, 1.8%–4.1%).⁷ Another included study showed that 19 patients in the RM group were diagnosed with AF over the 12-month study period, while only 5 in the group not receiving RM were diagnosed with AF (HR, 3.9; 95% CI, 1.4–10.4; P=0.007).

Recently, a large-scale assessment of a smartwatch used to identify AF was performed.⁸ Irregular pulse was noted in 0.52% of the patients. Among participants who were notified of an irregular pulse, the positive predictive value was 0.84 for detecting AF on ECG simultaneously with a subsequent irregular pulse notification and 0.71 for detecting AF on ECG simultaneously with a subsequent irregular tachogram.⁸ Given the relative low positive predictive value of wearable devices compared with CIEDs, patients were toward to be detected of atrial arrhythmia in wearable devices. Conversely, statistical significance was obtained only in CIEDs for incidence of stroke, although there was a trend of lower risk of stroke with wearable devices. The other reason is that the meta-analysis only included one study by Halcox et al⁹ with relatively small sample size, which would lower the statistical power. Stroke was not included as an outcome in the study by Steinhubl et al;⁷ however, they found an increased initiation of anticoagulants (5.7 vs. 3.7 per 100 person-years; difference, 2.0; 95% CI, 1.9–2.2). Further RCT-based data are required to clarify the risk of stroke in patients with RM of wearable devices.

Although our study did not show any differences in atrial arrhythmia detection in patients with CIED, we found a lower risk of stroke in those patients. Parthiban et al found a significant decrease in time to clinical decision/event detection in the RM group compared with the IO follow-up group, with a mean difference in days to clinical decision/event detection of −27.1 days (P<0.001).⁴ Early intervention might be the explanation for the lower risk of stroke in patients with CIED. Considering the economic aspect, most studies showed that RM of CIED combined with visits to clinics was more cost-effective than regular IO follow-ups.²³^–²⁵ In line with the Heart Rhythm Society Expert Consensus in 2015, all patients with CIEDs should be offered RM as part of their standard follow-up management.²⁶

Although high-rate atrial episodes, subclinical AF and silent AF increase the stroke risk, it remains unclear whether initiating anticoagulation for these patients is effective and safe. Two ongoing clinical trials (ARTESiA²⁷ and NOAH²⁸) comparing anticoagulation treatment vs. aspirin or no anticoagulation among patients with CIED-detected subclinical AF will likely provide further information. Without concrete evidence, clinicians should take into account the duration and burden of subclinical AF in combination with the CHA₂DS₂-VASc score to determine the use of anticoagulation.²⁹

Another important issue is whether intermittent use of anticoagulation according to RM contributes to better clinical equipoise on the best way to reduce stroke risk and bleeding risk in patients with paroxysmal AF. However, we could not draw a conclusion given that 2 of our included studies¹¹^,¹² showed opposite results. Based on the current evidence, the management of AF should be guided by the 2019 focused update of the AHA/ACC/HRS guideline for the management of patients with AF.³⁰

Study Limitations

First, there was a large inter-study variability, including differences in the inclusion criteria, RM devices, detection parameters and sensitivity, transmission frequency, and alerting systems used. However, we performed a subgroup analysis to reduce heterogeneity. Second, we included all types of atrial arrhythmia such as high-rate atrial episodes, atrial tachycardia, AF and atrial flutter, but the different definitions of atrial arrhythmia may affect the results regarding stroke and anticoagulation therapy. Because a higher AF burden is related to a higher risk of stroke, the direction of bias was towards the RM group if we included AF alone. Third, the study numbers were small especially for outcomes regarding RM-guided anticoagulation therapy. Further RCTs are needed to confirm the results.

Conclusions

Our meta-analysis showed that RM of wearable devices can enhance the detection of atrial arrhythmia, and RM might also reduce the risk of stroke in patients with CIED. RM of wearable devices and CIED appears to be a good choice for patients at high risk of stroke.

Acknowledgment

The authors thank the Taipei Medical University for financial support under grant no. TMU108-AE1-B11, which made this study possible.

Sources of Funding

Taipei Medical University grant no. TMU108-AE1-B11.

Disclosure

No conflicts of interest.

IRB Information

The Wan Fang Hospital of Taipei Medical University decided that the study did not need ethical approval.

References

1. Sanna T, Diener HC, Passman RS, Di Lazzaro V, Bernstein RA, Morillo CA, et al. Cryptogenic stroke and underlying atrial fibrillation. N Engl J Med 2014; 370: 2478–2486.
2. Hart RG, Benavente O, McBride R, Pearce LA. Antithrombotic therapy to prevent stroke in patients with atrial fibrillation: A meta-analysis. Ann Intern Med 1999; 131: 492–501.
3. Ono M, Varma N. Remote monitoring for chronic disease management: Atrial fibrillation and heart failure. Card Electrophysiol Clin 2018; 10: 43–58.
4. Parthiban N, Esterman A, Mahajan R, Twomey DJ, Pathak RK, Lau DH, et al. Remote monitoring of implantable cardioverter-defibrillators: A systematic review and meta-analysis of clinical outcomes. J Am Coll Cardiol 2015; 65: 2591–2600.
5. Amara W, Montagnier C, Cheggour S, Boursier M, Gully C, Barnay C, et al. Early detection and treatment of atrial arrhythmias alleviates the arrhythmic burden in paced patients: The SETAM study. Pacing Clin Electrophysiol 2017; 40: 527–536.
6. Lima C, Martinelli M, Peixoto GL, Siqueira SF, Wajngarten M, Silva RT, et al. Silent atrial fibrillation in elderly pacemaker users: A randomized trial using home monitoring. Ann Noninvasive Electrocardiol 2016; 21: 246–255.
7. 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: 146–155.
8. Perez MV, Mahaffey KW, Hedlin H, Rumsfeld JS, Garcia A, Ferris T, et al. Large-scale assessment of a smartwatch to identify atrial fibrillation. N Engl J Med 2019; 381: 1909–1917.
9. Halcox JPJ, Wareham K, Cardew A, Gilmore M, Barry JP, Phillips C, et al. Assessment of remote heart rhythm sampling using the AliveCor Heart Monitor to screen for atrial fibrillation: The REHEARSE-AF study. Circulation 2017; 136: 1784–1794.
10. Bumgarner JM, Lambert CT, Hussein AA, Cantillon DJ, Baranowski B, Wolski K, et al. Smartwatch algorithm for automated detection of atrial fibrillation. J Am Coll Cardiol 2018; 71: 2381–2388.
11. Stavrakis S, Stoner JA, Kardokus J, Garabelli PJ, Po SS, Lazzara R. Intermittent vs. Continuous Anticoagulation theRapy in patiEnts with Atrial Fibrillation (iCARE-AF): A randomized pilot study. J Interv Card Electrophysiol 2017; 48: 51–60.
12. Waks JW, Passman RS, Matos J, Reynolds M, Thosani A, Mela T, et al. Intermittent anticoagulation guided by continuous atrial fibrillation burden monitoring using dual-chamber pacemakers and implantable cardioverter-defibrillators: Results from the Tailored Anticoagulation for Non-Continuous Atrial Fibrillation (TACTIC-AF) pilot study. Heart Rhythm 2018; 15: 1601–1607.
13. Al-Khatib SM, Piccini JP, Knight D, Stewart M, Clapp-Channing N, Sanders GD. Remote monitoring of implantable cardioverter defibrillators versus quarterly device interrogations in clinic: Results from a randomized pilot clinical trial. J Cardiovasc Electrophysiol 2010; 21: 545–550.
14. Boriani G, Da Costa A, Ricci RP, Quesada A, Favale S, Iacopino S, et al. The MOnitoring Resynchronization dEvices and CARdiac patiEnts (MORE-CARE) randomized controlled trial: Phase 1 results on dynamics of early intervention with remote monitoring. J Med Internet Res 2013; 15: e167.
15. Crossley GH, Boyle A, Vitense H, Chang Y, Mead RH. The CONNECT (Clinical Evaluation of Remote Notification to Reduce Time to Clinical Decision) trial: The value of wireless remote monitoring with automatic clinician alerts. J Am Coll Cardiol 2011; 57: 1181–1189.
16. Guedon-Moreau L, Lacroix D, Sadoul N, Clementy J, Kouakam C, Hermida JS, et al. A randomized study of remote follow-up of implantable cardioverter defibrillators: Safety and efficacy report of the ECOST trial. Eur Heart J 2013; 34: 605–614.
17. Landolina M, Perego GB, Lunati M, Curnis A, Guenzati G, Vicentini A, et al. Remote monitoring reduces healthcare use and improves quality of care in heart failure patients with implantable defibrillators: The evolution of management strategies of heart failure patients with implantable defibrillators (EVOLVO) study. Circulation 2012; 125: 2985–2992.
18. Martin DT, Bersohn MM, Waldo AL, Wathen MS, Choucair WK, Lip GY, et al. Randomized trial of atrial arrhythmia monitoring to guide anticoagulation in patients with implanted defibrillator and cardiac resynchronization devices. Eur Heart J 2015; 36: 1660–1668.
19. Varma N, Epstein AE, Irimpen A, Schweikert R, Love C. Efficacy and safety of automatic remote monitoring for implantable cardioverter-defibrillator follow-up: The Lumos-T Safely Reduces Routine Office Device Follow-up (TRUST) trial. Circulation 2010; 122: 325–332.
20. Mabo P, Victor F, Bazin P, Ahres S, Babuty D, Da Costa A, et al. A randomized trial of long-term remote monitoring of pacemaker recipients (the COMPAS trial). Eur Heart J 2012; 33: 1105–1111.
21. Sardu C, Santamaria M, Rizzo MR, Barbieri M, di Marino M, Paolisso G, et al. Telemonitoring in heart failure patients treated by cardiac resynchronisation therapy with defibrillator (CRT-D): The TELECART study. Int J Clin Pract 2016; 70: 569–576.
22. Perl S, Stiegler P, Rotman B, Prenner G, Lercher P, Anelli-Monti M, et al. Socio-economic effects and cost saving potential of remote patient monitoring (SAVE-HM trial). Int J Cardiol 2013; 169: 402–407.
23. Ricci RP, Vicentini A, D’Onofrio A, Sagone A, Rovaris G, Padeletti L, et al. Economic analysis of remote monitoring of cardiac implantable electronic devices: Results of the Health Economics Evaluation Registry for Remote Follow-up (TARIFF) study. Heart Rhythm 2017; 14: 50–57.
24. Zanaboni P, Landolina M, Marzegalli M, Lunati M, Perego GB, Guenzati G, et al. Cost-utility analysis of the EVOLVO study on remote monitoring for heart failure patients with implantable defibrillators: Randomized controlled trial. J Med Internet Res 2013; 15: e106.
25. Klersy C, De Silvestri A, Gabutti G, Raisaro A, Curti M, Regoli F, et al. Economic impact of remote patient monitoring: An integrated economic model derived from a meta-analysis of randomized controlled trials in heart failure. Eur J Heart Fail 2011; 13: 450–459.
26. Slotwiner D, Varma N, Akar JG, Annas G, Beardsall M, Fogel RI, et al. HRS Expert Consensus Statement on remote interrogation and monitoring for cardiovascular implantable electronic devices. Heart Rhythm 2015; 12: e69–e100.
27. Lopes RD, Alings M, Connolly SJ, Beresh H, Granger CB, Mazuecos JB, et al. Rationale and design of the Apixaban for the Reduction of Thrombo-Embolism in Patients With Device-Detected Sub-Clinical Atrial Fibrillation (ARTESiA) trial. Am Heart J 2017; 189: 137–145.
28. Kirchhof P, Blank BF, Calvert M, Camm AJ, Chlouverakis G, Diener HC, et al. Probing oral anticoagulation in patients with atrial high rate episodes: Rationale and design of the Non-vitamin K antagonist Oral anticoagulants in patients with Atrial High rate episodes (NOAH-AFNET 6) trial. Am Heart J 2017; 190: 12–18.
29. Freedman B, Boriani G, Glotzer TV, Healey JS, Kirchhof P, Potpara TS. Management of atrial high-rate episodes detected by cardiac implanted electronic devices. Nat Rev Cardiol 2017; 14: 701–714.
30. January CT, Wann LS, Calkins H, Chen LY, Cigarroa JE, Cleveland JC, et al. 2019 AHA/ACC/HRS Focused Update of the 2014 AHA/ACC/HRS Guideline for the Management of Patients With Atrial Fibrillation: A Report of the American College of Cardiology/American Heart Association Task Force on Clinical Practice Guidelines and the Heart Rhythm Society. J Am Coll Cardiol 2019; 74: 104–132.

Corresponding author

Register with J-STAGE for free!