CAM 20208

Ambulatory Event Monitors and Mobile Cardiac Outpatient Telemetry

Category:Medicine   Last Reviewed:June 2019
Department(s):Medical Affairs   Next Review:June 2020
Original Date:July 1998    

Description: 
Various devices are available for outpatient cardiac rhythm monitoring. These devices differ in the types of monitoring leads used, the duration and continuity of monitoring, the ability to detect arrhythmias without patient intervention, and the mechanism of delivering the information from patient to clinician.  

These devices may be used to evaluate symptoms suggestive of arrhythmias (e.g., syncope, palpitations), and may be used to detect atrial fibrillation (AF) in patients who have undergone cardiac ablation of AF or who have a history of cryptogenic stroke.

Ambulatory Event Monitoring
For individuals who have signs and/or symptoms suggestive of arrhythmia(s) who receive patient- or autoactivated external ambulatory event monitoring or continuous ambulatory monitoring storing information for more than 48 hours, the evidence includes prospective and retrospective studies reporting on the diagnostic yield. Relevant outcomes are overall survival and morbid events. Studies have shown that continuous monitoring with longer recording periods clearly detects more arrhythmias than 24- or 48-hour Holter monitoring. Particularly for patients who, without the more prolonged monitoring, would only undergo shorter term monitoring, the diagnostic yield is likely to identify arrhythmias that may have therapeutic implications. The evidence is sufficient to determine that the technology results in a meaningful improvement in the net health outcome.

For individuals who have AF following ablation who receive long-term ambulatory cardiac monitoring, the evidence includes a randomized controlled trial (RCT) comparing ambulatory event monitoring with standard care and several observational studies. Relevant outcomes are overall survival, morbid events, medication use, and treatment-related morbidity. The RCT evaluating a long-term monitoring strategy after catheter ablation for AF reported significantly higher rates of AF detection. The available evidence has suggested that long-term monitoring for AF postablation is associated with improved outcomes. However, the specific type of monitoring associated with the best outcomes is not established, because different long-term monitoring devices were used across the studies. Trials demonstrating improved outcomes have used event monitors or implantable monitors. In addition, there are individual patient considerations that may make 1 type of monitor preferable over another. The evidence is sufficient to determine that the technology results in a meaningful improvement in the net health outcome.

For individuals who have cryptogenic stroke with a negative standard workup for AF who receive long-term ambulatory cardiac monitoring, the evidence includes systematic reviews of RCTs comparing ambulatory event monitoring with standard care. Relevant outcomes are overall survival, morbid events, medication use, and treatment-related morbidity. RCTs evaluating a long-term AF monitoring strategy poststroke have reported significantly higher rates of AF detection with longer term ambulatory monitoring. The available evidence has suggested that long-term monitoring for AF after cryptogenic stroke is associated with improved outcomes, but the specific type of monitoring associated with the best outcomes is not established, because different long-term monitoring devices were used across the studies. Trials demonstrating improved outcomes have used event monitors or implantable monitors. In addition, there are individual patient considerations that may make 1 type of monitor preferable over another. The evidence is sufficient to determine that the technology results in a meaningful improvement in the net health outcome.

For individuals who are asymptomatic with risk factors for AF who receive long-term ambulatory cardiac monitoring, the evidence includes an RCT and 2 nonrandomized studies. Relevant outcomes are overall survival, morbid events, medication use, and treatment-related morbidity. The studies showed use of the ambulatory monitors would result in higher AF detection compared with routine care. However, the RCT followed patients for 1 year and did not detect a difference in stroke occurrence between the monitored group and the standard of care group. The other studies did not discuss changes in patient management or health outcomes based on monitoring. Studies reporting on improved outcomes with longer follow-up are needed. The evidence is insufficient to determine the effects of the technology on health outcomes.

Implantable Loop Recording
For individuals who have signs and/or symptoms suggestive of arrhythmia with infrequent symptoms who receive patient- or autoactivated implantable ambulatory event monitoring, the evidence includes RCTs comparing implantable loop recorders with shorter term monitoring, usually 24- to 48-hour Holter monitoring. Relevant outcomes are overall survival, morbid events, medication use, and treatment-related morbidity. Studies assessing prolonged implantable loop recorders in patients have reported high rates of arrhythmia detection compared with shorter external event or Holter monitoring. These studies have supported use of a progression in diagnostics from an external event monitor to implantable loop recorder when longer monitoring is needed. The evidence is sufficient to determine that the technology results in a meaningful improvement in the net health outcome.

Outpatient Cardiac Telemetry
For individuals who have signs and/or symptoms suggestive of arrhythmia who receive outpatient cardiac telemetry, the evidence includes an RCT and nonrandomized studies evaluating rates of arrhythmia detection using outpatient cardiac telemetry. Relevant outcomes are overall survival and morbid events. The available evidence has suggested that outpatient cardiac telemetry is at least as good at detecting arrhythmias as ambulatory event monitoring. However, studies have not evaluated whether the real-time monitoring feature of outpatient cardiac telemetry leads to reduced cardiac events and mortality. The evidence is insufficient to determine the effects of the technology on health outcomes.

Background 
CARDIAC ARRHYTHMIAS
Cardiac monitoring is routinely used in the inpatient setting to detect acute changes in heart rate or rhythm that may need urgent response. For some conditions, a more prolonged period of monitoring in the ambulatory setting is needed to detect heart rate or rhythm abnormalities that may occur infrequently. These cases may include the diagnosis of arrhythmias in patients with signs and symptoms suggestive of arrhythmias as well as the evaluation of paroxysmal atrial fibrillation (AF).

Cardiac arrhythmias may be suspected because of symptoms suggestive of arrhythmias, including palpitations, dizziness, or syncope or presyncope, or because of abnormal heart rate or rhythm noted on exam. A full discussion of the differential diagnosis and evaluation of each of these symptoms is beyond the scope of this review, but some general principles on the use of ambulatory monitoring are discussed.

Arrhythmias are an important potential cause of syncope or near syncope, which in some cases may be described as dizziness. An electrocardiogram (ECG) is generally indicated whenever there is suspicion of a cardiac cause of syncope. Some arrhythmic causes will be apparent on ECG. However, for patients in whom an ECG is not diagnostic, longer monitoring may be indicated. The 2009 joint guidelines from the European Society of Cardiology and 3 other medical specialty societies suggested that, in individuals with clinical or ECG features suggesting an arrhythmic syncope, ECG monitoring is indicated; the guidelines also stated that the "duration (and technology) of monitoring should be selected according to the risk and the predicted recurrence rate of syncope."1 Similarly, guidelines from the National Institute for Health and Care Excellence (2014) on the evaluation of transient loss of consciousness, have recommended the use of an ambulatory ECG in individuals with a suspected arrhythmic cause of syncope. The type and duration of monitoring recommended is based on the individual's history, particularly the frequency of transient loss of consciousness.2 The Holter monitor is recommended if transient loss of consciousness occurs several times a week. If the frequency of transient loss of consciousness is every one to two weeks, an external event recorder is recommended; and if the frequency is less than once every two weeks, an implantable event recorder is recommended.

Similar to syncope, the evaluation and management of palpitations is patient-specific. In cases where the initial history, examination, and ECG findings are suggestive of an arrhythmia, some form of ambulatory ECG monitoring is indicated. A position paper from the European Heart Rhythm Association (2011) indicated that, for individuals with palpitations of unknown origin who have clinical features suggestive of arrhythmia, referral for specialized evaluation with consideration for ambulatory ECG monitoring is indicated.3

AF Detection
AF is the most common arrhythmia in adults. It may be asymptomatic or be associated with a broad range of symptoms, including lightheadedness, palpitations, dyspnea, and a variety of more nonspecific symptoms (eg, fatigue, malaise). It is classified as paroxysmal, persistent, or permanent based on symptom duration. Diagnosed AF may be treated with antiarrhythmic medications with the goal of rate or rhythm control. Other treatments include direct cardioversion, catheter-based radiofrequency- or cryo-energy-based ablation, or one of several surgical techniques, depending on the patient's comorbidities and associated symptoms.

Stroke in AF occurs primarily as a result of thromboembolism from the left atrium. The lack of atrial contractions in AF leads to blood stasis in the left atrium, and this low flow state increases the risk of thrombosis. The area of the left atrium with the lowest blood flow in AF, and therefore the highest risk of thrombosis, is the left atrial appendage. Multiple clinical trials have demonstrated that anticoagulation reduces the ischemic stroke risk in patients at moderate- or high-risk of thromboembolic events. Oral anticoagulation in patients with AF reduces the risk of subsequent stroke and was recommended by American Heart Association, American College of Cardiology, and Heart Rhythm Society (2014) joint guidelines on patients with a history of stroke or transient ischemic attack.4

Ambulatory ECG monitoring may play a role in several situations in the detection of AF. In patients who have undergone ablative treatment for AF, if ongoing AF can be excluded with reasonable certainty, including paroxysmal AF which may not be apparent on ECG during an office visit, anticoagulation therapy could potentially be stopped. In some cases where identifying paroxysmal AF is associated with potential changes in management, longer term monitoring may be considered. There are well-defined management changes that occur in patients with AF. However, until relatively recent the specific role of long-term (ie, >48 hours) monitoring in AF was not well-described.

Patients with cryptogenic stroke are often monitored for the presence of AF because AF is estimated to be the cause of cryptogenic stroke in more than 10% of patients, and AF increases the risk of stroke.5,6 Paroxysmal AF confers an elevated risk of stroke, just as persistent and permanent AF do. In individuals with a high-risk of stroke, particularly those with a history of ischemic stroke that is unexplained by other causes, prolonged monitoring to identify paroxysmal AF has been investigated.

Cardiac Rhythm Ambulatory Monitoring Devices
Ambulatory cardiac monitoring with a variety of devices permits the evaluation of cardiac electrical activity over time, in contrast to a static ECG, which only permits the detection of abnormalities in cardiac electrical activity at a single point in time.

A Holter monitor is worn continuously and records cardiac electrical output continuously throughout the recording period. Holter monitors are capable of recording activity for 24 to 72 hours. Traditionally, most Holter monitors have three channels based on three ECG leads. However, some currently available Holter monitors have up to 12 channels. Holter monitors are an accepted intervention in a variety of settings where a short period (24-48 hours) of comprehensive cardiac rhythm assessment is needed (eg, suspected arrhythmias when symptoms [syncope, palpitations] are occurring daily). These devices are not the focus of this review.

Various classes of devices are available for situations where longer monitoring than can be obtained with a traditional Holter monitor is needed. Because there may be many devices within each category, a comprehensive description of each is beyond our scope. Devices vary in how data are transmitted to the location where the ECG output is interpreted. Data may be transmitted via cellular phone or landline, or by direct download from the device after its return to the monitoring center. The device classes are described in Table 1.

Table 1. Ambulatory Cardiac Rhythm Monitoring Devices 

Device Class

Description

Device Examples

Noncontinuous devices with memory (event recorder)

Devices not worn continuously but rather activated by patient and applied to skin in the precordial area when symptoms develop

  • Zio® Event Card (iRhythm Technologies)
  • REKA E100™ (REKA Health)

Continuous recording devices with longer recording periods

Devices continuously worn and continuously record via ≥1 cardiac leads and store data longer than traditional Holter (14 d)

  • Zio® Patch system (iRhythm Technologies)

 

External memory loop devices (patient- or autotriggered)

Devices continuously worn and store a single channel of ECG data in a refreshedmemory. When the device is activated, the ECG is then recorded from the memory loop for the preceding 30-90 s and for next 60 s or so. Devices may be activated by a patient when symptoms occur (patient-triggered) or by an automated algorithm when changes suggestive of an arrhythmia are detected (autotriggered).

  • Patient-triggered: Explorer™ Looping Monitor (LifeWatch Services)
  • Autotriggered: LifeStar AF Express™ Auto-Detect Looping Monitor (LifeWatch Services)
  • Autotriggered or patient-triggered: King of Hearts Express® AF (Card Guard Scientific Survival)

Implantable memory loop devices (patient- or autotriggered)

Devices similar in design to external memory loop devices but implanted under the skin in the precordial region

  • Autotriggered or patient-triggered: Reveal® XT ICM (Medtronic) and Confirm Rx Insertable™ Cardiac Monitor (Abbott)
  • Autotriggered: BioMonitor, Biotronik)

Mobile cardiac outpatient telemetry

Continuously recording or autotriggered memory loop devices that transmit data to a central recording station with real-time monitoring and analysis

  • CardioNet MCOT (BioTelemetry)
  • LifeStar Mobile Cardiac Telemetry (LifeWatch Services)
  • SEEQ Mobile Cardiac Telemetry (Medtronic)

ECG: electrocardiogram. 

There are also devices that combine features of multiple classes. For example, the LifeStar ACT Ex Holter (LifeWatch Services) is a 3-channel Holter monitor, but is converted to a mobile cardiac telemetry system if a diagnosis is inconclusive after 24 to 48 hours of monitoring. The BodyGuardian® Heart Remote Monitoring System (Preventice Services) is an external autotriggered memory loop device that can be converted to a real-time monitoring system. The eCardio Verité™ system (eCardio) can switch between a patient-activated event monitor and a continuous telemetry monitor. The Spiderflash-T (LivaNova) is an example of an external autotriggered or patient-triggered loop recorder, but like the ZioÒ Patch, can record 2 channels for 14 to 40 days..  

Policy: 
The use of patient-activated or auto-activated external ambulatory event monitors OR continuous ambulatory monitors that record and store information for periods longer than 48 hours may be considered MEDICALLY NECESSARY as a diagnostic alternative to Holter monitoring in the following situations:

  • Patients who experience infrequent symptoms (less frequently than every 48 hours) suggestive of cardiac arrhythmias (i.e., palpitations, dizziness, presyncope or syncope).
  • Patients with atrial fibrillation who have been treated with catheter ablation, and in whom discontinuation of systemic anticoagulation is being considered.
  • Patients with cryptogenic stroke who have a negative standard work-up for atrial fibrillation, including a 24-hour Holter monitor (see Policy Guidelines section).

The use of implantable ambulatory event monitors, either patient-activated or auto-activated, may be considered MEDICALLY NECESSARY in the following situations:

  • In the small subset of patients who experience recurrent symptoms so infrequently that a prior trial of other external ambulatory event monitors has been unsuccessful.
  • In patients who require long-term monitoring for atrial fibrillation or possible atrial fibrillation (see Policy Guidelines section).

The use of outpatient cardiac telemetry (also known as mobile cardiac outpatient telemetry) as a diagnostic alternative to ambulatory event monitors in patients who experience infrequent symptoms (less frequently than every 48 hours) suggestive of cardiac arrhythmias (i.e., palpitations, dizziness, presyncope, syncope) is considered INVESTIGATIONAL.

Other uses of ambulatory event monitors, including outpatient cardiac telemetry and mobile applications, are considered INVESTIGATIONAL, including but not limited to monitoring effectiveness of antiarrhythmic medications and detection of myocardial ischemia by detecting ST-segment changes.

The use of long-term (greater than 48 hours) external ECG monitoring by continuous rhythm recording and storage (e.g., Zio Patch®) is established as MEDICALLY NECESSARY for the evaluation of patients syspected of having an arrhythmia 

  • Following Holter monitoring, when the results of the Holter monitoring were non-diagnostic.
  • Instead of Holter monitoring, for the evaluation of patients suspected of having an arrhythmia
    • Whose arrhythmias/symptoms occur so infrequently such that a Holter monitor is unlikely to provide a diagnosis; OR
    • When the patient is unlikely to recognize symptoms as being cardiac-related or the suspected cardiac changes may be asymptomatic; OR
    • When symptoms are so severe as to make the patient unable to activate an event monitor or the patient is unable to use the monitor due to cognitive or other patient-related factors; OR
    • Following a recent radiofrequency ablation for an arrhythmogenic focus to assess an arrhythmia that may be asymptomatic or that may occur beyond 48 hours post initiation of monitoring; OR
    • To assess the therapeutic effect of arrhythmia therapy and support there planning.

Policy Guidelines
The available evidence suggests that long-term monitoring for atrial fibrillation after cryptogenic stroke or postablation is associated with improved outcomes, but the specific type of monitoring associated with the best outcomes is not well-defined. Trials that have demonstrated improved outcomes have used either event monitors or implantable monitors. In addition, there are individual patient considerations that may make 1 type of monitor preferable over another.

Therefore, for the evaluation of patients with cryptogenic stroke who have had a negative standard workup for atrial fibrillation including 24-hour Holter monitoring, or for the evaluation of atrial fibrillation after an ablation procedure, the use of long-term monitoring with an external event monitor, OR a continuous ambulatory monitor that records and stores information for periods longer than 48 hours, OR an implantable ambulatory monitor may be considered medically necessary for patients who meet the criteria outlined above.

Coding
The implantation and removal of an insertable loop recorder are coded as follows:

33282: Implantation of patient-activated cardiac event recorder
33284: Removal of an implantable, patient-activated cardiac event recorder

The interpretation of the electrocardiograms (ECGs) recorded with ambulatory event monitors (AEMs) may be coded as follows:

93268: External patient and, when performed, auto activated electrocardiographic rhythm derived event recording with symptom-related memory loop with remote download capability up to 30 days, 24-hour attended monitoring; includes transmission, review and interpretation by a physician or other qualified health care professional

The above CPT code represents a bundled CPT code including all components of AEM monitoring, including ECG analysis of all recorded strips during a 30-day period.

Other CPT codes that can be used for AEM monitoring represent unbundling of the 93268 code. For example, CPT code 93270 describes the connection, recording and disconnection of an external device; CPT code 93271 describes the transmission download and analysis; and 93272 describes the physician review and interpretation of the ECG strips. AEM monitoring services may supply the monitoring, receipt of transmissions, and analysis of the ECGs (i.e., CPT codes 93271 and 93272), but the provider supplies the hookup and disconnection of the device (i.e., CPT code 93270). If this is the case, the unbundled codes may be used. It should also be noted that CPT code 93272 (physician review and interpretation) applies to all ECGs transmitted during a 30-day period; therefore, billing for each individual transmitted strip is not warranted.

There are specific CPT codes for mobile outpatient cardiac telemetry:

93228: External mobile cardiovascular telemetry with electrocardiographic recording, concurrent computerized real-time data analysis and greater than 24 hours of accessible ECG data storage (retrievable with query) with ECG triggered and patient selected events transmitted to a remote attended surveillance center for up to 30 days; review and interpretation with report by a physician or other qualified health care professional
93229 ; technical support for connection and patient instructions for use, attended surveillance, analysis and transmission of daily and emergent data reports as prescribed by a physician or other qualified health care professional

Both of these codes can only be reported once per 30 days of service.

There are category III CPT codes for devices with longer recording capabilities:

0295T External electrocardiographic recording for more than 48 hours up to 21 days by continuous rhythm recording and storage; includes recording, scanning analysis with report, review and interpretation
0296T ; recording (includes connection and initial recording)
0297T ; scanning analysis with report
0298T ; review and interpretation

Benefit Application
BlueCard®/National Account Issues
Plans may consider requiring the use of CPT code 93268, when possible, to avoid unbundling of services. However, aside from the hook-up and disconnection of the device, which is frequently performed by the provider, the actual monitoring and analysis of the electrocardiography (EKG) are frequently performed by a monitoring service. If this is the case, the various components of the ambulatory event monitors (AEMs) will be unbundled.

Because use of mobile cardiac outpatient telemetry (MCOT) is generally more costly than alternatives such as the autotrigger device, but has not been shown to lead to improved outcomes compared to those alternatives, it is considered not medically necessary using the Medical Policy Reference Manual definition of medical necessity.

For contracts that do not use this definition of medical necessity, other contract provisions may apply. For example, benefit or contract language describing the "least costly alternative" may also be applicable for this choice of testing.

Rationale 
This evidence review was created in July 1998 and has been updated regularly with searches of the MEDLINE database. The most recent literature update was performed through March 26, 2019.

Evidence reviews assess the clinical evidence to determine whether the use of technology improves the net health outcome. Broadly defined, health outcomes are the length of life, quality of life, and ability to function---including benefits and harms. Every clinical condition has specific outcomes that are important to patients and managing the course of that condition. Validated outcome measures are necessary to ascertain whether a condition improves or worsens; and whether the magnitude of that change is clinically significant. The net health outcome is a balance of benefits and harms.

To assess whether the evidence is sufficient to draw conclusions about the net health outcome of technology, two domains are examined: the relevance, and quality and credibility. To be relevant, studies must represent one or more intended clinical uses of the technology in the intended population and compare an effective and appropriate alternative at a comparable intensity. For some conditions, the alternative will be supportive care or surveillance. The quality and credibility of the evidence depend on study design and conduct, minimizing bias and confounding that can generate incorrect findings. The randomized controlled trial (RCT) is preferred to assess efficacy; however, in some circumstances, nonrandomized studies may be adequate. RCTs are rarely large enough or long enough to capture less common adverse events and long-term effects. Other types of studies can be used for these purposes and to assess generalizability to broader clinical populations and settings of clinical practice.

This review is structured around three questions: First, in what clinical situations, and with what classes, do ambulatory event monitors (AEMs) improve health outcomes? Second, under what circumstances are implantable AEMs associated with improved outcomes? Third, under what circumstances is real-time monitoring associated with improved outcomes?

For some of AEMs discussed herein, including those that include real-time monitoring and analysis, the technologies represent an enhancement to existing technology and are intended to improve outcomes compared with event monitors. As such, to demonstrate an improvement in health outcomes, there must be a clinically significant incremental benefit when the additional technology, such as real-time monitoring, is added.

AEMS IN THE DETECTION OF ARRHYTHMIAS
The first four sections of the policy focus on clinical situations for which the use of long-term AEMs may be associated with improved health outcomes. 

  • The use of long-term AEMs in the diagnosis of cardiac rhythm abnormalities in individuals with signs and/or symptoms of arrhythmias (eg, dizziness, syncope or near syncope, palpitations) is discussed. Specific arrhythmias may be relatively nonspecific in terms of the symptoms they cause. However, the diagnosis of some arrhythmias has well-defined management implications that are known to improve outcomes, such as the use of an implantable cardioverter defibrillator in individuals with potentially lethal arrhythmias, or antiarrhythmic drugs or pulmonary vein isolation for the treatment of atrial fibrillation (AF). Therefore, identification of an arrhythmia is considered a reasonable endpoint in this case.
  • The use of long-term AEMs for the detection of AF in patients following catheter ablation,  for which management (use of anticoagulation therapy) may be changed based on AF detection.
  • The use of long-term AEMs for the detection of AF in patients following cryptogenic stroke, for which management (use of anticoagulation therapy) may be changed based on AF detection.
  • The use of long-term AEMs for the detection of AF in asymptomatic patients.

The last two sections of the policy focus on types of long-term AEMs: implantable AEMs and outpatient cardiac telemetry.

AUTOACTIVATED EXTERNAL OR CONTINUOUS AMBULATORY EVENT MONITORING FOR PATIENTS WITH ARRHYTHMIA SYMPTOMS
Clinical Context and Test Purpose
The purpose of patient- or autoactivated external ambulatory event monitoring or continuous ambulatory event monitoring in patients who have signs and/or symptoms of arrhythmia is to provide an alternative detection method for AF.

The question addressed in this evidence review is: Does the use of patient- or autoactivated or continuous ambulatory event monitoring for patients with symptoms of arrhythmia improve net health outcome compared with electrocardiogram (ECG) only or 24 to 48 hour Holter monitoring?

The following PICOTS were used to select literature to inform this review.

Patients
The relevant population of interest are individuals with signs or symptoms suggestive of arrhythmia.

Interventions
The intervention being considered is patient- or autoactivated external event monitoring or continuous ambulatory event monitoring. Patient-activated devices are applied to the skin in the precordial area by the patient when symptoms are developing. Continuous event monitoring devices are worn continuously and are recording activity continuously and can store data longer than the Holter monitor.

Alternative AF detection methods that are used include an ECG or 24- to 48-hour Holter monitoring. An ECG provides information on cardiac electrical activity at one point in time.A Holter monitor is worn continuously and records cardiac electrical output continuously throughout the recording period. Holter monitors are capable of recording activity for 24 to 72 hours.

Outcomes
The general outcome of interest is diagnostic yield of the monitors in detecting arrhythmias. To measure incremental benefits of the patient-activated or continuous monitors, direct comparisons with the Holter monitor, or indirect comparisons of the number of detections in the first 48 hours with the number of detections during longer monitoring periods can be made.

Study Selection Criteria
For the evaluation of clinical validity of autoactivated or patient-activated external ambulatory event monitoring for patients with arrhythmia symptoms, studies that met the following criteria were considered:  

  • To assess the clinical validity, studies should report sensitivity, specificity, positive and negative predictive values. Alternatively, studies reporting on diagnostic yield are informative.
  • To assess the clinical utility, studies should demonstrate how results of the tests impacted treatment decisions and overall management of the patient.

Technically Reliable
Assessment of technical reliability focuses on specific tests and operators and requires review of unpublished and often proprietary information. Review of specific tests, operators, and unpublished data are outside the scope of this evidence review and alternative sources exist. This evidence review focuses on the clinical validity and clinical utility.

Clinically Valid
A test must detect the presence or absence of a condition, the risk of developing a condition in the future, or treatment response (beneficial or adverse). Below are studies providing evidence on the diagnostic yield of long-term AEMs in symptomatic patients.

Long-Term Ambulatory Event Monitoring in Symptomatic Patients
Newer devices are available that record cardiac rhythms continuously for longer periods of time than traditional Holter monitors. Several studies have evaluated the diagnostic yield of continuous monitoring for more than 48 hours, either directly through comparison with Holter monitoring or indirectly by calculating the proportion of arrhythmias detected in the first 48 hours of monitoring. The diagnostic yield of monitoring with external event monitors depends on the underlying population, the inherent sensitivity of the device, and the duration of monitoring.

Systematic Review
Hoefman et al (2010) published a systematic review on diagnostic tools for detecting cardiac arrhythmias.7 The literature search, conducted through March 2007, identified 28 studies for inclusion; 12 were single-arm studies and 16 were comparative studies. A meta-analysis was not possible due to the heterogeneity of the study populations and the devices tested. This review included studies of patients presenting with palpitations and compared the yield of remote monitoring for several classes of devices: Holter monitors, patient-activated event recorders, autotriggered event recorders, and implantable loop recorders (ILRs). The yield varied among devices, with autotrigger devices providing the highest range of detection (72%-80%), followed by patient-activated devices (17%-75%), and Holter monitors (33%-35%).

Randomized Controlled Trial
Steinhubl et al (2018) conducted an RCT comparing active home-based cardiac monitoring with the iRhythm Zio initiated immediately after study recruitment (n=1364) vs active monitoring after 4 months (n=1291).8 A cohort of patients (n=3476) without monitoring, matched by age, sex, and CHA2DS2VASc were part of a concurrent observational study. The primary endpoint was newly diagnosed AF at four months among those actively monitored at initiation vs those just beginning the monitoring. The secondary endpoint was newly diagnosied AF at one year among the actively monitored groups combined vs the matched observational controls. For the primary endpoint, at 4 months follow-up, 3.9% of the immediate group and 0.9% of the delayed group had newly diagnosed AF (absolute difference, 3.0%; 95% confidence interval [CI]: 1.8% to 4.1%). For the secondary endpoint, at 1 year followup, 6.7 per 100 person-years in the monitored group and 2.6 per 100 person-years in the control group had newly diagnosed AF. At one year, patients who were actively monitored were more likely to initiate anticoagulants, and have more cardiology visits and more primary care visits. There were no differences in emergency room visits or hospitalizations between the monitored and unmonitored groups after one year.

Observational Studies
Farris et al (2019) reviewed the records of patients who had undergone 30-day rhythm monitoring with the LifeWatch device at a single institution.9 A total of 3.4% of the patients had a new diagnosis of AF (402 per 1000 patient-years). The most common management response to the new diagnoses was to initiate anticoagulation therapy.

Turakhia et al (2013) evaluated the diagnostic yield of the Zio Patch.10 Data from the manufacturer were used to identify 26751 first-time users of the device. The most common clinical indications were palpitations (40.3%), AF (24.3%), and syncope (15.1%). Mean duration of use was 7.6 days, and 95.9% of patients wore the device for more than 48 hours. At least 1 episode of arrhythmia was detected in 16142 (60.3%) patients. The authors compared the detection rate in the first 48 hours with the detection rate over the entire time the device was worn, with 70.1% of patients having their arrhythmia detected within the first 48 hours and 29.9% having their first arrhythmia detected after the first 48 hours. The overall yield was significantly higher when comparing the total monitored period (62.2%) with the first 48 hours (43.9%; p<0.001). These data confirmed previous studies that had shown that while a substantial proportion of arrhythmias in symptomatic patients can be detected within a 48-hour period of monitoring, longer monitoring periods increase the detection rate.

Barrett et al (2014) compared arrhythmia detection rates in 146 patients who underwent simultaneous monitoring with a 24-hour Holter monitor and a 14-day Zio Patch monitor.11 Included were patients referred for evaluation of a suspected cardiac arrhythmia at a single institution. For the detection of atrioventricular block, sinus pause, polymorphic ventricular tachycardia, supraventricular tachycardia (SVT), or AF, Holter monitoring detected 61 arrhythmias, while the Zio Patch detected 96 (p<0.001). Over the monitoring period, the same 60 arrhythmia events were detected by both devices, with 36 only detected by the Zio Patch and 1 only detected by the Holter. The investigators conducted within-subject comparisons of arrhythmia detection for the 24-hour period during which both devices were worn. Holter monitoring detected 61 arrhythmia events compared with 52 detected by the Zio Patch (p=0.013). This study also suggested that extended monitoring may increase the diagnostic yield of cardiac monitoring. However, a relatively large number of missed events occurred with the Zio Patch during the period of simultaneous monitoring, which might have clinical significance if its performance is similar in nonresearch settings.

Solomon et al (2016) evaluated the diagnostic yield for potentially high-risk arrhythmias during 14 days of continuous recording with the Zio Patch among 122454 patients (122815 recordings) included in a manufacturer registry.12 Patients included in the series all underwent monitoring with the device from November 2011 to December 2013. Mean wear time was 9.6 days. Overall, there were 22443 (18%) patients with sustained ventricular tachycardia, 1766 (1.4%) patients with sinus pauses of 3 seconds or more, 521 (0.4%) patients with AF pauses of 3 seconds or more, 249 (0.2%) patients with symptomatic pauses, and 1468 (0.4%) with high-grade heart block, which were considered potentially high-risk arrhythmias. After 24 and 48 hours of monitoring, 52.5% and 65.5%, respectively, of potentially high-risk arrhythmias were detected. Seven days of monitoring identified 92.9% of potentially high-risk arrhythmias.

Bolourchi et al (2015) evaluated the diagnostic yield of 14 days of monitoring with the Zio Patch in a series of 3209 children included in a manufacturer registry.13 Patient age ranged from 1 month to 17 years. Indications for monitoring included palpitations (n=1138 [35.5%]), syncope (n=450 [14.0%]), unspecified tachycardia (n=291 [9.1%]), paroxysmal SVT (n=264 [8.2%]), and chest pain (n=261 [8.1%]). The overall prevalence of any arrhythmia was 12.1%, with 44.1% of arrhythmias occurring after the first 48 hours of monitoring. Arrhythmias were detected in 10.0% of patients referred for palpitations, 6.7% referred for syncope, 14.8% referred for tachycardia, 22.7% referred for paroxysmal SVT, and 6.5% referred for chest pain.

Single-center studies, summarized in Table 2, have reported on the diagnostic yield and timing of arrhythmia detection in patients monitored with the Zio Patch for a variety of arrhythmias. These studies generally have reported high rates of arrhythmia detection.

Table 2. Single-Center Studies Reporting on Zio Patch Yield  

Study

Population

Monitoring Indication

 

Main Findings

   

Indication

%

 

Eisenberg et al (2014)14

524 consecutive patients evaluated in an academic EP practice

  • Surveillance for unspecified arrhythmia or palpitations
  • Known/suspected AF
  • Syncope
  • Bradycardia surveillance
  • Tachycardia surveillance
  • Chest pain

47

30
8
4
5
2

  • Significant arrhythmias detected in 297 (57%)
  • 66% had 1st arrhythmia detected within 2 d of monitoring
  • 25% of patient-triggered events associated with clinically significant arrhythmias

Schreiber et al (2014)15

174 patients with symptoms suggestive of arrhythmia seen in an ED

  • Palpitations
  • Syncope
  • Unspecified arrhythmias detected in the ED

44.8
24.1
11.5

  • >1 significant arrhythmia other than chronic AF (≥4 beats VT, paroxysmal AF, ≥4 beats SVT, ≥3-s pause, 2nd-degree Mobitz II or 3rd-degree AV block, or symptomatic bradycardia) detected in 83 (47.7%)
  • Median time to arrhythmia detection:
    • Any arrhythmia: 1.0 d (IQR, 0.2-2.8 d)
    • VT: 3.1 d
    • Sinus pause: 4.2 d
    • Significant heart block: 5.8 d

AF: atrial fibrillation; AV: atrioventricular; ED: emergency department; EP: electrophysiology; IQR: interquartile range; SVT: supraventricular tachycardia; VT: ventricular tachycardia.  

Comparison of Devices
Eysenck et al (2019) compared 4 external cardiac monitors (Zio XT Monitor, NUUBO vest, Carnation Ambulatory Monitor, and Novacor R Test) with the gold standard of permanent pacemakers in the ability to detect AF.16 Patients who had permanent pacemakers (n=21) wore each of the external monitors for 2 weeks, in randomized order. A total of 1108 AF episodes were identified by the pacemakers during the study period. Results showed that the Zio, NUUBO, and Carnation monitors were more accurate in AF diagnosis compared with the Novacor R Test, when using the pacemaker detection episodes as the reference standard.

Health Quality Ontario (2017) published an assessment comparing long-term continuous AEMs with external cardiac loop recorders for detecting arrhythmias.17 The assessment included a systematic review of the literature on the effectiveness of both devices for detecting arrhythmias. No studies directly comparing long-term continuous AEMs with  external loop recorders (ELRs) were found, so indirect comparisons were constructed using 24-hour Holter monitors as the common comparator. Twelve cohort studies were included; seven addressed long-term AEMs and five addressed ELRs. Using a meta-regression model to control for variation in device-wearing time and baseline syncope rate, the estimated difference between the long-term continuous AEMs and ELRs in their ability to detect arrhythmias was small (risk difference, 0.01; 95%CI, -0.18 to 0.20). Both devices were more effective than a 24-hour Holter. However, the quality of evidence was evaluated as poor using GRADE criteria.

Some evidence suggests that autotriggered event monitors have an inherently higher yield than patient-activated AEMs. Several studies, including an analysis of a database of 100000 patients, have compared the diagnostic yield of automatic and patient-activated arrhythmia recordings and reported an improved yield with autotriggering devices.18-20

Clinically Useful
A test is clinically useful if the use of the results informs management decisions that improve the net health outcome of care. The net health outcome can be improved if patients receive correct therapy, or more effective therapy, or avoid unnecessary therapy, or avoid unnecessary testing.

Direct Evidence
Direct evidence of clinicalutility is provided by studies that have compared health outcomes for patients managed with and without the test. Because these are intervention studies, the preferred evidence would be from RCTs. No RCTs supporting clinical utility were identified.

Chain of Evidence
Indirect evidence on clinical utility rests on clinical validity. Clinical validity of long-term ambulatory monitoring in patients with arrhythmia symptoms was demonstrated in one RCT and in several large observational studies showing additional AF detection beyond the time frame of when a Holter monitor would be used (24 to 48 hours). When arrhythmia events are detected, management of patients typically involve antiarrhythmic or anticoagulant therapies, which are proven effective in stroke prevention. Therefore, longer term monitoring may improve health outcomes.

Section Summary: Autoactivated or Continuous Ambulatory Monitoring for Patients with Arrhythmia Symptoms
The available evidence on continuously worn cardiac monitors that can store data for longer periods of time than standard Holter monitors indicates that such devices typically detect greater numbers of arrhythmias during extended follow-up compared with 24- or 48-hour Holter monitoring. The RCT and several observational studies indicated that patients who had arrhythmias detected were more likely to receive anticoagulant therapy, antiarrhythmic therapy, and ablation or other cardiac procedures. Because these treatments have been proven effective for stroke prevention, it can be concluded that longer term monitoring of patients with cryptogenic stroke will improve outcomes.

LONG-TERM AMBULATORY CARDIAC MONITORING FOR PATIENTS WITH ATRIAL FIBRILLATION FOLLOWING ABLATION
Clinical Context and Test Purpose
All patients treated with ablation are given anticoagulation for up to three months postprocedure, with many patients remaining on long-term anticoagulation. In patients with an apparently successful ablation who do not show signs or symptoms of recurrent AF at time periods longer than three months postablation, a decision whether to continue treatment with anticoagulants needs to be made. Studies have demonstrated that late recurrences are not uncommon after ablation and that these recurrent episodes are often asymptomatic.21,22 However, the presence of recurrent episodes of AF is a predictor of future thromboembolic events. In a large observational study of 565 patients postablation, Chao et al (2011) found the 2 major predictors of thromboembolism were the CHADS2 score and the presence of recurrent episodes of AF.23

The purpose of AEMs (either patient-activated or continuous) in patients with AF following ablation is to provide an alternative detection method for recurrent AF in order to accurately assess the need for anticoagulation therapy.

The question addressed in this evidence review is: Does the use of AEMs (either patient-activated or continuous) improve the net health outcome of patients with AF following ablation compared with ECG only or 24- to 48-hour Holter monitoring?

The following PICOTS were used to select literature to inform this review.

Patients
The relevant population of interest are individuals with AF following ablation.

Interventions
The intervention being considered is patient- or autoactivated external event monitoring or continuous ambulatory event monitoring. Patient-activated devices are applied to the skin in the precordial area by the patient when symptoms are developing. Continuous event monitoring devices are recording activity continuously and can store data longer than the Holter monitor.

Comparators
Alternative surveillance methods that are used include an ECG or 24- to 48-hour Holter monitoring. An ECG provides information on cardiac electrical activity in one point in time. A Holter monitor is worn continuously and records cardiac electrical output continuously throughout the recording period. Holter monitors are capable of recording activity for 24 to 72 hours.

Outcomes
The general outcome of interest is diagnostic yield of the monitors in detecting arrhythmias. If arrhythmias do not recur following ablation, patients may consider discontinuing anticoagulation therapy.

Study Selection Criteria
For the evaluation of clinical validity of autoactivated or patient-activated external ambulatory event monitoring for patients with arrhythmia symptoms, studies that met the following criteria were considered: 

  • To assess the clinical validity, studies should report sensitivity, specificity, positive and negative predictive values. Alternatively, studies reporting on diagnostic yield are informative.
  • To assess the clinical utility, studies should demonstrate how results of the tests impacted treatment decisions and overall management of the patient.

Technically Reliable
Assessment of technical reliability focuses on specific tests and operators and requires review of unpublished and often proprietary information. Review of specific tests, operators, and unpublished data are outside the scope of this evidence review and alternative sources exist. This evidence review focuses on the clinical validity and clinical utility.

Clinically Valid
A test must detect the presence or absence of a condition, the risk of developing a condition in the future, or treatment response (beneficial or adverse).

Randomized Controlled Trial
In a prospective, randomized study, Kapa et al (2013) compared ILRs with conventional transtelephonic recorders in the assessment of arrhythmia burden after catheter ablation.24 Forty-four patients were enrolled and randomized; all patients received the ILR postablation. Six patients were excluded due to requests for device removal or loss to follow-up. During the first 6 months after ablation, all subjects underwent conventional monitoring that consisted of twice daily, 1-minute pulse rate assessments by the patient and 3, 30-day transtelephonic monitoring periods. At 6 months postablation, patients were allocated to the randomization arm (on a 1:1 basis at initial enrollment) of either the ILR (transmission of data every 31 days) or conventional monitoring (twice daily, 1-minute pulse rate assessment, 1 transtelephonic recording for 30 days at month 11). At 6 months postablation, conventional monitoring detected AF in 7 (18%) of 38 patients and the ILR confirmed AF in all of these patients. ILR monitoring also detected AF in an additional 11 (29%) patients. During the subsequent 6-month period, 5 of 18 patients in the conventional monitoring arm refused ongoing monitoring due to discomfort and lifestyle restrictions; of the remaining 13, 5 (38%) had a recurrence of AF. In the ILR group, 5 (25%) of 20 patients had recurrence of AF. During the randomization period, 71% of patients in the ILR group discontinued their antiarrhythmic drugs compared with 44% in the conventional monitoring group over the randomization period (p=0.04).

Observational Study
Reporting on the prospective Discerning Symptomatic and Asymptomatic Episodes Pre- and Post-Radiofrequency Ablation of AF study, Verma et al (2013) evaluated the incidence of asymptomatic AF episodes for 3 months before and 18 months after ablation in 50 patients implanted with a cardiac monitor.25  Patients were instructed to keep a standardized diary record of arrhythmia symptoms. Asymptomatic AF recurrences were defined as implantable cardiac monitor (ICM) events lasting two minutes or longer, without a corresponding diary entry. Based on diary reporting of symptoms, 29 (58%) of 50 patients were arrhythmia-free after ablation; based on monitor recordings from intermittent (every 3 month) ECG or Holter monitor, 28 (56%) patients were arrhythmia-free postablation. Patient detection of symptoms underestimates the AF occurrence rate following ablation, with 12% of patients having arrhythmias that were only detected through monitoring.

Clinically Useful
A test is clinically useful if the use of the results informs management decisions that improve the net health outcome of care. The net health outcome can be improved if patients receive correct therapy, or more effective therapy, or avoid unnecessary therapy, or avoid unnecessary testing.

Direct Evidence
Direct evidence of clinical utility is provided by studies that have compared health outcomes for patients managed with and without the test. Because these are intervention studies, the preferred evidence would be from RCTs. No RCTs were identified. Below is an observational study providing indirect evidence.

Several observational studies have followed patients who stopped anticoagulation after a comprehensive evaluation, which included ambulatory monitoring, that indicated the patient had a low-risk for recurrent episodes. These patients experienced a low subsequent rate of thromboembolic events. In 1 study, Themistoclakis et al (2010) evaluated 3355 patients from 5 clinical centers, of whom 2692 discontinued anticoagulation at 3 to 6 months postablation and 663 continued anticoagulation medication.26 During a mean follow-up of 28 months, 2 (0.07%) patients who discontinued anticoagulation experienced an ischemic stroke. This rate did not differ significantly from the stroke rate in patients who continued anticoagulation (0.45%). In addition, the adverse event rate of major hemorrhage was lower for patients who discontinued anticoagulation (0.04%) compared with those who continued (2%; p<0.001).

Chain of Evidence
Indirect evidence on clinical utility rests on clinical validity. An RCT and observational studies have shown that ambulatory monitoring was able to detect AF recurrences that were not detectable based on symptoms alone. No RCTs were identified that compared health outcomes for patients managed with and without ambulatory monitoring. However, there is a large observational study demonstrating that following ablation and a comprehensive evaluation including ambulatory monitoring that indicates a patient is low-risk, patients may consider discontinuing anticoagulation therapy. Patients who discontinued anticoagulation therapy following ablation experienced comparably low rates of stroke compared with patients remaining on anticoagulation therapy, and had statistically lower occurrences of major hemorrhage.

Section Summary: Long-term Ambulatory Monitoring for Patients with AF Following Ablation
Evidence includes an RCT and several observational studies that make a strong indirect argument that long-term monitoring for asymptomatic episodes of AF with AEMs will lead to changes in management with long-term anticoagulation. One study reported that patients who discontinued anticoagulation therapy after ambulatory monitoring was negative for recurrent episodes, experienced a low rate of stroke similar to patients who remained on anticoagulation therapy. In addition, patients discontinuing anticoagulants experienced fewer major hemorrhages. These changes in management based on ambulatory monitoring are likely to improve outcomes. Because different long-term monitoring devices were used across the studies, the specific type of monitoring associated with the best outcomes is not established.

LONG-TERM AMBULATORY CARDIAC MONITORING FOR PATIENTS WITH CRYPTOGENIC STROKE
Clinical Context and Test Purpose
Approximately 5% of patients with cryptogenic stroke will have AF diagnosed on ECG and/or telemetry monitoring in the hospital. Patients with a history of cryptogenic stroke who have had AF detected, are typically treated with anticoagulants. Studies comparing the use of continuous telemetry monitoring at the bedside with Holter monitoring for patients hospitalized for stroke or transient ischemic attack (TIA) have reported inconclusive results as to which is the preferred method for AF detection.27,28 Longer term ambulatory event monitoring has been shown to identify additional patients with asymptomatic episodes, with rates of detection estimated at 6% to 26% of patients.5,29,30

The purpose of long-term ambulatory cardiac monitoring in patients who have a history of cryptogenic stroke is to provide an alternative detection method for AF in order to accurately inform the decision to receive anticoagulation therapy.

The question addressed in this evidence review is: Does the use of longterm ambulatory cardiac event monitoring improve the net health outcome in patients with cryptogenic stroke compared with standard evaluation for stroke, including ECG and 24-hour Holter monitoring?

The following PICOTS were used to select literature to inform this review.

Patients
The relevant population of interest are individuals with a history of cryptogenic stroke with negative standard workup for AF.

Interventions
The intervention being considered is patient- or auto-activated external event monitoring or continuous ambulatory event monitoring. Patient-activated devices are applied to the skin in the precordial area by the patient when symptoms are developing. Continuous event monitoring devices are worn continuously and are recording activity continuously and can store data longer than the Holter monitor.

Comparators
The comparator is standard evaluation for stroke, including ECG or 24- to 48-hour Holter monitoring. An ECG provides information on cardiac electrical activity in one point in time. A Holter monitor is worn continuously and records cardiac electrical output continuously throughout the recording period. Holter monitors are capable of recording activity for 24 to 72 hours.

Outcomes
The general outcome of interest is diagnostic yield of the monitors in detecting arrhythmias. Accurate detection of arrhythmias may be used to inform management decisions concerning anticoagulation therapy.

Study Selection Criteria
For the evaluation of clinical validity of autoactivated or patient-activated external ambulatory event monitoring for patients with arrhythmia symptoms, studies that met the following criteria were considered:  

  • To assess the clinical validity, studies should report sensitivity, specificity, positive and negative predictive values. Alternatively, studies reporting on diagnostic yield are informative.
  • To assess the clinical utility, studies should demonstrate how results of the tests impacted treatment decisions and overall management of the patient. 

Technically Reliable
Assessment of technical reliability focuses on specific tests and operators and requires review of unpublished and often proprietary information. Review of specific tests, operators, and unpublished data are outside the scope of this evidence review and alternative sources exist. This evidence review focuses on the clinical validity and clinical utility.

Clinically Valid
A test must detect the presence or absence of a condition, the risk of developing a condition in the future, or treatment response (beneficial or adverse). Below are systematic reviews and RCTs providing evidence for the clinical validity of long-term ambulatory monitoring of patients with cryptogenic stroke.

Systematic Reviews
Sposato et al (2015) conducted a systematic review and meta-analysis of studies assessing rates of newly diagnosed AF after cryptogenic stroke or TIA based on cardiac monitoring, stratified into 4 sequential screening phases: phase 1 (emergency department) consisted of admission ECG; phase 2 (in-hospital) comprised serial ECG, continuous inpatient ECG monitoring, continuous inpatient cardiac telemetry, and in-hospital Holter monitoring; phase 3 (first ambulatory period) consisted of ambulatory Holter monitoring; and phase 4 (second ambulatory period) consisted of mobile cardiac outpatient telemetry (MCOT), external loop recording, and implantable loop recording.31 In total, 50 studies with 11658 patients met the inclusion criteria. Studies were mixed in their patient composition: 22 (28%) included only cryptogenic stroke cases, 4 (5%) stratified events into cryptogenic and noncryptogenic, and 53 (67%) included unselected patient populations. The proportion of patients diagnosed with poststroke AF during the ambulatory phases was 10.7% (95% CI, 5.6% to 17.2%) in phase 3, and 16.9% (95% CI, 13.0% to 21.2%) in phase 4. The overall AF detection yield after all phases of sequential cardiac monitoring was 23.7% (95% CI, 17.2% to 31.0%). In phase 4, there were no differences between the proportion of patients diagnosed with poststroke AF by MCOT (15.3%; 95% CI, 5.3% to 29.3%), ELR (16.2%; 95% CI, 0.3% to 24.6%), or ILR (16.9%; 95% CI, 10.3% to 24.9%; p=0.97).

Kishore et al (2014) conducted a systematic review and meta-analysis of prospective observational studies and RCTs that have reported detection rates of newly diagnosed AF in patients with ischemic stroke or TIA who had had any cardiac monitoring for at least 12 hours.32  Thirty-two studies were selected: 18 studies included patients with ischemic stroke only, 1 study included TIA only, and 13 studies included both ischemic stroke and TIA. Reviewers reported significant study heterogeneity. Among unselected patients (ie, selected on the basis of stroke pathogenesis, age, or prescreening for AF), the detection rate of any new AF was 6.2% (95% CI, 4.4% to 8.3%); among selected patients, it was 13.4% (95% CI, 9.0% to 18.4%). In cryptogenic strokes, new AF was detected in 15.9% of patients (95% CI, 10.9% to 21.6%). Among selected patients, the AF detection rate during 24-hour Holter monitoring was 10.7% (95% CI, 3.4% to 21.5%), while the detection rate during monitoring beyond 24 hours (including more prolonged Holter monitoring, implantable and nonimplantable loop recording, and MCOT) was 14.7% (95% CI, 10.7% to 19.3%).

The Kishore et al (2014) study and others suggest that longer periods of cardiac monitoring increase the likelihood of AF detection. However, many of these asymptomatic episodes of AF are brief and their relation to the preceding stroke uncertain. The ideal study to evaluate the role of cardiac monitoring in the management of patients with cryptogenic stroke would be trials that randomize patients to a strategy involving event monitoring or routine care with evaluation of rates of detection of AF and stroke-related outcomes.

Randomized Controlled Trials
Four RCTs were identified that evaluated ambulatory monitoring in patients with cryptogenic stroke. Two were small pilot trials. One small pilot RCT published by Kamel et al (2013) randomized 40 patients with cryptogenic ischemic stroke or high-risk TIA to usual care or to 21 days of MCOT (Table 3).33 There were no cases of AF detected in either group (Table 4).

A second small pilot trial published by Higgins et al (2013) randomized 100 patients with ischemic stroke and no history of AF presenting within 7 days of a cryptogenic ischemic stroke to either standard care, which included 12-lead ECG, 24-hour Holter monitoring, and/or echocardiography, at the discretion of the treating practitioner, or to standard care plus cardiac event monitoring with Novacor R-test Evolution 3, an ELR device (Table 3).34 Sustained AF (recorded for the complete 20-second rhythm strip after event triggering) was detected significantly more often with the ELR than with standard care at 14-day follow-up. The difference did not differ statistically at 90-day follow-up (Table 4).

Sanna et al (2014) reported on results from the Cryptogenic Stroke and underlying times Fibrillation (CRYSTAL AF) trial, an RCT that evaluated whether long-term monitoring with ICMs in patients who had cryptogenic stroke would lead to changes in anticoagulant management and/or improved outcomes (Table 3).35,36 The trial randomized 441 patients to continuous monitoring with the Reveal XT ICM or routine care. Eligibility criteria included no known history of AF, cryptogenic stroke, or TIA with infarct, and no mechanism determined after a workup that included 12-lead ECG, 24-hour Holter monitoring, transesophageal echocardiography, CT or magnetic resonance angiography of the head and neck, and hypercoagulability screening (for patients <55 years old). Analysis was intention-to-treat. Of the 441 patients randomized, 416 (94.3%) completed 6-month follow-up, 2 were lost to follow-up, 5 died, and 18 exited the trial before 6 months. Crossover occurred in 12 patients in the ICM group and 6 in the control group. AF was detected in 8.9% of the ICM group compared with 1.4% of the control group (hazard ratio [HR], 6.43; 95% CI, 1.90 to 21.74) (Table 4). Median time from randomization to detection of AF was 41 days (interquartile range, 14-84 days) in the ICM group and 32 days (interquartile range, 2-73 days) in the control group. Most AF episodes in the ICM group were asymptomatic (74%) compared with 33% in the control group. The rate of AF detection was similarly greater in the ICM group at the 12-month follow-up (Table 4). A majority of patients who had AF detected were prescribed anticoagulation therapy. Five (2.4%) of the 208 ICM inserted were removed due to infection or erosion of the device pocket.

Brachmann et al (2016) reported 3-year follow-up results from the CRYSTAL AF trial.37 At trial closure, 48 subjects had completed 3 years of follow-up (n=24 in each treatment group). By 3 years, the HR for detecting AF for ICM-monitored vs control patients was 8.8 (95% CI, 3.5 to 22.2; p<0.001).

Gladstone et al (2014) reported results from the Atrial Fibrillation in Patients with Cryptogenic Stroke study, an RCT that compared 30-day autotriggered external loop cardiac event monitors with conventional 24-hour monitors for the detection of AF in patients with cryptogenic stroke (Table 3).38 Patients were ages 55 years or older, with no known history of AF, and an ischemic stroke or TIA of undetermined cause within the prior 6 months. All patients underwent standard screening for AF with 1 or more ECGs and 1 or more 24-hour Holter monitors. In total, 572 patients were randomized to an ELR (ER910AF Cardiac Event Monitor, Braemar) or to a 24-hour Holter monitor. Among intervention group subjects, 82% completed at least 3 weeks of monitoring. AF was detected in 45 (16.1%) of 280 patients in the intervention group compared with 9 (3.2%) of 277 patients in the control group (risk difference, 12.9 percentage points; 95% CI, 8.0 to 17.6; p<0.001) (Table 4). At 90-day followup, patients in the intervention group (18.6%) were more likely to be treated with anticoagulants than those in the control group (11.1%; absolute treatment difference, 7.5 percentage points; 95% CI, 1.6 to 13.3; p=0.01).

Table 3. Summary of RCT Characteristics for AEM for Cryptogenic Stroke  

 

Interventions (n) 

Study

Country

Sites

Dates

Participants

Active

Comparator

Kamel et al (2013)33

U.S.

1

2009-2011

Cryptogenic ischemic stroke or high-risk TIA

MCOT (20)

Standard (20)

Higgins et al (2013)34

U.K.

2

2010-2011

Transient or persistent symptoms of acute TIA

ELR (50)

Standard (50)

Sanna et al (2014)36 & Brachmann et al (2016)37

Canada, Europe, U.S.

55

2009-2012

Cryptogenic ischemic stroke or TIA

ILR (221)

Standard (220)

Gladstone et al (2014)38

Canada

16

NR

Cryptogenic ischemic stroke or TIA

ELR (280)

Standard (277)

AEMs: ambulatory event monitors; ELR: external loop recorder; ILR: implantable loop recorder; MCOT: mobile cardiac outpatient telemetry; NR: not reported RCT: randomized controlled trial; TIA: transient ischemic attack.

Table 4. Summary of RCT Results for AEMs for Cryptogenic Stroke

Study

FU

AF Detection

Additional Findings

   

AEM, %

Standard, %

p

 

Kamel et al (2013)33

90 d

0

0

NS

  • MCOT identified atrial tachycardia in 2 patients (1 incorrectly labeled as AF by telemetry software)
  • MCOT identified 2 nonsustained ventricular tachycardia

Higgins et al (2013)34

14 d

90 d

18

22

2

8

<0.05

0.09

No difference between groups for recurrent stroke, TIA, or mortality

Sanna et al (2014)36; Brachmann et al (2016)37

6 mo

12 mo

3 y

8.9

12.4

30

1.4

2.0

3.0

<0.001

<0.001

<0.001

  • Percent patients on oral anticoagulation therapy significantly higher in ILR group vs standard group
  • At 3-y followup, recurrent stroke or TIA occurred in 20 patients in ILR group and in 24 in standard group

Gladstone et al (2014)38

90 d

16.1

3.2

<0.001

Atrial premature beats was identified in a regression model as a potential predictor of AF detection

AEM: ambulatory event monitor; AF: atrial fibrillation; FU: follow-up; ILR: implantable loop recorder; MCOT: mobile cardiac outpatient telemetry; RCT: randomized controlled trial; TIA: transient ischemic attack.

Nonrandomized Studies
Nonrandomized and noncomparative studies published before the RCTs described above have reported on AF detection rates after cryptogenic stroke and long-term monitoring with various devices, including ILRs,6,39,40 and continuous monitors with longer recording periods,41 along with a pilot study evaluating the Zio Patch for AF detection poststroke.42

Clinically Useful
A test is clinically useful if the use of the results informs management decisions that improve the net health outcome of care. The net health outcome can be improved if patients receive correct therapy, or more effective therapy, or avoid unnecessary therapy, or avoid unnecessary testing.

Direct Evidence
Direct evidence of clinical utility is provided by studies that have compared health outcomes for patients managed with and without the test. Because these are intervention studies, the preferred evidence would be from RCTs. No RCTs were identified demonstrating clinical utility.

Chain of Evidence
Indirect evidence on clinical utility rests on clinical validity. Clinical validity of long-term ambulatory monitoring in patients with cryptogenic stroke has been demonstrated in systematic reviews and RCTs that showed higher rates of AF detection with long-term monitoring. Because most patients with a history of stroke who have AF detected will be treated with anticoagulation, and because anticoagulation is an effective treatment for stroke prevention, it can be concluded that longer term monitoring of patients with cryptogenic stroke will improve outcomes.

Section Summary:  Long-term Ambulatory Cardiac Monitoring for Patients With Cryptogenic Stroke 
Randomized studies, including two large RCTs, have demonstrated that long-term monitoring is associated with higher rates of AF detection compared with Holter monitors among patients with cryptogenic stroke. Because most patients with a history of stroke who have AF detected will be treated with anticoagulation, and because anticoagulation is an effective treatment for stroke prevention, it can be concluded that longer term monitoring of patients with cryptogenic stroke will improve outcomes. Because different long-term monitoring devices were used across the studies, the specific type of monitoring associated with the best outcomes is not established.

LONG-TERM AMBULATORY CARDIAC MONITORING FOR ASYMPTOMATIC PATIENTS
Clinical Context and Test Purpose
Screening for AF in asymptomatic patients has been proposed to reduce burden of stroke. Evaluating the net benefit of screening for AF in asymptomatic patients requires considering: risk of stroke in absence of screening; incremental benefit of earlier vs later treatment for stroke when AF is detected; and potential harms of overdiagnosis.

Assessing the prevalence of asymptomatic AF is difficult because of the lack of symptoms. Approximately a third of all patients with AF are estimated to be asymptomatic.43 Studies have suggested that most paroxysmal episodes of AF are asymptomatic.44,45 It is uncertain whether patients with paroxysmal AF have a stroke risk comparable to those with persistent or permanent AF; some studies have suggested the risk of stroke is similar46,47 while in a systematic review of 12 studies (total n=99996 patients), Ganesan et al (2016) found that the risks of thromboembolism and all-cause mortality were higher with nonparoxysmal than with paroxysmal AF.48 The clinical management of symptomatic and asymptomatic AF is the same. Anticoagulation should be initiated if reduction in risk of embolization exceeds complications due to increased bleeding risk.

Screening for AF in asymptomatic patients could be either systematic or targeted to high-risk populations. European guidelines for screening for AF are based on a larger cluster RCT (Fitzmaurice et al [2007];n=14802) of opportunistic pulse taking vs systematic screening with 12 lead ECG or standard care in general practice.49 This RCT showed that systematic and opportunistic screening detected similar rates of AF and both were superior to standard care. The mechanisms of how and when to screen for AF in unselected populations have not been well-studied.

The purpose of long-term ambulatory cardiac monitoring in patients who are asymptomatic with risk factors for AF is to provide an alternative method of detecting AF.

The question addressed in this evidence review is: Does the use of long-term ambulatory cardiac monitoring in patients who are asymptomatic with risk factors for AF improve net health outcome compared with no additional evaluation or standard of care?

The following PICOTS were used to select literature to inform this review.

Patients
The relevant population of interest are asymptomatic individuals with risk factors for AF.

Interventions
The intervention being considered is patient- or autoactivated external event monitoring or continuous ambulatory event monitoring. Patient-activated devices are applied to the skin in the precordial area by the patient when symptoms are developing. Continuous event monitoring devices are worn continuously and are recording activity continuously and can store data longer than the Holter monitor.

Comparators
The comparators are no additional evaluation or standard care. Standard care may include an ECG and/or pulse palpation.

Outcomes
The general outcome of interest is diagnostic yield of the monitors in detecting arrhythmias. Accurate detection of arrhythmias may be used to inform management decisions of the asymptomatic patients.

Study Selection Criteria
For the evaluation of clinical validity of autoactivated or patient-activated external ambulatory event monitoring for patients with arrhythmia symptoms, studies that met the following criteria were considered:

  • To assess the clinical validity, studies should report sensitivity, specificity, positive and negative predictive values. Alternatively, studies reporting on diagnostic yield are informative.
  • To assess the clinical utility, studies should demonstrate how results of the tests impacted treatment decisions and overall management of the patient.

Technically Reliable
Assessment of technical reliability focuses on specific tests and operators and requires review of unpublished and often proprietary information. Review of specific tests, operators, and unpublished data are outside the scope of this evidence review and alternative sources exist. This evidence review focuses on the clinical validity and clinical utility.

Clinically Valid
A test must detect the presence or absence of a condition, the risk of developing a condition in the future, or treatment response (beneficial or adverse).

Randomized Controlled Trials
Halcox et al (2017) conducted an RCT, REmote HEArt Rhythm Sampling using the AliveCor heart monitor to scrEen for Atrial Fibrillation , which screened patients for AF using the AliveCor Kardia monitor (n=500) or routine care (n=501).50 Patients were 65 years and older, asymptomatic, with CHA2DS2-VASc scores of 2 or higher. Patients randomized to the Kardia monitor arm undertook twice-weekly, 30-second single-lead iECG recordings and uploaded the information to a secure server. Analysis was performed using an automated software system and forwarded to a physiologist reading service. Abnormal ECG readings were sent to cardiologists. Appropriate care was arranged when arrhythmias were detected. Patients in the routine care arm were followed by their general practitioners. All patients were contacted at 12, 32, and 52 weeks. At 52-week follow-up, 19 patients in the Kardia monitor arm and 5 patients in the routine care arm were diagnosed with AF (HR=3.9; 95% CI, 1.4 to 10.4; p=0.007). There were no significant differences in the rates of mortality; stroke, TIA, or spontaneous embolism; deep vein thromboembolism or pulmonary embolism; or other cardiovascular events between groups.

Observational Study
Turakhia et al (2015) reported on results for a single-center noncomparative study evaluating the feasibility and diagnostic yield of a continuous recording device with longer recording period (Zio Patch) for patients with risk factors for AF.51 The study included 75 patients older than age 55 with at least 2 risk factors for AF (coronary disease, heart failure, hypertension, diabetes, or sleep apnea), without a history of prior AF, stroke, TIA, implantable pacemaker or defibrillator, or palpitations or syncope in the prior year. Of the 75 subjects, 32% had a history of significant valvular disease and 9.3% had prior valve replacement. Most subjects (97%) were considered at moderate- to high-risk of stroke (CHA2DS2-VASc scores ≥2). After a mean follow-up of 7.6 days, AF was detected in 4 (5.3%) subjects, all of whom had CHA2DS2-VASc scores of 2 or greater. All patients with AF detected had an initial episode within the first 48 hours of monitoring. Five patients had detected episodes of atrial tachyarrhythmias lasting at least 60 seconds.

Clinically Useful
A test is clinically useful if the use of the results informs management decisions that improve the net health outcome of care. The net health outcome can be improved if patients receive correct therapy, or more effective therapy, or avoid unnecessary therapy, or avoid unnecessary testing.

Direct Evidence
Direct evidence of clinicalutility is provided by studies that have compared health outcomes for patients managed with and without the test. Because these are intervention studies, the preferred evidence would be from RCTs. No studies were identified providing direct evidence.

Chain of Evidence
Indirect evidence on clinical utility rests on clinical validity. If the evidence is insufficient to demonstrate test performance, no inferences can be made about clinical utility. No studies were identified which demonstrate that detecting AF in asymptomatic individuals will change management of these patients.

Section Summary: Long-term Ambulatory Cardiac Monitoring for Asymptomatic Patients
For the use of ambulatory monitoring in the diagnosis of AF in asymptomatic but higher risk patients, a small noncomparative study demonstrated that monitoring with the Zio Patch for a mean of eight days resulted in a small percentage (5%) of AF detection. An RCT reported that the Kardia monitor detected more arrhythmias than routine care. However, none of these studies evaluating asymptomatic patients determined whether these measurements changed patient management. The RCT, which followed patients for one year, did not detect a difference in health outcomes between patients monitored using Kardia or routine care. The use of population-based screening for asymptomatic patients is not well-established, and several studies are underway to evaluate population-based screening and may influence the standard of care for AF detection in those without symptoms or a history of stroke or TIA. To determine whether outcomes are improved for ambulatory monitoring for AF in patients without a history of stroke or TIA or treated AF, studies comparing the outcomes for various outpatient diagnostic screening strategies for AF would be needed.

IMPLANTABLE LOOP RECORDERS FOR PATIENTS WITH SYMPTOMS OF ARRHYTHMIA
Clinical Context and Test Purpose
This section discusses the use of ILR, with a focus on clinical situations when use of an ILR at the beginning of a diagnostic pathway is indicated. It is expected that a longer period of monitoring with any device category is associated with a higher diagnostic yield. A progression in diagnostics, from an external event monitor to ILR, in cases where longer monitoring is needed is considered appropriate. However, there may be situations where it is sufficiently likely that long-term monitoring will be needed and that an ILR as an initial strategy may be reasonable.

The purpose of ILRs in patients with signs or symptoms suggestive of arrhythmia with infrequent symptoms is to provide an alternative method of arrhythmia detection.

The question addressed in this evidence review is: Does the use of ILRs in individuals with signs or symptoms suggestive of arrhythmia with infrequent symptoms improve net health benefits compared with no additional evaluation, standard care, or external AEMs?

The following PICOTS were used to select literature to inform this review.

Patients
The relevant population of interest are individuals with signs or symptoms suggestive of arrhythmia with infrequent symptoms.

Interventions
The intervention of interest is an ILR. ILRs store electrical cardiac activity data. When activated (by patient or automatically), the cardiac activity is recorded from the memory loop. ILRs are implanted under the skin in the precordial area.

Comparators
Comparators of interest include no additional evaluation, standard care, or external AEMss. External AEMs may be patient- or autoactivated. Patient-activated devices are applied to the skin in the precordial area by the patient when symptoms are developing. Continuous event monitoring devices are worn continuously and are recording activity continuously, storing data longer than the Holter monitor.

Outcomes
The general outcome of interest is diagnostic yield of the ILRs in detecting arrhythmias. Accurate detection of arrhythmias may be used to inform management decisions of the individuals with infrequent symptoms.

Study Selection Criteria
For the evaluation of clinical validity of autoactivated or patient-activated external ambulatory event monitoring for patients with arrhythmia symptoms, studies that met the following criteria were considered:

To assess the clinical validity, studies should report sensitivity, specificity, positive and negative predictive values. Alternatively, studies reporting on diagnostic yield are informative.

To assess the clinical utility, studies should demonstrate how results of the tests impacted treatment decisions and overall management of the patient.

Technically Reliable
Assessment of technical reliability focuses on specific tests and operators and requires review of unpublished and often proprietary information. Review of specific tests, operators, and unpublished data are outside the scope of this evidence review and alternative sources exist. This evidence review focuses on the clinical validity and clinical utility.

Clinically Valid
A test must detect the presence or absence of a condition, the risk of developing a condition in the future, or treatment response (beneficial or adverse).

Systematic Reviews
Solbiati et al (2017) conducted a systematic review and meta-analysis on the diagnostic yield of ILRs in patients with unexplained syncope.52 The literature search, conducted through November 2015, identified 49 studies, published between 1998 and 2015, enrolling a total of 4381 patients. The methodologic quality of the studies was assessed using QUADAS and QUADAS-2. The diagnostic yield of ILR, defined as the proportion of patients in which ILR was useful in determining a syncope diagnosis was 44% (95% CI, 40% to 48%; I2=80%). Diagnoses included arrhythmic syncope, ventricular arrhythmia, supraventricular arrhythmia, and bradyarrhythmia. Reviewers noted that an important analytic limitation was the considerable heterogeneity among studies, partly because definitions of syncope and methods to assess unexplained syncope were inconsistent.

Burkowitz et al (2016) conducted a systematic review and meta-analysis of ILRs in the diagnosis of syncope and the detection of AF.53 For syncope diagnosis, the review identified three RCTs comparing ILRs with a conventional diagnosis strategy (Holter monitoring). In pooled analysis, an ILR diagnosis strategy was associated with a higher likelihood of the endpoint of diagnostic yield (relative risk, 4.17; 95% CI, 2.57 to 6.77; I2=14%). The RCTs (Da Costa et al [2013],54 Farwell et al [2004],55 and Krahn et al [2001]56) are described below.

Afzal et al (2015) reported on a systematic review and meta-analysis of studies comparing ILRs with wearable AEMs for prolonged outpatient rhythm monitoring after cryptogenic stroke.57 Reviewers included 16 studies (total n=1770 patients)-3 RCTs and 13 observational studies. For ILR-monitored patients, the median monitoring duration was 365 days (range, 50-569 days), while for wearable device-monitored patients, the median monitoring duration was 14 days (range, 4-30 days). Compared with wearable AEMs, ILRs were associated with significantly higher rates of AF detection (23.3% vs 13.6%; odds ratio, 4.54; 95% CI, 2.92 to 7.06; p<0.05).

Randomized Controlled Trials
Podoleanu et al (2014) reported on results of an open-label RCT comparing 2 strategies for evaluating syncope---an experimental strategy involving the early use of an ILR and a conventional evaluation strategy excluding an ILR (see Table 5).58 The trial included patients who had a single syncope (if severe and recent) or at least 2 syncopes in the past 12 months. The syncope had to be unexplained at the end of clinical examination and who had a workup with 12-lead ECG, echocardiography, and head-up tilt-test. Patients randomized to ILR received the Reveal or Reveal Plus device. After 14 months of follow-up, a definitive cause of syncope was established more frequently in the ILR group than in the standard care group (see Table 6). Arrhythmic causes of syncope in the ILR group included two (5%) cases of atrioventricular block, four (10%) cases of sinus node disease, one (2.5%) case of AF, one (2.5%) case of ventricular fibrillation, and three (8%) other tachycardias. In the conventionally managed group, eight patients had a diagnosis of presumed reflex syncope.

Da Costa et al (2013) compared use of an ILR with a conventional follow-up strategy in 78 patients with a first episode of syncope (Table 5).54 A significant number of patients had cardiomyopathy (23%), AF (15.4%), and/or bundle branch block (58%) on ECG. Twenty-one (27%) patients had at least 1 arrhythmia detected, with a significant difference in the detection rate for the ILR group compared with the conventional follow-up group (see Table 6).

Giada et al (2007) conducted an RCT assessing 2 diagnostic strategies in 50 patients with infrequent (≤1 episode per month) unexplained palpitations-an ILR strategy (n=26) and a conventional strategy (n=24) including 24-hour Holter, 4 weeks of ambulatory ECG monitoring with an external recorder, and an electrophysiologic study if the 2 prior evaluations were negative) (see Table 5).59 Prior cardiac evaluation in eligible patients included standard ECG and echocardiography. Rhythm monitoring was considered diagnostic when a symptom-rhythm correlation was demonstrated during spontaneous palpitations that resembled pre-enrollment symptoms. In the conventional strategy group, a diagnosis was made in 5 (21%) subjects, after a mean time to diagnosis of 36 days, based on external ECG monitoring in 2 subjects and electrophysiologic studies in 3 subjects. In the ILR group, a diagnosis was made in 19 subjects after a mean time to diagnosis of 279 days (Table 6).

Farwell et al (2004) reported on an RCT comparing the diagnostic yield of an ILR (Reveal Plus) with a conventional diagnostic strategy in 201 patients with unexplained syncope (Table 5).55 Eligible patients were evaluated at a single institution for recurrent syncope and had no definitive diagnosis after a basic initial workup (including 12-lead ECG, Holter monitoring in patients with suspected cardiac syncope, upright cardiac sinus massage, and tilt-table testing). At last follow-up, more loop recorder patients had an ECG diagnosis than control patients (HR for ECG diagnosis, 8.93; 95% CI, 3.17 to 25.19; p<0.001) (see Table 6). Seven of the loop recorder patients were diagnosed with the device's autotrigger feature. In the loop recorder group, 34 patients had an ECG-directed therapy initiated (vs 4 in the control group; HR=7.9; 95% CI, 2.8 to 22.3). No device-related adverse events were reported.

An earlier RCT by Krahn et al (2001) compared a conventional monitoring strategy (ELR monitoring for 2-4 weeks, followed by tilt-table and electrophysiologic testing) with at least 1 year of monitoring using an ILR in 60 subjects with unexplained syncope (n=30 per group) (Table 5).56 Eligible patients had a previous clinical assessment, at least 24 hours of continuous ambulatory monitoring or inpatient telemetry, and a transthoracic echocardiogram. A diagnosis was made in 20% of those in the conventional monitoring arm and in 52% of those in the ILR arm (see Table 6).

Table 5. Summary of RCT Characteristics for ILRs for Arrhythmia 

         

Interventions (n)

Study

Country

Sites

Dates

Participants

Active

Comparator

Podoleanu et al (2014)58

France

13

2004-2008

Single recent syncope or 2 in past 12 mo

ILR (39)

Standard (39)

Da Costa et al (2013)54

France

Multiple, NS

2005-2010

Single syncope

ILR (41)

Standard (37)

Giada et al (2007)59

Italy

Multiple, NS

NR

Unexplained palpitations

ILR (26)

Standard (24)

Farwell et al (2004)55

England

1

2000-2001

≥2 unexplained syncope in past 12 mo

ILR (103)

Standard (98)

Krahn et al (2001)56

England

1

NR

Single or recurrent unexplained syncope

ILR (27)

ELR (30)

ELR: external loop recorder; ILR: implantable loop recorder; NR: not reported; NS: not specified; RCT: randomized controlled trial.

Table 6. Summary of RCT Results for ILRs for Arrhythmia

Study

FU

Diagnosis Made, n (%)

Additional Findings

   

ILR

Standard

p

 

Podoleanu et al (2014)58

14 mo

18 (46)

2 (5)

<0.001

  • Advanced cardiology tests performed less frequently in ILR group vs standard (p=0.05)
  • No difference in QOL

Da Costa et al (2013)54

27 moa

15 (37)

4 (11)

0.02

Earlier diagnosis in ILR group permitted earlier pacemaker implantation. However, earlier implantation did not improve survival (potentially due to small sample)

Giada et al (2007)59

≥12 mo

19 (73)

5 (21)

<0.001

9 of 19 patients with negative results with standard care crossed over to ILR and 6 of them received a diagnosis

Farwell et al (2004)55

≥6 mo

34 (33)

4 (4)

<0.001

  • ECG-directed therapy was initiated quicker in the ILR group
  • No difference in syncopal episodes, mortality, or QOL

Krahn et al (2001)56

12 mo

14 (52)

6 (20)

0.012

  • Crossover offered to patients with negative results
  • 1 of 6 switching to ELR was diagnosed and 8 of 13 switching to ILR was diagnosed (p=0.07)

ECG: electrocardiogram; FU: follow-up; ILR: implantable loop recorder; QOL: quality of life; RCT: randomized controlled trial.
Mean.

Observational Studies
Several observational studies compared the diagnostic yield of ICMs to the Holster monitor. Other observational studies reported management outcomes following diagnoses, such as anticoagulation initiation or cardiac procedures.

Magnusson et al (2018) presented outcomes in a cohort of patients who received ILRs (n=173).60 Among the patients evaluated for syncope, 39 (27%) were diagnosed with an arrhythmia necessitating a pacemaker and 2 (1%) were diagnosed with an arrhythmia necessitating an implantable cardioverter defibrillator.

Maines et al (2017) described outcomes in 154 consecutive patients receiving the Reveal LINQ ILR.61 After a mean follow-up of 12 months, a diagnosis was made in 99 (64%) patients. In response to the diagnoses, 20 patients initiated anticoagulation therapy, 18 patients had pacemakers implanted, 22 patients underwent electrophysiological study and ablation, and 14 changed antiarrhythmic medications.

Ciconte et al (2017) published results from 66 patients with documented AF or symptoms attributable to AF, who were given an implantable monitoring device (BioMonitor).62 Recordings from the monitoring device were compared with 48-hour Holter monitoring results performed 4 weeks after implantation. Sensitivity and positive predictive value for AF detection of the implantable monitoring device were 95% and 76%, respectively.

Bhangu et al (2016) reported on the diagnostic yield of ILRs in a series of 70 elderly patients with unexplained falls.63 Cardiac arrhythmias were detected in 49 (70%) of the patients within a mean time of 47 days. A fall in 14 (20%) patients were attributed to the arrhythmia. Ten patients received a cardiac pacemaker.

Nolker et al (2016) published results of the Performance of an Implantable Cardiac Monitor to Detect Atrial Fibrillation  study, in which readings from an ICM (Confirm ICM, St. Jude Medical) were compared with readings from a Holter monitor used for 4 days at least 2 weeks post implant.64 Patients had either been diagnosed with or had a clinical suspicion of paroxysmal AF (n=90). Due to difficulties with synchronizing the Holter monitor and the implanted device, data from only 79 patients were used in calculations. Patient-level sensitivity, positive predictive value, specificity, and negative predictive value were 100%, 64%, 86%, and 84%, respectively. Episode-level sensitivity, positive predictive value, specificity, and negative predictive value were 95%, 64%, 87%, and 76%, respectively.

Sanders et al (2016) reported on the diagnostic yield for AF with the Reveal LINQ device, a miniaturized ILR with a detection algorithm designed to detect AF.65 This nonrandomized, prospective trial included 151 patients, most of whom (81.5%) were undergoing monitoring for AF ablation or AF management. Compared with Holter-detected AF, the ILR had a diagnostic sensitivity and specificity for AF of 97.4% and 97.0%, respectively.

Ziegler et al (2015) reported on a large (n=1247) set of patients identified from the manufacturer's registry undergoing ILR monitoring for AF detection after a cryptogenic stroke.66 Over a median follow-up of 182 days, 1521 episodes of AF were detected in 147 patients. Overall, 42 (29%) patients had a single episode of AF and 105 (71%) patients had multiple episodes. The overall detection rate (12.2% at 182 days) was somewhat higher than that reported in the CRYSTAL AF trial.

In a report from a registry of patients who received or were about to receive an ILR (the Reveal Plus, DX, or XT device) because of unexplained syncope, Edvardsson et al (2014) described the monitoring yield in 570 patients implanted and followed for at least a year or until diagnosis.67 Most (97.5%) patients had a standard ECG before implantation of the ILR, 11.8% had prior ELR, and 54.6% had in-hospital ECG monitoring. During the monitoring period, 218 (38%) patients had recurrent syncope. The proportion of specific diagnoses based on the ILR is not reported, but of subjects who had a recurrence, 42.2% had a pacemaker implanted, 4.6% had an implantable cardioverter defibrillator placed, 4.1% received antiarrhythmic drug therapy, and 3.7% underwent catheter ablation.

Hindricks et al (2010) evaluated the accuracy of an autotriggered ILR in 247 patients at high-risk for paroxysmal AF.68 All patients underwent simultaneous 46-hour continuous Holter monitoring, and the authors calculated the performance characteristics of the loop recorder using physician-interpreted Holter monitoring as the criterion standard. The sensitivity of the loop recorder for detecting AF episodes of 2 or more minutes was 88.2%, increasing to 92.1% for episodes of 6 or more minutes. AF was falsely identified by the loop recorder in 19 of 130 patients who did not have AF while on a Holter monitor, for a false-positive rate of 15%. AF burden was accurately measured by the loop recorder, with the mean absolute difference between the loop recorder and Holter monitor of 1.4%.

Hanke et al (2009) compared an autotrigger ILR with 24-hour Holter monitoring done at 3-month intervals in 45 patients who had undergone surgical ablation for AF.69 After a mean follow-up of 8.3 months, the ILR identified AF in 19 (42%) patients in whom Holter monitoring recorded sinus rhythm.

Safety of ILRs
Mittal et al (2015) reported on safety outcomes related to the use of an ILR, based on data from 2 studies, the Reveal LINQ Usability study and the Reveal LINQ Registry.70 The Usability study enrolled 151 patients at 16 European and Australian centers; adverse events were reported for the first month of follow-up. The Registry is a multicenter postmarketing surveillance registry, with a planned enrollment of at least 1200. At the time of analysis, 161 patients had been enrolled. For Registry patients, all adverse events were recorded when they occurred. The device is inserted with a preloaded insertion tool via a small skin incision. In the Usability study, one serious adverse event was recorded (insertion site pain); in the Registry study, two serious adverse events were recorded (one case each of insertion site pain and insertion site infection). The rates of infection and procedure-related serious adverse events in the Usability study were 1.3% and 0.7%, respectively, and 1.6% and 1.6%, respectively, in the Registry study.

Clinically Useful
A test is clinically useful if the use of the results informs management decisions that improve the net health outcome of care. The net health outcome can be improved if patients receive correct therapy, or more effective therapy, or avoid unnecessary therapy, or avoid unnecessary testing.

Direct Evidence
Direct evidence of clinical utility is provided by studies that have compared health outcomes for patients managed with and without the test. Because these are intervention studies, the preferred evidence would be from RCTs. No RCTs providing evidence for clinical utility were identified.

Chain of Evidence
Indirect evidence on clinical utility rests on clinical validity. Evidence for clinical validity was provided by several RCTs, which showed that significantly more diagnoses were made with ILRs compared with Holter monitors or other standard care. Many observational studies reported the initiation of treatment (for example, anticoagulation therapy or pacemaker implantation) following the confirmation of diagnoses with the ILR. Because these treatments are known to be effective, it can be concluded that long-term monitoring with ILRs will improve health outcomes.

Section Summary: ILRs for Patients with Symptoms of Arrhythmia
Several RCTs have reported high rates of arrhythmia detection with the use of ILRs compared with external event monitoring or Holter monitoring. These studies support the use of a progression in diagnostics from an external event monitor to ILR when longer monitoring is needed. Some available trials evaluating the detection of AF after ablation procedures or in patients with cryptogenic stroke used ILRs as an initial ambulatory monitoring strategy, after a negative Holter monitor. Many observational studies reported the initiation of treatment (for example, anticoagulation therapy or pacemaker implantation) following the confirmation of diagnoses with the ILR. Because these treatments are known to be effective, it can be concluded that long-term monitoring with ILRs will improve health outcomes.

MOBILE CARDIAC OUTPATIENT TELEMETRY FOR PATIENTS WITH SYMPTOMS OF ARRHYTHMIA
Clinical Context and Test Purpose
This section addresses whether the addition of real-time MCOT to ambulatory cardiac monitoring is associated with improved outcomes. Two factors must be addressed in evaluating MCOT: (1) the inherent detection capability of the monitoring devices and (2) whether the real-time transmission and interpretation of data confers an incremental health benefit. The proposed addition of real-time monitoring suggests that there may be a subset of individuals who require immediate intervention when an arrhythmia is detected. Because it is not clear which patients comprise that subset, or whether identification of those patients in the outpatient setting leads to improved outcomes (eg, reduced risks of sudden cardiac death), the evaluation of the second factor requires studies that directly assess outcomes, not just arrhythmia detection rates.

The purpose of outpatient cardiac telemetry in patients with signs or symptoms suggestive of arrhythmia is to provide an alternative method of transmitting electrical cardiac activity data to healthcare providers.

The question addressed in this evidence review is: Does the use of outpatient cardiac telemetry added to ambulatory cardiac monitoring improve net health outcome in patients with signs or symptoms suggestive of arrhythmia compared with ambulatory cardiac event monitoring alone?

The following PICOTS were used to select literature to inform this review.

Patients
The relevant population of interest are patients with signs or symptoms suggestive of arrhythmia.

Interventions
The therapy being considered is MCOT system which transmits ambulatory cardiac monitoring data in real-time to healthcare providers.

Comparators
The comparator of interest is ambulatory cardiac monitoring alone.

Outcomes
The general outcome of interest is the incremental benefit of transmitting the ambulatory cardiac monitoring data in real-time.

Study Selection Criteria
For the evaluation of clinical validity of autoactivated or patient-activated external ambulatory event monitoring for patients with arrhythmia symptoms, studies that met the following criteria were considered:

  • To assess the clinical validity, studies should report sensitivity, specificity, positive and negative predictive values. Alternatively, studies reporting on diagnostic yield are informative.
  • To assess the clinical utility, studies should demonstrate how results of the tests impacted treatment decisions and overall management of the patient.

Technically Reliable
Assessment of technical reliability focuses on specific tests and operators and requires review of unpublished and often proprietary information. Review of specific tests, operators, and unpublished data are outside the scope of this evidence review and alternative sources exist. This evidence review focuses on the clinical validity and clinical utility.

Clinically Valid
A test must detect the presence or absence of a condition, the risk of developing a condition in the future, or treatment response (beneficial or adverse).

Randomized Controlled Trials
An RCT by Rothman et al (2007) compared MCOT with standard event monitors (Table 7).71 This trial involved 305 patients randomized to the LOOP recorder or to MCOT (CardioNet) and monitored for up to 30 days. Patients were recruited from 17 centers. Investigators and patients were not blinded to randomization assignment. Monitor strips and diagnoses were reviewed by an electrophysiologist blinded to the monitoring device assignment. Most patients in the LOOP recorder group had a patient-triggered event monitor. Only a subset of patients (n=50) had autotrigger devices, thus precluding comparison between MCOT and autotrigger devices. Analyses were conducted on patients completing at least 25 days of monitoring. The primary endpoint was either confirmation or exclusion of arrhythmic cause of the patient's symptoms. Arrhythmias were classified as either clinically significant or clinically insignificant.  The diagnostic endpoint (confirmation or exclusion of arrhythmic cause of symptoms) was significantly different between the 2 groups (Table 8). The difference in rates was primarily due to detection of asymptomatic (not associated with simultaneous symptoms) arrhythmias in the MCOT group, symptoms consisting of rapid AF and/or flutter (15 patients vs 1 patient), and ventricular tachycardia defined as more than 3 beats and rate greater than 100 (14 patients vs 2 patients). These differences were thought to be clinically significant rhythm disturbances and the likely causes of the patients' symptoms. In this trial, median time to diagnosis in the total study population was seven days in the MCOT group and nine days in the LOOP group (Table 8). The trialists did not comment on the clinical impact (changes in management) of these findings in patients for whom the rhythm disturbance did not occur simultaneously with symptoms.

Table 7. Summary of RCT Characteristics

Study

Countries

Sites

Dates

Participants

Interventions

Duration
         

Active

Comparator

 

Rothman (2007)71

US

17

NR

Patients with a high clinical suspicion of a malignant  arrhythmia, with syncope, presyncope, or severe palpitations, and a nondiagnostic 24-hr Holter test

Mobile automated cardiac outpatient telemetry (CardioNet)

n=134

Patient-activated external looping event monitor

n=132

Confirmation of a diagnosis, up to 30 days

NR: not reported; RCT: randomized controlled trial.

Table 8. Summary of RCT Results

Study

Confirmation or Exclusion of Arrhythmic Cause of Symptoms, n (%)

Confirmation or Exclusion of Arrhythmic Cause of Symptoms in Subgroup with Syncope, n (%)

Confirmation or Exclusion of Arrhythmic Cause of Symptoms in Subgroup Autotriggered Recorder, n (%)

Time to Diagnosis median (95% CI)

Rothman (2007)71

263

113

50 263

MCOT

117 (88.0)

55 (88.7)

21 (87.5) 7 (4 to 11)

LOOP

98 (75.4

35 (68.6)

12 (46.2) 9 (7 to 15)
p-value 0.008 0.008 0.002 NR

CI: confidence interval; LOOP: looping event monitor; MCOT: mobile cardiac outpatient telemetry; NR: not reported; RCT: randomized controlled trial.

Observational Studies
Arrhythmia Detection
Derkac et al (2017) retrospectively reviewed the BioTelemetry database of patients receiving ambulatory ECG monitoring, selecting patients prescribed MCOT (n=69977) and patients prescribed AT-LER, an autotrigger looping event recorder (n=8513).72 Patients were diagnosed with palpitations, syncope and collapse, AF, tachycardia, and/or TIA. Patients given the MCOT were monitored for an average of 20 days and patients given the AT-LER were monitored an average of 27 days. The diagnostic yield using MCOT was significantly higher than that using AT-LER for several events: 128% higher for AF, 54% higher for bradycardia, 17% higher for ventricular pause, 80% higher for SVT, and 222% higher for ventricular tachycardia. Mean time to diagnosis for each asymptomatic arrhythmia was shorter for patients monitored by MCOT than by AT-LER. There was no discussion of management changes or health outcomes based on monitoring results.

Kadish et al (2010) evaluated the frequency with which events transmitted by MCOT represented emergent arrhythmias, thereby indirectly assessing the clinical utility of real-time outpatient monitoring.73 Medical records from 26438 patients who had undergone MCOT during a 9-month period from a single service provider were retrospectively examined. During a mean monitoring period of 21 days, 21% (5459) had an arrhythmic event requiring physician notification. Of these, 1% (260) had an event that could be considered potentially emergent. These potentially emergent events included 120 patients with wide-complex tachycardia, 100 patients with sinus pauses 6 seconds or longer, and 42 with sustained bradycardia at less than 30 beats per minute.

A number of uncontrolled case series have reported on arrhythmia detection rates of MCOT.74-77 One study (Joshi et al [2005]) described the outcomes of a consecutive case series of 100 patients.74 Included patients had the following symptoms: palpitations (47%), dizziness (24%), or syncope (19%). Patients being evaluated for the efficacy of drug treatment (25%) were also included. Clinically significant arrhythmias were detected in 51% of patients, but half of these patients were asymptomatic. The authors commented that the automatic detection resulted in an increased diagnostic yield, but there was no discussion of its unique features (ie, the real-time analysis, transmission, and notification of arrhythmia).

AF Detection
In the largest study evaluating the diagnostic yield of MCOT for AF, Favilla et al (2015) evaluated a retrospective cohort of 227 patients with cryptogenic stroke or TIA who underwent 28 days of monitoring with MCOT.78 AF was detected in 14% (31/227) of patients, of whom 3 reported symptoms at the time of AF. Oral anticoagulation was initiated in 26 (84%) patients diagnosed with AF. Of the remaining 5 (16%) not on anticoagulation therapy, 1 had a prior history of gastrointestinal bleeding, 3 were unwilling to accept the risk of bleeding related to the use of anticoagulants, and 1 failed to follow-up.

Miller et al (2013) retrospectively analyzed paroxysmal AF detection rates among 156 patients evaluated with MCOT within 6 months of a cryptogenic stroke or TIA.30 Over a median 21-day period of MCOT monitoring (range, 1-30 days), AF was detected in 17.3% of patients. Mean time to first occurrence of AF was 9 days (range, 1-21 days).

Tayal et al (2008) retrospectively analyzed patients with cryptogenic stroke who had not been diagnosed with AF by standard monitoring.77 In this study, 13 (23%) of 56 patients with cryptogenic stroke had AF detected by MCOT. Twenty-seven asymptomatic AF episodes were detected in the 13 patients; 23 of them were less than 30 seconds in duration. In contrast, Kalani et al (2015) reported a diagnostic yield for AF of 4.7% (95% CI, 1.5% to 11.9%) in a series of 85 patients with cryptogenic stroke.79 In this series, 82.4% of patients had completed transesophageal echocardiography, cardiac magnetic resonance imaging, or both, with negative results. Three devices were used and described as MCOT devices: 34% received LifeStar ACT ambulatory cardiac telemetry, 41% received the LifeStar AF Express autodetect looping monitor, and 25% received the Cardiomedix cardiac event monitor. While the authors reported that there was a system in place to transmit the data for review, it is unclear whether data were sent in "real-time."

Narasimha et al (2018) published results of a study in which 33 patients wore both an ELR and a Kardia monitor to screen for AF during a period of 14 to 30 days.80 Patients were 18 years or older, had palpitations less often than daily but more frequently than several times per month, and prior nondiagnostic ECGs. Exclusion criteria included myocardial infarction within the last three months, history of ventricular tachycardia/fibrillation, unstable angina, and syncope. Study personnel viewed the Kardia monitor recordings once daily and a physician was contacted if a serious or sustained arrhythmia was detected. Patients were also monitored by the ELR company, which notified a physician on call when necessary. All 33 patients had a diagnosis using the Kardia monitor and 24 patients received a diagnosis using the ELR (p=0.001).

Dorr et al (2019) compared the diagnostic accuracy of a smartwatch system with cardiologists' interpretation of an ECG in the diagnostic accuracy to detect AF.81 The smartwatch system uses an algorithm to enable rhythm analysis of the photoplethysmographic signals. The population consisted of 508 hospitalized patients who had interpretable ECG and photoplethysmographic recordings. The photoplethysmographic algorithm compared with the cardiologists' diagnoses had a sensitivity of 94% and a specificity of 98%. A limitation of the study was that many of the recordings were excluded due to insufficient signal quality (148 of 672). The investigators concluded that detection of AF is feasible with a smartwatch, though signal quality issues need to be resolved and a broader population needs to be tested.

Clinically Useful
A test is clinically useful if the use of the results informs management decisions that improve the net health outcome of care. The net health outcome can be improved if patients receive correct therapy, or more effective therapy, or avoid unnecessary therapy, or avoid unnecessary testing.

Direct Evidence
Direct evidence of clinical utility is provided by studies that have compared health outcomes for patients managed with and without the test. Because these are intervention studies, the preferred evidence would be from RCTs. No RCTs were identified that evaluated the management of patients with and without mobile cardiac monitoring.

Chain of Evidence
Indirect evidence on clinical utility rests on clinical validity. Evidence for clinical validity consists of one RCT and several observational studies. The RCT reported a larger proportion of patients receiving a diagnosis in the MCOT group compared with the LOOP group, though time to diagnosis was not significantly different. In addition, no studies demonstrated an incremental benefit of the real-time transmission and interpretation of data compared with the usual monitoring timeline.

Section Summary: MCOT for Patients with Symptoms of Arrhythmia
The available evidence has suggested that MCOT is likely to be at least as good at detecting arrhythmias as ambulatory event monitoring. Compared with ambulatory event monitoring, MCOT is associated with the theoretical advantage of real-time monitoring, permitting for emergent intervention for potentially life-threatening arrhythmias. One study reported that 1% of arrhythmic events detected on MCOT during a mean monitoring period of 21 days per patient could be considered potentially emergent. However, no studies were identified that addressed whether the use of MCOT is associated with differences in the management of or outcomes after these potentially emergent events. The addition of real-time monitoring to outpatient ambulatory monitoring is considered an enhancement to existing technology. Currently, the evidence does not demonstrate a clinically significant incremental benefit for MCOT.

Summary of Evidence
Ambulatory Event Monitoring
For individuals who have signs and/or symptoms suggestive of arrhythmia(s) who receive patient- or autoactivated external ambulatory event monitoring or continuous ambulatory monitoring storing information for more than 48 hours, the evidence includes 1 RCT) and prospective and retrospective studies reporting on the diagnostic yield. The relevant outcomes are overall survival (OS) and morbid events. The RCT and the observational studies have consistently shown that continuous monitoring with longer recording periods detects more arrhythmias than 24- or 48-hour Holter monitoring. Particularly for patients who, without the more prolonged monitoring, would only undergo shorter term monitoring, the diagnostic yield is likely to identify arrhythmias that may have therapeutic implications. The evidence is sufficient to determine that the technology results in a meaningful improvement in the net health outcome.

For individuals who have AF following ablation who receive long-term ambulatory cardiac monitoring, the evidence includes  one RCT comparing ambulatory event monitoring with standard care and several observational studies. The relevant outcomes are OS, morbid events, medication use, and treatment-related morbidity. The RCT evaluating a long-term monitoring strategy after catheter ablation for AF reported significantly higher rates of AF detection. The available evidence has suggested that long-term monitoring for AF postablation is associated with improved outcomes. However, the specific type of monitoring associated with the best outcomes is not established, because different long-term monitoring devices were used across the studies. Trials demonstrating improved outcomes have used event monitors or implantable monitors. In addition, there are individual patient considerations that may make one type of monitor preferable over another. The evidence is sufficient to determine that the technology results in a meaningful improvement in the net health outcome.

For individuals who have cryptogenic stroke with a negative standard workup for AF who receive long-term ambulatory cardiac monitoring, the evidence includes systematic reviews of RCTs comparing ambulatory event monitoring with standard care. The relevant outcomes are OS, morbid events, medication use, and treatment-related morbidity. RCTs evaluating a long-term AF monitoring strategy poststroke have reported significantly higher rates of AF detection with longer term ambulatory monitoring. The available evidence has suggested that long-term monitoring for AF after cryptogenic stroke is associated with improved outcomes, but the specific type of monitoring associated with the best outcomes is not established, because different long-term monitoring devices were used across the studies. Trials demonstrating improved outcomes have used event monitors or implantable monitors. In addition, there are individual patient considerations that may make one type of monitor preferable over another. The evidence is sufficient to determine that the technology results in a meaningful improvement in the net health outcome.

For individuals who are asymptomatic with risk factors for AF who receive long-term ambulatory cardiac monitoring, the evidence includes an RCT and a nonrandomized study. The relevant outcomes are OS, morbid events, medication use, and treatment-related morbidity. The studies showed use of the ambulatory monitors would result in higher AF detection compared with routine care. However, the RCT followed patients for one year and did not detect a difference in stroke occurrence between the monitored group and the standard of care group. The other studies did not discuss changes in patient management or health outcomes based on monitoring. Studies reporting on improved outcomes with longer follow-up are needed. The evidence is insufficient to determine the effects of the technology on health outcomes.

Implantable Loop Recording
For individuals who have signs and/or symptoms suggestive of arrhythmia with infrequent symptoms who receive patient- or autoactivated implantable ambulatory event monitoring, the evidence includes RCTs comparing ILRs with shorter term monitoring, usually 24- to 48-hour Holter monitoring, and many observational studies. The relevant outcomes are OS, morbid events, medication use, and treatment-related morbidity. Studies assessing prolonged ILRs in patients have reported high rates of arrhythmia detection compared with shorter external event or Holter monitoring. These studies have supported use of a progression in diagnostics from an external event monitor to implantable loop recorder when longer monitoring is needed. The evidence is sufficient to determine that the technology results in a meaningful improvement in the net health outcome.

Outpatient Cardiac Telemetry
For individuals who have signs and/or symptoms suggestive of arrhythmia who receive outpatient cardiac telemetry, the evidence includes an RCT and nonrandomized studies evaluating rates of arrhythmia detection using outpatient cardiac telemetry. The relevant outcomes are OS and morbid events. The available evidence has suggested that outpatient cardiac telemetry is at least as good at detecting arrhythmias as ambulatory event monitoring. However, studies have not evaluated whether the real-time monitoring feature of outpatient cardiac telemetry leads to reduced cardiac events and mortality. The evidence is insufficient to determine the effects of the technology on health outcomes.

CLINICAL INPUT FROM PHYSICIAN SPECIALTY SOCIETIES AND ACADEMIC MEDICAL CENTERS
While the various physician specialty societies and academic medical centers may collaborate with and make recommendations during this process, through the provision of appropriate reviewers, input received does not represent an endorsement or position statement by the physician specialty societies or academic medical centers, unless otherwise noted.

2014 Input
In response to requests, input was received from 3 physician specialty societies and 4 academic medical centers (3 reviews) while this policy was under review in 2014. Input was obtained to provide information on mobile cardiac outpatient telemetry and new devices. There was no consensus whether mobile cardiac outpatient telemetry is medically necessary. While reviewers agreed that mobile cardiac outpatient telemetry is comparable to event monitors for arrhythmia detection, they did not agree on whether the real-time monitoring provides incremental benefit over external event monitors or is associated with improved health outcomes compared with external event monitors. There was consensus on the medical necessity of externally worn event monitors with longer continuous recording periods as an alternative to Holter monitors or event monitors. For implantable memory loop devices that are smaller than older-generation devices, there was consensus that these devices improve the likelihood of obtaining clinically useful information due to improved ease of use, but there was no consensus that such devices improve clinical outcomes and are medically necessary.

2009 Input
In response to requests, input was received from 1 physician specialty society and 4 academic medical centers (5 reviews) while this policy was under review in 2009. There were differences among reviewers on outpatient cardiac telemetry, with some reviewers concluding it had a role in certain subsets of patients (eg, in those with sporadic atrial fibrillation). Other reviewers commented that the value of this technology should be considered in both providing a diagnosis and in making treatment decisions. At times, excluding arrhythmia as a cause of a patient's symptoms is an important finding. 

PRACTICE GUIDELINES AND POSITION STATEMENTS
International Society for Holter and Noninvasive Electrocardiology et al
The International Society for Holter and Noninvasive Electrocardiology and the Heart Rhythm Society (HRS; 2017) issued a consensus statement on ambulatory electrocardiogram and external monitoring and telemetry.82 Below are two summary tables from the consensus statement, detailing advantages and limitations of ambulatory electrocardiogram techniques (see Table 9) and recommendations for the devices that are relevant to this evidence review (see Table 10). 

Table 9. Advantages and Limitations of Ambulatory ECG Techniques, International Society for Holter and Noninvasive Electrocardiology/HRS 

ECG Monitoring Technique

Advantages

Limitations

Holter monitoring

 

  • Records and documents continuous 3- to 32-lead ECG signal simultaneously with biologic signals during normal daily activities
  • Physicians familiar with analysis software and scanning services
  • Frequent noncompliance with symptom logs and event markers
  • Frequent electrode detachments
  • Signal quality issues due to skin adherence, tangled wires, dermatitis
  • Absence of real-time data analysis
  • Poor patient acceptance of electrodes

Patch ECG monitors

 

  • Long-term recording of ≥14 d
  • Excellent patient acceptance
  • Limited ECG from closely spaced electrodes, lacking localization of arrhythmia origin
  • Inconsistent ECG quality due to body type variations

External loop recorders

 

  • Records only selected ECG segments marked as events either automatically or manually by patient
  • Immediate alarm generation on event detection
  • Single-lead ECG, lacking localization of arrhythmia origin
  • Cannot continuously document cardiac rhythm
  • Requires patient to wear electrodes continuously

Event recorders

 

  • Records only selected ECG segments after an event is detected by patient
  • Immediate alarm generation at event detected by patient
  • Well-tolerated by patient
  • Single-lead ECG, lacking localization of arrhythmia origin
  • Cannot continuously document cardiac rhythm
  • Diagnostic yield dependent on patient ability to recognize correct symptom

Mobile cardiac telemetry

 

  • Multilead, so higher sensitivity and specificity of arrhythmia detection
  • Streams data continuously; can be programmed to autodetect and autosend events at prescribed time intervals
  • Immediate alarm generation on event without patient interaction
  • Long-term patient acceptance is reduced due to requirement of daily electrode changes

ECG: electrocardiogram.

Table 10. Select Recommendations for Ambulatory ECG and External Monitoring or Telemetry, International Society for Holter and Noninvasive Electrocardiology/HRS

Recommendation

CORa

LOEb

Selection of ambulatory ECG

Holter monitoring when symptomatic events anticipated within 48 h

I

B-NR

Extended ambulatory ECG (15-30 d) when symptomatic events are not daily or are uncertain

I

B-R

Continuous monitoring (1-14 d) to quantify arrhythmia burden and patterns

!

B-NR

Specific conditions for use of ambulatory ECG

Unexplained syncope, when tachycardia suspected

I

B-R

Unexplained palpitation

I

B-R

Detection of atrial fibrillation, triggering arrhythmias, and postconversion pauses

IIa

B-NR

Cryptogenic stroke, to detect undiagnosed atrial fibrillation

I

B-R

ECG: electrocardiogram; COR: class of recommendation; LOE: level of evidence.
a COR definitions: I: strong recommendation; IIa: benefit probably exceeds risk.
b LOE definitions: B-NR: moderate level based on well-executed nonrandomized studies; B-R: moderate level based on randomized trials.   

American Heart Assoication, American College of Cardiology, and Heart Rhythm Society
The American College of Cardiology, the American Heart Association, and HRS (2019) updated guidelines initially issued in 20144 on the management of patients with atrial fibrillation (AF).83 These guidelines recommended the use of Holter or event monitoring if the diagnosis of the type of arrhythmia is in question, or as a means of evaluating rate control.

The same associations (2017) collaborated on guidelines on the evaluation and management of patients with syncope84 and patients with ventricular arrhythmias85. Cardiac monitoring recommendations are summarized below in Tables 11 and 12.

Table 11. Cardiac Monitoring Recommendations, AHA/ACC/HRS 

Recommendation

CORa

LOEb

Choice of a specific cardiac monitor should be determined on the basis of frequency and nature of syncope events.84

I

C-EO

To evaluate selected ambulatory patients with syncope of suspected arrhythmic etiology, the following external cardiac monitoring approaches can be useful: Holter monitor, transtelephonic monitor, external loop recorder, patch recorder, and mobile cardiac outpatient telemetry.84

IIa

B-NR

To evaluate selected ambulatory patients with syncope of suspected arrhythmic etiology, an implantable cardiac monitor can be useful.84

IIa

B-R

Ambulatory electrocardiographic monitoring is useful to evalute whether symptoms including palpitations, presyncope, or syncope, are caused by VA85 I B-NR
In patients with cryptogenic stroke (i.e., stroke of unknown cause), in whom external ambulatory monitoring is inconclusive, implantation of a cardiac monitor (loop recorder) is reasonable to optimize detection of silent AF.83 IIa B-R

ACC: American College of Cardiology; AHA: American Hearth Association; COR: class of recommendation; HRS: Heart Rhythm Society; LOE: level of evidence; VA: ventricular arrhythmia.
a COR definitions: I: strong recommendation; IIa: benefit probably exceeds risk.
b LOE definitions: B-NR: moderate level based on well-executed nonrandomized studies; B-R: moderate level based on randomized trials; C-EO: consensus of expert opinion based on clinical experience.

Table 12. Patient Selection Recommendations by Cardiac Rhythm Monitor, AHA/ACC/HRS

Type of Monitor

Patient Selection

Holter monitor

Symptoms frequent enough to be detected within 24 to 72 h

Patient-activated event monitor

  • Frequent spontaneous symptoms likely within 2 to 6 wk
  • Limited use when syncope associated with sudden incapacitation

External loop recorder (patient or auto-triggered)

Frequent spontaneous symptoms likely to occur within 2 to 6 wk 

External patch recorder

  • Alternative to external loop recorder
  • Leadless, so more comfortable, resulting in improved compliance
  • Offers only 1-lead recording

Mobile cardiac outpatient telemetry

  • Spontaneous symptoms related to syncope and rhythm correlation
  • High-risk patients needing real-time monitoring

Implantable cardiac monitor

Recurrent, infrequent, unexplained syncope 

ACC: American College of Cardiology; AHA: American Heart Association; HRS: Heart Rhythm Society.

Heart Rhythm Society et al
A consensus document on catheter and surgical ablation for AF was published in 2012 by HRS, the European Heart Rhythm Association, and the European Cardiac Arrhythmia Society86 and updated in 2017.87 This document did not contain formal practice guidelines, but provided general recommendations based on literature review and expert consensus. Use of ambulatory event monitors postablation was addressed in two sections of the document. First, in the section discussing use of anticoagulation following ablation, the following statement was made:

"Patients in whom discontinuation of systematic anticoagulation is being considered based on patient values and preferences should consider undergoing continuous or frequent ECG [electrocardiogram] monitoring to screen for AF recurrence."

In the section on postoperative rhythm monitoring of patients who are postablation, the following statements were made:

"The success of AF ablation is based in large part on freedom from AF recurrence based on ECG monitoring. Arrhythmia monitoring can be performed with the use of noncontinuous or continuous ECG monitoring tools."

The statement referenced a table of ambulatory cardiac monitoring devices (Holter, patch, external loop, implantable loop, wearable multisensors, Smartphone monitors), describing unique features of each. The table did not evaluate the safety or efficacy of these devices, nor recommend one over another.

European Heart Rhythm Association
The European Heart Rhythm Association (2009) published guidelines on the use of diagnostic implantable and external loop recorders.88 For the indications that the Association considered established at the time of publication, the guidelines made the following statements about indications for implantable and external recorders (see Table 13).

Table 13. Guidelines on Use of Diagnostic ILRs and ELRs 

Recommendation

COR

LOE

"ILR [implantable loop recorder] is indicated:

  • "In an early phase of evaluation of patients with recurrent syncope of uncertain origin who have:
  • "absence of high-risk criteria that require immediate hospitalization or intensive evaluation…"; and
  • "a likely recurrence within battery longevity of the device."

I

A

"ELRs are indicated in patients with recurrent palpitations, undocumented by conventional ECG techniques, who have: inter-symptom interval <4 weeks and absence of high-risk criteria…which require immediate hospitalization or intensive evaluation."

I

B

"ILR may be indicated to assess the contribution of bradycardia before embarking on cardiac pacing in patients with suspected or certain neurally mediated syncope presenting with frequent or traumatic syncopal episodes."

IIa

B

"ILRs may be indicated in selected cases with severe infrequent symptoms when ELRs and other ECG monitoring systems fail to document the underlying cause."

IIa

B

"ELRs [external loop recorder] may be indicated in patients with recurrent (pre)syncopes who have:

  • "inter-symptom interval of ≤4 weeks, and
  • "suspicion of arrhythmic origin and
  • "absence of high-risk criteria that require immediate hospitalization or intensive evaluation…."

IIa

B

COR: class of recommendations; ECG: electrocardiogram; ELR: external loop recorder; ILR: implantable loop recorder; LOE: level of evidence.

American Academy of Neurology
The American Academy of Neurology updated its guidelines on the prevention of stroke in patients with nonvalvular AF (NVAF).89 These guidelines made the following recommendations on the identification of patients with occult NVAF: 

A1. "Clinicians might obtain outpatient cardiac rhythm studies in patients with cryptogenic stroke without known NVAF, to identify patients with occult NVAF (Level C).
A2. Clinicians might obtain cardiac rhythm studies for prolonged periods (e.g., for 1 or more weeks) instead of shorter periods (e.g., 24 hours) in patients with cryptogenic stroke without known NVAF, to increase the yield of identification of patients with occult NVAF (Level C)."

U.S. PREVENTIVE SERVICES TASK FORCE RECOMMENDATIONS
Not applicable. 

ONGOING AND UNPUBLISHED CLINICAL TRIALS
Some currently unpublished trials that might influence this review are listed in Table 14.

Table 12. Summary of Key Trials  

NCT No.

Trial Name

Planned Enrollment

Completion Date

Ongoing

     

NCT02392754

Home-Based Screening for Early Detection of Atrial Fibrillation in Primary Care Patients Aged 75 Years and Older (SCREEN-AF)

822

Dec 2019

NCT02786940

Remote Cardiac Monitoring of Higher-Risk Emergency Department Syncope Patients after Discharge (REMOSYNC)

600

May 2019

NCT03001765

Impact of an Intensive Monitoring Strategy in Symptomatic Patients with Suspected Arrhythmia (IMPACT)

150

Jul 2019

NCT02428140a

Post-Embolic Rhythm Detection With Implantable Versus External Monitoring (PERDIEM)

300

Dec 2019

NCT03072693 Daily Ambulatory Remote Monitoring System vs Conventional Therapy for the Post-Discharge Management of Acute Decompensated Heart Failure 876 Jul 2020

NCT02793895

Detection of Atrial Fibrillation After Cardiac Surgery (SEARCH-AF)

396

Sep 2020

NCT02684825a

Detection of Silent Atrial Fibrillation aFter Ischemic StrOke (SAFFO)

424

Jun 2021

a Denotes industry involvement
NCT: national clinical trial.

References: 

  1. Task Force for the Diagnosis Management of Syncope, European Society of Cardiology, European Heart Rhythm Association, et al. Guidelines for the diagnosis and management of syncope (version 2009). Eur Heart J. Nov 2009;30(21):2631-2671. PMID 19713422.
  2. National Institute for Health and Care Excellence (NICE). Transient loss of consciousness ('blackouts') in over 16s [CG109]. 2014; https://www.nice.org.uk/guidance/cg109. Accessed March 15, 2018.
  3. Raviele A, Giada F, Bergfeldt L, et al. Management of patients with palpitations: a position paper from the European Heart Rhythm Association. Europace. Jul 2011;13(7):920-934. PMID 21697315.
  4. January CT, Wann LS, Alpert JS, et al. 2014 AHA/ACC/HRS guideline for the management of patients with atrial fibrillation: executive summary: a report of the American College of Cardiology/American Heart Association Task Force on practice guidelines and the Heart Rhythm Society. Circulation. Dec 2 2014;130(23):2071-2104. PMID 24682348.
  5. Mittal S, Movsowitz C, Steinberg JS. Ambulatory external electrocardiographic monitoring: focus on atrial fibrillation. J Am Coll Cardiol. Oct 18 2011;58(17):1741-1749. PMID 21996384.
  6. Christensen LM, Krieger DW, Hojberg S, et al. Paroxysmal atrial fibrillation occurs often in cryptogenic ischaemic stroke. Final results from the SURPRISE study. Eur J Neurol. Jun 2014;21(6):884-889. PMID 24628954.
  7. Hoefman E, Bindels PJ, van Weert HC. Efficacy of diagnostic tools for detecting cardiac arrhythmias: systematic literature search. Neth Heart J. Nov 2010;18(11):543-551. PMID 21113379.
  8. Steinhubl, SS, Waalen, JJ, Edwards, AA, Ariniello, LL, Mehta, RR, Ebner, GG, Carter, CC, Baca-Motes, KK, Felicione, EE, Sarich, TT, Topol, EE. Effect of a Home-Based Wearable Continuous ECG Monitoring Patch on Detection of Undiagnosed Atrial Fibrillation: The mSToPS Randomized Clinical Trial. JAMA, 2018 Jul 13;320(2). PMID 29998336.
  9. Farris, GG, Smith, BB, Oates, EE, Colon, CC, Doppalapudi, HH. New atrial fibrillation diagnosed by 30-day rhythm monitoring. Am. Heart J., 2019 Jan 15;209:29-35. PMID 30639611.
  10. Turakhia MP, Hoang DD, Zimetbaum P, et al. Diagnostic utility of a novel leadless arrhythmia monitoring device. Am J Cardiol. Aug 15 2013;112(4):520-524. PMID 23672988.
  11. 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. Jan 2014;127(1):95 e11-97. PMID 24384108.
  12. Solomon MD, Yang J, Sung SH, et al. Incidence and timing of potentially high-risk arrhythmias detected through long term continuous ambulatory electrocardiographic monitoring. BMC Cardiovasc Disord. Feb 17 2016;16(1):35. PMID 26883019.
  13. Bolourchi M, Batra AS. Diagnostic yield of patch ambulatory electrocardiogram monitoring in children (from a national registry). Am J Cardiol. Mar 1 2015;115(5):630-634. PMID 25591894.
  14. Eisenberg EE, Carlson SK, Doshi RH, et al. Chronic ambulatory monitoring: results of a large single-center experience. J Innovations Cardiac Rhythm Manage. Nov 2014;5:1818-1823.
  15. Schreiber D, Sattar A, Drigalla D, et al. Ambulatory cardiac monitoring for discharged emergency department patients with possible cardiac arrhythmias. West J Emerg Med. Mar 2014;15(2):194-198. PMID 24672611.
  16. Eysenck, WW, Freemantle, NN, Sulke, NN. A randomized trial evaluating the accuracy of AF detection by four external ambulatory ECG monitors compared to permanent pacemaker AF detection. J Interv Card Electrophysiol, 2019 Feb 12. PMID 30741360.
  17. Health Quality Ontario. Long-Term Continuous Ambulatory ECG Monitors and External Cardiac Loop Recorders for Cardiac Arrhythmia: A Health Technology Assessment. Ont Health Technol Assess Ser. Feb 2017;17(1):1- 56. PMID 28194254.
  18. Balmelli N, Naegeli B, Bertel O. Diagnostic yield of automatic and patient-triggered ambulatory cardiac event recording in the evaluation of patients with palpitations, dizziness, or syncope. Clin Cardiol. Apr 2003;26(4):173-176. PMID 12708623.
  19. Ermis C, Zhu AX, Pham S, et al. Comparison of automatic and patient-activated arrhythmia recordings by implantable loop recorders in the evaluation of syncope. Am J Cardiol. Oct 1 2003;92(7):815-819. PMID 14516882.
  20. Reiffel JA, Schwarzberg R, Murry M. Comparison of autotriggered memory loop recorders versus standard loop recorders versus 24-hour Holter monitors for arrhythmia detection. Am J Cardiol. May 1 2005;95(9):1055-1059. PMID 15842970.
  21. Dagres N, Kottkamp H, Piorkowski C, et al. Influence of the duration of Holter monitoring on the detection of arrhythmia recurrences after catheter ablation of atrial fibrillation: implications for patient follow-up. Int J Cardiol. Mar 18 2010;139(3):305-306. PMID 18990460.
  22. Pokushalov E, Romanov A, Corbucci G, et al. Ablation of paroxysmal and persistent atrial fibrillation: 1- year follow-up through continuous subcutaneous monitoring. J Cardiovasc Electrophysiol. Apr 2011;22(4):369-375. PMID 20958836.
  23. Chao TF, Lin YJ, Tsao HM, et al. CHADS(2) and CHA(2)DS(2)-VASc scores in the prediction of clinical outcomes in patients with atrial fibrillation after catheter ablation. J Am Coll Cardiol. Nov 29 2011;58(23):2380-2385. PMID 22115643.
  24. Kapa S, Epstein AE, Callans DJ, et al. Assessing arrhythmia burden after catheter ablation of atrial fibrillation using an implantable loop recorder: the ABACUS study. J Cardiovasc Electrophysiol. Aug 2013;24(8):875-881. PMID 23577826.
  25. Verma A, Champagne J, Sapp J, et al. Discerning the incidence of symptomatic and asymptomatic episodes of atrial fibrillation before and after catheter ablation (DISCERN AF): a prospective, multicenter study. JAMA Intern Med. Jan 28 2013;173(2):149-156. PMID 23266597.
  26. Themistoclakis S, Corrado A, Marchlinski FE, et al. The risk of thromboembolism and need for oral anticoagulation after successful atrial fibrillation ablation. J Am Coll Cardiol. Feb 23 2010;55(8):735-743. PMID 20170810.
  27. Gumbinger C, Krumsdorf U, Veltkamp R, et al. Continuous monitoring versus HOLTER ECG for detection of atrial fibrillation in patients with stroke. Eur J Neurol. Feb 2012;19(2):253-257. PMID 21895885.
  28. Lazzaro MA, Krishnan K, Prabhakaran S. Detection of atrial fibrillation with concurrent holter monitoring and continuous cardiac telemetry following ischemic stroke and transient ischemic attack. J Stroke Cerebrovasc Dis. Feb 2012;21(2):89-93. PMID 20656504.
  29. Cotter PE, Martin PJ, Ring L, et al. Incidence of atrial fibrillation detected by implantable loop recorders in unexplained stroke. Neurology. Apr 23 2013;80(17):1546-1550. PMID 23535493.
  30. Miller DJ, Khan MA, Schultz LR, et al. Outpatient cardiac telemetry detects a high rate of atrial fibrillation in cryptogenic stroke. J Neurol Sci. Jan 15 2013;324(1-2):57-61. PMID 23102659.
  31. Sposato LA, Cipriano LE, Saposnik G, et al. Diagnosis of atrial fibrillation after stroke and transient ischaemic attack: a systematic review and meta-analysis. Lancet Neurol. Apr 2015;14(4):377-387. PMID 25748102.
  32. Kishore A, Vail A, Majid A, et al. Detection of atrial fibrillation after ischemic stroke or transient ischemic attack: a systematic review and meta-analysis. Stroke. Feb 2014;45(2):520-526. PMID 24385275.
  33. Kamel H, Navi BB, Elijovich L, et al. Pilot randomized trial of outpatient cardiac monitoring after cryptogenic stroke. Stroke. Feb 2013;44(2):528-530. PMID 23192756.
  34. Higgins P, MacFarlane PW, Dawson J, et al. Noninvasive cardiac event monitoring to detect atrial fibrillation after ischemic stroke: a randomized, controlled trial. Stroke. Sep 2013;44(9):2525-2531. PMID 23899913.
  35. Sinha AM, Diener HC, Morillo CA, et al. Cryptogenic Stroke and underlying Atrial Fibrillation (CRYSTAL AF): design and rationale. Am Heart J. Jul 2010;160(1):36-41 e31. PMID 20598970.
  36. Sanna T, Diener HC, Passman RS, et al. Cryptogenic stroke and underlying atrial fibrillation. N Engl J Med. Jun 26 2014;370(26):2478-2486. PMID 24963567.
  37. Brachmann J, Morillo CA, Sanna T, et al. Uncovering atrial fibrillation beyond short-term monitoring in cryptogenic stroke patients: three-year results from the Cryptogenic Stroke and Underlying Atrial Fibrillation Trial. Circ Arrhythm Electrophysiol. Jan 2016;9(1):e003333. PMID 26763225.
  38. Gladstone DJ, Spring M, Dorian P, et al. Atrial fibrillation in patients with cryptogenic stroke. N Engl J Med. Jun 26 2014;370(26):2467-2477. PMID 24963566.
  39. Ritter MA, Kochhauser S, Duning T, et al. Occult atrial fibrillation in cryptogenic stroke: detection by 7-day electrocardiogram versus implantable cardiac monitors. Stroke. May 2013;44(5):1449-1452. PMID 23449264.
  40. Etgen T, Hochreiter M, Mundel M, et al. Insertable cardiac event recorder in detection of atrial fibrillation after cryptogenic stroke: an audit report. Stroke. Jul 2013;44(7):2007-2009. PMID 23674523.
  41. Tung CE, Su D, Turakhia MP, et al. Diagnostic yield of extended cardiac patch monitoring in patients with stroke or TIA. Front Neurol. Jan 2014;5:266. PMID 25628595.
  42. Rosenberg MA, Samuel M, Thosani A, et al. Use of a noninvasive continuous monitoring device in the management of atrial fibrillation: a pilot study. Pacing Clin Electrophysiol. Mar 2013;36(3):328-333. PMID 23240827.
  43. Savelieva I, Camm AJ. Clinical relevance of silent atrial fibrillation: prevalence, prognosis, quality of life, and management. J Interv Card Electrophysiol. Jun 2000;4(2):369-382. PMID 10936003.
  44. Israel CW, Gronefeld G, Ehrlich JR, et al. Long-term risk of recurrent atrial fibrillation as documented by an implantable monitoring device: implications for optimal patient care. J Am Coll Cardiol. Jan 07 2004;43(1):47-52. PMID 14715182.
  45. Page RL, Wilkinson WE, Clair WK, et al. Asymptomatic arrhythmias in patients with symptomatic paroxysmal atrial fibrillation and paroxysmal supraventricular tachycardia. Circulation. Jan 1994;89(1):224-227. PMID 8281651.
  46. Hart RG, Pearce LA, Rothbart RM, et al. Stroke with intermittent atrial fibrillation: incidence and predictors during aspirin therapy. Stroke Prevention in Atrial Fibrillation Investigators. J Am Coll Cardiol. Jan 2000;35(1):183-187. PMID 10636278.
  47. Hohnloser SH, Pajitnev D, Pogue J, et al. Incidence of stroke in paroxysmal versus sustained atrial fibrillation in patients taking oral anticoagulation or combined antiplatelet therapy: an ACTIVE W Substudy. J Am Coll Cardiol. Nov 27 2007;50(22):2156-2161. PMID 18036454.
  48. Ganesan AN, Chew DP, Hartshorne T, et al. The impact of atrial fibrillation type on the risk of thromboembolism, mortality, and bleeding: a systematic review and meta-analysis. Eur Heart J. May 21 2016;37(20):1591-1602. PMID 26888184.
  49. Fitzmaurice DA, Hobbs FD, Jowett S, et al. Screening versus routine practice in detection of atrial fibrillation in patients aged 65 or over: cluster randomised controlled trial. BMJ. Aug 25 2007;335(7616):383. PMID 17673732.
  50. Halcox JPJ, Wareham K, Cardew A, et al. Assessment of Remote Heart Rhythm Sampling Using the AliveCor Heart Monitor to Screen for Atrial Fibrillation: The REHEARSE-AF Study. Circulation. Nov 7 2017;136(19):1784-1794. PMID 28851729.
  51. Turakhia MP, Ullal AJ, Hoang DD, et al. Feasibility of extended ambulatory electrocardiogram monitoring to identify silent atrial fibrillation in high-risk patients: the Screening Study for Undiagnosed Atrial Fibrillation (STUDY-AF). Clin Cardiol. May 2015;38(5):285-292. PMID 25873476.
  52. Solbiati M, Casazza G, Dipaola F, et al. The diagnostic yield of implantable loop recorders in unexplained syncope: A systematic review and meta-analysis. Int J Cardiol. Mar 15 2017;231:170-176. PMID 28052814.
  53. Burkowitz J, Merzenich C, Grassme K, et al. Insertable cardiac monitors in the diagnosis of syncope and the detection of atrial fibrillation: A systematic review and meta-analysis. Eur J Prev Cardiol. Aug 2016;23(12):1261-1272. PMID 26864396.
  54. Da Costa A, Defaye P, Romeyer-Bouchard C, et al. Clinical impact of the implantable loop recorder in patients with isolated syncope, bundle branch block and negative workup: a randomized multicentre prospective study. Arch Cardiovasc Dis. Mar 2013;106(3):146-154. PMID 23582676.
  55. Farwell DJ, Freemantle N, Sulke AN. Use of implantable loop recorders in the diagnosis and management of syncope. Eur Heart J. Jul 2004;25(14):1257-1263. PMID 15246645.
  56. Krahn AD, Klein GJ, Yee R, et al. Randomized assessment of syncope trial: conventional diagnostic testing versus a prolonged monitoring strategy. Circulation. Jul 3 2001;104(1):46-51. PMID 11435336.
  57. Afzal MR, Gunda S, Waheed S, et al. Role of outpatient cardiac rhythm monitoring in cryptogenic stroke: a systematic review and meta-analysis. Pacing Clin Electrophysiol. Oct 2015;38(10):1236-1245. PMID 26172621.
  58. Podoleanu C, DaCosta A, Defaye P, et al. Early use of an implantable loop recorder in syncope evaluation: a randomized study in the context of the French Healthcare System (FRESH study). Arch Cardiovasc Dis. Oct 2014;107(10):546-552. PMID 25241220.
  59. Giada F, Gulizia M, Francese M, et al. Recurrent unexplained palpitations (RUP) study comparison of implantable loop recorder versus conventional diagnostic strategy. J Am Coll Cardiol. May 15 2007;49(19):1951-1956. PMID 17498580.
  60. Magnusson, PP, Olszowka, MM, Wallhagen, MM, Koyi, HH. Outcome of implantable loop recorder evaluation. Cardiol J, 2017 Aug 26;25(3). PMID 28840588.
  61. Maines, MM, Zorzi, AA, Tomasi, GG, Angheben, CC, Catanzariti, DD, Piffer, LL, Del Greco, MM. Clinical impact, safety, and accuracy of the remotely monitored implantable loop recorder Medtronic Reveal LINQTM. Europace, 2017 Oct 11;20(6). PMID 29016753.
  62. Ciconte G, Saviano M, Giannelli L, et al. Atrial fibrillation detection using a novel three-vector cardiac implantable monitor: the atrial fibrillation detect study. Europace. Jul 1 2017;19(7):1101-1108. PMID 27702865.
  63. Bhangu J, McMahon CG, Hall P, et al. Long-term cardiac monitoring in older adults with unexplained falls and syncope. Heart. May 1 2016;102(9):681-686. PMID 26822427.
  64. Nolker G, Mayer J, Boldt LH, et al. Performance of an Implantable Cardiac Monitor to Detect Atrial Fibrillation: Results of the DETECT AF Study. J Cardiovasc Electrophysiol. Dec 2016;27(12):1403-1410. PMID 27565119.
  65. Sanders P, Purerfellner H, Pokushalov E, et al. Performance of a new atrial fibrillation detection algorithm in a miniaturized insertable cardiac monitor: Results from the Reveal LINQ Usability Study. Heart Rhythm. Jul 2016;13(7):1425-1430. PMID 26961298.
  66. Ziegler PD, Rogers JD, Ferreira SW, et al. Real-world experience with insertable cardiac monitors to find atrial fibrillation in cryptogenic stroke. Cerebrovasc Dis. 2015;40(3-4):175-181. PMID 26314298.
  67. Edvardsson N, Garutti C, Rieger G, et al. Unexplained syncope: implications of age and gender on patient characteristics and evaluation, the diagnostic yield of an implantable loop recorder, and the subsequent treatment. Clin Cardiol. Oct 2014;37(10):618-625. PMID 24890550.
  68. Hindricks G, Pokushalov E, Urban L, et al. Performance of a new leadless implantable cardiac monitor in detecting and quantifying atrial fibrillation: Results of the XPECT trial. Circ Arrhythm Electrophysiol. Apr 2010;3(2):141-147. PMID 20160169.
  69. Hanke T, Charitos EI, Stierle U, et al. Twenty-four-hour holter monitor follow-up does not provide accurate heart rhythm status after surgical atrial fibrillation ablation therapy: up to 12 months experience with a novel permanently implantable heart rhythm monitor device. Circulation. Sep 15 2009;120(11 Suppl):S177-184. PMID 19752365.
  70. Mittal S, Sanders P, Pokushalov E, et al. Safety profile of a miniaturized insertable cardiac monitor: results from two prospective trials. Pacing Clin Electrophysiol. Dec 2015;38(12):1464-1469. PMID 26412309.
  71. Rothman SA, Laughlin JC, Seltzer J, et al. The diagnosis of cardiac arrhythmias: a prospective multi-center randomized study comparing mobile cardiac outpatient telemetry versus standard loop event monitoring. J Cardiovasc Electrophysiol. Mar 2007;18(3):241-247. PMID 17318994.
  72. Derkac WM, Finkelmeier JR, Horgan DJ, et al. Diagnostic yield of asymptomatic arrhythmias detected by mobile cardiac outpatient telemetry and autotrigger looping event cardiac monitors. J Cardiovasc Electrophysiol. Dec 2017;28(12):1475-1478. PMID 28940881.
  73. Kadish AH, Reiffel JA, Clauser J, et al. Frequency of serious arrhythmias detected with ambulatory cardiac telemetry. Am J Cardiol. May 1 2010;105(9):1313-1316. PMID 20403485.
  74. Joshi AK, Kowey PR, Prystowsky EN, et al. First experience with a Mobile Cardiac Outpatient Telemetry (MCOT) system for the diagnosis and management of cardiac arrhythmia. Am J Cardiol. Apr 1 2005;95(7):878-881. PMID 15781022.
  75. Olson JA, Fouts AM, Padanilam BJ, et al. Utility of mobile cardiac outpatient telemetry for the diagnosis of palpitations, presyncope, syncope, and the assessment of therapy efficacy. J Cardiovasc Electrophysiol. May 2007;18(5):473-477. PMID 17343724.
  76. Saarel EV, Doratotaj S, Sterba R. Initial experience with novel mobile cardiac outpatient telemetry for children and adolescents with suspected arrhythmia. Congenit Heart Dis. Jan-Feb 2008;3(1):33-38. PMID 18373747.
  77. Tayal AH, Tian M, Kelly KM, et al. Atrial fibrillation detected by mobile cardiac outpatient telemetry in cryptogenic TIA or stroke. Neurology. Nov 18 2008;71(21):1696-1701. PMID 18815386.
  78. Favilla CG, Ingala E, Jara J, et al. Predictors of finding occult atrial fibrillation after cryptogenic stroke. Stroke. May 2015;46(5):1210-1215. PMID 25851771.
  79. Kalani R, Bernstein R, Passman R, et al. Low yield of mobile cardiac outpatient telemetry after cryptogenic stroke in patients with extensive cardiac imaging. J Stroke Cerebrovasc Dis. Sep 2015;24(9):2069-2073. PMID 26139455.
  80. Narasimha D, Hanna N, Beck H, et al. Validation of a smartphone-based event recorder for arrhythmia detection. Pacing Clin Electrophysiol. Mar 1 2018. PMID 29493801.
  81. Dörr, MM, Nohturfft, VV, Brasier, NN, Bosshard, EE, Djurdjevic, AA, Gross, SS, Raichle, CC, Rhinisperger, MM, Stöckli, RR, Eckstein, JJ. The WATCH AF Trial: SmartWATCHes for Detection of Atrial Fibrillation. JACC Clin Electrophysiol, 2019 Feb 21;5(2). PMID 30784691.
  82. Steinberg JS, Varma N, Cygankiewicz I, et al. 2017 ISHNE-HRS expert consensus statement on ambulatory ECG and external cardiac monitoring/telemetry. Heart Rhythm. Jul 2017;14(7):e55-e96. PMID 28495301.
  83. January, CC, Wann, LL, Calkins, HH, Field, MM, Chen, LL, Furie, KK, Cigarroa, JJ, Heidenreich, PP, Cleveland, JJ, Murray, KK, Ellinor, PP, Shea, JJ, Ezekowitz, MM, Tracy, CC, Yancy, CC. 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. Heart Rhythm, 2019 Feb 1. PMID 30703530.
  84. Shen WK, Sheldon RS, Benditt DG, et al. 2017 ACC/AHA/HRS Guideline for the Evaluation and Management of Patients With Syncope: Executive Summary: 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. Aug 1 2017;70(5):620-663. PMID 28286222.
  85. Al-Khatib, SS, Stevenson, WW, Ackerman, MM, Bryant, WW, Callans, DD, Curtis, AA, Deal, BB, Dickfeld, TT, Field, MM, Fonarow, GG, Gillis, AA, Granger, CC, Hammill, SS, Hlatky, MM, Joglar, JJ, Kay, GG, Matlock, DD, Myerburg, RR, Page, RR. 2017 AHA/ACC/HRS guideline for management of patients with ventricular arrhythmias and the prevention of sudden cardiac death: Executive summary: A Report of the American College of Cardiology/American Heart Association Task Force on Clinical Practice Guidelines and the Heart Rhythm Society. Heart Rhythm, 2017 Nov 4;15(10). PMID 29097320.
  86. Calkins H, Kuck KH, Cappato R, et al. 2012 HRS/EHRA/ECAS expert consensus statement on catheter and surgical ablation of atrial fibrillation: recommendations for patient selection, procedural techniques, patient management and follow-up, definitions, endpoints, and research trial design. J Interv Card Electrophysiol. Mar 2012;33(2):171-257. PMID 22382715.
  87. Calkins H, Hindricks G, Cappato R, et al. 2017 HRS/EHRA/ECAS/APHRS/SOLAECE expert consensus statement on catheter and surgical ablation of atrial fibrillation: Executive summary. J Arrhythm. Oct 2017;33(5):369-409. PMID 29021841.
  88. Task Force members, Brignole M, Vardas P, et al. Indications for the use of diagnostic implantable and external ECG loop recorders. Europace. May 2009;11(5):671-687. PMID 19401342.
  89. Culebras A, Messe SR, Chaturvedi S, et al. Summary of evidence-based guideline update: prevention of stroke in nonvalvular atrial fibrillation: report of the Guideline Development Subcommittee of the American Academy of Neurology. Neurology. Feb 25 2014;82(8):716-724. PMID 24566225.
  90. Centers for Medicare & Medicaid Services (CMS). National Coverage Determination (NCD) for Electrocardiographic Services (20.15). 2004; https://www.cms.gov/medicare-coverage-database/details/ncd- details.aspx?MCDId=16&ExpandComments=n&McdName=Thomson+Micromedex+DrugDex+%C2%AE+Compe ndium+Revision+Request+-+CAG-00391&NCDId=179. Accessed April 9, 2019.

Coding Section

Codes Number  Description 
 CPT 33282**  Implantation of patient-activated cardiac event recorder (**CODE WILL BE DELETED EFFECTIVE 01/01/2019)
  33284** Removal of an implantable, patient-activated cardiac event recorder (**CODE WILL BE DELETED EFFECTIVE 01/01/2019)
  33285 (effective 1/1/2019) Insertion, subcutaneous cardiac rhythm monitor, including programming 
  33286 (effective 1/1/2019)  Removal, subcutaneous cardiac rhythm monitor 
  93228 External mobile cardiovascular telemetry with electrocardiographic recording, concurrent computerized real time data analysis and greater than 24 hours of accessible ECG data storage (retrievable with query) with ECG triggered and patient selected events transmitted to a remote attended surveillance center for up to 30 days; review and interpretation with report by a physician or other qualified health care professional
  93229 ;technical support for connection and patient instructions for use, attended surveillance, analysis and transmission of daily and emergent data reports as prescribed by a physician or other qualified health care professional
  93268 External patient and, when performed, auto-activated electrocardiographic rhythm derived event recording with symptom-related memory loop with remote download capability up to 30 days, 24-hour attended monitoring; includes transmission, review, and interpretation by a physician or other qualified health care professional 
  93270 ;recording (includes connection, recording, and disconnection) 
  93271 ;transmission download and analysis 
  93272 ;review and interpretation by a physician or other qualified health care professional 
  0295T External electrocardiographic recording for more than 48 hours up to 21 days by continuous rhythm recording and storage; includes recording, scanning analysis with report, review and interpretation 
  0296T ;recording (includes connection and initial recording) 
  0297T ;scanning analysis with report 
  0298T  ;review and interpretation 
ICD-9 Procedure 89.50  Ambulatory cardiac monitoring 
ICD-9 Diagnosis  426.9  Unspecified conduction disorder 
  427.31  Atrial fibrillation 
  427.60 Premature beats, unspecified 
  427.89  Other specified cardiac dysrhythmias 
  427.9 Cardiac dysrhythmia, unspecified 
  427.91 Cerebral artery occlusion, unspecified with cerebral infarction (stroke) 
  780.2  Syncope and collapse 
  780.4 Dizziness 
  785.1  Palpitations
HCPCS     
ICD-10-CM (effective 10/01/15)  144.0-144.7  Atrioventricular and left bundle-branch block code range 
  145.0-145.9 Other conduction disorders code range 
  147.0-147.9 Paroxysmal tachycardia code range 
  148.0-148.1  Atrial Fibrillation and flutter code range 
  149.01-149.9  Other cardiac arrhythmias code range 
  163.9  Cerebral infarction, unspecified (Stroke NOS) 
  R00.2  Palpitations 
  R42  Dizziness and giddiness 
  R55 Ayncope and collapse
ICD-10-PCS (effective 10/01/15)   ICD-10-PCS codes are only used for inpatient services. 
  4A12X45  Measurement and monitoring, physiological systems, monitoring, cardiac, external, electrical activity, ambulatory 
Type of Service  Cardiology   
Place of Service Outpatient   

This medical policy was developed through consideration of peer-reviewed medical literature generally recognized by the relevant medical community, U.S. FDA approval status, nationally accepted standards of medical practice and accepted standards of medical practice in this community, Blue Cross and Blue Shield Association technology assessment program (TEC) and other non-affiliated technology evaluation centers, reference to federal regulations, other plan medical policies and accredited national guidelines.

"Current Procedural Terminology© American Medical Association.  All Rights Reserved" 

History From 2014 Forward     

06/14/2019 

Annual review, no change to policy intent. Updating background, rationale and references. 

12/05/2018 

Update policy with 2019 coding. No other changes 

07/07/2018 

Annual review, adding mobile apps as an investigational modality. Also updating background, description, rationale, and references. 

06/12/2017 

Annual review, no change to policy intent. 

06/14/2016 

Interim review, policy statements edited for simplicity to group continuous ambulatory monitors with longer recoding periods with external event monitors and to move language regarding the use of long-term outpatient monitoring for AF to policy guidelines. Updating background, description, guidelines, rationale and references. 

03/22/2016 

Interim review, no change to policy intent, updating Zio Patch coverage statement but NO change to intent. 

10/19/2015 

Annual review, no change to policy intent. Updated background, description, rationale and references. 

06/25/2015 

Corrected the work stroke in the policy section. No other change was made. 

10/29/2014

Annual review. Added policy guidelines and coding. Updated references and rationale. Added policy verbiage that indicates the Zio Patch may be medically necessary under certain conditions.  


Go Back