Coverage Policy Manual
Policy #: 1998153
Category: Surgery
Initiated: July 1998
Last Review: October 2018
  Cardiac Event Recorder, Insertable Loop Recorder

Description:
There are a wide variety of devices available for outpatient cardiac rhythm monitoring. The primary purpose of these devices is the evaluation of suspected arrhythmias that have not been detected by office- or hospital-based 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 delivery of the information from patient to clinician.
 
An insertable or implantable loop recorder device is inserted just under the patient’s skin in the chest area during an outpatient surgical procedure. When symptoms are felt, the patient places a hand-held activator over the recorder to activate the storage of cardiac rhythms. The Reveal® Insertable Loop Recorder (Medtronic) is an implantable memory loop device recently approved by the U.S. Food and Drug Administration (FDA). The Reveal ILR is a device 61 mm x 19 mm x 8 mm weighing 17 grams. It is implanted subcutaneously in a single incision procedure in a left pectoral or mammary location.  Its projected longevity is 14 months to a low battery condition.  The manufacturer recommends that the device be explanted when it is no longer clinically necessary or when the battery is depleted.
 
Implantable continuous “memory loop” devices with autotrigger combine the long-term monitoring available with implantable devices with the autotriggers seen on newer event monitors. These devices contain algorithms that are programmed to detect heart rates exceeding an upper or lower limit, asystole of greater than 3 seconds. They typically contain other autotriggers, such as a variable RR interval seen with atrial fibrillation.
 

Policy/
Coverage:
Effective October 2015
 
Meets Primary Coverage Criteria Or Is Covered For Contracts Without Primary Coverage Criteria
 
An insertable loop recorder meets member benefit certificate primary coverage criteria that there be scientific evidence of effectiveness in improving health outcomes when all of the following conditions have been met:
 
    • Recurrent unexplained syncope, or a single episode of unexplained syncope with injury or accident;  AND
    • Nondiagnostic complete history and physical examination, including ECG and orthostatic blood pressure recording; AND
    • Two negative or nondiagnostic 30-day event recordings, or equivalent;
 
Note: The ILR device insertion procedure is considered to be a physician service. It carries a 90 day global and does not require an assistant surgeon.
 
Note: Removal of an ILR device on the same day as the insertion of a cardiac pacemaker in any given patient is considered to be a part of the pacemaker insertion procedure and will not be covered separately.
 
Does Not Meet Primary Coverage Criteria Or Is Investigational For Contracts Without Primary Coverage Criteria
 
Use of ILR does not meet member benefit certificate primary coverage criteria that there be scientific evidence of effectiveness in improving health outcomes in the following conditions:
 
    • Patients with presyncopal episodes;
    • Patients failing to fulfill the indications for coverage of this policy;
    • Patients receiving another ILR device within the last two years
 
For members with contracts without primary coverage criteria, use of ILR in the conditions listed below is considered not medically necessary. Services that are considered not medically necessary are specific contract exclusions in most member benefit certificates of coverage.
 
    • Patients with presyncopal episodes;
    • Patients failing to fulfill the indications for coverage of this policy;
    • Patients receiving another ILR device within the last two years
 
Effective Prior to October 2015
 
 
Meets Primary Coverage Criteria Or Is Covered For Contracts Without Primary Coverage Criteria
 
An insertable loop recorder meets member benefit certificate primary coverage criteria that there be scientific evidence of effectiveness in improving health outcomes in patients with syncope who have experienced infrequent syncopal episodes which have defied diagnosis by conventional means; when all of the following conditions have been met:
  
    • Complete history and physical examination;
    • Electrocardiogram (ECG);
    • Two negative or nondiagnostic 30-day pre-symptom memory loop patient demand recordings (may be either single or multiple event recordings);
    • Negative or nondiagnostic tilt table testing.  
 
Note: The ILR device insertion procedure is considered to be a physician service.  It carries a 90 day global and does not require an assistant surgeon.
 
Note: Removal of an ILR device on the same day as the insertion of a cardiac pacemaker in any given patient is considered to be a part of the pacemaker insertion procedure and will not be covered separately.
 
Does Not Meet Primary Coverage Criteria Or Is Investigational For Contracts Without Primary Coverage Criteria
  
Use of ILR does not meet member benefit certificate primary coverage criteria that there be scientific evidence of effectiveness in improving health outcomes in the following conditions:
 
    • Patients with presyncopal episodes;
    • Patients failing to fulfill the indications for coverage of this policy;
    • Patients for whom compliance or lifestyle makes use of external monitoring systems inconvenient.
    • Patients receiving another ILR device within the last two years
 
For members with contracts without primary coverage criteria, use of ILR in the conditions listed below is considered not medically necessary. Services that are considered not medically necessary are specific contract exclusions in most member benefit certificates of coverage.
 
    • Patients with presyncopal episodes;
    • Patients failing to fulfill the indications for coverage of this policy;
    • Patients for whom compliance or lifestyle makes use of external monitoring systems inconvenient.
    • Patients receiving another ILR device within the last two years
 
 
 

Rationale:
Hoefman et al. (Hoefman, 2010) published a systematic review on diagnostic tools for detecting cardiac arrhythmias. This analysis 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; auto-triggered event recorders; and implantable loop recorders. The yield varied among devices, with the autotrigger devices offering the highest range of detection (72-80%), followed by the patient-activated devices (17-75%), and Holter monitors (33-35%). No combined analysis was performed due to heterogeneity in patient population and study design. Limitations in the evidence base precluded any specific recommendations on selection of devices. The authors concluded that the choice of device should be driven largely by the presence, type and frequency of symptoms experienced by each individual patient.
 
Implantable autotrigger loop recorders have also been developed that are specifically geared toward detection of atrial fibrillation through the use of atrial fibrillation detection algorithms. . Hindricks et al.  evaluated the accuracy of an implantable autotriggered loop recorder in 247 patients at high risk for paroxysmal atrial fibrillation. 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 gold standard. The sensitivity of the loop recorder for detecting atrial fibrillation episodes of 2 minutes or more in duration was 88.2%, rising to 92.1% for episodes of 6 minutes or more. Atrial fibrillation was falsely identified by the loop recorder in 19 of 130 patients who did not have atrial fibrillation on Holter monitoring, for a false-positive rate of 15%. The atrial fibrillation burden was accurately measured by the loop recorder, with the mean absolute difference between the loop recorder and Holter monitor of 1.4 +/- 6.4% (Hindricks, 2010).
 
Hanke et al. compared an implantable autotrigger device with 24-hour Holter monitoring done at 3-month intervals in 45 patients who had undergone surgical ablation for atrial fibrillation. (10) After a mean follow-up of 8.3 months, the implantable loop recorder identified atrial fibrillation in 19 patients (42%) in whom Holter monitoring recorded sinus rhythm (Hanke, 2009).
 
2012 Update
A literature search conducted through September 2012 did not reveal any new information that would prompt a change in the coverage statement.
 
2013 Update
A search of the MEDLINE database was conducted through September 2013. There was no new information identified that would prompt a change in the coverage statement.
 
One small RCT was identified that compared the use of an implantable loop recorder with conventional follow-up in 78 patients with a first episode of syncope (Da Costa, 2013). A significant number of patients had cardiomyopathy (23%), atrial fibrillation (15.4%), and/or bundle branch block on electrocardiography (ECG, 58%). Mean follow-up time was 27 months. A total of 21 patients (27%) had at least one arrhythmia detected, with a significant difference in detection rate for the implantable loop recorder group (36.6%) compared to the conventional follow-up group (10.8%, p=0.02).
   
2014 Update
A literature search conducted through September 2014 did not reveal any new information that would prompt a change in the coverage statement. The key identified literature is summarized below.
 
Barrett et al published a comparison of arrhythmia detection rates in 146 patients who underwent simultaneous monitoring with a 24-hour Holter monitor and a 14-day Zio Patch monitor (Barrett, 2014). Included were patients referred for evaluation of a suspected cardiac arrhythmia at single institution for the detection of atrioventricular block, pause, polymorphic ventricular tachycardia, supraventricular tachycardia, or AF. Holter monitoring detected 61 arrhythmias, while the Zio Patch detected 96 (p<0.001). Over the course of the monitoring period, 60 arrhythmias were detected by both devices, with 36 detected by the Zio Patch that were not detected by Holter monitoring and 1 detected by the Holter that was not detected by the Zio Patch. 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 further suggests 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 may have clinical significance if its performance is similar in nonresearch settings.
 
Autotriggered Event Monitors and Loop Recorders
The most appropriate indications for an implantable loop recorder (compared with an external loop recorder) are not well established. Locati et al evaluated the diagnostic yield of an external loop recorder with extended memory capacity in the evaluation of patients with syncope, presyncope, or sustained palpitations in an effort to determine their role in the diagnostic workup of these conditions (Locati, 2014). The authors evaluated 307 consecutive patients with external loop recorders (the SpiderFlash-A®, a patient-triggered device, and SpiderFlash-T® device, which has autotrigger capacity) who were enrolled in a registry at a single institution. The mean duration of recording was 24.1 days, with 85% of subjects having recording for 3 to 5 weeks. Ninety-two patients had syncope as their initial presentation; of those, a typical syncopal event occurred during the study monitoring period in 17 patients and an ECG recording during syncope was available in 16 patients. In 7 of these patients (44%), significant arrhythmias were recorded during syncope: bradycardia and pauses requiring pacemaker implant in 3 patients and fast supraventricular tachyarrhythmia (paroxysmal AF or paroxysmal supraventricular tachycardia) in 4 patients. Two hundred fifteen patients had palpitations or presyncope as their initial presentation; of those, a typical episode occurred during the study monitoring period in 184 patients, and many had multiple episodes (median 3 episodes per patient). In SpiderFlash-A recordings, sinus rhythm or sinus  tachycardia was recorded in about one third of the cases, and about one third had sustained supraventricular tachycardia or AF at the time of the symptoms. In the SpiderFlash-T recordings, supraventricular tachycardia or AF was recorded in about 46% of the cases, while bradycardia or pauses was recorded in about 13% of the cases.
 
Patients with AF treated with catheter ablation
In a prospective, randomized study, Kapa et al compared implantable loop monitors with conventional trans-telephonic recorders in the assessment of arrhythmia burden after catheter ablation of AF (Kapa, 2013). Forty-four patients were enrolled and randomized; all patients received the implantable loop recorder 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 three 30-day trans-telephonic monitoring periods. At 6 month postablation, patients were allocated to the randomization arm (decided in a 1:1 manner at initial enrollment) of either the implantable loop recorder (transmission of data every 31 days) or conventional monitoring (twice daily 1-minute pulse-rate assessment, and 1 trans-telephonic recording for 30 days at month 11). Over the first 6 months after ablation, conventional monitoring revealed AF in 7/38 patients (18%) and the implantable loop recorder confirmed AF in all of these patients. In an additional 11 patients (29%), AF was detected on implantable loop recorder. During the subsequent 6-month period, 5/18 patients in the conventional monitoring arm refused ongoing monitoring due to discomfort and lifestyle restrictions; of the remaining 13, 5 had a recurrence of AF (38%). In the implantable loop recorder group, 5 of 20 patients had recurrence of AF. In the implantable loop recorder arm, 71% patients had their antiarrhythmic drugs discontinued compared with 44% in the conventional monitoring group over the randomization period (p=0.04).
 
Christensen et al reported results of long-term cardiac monitoring with implantable loop recorders in a population of 85 patients with cryptogenic stroke (Christensen, 2014). The device was explanted early in 5 patients, 3 due to a skin reaction and 2 due to discomfort; after more than 1 year of monitoring, an additional 3 patients chose early removal of the device. In 18 patients (20.7%), paroxysmal AF was detected during the study period, 4 by ECG (2 obtained in preparation for implantation procedure, 1 on ECG for pacemaker placement for a non-AF arrhythmia, and 1 on ECG due to symptomatic tachycardia), and 14 on the basis of the implantable loop recorder monitoring. The mean time from stroke onset to the first episode of AF on the loop recorder was 109 days. Although patients with detected AF received anticoagulation, rates of stroke or TIA were higher in the AF group than the non-AF group (33.3% vs 10.1%, p=0.024).
 
Patients with cryptogenic stroke
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. Oral anticoagulation in patients with AF reduces the risk of subsequent stroke and is recommended by American Heart Association/American College of Cardiology guidelines for patients with a history of stroke or transient ischemic attack (TIA) (January, 2014a).
 
Kishore et al conducted a systematic review and meta-analysis of prospective observational studies and RCTs that reported rates of detection of newly-diagnosed AF in patients with ischemic stroke or TIA who underwent any cardiac monitoring for at least 12 hours (Kishore, 2014). Thirty-two studies were included: 18 studies that included patients with ischemic stroke only, 1 study that included TIA only, and 13 studies included both ischemic stroke and TIA. The authors 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%) and among selected patients was 13.4% (95% CI, 9.0% to 18.4%). In cryptogenic strokes, new AF was detected in 15.9% (95% CI, 10.9% to 21.6%). Among selected patients, the detection rate of AF 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 recorder, and MCOT) was 14.7% (95% CI, 10.7% to 19.3%).
 
The Kishore and other studies suggest that longer periods of cardiac monitoring increase the likelihood of AF detection. However, many of these asymptomatic episodes of AF are brief and the relationship to the preceding stroke uncertain, as there are other potential causes of asymptomatic stroke. 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.
 
Two larger RCTs were published in 2014. Sanna et al reported results from the CRYSTAL-AF study, an RCT to evaluate whether long-term monitoring of patients with cryptogenic stroke with implantable cardiac monitors (ICM) leads to changes in anticoagulant management and/or improved outcomes (Sinha, 2010; Sanna, 2014). The study 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 seen on computed tomography (CT) scan or magnetic resonance imaging, 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 randomly assigned patients, 416 (94.3%) completed 6 months of follow-up, 2 were lost to follow-up, 5 died, and 18 exited the study 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). The median time from randomization to detection of AF was 41 days (interquartile range [IQR], 14-84) in the ICM group and 32 days (IQR, 2-73) in the control group. Most AF episodes in the ICM group were asymptomatic (74%), compared with 33% of those in the control group. The rate of AF detection was similarly greater in the ICM group at the 12-month follow-up point (12.4% vs 2.0%; HR=7.3; 95% CI, 2.6 to 20.8; p<0.001). The rate of use of oral anticoagulants was 10.1% in the ICM group versus 4.6% in the control group at 6 months (p=0.04) and 14.7% versus 6.0% at 12 months (p=0.007). Five of the 208 ICMs (2.4%) that were inserted were removed due to infection or erosion of the device pocket.
 
In 2014, Gladstone et al reported results from the EMBRACE study, an RCT that compared 30-day autotriggered cardiac event monitors with conventional 24-hour monitors for the detection of AF in patients with cryptogenic stroke (Gladstone, 2014). Included patients were aged 55 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. Five hundred seventy-two patients were randomized to receive an external event recorder (ER910AF Cardiac Event Monitor, Braemar) or 24-hour Holter monitoring. Among the intervention group subjects, 82% completed at least 3 weeks of monitoring. AF was detected in 45 of 280 patients (16.1%) in the intervention group, compared with 9 of 277 (3.2%) in the control group (risk difference, 12.9 percentage points; 95% CI, 8.0 to 17.6; p<0.001). At 90 days of follow-up, patients in the intervention group were more likely to be treated with anticoagulants than the control group (18.6% vs 11.1%; absolute treatment difference, 7.5 percentage points; 95% CI, 1.6 to 13.3; p=0.01).
 
Several nonrandomized studies have evaluated the role of implantable loop recorders in the diagnosis of paroxysmal AF in cryptogenic stroke. Ritter et al compared 7-day Holter monitoring with an implantable loop recorder (Ritter, 2013).) A total of 60 patients with an acute cryptogenic stroke that was consistent with an embolic event were included. All patients received 7-day Holter monitoring, as well as an ICM. Patients were monitored with the ICM for a minimum of 1 year, or until an episode of AF was detected. A total of 10 patients (17%; 95% CI, 7% to 26%) had AF detected by ICM compared with 1 patient (1.7%; 95% CI, 0% to 5%) who had AF detected by Holter monitor (between-group comparison of detection rate, p<0.001). The average time to detection with ICM was 64 days (range, 1-556 days). All patients who had AF detected were treated with anticoagulation, and there were no recurrent strokes in either group.
 
In a study by Etgen et al, patients with cryptogenic, MRI-proven stroke who were eligible for oral anticoagulation were offered evaluation with an implantable cardiac loop recorder (Reveal XT; MedtronicInc.) and followed for the development/diagnosis of AF (Etgen, 2013). Evaluation for causes of stroke included MRI, 12-lead ECG, 24 to 72 hour continuous cardiac monitoring in a stroke unit, at least one 24-hour Holter monitor, extra- and transcranial neurosonography, echocardiography, CT/MRI angiography, and laboratory screening for prothrombotic states in patients aged younger than 55 years. Of 65 patients diagnosed with cryptogenic stroke at a single institution, 22 (33.8%) patients were implanted with a loop recorder, while the remaining patients were considered “not feasible” for the event recorder due to a contraindication to anticoagulation (n=20), cognitive problem (n=7), lack of sufficient cardiac follow-up (n=7), noncompliance (n=5), or refusal of insertion (n=4). Over 1 year of follow-up, paroxysmal AF was detected in 6 patients (27.3%).
 
2016 Update
A literature search conducted through September 2016 did not reveal any new information that would prompt a change in the coverage statement. The key identified literature is summarized below.
 
In 2015, Sposato and colleagues reported results of a systematic review and meta-analysis of studies reporting rates of new AF diagnosed after cryptogenic stroke or TIA based on cardiac monitoring, stratified into 4 sequential phases of screening: 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; and phase 4 (second ambulatory period) consisted of mobile cardiac outpatient telemetry (MCOT), external loop recording, and implantable loop recording (Sposato, 2015).
 
In total, 50 studies with 11,658 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 summary proportion of patients diagnosed with poststroke AF was 7.7% (95% confidence interval [CI], 5.0 to 10.8) in phase 1, 5.1% (95% CI, 3.8% to 6.5%) in phase 2, 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%), external loop recording (16.2%; 95% CI, 0.3% to 24.6%), or ILR (16.9%; 95% CI, 10.3% to 24.9%; p=0.97).
 
In 2016, Burkowitz and colleagues reported on a systematic review and meta-analysis of ILRs in the diagnosis of syncope and the detection of AF (Burkowitz, 2016). These indications are discussed separately in this review. For the indication of syncope diagnosis, the review identified 3 RCTs comparing ILRs with a conventional diagnosis strategy, which was Holter monitoring in all 3 studies. In pooled analysis, an ILR diagnosis strategy was associated with a higher likelihood of the end point of diagnostic yield (relative risk, 4.17; 95% CI, 2.57 to 6.77; I2=14%).
 
In 2014, Podoleanu and colleagues reported 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 strategy (Podoleanu, 2014). The study 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 a workup including 12-lead ECG, echocardiography, and head-up tilt-test. The 78 included patients were randomized to receive an ILR (the Reveal or Reveal Plus device; Medtronic, Minneapolis, MN; n=39) immediately or to be assessed using the conventional evaluation strategy (n=39), excluding the use of an ILR. After 14 months of follow-up, a certain cause of syncope was established in 18 (46.2%) of patients in the ILR group and in 2 (5%) of conventionally managed patients (p<0.001). Arrhythmic causes of syncope in the ILR group included 2 (5%) cases of atrioventricular (AV) block, 4 (10%) cases of sinus node disease, 1 (2.5%) case of AF, 1 (2.5%) case of ventricular fibrillation, and 3 (8%) other tachycardias. In the conventionally managed group, 8 patients had a diagnosis of presumed reflex syncope.
 
One small RCT compared use of an ILR to conventional follow-up in 78 patients with a first episode of syncope (Da Costa, 2013). A significant number of patients had cardiomyopathy (23%), AF (15.4%), and/or bundle branch block on ECG (58%). Mean follow-up time was 27 months. Twenty-one (27%) patients had at least 1 arrhythmia detected, with a significant difference in detection rate for the ILR group (36.6%) compared with the conventional follow-up group (10.8%, p=0.02).
 
In 2015, Afzal and colleagues reported on a systematic review and meta-analysis of studies comparing ILRs with wearable AEMs for prolonged outpatient rhythm monitoring after cryptogenic stroke (Afzal, 2015). The review 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 device 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).
 
In the Burkowitz systematic review and meta-analysis (described above), for the indication of cryptogenic stroke, 1 RCT and 5 noncomparative studies met inclusion criteria (Burkowitz, 2016). The sole RCT identified by Sanna and colleagues is described above.
 
In 2015, Ziegler and colleagues reported on a large (N=1247) set of patients undergoing ILR monitoring for AF detection after a cryptogenic stroke who were identified from the manufacturer’s registry (Ziegler, 2015). Over a median follow-up of 182 days, a total of 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 CRYSTAL AF.
 
Sanders and colleagues reported on the diagnostic yield for AF with the Reveal Linq device, a miniaturized ILR with a detection algorithm designed to detect AF in a nonrandomized, prospective trial (Sanders, 2016). The study 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%.
 
In 2015, Mittal and colleagues reported on safety outcomes related to the use of an ILR, the Reveal LINQ device, based on data from 2 studies, the Reveal LINQ Usability study and the Reveal LINQ Registry (Mittal, 2015). 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 version used in these studies measures 7 × 45 × 4 mm3, and is inserted with a preloaded insertion tool via a small skin incision. In the Usability study, 1 serious adverse event was recorded (insertion site pain); in the Registry study, 2 serious adverse events were recorded (1 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 were 1.6% and 1.6%, respectively, in the Registry study
 
Studies of prolonged use of ILRs in patients have reported high rates of arrhythmia detection 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.
 
2017 Update
A literature search conducted using the MEDLINE database did not reveal any new information that would prompt a change in the coverage statement.
  
2018 Update
A literature search was conducted through September 2018.  There was no new information identified that would prompt a change in the coverage statement.  The key identified literature is summarized below.
 
AEMs and AF Detection in Asymptomatic Patients
 
Randomized Controlled Trials
Halcox et al conducted an RCT (REHEARSE-AF), which screened patients for AF using the AliveCor Kardia monitor (n=500) or routine care (n=501) (Halcox, 2017). 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 Studies
Narasimha et al 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 (Narasimha, 2018). Patients were 18 years or older, had palpitations less often than daily but more frequently than several times per month, and prior non-diagnostic ECGs. Study personnel viewed the Kardia monitor recordings once daily. A physician was contacted if a serious or sustained arrhythmia was detected. Patients were also monitored by the external loop recording company, which notified a physician on call when necessary. All 33 patients had a potential diagnosis using the Kardia monitor and 24 patients received a diagnosis using the ELR (p=0.001).
 
ILR Use for Individuals With Signs and/or Symptoms of Arrhythmia
 
Systematic Reviews
Solbiati et al conducted a systematic review and meta-analysis on the diagnostic yield of ILRs in patients with unexplained syncope (Solbiati, 2017). 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.
 
ILRs in the Detection of AF
 
Observational Studies
Ciconte et al published results from 66 patients with documented AF or symptoms attributable to
AF, who were given an implantable monitoring device (BioMonitor) (Ciconte, 2017). 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.
 
Nolker et al published results of the DETECT AF study, in which readings from an implantable cardiac monitor (Confirm ICM, St. Jude Medical) were compared with readings from a Holter monitor used for 4 days at least 2 weeks post implant (Nolker, 2016). 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.
 
MOBILE CARDIAC OUTPATIENT TELEMETRY
 
Observational Studies
 
Arrhythmia Detection
Derkac et al retrospectively reviewed the BioTelemetry database of patients receiving ambulatory
ECG monitoring, selecting patients prescribed MCOT (n=69,977) and patients prescribed AT-LER, an auto-trigger looping event recorder (n=8513) (Derkac, 2017). 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.
 

CPT/HCPCS:
33282Implantation of patient-activated cardiac event recorder
33284Removal of an implantable, patient-activated cardiac event recorder
33285Insertion, subcutaneous cardiac rhythm monitor, including programming
33286Removal, subcutaneous cardiac rhythm monitor
33999Unlisted procedure, cardiac surgery
93799Unlisted cardiovascular service or procedure
E0616Implantable cardiac event recorder with memory, activator, and programmer
E1399Durable medical equipment, miscellaneous

References: Afzal MR, Gunda S, Waheed S, et al.(2015) 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

Barrett PM, Komatireddy R, Haaser S et al.(2014) Comparison of 24-hour Holter monitoring with 14-day novel adhesive patch electrocardiographic monitoring. Am J Med 2014; 127(1):95 e11-7.

Burkowitz J, Merzenich C, Grassme K, et al.(2016) Insertable cardiac monitors in the diagnosis of syncope and the detection of atrial fibrillation: A systematic review and meta-analysis. Eur J Prev Cardiol. Feb 10 2016. PMID 26864396

Christensen LM, Krieger DW, Hojberg S et al.(2014) Paroxysmal atrial fibrillation occurs often in cryptogenic ischaemic stroke. Final results from the SURPRISE study. Eur J Neurol 2014; 21(6):884-9.

Ciconte G, Saviano M, Giannelli L, et al.(2017) 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

Da Costa A, Defaye P, Romeyer-Bouchard C et al.(2013) 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 2013; 106(3):146-54.

Da Costa A, Defaye P, Romeyer-Bouchard C, et al.(2013) 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

Derkac WM, Finkelmeier JR, Horgan DJ, et al.(2017) 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

ECRI Evidence-based Practice Center(2007) Remote cardiac monitoring 12-12-2007. Agency for Healthcare Research & Quality. http://www.cms.hhs.gov/mcd/viewtechassess.asp?where=index&tid=51, accessed 1-14-08.

Etgen T, Hochreiter M, Mundel M et al.(2013) Insertable cardiac event recorder in detection of atrial fibrillation after cryptogenic stroke: an audit report. Stroke 2013; 44(7):2007-9.

Gladstone DJ, Spring M, Dorian P et al.(2014) Atrial fibrillation in patients with cryptogenic stroke. N Engl J Med 2014; 370(26):2467-77.

Halcox JPJ, Wareham K, Cardew A, et al.(2017) 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

Hanke T, Charitos EI, Stierle U et al.(2009) 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 2009; 120(11 Suppl):S177-84.

Hindricks G, Pokushalov E, Urban L et al.(2010) Performance of a new leadless implantable cardiac monitor in detecting and quantifying atrial fibrillation: Results of the XPECT trial. Circ Arrhythm Electrophysiol 2010; 3(2):141-7.

Hoefman E, Bindels PJ, van Weert HC.(2010) Efficacy of diagnostic tools for detecting cardiac arrhythmias: systematic literature search. Neth Heart J 2010; 18(11):543-51.

January CT, Wann LS, Alpert JS et al.(2014) 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 2014a.

January CT, Wann LS, Alpert JS et al.(2014) 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 Practice Guidelines and the Heart Rhythm Society. Journal of the American College of Cardiology 2014b.

Kalani R, Bernstein R, Passman R, et al.(2015) 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

Kapa S, Epstein AE, Callans DJ et al.(2013) Assessing arrhythmia burden after catheter ablation of atrial fibrillation using an implantable loop recorder: the ABACUS study. J Cardiovasc Electrophysiol 2013; 24(8):875-81.

Kishore A, Vail A, Majid A et al.(2014) Detection of atrial fibrillation after ischemic stroke or transient ischemic attack: a systematic review and meta-analysis. Stroke 2014; 45(2):520-6.

Krahn AD, Klein GJ, Norris C, et al.(1995) The etiology of syncope in patients with negative tilt table and electrophysiological testing. Circ 1995; 92:1819-182.

Krahn AD, Klein GJ, Yee R.(1997) Recurrent Syncope. Experience with an implantable loop recorder. Card Clin 1997; 2:313-326.

Locati ET, Vecchi AM, Vargiu S et al.(2014) Role of extended external loop recorders for the diagnosis of unexplained syncope, pre-syncope, and sustained palpitations. Europace 2014; 16(6):914-22.

Mittal S, Sanders P, Pokushalov E, et al.(2015) Safety profile of a miniaturized insertable cardiac monitor: results from two prospective trials. Pacing Clin Electrophysiol. Dec 2015;38(12):1464-1469. PMID 26412309

Narasimha D, Hanna N, Beck H, et al.(2018) Validation of a smartphone-based event recorder for arrhythmia detection. Pacing Clin Electrophysiol. Mar 1 2018. PMID 29493801

Nolker G, Mayer J, Boldt LH, et al.(2016) 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

Podoleanu C, DaCosta A, Defaye P, et al.(2014) 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

Ritter MA, Kochhauser S, Duning T et al.(2013) Occult atrial fibrillation in cryptogenic stroke: detection by 7-day electrocardiogram versus implantable cardiac monitors. Stroke 2013; 44(5):1449-52.

Sanders P, Purerfellner H, Pokushalov E, et al.(2016) Performance of a new atrial fibrillation detection algorithm in a miniaturized insertable cardiac monitor: Results from the Reveal LINQ Usability Study. Heart Rhythm. Mar 4 2016. PMID 26961298

Sanna T, Diener HC, Passman RS et al.(2014) Cryptogenic stroke and underlying atrial fibrillation. N Engl J Med 2014; 370(26):2478-86.

Shen WK, Sheldon RS, Benditt DG, et al.(2017) 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

Sinha AM, Diener HC, Morillo CA et al.(2010) Cryptogenic Stroke and underlying Atrial Fibrillation (CRYSTAL AF): design and rationale. Am Heart J 2010; 160(1):36-41 e1.

Solbiati M, Casazza G, Dipaola F, et al.(2017) 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

Sposato LA, Cipriano LE, Saposnik G, et al.(2015) 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

Strickberger SA, Benson DW, et al.(2006) AHA/ACCF scientific statement on the evaluation of syncope. J Am Coll Cardiol, 2006; 47:473-84.

Ziegler PD, Rogers JD, Ferreira SW, et al.(2015) Real-world experience with insertable cardiac monitors to find atrial fibrillation in cryptogenic stroke. Cerebrovasc Dis. 2015;40(3-4):175-181. PMID 26314298


Group specific policy will supersede this policy when applicable. This policy does not apply to the Wal-Mart Associates Group Health Plan participants or to the Tyson Group Health Plan participants.
CPT Codes Copyright © 2019 American Medical Association.