Author + information
- Received June 19, 2018
- Revision received September 11, 2018
- Accepted September 19, 2018
- Published online February 18, 2019.
- Vassilios J. Bezzerides, MD, PhDa,
- Amy Walsh, RNa,
- Maria Martuscello, RNa,
- Carolina A. Escudero, MDb,
- Kimberlee Gauvreau, ScDa,
- Geralyn Lama,
- Dominic J. Abrams, MBBSa,
- John K. Triedman, MDa,
- Mark E. Alexander, MDa,
- Laura Bevilacqua, MDa and
- Douglas Y. Mah, MDa,∗ ()
- aElectrophysiology Division, Department of Cardiology, Boston Children’s Hospital, Boston, Massachusetts
- bDepartment of Pediatrics, Harvard Medical School, Boston, Massachusetts
- ↵∗Address for correspondence:
Dr. Douglas Y. Mah, Department of Cardiology, Boston Children’s Hospital, 300 Longwood Avenue, Boston, Massachusetts 02115.
Objectives This study sought to determine the practical use of the recently introduced LINQ implantable loop recorder (LINQ-ILR) in a cohort of pediatric and adult congenital arrhythmia patients.
Background Correlating symptoms to a causative arrhythmia is a key aspect of diagnosis and management in clinical electrophysiology.
Methods Retrospective review of clinical data, implantation indications, findings, and therapeutic decisions in patients who underwent LINQ-ILR implantation from April 1st, 2014 to January 30th, 2017 at Boston Children’s Hospital.
Results A total of 133 patients were included, of which 76 (57%) were male. The mean age at implantation was 15.7 ± 9.1 years with a duration of follow-up of 11.8 months. Congenital heart disease was present in 34 patients (26%), a confirmed genetic diagnosis in 50 (38%), and cardiomyopathy in 22 (26%), and the remainder were without a previous diagnosis. Syncope was the most common indication for LINQ-ILR implantation, occurring in 59 patients (44%). The median time to diagnosis was 4.5 months, occurring in 78 patients (59%). Cardiac device placement occurred in 17 patients (22%), a medication change in 9 (12%), electrophysiology study/ablation in 5 (6%), or LINQ-ILR explantation in 42 (54%). Infection or erosion occurred in 5 patients. Syncope was correlated with a diagnostic transmission (54% vs. 31%, p = 0.01).
Conclusions The LINQ-ILR is an important diagnostic tool, providing useful data in more than one-half of patients in <6 months. Adverse events are low. Patient selection is critical and undiagnosed syncope represents an important presenting indication for which a LINQ-ILR implant should be considered.
The correlation of symptoms to a causative arrhythmia is critical for patient diagnosis and management in clinical electrophysiology (EP). This is especially important in pediatric patients and adults with congenital heart disease (ACHD) where associated arrhythmias can complicate their clinical management (1–3). Short-term noninvasive monitoring can be effective in correlating clinically relevant arrhythmias with symptoms, but it may be limited in situations where symptoms are infrequent. Both adult- and pediatric-focused series demonstrate markedly decreased yield with these devices after 2 weeks of monitoring (4). Implantable loop recorders (ILR) can be an effective strategy in detecting paroxysmal arrhythmias in both pediatric (5) and adult patients (6–8).
The Reveal LINQ (Medtronic Inc., Minneapolis, Minnesota) was approved in the United States in February 2014. With a volume of approximately 1 cc, it represented a significant reduction in size from previous models (9). This decrease in device size reduces procedural and recovery time as well as potentially expanding the patient population eligible for implantation (9). This is especially relevant for younger pediatric patients where the larger size of previous ILR limited its use despite the potential to refine clinical decision making in the setting of concerning symptoms such as atypical syncope (10). Because of increased complications in pediatric patients with implantable cardioverter-defibrillators (ICD) (11,12), ILR have the potential to improve patient selection for ICD or other therapies and to improve clinical outcomes.
Implantation of ILR has proven an effective strategy in the management of pediatric patients with inherited arrhythmias (13), but little is known about the effectiveness of LINQ-ILR in arrhythmia management of patients with more diverse underlying diagnoses or with ACHD. The goal of this study was to determine the use of the new Medtronic LINQ-ILR in the clinical diagnosis and management of pediatric and adult patients in a real-world setting.
After obtaining approval from the Institutional Review Board at Boston Children’s Hospital, a retrospective chart review was undertaken. Patients who underwent ILR implantation with a Medtronic LINQ-ILR from March 2014 to January 2017 were identified from hospital records. All patients had undergone at least comprehensive ambulatory cardiac evaluation for their symptoms including echocardiography and exercise testing at appropriate ages. Some patients underwent previous attempts at arrhythmia diagnosis with looping or event recorders at the discretion of the treating physician. Patients with atypical syncope did not routinely undergo previous monitoring with event recorders, as this is not standard practice in our institution. The judgment to insert an ILR was made in collaboration with 1 of the 6 staff pediatric electrophysiologists. Data collected included patient demographics, underlying genetic diagnosis, the presence of congenital heart disease, structural heart disease, or cardiomyopathy. Additional data collected included categorization of diagnosis, current treatment, complications, time to diagnostic transmission, recorded diagnosis, and change in patient management. Transmissions were reviewed daily by 2 experienced nurses (M.M., A.M.) and possibly diagnostic recordings were flagged for additional review by the treating or consulting electrophysiologist. Transmissions were determined to be diagnostic by the treating clinician and confirmed by reviewing the original transmission. R-wave recordings from the first post-implantation transmission were analyzed for total R-wave amplitude by digitization and calibrated analysis using ImageJ (National Institutes of Health, Bethesda, Maryland). At least 5 QRS waveforms for each transmission were averaged and reported as a mean ± SE.
Results are expressed as percentages, with mean ± SD for continuous variables unless otherwise stated. The interquartile ranges (IQR) are included for median values. Differences between groups were evaluated using Fisher exact test for categorical variables, the unpaired Student t-test for continuous variables with a normal distribution, and the Wilcoxon rank sum test for skewed continuous variables.
A total of 133 patients were included in the study. Baseline patient characteristics and indications for implantation are listed in Table 1. The average age at implantation was 15.7 ± 9.1 years (3.6 months to 44.6 years), with 11 patients (8.3%) under the age of 5 years. The largest proportion of patients had an underlying genetic diagnosis (n = 50, 38%) with additional diagnoses of congenital heart disease (n = 34, 26%), and cardiomyopathy/ventricular dysfunction (n = 22, 17%). Of the underlying genetic diagnoses, channelopathies were the most common with a total of 26 patients: long QT syndrome (n = 15), catecholaminergic polymorphic ventricular tachycardia (n = 9) and Brugada syndrome (n = 2). Cardiomyopathies included arrhythmogenic cardiomyopathy (n = 7), hypertrophic cardiomyopathy (n = 6), and dilated cardiomyopathies (n = 2). In terms of symptom-based indications for LINQ implantation, syncope was the most common indication occurring in 59 patients (44%), with 21 patients (16%) presenting with palpitations. Other indications for LINQ-ILR implantation are listed in Table 1.
Conscious sedation was the most common form of anesthesia during LINQ-ILR implantation (n = 66, 50%), with general anesthesia in 41 patients (31%) and local anesthesia only in 21 patients (16%). LINQ-ILR implantations were also routinely performed in conjunction with other EP procedures, such as ablations, under a single anesthetic (n = 39, 29%). This often occurred in the setting of a negative EP study or incomplete ablation with frequent clinical symptoms. LINQ-ILR implantations were performed in nonstandard sites such as the axilla (n = 41, 31%), which occurred more frequently in younger patients (axillary location, mean age = 142 ± 17 months vs. precordial location, mean age = 206 ± 12 months, p < 0.005).
Diagnostic yield and outcomes
The median follow-up was 11.8 months (IQR: 5.6 to 20.6 months, N = 133). A total of 78 patients recorded diagnostic transmissions with a positive diagnosis made in 31 patients (40%) and a negative diagnosis made in 47 patients (60%) (Table 2). The median time to either a positive or negative diagnostic transmission was 4.5 months post-implantation (IQR: 1.7 to 13.7 months). Diagnostic transmissions included both symptom-triggered and device-triggered arrhythmias (Figure 1). The diagnosis changed management in 36 patients, with resultant pacemaker or ICD implantation in 17 patients (22%), as shown in Table 3. The remaining 55 patients had a pending diagnosis and were still being actively monitored at the conclusion of the study.
Details regarding clinical characteristics and the diagnostic ILR transmission in patients who underwent ICD or pacemaker implantation are provided in Table 4. In the 10 patients who underwent pacemaker implantation, 7 had symptomatic sinus pauses all >10 s in duration with symptomatic sinus node dysfunction in 2 patients. One patient demonstrated intermittent complete heart block in the setting of tetralogy of Fallot. An additional patient with chronic dual-chamber pacing recorded several episodes of nonsustained ventricular tachycardia in the setting of worsening ventricular function, so an additional ventricular lead was inserted. ICDs were inserted in 7 patients with previously placed ILRs (Table 4). Ventricular arrhythmias were recorded in 6 of these patients including Torsades de pointes, monomorphic ventricular tachycardia, and nonsustained ventricular tachycardia. Spontaneous atrioventricular block was recorded in 1 patient with syncope in the setting of hypertrophic cardiomyopathy and 2 patients had increasing ventricular ectopy that prompted ICD implantation in the setting of ventricular dysfunction.
A diagnosis obtained from an ILR transmission prompted a medication change in 9 patients with a referral for EP study in 5 patients. In 42 patients, the treating electrophysiologist determined that no further intervention was necessary and the LINQ-ILR was removed or the patient was referred for explantation during the study period.
We analyzed patient characteristics that were associated with receiving a diagnostic transmission after ILR implantation. The presence of congenital heart disease, an underlying genetic diagnosis, or cardiomyopathy were not associated with diagnostic transmissions (Table 2). Those patients with a positive diagnostic transmission were older than those with a negative diagnostic transmission (18 ± 11.3 years vs. 13.2 ± 7.1 years, p = 0.04). Analysis of implant indications revealed that only syncope was associated with a diagnostic transmission during the study period, with syncope in 42 patients (54%) versus 17 (31%) without syncope, giving an odds ratio of 3.11 (95% confidence interval: 1.52 to 6.32; p = 0.002). Interestingly, in those patients with syncope where a diagnostic transmission was received, there was no difference between the proportions of patients with a positive or negative diagnosis (Table 2).
Implant location and complications
In addition to the traditional precordial location, we included patients with LINQ implants in nontraditional locations such as the axilla. A total of 41 patients (31%) had an axillary implant with 84 patients (63%) undergoing traditional precordial implantation. Although there was no age or weight dependence, axillary implants were more common in female patients (27 of 41 [66%]) versus male patients (14 of 41 [34%]; p = 0.004 for both) (Table 3). R-wave measurements of representative transmissions revealed a significant reduction in R-wave amplitude from axillary implants as compared to those in the precordium (0.75 ± 0.08 mV [SE, n = 32] vs. 1.3 ± 0.09 mV [SE, n = 75], respectively; p < 0.001) (Figure 2). In addition, the frequency of patients with under- or oversensing by the LINQ algorithm resulting in falsely classified abnormal recordings occurred entirely in patients with axillary implants as compared to precordial implants (11 of 32 [34%] vs. 0 of 75 [0%]; p < 0.001). These oversensing and undersensing events often occurred in the same patient, falsely indicating both high-rate and pause events.
There were no acute complications during implantations in our patient cohort. There were 6 patients with ILR-related complications that required explantation of the device. This included 3 patients with infected devices that did not respond to antibiotics, 2 patients with device erosion, and 1 device removed because of excessive pain at the site of implant. Complications occurred much more frequently in patients <5 years of age at the time of implantation than the remainder of the cohort (3 of 11 [27%] vs. 2 of 121 [2.5%]; p = 0.003), giving an odds ratio of 14.8 (95% confidence interval: 2.56 to 85.2; p = 0.003). All 6 devices were successfully removed without any residual infection or additional complications, with 1 device that was reinserted.
This large pediatric study examines the use of the Medtronic LINQ-ILR in pediatric and ACHD patients. Because of the smaller size of the LINQ-ILR device, our cohort included a higher percentage of younger patients and inclusion of a novel implant site in the axilla. Similar to other studies (5,14), the pre-implantation symptom of syncope was strongly associated with a confirmatory diagnosis made by ILR recordings during the study period, but did not predict a positive diagnosis. The average time to diagnosis was just under 5 months, which suggests that clinicians regularly identify patients with concerning symptoms where more time-limited monitoring has been ineffective or is appropriately judged to be a waste of resources. Making a diagnosis by ILR recording, either positive or negative, changed management in most patients and prompted implantation of either a pacemaker or ICD in nearly one-quarter of patients with a diagnosis. This is particularly important given the desire to refine the selection criteria for device implantation in pediatric patients, due to the increased risk of complications (11).
We also examined the outcomes in patients with atypical implant sites such as the axilla, which were performed in smaller patients or for patient-driven aesthetic concerns. Whereas there was no difference in the diagnostic yield in axillary implanted LINQ as compared to those inserted in the precordial position, the risk of transmissions with significant noise, resulting in false recordings was significantly higher in patients with axillary implants and therefore increased the need for review by clinical staff. In addition to the axillary implantation, our study included patients <5 years of age, an age group that has not been previously studied (5,9,13). Whereas the diagnostic yield and refinement of clinical management was similar to older patients, the risk for complications including erosion and infection was significantly higher. Considerable caution must be exercised when considering ILR implantation in younger or smaller patients. Although surgical closure in multiple layers may reduce the risk of complication, the benefit of this technique was not examined in this study and could be an area for further research. However, implantation of a LINQ-ILR in a younger patient may still be indicated, especially in the case of atypical syncope or underlying genetic diagnosis with negative cardiovascular and neurological workups.
Similar to other studies, a pre-implantation symptom of atypical syncope was strongly associated with a diagnostic recording during the study period. However, syncope did not predict a positive diagnosis. The lack of recorded arrhythmia during a patient’s characteristic symptoms can be extremely valuable and prevents unnecessary therapies or procedures. Our data suggest that a change in management occurred in a majority of patients and that most diagnoses could be made in <5 months. This diagnostic determination includes a negative diagnosis and possible discharge from routine cardiology follow-up. In addition to making a specific positive or negative symptom-rhythm correlation, LINQ-ILRs were also inserted for long-term monitoring of patients with underlying arrhythmogenic diagnoses, such as long QT syndrome or hypertrophic cardiomyopathy. In these patients, LINQ-ILR implantation is used to titrate medication, monitor the progression of a known arrhythmia diagnosis, or differentiate benign syncope from arrhythmogenic syncope, potentially leading to a change in clinical management.
Limitations of this study include the single-center design and the associated possible limited application to other populations. Another limitation is that this analysis was a retrospective review. A prospective study could further define which patients would benefit from LINQ implantation, such as those with syncope, and those who should undergo other diagnostic or therapeutic procedures such as EP study.
This large retrospective study has examined the practical use of the LINQ-ILR in both pediatric and ACHD patients. We have demonstrated that the diagnostic yield is highest in patients presenting with atypical syncope. The LINQ-ILR can be used in younger patients but caution must be exercised as there is an increased risk of complications. In addition, atypical implant sites in the axilla are associated with false recordings and should be avoided unless the benefit significantly outweighs the risk. Overall, the LINQ-ILR is a valuable diagnostic and clinical management tool for the pediatric and adult congenital electrophysiologist.
COMPETENCY IN MEDICAL KNOWLEDGE: ILR implantation in specialized populations such as children and AHCD can be effective in diagnosing clinically relevant arrhythmias in a variety of situations. The risk of infection and erosion is more common in younger patients. Atypical implantation location such as in the axilla can increase the risk of inaccurate sensing. Careful patient selection before implantation is critical to increase the effectiveness and reduce complications with LINQ-ILR use.
TRANSLATIONAL OUTLOOK: While the LINQ-ILR can be an effective tool for arrhythmia diagnosis in younger patients, additional caution is necessary to avoid complications. Developing methods to avoid this complications, including wound closure techniques, should be areas for further research.
The authors have reported that they have no relationships relevant to the contents of this paper to disclose.
All authors attest they are in compliance with human studies committees and animal welfare regulations of the authors’ institutions and Food and Drug Administration guidelines, including patient consent where appropriate. For more information, visit the JACC: Clinical Electrophysiology author instructions page.
- Abbreviations and Acronyms
- adults with congenital heart disease
- implantable cardioverter-defibrillator
- implantable loop recorder
- interquartile range
- Received June 19, 2018.
- Revision received September 11, 2018.
- Accepted September 19, 2018.
- 2019 American College of Cardiology Foundation
- Walsh E.P.,
- Cecchin F.
- Pierre B.,
- Fauchier L.,
- Breard G.,
- Marie O.,
- Poret P.,
- Babuty D.
- Nguyen H.H.,
- Law I.H.,
- Rudokas M.W.,
- Lampe J.,
- Bowman T.M.,
- Van Hare G.F.,
- Avari Silva J.N.
- Paris Y.,
- Toro-Salazar O.H.,
- Gauthier N.S.,
- et al.,
- for New England Congenital Cardiology Association
- DeWitt E.S.,
- Triedman J.K.,
- Cecchin F.,
- et al.
- Maron B.J.,
- Spirito P.,
- Ackerman M.J.,
- et al.
- Avari Silva J.N.,
- Bromberg B.I.,
- Emge F.K.,
- Bowman T.M.,
- Van Hare G.F.
- Rossano J.,
- Bloemers B.,
- Sreeram N.,
- Balaji S.,
- Shah M.J.
- Sakhi R.,
- Theuns D.A.M.J.,
- Bhagwandien R.E.,
- et al.