Single-centre use of implantable loop recorders in patients with congenital heart disease

Abstract

Aims

To determine the efficacy of the implantable loop recorder (ILR) in patients with congenital heart disease.

Methods and results

Retrospective data analyses of all patients with congenital heart disease undergoing ILR implantation in a congenital cardiac centre. Of 78 ILRs implanted between July 2000 and October 2007, 22 (15 male) patients had congenital heart disease. The median age at implantation was 15.5 years (7.0–46.5). Six patients had neurodevelopmental delay. Indications for implantation were syncope ( n = 15), palpitations ( n = 4), and pre-syncope ( n = 3). Median implantation time was 19 months (1–30). Explantation data were available in 16 patients (73%). Implantable loop recorders were explanted following establishment of positive diagnosis ( n = 3), negative diagnosis ( n = 7), and at the end of battery life ( n = 4). One patient died with the ILR in situ . One device eroded after 2 months without further complication. Following ILR interrogation, diagnosed arrhythmias included ventricular tachycardia ( n = 2) and sinus node dysfunction ( n = 1). On the basis of these findings, two patients had intracardiac defibrillators and one patient had a dual-chamber pacemaker implanted.

Conclusion

Implantable loop recorders lead to a positive or negative arrhythmic diagnosis following explantation in 71% of patients with congenital heart disease. This is particularly beneficial in patients with neurodevelopmental delay where accurate history is often limited.

Introduction

Patients with treated and untreated congenital heart disease have an increased incidence of malignant arrhythmias. ^1–3 These troublesome and often dangerous rhythm disturbances arise due to surgical scars, extensive suture lines, and abnormal haemodynamics. Advances in interventional treatment strategies such as radiofrequency catheter ablation and the implantable defibrillator provide effective treatment for malignant tachycardias. ^{4 , 5} Similarly modern pacing strategies for sinus node dysfunction and atrioventricular block are targeting the long-term complications previously seen with endocardial pacing systems. ⁶ However, ascertaining which patients are suffering from a symptomatic arrhythmia may be difficult. The implantable loop recorder (ILR) has been shown to be effective in diagnosing syncope in both an unselected population and adults with structural heart disease. ^{7 , 8} However, there is only one series (four patients) reporting the use of the ILR in patients with congenital heart disease ⁹ with other authors reporting small numbers of patients with congenital heart disease as part of an overall paediatric cohort. ^10–14 We report on the efficacy of ILRs in 22 children and adults with congenital heart disease.

Methods

Implantation

Two investigators performed device implantation (Reveal ^® or Reveal ^® Plus Medtronic, Inc., Minneapolis, MN, USA) in a dedicated cardiac catheterization laboratory based in a regional congenital heart centre. The indication for device implantation was recurrent syncope, near syncope, or recurrent palpitations that remained undiagnosed after standard tests including 24 h Holter monitor ( n = 17) and/or exercise testing ( n = 13). General anaesthesia was used in 17 patients. Eight patients had an additional procedure. Implantation site was prepectoral or in the left axilla with the site of implantation guided by the patient's preference. Standard ECG surface mapping was carried out before implantation. ¹¹ After the skin was infiltrated with local anaesthetic, a 2.5 cm incision was made and a subcutaneous or subpectoral pocket fashioned by blunt dissection. The device was anchored with non-absorbable sutures and the skin wound closed with a continuous subcuticular suture. The device was set to record three patient activations and five automatic activations with automatic activation parameters set to detect bradycardia <40/min, tachycardia >220/min, and pauses >4.5 s in all cases.

The patient along with a family member was instructed in the use of the activator prior to discharge. The device remained in situ until a positive diagnosis or negative diagnosis (symptoms without any rhythm disturbance) was reached or at the end of battery life. All patients completed a 5 day course of oral antibiotics following implantation.

Data collection

Data were obtained by retrospective review of the case records, unit database (HeartSuite), and cardiac catheterization records. We identified all patients undergoing ILR implantation and selected those with congenital heart disease including patients with congenital cardiomyopathy for review.

Results

Seventy-eight ILRs were implanted (77 patients) between July 2000 and October 2007. Of these, 22 patients (15 male) had congenital heart disease ( Table  1 ).

Fourteen patients had undergone cardiac surgery. Six patients had neurodevelopmental delay including trisomy 21 ( n = 4) and Barth syndrome ( n = 1) with one further patient having neurodevelopmental delay of unknown origin. Fifteen patients were <18 years. Median age at implantation was 15.5 years (7.0–46.5). Median weight was 51.3 kg (21.4–104). The ILR was implanted in the left infraclavicular region ( n = 18) and the left axillary position ( n = 4). Additional procedures at implantation included diagnostic cardiac catheterization ( n = 6), electrophysiology study ( n = 3), and inter-atrial baffle angioplasty post-Mustard operation ( n = 1). The median duration of implantation was 19 months (1–30). Explantation data were available on 16 patients (73%) with six devices still in situ . Patient 2 died suddenly 1 month following implantation of the device. Interrogation of the device following explantation demonstrated ventricular fibrillation at the time of death. ¹⁵ This patient had undergone previous Mustard procedure and was paced with a femoral venous pacemaker due to heart block. He was having syncope; however, pacemaker interrogation was repeatedly normal. Exercise test demonstrated occasional ventricular ectopics. It was uncertain whether his symptoms were due to intermittent lead microdisplacement or a ventricular dysrhythmia. The patient had poor haemodynamics and was advised to consider transplant assessment but declined. There was no evidence of a ventricular arrhythmia and thus an ILR was implanted in an attempt to further risk stratify prior to considering an ICD which would have been technically difficult due to lack of venous access (occluded baffle) and likely to be associated with significant morbidity. The device in Patient 13 (neurodevelopmental delay) eroded within 2 months of implantation. There were no other device-related complications.

The median number of ILR downloads was 3 (range 1–11). There was a median of 2 (range 0–10) patient-activated events and a median of 5 (range 0–35) automatic activated events ( Table  2 ). Three patients did not have an event while the ILR was implanted. One patient had an episode of palpitations but did not activate the device and the device did not auto-activate.

Five patients had artefact on at least one recording. Two patients had auto-activation of the device with subsequent downloads demonstrating ‘drop-out’ or loss of signal ( Figure  1 ). Three patients had auto-activation due to artefact from oversensing chest wall muscle potentials. None of these patients had symptoms during these auto-activations.

Following explantation, a positive arrhythmic diagnosis was reached in three patients. Patient 7 had episodes of bradycardia with low atrial rates during syncopal episodes. He had a dual chamber pacemaker implanted and has not had further syncopal episodes. Patient 16 had monomorphic broad complex tachycardia on ILR during symptoms. This was confirmed on subsequent electrophysiology study (EPS) as ventricular tachycardia originating from the right ventricular outflow tract. He underwent implantation of a pacemaker-defibrillator, but required frequent overdrive pacing for recurrent ventricular tachycardia and subsequently underwent a successful radiofrequency ablation of a right ventricular outflow tract tachycardia. Patient 19 also had monomorphic broad complex tachycardia on ILR during symptoms with confirmed ventricular tachycardia during EPS and had an implantable pacemaker-defibrillator inserted. He has had no further symptoms to date and no discharges from his defibrillator. Patient 6 had suspected diagnosis of atrioventricular nodal re-entry tachycardia (AVNRT) following initial corrective surgery for tetralogy of Fallot; however, EPS did not provoke tachycardia (although there was evidence of dual atrioventricular node physiology). An ILR inserted at the end of the EPS has demonstrated a narrow complex tachycardia with subsequent 12-lead ECG again suggesting AVNRT. Overall monitoring of the frequency of the events on ILR has been useful in guiding medical management as the patient has not always been aware of episodes of tachycardia.

Seven patients had a negative arrhythmic diagnosis with no rhythm abnormality during symptoms. Six of these have been reassured and have remained well including one patient with mild sinus node dysfunction. Patient 12 had hypertrophic cardiomyopathy with a strong family history of sudden death. Although the ILR did not demonstrate any arrhythmia, an implantable defibrillator has subsequently been inserted based on his risk stratification. Patient 1 did not have any events during the 27 months of implantation; however, he had a repeat collapse 4 years after explantation with evidence of non-sustained ventricular tachycardia. An implantable defibrillator has now been inserted. Of the six devices still in situ , four patients have had symptoms without rhythm disturbance and one patient has had no symptom recurrence.

Discussion

Patients with congenital heart disease have an increased prevalence of both atrial ^{2 , 16} and ventricular tachycardia. ^{3 , 17} These arrhythmias are frequently associated with syncope. ^{18 , 19} Determining whether onset of new symptoms in such patients is significant is difficult, but must be taken seriously, considering the high prevalence of arrhythmia and the potentially devastating consequences. The diagnostic yield from conventional methods of evaluation such as external recorders and exercise testing is low. ²⁰ The ILR provides an excellent tool for determining whether symptoms are associated with rhythm disturbance and also in identifying the type of arrhythmia and hence guide further therapy. Thus, the ILR formed part of a risk stratification in these patients and was particularly useful in the two patients with documented ventricular tachycardia. Implantation of an ICD in children and adolescents is not without morbidity and should only be considered when there is a documented ventricular arrhythmia. ICD implantation may be technically difficult as outlined by Patient 2. This patient with previous Mustard procedure had an occluded baffle and therefore no venous access. An epicardial system in a patient with multiple previous sternotomies was not straightforward, and therefore, following extensive discussion, we implanted an ILR to investigate his symptoms further before subjecting him to a higher risk procedure. We did not implant an ILR if there was evidence of life-threatening arrhythmias or if we judged the cause of the symptoms could be established by non-invasive means.

The device is easy to use, well tolerated, and despite previously reported high complication rates in children, ¹³ the complication rate in our series was low. The ILR is also attractive in patients with congenital heart disease and associated chromosomal abnormalities such as trisomy 21 when a detailed history may be absent and compliance with further investigations limited. Six patients (27%) in our cohort had neurodevelopmental delay and positive arrhythmic diagnosis has been achieved in one patient to date. This patient had an endocardial pacemaker implanted with resolution of syncope. Negative arrhythmic diagnosis has allowed reassurance in another with resolution of symptoms. Three devices are still in situ . Two of these patients have had further symptoms (syncope n = 1 and palpitations n = 1), but the ILR readings have demonstrated sinus tachycardia only. This has allowed the families to be reassured. In a 27-year-old patient with trisomy 21, the disappearance of symptoms coincided with the insertion of the ILR and the device has been left in place. The patient attributes his improvement to the device and his family has requested that it be left in situ wishing to take advantage of this placebo effect. The device in the sixth patient eroded and self-explanted.

We have established either positive or negative arrhythmic diagnoses in 10 of the 14 devices electively explanted (71%). Previous studies looking at this selected subgroup of patients report positive and negative diagnostic yields of between 42 and 75%, although numbers have been limited. ^{10 , 13} Four patients (27%) became symptom-free after implantation of the ILR and this is lower than previously reported event-free rates following implantation in both children (44%) and adults (57%). ^{7 , 13} Of these, only one patient has had further symptoms 4 years after explantation of the ILR, requiring an implantable defibrillator.

It is our departmental protocol to use general anaesthesia for ILR implantation in children, although implantation under local anaesthesia has been described. ¹³ These patients are admitted as day cases with discharge home the evening of the procedure. We also used general anaesthesia in adult patients with neurodevelopmental delay and when planning an associated procedure. Two patients had an ILR implanted after a negative EPS and the diagnostic potential of ILR in this subgroup has been described before. ²¹ This was confirmed in one patient in our series with multiple episodes of tachycardia noted on ILR, despite negative EPS. This has been useful in guiding therapy.

Prior to implantation, the optimal position of the device is discussed with the patient. In general, this is sited in the prepectoral position in a subcutaneous pocket. In patients who participate in contact sport, the device is placed in a subpectoral position. This position may also be preferable in patients with neurodevelopmental delay to provide added protection against trauma. The device that eroded was in a subcutaneous position in a patient with who repeatedly traumatized the wound insertion site and based on this experience, we would currently recommend subpectoral implantation for such patients. In younger children and in older girls who prefer to avoid a prepectoral scar, the device is placed in a left axillary position. This was not associated with increased artefact in the four patients with axillary ILRs from our cohort. Sensing artefact is reported to be common with 89% of children in one study having false-positive activations as a result of either artefact or sinus tachycardia. ¹¹ We have also noted ‘drop-out’ during auto-activation in two patients in our cohort and it can be challenging to differentiate this from asystole. Failure of auto-activation on three occasions in a patient with polymorphic ventricular tachycardia has also been reported and is more concerning. ¹¹ An updated device (Reveal ^® DX, Medtronic, Inc.) promoting modifications to the QRS detection algorithms leading to more advanced sensing and detection is now available. Reports are awaited on whether this will improve both false-positive and false-negative detection rates.

Implantable loop recorders are safe and useful in the management of suspected arrhythmias in selected children and adults with congenital heart disease leading to positive or negative arrhythmic diagnoses in 71% and potentially life-saving therapy (pacemaker/defibrillator implantation) in 20% of patients. The ILR is a useful tool to prevent unwanted therapeutic interventions and provide reassurance to the patient with non-cardiogenic syncope spectrum. This is of particular value in patients with neurodevelopmental delay where diagnosis with non-invasive investigations may be difficult.

Conflict of interest: none declared.

References