Author + information
- Received October 6, 2016
- Revision received March 5, 2018
- Accepted March 15, 2018
- Published online July 16, 2018.
- Thomas Kriebel, MDa,∗ (, )
- Matthias J. Müller, MDa,
- Wolfgang Ruschewski, MDb,
- Ulrich Krause, MDa,
- Thomas Paul, MDa and
- Heike Schneider, MDa
- aDepartment of Pediatric Cardiology and Intensive Care Medicine, George August University Medical Center, Göttingen, Germany
- bDepartment of Thoracic and Cardiovascular Surgery, Georg August University Medical Center, Göttingen, Germany
- ↵∗Address for correspondence:
Dr. Thomas Kriebel, Chefarzt der Klinik für Kinder- und Jugendmedizin und Kinderkardiologie, Westpfalz-Klinikum Kaiserslautern, Hellmut-Hartert-Strasse 1,67655 Kaiserslautern, Germany.
Objectives The purpose of this study was to analyze course of defibrillation threshold (DFT) with growth.
Background Data on regular DFT testing after extracardiac implantable cardioverter-defibrillator (ICD) placement in infants and small children is still limited.
Methods An extracardiac ICD was placed in 23 pediatric patients (median age 6.1 years; median body weight 21 kg, median length 120 cm). The defibrillator lead was tunneled pleurally, and the device was placed as “active can” in the right upper abdomen or in a horizontal position between the diaphragm and the pericardium, respectively. DFT was verified intraoperatively, 3 months later, and every 12 months thereafter. The aim was to achieve DFT <15 J allowing ICD programming with a double safety margin above DFT.
Results In all 23 patients, an intraoperative DFT <15 J could be accomplished. Serial DFT testing showed an increase from a median DFT of 10 J intraoperatively to 15 J after 1 year. During mean follow-up of 2.0 years, a significant correlation between DFT and body length, but not body weight, was observed. In 4 of 23 (17%) patients, surgical revision was required because of a DFT increase >20 J during regular DFT testing. No complications regarding DFT testing were noted.
Conclusions After extracardiac ICD placement in infants and small children, DFT increase related to body length was evident during mid-term follow-up. Routine serial DFT testing was a safe procedure and identified a significant DFT increase in 4 of 23 patients. Serial DFT testing during follow-up in these patients is recommended.
- infants and children
- implantable cardioverter defibrillator
- extracardiac technique
- defibrillation threshold
- pleural shock elect
Until now, placement of implantable cardioverter-defibrillators (ICDs) in children and adolescents was a rare procedure compared with its more frequent use in the adult population. However, despite a 3-fold increase in pediatric ICD placement from 1997 to 2006 (1), ICD use in children with primary electrical disease has tapered off significantly over the past decade owing to many factors, but the most important is that we know how to better risk-stratify these children. ICD placement in infants and small children accounts for <1% of total ICDs (2).
The majority of ICD placements in older children and adolescents with biventricular hearts have devices implanted using the transvenous route. There is, however, still no established concept for implantation in infants and small children. Recently, the subcutaneous ICD (S-ICD) was introduced into clinical practice. Because of the size of the S-ICD, this system is not suitable for use in small children, as there is an increased risk of erosion or wound dehiscence. In addition, sensing problems with inappropriate discharges as well as delayed detection of ventricular fibrillation have been reported (3).
During the last decade, a variety of implantation techniques have been introduced into clinical practice for ICD placement in infants and small children as well as in patients with congenital heart defects lacking venous access to the heart. Up to now, there has been only a limited number of studies covering more than 3 patients (4–10). Surgical techniques applied varied in positioning of the shock electrode and the device and procedures were accomplished by open thoracotomy and minimal invasive procedures, respectively. However, follow-up data after ICD placement using these novel techniques is still limited. In addition, course of DFT has not been reported in the majority of these studies, which is, however, of paramount importance with respect to further growth and safety of the patients, as shift of the electrical field and shock vector may be inferred.
In 2006, we reported our initial data on the extracardiac ICD-placement technique using a subcutaneous position of the shock electrode and insertion of the device within the right anterior abdominal wall (10). During further post-operative course, we had, however, to observe need for surgical revisions in almost all patients (83%) during a mean follow-up of 2.9 years due to dislocation or fracture of the shock electrode. Therefore, in 2009, we changed our implantation technique to a pleural position of the shock electrode, as early data from other centers applying this technique were promising (4,11). Follow-up data, however, are sparse.
The purpose of the current study was therefore to analyze safety and efficacy by serial DFT assessment in a considerable number of patients over mid-term follow-up.
From 2007 to 2014, an ICD was placed using a defibrillation coil in the pleural space in 23 patients with body weight <40 kg, including 3 patients with former subcutaneous systems (10). In these 3 patients, the system had failed because of dislocation and fracture of the ICD coil in 2 patients and battery depletion in the remaining patient. The subcutaneous coil was completely removed in these patients. Median age of the study patients was 6.1 ± 3.3 (0.2 to 11.5) years, median body weight was 21.0 ± 9.6 (4.5 to 39.0) kg, and median body length was 120.0 ± 26.5 (61 to 147) cm. The case report of 1 patient in this series had been published previously (12). The study protocol was approved by the institutional committee of the Göttingen Heart Center.
Patients' diagnoses are illustrated in Figure 1. ICD placement was performed in 12 patients for primary prevention and in 11 patients for secondary prevention, according to current guidelines (13,14). Patients received additional antiarrhythmic medication, depending on their underlying disease.
In all study patients, a modification of the technique as reported by Bauersfeld et al. (4) was applied. During general anesthesia, the defibrillation lead (Medtronic Transvene 6937 SN, 35 or 52 cm, coil length 8 cm; Medtronic Inc., Minneapolis, Minnesota) was inserted into the pleural space via a small left lateral thoracotomy within the 4th intercostal space until almost reaching the vertebral column and was secured with several sutures. The proximal end of the electrode was fixated to the anterior thoracic wall with the anchoring sleeve. Bipolar steroid eluting epicardial sensing and pacing leads (Capsure Epi 4968, 25 or 35 cm, Medtronic Inc.) were sutured to the ventricular myocardium, and atrial myocardium in selected patients who needed dual-chamber pacing and/or sensing.
Until 2008, the generator (Virtuoso VR D 164, Secura VR D 234, Protecta VR D 364 or XT DR 354, Medtronic Inc.) was placed in the right upper abdominal wall as described in our first series (10). Since 2009, the generator was inserted in a horizontal position between the diaphragm and the pericardium (Figure 2) via a subxiphoid incision. In these patients, the shock electrode and the sensing and pacing leads were tunneled under the left costal arch and connected to the device. Finally, the device was fixated at the right or left costal arch to prevent migration.
Intraoperative DFT assessment
Sensing and pacing thresholds, as well as electrode impedances, were determined. The defibrillation electrode was used as cathode and the “active can” device as anode. For the purpose of this study, the aim was to achieve a DFT <15 J, allowing ICD programming with a double safety margin above DFT.
Ventricular tachycardia or fibrillation was induced by ventricular burst pacing or T-wave shocks. DFT was determined starting with 5 J in patients who weighed <10 kg and with 10 J in patients >10 kg. If ventricular tachycardia/fibrillation was terminated twice with this setting, no further testing was performed. Otherwise, energy output was increased in 5-J increments to a maximum of 15 J. If a 15-J shock was not successful, the position of the shock electrode was modified. First shock energy was programmed individually at the level of double DFT subsequently.
Standards of post-operative care of our institution have been described previously (10). Chest X-rays in the posterior-anterior and lateral projections were performed to delineate lead and device positions before discharge and according to clinical demands.
DFT testing and outpatient visits
During post-operative follow-up, patients were seen on a regular basis by their referring pediatric cardiologists or in our outpatient department. Three months after extracardiac ICD placement, DFT was reassessed under sedation with propofol, as described above, starting with the individual DFT as determined at implantation. If DFT was unchanged, value was defined as minimum DFT. Therefore, the “true” DFT was not determined during follow-up. If DFT had remained at a stable value, DFT testing was repeated annually thereafter. Follow-up DFT testing was performed only once to prevent harm to the patient.
If the first shock at the previous DFT was unsuccessful, successive shocks were delivered in 5-J increments. In patients with DFT increases >5 J, retesting of the DFT was performed again after 3 months. If a 20-J shock was unsuccessful, surgical revision was indicated to reestablish a sufficient safety margin, as stated above.
Continuous data were compared using the 2-tailed t-test. Survival probabilities free of need for surgical revision were estimated using the Kaplan-Meier method. Influence of independent parameters (such as length and body weight) on DFT was assessed by a multiple regression model for repeated measures. All analyses were performed using the free software R (version 2.14, The R Foundation for Statistical Computing, Vienna, Austria). A level of significance was set to p < 0.05 for all analyses.
ICD placement could be accomplished successfully according to our protocol in all 23 patients. Median intraoperative DFT was 10 (5 to 15) J. Impedances, sensing, and pacing thresholds were within acceptable limits in all patients. DFT course during follow-up related to body weight and length is displayed in Figure 3A. In summary, median DFT increased from 10 J intraoperatively (n = 23) to 15 J after 1 year (n = 17) and remained stable at 15 J after 3 years (n = 8) (Figure 3B). Using multiple regression analysis no correlation between DFT and body weight (Table 1) but between DFT and body length (Figure 4) as well as body surface area and impedance of the shock electrode was found. However, no correlation was noted between DFT and increase of body weight or length (Table 1). The course of the median impedance of the shock electrode is displayed in Table 2.
In 4 of 23 patients (long QT syndrome n = 3, catecholaminiergic polymorphic ventricular tachycardia n = 1), routine DFT testing identified a DFT increase >20 J, requiring surgical revision (Table 3). Impedance of the shock electrode or biplane chest X-ray did not provide any evidence for DFT increase in these 4 patients. In 2 of 4 patients, the device was repositioned from the right abdominal wall to a subcardiac position. In the third patient, position of the subcardiac device was modified to a left-sided position to optimize the vector for defibrillation (Figure 5), resulting in a DFT <10 J. In the fourth patient, no DFT <15 J could be achieved, despite repeated repositioning of the shock electrode and the device. Finally, after implantation of an additional intrapericardial electrode, a DFT <10 J could be achieved.
During a mean follow-up of 2.0 ± 1.6 (0.25 to 5.20) years, surgical revision was required in a fifth patient, owing to inadequate sensing parameters. New epicardial sensing and pacing leads were subsequently implanted. Surgical revisions in these 5 patients (22%) were needed after a mean interval of 1.1 (0.25 to 2.00) years after implantation.
Safety of serial DFT testing and loss of battery capacity
In the 23 patients, a total of 87 DFT testings were performed during the study period. Significant complications did not occur in any of the patients. Mean charging time was 51 ± 24 s/year with serial DFT testing. These data were compared with 18 pediatric patients from our department with an endocardial ICD system of the same manufacturer without serial DFT testing and a mean charging time of 36 ± 15 s/year (p < 0.05), indicating a significant loss of battery capacity due to serial DFT testing.
Up to now, satisfactory short-term results have been reported from several centers for implantation of extracardiac ICD applying a variety of modifications in young patients (4–9,15,16).
In our series of 23 patients with a pleural shock electrode, median DFT at implant was 10 J. Data corresponds well with previous reports, although DFT at implantation may vary depending on the surgical technique. Radbill et al. (15) did not find any difference between intraoperative DFT in 12 young transvenous ICD patients when compared with 8 patients with pericardial ± subcutaneous coils. Hsia and coworkers (6) reported DFTs <15 J at implantation, comparable to our study, using an intrapericardial position of the shock electrode in 7 children. Stephenson and coworkers (5), however, reported significantly lower DFTs in patients with shock coils placed directly on the epicardium when compared with subcutaneous finger arrays. It may be speculated that lower DFTs can be achieved with an intrapericardial shock electrode, compared with a pleural position, as less lung tissue is located within the electrical field.
Data on DFT during mid-term follow-up in extracardiac ICD, however, are lacking, particularly with respect to DFT course with growth of the patients. In our series, a significant DFT increase was found in 4 of 23 patients during routine follow-up DFT testing. A study by Stephenson and coworkers (17) reviewed 155 patients with a transvenous ICD. Mean age at implantation was 20.4 years and included a significant number of adult patients. DFT testing in this study detected a clinically significant DFT increase in 24%. DFT changes were compared between patients who had routine DFT testing and those who had testing at generator change versus patients with DFT testing prompted by clinical demands. A significant increase was evident in patients with a clinical indication. However, detailed DFT values were not provided. In addition, there was a trend toward a greater DFT shift in smaller patients, but no significant difference was evident based on body surface area.
Radbill and coworkers (15) assessed DFT in 5 patients before and after a median interval of 56 months after ICD placement, using a binary search protocol. In 2 of 5 patients, a significant DFT increase was noted. This finding, however, was not associated with growth.
In the majority of reports focusing on extracardiac ICD in infants and small children, DFT retesting was not performed at all or only in a small portion of the patients (4,6–7,9,11).
To the best of our knowledge, we are the first to present data on serial DFT testing during postoperative follow-up in a uniform group of small children. Results demonstrate that a DFT increase may occur during follow-up in a particular patient that would have been missed without regular DFT testing. A significant correlation between DFT and length as well as body surface area was evident. All 4 patients with a DFT increase had channelopathy, which makes DFT changes related to the underlying disease very unlikely.
When performing serial DFT testing in a high-risk population, potential risks—including electrical storm, cerebrovascular accident or transient ischemic attack, prolonged resuscitation, and even death—must be taken into account (18). Our results, however, strongly support the recently published Class IIa recommendation for regular DFT testing of nontransvenous ICD systems (19).
Our data correspond well with experimental findings applying a computer model to DFT in different torsos with novel electrode configurations (20,21). In these experiments, a median DFT increase from 14 J in a 10-kg torso to 48 J in a 32-kg torso and up to 85 J in a 75-kg torso was reported, respectively. In our study, body length but not body weight, correlated with higher DFTs. Median body weight gain was only 3 kg after 1 year and 6.5 kg after 3 years, respectively. Although there was only a small weight gain during the study period, median gain in length was 6 cm after 1 year and 17 cm after 3 years, respectively. Length may be more important than gross body weight, as growth curves in this age range typically show more pronounced changes in length. These changes in stature may influence defibrillation vectors more by modifying the spatial relationship of coil to can and the amount of lung tissue between the 2, resulting in significant DFT rise. It is of note that lung capacitance is a huge factor in DFT determination (22). Therefore, body length seems to be a more sensitive parameter than body weight to predict DFT during growth. The findings may not be transferable to children with pericardial leads, as, in the pleural position, the amount of lung between the device and electrode is an important DFT determinant. Also, the study is limited by the length of follow-up. Only 8 out of 23 patients had follow-up at 3 years.
It needs to be emphasized, however, that increase of body weight and body length were not correlated with DFT. It may be speculated that longer body length itself, and not the increase in body length, reflects more the actual status of the amount of lung tissue and the defibrillation vector. It is of note that there was a strong correlation between impedance of the shock electrode and DFT. It may be speculated that increased lung volume had an impact on the impedance of the shock electrode.
We used a “relatively high” coil position in our study. In this primary setting, we gained DFTs of 10 J in most of the patients. The high coil position, as used in the current study, may not be best. Choosing a lower coil position in the future may perhaps reduce intrathoracic current and thus increase transcardiac current.
DFT course in our patients with extracardiac ICD could not be predicted by any noninvasive parameter. This underlines the need of regular DFT testing to achieve a sufficient safety margin between DFT and maximal ICD output.
DFT, as defined according to our protocol, reflected the lowest energy tested and not the “true” DFT. Measuring the “true” DFT would have required additional testing and might have been associated with additional harm to the patient. Although we did not observe any complications during serial DFT testing, it needs to be emphasized that this procedure can be life threatening (18) and causes significant consumption of battery capacity. Finally, long-term effects like scarring at the lung/pleura interface may occur after inserting the shock electrode into the pleural space impeding pulmonary function.
In our young patients with extracardiac ICDs using pleural defibrillation leads, serial DFT testing revealed a DFT increase >20 J in 4 of 23 patients, requiring surgical revision during mid-term follow-up. A correlation between length and DFT was evident.
Because of the unpredictability of ICD performance with growth of patients with extracardiac ICDs, serial DFT testing is recommended to recognize ICD failure in a timely fashion. This approach, however, is associated with a significant loss of battery capacity.
COMPETENCY IN MEDICAL KNOWLEDGE: The use of alternative ICD placement techniques, such as the extracardiac system in toddlers and infants, is feasible. Our data strongly support the need for serial DFT testing with further growth in these patients. Based on the results of this study, yearly defibrillation tests should be considered, and revision of the system should be performed if an adequate safety margin of >10 J is no longer present.
TRANSLATIONAL OUTLOOK: Additional studies are needed to identify clinical parameters and potential clinical thresholds associated with a critical DFT related to these innovative ICD implantation techniques. Data presented suggest that body length has a greater impact on DFT changes than body weight.
The authors have reported that they have no relationships relevant to the contents of this paper to disclose.
All authors attest that 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
- defibrillation threshold
- implantable cardioverter-defibrillator
- Received October 6, 2016.
- Revision received March 5, 2018.
- Accepted March 15, 2018.
- 2018 American College of Cardiology Foundation
- Proclemer A.,
- Zecchin M.,
- D'Onofrio A.,
- et al.
- Epstein A.E.,
- DiMarco J.P.,
- Ellenbogen K.A.,
- et al.
- Epstein A.E.,
- DiMarco J.P.,
- Ellenbogen K.A.,
- et al.
- Kanaya T.,
- Nishigaki K.,
- Ishimaru K.,
- et al.
- Priori S.G.,
- Blomström-Lundqvist C.,
- Mazzanti A.,
- et al.
- Lerman B.B.,
- Deale O.C.