Benefits and risks of implantable cardioverter-defibrillators in children and young adults with congenital heart disease, primary electrical disease or paediatric cardiomyopathy^†

Abstract

OBJECTIVES

This study aimed to investigate the benefits and potential risks of implantable cardioverter-defibrillators in paediatric and congenital heart disease (CHD) patients in the current era.

METHODS

All patients with CHD, paediatric cardiomyopathy or primary electrical disease, who underwent implantation of a defibrillator from 2001 to 2023, were examined. The occurrence of appropriate therapy, unplanned surgeries due to device complications, inappropriate shocks and associated risk factors were analysed.

RESULTS

A total of 214 patients were included, with 61% having CHD, 17% having paediatric cardiomyopathy and 22% having primary electrical disease. The most common diagnoses were transposition of the great arteries and tetralogy of Fallot (each 12%). The median age at implantation was 23 years (interquartile ranges 15–38), and the median follow-up was 5.7 years (95% confidence interval 4.9—7.3). A total of 196 patients met the criteria for outcome analysis, where appropriate therapy was observed in 41% (n = 80), occurring more often in patients with indications for secondary prevention than primary prevention (56% vs 26% at 5 years, P = 0.003). The cumulative incidence of inappropriate shocks was 13% (n = 26), with patients with CHD being more frequently affected. Unplanned surgeries were required in 36% (n = 71), predominantly due to lead-related issues in those with abdominal generator placement.

CONCLUSIONS

The high rate of appropriate therapies underscores the critical importance of risk assessment in ICD selection, particularly to mitigate lead failures and unnecessary shocks. However, defibrillator therapy has a relevant rate of unplanned surgeries, with abdominal generators and epicardial/extracardiac leads being risk factors.

Open in new tabDownload slide

INTRODUCTION

Implantable cardioverter-defibrillators (ICDs) are the most important tools for preventing sudden cardiac arrest due to malignant ventricular arrhythmias (VAs) in children and young adults with congenital heart disease (CHD). Compared to the adult population, children and young adults with CHD present unique anatomical challenges for the placement of ICD systems and represent a small minority of ICD recipients compared to the adult population. Retrospective studies showed a survival benefit for ICD in this cohort, with approximately 20–30% of the patients receiving appropriate ICD therapy [1–4]. However, inappropriate shocks occur in 20–45% of paediatric and CHD patients compared to only 10–15% in the adult population [1–5]. Furthermore, a high incidence of device-related complications has also been reported in this cohort, up to 33–36% [1, 3, 5, 6].

Thanks to the technological evolution of ICD therapy in the last decades, such as extracardiac coils (subcutaneous, pericardial or pleural) and subcutaneous ICD (sICD) systems [7–9], there are more choices in the device and lead type and more possibilities to reduce the device-related complications in this cohort. Transvenous leads are commonly used but are prone to dislodge, fracture or insulation breaches. Conversely, epicardial ICD systems with epicardial leads or extracardiac coils showed even worse lead results and had a higher risk of lead fracture due to body growth [10]. Meanwhile, sICD systems have emerged as a favourable option for paediatric and CHD patients over the last decade [9].

This study aims to examine the incidence and risk factors for both appropriate ICD therapies and complications, such as inappropriate shocks and unplanned surgeries, in a cohort of paediatric and young adult patients with CHD. Additionally, we analysed potential risk factors for these events.

PATIENTS AND METHODS

Ethical statement

The study was approved by the Institutional Review Board of the Technical University of Munich (approval number 2024–334-S-CB on 8 July 2024). Any collection and storage of data from research participants for multiple and indefinite use is consistent with requirements outlined in the WMA Declaration of Taipei. The ethics committee approved the establishment and monitor the ongoing use of databases.

Patients and data collection

This study encompassed all patients who underwent primary ICD implantation at the German Heart Center Munich from January 2001 to October 2023. The patient selection criteria included individuals with a prior diagnosis of (a) CHD regardless of age at ICD implantation, (b) paediatric cardiomyopathy diagnosed before the age of 18, with ICD implantation at any age or (c) primary electrical disease with ICD implantation before reaching 18 years of age. Data were collected from the digital medical records and hospital device database, capturing baseline patient characteristics, ICD therapy indications and specific device details.

Indication and practices for ICD implantation

An indication for primary prevention was classified as such if the patient had not previously experienced an episode of sudden cardiac arrest or sustained VA. Lead placement was categorised as (a) transvenous leads, (b) epicardial/extracardiac if the ICD system contained at least 1 epicardial lead, or an extracardiac array (intrapericardial, subcutaneous, pleural, or finger electrode), or (c) sICD leads. Generator placement was classified as (I) infraclavicular left, (II) infraclavicular right, (III) abdominal or (IV) sICD, with its typical location on the left thoracic side.

Effects and complications after ICD implantation

In terms of ICD outcomes, only patients with a follow-up duration of at least 8 weeks post-implantation or those experiencing death or any ICD-related event at any point post-implantation were included. We defined the follow-up duration from the first ICD implantation until the last follow-up appointment registered in our clinic. If death occurred at any point, the follow-up duration was considered until the time of death. If any other events occurred and the patient had another follow-up appointment, we considered the duration until the last appointment. If an ICD system was explanted and no other was implanted, the explantation time was considered the last follow-up.

Appropriate therapy was defined as shock or anti-tachycardia pacing (ATP) for a VA faster than the programmed detection criteria and accurately detected by the ICD. Inappropriate shock was defined as a defibrillation therapy for anything other than a VA faster than the programmed detection criteria. The appropriateness of ICD-delivered therapy was verified through the review of device electrograms by a team of cardiologists specialized in cardiac device therapy. The causes of inappropriate shocks were classified as oversensing due to lead failure, which led to immediate surgery for revision, or inappropriate discrimination of supraventricular tachycardia (SVT) and T-Wave Oversensing, which was primarily addressed with adjustments of ICD programming. Unplanned surgeries were defined as any operative interventions for ICD system management, excluding generator changes for battery depletion, encompassing lead revision, system explantation or system change (upgrade, downgrade or change of the generator placement). Surgeries for other reasons (e.g. surgeries in patients with CHD), where lead revisions or system changes were also performed, have been classified as unplanned surgeries. Generator changes, even premature, have been considered elective surgeries.

Statistical analysis

Statistical analysis included Kaplan–Meier time-to-event analysis for the primary outcomes, with differences between groups determined via the log-rank test. Continuous variables were expressed as medians with interquartile ranges (IQRs) and categorical variables as absolute numbers and percentages. Median follow-up for the entire cohort was estimated using the reverse Kaplan–Meier method, where patients were censored at the time of unplanned surgical device explantation or death, and loss to follow-up was counted as an event. Cox proportional hazard models were used to explore associations between patient risk factors and ICD outcomes. The proportional hazards assumption for all Cox regression models was tested based on weighted residuals, as described by Grambsch and Therneau [11]. Risk factors were examined in univariable and multivariable analyses. Data analysis and graphing were performed with the Statistical Package for the Social Sciences (SPSS) version 28.0 for Windows (IBM, Ehningen, Germany).

RESULTS

Patient characteristics and initial ICD-specific data

A total of 214 patients were included. Median age at the initial procedure was 23 years (IQR: 15–38), and median follow-up estimated using the reverse Kaplan–Meier method was 5.62 (95% confidence interval 4.71—7.02) years (Supplementary Material, Fig. S1). During our follow-up, there were 27 deaths registered from the total of 214 inclusions. None of the deaths were device-related during the implantation procedure or an unplanned surgery.

The distribution of all diagnoses is illustrated in Fig. 1: 131 (61%) with CHD, 35 (17%) with paediatric cardiomyopathy and 48 (22%) with primary electrical disease.

Baseline characteristics, ICD system data and subgroup specifics are presented in Table 1. ICD implantation was indicated for primary prevention in 73 patients (34%) and for secondary prevention in 141 patients (66%). The placement of the ICD generators was as follows: infraclavicular left side for 139 patients (65%), infraclavicular right side for 16 patients (7%), abdominal for 30 patients (14%) and on the left lateral chest for sICD systems in 29 patients (14%). The majority of ICD systems contained only transvenous leads (71%), while 15% had at least 1 epicardial lead or extracardiac array, and 14% had only subcutaneous sICD leads. The results of the era analysis between the early era (2001–2011) and the late era (2012–2023) are shown in Supplementary Material, Table S1. In the late era, we noted a higher indication for secondary prevention, the evolution of the sICD systems, and a higher use of atrial leads, compared to the early era.

Outcomes

The study outcomes, including subgroup analysis, are detailed in Table 2. Out of the 214 initially registered patients, only 196 were deemed eligible, as they had a follow-up time of at least 8 weeks post-implantation or had experienced an adverse event or death at any point. Of these 196 patients, 80 (40.8%) required at least 1 appropriate therapy during their follow-up time, 26 (13.3%) an inappropriate shock and 71 (36.2%) an unplanned surgery. The overview of time-to-first event for appropriate therapy, inappropriate shock and unplanned surgery is illustrated in Fig. 2.

Appropriate ICD therapy

Of the 196 patients, 80 (41%) experienced at least 1 appropriate ICD therapy (shock or ATP), with a median time to the first appropriate therapy of 2.3 years (IQR 0.8–5.0). Notably, patients with ICD indication for secondary prevention had a significantly higher rate of appropriate therapy compared to those with ICD indication for primary prevention (56% vs 26% at 5 years, P = 0.003), as illustrated in Fig. 3. The hazard ratios (HRs) estimated from the Cox models also revealed a significant association between ICD indication for secondary prevention and time to first appropriate shock with the univariable model and the multivariable model (HR: 2.24, 95% CI: 1.31–3.83, P = 0.003) as an independent risk (Table 3A). When considering subgroups based on lead placement and those based on generator placement, no significant differences were observed regarding appropriate therapy.

Inappropriate shocks

Overall, 26 patients (13%) received at least 1 inappropriate shock, with a median time to the first inappropriate shock being 1.7 years (IQR 0.6–5.6). Eight of them were due to lead failure and 18 were due to SVT/T-Wave-Oversensing. No significant difference regarding the causes of inappropriate shocks was observed in any subgroups. Patients with cardiomyopathy experienced fewer inappropriate shocks than those with CHD or primary electrical disease, although without significant difference (P = 0.253). Furthermore, the subgroups categorized by indication for ICD therapy, lead placement or generator placement were all similar regarding inappropriate shocks.

Unplanned surgeries due to device complications

Unplanned surgery was required in 71 patients (36%), with a median time to first surgery of 3.6 years (IQR 0.9–7.2). When considering subgroups based on lead placement, patients with at least 1 epicardial lead or extracardiac array required significantly more unplanned surgeries (P = 0.03) than patients with just transvenous leads or sICD leads, mainly due to lead revisions (62 patients). Furthermore, ICD systems with a generator located in an abdominal pocket were significantly more susceptible to device-related unplanned surgeries (P < 0.001) when compared to the other ICD systems—Fig. 4. This was valid for all indications regarding unplanned surgeries due to device complications: lead revisions, systems explantations, and system changes, as seen in Table 2.

The utilization of epicardial/extracardiac leads (HR: 1.97, 95%-CI: 1.16–3.34, P = 0.013) and the placement of the generator in an abdominal (HR: 3.0, 95% CI: 1.77–5.09, P < 0.001) were risk factors for shorter time to unplanned surgery with univariable analysis (Table 3B). The multivariable analysis revealed the generator placement in an abdominal pocket (HR: 3.0, 95% CI: 1.77–5.09, P < 0.001) as an independent risk for unplanned surgeries (Table 3B). No differences regarding unplanned surgery were observed in the subgroups categorized by the main diagnosis or by the indication for ICD therapy.

Subcutaneous ICD

There were 29 sICD implantations during our observation period, starting in 2012. Appropriate therapy occurred in 8 patients (31%), inappropriate shocks in 4 patients (15%) and unplanned surgery in 4 patients (15%) over a median follow-up of 3.3 years. Although the sICD group showed a trend towards fewer inappropriate shocks and unplanned surgeries, these differences were not statistically significant, potentially due to the smaller sample size in this subgroup.

DISCUSSION

Necessity of ICD systems in children and young adults with CHD

The benefits of ICD therapy in this population are unequivocal, providing a lifesaving mechanism that addresses the heightened risk of fatal arrhythmic events. Retrospective studies have consistently demonstrated a survival benefit associated with ICD implantation with approximately 20–30% of patients receiving appropriate ICD therapy [1–4].

The current study reinforces the necessity of ICD systems by revealing that an impressive 41% of patients experienced appropriate ICD interventions. A significant contributing factor to this elevated rate of appropriate therapy may be the high rate of implantation for secondary prevention purposes—64%. This aligns with existing evidence that secondary prevention indications are associated with higher rates of appropriate ICD therapy [2].

We had a high proportion of patients with CHD, likely due to the large CHD programme at this centre and the many late references of patients from outside Germany. Furthermore, we have less biventricular systems than other similar cohorts, due to the relevant number of patients with PED in our cohort, none of whom indicated cardiac resynchronization, but also due to the long data collection period, where CRT was not established as a valid option yet.

Risks after ICD implantation

The occurrence of inappropriate shocks is a relevant risk associated with ICDs. In paediatric and CHD patients, the rate of inappropriate shocks ranges from 20% to 45% [1–5]. These inappropriate shocks are not only physically distressing but also have profound psycho-social consequences. Its occurrence can lead to anxiety, depression, and a constant fear of receiving shocks and can decrease the quality of life compared to peers with other chronic illnesses [12]. The severity of the psycho-social consequences after inappropriate shocks, especially in children and young adults, is underrated in the long term and needs more attention from the attending physicians. The lack of psychological support for this cohort is an issue that needs to be addressed over the following years.

Moreover, inappropriate shocks have been associated with adverse clinical outcomes, including an increased risk of mortality in adult populations [13]. The lower-than-expected rate of inappropriate shocks in our study may be attributable to the individualised ICD programming practices at our institution and to the relatively high rate of atrial leads, known to enhance the detection of VAs. However, the study did not analyse this aspect.

Device-related complications represent another critical long-term risk for patients with ICDs. Studies, including ours, have reported that up to 33–36% of paediatric and CHD patients require unplanned surgeries due to complications such as lead revision or system explantation [1, 3, 6–8].

The emergence of sICDs

Noteworthy findings from 6 studies involving 221 patients with sICDs reported lower occurrences of appropriate shocks (20%), inappropriate shocks (15%) and surgical complications (13%) after a 3-year follow-up period [14–19]. Our study’s separate data on sICDs over a median follow-up of 3.3 years align with these results regarding the adverse events, while indicating a higher rate of appropriate therapies which affirms the effectiveness and safety of sICDs. However, a 2022 European registry of paediatric and young adult patients with CHD highlighted limitations in the use of sICDs due to the perceived large size of the generator for small children or thin adolescents and its lack of pacing capabilities [9]. Nevertheless, these restrictions are offset by the simplified implantation procedure, device programming, absence of endocardial leads and ease of extraction procedures associated with sICDs [9].

Strategies for opimising ICD therapy in young patients

The identification of abdominal generator and epicardial leads as risk factors for unplanned surgeries should prompt clinicians to consider alternative strategies, such as sICDs, in patients with high-risk profiles. The following suggestions could help develop strategies that reduce the risks associated with ICD therapy:

Patient selection and risk stratification

The first step in optimizing ICD therapy is the careful selection of candidates based on a thorough assessment of the risk-benefit ratio. For patients with a history of life-threatening arrhythmic events (secondary prevention), the benefit of ICD therapy is generally well-established [2, 4]. However, in cases where ICD implantation is considered for primary prevention, a more cautious approach is warranted. DeWitt et al. [3] demonstrated that the risk-benefit ratio of ICD therapy, especially in patients with an indication for primary prevention, may shift over time, requiring reconsideration of ICD protection as the patient ages and the risk for adverse events increases.

Individualized ICD system selection

A tailored selection strategy of the ICD systems, as postulated by Silvetti [10], suggests the use of epicardial ICD systems in infants and small children, transvenous ICDs in children weighing more than 30 kg (with a preference for postponing dual-chamber systems until after puberty) and sICDs in patients with a body mass index exceeding 20 kg/m² unless contraindicated. This approach aligns with the findings of our study and aims to minimize complications while maintaining the effectiveness of ICD therapy.

Optimizing ICD programming

Individualized ICD programming, tailored to the specific arrhythmic patterns of the patient, can enhance the accuracy of arrhythmia detection and reduce the incidence of inappropriate shocks. Moreover, regular follow-ups or remote ICD monitoring are essential to ensure that ICD programming remains optimal. Due to the technological advancements over the last decade, more ICD generator models can be monitored remotely. This increases the chances for early recognition of heart arrhythmias or deteriorating lead values.

Comprehensive patient and family education

Education and counselling for both patients and their family are critical components. They should be thoroughly informed about the benefits and risks of ICD therapy, including the possibility of inappropriate shocks, unplanned surgeries and psychological impact.

Limitations

The single-centre nature of the study and the potential for selection bias may limit the generalizability of these results to other populations. Additionally, the retrospective design could have led to underreporting of certain complications. Data were only as complete as the documentation in the medical record. The lack of completeness of the follow-up data was a limitation and led to a bias in case selection. As the main focus of this study remained the comparison of the technical abilities and restrictions of each ICD system, we did not include death related to the primary diagnosis as an end-point or consider it as a competing risk in our analysis. It is pertinent to note that the implantation procedures were performed by various specialists from the hospital, including cardiologists, heart surgeons and paediatric heart surgeons, each employing distinct procedural techniques. Additionally, the data solely encompass the initial occurrence of the defined outcomes, thereby underestimating the recurrence of appropriate ICD therapy and associated complications.

CONCLUSION

Our study underscores the vital role of ICD therapy in preventing life-threatening arrhythmias in young patients but also highlights the need for vigilant monitoring and strategic planning to mitigate the risk of device-related complications. The occurrence of such complications in ICD recipients, mainly unplanned surgeries, was notably high. Placement of the ICD generator in an abdominal pocket and utilization of epicardial/extracardiac leads were identified as significant risk factors for unplanned surgeries. Future research should focus on multicentre prospective studies to validate these findings and explore the long-term outcomes of newer ICD technologies, such as sICDs, while also developing specific recommendations for ICD programming for every type of arrhythmia and underlying disease.

SUPPLEMENTARY MATERIAL

Supplementary material is available at EJCTS online.

FUNDING

This study was supported by grants from the Förderverein des Deutschen Herzzentrums München.

Conflict of interest: The authors declare no potential conflicts of interest with respect to the research, authorship, or publication of this article.

DATA AVAILABILITY

The data underlying this article will be shared by the corresponding author upon reasonable request.

Author contributions

Andrei Fetcu, MD: Conceptualization; Data curation; Formal analysis; Methodology; Visualization; Writing—original draft. Thibault Schaeffer: Writing—review & editing. Carolin Niedermaier: Writing—review & editing. Jonas Palm: Writing—review & editing. Takuya Osawa: Writing—review & editing. Muneaki Matsubara: Writing—review & editing. Paul Philipp Heinisch: Writing—review & editing. Lena Friedrich: Writing—review & editing. Carsten Lennerz: Writing—original draft. Alfred Hager: Conceptualization; Supervision; Validation; Writing—review & editing. Peter Ewert: Data curation; Investigation; Project administration; Supervision; Writing—review & editing. Jurgen Hoerer: Conceptualization; Project administration; Supervision; Validation; Writing—review & editing. Masamichi Ono: Conceptualization; Data curation; Investigation; Project administration; Supervision; Visualization; Writing—review & editing

Reviewer information

European Journal of Cardio-Thoracic Surgery thanks Tai Fuchigami and the other anonymous reviewers for their contribution to the peer review process of this article.

Footnotes

REFERENCES

ABBREVIATIONS

ABBREVIATIONS
 
• ATP

  Anti-tachycardia pacing

 
• CHD

  Congenital heart disease

 
• HR

  Hazard ratio

 
• ICD

  Implantable cardioverter-defibrillator

 
• IQR

  Interquartile range

 
• sICD

  Subcutaneous implantable cardioverter-defibrillator

 
• SVT

  Supraventricular tachycardia

 
• VA

  Ventricular arrhythmia