Durability and coil positioning in paediatric epicardial ICD systems: insights from a 20-year retrospective study

Background Paediatric epicardial implantable cardioverter defibrillator (ICD) implantation remains technically challenging due to small body size, evolving anatomy, and unique arrhythmic substrates. There was a diversity of practice in the placement location of the defibrillation coil. Long-term data, particularly in Asian patients, are scarce. Purpose The study aimed to evaluate the long-term outcomes of epicardial ICD in children, and provides insights on ICD coils positions and durability of epicardial systems during growth. Methods A 20-year retrospective cohort analysis was conducted for all epicardial paediatric ICD implantations in the single tertiary paediatric cardiology, from 2005 to 2025. Demographics, clinical diagnosis, ICD indication, defibrillation threshold testing (DFT), lead configuration were recorded. Follow-up events including appropriate and inappropriate therapies, system revisions were analysed. Results Nine patients (M:F = 7:2) were included. Median age at implantation was 4.5 years (range 0.2–7) and median body weight 15.4 kg (range 5.3-19.7). Eight patients received ICDs for secondary prevention. Congenital heart disease was present in two patients. All devices were implanted as epicardial systems via midline sternotomy except two cases using a small subxiphoid incision, and the ICD coil was advanced posteriorly with active screw fixation at the lead tip. For those undergoing DFT (n=6), successful defibrillation energies used ranged from 10-20 J, with safety margins 15-27J and shock impedance 46-80 ohm. There were two minor complications of post-operative pleural and pericardial effusion. During a median follow-up of 9.1 years (IQR 7.6-11.0), there was no mortality. Six patients received appropriate ICD therapies, the earliest within 1 year after implantation. Three patients experienced inappropriate shocks due to sensing lead failure. Late revisions of ICD system were required in six cases (66.7%), at a median time of 6.4 years (IQR 5.2-8.0). Two revisions were attributed to suboptimal coil position. One ICD coil was placed subcutaneously over the back, and the other the coil was found at the diaphragmatic surface of the pericardial space (Figure 1). DFT was successful in early childhood but there was significant migration of ICD coil or failure in subsequent DFT with the growth of the children. Four patients were still using the initial implanted ICD coil as part of the epicardial ICD system at 8.4 years (IQR 7.6-10.4), and these ICD coils were placed over posterior pericardial space or transverse pericardial sinus (Figure 2). Conclusion Paediatric epicardial ICD implantation provides effective long-term protection against sudden cardiac death in small children and infants. Late revision of epicardial system is common, with majority related to epicardial sensing lead failure. Coil positioning within the posterior or transverse pericardial sinus demonstrated stable system performance without revision need.