Inadvertent Placement of an Extravascular Implantable Cardioverter‐Defibrillator Lead in the Pericardial Space: Recognition and Successful Management at the Time of Implantation

We here describe the case of a 48 years old man with a history of arterial hypertension and hypercholesterolemia, with no known comorbidities or allergies and no family history of cardiomyopathy or sudden cardiac death. In 2002, an ECG revealed incomplete right bundle branch block and early repolarization abnormalities, leading to regular cardiologic follow-up, all initially unremarkable. A cardiac MRI performed in November 2015 showed thinning and hypokinesia of the mid-to-distal dorsal-lateral wall with subepicardial late gadolinium enhancement (LGE) involving the entire free wall of the left ventricle (LV), with preserved systolic function (EF 60%). Bisoprolol therapy was initiated. Genetic testing revealed a possibly pathogenic variant in plakophilin-2 (PKP2) and a variant of uncertain significance in myopalladin gene (MYPN). A repeat cardiac MRI in April 2018 confirmed findings consistent with arrhythmogenic left ventricular cardiomyopathy (ALVC). An electrophysiologic study in November 2021 was negative, and a loop recorder was implanted; subsequent follow-ups showed clinical stability and no arrhythmic events. At the most recent follow-up on July 19, 2025, a cardiology consultation and loop recorder interrogation revealed isolated premature ventricular beats (PVCs), a brief episode of ventricular bigeminy, and one run of non-sustained ventricular tachycardia (17 beats, cycle length ~360 ms) in an asymptomatic patient. In view of this arrhythmic event in the context of structural heart disease, an elective hospital admission was scheduled for repeat cardiac MRI, loop recorder explantation, and implantation of an ICD. The cardiac MRI showed a non-dilated LV with preserved systolic function (EF 58%) and stable, extensive non-ischemic LGE involving mainly the anteroseptal and subepicardial regions (Figure 1, left panel). In consideration of the patient's young age, the presence of a disease associated with a high risk of monomorphic ventricular tachycardia potentially responsive to antitachycardia pacing (ATP), and after discussion with the patient, the decision was made to proceed with implantation of an extravascular ICD 1, 2. Standard clinical protocols were adhered for both pre-procedural planning and procedural techniques 3. The lateral chest MRI demonstrated the presence of a narrow space between the sternum and the epicardial surface of the right ventricle, a prominent xiphoid process, and a pectus excavatum deformity, all suggesting a potentially challenging implantation (Figure 1, right panel). An initial horizontal subxiphoid skin incision in the standard position did not allow for easy advancement of the tunneling tool into the retrosternal plane. Therefore, the incision was extended caudally with an additional T-shaped extension. This modification enabled a successful passage of the dedicated retrosternal tunneling tool equipped with a peel-away introducer sheath. The first tunneling was performed without significant resistance, and the defibrillation lead was positioned on the cardiac surface at the epicardial level. After withdrawal of the introducer, sensing parameters were satisfactory (R-wave amplitude: 4 mV) with no evidence of P-wave oversensing. However, synchronous movement of the lead with the cardiac cycle raised suspicion of inadvertent placement of the lead within the pericardial space. We decided to re-advance the introducer sheath over the lead body, remove the lead, and inject contrast medium. The contrast injection clearly delineated the pericardial space (Figure 2, left panel) confirming the suspected inadvertent placement of the introducer within the pericardial space. The introducer was gradually withdrawn while small amounts of contrast dye were repeatedly injected, confirming its position within the pericardial space and suggesting that the site of pericardial entry was located in the lower retrosternal area (Figure 2, middle panel). During introducer retraction, no blood could be aspirated, excluding injury to either vessels or myocardium. In addition, the cardiac silhouette on fluoroscopy did not show signs of cardiac tamponade. These findings supported the decision to safely proceed with the procedure. For safety reasons, we waited 20 min, during which both the patient's hemodynamic status and cardiac imaging remained stable. This further supported the decision not to place a pericardial drain. The initial introducer was then removed and replaced with a new peel-away introducer. The residual pericardial contrast then served as a visual guide for a second, more superficial retrosternal tunneling, allowing for correct positioning of the original defibrillation lead over the right ventricular epicardium but outside the pericardial cavity (Figure 2, right panel). Final electrical parameters were acceptable (R-wave amplitude: 1.5 mV, no P-wave oversensing). Lastly, the proximal portion of the lead was tunnelled and connected to the ICD generator, which was placed in a subcutaneous pocket in the left anterior mid-axillary position. Chest X-ray performed the day after implantation in two projections confirmed the appropriate positioning of the ICD system (Figure 3), and defibrillation threshold testing demonstrated good device function. At the 1-month post-implantation follow-up, the sensing parameters were satisfactory (R-wave amplitude: 2 mV, no P-wave oversensing), with no arrhythmic events recorded in the device memory. This case highlights that, even when performing an extravascular ICD implantation with the utmost care, inadvertent lead placement within the pericardial space may still occur. In our case, the use of contrast medium to verify the introducer position allowed confirmation of the clinical suspicion and guided a second, more accurate retrosternal tunneling, leading to successful implantation without adverse clinical consequences. Transvenous implantable cardioverter-defibrillators (ICDs) are a well-established intervention for lowering the risk of sudden cardiac death in susceptible patients 4, 5. However, placing ICD leads through the venous system can lead to significant short- and long-term complications, such as vascular trauma, venous occlusion, bloodstream infections, cardiac perforation, and risks associated with extracting chronic leads. The subcutaneous ICD (S-ICD) was designed as an alternative to traditional transvenous devices, positioning its lead in the tissue layer between the skin and the sternum. By avoiding the vascular system altogether, the S-ICD lowers both the incidence and severity of complications related to transvenous leads, while still providing the same protection to the patient 6, 7. Its limitations include the inability to deliver ATP and the need for a bulkier generator—located outside the thoracic cavity—to produce higher-energy shocks, which can shorten battery life. The EV ICD was subsequently developed to bypass intravascular placement while preserving many of the capabilities of a transvenous system. With its lead positioned in the substernal space, the EV ICD can offer both ATP and defibrillation therapy from a single device, with size, expected longevity, and defibrillation thresholds comparable to those of transvenous ICDs, all while remaining outside the vasculature. (1) More recently, 3 years follow-up data of the Pivotal study confirmed effectiveness and safety of the EV-ICD; no intraprocedural complications were reported among the 299 patients with a successful implantation. A few patients required system revisions and in some cases lead dislodgment was reported; nevertheless, whether inadvertent placement in the pericardial space was suspected or diagnosed was not specified. (2) Very recently, a case of inadvertent placement of an ICD lead diagnosed 6 months after implantation was reported, but information on how this case was managed is lacking 8 To the best of our knowledge, this is the first description of accidental placement of an extravascular ICD lead within the pericardial space that was recognized and managed during the implant procedure. A heartfelt thanks to Sabrina de Cicco and Vincenzo Vumbaca from Medtronic Italy for invaluable assistance in managing this case. The authors declare no conflicts of interest.