1 INTRODUCTION

Infants and small children with congenital or acquired bradyarrhythmias or tachyarrhythmias often require cardiac device therapy to maintain a stable rhythm. Complete atrioventricular block (CAVB) can be congenital, occurring in 1 out of 15 000 patients, most requiring a permanent pacemaker during infancy or childhood.^1-4 Postoperative CAVB following congenital heart disease (CHD) surgery has an incidence of 1%–6% and is another scenario that requires a pacemaker.^5-9 Patients with CHD may also develop sinus node dysfunction (SND) or conduction abnormalities necessitating pacemaker placement.^{10, 11} Additionally, patients both with and without CHD can suffer from ventricular tachycardia or fibrillation events that require an implantable cardioverter defibrillator (ICD), either for primary or secondary prevention.

Transvenous pacemaker and ICD implantation is routinely an outpatient procedure in adults. However, the transvenous approach is prohibitive in younger children due to small stature, rapid growth, and the risk of venous stenosis.^{12, 13} Transvenous lead placement may also be challenging in older children and adults with congenital heart disease, either due to nonstandard venous anatomy such as superior vena cava stenosis or duplication, or previous surgeries such as a Glenn operation.^{14, 15} The current standard for these patients is lead placement directly on the epicardium via open chest surgery.^{14, 16} The sternotomy or thoracotomy from this procedure results in significant pain and a longer hospital stay, including intensive care.^{6, 17, 18} A percutaneous method for epicardial lead implantation could be a novel solution for cardiac device therapy in these patients.

Previous studies have proposed alternative surgical methods to decrease the invasiveness of lead implantation, either with different, smaller surgical incisions, or thoracoscopy via multiple ports.^19-24 Our group has previously established the feasibility of epicardial lead placement using a single incision access port to successfully implant ICDs and prototype miniature pacemakers under direct visualization in acute and chronic porcine studies.^25-28 However, all of the prior implants still required a small surgical incision. A novel, truly percutaneous approach that requires just a single needle stick represents a significant improvement in the invasiveness of this procedure for infants and small children. The objective of this study is to assess the feasibility of PeriScope, a novel percutaneous access kit for epicardial lead implantation under direct visualization in an infant animal model.

2 METHODS

This study adheres to the National Institutes of Health's Guide for the Care and Use of Laboratory Animals and was approved by the Institutional Animal Care and Use Committees at Children's National Hospital (CNH) (#00030442) and Nationwide Children's Hospital (NCH) (#AR20-00155). Two acute studies were performed at CNH Research Animal Facility, and four additional acute studies were performed at NCH Animal Research Core.

2.1 Device design

Preliminary designs of this device using a flexible fiberscope within the lumen of the pericardial access needle resulted in a relatively small field of view with poor visualization and inability to confirm sheath placement or lead fixation (Figure 1A). The current version of our access kit uses a custom illumination and insufflation sheath through which the procedure is performed. A 4.3 mm diameter trocar (Ethicon) was lined with thousands of optical fibers (Myriad Fiber) forming a ring light at the distal end of the sheath (Figure 2). This provided sufficient intrathoracic illumination. A Veress needle with spring-loaded action is inserted through this illumination sheath for intrathoracic deployment. Additionally, a small micro-CMOS sensor camera (Omnivision) was installed at the end of each fiberscope (Myriad Fiber) to improve resolution and field of view (Figure 1B). Moreover, the 8 French dilator and sheath used for pericardial access (Merit Medical) was customized by tunneling a groove down the dilator for fiberscope placement (Figure 3). This allowed for visual confirmation of sheath insertion. Lastly, a splitter tool was developed to allow both a pacing lead and a fiberscope to enter the 8 French sheaths simultaneously to provide direct visualization of lead attachment.

2.2 Animal study

An immature porcine model consisting of 1–2 week-old female Yorkshire piglets was chosen for this study due to similar size to infants with similar external cardiac anatomy, including coronary vessels with a relative preference for a right-dominant coronary system.^{29, 30} While the heart occupies a slightly mesocardic position compared to humans, the intracardiac anatomy and size of the heart relative to the rest of the body is analogous, making these animals ideal subjects for the procedure.^{18, 29, 31}

Each Yorkshire piglet was weighed before the surgery and given intramuscular ketamine (20 mg/kg) and xylazine (2 mg/kg) for preprocedural sedation. Induction and maintenance of anesthesia were performed with isoflurane (1%–5%), and an endotracheal tube was inserted and secured in place for mechanical ventilation. Each piglet was then placed in the supine position and secured with 4-point restraints. Color, heart rate, respiratory rate, and pulse oximetry were monitored continuously during the procedure.

A 2 mm superficial nick was made in the subxiphoid space to the left of midline with a surgical scalpel (Figure 4A). A Veress needle was loaded into the illumination sheath and inserted through the skin nick toward the left thorax (Central Illustration). The spring-loaded action of the Veress needle allows the needle to pass through the soft tissue and diaphragm without damaging intrathoracic structures (Figure 4B). The needle was then removed from the trocar, and the thorax was insufflated with CO₂ (pressure 0.4 mmHg; gas flow 2.5 L/min). A 14-gauge needle with a 1.6 mm diameter fiberscope with embedded micro-CMOS camera was inserted through the illumination sheath and used to access the pericardial space under direct visualization (Figure 5A). The camera was removed, and a guidewire was advanced through the needle. An 8-French 13 cm peel-away sheath and dilator with a second, 0.65 mm diameter fiberscope with embedded micro-CMOS camera was then loaded onto the wire. The entire assembly was then advanced over the wire into the pericardial space with visual confirmation provided by the camera (Figure 5B,C). The dilator was then removed, providing sheath access of the pericardial space.

To visualize the implantation of a pacing lead within the pericardial space, a third fiberscope with micro-CMOS camera and a commercially available 4796 or 4798 pacing lead with side-biting helix (Medtronic Inc, Mounds View, MN) were loaded onto the splitter, and inserted into the sheath (Figure 6A). The splitter assembly was advanced into the pericardial space, and the lead was visualized as it was secured by clockwise rotation (Figure 6B). Pacing parameters were measured to ensure adequate R-wave sensing and capture thresholds. The sheath was then peeled away and removed from the pericardial space along with the camera and splitter. Final R-wave amplitudes, lead impedances, and capture thresholds at both 0.4 ms and 1.0 ms pulse widths were then measured using a Medtronic pacemaker analyzer. After humane euthanasia, a limited necropsy was performed.

Procedural times were recorded for each surgery. Piglet weights are reported as mean and range. Median values with interquartile ranges (IQR) were calculated for R-wave amplitudes, lead impedances, capture thresholds, and procedure times. Statistical analyses were done in MedCalc 12.2.1.0 statistical software (MedCalc Software).

3 RESULTS

Percutaneous implantation of an epicardial pacemaker lead was performed in 6 Yorkshire piglets weighing 4.0 (3.7–4.4) kg under general anesthesia and veterinary supervision as described above. All piglets had successful lead implantation and tolerated the procedure without medical or surgical complications.

Median time from skin nick to sheath access of the pericardium was 9.5 (IQR 8–11) min. Median total procedure time was 16 (IQR 14–19) min. There was one outlier in procedure times. Sheath access in the first piglet studied took 46 min due to difficulty with pericardial access. This was likely due to the learning curve associated with use of a novel tool and technique. Imaging initially confirmed needle access of the pericardium, but between camera removal and guidewire insertion, the needle was accidentally advanced, resulting in a through-and-through puncture of the pericardium. Once the sheath was inserted, the lead was unable to be passed under the pericardium. Thus, the sheath was removed and the access needle was reinserted through the light sheath trocar to reestablish pericardial access. There were no complications to the animal and the second attempt was successful. The subsequent five piglets did not have this problem, and the total procedure time for each of those five animals was less than 20 min.

Final values for lead impedances, R-wave amplitudes, and capture thresholds are reported in Table 1. Initial position of the lead demonstrated atrial pacing or high thresholds in two animals. The lead was subsequently adjusted to provide appropriate ventricular pacing. Median R wave sensing was 5.4 (IQR 4.0–7.3) mV. Median capture threshold was 2.1 (IQR 1.7–2.4) V at 0.4 ms and 1.3 (IQR 1.2–2.0) V at 1.0 ms. The first two piglets had a bipolar 4796 model lead implanted. Bipolar sensing and capture were tested. The subsequent four piglets had a model 4798 lead implanted, which is a quadripolar lead with differential spacing between poles. We noted that unipolar testing of one of the more proximal poles generally resulted in better capture thresholds and clear ventricular pacing instead of atrial capture. For piglet #6, atrial pacing had lower capture thresholds, but ventricular pacing values were used in the analysis for consistency. Limited necropsy following each procedure confirmed that the lead was appropriately affixed against the epicardium (Figure 7A). The resultant skin defect was no larger than 5 mm (Figure 7B).

4 DISCUSSION

Percutaneous epicardial pacemaker lead implantation via this novel minimally invasive thoracoscopic approach was successful in an infant porcine model. All six animals had clinically acceptable acute sensing and pacing lead parameters as well as normal lead impedances. The median total time for the procedure was less than 20 min from needle puncture of the skin to lead implantation. Converting what is currently an open chest surgery into a short percutaneous procedure has the potential to significantly improve the lives of infants and young children who need epicardial pacing.

Other groups have attempted minimally invasive approaches to epicardial lead implantation, but the majority have been performed in adults and required either multiple incisions or an incision at least 1 cm in length. In 1997, Furrer et al. used video-assisted thoracoscopic surgery (VATS) to place epicardial leads in an animal model, but their approach required 4 incisions, including one 4 cm long.¹⁹ Navia et al. demonstrated epicardial lead placement with three separate 1 cm incisions, each containing an access port.²⁰ More recently, Schneider et al. evaluated minimally invasive ICD placement via 2–3 cm inframammary incision in children, and Costa et al. evaluated the feasibility of a minimally invasive 3 cm subxiphoid incision to implant epicardial leads in neonates.^{21, 22} Nellis et al. reported the use of three separate 5 mm ports to suture leads to the epicardium in five patients, aged 9–11 years.²³ In a younger population, Termosesov et al. performed VATS in five children aged 2–4 years, but their approach also required three separate incisions.²⁴ Bar-Cohen et al. used a percutaneous technique to implant a micropacemaker into adult pigs. However, this technique required an 18-French delivery system and the use of dye and fluoroscopy, rather than direct visualization, and was only successful in 3 out of 6 animals.³² The novel thoracoscopic approach described in this manuscript is a significant improvement over current techniques because it is truly percutaneous without surgical incisions, allows for direct visualization of pericardial access and lead placement, only requires a single puncture instead of multiple incisions or multiple ports, and is applicable to infants as well as older children.

One additional advantage of a single needlestick approach is that the needle can be directed toward multiple locations on the epicardium. Transvenous leads traditionally provide chronic ventricular pacing from the right ventricle (RV). However, the percutaneous access technique allows the operator to angle the needle toward the left ventricle (LV) under direct visualization. Chronic RV pacing can lead to cardiomyopathy, diminished exercise capacity, abnormal histological changes, and ventricular remodeling.^{3, 33-35} On the other hand, LV pacing can provide hemodynamic and functional advantages such as better synchrony, increased efficiency, and a lower propensity for cardiomyopathy.^{3, 33, 36}

The sheath can also be directed toward the atrium (Figure 5A). The ability to provide atrial pacing significantly increases the applicability of this technique. Multiple needlesticks can provide dual chamber pacing or even cardiac resynchronization therapy. Atrial pacing can also help prevent Fontan failure in single ventricle CHD patients with a Fontan circulation who suffer from sinus node dysfunction and junctional rhythm.^{37, 38} Indeed, atrial pacing was successful from the distal poles of the 4798 pacing lead in multiple animals in this study.

A relatively large proportion of patients who need epicardial pacing are those with postoperative AV block following CHD surgery. Some patients may even develop late heart block months to years after their surgeries.^{17, 39} Many CHD patients also undergo multiple redo sternotomies over the course of their lives. This results in significant intrathoracic adhesions that can complicate future thoracic surgeries and are associated with significant morbidity and mortality.⁴⁰ Epicardial lead implantation via minimally invasive single-incision access port has been previously established in an adhesion animal model that mimics postoperative intrathoracic adhesions.⁴¹ This study's single needle-stick approach advances this technique to make it truly percutaneous, and should be able to provide a safer solution for patients who may suffer from intrathoracic adhesions and are otherwise unable to undergo a redo sternotomy.

Another potential advantage of this thoracoscopic approach is its potential effect on length of hospital stay. The addition of an epicardial pacemaker during the postoperative period usually results in a longer hospital stay.^{6, 17} While there is limited data on thoracoscopic procedures, laparoscopic procedures in the abdomen have been shown to have less pain, shorter hospital stays, and less morbidity and mortality than open procedures.⁴² It is reasonable to conclude that a thoracoscopic epicardial lead implantation will also result in less pain and a shorter stay compared to a procedure requiring a sternotomy or thoracotomy.

Access to the pericardial space may be used for applications beyond pacemaker and ICD leads. Epicardial ablation is a tool utilized more commonly in the adult population than in pediatrics.⁴³ However, the pericardial access technique for these ablations remains risky with a significant chance for major complications. These include inadvertent ventricular perforation, pleural injury, phrenic nerve injury or coronary injury that can lead to pericardial effusions, tamponade, or even death.^44-46 The ability to achieve pericardial access under direct visualization could mitigate many of these potential complications. Other possible applications for this tool include temporary pacing wires or even to attempt cardiac biopsies.

5 LIMITATIONS

There remain some challenges with this technique. A larger gauge needle was necessary to accommodate the camera and scope within the needle. However, this resulted in an anecdotally higher incidence of going “through and through” the pericardium, necessitating a repeat pericardiotomy in a different location. As with all new techniques, there is a learning curve, as demonstrated by the first procedure taking over 45 min and the remainder of the procedures taking less than 20 min overall from incision to lead fixation. Lastly, there is no commercially available pacing or ICD lead that is specifically designed for this approach. A Medtronic 4798 quadripolar pacing lead with a side-biting helix for fixation was used to test pacing in this study. However, previous studies have shown that the side biting helix may not be strong enough to withstand the rapid somatic growth of piglets in chronic porcine animal studies.^{27, 47} Development of a new lead or stronger attachment would significantly improve the applicability of our approach.

6 CONCLUSION

Percutaneous epicardial pacing is feasible using a needlestick minimally invasive pericardial access tool kit in an infant porcine study. There were no complications and pacing parameters were stable and within clinical limits for each animal. While there are significant advantages to this approach and technique, a few improvements remain before trials in human subjects.

CLINICAL PERSPECTIVES

Competency in Patient Care and Procedural Skills: A percutaneous approach to epicardial lead implantation under direct visualization could result in decreased pain and a shorter hospital length-of-stay for infants and small children who require cardiac device therapy.

TRANSLATIONAL OUTLOOK

Medium and long-term animal studies are necessary to evaluate the stability of leads implanted via this approach over time.