Over the past six decades, there have been considerable advances in permanent pacemaker therapy, which has become an established intervention for managing bradyarrhythmia, symptomatic heart block and heart failure, contributing to reduced mortality and improved quality of life.1,2 Currently, transvenous pacing is the predominant method for cardiac pacing, with central venous access typically obtained through the subclavian, cephalic or axillary vein for electrode placement into the right chamber and coronary sinus. However, this approach is limited due to difficulty accessing the superior or inferior vena cava because of chronic occlusion or previous surgeries, and is associated with multiple short- and long-term complications related to the transvenous leads.3–7 Epicardial lead placement, often deployed intraoperatively during cardiac surgery, involves suturing pacing leads directly onto the ventricular epicardium.1,8,9 This procedure requires thoracotomy or sternotomy and may result in significant pain, prolonged hospital stay and increased upfront morbidity and mortality.10–12 Leadless pacemaker systems, delivered via a percutaneous femoral, catheter-based route, have the advantage of mitigating several complications associated with conventional leads and generator pockets, providing improved patient comfort and aesthetics.13,14 However, their widespread adoption remains limited due to high cost, limited retrievability and an increased incidence of short-term complications.15,16
In the present study we propose a novel echocardiography-guided percutaneous transapical intraventricular septal (PTAIVS) pacing technique for delivering electrodes into the interventricular septum (IVS). Inspired by the Liwen RF ablation system (Hangzhou Nuo Cheng Medical Instrument Co. Ltd), originally designed for treating drug-resistant hypertrophic obstructive cardiomyopathy, this approach offers a minimally invasive alternative when conventional endovascular access is not possible, potentially mitigating transvenous access complications.17
The aims of this study were to evaluate the technical feasibility, safety and 3-month follow-up outcomes of the PTAIVS procedure in a canine model. The primary endpoint was procedural success, defined as successful electrode implantation confirmed through unipolar pacing.
Methods
This was a translational research design in a canine model, conducted in a phased manner (SET1, the first six dogs with pacing; SET2, the next six dogs with pacing) to iteratively improve device design and test procedural feasibility and safety. The study protocol was approved by the Institutional Ethics Committee of Animal Experimental Center in Shanghai Tong Ren Hospital and the study was performed in accordance with institutional and national guidelines for the care and use of laboratory animals.
Animal Preparation
In this study, we used 12 adult beagle dogs (either sex). The dogs were weighed and administered intramuscular ketamine 20 mg/kg and xylazine 2 mg/kg for preprocedural sedation. Anaesthesia was induced with isoflurane 2% and maintained with isoflurane 1–5% throughout the procedure. Each dog was placed in the left lateral position and secured with four-point restraints. The hair over the heart region was shaved before the operation. Continuous 12-lead ECG (LABSYSTEM PRO EP Recording System; Boston Scientific) monitoring was maintained throughout the procedure. An intracardiac electrogram was recorded to assess paced QRS morphology and electrical activation of the local myocardium. A regular setting was applied with a band-pass filter of 30–500 Hz to collect high-frequency signals and a band-pass filter of 0.5–500 Hz especially to visualise low-frequency signals (e.g. current of injury).
Percutaneous Transapical Intraventricular Septum Pacing Procedure
The PTAIVS pacing procedure is shown in Figure 1. Transthoracic (TTE) and transoesophageal (TOE) echocardiography were used to guide the procedure. The apical region for initial insertion was identified using TTE (CX50; Philips) with a 5–1 MHz broadband sector array transducer. Then, under real-time TOE guidance using an X8-2T transducer (Philips), a coaxial introducer needle (17 g × 10 cm; CareFusion) was inserted from the designated apical region and navigated towards the IVS. Upon reaching the mid-to-basal segment of the IVS, approximately 2 cm below the aortic valve, the inner needle was retracted from the stylet, allowing for the deployment of a pacing electrode lead (Model 3830; Medtronic) through the needle cannula for fixation.
Two sets of pacing tools were used during the study. The first six dogs were implanted using an initial pacing set (SET1) comprising the aforementioned 17 g × 10 cm coaxial introducer needle and the Model 3830 pacing electrode (1.8-mm helix). However, early observations revealed that the short helix of the SET1 system was inadequate for secure anchoring within the myocardium, as evidenced by lead dislodgement during the follow-up assessments. Based on these findings, a modified system (SET2) was used in the remaining six dogs, incorporating a 15 g × 10 cm coaxial introducer needle (CareFusion) and a pacing lead with a longer, 10-mm helix (SMCSP2023F; Singular Medical). This adjustment aimed to improve lead fixation within the myocardial tissue. Detailed lead configuration is presented in Supplementary Table 1.
Following fixation, unipolar pacing was performed. Procedural time, ECG morphology and lead parameters, including capture thresholds (at a 0.5 ms pulse width), pacing impedances and R-wave amplitudes, were recorded (Figure 2A). Once successful pacing was confirmed, the rear end of the lead was capped and embedded subcutaneously on the chest wall at the apical insertion site using sutures. The lead was not tunnelled to a generator pocket or connected to an active pulse generator. CT scans and fluoroscopic imaging in anterior–posterior and lateral views were performed immediately after the procedure to document lead position (Figures 2B and 2C). Follow-up assessments at 4 and 12 weeks included electrical parameters (pacing thresholds, impedances and intrinsic amplitudes) and monitoring for pericardial effusion via TTE.
Of the 12 dogs, the first three (SET1 group) were killed immediately after implantation and acute testing to evaluate lead placement and acute pathological findings. Three additional dogs (SET1 group) were killed at the 4-week follow-up to assess lead stability. The remaining six dogs (SET2 group) were killed at 12 weeks after completion of an extended observation period.
Pathology
Limited necropsy was performed immediately after dogs were killed, and the hearts were excised with leads intact for gross examination. After examining the heart surface and leads, the right and left ventricles were opened along the longitudinal myocardial fibres from the tricuspid and the mitral valves, respectively, down to the apex, allowing for the isolation of the IVS from the rest of the heart. The IVS was then incised to expose the pacing electrode embedded within the septal tissue. The location of the electrode was photographed using a digital camera (D700; Nikon). The length and thickness of the IVS and the depth of the pacing lead’s advancement into the septum (from the tip of the electrode to the edge of the apical insertion point) were measured.
For pathological examination, the electrode was removed and the implantation site was excised transmurally and fixed in 10% neutral buffered formalin. The fixed tissue samples were sectioned transmurally (300 µm) and then stained with haematoxylin and eosin and Masson’s trichrome for microscopic examination. Tissue fibrosis surrounding the implantation site was measured by a pathologist blinded to study design.
Statistical Analysis
Data distribution was assessed using the Shapiro–Wilk test. Normally distributed variables were analysed using Student’s t-test and are expressed as the mean±SD. Non-normally distributed variables were compared using the Mann–Whitney U-test and are presented as median values with the interquartile range (IQR). Two-tailed p<0.05 was considered statistically significant. All analyses were performed using SPSS 26.0 (IBM).
Results
PTAIVS was successfully performed in all 12 adult dogs (mean weight 16.0 ± 1.9 kg), with successful lead placement within the IVS and no acute complications. The median length and thickness of the IVS were 6.5 (IQR 6.30–7.23) cm and 0.5 ± 0.1 cm, respectively. Minor bleeding occurred with each needle insertion, but no significant bleeding or arrhythmias were observed intraoperatively. The time from needle insertion to lead fixation was 28.8 ± 4.8 min. The median total procedure time from locating the cardiac apex to suture closure was 42.0 (IQR 39.5–47.0) min (Supplementary Table 2). There were no significant differences in procedure times within the SET1 and SET2 groups (p=0.251 and 0.394). However, the insertion-to-fixation time in the first two procedures in the SET1 group was notably longer (39 min and 36 min, respectively), due to multiple attempts to adjust the insertion route within the IVS, reflecting the learning curve associated with the technique and the use of TOE.
Electrical Parameters
Table 1 summarises the electrical parameters at implantation. The median pacing threshold was 1.7 (IQR 0.85–2.50) V, the median R-wave amplitude was 6.8 (IQR 6.13–13.00) mV and the median impedance was 536 (IQR 510–922) Ω, with no significant differences between the SET1 and SET2 groups (p=0.310, 0.589 and 0.485, respectively). At 4 weeks, SET2 leads showed stable performance, with a pacing threshold of 1.6 ± 0.8 V, R-wave amplitude of 8.9 ± 2.4 mV and impedance of 592 ± 66 Ω. By the 12-week follow-up, only one lead (animal no. 7) remained functional, with a threshold of 5.9 V, R-wave amplitude of 2.5 mV and impedance of 600 Ω.
Postimplantation Observations
Three dogs were killed after the lead implantation and testing. The depth of the pacing electrode insertion into septum, measured as the distance between the tip of the electrode and the edge of the apical insertion site, was 3.7 ± 0.3 cm, indicating that the leads were positioned appropriately within the mid-to-basal region of the septum. Figure 3 shows the gross appearance of an explanted heart with the pacing lead attached (Figure 3A) and the precise location of the pacing electrode within the septum (Figure 3D). Based on pathological examinations, there was no evidence of injury to the ventricle papillary muscles, chordae tendineae or the aortic valve. In addition, no clinically significant effusions or tamponade were observed.
Lead Stability
Echocardiographic imaging at the 4-week follow-up showed that pacing leads had dislocated from the septum in all three dogs implanted with the SET1 system. For the six dogs implanted with the modified SET2 system, no identifiable displacement was observed at 4 weeks. However, by the 12-week postimplant evaluation, all leads exhibited varying degrees of dislodgement. Specifically, one lead (animal no. 7) was slightly displaced while remaining within the myocardial tissue and retained stable electrical function. The remaining five leads were either too displaced to maintain electrical function or had exited the heart (Figure 4A).
Follow-up Pathology
Six dogs implanted with the SET2 system reached the study endpoint with a mean follow-up of 98 ± 7 days. Mild inflammatory changes were seen in the explants. Based on pathological examinations, the depth of the tissue fibrosis measured from the apical insertion point to the edge of the tissue fibrosis was 3.2 ± 1.1 mm. The area of the tissue fibrosis was 12.7 ± 3.9 mm2 (Figure 4B).
Discussion
This study demonstrates, for the first time, the technical feasibility and procedural safety of PTAIVS lead placement guided solely by echocardiography in an animal model. This approach eliminates the need for thoracotomy or fluoroscopic guidance and can be completed with a relatively short procedural time. The pacing electrodes were successfully positioned within the IVS, achieving clinically acceptable pacing thresholds and impedance. No major complications occurred during or immediately after the procedure, and no injury to the aortic valve or the ventricles was observed, as confirmed by pathological examination. Overall, this study presents a novel and feasible approach to IVS pacing via a percutaneous transapical route. Nevertheless, the issue of lead stability persists and requires further refinement of the pacing tools.
Despite the widespread adoption of transvenous implants, the incidence of complications, such as vascular laceration, stenosis, occlusion, tricuspid valve regurgitation and endocarditis, has increased over the years.18,19 Among these complications, the reported incidence of life-threatening thrombotic and embolic events, such as superior vena cava syndrome, ranges from 0.2% to 3.3% of all implants, with lead-induced endocarditis carrying a mortality risk as high as 12–31%.20–26 Leadless pacemakers, although addressing most lead-related issues, are associated with increased site thrombosis via the femoral access route and tricuspid valve dysfunction, and remain unsuitable for patients with restricted venous pathways due to non-standard venous anatomy or previous surgeries (e.g. inferior vena cava filter placement).27–29 Hence, a percutaneous lead delivery method has potential value.
In this study, we used two types of echocardiograms to guide the electrode directly from the chest skin to the target pacing site, with fluoroscopy used only after the procedure to document lead placement. The major advantage of this approach lies in the selection of the insertion site and wire delivery path, mitigating transvenous limitations and shortening the duration of the procedure. Eliminating fluoroscopy guidance also allows for greater flexibility, including the potential for bedside implantation.
Lead Stability and Contributing Factors
Two versions of pacing lead were used in this study. Despite the modification, lead stability issue persisted. During the initial phase, the Model 3830 electrode, with a 1.8-mm helix, was used. Although effective for traditional pacing, where it is securely fixed by penetrating the endocardium, this lead proved inadequate for transapical IVS placement. The short helix was not able to achieve deep electrode–myocardium adhesion, resulting in dislodgement in all six dogs within the first 4 weeks. In response, we used a modified implantation set (SET2) that incorporated a lead with a longer (10-mm) helix, designed to penetrate deeper into the myocardial tissue for enhanced fixation. This modification led to no lead dislodgement at the 4-week assessment, demonstrating improved initial adhesion. However, by the 12-week follow-up, all six leads using the SET2 system exhibited varying degrees of dislodgement, indicating that helix length alone is insufficient for long-term fixation.
Several factors may have contributed to persistent lead instability. Zhang et al. demonstrated that leads with smaller outer diameters require greater torque to penetrate the myocardium.30 This is likely because thinner leads are usually more flexible and have lower torsional stiffness, which causes rotational energy to be lost as the lead twists along its length rather than being efficiently transmitted to the tip. Such mechanical inefficiency could compromise precise control of tip engagement during active fixation, especially in deep septal penetration like PTAIVS, where more rotation force is needed, potentially contributing to incomplete anchoring and subsequent dislodgement.
In addition, Chapman et al. demonstrated that different lead designs exhibit distinct rotational response characteristics and that solid-core leads, such as the 3830 Model, may transmit torque less efficiently than stylet-driven leads, further challenging controlled tip fixation.31
Intrinsic myocardial properties and dynamic forces at the implantation site may also influence lead stability. Zhang et al. found substantial variation in lead–myocardial adherence depending on implantation site, noting that different lead designs perform optimally in specific cardiac regions.30 In our model, the canine intraventricular septum may be subject to more pronounced mechanical stress than anticipated, requiring greater resistance to motion for stable lead positioning over time. In addition, canine hearts exhibit faster contraction and relaxation relative to human hearts, resulting in different strain dynamics and potentially greater mechanical stress on the lead.32 This could increase the difficulty of maintaining stable lead fixation in the canine model.
Future Directions: Refinement of Tools and Techniques
Given previous findings and the results of the present study, future efforts should focus on refining both pacing tools and the procedural technique.
Tool refinements should explore alternative lead designs that optimise helix length and outer diameter to enhance mechanical engagement with myocardial tissue. For potential clinical translation in larger species or to humans, leads with standard or slightly increased diameters (5–7 Fr) may improve torque transmission and increase surface contact, adding fixation without causing excessive tissue trauma or complications during lead extraction. Material composition and lead stiffness should also be considered to reduce torque loss during rotation and to avoid abrupt dislodgement due to excessive torque build-up.
Furthermore, integrating supplementary anchoring systems, such as small tines or a mesh structure at the distal tip of the electrode, may provide additional mechanical support and help maintain long-term stability within the cardiac tissue. For instance, a side helix fixation mechanism, as reported by Keilegavlen et al. in left ventricular lead placement, offers enhanced mechanical engagement and could be adapted to improve anchoring in transapical IVS pacing.33
Technical refinements may include adjusting the lead trajectory or targeting more specific pacing sites within the septum. In this study, by delivering the pacing electrode into the mid-to-basal segment of the IVS, we achieved deep septal pacing. Deep septal pacing is defined as the mode in which a pacing lead is positioned deep in the septum but lacks a terminal R-wave in V1 and fails to reach the left bundle branch area.34 Although not capturing the conduction system, this type of pacing produces faster and more physiological activation of the ventricles than right apical pacing or right ventricular septal pacing. Although the variability of pacing performance within the IVS was not the primary focus of the present study, it would be worth exploring the possibility of achieving left ventricular septal pacing or even left bundle branch pacing in future investigations by adjusting the lead trajectory towards the basal inferoseptal segment of the left ventricle during the manoeuvring. This could engage the left-sided conduction system, further improving the efficacy of the pacing approach. Future studies should also consider the balance between procedural simplicity and precision. Techniques such as high right ventricular septal pacing, as reported by Katritsis and Calkins, may offer simpler, more accessible options that enhance the applicability and safety of the procedure, especially in complex or high-risk clinical scenarios.35
Importantly, anatomical differences between canine and human hearts should guide refinement of the implantation strategy. Transapical access is relatively straightforward in dogs due to the thoracic anatomy and apical position. In humans, however, navigating a steeper trajectory through thicker thoracic tissue while avoiding critical structures such as the lungs and internal mammary vessels could pose a significant challenge. Modifying the insertion trajectory, such as using a more basal entry site or an oblique angle, may improve safety and reduce mechanical stress on the lead from myocardial motion. Advanced imaging modalities, such as CT-guided navigation, could enhance precision. These species-specific variations require follow-up studies in human-specific models or cadaveric hearts to validate the safety and performance of this pacing approach prior to clinical translation.
Limitations
The primary focus of this study was on evaluating the technical feasibility of PTAIVS pacing, emphasising the route and method of lead placement. Coaxial needles, originally designed for soft tissue biopsy, were used as sheaths for electrode delivery. Although the needle facilitated straightforward skin penetration and access to the septum, challenges arose during its retrieval due to its metallic construction, which meant it could not be detached from the pacing lead. As a result, the lead was capped and secured subcutaneously at the apical insertion site rather than tunnelled to a generator pocket, which may have influenced external forces from subcutaneous movement and altered lead stability and affected durability assessments. Future studies should explore alternative sheaths or modify the existing needle design to enable seamless lead delivery and retrieval, allowing for full pacemaker implantation. In addition, this study focused on the short-term response of PTAIVS pacing in an animal model. Future investigations should include long-term observations in diverse pathological models to assess the safety and effectiveness of this approach.
Conclusion
Echocardiography-guided PTAIVS pacing offers an alternative access path for ventricular pacing with the potential to overcome the limitations of conventional transvenous approaches. Our study demonstrates the feasibility and initial safety of this method in an animal model. Further refinements of the pacing tools and implantation techniques are required to ensure sustained lead positioning and effective pacing.
Clinical Perspective
- Using dual echocardiographic guidance, pacing leads were introduced percutaneously through the cardiac apex and successfully delivered into the mid-to-basal region of the IVS in 12 dogs, without acute complications.
- PTAIVS pacing eliminates the need for intraprocedural fluoroscopy and allows for precise septal pacing, providing a novel pacing route for patients with inaccessible venous pathways.
- Although short-term feasibility and safety were demonstrated, further pacing tool refinements are needed to improve long-term lead stability and efficacy before clinical translation.