Impact of contact parameters on lesion dimensions during circular Pulsed-Field ablation in ex vivo and in vivo models

Abstract

Pulsed field ablation (PFA) is a novel nonthermal therapy for cardiac arrhythmias. The influence of contact force (CF), contact status, and application count on lesion characteristics in bipolar biphasic circular PFA remains uncertain. This study investigated lesion dimensions and acute histopathological changes using potato and porcine myocardial models. PFA was delivered in potato models under non-contact conditions or stratified contact forces (0–25 g, 25–50 g, > 50 g; 5 applications each). Porcine myocardium lesions were created with 1, 3, or 5 applications under contact or non-contact. Contact conditions yielded deeper lesions than non-contact in potato models (p < 0.001), with no significant differences among CF subgroups. In porcine myocardium, contact lesions were significantly deeper after three (4.82 ± 0.91 mm vs. 3.33 ± 0.74 mm; p < 0.05) and five applications (5.80 ± 0.49 mm vs. 3.55 ± 0.45 mm; p < 0.001) compared to non-contact. Lesion width was unaffected by contact status or application number. Histology demonstrated contraction-band necrosis, neutrophilic infiltration, edema, and hemorrhage with preserved microvasculature; these changes were more pronounced under contact. In the phantom model, lesion dimensions were predominantly governed by contact stability rather than contact force magnitude. In vivo validation confirmed that maintaining stable contact is a prerequisite for effective lesion formation. Furthermore, increasing the number of applications augmented lesion depth in porcine myocardium, while width remained stable.

Supplementary Information

The online version contains supplementary material available at 10.1038/s41598-026-42503-1.

Introduction

Catheter ablation has become a cornerstone therapy in the invasive management of cardiac arrhythmias^1,2. Among the available energy sources, radiofrequency ablation, cryoablation, and, more recently, pulsed-field ablation (PFA) are the most widely used modalities^3–5. PFA is a nonthermal technique that delivers high-voltage electric fields to induce irreversible electroporation of myocardial cells^6–8. This emerging technology has garnered significant interest due to its promising safety profile and procedural efficacy⁹.

Although the choice of energy source is central to lesion formation, successful ablation outcomes depend on a multifactorial interplay of biophysical parameters^10–14. Among these, catheter-tissue contact force (CF) and the number of applications are well-established determinants of lesion size and transmurality in thermal ablation^15,16, particularly with radiofrequency energy^17,18. However, in the context of PFA, especially with bipolar biphasic circular ablation catheters, data characterizing the impact of CF on lesion dimensions remain limited. Additionally, the acute histopathological features associated with varying contact conditions specifically using circular PFA catheters have not been fully elucidated. Moreover, recent studies on multi-electrode catheters have utilized tissue proximity index (TPI) as an alternative to traditional contact force to indicate catheter-tissue contact^16,19–21. Therefore, in this experiment, we also adopted a circular catheter compatible with TPI for the animal model to assess tissue contact.

A clearer understanding of how catheter–tissue CF influences lesion morphology and tissue response in PFA may inform procedural optimization and improve clinical outcomes. In this study, we systematically investigated the effects of CF on lesion depth and width, as well as the corresponding histopathological changes, using both potato and porcine models with a bipolar biphasic circular PFA catheter.

Methods

Study Design

Lesion dimensions produced by a bipolar, biphasic, circular PFA catheter were evaluated using both potato and porcine models with identical pulse width (Fig. 1). In the porcine model, ablations were performed via an endocardial approach. We systematically assessed the effects of CF and the number of applications (1, 3, or 5) on lesion morphology, building upon preliminary findings from the potato model. Myocardial tissue containing ablation lesions was harvested within two hours post-ablation for histopathological analysis, including hematoxylin-eosin (HE) and Masson’s trichrome staining to characterize acute myocyte injury and fibrotic changes. All experimental procedures involving porcine myocardial tissues were reviewed and approved by the Animal Ethics Committee of West China Hospital (Approval No. 20250227083). The study was conducted in accordance with institutional guidelines and the ARRIVE 2.0 framework for reporting animal research.

Fig. 1.

Circular Pulsed Ablation Catheter and Pulse Schematic.

Pulsed-field ablation system

The PFA system used for both models comprised the PulsedFA^® Cardiac PFA Catheter (loop diameter: 15 mm; 7 electrodes; 4-mm interelectrode spacing), a generator (LEAD-PFA^® Cardiac PFA System), and a 3D electroanatomic mapping system (LEAD-Mapping^® Cardiac Mapping System), all manufactured by Sichuan Jinjiang Electronic Medical Device Technology Co., Ltd. In the potato model, ablation parameters were set at 1800 V (amplitude) with an effective duration of 150 µs per application (repeated for 5 applications), whereas the porcine model received ablation at 1800 V (amplitude) with 400 µs effective duration, both protocols employing identical pulse width.

Potato model

Russet potatoes (Solanum tuberosum) were sliced transversely into 10-mm-thick cylindrical sections and stored in distilled water to minimize enzymatic browning. During the experiment, specimens were immersed in 0.9% saline at room temperature in a custom electrophysiologic chamber. A circular PFA catheter was mounted on a robotic arm integrated with a triaxial force transducer to precisely control CF. A total of 58 potato slices were used to evaluate lesion formation. Ablation was conducted under four predefined contact conditions:

1. non-contact (suspension 1.0 mm above tissue in 0.9% saline).

2. 0 g < CF ≤ 25 g.

3. 25 g < CF ≤ 50 g.

4. CF > 50 g.

Following ablation, potato specimens were incubated in 2,3,5-triphenyltetrazolium chloride solution for 18 h to delineate lesion boundaries. Lesion width and depth were measured with digital calipers.

Porcine model

Three Landrace swine were acclimatized for at least 7 days in a standard facility maintained at 20–25 °C with 50–60% relative humidity under a 12-hour light/dark cycle. To facilitate postoperative monitoring, animals were individually housed in enclosures complying with large animal welfare standards, with ad libitum access to water and twice-daily standard feeding. For the ablation procedure, general anesthesia initiated with intramuscular atropine (2 mg) 15 min before induction. Anesthesia induction was performed via combined intramuscular injection of Xylazine Hydrochloride (400 mg) and Zoletil (250 mg, a compound formulation containing dihydroetorphine hydrochloride to ensure adequate analgesia). Anesthesia maintenance was sustained with continuous intravenous infusion of Propofol (2 mg/kg/h) and Xylazine Hydrochloride (1.5 mg/kg/h), titrated based on real-time physiological monitoring. Endotracheal intubation was implemented to deliver continuous mechanical ventilation throughout the procedure. Systemic anticoagulation was maintained with intravenous heparin to achieve an activated clotting time of 250–350 s. Ventricular PFA was performed using the same catheter system. Catheter-tissue contact was monitored in real time using the TPI, an impedance-based metric calculated as the ratio of the real-time impedance difference between adjacent electrodes to the baseline impedance difference. TPI values were visually represented by a color-coded display (green: no contact; yellow/orange: contact) to facilitate intraprocedural assessment. Each ablation site was recorded on the 3D electroanatomic mapping system and subsequently harvested for analysis (Supplementary Fig. 1). Animals were euthanized within two hours post-ablation for tissue collection.

Lesion morphometry and histopathological assessment

Two hours after PFA, a 2% solution of 2,3,5-triphenyltetrazolium chloride was administered intravenously at a dose of 1 mL/kg. Following a 15-minute circulation period, animals were euthanized. Ventricular tissue was excised and fixed in 4% neutral buffered formalin for seven days. Longitudinal sections were made along the ventricular axis for gross inspection. Lesions were localized based on mapping data and sectioned at 3-mm intervals perpendicular to the endocardial surface. Only non-transmural lesions were included in morphometric analysis. Lesion depth and width were measured using digital calipers. Representative tissue samples were embedded in paraffin, sectioned at 4 μm thickness, and stained with HE and Masson’s trichrome. Histological features, including myocyte necrosis, were evaluated by a board-certified veterinary pathologist blinded to experimental conditions.

Statistical analysis

All statistical analyses were conducted using R Studio (version 4.4.3). Continuous variables are expressed as mean ± standard deviation. Group comparisons were performed using one-way ANOVA followed by Tukey’s post hoc test. A two-tailed p value < 0.05 was considered statistically significant.

Results

Potato samples

As shown in Fig. 2, CF had no significant effect on the maximum lesion width, except for a notable increase in the 25 g < CF ≤ 50 g group compared to the non-contact group (p < 0.01). All lesions were non-transmural and confined within the 10-mm tissue thickness. For the minimum lesion width, all contact groups produced significantly wider lesions than the non-contact group (all p < 0.001), with no significant differences among the contact groups. A similar trend was observed for lesion depth, where all contact groups generated significantly deeper lesions than the non-contact group (all p < 0.001), with no significant inter-group differences within the contact groups.

Fig. 2.

Bar Charts of Lesion Dimensions under Different Catheter-Tissue Contact Force in the Potato Model. Legend: The height of each bar represents the mean value of lesion depth or width. The horizontal line above each bar indicates Mean + SD, and the line below indicates Mean - SD. Each point represents a single sample. * indicates p < 0.05, ** indicates p < 0.01, and *** indicates p < 0.001.

Figure 3 summarizes the lesion dimensions under different contact conditions. In the non-contact group, minimum and maximum lesion depths were 0.98 ± 0.32 mm and 1.78 ± 0.70 mm, respectively. In the contact groups, the 25 g < Force ≤ 50 g group exhibited the deepest lesions, with a minimum depth of 2.47 ± 0.88 mm and a maximum depth of 3.62 ± 0.77 mm.

Fig. 3.

Lesion Dimension Variability by Catheter-Tissue Contact Force Under PEF in the Potato Model.

Porcine myocardium

A total of 36 ventricular lesions were created in vivo using the circular PFA catheter, stratified by contact condition (contact vs. non-contact) and number of applications (1, 3, or 5). Figure 4 demonstrates that lesion width did not differ significantly between contact and non-contact groups at any application count. In contrast, lesion depth was significantly greater in the contact group compared to the non-contact group after both 3 applications (p < 0.05) and 5 applications (p < 0.001). Figure 5 illustrates that lesion width remained essentially unchanged across different ablation counts in both the non-contact and contact groups. However, lesion depth increased significantly with additional applications. In the non-contact group, lesions created with 3 and 5 applications were significantly deeper than those created with a single application (p < 0.001). The contact group demonstrated a similar trend, with lesions after 5 applications producing deeper lesions than 3 applications. Figure 6 compares lesion dimensions across both contact forces and application numbers. The deepest lesion (5.8 ± 0.49 mm) occurred in the contact group following 5 applications, while the shallowest lesion (1.66 ± 0.65 mm) was seen after a single application in the non-contact group.

Fig. 4.

Bar Charts of Lesion Dimensions under Different Catheter-Tissue Contact Status in the Porcine Model. Legend: The height of each bar represents the mean lesion depth or width. The horizontal line above each bar indicates Mean + SD, and the line below indicates Mean - SD. Each point represents a single sample. * indicates p < 0.05, ** indicates p < 0.01, and *** indicates p < 0.001.

Fig. 5.

Bar Charts of Lesion Dimensions under Different Ablation Applications in the Porcine Model. Legend: The height of each bar represents the mean lesion depth or width. The horizontal line above each bar indicates Mean + SD, and the line below indicates Mean - SD. Each point represents a single sample. * indicates p < 0.05, ** indicates p < 0.01, and *** indicates p < 0.001.

Fig. 6.

Pulsed Field Ablation Lesion Modulated by Contact Status and Ablation Count in the Porcine Model.

Histological findings

Both HE and Masson’s trichrome staining were performed on the ablation lesions. Figure 7 displays tissue characteristics following 5 contact applications. Figure 7A shows the ablation zone with HE staining, and Fig. 7B and D provide higher-magnification views of the region. Figure 7F displays the corresponding plane stained with Masson’s trichrome, with Fig. 7C and E showing magnified insets. In the enlarged views, contraction-band necrosis, neutrophilic inflammatory infiltrates, interstitial edema, and hemorrhage were observed. Notably, small blood vessels remained intact.

Fig. 7.

Histopathological Staining of Ventricular Myocardium Following 1800 V (amplitude), 400µs Effective Duration × 5 Applications under Contact Status. Legend: Panel A shows HE staining with inflammatory infiltration and interstitial edema. Panels B and D are magnified views of Panel A. Panel F shows Masson’s trichrome staining, and Panels C and E are magnified views of Panel F. Panel B illustrates contraction band necrosis (black arrow) with preserved small vessel structure (hollow triangle). Panel C displays fibrous connective tissue surrounding small vessels (solid triangle). Panel D highlights contraction-band necrosis (black arrow), localized hemorrhage, and neutrophilic infiltration (white arrow). Panel E and F show fibrous connective tissue (asterisks).

Figure 8 shows histology after 5 non-contact applications. Contraction-band necrosis was observed, but inflammatory infiltrates were more superficial, with minimal edema and hemorrhage.

Fig. 8.

Histopathological Staining of Ventricular Myocardium Following 1800 V (amplitude), 400µs Effective Duration × 5 Applications under Non-Contact Status.

Discussion

This study provides a detailed assessment of the relationship between CF and acute lesion dimensions following PFA using a bipolar, biphasic circular catheter. Our findings demonstrate that catheter-tissue contact status exerts a more significant influence on lesion depth than on lesion width. Among the contact groups, incremental increases in CF did not lead to significantly differences in lesion dimensions in the potato model. Additionally, increasing the number of PFA applications was associated with deeper lesions in myocardial tissue. Histopathological examination of acute-phase lesions consistently showed contraction-band necrosis, neutrophilic infiltration, interstitial edema, focal hemorrhage, and early fibrotic changes.

Prior studies using unipolar pulsed focal catheters have reported that increased CF correlates with greater lesion depth^15,22,23. In contrast, investigations of bipolar PFA systems^24–26, whether using focal point or circular catheter configurations, demonstrate a stronger correlation between contact status and lesion depth rather than magnitude of CF, as the primary determinant of lesion depth. Notably, a recent study of a bipolar biphasic circular PFA catheter reported a significant association between lesion depth and CF magnitude²¹. In our study, catheter-tissue contact had a greater impact on lesion depth than did incremental increases in CF. Furthermore, in the potato model, CF only modestly influenced minimum lesion width, with no appreciable effect on maximum lesion width. This suggests that lesion width, particularly the maximal dimension, may be more dependent on factors such as applied voltage, pulse duration, and the number of applications rather than CF.

With respect to the number of PFA applications, prior research has shown that lesion depth increases with additional applications, regardless of electrode configuration (unipolar or bipolar) or catheter design^21,27. In our investigation, although the groups were limited to 1, 3, or 5 applications, the general trend observed was consistent with previous findings and remained independent of catheter-tissue contact status. Notably, in circular PFA catheters, we observed no significant correlation between lesion width and the number of applications within the defined parameters.

In previous study, cardiac computed tomography data suggest the average left atrial wall thickness is approximately 1.4 ± 0.2 mm²⁸, a value that has been linked to long-term outcomes following pulmonary vein isolation in atrial fibrillation. In the study by Teres et al.²⁹, the anterior wall thickness ranged from 0.3 to 4.5 mm, and the posterior wall from 0.3 to 2.3 mm. Other contemporary studies have reported similar ranges: 0.86–4.43 mm anteriorly and 0.51–2.21 mm posteriorly³⁰. Based on our data, bipolar biphasic circular PFA using 3 applications (Pulse duration: 400 µs, Voltage amplitude: 1800 V) under contact conditions resulted in a mean lesion depth of 4.82 ± 0.91 mm. Experimental models suggest that three applications under contact conditions may generate lesion depths approaching atrial wall thickness dimensions.

The histological changes following PFA exhibit a marked temporal progression. Previous studies have shown that by 4 weeks post-ablation, lesions typically exhibit extensive fibrosis, mild focal inflammation, and preserved microvascular architecture³¹. At 7 days post-ablation, contraction-band necrosis, inflammatory cell infiltration, collagen deposition, and early fibrotic remodeling have been consistently observed²¹. Compared to prior studies examining 7-day post-ablation specimens, the samples collected 2 h post-PFA in our study demonstrated more robust acute inflammatory responses, although fibrotic remodeling remained in its early stages. Some reports have also identified superficial coagulative necrosis at 7 days post-PFA, suggesting residual thermal injury^22,32. Our histopathological analysis of the acute phase revealed consistent features of contraction-band necrosis, neutrophilic infiltration, hemorrhage, and interstitial edema. These findings are in line with existing literature on acute-phase PFA-induced tissue responses^23,27,33. Regarding vascular impact, while our histological analysis showed that microvasculature within the lesion zone remained structurally intact acutely, Tam et al. recently reported a 17.1% increase in vessel wall thickness and a 10.1% reduction in luminal diameter at 3 months post-ablation³⁴. These findings suggest that vascular safety cannot be assessed solely based on acute data, and future long-term studies are warranted to evaluate potential chronic remodeling.

Limitations

This study has several limitations. First, the potato model, while useful for initial visualization and quantification, does not replicate the mechanical properties, catheter stability, or perfusion dynamics of in vivo cardiac tissue. Second, regarding contact verification in the porcine model, we utilized TPI rather than quantitative contact force sensing. While impedance is increasingly recognized as a superior metric in multipolar PFA catheters, this methodology precludes a direct correlation analysis between mechanical force values and lesion dimensions. Third, the limited sample size may have introduced Type II error regarding the lack of significant differences between contact force subgroups, warranting validation in larger sample sizes. Finally, our histological analysis was restricted to the acute post-ablation period. It should be noted that acute PFA lesions may encompass a zone of reversible electroporation, and thus acute dimensions may not linearly translate to chronic durable lesion sizes. Consequently, future survival studies are warranted to validate these findings long-term and evaluate lesion maturation.

Conclusions

This study demonstrates that catheter-tissue contact status is a prerequisite for effective lesion formation using a bipolar, biphasic circular PFA catheter. In the phantom model, lesion dimensions remained stable across varying CF levels, indicating that contact stability, rather than the magnitude of force, was the primary determinant of lesion size in this setting. In vivo validation further confirmed that maintaining stable contact is critical for maximizing lesion depth. Acute-phase histopathology revealed characteristic contraction-band necrosis, accompanied by neutrophilic infiltration, interstitial edema, and focal hemorrhage, with preservation of the microvascular architecture. Collectively, these findings highlight that ensuring stable catheter-tissue coupling is fundamental for optimizing lesion formation with this PFA technology.

Supplementary Information

Below is the link to the electronic supplementary material.

Author contributions

Xianjin Hu conceptualized the study, curated and analyzed data, and wrote the original draft; Wenjie Li, Bangjiaxin Ren, Fanghui Li, and Aobo Gong contributed to data analysis and validation; Xianzhao Zhu and Luyang Xie conducted investigation and data curation; Ying Cao and Zexi Li performed visualization and assisted in manuscript writing; Rui Zeng designed the methodology, supervised the project, and revised the manuscript. All authors reviewed and approved the final version.

Funding

This work was supported by Science and Technology Department of Sichuan Province (Grant Number: 2024YFFK0046), 1∙3∙5 Project for Disciplines of Excellence-Clinical Research Incubation Project, West China Hospital of Sichuan University (Grant Number: 2023HXFH002). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Data availability

The data that support the findings of this study are available from the corresponding author on reasonable request.

Declarations

Competing interests

Xianzhao Zhu and Luyang Xie are employees of Sichuan Jinjiang Electronic Medical Device Technology Co., Ltd. Their roles in the study were limited to: (1) Technical assistance in configuring and calibrating the PFA equipment during experiments. (2) Operational support for the 3D Electroanatomic Mapping System. (3) Data collection and preliminary organization of raw experimental outputs. It is important to note that these two authors were not involved in the design of the experimental protocol, data analysis/interpretation, manuscript preparation, or any decision-making processes related to the study outcomes. Their participation was strictly confined to technical execution under the supervision of the independent research team. All remaining authors have declared no conflicts of interest.

Conflict of interest

Xianzhao Zhu and Luyang Xie are employees of Sichuan Jinjiang Electronic Medical Device Technology Co., Ltd. All remaining authors have declared no conflicts of interest.