Optimizing radiofrequency ablation parameters for perianal fistula treatment: A porcine model study

Abstract

BACKGROUND

The effective treatment of perianal fistulas while ensuring sphincter preservation remains challenging. Radiofrequency ablation (RFA) is a promising approach; however, its clinical translation is limited because energy settings have been extrapolated from non-fistulous tissues, and most probes deliver only focal heating. We designed a 50-mm segmental RFA catheter to achieve maximal granulation tissue ablation while maintaining uniformity along the tract.

AIM

To identify the optimal temperature-time parameters for uniform, effective endoluminal RFA of perianal fistulas.

METHODS

Ex vivo screening in fresh pork loin selected four candidate energy settings defined by temperature and duration: 120 °C/20 seconds, 120 °C/24 seconds, 100 °C/20 seconds, 100 °C/24 seconds. These were applied endoluminally to the perianal fistula tracts created in four pigs. After RFA, each tract was divided into the inner, middle, and outer thirds for histological analysis. The primary endpoint was the ablation-to-granulation tissue area ratio, and burn depth was summarized as the equivalent radius (r_e).

RESULTS

The 120 °C/24 seconds setting achieved the highest ablation-to-granulation ratio (70.9 ± 7.5%), exceeding 120 °C/20 seconds (51.9 ± 8.3%), 100 °C/20 seconds (46.9 ± 4.0%), 100 °C/24 seconds (43.7 ± 6.1%). It also produced the greatest burn depth (r_e 0.46 ± 0.05 mm vs 0.30 ± 0.04 mm, 0.24 ± 0.03 mm, and 0.19 ± 0.03 mm; all pairwise P < 0.001). The depth was consistent across segments.

CONCLUSION

In a porcine fistula model, segmental RFA at 120 °C for 24 seconds produced the deepest, most homogeneous ablation, supporting its use as a reference setting for further studies.

Key Words: Fistula; Perianal fistula; Radiofrequency ablation; Swine; Thermal therapy

Core Tip: Endoluminal radiofrequency ablation (RFA) is a sphincter-preserving option for perianal fistulas. However, standardized energy settings validated in true fistula tissue are lacking. Using a porcine model of perianal fistula, we performed ex vivo prescreening and in vivo histological validation of four candidate settings with a 50-mm segmental RFA catheter. The 120 °C/24 seconds setting produced the most effective ablation (highest ablation/granulation ratio) with the deepest and most consistent burn depth along the tract, providing a practical reference parameter for further translational studies.

INTRODUCTION

Perianal fistula is a chronic and complex condition that poses significant challenges in treatment. The primary objective of therapy is the complete elimination of the fistula tract. Although surgical interventions can be effective, they frequently lead to complications such as fecal incontinence, which can severely affect the quality of life of the patient. Traditional methods, such as fistulotomy, have progressively transitioned towards to more minimally invasive techniques that aim to maintain sphincter integrity and reduce postoperative wounds. Recent innovations, including thermal therapies, such as video-assisted anal fistula treatment (VAAFT)[1,2] and fistula-tract laser closure (FiLaC)[3,4], have gained prominence because of their efficacy in destroying fistula tissue and facilitating tract shrinkage. Building on these advancements, our previous research introduced a novel radiofrequency ablation (RFA) device that has shown promise as a therapeutic option for the management of perianal fistulas[5]. This RFA device allows endoluminal treatment, offering a minimally invasive strategy that specifically targets the fistula tract while minimizing damage to adjacent tissues. Notably, our system is based on a 50-mm segmental electrode rather than a focal tip “dot-ablation” design, enabling more uniform longitudinal energy delivery.

RFA is based on the principle of inducing thermal damage, whereby the tissue temperature is elevated to a predetermined range, resulting in protein denaturation, irreversible disruption of cellular membranes, and subsequent cell death[6]. The characteristics of the ablation zone can be accurately controlled by manipulating various parameters, including temperature, electrode size, duration of operation, and total energy delivered (measured in joules, J)[6]. RFA utilizes a high-frequency alternating current, which is particularly beneficial in medical contexts owing to its reduced stimulation of adjacent tissues and lower risk of electrocution, thereby enhancing safety[7-9]. This technique has been extensively and safely employed in a range of medical applications, such as the treatment of hepatic tumors[8-10], thyroid nodules[7], and varicose veins[11,12], demonstrating proven clinical efficacy and safety profiles in these organ systems. Given its proven safety and effectiveness in various applications, RFA is a promising option for minimally invasive treatment of perianal fistulas.

Our previous porcine study evaluated the feasibility of RFA for perianal fistulas using endovenous (varicose vein) temperature-time settings and yielded suboptimal outcomes[5]. Building on this finding, the present work was designed to address a distinct and clinically relevant question: What fistula-specific thermal dosing improves ablation effectiveness with a 50-mm segmental RFA catheter? We postulated that the prior negative results may be attributable to insufficient thermal damage caused by the non-fistula-optimized parameters[11]. Recent clinical studies involving human subjects have reported limited success rates following RFA interventions, suggesting that inadequate energy dosing and/or inconsistent energy delivery may contribute to limited efficacy[13,14]. Given this context, we hypothesized that identifying optimal perianal fistula-specific thermal parameters would improve ablation effectiveness. Therefore, the primary objective of the present study was to compare candidate temperature-time settings and determine the optimal RFA parameters for perianal fistula treatment. As a secondary outcome, we assessed whether the segmental RFA system could achieve uniform ablation along the entire fistula tract.

MATERIALS AND METHODS

Four distinct RFA conditions were evaluated in a porcine perianal fistula model using the energy settings determined based on the preliminary findings described below (Figure 1A).

Figure 1 Experimental set-up and novel radiofrequency ablation device. A: Schematic of the porcine fistula model used for in-vivo optimization of radiofrequency ablation (RFA) parameters. Three fistula tracts were surgically created in each animal and kept patent with setons, as illustrated in the enlarged inset. For ablation, seton was removed and the RFA catheter introduced endoluminally, as confirmed by the intra-operative photograph; B: Prototype segmental RFA catheter. The distal 50 mm copper-colored segment constitutes the active ablation blade, and graduations at 10 mm intervals along the shaft allow precise depth control. This 50 mm active segment delivers a uniformly distributed ablation along its entire length, thereby overcoming the focal-tip limitation of conventional devices.

RFA system

The RFA system employed in this study consisted of a proprietary RFA generator (PRO-GEN™ PG-50, RF Medical Ltd., Seoul, Republic of Korea) and a novel, single-use bipolar RFA catheter (FISTU-TIP™, RF Medical Ltd., Seoul, Republic of Korea) (Figure 1B). The generator operated at a standard frequency of 480 kHz and delivered a maximum power output of 50 W. The FISTU-TIP catheter was specifically designed for endoluminal treatment of fistula tracts. It featured a flexible shaft with an outer diameter of 2.0 mm at the shaft and 3.0 mm at the tip to facilitate insertion into the fistula tract. The distal end of the catheter incorporated a unique 50 mm long segmental heating element designed to deliver circumferential thermal energy uniformly along the treated segment. The system was operated in temperature-controlled mode. Thermocouples integrated within the heating element segment continuously monitored the tissue interface temperature. The generator automatically modulated the power output to precisely maintain the pre-set target temperature (5-200 °C) within ± 3 °C for the selected duration (5-300 seconds).

Selection and experimental validation of candidate parameters for optimal RFA settings

The perianal fistula tract wall typically presents a thickness ranging from approximately 1-3 mm[15], with variations influenced by factors such as inflammation, chronicity, and individual patient characteristics. In instances of chronic or recurrent fistulae, the wall thickness may increase significantly, reaching approximately 3-5 mm[16]. This increase can adversely affect the efficacy of thermal ablation treatments, thereby necessitating a customized approach to ensure the appropriate delivery of energy[13]. Consequently, an ex vivo validation study was conducted using pork loins to identify suitable RFA parameters that could effectively ablate the full thickness of the tract wall while minimizing collateral damage. Various combinations of temperature and duration were systematically tested to determine optimal thermal conditions. Segments of pork loin were evenly sectioned to mimic the thickness of the fistula wall, and the RFA probe was positioned centrally within the folded pork loin tissue to simulate the conditions of the fistula tract (Figure 2A). Following activation of the RFA device under each experimental condition, the area of the thermally ablated tissue was quantified (Figure 2B). Each experimental condition was repeated eight times, and the mean ablation area was analyzed to determine the parameters that effectively balanced the ablation depth with minimal thermal injury.

Figure 2 Ex vivo platform for preliminary screening radiofrequency ablation temperature-time settings. A: Schematic of the segmental radiofrequency ablation catheter positioned along the midline of a fresh porcine loin; the loin was then folded over the catheter so that the loin enveloped the electrode and maintained continuous, uniform contact during energy delivery; B: Loin re-opened after ablation (representative condition: 100 °C/28 seconds), showing a well-demarcated linear ablation tract. This configuration enabled direct caliper measurement of ablation depth and length for each candidate settings.

The preliminary findings indicated that exposure to temperatures of 100 °C and 120 °C for durations exceeding 24 seconds (for instance, 28 seconds) consistently resulted in significant thermal damage surpassing 5 mm, thereby presenting a risk of inadvertent injury. In contrast, durations of 20 seconds and 24 seconds at these temperatures facilitated controlled ablation depths while effectively minimizing excessive damage. Consequently, four candidate RFA conditions, 120 °C for 20 seconds, 120 °C for 24 seconds, 100 °C for 20 seconds, and 100 °C for 24 seconds, were identified as optimal candidates for further in vivo evaluation, reflecting a judicious balance of efficacy, safety, and potential clinical applicability.

Creation of the porcine perianal fistula model and histological assessment of ablation depth

A porcine model for the study of perianal fistulas was developed following the validated protocol reported in our previous study[5]. Four healthy pigs with a mean weight of 48 kg (range, 46-59 kg) were procured from Cronex Inc. (Seoul, Republic of Korea). For each subject, three iatrogenic fistula tracts were surgically created and maintained using rubber setons for a period of four weeks to facilitate the development of chronic fistulous changes.

All experimental procedures were approved by the Institutional Animal Care and Use Committee of Cronex Inc. (CRONEX-IACUC: 202302002) and adhered to the guidelines set forth by the American Veterinary Medical Association regarding the humane euthanasia of animals[17]. Surgical interventions, including the creation of fistulas, placement and removal of setons, and administration of RFA, were performed under general anesthesia and were designed to minimize pain or discomfort to the animals. Anesthesia was initiated through intramuscular injections of Zoletil^® 50 (0.5 mL/kg) and Rompun^® (0.2 mL/kg), followed by intravenous administration of Zoletil^® 50 (0.3 mL/kg) and Rompun^® (0.5 mL/kg). General anesthesia was sustained with inhaled isoflurane after endotracheal intubation, and the animals were maintained in a prone position throughout the surgical procedures.

Two weeks post-seton removal, each fistula tract underwent treatment with RFA under one of four experimental conditions: 120 °C for 20 seconds, 120 °C for 24 seconds, 100 °C for 20 seconds, and 100 °C for 24 seconds. To evaluate the tissue response and the extent of ablation, the treated fistula tracts were excised immediately after RFA on the same day. Each tract was meticulously dissected into three anatomical segments (inner, middle, and outer) along the longitudinal axis. Histological analysis was performed using hematoxylin and eosin (H&E) staining. A total of 36 specimens were examined, with each experimental group comprising nine samples (three tracts per porcine subject divided into three segments). The fistula tract was considered an independent experimental unit, and the inner, middle, and outer specimens were treated as within-tract repeated measures to evaluate longitudinal uniformity rather than independent samples.

Methodology for measuring depth of thermal ablation

The extent of thermal ablation was assessed by histological analysis. H&E staining was used to evaluate the tissue changes induced by RFA, with a specific focus on the depth of thermal injury within the granulation tissue lining the fistula tract. The primary outcome measure was the ratio of the ablation area to the granulation tissue area, which served as a quantitative indicator of the ablation efficacy. Histological images were independently examined by two independent examiners, Hong KD and Yoon S, who manually annotated the regions of ablation and granulation using ImageJ software. Image segmentation was performed manually, and the mean values from both observers were used for subsequent analyses (Figure 3). The ablation area was characterized by distinct features of coagulative necrosis, including eosinophilic cytoplasm, nuclear pyknosis or karyolysis, loss of normal tissue architecture, and the absence of vascular structures[18]. Conversely, granulation tissue was identified by the presence of capillary proliferation, fibroblast infiltration, and a loose extracellular matrix that typically appeared more basophilic and structurally organized[19]. The differentiation in color under H&E staining further facilitated identification: Granulation tissue generally exhibited a bright pink hue, whereas burned or necrotic areas appeared violet to deep red owing to protein denaturation and cellular degradation[20].

Figure 3 Representative histological section illustrating Image J-based quantification of ablation and granulation tissue. A: Hematoxylin and eosin -stained cross-section of an outer fistula-tract segment harvested after radiofrequency ablation ablation (120 °C/24 seconds), showing a centrally obliterated lumen surrounded by a concentric zone of thermal injury; B: The same section processed in Image J, with color overlays delineating tissue compartments used for morphometry: Residual lumen (black), carbonized/charred core (violet), coagulative necrosis (red), transition zone (orange) and residual granulation tissue (yellow).

To evaluate the uniformity of ablation across the length of the fistula tract, each tract was divided into three longitudinal segments: Inner, middle, and outer. The ablation-to-granulation tissue area ratio was calculated separately for each segment, thereby enabling the assessment of both the consistency of treatment and the depth of tissue destruction.

Statistical comparison and methodology

The ratio of ablation area to granulation tissue area was compared across the four RFA treatment groups using a one-way analysis of variance (ANOVA); when the omnibus F-test was significant, pairwise differences were explored with Tukey’s honestly significant difference (HSD) procedure (α = 0.05). The absolute burn depth was then evaluated by recording the long and short axes of each ablation ellipse and calculating the area-preserving equivalent radius (r_e), which was defined as the radius of a circle with an area equal to that of the measured ellipse, using the formula r_e = sqrt (long × short)/2. Assumptions of normality (Shapiro-Wilk) and homogeneity of variance (Levene’s test) were assessed. For each group, descriptive statistics (mean ± SD and 95% confidence intervals) were tabulated for the long and short axes of each ablation ellipse, and for r_e. Finally, intra-tract uniformity was assessed by analyzing the ablation-to-granulation ratio across the inner, middle, and outer segments of each fistula tract with a repeated-measures ANOVA that accounted for within-specimen dependency. Significant segment effects were dissected using Bonferroni-adjusted post-hoc tests. Throughout, two-sided P values < 0.05 were considered statistically significant. All statistical analyses were performed using R software (version 4.2.0; R Foundation for Statistical Computing, Vienna, Austria).

RESULTS

The study included four pigs, each subjected to seton placement at three distinct anatomical positions (3, 9, and 12 o’clock). Due to inadvertent displacement of the seton in two pigs, the final analysis was conducted on nine fistula tracts. Each tract was further divided into three segments (inner, middle, and outer), yielding 27 segments for histological examination.

Comparison of ablation-to-granulation tissue area ratios across experimental groups

The mean ratios of ablation-to-granulation tissue area for the four experimental groups, 120 °C for 20 seconds (group 1, 120 °C/20 seconds), 120 °C for 24 seconds (group 2, 120 °C/24 seconds), 100 °C for 20 seconds (group 3, 100 °C/20 seconds), and 100 °C for 24 seconds (group 4, 100 °C/24 seconds), were recorded as 51.9 ± 8.3%, 70.9 ± 7.5%, 46.9 ± 4.0%, and 43.7 ± 6.1%, respectively. One-way ANOVA revealed a statistically significant difference in ablation efficacy among the groups (P < 0.001, F = 18.7). Subsequent post-hoc Tukey HSD analysis indicated that group 2 (120 °C for 24 seconds) exhibited significantly higher ablation/granulation ratios compared to group 1 (P = 0.00015), group 3 (P = 0.00002), and group 4 (P = 0.000003), thereby demonstrating superior tissue ablation under this specific parameter. No significant differences were observed between group 1 and group 3 (P = 0.525) or between group 3 and group 4 (P = 0.856), suggesting comparable efficacy within these lower-energy conditions (Figure 4).

Figure 4 Box plot illustrating the ablation-to-granulation tissue area ratios across the four experimental groups. Group 2 (120 °C for 24 seconds) exhibited the highest ablation/granulation ratios. Group 2 was significantly higher than groups 1, 3, and 4 by Tukey’s honestly significant difference post hoc test (all P < 0.001).

Complementing the ratio analysis, the absolute depth of thermal injury, expressed as the area-preserving equivalent radius (r_e), also differed across the treatment arms. The mean ± SD values were 0.30 ± 0.04 mm for group 1, 0.46 ± 0.05 mm for group 2, 0.24 ± 0.03 mm for group 3, and 0.19 ± 0.03 mm for group 4 (Table 1). One-way ANOVA demonstrated a significant group effect (F = 24.3, P < 0.001), and the Tukey-HSD post-hoc test confirmed that group 2 achieved a significantly greater burn depth than the other three groups (all pairwise P < 0.001).

Table 1 Absolute ablation-depth parameters across radiofrequency ablation treatment groups: Major(long) axis, minor(short) axis, and area-preserving equivalent radius (r_e).

Group		mean ± SD (mm)	95%CI
1	Long axis	1.0396 ± 0.4924	0.6611-1.4180
	Short axis	0.3602 ± 0.1793	0.2224-0.4981
	equivalent radius (re)	0.3006 ± 0.1310	0.1999-0.4013
2	Long axis	1.4168 ± 0.2584	1.1457-1.6880a
	Short axis	0.5900 ± 0.1814	0.3997-0.7803a
	equivalent radius (re)	0.4550 ± 0.0995	0.3506-0.5595a
3	Long axis	0.8472 ± 0.3273	0.5037-1.1906
	Short axis	0.2708 ± 0.0997	0.1662-0.3754
	equivalent radius (re)	0.2377 ± 0.0823	0.1514-0.3240
4	Long axis	0.8218 ± 0.3913	0.4112-1.2324
	Short axis	0.1813 ± 0.0889	0.0880-0.2746
	equivalent radius (re)	0.1925 ± 0.0915	0.0965-0.2885

^aP < 0.05 significantly greater than groups 1, 3, and 4.

CI: Confidence interval.

Uniformity of RFA-induced ablation within individual fistula tracts

The ablation/granulation tissue ratios were compared among the inner, middle, and outer segments to evaluate the consistency of ablation along the length of each fistula tract. The repeated-measures ANOVA indicated no statistically significant differences between the segments (P = 0.138, F = 2.25), suggesting that the RFA device achieved relatively uniform thermal ablation throughout the tract (Figure 5).

Figure 5 Line plot depicting the ablation-to-granulation tissue ratios within individual fistula tracts, segmented into inner, middle, and outer regions. No consistent pattern of variation was observed, further supporting the uniformity of the radiofrequency ablation effect across the tract.

DISCUSSION

In recent decades, advances in sphincter-sparing surgical techniques have emerged to address perianal fistulas and reduce the likelihood of anal incontinence. Techniques such as rectal advancement flaps[21], ligation of the intersphincteric fistula tract[22], fistula plugs[23], and application of fibrin glue[24] are designed to obliterate the fistula tract while maintaining continence. Minimally invasive methods that thermally ablate the tract wall, including VAAFT[1], FiLaC[3], and RFA[13], have been proposed as alternatives to preserve sphincter function. Despite these innovations, the clinical outcomes associated with RFA have been unsatisfactory. Taken together, these evolving approaches highlight the need to clarify where optimized and uniform RFA dosing may offer incremental benefits within the current treatment landscape. A multicenter prospective trial utilizing a temperature of 120 °C for a duration of 12 seconds reported a fistula healing rate of only 34.7% at both 6 and 12 months[13]. Another study reported that RFA did not confer significant advantages in the treatment of complex fistulas with sphincter-sparing surgery, with a recurrence rate exceeding 70% at 12 months[14]. These results indicate that the parameters established from non-fistulous applications, such as those used in venous ablation, may be insufficient to achieve adequate destruction of the tract wall in perianal fistulas. Our prior research further highlighted that the fistula tract may necessitate deeper and broader ablation than the vein walls, thereby underscoring the importance of optimizing parameters specifically tailored to this pathology[5]. Consequently, the current study sought to experimentally assess a variety of RFA settings, particularly in terms of temperature and duration, using a validated porcine model to identify an optimal setting that balances therapeutic efficacy with tissue safety.

To guide parameter selection prior to in vivo experiments, we performed an ex vivo study utilizing pork loin tissue to assess thermal damage profiles across a wide array of temperature-time combinations. This initial analysis indicated that a 28 seconds application at either 100 °C or 120 °C resulted in excessive thermal diffusion beyond the designated ablation zone, raising significant safety concerns. As a result, extended durations were omitted from subsequent testing. Drawing from the ex vivo results, we identified four refined candidate conditions for in vivo validation: 120 °C/20 seconds, 120 °C/24 s, 100 °C/20 seconds, and 100 °C/24 seconds. Histological evaluation demonstrated that the 120 °C/24 seconds condition achieved the highest ratio of ablation to granulation tissue area, exhibiting statistically significant superiority over the other conditions. Furthermore, this parameter ensured consistent ablation across the inner, middle, and outer segments of the tract, thereby confirming its reliability and uniformity. These results indicate that the 120 °C/24 seconds setting represents an optimal compromise between ablation effectiveness and thermal safety. Further validation through clinical studies is crucial to ascertain whether this parameter can be established as a standardized setting for RFA treatment of perianal fistulas.

In addition to the optimization of the thermal parameters, the design and methodology of energy delivery are crucial for the efficacy of RFA in the treatment of perianal fistulas. Traditional FiLaC techniques and the majority of commercially available RFA devices currently rely on a focal-tip “dot-ablation” profile, which is typically limited to an approximately 5-mm ablation zone per application[3,13]. As a result, ablation is executed in a stepwise manner. FiLaC requires manual retraction of the probe at a rate of 1 cm every 3 seconds[3], whereas earlier-generation RFA probes recommend a 5 mm withdrawal every 12 seconds[13]. Such stepwise “dot-ablation” inherently increases the risk of longitudinal heterogeneity, potentially leading to inadequate treatment of certain areas and excessive treatment of others, which may contribute to the suboptimal outcomes reported in previous studies[13,14]. In contrast, our innovative RFA device was designed with a 50-mm segmental electrode, which increases the ablation coverage per application fourfold compared with conventional 5 mm tips. Given that most perianal fistula tracts measure ≤ 50 mm in length[25,26], this design allows complete tract ablation with a single energy application in most instances. For fistulas > 50 mm, the device incorporates calibrated shaft markings to facilitate precise repositioning, enabling operators to withdraw the catheter at a specified distance and administer a second ablation cycle. This approach ensures comprehensive coverage without overlapping thermal injuries or skipping areas. To substantiate this theoretical advantage, we categorized the fistula tract into inner, middle, and outer segments, and performed histological analyses across all levels. These findings revealed a uniform ablation throughout the tract, corroborating the assertion that our segmental ablation design fosters consistent and effective energy delivery. When combined with the optimized parameters (120 °C/24 seconds), this technical advancement addressed two primary limitations of existing RFA systems: Restricted energy delivery and procedural inconsistency. Collectively, these enhancements present a promising solution for achieving a predictable and complete destruction of the fistula tract in a reproducible manner.

The therapeutic justification for employing endoluminal thermal ablation in the management of perianal fistulas is based on its proven effectiveness in treating other tubular structures. For instance, in the management of varicose veins, both RFA and laser therapy can achieve sustained closure by inducing thermal damage to the venous wall, resulting in collagen denaturation, wall contraction, and subsequent fibrotic obliteration of the lumen[27]. This principle has been similarly applied to bile ducts, whereby endobiliary RFA induces transmural coagulative necrosis and controls ductal collapse in porcine models[28]. Similarly, laser ablation of the pilonidal sinus tract can effectively eradicate the epithelial lining and facilitate tract closure without necessitating extensive excision[29]. Notwithstanding these precedents, direct evidence substantiating the structural shrinkage or complete obliteration of perianal fistula tracts via endoluminal energy delivery remains scarce. Current minimally invasive techniques, such as FiLaC and RFA, have exhibited inconsistent clinical outcomes, and to date, no study has definitively demonstrated histological closure or luminal remodeling in the context of fistulas. Consequently, the hypothesis that the thermal destruction of the tract wall alone is adequate to promote long-term healing remains unverified. Accordingly, we used the ablation-to-granulation tissue area ratio as a mechanistic surrogate parameter reflecting the extent of the targeted wall injury, which is hypothesized to drive subsequent contraction/fibrotic sealing, while acknowledging that a direct correlation with healing outcomes warrants validation in chronic models. In this study, we aimed to fill this knowledge gap using a controlled experimental model. Among the various parameters evaluated, an RFA setting of 120 °C/24 seconds yielded the most favorable equilibrium between thermal penetration and ablation coverage. This specific setting resulted in the highest ratio of ablation-to-granulation tissue while ensuring consistency throughout the tract, indicating that a precisely calibrated thermal ablation may substantiate the underlying hypothesis regarding tract obliteration. Of note, our study did not assess post-treatment tract closure or long-term healing, as tissue analysis was conducted immediately after RFA. Accordingly, the clinical efficacy of the ablation parameters remains hypothetical based on acute histology alone and requires validation by delayed histology and longitudinal assessments of tract closure in long-term follow-up models. Nevertheless, the observed ablation patterns provided preliminary mechanistic insights into the degree of thermal injury, which may be necessary for achieving clinical success. Future research should include delayed histological assessments and radiological follow-ups, which are crucial to determine whether this level of ablation leads to sustained fistula closure.

This study had several limitations that merit careful consideration. First, although a wider array of RFA parameter settings was initially proposed, the investigation focused on only four specific temperature-duration combinations. This restricted approach was adopted to maintain experimental consistency and to adhere to animal welfare standards; however, it raises the possibility that other potentially effective configurations were not examined. Secondly, the overall sample size was limited to four porcine subjects. Additionally, unintentional displacement of the seton in two animals resulted in a decrease in the number of analyzable fistula tracts from twelve to nine. To address this limitation, each tract was further divided into inner, middle, and outer segments, resulting in 27 histological specimens. This segmentation facilitated intratract comparisons and partially compensated for the reduction in sample size at the tract level. However, segmentation increases the number of histologic sections without increasing the number of independent units. Therefore, despite the repeated-measures approach, the limited tract-level sample size may reduce statistical power and constrain generalizability. Third, although we posited that a more extensive ablation of granulation tissue would enhance tract shrinkage and promote healing, this relationship has not been empirically substantiated. Although this hypothesis is consistent with established surgical principles such as those underpinning fistulectomy and fistulotomy, the degree to which granulation tissue ablation alone leads to long-term closure of the fistula remains uncertain. Fourth, the histological assessments in this study were conducted immediately following RFA, precluding the evaluation of post-treatment healing, tissue remodeling, or tract closure. Although these short-term findings provide mechanistic insights, they do not directly confirm the clinical efficacy or durability of treatment. Finally, the translational relevance of our findings is limited by the use of an acute, non-chronic porcine fistula model. In contrast to human perianal fistulas, which are typically chronic, fibrotic, and often epithelialized, the experimental tracts in this study were recent and predominantly composed of granulation tissue. Such disparities may influence both the thermal response and the translation to clinical practice. The absolute r_e values observed in this acute porcine model may underestimate the injury dimension required for chronically fibrotic human fistulas, and extrapolation should be made with caution. Future investigations utilizing chronic animal models and incorporating long-term follow-ups are crucial to validate the proposed RFA settings within a clinically pertinent framework.

CONCLUSION

In the evaluation of various RFA settings, a temperature of 120 °C maintained for 24 seconds was found to be the most effective energy setting for ablating perianal fistula tract tissue in a porcine model. This setting achieved the highest ratio of the ablation-to-granulation tissue area, accompanied by a consistent thermal distribution. These results indicate that this setting warrants further investigation. Subsequent animal studies and clinical trials are necessary to assess its safety and long-term effectiveness with the aim of establishing the proposed method as a standardized protocol for RFA-based treatment of fistulas.

ACKNOWLEDGEMENTS

During the preparation of this manuscript, the authors used ChatGPT-4o (OpenAI, April 2024 version) for the purposes of English language editing and clarity enhancement. The authors have thoroughly reviewed and revised the AI-generated output and take full responsibility for the content of this publication.