Ventricular pulsed field ablation from bench to bedside: preclinical and clinical evidence from a systematic review and meta-analysis Authors: Paolo Compagnucci, Luca Finori, Michela Casella, Alessio Gasperetti, Henry D Huang, Yari Valeri, Leonardo D’Angelo, Francesca Campanelli, Giorgio Giacomini, Giovanni Volpato, Laura Cipolletta, Quintino Parisi, Luigi Di Biase, Andrea Natale, Pasquale Santangeli, Antonio Dello Russo Published: 2026 ABSTRACT AbstractAimsEvidence supporting the use of pulsed field ablation (PFA) for ventricular arrhythmia (VA) ablation remains limited and heterogeneous, and the relationship between preclinical lesion characteristics and clinical outcomes is incompletely defined. We sought to systematically summarize preclinical lesion metrics and clinical outcomes of ventricular PFA.Methods and resultsWe performed a systematic review and meta-analysis of preclinical and clinical studies of ventricular PFA. Preclinical studies reporting lesion depth or transmurality and clinical studies detailing procedural efficacy and safety were included. Random-effects models were used to pool lesion characteristics and clinical outcomes. Seventy-five studies were included (33 preclinical, 43 clinical; 355 patients). In preclinical models, mean lesion depth was 5.96 mm(95%CI, 5.07–6.84) in healthy endocardium and 5.19 mm (95%CI, 2.58–7.79) in healthy epicardium. Notably, lesion depth was quantitatively similar in models of scarred myocardium (5.96 mm; 95%CI, 5.39–6.53). Lesion transmurality was variable in healthy tissues (32%; 95%CI, 15–52%) but higher within scar (53%; 95%CI, 36–70%). Consistently, lesion depth was similar between PFA and RF in healthy myocardium, but greater with PFA in scarred models (within-study mean depth difference, 2.09 mm; 95%CI, 0.84–3.34). Greater contact force, repeated applications, and higher energy/duration settings were associated with increased lesion depth. Clinically, acute success was high for both premature ventricular complexes (PVCs,90%; 95%CI, 83–96%) and ventricular tachycardia (VT;92%; 95%CI, 87–96%) in studies with ≥3patients. Over a mean follow-up of approximately 5months, recurrence rates were 19% (95%CI, 12–28%) for PVCs and 21% (95%CI, 10–35%) for VT. The overall complication rate was 10% (95%CI, 5–17%) in studies with ≥3patients, with coronary artery spasm and conduction disturbances representing the most frequent adverse events.ConclusionVentricular PFA demonstrates high acute efficacy and encouraging mid-term outcomes, supported by preclinical evidence of favourable lesion penetration within scarred myocardium. Safety considerations-particularly coronary and conduction system effects-and the heterogeneity of preclinical evidence highlight the need for further investigation.Keywords: Pulsed field ablation, Ventricular tachycardia, Premature ventricular complexes, Catheter ablation, Meta-analysis, Safety, Recurrence, Electroporation Introduction Catheter ablation is increasingly recommended to improve symptoms and reduce implantable cardioverter-defibrillator shocks and arrhythmia recurrence in patients with ventricular arrhythmias (VAs).1–3 Conventional catheter ablation relies on radiofrequency energy (RF), with effective lesion formation critically depending on adequate catheter-tissue contact, appropriate power and irrigation settings, and tissue impedance.2,4,5 In thick ventricular myocardial regions and in substrates characterized by extensive scarring or intramural location, achieving durable and deep lesions remains challenging.2,5,6 These limitations may translate into an increased risk of post-ablation arrhythmia recurrence, which persists even in the contemporary era.5,7 Pulsed field ablation (PFA) is an emerging energy source characterized by a minimally thermal profile and relative cardiomyocyte selectivity, potentially expanding safety margins near non-target tissues and mitigating heat-related complications, such as steam pops.8–10 Rapid adoption of PFA in atrial fibrillation ablation has stimulated interest in ventricular applications, particularly given emerging data suggesting greater lesion depth compared to thermal energy sources.11,12 However, despite the potential utility of PFA for the treatment of VAs, its real-world use remains limited and at an early stage of adoption, being largely confined to expert centres, as highlighted by a 2024 survey in which PFA was reported to be rarely used for ventricular tachycardia (VT) ablation, accounting for only 1.7% of respondents.13 Furthermore, the current evidence base for ventricular PFA remains fragmented, consisting largely of first-in-human experiences, small case series, and registries employing heterogeneous systems and energy delivery protocols.11 Moreover, the relationship between preclinical lesion characteristics, particularly lesion depth and transmurality, and clinical outcomes also remains incompletely defined. In this systematic review and meta-analysis, we aimed to summarize the available evidence from preclinical studies of PFA in ventricular settings and to synthesize clinical data on acute safety, procedural effectiveness, and mid-term arrhythmia recurrence following ventricular PFA. Methods Study design and registration We performed a systematic review and meta-analysis of preclinical and clinical studies evaluating ventricular PFA. Study protocols were registered as follows: the clinical component in PROSPERO (ID: CRD420251138923) and the preclinical component in the Open Science Framework (DOI: 10.17605/OSF.IO/TZ28H). The present manuscript was reported in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement.14 Eligibility criteria Preclinical studies included animal models of ventricular PFA reporting lesion depth or transmurality assessed by pathology. In the absence of randomized controlled trials evaluating ventricular PFA, eligible clinical studies included non-randomized comparative studies, single-arm registries, and case reports or series detailing at least one outcome of interest in patients with premature ventricular complexes (PVCs) and/or VT undergoing ventricular PFA. Conference abstracts without full data were excluded. Search strategy and study selection MEDLINE/PubMed, Scopus, medRxiv, and ClinicalTrials.gov were searched from inception to 10 October 2025. The full search strategy is provided in the Supplementary Methods. Reference lists of eligible studies were manually screened. Duplicate records were removed using Rayyan and Zotero. Two reviewers (L.F. and P.C.) independently screened titles and abstracts and subsequently assessed full texts for eligibility; disagreements were resolved by consensus or third-party adjudication (M.C.). The study selection process is summarized in the PRISMA flow diagram (Figure 1). PRISMA flow diagram. *One study reported both preclinical and clinical data. Data extraction For preclinical studies, extracted variables included animal species, ventricular chamber, procedural approach (endocardial or epicardial), dosing parameters, lesion depth, and transmurality. When studies reported medians with interquartile ranges or ranges, means and standard deviations were derived using established methods (Luo et al.15; Wan et al.16) For pooling of lesion depth and transmurality, when multiple contact force settings were reported, we prioritized adequate-contact conditions, defined as contact force ≥10 g, in accordance with the HRS expert consensus on catheter ablation of VAs.2 The effects of contact, number of applications, and amplitude/duration settings were further explored using within study comparisons between high and low settings, when feasible. For clinical studies, extracted data included study design, sample size, patient demographics, arrhythmia type, procedural approach, ablation system, acute procedural success, intraprocedural or ≤30-day adverse events, follow-up duration, and arrhythmia recurrence. When available, individual patient-level data were averaged; otherwise, aggregated data were extracted. Outcomes For the preclinical component, the primary outcomes were lesion depth and transmurality assessed on pathology. For the clinical component, primary outcomes were: (i) acute success (PVC elimination and any-clinical and non-clinical-VT non-inducibility), (ii) procedure-related complications occurring intra-procedurally or within 30 days, and (iii) arrhythmia recurrence during follow-up. Risk of bias assessment Clinical studies were appraised using Joanna Briggs Institute (JBI) critical appraisal tools (Prevalence, Case Series, and Case Reports). Preclinical studies were assessed using the SYRCLE risk-of-bias tool and ARRIVE 2.0 guidelines. Formal assessment of publication bias (e.g. funnel plots or Egger's test) was not performed. For meta-analyses of proportions, these methods have recognized methodological limitations because they were developed primarily for comparative effect measures.17,18 In addition, the small number of studies available for the comparative analyses precluded meaningful assessment of publication bias.17,18 Statistical analysis Continuous variables (lesion depth) were pooled using inverse variance weighted random-effects models with restricted maximum likelihood (REML) estimation. As a sensitivity analysis, lesion depth was additionally analysed on the logarithmic scale to assess the robustness of the findings and the impact of the skewed distribution of a strictly positive variable. Proportions were pooled using inverse-variance weighted random-effects models after variance stabilization with the Freeman–Tukey double arcsine transformation. Pooled estimates were back-transformed for presentation. Between-study heterogeneity was quantified using the I2 statistic. Pre-specified subgroup analyses were conducted for both preclinical and clinical datasets. Given the early-stage nature of ventricular PFA, all available clinical evidence was included in the main analysis; however, sensitivity analyses restricted to studies with ≥3 patients were performed to enhance robustness. Additionally, leave-one-out analyses were performed in the presence of substantial heterogeneity (I2 > 50%). Within-study mean differences in lesion depth were calculated for: (i) high vs. low catheter–tissue contact, (ii) high vs. low number of applications, (iii) high vs. low energy settings, and (iv) PFA vs. RF, when both modalities were compared within the same study. Mean differences were defined as high minus low for parameter-based comparisons and as PFA minus RF for energy source comparisons. Study-specific mean differences and corresponding standard errors were calculated and pooled using random-effects models. For transmurality, proportions were extracted for each comparison group, and pooled analyses were performed using random-effects models. Between-group differences were summarized as pooled risk differences with 95% confidence intervals. All tests were two-sided, with statistical significance defined as P < 0.05. Analyses were performed using the meta package in R (R Foundation for Statistical Computing, Vienna, Austria). Results Study selection and characteristics The search yielded 551 records; after deduplication, 388 unique titles/abstracts were screened, and 92 full texts were assessed. Seventy-five studies met criteria, 33 preclinical19–51 and 43 clinical;52–69,69–92 one report included both domains (Figure 1; Tables 1 and 2; Supplementary material online, Tables S1–S3).48 The principal ablation strategies and the relative translational outlooks are described in Table 1 (tools with published clinical evidence) and Supplementary material online, Table S2 (tools without published clinical evidence at the time of literature search). Table 1Translational and descriptive summary of the strategies and devices with published clinical evidence on ventricular pulsed field ablation at the time of literature searchPFA-based strategy or technologyTechnological conceptRationalePotential advantagesPotential limitationsMain preclinical findingsClinical evidenceCentauri PFA generator coupled with irrigated catheters,4,27,29,30,37,54,58,63–65,73–75,81,92 CardioFocus Generator coupled with 8-Fr, uni−/bidirectional, irrigated catheters (ThermoCool SmartTouch, Tacticath SE, or StablePoint)Monopolar PFA modeBiphasic, microsecond pulsesApplication duration varies with amplitude: 3 s for 19 A, 5 s for 22 A, 7 s for 25 APulse amplitude/duration 19–25A/µs Open platform to be used with available electroanatomical mapping systems; it can be used with focal contact-force sensing, irrigated RF catheters Compatible with standard sheathsIts use with focal catheters facilitates ventricular, epicardial and coronary sinus navigationCE marked Focal tip design requires multiple applications to cover large substratesReported instances of transient conduction system blocks attributed to current leakage from the proximal electrode when used with the ThermoCool SmartTouch catheter in the retrograde aortic approach to the left ventricular endocardiumNeed to manually place ablation tags on the mapping system With 8 packets of 25 A pulses, PFA can penetrate within and beyond scarLesion depth is smaller in scar compared to healthy myocardium (5.3 vs. 7.2 mm)27Dose-dependent increase in lesion depth with amplitude settings (5.7 mm with 19 A, 7.2 mm with 22 A, 8.2 mm with 25 A)29Dose-dependent increase in lesion depth with contact force37PFA lesions are characterized by contraction band necrosis and cardiomyocytolysis, with less coagulative necrosis compared to thermal ablation Large clinical experience (n = 107)The sole system used for PFA applications within the coronary venous system (n = 19); high impedance did not preclude PFA delivery in any caseHigh acute success rate in both PVC and VT/VF cohortsSubstantial risk of mid-term recurrence (meta-analytic rate of 48% in both PVC and VT/VF cohorts) Farapoint,26,40,89 Boston Scientific 8-Fr bidirectional, non-irrigated focal catheter4 electrodesBipolar PFA modeBiphasic wave2.5 s duration per applicationAmplitude/duration 1.4/2.0 kV/µs Focal catheter with electroanatomical integration; PFA delivery across all 4 electrodes; no direct contact force sensing capability (local impedance-based) Compatible with a standard sheathThe focal design facilitates ventricular and epicardial navigationEnergy delivery across all 4 electrodes can facilitate deployment of linear lesions when the catheter is parallel to tissue (e.g. in the septum)Lack of irrigation can reduce risk of fluid overloadCE marked and FDA approved for atrial flutter Very limited information on clinical performanceFocal tip design requires multiple applications to cover large substrates although linear ablation-with the catheter parallel to tissue-can be useful to connect scar regions to anatomic boundariesNo dual-energy capability With 3 applications/location, successful lesion penetration of scar was observed, with similar lesion depth at the infarct border (5.9 mm) and in healthy myocardium (5.7 mm)26With 4 applications/location, similar lesion depth over chronic RF lesions (7.1 mm) and in healthy myocardium (7.0 mm)40Long linear lesions where the catheter is parallel to tissueWell demarcated lesions characterized by a haemorrhagic central core Published case series of 2 patients undergoing papillary muscle PVC ablationFarapulse,22,52,53,55–57,60,62,67–69,72,77–80,82,84,87,88 Boston Scientific 12-Fr over-the-wire catheter31/35 mm tipBipolar PFA modeBiphasic, micro- to sub-millisecond pulses2.5 s duration per applicationPulse amplitude/duration 1.8–2.0 kV/µs Multielectrode single-shot, regional system primarily designed for pulmonary vein isolation; electroanatomical integration with the Faraview catheter Regional catheter that allows to rapidly cover large substratesCE marked and FDA approved Catheter size and over-the-wire design require special care for use in the ventricle and epicardial application was only reported with surgical approachOptimal ventricular waveform and dosing less well established than for atrial applications With 4 stacked lesions: 6 mm lesion depth in healthy myocardium; 4.8 mm lesion depth in scarred myocardiumIn myocardial infarct, histologically deeper lesions compared to radiofrequency ablationPFA spares arterioles, veins, and nerves 48 patients treated for ventricular arrhythmiasExcellent safety profile, with only 2 complications reportedPreliminary acute (PVC, n = 15; VT/VF, n = 27) and follow-up (PVC, n = 15; VT/VF, n = 30) efficacy information available FieldForce,48 Field Medical 8.5 Fr bidirectional, irrigated focal catheter3 electrodes;Monophasic, nano- to micro-second pulsesEach lesion consists of 5 applications gated to the QRS, each <200 ms at >10 kV amplitude High-energy, focal catheter optimized to maximize lesion depth; the device is compatible with available electroanatomical mapping systems but not fully integrated; it has contact force-sensing capability Compatible with standard sheathsThe focal design facilitates ventricular and epicardial navigationWaveform optimized to produce ventricular transmural lesions, which may address deep/epicardial substrates with an endocardial-only approach The device is not integrated with 3D mapping systems, requiring manual annotation of ablation pointsReported instance of cardiac resynchronization therapy-defibrillator interferenceNo dual-energy capabilityNot approved in Europe or the United States Lesion depth of 12.5 mm in healthy myocardium with 5 applications/lesionHomogeneous lesions with well demarcated margins and preserved arterioles Only device evaluated in a small (n = 26) prospective non-randomized trialFavourable efficacy outcomes in a complex population enriched in redo cases3 safety events (11.5%)Direct clinical evidence of transmural tissue homogenization with endocardial-only PFA PulsedFA FocalPoint catheter combined with the Lead-Mapping system,90 Jinjiang Electronic Science and Technology Co. 8-Fr, focal unidirectional catheter4 electrodes (1 head and 3 ring)Bipolar PFA modeBiphasic millisecond pulses288 or 480 μs durationPulse amplitude 0.5–2.0 kV (generally 800–1.000–1.200–1.800 V) Focal, dual-energy catheter coupled with electrogram recording capability and electroanatomical integration; has contact-force sensing capability Can be introduced into a standard sheathThe focal design facilitates ventricular navigationThe dual energy capability enhances device versatility and allows RF/PFA stacking Focal tip design requires multiple applications to cover large substratesNo published data on VT/VF ablation proceduresOptimal ventricular waveform and dosing not well establishedNot approved in Europe or the United States No published preclinical data in ventricular models Limited clinical data (12 patients)Only PVCsFavourable acute (10/12) and 6-month follow-up (9/12) PVC elimination1 complication reported—a transient complete atrioventricular block with para-Hisian PFA application Sphere-9,20,23,25,39,59,61,66,70,71,76,83,85,86,91Medtronic 8-Fr bidirectional, irrigated catheter9 mm nitinol/lattice tip with 9 minielectrodes, 1 noncontact indifferent electrode within the sphere, and 2 distal shaft ring electrodesMonopolar PFA modeBiphasic, microsecond wave4 s application durationAmplitude/duration 2.0 kV/µs Large-tip focal, dual-energy catheter coupled with high-quality electrogram recording capability (9 mini-electrodes) and full electroanatomical integration; flexible lattice tip; impedance-based contact assessment tool Compatible with a standard 8.5 sheathAllows high-quality mapping and zero-exchange workflowsAllows rapid ablation of large substrates (large-tip focal design)The dual energy capability enhances device versatility and allows RF/PFA stackingCE marked and FDA approved The large tip design carries risk of mitral/tricuspid valve entrapment with need to recapture the device within the long sheath (problematic for retrograde aortic approaches)The spheric tip can entrap epicardial fat during epicardial procedures—potential need to replace the catheter before remapping the endocardiumReported instances of defibrillator interference/failureHRS safety alert regarding RF delivery with Sphere-9 in Biotronik defibrillator carriers (doi: https://www.hrsonline.org/resource/device-interactions-medtronic-biotronik/) Lesion depth increases with the number of applications (from 5.6 mm with 1 application to 8.8 mm20 or 6.1 mm23 with 4 applications)Similar penetration in healthy myocardium (4.8 mm) and healed scar (5.6 mm) with 4 applicationsPFA with Sphere-9 (4 applications) is associated with greater lesion depth within heterogeneous scar compared to RF ablation with ThermoCool ST-SF (6.4 mm vs. 5.4 mm)Successful ablation of intracavitary structures (such as papillary muscles and moderator band) Largest published clinical experience (n = 160 overall), including the largest observational study (AVAAR, n = 126)Frequent hybrid PFA/RF workflowsHigh acute success rate in both PVC and VT/VF proceduresLowest meta-analytic rate of mid-term VT/VF recurrence (1%)14% meta-analytic rate of any procedure-related adverse events, most commonly ST-segment changes FDA, Food and Drug Administration; PFA, pulsed field ablation; PVC, premature ventricular complex; RF, radiofrequency ablation; VF, ventricular fibrillation; VT, ventricular tachycardia Translational and descriptive summary of the strategies and devices with published clinical evidence on ventricular pulsed field ablation at the time of literature search Generator coupled with 8-Fr, uni−/bidirectional, irrigated catheters (ThermoCool SmartTouch, Tacticath SE, or StablePoint) Generator coupled with 8-Fr, uni−/bidirectional, irrigated catheters (ThermoCool SmartTouch, Tacticath SE, or StablePoint) Monopolar PFA mode Monopolar PFA mode Biphasic, microsecond pulses Biphasic, microsecond pulses Application duration varies with amplitude: 3 s for 19 A, 5 s for 22 A, 7 s for 25 A Application duration varies with amplitude: 3 s for 19 A, 5 s for 22 A, 7 s for 25 A Pulse amplitude/duration 19–25A/µs Pulse amplitude/duration 19–25A/µs Compatible with standard sheaths Compatible with standard sheaths Its use with focal catheters facilitates ventricular, epicardial and coronary sinus navigation Its use with focal catheters facilitates ventricular, epicardial and coronary sinus navigation CE marked CE marked Focal tip design requires multiple applications to cover large substrates Focal tip design requires multiple applications to cover large substrates Reported instances of transient conduction system blocks attributed to current leakage from the proximal electrode when used with the ThermoCool SmartTouch catheter in the retrograde aortic approach to the left ventricular endocardium Reported instances of transient conduction system blocks attributed to current leakage from the proximal electrode when used with the ThermoCool SmartTouch catheter in the retrograde aortic approach to the left ventricular endocardium Need to manually place ablation tags on the mapping system Need to manually place ablation tags on the mapping system With 8 packets of 25 A pulses, PFA can penetrate within and beyond scar With 8 packets of 25 A pulses, PFA can penetrate within and beyond scar Lesion depth is smaller in scar compared to healthy myocardium (5.3 vs. 7.2 mm)27 Lesion depth is smaller in scar compared to healthy myocardium (5.3 vs. 7.2 mm)27 Dose-dependent increase in lesion depth with amplitude settings (5.7 mm with 19 A, 7.2 mm with 22 A, 8.2 mm with 25 A)29 Dose-dependent increase in lesion depth with amplitude settings (5.7 mm with 19 A, 7.2 mm with 22 A, 8.2 mm with 25 A)29 Dose-dependent increase in lesion depth with contact force37 Dose-dependent increase in lesion depth with contact force37 PFA lesions are characterized by contraction band necrosis and cardiomyocytolysis, with less coagulative necrosis compared to thermal ablation PFA lesions are characterized by contraction band necrosis and cardiomyocytolysis, with less coagulative necrosis compared to thermal ablation Large clinical experience (n = 107) Large clinical experience (n = 107) The sole system used for PFA applications within the coronary venous system (n = 19); high impedance did not preclude PFA delivery in any case The sole system used for PFA applications within the coronary venous system (n = 19); high impedance did not preclude PFA delivery in any case High acute success rate in both PVC and VT/VF cohorts High acute success rate in both PVC and VT/VF cohorts Substantial risk of mid-term recurrence (meta-analytic rate of 48% in both PVC and VT/VF cohorts) Substantial risk of mid-term recurrence (meta-analytic rate of 48% in both PVC and VT/VF cohorts) 8-Fr bidirectional, non-irrigated focal catheter 8-Fr bidirectional, non-irrigated focal catheter 4 electrodes 4 electrodes Bipolar PFA mode Bipolar PFA mode Biphasic wave Biphasic wave 2.5 s duration per application 2.5 s duration per application Amplitude/duration 1.4/2.0 kV/µs Amplitude/duration 1.4/2.0 kV/µs Compatible with a standard sheath Compatible with a standard sheath The focal design facilitates ventricular and epicardial navigation The focal design facilitates ventricular and epicardial navigation Energy delivery across all 4 electrodes can facilitate deployment of linear lesions when the catheter is parallel to tissue (e.g. in the septum) Energy delivery across all 4 electrodes can facilitate deployment of linear lesions when the catheter is parallel to tissue (e.g. in the septum) Lack of irrigation can reduce risk of fluid overload Lack of irrigation can reduce risk of fluid overload CE marked and FDA approved for atrial flutter CE marked and FDA approved for atrial flutter Very limited information on clinical performance Very limited information on clinical performance Focal tip design requires multiple applications to cover large substrates although linear ablation-with the catheter parallel to tissue-can be useful to connect scar regions to anatomic boundaries Focal tip design requires multiple applications to cover large substrates although linear ablation-with the catheter parallel to tissue-can be useful to connect scar regions to anatomic boundaries No dual-energy capability No dual-energy capability With 3 applications/location, successful lesion penetration of scar was observed, with similar lesion depth at the infarct border (5.9 mm) and in healthy myocardium (5.7 mm)26 With 3 applications/location, successful lesion penetration of scar was observed, with similar lesion depth at the infarct border (5.9 mm) and in healthy myocardium (5.7 mm)26 With 4 applications/location, similar lesion depth over chronic RF lesions (7.1 mm) and in healthy myocardium (7.0 mm)40 With 4 applications/location, similar lesion depth over chronic RF lesions (7.1 mm) and in healthy myocardium (7.0 mm)40 Long linear lesions where the catheter is parallel to tissue Long linear lesions where the catheter is parallel to tissue Well demarcated lesions characterized by a haemorrhagic central core Well demarcated lesions characterized by a haemorrhagic central core 12-Fr over-the-wire catheter 12-Fr over-the-wire catheter 31/35 mm tip 31/35 mm tip Bipolar PFA mode Bipolar PFA mode Biphasic, micro- to sub-millisecond pulses Biphasic, micro- to sub-millisecond pulses 2.5 s duration per application 2.5 s duration per application Pulse amplitude/duration 1.8–2.0 kV/µs Pulse amplitude/duration 1.8–2.0 kV/µs Regional catheter that allows to rapidly cover large substrates Regional catheter that allows to rapidly cover large substrates CE marked and FDA approved CE marked and FDA approved Catheter size and over-the-wire design require special care for use in the ventricle and epicardial application was only reported with surgical approach Catheter size and over-the-wire design require special care for use in the ventricle and epicardial application was only reported with surgical approach Optimal ventricular waveform and dosing less well established than for atrial applications Optimal ventricular waveform and dosing less well established than for atrial applications With 4 stacked lesions: 6 mm lesion depth in healthy myocardium; 4.8 mm lesion depth in scarred myocardium With 4 stacked lesions: 6 mm lesion depth in healthy myocardium; 4.8 mm lesion depth in scarred myocardium In myocardial infarct, histologically deeper lesions compared to radiofrequency ablation In myocardial infarct, histologically deeper lesions compared to radiofrequency ablation PFA spares arterioles, veins, and nerves PFA spares arterioles, veins, and nerves 48 patients treated for ventricular arrhythmias 48 patients treated for ventricular arrhythmias Excellent safety profile, with only 2 complications reported Excellent safety profile, with only 2 complications reported Preliminary acute (PVC, n = 15; VT/VF, n = 27) and follow-up (PVC, n = 15; VT/VF, n = 30) efficacy information available Preliminary acute (PVC, n = 15; VT/VF, n = 27) and follow-up (PVC, n = 15; VT/VF, n = 30) efficacy information available 8.5 Fr bidirectional, irrigated focal catheter 8.5 Fr bidirectional, irrigated focal catheter 3 electrodes; 3 electrodes; Monophasic, nano- to micro-second pulses Monophasic, nano- to micro-second pulses Each lesion consists of 5 applications gated to the QRS, each <200 ms at >10 kV amplitude Each lesion consists of 5 applications gated to the QRS, each <200 ms at >10 kV amplitude Compatible with standard sheaths Compatible with standard sheaths The focal design facilitates ventricular and epicardial navigation The focal design facilitates ventricular and epicardial navigation Waveform optimized to produce ventricular transmural lesions, which may address deep/epicardial substrates with an endocardial-only approach Waveform optimized to produce ventricular transmural lesions, which may address deep/epicardial substrates with an endocardial-only approach The device is not integrated with 3D mapping systems, requiring manual annotation of ablation points The device is not integrated with 3D mapping systems, requiring manual annotation of ablation points Reported instance of cardiac resynchronization therapy-defibrillator interference Reported instance of cardiac resynchronization therapy-defibrillator interference No dual-energy capability No dual-energy capability Not approved in Europe or the United States Not approved in Europe or the United States Lesion depth of 12.5 mm in healthy myocardium with 5 applications/lesion Lesion depth of 12.5 mm in healthy myocardium with 5 applications/lesion Homogeneous lesions with well demarcated margins and preserved arterioles Homogeneous lesions with well demarcated margins and preserved arterioles Only device evaluated in a small (n = 26) prospective non-randomized trial Only device evaluated in a small (n = 26) prospective non-randomized trial Favourable efficacy outcomes in a complex population enriched in redo cases Favourable efficacy outcomes in a complex population enriched in redo cases 3 safety events (11.5%) 3 safety events (11.5%) Direct clinical evidence of transmural tissue homogenization with endocardial-only PFA Direct clinical evidence of transmural tissue homogenization with endocardial-only PFA 8-Fr, focal unidirectional catheter 8-Fr, focal unidirectional catheter 4 electrodes (1 head and 3 ring) 4 electrodes (1 head and 3 ring) Bipolar PFA mode Bipolar PFA mode Biphasic millisecond pulses Biphasic millisecond pulses 288 or 480 μs duration 288 or 480 μs duration Pulse amplitude 0.5–2.0 kV (generally 800–1.000–1.200–1.800 V) Pulse amplitude 0.5–2.0 kV (generally 800–1.000–1.200–1.800 V) Can be introduced into a standard sheath Can be introduced into a standard sheath The focal design facilitates ventricular navigation The focal design facilitates ventricular navigation The dual energy capability enhances device versatility and allows RF/PFA stacking The dual energy capability enhances device versatility and allows RF/PFA stacking Focal tip design requires multiple applications to cover large substrates Focal tip design requires multiple applications to cover large substrates No published data on VT/VF ablation procedures No published data on VT/VF ablation procedures Optimal ventricular waveform and dosing not well established Optimal ventricular waveform and dosing not well established Not approved in Europe or the United States Not approved in Europe or the United States Limited clinical data (12 patients) Limited clinical data (12 patients) Only PVCs Only PVCs Favourable acute (10/12) and 6-month follow-up (9/12) PVC elimination Favourable acute (10/12) and 6-month follow-up (9/12) PVC elimination 1 complication reported—a transient complete atrioventricular block with para-Hisian PFA application 1 complication reported—a transient complete atrioventricular block with para-Hisian PFA application 8-Fr bidirectional, irrigated catheter 8-Fr bidirectional, irrigated catheter 9 mm nitinol/lattice tip with 9 minielectrodes, 1 noncontact indifferent electrode within the sphere, and 2 distal shaft ring electrodes 9 mm nitinol/lattice tip with 9 minielectrodes, 1 noncontact indifferent electrode within the sphere, and 2 distal shaft ring electrodes Monopolar PFA mode Monopolar PFA mode Biphasic, microsecond wave Biphasic, microsecond wave 4 s application duration 4 s application duration Amplitude/duration 2.0 kV/µs Amplitude/duration 2.0 kV/µs Compatible with a standard 8.5 sheath Compatible with a standard 8.5 sheath Allows high-quality mapping and zero-exchange workflows Allows high-quality mapping and zero-exchange workflows Allows rapid ablation of large substrates (large-tip focal design) Allows rapid ablation of large substrates (large-tip focal design) The dual energy capability enhances device versatility and allows RF/PFA stacking The dual energy capability enhances device versatility and allows RF/PFA stacking CE marked and FDA approved CE marked and FDA approved The large tip design carries risk of mitral/tricuspid valve entrapment with need to recapture the device within the long sheath (problematic for retrograde aortic approaches) The large tip design carries risk of mitral/tricuspid valve entrapment with need to recapture the device within the long sheath (problematic for retrograde aortic approaches) The spheric tip can entrap epicardial fat during epicardial procedures—potential need to replace the catheter before remapping the endocardium The spheric tip can entrap epicardial fat during epicardial procedures—potential need to replace the catheter before remapping the endocardium Reported instances of defibrillator interference/failure Reported instances of defibrillator interference/failure HRS safety alert regarding RF delivery with Sphere-9 in Biotronik defibrillator carriers (doi: https://www.hrsonline.org/resource/device-interactions-medtronic-biotronik/) HRS safety alert regarding RF delivery with Sphere-9 in Biotronik defibrillator carriers (doi: https://www.hrsonline.org/resource/device-interactions-medtronic-biotronik/) Lesion depth increases with the number of applications (from 5.6 mm with 1 application to 8.8 mm20 or 6.1 mm23 with 4 applications) Lesion depth increases with the number of applications (from 5.6 mm with 1 application to 8.8 mm20 or 6.1 mm23 with 4 applications) Similar penetration in healthy myocardium (4.8 mm) and healed scar (5.6 mm) with 4 applications Similar penetration in healthy myocardium (4.8 mm) and healed scar (5.6 mm) with 4 applications PFA with Sphere-9 (4 applications) is associated with greater lesion depth within heterogeneous scar compared to RF ablation with ThermoCool ST-SF (6.4 mm vs. 5.4 mm) PFA with Sphere-9 (4 applications) is associated with greater lesion depth within heterogeneous scar compared to RF ablation with ThermoCool ST-SF (6.4 mm vs. 5.4 mm) Successful ablation of intracavitary structures (such as papillary muscles and moderator band) Successful ablation of intracavitary structures (such as papillary muscles and moderator band) Largest published clinical experience (n = 160 overall), including the largest observational study (AVAAR, n = 126) Largest published clinical experience (n = 160 overall), including the largest observational study (AVAAR, n = 126) Frequent hybrid PFA/RF workflows Frequent hybrid PFA/RF workflows High acute success rate in both PVC and VT/VF procedures High acute success rate in both PVC and VT/VF procedures Lowest meta-analytic rate of mid-term VT/VF recurrence (1%) Lowest meta-analytic rate of mid-term VT/VF recurrence (1%) 14% meta-analytic rate of any procedure-related adverse events, most commonly ST-segment changes 14% meta-analytic rate of any procedure-related adverse events, most commonly ST-segment changes FDA, Food and Drug Administration; PFA, pulsed field ablation; PVC, premature ventricular complex; RF, radiofrequency ablation; VF, ventricular fibrillation; VT, ventricular tachycardia Table 2Summary of clinical studies with ≥3 patientsAuthor (year)Patients (n)ArrhythmiaAblation locationCatheterPulse amplitudePeak current N of PFA applicationsLozano-Granero et al.57 (2023)3VTCase 1: midapical lateral LV; Case 2: LV inferolateral apical aneurysm; Case 3: basal inferolateral RV free wallFarawave2.0 kVNR13Pannone et al.70 (2024)4VT/PVCMixed VT/PVC substrates: RVOT, peri-tricuspid RV scar, LV septal/inferior/lateral mid-basal scar, basal septal/peri-aortic LVOT/lateral basal scarSphere-9 lattice-tipNRNR45 ± 29.6Peichl et al.71 (2024)18VTAnterior/anteroseptal LV wall 28%; inferior LV wall 28%; lateral LV wall 22%; RV 17%; periaortic region 6%Sphere-9 lattice-tipNRNR8 ± 9Peichl et al.92 (2024) [Cohort 44]44PVC/VTMixed ventricular sites: PVC group from LVOT (52%), RVOT (33%), posteromedial LV papillary muscle (10%), posterobasal LV (5%); VT group from lateral LV (35%), LVOT/great cardiac vein (26%), anterior LV (13%), lateral RV (9%), inferior LV (9%), LV papillary muscle (4%), RVOT (4%)ThermoCool SmartTouch + CENTAURINR25 A16 ± 15Della Rocca et al.75 (2024)20PVCMixed ventricular sites: RVOT (n = 8), LVOT (n = 4), posteromedial papillary muscle (n = 2), moderator band (n = 2), tricuspid annulus (n = 1), right posterior papillary muscle (n = 1), left posterior fascicle (n = 1), aortomitral continuity (n = 1)TactiCath Contact Force or Thermocool SmartTouch + CENTAURINR22 A (55%)/25 A (45%)8 (range 1–20)De Silva et al.80 (2025)3PVC/VFRV moderator band; RV anterior papillary muscle/moderator band complex; LV posteromedial papillary muscleFarawave2.0 kVNR22 ± 12Ruwald et al.81 (2025)35PVCMixed ventricular sites: PVCs mainly RVOT/LVOT, plus tricuspid annulus, RV moderator band, aortomitral continuity, papillary muscles, LV septum/fascicle, mitral annulus; VTs from mixed scar-related and idiopathic ventricular substratesNavistar SmartTouch + CENTAURINR25 A17 ± 15Younis et al.82 (2025)11VT/PVCLarge LV/RV scar substrates (anterior, septal, inferior, posterior-inferior, RVOT) and LV papillary musclesFarawave2.0 kVNR29 (16–51)Arnaud et al.83 (2025)5VFPurkinje-related VF substrates involving intracavitary structures, including LV false tendon, hypertrophic septum, and moderator band/anterior tricuspid papillary muscle junctionSphere-9 lattice-tipNRNR22.7 ± 15.1Padisak et al.84 (2025)12VT/PVCMixed ventricular sites: idiopathic RVOT PVCs and scar-related VT (RVOT/RV posteroseptal, RV anterolateral, LV apical, LV anterior/inferior/posterior, RV anteroapical)Farawave2.0 kVNR8 (8–12)Sacher et al.85/AVAAR Registry (2025)126VT/PVC/VFMixed ventricular substrates: VT, PVC, VF; LV and RV endocardial ± epicardial sites; ischemic, non-ischemic, RVOT, papillary, septal, peri-annular substratesSphere-9 lattice-tipNRNR33 ± 30Reddy et al.48 (2025)26VTMixed scar-related VT substrates, predominantly ventricular scar targeted endocardially; endo + epi mapping in a subsetFieldForce 8.5-F force-sensing focal PFA catheter>10 kVNR98 (58–121)Shokr88 (2025)3PVCPosteromedial papillary muscle (LV)FarawaveNRNR9 ± 5Cai et al.90 (2025)12PVCRVOT (n = 7), LVOT (n = 3), left anterior papillary muscle (n = 1), para-Hisian septum (n = 1)Focal contact-force sensing PFA catheter (Jinjiang system)1800VNR23 (11–35)LV, left ventricle; LVOT, left ventricular outflow tract; NR, not reported; PFA, pulsed field ablation; PVC, premature ventricular complex; RV, right ventricle; RVOT, right ventricular outflow tract; VF, ventricular fibrillation; VT, ventricular tachycardia. Summary of clinical studies with ≥3 patients LV, left ventricle; LVOT, left ventricular outflow tract; NR, not reported; PFA, pulsed field ablation; PVC, premature ventricular complex; RV, right ventricle; RVOT, right ventricular outflow tract; VF, ventricular fibrillation; VT, ventricular tachycardia. Preclinical corpus Forty-two animal studies were initially identified; nine studies reported only qualitative analyses of ventricular PFA and were therefore excluded (see Supplementary material online, Table S4). Quantitative depth or transmurality assessments were available for 33 studies involving n = 301 animals (n = 293[97%] swine) receiving ventricular PFA, for a total of n = 1579 lesions (healthy myocardium, n = 1428[90%], scarred myocardium, n = 151[10%]).19–51 Most applications were endocardial, with a smaller epicardial cohort (n = 226[14%]).21,23,26,30,39,41,45 Lesion depth and transmurality In healthy myocardium, the pooled mean lesion depth was 6.05 mm (95%CI, 5.27–6.84; I2 = 100%; Supplementary material online, Figure S1),19–24,26–51 with consistent results in healthy endocardial (5.96 mm [95%CI, 5.07–6.84; I2 = 99%])19,20,22,24,26,27,29,31–42,46–51 and healthy epicardial tissues (5.19 mm [95%CI, 2.58–7.79; I2 = 99%]; Figure 2).21,26,30,39,41,45 Lesion thicknesses were slightly higher in healthy left ventricular (6.79 mm [95%CI, 5.42–8.16; I2 = 98%])22,27,29,31,35,37,39–42,48,49,51 than in healthy right ventricular tissues (5.31 mm [95%CI, 4.38–6.24; I2 = 95%; Supplementary material online, Figure S2).23,27,33,37,39,42,51 Importantly, pooled mean lesion depths were quantitatively similar in scarred myocardial models (5.96 mm [95%CI, 5.39–6.53; I2 = 56%; Figure 2).22,23,25–27,35,40,49 Marked heterogeneity was observed in all these analyses, which could not be attributed to any individual study. The results of leave-one-out analyses are displayed in the Supplementary material (Supplementary material online, Figures S3–S8), together with analyses restricted to commercially available systems (see Supplementary material online, Figures S9–S11). Additionally, sensitivity analyses using log-transformed lesion thickness yielded similar pooled estimates and heterogeneity (see Supplementary material online, Figures S12–S15). Overall lesion thickness in healthy endocardium, healthy epicardium, and scarred ventricular myocardium. (A) Lesion thickness (mm) in healthy ventricular endocardium. (B) Lesion thickness (mm) in healthy ventricular epicardium. (C) Lesion thickness (mm) in scarred ventricular myocardium. Several studies reported lesion depth according to contact, number of applications/trains, and amplitude/duration settings (Figure 3). Within-study mean lesion depth difference between high contact (defined as a minimum of >30 g or as 0-mm offset or as stable contact or as a Δlocal impedance ≥30 Ohm) and low contact (defined as a maximum of <25 g or as 4-mm offset or as intermittent contact or as a Δlocal impedance ≤10 Ohm) lesions was 2.56 mm (95%CI, 1.61–3.52; I2 = 97%).21,32–34,36,37,39,42,46 Importantly, a positive association between increasing contact and greater lesion depth was reported with most systems in which this aspect was formally evaluated, including a multielectrode catheter,32 a dual-energy focal device,36 and the lattice-tip catheter,39 with the exception of an investigational PFA generator coupled with a contact force-sensing catheter.46 Similarly, within-study mean lesion depth difference between high number of applications (defined as 4–24 applications/trains) and low number of applications (defined as 1–12 applications/trains) averaged 2.31 mm (95%CI, 1.44–3.18; I2 = 90%).23,36,39,47,49 Furthermore, within-study mean lesion depth difference between high amplitude/duration (defined as 25A, 1500 V, or 15 s) and low amplitude/duration (defined as 19A, 1300 V, or 5 s) was 1.93 mm (95%CI, 1.21–2.65; I2 = 6%).29,38,49 Finally, seven studies reported lesion depth according to energy source (PFA vs. RF; additional details in the Supplementary Results).20,22,25,40,41,44,50 Within-study mean lesion depth difference between PFA and RF was −0.02 mm (95%, CI, −1.15–1.12; I2 = 89%) in healthy myocardium20,22,40,41,44,50 and 2.09 mm (95%CI, 0.84–3.34; I2 = 58%) in scarred myocardium models (Figure 4).22,25,40 The results of leave-one-out analyses are displayed in the Supplementary material online, Figures S16–S19. Within-study mean lesion depth difference between high and low contact, number of applications, amplitude or duration. (A) High contact was defined as a minimum of >30 g or as 0-mm offset or as stable contact or as a Δlocal impedance ≥30 Ohm; low contact was defined as a maximum of <25 g or as 4-mm offset or as intermittent contact or as a Δlocal impedance ≤10 Ohm). (B) High number of applications was defined as 4–24 applications/trains; low number of applications was defined as 1–12 applications/trains. (C) High amplitude/duration was defined as 25A, 1500 V, or 15 s; low amplitude/duration was defined as 19A, 1300 V, or 5 s. The mean lesion depth difference is expressed in mm. Within-study mean lesion depth difference between pulsed field ablation and radiofrequency energy. The mean lesion depth difference is expressed in mm. PFA, pulsed field ablation; RF, radiofrequency energy. Concerning the proportion of transmural lesions (which was assessed in 749 lesions from healthy myocardium models and 83 lesions in scarred myocardium models), it was 32% (95%CI, 15–52%; I2 = 97%) in healthy tissue (Figure 5),19,23,24,26,28,33,35–39,44,45,47,49,50 and 53% (95%CI, 36–70%; I2 = 52%; Figure 5) in scarred myocardium.23,25,26,35,49 Additional details are provided in the Supplementary Results and Supplementary material online, Figures S20–S27. Proportion of transmural lesions in healthy and scarred ventricular myocardium. Note that the proportion of transmural lesions is substantially higher in scarred myocardial model (B) than in healthy myocardium (A). Clinical population Among 355 patients, 122(34.4%) underwent PFA for PVCs and 233 (65.6%) for VT (n = 225) or VF (n = 8). The weighted mean age was 58 years (range 27–78), and 86 patients (24%) were women. Among VT/VF cases, the underlying substrate was ischaemic cardiomyopathy in 115 patients (49%), non-ischaemic cardiomyopathy in 82(35%), mixed (ischaemic and valvular) in 1(0.4%), and unspecified structural heart disease in 22(9%), while the remaining 13 patients(6%) had idiopathic VT/VF; the procedural approach was epi-endocardial in 38(16%) patients, and epicardial in 3(1%) cases.4,48,52–57,59–71,76–86,92 Among PVC cases, idiopathic PVCs accounted for most cases (n = 9 679%), followed by non-ischaemic (n = 2 016%) and ischaemic cardiomyopathy(n = 6,5%); the procedural approach was epi-endocardial in 2(2%) cases and involved ablation within the coronary venous system in 16(13%).58,72–75,80–82,84,85,87–92 PFA was used in redo procedures in 42 PVC cases (34%) and 142 VT cases (61%). Hybrid PFA/RF workflows were common (n = 142 overall; n = 97 for VT/VF; n = 45 for PVC). A detailed description of epi-endocardial/epicardial procedures and of cases involving ablation within the coronary venous system is provided in the Supplementary Results. Six PFA systems were represented in Table 1. The lattice-tip catheter was used in 160 cases (45%),70,71,85 a PFA generator with irrigated catheters in 107(30%),75,92 the pentaspline catheter in 48(14%),68,82,84 a contact force-sensing high-voltage investigational catheter in 26(7%),48 and two focal systems accounted for the remaining cases.89,90 Safety Procedure-related complications were reported in 63 patients overall (see Supplementary material online, Table S5). Of these, 59 occurred within 1 day of the procedure, whereas 9 were reported between Days 1 and 30. The most common complication was ST-segment changes, occurring in 16 patients (5%), followed by transient (n = 11, 3%)/persistent (n = 3, 1%) conduction system disorders, vascular access complications (n = 8, 2%), stroke/thromboembolic events (n = 5, 1%), and heart failure exacerbation/cardiogenic shock (n = 5, 1%). ST-segment changes occurred during epicardial PFA application in nine cases, after ablation within the middle cardiac vein in one patient (see Supplementary Results), and during endocardial ablation at the left/right ventricular outflow tract (n = 3), LV anterior wall (n = 2), and RV moderator band (n = 1), suggesting that proximity-related coronary artery spasm may have contributed to most cases. The proportion meta-analysis of all clinical studies showed a 4% (95%CI, 0–10%; I2 = 0%) rate of any procedure-related complications (see Supplementary material online, Figure S28). Restricting the analysis to studies with ≥3 patients yielded a complication rate of 10% (95%CI, 5–17%; I2 = 39%; Figure 6). Notably, the complication rate appeared lower with the pentaspline catheter (0%; 95%CI, 0–3%; I2 = 0%) and with the PFA generator coupled with irrigated catheters (3%; 95%CI, 0–19%; I2 = 21%) compared to the lattice-tip catheter (14%; 95%CI, 6–24%; I2 = 0%; Supplementary material online, Figure S29). Rate of any procedure-related complications in studies with ≥3 patients. In aggregated counts, six deaths (1.6%) were reported-all in VT cohorts. A structured table with timing, causes, and adjudicated attribution is provided in the Supplementary material online, Table S6. Efficacy—PVC cohort Among patients undergoing PVC ablation, the proportion meta-analysis showed an acute success rate (see Supplementary material online, Table S7 for study-specific definitions) of 98% (95%CI, 91–100%; I2 = 0) in the overall sample and 90% (95%CI, 83–96%; I2 = 0) in studies with ≥3 patients, with similar results irrespective of the ablation system (see Supplementary material online, Figures S30–S32). Over a weighted mean follow-up duration of 5.2 months (range 0.9–9.6 months), PVC recurrence was reported in 25 patients, corresponding to a meta-analytic recurrence proportion of 12% (95%CI, 4–22%; I2 = 0%) in the overall sample (see Supplementary material online, Figure S33) and 19% (95%CI, 12–28%; I2 = 0%) in studies with ≥3 patients (Figure 7). Regarding individual technologies, recurrence rates were highest with the PFA generator coupled with irrigated catheters (Figure 7).58,73–75,81,92 Mid-term recurrence rates in index and redo procedures are shown in the Supplementary material online, Figure S34. Rate of recurrence during mid-term follow-up—PVC Ablation. PVC recurrence rates during a weighted mean follow-up duration of 5.2 months are reported for studies with ≥3 patients (A) and individually for the principal ablation systems (B, pentaspline catheter; C, PFA generator coupled with irrigated catheters; D, lattice-tip catheter). PFA, pulsed field ablation; PVC, premature ventricular complex. Efficacy—VT cohort Among patients undergoing VT/VF ablation, 205(88%) underwent PVS at the end of the procedure to assess acute efficacy (see Supplementary material online, Table S8).4,48,52–54,56,57,59,59,61–63,65,66,68–71,76,77,79,81,82,84–86,92 In the overall proportion meta-analysis, which included several small studies and case reports with uniformly successful acute outcomes, the complete acute procedural success rate was 100% (95%CI, 97–100%; I2 = 0; Supplementary material online, Figure S35), while it was 92% (95%CI, 87–96%; I2 = 0%) considering studies enrolling ≥3 patients (see Supplementary material online, Figure S36), without significant differences across the principal ablation systems (see Supplementary material online, Figure S37). During a weighted mean follow-up of 5.1months (range 0.1–9.6 months), 45 patients experienced VT. The proportion meta-analysis showed a follow-up VT rate of 13% (95%CI, 3–26%; I2 = 18%; Supplementary material online, Figure S38) in the overall sample and 21% (95%CI, 10–35%; I2 = 63%) when restricting the analysis to studies with ≥3 patients (Figure 8; Supplementary material online, Figure S39). Rate of recurrence during mid-term follow-up—VT/VF ablation. VT/VF recurrence rates during a weighted mean follow-up duration of 5.1 months are reported for studies with ≥3patients (A) and individually for the principal ablation systems (B, pentaspline catheter; C, PFA generator coupled with irrigated catheters; D, lattice-tip catheter). PFA, pulsed field ablation; VF, ventricular fibrillation; VT, ventricular tachycardia. When analysing the three principal ablation systems individually, follow-up VT rates were highest for the PFA generator coupled with irrigated catheters (Figure 8). Subgroup analyses of studies enrolling ≥3 patients showed follow-up VT rates of 26% (95%CI, 12–43%, I2 = 42%) for index procedures and 33% (95%CI, 22–45%, I2 ≈ 3%) for redo procedures (see Supplementary material online, Figure S40). Risk of bias Preclinical Using SYRCLE and ARRIVE 2.0 tools (see Supplementary material online, Table S9), the overall risk of bias was moderate, primarily driven by frequent ‘unclear’ ratings for randomization, allocation concealment, and blinding, as well as incomplete reporting (sample-size justification, exclusions). Additional concerns included heterogeneity in catheter/waveform parameters, tissue conditions, and survival-to-histology intervals. Clinical The first-in-human VCAS study and the AVAAR registry were judged to have low-to-moderate risk of bias (see Supplementary material online, Table S10), with concerns related to small sample size and early-feasibility design (VCAS),48 as well as non-consecutive enrolment and short follow-up (AVAAR).85 Among case series/reports, quality was variable, ranging from moderate to high (see Supplementary material online, Tables S11 and S12). The main limitations included unclear inclusion criteria, heterogeneous endpoint definitions, and limited handling of missing data/confounding. Discussion This systematic review and meta-analysis provide a comprehensive evaluation of ventricular PFA, integrating preclinical lesion metrics with emerging clinical outcomes across a range of PFA systems and dosing strategies.11 The main results can be summarized as follows. First, preclinical data demonstrate that lesion depth associated with ventricular PFA is generally consistent regardless of the presence of scar.26,27 Furthermore, PFA produces lesion depths comparable to RF in healthy myocardium, whereas in scarred ventricular models PFA achieves greater lesion depth than RF. Nonetheless, transmural lesion formation is not universally achieved with endocardial-only or epicardial-only PFA applications. Transmurality appears more likely in scarred myocardium and within the RV endocardium compared with the LV.23 These observations suggest that, based on current evidence, endocardial-only PFA is unlikely to eliminate the need for endo-epicardial procedures when targeting complex, intramural or subepicardial substrates.2,67,93 Second, early clinical experience demonstrates high acute success rates in complex and selected patient populations, frequently including a substantial proportion of redo procedures.85 Mid-term outcomes are encouraging, particularly considering that most available data derive from technologies originally designed for atrial ablation.8,12,48,85 However, complication rates are not negligible, particularly when PFA is compared with contemporary RF ablation technologies,5 with additional PFA-specific considerations.85 Preclinical data Across experimental models, mean lesion depth with PFA is approximately 6 mm, translating into variable rates of lesion transmurality (Figures 2–4). Our exploratory direct comparative analysis demonstrated similar lesion depth between PFA and RF in healthy myocardium,20,22,40,41,44,50,94 whereas PFA produced significantly deeper lesions in scarred ventricular models.22,25,40 Although based on a limited number of heterogeneous studies, these findings suggest a potential advantage of PFA for targeting deep intramural substrates in scar-related VT.22,48 Mechanistically, while fibrotic tissue may limit conductive heating during RF delivery, it may conversely facilitate the distribution of PFA electric fields towards surviving myocyte bundles within heterogeneous scar.22,40,95 This property may enhance lesion effectiveness in regions that are traditionally challenging to treat with thermal energy.22,40,95 Importantly, the high heterogeneity in preclinical lesion thickness, resulting in wide prediction intervals, suggests that differences among ablation systems have a major influence on ventricular lesion formation and penetration. These differences involve not only catheter design (e.g. circumferential regional catheters developed for pulmonary vein isolation vs. large-tip focal or focal devices that may be better suited for ventricular navigation), but also PFA waveform characteristics, pulse parameters, and energy dosing strategies.8,10,11 Accordingly, the reported lesion metrics should not be generalized to PFA as a technology per se, but are instead platform-, system-, and dose-specific, representing an important caveat for clinical application (Table 1, Supplementary material online, Table S2).8,10,11 Whereas some systems use catheters originally designed for RF ablation and subsequently adapted for PFA delivery,75,92 others have been specifically developed for PFA applications, such as the high-voltage system, which was optimized for ventricular use.48 Substantial variability also exists in catheter features. Some systems retain saline irrigation, such as dual-energy devices and the high-voltage system, whereas others do not require irrigation.22,26 While irrigation is essential for lesion formation and safety during RF ablation and may mitigate residual thermal effects associated with some PFA systems,96 its impact on PFA lesion formation remains incompletely understood. As with RF ablation, several ablation parameters can be adjusted to increase lesion depth. Some evidence suggests that catheter-tissue proximity is sufficient for effective lesion formation,21 whereas other studies have reported an association between increasing contact force and greater lesion depth.32–34,37,39,42,46 The number of PFA applications20,31,36,47,49 and voltage/duration settings29,38,49 also seem to represent important determinants of lesion thickness. These variables may be combined into composite lesion indices,32 although their utility in the ventricular setting remains to be established. Additional strategies to augment lesion depth include bipolar PFA, in which energy is delivered between two catheters positioned across the myocardial wall,24,39 and RF/PFA stacking, which involves the sequential, colocalized application of RF and PFA using dual-energy devices.50 This approach may enhance lesion depth through an RF-mediated increase in local tissue conductivity, a phenomenon that appears to be driven by the rise in tissue temperature following RF application.50,97 While promising, these approaches require further validation, particularly in view of the potential risks associated with deeper or more extensive myocardial injury.24 Clinical data Despite advances in mapping and ablation strategies, recurrence after VA ablation remains substantial, with one-year freedom from recurrence approximating 60–80% for VT2,5 and 70–80% for PVCs.2,98 This suboptimal long-term efficacy has been attributed to several factors, including incomplete mapping, incomplete substrate/trigger elimination, and substrate progression.2,6 In this context, PFA may offer a complementary approach for improving substrate modification, particularly in deep myocardial substrates or within scar. Nonetheless, insufficient lesion depth in healthy myocardium, inadequate target identification, and lesion regression due to reversible electroporation may represent significant issues with current technologies adopted for ventricular PFA. Lesion reversibility appears particularly relevant, as both preclinical studies and a case report describing the histologic lesion analysis of an explanted heart after ventricular pentaspline PFA have shown that ablation lesions are characterized by a central irreversibly electroporated core surrounded by a zone of reversible injury, which may recover conduction over time.29,36,37,52 This phenomenon may have clinical implications for acute procedural endpoints, particularly the elimination of abnormal local electrograms and acute suppression of PVCs or non-inducibility of VT. Accordingly, preclinical studies have shown that the immediate post-ablation reduction in local electrogram amplitude does not correlate with lesion depth,27,35 while studies employing serial post-ablation electroanatomical mapping43 or magnetic resonance imaging have demonstrated a reduction in ablation lesion volume beyond the first 24 h.35,49 A similar pattern has also been reported in a clinical study of PVC ablation using a focal PFA system.90 Our meta-analysis confirms and expands these findings by reporting a discrepancy between the almost universal acute procedural success and the non-negligible risk of VA recurrence during mid-term follow-up, a finding that was particularly pronounced for the PFA generator coupled with irrigated catheters.92 Therefore, the optimal post-ablation waiting time for assessing acute procedural success with individual technologies should be better delineated as ventricular PFA enters the clinical arena. Whether conventional acute endpoints for VT ablation remain equally informative after PFA, or whether complementary strategies—for instance, more widespread use of post-ablation non-invasive programmed ventricular stimulation—may improve assessment of procedural efficacy, should be further investigated in future studies.2 Notwithstanding this issue, the high acute success and favourable follow-up outcomes observed in this meta-analysis should be interpreted in light of the complexity of the treated population, with 34% of PVC cases and 61% of VT cases representing redo procedures after prior RF ablation. The continuous development of dedicated PFA or dual-energy tools specifically for targeting ventricular tissue may further enhance procedural efficacy.48,89,90 From a safety perspective, the overall complication rate of approximately 10% in larger studies is consistent with available data on RF ablation.2 However, coronary events emerge as a specific concern during ventricular PFA procedures, particularly during epicardial ablation.99 While nitrate pre-treatment may mitigate acute spasm, the long-term consequences of PFA applications in proximity to coronary arteries remain incompletely defined and warrant further investigation.71,100 In addition, although the effects of PFA on the specialized cardiac conduction system are incompletely understood, the observed rates of transient/persistent conduction disturbances suggest that caution should be applied when delivering PFA near the atrioventricular node.24,75 A single study reported that the risk of atrioventricular block may be particularly high when using a linear catheter(ThermoCool-SmartTouch) coupled with a PFA generator via a retrograde-aortic approach due to current leakage from the ring electrode, highlighting the importance of technical factors.92 In most other reported cases, however, conduction disturbances were attributed to PFA delivery in close proximity to the cardiac conduction system rather than to energy shunting, suggesting that anatomical location may represent the predominant mechanism. Finally, although preclinical evidence supporting colocalized, sequential application of RF and PFA remains limited,50 hybrid workflows with RF/PFA stacking are increasingly reported in the literature, largely driven by the availability of dual-energy platforms.85 The relative safety and efficacy of RF-only, PFA-only, and combined approaches for ventricular arrhythmia ablation should be systematically evaluated in future studies. Limitations These findings should be interpreted in light of several limitations. Preclinical evidence is characterized by substantial heterogeneity in ablation platforms, energy delivery protocols, and the timing and methods used for lesion assessment. However, rigorous sensitivity analyses, including leave-one-out analyses and log-transformed meta-analyses, demonstrated that the findings were robust to both the exclusion of individual studies and the statistical approach used. Clinical data are limited by small sample sizes and relatively short-follow-up durations. Accordingly, differences in safety and efficacy outcomes across individual ablation systems should be considered hypothesis-generating. Notably, the systematic differences in efficacy and safety outcomes observed between overall estimates and those restricted to cohorts with ≥3 patients suggest the possibility of a small-study effect. This may reflect a combination of selection bias, publication bias (preferential reporting of technical successes), and the involvement of high-volume, pioneering centres. Standardization of procedural parameters, lesion assessment methodologies, and clinical endpoints will be essential to better define the role of PFA in VA ablation. Future studies will need to further address several open questions for ventricular PFA, including the impact of catheter irrigation and catheter design-especially regarding the comparison between large-focal and conventional focal catheters-PFA dosing, PFA/PFA or RF/PFA stacking, and the value of conventional acute procedural endpoints. Conclusions Ventricular PFA demonstrates high acute efficacy, acceptable near-term safety, and consistent mid-term durability across both PVC and VT ablation. Preclinical lesion metrics provide a mechanistic rationale for these findings, suggesting favourable penetration of PFA within scarred myocardium, where it appears to achieve greater lesion depth than RF, and emphasizing the role of PFA dosing. Future studies should systematically validate strategies to mitigate PFA-specific complications-such as coronary artery spasm-in VA ablation procedures, and to further define the safety and efficacy of combined RF/PFA approaches, including lesion stacking. Supplementary Material Acknowledgements None. Contributor Information Paolo Compagnucci, Cardiology and Arrhythmology Clinic, University Hospital ‘Azienda Ospedaliero-Universitaria Delle Marche’, Via Conca 71, Ancona 60126, Italy. Luca Finori, Cardiology and Arrhythmology Clinic, University Hospital ‘Azienda Ospedaliero-Universitaria Delle Marche’, Via Conca 71, Ancona 60126, Italy; Department of Biomedical Sciences and Public Health, Marche Polytechnic University, Ancona, Italy. Michela Casella, Department of Clinical Sciences, Marche Polytechnic University, Ancona, Italy; IRCCS Maria Cecilia Hospital - GVM Care & Research, Cotignola, Italy. Alessio Gasperetti, Department of Cardiovascular Medicine, Cleveland Clinic, Cleveland, OH, USA. Henry D Huang, Division of Cardiovascular Medicine, University of Chicago, Chicago, IL, USA. Yari Valeri, Cardiology and Arrhythmology Clinic, University Hospital ‘Azienda Ospedaliero-Universitaria Delle Marche’, Via Conca 71, Ancona 60126, Italy; Department of Biomedical Sciences and Public Health, Marche Polytechnic University, Ancona, Italy. Leonardo D’Angelo, Cardiology and Arrhythmology Clinic, University Hospital ‘Azienda Ospedaliero-Universitaria Delle Marche’, Via Conca 71, Ancona 60126, Italy; Department of Biomedical Sciences and Public Health, Marche Polytechnic University, Ancona, Italy. Francesca Campanelli, Cardiology and Arrhythmology Clinic, University Hospital ‘Azienda Ospedaliero-Universitaria Delle Marche’, Via Conca 71, Ancona 60126, Italy; Department of Biomedical Sciences and Public Health, Marche Polytechnic University, Ancona, Italy. Giorgio Giacomini, Cardiology and Arrhythmology Clinic, University Hospital ‘Azienda Ospedaliero-Universitaria Delle Marche’, Via Conca 71, Ancona 60126, Italy; Department of Biomedical Sciences and Public Health, Marche Polytechnic University, Ancona, Italy. Giovanni Volpato, Cardiology and Arrhythmology Clinic, University Hospital ‘Azienda Ospedaliero-Universitaria Delle Marche’, Via Conca 71, Ancona 60126, Italy. Laura Cipolletta, Cardiology and Arrhythmology Clinic, University Hospital ‘Azienda Ospedaliero-Universitaria Delle Marche’, Via Conca 71, Ancona 60126, Italy. Quintino Parisi, Cardiology and Arrhythmology Clinic, University Hospital ‘Azienda Ospedaliero-Universitaria Delle Marche’, Via Conca 71, Ancona 60126, Italy. Luigi Di Biase, Cardiac Arrhythmia Center, Division of Cardiology at Montefiore-Einstein Center, Bronx, NY, USA. Andrea Natale, St David’s Medical Center, Texas Cardiac Arrhythmia Institute, Austin, TX, USA; Department of Biomedicine and Prevention, Division of Cardiology, University of Rome Tor Vergata, Rome Italy; Interventional Electrophysiology, Scripps Clinic, San Diego, CA, USA; Metro Health Medical Center, Case Western Reserve University School of Medicine, Cleveland, OH, USA. Pasquale Santangeli, Department of Cardiovascular Medicine, Cleveland Clinic, Cleveland, OH, USA. Antonio Dello Russo, Cardiology and Arrhythmology Clinic, University Hospital ‘Azienda Ospedaliero-Universitaria Delle Marche’, Via Conca 71, Ancona 60126, Italy; Department of Biomedical Sciences and Public Health, Marche Polytechnic University, Ancona, Italy. Supplementary material Supplementary material is available at Europace online. Funding This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. Data availability All data were obtained from published sources; extraction worksheets and analysis code are available upon reasonable request. (10.17605/OSF.IO/TZ28H) Ethics statement This study synthesized published, de-identified data; IRB approval was waived. References Associated Data Supplementary Materials Data Availability Statement All data were obtained from published sources; extraction worksheets and analysis code are available upon reasonable request. (10.17605/OSF.IO/TZ28H)