If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Address reprint requests and correspondence: Dr David E. Haines, OUWB School of Medicine, Heart Rhythm Center, Beaumont Health System, 3601 West 13 Mile Road, Royal Oak, MI 48073.
Radiofrequency (RF) has become an accepted energy source for myocardial ablation but may result in discontinuous lesions and nontargeted tissue injury. We examined the feasibility and safety of lesion formation using high-amplitude, bipolar pulsed electric fields delivered from a multielectrode array catheter.
Objective
The purpose of this study was to compare duty-cycled radiofrequency ablation (RFA) to pulsed field ablation (PFA) in terms of acute electrical effects, 2-week lesion formation, and injury to nontargeted tissues.
Methods
Intracardiac ablations were performed in 6 pigs using a circular pulmonary vein ablation catheter. The energy source for ablation delivery was randomized to deliver either PFA or RFA to 3 atrial endocardial sites. Bipolar pace capture and electrogram amplitude measurements were recorded at each site. Histopathology and necropsies were performed after 2 weeks.
Results
The circular pulmonary vein ablation catheter was used to deliver pulsed electric fields to produce cardiac lesions without skeletal muscle stimulation. Evaluating all ablations in each site, electrogram amplitudes were reduced to <0.5 mV in 67.5% of PFA vs 27.0% of RFA deliveries (P <.001). Bipolar cardiac capture was lost after 100% vs 92.0% of PFA vs RFA (P = .005). At 2 weeks, PFA resulted in consistent transmural and homogeneous replacement fibrosis devoid of lingering myocyte “sequesters.” RFA lesions showed a stronger inflammatory response extending to the epicardial fat, arterial injury, and thrombosis. Neither PFA nor RFA lesions showed endocardial thrombus.
Conclusion
Intracardiac PFA can be feasibly delivered from a circular catheter to create fibrotic lesions that have acute electrical effects, without injury to nontargeted tissue.
Percutaneous catheter ablation using radiofrequency ablation (RFA) and cryothermy to achieve pulmonary vein (PV) isolation in atrial tissue have become widely accepted procedures for treatment of atrial fibrillation (AF).
Additional energy forms developed for catheter ablation include microwave, high-intensity focused ultrasound, low-intensity collimated ultrasound, lasers, cryoenergy, and heated saline. Many of the hyperthermal ablation technologies have encountered safety challenges, most commonly related to a lack of control over the extent of lesion formation, which can produce unintended injury to nontargeted tissues.
The impact of power output during percutaneous catheter radiofrequency ablation for atrial fibrillation on efficacy and safety outcomes: a systematic review.
Years after DC shocks had been abandoned because of safety issues, the mechanism of lesion formation from DC shocks was determined to be irreversible electroporation (IRE).
The mechanism of lesion formation in IRE is a function of electric field exposures that break down cell membrane permeability, leading to cell death. In the past decade, IRE has become an accepted treatment option for nonresectable tumors in close proximity to major blood vessels or nerves, as these structures seem to be relatively resistant to injury from IRE.
In recent studies, IRE has been applied to the heart in preclinical experiments demonstrating lesion formation in myocardium while preserving the integrity and function of nearby structures, such as the esophagus, lungs, coronary arteries, PVs, and phrenic nerve.
Pulsed field ablation (PFA) is a form of IRE that uses a train of bipolar and biphasic pulses of high voltage and short duration to create tissue injury without significant heating. It was hypothesized that PFA delivery to a 9-electrode circular array catheter in this preclinical feasibility trial could achieve atrial myocardial injury comparable to that achieved by duty cycled RFA, with reduced injury to nontargeted tissues.
Methods
Porcine ablation model
This preclinical study was conducted at Medtronic’s Physiological Research Laboratories (Minneapolis, MN) and was certified by the Association for Assessment and Accreditation of Laboratory Animal Care. The study protocol was approved by the Institutional Animal Care and Use Committee and conformed to the Guide for the Care and Use of Laboratory Animals. The porcine ablation model has been previously described.
Predictability of lesion durability for AF ablation using phased radiofrequency: power, temperature, and duration impact creation of transmural lesions.
In brief, six 4-month-old female Yorkshire-mix breed pigs (70.6 ± 3.5 kg) underwent general anesthesia. Amiodarone (150 mg intravenous bolus) was given, and heparin was administered to maintain an activated clotting time at a minimum of 350 seconds. A 7F steerable catheter (Marinr, Medtronic) was placed via femoral access in the right ventricular apex for sensing. The catheter was connected to the atrial channel of a temporary pacer (model 5388, Medtronic), which sent a ventricular pace signal to the triggering channel of the high-voltage pulse generator to gate the PFA deliveries during the ventricular refractory period. A 10F fixed-curve sheath (79 cm; Arrive, Oscor, distributed by Medtronic) was placed in the right femoral vein for access to the atria. Left atrial access was achieved by transseptal puncture with the fixed-curve sheath and a hypercurved dilator.
After all ablations were completed, the pigs recovered under veterinary supervision. Enoxaparin 60 mg intravenous twice daily was administered for anticoagulation, with clopidogrel 75 mg orally once daily as an alternative if venous access was not available. All pigs remained healthy and survived until their scheduled euthanasia (14–17 days) without clinical symptoms. Each pig was euthanized by introducing a quadripolar electrophysiological catheter and inducing ventricular fibrillation with 9-V DC applied in the right ventricle.
Catheter ablation
For ablation, a 9-electrode circular array pulmonary vein ablation catheter (PVAC GOLD, Medtronic) was powered by either a duty-cycled radiofrequency (RF) generator (GENius, version 15.1, Medtronic) or a custom-built PFA research generator. The duty-cycled RF system has been previously described.
Predictability of lesion durability for AF ablation using phased radiofrequency: power, temperature, and duration impact creation of transmural lesions.
Microembolism and catheter ablation I: a comparison of irrigated radiofrequency and multielectrode-phased radiofrequency catheter ablation of pulmonary vein ostia.
The RF generator simultaneously delivers pulses of unipolar and bipolar RF energy to the 9 electrodes of the circular array catheter (Figure 1A). RF energy delivered via the circular array catheter has been shown to efficiently create contiguous, transmural lesions capable of PV isolation with documented clinical efficacy and safety.
and reports on catheter deliveries of DC shocks, we hypothesized that pulsed electric field delivery through the circular array catheter could create contiguous transmural myocardial lesions akin to RF energy. The PFA research pulse generator was a novel, custom-designed generator that was originally designed to create linear atrial cardiac lesions with bipolar tissue clamp ablation devices.
It delivers high-voltage, biphasic pulse trains to the multielectrode catheter through a cable that connects electrodes 1, 3, 5, 7, and 9 as one polarity and electrodes 2, 4, 6, and 8 as the opposite polarity.
Figure 1A: Multielectrode circular array catheter used for bipolar pulsed field ablation (PFA) deliveries, shown with electrode numbering designation. Odd-numbered electrodes are connected to one polarity, whereas all even-numbered electrodes are connected to the opposite polarity for PFA deliveries. B, C: Detail of 3 biphasic pulses (B) delivered at the start (red box) of a train of 60 pulses (C) with a train delivery duration of 36 ms. D–G: Examples of electrogram recordings before (D, F) and after (E, G) ablation.
Six pigs were randomized to receive either RFA or PFA deliveries at each of the 3 endocardial sites (right superior PV ostium, left atrial appendage [LAA], and right atrial appendage [RAA]; n = 18). The sites were selected for the purpose of evaluating future clinical applications, especially in cases of complex trabeculated anatomy, for which direct tissue contact may be challenging. To ensure accurate and consistent positioning, catheter placement was guided by fluoroscopic imaging and was performed by an experienced veterinary surgeon who performed all procedures within a 2-day period. Furthermore, adequate tissue contact was confirmed by pace capture at 5 V before ablation.
At each anatomic site, the circular array catheter was applied 4 or 5 times, and either RF or PFA energy delivery was performed. Duty-cycled RF energy was delivered in a 2:1 bipolar/unipolar ratio, a maximum power of 10 W per electrode, and 60°C temperature setpoint for 60 seconds at each placement. PFA was delivered as biphasic pulse trains with a pulse width of 100 μs for each phase and with 200-μs interpulse pauses (Figure 1). Dosing of PFA mirrored previous work,
in which each placement of the circular array catheter consisted of 5 pulse trains of 60 pulses per train at 500 V delivered with R-wave gating over approximately 10 seconds. Each of the pulse trains had a duration of 36 ms. All PFA deliveries were made in a purely bipolar manner, energizing all odd-numbered electrodes at one polarity while the even numbered electrodes were the opposite polarity (Figure 1A). No energy was passed to a ground patch with PFA.
Electrogram data acquisition and analysis
Electrical data were collected before and after each ablation. Intracardiac bipolar electrograms (EGMs) were filtered from 30 and 500 Hz and recorded on a GE Cardiolab System (GE Healthcare, Chicago, IL) (Figure 1) from 5 electrode pairs on the circular array catheter (E1–2, E3–4, E5–6, E7–8, and E8–9). The presence or absence of pacing capture at 5.0 V was recorded from each pair. EGM amplitudes before and after ablation were analyzed. From the first ablation performed at each anatomic site, the pairs having a preablation EGM amplitude of at least 0.5 mV were selected as representative of having good tissue contact in the targeted area. Using data from only the first ablation at each site ensured that only nonablated tissue would be underlying the electrode array for this analysis. The postablation EGM amplitude was then used to calculate a percent reduction in EGM, and no further testing, such as entrance/exit block, was completed for this study.
Pathology
Pathology analysis was performed by 1 board-certified veterinary pathologist experienced in device pathology who was blinded to the assignment of ablation modality for each sample. At necropsy, the heart, adjacent lung lobes and nerves, esophagus, and kidneys were examined grossly. Ablated PVs and atria were examined and imaged before and after triphenyltetrazolium chloride staining. After samples were fixed in 10% neutral buffered formalin, the lesion size was measured (maximum width and length), multiple longitudinal sections were trimmed, and specimens were forwarded for histologic analysis.
Specimens were dehydrated, embedded in paraffin, and sectioned at approximately 4 μm. Slides were stained with hematoxylin and eosin and with Masson trichrome. Histopathologic assessment via bright-field microscopy included the following parameters: maximum depth and transmurality of ablation lesion, scoring of neointimal thickness, presence of endocardial thrombus, and presence of surviving or necrotic cardiomyocytes (“sequesters”) within lesions. Moreover, intralesional and collateral pathologic findings interpreted as ablation sequelae were diagnosed.
Statistical analysis
A combined analysis was performed on the circular array catheter ablations performed in the right PV, LAA, and RAA to compare RFA and PFA ablation modalities using quantitative EGM reduction and loss of pace capture as metrics for lesion efficacy. For each of these electrode pairs, the percentage change in EGM was calculated, and mean (±SD) was used to compare EGM reductions due to PFA and RF ablation using Minitab 17 software (Minitab Inc, State College, PA) in a 2-sample t test. For comparison of categorical factors, such as loss of pace capture, a χ2 test was used. P <.05 was considered significant.
Results
Acute procedure outcomes
Ablations were successfully performed at all protocol designated sites. None of the ablative energy deliveries produced any observable skeletal muscle or diaphragm stimulation. Neither ventricular arrhythmias nor AF was initiated by either energy source. At sites with consistent pace capture preablation, 100% had loss of capture after PFA vs 92.2% after RFA (P = .005). EGM amplitude reduction >50% for the first ablation in each site was seen in 91.1% of PFA vs 73.2% of RFA lesions (P = .027), whereas PFA resulted in a larger average percent reduction in EGM amplitude compared to RFA (72.2% ± 23.1%, n = 45; vs 56.4% ± 27.5%, n = 41; P = .005). Considering all ablations in each site, amplitudes were reduced to <0.5 mV in 67.5% of PFA vs 27.0% of RFA (P <.001) and were reduced across electrode pairs for each anatomic site targeted for ablation (Table 1 and Figure 2).
Figure 2Bipolar electrogram (EGM) amplitudes before and after ablation at only the first placements at each anatomic site for each ablation technology. Five electrode pairs were evaluated in each of 3 animals for each of the plots. LAA = left atrial appendage; PFA = pulsed field ablation; RAA = right atrial appendage; RFA = radiofrequency ablation; RPV = right pulmonary vein.
All PFA and RFA deliveries to the endocardial atrial sites resulted in early replacement fibrosis of the treated tissue, were easy to detect during necropsy, and were well demarcated from the surrounding, normal tissue. No endocardial disruption, thrombus, or charring was present, regardless of technology. The esophagus and the phrenic nerves were distant to all ablations and were devoid of ablation effects. Gross images of PFA lesions in the RAA are shown in Figure 3 and RFA lesions in Figure 4. The gross surface lengths at all sites were comparable between PFA and RFA (Table 1). Mean lesion length for PFA was 29.9 ± 11.8 mm (n = 8) and for RFA was 24.2 ± 15.7 mm (n = 6) (P = .479). In contrast, lesion width was significantly different between them. Mean lesion width for PFA was 12.0 ± 3.9 mm (n = 8) and for RFA was 6.5 ± 4.6 mm (n = 7) (P = .033). Lesion depths were determined via histology and yielded similar average depths at all sites for PFA and RFA (mean lesion depth for PFA: 2.25 ± 0.85 mm, n = 20; for RFA 2.47 ± 1.01 mm, n = 21) (P = .447). The range of lesion depths was 0.59–3.4 mm for PFA and 1.2–3.9 mm for RFA. Histopathology showed that 16 of 21 PFA specimens and 13 of 20 RFA specimens were focally transmural. Site-specific findings (Table 1) show the LAA to be the most challenging in which to achieve transmurality with either technology.
Figure 3Right atrial appendage of a pig that underwent pulsed field ablation 2 weeks earlier. Gross appearance of the endocardial surfaces of everted appendage in 2 different aspects (A, B), after opening of the appendage (C) and after triphenyltetrazolium chloride staining (D). Asterisks are shown along the ablation lesion. Dashed line in D indicates origin of the histology section (E). Note the sparing of the epicardial fat (arrowhead), presence of several fully patent intralesional arteries (arrows), and uniform appearance of the immature replacement fibrosis (blue). Masson trichrome stain.
Figure 4Right atrial appendage of a pig that underwent radiofrequency ablation 2 weeks earlier. Gross appearance of the everted appendage in 2 different aspects (A, B), after incision (C) and after triphenyltetrazolium chloride staining (D). Asterisks are shown along the ablation lesion. Stippled line in D indicates origin of the histology section (E). Note the effacement of the epicardial fat (arrowhead) with fibrosis and inflammation, presence of thrombosed intralesional arteries (arrow), and irregular appearance of the immature replacement fibrosis (blue) that is focally admixed with hemorrhage (open arrows). Masson trichrome stain.
Histopathologic examination yielded several pathognomonic features of PFA and RFA-induced lesions. Remodeling of the cardiac wall into fibrotic tissue because of PFA delivery was more homogeneous compared to RFA lesions (Figures 3E and 4E). Few islands of sequestered viable myocytes were seen with PFA ablations but were consistently observed at RFA sites (Figures 5A, 5B, and 6). Compared to RFA, PFA deliveries did not cause epicardial fat tissue inflammation in the treated area or noteworthy arteriolar remodeling (Figures 5C–5F and 7) as intralesional arteries were mostly within normal limits. Intima and media hyperplasia as well as thrombosis were frequent observations at the RFA sites. Additional findings for both ablation modalities comprised expected neointimal thickening on an undisrupted endocardium, absence of endocardial thrombus, and fibrosis of intralesional autonomic nerves. Finally, in contrast to RFA, there was no evidence of myocardial sparing around large intralesional arteries or trabeculae due to convective cooling by arterial or intracavitary blood flow in the PFA lesions (Figure 3, Figure 4, Figure 5, and 7).
Figure 5Histologic sections of various porcine atria that underwent radiofrequency ablation (RFA) (A, C, E) or pulsed field ablation (PFA) treatment (B, D, F) 2 weeks earlier. “Sequesters” (asterisks), hemorrhage, and thrombosis are present in A and homogeneous fibrosis and normal patent vessels in B. Larger magnification of remodeled arterioles with scant interstitial hemorrhage is shown in C and normal arterioles and no hemorrhage in D. Fibrosis and lipogranulomatous inflammation are shown in E, whereas the epicardial fat is normal in F.
Figure 6Low-magnification histologic cross-sections of the right atrial appendage (RAA) (A, B) and left atrial appendage (LAA) (C, D) showing radiofrequency ablation (RFA) lesions (A, C) and pulsed field ablation (PFA) lesions (B, D). Note the presence of hemorrhage and groups of lingering, necrotic cardiomyocytes (“sequesters,” asterisks) resulting in a heterogeneous replacement fibrosis of the RFA lesions. PFA lesions are more evenly healed and devoid of hemorrhage and “sequesters.” Also note that the epicardial fat over the transmural PFA lesion is widely normal (B:arrow) and that there is endocardial sparing at the RFA lesion. Masson trichrome stain.
Figure 7Histologic section of the left atrium after pulsed field ablation treatment 2 weeks earlier. Note homogeneous fibrosis that is not curtailed by the presence of several large-caliper arteries (arrows) in the ablation periphery. Masson trichrome stain.
This study was designed to compare intracardiac delivery of radiofrequency and pulsed electric fields delivered from the same multielectrode catheter platform in randomized atrial sites with the objectives of comparing acute measures of ablation technology performance as well as lesion formation and collateral damage at 2 weeks postablation. We observed that using the selected parameters, PFA lesion size was comparable to RFA and that there was no evidence of collateral damage. Compared to RF, there were microscopic differences in the healing characteristics of PFA at the 2-week timepoint, with fewer intralesional cardiomyocyte “sequesters” and absence of epicardial fat tissue inflammation. Moreover, these results were achieved without specifically tuning the electric field pulse parameters for intracardiac use, although the pulse delivery design is a key component of ablation efficacy.
The PFA intracardiac delivery design includes a number of parameters, such as pulse amplitude (applied voltage), pulse width, number of pulses to be delivered in a pulse train, and number of pulse trains to be delivered at each placement of the electrode array. As reported in previous studies of IRE ablations, high-voltage deliveries driving high current may produce heat in the immediate vicinity of electrodes,
However, no such material was observed on the energy delivery electrodes in this study, therefore suggesting that the mode of delivery of PFA did not produce heating that was of clinical significance. PFA energy delivery duration was also extremely fast, with energy deliveries on the order of milliseconds, compared to minutes with RFA. Regardless, safe energy delivery parameters for the PFA of myocardium still need to be evaluated in greater detail, because this study did not directly measure local temperature or impedance changes during ablation.
Energy vectoring
Other investigators have reported lesion creation through DC shock deliveries from large-area circular multielectrode catheters to create lesions through IRE.
Those studies reported similar findings to the present results, with transmural lesion creation and a lack of collateral injury. However, the energy vectoring used in those deliveries was unipolar, between all electrodes of the array catheter and a large area ground patch on the animal. The present study used purely bipolar energy delivery, forming a toroidal ablative field surrounding the circular electrode array. Such bipolar electric field vectoring avoided stimulation of skeletal muscle in this study. Tissues surrounding the array underwent IRE, which was observed to be proximity based and not affected by circulatory “heat sinks,” such as blood vessels, which are known to limit lesion formation with conventional ablations.
Conversely, lesions created with PFA within the atrial appendages in this study were observed to extend deeply and around trabeculae and muscle bundles, unaffected by blood flow cooling (Figures 3A–3C). These results suggest that PFA may be less dependent on direct energy conduction through the tissue, which may confer procedural benefits in that lesions could potentially be created even in the absence of uniform electrode–tissue contact. However, the importance of direct electrode–tissue contact with PFA requires further study because a similar number of specimens from both technologies were not circumferentially transmural.
Electrical findings
This study demonstrated that PFA was able to effectively reduce local EGM amplitude across all electrode pairings (Figure 2), suggesting the technology can create circumferential lesions within the heart. Furthermore, the circular array catheter was effective in safely delivering purely bipolar, intracardiac, pulsed electric fields to produce fibrotic intracardiac lesions without observed skeletal muscle stimulation.
Both PFA and RFA deliveries produced repeatedly transmural atrial lesions that were well demarcated, comparable in surface area, and devoid of endocardial thrombus. However, an important pathomorphologic observation was that PFA had more quiescent healing characteristics when compared microscopically to RFA (Figure 5). PFA sites presented with fewer lingering “sequestered” myocytes and reduced injury to intralesional arteries compared to RFA settings, which resulted in more homogeneous healing responses through the PFA lesions.
The mechanisms behind selective electroporation-induced cell death in tissues exposed to PFA need further research, but the fundamental process involves pulsed electric field applications to initiate membrane pore changes in the lipid bilayer until the cell membrane is irreversibly hyperpermeablized.
Cell characteristics of size and aspect ratio are known to play a role in determining the lethal threshold for cell death. Electroporation therapy has been explored in a variety of noncardiac cell types, with minimal reported adverse effects and safety risks.
The threshold for permeabilization of myocytes is unknown, and the mechanism of cell death may vary between skeletal, cardiac, and smooth muscle based on structural and electrical transmembrane differences.
Considering the findings of our study, PFA was observed to preferentially kill targeted cardiomyocytes while sparing other nontargeted intralesional cells (Figure 4, Figure 5, Figure 6) . Finally, PFA energy delivery duration is extremely fast, with energy deliveries on the order of milliseconds, compared to minutes with RFA.
Study limitations
The energy delivery parameters chosen for PFA have not been optimized for human intracardiac use, and the threshold pacing used in this study may not completely reflect applications in diseased myocardium. Variability in the baseline EGM amplitude was seen, particularly in appendage locations. This might limit the ability to use EGM amplitude reduction as a definitive indicator of successful ablation. In addition, the model presented here is limited by the 2-week follow-up, even though the healing and replacement processes were well advanced (Figure 4, Figure 5, Figure 6). Together, the impact on outcomes for human cardiac lesion formation, including risk of collateral injury to noncardiac structures, using this form of ablation energy is unknown, so additional preclinical and clinical evaluation is required to determine the most therapeutic and safe range for intracardiac pulsed electric field ablation deliveries in diseased human hearts.
Conclusion
PFA technology produced targeted cardiomyocyte death, reduced EGM amplitude, and resulted in lasting atrial lesions when delivered from the multielectrode circular array catheter. Compared to duty-cycled RF ablations, the healing characteristics of PFA lesions were devoid of a thermal signature, had an absence of lingering “sequestered” cardiomyocyte groups, had more uniform replacement fibrosis, showed significantly reduced epicardial fat inflammation, and resulted in less intralesional blood vessel remodeling, whereas both PFA and RFA deliveries were devoid of collateral damage. Further research on this new catheter ablation energy source is needed to verify reduced specific safety risks and potentially improved efficacy over existing ablation technologies.
Acknowledgment
The authors wish to thank Professor Damijan Miklavčič from the University of Ljubljana for proofreading of this manuscript and for valuable suggestions during the final stage of manuscript preparation.
References
Calkins H.
Hindricks G.
Cappato R.
et al.
2017 HRS/EHRA/ECAS/APHRS/SOLAECE expert consensus statement on catheter and surgical ablation of atrial fibrillation.
The impact of power output during percutaneous catheter radiofrequency ablation for atrial fibrillation on efficacy and safety outcomes: a systematic review.
Predictability of lesion durability for AF ablation using phased radiofrequency: power, temperature, and duration impact creation of transmural lesions.
Microembolism and catheter ablation I: a comparison of irrigated radiofrequency and multielectrode-phased radiofrequency catheter ablation of pulmonary vein ostia.
This study was fully funded by Medtronic PLC. Drs Kirchhof, Barka, Grassl, and Howard, and Mr. Stewart are employees of Medtronic, plc. Drs Haines and Verma are consultants for Medtronic, plc.
31116-0&id=fx1.jpg){kind=link}
31116-0&id=gr1.jpg){kind=link}
31116-0&id=gr2.jpg){kind=link}
31116-0&id=gr3.jpg){kind=link}
31116-0&id=gr4.jpg){kind=link}
31116-0&id=gr5.jpg){kind=link}
31116-0&id=gr6.jpg){kind=link}
31116-0&id=gr7.jpg){kind=link}