Comparison of pulsed field ablation and cryoballoon ablation for pulmonary vein isolation

Pulmonary vein isolation (PVI) remains the cornerstone in the treatment of atrial fibrillation (AF). PVI using cryoballoon (CB) technology has emerged as a standard procedure in many centers. Recently, pulsed field ablation (PFA) has been introduced and used to achieve PVI. First data show high acute and favorable long‐term outcomes. So far, data comparing these new “single shot” devices are sparse. We sought to compare procedural and outcome data for first time PFA users versus CB in patients undergoing de novo PVI. Furthermore, potentially postprocedural discomfort and affection of autonomic ganglia were assessed.


Conclusion:
The new PFA technology is equally effective and safe as compared to CB for complete PVI with potentially shorter procedure time and significantly less contrast medium.However, AF recurrence rates after PFA PVI seem to be comparable to CB PVI.

| INTRODUCTION
Pulmonary vein isolation (PVI) has been proven to be more effective in maintaining sinus rhythm than noninvasive strategies. 1,2Alongside several different ablation techniques, the cryoballoon (CB) has become one of the most often used devices to achieve PVI.Its steep learning curve, good reproducibility, therefore favorable safety profile and its significantly shorter procedure time compared to conventional radiofrequency (RF) PVI 3 are reasons for the widespread use of CB ablation.Recent data emphasized on the efficacy of CB PVI compared to antiarrhythmic drug therapy. 1,2Although alternative options have been presented, 4,5 the vast majority of physicians use contrast medium in combination with fluoroscopy to obtain ideal PV occlusion and therefore successful PVI limiting its use in patients for example, with impaired renal function.Furthermore, this thermal ablation technique is of potential harm for adjacent structures such as the esophagus and phrenic nerve. 6,7tely, pulsed field ablation (PFA) was implemented into clinical routine as a new non-thermal ablation modality to achieve PVI. 8 Very short high voltage pulsed fields induce electroporation causing an increased permeability of cell membranes consecutively resulting in death of cardiomyocytes. 9In theory, this high tissue specificity potentially omits damage to adjacent structures like the esophagus and phrenic nerve. 8First data show promising acute 10 and favorable long-term outcomes after 12 months [11][12][13] and safety profiles. 14After streamlining PFA procedure learning curve, complete PVIs are performed quite fast 15 with high long-term durability. 16erefore, shorter procedure times, fast acute PVI, a potentially lower complication rate due to myocyte specificity and use of less/no contrast medium and fluoroscopy in PFA are potential advantages over CB PVI which still is the gold standard in many centers for de novo PVI.So far, data comparing PFA and CB ablation for de novo PVI are sparse.

| Study population
In this retrospective single-center analysis, all consecutive patients undergoing PFA-based de novo PVI between September 2021 and October 2022 in the setting of paroxysmal and persistent atrial fibrillation (PAF and persAF) were analyzed and compared to consecutive patients who underwent de novo PVI using the Boston Scientific POLARx™ CB system in the same time period.After implementation of PFA, less CB PVI were performed hence, for equal comparison one-third of the enrolled patients underwent CB PVI between February and September 2021.All physicians (five different operators) used PFA for the first time.Data acquisition was performed using an electronic data capture system 17 (RedCap Database).
The study was approved by the local Ethics Committee of the University of Cologne and complied with the Declaration of Helsinki.
All patients provided written informed consent.

| Procedure setting
In all procedures, oral anticoagulation was interrupted on the day of CA and continued on the evening of PVI.An international normalized ratio value of 2−3 was targeted if the patients received vitamin K antagonists.Propofol, midazolam and fentanyl were used for deep analgosedation. 18Transesophageal echocardiography (TEE) was performed if patients presented in an atrial arrhythmia.In SR, TEE was conducted only in patients with interrupted OAC before CA and CHA 2 DS 2 -VASc-Score > 2. After positioning two femoral sheaths, a decapolar reference catheter (Dynamic XT™, Boston Scientific) was placed in the coronary sinus (CS).Under fluoroscopic guidance, transseptal puncture was performed using a transseptal delivery system (TSX, Boston Scientific).After TSP, a weight-adapted heparin bolus was administered.To maintain an activated clotting time (ACT) > 300 ms, repetitive heparin boli were administered.Procedure time was defined as the time from groin puncture to sheath removal (skin-to-skin time).A figure-of-eight suture for venous closure with additional compression badge (6 h) was applied after sheath removal. 19Directly after the procedure, pericardial effusion was ruled out by transthoracic echocardiography (TTE) in the electrophysiology (EP) laboratory.Repeat TTE examinations were performed 2 and 24 h after the procedure.

| PFA PVI
All PFA procedures were performed with the FARAPULSE™ system (Boston Scientific).After TSP, the 8.5-F TSX sheath was replaced by the 13-F PFA sheath (FARADRIVE™, Boston Scientific).In the first 30 PFA procedures, an extra stiff straight guidewire (Amplatz extra stiff straight wire™, Cook Group Incorporated) was used to intubate the PVs followed by overthe-wire positioning of the PFA catheter (FARAWAVE™; Boston Scientific) at the respective PV antrum.After 30 procedures, a J-tip guidewire (InQwire™, Merit Medical Systems) was used instead of the straight guidewire according to the manufacturer's recommendation.To prevent vagal reactions potentially leading to bradycardias and asystole during PFA, 0.5 mg atropine were administered before the first PFA impulse.Also, if applicable, the CS catheter was placed at the LV summit or within a ventricular CS-branch to provide ventricular pacing if needed.At every PV antrum, 4 PFA impulses were delivered in the "basket" and "flower" configuration of the catheter, respectively.The delivery of more PFA applications was at operators' discretion.The FARAWAVE™ catheter was used to display the local electrograms at the PV antrum (between the PFA impulses) and inside the PV (after the impulses) to reveal entrance block.In 14 (26%) procedures, an electroanatomical 3D voltage map (CARTO3™, Biosense Webster or EnsiteX™, Abbott Medical) of the left atrium (LA) was acquired with a multipolar mapping catheter (PENTARAY™, Biosense Webster Lasso™; Biosense Webster; Advisor™ HD Grid, Abbott Medical) before and after treatment of all PVs to verify entrance block and change in displayed voltage.An analysis of the prescription of analgesics due to postprocedural pain in the first 48 h after PVI was conducted using a fully digital patient file.

| CB PVI
In all CB procedures, the 31 mm POLARx™ cryoablation system (Boston Scientific) was used.Monitoring of the esophageal temperature was established using a temperature probe (S-Cath, Esophageal Temperature Probe, Circa Scientific Inc.).Ablation was stopped when esophageal temperature dropped below 16°C to prevent atrioesophageal fistula formation or esophageal erosion.TSP was performed as previously described followed by introducing the POLARMAP™ circular mapping catheter and the POLARx™ CB (Boston Scientific) into the LA using the steerable POLARSHEATH™ 12.5-F sheath (Boston Scientific).Sufficient PV occlusion was assessed fluoroscopically by application of contrast medium distally of the inflated CB.
Again, after PVI entrance block was assessed using the POLARMAP™ catheter.Phrenic nerve capture using diaphragmatic compound motor action potentials (CMAP) was ensured during freezes of the right PVs.The duration of freezes was determined by either the timeto-isolation (TTI), abolishment of PV potentials if displayed on the circumferential mapping catheter (POLARMAP™) or by reaching nadir temperature.If TTI was within 60 s, a freeze of 180 s was applied.In case TTI was achieved after 60 s, a longer freeze of 240 s was administered.Nadir temperature was defined as −60°C and led to abortion of the freeze.As stated before, prescription of analgesics was also analyzed in the CB group.

| Follow-up
Follow-up was obtained during outpatient clinic visits at 3 and 12 months after PVI.Before every visit, a 24-h Holter electrocardiogram (ECG) was performed to detect atrial arrhythmia recurrence and if applicable patients were encouraged to use their smart devices.
During outpatient clinic visits, an additional 12-lead ECG was obtained.Heart rates (HR) in SR before PVI and at 3-month followup were retrospectively analyzed.For better comparison of HR, patients were excluded from HR analysis if no ECG in sinus rhythm was available before ablation.Furthermore, tele-consultations and interrogation of cardiac implantable electronic devices complemented the follow-up if applicable.Recurrences of atrial arrhythmias (AF, atrial tachycardia [AT], atrial flutter [AFL]) lasting for longer than 30 s after a 90-day blanking period were documented and considered as arrhythmia recurrence.

| Endpoints
Primary endpoints consisted of procedural data, reported postprocedural discomfort and arrhythmia-free survival.Furthermore, we assessed complication rates for PFA and CB PVI and a learning curve for PFA in particular.Moreover, PFA procedures were divided into tertiles (18 patients each) and directly compared regarding differences in procedure and fluoroscopy durations.
In the PFA group, 33 ± 4 applications were needed to achieve complete PVI compared to 6 ± 2 freezes in the CB group.For CB PVI, mean TTI was 87 ± 145 s and mean total freeze time per vein was 279 ± 160 s.

| Complications
Two pericardial tamponades occurred in the PFA group (4%) compared to none in the CB group (p = 0.495).Of those two

| DISCUSSION
We compared the novel non-thermal PFA modality to the wellestablished thermal CB ablation technique for de novo PVI in the setting of PAF and persAF.

| Main findings
1. Non-thermal PFA PVI is equally effective and safe as compared to the well-established CB PVI in PAF and persAF.
2. Procedure duration trends to be shorter for PFA PVI.
3. Postprocedural patient discomfort seems to be equal in both modalities.
4. A steep learning curve, high efficacy and favorable safety profile of PFA potentially supersedes CB ablation for PVI.

| Procedural characteristics
This study compared procedural data and outcome data of two "singleshot" devices for PVIthe novel non-thermal PFA with thermal CB ablation.PFA PVI revealed a trend to shorter procedure times although all operators used PFA for the first time and were experienced CB users.
During the first cases EAM was performed to evaluate PFA ablation extension which added procedure time to the overall analysis.Excluding EAM PFA procedure times were significantly shorter.Of note, a steep learning curve and streamlining the procedure potentially leads to even shorter procedure times as shown by other groups. 15uoroscopy times were longer in PFA in comparison to CB which might be explicable by the learning curve and the necessity of Compared to CB in PFA complete tissue contact is not as crucial. 20In contrast, midterm follow-up data suggest worse outcomes after CB PVI in patients with atypical PV anatomy. 21erefore, PFA might be of advantage in challenging anatomies.

| Safety
Overall, both technologies revealed a favorable safety profile for PVI in the setting of PAF and persAF.However, types of reported complications differed between both groups: After PFA PVI, 2 of 54 patients suffered from a pericardial tamponade indicating a risk of 4%.Compared to the large MANIFEST and EU-PORIA registry (respective 1%), the rate of tamponades was higher in our study. 14,22The occurrence of the latter, also reported in other studies, 14  The rate of zero phrenic nerve palsies for PFA in our study is completely in line with previous studies and emphasizes on the tissue specificity for myocytes in the setting of PFA-PVI. 14Also, the observed rate of 4% phrenic nerve injuries after CB-PVI with complete recurrence has been reported recently in a large retrospective multicenter registry. 6Of note, in both groups no atrioesophageal fistulas or PV stenoses were reported during FU.
Larger studies are needed to evaluate, whether PFA reduces the damages of adjacent tissue during CA in comparison to thermal modalities.All reported findings are to appreciate in the setting of new PFA users.

| Outcomes
Arrhythmia-free survival after 273 ± 129 days was equal in both study groups (PFA: 74%, CB: 72%, HR: 0.98, p = 0.88) consisting of patients suffering from PAF and persAF matching recently published data from Urbanek et al. showing comparable procedural efficacy between both modalities. 23 far, long-term outcome data after PFA PVI are sparse.
However, our results are in line with the large non-randomized MANIFEST and EU-PORIA registries including patients undergoing PFA PVI with the FARAPULSE™ system and reporting an arrhythmiafree survival in PAF and persAF of 78% and 76% after 12 months. 13,22Moreover, our data are comparable to results from a recently published study analyzing outcomes using a lattice-tip catheter. 24Of note, long-term freedom from atrial arrhythmias in our study was favorable compared to findings from a trial enrolling 150 patients undergoing PFA PVI with a circular PFA catheter. 11wever, our findings using the FARAPULSE™ system are not transferable to other PFA systems.Each system using PFA applies distinct waveforms and energy settings with potentially different outcomes and safety profiles. 24e impact of ablation of ganglionated plexi (GP) in AF remains debatable. 25,26GP ablation can be recognized by higher resting HR in SR. 27 Our data did not reveal significant differences in heat rate changes between non-thermal PFA and thermal CB PVI.Contrary to our data, Musikantow et al. reported a significantly increased HR in SR in thermal CB and RF-balloon PVI compared to PFA. 28

| LIMITATIONS
This study is limited by its retrospective design and limited sample size.As in many other studies, assessment of arrhythmia recurrence was limited to 24 h-Holter ECGs in the majority of patients.However, a continuous rhythm monitoring using for example, implantable loop recorders would have been desirable.Due to a certain number of patients lost to FU some potentially late occurring complications might be underestimated.Further prospective randomized trials should evaluate potential benefits of PFA compared to CB for AF ablation.

| CONCLUSION
In PFA beginners achieving complete PVI in the setting of PAF and persAF is comparable with CB PVI without significant differences in efficacy and safety.Of note, the learning curve for PFA appears quite steep leading to potentially significant shorter procedure duration as compared to the already fast CB PVI.The previous reported tissue specificity for PFA in the setting of PVI was confirmed and is of potential advantage as opposed to CB.
Statistical analysisStatistical analyses were performed with Microsoft Excel (Microsoft Corporation, Redmond) and GraphPad Prism (Version 9.5.1 for Mac, GraphPad Software).Continuous variables are presented as mean ± standard deviation, categorical variables were summarized as counts and percentages.The D'Agostino-Pearson test was performed for normality tests.Unpaired t-tests were used for continuous variables if they were normally distributed.Otherwise, Mann−Whitney U tests were performed.Fisher's exact test was created in case of dichotomous variables.For the outcome analysis we conducted Kaplan−Meier analyses.Here, the Gehan−Breslow−Wilcoxon test for p Value and the Mantel−Haenszel test for hazard ratio were conducted.A p Value < 0.05 was generally defined as statistically significant.
Complete PVI was achieved in 216 of 216 (100%) PVs in the PFA group and 215 of 216 (99.5%)PVs in the CB group.Due to occurrence of transient phrenic nerve palsy completion of PVI was omitted in one patient.

1 2
Development of procedure (A) and fluoroscopy (B) time in PFA procedures over time.PFA, pulsed field ablation.Comparison of procedure (A) and fluoroscopy (B) times of the first, second and third tertile of PFA procedures (****p ≤ 0.0001, *p ≤ 0.05, ns, non-significant).
was mainly explicable due to the initially recommended extra stiff straight tip guidewire (Amplatz extra stiff straight wire™) used for transseptal PFA catheter positioning in both patients suffering from pericardial tamponade and therefore representing a possible cause of this complication.After switching to a J-tip shaped guidewire (InQwire™), no further wire-related pericardial tamponades occurred.Although one pericardial tamponade was most likely related to complex TSP, the large 13F-sheath used for transseptal delivery and the PFA catheter itself could be considered as a cause for the observed tamponades.

Table 2 .
Note: Values are presented as mean ± SD.Categorical data are given as n (%).Abbreviations: CAD, coronary artery disease; GFR, glomerular filtration rate; LVEF, left ventricular ejection fraction.group detected during phrenic nerve pacing.After a waiting time of 15 min in both cases phrenic nerve capture was restored.Therefore, no persistent phrenic nerve injury was reported in the overall cohort.All procedural complications are displayed in