Myocardial injury and inflammation following pulsed‐field ablation and very high‐power short‐duration ablation for atrial fibrillation

Pulmonary vein isolation (PVI) using radiofrequency ablation (RFA) is an established treatment strategy for atrial fibrillation (AF). To improve PVI efficacy and safety, high‐power short‐duration (HPSD) ablation and pulsed‐field ablation (PFA) were recently introduced into clinical practice. This study aimed to determine the extent of myocardial injury and systemic inflammation following PFA, HPSD, and standard RFA using established biomarkers.


| INTRODUCTION
Pulmonary vein isolation (PVI) represents the cornerstone of catheter ablation for atrial fibrillation (AF). 1 Radiofrequency (RF) is the most commonly used ablation modality and employs thermal energy to create transmural, contiguous, and irreversible lesions.3][4] In an effort to improve PVI efficacy and safety, high-power short-duration (HPSD), RFA, and pulsed-field ablation (PFA) were recently introduced into clinical practice.
Whereas conventional RFA commonly employs 20-40 W for 15-30 s, HPSD uses higher power (>45 W) for a shorter duration (<20 s).9][10] PFA is a novel, nonthermal cardiac ablation modality using highvoltage electric fields to create nanoscale pores in the cell membrane through a mechanism known as irreversible electroporation. 11,12Cell membrane disruption compromises cell homeostasis eventually leading to cell death.Currently available PFA protocols were designed to efficiently and selectively electroporate cardiac tissue, while sparing adjacent tissue such as vessels, nerves, and the esophagus. 11,12PFA lesions are homogeneous with a preserved tissue architecture and are reported to be associated with less tissue inflammation than RFA in animal models. 13,14PFA thus bears the potential of reducing the postablation inflammatory reaction seen with RF technologies.
While biomarkers of myocardial injury and inflammation have been extensively studied with conventional RFA, 3,4,15 clinical data on PFA and very HPSD with >50 W (vHPSD) is currently lacking.Highsensitivity cardiac troponin T (hs-cTnT) is a specific biomarker of myocardial injury which is released after PVI and correlates with RF duration. 3,4,16The aim of this study was to compare the extent of myocardial tissue damage and systemic inflammation of PFA, HPSD-70W, and HPSD-90W with standard RFA using established biomarkers.All patients provided written informed consent for the ablation procedure.The study was approved by the local ethics committee (approval #2020-348_1-S-NP).

| Ablation procedure
An uninterrupted oral anticoagulation regimen was applied in all patients.Left atrial thrombus was excluded ≤48 h before the ablation using contrast-enhanced cardiac computed tomography (CT).Transesophageal echocardiography was performed in case of contraindications for contrast agent administration or CT unavailability.
Antiarrhythmic drugs (AAD) were discontinued >5 half-lives before the procedure and were not restarted afterward.
Ablation was performed under sedation using midazolam, propofol, and fentanyl.After gaining venous femoral access, fluoroscopy-guided single transseptal puncture was performed using the Agilis™ steerable sheath (Abbott) in the standard RFA and HPSD-70W group, the CARTO VIZIGO™ (Biosense Webster) sheath in the HPSD-90W group or the Swartz™ Braided LAMP™ 45 sheath (Abbott) in the PFA group.Double access to the left atrium (LA) was obtained in the RF groups, while PFA ablation was performed using a single access.Unfractionated heparin was administered after transseptal puncture by continuous infusion to achieve an activated clotting time ≥300 s throughout the procedure.

| RFA
High-density electroanatomical mapping was performed using EnSite™ Precision and the Advisor circular catheter (Abbott) in standard RFA and HPSD-70W.The CARTO ® 3 system and the Lasso catheter (Biosense Webster) were employed in HPSD-90W.
Wide antral circumferential PVI was performed using point-bypoint-lesions with the 4 mm irrigated-tip catheter FlexAbility™ SE and the Ampere ® RF generator (Abbott; standard RFA and HPSD-70W) or the 3.5 mm irrigated-tip catheter QDOT MICRO™ and the nGEN™ RF generator (Biosense Webster; HPSD-90W).The ablation catheter was not placed in the steerable sheath in any of the groups.
In the standard RFA group, 30-40 W were applied for 30 s on the anterior wall and for 20 s on the posterior wall.In the HPSD-70W group, 70 W/7 s (anterior wall) or 70 W/5 s (posterior wall) were applied.Ablation was performed in both groups using a powercontrolled mode (automatic temperature cut-off 42°C, irrigation 17 mL/min [standard RFA] and 20 mL/min [HPSD-70W]) and each RF delivery was limited by an automated duration control.An interlesion distance of ≤6 mm (standard RFA) or 5-6 mm (HPSD-70W) was targeted.In case first-pass PVI was not achieved, touch-up lesions were applied using the same ablation settings.
In the HPSD-90W group, 90 W/4 s was applied on the anterior and posterior wall in a temperature-controlled mode (QMODE+) with a target temperature of 60°C.In case first-pass PVI was not achieved, touch-up RF lesions using 50 W/15 s were applied in the conventional QMODE.An interlesion distance of 5-6 mm was targeted.
PVI was confirmed by entrance block by placing the circular mapping catheter in each PV antrum in all RF groups.Acute reconnection was then assessed in each patient either by rechecking entrance block after ≥20 min waiting time or by adenosine administration.In case of documented PV reconnection, ablation was performed at the site of earliest activation.

| PFA
After transseptal puncture, the Swartz™ Braided LAMP™ 45 sheath was replaced by the FaraDrive™ sheath (Farapulse) and the pentaspline Farawave™ catheter (Farapulse) was advanced in the LA over a straight 0.035″ stiff guidewire.Atropine (0.5-1 mg) was administered intravenously before the first pulse to prevent vagal reactions.Intermittent RV pacing was performed with a quadripolar catheter (Supreme, St. Jude Medical) placed in the RV apex in case of vagal reactions involving transient asystole.For each PV, we applied multiple biphasic and bipolar pulse cycles of 2 kV with the Farastar™ PFA generator (Farapulse).Specifically, four pulse cycles were applied in the "basket" configuration, followed by four pulse cycles in the "flower" configuration.After two consecutive cycles in the same position, the catheter was slightly rotated before applying the two following cycles.Additional pulses were only applied in case elimination of all PV potentials by 4 + 4 pulse cycles was not achieved.Entrance block was confirmed at the end of the procedure by placing the Farawave™ catheter in each PV ostium.

| Blood sampling
Blood collection from an antecubital vein is routinely performed at our center the morning of the day before and after the ablation procedure.Biomarkers were therefore measured 24 ± 2 h before (baseline) and 24 ± 2 h after PVI (Day 1).Blood sampling was also performed 48 ± 2 h after PVI (Day 2) in patients not discharged the day after the procedure.White blood cell (WBC) count was assessed using an automated blood cell counter.Serum hs-cTnT was quantified using the Elecsys Troponin T high-sensitive assay (cut-off value 14 ng/L, Roche Diagnostics).Assay-specific cut-off values for creatine kinase (CK) and CK-MB activity were 170 and 24 U/L, respectively.Hs-cTnT release, CK release, CK-MB release, and WBC elevation were quantified by subtracting biomarker levels measured on Days 1 and 2 postablation from baseline levels.

| Follow-Up
Pericardial effusion was routinely excluded at the end of the procedure and the following day using transthoracic echocardiography.The groin was clinically inspected the day after the procedure.In case of abnormalities, a duplex sonography was performed.Follow-up visits were scheduled at 3 and 6 months after ablation in our outpatient clinic for all patients and included a 12-lead electrocardiogram (ECG), a 5-day Holter ECG, and a review of any other available external ECGs.Arrhythmia recurrence was defined as any atrial arrhythmia (AF or atrial tachycardia) with a duration ≥30 s documented after a 2-month blanking period following the ablation procedure.

| Changes in biomarkers of myocardial injury and inflammation
Baseline levels of hs-cTnT, CK, CK-MB, WBC, C-reactive protein and creatinine were comparable between groups and are presented in Abbreviations: HPSD, high-power short-duration; PFA, pulsed-field ablation; PV, pulmonary vein; RFA, radiofrequency ablation.
The day after ablation, hs-cTnT release, CK release, CK-MB release, and WBC count elevation significantly differed between groups (Table 2).
Hs-cTnT release was similar between PFA and vHPSD groups.Postablation CK and CK-MB release was higher in the PFA group

| Correlation between cardiac injury and inflammation
Pearson correlation analyses were performed to assess whether myocardial injury is associated with RF duration and systemic inflammation.Hs-cTnT release significantly correlated with RF duration in all RFA groups (Figure 4), whereas the best correlation was found in the HPSD-70W group (r = .542,p < .001).A significant correlation between WBC and hs-cTnT release was found in the standard RFA group only (r = .410,p = .003),whereas hs-cTnT release and WBC did not correlate in the HPSD-70W, HPSD-90W, and PFA groups.

| Procedural safety and rhythm outcomes
Procedural complications were similar between groups and are listed in Table 3.One patient with a history of two previous strokes presented with a minor stroke 4 days after PFA and showed no persisting neurological deficit on follow-up.Freedom of any atrial arrhythmia at 6 months off AAD was 73.1% (standard RFA) versus 67.8% (HPSD-70W) versus 70.0% (HPSD-90W) versus 84.4% (PFA; p = .391,Table 3).The predictive value of the studied biomarkers for arrhythmia recurrence was assessed using logistic regression.Hs-cTnT release was associated with a reduced risk of arrhythmia recurrence in the standard RF group (odds ratio 0.821, confidence interval 0.675-0.997,p = .047),while no significant association between the studied biomarkers and rhythm outcome was found in any other group (Table 4).

| PFA is associated with a higher myocardial injury
PFA is a novel ablation modality which is increasingly being used in the clinical setting with promising efficacy and safety results. 17,18wever, the extent of myocardial injury and inflammation induced  Abbreviations: HPSD, high-power short-duration; PFA, pulsed-field ablation; RFA, radiofrequency ablation.
POPA ET AL.
| 323 by PFA as compared to the gold-standard RFA has remained poorly characterized in patients.
In this study, PFA was associated with a significantly higher release of myocardial injury biomarkers hs-cTnT (1.4-fold), CK (3.4-fold), and CK-MB (5.8-fold) as compared to conventional RFA (30-40 W).The most intuitive explanation for this finding is that PFA generates a larger volume of cardiac tissue damage.In agreement with this, a previous magnetic resonance imaging (MRI) study reported 60% higher acute late gadolinium enhancement with PFA than with thermal ablation. 19This is likely due to a combination of factors such as catheter tip design (pentaspline PFA catheter generating a larger lesion surface than point-by-point RFA creating linear lesions) and an improved lesion efficacy of PFA determined by less reliance on contact force, catheter orientation and blood flow velocity than RFA. 11,14,20,21 increased hs-cTnT release was recently reported with PFA versus standard RFA (25-50 W) 22 and PFA versus HPSD-50W. 23Our study is in line with these findings, while it additionally provides novel evidence that a significant increase in CK and CK-MB activity is also characteristic of PFA and that biomarkers of myocardial injury after PFA are regressive on the second day after ablation, which is suggestive of a similar time course of tissue injury as RFA.As opposed to the previous study by Badertscher et al., 23 we found that PFA is actually equivalent to vHPSD in terms of hs-cTnT release.This may be due to greater lesion volumes and improved lesion efficacy with HPSD-70W and HPSD-90W as compared to HPSD-50W.
However, the highly distinct lesion formation mechanisms of the two ablation modalities should also be considered when interpreting the current results, since they may impact biomarker release kinetics.
While RFA induces immediate cardiac tissue necrosis by Joule heating, cell death following irreversible electroporation is multifaceted and likely occurs over a time course of several hours. 12,24A delayed release of cardiac proteins into the circulation after PFA could defer biomarker peak levels and thus account for higher levels measured the day after ablation.Furthermore, the preserved microvasculature observed in PFA lesions 14,19 may also facilitate wash-out of cardiac proteins into the circulation, whereas this could be impaired in the necrotic core of RFA lesions due to vascular disruption.Thermal inactivation of CK in the tissue has been previously demonstrated after RFA 25 and may account for the merely slight CK and CK-MB activity increase usually observed with this technology.Note: Odds ratio is expressed per 100 unit change for hs-cTnT release and per 1 unit change for all other variables.
during PFA represents another putative mechanism for the significant CK and CK-MB release.However, the CK-MB/CK ratio postablation was not different with RFA and PFA in our study, rendering this mechanism unlikely.

| PFA is associated with reduced inflammation
3][4] The extent of inflammation following PFA has not been described so far in the clinical setting.
The current study found postprocedural WBC elevation to be reduced by 60% after PFA despite higher levels of myocardial injury biomarkers.Importantly, while a positive correlation between myocardial injury and inflammation was found with standard RFA, as previously described, 3 there was no such correlation with PFA.
Indeed, previous preclinical studies have shown reduced immune cell infiltrates in PFA lesions as compared to RFA lesions on histopathologic analysis. 13,14It is conceivable that selective electroporation of cardiomyocytes generates a reduced proinflammatory cytokine release in the tissue which then triggers less systemic mobilization of WBC into the circulation than coagulative necrosis induced by RF energy.The current results thus suggest that the reduced tissue inflammation seen with PFA in animal models also translates to a reduced systemic inflammatory reaction in the clinical setting.In agreement with this, PFA is associated with 20% less tissue edema than thermal ablation on MRI. 19Further studies using more specific biomarkers of inflammation are warranted to confirm these findings.
4.3 | vHPSD is associated with a higher hs-cTnT release than standard RFA When comparing the three RF technologies, our results are suggestive of an improved lesion efficacy of HPSD-70W and HPSD-90W, since they generated a significantly higher hs-cTnT release than standard RFA.Particularly, HPSD-70W showed the best correlation with RF duration, which is well in accordance with the highest first-pass isolation rate seen in this group and, therefore, suggestive of an optimized ablation effectiveness.HPSD is reported to create larger and shallower lesions, while reducing reversible tissue injury by limiting the conductive heating phase. 5,28The level of inflammation was not different between vHPSD and standard RFA, which is not surprising considering the same mechanism of lesion formation by coagulative necrosis.Interestingly, hs-cTnT levels did not correlate with WBC in the vHPSD groups.This finding might be related to the marked reduction in conductive heating typical of vHPSD 5,28 which could account for reduced thermal injury and associated inflammation of adjacent tissue (pericardium, lung, esophagus).

| Clinical implications
Both PFA and vHPSD (70 and 90 W) provide an improved procedural efficiency than standard RFA by generating a higher hs-cTnT release during a significantly reduced procedure time with similar safety and midterm rhythm outcomes.Particularly, the greatest reduction in procedure duration (−42%) and LA dwell time (−50%) was observed with PFA.Despite generating the highest myocardial injury, PFA is actually associated with a reduced inflammatory reaction.This is reassuring and warrants further studies to define whether this finding also translates to less early arrhythmia recurrences and inflammationrelated symptoms in the first weeks following ablation.The slightly higher rates of groin complications for HPSD-90W and PFA in our study are likely related to an initial learning curve of using sheaths with a larger diameter (VIZIGO and FaraDrive) than the Agilis sheath, which had been otherwise used in our routine clinical practice.
Concerning the predictive value of myocardial injury and inflammation biomarkers for rhythm outcomes, we found that hs-cTnT release is associated with a decreased risk of arrhythmia recurrence in the standard RFA group.This is in agreement with previous studies on RFA 29,30 and suggests that a more extensive ablation is beneficial for rhythm outcomes, probably due to more durable lesions preventing PV reconnections.While hs-cTnT release is significantly higher with PFA and vHPSD, it is not predictive of rhythm outcome within these groups.This is likely due to an improved lesion homogeneity and efficacy seen with these optimized technologies, as opposed to standard RFA, where lesion quality might be more heterogenous and depend to a greater extent on different factors such as catheter-tissue contact, catheter stability, and operator experience.Therefore, postablation hs-cTnT release may be a good surrogate parameter for efficient PVI only for ablation settings generating a greater variability of lesion quality.With optimized technologies rendering ablation more homogeneous and efficient, it loses its sensitivity within the groups.

| LIMITATIONS
This was a retrospective analysis with the inherent limitations of this study design, including a potential selection bias.Nevertheless, patients were consecutively included in each group based on the same criteria as soon as each new technology became available in our center, while baseline characteristics remained well-balanced between groups.Data was available for biomarkers routinely assessed at our center in clinical practice.However, more specific markers of inflammation are required to confirm the current findings.
While the primary aim of this study was to quantify differences in myocardial injury and inflammation, it was not powered to detect differences in procedural safety and outcomes.Finally, the results may not be transferred to different catheter tip designs, HPSD, and PFA protocols.
POPA ET AL.
| 325 PFA is associated with a higher myocardial injury and a lower level of systemic inflammation as compared to RFA.Whether these findings are reflective of improved lesion efficacy and associated with improved long-term rhythm outcomes remains to be addressed in future studies.

2 | METHODS 2 . 1 |
Study designWe retrospectively included 179 consecutive patients with symptomatic paroxysmal AF who received first-time PVI at our center between 11/2020 and 10/2022.Four groups were defined according to the ablation modality and power/duration settings employed: standard RFA with 30-40 W (n = 52), HPSD-70W (n = 60), HPSD-90W (n = 32), and PFA (n = 35).The choice of the ablation modality was based on technology availability and/or the operator's preference.Patients requiring any ablation in addition to PVI or a cardioversion were excluded.Clinical, procedural, and laboratory data were collected prospectively in a dedicated computerized database.

3 . 1 |
Continuous variables are expressed as mean ± standard deviation and categorical variables are presented as frequencies or percentages.Baseline characteristics, procedural data, and outcome data were compared using ordinary one-way analysis of variance (ANOVA) for continuous variables and the χ 2 test for categorical variables.Multiple comparisons of procedural data and biomarker values in the four groups (standard RFA, HPSD-70W, HPSD-90W, PFA) were performed by comparing the mean of each group with the mean of every other group using ordinary one-way ANOVA with Tukey's multiple comparisons test.Correlation between biomarker values was assessed using Pearson correlation coefficient.The predictive value of biomarkers for atrial arrhythmia recurrence was assessed using simple logistic regression for each individual group.Statistical tests and confidence intervals with two-tailed p < .05were considered statistically significant.Statistical analysis was performed using the GraphPad Prism version 9.5.1 for Windows (GraphPad software).POPA ET AL. | 319 Patient population and procedural characteristics The study population consisted of 179 patients with paroxysmal AF (mean age 63.1 ± 10.3 years, 110 [61.5%] male, mean body mass index 26.6 ± 4.3 kg/m 2 , mean left ventricular ejection fraction 58.5 ± 5.0%, and median CHA 2 DS 2 VASC Score 2 [interquartile range 1; 3]).Baseline characteristics were well-balanced between the four groups (Table1

| 321 1 . 4 -
fold increased hs-cTnT release, a 3.4-fold increased CK release, a 5.8fold increased CK-MB release, and a 2.5-fold decreased WBC difference as compared to standard RFA.Postprocedural CK-MB/CK ratio did not differ between groups.Biomarker dynamics are shown in Figure 3.The highest levels of all studied biomarkers were measured on Day 1 postablation and were regressive on Day 2 in all groups.

F I G U R E 2
Biomarkers of myocardial injury and inflammation 1 day after ablation.Individual values for hs-cTnT release, CK release, CK-MB release, and WBC difference are shown as scatter plots.Bars represent mean ± standard deviation.*p < .05,**p < .01,***p < .001.CK, creatine kinase; CK-MB, creatine kinase MB isoform; hs-cTnT, high-sensitivity cardiac troponin T; WBC, white blood cell.F I G U R E 3 Dynamics of myocardial injury and inflammation after ablation.Mean values of hs-cTnT levels, CK activity, CK-MB activity, and WBC count are shown at baseline, Days 1 and 2 after ablation.Biomarker values on Day 2 after ablation were available in 37% of patients.Bars represent mean ± standard deviation.CK, creatine kinase; CK-MB, creatine kinase MB isoform; hs-cTnT, high-sensitivity cardiac troponin T; WBC, white blood cell.The current study compared the extent of myocardial injury and inflammation following PVI with PFA and RFA using established biomarkers.The main findings of this study are (1) PFA is associated with the highest myocardial injury and with the lowest systemic inflammation as compared to standard RFA, HPSD-70W, and HPSD-90W.(2) HPSD-70W and HPSD-90W are associated with a higher hs-cTnT release than standard RFA, while inflammation levels are similar.(3) Hs-cTnT release is associated with a reduced risk of arrhythmia recurrence in the standard RFA group, while it is not predictive of rhythm outcomes in the other groups.(4) Biomarker dynamics after PFA and RFA are similar, with a peak on the first day postablation.

F I G U R E 4
Correlation between RF duration, myocardial injury, and inflammation.(A) Correlation between hs-cTnT and RF duration.(B) Correlation between hs-cTnT and WBC difference.Pearson correlation coefficient (r) and p value are shown for each group.hs-cTnT, high-sensitivity cardiac troponin T; RF, radiofrequency; WBC, white blood cell.T A B L E 3 Outcomes.

Table 2 .
T A B L E 1 Baseline and procedural characteristics.
4,26,27Finally, skeletal muscle engagement occurring T A B L E 4 Univariate analysis of biomarkers and arrhythmia recurrence at 6 months.