Impact of pulsed field ablation on intraluminal esophageal temperature

Atrio‐esophageal fistula after esophageal thermal injury (ETI) is one of the most devastating complications of available energy sources for atrial fibrillation (AF) ablation. Pulsed field ablation (PFA) uses electroporation as a new energy source for catheter ablation with promising periprocedural safety advantages over existing methods due to its unique myocardial tissue sensitivity. In preclinical animal studies, a dose‐dependent esophageal temperature rise has been reported. In the TESO‐PFA registry intraluminal esophageal temperature (TESO) changes in a clinical setting are evaluated.


Conclusion:
A small but significant intraluminal esophageal temperature rise can be observed in most patients during PFA.TESO rise over 40°C is rare.The clinical implications of the observed findings need to be further evaluated.

K E Y W O R D S
atrial fibrillation, esophageal temperature, esophageal thermal injury, pulsed field ablation

| INTRODUCTION
Atrial fibrillation (AF) is the most common arrhythmia.Catheter ablation (CA) is an established, guideline directed treatment for symptomatic AF. 1 Historically, thermal ablation by cryothermal, laser or radiofrequency (RF) energy has been used to achieve pulmonary vein isolation (PVI) and effective rhythm control.The risk of esophageal thermal injury (ETI) that may lead to the life-threatening complication of atrio-esophageal fistula is a major drawback of conventional thermal ablation energy sources. 2 With pulsed field ablation (PFA), irreversible electroporation has recently been introduced as a novel form for ablation, rethinking our established energy sources.In RF ablation, low voltage currents are continuously applied over seconds, whereas in PFA several hundreds of volts or tens of kilovolts are repeatedly applied over an ultra-short time of micro to nanoseconds.Instead of unselective penetration through all types of tissue, PFA is designed to induce tissue-specific irreversible electroporation in the cell membrane of cardiomyocytes.
Noncardiac tissue such as nerves, blood vessels, blood cells and the esophagus are not affected, which promises safety advantages over existing methods. 3][6][7][8][9][10][11][12] Therefore, catheter design and technical aspects of PFA technology have been optimized in the past to avoid collateral damage of noncardiac structures like the esophagus. 13rrently, there are only limited clinical data on a small number of patients that assessed intraluminal esophageal temperature (TESO) changes during PFA in humans. 14Therefore, we sought to investigate intraluminal TESO changes in patients undergoing AF ablation with PFA.

| Patients
Between August 2021 and May 2022, consecutive patients with symptomatic AF who were treated with CA using the novel PFA system (Boston Scientific, Farawave) were prospectively enrolled.
The registry was approved by the local ethics committee (Lübeck ablation registry ethical review board number: WF-028/15) and all participants provided written informed consent.All investigation were performed in compliance with the ethical standards laid down in the 1964 Declaration of Helsinki and its later amendments.

| Periprocedural management
All patients underwent preablation investigation as of the center's standard of care.In brief, intracardiac thrombi were ruled out using transesophageal echocardiography.In patients on vitamin K antagonists, the procedure was performed under therapeutic INR values of 2−3.In patients on direct oral anticoagulants (DOAC) the morning dose was omitted on the procedure day.All ablations were performed under analgosedation using propofol, midazolam and fentanyl following the recommendations of a position paper by the German Cardiac Society. 15trasound guided femoral vein punctures were performed (via 1−2 × 8 F sheaths).One diagnostic catheter was introduced and positioned in the coronary sinus if two punctures were used.A single transseptal puncture (TSP) was performed under fluoroscopic guidance using a modified Brockenbrough technique.After TSP, heparin boluses were administered targeting an activated clotting time of >300 s.A SL1 sheath was used for antegrade LA access.
Before first pulse delivery, 1 mg atropine was given intravenously to avoid vagal reaction such as sinus arrest or intermittent atrioventricular block.

| Ablation protocol
Detailed description of the optimized PFA procedure has been described before. 13In brief, eight pulse trains (2 kV/2.5 s, bipolar, biphasic, x4 basket/flower configuration each) were delivered to each PV starting on the left-sided veins.Two extra pulse trains per PV were applied in a flower configuration and more antral angulation were added for wide antral circumferential ablation (WACA) as per the center's specific protocol. 16The acute procedural endpoint was first-attempt-all-veins-isolated (FAAVI). 17

| Esophageal temperature monitoring
Real-time continuous intraluminal esophageal temperature changes were closely monitored with a copper plated s-shaped 12-pole flexible and self-expanding temperature probe (CIRCA S-CATH™, CIRCA Scientific, Inc.).The proper positioning of the probe is illustrated in Figure 1 and was correct when the entire LA was covered by the 12 poles as confirmed by fluoroscopy.Intraluminal esophageal temperatures were measured during the entire ablation procedure to assess the maximal TESO increase.In addition, TESO was measured at baseline and at the end of each-sided PV ablation as well as during WACA.Esophagoscopy was performed in case of symptoms suggestive of esophageal irritation, including pain during swallowing or ongoing chest discomfort unresponsive to mild analgetics, or a TESO cut off > 41°C. 18 1.    3.

| Procedural characteristics
When analyzed with respect to the position of the temperature probe/course of the esophagus the following result were found: There was no difference in mean ΔTESO for a left (L), right (R) or middle (M) sided temperature probe position which were 0.53 ± 0.39, 0.48 ± 0.34, 0.44 ± 0.35°C; p = .758,respectively.
When ablating on the site of the probe (on-probe) in comparison to ablation remote to the probe (remote-probe), a significant different temperature rise of 0.49 ± 0.36 versus 0.24 ± 0.28°C;    naturally with a delay and plateau phase, cannot be stopped once the energy has been applied and potentially results in collateral damage to any kind of tissue in close proximity to the ablation site.This can have devastating effects to the esophagus resulting in a lifethreatening complication called atrio-esophageal fistula which is associated with a high mortality up to 50%. 19rrent PFA systems are not equipped with a feedback loop for temperature monitoring or contact measurement.Further research is needed to evaluate the potential of thermal injury from a PFA system.
As investigated in mathematical simulations a clinically relevant thermal effect from PFA technology is possible [4][5][6] .Focusing on a specific technical setting, a biological significant temperature rise > 50°C was predicted in 5% of cases. 20The highest TESO measured in our cohort was 40.3°C lasting over only a few seconds.Regarding the aforementioned aspects, this should not have led to biological relevant thermal effects to the esophageal tissue.However, data about the impact of TESO rise is based on RF ablation and can only be extrapolated on PFA technology.Further studies are needed to evaluate the impact of PFA on esophageal injury.
The physics and technical aspects behind PFA are complex.
Various parameters like electrode size, electrode spacing, electric field strength, duration and repetition rate of the pulse train, organ blood flow, tissue histoarchitecture and even fiber orientation influence the efficiency of irreversible electroporation. 20,21th respect to our observation, the manufacturer specific settings of the only currently available CE-proven PFA catheter with the most preclinical and clinical evaluation, seem to be well balanced to achieve fast effective acute and chronic electrical PVI on the one They reported a maximum TESO of 36.14 ± 0.34°C, with a mean change of 0.06°C from baseline which is in contrast to our findings. 14 line with our data, is the fact, that no patients demonstrated any symptoms of ETI during follow-up.These data however are limited by the small number of investigated patients (n = 7) and no information about the technology used for temperature monitoring.
As with any physical effect in nature it looks like that with PFA the question of which "dose makes the poison" has now been solved in context to AF ablation and ETI.Our data prove, that with the current PFA setting, objectively significant thermal effects from the electric field and Joule heating are present.We therefore do not recommend to use "non-thermal" in context with PFA either.
However, they are technically reduced to a clinically minimal amount with almost no temperature overshoot from conductive heating in comparison to conventional ablation energy technologies.This is a major advantage over conventional energy sources for AF ablation.

| Preclinical and clinical data
In early preclinical experiments, thermal energy release from electroporation was investigated and a dose dependent temperature rise was observed when different modalities were used. 4,6,9,20These observations led to a specific PFA catheter engineering and design to balance between effective IRE and biologically insignificant thermal effects.
In a first animal experiment, a 200 Joule electric pulse was directly applied to a porcine esophagus.On histology, intraepithelial vesicle formation within the esophageal adventitia was observed with complete normalization after 1 week. 7Hong et al. showed PFA to minimize the ETI risk, involving only the muscular layer of the esophagus, while epithelium and mucosal lamina muscularis remained without pathological changes. 23 a preclinical animal study investigating six female swine, a 3-pole esophageal temperature monitoring probe was placed behind the heart and showed no TESO changes during PFA of the LA, right atrial appendage and superior vena cava. 8,12This is in contrast to our findings and emphasizes the proper placement of the temperature probe and coverage of the ablation site.On further assessment by cross pathology or histology, no thermal lesions were seen in these animals. 12One animal study reported about an average TESO increase by 0.7 ± 0.35 up to 5.0 ± 1.83°C depending on the pulse energy used (700−1500 Volts). 9fety aspects have also already been studied in first clinical trials (IMPULSE, PEFCAT, PEFCATII, PersAFOne trial).In a moderate number of patients (n < 150), ETI evaluated by esophagoscopy was absent. 13,24Noninvasive investigation by cardiac magnetic resonance (CMR) imaging showed absence of late gadolinium enhancement (LGE) as a surrogate for ETI in comparison to thermal energy ablation technologies. 25In our study we did not see a delay or an overshoot nor a longer high temperature plateau phase as a result of conductive vascular necrosis especially to the anterior esophageal arteries as the pathophysiological prerequisite for the development of edema, scar or manifest esophageal lesions were absent. 7,12nsistent with the aforementioned findings, the multi-national MANIFEST-PF registry reported about the first clinical data from a "real world" performance of PFA in n > 1700 unselected AF patients and could not report any evidence of ETI. 22This is of clinical importance, because no predefined framework for esophageal management was applied during the procedure (esophageal temperature monitoring, mechanical deviation, cooling or energy reduction during ablation on the posterior LA wall).So far, no case of atrioesophageal fistula after a PFA procedure has been reported.This also questions the need for postablation treatment with proton pump inhibitors as there are known side-effects like a slight prolongation of the QT-interval, especially in case of co-medication with antiarrhythmic drugs.

| Limitations
The presented data reflect a single-center registry experience and encompass a small number of patients with early AF stages

| CONCLUSIONS
To the best of our knowledge, this is the first study to report about the impact of PFA on intraluminal esophageal temperature changes in a larger series of human.A small but significant intraluminal esophageal temperature rise can be observed in most patients during PFA.TESO rise > 40°C is rare.Clinically, this was not associated with any short-and midterm symptoms related to potential ETI.The clinical implication of the observed findings needs to be further evaluated.

A
vascular closure device or a figure-of-eight suture and a pressure bandage were used to prevent femoral bleeding.The pressure bandage was removed after 1−4 h depending on the closure technique used.The suture was removed the next day.Following ablation, all patients underwent transthoracic echocardiography immediately, after 2 h, and at Day 1 postablation to rule out pericardial effusion.Oral anticoagulants were reinitiated 6 h postablation and continued for at least 3 months and thereafter according to CHA 2 DS 2 -VASc-score.If not otherwise contra-indicated, antiarrhythmic drugs were prescribed for 3 months postablation to prevent early AF recurrence during the blanking period.Symptoms related to ETI were assessed 2 and 4 h postablation and daily until discharge.Proton-pump-inhibitors were orally administered twice daily until discharge and once daily for 6 weeks postablation as part of the center's standard of care.
Continuous variables were tested for normal distribution using the Shapiro-Wilk test.In case of normal distribution, data are presented as mean ± standard deviation (SD), otherwise as median and interquartile ranges (IQR): first quartile [Q1] and third quartile [Q3].Comparison of continuous data was performed using the student's t-test if normally distributed, otherwise with the Wilcoxon signed-rank test.Categorical data are reported as absolute (n) and relative frequencies (%) and were compared using the chi-square test or Fisher's exact test (small expected frequencies).Statistical significance of the observed difference was expressed in terms of p Values.Exact two-sided statistical significance was considered for a p < .05.All calculations were performed with the statistical analysis software STATA version 15 (Stata Corporation).
Patient characteristics 43 consecutive patients (62 years, 67% male, 58% CHA 2 DS 2 Vasc Score 2) with predominantly symptomatic paroxysmal AF (61%), a preserved left ventricular ejection fraction and low burden of comorbidities were prospectively enrolled into the registry.Prior pharmacological or electrical rhythm control using antiarrhythmic drugs or electrical cardioversion had been attempted in 49% and 35% of patients, respectively.Oral anticoagulation with a DOAC was established in the majority of patients (93%).Detailed patient baseline characteristics are shown in Table All PFA procedures were performed by two experienced EP operators with over 10 years of invasive training.During a median procedure time of 35 [26−64] min, left atrial catheter dwelling time of 16 [12−18] min and fluoroscopy time of 7 [5−11] min, the acute procedural endpoint (FAAVI) was achieved in all patients.Procedural data are depicted in Table 2 .

3. 3 |
Intraluminal esophageal temperature monitoring Intraluminal placement of the temperature probe into the esophagus was feasible in all patients.On fluoroscopy the esophagus mainly F I G U R E 1 Heart model from posterior view showing the proper positioning of the esophageal temperature probe with coverage of the entire posterior left atrial wall.Esophagus has been removed for better visualization.White arrows pointing to 6 of the 12 copperplated temperature poles.Ao, Aorta; LA, left atria; LIPV, left inferior pulmonary vein; LSPV, left superior pulmonary vein; PA, pulmonary artery; RA, right atria; RIPV: right inferior pulmonary vein; RSPV, right superior pulmonary vein; SVC, superior vena cava.coursed near the left sides PVs (20/43, 47%).A right-sided esophagus course and a coursing directly behind the LA posterior wall was observed in 6/43, 14% and 17/43, 39% patients, respectively.In Figure 2 we evaluated TESO changes when measured pulse-by-pulse in one patient with a left-sided esophagus course, showing prompt temperature increase and decrease during PFA of the left-sided PVs.The mean ΔTESO change between baseline TESO and maximum TESO was small but statistically significant and measured 0.8 ± 0.6°C, p = <.001.A TESO increase ≥ 1°C was observed in 10/43 (23%) patients.The highest TESO measured 40.3°C during WACA-PFA in a patient with the esophagus coursing direct behind the LA.The PFA catheter position in relation to the temperature probe as well as the distance between them in the patient with the largest ΔTESO of 3.7°C is given in Figure 3. Detailed information about TESO changes after PFA of each-sides veins and after WACA is shown in Table p < .001was observed.When comparing ΔTESO between baseline and maximum TESO reached in comparison to the probe position, a course right behind the LA in the middle of the ipsilateral PVs was associated with the highest temperature change of about 0.84 ± 0.88°C in comparison to 0.75 ± 0.43°C on the left side and 0.55 ± 0.37°C on the right side.However, this was not statistically significant p = .833and is limited due to the different sample sizes in each category.

3. 4 |DISCUSSIONS 4 . 1 |
Clinical and rhythm outcomeClinically, no patient reported early symptoms related to potential ETI.No esophagoscopy was indicated.On telephone follow-up 6 months postablation no patient reported suggestive symptoms or diagnosed ETI or atrio-esophageal fistula.Freedom from AF/AT recurrence after 12 months was 88% (n = 38/43).One patient received cavotricuspid isthmus ablation due to typical atrial flutter during follow-up.4| Main findings of the PFA-TESO registryIn our prospective PFA-TESO registry, we closely monitored intraluminal esophageal temperature changes in patients undergoing PFA for the treatment of symptomatic AF.A small but statistically significant TESO increase by less than 1°C was observed.A TESO rise ≥ 1 C°was found in every fourth patient.Although labeled as a non-thermal ablation energy, TESO increased by 3.7°C up to a maximal observed TESO of 40.3°C when PFA was performed in close proximity to the esophagus.In a pulse-by-pulse analysis, immediate temperature changes were observed with short-term transient thermal effects lasting only a few seconds (Figure2).Temperature overshoot with latency in temperature rise and a plateau phase, as seen with conventional thermal energy sources for AF ablation were absent.T A B L E 1 Baseline patient characteristics (N = 43).

4. 2 |
Physics behind PFAThe application of an electric field to biological tissue naturally results in Joule heating with a certain amount of temperature rise.The extent of the thermal effects and the potential tissue injury are doseand time-dependent.Whereas temperatures over 40°C led to thermal injury after several hours, the same effect can be observed after only a few seconds at temperatures greater than 50°C.With conventional AF ablation energy sources temperature overshoot results in increased conductivity and further conductive heating throughout the periphery.The temperature overshoot which comes F I G U R E 2 Diagram showing the time course and change of the intraluminal esophageal temperature (TESO) by 0.7°C during pulsed field pulmonary veins isolation with wide antral circumferential ablation (WACA) in a patient with an esophagus coursing behind the left pulmonary veins.LIPV, left inferior pulmonary vein; LSPV, left superior pulmonary vein; RIPV, right inferior pulmonary vein.RSPV, right superior pulmonary vein.F I G U R E 3 Intraprocedural fluoroscopy showing the pulsed field catheter position in relation to the intraluminal esophageal temperature probe as well as the distance between them in a patient with an esophagus coursing in the middle of the ipsilateral pulmonary veins experiencing the largest ΔTESO of 3.7°C during pulmonary veins isolation with wide antral circumferential ablation (WACA).LIPV, left inferior pulmonary vein; LSPV, left superior pulmonary vein; RIPV, right inferior pulmonary vein; RSPV, right superior pulmonary vein.
hand and clinically negligible thermal effects in comparison to conventional ablation techniques on the other hand. 22This is of utmost importance as several other PFA catheter technologies and designs are under development as well as extension of PFA to non-PV sides such as the posterior LA wall.The non-randomized, prospective, multicenter, global, premarket clinical PULSED AF study investigated a novel not yet market released spiral PFA catheter.
and less further co-morbidities.Esophageal injury might have been present but did not result in clinical symptoms or complications.Thus, routine invasive or noninvasive investigation of the esophagus by endoscopy or LGE-CMR was not performed.Data are only applicable to the vendor-specific PFA catheter design and technology.Potential bias might result from using a specific non-PFA optimized temperature probe.A temperature measurement that respects the specific pulse configuration with high voltage over a short duration of the electrical energy would have required temperature sensors with a very high response time and sampling rate that is twice the maximal rate of change of the temperature.The CIRCA temperature probe delivers 240 data points per second; 12 thermocouple temperature sensors update 20 times per second.While these specifications are appropriate for RF energy, they might have under recorded the peak rapid rise and decay of the temperature change generated by PFA.Flexible monolayer molybdenum disulfide (MoS2) temperature sensors and arrays can detect temperature changes within a few microseconds over 100-times faster than flexible thin-film metal sensors but were not available for investigation in this study.
heating like with RF energy.This should not have promoted thermal damage to deeper noncardiac tissues that translates into remote edema or scar detectable by current CMR methods.Furthermore, the aforementioned histological studies found scar only several millimeters from the catheter which was cardiomyocyte specific.Thermal T A B L E 3