Posterior wall ablation characteristics and esophageal safety of an optimized protocol for voltage-guided, high-power short--duration ablation of atrial fibrillation

Introduction: The ablation index (AI) standardizes lesions in radiofreqency ablation (RF) of atrial fibrillation (AF). High-power, short-duration ablation (HPSD) shortens procedures, but AI is histologically unvalidated in HPSD. We evaluated esophageal safety with voltage-adapted AF ablation using an optimized, AI-guided HPSD (AI-HPSD) protocol. Methods: Consecutive AF patients undergoing AI-HPSD at 50W (AI=400-450 for posterior left atrium (LA); AI ≥500 anteriorly) were compared with a recent AF cohort who underwent AI-guided low-power, long-duration ablation (AI-LPLD). All participated in our ablation registry. Esophageal endoscopy was performed 1-3 days post-ablation. Posterior wall lesion characteristics and endoscopically-detected esophageal lesions (EDEL) were compared. Results: AI-LPLD (n=100) and AI-HPSD (n=100) groups had similar baselines, except AI-LPLD had slightly larger LA-area. VISITAG numbers were comparable, but total posterior wall RF time was shorter with AI-HPSD versus AI-LPLD (4.5±1.6 versus 11.9±5.1 s, p<0.001). Mean AI and mean impedance drop were higher in AI-HPSD (449±17 versus 401±39; p<0.001 and 8.8±2.3 versus 7.8±2.5; p<0.001). AI standard deviation was lower with AI-HPSD (30±15 versus 49±19; p<0.001). Although Max AI values were higher in AI-HPSD (517±46 versus 496±55; p=0.003), maximum impedance drop did not differ significantly. EDEL rate was comparably low at 2% (AI-HPSD) and 4% (AI-LPLD), but Max AI >520 occurred more with AI-HPSD (3.5±5.5 versus 2.5±5.2; p=0.048). Conclusion: In routine practice, optimized AI-HPSD at 50W with posterior LA target AI =400-450 was not associated with more EDEL. However, a signal for more unacceptable Max AI values existed. We recommend limiting AI to 400 and developing an automatic, AI-guided ablation stop. Posterior wall ablation characteristics and esophageal safety of an optimized protocol for voltage-guided, high-power short–duration ablation of atrial fibrillation Lisa C. Costello-Boerrigter 1, Frank Steinborn 2, Mykhaylo Chapran 2, Kourosh Vathie 2, Nemanja Milisavljevic 3, Mohamad Assani 2, Ralf Surber 4, Christina Grieger 2, Jens Martin Kittner 3, Anja Schade 2,5 * 1 Department of Cardiology, Heart Center, Zentralklinik Bad Berka and Clinical Research, Rhön Klinikum AG; Robert-Koch-Allee 9, 99437 Bad Berka, Germany; [email protected] 2 Department of Cardiology/Interventional Electrophysiology, Helios Hospital Erfurt, Universitärer Campus der Health and Medical University, Nordhäuser Str. 74, 99089 Erfurt, Germany 3 Department of Internal Medicine 2, Helios Hospital Erfurt, Nordhäuser Str. 74, 99089 Erfurt, Germany 4 Department of Internal Medicine 1 /Cardiology, Jena University Hospital, Am Klinikum 1, 07747 Jena, Germany 5 Clinic for Rhythmology and Interventional Electrophysiology, Rhön Campus Bad Neustadt, Von-Guttenberg-Str. 11, 97616 Bad Neustadt/ Saale, Germany; [email protected] * Corresponding author: PD Dr. med. Anja Schade Chief of Rhythmology and Interventional Electrophysiology Heart Center Rhön-Klinikum – Campus Bad Neustadt Von-Guttenberg-Str. 11 97616 Bad Neustadt a. d. Saale Germany E-mail: [email protected] Phone: +49 (0) 9771 66-23250 Fax: +49 (0) 9771 66-98-23250

The ablation index (AI) standardizes lesions in radiofreqency ablation (RF) of atrial fibrillation (AF). High-power, short-duration ablation (HPSD) shortens procedures, but AI is histologically unvalidated in HPSD. We evaluated esophageal safety with voltage-adapted AF ablation using an optimized, AI-guided HPSD (AI-HPSD) protocol.

Consecutive AF patients undergoing AI-HPSD at 50W (AI=400-450 for posterior left atrium (LA); AI ≥500 anteriorly) were compared with a recent AF cohort who underwent AI-guided low-power, long-duration ablation (AI-LPLD). All participated in our ablation registry. Esophageal endoscopy was performed 1-3 days post-ablation. Posterior wall lesion characteristics and endoscopically-detected esophageal lesions (EDEL) were compared.

AI-LPLD (n=100) and AI-HPSD (n=100) groups had similar baselines, except AI-LPLD had slightly larger LA-area. VISITAG numbers were comparable, but total posterior wall RF time was shorter with AI-HPSD versus AI-LPLD (4.5±1.6 versus 11.9±5.1 s, p<0.001). Mean AI and mean impedance drop were higher in AI-HPSD (449±17 versus 401±39; p<0.001 and 8.8±2.3 versus 7.8±2.5; p<0.001). AI standard deviation was lower with AI-HPSD (30±15 versus 49±19; p<0.001). Although Max AI values were higher in AI-HPSD (517±46 versus 496±55; p=0.003), maximum impedance drop did not differ significantly. EDEL rate was comparably low at 2% (AI-HPSD) and 4% (AI-LPLD), but Max AI p=0.048).

In routine practice, optimized AI-HPSD at 50W with posterior LA target AI =400-450 was not associated with more EDEL. However, a signal for more unacceptable Max AI values existed. We recommend limiting AI to 400 and developing an automatic, AI-guided ablation stop. Conflicts of Interest: None declared Acknowledgments and Support: This work was completed without the aid of external funding. Likewise, no drugs or devices were supplied as means of support. Data Availability Statement: The data underlying this article will be shared on reasonable request to the corresponding author. Key words: esophageal injury, ablation index, high-power short-duration ablation, voltage adapted ablation

Antral pulmonary vein isolation (PVI) is an effective treatment for atrial fibrillation (AF). Given the anatomical location of the posterior wall of the left atrium (LA) relative to the anterior esophagus, iatrogenic esophageal thermal injury is possible. Rarely, the esophageal wall develops a fistula towards the LA. (1) An atrioesophageal fistula (AEF) is a potentially fatal AF ablation complication. Higher grades of endoscopically-detected esophageal lesions (EDEL) seen on post-ablation endoscopy are significant predictors of AEF development. (2,3) Recently, the ERASE study confirmed the additional benefit of a voltage adapted ablation concept beyond PVI only. (4) However, more extensive ablation concepts could pose more risks. The ablation index (AI) can be used to create homogeneous lesions. AI is calculated using a weighted formula that includes contact force, duration of ablation, and power and predicts the lesion size induced by RF ablation. (5). Lesions produced by various combinations of power, contact force, and ablation duration but reaching the same AI have the same lesion size. Recently, we reported a comparison of conventional contact force-guided ablation versus AI-guided ablation. Use of AI to guide lesion application, resulted in more homogeneous energy application to the LA posterior wall; however, the number of EDEL seen with the two methods did not differ. Nevertheless, maximum AI values above 520 were predictors of EDEL, including category 2 EDEL. Conversely, none of the patients with maximum AI <450 developed category 2 EDEL.(6) Das and Phlips found higher reconnection rates posteriorly with an AI our findings with those of Phlips, we defined an AI target range of 400 to 450 for routine use in our center. High-power short-duration (HPSD) ablation has improved lesion application so that there is less collateral tissue damage and edema. (8,9) However, AI has not been extensively or histologically validated for HPSD ablation. In this study, we aimed to compare lesion application characteristics and esophageal safety of AI-guided HPSD (AI-HPSD) using 50W and the new posterior AI target range of 400 to 450 with conventional, AI guided low power long duration ablation (AI-LPLD) in the context of voltage adapted AF ablation as part of everyday clinical practice.

Patient selection All patients were participants in the Erfurt Prospective AF Ablation Registry and provided written informed consent. This study was approved by the Ethics Committee of the State Medical Association of the Federal State of Thuringia, and enrolled patients with paroxysmal, persistent, and long-term persistent AF. PVI using CARTO 3D with additional voltage-guided ablation was used in all patients with persistent AF (i.e. persistent and long-term persistent considered together) and in paroxysmal AF cases with additional regular atrial tachycardias (AT) or structural heart disease. Clinical and procedural data from 100 consecutive patients undergoing their first AF ablation with 50 W AI-HPSD were compared with 100 patients in whom AI-LPLD ablation had been used for a first AF ablation procedure. The AI-LPLD patients were enrolled consecutively between June 2019 and June 2020. They have been previously described by us and are serving as a control. (5) The AI-HPSD patients were enrolled from December 2021 to August 2022. A THERMOCOOL SMARTTOUCH® SF catheter (Johnson & Johnson, New Brunswick, NJ, USA) was used for all ablations. Ablation technique After LA thrombus exclusion by transesophageal echocardiography, the procedure was performed under deep sedation with propofol and sufentanil and with intravenous unfractionated heparin to keep ACT >300. Using fluoroscopic guidance, two sheathes were inserted into the LA via a single transseptal puncture. As previously described (5), anatomical and high-resolution voltage mapping of the LA was performed using LASSO TM Nav (Johnson & Johnson, New Brunswick, NJ, USA) under continuous pacing from a coronary sinus catheter (Webster CS, Johnson & Johnson New Brunswick, NJ, USA). Rapid anatomical mapping was done with an LA geometry interpolation threshold of 20. Circumferential PVI lesions were applied at the antral level of the PV ostium starting with encircling on the left side and then proceeding to the right after achieving bidiectional conduction block. For the AI-LPLD group, the PVI endpoint was bidirectional conduction block after a 15-minute waiting period. For the AI-HPSD group, bidirectional block was proven after both circlings were finished and an adenosine bolus given as a challenge for the right PVs. In patients with significant low voltage zones (LVZs), i.e. LVZ applied to encircle the LVZ or to draw a voltage adapted line to avoid critical isthmus sites for typical macroreentries as described previously. (10) Typical voltage maps with corresponding ablation sets are shown in Figure 1. Subsequent inducible atrial tachycardias (AT) were mapped and ablated. Radiofrequency ablation settings All lines were applied using point-by-point RF ablation with an interlesion distance of <6 mm. In the AI-LPLD group, AI targets were ≥500 anteriorly and ≥350 or 400 posteriorly, depending upon the operator. As previously described, our operator-dependent AI targets for AI-LPLD ablation were derived from the blinded AI values in contact force-guided procedures of each operator. Targets were defined here for the AI-LPLD group as this mean blinded AI value minus 10. (6) For AI-HPSD patients, an AI target range of 400 to 450 was used for posterior ablation. When gaps between the visitags inadvertently occurred, gap closure at posterior wall was delayed by a few minutes to avoid heat accumulation in the AI-HPSD group. In the AI-LPLD group, a power of 30–35 watts (flow rate 15 ml/min) was used for anterior wall RF ablation and 20–25 watt (8 ml/min flow rate) for the posterior wall. In the AI-HPSD group, a power of 50 W (flow rate 20 ml/min, and 3 seconds of pre-flushing) was applied independent of ablation lesion location. The targeted contact force ranges were 5 - 40 g for AI-LPLD and 5 - 30 g for AI-HPSD. The CARTO 3 VISITAG™ module was used for the automated display of RF applications. VISITAG™ parameters were tag size of 3 mm, minimum time of 3 s, force over time of 25%, minimum force of 3 g. Location stability was set on 2.5 mm for the AI-LPLD group and 5 mm for the AI-HPSD group. VISITAG colouring thresholds for posterior ablation were set on 250/350 in the AI-LPLD group and 250/380 in the AI-HPSD group. Esophageal protection No esophageal temperature probe was used. All patients received pantoprazole 40 mg twice daily for eight weeks. Upper endoscopy and evaluation of EDEL Esophageal endoscopy was performed 1–3 days post-ablation. All abnormalities were documented. However, only lesions at the esophageal-atrial contact region were considered thermal EDELs. As previously described, ulcers ≤5 mm in diameter and erythema or erosions were classified as Category 1 lesions, and ulcers >5 mm in diameter were classified as Category 2 lesions. (11) (Figure 2) Evidence of hemorrhage or fibrinous coverage were likewise documented. Patients with EDEL were treated according to an institutional protocol and followed endoscopically until esophageal injuries were in remission as described earlier in detail (5). In brief, all patients with ulcers received a liquid diet and were monitored for clinical and laboratory signs of infection. A CT scan was performed when there were signs of infection. Repeat endoscopy was done 3-4 days later and until healing was obvious. Follow up Patients were to obtain emergent medical attention for signs or symptoms of esophageal injury, inflammation, fever, or stroke. Patients had 72-hour Holter monitoring three and 12 months post-procedure. At the same time points, other complications were documented. Antiarrhythmic medications were stopped 8-10 weeks post-ablation. Ablation success was defined as freedom from a documented episode of AT or AF of > 30 s, after a three-month blanking period. Post-procedural analysis of posterior wall ablation data Ablation characteristics of the LA posterior wall were analysed after procedure completion. The VISITAGS TM of both groups were examined again and labelled as posterior or non-posterior points. Contact force, AI, impedance drop, and ablation duration were then determined for all posterior wall VISITAGS TM . Statistical analysis Windows Stata/IC 16.1 was used for statistical analyses. A two-sided P<0.05 was considered significant. Normally distributed metric parameters were presented as mean ± standard deviation. For non-normally distributed metric parameters, the median, maximum, and minimum values were determined. Absolute and relative frequencies were calculated for categorical parameters. For comparison of groups, the Mann–Whitney U test was used for metric parameters, and the Fisher’s exact test was used for categorical parameters.

Baseline parameters The AI-LPLD group patients were 56% male with a mean age of 68.2+8.9 years. The AI-HPSD patients were 53% male with a mean age of 69.3 ±8.1 years. In both groups, 82% of patients had persistent AF. Detailed baseline characteristics are in Table 1. The AI-LPLD group had a slightly larger LA area, but there were no other significant baseline differences. Table 1: Baseline parameters Procedural characteristics PVI was performed in all patients. Additional lesions beyond PVI were needed in 33% of AI-LPLD patients versus 28% of AI-HPSD (p= 0.44). Procedure duration, fluoroscopy time, and dose area product were significantly lower in the AI-HPSD group, as shown in Table 2. Table 2: Procedural parameters LA left atrium; PVI pulmonary vein isolation Posterior wall ablation characteristics RF time per lesion and total RF time at the posterior wall were both significantly shorter in the AI-HPSD group than in the AI-LPLD group (p < 0.001 and p <0.001, respectively). Mean AI and mean impedance drop were significantly higher in the AI-HPSD group with the new AI target range (both with p <0.001). However, the AI-HPSD group had a significantly lower standard deviation for AI (p group (p=0.003), but this did not result in higher maximum impedance drop (p=0.187). Total AI was comparable between the groups. All relevant data are summarized in Table 3. Table 3: Posterior wall ablation characteristics (AI ablation index; SD standard deviation; Max maximum, Min minimum; RF radiofrequency) Complications Major procedural and peri-procedural complications occurred in 2% of AI-LPLD group and 3% of the AI-HPSD group (p=1.000). Specifically, in the AI-LPLD group one patient experienced a transitory ischemic attack and one developed a pseudoaneurysm needing surgery. In the AI-HPSD group there was one stroke, one tamponade, and one pseudoaneurysm requiring surgery. Minor complications occurred in 1% of the AI-LPLD group and 2% of the AI-HPSD group (p=1.000). Esophageal lesions Post-procedural EDEL occurred in 4% of the AI-LPLD patients versus 2% of the AI-HPSD (p=0.683). No AEF developed in any patient. All lesions were healed at the time of final endoscopy. Table 4 provides EDEL data for both groups. Table 4: EDEL in AI-LPLD versus AI-HPSD Characteristics of patients with AI-HPSD ablation and esophageal lesions Tables 5 A-C compare characteristics of the two AI-HPSD patients with EDEL to those of the AI-HPSD patients without EDEL. Patient 1 had a higher rate of intraprocedural posterior lesions with an AI>520. Patient 2, the only patient with a category 2 lesion, was older, had a larger LA, and had more extensive posterior wall ablation than the unaffected AI-HPSD patients. These differences were all outside the standard deviation of the group characteristics. Of note, in both of these patients maximum AI values >450 were identified. Table 5. Characteristics of the two AI-HPSD patients with EDEL and those AI-HPSD patients without EDEL. A) Baseline data. B) Procedure characteristics. C) Posterior ablation characteristics. For those patients with EDEL, values outside the standard deviation of the group characteristics are highlighted in red. AI ablation index; BMI body mass index; GFR glomerular filtration rate; LAESVI left atrial endsystolic volume index; LVEF left ventricular ejection fraction; Max Maximum; Min Minimum; Mean Mean value; PVI pulmonary vein isolation; RF: radiofrequency Long-term results Long-term follow-up was available for 81% of the patients in AI-LPLD group and 70% of patients in AI-HPSD group. Two patients died during the 12 month follow-up period from causes unrelated to the procedure. Freedom from recurrence of AF and AT was 80% versus 83% (p=1.000) in AI-LPLD versus AI-HPSD, and freedom of recurrence off drugs was 79% versus 78% (p=1.000).

Main results This study is, to the best of our knowledge, the first to analyze in detail posterior wall ablation characteristics and esophageal safety after AI-HPSD ablation using 50 W with an optimized posterior AI target range between 400 and 450, and to compare it with AI-LPLD ablation. We demonstrated here that our new, optimized protocol did not result in a higher EDEL rate than a standard AI-LPLD protocol. The EDEL occurrence rate was equally low at 2% in the AI-HPSD group and 4% in the AI-LPLD group, and no AER occurred in either group. Of the two AI-HPSD patients with EDEL, one had multiple clinical risk factors for EDEL and the other had many applications with a maximum AI far outside the desired target range. Interestingly, we have since noticed that use of the AI target range of 400-450 during everyday clinical practice is associated with a relevant number of applications outside of the AI target range, and we advocate for the development of ablation systems with an automatic stop upon reaching target-maximum AI. Consistent with other groups, we observed a dramatic reduction in procedure duration and RF ablation time (12,13,14,15), and consequently, the fluoroscopy time and radiation delivered were also less in the AI-HPSD group. Complication rates were comparably low in both groups. EDEL and esophageal safety in an optimized AI-HPSD protocol Computer modeling and ex vivo animal studies have shown that HPSD ablation creates not only more homogeneous and wider lesions, but also shallower lesions. (7) This might decrease the risk of esophageal injury; however, an MRI study evaluating esophageal injury found no differences in esophageal late gadolinium enhancement after PVI by HPSD versus LPLD ablation. (16) The EDEL rates reported after conventional RF ablation using LPLD protocols have varied from 1.2% to 30%. (17,18,19,20) This wide range might be secondary to some studies using esophageal temperature probes, which have been associated with higher EDEL rates. (19) In this study, we did not use an esophageal probe, and we found EDEL in 4% of AI-LPLD group patients and 2% of the AI-HPSD group patients. However, in the POWER-FAST PILOT study, the incidence was higher at 28% and 22% in 30 W and 50 W groups, respectively, and steam pops were heard in 8% of the HPSD. (5,21) Of note, the POWER-FAST PILOT monitored intraprocedural temperature via an esophageal probe (21). Sternick and colleagues insightfully noted that stainless steel heats up more easily than skin and transfers heat two orders of magnitude more than endothelium disperses it; therefore, RF inductive heating of the probe’s stainless steel thermocouple might promote esophageal ulcer formation. (22) Clinical studies have subsequently given support to this concept. (23, 24) Few publications have examined EDEL in AI-HPSD ablation performed without esophageal temperature monitoring. The EDEL rate of 2% that we found after targeting AI values of 400 to 450 is very low. Wolff et al. had a comparably low rate of 4.2% using AI-HPSD ablation, but the maximum posterior AI target was lower at 350 posteriorly and up to 400 for left and right roof and inferior lines. (25) Müller et al. also reported a low EDEL rate of 3 - 7 %, but with formation of one AEF; however, their AI protocol also had lower posterior AI target levels (300 and 350) than ours. Details regarding the maximum AI data of the patient with the AEF were not described. (26,27) Francke et al. documented a higher EDEL rate of 11.7% with an AI target of 400. (14) Unlike our study, all posterior lesions were set consecutively without time spacing. Also, that study did not report the posterior maximum AI values achieved. During AI-HPSD, we briefly delayed closure of accidentally occuring gaps between the visitags to avoid heat accumulation. This might have contributed to our lower EDEL rate. None of the other studies documented pre-flushing before starting RF application. Pre-flushing is recommended by Johnson & Johnson, but is not performed in every center. Our protocol consistently used pre-flushing of the ablation catheter. Inconsistency regarding this recommendation might contribute to variability in reported EDEL rates. Similarly, slight variations in VISITAG pre-settings such as “location stability“ and “respiration adjustment“ might affect esophageal safety. In our AI-HPSD group, location stability was set to 5 instead of 2.5 as in the AI-LPLD group. This might have contributed to the comparable safety. Careful consideration of all Visitag setting parameters is necessary when adopting the ablation targets of published “safe protocols“. Posterior ablation characteristics with the optimized AI-HPSD protocol A wide range of protocols for HPSD ablation exists. Whereas some protocols, such as the one of Baher et al., define fixed ablation time targets, others use AI or lesion size index. (15,16) Nakagawa et al. showed that achieving the same AI with different values for the power-time-contact force variables results in lesions of the same size. (28) However, AI use is not validated for > 45W applications. (29) Thus, carefully examining the use of power > 45 W at the posterior LA wall as part of AI-HPSD ablation is of clinical interest. We used 50W regardless of the LA anatomical location in our protocol and still had a low EDEL rate. We previously reported posterior lesion characteristics after AI-LPLD ablation with a target >350 to 400 posteriorly. We found large standard deviations for the AI, so that some maximum AI values of > 520 occurred. Posterior wall maximum AI values category 2 EDEL, the posterior wall maximum AI values were always 400-450 for posterior ablation in the AI-HPSD group. With that, the standard deviation for the AI values was found to be significantly lower in the AI-HPSD group. Thus, focusing on a target range led to a more homogeneous lesion application. Mean AI and mean impedance drop were higher, which is understandable due to the lower minimal AI target level of some AI-LPLD patients. Maximum AI was higher in the AI-HPSD group, but not maximum impedance drop. HPSD shortens the time per RF lesion and also the reaction time necessary to stop the ablation when the AI target is reached. Any given ablation duration results in a faster AI rise with HPSD than with LPLD ablation. On the other hand, HPSD ablation results in shallower lesions. Consequently, a higher AI might not produce the same increase in lesion depth as in LPLD and thus be safer. Experimental and histological validation of AI-HPSD ablation is still needed. In light of the RESCUE-AF Trial, strict avoidance of contact force values > 20g could also be an important feature for an optimized AI-HPSD protocol. Although not a parameter focused on in our study, the mean contact force for the AI-HPSD group was < 20 g (19 + 4.3 g). The RESCUE-AF Trial results suggest that restricting the contact force to < 20g at the posterior LA wall decreases esophageal injury risk. (30) Although, contact force target range was slightly more strict in our AI-HPSD group, the mean contact force values were only slightly different (Table 3). Influence of AI and patient risk factors on EDEL Our low EDEL rate does not permit the identification of EDEL predictors. However, examination of the two EDEL cases in the AI-HPSD group found that the patients had procedural and clinical risk factors. Procedural risk factors included several posterior lesions with maximum AI values clinical risk factors included known risk factors such as advanced age and an enlarged LA, both of which fell outside of the standard deviation for the group. (3,5,27,31,32) Since the two EDEL patients had maximum AI values outside of the target range, these EDEL might have been avoided by stricter target range adherence. To facilitate this, the development of an ablation mode with an automatic stop upon reaching the target-maximum AI would be desirable. Additionally, patients with multiple clinical risk factors might require an individualized target AI reduction. Procedural success and all cause complication rate Our mean procedure time difference between the AI-LPLD and AI-HPSD groups was relatively large at 56 minutes . This difference in procedure time is partially explained by the abandonment of the 15 minute wait in favor of an adenosine bolus challenge in the AI-HPSD group. This large difference in procedure time is also explained in part by the degree of ablation in voltage guided concepts. The time-saving feature of HPSD ablation is especially evident with more extensive ablation procedures. We likewise found that fluoroscopy time was slightly but significantly shorter in the AI-HPSD group, which has not been a constant finding in the literature. (12,15) Complication rates were low in both groups and comparable to other studies. Finally, comparing AI-LPLD and AI-HPSD, we found no difference in longterm-freedom from AF and AT, similar to other studies. (33)

This study has design limitations; namely, it is not randomized. Ratherr, the data came from a prospective registry. During the AI-LPLD inclusion period and during the AI-HPSD inclusion period all patients were ablated according to the same protocol in use for the respective periods. Slight differences among the included patients cannot be excluded; however, there was only an 18 month difference between the last treated AI-LPLD patient and the first AI-HPSD patient. Thus, such influences should be minimal. Baseline characteristics were similar apart from a slightly larger mean LA area in the AI-LPLD group. Although comparable in size to other HPSD studies, the rarity of EDEL and other complications could make this study underpowered to detect minor differences. Furthermore, the low EDEL rate does not allow any conclusions regarding baseline or procedural predictors.

This new, optimized protocol for AI-HPSD performed at 50W used a specifically chosen AI target range of 400-450 for the posterior LA and followed carefully set procedures such as catheter pre-flushing for an extended voltage guided ablation concept. In terms of EDEL and the 12-month freedom from recurrent AF and AT, we found no differences in comparison to an historical AI-LPLD control. The procedure required less time and less fluoroscopy. Although EDEL occurence was extremely low, it could be improved by the development of an ablation mode with an automatic stop upon achieving the target-maximum AI of 400. Indeed, maximum AI values above this new target range have been found by us in our routine clinical practice, and this could be avoided with such a development.

1. Kapur S, Barbhaiya C, Deneke T, Michaud GF Esophageal Injury and Atrioesophageal Fistula Caused by Ablation for Atrial Fibrillation. Circulation 2017; 136:1247-1255. 2. Halbfass P, Pavlov B, Müller P, Nentwich K, Sonne K, Barth S, et al. Progression from esophageal thermal asymptomatic lesion to perforation complicating atrial fibrillation ablation a single-center Registry. Circ Arrhythm Electrophysiol 2017; 10:e00523. 3. Yarlagadda B, Deneke T, Turagam M, Dar T, Paleti S, Parikh V, et al. Temporal relationships between esophageal injury type and progression in patients undergoing atrial fibrillation catheter ablation. Heart Rhythm 2019; 16:204–212. 4. Huo Y, Gaspar T, Schönbauer R, Wójcik M, Fiedler L, Roithinger FX et al. Low-Voltage Myocardium-Guided Ablation Trial of Persistent Atrial Fibrillation. NEJM Evid. 2022; 1(11) 5. Das M, Loveday JJ, Wynn GJ, Gomes S, Saeed Y, Bonnett LJ et al. Ablation index, a novel marker of ablation lesion quality: prediction of pulmonary vein reconnection at repeat electrophysiology study and regional differences in target values. Europace 2017;19: 775–783. 6. Schade A, Costello-Boerrigter L, Deneke T, Steinborn F, Chapran M, Vathie K, et al. Oesophageal safety in voltage-guided atrial fibrillation ablation using ablation index or contact force only: a prospective comparison. Europace 2022; 24:1909-1916. 7. Phlips T, Taghji P, El Haddad M, Wolf M, Knecht S, Vandekerckhove Y, et al. Improving procedural and one-year outcome after contact force-guided pulmonary vein isolation: the role of interlesion distance, ablation index, and contact force variability in the ‘CLOSE’-protocol. Europace 2018; 20: f419-f427. 8. Bourier F, Duchateau J, Vlachos K, Lam A, Martin CA, Takigawa M, et al. High-power short-duration versus standard radiofrequency ablation: Insights on lesion metrics J Cardiovasc Electrophysiol. 2018; 29:1570-1575. 9. Kotadia, ID, Williams SE, O’Neill M High-power, Short-duration Radiofrequency Ablation for the Treatment of AF Arrhythmia & Electrophysiology Review 2019; 8:265-272. 10. Schade A, Costello-Boerrigter L, Steinborn F, Bayri AH, Chapran M, Surber R, et al. Voltage-guided ablation in persistent atrial fibrillation-favorable 1-year outcome and predictors. J Interv Card Electrophysiol, 2021; 62:249-257. 11. Halbfass P, Berkovitz A, Pavlov B, Sonne K, Nentwich K, Ene E et al. Incidence of acute thermal esophageal injury after atrial fibrillation ablation guided by prespecified ablation index. J Cardiovasc Electrophysiol 2019;30:2256–61. 12. Shin DG, Ahn J, Han S-J, Lim HE Efficacy of high-power and short-duration ablation in patients with atrial fibrillation: a prospective randomized controlled tria.l Europace 2020; 22:1495-1501. 13. Wielandts J-Y, Kyriakopoulou M, Almorad A, Hilfiker G, Strisciuglio T, Phlips T, et al. Prospective Randomized Evaluation of High Power During CLOSE-Guided Pulmonary Vein Isolation The POWER-AF Study. Circ Arrhythm Electrophysiol. 2021;14: e009112. 14. Francke A, Naumann G, Weidauer MC, Scharfe F, Schoen S, Wunderlich C, et al.Esophageal safety in CLOSE-guided 50 W high-power-short-duration pulmonary vein isolation: The PREHEAT-PVI-registry. J Cardiovasc Electrophysiol. 2022; 33::2276-2284. 15. Chieng D, Segan L, Suguma H, Al-Kaisey A, Hawson J, Moore BM, et al. Higher power short duration vs. lower power longer duration posterior wall ablation for atrial fibrillation and oesophageal injury outcomes: a prospective multi-centre randomized controlled study (Hi-Lo HEAT trial) Europace. 2023; 25: 417–424. 16. Baher A, Kheirkhahan M, Rechenmacher SJ, Marashly Q, Kholmovski EG, Siebermair J, et al. High-Power Radiofrequency Catheter Ablation of Atrial Fibrillation: Using Late Gadolinium Enhancement Magnetic Resonance Imaging as a Novel Index of Esophageal Injury. JACC Clin Electrophysiol. 2018; 4:1583-1594. 17. Wolf M, El Haddad M, De Wilde V, Phlips T, De Pooter J, Almorad A, et al. Endoscopic evaluation of the esophagus after catheter ablation of atrial fibrillation using contiguous and optimized radiofrequency applications. Heart Rhythm. 2019; 16:1013-1020. 18. Contreras‐Valdes FM, Heist EK, Danik SB, Barrett SB, Blendea D, Brugge WR, et al. Severity of esophageal injury predicts time to healing after radiofrequency catheter ablation for atrial fibrillation. Heart Rhythm. 2011;8(12):1862‐1868. 19. Müller P, Dietrich JW, Halbfass P, Abouarab A, Fochler F, Szöllösi A, et al. Higher incidence of esophageal lesions after ablation of atrial fibrillation related to the use of esophageal temperature probes. Heart Rhythm 2015;12:1464-1469. 20. Schoene K, Arya A, Grashoff F, Knopp H, Weber A, Lerche M, et al. Oesophageal Probe Evaluation in Radiofrequency Ablation of Atrial Fibrillation (OPERA): results from a prospective randomized trial. Europace 2020; 22:1487-1494. 21. Castrejón-Castrejón S, Martínez Cossiani M, Ortega Molina M, Escobar C, Froilán Torres C, Gonzalo Bada N, et al. Feasibility and safety of pulmonary vein isolation by high-power short-duration radiofrequency application: short-term results of the POWER-FAST PILOT study. Journal of Interventional Cardiac Electrophysiology 2020; 57:57–65. 22 Sternick EB, Persiano AC, Arantes V Is it safe to monitor oesophageal temperature during AF ablation? Europace 2012; 14:1432. 23. Müller P, Dietrich JW, Halbfass P, Abouarab A, Fochler F, Szöllösi A, et al. Higher incidence of esophageal lesions after ablation of atrial fibrillation related to the use of esophageal temperature probes. Heart Rhythm 2015; 12:1464-1469. 24. Grosse Meininghaus D, Blembel K, Waniek C, Kruells-Muench J, Ernst H, Kleemann T et al. Temperature monitoring and temperature-driven irrigated radiofrequency energy titration do not prevent thermally induced esophageal lesions in pulmonary vein isolation: a randomized study controlled by esophagoscopy before and after catheter ablation. Heart Rhythm 2021; 18:926–934. 25. Wolff C, Langenhan K, Wolff M, Efimova E, Zachäus M, Darma A, et al. Incidence and predictors of thermal oesophageal and vagus nerve injuries in Ablation Index-guided high-power-short-duration ablation of atrial fibrillation: a prospective study. Europace. 2024; 26: euae107. 26. Müller J, Berkovitz A, Halbfass P, Nentwich K, EneE, Sonne K, et al. Acute oesophageal safety of high-power short duration with 50Wfor atrial fibrillation ablation. Europace 2022; 24: 928–937. 27. Müller J, Nentwich K, Berkovitz A, Ene E, Sonne K, Zhuravlev V, et al. Acute oesophageal safety and long-term follow-up of AI-guided high-power short-duration with 50 W for atrial fibrillation ablation. Europace 2023; 25: 1379–1391. 28. Nakagawa H, Ikeda A, Govari A, Papaioannou T, Constantine G, Bar-Tal M, et al. Prospective study using a new formula incorporating contact force, radiofrequency power and application time (Force-Power-Time Index) for quantifying lesion formation to guide long continuous atrial lesions in the beating canine heart. Circulation. 2013;128: Creation of Continuous and Transmural Radiofrequency Lesions A12104 29. Nakagawa H. Creation of Continuous and Transmural Radiofrequency Lesions. Circ Arrhythm Electrophysiol. 2018; 11: e006378. 30. Zhang X, Kuang X, Gao X, Xiang H, Wei F, Liu T, et al. RESCUE-AF in Patients Undergoing Atrial Fibrillation Ablation: The RESCUE-AF Trial. Circ Arrhythm Electrophysiol. 2019; 12: e007044. 31. Stauber A, Kornej J, Bollmann A, Hindricks G, Sommer P. Relevance of esophageal position and temperature on thermal injuries and rhythm outcome in atrial fibrillation ablations. Pacing Clin Electrophysiol 2020; 43:194–200. 32. Assis FR, Shah R, Narasimhan B, Ambadipudi S, Bhambhani H, Catanzaro JN, et al. Esophageal injury associated with catheter ablation for atrial fibrillation: Determinants of risk and protective strategies. J Cardiovasc Electrophysiol 2020; 31:1364-1376. 33. Amin AM, Ghaly R, Ibrahim AA, Ali MA, Almaadawy O, Elzahaby A, et al. Efficacy and safety of high-power short-duration ablation for atrial fibrillation: a systematic review and meta-analysis of randomized controlled trials. J Interv Card Electrophysiol. 2024; 67:1445-1461. Figure 1 Figure 1: CARTO® 3 views with types of low voltage distribution and typical lesion sets a) no low voltage (PVI only), b) small low voltage area (direct ablation and linear lesion between mitralanlus and LSPV to avoid critical isthmus), c) large low voltage area at anterior wall (linear lesions to encircle the area and avoid typical macroreentries) Figure 2 Figure 2: Endoscopically detected esophageal lesions. A) Category 1 lesion (erosion), B) Category 2 lesion (ulcer 7 mm) Information & Authors Information Version history Peer review timeline Published Journal of Cardiovascular Electrophysiology Version of Record9 Jul 2025Published Copyright This work is licensed under a Non Exclusive No Reuse License. Collection

Authors Metrics & Citations Metrics Article Usage 250views 171downloads Citations Download citation Lisa Costello-Boerrigter, Frank Steinborn, Mykhaylo Chapran, et al. Posterior wall ablation characteristics and esophageal safety of an optimized protocol for voltage-guided, high-power short--duration ablation of atrial fibrillation. Authorea. 13 March 2025. DOI: https://doi.org/10.22541/au.174182957.71262166/v1 DOI: https://doi.org/10.22541/au.174182957.71262166/v1 If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. Simply select your manager software from the list below and click Download. For more information or tips please see 'Downloading to a citation manager' in the Help menu.