Incidence of ablation‐induced esophageal injury associated with high‐power short duration temperature‐controlled pulmonary vein isolation using a specialized open‐irrigated ablation catheter: A retrospective single‐center study

To evaluate short‐term efficacy and incidence of ablation‐induced endoscopically detected esophageal injury in patients undergoing high‐power, short‐duration (HPSD) pulmonary vein isolation using a novel irrigated radiofrequency ablation catheter and ablation generator setup.


| INTRODUCTION
Efficacy and safety of catheter ablation in patients with paroxysmal and persistent atrial fibrillation have increased substantially over the last two decades. However, pulmonary vein reconduction limiting long-term success on the one hand and relevant peri-procedural complications like cardioembolic events and ablation-induced esophageal injury on the other hand are clinically relevant procedurerelated problems.
Recent efforts have focused on modifying ablation energy, ablation time, and combinations of different ablation parameters to increase acute and long-term success and limit the incidence of esophageal lesions. Endoscopy is feasible to detect ablation induced esophageal lesions. Endoscopically detected esophageal lesions (EDEL) are considered a surrogate parameter for the risk of clinically relevant esophageal complications like perforation and fistula. 1,2 According to pre-existing literature the incidence of atrio-esophageal fistula after atrial fibrillation (AF) ablation was below 0.2%, whereas another study found an incidence of atrio-esophageal fistula and esophageal perforation of up to 0.6% in patients undergoing postablation esophageal endoscopy. 1,3,4 A novel approach aims to overcome these limitations by using high-power and short duration (HPSD) radio-frequency ablation. The QDOT Fast Trial, using an ablation setup of 90 W and 4 s per lesion, one endoscopically detected injury-that is, an esophageal ulceroccurred, resulting in a relatively low rate of esophageal lesions of only 2%. 5 To the best of our knowledge no studies evaluating the incidence of EDEL when using the QDOT Micro ablation catheter in conjunction with different energy and ablation time settings have been published so far. A unique characteristic of the QDOT ablation catheter is, that thermocouples are placed directly at the catheter tip and tissue interface. This enables acquisition of tissue temperature directly beneath the catheter tip. In a temperature controlled ablation mode the risk of tissue-overheating or steam-pops could possibly be reduced compared with the use of conventional ablation catheters.
Aim of the current study was to characterize the safety profile of a HPSD strategy using the same ablation catheter but lower energy settings compared with the QDOT Fast Trial.

| METHODS
Consecutive patients undergoing an AF ablation procedure using a novel ablation catheter (QDOT Micro; Biosense Webster) from November 2019 to February 2020 at our institution were included. The study was approved by the local institutional review board. All patients gave informed consent to the ablation procedure and postablation diagnostics. According to our standard approach all patients underwent postablation endoscopy on the day after ablation to identify thermal esophageal injury related to AF ablation.

| Atrial fibrillation ablation procedure
All procedures were performed using a three-dimensional electroanatomic mapping system (CARTO 3, BiosenseWebster) under analgo-sedation using continuous propofol infusion in conjunction with morphine derivatives. In Group 1 patients (de-novo AF ablation) all patients were treated with complete wide antral pulmonary vein isolation (PVI). Isolation of the ipsilateral PVs was performed with point-by-point lesions. 4 Group 2 patients (redo AF ablation) underwent re-isolation of PVs by ablating all identified gaps within the previous antral ablation lines to re-isolate all reconnected PVs. Following our standard approach the superior caval vein was isolated additionally in this group. In both groups ablations were done using a

| Postablation esophageal endoscopy
Esophageal endoscopy (EE) was performed by experienced operators (having performed >3000 post-AF-ablation endoscopies) on the next day after ablation to assess for the presence and extent of EDEL.
EDEL were categorized into mild (Category 1 lesion: erythema/ erosion or ulcers ≤ 5 mm diameter) or severe (Category 2 lesion: ulcer > 5 mm diameter) based on our prior clinical experience. 4

| Postprocedural treatment
All patients received proton pump inhibitors in double standard dose for 6 weeks postprocedurally. In case of EDEL, control endoscopy was performed within 7 days. Patients with Category 2 lesions received a liquid diet until regression of EDEL was confirmed in a repeat endoscopy during the initial hospital stay.

| Statistical analysis
The data are expressed as mean ± SD for continuous variables or as frequencies and percentages for categorical variables. To compare continuous and categorical variables between those with and without EDEL, we used t-test or Fisher's exact test as appropriate. Two-sided p-values of less than .05 were considered statistically significant. All statistical analyses were performed using GraphPad Prism 8 (GraphPad Software).

| RESULTS
A total of 45 patients (67 ± 10 years; 58% male; 42% paroxysmal AF) were included in the study. In total 14/45 (31%) patients underwent a redo AF ablation procedure due to AF recurrence. In 9/14 (64%) patients additional left and right atrial RF ablation was performed. In eight patients isolation of the superior caval vein was done additionally to re-isolation of PVs. In two patients a box lesion at posterior left atrial (LA) wall and in six patients additional lines not at posterior wall sites were applied due to occurrence of a LA macro reentrant tachycardia. Of note, 31/45 (69%) patients receiving a first AF ablation procedure were included in the final analysis (67 ± 11 years; 55% male; 48% paroxysmal AF). Patient characteristics of the complete study cohort are shown in Table 1 and the group of patients undergoing first-time AF ablation are shown in Table 2. In all 31 patients effective pulmonary vein isolation with bidirectional block was achieved at the end of the procedure.

| Incidence of EDEL
In Group 1 patients endoscopy was done on the day after the procedure in all 31 patients. In 5/31 (16%) patients EDEL was detected on postablation endoscopy (Category 1: erosion n = 2, Category 2: ulcer n = 3; Figure 3). The 3/14 (21%) Group 2 patients did not undergo postprocedural endoscopy due to a lack of ablation at posterior wall. None of 11 patients with postprocedural endoscopy showed a thermal lesion. The group of patients undergoing first-time PVI and the group undergoing a redo ablation procedure did not differ significantly regarding procedural parameters (Table 3).
Patient characteristics in the group of patients with postprocedural EDEL and in the group of patients without EDEL did not differ significantly regarding patient characteristics ( Table 2). Of

| Comparison of procedural parameters in EDEL positive and EDEL negative patients
Regarding patient characteristics including comorbidities like coronary artery disease, diabetes mellitus and arterial hypertension, no statistically significant differences were found between patients with and without EDEL ( Table 2). When comparing procedural parameters, that is, procedure time, total ablation time and ablation time at posterior wall, mean ablation index (AI) per lesion, mean ablation temperature, and mean contact force at posterior wall in patients with and without EDEL only the variable mean contact force at posterior wall showed a significant difference. Interestingly, mean contact force at posterior wall was significantly higher in patients without esophageal lesions in postprocedural esophageal endoscopy compared with patients with EDEL (15.6 ± 4.7 g vs. 11.9 ± 0.8 g; p = .001).

| Adverse events and follow-up
No clinically relevant intra-procedural adverse events were reported. Using radiofrequency catheter ablation with moderate ablation power and longer ablation times results in a relevant number of EDEL. 4 Very high-power short duration ablation seemed not to be associated with EDEL according to a small prospective study. 5 However, whether this holds true for a larger number of patients treated in a real-world setting still has to be proven in larger studies and registries. The approach of our study was to reduce the target ablation power at the cost of an increase in ablation time compared with very high-power short duration ablation strategies. A target ablation index of 350 at posterior wall ablation sites in conjunction with a target ablation power of 50 W was used. An ablation power of 50 W was higher compared with standard ablation protocols using between 20 and 35 W at posterior wall but relevantly lower compared with 70-90 W used in very high-power short duration ablation studies. 5,7 Of note, the incidence of EDEL resulting from an ablation approach using one set of ablation parameters and one ablation catheter type is not transferrable to other approaches using different sets of parameters and ablation catheters. 5,7-9 Furthermore, ablation index was not evaluated in the setting of a high-power short-duration ablation strategy so far. Therefore, the relatively high incidence of EDEL might also be a result of a longer ablation duration per lesion compared with other high-power short-duration ablation trials.
However, the rational of using a target ablation index value for posterior wall ablation was to objectify the amount of RF-energy applied to each posterior wall ablation site. Kottmaier et al. 7 chose a very-high-power short-duration ablation strategy (70 W) and a strict time limit of 5 s resulting in a reasonable high 1-year-success rate of 83%. Of note, in contrast to our trial this study did not include a rigorous endoscopic follow-up. 7 Therefore, the incidence of EDEL associated with a high-power short-duration ablation protocol using  Abbreviation: EDEL, endoscopically detected esophageal lesion.
ablation lesions result in a lower rate of acute PV reconduction and necessity of repeat ablation at the same ablation sites. Therefore, high-power short duration ablation could potentially result in a lower risk of esophageal injury while maintaining a comparable or even higher clinical success rate. Reddy et al. 5 reported a short-term success rate of 94% at 3 months FU.
4.2 | Pre-existing data supporting the use of a short-duration high-power ablation approach A couple of retrospective clinical trials have already evaluated a strategy of increased ablation power of 50 W and higher in conjunction with standard ablation catheters resulting in a favorable safety and efficacy profile. [7][8][9] Of note, according to published data of our own group the incidence of EDEL in a subgroup of 25 patients undergoing high-power short duration ablation using 50 W and an AI of 350 at posterior wall in conjunction with standard irrigated contact-force ablation catheter was 12%. 11 Patients in this trial were ablated according to the same ablation protocol using the same ablation parameters as in the cohort of 25 patients mentioned above.
The only differences was the use of a specialized ablation catheter designed for a "very high-power short-duration" ablation approach.

| Rational for omission of luminal esophageal temperature monitoring
Temperature probes measuring intraluminal esophageal temperature were not used in this study. The value of intraluminal esophageal temperature measurement using conventional thermocouples is highly questionable as a measurable intraluminal temperature increase might be significantly too late to prevent esophageal injury, especially in the setting of a "high-power short-duration" strategy.
Many studies including a recent meta-analysis have questioned the usefulness of standard temperature probes for preventing esophageal injury. 14

| LIMITATIONS
A major limitation of this study is the limited patient number. Despite the low sample size, it's strength is the timely and consistent manner of postprocedural endoscopy in all patients included in the study and the uniform ablation protocol.
The follow-up period for recording esophageal complications was limited to three months after ablation. However, perforating esophageal complications usually become clinically evident within 2-4 weeks, but potentially might evolve later during follow-up.
We used AI values to quantify and limit the amount of ablation at posterior wall. Of note, AI has neither been explicitly validated for ablation with high-power short duration ablation strategies nor va-

DATA AVAILABILITY STATEMENT
The data that support the findings of this study are available from the corresponding author upon reasonable request.