Acute and long‐term efficacy of ablation index‐guided higher power shorter duration ablation in patients with atrial fibrillation: A prospective registry

Abstract Background Theoretically, targeting the same ablation index (AI) using higher power may achieve the same lesion size with a shorter ablation time. We evaluated the acute and long‐term efficacy of higher‐powered ablation guided by ablation index (HPAI) compared with conventional‐powered ablation guided by AI (CPAI) for pulmonary vein isolation (PVI) in patients with atrial fibrillation (AF). Methods Drug refractory symptomatic AF patients who had been ablated with 40 W on the anterior/roof segments and 30 W on the posterior/inferior/carina segments were enrolled (HPAI group). We compared the HPAI group with the CPAI group who were ablated with 30 W on the anterior/roof segments and 25 W on the posterior/inferior/carina segments. The same AI was targeted (≥450 on the anterior/roof segments and ≥350 on the posterior/inferior/carina segments). We compared ablation time, acute pulmonary vein reconnection (PVR) and 1‐year AF recurrence between the two groups. Results A total of 118 patients were included (86 in the HPAI group and 32 in the CPAI group, paroxysmal AF, 73%). There was no significant difference in the acute PVR rate between the HPAI and the CPAI groups (3.7% vs. 4.2%, P = .580) with a 41% reduction in ablation time for PVI (38.7 ± 8.3 vs. 65.8 ± 13.7 minutes, P < .001). The 1‐year AF recurrence rate was not significantly different between HPAI and CPAI groups (12.8% vs. 21.9%, Log‐rank P = .242). There were no major complications in either group. Conclusions Increased power during AF ablation, using the same AI targets, reduced the procedure and ablation times, and showed a comparable acute and long‐term outcome without compromising safety. Clinical Trial Registration https://www.clinicaltrials.gov. Unique identifier: NCT 04379557.


| INTRODUC TI ON
Catheter ablation of atrial fibrillation (AF) to achieve pulmonary vein isolation (PVI) is an effective treatment for rhythm management. 1,2 Recently, there have been several technological advances in catheter ablation of AF that have helped to achieve a more effective and safer PVI. [3][4][5] The ablation index (AI) is a recently developed weighted formula, which includes contact force (CF), ablation time, and radiofrequency (RF) power that has a good correlation with lesion formation. 5,6 After the introduction of AI in clinical practice, several studies reported the efficacy and safety of AI-guided PVI. [7][8][9][10] Previously, we reported that optimal AI targeted PVI (anterior/roof segments, AI 450 and posterior/inferior/carina segments, AI 350), the so-called "OPTIMUM" protocol, showed an improved acute outcome of PVI than conventional CF-guided strategy. 11 A recently developed ablation catheter with surround flow (SF) and 56 irrigation holes allows safe ablation at higher RF power and may lead to rapid and effective lesion formation. [12][13][14] Theoretically, targeting the same AI using higher power may achieve the same quality and size of lesion with shorter ablation times. Several studies have recently reported that PVI using higher power reduced procedure time with comparable efficacy. [15][16][17][18][19] However, these studies applied different target AIs and different ranges of "high power," and still, there is concern about the risk of complications, such as steam pop in a higher-power ablation. [15][16][17][18][19] There are limited data regarding the acute and long-term efficacy and safety of applying higher power than conventional power targeting optimal AI value in our daily routine PVI procedure.
In this study, we aimed to evaluate how to apply a higher power in PVI guided by optimal target AI and compare the ablation time, acute and long-term efficacy and safety between higher-powered ablation and conventional-powered ablation in patients with AF.

| Study design and study population
From July 2018 to November 2019, 86 consecutive patients with drug-refractory symptomatic AF underwent optimal AI-guided PVI with higher RF power (higher-powered AI-guided ablation, HPAI group). The results of the HPAI group were compared with those of the 32 patients enrolled in the "OPTIMUM" phase 2 study, 11 undergoing optimal AI-guided PVI with conventional RF power (conventional-powered AI-guided ablation, CPAI group). We planned to compare the HPAI group that prospectively collected with the CPAI group that pre-registered cases to enroll the total study population more easily and efficiently as mentioned and reduce the total study duration. Immediately after the completion of the OPTIMUM study (the CPAI group), 11 the LESS study (the HPAI group) was conducted, so there was no significant change in treatment practice, operator proficiency, and the device. Patients younger than 20 years or older than 80 years, with left atrial (LA) diameter >50 mm, and severe left ventricular systolic dysfunction (left ventricle ejection fraction <35%) were excluded. All procedures were performed by experienced operators in two tertiary centers. The study was approved by the Institutional Review Board of each center. The study was conducted according to the Declaration of Helsinki. All patients provided written informed consent before undergoing the AF ablation procedure. When the esophageal temperature increased to more than 38℃, we stopped the RF energy delivery and moved to other sites.

| Protocol for the PVI procedure and outcome measurement
The target AI was based on the OPTIMUM protocol. 11 Briefly, the OPTIMUM protocol was a point-by-point RF ablation for PVI with contiguous lesions. RF energy was delivered until an AI of ≥450 was attained at the anterior/roof segments, and an AI of ≥350 was attained at the posterior/inferior/carina segments. The target CF was between 5 and 20 g. Visitag (Biosense Webster Inc) was used with a predefined set of catheter stability (maximum range 2.5 mm for a minimum time of 5 seconds) and a minimum CF of 5 g with force over time 25% (3 g at the left superior and inferior ridges). Each Visitag annotation point was presented according to the AI as a lesion tag K E Y W O R D S ablation index, atrial fibrillation, high power ablation, pulmonary vein isolation size of 2 mm (radius 2 mm ball), and the maximal interlesion distance between neighboring lesion was ≤4 mm. If the catheter dislocated before reaching the target AI, a new RF ablation was applied to reach the AI target. The study flow is summarized in Figure 1.
After the first pass ablation of all PVs, residual potential (RP) was assessed as the remaining PV potential. When the PV was not isolated after the first encirclement, additional touch-up ablation was delivered. After a 20-minute observation of PVI achievement, spontaneous early reconnection (ER) of PV was evaluated. Acute PVR was defined as the presence of ER. In the case of the presence of acute PVR, additional touch-up ablation was performed to achieve PVI. Additional linear ablation or ablation for non-PV trigger foci was performed according to the discretion of the operator. If AF persisted after ablation, internal or external electrical cardioversion was performed.

| Demographic, procedural, and ablation lesion data analysis
Patient data on age, sex, comorbidities, CHA 2 DS 2 -VASc score, and echocardiographic parameters were collected. Covariates, including the ablation procedure (PVI only or whether performing additional ablation other than PVI), total procedure time, total ablation time, ablation time for PVI, and fluoroscopy time, were analyzed.
Segments with RP and acute PVR were analyzed by predefined 14 PV segments ( Figure S1). The VisiTag data from each ablation point during PVI were analyzed. Mean CF, power, impedance drop, forcetime integral (FTI), AI, mean, and total ablation duration at each point was analyzed for each segment.

| Acute and long-term efficacy outcomes
We compared total procedure time, total ablation time, ablation time for PVI, and fluoroscopy time between the HPAI and CPAI groups.
For assessment of acute efficacy, the rate of RP and acute PVR among PV segments were compared between the two groups.
After the index procedure, follow-up visits were arranged after 2 weeks and 3, 6, 9, and 12-month. A 12-lead electrocardiogram was performed at each follow-up visit, and a 24-hour Holter monitoring was performed at 3-and 12-month visit. The long-term efficacy endpoint was defined as the recurrence of atrial tachycardia (AT) or AF evaluated by documented any AT or AF lasting longer than 30 seconds within 1-year follow-up after the 3-month blanking period after the index procedure. 1

| Safety outcomes
Procedure-related complications were defined as safety outcomes, including steam pop, cardiac tamponade, atrial-esophageal fistula, stroke, transient ischemic attacks, and death within 4-weeks and 1year after the index procedure.

| Preclinical study: comparison of the lesion size between 40W and 30W RF ablation at the same target AI of 450 in canine models
To provide complementary information on the ablation parameters and the lesion sizes between 40 W and 30 W RF ablation at the target AI of 450, we conducted a preclinical in vivo experiment using four mongrel dogs (32 ± 6.8 kg, males). General anesthesia was induced using intravenous tiletamine/zolazepam (Zoletil ® , 5 mg/kg; Virbac S/A) and maintained using isoflurane gas (1%-2% oxygen). The animals were intubated and mechani-

| Statistical analysis
Categorical variables are presented as numbers and percentages, and continuous variables with normal distribution are presented as mean ± SD. The chi-square test was performed to compare categorical variables. Continuous variables were compared with the Student's t test. Continuous variables with non-normal distribution are presented as the median and interquartile range (IQR) and compared by Wilcoxon test or the Kruskal-Wallis test. The AT/AF free survival at 1-year between the two groups was analyzed using Kaplan-Meier curves and log-rank test. All tests were 2-sided, and a F I G U R E 1 Study design. AI, ablation index; PVAI, pulmonary vein antral isolation P < .05 was considered statistically significant. All statistical analyses were performed using SPSS 25.0 (IBM Corp.).

| Baseline characteristics
A total of 86 patients (mean age, 59.5 ± 9.0 years, median CHA 2 DS 2 -VASc score, 1 [IQR [1][2], and 77% of paroxysmal AF) were consecutively enrolled in the HPAI group. There were no significant differences between the two groups in their baseline characteristics (Table 1 and Table S1).

| Procedure and ablation times in HPAI and CPAI groups
The HPAI group showed a shorter PVI time than the CPAI group (Table 1 and Figure 2). The shorter PVI time resulted in the reduction of total procedure time, total ablation time, and fluoroscopy time in the HPAI group compared with the CPAI group (Table 1 and Figure 2). The mean PVI time was reduced by 41%, and the mean total procedure time reduced by 30% in the HPAI group compared with the CPAI group.

| Ablation parameters in HPAI and CPAI groups
Compared with the CPAI group, there were fewer RF applications for PVI and ablation points by each segment in the HPAI group (Table 2).
In the HPAI group, the mean ablation time for each ablation was shorter and the total ablation time for each segment was shorter, whereas the mean AI was not significantly different between the two groups ( Table 2, Table S2, and Figure S2). In the HPAI group, the mean power was higher, whereas CF, FTI, and impedance drop were lower than in the CPAI group. These results were consistently observed in anterior/roof segments and posterior/inferior/carina segments (Table S2). Figure 3 shows the rates of segments with RP and acute PVR in the HPAI and CPAI groups. The HPAI group showed a lower RP rate than and ER rates of each segment in two study groups are presented in Figure S3. The rate of acute PVR was similar between the HPAI and the CPAI group (3.7% vs. 4.2%, P = .580). The distributions of RP and ER by predefined PV segments are presented in Figure S3.

| Comparison of acute and long-term efficacy outcomes between HPAI and CPAI groups
In the period from the end of the blanking period to the end of 12-month follow-up, AT/AF recurrence was documented in 12.8% of the HPAI group and 21.9% of the CPAI group. Figure 4 shows Kaplan-Meier curves of AT/AF free survival for the two groups.
There was no significant difference between the two groups (logrank P = .367). Patients with paroxysmal AF showed a higher 1-year AF-free survival rate than persistent AF (Figure 4). In both paroxysmal AF and persistent AF, 1-year AT/AF survival was not different between HPAI and CPAI groups.

| Safety outcomes
Procedural complications had not occurred within 4-weeks and 1- year of the procedure in the HPAI and CPAI groups.

| Preclinical study: lesion size at target AI of 450 for 40 W and 30 W
We performed a preclinical study to compare the lesion size and ablation parameters between 40 W and 30 W deliveries with a target AI of 450 (Table S3, Figures S4 and S5)

| D ISCUSS I ON
In this study, we found that higher-than-conventional RF power PVI guided by optimal target AI showed comparable acute and longterm efficacy of PVI with reducing the procedure and ablation times After introducing values expecting the lesion size, such as AI, high power ablation was interpreted in the context of the same AI in several previous studies, as summarized in Figure 5. [15][16][17][18][19] Rather than universal use of "5 seconds" as short duration, different ablation times derived from the formula were applied. 4,[15][16][17][18][19] Although these studies shared the same hypothesis and results: by targeting the same AI with a higher power, they achieved comparable efficacy and safety with reduction of procedure and ablation times, the detailed ablation strategies were varied. Previous studies based on the "CLOSE protocol" used a higher target AI than our study: AI 550 at anterior and 400 at posterior. 16,17,19 In addition, the definition of high RF power varied in these studies: variously, anterior 40, 45, and 50 W. 16  in conventional ablation). 29 In the context of our study results, operators can apply a high power strategy during AF ablation using target AI on the basis of conventional power PVI.

| Study limitations
This study has several limitations. First, this is a prospective registry study when compared with a previous study group; we sequentially enrolled patients in two study groups. Therefore, the two groups were taken from two different prospective registries instead of one randomized prospective study. Although baseline characteristics of HPAI and CPAI groups were not significantly different, the results should be carefully interpreted with understanding the inherent limitation of the study design. Second, ablation was performed with a different type of catheter in the two groups (Thermocool SmartTouch catheter in the CPAI group vs. Thermocool SmartTouch SF catheter in the HPAI group). In previous studies, the SF catheter was more frequently used for high-power PVI groups. [15][16][17]19 Results from a recent study, using the same catheter (Thermocool SmartTouch catheter) in both groups are consistent with ours: the high power group had reduced procedure and ablation times without any differences in the efficacy and safety of the PVI procedure. 20 Third, the main purpose of this study was to compare the different power strategies for PVI. However, a substantial portion of patients received additional ablation other than PVI during the procedure (Table S1) Figure S6). To provide a supplementary analysis for the difference in LAVI between two groups, we matched 2:1 of the HPAI and CPAI group using similar LAVI (CPAI, n = 26, and HPAI, n = 52). After matching LAVI between two groups, the mean LAVI values of the two groups were not different (47.2 ± 13.6 in the CPAI group and 48.1 ± 13.6 in the HPAI group, P = .786, Table S4). In this matched cohort, the HPAI group consistently showed significantly shorter PVI time than the CPAI group (38.6 ± 9.2 vs. 66.1 ± 14.7 minutes, P < .001). Thus, higher power ablation with the same target AI could markedly reduce PVI time. Sixth, our strategy for LIPV inferior/posterior lesions was not different in the two groups, and the acute reconnection rates of the two groups were not significantly different (HPAI group vs. CPAI group, 1.2% vs. 3.1%, P = .299). We applied 25 W for a 15-seconds ablation with esophageal temperature monitoring. The acute reconnection rate at LIPV posterior and inferior segments was 1.7%, which was slightly lower, but not significantly iffered from other segments (P = .066). Lastly, this study included a small number of patients; thus, the comparison of safety between the two groups might not be conclusive. A high-power ablation strategy is not a standard technique, and the risk of perforation is an important issue to overcome. We found no complications among 87 patients who received PVI using higher power RF ablation, which would provide significant clinical implications.
However, more safety-related data is needed for higher power strategies to become common.

| CON CLUS ION
Higher-powered ablation can be safely performed using an AIguided strategy. Higher-powered AI-guided PVI reduced total ablation and total procedure times by a significant reduction in ablation time for PVI, and showed comparable acute PVR rate and long-term AF-free survival without significant complications, compared to conventional-powered AI-guided PVI.

ACK N OWLED G M ENTS
We thank Junyeop Yeo for the assistance.