Comparison of Spatio‐Temporal Dispersion Electrogram Ablation and Low Voltage Zone Ablation in Patients With Persistent Atrial Fibrillation

ABSTRACT

Background

Pulmonary vein isolation (PVI) alone was less effective for patients with persistent atrial fibrillation (AF) compared to those with paroxysmal AF. Although the TAILORED AF trial demonstrated the efficacy of spatio‐temporal dispersion electrogram (STDE) ablation as additional treatment for PVI, comparisons between STDE and low‐voltage zone (LVZ) remain unclear. We compared the efficacy of two different methods beyond PVI: the STDE and the LVZ ablation.

Methods

This retrospective observational study included 107 consecutive patients (median age, 71 years; 79 male) who underwent initial catheter ablation for persistent AF at our institution. STDE evaluation and STDE‐guided substrate modification were performed in the STDE ablation group (n = 53). The historical control group (n = 54) did not undergo STDE evaluation and received additional ablation based on the LVZ. The primary endpoint was the recurrence of atrial arrhythmias after the ablation up to 24 months, with a 3‐month blanking period applied.

Results

The number of additional ablations, including linear ablation, posterior wall isolation, and focal ablation, was significantly higher in the STDE group. The AF‐free survival rate at 24 months follow‐up after ablation was higher in the STDE group (76.3% vs. 58.4%; p = 0.027). In the multivariate Cox proportional regression analysis, prolonged AF duration (hazard ratio [HR] 1.37, 95% confidence interval [CI] 1.16–1.61; p < 0.001) and larger left atrial size (HR: 1.11, 95% CI: 1.02–1.20; p = 0.012) were independent predictors of AF recurrence. Conversely, STDE ablation was independently associated with a significantly lower risk of recurrence (HR: 0.29, 95% CI: 0.12–0.66; p = 0.004).

Conclusion

STDE was associated with a lower risk of AF recurrence than PVI with LVZ ablation.

This study compared STDE‐guided versus LVZ‐guided ablation beyond PVI in 107 persistent AF patients. At 24 months, the AF‐free survival rate was significantly higher in the STDE group (76.3% vs. 58.4%, p=0.027). STDE ablation independently reduced recurrence risk (HR: 0.29, p=0.004), providing superior clinical efficacy over LVZ ablation.

Abbreviations

AT

atrial tachycardia

CFAE

continuous fractionated atrial electrograms

EPVI

extended pulmonary vein isolation

FAAM

fractionated signal area in the atrial muscle

LA

left atrium

LVZ

low voltage zone

PVI

pulmonary vein isolation

RA

right atrium

STDE

spatio‐temporal dispersion electrogram

1. Introduction

For paroxysmal atrial fibrillation (AF), the efficacy of pulmonary vein isolation (PVI) is well established, with a reported AF‐free survival rate of approximately 80% [1]. However, PVI alone has been reported to have a lower AF‐free survival rate of approximately 60% for persistent AF [2]. Both the DECAAF AF II and CAPLA trials failed to show the prognostic impact of additional procedures including magnetic resonance imaging‐guided fibrosis ablation and posterior wall isolation on the recurrence of AF after ablation in patients with persistent AF [3, 4]. Efforts are in progress to improve the outcomes of ablation for persistent AF with substrate ablation in addition to PVI.

Seitz et al. found that assessment of intra‐atrial potentials during AF using a multipolar mapping catheter revealed potentials that were distributed both spatially and temporally between electrodes at different sites, known as spatio‐temporal dispersion electrograms (STDEs), and that ablation targeting STDE in patients with persistent AF stopped AF during ablation in 95% of cases. The study showed that STDE ablation resulted in a lower rate of AF recurrence in comparison to conventional methods during the 1‐year post‐ablation follow‐up period [5]. Artificial intelligence software has been developed to standardize STDE‐based ablation. Furthermore, the TAILORED‐AF trial has demonstrated that PVI combined with STDE ablation significantly reduces the recurrence of AF compared to PVI alone [6]. However, no studies have yet compared STDE ablation with a conventional additional strategy, such as ablation in the low voltage zone (LVZ). This study compared STDE and LVZ as additional ablation for PVI to test the hypothesis that STDE ablation is superior for treating persistent AF.

2. Methods

2.1. Study Design

This is single‐center retrospective observational study. One hundred and seven consecutive patients with persistent AF who underwent their first catheter ablation at our hospital between July 2018 and March 2021 were retrospectively evaluated. All patients underwent radiofrequency PVI, and as an additional procedure, ablation targeting the LVZ defined as bipolar voltage of < 0.5 mV during sinus or atrial pacing rhythm was performed until August 2019. The STDE‐guided ablation strategy was introduced in August 2019. This study compared two groups: those who underwent PVI plus STDE guided ablation (the STDE group) and those who underwent PVI plus LVZ guided ablation (the LVZ group). The indication for ablation was determined according to Japanese guidelines [7]. The primary endpoint was the recurrence of any atrial tachyarrhythmia during the follow‐up period of up to 24 months, applying to the blanking period of 3 months. This single‐center study was approved by our institutional review board (approval number #22‐083). The requirement for written informed consent was waived owing to the retrospective nature of this study.

2.2. Ablation Procedure

In all cases, antiarrhythmic agents were uninterrupted. Warfarin and dabigatran were uninterrupted. However, apixaban, rivaroxaban, and edoxaban were skipped in the morning on the day of the procedure if the patients took them in the morning. Electrocardiogram‐gated cardiac computed tomography was performed before the ablation procedure unless patients had severe renal impairment, bronchial asthma, or a history of contrast agent allergy. The entire procedure was performed under deep sedation with midazolam and dexmedetomidine hydrochloride. An adaptive servo ventilator was used in all patients, and a nasal airway was introduced, if needed. Activated clotting time was maintained at 300–400 s after intravenous heparin administration. A steerable 6‐F 20‐polar catheter (BeeAT; Japan Lifeline, Tokyo, Japan) was placed in the coronary sinus via the right jugular or femoral vein and served as a reference for the 3D electroanatomical mapping system (EnSite Precision; Abbott, St Paul, MN or CARTO; Biosense Webster, Diamond Bar, CA, USA). A single transseptal puncture with navigation by intracardiac ultrasound (ViewFlex; Abbott or Soundstar; Biosense Webster) was used to deliver the ablation and mapping catheter to the left atrium (LA) (one puncture, two sheaths). In the case of EnSite, a TactiCath irrigation catheter (Abbott) was delivered via an 8.5‐F steerable sheath (Agilis; Abbott). Detailed left atrial and pulmonary vein geometry and electrical maps were collected using a multi‐spline (HD grid; Abbott) mapping catheter via an 8.5‐F long sheath (SL‐0; Abbott). In the case using CARTO, a THERMOCOOL SMARTTOUCH SF Catheter (Biosense Webster) was delivered via an 8.5‐F steerable sheath (Agilis; Abbott or VIZIGO; Biosense Webster). The mapping catheter was a PENTARAY™ NAV eco catheter (Biosense Webster) delivered via the SL‐0.

2.3. Extended Pulmonary Vein Isolation

Intracardiac cardioversion was performed before starting EPVI to assess the voltage map during sinus rhythm or atrial pacing in most of the patients. We set the target lesion index or ablation index to 4.5/400 at the posterior PV antrum and 5.2/480 at the anterior PV antrum, respectively. The inter‐lesion distance was set to < 5 mm at the points targeting a lesion index of 4.5 and < 6 mm targeting a lesion index of 5.2 using EnSite, and < 5 mm using CARTO. The energy was set to 25–35 W in the posterior direction and 35–45 W in the anterior direction. The goal of EPVI is to enclose the pulmonary veins with contiguous and optimized RF lesions indicated by the lesion index/ablation index. A bidirectional conduction block between the PV and the LA was confirmed after EPVI.

2.4. STDE/LVZ Evaluation and Additional Ablation

In the STDE group, mapping during AF was performed to detect STDE according to the original method [5]. STDE was defined as clusters of fractionated or non‐fractionated electrograms with time and space dispersion (Figure 1). The STDE evaluations were performed by two experienced electrophysiologists to ensure consensus and reproducibility. We analyzed the electrograms visually rather than using automated or AI‐based methods. A bipolar voltage of < 0.2 mV during AF was considered as low voltage in the STDE group [8]. In the control group (LVZ group), a voltage map was created during sinus rhythm or atrial pacing after intracardiac electrocardioversion using a BeeAT catheter. The LVZ was defined as ≤ 0.5 mV during sinus rhythm or atrial pacing. If STDEs and LVZs were found, all detected sites were strictly targeted for ablation. The PVI line was shifted towards the area of STDE/LVZ to enclose both PV and STDE/LVZ if STDE/LVZ was confirmed to be near the PV. When the targets were located on the posterior wall of the LA, isolation of the posterior wall was performed. The STDE/LVZ at the other sites was ablated focally to achieve homogenization of the region of interest. This focal ablation was performed with a target lesion index/ablation index of 4.5/400, covering the area of interest with an inter‐lesion distance of 7–8 mm. If the targets were located on the LA roof, we performed linear ablation on the LA roof to avoid roof‐dependent atrial tachycardia (AT). Furthermore, if the targets were adjacent to insulators, such as the PVI line and mitral annulus, additional ablation was performed between the targeted area and the PVI line or mitral annulus to prevent iatrogenic AT. The earliest propagation site in the LA was assessed by activation mapping during sinus rhythm or atrial pacing. When the LVZ or STDE was located at the earliest propagation site in the LA, additional ablation was not performed to avoid intra‐atrial conduction delays. In both additional ablations, the endpoint was defined as ablation of the area of interest regardless of whether the AF was terminated.

Figure 1.

(A) An example of spatio‐temporal dispersion electrogram. Variability in the potential between electrodes was observed within the cycle length of atrial fibrillation. (B) An example of STDE detection using the EnSite system. White tags indicate identified STDE sites. (C) An example of STDE detection using the CARTO system. Blue tags indicate identified STDE sites. (D) An example of the relationship between identified STDE sites and ablation lesions. Blue tags represent STDE sites, and other tags represent ablation lesions.

2.5. Post‐Ablation Mapping

After PVI and additional ablation, if AF persisted, electrical cardioversion was performed, with mapping during SR or atrial pacing. We confirmed the disappearance of potential in each pulmonary vein and exit block, and created a voltage map of the atrium. When AT was induced by atrial burst pacing, with or without isoproterenol provocation, additional mapping and ablation were performed to eliminate the induced AT.

2.6. Follow‐up After Ablation

Three months postoperatively, cardiac computed tomography was performed in all patients, if possible, and the presence of PV stenosis and LA volume were assessed. Antiarrhythmic agents were interrupted or uninterrupted during follow‐up at the physician's discretion. A Holter electrocardiogram analysis was performed at 3 months, 6 months, 1 year, and 2 years in all patients whenever possible to evaluate for the recurrence of any atrial arrhythmias, and additional electrocardiograms were performed as needed if the patient had symptoms. Recurrence was defined as any atrial tachyarrhythmia lasting > 30 s over a blanking period of 3 months.

2.7. Statistical Analysis

All statistical analyses were performed using R (ver. 3.6.3, The R Foundation for Statistical Computing, Vienna, Austria). Continuous variables are presented as the mean ± standard deviation or median and interquartile range [IQR] if not normally distributed and were compared using the unpaired t‐test or Mann‐Whitney U test. All categorical variables were expressed as raw numbers and percentages and analyzed using Fisher's exact test. Freedom from atrial tachyarrhythmia recurrence was estimated using the Kaplan‐Meier method and compared between groups using the log‐rank test. Furthermore, sensitivity analysis was performed focusing on the subgroup of patients who underwent additional ablation. A Cox proportional hazards regression analysis was performed to detect any significant predictors of atrial tachyarrhythmia recurrence (reported as hazard ratio [HR] with 95% confidence interval [CI]). p values of < 0.05 were considered to indicate statistical significance.

3. Results

3.1. Baseline Clinical Characteristics

Table 1 presents the baseline patient characteristics. We retrospectively enrolled 107 consecutive patients (median age, 71.0 [IQR 66.0–76.5] years; 79 male) with persistent AF who underwent their first catheter ablation at our hospital from July 2018 to March 2021. The 53 and 54 patients were assigned to the STDE and LVZ groups, respectively. There were no significant differences in age, sex, body mass index, left ventricular ejection fraction measured by transthoracic echocardiography, or LA volume on computed tomography between the two groups. The duration of AF was significantly longer in the STDE group, and the prevalence of hypertension and diabetes mellitus was higher in the LVZ group. The CHADS‐VASc and HAS‐BLED scores of the two groups did not differ to a statistically significant extent and there was no difference in the use of antiarrhythmic agents.

Table 1.

Baseline clinical characteristics.

	STDE (N = 53)	LVZ (N = 54)	p value
Age (years)	66.1 ± 11.0	67.3 ± 10.4	0.549
Male sex, n (%)	40 (75.5)	39 (72.2)	0.827
BMI (kg/m2)	24.9 ± 4.1	25.3 ± 3.7	0.618
Duration of AF (months)	14 [6, 36]	10 [4, 13]	0.027
History of AF (months)	23 [12, 60]	18 [10, 53]	0.420
Comorbid disease, n (%)
CHF	25 (48.1)	23 (42.6)	0.697
IHD	3 (5.9)	6 (11.1)	0.490
Hypertension	21 (40.4)	36 (66.7)	0.011
DM	11 (21.2)	24 (44.4)	0.013
CKD	29 (54.7)	24 (44.4)	0.336
CHADS‐VASc	2 [1, 3]	3 [1, 4]	0.210
HAS‐BLED	1 [1, 2]	2 [1, 2]	0.815
AAD at addmission, n (%)
Type Ⅰ	11 (21.2)	18 (33.3)	0.194
Type Ⅱ	40 (75.5)	39 (73.6)	1.0
Type Ⅲ	4 (7.7)	9 (16.7)	0.237
Type Ⅳ	22 (42.3)	22 (40.7)	1.0
Data of TTE
LVEF (%)	59 [52, 67]	61 [51, 67]	0.740
LA diameter (mm)	41.9 [38.0, 46.0]	43.8 [39.0, 46.8]	0.233
LA volume (CT) (mL)	156.5 [132.7, 174.8]	151.3 [114.6, 166.6]	0.127

Note: Data are presented as the mean ± SD, median [IQR], or n (%). Duration of AF, the longest sustained episode of AF; History of AF, the total history of AF since diagnosis.

Abbreviations: AAD, antiarrhythmogenic drug; AF, atrial fibrillation; BMI, body mass index; CHF, chronic heart failure; CKD, chronic kidney disease; CT, computed tomography; DM, diabetes mellitus; IHD, ischemic heart disease; IQR, interquartile range; LA, left atrium; LVEF, left ventricular ejection fraction; LVZ, low voltage zone; STDE, spatiotemporal dispersion electrogram; TTE, transthoracic echocardiography.

3.2. Ablation Procedure

Table 2 shows the details of the ablations in each group. The acute success rate of EPVI was 100% in the STDE group and 98% in the LVZ group. Sixteen patients (30.2%) in the STDE group and 32 patients (59.3%) in the LVZ group received only PVI (p = 0.004). Cavotricuspid isthmus linear ablation was more common in the LVZ group. Rather than empiric cavotricuspid isthmus linear ablation (5.7% vs. 14.8%, p = 0.202), ablation of clinically documented cavotricuspid isthmus dependent atrial flutter before or during the ablation procedure (1.9% vs. 16.7%, p = 0.016) was more frequently performed in the LVZ group. Additional ablations, including roof linear ablation and posterior wall isolation, were more frequently performed in the STDE group. The rates of non‐PV foci ablation and superior vena cava isolation in the two groups did not differ to a statistically significant extent. Ablation of non‐PV foci was performed in only two cases within the STDE group. These interventions were conducted for frequent, spontaneously occurring atrial premature contractions during the procedure; notably, no specific induction protocols, such as pharmacological provocation, were employed. Although the operation time was significantly longer in the STDE group, there was no difference in fluoroscopy time. There was no significant difference in the incidence of major complications between the two groups (STDE group: 5 cases vs. LVZ group: 2 cases, p = 0.270). In the STDE group, the complications included three cases of transient ST‐segment elevation suspected to be due to air embolism, all of which resolved spontaneously without intervention, and two cases of cardiac tamponade. In the LVZ group, two cases of cardiac tamponade were observed.

Table 2.

Details of ablation in each group.

	STDE (N = 53)	LVZ (N = 54)	p value
EnSite, n (%)	37 (69.8)	26 (48.1)	0.031
Successful PVI, n (%)	53 (100)	53 (98.1)	1
Only PVI, n (%)	14 (26.4)	27 (50.0)	0.017
CTI linear ablation, n (%)	4 (7.5)	17 (31.5)	0.011
Only PVI or PVI + CTI, n (%)	16 (30.2)	32 (59.3)	0.004
Additional ablation, n (%)
Focal LVZ	2 (3.8)	2 (3.7)	1
Focal STDE	29 (54.7)	NA	NA
Non PV foci	2 (3.8)	0	0.196
SVC isolation	3 (5.7)	0	0.079
Linear ablation	29 (54.7)	20 (37.0)	0.029
Anterior line	0	2 (3.7)	0.503
Roof line	29 (54.7)	18 (33.3)	0.006
BOX isolation	24 (45.3)	12 (22.2)	0.014
Operation time, (min)	246.7 ± 61.6	206.0 ± 67.5	0.002
Fluoroscopy time, (min)	34.7 ± 12.7	35.9 ± 13.6	0.642
Major complications, n (%)	5 (9.5)	2 (3.7)	0.270

Note: Data are presented as the mean ± SD, or n (%). Major complications include cardiac tamponade, air embolism, major bleeding, stroke, and injury to the phrenic nerve and esophagus.

Abbreviations: CTI, cavotricuspid isthmus; LVZ, low‐voltage zone; PVI, pulmonary vein isolation; STDE, spatiotemporal dispersion electrogram; SVC, superior vena cava.

In the LVZ group, five cases underwent only cavotricuspid isthmus linear ablation, and six cases only the roof line ablation for preventing iatrogenic roof‐dependent AT due to proximity to the left and right posterior PVI lines. These 11 cases did not have any LVZs. Therefore, LVZ‐guided ablation was actually performed in 16 cases (29.6%).

3.3. Distribution and Amplitude of STDEs

Out of 53 patients in the STDE group, 5 were examined for STDE not only in the LA but also in the right atrium (RA). RA mapping was performed by the operator's discretion due to the enlarged RA or no termination of AF following PVI plus STDE ablation in the left atrium. STDE in the LA was identified in 37 patients (69.8%) and in the RA in 4 patients (80%). In addition, the distribution of STDE was classified into six sites: the posterior wall, anterior wall, roof, and bottom in the LA, the free wall in the RA, and the LA/RA septum. The LA posterior wall (n = 26) and the LA anterior wall (n = 25) were common locations for STDE. STDE was obtained at the LA roof in 9 cases, LA bottom in 12 cases, RA free wall in 1 case, and LA/RA septum in 11 cases. For the LVZ group, the distribution of LVZ included the LA anterior wall in 10 cases, posterior wall in 9, roof in 3, and 1 case each in the LA bottom and septum. Notably, empirical BOX isolation was performed in 3 cases within the LVZ group.

An additional analysis was performed on 28 cases, in which a detailed retrospective analysis of STDE points was possible. The average voltage of all STDE points (981 points) during AF was 0.91 ± 0.61 mV. Bipolar voltage of < 0.2 mV, 0.2–0.5 mV, and > 0.5 mV was observed in 68 (6.9%), 229 (23.3%), and 684 (69.7%) points, respectively.

3.4. Results During the Follow‐up Period

There were 5 dropouts before 24 months in the STDE group and 10 in the LVZ group, with a median follow‐up of 13 [12–19] months and 12 [11–13] months respectively.

The AF/AT free survival rate at twenty‐4 months follow‐up was 76.3% (95% confidence interval [CI] 62.1%–85.8%) in the STDE group and 58.4% (95% CI: 43.8–70.4%) in the LVZ group; the log‐rank test revealed that the rate was significantly higher in the STDE group (p = 0.027, Figure 2). The multivariate Cox proportional regression analysis demonstrated that AF duration, LA volume, and additional STDE ablation were independently associated with AF/AT recurrence after ablation (Table 3). The associations between the STDE group and AF/AT recurrence were observed in the multivariate analysis, even when adjusting for other factors such as mapping system, additional ablation and CKD (Table 4). When assessing AF and AT recurrence separately, the recurrence of AT did not differ significantly between the two groups (STDE group, n = 2; LVZ group, n = 4; 3.8% vs. 7.4%, p = 0.691). In contrast, the prevalence of AF recurrence was lower in the STDE group than that in the LVZ group.

Figure 2.

Kaplan‐Meier curves show post‐procedural atrial fibrillation (AF)/atrial tachycardia (AT) free rates in the STDE and LVZ groups.

Table 3.

Cox proportional regression analysis.

Variables	Univariate analysis			Multivariate analysis
	HR	95% CI	p value	HR	95% CI	p value
Age	1.00	0.97–1.04	0.941
Sex (male)	1.03	0.48–2.20	0.943
CHF	1.36	0.69–2.67	0.371
AF duration per year	1.25	1.07–1.45	0.005	1.37	1.16–1.61	< 0.001
LA volume per 10 mL	1.10	1.01–1.19	0.029	1.11	1.02–1.20	0.012
STDE group	0.47	0.23–0.95	0.034	0.29	0.12–0.66	0.004

Abbreviations: AF, atrial fibrillation; CI, confidence interval; CHF, chronic heart failure; HR, hazard ratio; LA, left atrium; STDE, spatio‐temporal dispersion electrogram.

Table 4.

Cox proportional regression analysis.

Variables	Model 1 HR (95% CI)	p value	Model 2 HR (95% CI)	p value	Model 3 HR (95% CI)	p value
STDE group	0.33 (0.15–0.72)	0.005	0.34 (0.15–0.74)	0.006	0.30 (0.14–0.64)	0.002
AF duration per year	1.36 (1.15–1.60)	< 0.001	1.38 (1.16–1.64)	< 0.001	1.40 (1.18–1.66)	< 0.001
Mapping system (EnSite)	0.91 (0.45–1.81)	0.779
Additional ablation			0.86 (0.40–1.89)	0.712
CKD					1.53 (0.75–3.12)	0.241

Abbreviations: CI, confidence interval; CKD, chronic kidney disease; HR, hazard ratio; STDE, spatio‐temporal dispersion electrogram.

LA volume reduction at 3 months postoperatively was greater in the STDE group than in the LVZ group (34.6 [19.3–79.9] ml vs. 13.7 [0.0–23.6] mL, p < 0.001). There was no significant difference in the prescription rate of antiarrhythmic agents, including the Na channel blockers, amiodarone and bepridil between the two groups (28.3% vs. 24.1%, p = 1.0). In a subgroup analysis restricted to patients not receiving antiarrhythmic agents, the STDE group showed a significantly higher AF/AT free survival rate compared with the LVZ group (log‐rank p = 0.004, Figure S1). Redo ablation was performed in 28 patients (STDE group, n = 12; LVZ group, n = 16).

3.5. Subgroup Analysis in Patients With Additional Ablation

We analyzed 37 cases in the STDE group and 22 cases in the LVZ group who underwent additional ablation, excluding those with only CTI linear ablation. At the twenty‐4 months follow‐up, the AF/AT free rates were 78.0% (95% CI: 0.61–0.88) in the STDE group and 53.7% (95% CI: 0.31–0.72) in the LVZ group. The log‐rank test showed that the AF/AT free rate was significantly higher in the STDE group than in the LVZ group (p = 0.043, Figure 3). In a subgroup analysis restricted to patients who underwent only PVI or CTI ablation, no significant difference in the non‐recurrence rate was observed between the two groups. However, the STDE group showed a trend toward a higher non‐recurrence rate compared with the LVZ group (Figure S2).

Figure 3.

Kaplan‐Meier curves from subgroup analysis of the STDE and LVZ groups divided into PVI only and additional ablation.

4. Discussion

4.1. Main Findings

The main findings of this study were as follows: (i) STDE was detected in approximately two‐thirds of the patients with persistent AF; (ii) PVI plus STDE ablation for persistent AF reduced the cumulative recurrence rate during follow‐up periods of up to 24 months in comparison to PVI plus LVZ ablation; and (iii) the number of additional ablations, including linear ablation, posterior wall isolation, and focal ablation, was significantly higher in the STDE group.

4.2. Clinical Impact of STDE Ablation on Reducing AF Recurrence

Catheter ablation for persistent AF has a limited effect, and the STAR AF II trial reported that additional staged ablation, such as continuous fractionated atrial electrograms (CFAE) and linear ablation, did not improve outcomes [9]. On the other hand, the EARNEST‐PVI trial suggested that additional ablation, including CFAE and linear ablation, may be effective [10]. The recent improvements in the durability of PVI due to proficiency in technique and advances in devices could raise the significance of additional ablation targeting the substrate of AF.

Our findings showed that the addition of STDE ablation was independently associated with a decreased risk of AF/AT recurrence in the multivariate analysis, suggesting that STDE may be an effective treatment for persistent AF. In this study, significantly more focal ablation, linear ablation, and LA posterior wall isolation were performed in the STDE group, which may have contributed to the improved outcomes. Our findings showed that most of the STDE‐positive points had a preserved bipolar voltage amplitude. This suggests that STDE‐based substrate detection is possible even in cases without an LVZ. Furthermore, even when focusing solely on the subgroup that underwent additional ablation as beyond PVI, the STDE group showed a lower rate of AF/AT recurrence compared to the LVZ group (Figure 3). These results suggest that STDEs may be superior to LVZs in detecting AF substrates that are eligible for ablation. The clinical benefit observed in the STDE group is a synergistic effect of superior substrate identification and subsequent targeted ablation. The lack of statistical significance in subgroup analysis restricted to patients who underwent only PVI or CTI ablation might be attributed to the limited sample size in this subgroup. Additionally, it may reflect the high specificity of the STDE‐guided approach, which successfully identified a small subset of patients with genuinely low atrial remodeling who required no further substrate modification, whereas the historical LVZ‐guided control group might have underestimated the potential substrates (Figure S2). In fact, previous research showed that AF/AT recurrence after ablation was higher in patients with LVZs than in those without [11] and PVI plus LVZ‐guided ablation did not show a clinical impact on reducing AF recurrence in comparison to PVI alone [12, 13, 14]. ERASE‐AF trial [15], which showed a modest benefit of additional ablation based on LVZ, had a major limitation: approximately 70% of patients in both arms had no LVZ and were treated with PVI only. Therefore, the difference in the primary endpoint in the ERASE‐AF trial should be interpreted with caution. Additionally, our study revealed no significant difference in PV reconduction during the redo session between the two groups. These findings indicated that the durability of PVI did not affect the difference in AF recurrence between the STDE and LVZ groups.

The TAILORED AF trial showed that artificial intelligence‐guided ablation of STDE in addition to PVI was able to suppress the recurrence of AF in comparison to PVI alone [6]. The results of the present study were in line with the TAILORED AF trial. In contrast, AT recurrence was increased in the artificial intelligence‐guided STDE ablation group in the TAILORED AF trial. In our study, there was no significant difference in AT recurrence between the STDE and LVZ groups, despite the STDE group undergoing more additional ablation than the LVZ group. As mentioned in the Methods section, we have been attempting to avoid creating an iatrogenic isthmus of AT during STDE ablation. This additional ablation strategy beyond PVI may contribute to suppressing the recurrence of AT. The TAILORED AF trial demonstrated significant superiority of STDE‐guided ablation in the prespecified subgroup of patients with persistent AF that had sustained for more than 6 months. Our findings were in line with the TAILORED AF trial because the duration of AF in the STDE group was 14 (6–36) months in the STDE group.

Spatial and temporal dispersion are not included in the definition of the CFAE, so STDE and CFAE are not identical. CFAE only examines the local fractionation, whereas STDE is considered superior for detecting critical substrates associated with the rotor. Furthermore, in our previous study using EnSite for STDE detection, a fractionation count > 25 was the cut‐off for STDE positivity [16]. Converting the CFAE definition of CL < 120 ms to fractionation count yields approximately 8–16. This result suggests that STDE often detects regions with even higher frequency than CFAE and the area of STDE would be smaller than that of CFAE.

4.3. Study Limitations

The present study was limited by several factors. First, this was a single‐center retrospective and non‐randomized study with a small sample size. This study compared two adjunctive ablation strategies (STDE vs. LVZ) based on our historical cohorts. Because PVI combined with LVZ ablation was the standard clinical strategy at our institution during the baseline period, historical data for patients undergoing PVI only were extremely limited. Consequently, a direct comparison with patients undergoing PVI alone was not conducted, and the incremental benefit of additional ablation over isolated PVI remains to be further elucidated. Second, the lack of stringency in follow‐up, including the rare use of an implantable cardiac monitor, and the fact that three patients dropped out within 12 months after the ablation, may underestimate the recurrence of atrial arrhythmias. Third, as the LVZ group underwent ablation earlier than the STDE group, there was possibility of unavoidable era effects, including changes in mapping systems, advances in workflow, catheter technology and lesion delivery. This also suggests that improvement in the operator's skill may have contributed to the better outcome of the STDE group, although there was no significant difference in the use of antiarrhythmic agents and the prevalence of PV reconduction in the redo procedure. Fourth, we could not assess the effect of STDE‐guided additional ablation on AF termination because intracardiac cardioversion was performed before starting PVI in most patients. Fifth, we did not consider STDEs consisting of all non‐fractionated potentials as STDEs because it was difficult for us to visually assess them. This fact may result in a lower prevalence of additional STDE ablation beyond PVI in this study compared to the TAILORED AF trial. Therefore, further prospective multicenter randomized controlled trials are required to confirm and enhance our results. Finally, the identification of STDE sites may involve subjective judgment. In the present study, all evaluations were performed by two experienced electrophysiologists to ensure consensus and reproducibility. Nevertheless, the potential for inter‐observer variability cannot be entirely excluded, and further standardized, automated algorithms may be required to enhance objectivity.

5. Conclusion

Additional ablation of PVI targeting STDE may be effective in patients with persistent AF during mid‐term follow‐up.

Funding

The authors have nothing to report.

Ethics Statement

This study was approved by the institutional review board (approval number #22‐083).

Consent

The requirement of written informed consent was waived due to the retrospective nature of this study project.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Figure S1: Kaplan‐Meier curves from subgroup analysis of patients without antiarrhythmic agents. Figure S2: Kaplan‐Meier curves from subgroup analysis of patients who underwent PVI or PVI plus CTI ablation only.

Data Availability Statement

The data that support the findings of this study are available from the corresponding author or the first author upon reasonable request.