ABSTRACT
Background
Substrate mapping for atrial fibrillation (AF) ablation commonly relies on sinus rhythm (SR) maps, which may overlook rate‐dependent electrophysiological changes.
Objective
This study aimed to evaluate the diagnostic and prognostic value of rate‐dependent substrate characterization using high‐density omnipolar mapping under rapid pacing.
Methods
In a prospective cohort of 46 patients (5 controls, 22 paroxysmal AF [PAF], 19 persistent AF [PeAF]), bi‐atrial conduction velocity (CV) and low‐voltage zones (LVZ) were assessed across seven mapping conditions: SR and graded pacing (600, 400, 300 ms) from the distal coronary sinus (CSD) and high right atrium (HRA).
Results
CV decreased with faster pacing across all cohorts. At 300 ms, PeAF patients exhibited significantly slower CV than PAF and controls (0.60 ± 0.07 vs. 0.72 ± 0.06 vs. 0.90 ± 0.05 m/s; p < 0.001). The global LVZ burden was greater in PeAF than PAF at 300 ms (58.71 [39.84–68.69] vs. 26.18 [20.94–37.40]; p = 0.009). Decremental pacing was associated with progressive bi‐atrial conduction slowing in AF, most pronounced in PeAF (0.76 ± 0.08 vs. 0.74 ± 0.06 vs. 0.69 ± 0.02 vs. 0.60 ± 0.07 m/s; p < 0.001). Rapid pacing at 300 ms showed the numerically highest discriminative capacity. Patients with global CV < 0.67 m/s (AUC 0.749; p = 0.019) or LVZ% > 43.4% (AUC 0.681; p = 0.008) at 300 ms were associated with a higher risk of recurrence.
Conclusion
Rapid pacing at 300 ms reveals latent bi‐atrial substrate occult during SR and may improve risk stratification.
Keywords: ablation, atrial fibrillation, conduction velocity, high‐resolution mapping, low‐voltage zone, pacing cycle length, recurrence
Abbreviations
- AF
atrial fibrillation
- CSD
distal coronary sinus
- CV
conduction velocity
- HRA
high right atrium
- LA
left atrial
- LVZs
low‐voltage zones
- PCL
pacing cycle length
- PVI
pulmonary vein isolation
- RA
right atrial
- SR
sinus rhythm
- 术语
定义
1. Introduction
Pulmonary vein isolation (PVI) remains the cornerstone of rhythm control for atrial fibrillation (AF) [1]. However, durable success—particularly in persistent AF (PeAF)—depends on accurately defining and addressing extra‐pulmonary atrial substrate [2, 3]. High‐density electroanatomic mapping enables fine‐grained characterization of left atrial (LA) voltage and conduction, and low‐voltage zones (LVZs) are widely used as surrogates for atrial fibrosis [4, 5, 6]. Yet substrate mapping is commonly performed in sinus rhythm (SR) or at pacing cycle lengths (PCLs) chosen at operator discretion (often ≥ 600 ms) [7, 8, 9, 10]. This raises a key clinical question: does the pacing rate during mapping materially change substrate characterization, and does it affect outcomes?
Atrial electrophysiology is inherently rate‐dependent. Remodeled atrial tissue can exhibit impaired conduction reserve: shorter PCLs unmask lower voltages, fractionated electrograms, and conduction slowing that appear normal at longer cycle lengths [11, 12, 13]. Prior studies have reported PCL‐dependent reductions in CV and voltage [14], but most used conventional bipolar catheters. Traditional bipolar mapping depends on wavefront directionality; misalignment can produce artificially low voltages and inaccurate local activation times, confounding true remodeling with measurement artifact [15, 16]. As a result, the full extent of dynamic substrate—functional abnormalities revealed only under stress—has been poorly defined.
Omnipolar technology with high‐density grid catheters addresses these limitations by resolving local electric fields independent of catheter orientation, enabling more precise discrimination between healthy and diseased tissue [16, 17, 18]. This capability motivates a systematic re‐evaluation of PCL dependency using modern systems, transforming PCL from a passive setting into an active diagnostic lever.
We performed high‐density omnipolar mapping in a prospective cohort spanning control, PAF, and PeAF to: (1) quantify PCL‐dependent changes in voltage and CV by comparing baseline (SR/600 ms) versus rapid (600, 400, and 300 ms) pacing; (2) characterize dynamic substrate features unique to PeAF; and (3) evaluate whether dynamic indices—such as CV and LVZ burden at rapid pacing—better predict post‐ablation recurrence than baseline metrics.
2. Methods
2.1. Study Population and Design
This single‐center, prospective observational study was conducted at West China Hospital, Sichuan University (Chengdu, China). From August 2024 to November 2025, we prospectively enrolled 59 patients who underwent electrophysiological study (EPS) and ablation due to access pathway, paroxysmal and persistent AF. All participants provided written informed consent for the use of anonymized data. The study protocol was approved by the Ethics Committee of West China Hospital.
Exclusion criteria included: (1) contraindications to catheter ablation (e.g., active infection, coagulopathy, or severe comorbidity); (2) prior AF or atrial arrhythmia ablation (to avoid procedure‐induced substrate modification); (3) significant structural heart disease (moderate‐to‐severe valvular disease, congenital heart disease, systolic heart failure, prior myocardial infarction); and (4) inadequate LA mapping density precluding predefined spatial coverage and quality control, e.g. rapid pacing‐induced hemodynamic instability, etc.
Baseline demographic, clinical, and echocardiographic data were extracted from electronic medical records: age, sex, body mass index, AF type/duration, comorbidities, medications, LA diameter, and left ventricular ejection fraction.
Following intracardiac electrophysiological evaluation and transseptal access, patients underwent PVI only to avoid ablation‐induced substrate modification that could confound mapping analyses. Electrical cardioversion was performed for PeAF patients who did not restore SR after PVI. All patients completed seven mapping sessions: SR and constant pacing from the CSD and HRA at PCLs of 600, 400, and 300 ms. Post‐procedure follow‐up was 12 months, with prospective assessment of atrial arrhythmia recurrence (≥ 30 s).
2.2. Radiofrequency Ablation Procedure
After the EPS, transseptal puncture was performed under fluoroscopic and/or intracardiac echocardiographic guidance. Unfractionated heparin was administered immediately after transseptal access, with activated clotting time maintained at 250–350 s and measured every 30 min during LA instrumentation.
In patients who underwent left‐sided access pathway ablation, LAT map‐guided ablation was performed to eliminate the access pathway.
In patients who underwent AF ablation, pulmonary venography delineated LA and pulmonary venous anatomy, followed by high‐resolution electroanatomic mapping and LA model construction. Radiofrequency ablation lesion sets were created as contiguous wide‐antral lesions around ipsilateral pulmonary veins to achieve circumferential isolation (PVI). No additional substrate modification was performed; electrical cardioversion was conducted if necessary to restore SR. The acute procedural endpoint was bidirectional PV isolation.
2.3. High‐Resolution Mapping Under Graded Rapid Pacing
High‐density LA and right atrial (RA) mapping were performed using the EnSite X EP System (Abbott, St. Paul, MN) with the Advisor HD Grid multipolar mapping catheter (Abbott). The proprietary omnipolar technology algorithm leverages triangular electrode cliques to compute, across 360°, the maximal local voltage vector independent of catheter orientation, yielding direction‐agnostic, high‐fidelity local signals. After restoration of SR and confirmation of PVI, a systematic mapping protocol interrogated electrophysiological properties across graded frequency loads. Geometry and data were collected under seven conditions: SR; CSD pacing at 600, 400, and 300 ms; and HRA pacing at 600, 400, and 300 ms.
Geometry and acquisition settings were kept identical across PCLs within each patient. Mapping commenced after steady‐state capture, with the pacing catheter position confirmed unchanged. Automated point collection imposed strict beat‐acceptance criteria: (i) cycle length stability within ± 5 ms; (ii) stable secondary reference local activation time; (iii) consistent local electrogram morphology; (iv) catheter movement below a prespecified speed threshold with respiration gating; and (v) absence of premature beats or fusion.
Mapping points within 3 mm of the atrial geometry shell were retained. Points with poor contact, excessive noise, or far‐field contamination were excluded. Minimum total point counts and regional density thresholds were enforced per PCL; if unmet, the PCL was re‐mapped until criteria were satisfied or excluded from primary analysis.
To differentiate native pathological substrate from ablation‐induced lesions, data points within pulmonary veins, appendages, or valve annuli were excluded. Low‐voltage areas strictly confined to ablation lines (PV antrum) created during PVI were excluded from substrate burden calculations to evaluate underlying atrial myopathy.
CV was automatically computed by the EnSite X system in omnipolar mode per prior methods. For each atrial region under each mapping strategy, 20 valid points were randomly selected; the regional mean CV was calculated from these values. Global CV was defined as the mean value of local conduction velocities obtained from all analyzable points in both the left and right atria. Preliminary analyses confirmed no significant differences attributable to pacing location, indicating that conduction anisotropy did not affect CV measurements. Given the lack of a universally established consensus for “slow conduction” in predicting prognosis, a data‐driven approach was adopted to define the optimal threshold. Specifically, receiver operating characteristic (ROC) curve analyses were performed to evaluate the discriminative capacity of CV for AF recurrence, and the optimal cut‐off value was determined using the Youden index (maximum vertical distance between the ROC curve and the diagonal line) to stratify patients into low CV and high CV groups.
LVZs were defined as bipolar voltage ≤ 0.5 mV. The same threshold was applied across SR and all paced mapping conditions. Using the EnSite X area measurement function, regions corresponding to ≤ 0.5 mV were outlined and their area calculated. LVZ incidence was defined as the proportion of patients with any detectable LVZ under a given pacing condition. Among patients with detectable LVZ, LVZ burden was quantified as the percentage of total mapped atrial surface area occupied by LVZ (LVZ%) (Figure 1).
Figure 1.
Representative high‐density maps of conduction velocity and voltage substrate across different pacing cycle lengths. Representative maps from control and AF patients arranged in three rows showing Conduction Velocity (Top) and Voltage (Bottom) during Sinus Rhythm, CSD 600 ms, 400 ms, and 300 ms (left to right). During mapping, the pulmonary veins and valve annuli were cut out at the level of the lesion set and excluded to ensure normalization of geometry and surface area. The maps demonstrate a significant progressive increase in low‐voltage zones (≤ 0.5 mV) and global conduction slowing in response to each subsequent pacing strategy in AF patients, highlighting the critical impact of pacing cycle length on the potential atrial area defined as substrate.
Regional Segmentation: The atria were segmented into nine regions (LA: posterior, anterior, roof, septum, lateral, inferior; RA: posterior, anterior, septal) for regional assessment.
2.4. Statistical Analysis
Continuous variables are reported as mean ± standard deviation (SD) or median (interquartile range [IQR]) as appropriate. Normality was assessed using Shapiro–Wilk, and homogeneity of variance with Levene's test. Categorical variables are presented as n (%).
Within‐subject changes in CV across PCLs were first assessed using repeated‐measures ANOVA when parametric assumptions (including sphericity) were satisfied, or the Friedman test otherwise. When the overall test was significant, planned pairwise comparisons between SR and each paced condition were performed using paired t‐tests or Wilcoxon signed‐rank tests, with multiplicity adjustment. LVZ incidence across PCLs was compared using Cochran's Q test, with exact McNemar tests for pairwise comparisons versus SR. Because LVZ burden (LVZ%) was predefined as the percentage of mapped atrial surface occupied by LVZs among patients with detectable LVZ, LVZ% was summarized as median (IQR]) in LVZ‐positive patients and compared across pacing conditions using Kruskal–Wallis tests; prespecified pairwise comparisons versus SR were performed using Mann–Whitney U tests with Holm correction. For between‐group comparisons at a given pacing condition, LVZ incidence was compared using Fisher's exact test and LVZ% among LVZ‐positive patients was compared using Mann–Whitney U tests. Categorical variables used the chi‐square test or Fisher's exact test, as appropriate. Because the control cohort was small, exact Mann–Whitney U tests were additionally performed as sensitivity analyses for the key control‐related pairwise comparisons of global bi‐atrial CV. Comparisons between CSD and HRA were performed using paired t‐tests.
ROC curves assessed the discriminative ability of CV and LVZ% at different pacing modes for predicting arrhythmia recurrence. Area under the curve (AUC) was calculated; optimal cut‐offs were determined by the Youden index (sensitivity + specificity − 1).
Arrhythmia‐free survival was estimated by Kaplan–Meier analysis. Patients were stratified by ROC‐derived cut‐offs (not tertiles), and survival distributions compared using the log‐rank test. Follow‐up duration was calculated from the index procedure, with a standard 3‐month blanking period.
Two‐sided P‐values < 0.05 were considered statistically significant. Analyses were performed using R (version 4.3.0 or higher) and SPSS (version 26.0).
3. Results
3.1. Baseline Characteristics
Fifty‐nine patients were prospectively enrolled from August 2024 to November 2025, in which 54 with AF scheduled for radiofrequency catheter ablation guided by the EnSite X three‐dimensional electroanatomic mapping system (Abbott), and five controls with a left‐sided accessory pathway undergoing elective ablation. Thirteen AF patients were excluded after enrollment (either withdrew or did not meet the inclusion criteria). Forty‐six were included in the final analysis: 5 controls, 22 PAF, and 19 PeAF (Figure 2). All subjects completed mapping across SR, 600 ms, 400 ms, and 300 ms. Patients presenting AF during peri‐mapping phase were cardioverted to SR to complete the designated extensive mapping process, including either sustained AF post‐PV‐isolation or induced AF due to mapping/rapid‐pacing. No significant procedure‐related complications occurred within 30 days post‐procedure, and all patients completed the 12‐month follow‐up.
Figure 2.
Study design and flowchart. CSD, distal coronary sinus; HRA, high right atrium; PAF, paroxysmal atrial fibrillation; PeAF, persistent atrial fibrillation; PVI, pulmonary vein isolation; SR, sinus rhythm.
Table 1 summarizes baseline characteristics. Mean age was 61.8 ± 9.0 years; 60.9% were male. Baseline characteristics, including age, sex, body mass index, and comorbidities, were comparable among the study groups. LA diameter differed significantly across groups (31.8 ± 2.7 vs. 36.7 ± 5.3 vs. 42.1 ± 3.0 mm; p < 0.001). Tukey‐adjusted post hoc comparisons showed that LA diameter was larger in PeAF than in PAF (p < 0.001) and controls (p < 0.001), whereas the difference between PAF and controls did not reach statistical significance after multiple‐comparison adjustment (p = 0.064). All antiarrhythmics were discontinued for at least five half‐lives prior to the procedure, except Amiodarone.
Table 1.
Baseline characteristics.
| ALL (n = 46) | Control (n = 5) | PAF (n = 22) | PeAF (n = 19) | P value | |
|---|---|---|---|---|---|
| Recurrence | 13 (28.3%) | 0 (0.00) | 6 (27.3) | 7 (36.8) | 0.411 |
| Male | 28 (60.9%) | 3 (60.0) | 11 (50.0%) | 14 (73.7%) | 0.346 |
| Age, year | 61.76 ± 9.03 | 55.20 ± 12.28 | 63.18 ± 9.61 | 61.84 ± 6.93 | 0.206 |
| BMI, kg/m2 | 25.21 [22.90–27.08] | 26.00 [25.00–28.00] | 24.81 [22.29–26.21] | 26.54 [24.37–27.86] | 0.114 |
| Smoking | 15 (32.6%) | 2 (40.0) | 5 (22.7) | 8 (42.1) | 0.444 |
| Drinking | 11 (23.9%) | 1 (20.0) | 4 (18.2) | 6 (31.6) | 0.700 |
| Symptoms | 24 (52.2%) | 1 (20.0) | 11 (50.0%) | 12 (63.2%) | 0.233 |
| LA, mm | 38.4 ± 5.41 | 31.8 ± 2.68 | 36.7 ± 5.26 | 42.1 ± 3.03 | < 0.001 |
| RA, mm | 39.0 [34.0–46.0] | 35.0 [34.0–36.0] | 35.0 [33.0–40.5] | 46.0 [42.0–49.0] | < 0.001 |
| LVEF, % | 66.0 [61.0, 70.0] | 71.0 [69.0, 72.0] | 66.0 [61.2, 69.8] | 64.0 [60.0, 69.5] | 0.131 |
| Hypertension | 17 (37.0%) | 1 (20.0) | 8 (36.4) | 8 (42.1) | 0.685 |
| Diabetes mellitus | 5 (10.9%) | 2 (40.0) | 1 (4.55) | 2 (10.5) | 0.120 |
| Coronary heart disease | 6 (13.0%) | 1 (20.0) | 3 (13.6) | 2 (10.5) | 0.841 |
| Prior TIA/Stroke | 1 (2.17%) | 0 (0.00) | 1 (4.55) | 0 (0.00) | 1.000 |
| Heart failure | 7 (15.2%) | 0 (0.00) | 2 (9.09) | 5 (26.3) | 0.332 |
| Class I/III AADs | 4 (8.70%) | 0 (0.00) | 2 (9.09) | 2 (10.5) | 1.000 |
| ACEi or ARB | 12 (26.1%) | 1 (20.0) | 5 (22.7) | 6 (31.6) | 0.889 |
| Beta‐blocker | 15 (32.6%) | 1 (20.0) | 9 (40.9) | 5 (26.3) | 0.595 |
| CCB | 12 (26.1%) | 0 (0.00) | 8 (36.4) | 4 (21.1) | 0.241 |
Note: Continuous variables are reported as mean ± standard deviation (SD) for normally distributed data or median [IQR] for non‐normally distributed data. Categorical variables are presented as frequencies and percentages (n [%]). For comparisons among the three groups, one‐way Analysis of Variance (ANOVA) was performed for continuous variables that met the assumptions of normality and homogeneity of variance. For non‐normally distributed variables, the Kruskal–Wallis test was used. Categorical variables were compared using Fisher's exact test (two‐sided).
Abbreviations: AADs, antiarrhythmic drugs; ACEi, angiotensin converting enzyme inhibitors; ARB, angiotensin receptor blockers; BMI, body mass index; CCB, calcium channel blockers; LA, left atrium; RA, right atrium; TIA, transient ischemic attack.
3.2. Characterization of Atrial Substrate Across AF Types
3.2.1. Validation of Mapping Strategy
Preliminary analysis evaluated pacing site effects (HRA vs. CSD). Consistent with omnipolar technology's direction‐agnostic capability, global CV and LVZ burden did not differ significantly between pacing sites across cycle lengths (Table S1). Subsequent analyses focus on CSD pacing, which provided stable capture in all subjects.
3.2.2. Global and Regional Conduction Velocity
Mean bi‐atrial CV differed significantly among groups during SR (0.95 ± 0.05 vs. 0.82 ± 0.07 vs. 0.76 ± 0.08 m/s; overall p < 0.001). Pairwise comparisons showed significantly lower CV in PAF (p = 0.002) and PeAF (p < 0.001) versus Control, and in PeAF versus PAF (p = 0.045). Consistently, CV decreased with faster pacing, with the largest inter‐group differences observed at 300 ms (0.90 ± 0.05 vs. 0.72 ± 0.06 vs. 0.60 ± 0.07 m/s; overall p < 0.001). Notably, all pairwise comparisons remained statistically significant at this cycle length (all p < 0.001). PeAF exhibited slower conduction than PAF and controls across all pacing modes (Figure 3). Exact Mann‐Whitney U sensitivity analyses for global bi‐atrial CV showed consistent control‐related pairwise differences across SR, 600, 400, and 300 ms (Supporting Information S1: Table S5).
Figure 3.
Conduction velocity and low‐voltage zone across pacing conditions in control, PAF, and PeAF patients. Box plots showing bi‐atrial conduction velocity and percentage of low‐voltage zones (LVZ%) across four pacing conditions: sinus rhythm, CSD 600 ms, CSD 400 ms, and CSD 300 ms. Differences in conduction velocity and LVZ% between groups varied across pacing conditions. The line plots below show the trends of conduction velocity (left) and LVZ% (right) across pacing conditions for each group. Conduction velocity is presented as mean ± standard deviation, and LVZ% as median [IQR] among patients with detectable LVZ; *p < 0.05, **p < 0.01, ***p < 0.001. CSD, distal coronary sinus; LVZ%, percentage of low‐voltage zones; PAF, paroxysmal atrial fibrillation; PeAF, persistent atrial fibrillation.
Regional analysis highlighted marked disparities in the LA anterior and posterior walls and the RA posteroinferior region (Figure 4). Significant overall differences were observed in the LA roof (p = 0.047) and RA (p = 0.009). Specifically, PAF maintained relatively preserved CV in the LA roof (0.84 ± 0.15 vs. 1.01 ± 0.08 m/s; p = 0.082) and RA mean (0.76 [0.68–0.78] vs. 0.80 [0.78–0.80] m/s; p = 0.11) compared with controls, whereas PeAF showed significant slowing in these regions (Roof: 0.82 ± 0.16 vs. 1.01 ± 0.08 m/s; p = 0.038; RA mean: 0.66 [0.62–0.73] vs. 0.80 [0.78–0.80] m/s; p = 0.013). At 300 ms, CV differences were significant across most regions (Table 2).
Figure 4.
Segmental comparison of conduction velocity at different pacing cycle length. The bar charts show mean and standard error values. LA, left atrial; RA, right atrial.
Table 2.
Comparison of conduction velocities across control, PAF, and PeAF groups.
| Pacing | Region | Control | PAF | PeAF | P1 value | P2 value | P3 value | P4 value |
|---|---|---|---|---|---|---|---|---|
| SR | LA mean, m/s | 1.02 (0.99–1.04) | 0.88 (0.82–0.92) | 0.81 (0.72–0.88) | < 0.001 | 0.007 | < 0.001 | 0.088 |
| Anterior, m/s | 1.07 ± 0.09 | 0.88 ± 0.15 | 0.83 ± 0.16 | 0.013 | 0.042 | 0.009 | 0.555 | |
| Posterior, m/s | 1.09 (1.09–1.11) | 0.88 (0.80–0.92) | 0.78 (0.73–0.89) | 0.005 | 0.026 | 0.004 | 0.204 | |
| Inferior, m/s | 0.99 (0.98–1.11) | 0.88 (0.77–0.98) | 0.82 (0.76–0.88) | 0.052 | 0.124 | 0.045 | 0.343 | |
| Roof, m/s | 1.01 ± 0.08 | 0.84 ± 0.15 | 0.82 ± 0.16 | 0.047 | 0.082 | 0.038 | 0.82 | |
| Lateral, m/s | 0.97 ± 0.08 | 0.86 ± 0.16 | 0.80 ± 0.14 | 0.077 | 0.307 | 0.071 | 0.411 | |
| LA septum, m/s | 0.94 ± 0.10 | 0.86 ± 0.17 | 0.76 ± 0.17 | 0.041 | 0.573 | 0.073 | 0.119 | |
| RA mean, m/s | 0.80 (0.78–0.80) | 0.76 (0.68–0.78) | 0.66 (0.62–0.73) | 0.009 | 0.11 | 0.013 | 0.08 | |
| Anterolateral, m/s | 0.86 ± 0.19 | 0.81 ± 0.18 | 0.77 ± 0.16 | 0.498 | 0.871 | 0.552 | 0.651 | |
| RA septum, m/s | 0.90 (0.87–0.90) | 0.78 (0.69–0.90) | 0.68 (0.60–0.77) | 0.033 | 0.373 | 0.088 | 0.088 | |
| Posteroinferior, m/s | 0.73 ± 0.09 | 0.60 ± 0.10 | 0.54 ± 0.12 | 0.005 | 0.044 | 0.003 | 0.24 | |
| Bi‐atrial global, m/s | 0.95 ± 0.05 | 0.82 ± 0.07 | 0.76 ± 0.08 | < 0.001 | 0.002 | < 0.001 | 0.045 | |
| CSD 600 ms | LA mean, m/s | 1.00 ± 0.07 | 0.84 ± 0.07 | 0.79 ± 0.06 | < 0.001 | < 0.001 | < 0.001 | 0.048 |
| Anterior, m/s | 1.04 (1.01–1.11) | 0.83 (0.78–0.93) | 0.77 (0.76–0.80) | 0.002 | 0.058 | 0.002 | 0.058 | |
| Posterior, m/s | 1.04 (1.03–1.12) | 0.85 (0.73–0.94) | 0.76 (0.74–0.80) | 0.002 | 0.008 | 0.001 | 0.293 | |
| Inferior, m/s | 0.96 (0.90–1.09) | 0.79 (0.70–0.96) | 0.76 (0.72–0.86) | 0.09 | 0.107 | 0.107 | 0.911 | |
| Roof, m/s | 1.02 (0.95–1.09) | 0.78 (0.71–0.86) | 0.79 (0.69–0.88) | 0.033 | 0.037 | 0.032 | 0.705 | |
| Lateral, m/s | 0.98 (0.79–1.09) | 0.76 (0.71–0.96) | 0.74 (0.72–0.83) | 0.091 | 0.123 | 0.088 | 0.588 | |
| LA septum, m/s | 0.91 (0.85–1.06) | 0.78 (0.68–0.99) | 0.70 (0.66–0.80) | 0.04 | 0.216 | 0.051 | 0.216 | |
| RA mean, m/s | 0.82 ± 0.06 | 0.71 ± 0.09 | 0.66 ± 0.08 | < 0.001 | 0.029 | < 0.001 | 0.092 | |
| Anterolateral, m/s | 0.92 (0.76–0.92) | 0.73 (0.66–0.95) | 0.69 (0.64–0.77) | 0.256 | 0.588 | 0.362 | 0.588 | |
| RA septum, m/s | 0.90 (0.74–0.98) | 0.78 (0.68–0.85) | 0.65 (0.63–0.70) | 0.016 | 0.232 | 0.035 | 0.063 | |
| Posteroinferior, m/s | 0.73 (0.71–0.80) | 0.54 (0.50–0.64) | 0.45 (0.42–0.61) | 0.003 | 0.078 | 0.003 | 0.078 | |
| Bi‐atrial global, m/s | 0.94 ± 0.06 | 0.80 ± 0.07 | 0.74 ± 0.06 | < 0.001 | < 0.001 | < 0.001 | 0.022 | |
| CSD 400 ms | LA mean, m/s | 0.99 (0.93–1.03) | 0.80 (0.77–0.86) | 0.72 (0.70–0.73) | < 0.001 | 0.024 | < 0.001 | < 0.001 |
| Anterior, m/s | 1.05 (0.92–1.11) | 0.80 (0.77–0.88) | 0.71 (0.69–0.72) | < 0.001 | 0.049 | < 0.001 | < 0.001 | |
| Posterior, m/s | 1.03 (0.96–1.09) | 0.83 (0.72–0.88) | 0.73 (0.68–0.76) | < 0.001 | 0.024 | < 0.001 | 0.042 | |
| Inferior, m/s | 0.96 (0.91–1.06) | 0.75 (0.69–0.89) | 0.75 (0.67–0.80) | 0.012 | 0.039 | 0.009 | 0.287 | |
| Roof, m/s | 0.97 (0.92–1.05) | 0.76 (0.67–0.84) | 0.72 (0.64–0.76) | 0.003 | 0.045 | 0.003 | 0.083 | |
| Lateral, m/s | 0.97 (0.85–0.98) | 0.72 (0.66–0.93) | 0.69 (0.64–0.73) | 0.014 | 0.094 | 0.012 | 0.142 | |
| LA septum, m/s | 0.91 (0.84–1.04) | 0.70 (0.64–0.92) | 0.66 (0.63–0.71) | 0.014 | 0.118 | 0.014 | 0.118 | |
| RA mean, m/s | 0.81 ± 0.09 | 0.69 ± 0.08 | 0.62 ± 0.07 | < 0.001 | 0.009 | < 0.001 | 0.018 | |
| Anterolateral, m/s | 0.80 (0.74–0.95) | 0.71 (0.63–0.93) | 0.63 (0.62–0.73) | 0.123 | 0.395 | 0.161 | 0.395 | |
| RA septum, m/s | 0.87 (0.78–0.95) | 0.72 (0.67–0.77) | 0.63 (0.61–0.68) | < 0.001 | 0.143 | 0.002 | 0.003 | |
| Posteroinferior, m/s | 0.72 (0.70–0.79) | 0.54 (0.47–0.59) | 0.43 (0.40–0.57) | 0.002 | 0.055 | 0.003 | 0.055 | |
| Bi‐atrial global, m/s | 0.91 (0.90–0.95) | 0.78 (0.72–0.81) | 0.68 (0.68–0.69) | < 0.001 | 0.026 | < 0.001 | < 0.001 | |
| CSD 300 ms | LA mean, m/s | 0.95 ± 0.06 | 0.75 ± 0.07 | 0.63 ± 0.09 | < 0.001 | < 0.001 | < 0.001 | < 0.001 |
| Anterior, m/s | 1.00 ± 0.14 | 0.79 ± 0.14 | 0.64 ± 0.15 | < 0.001 | 0.013 | < 0.001 | 0.005 | |
| Posterior, m/s | 1.02 ± 0.12 | 0.80 ± 0.10 | 0.68 ± 0.12 | < 0.001 | < 0.001 | < 0.001 | 0.008 | |
| Inferior, m/s | 0.97 (0.87–0.99) | 0.74 (0.66–0.86) | 0.69 (0.48–0.77) | 0.002 | 0.04 | 0.001 | 0.056 | |
| Roof, m/s | 0.94 ± 0.12 | 0.76 ± 0.12 | 0.61 ± 0.15 | < 0.001 | 0.024 | < 0.001 | 0.004 | |
| Lateral, m/s | 0.92 ± 0.10 | 0.73 ± 0.14 | 0.63 ± 0.14 | < 0.001 | 0.021 | < 0.001 | 0.035 | |
| LA septum, m/s | 0.90 ± 0.09 | 0.71 ± 0.12 | 0.60 ± 0.14 | < 0.001 | 0.011 | < 0.001 | 0.03 | |
| RA mean, m/s | 0.76 (0.76–0.79) | 0.66 (0.62–0.71) | 0.54 (0.49–0.57) | < 0.001 | 0.02 | < 0.001 | 0.001 | |
| Anterolateral, m/s | 0.82 ± 0.08 | 0.73 ± 0.14 | 0.62 ± 0.14 | 0.004 | 0.38 | 0.012 | 0.026 | |
| RA septum, m/s | 0.86 ± 0.05 | 0.68 ± 0.14 | 0.57 ± 0.15 | < 0.001 | 0.043 | < 0.001 | 0.038 | |
| Posteroinferior, m/s | 0.70 ± 0.08 | 0.53 ± 0.14 | 0.42 ± 0.14 | < 0.001 | 0.038 | < 0.001 | 0.02 | |
| Bi‐atrial global, m/s | 0.90 ± 0.05 | 0.72 ± 0.06 | 0.60 ± 0.07 | < 0.001 | < 0.001 | < 0.001 | < 0.001 |
Note: Continuous variables are reported as mean +/‐ standard deviation (SD) for normally distributed data or median (interquartile range [IQR]) for non‐normally distributed data. Comparisons were performed using one‐way ANOVA with Tukey's post‐hoc test for normally distributed variables and the Kruskal–Wallis test with Dunn's post‐hoc test for non‐normally distributed variables.
Abbreviations: CSD, distal coronary sinus; LA, left atrium; PAF, paroxysmal atrial fibrillation; PeAF, persistent atrial fibrillation; RA, right atrium; P1 value = overall comparison (Control vs. PAF vs. PeAF); P2 value = Control vs.versus PAF; P3 value = Control vs.versus PeAF; P4 value = PAF versus PeAF.
3.2.3. Low Voltage Zone Characteristics
No LVZs were identified in controls. During SR, LVZs were observed in 22.7% (5/22) of PAF and 42.1% (8/19) of PeAF (p = 0.198). With 300 ms pacing, LVZ incidence increased to 31.8% (7/22) in PAF and 57.9% (11/19) in PeAF (p = 0.108) (Figure 3). Among patients with detectable substrate, LVZ% was similar between PAF and PeAF during SR (18.13 [15.01‐26.71] vs. 26.47 [20.39‐38.92]; p = 0.164) but was significantly larger in PeAF at 300 ms (58.71 [39.84‐68.69] vs. 26.18 [20.94‐37.40]; p = 0.009).
Regionally, LVZs were most commonly observed in the LA anterior and posterior walls. At 300 ms, PeAF showed a trend toward higher LVZ incidence in the LA posterior wall compared with PAF (47.4% vs. 27.3%; p = 0.211), consistent with more extensive extra‐PV substrate (Table S2). Figure 5 provides a descriptive illustration of the regional LVZ distribution across pacing conditions. Table 3
Figure 5.
Segmental distribution of low voltage zones at different pacing cycle length. Plots illustrate the regional distribution of LVZs across pacing conditions, expressed as the number of LVZ‐positive observations in each segment. LA, left atrial; LVZ, low‐voltage zone; RA, right atrial.
Table 3.
ROC curves and cut‐off points of CV and LVZ% for associations with AF recurrence.
| Pacing | AUC | Cutoff | Sensitivity | Specificity | Log‐rank P |
|---|---|---|---|---|---|
| Conduction velocity (m/s) | |||||
| Sinus rhythm | 0.703 | 0.85 | 0.615 | 0.821 | 0.042 |
| CSD 600 ms | 0.714 | 0.79 | 0.846 | 0.536 | 0.03 |
| CSD 400 ms | 0.721 | 0.73 | 0.923 | 0.464 | 0.032 |
| CSD 300 ms | 0.749 | 0.67 | 0.615 | 0.821 | 0.019 |
| Low voltage zone (%) | |||||
| Sinus rhythm | 0.639 | 27.29 | 0.462 | 0.893 | 0.244 |
| CSD 600 ms | 0.641 | 28.41 | 0.462 | 0.893 | 0.02 |
| CSD 400 ms | 0.663 | 37.1 | 0.538 | 0.821 | 0.045 |
| CSD 300 ms | 0.681 | 43.38 | 0.538 | 0.893 | 0.008 |
Abbreviations: AUC, area under the curve; CSD, distal coronary sinus; ROC, receiver operating characteristic; SR versus 300‐ms AUCs were compared using the DeLong test (Supplementary Table S6). Cutoffs were internally derived using the Youden index and should be considered exploratory.
3.3. Dynamics of Atrial Substrate Across PCLs
3.3.1. Global and Regional Conduction Velocity
In the control group, bi‐atrial conduction velocity remained stable across all pacing modes, as well as at SR. In contrast, both PAF and PeAF cohorts exhibited significant rate‐dependent slowing with decremental pacing (SR → PCL600 → PCL400 → PCL300): PAF decreased from 0.82 ± 0.07 → 0.80 ± 0.07 → 0.77 ± 0.06 → 0.72 ± 0.06 m/s; PeAF from 0.76 ± 0.08 → 0.74 ± 0.06 → 0.69 ± 0.02 → 0.60 ± 0.07 m/s (all p < 0.001).
Rate‐dependent slowing was greater in PeAF. Regionally, heterogeneous responses were observed: LA inferior, lateral, and septum were particularly sensitive to rapid pacing. PAF maintained stable conduction in the LA anterior, RA anterolateral, and RA posteroinferior regions, whereas PeAF exhibited significant reductions at 300 ms (LA anterior: 0.86 [0.66–0.95] → 0.69 [0.49–0.74] m/s; p = 0.004; RA anterolateral: 0.75 [0.69–0.80] → 0.60 [0.55–0.71] m/s; p = 0.016; RA posteroinferior: 0.56 [0.49–0.66] → 0.44 [0.31–0.55] m/s; p = 0.047) (Table S3).
3.3.2. Low Voltage Zone Characteristics
In PeAF, LVZ incidence showed an upward trend with shorter PCLs, although the overall within‐group change did not reach statistical significance (42.1% in SR, 42.1% at 600 ms, 47.4% at 400 ms, and 57.9% at 300 ms; p = 0.066). Among patients with detectable LVZ, LVZ burden differed significantly across pacing conditions in PeAF, with the highest burden observed at 300 ms (58.71 [39.84‐68.69] at 300 ms vs. 34.18 [32.33–55.19] at 400 ms vs. 25.93 [20.66–40.28] at 600 ms vs. 26.47 [20.39–38.92] in SR; p = 0.023). Pairwise comparisons versus SR showed a significant increase only at 300 ms (p = 0.045), whereas no significant differences were observed at 600 ms (p = 0.875) or 400 ms (p = 0.458). This expansion suggests a larger potential substrate burden under rapid pacing in PeAF. In PAF, LVZ incidence and LVZ burden also trended upward with shorter PCLs, although statistical significance was not reached (Table S4).
3.4. Prognostic Performance and Risk Stratification
ROC analyses showed numerically higher AUCs with faster pacing. Bi‐atrial mean CV at 300 ms showed the numerically highest discriminative capacity (AUC 0.749) compared with SR (0.703) and intermediate PCLs (600 ms: 0.714; 400 ms: 0.721). LVZ% at 300 ms similarly showed a higher AUC than baseline mapping in SR (AUC 0.681 vs. 0.639). At 300 ms, bi‐atrial mean CV < 0.67 m/s yielded specificity 82.1% and sensitivity 61.5%; LVZ% > 43.4% yielded specificity 89.3% and sensitivity 53.8%. Although CV sensitivity was highest at 400 ms (92.3%), the 300 ms mode offered a more balanced profile with superior specificity, making it a potentially informative pacing condition in this cohort. However, DeLong testing showed no significant difference between SR and 300‐ms AUCs for CV or LVZ% (Supporting Information S1: Table S6). These thresholds were internally derived and should be considered exploratory.
Kaplan–Meier analysis confirmed that patients with global CV < 0.67 m/s at 300 ms had significantly lower arrhythmia‐free survival (log‐rank p = 0.019). LVZ% predicted outcomes only under rapid pacing: SR LVZ% was not significant (log‐rank p = 0.244), while 300 ms LVZ% > 43.4% was associated with increased recurrence (log‐rank p = 0.008) (Figure 6).
Figure 6.
Freedom from recurrence by CV and LVZ. Kaplan–Meier estimates of recurrence‐free survival at 12 months stratified by conduction velocity (left) and LVZ% (right). Left: global conduction velocity cut‐off 0.67 m/s. Right: percentage of low‐voltage zones cutoff 43.4%. CV, conduction velocity; LVZs, low‐voltage zones; PVI, pulmonary vein isolation.
4. Discussion
4.1. Major Findings
Using high‐density omnipolar mapping under rapid pacing, we demonstrate: (1) a pathological gradient with slower bi‐atrial conduction and greater LVZ burden in PeAF versus PAF and controls; (2) rapid pacing at 300 ms can unmask latent abnormalities, consistent with rate‐dependent substrate remodeling; and (3) higher prognostic value for post‐ablation recurrence during rapid pacing than during SR, particularly for global CV < 0.67 m/s and LVZ% > 43.4% at 300 ms.
4.2. Progressive Atrial Remodeling
AF appears to reflect a progressive atrial cardiomyopathy. Electrical remodeling—slowed CV and reduced voltage—parallels clinical progression [19, 20, 21]. Slowed conduction fosters localized activation delays, wavefront fragmentation, and reentry, facilitating AF maintenance [22]. LVZs serve as electrophysiological surrogates for structural remodeling, i.e. fatty infiltration, fibrosis, etc [23]. Our high‐density, anatomically segmented mapping reveals bi‐atrial involvement, including the RA and LA roof, consistent with diffuse myopathy [24] and explaining the limited success of PVI‐only strategies in PeAF [25, 26].
4.3. Mechanisms of Dynamic Changes
Rapid pacing expands abnormal substrate by exhausting conduction reserve and revealing functional block [14, 27, 28, 29, 30]. Healthy tissue maintains CV at high rates via rapid sodium channel recovery and strong cell coupling; remodeled tissue exhibits steep CV restitution and fails at high rates [14, 30]. Additionally, heterogeneous fibrosis can appear as normal voltage during SR but fractionates at rapid rates, attenuating amplitude below the voltage cut‐off [14]. This rate‐dependent “dynamic substrate” has been described previously and is relevant to AF onset/maintenance [27, 28, 29].
4.4. Role of Omnipolar Technology
Omnipolar technology minimizes directional bias that confounds conventional bipolar measurements [24, 27]. By computing maximal local voltage vectors across 360°, omnipolar technology provides orientation‐independent signals with improved consistency, noise immunity, and spatial precision [31, 32, 33, 34]. This methodological approach enhances the robustness of our data.
4.5. Prognostic Value and Clinical Implications
Static SR mapping is insufficient for risk stratification [24]. Incorporating rapid pacing (300 ms) into routine substrate evaluation may help identify a higher‐risk phenotype (global CV < 0.67 m/s or LVZ% > 43%) warranting closer monitoring and informing future studies of tailored intervention [14]. Whether dynamic LVZ regions represent effective targets for substrate modification remains untested and warrants prospective evaluation [35].
5. Limitations
This single‐center study with modest sample size may limit generalizability and multivariable analyses of rare subgroups. The control cohort was small (n = 5) and should be interpreted as an invasive reference cohort rather than a definitive normative population, although exact Mann‐Whitney U sensitivity analyses supported the main control‐related global CV comparisons. In addition, the ROC‐derived cutoffs and AUC estimates were generated within the same cohort and may therefore be optimistic. We did not correlate dynamic LVZs with histology or MRI, and inter‐modality divergence has been reported. Follow‐up was limited to 12 months.
6. Conclusion
Rapid 300 ms pacing during high‐density omnipolar mapping reveals latent bi‐atrial substrate, including conduction slowing and functional low voltage, that may be occult during SR. Dynamic substrate burden may improve post‐ablation risk stratification compared with baseline mapping and may help inform future validation and interventional studies. Whether targeting these dynamic abnormalities improves outcomes remains to be determined.
Ethics Statement
The study protocol was approved by the Ethics Committee of West China Hospital, Sichuan University ((No. 2022‐Review (383)).
Consent
Written informed consent was obtained from all patients.
Conflicts of Interest
The authors declare no conflicts of interest.
Supporting information
Supporting File
Acknowledgments
We thank our hospital's clinical engineers for their support. We also acknowledge Yan Huo for technical support related to the mapping system. This work was supported by the National Natural Science Foundation of China (Grant No. 82570391), the Natural Science Foundation of Sichuan (Grant No. 2025ZNSFSC0714), and a research grant from Abbott for the OCTOPUS‐AF study (Grant No. EP‐2025‐000723).
Wang X., Chan J., Jia R., et al., “Substrate Unmasking via High‐Density Omnipolar Mapping Under Rapid Pacing in Atrial Fibrillation,” Journal of Cardiovascular Electrophysiology 37 (2026): 1482‐1495. 10.1111/jce.70372.
Substrate Unmasking in Atrial Fibrillation
Xinmeng Wang and Juan Chan contributed equally.
Contributor Information
Jiong Li, Email: lijionghh@scu.edu.cn.
Kaijun Cui, Email: cuikaijun@scu.edu.cn.
Data Availability Statement
Data underlying this article are available on reasonable request to the corresponding author.
References
- 1. Calkins H., Hindricks G., Cappato R., et al., “2017 HRS/EHRA/ECAS/APHRS/SOLAECE Expert Consensus Statement on Catheter and Surgical Ablation of Atrial Fibrillation,” Heart Rhythm: the Official Journal of the Heart Rhythm Society 14 (2017): e275–e444. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Bentley R., Logantha S. J. R. J., Sharma P., Rainbow R. R., and Lip G. Y. H., “Pathophysiological Insights Into Atrial Fibrillation: Revisiting the Electrophysiological Substrate, Anatomical Substrate, and Possible Insights From Proteomics,” Cardiovascular Research 117 (2021): e41–e45. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Kottkamp H., “On the Atrial Fibrillation Substrate,” Journal of the American College of Cardiology 74 (2019): 1348–1351. [DOI] [PubMed] [Google Scholar]
- 4. Sohns C. and Marrouche N. F., “Atrial Fibrillation and Cardiac Fibrosis,” European Heart Journal 41 (2020): 1123–1131. [DOI] [PubMed] [Google Scholar]
- 5. Oakes R. S., Badger T. J., Kholmovski E. G., et al., “Detection and Quantification of Left Atrial Structural Remodeling With Delayed‐Enhancement Magnetic Resonance Imaging in Patients With Atrial Fibrillation,” Circulation 119 (2009): 1758–1767. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Mahnkopf C., Badger T. J., Burgon N. S., et al., “Evaluation of the Left Atrial Substrate in Patients With Lone Atrial Fibrillation Using Delayed‐Enhanced MRI: Implications for Disease Progression and Response to Catheter Ablation,” Heart Rhythm: The Official Journal of the Heart Rhythm Society 7 (2010): 1475–1481. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Kottkamp H., Berg J., Bender R., Rieger A., and Schreiber D., “Box Isolation of Fibrotic Areas (BIFA): A Patient‐Tailored Substrate Modification Approach for Ablation of Atrial Fibrillation,” Journal of Cardiovascular Electrophysiology 27 (2016): 22–30. [DOI] [PubMed] [Google Scholar]
- 8. Verma A., Jiang C., Betts T. R., et al., “Approaches to Catheter Ablation for Persistent Atrial Fibrillation,” New England Journal of Medicine 372 (2015): 1812–1822. [DOI] [PubMed] [Google Scholar]
- 9. Yamaguchi T., Tsuchiya T., Nagamoto Y., et al., “Long‐Term Results of Pulmonary Vein Antrum Isolation in Patients With Atrial Fibrillation: An Analysis in Regards to Substrates and Pulmonary Vein Reconnections,” Europace: European Pacing, Arrhythmias, and Cardiac Electrophysiology: Journal of the Working Groups on Cardiac Pacing, Arrhythmias, and Cardiac Cellular Electrophysiology of the European Society of Cardiology 16 (2014): 511–520. [DOI] [PubMed] [Google Scholar]
- 10. Yang G., Zheng L., Jiang C., et al., “Circumferential Pulmonary Vein Isolation Plus Low‐Voltage Area Modification in Persistent Atrial Fibrillation,” JACC: Clinical Electrophysiology 8 (2022): 882–891. [DOI] [PubMed] [Google Scholar]
- 11. Heida A., van der Does W. F. B., van Staveren L. N., et al., “Conduction Heterogeneity,” JACC: Clinical Electrophysiology 6 (2020): 1844–1854. [DOI] [PubMed] [Google Scholar]
- 12. Prabhu S., Voskoboinik A., McLellan A. J. A., et al., “Biatrial Electrical and Structural Atrial Changes in Heart Failure,” JACC: Clinical Electrophysiology 4 (2018): 87–96. [DOI] [PubMed] [Google Scholar]
- 13. Thijssen V., “Structural Remodelling During Chronic Atrial Fibrillation: Act of Programmed Cell Survival,” Cardiovascular Research 52 (2001): 14–24. [DOI] [PubMed] [Google Scholar]
- 14. Sekihara T., Oka T., Ozu K., Yoshida A., and Sakata Y., “Pacing Cycle Length‐Dependent Electrophysiologic Changes in Left Atrium: Poor Validity of Using Low‐Voltage Area and Slow Conduction Area Under Specific Pacing Cycle Length as Absolute Substrates of Atrial Fibrillation,” Heart Rhythm: The Official Journal of the Heart Rhythm Society 22 (2025): 1659–1666. [DOI] [PubMed] [Google Scholar]
- 15. Takigawa M., Relan J., Martin R., et al., “Effect of Bipolar Electrode Orientation on Local Electrogram Properties,” Heart Rhythm: the Official Journal of the Heart Rhythm Society 15 (2018): 1853–1861. [DOI] [PubMed] [Google Scholar]
- 16. Takigawa M., Relan J., Kitamura T., et al., “Impact of Spacing and Orientation on the Scar Threshold With a High‐Density Grid Catheter,” Circulation: Arrhythmia and Electrophysiology 12 (2019): e007158. [DOI] [PubMed] [Google Scholar]
- 17. Dittrich S., Scheurlen C., van den Bruck J. H., et al., “The Omnipolar Mapping Technology‐A New Mapping Tool to Overcome “Bipolar Blindness” Resulting in True High‐Density Maps,” Journal of Interventional Cardiac Electrophysiology 67 (2024): 399–408. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Haldar S. K., Magtibay K., Porta‐Sanchez A., et al., “Resolving Bipolar Electrogram Voltages During Atrial Fibrillation Using Omnipolar Mapping,” Circulation: Arrhythmia and Electrophysiology 10 (2017): e005018. [DOI] [PubMed] [Google Scholar]
- 19. Frontera A., Pagani S., Limite L. R., et al., “Slow Conduction Corridors and Pivot Sites Characterize the Electrical Remodeling in Atrial Fibrillation,” JACC: Clinical Electrophysiology 8 (2022): 561–577. [DOI] [PubMed] [Google Scholar]
- 20. Huo Y., Gaspar T., Pohl M., et al., “Prevalence and Predictors of Low Voltage Zones in the Left Atrium in Patients With Atrial Fibrillation,” EP Europace 20 (2018): 956–962. [DOI] [PubMed] [Google Scholar]
- 21. Kurata N., Masuda M., Kanda T., et al., “Left Atrial Localized Low‐Voltage Areas Indicate Whole Left Atrial Electrophysiological Degeneration in Atrial Fibrillation Patients,” Circulation Journal 86 (2022): 192–199. [DOI] [PubMed] [Google Scholar]
- 22. Qi D., Guan X., Liu X., Liu L., Liu Z., and Zhang J., “Slow Conduction Velocity Predicts Atrial Fibrillation Recurrence After Radiofrequency Ablation,” Journal of Cardiovascular Electrophysiology 35 (2024): 461–468. [DOI] [PubMed] [Google Scholar]
- 23. Yang G., Yang B., Wei Y., et al., “Catheter Ablation of Nonparoxysmal Atrial Fibrillation Using Electrophysiologically Guided Substrate Modification During Sinus Rhythm After Pulmonary Vein Isolation,” Circulation: Arrhythmia and Electrophysiology 9 (2016): e003382. [DOI] [PubMed] [Google Scholar]
- 24. Starek Z., Di Cori A., Betts T. R., et al., “Baseline Left Atrial Low‐Voltage Area Predicts Recurrence After Pulmonary Vein Isolation: WAVE‐MAP AF Results,” Europace 25 (2023): euad194. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Chen J., Arentz T., Cochet H., et al., “Extent and Spatial Distribution of Left Atrial Arrhythmogenic Sites, Late Gadolinium Enhancement at Magnetic Resonance Imaging, and Low‐Voltage Areas in Patients With Persistent Atrial Fibrillation: Comparison of Imaging Vs. Electrical Parameters of Fibrosis and Arrhythmogenesis. Europace: European Pacing, Arrhythmias, and Cardiac Electrophysiology: Journal of the Working Groups on Cardiac Pacing, Arrhythmias, and Cardiac Cellular Electrophysiology of the,” European Society of Cardiology 21 (2019): 1484–1493. [DOI] [PubMed] [Google Scholar]
- 26. Eichenlaub M., Mueller‐Edenborn B., Minners J., et al., “Comparison of Various Late Gadolinium Enhancement Magnetic Resonance Imaging Methods With High‐Definition Voltage and Activation Mapping for Detection of Atrial Cardiomyopathy. Europace: European Pacing, Arrhythmias, and Cardiac Electrophysiology: Journal of the Working Groups on Cardiac Pacing, Arrhythmias, and Cardiac Cellular Electrophysiology of the,” European Society of Cardiology 24 (2022): 1102–1111. [DOI] [PubMed] [Google Scholar]
- 27. Wong G. R., Nalliah C. J., Lee G., et al., “Dynamic Atrial Substrate During High‐Density Mapping of Paroxysmal and Persistent AF,” JACC: Clinical Electrophysiology 5 (2019): 1265–1277. [DOI] [PubMed] [Google Scholar]
- 28. Gong Y., Xie F., Stein K. M., et al., “Mechanism Underlying Initiation of Paroxysmal Atrial Flutter/Atrial Fibrillation by Ectopic Foci: A Simulation Study,” Circulation 115 (2007): 2094–2102. [DOI] [PubMed] [Google Scholar]
- 29. Williams S. E., Linton N., O'Neill L., et al., “The Effect of Activation Rate on Left Atrial Bipolar Voltage in Patients With Paroxysmal Atrial Fibrillation,” Journal of Cardiovascular Electrophysiology 28 (2017): 1028–1036. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30. Lalani G. G., Schricker A., Gibson M., Rostamian A., Krummen D. E., and Narayan S. M., “Atrial Conduction Slows Immediately Before the Onset of Human Atrial Fibrillation,” Journal of the American College of Cardiology 59 (2012): 595–606. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31. Deno D. C., Balachandran R., Morgan D., Ahmad F., Masse S., and Nanthakumar K., “Orientation‐Independent Catheter‐Based Characterization of Myocardial Activation,” IEEE Transactions on Biomedical Engineering 64 (2017): 1067–1077. [DOI] [PubMed] [Google Scholar]
- 32. Massé S., Magtibay K., Jackson N., et al., “Resolving Myocardial Activation With Novel Omnipolar Electrograms,” Circulation: Arrhythmia and Electrophysiology 9 (2016): e004107. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33. Jiang R., Beaser A. D., Aziz Z., Upadhyay G. A., Nayak H. M., and Tung R., “High‐Density Grid Catheter for Detailed Mapping of Sinus Rhythm and Scar‐Related Ventricular Tachycardia,” JACC: Clinical Electrophysiology 6 (2020): 311–323. [DOI] [PubMed] [Google Scholar]
- 34. Masuda M., Asai M., Iida O., et al., “Left Atrial Voltage Mapping With a Direction‐Independent Grid Catheter: Comparison With a Conventional Circular Mapping Catheter,” Journal of Cardiovascular Electrophysiology 30 (2019): 2834–2840. [DOI] [PubMed] [Google Scholar]
- 35. Schmidt B., Bordignon S., Metzner A., et al., “Ablation Strategies for Repeat Procedures in Atrial Fibrillation Recurrences Despite Durable Pulmonary Vein Isolation: The Prospective Randomized ASTRO AF Multicenter Trial,” Circulation 150 (2024): 2007–2018. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Supporting File
Data Availability Statement
Data underlying this article are available on reasonable request to the corresponding author.