BACKGROUND Pulmonary vein (PV) isolation is a well-established treatment for atrial fibrillation (AF), however, strategies for patients with recurrent AF and isolated PVs remain elusive. OBJECTIVE This study aimed to evaluate the effectiveness of a personalized artificial intelligence (AI)-guided spatiotemporal dispersion mapping and ablation in patients undergoing a repeat ablation and whose PVs remain isolated from previous ablation procedures. METHODS The Re-Ablation Using a Tailored Approach Targeting EGM-Dispersion (RESTART) trial (NCT05477147) was an interventional, prospective, single-arm, multicenter clinical trial. Patients with previous catheter or surgical ablation for paroxysmal, persistent, or long-standing persistent AF and with documented symptomatic AF recurrences were enrolled. Patients with documented reconnected PVs during the procedure were prospectively withdrawn from the trial. Follow-up included 3-, 6-, and 12-month visits with 12-lead electrocardiogram and 24-hour Holter. The primary endpoint of the study was freedom from documented AF at 12 months after a single AI-guided repeat ablation procedure. RESULTS Of the 213 patients enrolled, the main causes for premature study exit were PV reconnection (n = 80) or noninducibility of AF (n = 14). The primary endpoint was achieved in 83% of patients (77 of 93), and freedom from any atrial arrhythmia after a single procedure was 70% (65 of 93). Notably, AF termination during ablation was achieved in 54% of patients, and right atrial dispersion was identified in 59%. The procedure was associated with improvements in quality-of-life metrics (AF Effect on Quality-of-Life questionnaire and 36-Item Short Form Survey scores). CONCLUSION In patients with documented isolated PVs, AI-guided dispersion ablation is safe and effective and represents an alternative to other ablative strategies.

BACKGROUND:The TAILORED-AF randomized trial demonstrated that artificial intelligence-guided ablation of spatiotemporal dispersion in addition to pulmonary vein isolation significantly reduced atrial fibrillation (AF) recurrence at 1-year follow-up in patients with persistent AF. Still, approximately one-third of the population experienced atrial tachycardia (AT) recurrences, most of which were successfully ablated. OBJECTIVE:In this post hoc analysis of the TAILORED-AF trial, we determined the exact mechanism and locations of ATs and examined the adherence to protocol-recommended ablation and its relation to the occurrence of ATs. METHODS:During index procedures, post-AF ATs were diagnosed based on common mapping and pacing maneuvers. Maps pertaining to activation and ablation lesions were evaluated by a panel of experts to assess compliance and completion of the TAILORED-AF protocol recommendations. RESULTS:During the index procedure, 140 ATs (96 patients) were characterized including 76 macroreentrant ATs (29% perimitral, 42% peritricuspid, 20% roof dependent, 9% other), 37 localized ATs, 12 ATs with undetermined mechanisms, and 15 unstable ATs. The percentage of repeat ablations was significantly lower in patients who had a complete ablation set in common macroreentry AT regions (roof, mitral) ablation than patients with an incomplete ablation set (22% vs 48%; P = .02; odds ratio 0.31; 95% confidence interval 0.12-0.79). CONCLUSION:Most ATs observed after artificial intelligence-guided, dispersion-based catheter ablation are targetable. The prevalence of these ATs and the repeat procedure rate are reduced when strategic completion of dispersion-based ablation was obtained.

Substrate ablation strategies in addition to pulmonary vein isolation (PVI) for the maintenance of sinus rhythm (SR) are still debated. Targeting low voltage areas (LVA) in addition to PVI may represent an efficient strategy for the ablation of persistent atrial fibrillation (AF). SCAR-AF study (ablation of LVA during sinus rhythm) was a multicenter, prospective, randomized trial, evaluating the effect of LVA ablation in addition to PVI for persistent AF on SR maintenance. From September 2019 to August 2021, patients with de novo persistent AF were included in the study. After LA mapping guided by a 3D mapping system, patients were divided according to the presence or absence of LVA. Patients without LVA were treated by PVI only (Group A). Patients with LVA were randomized to PVI only (Group B) or PVI + ablation of LVA (Group C). LVA was defined as voltage mapping with bipolar atrial voltage <0.5 mV. The primary endpoint was freedom from atrial arrhythmias, after a single ablation procedure. A total of 211 patients (Sex male: 73%, Mean Age 63.8+/-9.3 years, CHADS-VASC 2.1, long standing AF 33.5%). After 18 months FU, atrial-arrhythmia-free survival did not differ significantly between the 3 groups, 79% in Group A, 75.7% in Group B, 73.1% in Group C (Group A vs Group B: HR: 1.28; 95% CI: 0.64-2.55, p = .48, Group B vs Group C: HR 95%CI: 0.67-2.45; p = 0.45). Multivariate analysis showed that presence of LVA was associated with age (years) (HR 1.11 CI 1.06 – 1.16, p < 0.001) and inversely correlated with BMI (kg/m²) (HR 0.93, CI 0.87 – 0.99, p 0.029) and current smoking. In this randomized trial, PVI plus ablation of LVA did not significantly improved outcomes in patients with persistent AF. LVA may represent a marker of severe atrial cardiomyopathy.

Recently, the TAILORED-AF randomized clinical trial (NCT04702451) demonstrated superiority of targeting dispersion in addition to PVI identified by an artificial intelligence (AI) software over a PVI-only strategy in persistent atrial fibrillation (AF). Ablation at the left atrial anterior wall is common with this ablation approach. Atrial contractility after such an ablation approach has never been investigated. To investigate the left atrial contractility after a TAILORED-AF ablation approach performed for persistent AF. We conducted a single-center, non-randomized retrospective study of persistent AF patients who underwent a first-time catheter ablation in stable sinus rhythm. All patients underwent the TAILORED-AF approach of AI-guided spatiotemporal dispersion ablation in addition to PVI. First, electro-anatomical biatrial mapping was performed and the AI-guided spatiotemporal dispersion software was employed to detect dispersion area in real time intracardiac electrograms. Then, PVI and dispersion ablation were performed with the open-irrigated contact force-sensing radiofrequency ablation catheter using high-power short-duration ablation. All patients underwent a transmitral flow evaluation using pulse-wave doppler in stable sinus rhythm at 22 ± 10 months. Of the 24 patients included (age: 69.73±8.4 years; female: 26%; body mass index: 28.03±4.32 kg/m²; left ventricular ejection fraction: 59±11%; cardiomyopathy: 21%; CHADS2-VASc: 1.89+-1.33, All patients had persistent AF (52% had long-standing persistent AF) and thus had no atrial contraction. Ablation at the anterior wall was performed in all patients with an extensive ablation in 9/19 (47%) and a complete anterior line in 7/19 (37%). All patients were in stable sinus rhythm without symptoms at 22 ± 10 months follow up. In all patients we observed an A wave recovery during the evaluation with an A wave velocity equal to 0.5±0.2 m/s and an E/A ratio equal to 1.8±0.4 (figure below). This study demonstrates that despite an ablation at anterior wall in all patients, the recovery of left atrial contraction after TAILORED-AF procedures was observed in 100% of the patients.

AIMS:There is still conflicting evidence if women with persistent atrial fibrillation (AF) profit from a pulmonary vein isolation (PVI) plus strategy. We evaluated the efficacy of a spatio-temporal dispersion-targeted ablation strategy in women from the TAILORED-AF trial. METHODS AND RESULTS:In TAILORED-AF, 370 patients were randomised to either a personalised, artificial intelligence (AI)-guided tailored ablation or to PVI-only. AF substrate mapping data and 12-month ablation outcomes were compared between women and men. Overall, 21% patients were female (70.4 ± 6.9 vs. 64.5 ± 8.5 years for men, P < 0.001). While spatio-temporal dispersion extent was similar between groups, left atrial low-voltage surface area (<0.2 mV) was significantly larger in women (P < 0.01). In women, the single-procedure freedom from AF (76% vs. 50%, log-rank P < 0.001) and any atrial arrhythmia (56% vs. 38%, log-rank P < 0.05) were significantly superior to PVI alone with a tailored procedure. In the PVI-only group, the single-procedure freedom from AF (50% vs. 70%, log-rank P < 0.001) and any atrial arrhythmia (38% vs. 61%, log-rank P < 0.001) were significantly lower in women. After a tailored ablation, no significant differences were observed between women and men regarding freedom from AF (76% vs. 91%, log-rank P = 0.07) or any atrial arrhythmia (56% vs. 62%, log-rank P = 0.69) free survival. CONCLUSION:Compared to men, PVI-only in women with persistent AF leads to a significantly lower freedom from atrial arrhythmia. A personalised spatio-temporal dispersion-targeted ablation strategy led to a higher rate of freedom from any atrial arrhythmia than standard PVI after a single procedure in women and comparable outcomes between women and men. REGISTRATION IDENTIFICATION:clinicaltrials.gov NCT04702451.

Recently, the TAILORED-AF randomized clinical trial (NCT04702451) demonstrated superiority of ablating persistent atrial fibrillation (AF) with the targeting of dispersion — identified by an artificial intelligence (AI) software — in addition to PVI vs. a PVI-only strategy. Previously, a contact force–sensing catheter optimized for temperature-controlled, very high-power short-duration (vHPSD) radiofrequency (RF) ablation demonstrated safety and efficacy for AF ablation. To evaluate the feasibility and procedural efficiency of using a vHPSD catheter for an electrogram-based, artificial intelligence-guided patient-tailored persistent AF ablation approach. We conducted a pilot single-center, non-randomized study of persistent AF patients who underwent a first-time catheter ablation. All patients underwent a TAILORED-AF-like approach consisting of an AI-guided spatiotemporal dispersion ablation in addition to PVI. First, electro-anatomical biatrial mapping was obtained while an AI-guided spatiotemporal dispersion software was employed to highlight dispersion regions in real time. Then, PVI and dispersion ablation were performed with the vHPSD catheter using region-adapted maximal power (posterior wall, 90W and anterior wall, 50W). Procedural characteristics and acute outcomes of a 25-patient cohort (vHPSD cohort) were then compared to a sub-group of patients enrolled in the TAILORED-AF trial who were randomized to dispersion-based ablation with a variety of non-vHPSD ablation catheters (non- vHPSD cohort). Of the 25 patients included in the vHPSD cohort (age: 72±8 years; female: 44%; body mass index: 27±5 kg/m²; left ventricular ejection fraction: 50%; cardiomyopathy: 44%; CHADS2-VASc: 2 [1.5-3.0]), 72% had persistent AF and 28% had long-standing persistent AF and no major complication was observed. AF termination occurred in 21/25 patients (84%) in the vHPSD cohort and in 110/170 patients (65%) in the non-vHPSD cohort (p=0.06). Procedure, RF and fluoroscopy times were significantly shorter in the vHPSD vs. in the non-vHPSD TAILORED cohort (127±32 min vs 183±61 min, p<0.001; 23±12 vs 43±17, p<0.001; 4±3 vs 10±10, p<0.001). This pilot study demonstrates that the use of a vHPSD catheter is safe and efficient to follow a spatiotemporal dispersion-based ablation procedure. Notably, the use of a vHPSD catheter allows for a significant reduction in TAILORED-AF-like procedure times.

BACKGROUND:Targeting low-voltage areas (LVAs) in addition to pulmonary vein isolation (PVI) can improve ablation outcome in persistent atrial fibrillation (AF). OBJECTIVE:SCAR-AF was a multicenter, prospective, randomized trial, evaluating LVA ablation plus PVI for persistent AF. METHODS:Patients with de novo persistent AF were recruited (9 referral centers in France) into the SCAR-AF study (September 2019 to August 2021). Patients without LVA were treated with PVI alone (PVI-), and those with LVA were randomized to either PVI alone (PVI+) or PVI plus LVA ablation (PVI + LVA). The primary outcome was freedom from atrial arrhythmias (AF/atrial tachycardia) after a single procedure. RESULTS:A total of 211 patients (73.5% men, mean ± standard deviation age 63.8 ± 9.3 years, CHA2DS2-VASc score 2.1, long-standing AF 44.5%) were included. At 18-month follow-up, the incidence of atrial-arrhythmia-free survival did not differ between groups (79.0% in PVI-, 75.7% in PVI+, and 73.1% in PVI + LVA; PVI- vs PVI+, hazard ratio (HR) 1.28, 95% confidence interval [CI] 0.64-2.55, P = .48; PVI+ vs PVI + LVA, HR 1.28; 95% CI 0.67-2.45, P = .45). On multivariable analysis, presence of LVA was associated with advancing age (HR 1.11, 95% CI 1.06-1.16, P < .001) and was inversely correlated with body mass index (HR 0.93, 95% CI 0.87-0.99, P = .029) and smoking. CONCLUSION:In this randomized trial, PVI plus LVA ablation did not improve outcomes in patients with persistent AF. LVA may represent a marker of atrial cardiomyopathy, but its presence does not seem to be an effective target in persistent AF.

BACKGROUND Spatiotemporal dispersion-guided ablation is a tailored approach for patients in persistent atrial fi brillation (PsAF). The characterization of dispersion extent and distribution and its association with common clinical descriptors of PsAF patients has not been studied. OBJECTIVES Artificial fi cial intelligence-adjudicated dispersion extent and distribution (AI-DED) was obtained with a machine/deep learning classifier fi er (VX1 Software, Volta Medical) in PsAF patients undergoing ablation. The purpose of this study was to test the hypothesis that AI-DED is unique to each patient and independent of common procedural and clinical parameters. METHODS In a subanalysis of the Ev-AIFib study (NCT03434964), spatiotemporal dispersion maps were built with VX1 software in 78 consecutive persistent and long-standing PsAF patients. AI-DED was quantified fi ed using 2 distinct approaches (visual regional characterization or automated global quantification fi cation of AI-DED). RESULTS AI-DED paired-subregion Euclidean distance measurements between 78 patients (average distance 5.07 +/- 0.60; min 2.23; max 9.75) demonstrate that AI-DED is a patient-unique characteristic of PsAF. Importantly, both AF type and AF history do not correlate with AI-DED levels (R2 2 = 0.006, P = .53; and R2 2 = 0.03, P = .25, respectively). The most extensive AI-DED levels are not associated with poorer procedural (83%, 81%, and 83% of AF termination in low, medium, and high dispersion groups, respectively; P = .954) and long-term (88%, 75%, and 91% of freedom from AF/atrial tachycardia after multiple procedures; P = .517) outcomes. CONCLUSION The atrial distribution and extent of multipolar electrogram spatiotemporal dispersion follow a nonrandom, albeit patient-unique, distribution in PsAF patients. AI-DED may represent a procedure-implementable fi ngerprint of the PsAF substrate.

Abstract Background While pulmonary vein isolation (PVI) is the gold standard approach of treatment of atrial fibrillation (AF), the optimal strategy in case of AF recurrences for redo patients is not well established. Several studies described the usefulness of ablating in atrial regions exhibiting abnormal electrograms, i.e. spatio-temporal dispersion, during AF and an artificial intelligence (AI) electrograms (EGMs)-based software solution has been developed and validated to obtain real-time adjudications of multipolar electrograms in de novo patients. Purpose This study sought to evaluate the novel AI-EGMs-based software approach for redo-AF ablation procedures. Methods This study was a retrospective, multicentric, nonrandomized study. Patients with recurrent symptomatic AF were enrolled in 7 European centers. Redo ablation procedures were performed using the AI software and acute and long-term outcomes after ablation were evaluated. Results A total of 117 patients with refractory symptomatic AF admitted for redo procedure were included with 30% paroxysmal, 42% short-standing persistent and 28% long-standing persistent AF at the time of the redo procedure. During the index procedures, a previous PVI ablation was performed in all patients while additional ablation such as lines and EGM-based ablation was performed in 64 patients (55%). Out of 117 patients, 30 (26%) presented with structural heart disease, 61 (52%) with hypertension, 31 (26%) with a BMI >30, 16 (14%) a sleep apnea syndrome and 24 (20%) had an antiarrhythmic treatment at the time of the procedure. The average AF history was 6.54 ± 4.15 years with an average sustained duration equal to 25 ± 62 months. AI-EGMs-based AF ablation was performed in all patients and easily integrated in the physicians’ workflow during redo procedures. AF termination, i.e. AF regularization or sinus rhythm conversion, occurred in 82/117 patients (71%) with sinus rhythm conversion by ablation in 79/117 patients (68%). After a mean follow-up of 11 ± 5 months, 81% of patients were free from documented AF, with or without antiarrhythmic drugs (AADs) after one redo procedure and 60% of patients were free from any documented atrial arrhythmia, with or without AADs, after one redo procedure per patient. Furthermore, we observed that 77%, 84% and 82% of patients were free from documented AF with or without AADs in paroxysmal, short-standing and long-standing persistent AF patients respectively, and that 71%, 59% and 49% of patients were free from any documented atrial arrhythmia with or without AADs in paroxysmal, short-standing and long-standing persistent AF patients respectively. Conclusions This retrospective study reports that patient-tailored spatiotemporal dispersion ablation is a relevant and standardized approach for repeat AF ablation. This was obtained with an AI-EGMs-based software solution, which may be added to the operative workflow of redo AF procedures.

Background: Recurrences of atrial fibrillation (AF) after pulmonary vein isolation (PVI) are mainly due to pulmonary vein reconnection. However, a growing number of patients have AF recurrences despite durable PVI. The optimal ablative strategy for these patients is unknown. We analyzed the impact of current ablation strategies in a large multicenter study. Methods: Patients undergoing a redo ablation for AF and presenting durable PVI were included. The freedom from atrial arrhythmia after pulmonary vein-based, linear-based, electrogram-based, and trigger-based ablation strategies were compared. Results: Between 2010 and 2020, 367 patients (67% men, 63±10 years, 44% paroxysmal) underwent a redo ablation for AF recurrences despite durable PVI at 39 centers. After durable PVI was confirmed, linear-based ablation was performed in 219 (60%) patients, electrogram-based ablation in 168 (45%) patients, trigger-based ablation in 101 (27%) patients, and pulmonary vein-based ablation in 56 (15%) patients. Seven patients (2%) did not undergo any additional ablation during the redo procedure. After 22±19 months of follow-up, 122 (33%) and 159 (43%) patients had a recurrence of atrial arrhythmia at 12 and 24 months, respectively. No significant difference in arrhythmia-free survival was observed between the different ablation strategies. Left atrial dilatation was the only independent factor associated with arrhythmia-free survival (HR, 1.59 [95% CI, 1.13–2.23]; P =0.006). Conclusions: In patients with recurrent AF despite durable PVI, no ablation strategy used alone or in combination during the redo procedure appears to be superior in improving arrhythmia-free survival. Left atrial size is a significant predictor of ablation outcome in this population.

Abstract Funding Acknowledgements Type of funding sources: Other. Main funding source(s): Volta Medical Saint Joseph Hospital Marseille, France Background Spatiotemporal dispersion is an electrical footprint of atrial fibrillation (AF) drivers that has been successfully implemented to target extra-pulmonary veins (PVs) regions during persistent AF ablation. Purpose The aim of the study is to characterize spatiotemporal dispersion extent and location and to compare dispersion atrial localization and extent between pacing-induced AF and spontaneous AF. Methods Spatiotemporal dispersion maps were built with an artificial intelligence software (VX1, Volta Medical) and analyzed in 71 consecutives persistent (66%) and long-standing persistent (34%) AF patients admitted for a first ablation procedure. Fifty-two patients were in spontaneous AF (73%) at the outset of the procedure while AF was induced in 19 patients (27%) by burst pacing ± isoproterenol infusion. A semi-quantitative visual quantification of dispersion extent was conducted by implementing an atrial segmentation into 22 regions and a region-centered score from 0 (no dispersion) to 3 (high dispersion). Also, an automated quantification was performed as follows: (i) a gradient filter designed to extract atrial shapes from background was applied, (ii) dispersion areas were segmented using a color thresholding method, and (iii) a structuring element was used to connect segmented dispersion areas. Results The regional characterization of dispersion shows that dispersion distribution follows a similar pattern and is present in similar atrial regions, regardless of whether AF is spontaneous or induced (Figure 1). Global dispersion extent, however, is significantly higher in the left atrium of patients in spontaneous AF compared to patients with induced AF (15.43% ± 9.04 versus 9.86% ± 6.41, P=0.0025, Figure 2). Accordingly, the regional dispersion score tends to be lower when AF is induced. Dispersion hotspots are: LSPV anterior antrum -ridge (R1), RSPV anterior antrum (R5), anterior wall (R9), roof (R10), posterior wall (R11), and low left atrial septum (R15) in the left atrium. In the right atrium: low right atrial septum (R15) and posterior right atrium (R20) (Figure 1). Conclusions Artificial intelligence-enabled dispersion persistent AF mapping indicates that dispersion is reduced in induced vs. spontaneous-AF but that its distribution follows a similar pattern.

Objective: To provide insight into first clinical use cases of NF annotation algorithm and PF data.Methods: Acute procedure data were prospectively collected from 32 operators at 16 institutions in the US and Europe.All cases were performed on EnSite X system with EnSite OT Near Field software (Abbott).PF information was used as an adjunct to activation or voltage maps via an emphasis layer or as a standalone map (Figure).Choice of mapping tools and technique was at physician discretion.Results: Analysis included ablation cases (n5175) with the following indications: AF (60.6% [54.7% PAF, 45.3%, PsAF; 34.9% redo]), SVT (20.6%), and VT/PVC (18.9%).The NF algorithm was the roving detection option of choice for automated point collection in most cases (96.2% of AF cases, 91.7% SVT, 69.7% VT/PVC).Alternative detections were Last Deflection in VT cases (33%) and First Deflection in PVC cases (29%).When NF was used, correctness of EGM annotation was noted as highly accurate in 93.5% of cases.Standalone PF maps were used as a diagnostic tool in 48 (27.4%)Depth of Activation Insights cases.Bounding values for map display were variable based on use-case to highlight various electrophysiologic phenomena (Table ).In an AF use case, PF was used for gap identification (n56); values were applied at 225628 Hz, consistent with previously published value of 240Hz (Merino, et al.EHRA 2022).Alternatively in VT/PVC cases, low PF far-field zones were highlighted to assess activation at depth in tissue; mean values were 198610 Hz.Conclusion: Peak frequency mapping helps distinguish near field from far-field electrograms.PF mapping can be used as (1) a standalone mapping technique to accurately identify near field components and (2) PF mapping can be used as an adjunctive technique to complement activation and voltage maps.PF mapping appears to assist in substrate differentiation, pathway localization, and contact assessment.Further research is warranted on optimal settings for various tissue types and rhythms.

Real-time artificial intelligence (AI) software design to detect dispersed electrograms (EGMs) during atrial fibrillation (AF) procedures has previously been described but a communication with a 3D mapping environment was missing.

The normal electrical potential propagates throughout the atria periodically. During atrial arrhythmias its prop-agation is modified because the substrate is not homoge-neous and new sources of punctual electrical activity appear. The periodic behavior of activation remains predom-inant, but becomes local in different parts of the atria. It is characterized by cycle length (CL) which measures the frequency of activation and can be computed from intrac-ardiac bipolar electrograms (EGM) recorded by a mapping catheter during the catheter ablation procedure. The CL value of different mapped zones is an extremely important resource for physicians when performing persis-tent Atrial Fibrillation (AF) ablation because it helps to identify pathological zones and define the ablation strat-egy. Thus, a reliable estimation of the CL of atrial tissue is essential. The complexity of this task stems from the large variability in EGM morphology influenced by mul-tiple wavefronts, fragmentation and added noise. In this work, we propose a cycle length estimator that can process the complex mapping signals recorded during atrial arrhythmias ablation and reliably provide the frequency of their periodic activity.