Artificial intelligence-intracardiac echocardiography-guided pulsed field ablation of atrial fibrillation with variable-loop circular catheter.

AIMS: Intracardiac echocardiography (ICE) facilitates left atrial (LA) reconstruction during atrial fibrillation (AF) ablation. The artificial intelligence-based CARTOSOUND FAM (AIFAM) module enables automated three-dimensional LA reconstruction without the need for a dedicated pre-ablation mapping catheter. While this workflow has been described previously in radiofrequency ablation, its application and outcome in pulsed field ablation (PFA) remain limited. The objective of this study is to evaluate the feasibility, safety, and arrhythmia-free outcomes of a no-pre-mapping AIFAM-guided workflow during PFA using the variable-loop circular catheter (VLCC).

METHODS AND RESULTS: Patients undergoing PFA AF ablation using VLCC were screened and those using AIFAM-guided ablation without pre-map were included in the analysis. The LA shell was created using the AI-based ICE module from the right atrium without the need for a LA mapping catheter and was used as the only reference shell for PFA energy delivery. Baseline demographics, AF subtype, comorbidities, procedural details, and follow-up outcomes were analysed. Recurrence was assessed using a 60-day blanking period and Kaplan-Meier survival analysis. A total of 412 patients underwent AF ablation using VLCC during the study period. Among them, 210 patients (mean age 68.1 ± 10.2 years; 54.8% male) underwent AIFAM-guided ablation without pre-map and were analysed. Atrial fibrillation subtype was paroxysmal in 52% and non-paroxysmal in 48%. Concomitant AF ablation and left atrial appendage occlusion (LAAO) were performed in 76 patients (36%). The mean CHA2DS2-VASc score was 3.4 ± 1.7, and the mean left ventricular ejection fraction was 56.2 ± 11.1%. All patients underwent pulmonary vein isolation, with additional PF lesions including posterior wall isolation in 87.5% and superior vena cava ablation in 60.5%. Skin-to-skin procedure time and LA dwell time were 78.8 ± 29.1 min and 47.1 ± 18.8 min for ablation alone and 87.3 ± 22.9 min and 54.6 ± 13.0 min for concomitant AF ablation and LAAO. Acute procedural success was achieved in 100% of cases. One acute non-catheter-related vascular complication (0.5%), and no late complications were documented. Among 184 patients with at least one follow-up, freedom from arrhythmia survival after the 60-day blanking was 77.7% in paroxysmal AF and 71.8% in non-paroxysmal AF at 270 days.

CONCLUSION: AIFAM enables safe and reliable LA reconstruction for successful PFA ablation, including concomitant AF ablation and LAAO using VLCC, and reduces the need for pre-ablation mapping in select patients undergoing index ablation. This strategy demonstrated high acute success, low complication rates, and favourable arrhythmia freedom in a contemporary cohort.