Abstract
AIMS: This prospective randomized controlled trial investigated the comparative efficacy and safety of circumferential pulmonary vein isolation (CPVI) combined with modified linear ablation (CPVI-MLA) vs. standalone CPVI in patients with long-standing persistent atrial fibrillation (LSPAF). METHODS AND RESULTS: In this single-centre pilot trial, 134 LSPAF patients were randomized to the CPVI-MLA (n = 67) or CPVI-only (n = 67) groups. The CPVI-MLA protocol integrated four components: (i) ethanol infusion targeting the ligament of Marshall; (ii) complete CPVI; (iii) extended lesion sets (posterior wall isolation, dual isthmus ablation); and (iv) substrate modification [left atrial intima adjoining coronary sinus (LAI-CS) and superior vena cava isolation (SVCI)]. A 24 h Holter monitoring was performed at the 1st, 3rd, and 6th month follow-up visits, with 7-day Holter monitoring at the 12th month follow-up visit. The primary endpoint was freedom from atrial tachyarrhythmias (≥ 30 s) after the initial 3-month blanking period post-index procedure, without antiarrhythmic drugs. After a mean follow-up of 14.5 ± 9.1 months, 76.1% (51/67) in the CPVI-MLA group and 65.7% (44/67) in the CPVI-only group achieved the primary endpoint (P = 0.32). However, the CPVI-MLA group demonstrated significantly higher atrial fibrillation (AF)-free survival rate (91.0 vs. 76.1%, P = 0.049), while atrial tachycardia/atrial flutter-free survival rates were comparable (83.5 vs. 88.1%, P = 0.45). The CPVI-MLA strategy required longer ablation time (68.6 ± 12.3 vs. 49.4 ± 10.3 min, P < 0.001) and fluoroscopy exposure (14.9 ± 9.8 vs. 9.3 ± 6.7 min, P < 0.001). Serious adverse events were rare and similar between groups (1.5 vs. 0%, P = 1.00). CONCLUSION: In patients with LSPAF, the CPVI-MLA strategy significantly improved freedom from AF compared with CPVI alone, although it did not improve overall sinus rhythm maintenance rate. This strategy may offer a refined approach for complex AF ablation, warranting further validation in larger trials.