PAM-VT 2 Study: Long-Term Scar Evolution and Ablation Lesion Assessment by Late Gadolinium Enhancement Cardiac Magnetic Resonance After Ventricular Tachycardia Ablation.

Abstract

Late gadolinium enhancement cardiac magnetic resonance (LGE-CMR) is useful for identifying ventricular tachycardia (VT) substrate in patients with structural heart disease. While preprocedural LGE-CMR is widely used for planning, the role of postprocedural LGE-CMR in evaluating VT ablation success and long-term scar evolution has been less explored. This study aimed to prospectively and systematically assess the long-term evolution of scar and ablation lesions using serial postablation LGE-CMR with long-term follow-up. This prospective study included 51 patients (mean age, 65.2±9.8 years; men, 95.8%; ischemic heart disease, 83%; left ventricular ejection fraction, 34.5±10.4%) undergoing their first substrate-based VT ablation between March 2019 and July 2020. Preprocedural LGE-CMR and 2 postprocedural scans at 3 to 6 months (CMR-1) and 18 to 24 months (CMR-2) were performed. Scar characteristics, including core scar, border zone, and conducting channels, were analyzed. VT recurrence was monitored, and factors associated with recurrence were evaluated using a Cox proportional hazards model. A Kaplan-Meier curve was used to represent the VT-free survival function. Core scar mass increased significantly from baseline to CMR-1 (12.2±1.5 to 19.8±1.6 g, <i>P</i><0.01) and remained stable at CMR-2. In contrast, the border zone decreased significantly over time (pre, 25.3±1.8 g; CMR-1, 20.8±2.0 g; CMR-2, 16.7±2.1 g; <i>P</i><0.01). A significant decrease in conducting channels was noted after ablation and persisted at CMR-2 (pre, mean 2.4±0.2/median 2 [interquartile range, 1-3]; CMR-1, mean 1.4±0.2/median 1 [interquartile range, 0-1]; CMR-2, mean 1.6±1.0/median 1 [interquartile range, 0-1]; <i>P</i><0.001). VT recurrence occurred in 29.4% of patients during a median follow-up of 3.1 years. The number of conducting channels at CMR-1 and their relative reduction from baseline were related to VT recurrence. The persistence of 2 or more conducting channels at CMR-1 was associated with a higher recurrence rate: 75.6 versus 19.5% (hazard ratio [HR]; 4.1; 95% CI, 2.4-12.1; <i>P</i>=0.012). Evidence of favorable left ventricular remodeling was observed, with a significant reduction in left ventricular volume at CMR-2 (131.8±8.6 mL; CMR-1, 156.7±8.1 mL and 160.8±7.6 mL at baseline <0.01). Postablation LGE-CMR reveals durable changes in scar characteristics, with early scar evaluation at 3 to 6 months strongly associated with long-term VT recurrence. The reduction in scar heterogeneity and conducting channels is sustained over time, underscoring the usefulness of LGE-CMR for assessing ablation success. Additionally, VT ablation appeared to be associated with favorable reverse remodeling, highlighting potential benefits beyond arrhythmia control. These findings support the use of LGE-CMR for personalized management following VT ablation.