Surpoint algorithm for improved guidance of ablation for ventricular tachycardia (SURFIRE-VT): a propensity-matched study

Abstract

Abstract Background Ablation Index (AI) is a novel multiparametric radiofrequency (RF) ablation lesion quality metric, which incorporates contact force, time, and power in a weighted formula. AI guided strategies have been shown to be clinically effective for catheter ablation of atrial arrhythmias. The utility of AI to guide ablation of ventricular tachycardia (VT) in patients with structural heart disease is not known. Objectives To assess RF ablation and procedural characteristics, safety, and long-term clinical outcomes in patients undergoing VT ablation using either an AI guided strategy (maximum AI target value 550) or conventional approach using traditional parameters at a large tertiary medical center. Methods Consecutive patients (n=91) undergoing initial VT ablation for both ischemic (ICM) and non-ischemic (NICM) VT at single tertiary medical center between 2017 and 2022 were examined. Patients were 1:1 propensity matched for baseline characteristics, co-morbid conditions, and antiarrhythmic use depending on whether the ablation was guided by AI or traditional lesion quality parameters. Baseline categorical variables were compared using chi square test or Fisher exact test, and continuous measures were evaluated using Wilcoxon rank sum test. Linear regression was used to assess the impact of the use of AI on ablation characteristics and Cox proportional hazard model was used evaluate clinical outcomes. Results After propensity matching, baseline characteristics were well balanced between AI (n=37) and non-AI (n=37) groups (Table 1). Patients undergoing an AI guided strategy had shorter average procedure duration (-30.9 minutes, 95% CI [-60.4, -1.4], p=0.04) and average time per lesion (-13.7 seconds, 95% CI [-17.9, -9.5], p <0.001), as well as a trend toward significantly lower total RF ablation duration (357 seconds, 95% CI [-776, 61.9] p=0.09) and fewer steam pops (-0.19, 95% CI [-0.40, 0.02] p = 0.08). Lesion sets were similar between AI and non-AI groups (Scar homogenization: 41% vs 27%; p=0.22, Linear: 23% vs 73%; p=0.31, elimination LPs/LAVAs: 50% vs 43%; p=0.25, Scar dechanneling: 19% vs 8%; p=0.19, Core isolation: 5% vs 11%; p=0.4) as was use of epicardial mapping and ablation (11% vs 14%; p=0.73) Inducibility of VT (70% vs 67%; p=0.8) and termination of clinical VT during ablation (11% vs 16%; p=0.17) were also similar between AI and non-AI groups. When compared to a traditional strategy, patients undergoing an AI strategy had similar hazard for VT recurrence (0.97, 95% CI [0.42 – 2.2], p=0.93) and appropriate device therapy (0.79, 95% CI [0.32 – 1.95] p=0.61). Conclusions Compared to a strategy guided by traditional lesion parameters, use of AI of 550 for VT ablation in patients with structural heart disease resulted in shorter procedure time and average time of RF delivery per lesion with similar intermediate term clinical outcomes.Table 1Figure 1