Utility and Limitations of Ablation Index for Guiding Therapy in Ventricular Myocardium

Background: Ablation index (AI) is used for guiding therapy during pulmonary vein isolation. However, its potential utility in ventricular myocardium is unknown. Objectives: This study sought to examine the correlation between AI and lesion dimensions in healthy and infarcted ventricles. Methods: In ex vivo experiments using healthy swine ventricles, the correlation between AI (400-1,200) and lesion dimensions was examined at fixed power (30 W) and contact force (CF) (15 g). To examine the accuracy of AI in predicting lesion dimensions created by different combinations of ablation parameters, applications with a similar prespecified AI value created using different power (30 vs 40 W), CF (15 vs 25 g) or impedance (130-170 Ω) were created. In in vivo experiments, the correlation between AI and lesion dimensions was examined in healthy and infarcted myocardium. Results: Ex vivo experiments (247 lesions, 36 hearts) showed good correlation between AI and lesion depth (R = 0.93; P < 0.001). However, in vivo experiments (9 healthy swine and 10 infarcted swine) showed moderate correlation in healthy myocardium (R = 0.64; P < 0.01) and poor correlation in infarcted myocardium (R = 0.23; P = 0.61). AI values achieved using different combinations of power, CF, and baseline impedance resulted in different lesion depths: Ablation at 30 W produced deeper lesions compared with 40 W, ablation with CF of 15 g produced deeper lesions compared with CF of 25 g, and ablation at lower impedance produced larger lesions at similar prespecified AI values (P < 0.01 for all). Conclusions: AI has limited value for guiding ablation in ventricular myocardium, particularly scar. This may be related to small proportional significance of application duration and complex tissue architecture.