Predicting Ablation Lesion Depth in the Swine Ventricle using Ablation Index – Differences between Healthy and Scarred Tissue

Background:

Ablation index(AI) using a logarithmic algorithm incorporating Force-Power-Time-Index (FPTI) has been shown to predict lesion depth in the atrium and has gained popularity in atrial ablation. Data regarding the potential use of ablation index in the ventricles is limited.

Methods:

A total of 21 swine(11 healthy;10 after recovery from anterior myocardial infarction) were studied. A complete bi-ventricular voltage map was created, including the valve regions to ensure adequate anatomical localization of ablation lesions. LV ablation lesions were delivered at 40W targeting an AI of 500, 700 and 900, and RV lesions delivered at 25W targeting an AI of 300 and 500. Scar tissue ablation was performed at 40W and targeted an AI of 500 and 900. Following fixation, the lesions were identified and measured and lesion depth compared to targeted AI values.

Results:

240 ablation lesion were analyzed(41 lesions in scar) and measured. The AI accurately predicted lesion depth(±1.0mm) in the majority of lesion applications(184/199; 92.4%) in normal tissue. Outliers were all within 2mm of the achieved index target in healthy tissue. In myocardial scar tissue AI accurately predicted lesion depth(±1.0mm) in only 31% of the lesions and within(±3.0mm) in 70%(29/41). Only 1 lesion(2.4%) in scar exceeded the AI prediction depth.

Conclusion:

AI has a high ability to predict lesion depth when utilized in healthy ventricular myocardium. Delivery of radiofrequency in scarred myocardium results in AI overestimation of lesion depth, mainly due to the different tissue properties, hence the presence of fibrosis and adipose tissue in healed post-infarct scar.