Response by Handa et al to Letter Regarding Article, “Granger CausalityBased Analysis for Classification of Fibrillation Mechanisms and Localization of Rotational Drivers”

摘要

In Response: We thank Drs Masé, Faes, and Ravelli for their interest in our article.1 We agree that suboptimal outcomes from atrial fibrillation ablation, especially in the context of persistent atrial fibrillation, can in part be attributed to the limitations of current mapping and signal processing techniques and their application to limited coverage multisite mapping. Our recent article describes the application of Granger causality (GC) analysis to fibrillatory signals, and as highlighted by Dr Masé et al, there are a limited number of previous publications that propose its application to fibrillatory signals. However, these studies have predominantly been carried out on in silico models without direct validation in a human or animal model. In these previous studies, the application to electrograms acquired during atrial fibrillation mapping was limited to a small number of patients. Our study builds upon this previous work but is unique in providing extensive experimental validation for the application of GC analysis, by benchmarking against phase analysis of high-resolution optical mapping data. In addition, we provided a validated framework for application of GC analysis using data of low spatial resolution before applying it to atrial fibrillation electrograms acquired from multisite mapping. Another distinct and novel aspect of our study specifically applies to the utilization of GC-based analysis, using data of low-spatial resolution,2 to calculate the global organization of fibrillation using the causality pairing index and for identification of sites harboring rotational drivers using the circular interdependence value. We recently demonstrated that a range of fibrillation mechanisms and global organizational complexity exist, with only hearts demonstrating high fibrillatory organization having an underlying driver region. In this regard, GC-based causality pairing index quantification of fibrillatory organisation may hold potential in identifying the specific fibrillation electrophenotype to tailor treatments3.