Individual spatio-temporal atrial fibrillation dynamics predict arrhythmia recurrence - insights from non-invasive global mapping

Abstract Background Mechanisms of atrial fibrillation and their therapeutic implications remain incompletely understood. This is in part owed to the invasive nature of established mapping systems and their lacking capacity to assess the spatio-temporal dynamics of this complex arrhythmia. Arrhythmogenic substrate in terms of local slow-conduction has been shown to play an important role in the AF pathomechanism and to predict risk of recurrence after AF ablation (1). It is unknown however, to what extent individual AF patterns are determined by local conduction velocities and how they impact the course of the disease. Purpose Here we use non-invasive global mapping by ECGi which enables comprehensive assessment local conduction velocities and spatio-temporal dynamics that maintain AF. We aim to identify AF drivers that are stable over time and link them to the underlying individual substrate and arrhythmia recurrence. Methods In this prospective study consecutive patients who presented in AF and underwent AF ablation or simple electrical cardioversion were included. All patients received continuous periprocedural ECGi-mapping in AF (before cardioversion) and in sinus rhythm (after cardioversion). Stable AF drivers were defined as locations that displayed the lowest cycle lengths (below the 10th percentile of all local cycle lengths at a given timepoint) over more than 50% of the recording time frame o 20-40 seconds. Maps were visualized based on mean local cycle lengths and their spatio-temporal stability during AF ("driver-stability maps"), and on local conduction velocities in sinus rhythm (Fig. 1). In a regional analysis based on 15 predefined left and right atrial areas the number, localization and stability of AF drivers was compared with local conduction velocities of the same patient and linked to arrhythmia-free survival over a 12-months follow-up. Results 52 patients were included. A mean number of 2.8 slow-conduction areas and 2.2 stable drivers were detected per patient. Stable drivers were predominantly localized in areas of slow-conduction (76,8%). The extent of slow-conduction determined the complexity of AF patterns, where higher numbers of slow-conduction areas are associated with more abundant but less stable drivers. Both, the number of slow-conduction areas (HR 5.2, p=0.033) and the complexity of AF patterns (HR 4.3, p=0,0025) were independent predictors of arrhythmia recurrence. Patients with more than 2 drivers had significantly higher recurrence (45% vs. 24%, p=0,001). Conclusions Local conduction velocities determine the complexity of individual AF patterns and the localization of stable drivers. As both, local conduction velocities and AF complexity predict arrhythmia-free survival, non-invasive ECGi mapping may become a powerful tool to define ablation targets and personalized AF therapies.Figure 1