Predicting Arrhythmia Dynamics and Ablation Targets in Persistent Atrial Fibrillation

Atrial fibrillation (AF) is a major cause of morbidity and mortality, and is the most prevalent cardiac arrhythmia, affecting 1-2% of the worldwide population. Over the past couple decades, catheter ablation has emerged as a potential treatment for AF by electrically isolating pro-arrhythmic triggers in the pulmonary veins. Although this strategy is more effective than antiarrhythmic drugs in suppressing AF, the outcomes of this procedure are poor, particularly in persistent AF (PsAF) where recurrence rates are approximately 50%. In patients with PsAF, there is often extensive atrial fibrosis, which confounds strategies to determine AF-suppressing ablation targets. This challenge is exacerbated by variability in the fibrosis spatial pattern between patients. In this dissertation, we used a combination of patient-specific computational modeling, multielectrode catheter mapping, and inverse electrocardiography to uncover the mechanistic links between the fibrosis spatial pattern and PsAF dynamics, and use these insights to improve catheter ablation treatment planning for this arrhythmia. In the first half of this dissertation, we established the framework for building patient-specific models of fibrotic atria reconstructed from late-gadolinium enhanced magnetic resonance imaging, which incorporate biophysically-realistic representations of remodeled atrial electrophysiology. After applying programmed electrical stimulation in these models, we discovered that AF was perpetuated by spatially-confined reentrant drivers that localize to specific regions, characterized by boundary zones between fibrotic and non-fibrotic tissue where there is high fibrosis density and entropy. We compared these findings to studies of reentrant drivers using inverse electrocardiography and discovered that there was a strong correlation between the reentrant behavior and locations detected by computer simulations and inverse electrocardiography, and that ablation of sites observed by both methodologies may improve post-procedural outcomes. In the second half of this dissertation, we evaluated ablation strategies that eliminate AF-perpetuating regions in the fibrotic substrate. We first demonstrated that ablation of reentrant drivers that manifest during an AF episode is insufficient to terminate arrhythmia and prevent AF recurrence. We then test a novel simulation-guided ablation strategy that prevents the formation of any potential reentrant driver from persisting in the fibrotic substrate with a prospective clinical trial