Reentry-Driven Model of Atrial Fibrillation Is Maintained by Paired Reentries and Terminated by Strategic Pairwise Virtual Ablation

Arthur Bezerra^1*Robin Van Den Abeele¹Bjorn Verstraeten¹Sebastiaan Lootens¹Arstanbek Okenov¹Timur Nezlobinsky¹Sander Hendrickx¹Vincent FM Segers²Nele Vandersickel¹

• ¹Ghent University, Ghent, Belgium
• ²Universiteit Antwerpen, Antwerp, Belgium

Atrial fibrillation (AF) is a very common cardiac arrhythmia whose mechanisms are still a topic of debate. This work applied concepts of topology to gain new insights into reentry-based simulated AF, similar to our previous work in atrial tachycardia (AT). We demonstrate that the Index Theorem – which states reentries must come in pairs of opposite rotations – applies to a model of AF, even when the complex dynamics change over time. Additionally, we tested the hypothesis that connecting opposite pairs of singularities can terminate simulated AF in the same way as clinical and simulated AT. To do so, we applied a modified phase mapping capable of detecting both functional and anatomical reentry to a dataset of 600 AF simulations based on clinical data. We then compared three virtual ablation strategies: random lines, straight lines, and heuristic lines. Straight lines connected pairs of opposite singularities through the shortest path; heuristic lines connected them in such a way that prioritized blocking the conduction path; and random lines connected randomly selected pairs of points with comparable distance to the other methods. We showed that our algorithm could verify the predicted paired reentries for 99% of the simulation duration on average, and 93% for the worst-performing case. The heuristic virtual ablation method terminated activity for 90% of cases, a marked improvement against the straight line method (55%) and the random method (0.5%). This work provides mechanistic insights into AF, and points towards pitfalls of ablation strategies, both of which have the potential to improve our understanding and ability to treat this condition.