High Density Mapping of Atrial Fibrillation During Vagal Nerve Stimulation in the Canine Heart: Restudying the Moe Hypothesis

Abstract

Moe et al. hypothesized that multiple wavelets (random reentry) were the mechanism of atrial fibrillation (AF) based on studies in a vagal nerve stimulation (VNS) canine model and a computer model of AF, but atrial mapping during AF in this model has not been done. We restudied this model using high density, simultaneous site mapping to test the hypothesis that AF was due to multiple wavelets. During pacing induced AF during VNS in 10 dogs, 512 unipolar atrial electrograms were recorded simultaneously from both atria. AF activation maps were produced including through AF termination after VNS cessation. During sustained AF, multiple foci (persistent and transient) of different cycle lengths (CLs) were present in both atria. Persistent foci of short (mean 112 ± 25 milliseconds), regular (standard deviation 5.3 ± 3 milliseconds) CLs were predominantly found in the left atria, near the pulmonary veins and coronary sinus. Both types of foci acted as drivers, and each produced wave fronts that largely resulted in collision or merging with each other at variable sites. No random reentry (multiple wavelets) was demonstrated. Ordered reentry (circus movement with head-tail interaction) was infrequently seen. With cessation of VNS, focal firing slowed and disappeared, followed by resumption of sinus rhythm after a prolonged pause. In contrast to the prediction of the multiple wavelet hypothesis, during AF in the Moe model, multiple foci drove the atria, producing and maintaining AF. Reentry played little, if any, role.