Selecting repetitive focal and rotational activation patterns with the highest probability of being a source of atrial fibrillation

Introduction Repetitive focal and rotational activation patterns are currently used as additional ablation targets for atrial fibrillation (AF). However, there is no evidence that all these detected targets are actual sources of AF. In this paper, we present an approach that detects and ranks AF activation patterns not only based on the degree of pattern repetitiveness but also on the extent to which they are able to entrain their vicinity. This new technique might enable selecting the site with the highest probability of being a source for AF. Methods We retrospectively analyzed high-density bi-atrial sequential mapping in ablation-naive persistent AF patients (n = 13, PentaRay catheter, 30s recordings). Repetitive focal and rotational activation patterns were detected based on local activation time annotation of unipolar electrograms. The spatial stability was determined as local repetitive pattern duration. The entrainment capability was defined as the average time a directionally coherent repetitive activation pattern was observed in adjacent recordings. Results A total of 459 recordings were analyzed (35 ± 5 per patient). We detected 131 repetitive focal (10 ± 4 per patient) and 56 rotational activation patterns (4 ± 3 per patient) in total. Focal patterns were more repetitive than rotational patterns (median [IQR] 0.7 [0.4–1.3] seconds vs. 0.5 [0.4–0.6] seconds, p < 0.001 Mann-Whitney U test). By applying a 90th percentile threshold to both local and directionally coherent adjacent repetitiveness, we identified 10 sites (9 focal and 1 rotational) in 7 patients as the most probable sources. The majority of these sites were in the upper right atrium or left pulmonary vein region. Notably, in 6 patients (46 %), no probable sources were detected using this threshold. Conclusion This study introduces a novel technique to select the repetitive focal or rotational pattern with the highest probability of being a source. We observed that only a minority of repetitive focal or rotational patterns seem to be able to entrain their vicinity and thereby are likely to serve as sources of AF.