Abstract 16340: Atrial Wall Thickness Features Predefine Human Right Atrial Driver: Insights Gained From Novel Structural Analysis

Abstract

Introduction: Atrial fibrillation (AF) is not effectively treated in clinics due to a lack of adequate tools for identifying AF drivers as targets in ablation treatment. Recent studies revealed that 30-50% of patients with persistent AF have drivers not only in the left atrium (LA) but also in the right atrium (RA). However, the precise substrate underlying RA AF drivers remains elusive. Hypothesis: High-resolution analysis of human atrial wall 3D structure can predict the occurrence of right atrial drivers. Methods: Coronary perfused explanted human atria (N=10, 46.2±14.7 y.o.;40% female) were optically mapped with high-resolution CMOS cameras. Sustained AF was induced by fast pacing and perfusion of adenosine and/or isoproterenol. Hearts were imaged at ~170μm 3 resolution by 9.4T gadolinium-enhanced MRI (Fig. A). A coupled PDE approach was utilized to estimate the atrial wall thickness (Fig. B). Fibrosis was identified above a signal intensity threshold validated by histology; and fibrosis density was constructed for each voxel by determining the percent of neighboring fibrotic voxels within a radius of 5 voxels. Results: Optical mapping identified 9 RA drivers in 6 hearts (H1-6) and 7 LA drivers in 5 hearts (H5, 6 and 8-10, Fig. C). All 9 RA AF drivers were anchored in myobundle structure in close proximity to the crista terminalis (CT). Wall thickness distribution in RA excluding the CT was similar to that in the LA (N=10, Fig. D). Structural analysis showed that RA with AF drivers (N=6) was thicker with greater wall variation and higher fibrosis density than RA without AF drivers (N=4, Fig. E). Furthermore, RA driver regions (N=9) were thicker and varied more than other RA regions (4.64±2.86 mm vs 4.56±2.57mm, p<0.05). Conclusions: For the first time, the 3D structural analysis demonstrates that structural features including atrial wall thickness and its variation near the CT can predict RA driver occurrences in human hearts and be used to improve targeted ablation.