APS Logo

Changes in Left Atrial Flow and Stasis After Rhythm Restoration via Cardiac Ablation

ORAL

Abstract

Atrial fibrillation (AF) increases stroke risk by promoting blood stasis and thrombus formation

in the left atrial appendage (LAA). Because AF is associated with both functional and structural

changes in the left atrium (LA), it is challenging to isolate the hemodynamic effects of rhythm

alone. In patient-specific flow modeling, rhythm cannot be controlled at the time of imaging,

and most studies compare AF and sinus rhythm (SR) across different patients, confounding rhythm

with anatomy. To overcome this, we studied patients imaged in AF prior to catheter ablation and

again in stable SR more than three months post-ablation, enabling direct comparison of flow and

stasis under both rhythms in the same anatomical configuration.

For each patient, LA anatomical meshes were automatically generated from 4D cardiac CT

segmentations using a custom-trained nnU-Net. Time-resolved atrial wall motion was extracted via

non-rigid Coherent Point Drift (CPD) registration across frames. To enable consistent comparison

of wall dynamics across rhythms and patients, we used a Universal Atrial Coordinate (UAC)

framework based on Signed Distance Fields (SDF), followed by Fourier analysis to characterize

rhythm-dependent wall motion in a standardized reference space.

Patient-specific LA flow and LAA stasis were simulated using an in-house single-GPU immersed

boundary CFD solver in both AF and SR conditions. Blood was modeled as a non-Newtonian fluid via

a Carreau–Yasuda relation, activated in regions of high residence time. Simulations spanned 20

cardiac cycles to resolve stasis regions. Results show that while

the impact of AF on LA flow is patient-specific, it consistently leads to increased LAA stasis

relative to SR when controlling for LA and LAA anatomy.

Presenters

  • Savier Sharda

    University of Washington

Authors

  • Savier Sharda

    University of Washington

  • Akshaykumar Gupta

    University of Washington

  • Alejandro Gonzalo

    University of Washington

  • Manuel Guerrero-Hurtado

    Universidad Carlos III de Madrid

  • Pablo Martinez-Legazpi

    Universidad Nacional de Educación a Distancia, Universidad Nacional de Educación a Distancia & CIBERCV

  • Javier Bermejo

    Hospital General Universitario Gregorio Maranon

  • Manuel García-Villalba

    TU Wien, Technical University of Vienna

  • Oscar Flores

    University Carlos III De Madrid

  • Elliot McVeigh

    University of California San Diego

  • Andrew M Kahn

    University of California San Diego

  • Juan C del Alamo

    Department of Mechanical Engineering, University of Washington, Seattle, Washington; Center for Cardiovascular Biology, University of Washington, Seattle, Washington, University of Washington