Assessment of thromboembolic risk in the Left Atrium via 4D flow MRI and Computational Fluid Dynamics
Description
Interest on the biomechanics of the Left Atrium (LA) has grown over the last few years [1], as impaired atrial function is strongly associated with increased risk of stroke. Atrial Fibrillation (AF) is responsible for about one-third of all strokes [2]. Though the link between AF and stroke is well established, AF is not the only source of stroke associated with the LA [3] and about 20% of strokes remain without a clearly defined cause [4]. It is believed that areas of blood stagnation may develop, consecutive to changes in the mechanics of the LA wall (especially fibrosis), thereby increasing thrombo-embolic risk. The complex coupling between the LA wall biomechanics (and its remodeling) and intra-atrial hemodynamics is essential for evaluating atrial function and predicting the risk of strokes [5]. However, this relationship is not fully understood making the evaluation of the atrial function challenging. This project aims at developing biomarkers for the evaluation of atrial function and its associated risk of strokes. These markers will be based on the combination of 4D flow MRI and computational fluid dynamics (CFD). While new imaging techniques have made essential contributions to better understand and characterize LA function [6] [7], modelling approaches play a complementary role by providing predictive insights [8] [9].
This project will be based on a dataset of 4D flow MRI images of patients (n>50) providing 3D blood velocity fields and LA wall motion. The approach will be divided into the following steps:
- Post-processing of the 4D flow MRI images for the cohort: segmentation of the images using already developed tools [10] and extraction of global (flow rates, volume variations, …) and local (stagnation index, vorticity, …) metrics
- Adaptation and application of high-fidelity CFD simulations of intra-atrial hemodynamics: adaptation of the existing in-house solver (used for CT-based wall motion) to MRI-derived motion; computation of similar flow metrics for comparison with 4D flow MRI and computation of more advanced metrics (residence time distribution, thrombin concentration fields,…) [9]
- Evaluation of the uncertainties of the 4D flow imaging approach: down-sampling of the CFD simulations and comparison to in vivo 4D flow MRI data
- Evaluation of the predicting potential of the derived metrics: contrasting the in silico derived metrics with the clinical outcomes (occurrence of thrombo-embolic events).
Combining insights from biomechanics and cardiovascular fluid dynamics with clinical data, this project aims to substantially enhance our understanding of Left Atrium biomechanics, while setting the foundation for a clinically translational approach to predict thrombosis risk in patients.
Bibliography
Bibliography
[1] F. Bisbal, A. Baranchuk, E. Braunwald, A. Bayés de Luna, and A. Bayés-Genís, “Atrial Failure as a Clinical Entity,” J. Am. Coll. Cardiol., vol. 75, no. 2, pp. 222–232, Jan. 2020, doi: 10.1016/j.jacc.2019.11.013.
[2] B. Freedman, T. S. Potpara, and G. Y. H. Lip, “Stroke prevention in atrial fibrillation,” The Lancet, vol. 388, no. 10046, pp. 806–817, Aug. 2016, doi: 10.1016/S0140-6736(16)31257-0.
[3] K. Tandon, D. Tirschwell, W. T. Longstreth, B. Smith, and N. Akoum, “Embolic stroke of undetermined source correlates to atrial fibrosis without atrial fibrillation,” Neurology, vol. 93, no. 4, pp. e381–e387, July 2019, doi: 10.1212/WNL.0000000000007827.
[4] C. O. Johnson et al., “Global, regional, and national burden of stroke, 1990–2016: a systematic analysis for the Global Burden of Disease Study 2016,” Lancet Neurol., vol. 18, no. 5, pp. 439–458, May 2019, doi: 10.1016/S1474-4422(19)30034-1.
[5] L. E. Sade et al., “Left atrial mechanics for secondary prevention from embolic stroke of undetermined source,” Eur. Heart J. - Cardiovasc. Imaging, vol. 23, no. 3, pp. 381–391, Feb. 2022, doi:10.1093/ehjci/jeaa311.
[6] T. Sekine et al., “4D Flow MR Imaging of the Left Atrium: What is Non-physiological Blood Flow in the Cardiac System?,” Magn. Reson. Med. Sci., vol. 21, no. 2, pp. 293–308, 2022, doi:10.2463/mrms.rev.2021-0137.
[7] J. Garcia et al., “Left atrial vortex size and velocity distributions by 4D flow MRI in patients with paroxysmal atrial fibrillation: Associations with age and CHA2DS2-VASc risk score,” J. Magn. Reson. Imaging, vol. 51, no. 3, pp. 871–884, Mar. 2020, doi: 10.1002/jmri.26876.
[8] L. Parker et al., “A multi-modal computational fluid dynamics model of left atrial fibrillation haemodynamics validated with 4D flow MRI,” Biomech. Model. Mechanobiol., vol. 24, no. 1, pp. 139–152, Feb. 2025, doi: 10.1007/s10237-024-01901-y.
[9] M. Guerrero-Hurtado et al., “Hemodynamics affects factor XI/XII anticoagulation efficacy in patient-derived left atrial models,” Comput. Methods Programs Biomed., vol. 267, p. 108761, July 2025, doi:10.1016/j.cmpb.2025.108761.
[10] X. Morales et al., “A Computational Pipeline for Advanced Analysis of 4D Flow MRI in the Left Atrium,” 2025, arXiv. doi: 10.48550/ARXIV.2505.09746.
[11] J.-B. Guichard and S. Nattel, “Atrial Cardiomyopathy,” J. Am. Coll. Cardiol., vol. 70, no. 6, pp. 756–765, Aug. 2017, doi: 10.1016/j.jacc.2017.06.033.
[12] R. M. Inciardi and A. Rossi, “Left atrium: a forgotten biomarker and a potential target in cardiovascular medicine,” J. Cardiovasc. Med., vol. 20, no. 12, pp. 797–808, Dec. 2019, doi:10.2459/JCM.0000000000000886.
