Numerical modeling of arterial pulse wave propagation to characterize aortic hemodynamic: Validation using magnetic resonance data - 05/03/13

Doi : 10.1016/j.irbm.2012.12.008

E. Bollache ^a,^b,^⁎ , N. Kachenoura ^a, F. Frouin ^a, A. Redheuil ^a,^c, E. Mousseaux ^a,^c, D. Lucor ^b
^a Laboratoire d’imagerie fonctionnelle, UMR_S 678 Inserm, 75013 Paris, France
^b Institut Jean le Rond d’Alembert, UMR CNRS 7190, 75005 Paris, France
^c Service de radiologie cardiovasculaire, Hôpital Européen Georges Pompidou, AP–HP, 75015 Paris, France

Corresponding author. Laboratoire d’imagerie fonctionnelle, CHU Pitié-Salpêtrière, 91, boulevard de l’Hôpital, 75634 Paris cedex 13, France.

Bienvenue sur EM-consulte, la référence des professionnels de santé.
L’accès au texte intégral de cet article nécessite un abonnement.

pages	4
Iconographies	3
Vidéos	0
Autres	0

Abstract

Objectives

Arterial stiffness, which is caused by aging and other cardiovascular risk factors and primarily affects the aorta, is associated with cardiac and cerebral morbidity and mortality. The objective of our study was to non-invasively estimate local biomechanical and hemodynamic biomarkers related to proximal aortic stiffness, by combining cardiovascular magnetic resonance (CMR) data and numerical simulations.

Materials and methods

To achieve this aim, we used a numerical 1D fluid-structure model to simulate blood flow in the descending aorta, and we combined this model with clinical data (areas and velocities in three levels of the descending aorta, carotid pressures) acquired in two healthy subjects using CMR and applanation tonometry.

Results

First, we studied the sensibility of our model on an idealized aorta and showed that our model was able to characterize age-related arterial alterations, when compared to established physiological knowledge. Furthermore, while comparisons of simulations against clinical data revealed low errors (<20%) in terms of aortic areas and velocities for the two subjects, more important errors were found for pulse pressures (up to 20%). Importantly, errors in terms of velocity and area were lower than their variations occurring with aging.

Conclusions

Thus, our fast method could enable the non-invasive estimation of aortic functional parameters and a more realistic version of our numerical model could also provide a reliable estimation of central pressure.

Le texte complet de cet article est disponible en PDF.