Including aortic valve morphology in computational fluid dynamics simulations: Initial findings and application to aortic coarctation

• Published in: Medical engineering & physics Vol. 35; no. 6; pp. 723 - 735
• Main Authors: Wendell, David C, Samyn, Margaret M, Cava, Joseph R, Ellwein, Laura M, Krolikowski, Mary M, Gandy, Kimberly L, Pelech, Andrew N, Shadden, Shawn C, LaDisa, John F
• Format: Journal Article
• Language: English
• Published: England Elsevier Ltd 01-06-2013
• Subjects: Adolescent; Adult; Aortic Coarctation - pathology; Aortic Coarctation - physiopathology; Aortic Valve - pathology; Aortic Valve - physiopathology; Bicuspid aortic valve; Computer Simulation; Congenital heart disease; Female; Hemodynamics; Humans; Hydrodynamics; Kinetic energy; Kinetics; Magnetic resonance imaging (MRI); Male; Phase-contrast MRI; Radiology; Shear stress; Stress, Mechanical; Turbulence; Turbulence; Bicuspid aortic valve; Magnetic resonance imaging (MRI); Shear stress; Congenital heart disease; Hemodynamics; Kinetic energy; Phase-contrast MRI
• ISSN: 1350-4533; 1873-4030
• DOI: 10.1016/j.medengphy.2012.07.015

Abstract Computational fluid dynamics (CFD) simulations quantifying thoracic aortic flow patterns have not included disturbances from the aortic valve (AoV). 80% of patients with aortic coarctation (CoA) have a bicuspid aortic valve (BAV) which may cause adverse flow patterns contributing to morbidity. Our objectives were to develop a method to account for the AoV in CFD simulations, and quantify its impact on local hemodynamics. The method developed facilitates segmentation of the AoV, spatiotemporal interpolation of segments, and anatomic positioning of segments at the CFD model inlet. The AoV was included in CFD model examples of a normal (tricuspid AoV) and a post-surgical CoA patient (BAV). Velocity, turbulent kinetic energy (TKE), time-averaged wall shear stress (TAWSS), and oscillatory shear index (OSI) results were compared to equivalent simulations using a plug inlet profile. The plug inlet greatly underestimated TKE for both examples. TAWSS differences extended throughout the thoracic aorta for the CoA BAV, but were limited to the arch for the normal example. OSI differences existed mainly in the ascending aorta for both cases. The impact of AoV can now be included with CFD simulations to identify regions of deleterious hemodynamics thereby advancing simulations of the thoracic aorta one step closer to reality.