Abstract
The long-term clinical efficacy of coronary stents is limited by restenosis. Coronary stenting results in altered arterial geometry, local blood flow patterns, and wall shear stress (WSS), all of which can influence restenosis. Computational fluid dynamics (CFD) simulations employ assumptions about blood properties and boundary conditions, which also influence WSS alterations from stenting. Our objective was to evaluate three common assumptions applied with stented coronary artery CFD simulations (inlet velocity profile, outlet boundary conditions, and viscosity) to provide insight for future studies. A patient-specific right coronary artery was reconstructed from intravascular ultrasound and computed tomography imaging. Time-averaged WSS (TAWSS) and oscillatory shear index (OSI) were compared for CFD simulations using parabolic and Womersley (α = 2.5) velocity profiles and Newtonian versus non-Newtonian (Carreau-Yasuda) viscosity. TAWSS and OSI differences were quantified for 5-element lumped parameter network (LPN) outlet boundary conditions as compared to a 3-element Windkessel approach previously applied by neglecting ventricular contraction. Differences in velocity profiles were negligible beyond two diameters from the inlet. Differences between Womersley inlet 3-element and 5-element LPNs were ~ 5% for TAWSS and ~ 200% for OSI, and most pronounced near stent struts. OSI differences were due to pressure differences between outlet boundary conditions inducing different near-wall velocity gradients. TAWSS differences between viscosity models were greatest near struts (~ 100%). Collectively these results suggest certain assumptions commonly applied for simulations of stented coronary arteries have a greater impact on TAWSS and OSI than others, with outlet boundary conditions being paramount, followed by viscosity model and inlet velocity profile.