Study of a Nitinol Stent Deployed into Anatomically Accurate Artery Geometry and Subjected to Realistic Service Loading

Abstract

Realistic simulation of a stent in service loading is necessary for the design of Nitinol stents. This type of analysis is complicated by several issues: artery geometry and material properties are largely unknown and may vary from person to person; the loading on the stent is due to interactions between pulsatile blood flow, the artery, and stent. This study addresses the first issue by obtaining realistic artery geometry from magnetic resonance imaging (MRI) scans of a patient in the fetal and supine positions, and comparing deformation of the stent deployed into the two different artery configurations. The second issue is addressed by investigating two different methods of applying service loading. The first method applies sinusoidal pressure waves directly to the artery wall. The second method considers the effect of blood flow within the artery by performing a fluid-structure interaction (FSI) simulation. Simulation of expansion, annealing, crimping, and deployment of a Nitinol stent into the two artery configurations was performed using Abaqus. Following the deployment, sinusoidal pressure pulses, emulating systolic and diastolic blood pressures, were applied to the inner surface of the artery. To simulate the effect of blood within the artery, the coupled Eulerian-Lagrangian method in Abaqus was used. The blood was modeled as a partially filled Eulerian domain, whose material boundary contacted the Lagrangian artery. Transient inlet velocity and outlet pressure conditions were applied to the Eulerian domain to simulate blood flow. Since fatigue life calculations are important for the design of Nitinol stents, the maximum principal strains at the diastolic pressure troughs and the difference between the maximum principal strains at diastolic and systolic pressure were investigated for five cycles. Different maximum principal strains resulted for the different artery geometries (fetal or supine), while the difference between maximum principal strains was similar. For the simulations including the effect of blood flow, the maximum principal strains at the diastolic pressure troughs were somewhat higher than the simulations in which pressure was directly applied to the artery. In conclusion, this study established and compared various methods of improving the service loading in the simulation of Nitinol stents, specifically including anatomically correct artery geometry and the effect of blood flow.