Computational Evaluation of Shear Stress and Restenosis in Stented Coronary Arteries Using Optical Coherence Tomography
Date of Award
Master of Science (MS)
The cause of coronary artery neointimal thickness (NT) leading to restenosis in ~10% of drug-eluting stents (DES) is unknown, but adverse wall shear stress (WSS) may contribute. Prior studies comparing WSS to restenosis for first generation DES yielded conflicting results, and cited different mechanisms of action for DES agents. Studies to date have not accounted for stent geometry, which dictates local WSS patterns influencing drug concentration. The objective of this investigation was to evaluate current generation stent platforms via their WSS patterns and their respective impact on NT. We prospectively enrolled 19 patients, who were randomized to thin-strut (81μm) 2-link Promus element (PE: n=7), thin-strut 3-link Xience Prime (XP: n=6), or thick-strut (145μm) 2-link Nobori stents (NO: n=6), and underwent optical coherence tomography (OCT) and coronary CT angiography (CTA) post-stenting (PS) and after 9-months follow-up (FU). PS and FU WSS were calculated using computational fluid dynamics (CFD) simulations of patient-specific 3D arteries reconstructed by the fusion of OCT and CTA data, and normalized to WSS in the proximal unstented region. Stent-to-vessel area ratios were less in PE and XP than NO (0.24±0.04 and 0.21±0.03 vs. 0.34±0.04, respectively; P<0.05). The difference between the NO and XP was significant. PS normalized stent-induced low WSS area/stent length was least in PE, followed by XP and NO (1.05±0.27, 1.10±0.20 and 2.00±0.41 mm2/mm, respectively, P<0.05). The difference between the NO and PE was significant. When combining the data from the proximal, middle, and distal regions, an inverse correlation between PS WSS and NT at FU was greatest for the NO (-0.019±0.001 mm/Pa) followed by PE (-0.017±0.003 mm/Pa) and XP (-0.008±-0.003 mm/Pa). The difference in inverse correlation between the NO and XP was significant. The difference in correlations of PS normalized WSS and NT at FU between the biolimus-eluting NO and everlimus-eluting XP stents may be attributed to the mechanisms of action for the specific elution agents. CFD simulations using OCT and CTA suggest that DES with thicker struts and a greater stent-to-vessel area ratio produce more adverse WSS, which may lead to a higher risk for restenosis and be influenced by DES agent.