Effects of stent shape on focal hemodynamics in intracranial atherosclerotic stenosis: A simulation study with computational fluid dynamics modeling

Abstract

Background and aims: The shape of a stent could influence focal hemodynamics and subsequently plaque growth or in-stent restenosis in intracranial atherosclerotic stenosis (ICAS). In this preliminary study, we aim to investigate the associations between stent shapes and focal hemodynamics in ICAS, using computational fluid dynamics (CFD) simulations with manually manipulated stents of different shapes.

Methods: We built an idealized artery model, and reconstructed four patient-specific models of ICAS. In each model, three variations of stent geometry (i.e., enlarged, inner-narrowed, and outer-narrowed) were developed. We performed static CFD simulation on the idealized model and three patient-specific models, and transient CFD simulation of three cardiac cycles on one patient-specific model. Pressure, wall shear stress (WSS), and low-density lipoprotein (LDL) filtration rate were quantified in the CFD models, and compared between models with an inner- or outer-narrowed stent vs. an enlarged stent. The absolute difference in each hemodynamic parameter was obtained by subtracting values from two models; a normalized difference (ND) was calculated as the ratio of the absolute difference and the value in the enlarged stent model, both area-averaged throughout the arterial wall.

Results: The differences in focal pressure in models with different stent geometry were negligible (ND<1% for all cases). However, there were significant differences in the WSS and LDL filtration rate with different stent geometry, with ND >20% in a static model. Observable differences in WSS and LDL filtration rate mainly appeared in area adjacent to and immediately distal to the stent. In the transient simulation, the LDL filtration rate had milder temporal fluctuations than WSS.

Conclusions: The stent geometry might influence the focal WSS and LDL filtration rate in ICAS, with negligible effect on pressure. Future studies are warranted to verify the relevance of the changes in these hemodynamic parameters in governing plaque growth and possibly in-stent restenosis in ICAS.

Keywords: computational fluid dynamics (CFD); hemodynamics; in-stent restenosis (ISR); intracranial atherosclerotic stenosis (ICAS); low-density lipoprotein (LDL); stent geometry; wall shear stress (WSS).

Figure 1

Idealized (A) and patient-specific (B) ICAS and stenting models. d, diameter; r, radius; R, radius of curvature. (A) Left: Idealized geometry of an intracranial artery with ICAS. Right: simulated variations of the vessel geometry after stenting, with 3 different stent shapes. (B) Geometric models of ICAS lesions in M1 segment of MCA in 3 patients, and the simulated variations of the vessel geometry after stenting, with 3 different stent shapes. The arterial segment adjacent to the ICAS lesion that is virtually stented is magnified with red rectangle in each scenario.

Figure 2

The Windkessel model applied in a patient-specific ICAS model (upper panel) and arterial geometry simulated with different stent shapes (lower panel). The red arrows show the blood flow directions. ICA, internal carotid artery; MCA: middle cerebral artery; ACA, anterior cerebral artery. R, C, p, d, m, a in the Windkessel elements denote resistance, capacitance, proximal, distal, MCA, and ACA, respectively.

Figure 3

The distribution of differences of WSS and LDL filtration rate, of the inner-narrowed and outer-narrowed models compared with the enlarged model, in patient-specific static models. A red color indicates a larger difference and blue or green color indicates a smaller difference, in the parameter of the inner-narrowed or outer-narrowed model against the enlarged model (as a reference model that is not shown in the figure). For instance, there is larger difference in WSS in the stented segment and an immediate branch (in red color), but smaller difference in WSS in other regions (blue color), in the inner-narrowed model in case 1 in the left panel, as compared with WSS in the enlarged model.

Figure 4

The normalized differences of pressure, WSS and LDL filtration rate, in the inner-narrowed and outer-narrowed models compared with the enlarged model, in the idealized and patient-specific static CFD models. (A) Results in idealized model. (B) Results in patient-specific models.

Figure 5

The distribution of the differences in WSS and LDL filtration rate, of the inner-narrowed and outer-narrowed models compared with the enlarged model. The data were measured end-of-diastole and systole of the second cardiac cycle (0.86 s and 1.00 s in simulation) where the inlet blood pressure was minimal and maximal.

Figure 6

The normalized differences of pressure, WSS and LDL filtration rate, of the inner-narrowed and outer-narrowed models compared with the enlarged model. The data were measured at end-of-diastole and systole of the second cardiac cycle (0.86 s and 1.00 s in simulation) where the inlet blood pressure was minimal and maximal.

Figure 7

The transient, area-averaged WSS and LDL filtration rate and inlet pressure during the second cardiac cycle.