2018 年 12 巻 8 号 p. 376-385
Objective: The hemodynamics of cerebral aneurysms was evaluated by computational fluid dynamics (CFD) analysis using the non-Newtonian (Casson’s) fluid model and the Newtonian fluid model obtained from measurements. The two fluid models were examined to clarify the influence of blood viscosity on hemodynamic parameters.
Methods: We measured blood viscosity of blood obtained from 50 healthy adults at 12 shear rate ranges using a compact-sized falling needle rheometer. Blood viscosity was set as the Newtonian and Casson’s fluid models determined using these measurements. In all, 12 cerebral aneurysms were evaluated by transient analysis to calculate the wall shear stress (WSS), wall shear stress gradient (WSSG), flow velocity (FV), oscillatory shear index (OSI), and parameters that facilitate the quantitative assessment of the fluctuations of individual vectors, including the gradient oscillatory number (GON) and oscillatory velocity index (OVI). Bland–Altman analysis was performed to compare the two models, and systematic errors were examined.
Results: The relationship between the apparent viscosity and the shear rate obtained from blood samples of 50 healthy adults revealed the characteristics of Casson’s fluid. The systematic errors in hemodynamic parameters for the two fluid models were small, and the correlation coefficients of the WSS, WSSG, FV, OSI, GON, and OVI were 0.9999, 0.9999, 0.9985, 0.9734, 0.9758, and 0.9258, respectively. Furthermore, the means of these hemodynamic parameters for the entire aneurysm showed a high consistency rate between the two groups, whereas different values were observed in focal hemodynamics including blebs.
Conclusion: Newtonian fluid numerical modeling may be useful for analyzing the entire aneurysm. On the other hand, these results indicated that hemodynamics analyzed using non-Newtonian blood viscosity could have certain effects on focal hemodynamics that may be related to aneurysm growth and rupture.