Article Text
Abstract
Flow Diverter (FD) stenting is now widely used to treat large cerebral aneurysms. It induces thrombosis inside the aneurysmal sac. However, thrombosis formation does not always take place and it is believed that the formation of the thrombus is strongly related to hemodynamic factors around the aneurysm. In our previous study, computational fluid dynamics (CFD) has been applied to identify risk factors for aneurysm rupture. In this study, we apply CFD simulations on aneurysms treated by FD to investigate hemodynamic factors that may influence thrombosis formation. We selected 10 FD treated aneurysms which had been followed up for at least 6 months and evaluated their treatment outcome at 6-month follow-up. Five cases were completely occluded (CO) and 5 cases showed residual blood flow (RE). First, we reconstructed patient-specific vessel geometries from digital subtraction angiography data. Next, we deployed FDs based on actual FDs using our original virtual stenting program. Then, we simulated the blood flow and calculated hemodynamic parameters previously identified as risk factors for aneurysm rupture. Various parameters were estimated before and after FD stenting to obtain their change rate. We used the Mann-Whitney U test to compare differences of each hemodynamic parameter between the CO and RE groups. In both groups, the average blood flow velocity was reduced after FD stenting (see figure 1), but the difference between the groups was small. The RE cases had statistically significantly lower pressure loss coefficient (PLc) values (i.e., the blood flow entered more easily into the aneurysmal sac for the RE cases) (p=0.04). In addition, the CO cases had a greater change rate of minimum wall shear stress (WSS) acting on the aneurysmal wall. The averaged minimum-WSS change rate of the CO cases was 5 times higher than that of the RE cases, but the difference was not statistically significant (p=0.30). In terms of rupture risk, the values of WSS and PLc suggested that the CO cases had a lower rupture risk than the RE cases (see figure 2). CFD simulations revealed that PLc and WSS were different for CO cases and RE cases. These hemodynamic parameters were also related to aneurysmal rupture risk. Thrombosis formation may thus be related to aneurysmal rupture risk and CO cases appeared to have a lower rupture risk. These hemodynamic parameters may be useful for the outcome prediction of FD stenting.
Streamline changes
Scatter diagram of hemodynamics parameters
Disclosures Y. Uchiyama: None. H. Takao: None. S. Fujimura: None. T. Suzuki: None. H. Ono: None. T. Ishii: None. T. Okudaira: None. K. Otani: None. K. Fukudome: None. T. Ishibashi: None. M. Yamamoto: None. Y. Murayama: None.