13th Annual Global Embolization Symposium & Technologies
Purpose : The minimally-invasive treatment of liver cancer by particle embolization has shown promising outcomes. Having a predictive understanding of how particles travel through bifurcating vessels is an integral part of a successful procedure. The goal of this in vitro study is to investigate the physics of the transport of microspheres in successive bifurcating vessels.
Material and Methods : The in vitro model is a parent vessel that bifurcates symmetrically four times into sixteen terminal branches as shown in the Figure. This simplified model is chosen to reduce the significant number of variables. The model fabricated using high-resolution 3D-printing is placed in a flow circuit where a physiologically relevant flow is pushed through it. The parent vessel size and the inflow condition is based on the common hepatic artery. Particles representing embolic microspheres injected into the vessel by means of a syringe pump are tracked in real time using a high-speed imaging system. Two flow regimes and two particle sizes at different flow distribution cases are considered. In addition to the optical imaging method, phase-contrast MRI is used to measure the 3D flow velocity field with submillimeter resolution in a 3-Tesla magnet.
Results : In our analysis, we first consider the fundamental question of how particles distribute between the distal branches when a uniform flow division is imposed. The assumption that the particles follow the volumetric flow distribution, each branch should receive 6.25% of the total. However, a large variation away from this value is observed, i.e. between 2% and 9% as shown in the Figure. We next consider the effect of changing the volumetric flow distribution away from uniformity. Results show that in branches where double the flow rate (with respect to others) is imposed, the particle count is interestingly between 4% and 10% instead of 11.1%. While the MRI-based results are consistent with the particle tracking results, further insight of the complex flow pattern is obtained.
Conclusions : Our statistically converged results indicate that the common assumption that particles follow the blood flow distribution at a bifurcation is not necessarily true.