A resolved two-way coupled CFD/6-DOF approach for predicting embolus transport and the embolus-trapping efficiency of IVC filters

Abstract

Inferior vena cava (IVC) filters are medical devices designed to provide a mechanical barrier to the passage of emboli from the deep veins of the legs to the heart and lungs. Despite decades of development and clinical use, IVC filters still fail to prevent the passage of all hazardous emboli. The objective of this study is to (1) develop a resolved two-way computational model of embolus transport, (2) provide verification and validation evidence for the model, and (3) demonstrate the ability of the model to predict the embolus-trapping efficiency of an IVC filter. Our model couples computational fluid dynamics simulations of blood flow to six-degree-of-freedom simulations of embolus transport and resolves the interactions between rigid, spherical emboli and the blood flow using an immersed boundary method. Following model development and numerical verification and validation of the computational approach against benchmark data from the literature, embolus transport simulations are performed in an idealized IVC geometry. Centered and tilted filter orientations are considered using a nonlinear finite element-based virtual filter placement procedure. A total of 2048 coupled CFD/6-DOF simulations are performed to predict the embolus-trapping statistics of the filter. The simulations predict that the embolus-trapping efficiency of the IVC filter increases with increasing embolus diameter and increasing embolus-to-blood density ratio. Tilted filter placement is found to decrease the embolus-trapping efficiency compared with centered filter placement. Multiple embolus-trapping locations are predicted for the IVC filter, and the trapping locations are predicted to shift upstream and toward the vessel wall with increasing embolus diameter. Simulations of the injection of successive emboli into the IVC are also performed and reveal that the embolus-trapping efficiency decreases with increasing thrombus load in the IVC filter. In future work, the computational tool could be used to investigate IVC filter design improvements, the effect of patient anatomy on embolus transport and IVC filter embolus-trapping efficiency, and, with further development and validation, optimal filter selection and placement on a patient-specific basis.

Fully-resolved simulations of particle-laden viscoelastic fluids using an immersed boundary method

“…In this section we present the mathematical formulation for an algorithm that is able to efficiently handle the rigid body motion of particulate spherical bodies surrounded by a viscoelastic fluid. The algorithm considers a constraint-based formulation, which provides a rigorous basis for the IB implementation performed in the open-source framework code CFDEMcoupling [34,35] .…”

“…The use of a common opensource computational library to perform these new developments is expected to increase the accessibility and utility of the contribution. The numerical method implemented is based on the approaches described in Shirgaonkar et al [7] , Hager et al [34] and Aycock et al [35] , where the latter study presents an improved IB method to describe the flow of Newtonian fluids around spherical particles. To verify the robustness of the implemented numerical approach, a number of benchmark problems with known published results were simulated.…”

“…Section 3 provides the details of the numerical discretization and solution procedure for the overall algorithm that is described by the mathematical formulation presented in Section 2 . In Section 4 we present a number of case studies involving Newtonian and viscoelastic flows around spherical particles, namely the sedimentation of a sphere in a bounded domain [41,42] , the rotation of a sphere suspended in a homogeneous steady shear flow of an Oldroyd-B fluid [43] , and the cross-stream migration of a neutrallybuoyant sphere in Poiseuille flow [35,44] . In each case, a comparison of the results obtained with the newly-developed code is performed with data reported in the literature, in order to assess both accuracy and robustness.…”

An immersed interface method for discrete surfaces

“…Fig. 28 compares the lift coefficient generated by the present method to previous work [94,95,74,96,97] for α in the range 0-3 at Re = 5 and 20. The C L values generated by the present IIM are in excellent agreement with prior work for the entire range of α values considered.…”

“…29 demonstrates that these flow conditions are accurately predicted using the present IIM. Badr et al [96] Aycock et al [95] Ingham & Tang [94] D'alessio & Dennis [97] Ingham & Tang [94] We set ρ = 1 and use the inflow velocity U as the characteristic velocity. The Reynolds number is Re = ρU D µ , and we consider Reynolds numbers from 20 to 500.…”