A Volume-of-Fluid Method of Modeling Thrombus Embolization Applied to Flow Through a Plain Tube

  • Tobin, Nicolas (The Pennsylvania State University)
  • Manning, Keefe (The Pennsylvania State University)

Please login to view abstract download link

Thrombosis and thromboembolism (T&TE) are major life-threatening complications in many blood-contacting biomedical devices including artificial heart valves, stents, and blood pumps and can lead to heart attack, pulmonary embolism, or stroke. Despite the central role of adhesion and fluid forces on the process of clots embolizing, very little is understood about the mechanics of T&TE due to the challenges associated with in-vitro and in-vivo characterization. Computational investigation of T&TE has, however, proven promising for gaining mechanistic insights on the formation of blood clots [1]. Objectives: To investigate the mechanics of thromboembolism in a canonical geometry, a volume-of-fluid model is implemented into OpenFOAM, where blood clots are modeled as a Phan-Thien-Tanner (PTT) [2] viscoelastic material, and phostphate-buffered saline is modeled as Newtonian. Simulations are performed of a 160 μl clot in a plain tube with a diameter of 12.7 mm. Multi-domain large-eddy simulations are run at flow rates ranging from 1.9 to 10 liters per minute, corresponding to Reynolds numbers between 3,175 and 16,713. At the wall, the PTT model is modified with a locally increased extensibility parameter. Values of the extensibility parameter are bounded by agreement with experimental embolization data. Results: Results show that the threshold wall extensibility parameter for embolization varies as a function of flow rate, approximately inversely with the drag exerted on the clot. Based on experimental data, results indicate that clots adhering to polycarbonate may be modeled with a wall extensibility parameter of around 3. The evolution of local clot strain suggests that embolization is initiated by a stretching until failure at the leading edge. The detachment from the surface is then carried downstream until the point of full embolization. Conclusions: The developed modeling approach matches the observed embolization dynamics in the in-vitro experiments. Work is ongoing to validate the adhesion modeling with tensile testing and embolization in more complicated geometries.