(148i) A Novel Viscoelastic Thrombogenesis Model from High Performance Lattice Boltzmann Method Yield-Stress Calculations Based on Intravital Images of Clot Formation in Live Mice

Thrombo-embolic infarction is the major cause of mortality and morbidity in the United States, causing over 1 million strokes, heart attacks and other life-threatening thrombotic events each year in the United States. Conversely, deficiencies in these processes result in severe bleeding risks. Unfortunately, despite tremendous efforts in understanding thrombosis, the viscoelastic properties of thrombi that are responsible for embolism are not well understood. The ability to access hydrodynamics stresses at which thrombus structure yield to deformation in a blood vessel can provide meaningful information on when the thrombus is likely to embolize. Lack of an accurate method/computational model for estimating the stresses acting on a thrombus in vivo prompted us in developing a robust approach that combines intravital imaging and simulation. Interestingly, the viscoelastic behavior exhibited by the thrombus resembles that of a Bingham fluid - a material that behaves as a rigid body at low stresses but flows as a viscous fluid when it experiences stresses in excess for its “critical yield stress”. Thus, we ventured to measure the critical yield stress at which the thrombi yield behavior occurs.

All experiments were performed at Dr. Skip Brass’s Lab, University of Pennsylvania. To that end, thrombi were induced by means of a laser injury in cremaster muscle microcirculation of live mice and tracked via fluorescence markers using confocal microscopic imaging. The images are stabilized using Fiji plugins StackReg and Image Stabilization to fix translation. An in-house Matlab® code is used to produce 3D reconstruction by interpolating the thrombus height to width ratio and fitting a parabola across the edge in the direction perpendicular to blood flow. An in-house D3Q15 Lattice Boltzmann Method code is used to simualte convective blood flow on supercomputers, due to the large simulation sized. The simulations are performed for a constant pressure drop across the channel and the pressure drop was estimated based on the average blood velocity obtained in vivo by the Doppler velocimetry. A blood viscosity of μ ≈ 0.03 is assumed based on literature, and no-slip boundary condition is applied at the wall faces.

Our results indicate that although each clot evolves uniquely, all of the data can be collapsed on a single dimensionless scale. The non-dimensionalization of the data is done by taking the critical time at which the clot’s structure yields to the surround blood flow as a reference. At this special point, the clot stops undergoing uniform growth, yields to the surrounding blood flow and rearranges its shape to minimize the drag on its surface. Consequently, we present results for both wild type mice and various mutants on this master scale.

To our knowledge, this is the first image-based in-vivo assessment of blood clots viscoelastic nature. Collectively, the outcome of this work is expected to provide insight towards an understanding of clot generation, its structure, stability, thromboembolism and coagulopathy. Moreover, they can assist in creating simpler thrombogenesis models that can help improve the understanding of risk factors associated with blood clotting, and ideally help researchers to reduce risks of occlusion and embolism in patients.