The influence of blood composition and loading frequency on the behavior of embolus analogs

Bryan C Good
DOI: https://doi.org/10.1016/j.jmbbm.2023.105738
Abstract:In cases of acute ischemic stroke (AIS), mechanical thrombectomy (MT) can be used to directly remove lodged thromboemboli. Despite improvements in patient outcomes, one of the key factors affecting MT success is the mechanical properties of the occlusive thrombus. Therefore, the goal of this study was to investigate the viscoelastic properties of embolus analogs (EAs) and determine the influence of EA hematocrit and loading frequency. Bovine blood EAs were created over a range of physiological hematocrits (0-60%) and cyclic uniaxial compression testing was performed at three loading frequencies to mimic in vivo loading conditions, followed by stress-relaxation testing. It was found that EAs exhibited behaviors typical of hyper-viscoelastic materials and that EA hematocrit played a large role in both EA stiffness and relaxation, with both parameters decreasing as hematocrit increased from 0 to 60%. The viscoelastic behavior of the EAs was also affected by the frequency at which they were loaded, with significant increases in peak stresses between the 0.5 and 2 Hz loaded EAs. Lower hematocrit EAs had very dense fibrin networks while the higher hematocrit EAs consisted of closely packed RBCs with little fibrin present. These results suggest that fibrin contributes to EA stiffness and relaxation behaviors while RBCs play a role in decreasing the overall viscous response and strain-rate dependency. An Ogden hyperelastic model was found to best reproduce the EA loading data while a 3-term Prony series was fit to the stress relaxation data. A hyper-viscoelastic modeling framework was then implemented combining the loading and stress-relaxation fits and the results could match the full cyclic loading data for EAs of varying hematocrit and loading frequency. The results of the experimental mechanical characterization and hyper-viscoelastic curve fitting can be incorporated in future modeling efforts to optimize mechanical thrombectomy for AIS patients.
What problem does this paper attempt to address?