Demitria A. Poulos ¹ Michael T. Froehler ² Bryan C. Good ^1*

• ¹ The University of Tennessee, Knoxville, Knoxville, United States
• ² Vanderbilt University Medical Center, Nashville, Tennessee, United States

Introduction: Mechanical thrombectomy becomes more complex when the occlusion occurs in a tortuous cerebral anatomy, increasing the puncture to reperfusion time and the number of attempts for clot removal. Therefore, an understanding of stent retriever performance in these locations is necessary to increase the efficiency and safety of the procedure. An in vitro investigation into the effects of occlusion site tortuosity, blood clot hematocrit, and device geometry was conducted to identify their individual influence on stent retriever removal forces. Methods: Embolus analogs were used to create occlusions in a mock circulatory flow loop, and in vitro mechanical thrombectomies were performed in arterial models of increasing tortuosity. The stent retriever removal forces of Solitaire Platinum and EmboTrap II devices were recorded through each geometry with and without embolus analogs present. Similar experiments were also conducted with Solitaire stent retrievers of varying lengths and diameters and 0, 25, and 50% hematocrit embolus analogs.The removal force increased as model tortuosity increased for both the Solitaire Platinum and EmboTrap II stent retriever devices. The average removal forces in the simplest geometry with the Solitaire Platinum and EmboTrap II were 0.24±0.01 N and 0.37±0.02 N, respectively, and increased to 1.2±0.08 N and 1.6±0.17 N, respectively, in the most complex geometry. Slight increases in removal force were found with 0% hematocrit embolus analogs, however, no statistical significance between removal force and EA hematocrit was observed. A comparison between stent retriever removal forces between devices of different diameters also proved to be significant (p<0.01), while forces between devices of varying lengths were not (p>0.05). Conclusions: Benchtop mechanical thrombectomies performed with commercial stent retrievers of varying geometry showed that device removal forces increase with increasing model tortuosity, clot hematocrit does not play a significant role in device removal force, and that a stent retriever's diameter has a greater impact on removal forces compared to its length. These results provide an improved understanding of the overall forces involved in mechanical thrombectomy and can be used to develop safer and more effective stent retrievers for the most difficult cases.