Abstract Number ‐ 169: Intraprocedural angiographic sign for assessing the stent‐clot interaction during mechanical thrombectomy

Abstract

The characteristics of the occlusive clot affect the clot integration with a stent retriever (SR). This relationship, stent‐clot interaction, is considered to be a major factor in the technical success of mechanical thrombectomy. To date, numerous studies analyzing the retrieved clots have shown that both soft erythro‐rich clots and hard fibrin‐rich clots make clot retrieval challenging. Several studies have successfully obtained information on this interaction using three‐dimensional (3D) rotational angiography. However, these 3D imaging technologies have not been utilized in clinical practice due to the time‐consuming nature of image acquisition and processing. Our previous clinical study demonstrated that the angiographic findings about the deployed stent morphology obtained from conventional two‐dimensional (2D) angiography could predict recanalization (1). The greater stent expansion at the occlusion was strongly associated with recanalization after the procedure. This intraprocedural angiographic sign allows us to know the stiffness of the clots in real‐time and to choose the optimal technique. The purpose of this study was to evaluate whether the stent expansion assessed by a 2D angiographical image reflects the actual stent dilation at the occlusion (Figure). We investigated the correlations between 2D images and 3D structures of the deployed SR using an experimental occlusion model. Using occlusion models created with pseudo‐clot with 9 hardness levels (n = 3/clot type), images of the deployed Trevo SR were obtained by cone‐beam computed tomography.As the measurement metric for the 2D images, we used the degree of stent expansion obtained from a plane along the long axis of the device. In clinical practice, however, this 2‐D image is usually obtained from one viewing angle. Therefore, to investigate the difference in measurement by viewing angle, different angle 2D images were created to evaluate the stent expansion. For the 3D structures, we used the stent area obtained from the short‐axis plane of the vascular model, considering this as a surrogate for actual stent expansion. We evaluated the correlation between the 2D images and the 3D structure. A total of 27 model image sets were obtained, showing graduated stent expansion (range: 21–79%) depending on the clot type. The median variation in the degree of stent expansion for each model measured at different angles, which means the differences by viewing angles, was 9% (range: 5–20%). The median degree of stent expansion was strongly correlated with the stent area (Pearson’s coefficient: 0.98), indicating that the degree of stent expansion could reflect the 3D structure. This study showed that the stent expansion on 2D angiography, even assessed from one direction, could be used as the approximation of the actual stent dilatation at the occlusion. This angiographic sign provides real‐time feedback on the clot characteristics at the occlusion.