Computational Investigation and Histopathological Validation of Interaction Between Stent Graft and Aorta in Retrograde Type A Dissection After TEVAR in Canine Models

Abstract

Purpose: Retrograde type A dissection (RTAD) after thoracic endovascular aortic repair (TEVAR) has been a major drawback of endovascular treatment. To our knowledge, no studies have simulated and validated aortic injuries caused by stent grafts (SGs) in animal models. Therefore, the aim of this study was to evaluate and quantify the SG–aorta interaction through computational simulations and to investigate the underlying mechanism through histopathological examinations. Methods: Two custom-made Fabulous® (DiNovA Meditech, Hang Zhou, China) SGs were implanted in 2 canine aortas with a 5-mm difference in the distance in landing locations. The aortic geometries were extracted from RTAD and non-RTAD cases. A computational SG model was assembled based on the implanted SG using the software Pro-ENGINEER Wildfire 5.0 (PTC Corporation, Needham, Mass). TEVAR simulations were performed 7 times for each canine model using Abaqus software (Providence, RI, USA), and the maximum aortic stress (MAS) was calculated and compared among the groups. Three months after SG implantation, the canine aortas were harvested, and were examined using hematoxylin and eosin staining and Elastica Van Gieson (EVG) staining to evaluate histopathological changes. Results: In the computational models for both canines, MAS was observed at the proximal bare stent (PBS) at aortic greater curve. The PBS generated higher stress toward the aortic wall than other SG parts did. Moreover, the MAS was significantly higher in canine No.1 than in canine No.2 (0.415±0.210 versus 0.200±0.160 MPa) (p<0.01). Notably, in canine No.1, an RTAD developed at the MAS segment, and histopathological examinations of the segment showed an intimal flap, a false lumen, elastin changes, and medial necrosis. RTAD was not observed in canine No.2. In both SG-covered aortas, medial necrosis, elastic fiber stretching, and inflammatory infiltration were seen. Conclusion: The characteristic MAS distribution remained at the location where the apex of the PBS interacted with the aortic wall at greater curve. RTAD histopathological examinations showed intimal damage and medial necrosis at the proximal landing zone, at the same MAS location in computational simulations. The in vivo results were consistent with the computational simulations, suggesting the MAS at greater curve may cause RTAD, and the potential application of computational simulation in the mechanism study of RTAD.