O-107 - NEAR-WALL HEMODYNAMIC CHANGES INDUCED BY RETROGRADE INNER BRANCHED STENT GRAFT IMPLANTATION FOR LEFT SUBCLAVIAN ARTERY PERFUSION

TOPIC:
New vascular techniques and devices
AUTHORS:
Yoon W. (Section of Vascular Surgery, Department of Surgical Sciences, Uppsala University ~ Uppsala ~ Sweden) , Lee C. (Division of Vascular Surgery, Department of Surgery, NorthShore University Health System ~ Evanston ~ United States of America) , Han S. (Comprehensive Aortic Center, Keck Medical Center of University of Southern California ~ Los Angeles ~ United States of America) , Mani K. (Section of Vascular Surgery, Department of Surgical Sciences, Uppsala University ~ Uppsala ~ Sweden) , Wanhainen A. (Section of Vascular Surgery, Department of Surgical Sciences, Uppsala University ~ Uppsala ~ Sweden)
Introduction:
Retrograde inner branched endografts for the left subclavian artery (LSA) offer a complete endovascular solution when LSA preservation is required during thoracic endovascular aortic repair (TEVAR). Implications of this geometric configuration on LSA hemodynamics, however, have not been well investigated. We compared near-wall hemodynamic parameters before and after LSA branched endograft (LBE) implantation using computational fluid dynamic (CFD) simulations.
Methods:
Eleven patients who had undergone successful implantation of a LBE (Gore TAG thoracic branch endoprosthesis, W.L.Gore & Associates, Inc. Flagstaff, Az) were included. Three-dimensional (3D) aortic arch geometries were constructed from the pre- and post-LBE implantation computed tomography (CT) images of these patients. The resulting twenty-two 3D aortic arch geometries were then discretized into finite element meshes for CFD simulations. The present study aimed at a generic investigation of patient specific LBE-blood flow interaction. In-flow boundary conditions were prescribed using normal physiologic pulsatile circulation. Outlet boundary conditions consisted of Windkessel models with previously published values. Blood flow, modeled as Newtonian fluid, simulations were performed with rigid wall assumptions utilizing SimVascular's incompressible Navier-Stokes solver. We compared the most established hemodynamic descriptors: pressure, flow rate, time-averaged wall shear stress (TAWSS), oscillating shear index (OSI), and percent area with OSI >0.2 (%A OSI>0.2). Data were presented on the stented portion of LSA.
Results:
LBE implantation was associated with a small decrease in mean LSA peak pressure (153.3±2.7 mmHg vs 158.8±1.0 mmHg; p=0.005) and LSA to RSA (right subclavian artery) index (LSA peak pressure/RSA peak pressure) (0.97±0.03 vs 1.01±0.02; p=0.005) compared with pre-implantation simulations. No difference was observed in mean peak LSA flow rates between pre- and post-implantation (41.9±5.0 vs 41.4±3.7, p=0.59). There was a significant post-implantation increase in TAWSS (15.3±4.5 dynes/cm2 vs 7.9±3.3 dynes/cm2; p=.003) (Figure 1), leading to significant decrease in both OSI (0.085±0.026 vs 0.12±0.037; p=.01) and %A OSI >0.2 (11.4±8.3 vs 20.5±11.9; p=.03) (Figure 2). The angles between the centerlines of the main stent graft and LSA bridging stent graft (LBSG) ranged from 58° to 157° (median, 81°; IQR, 32°). When comparing LBSGs with <90° (n=6) versus those with >90° (n=5), no difference was observed in the %A OSI >0.2 (11.5±6.0 vs 11.2±11.3; p=0.65). Additionally, LSA lumen compression (stenosis), ranged from 16% to 57.7% (median, 32.1%; IQR, 29.2%), was observed. When simulations of >50% stenosis cases were compared with those with <50% stenosis, there was no difference in the %A OSI >0.2 (12.4±5.3 vs 11.0±9.5; p=0.81).
Conclusion:
LSA branched endografts with retrograde inner branch configurations produce modest hemodynamic disturbances which are unlikely to result in clinically relevant changes to the stented LSA environment.
ATTACHMENTS: