COMPUTATIONAL ACCESS FLOW REDUCTION EFFECT ON WALL SHEAR STRESS IN BRACHIOCEPHALIC FISTULAE

The population of patients with end stage renal disease (ESRD) is growing at a rate higher than the global population. The only viable treatment for these patients is a kidney transplant. However, in the absence of a suitable kidney donor, renal patients are left with hemodialysis as a renal replacement therapy. Hemodialysis is facilitated through arteriovenous fistula (AVF), and the particular interest in this investigation is brachiocephalic fistula (BCF). The survival of dialysis patients depends on maintaining patency of fistula over a prolonged period of time. The extreme hemodynamic environment that is created by BCF triggers the onset of neointimal hyperplasia (NH) in most renal dialysis patients, which leads to access failure via stenosis. This is because the hemodynamics in AVF are well outside the normal physiological range. Computational fluid dynamics (CFD) along with shape optimization allows for the study of the hemodynamic parameters such as wall shear stress that have been shown to be detrimental in the future occurrence of cephalic arch stenosis. In this study, CFD modeling and identification of hemodynamic patterns was possible in three dimensions due to advanced post processing of IVUS patient-specific geometries. A method utilizing 3D CFD and shape optimization has been developed to implement Miller’s banding method used in clinical practice to evaluate its impact on WSS and onset of neointimal hyperplasia. The level of banding represented by a constriction is in fact a patient specific value and is not a trivial solution of minimum flow rate necessary to conduct hemodialysis; hence suggesting that even restoring the inlet velocity to the velocity pre-fistula creation might not reduce incidence of cephalic arch stenosis. The findings of this study support the previous hypothesis that non-homeostatic WSS distributions trigger neointimal hyperplasia and resulting venous stenosis. The important outcome is that the presented computational framework allows for evaluation of Miller's banding method for reducing the blood flow rate via surgical constriction and identification of a patient-specific banding level that restores the WSS to the normal physiological range.M.S. in Mechanical and Aerospace Engineering, May 201