Tissue Engineering of Vascular Access: A Feasibility Study

Background: Over 1.5 million people in the United States are dialysis dependent with the vast majority on hemodialysis. Approximately half of these patients require prosthetic grafts for access as suitable vein is not available. Prosthetic grafts fail regularly from venous limb stenosis, thrombosis, and infection resulting in graft revision or failure an average of 18 months after implantation. We have previously demonstrated that bioengineered blood vessels constructed from autologous progenitor cell-derived endothelial cells can be used for arterial reconstruction with patency for more than 4 months in sheep. In the present study we investigated the feasibility of using bioengineered blood vessels as an alternative to prosthetic arteriovenous grafts in a pre-clinical model of vascular access.
Methods: Bioengineered vessels were created and implanted between the carotid artery and the external jugular vein of female sheep (n=5). Using a high efficiency isolation procedure, autologous endothelial progenitor cells were collected from peripheral blood of each animal and differentiated in culture into mature endothelial cells. Endothelial phenotype was confirmed using immunohistochemistry (IHC) for differentiation markers. Cells were then seeded on the lumen of collagen-based vascular scaffolds and conditioned to pulsatile flow and pressure for two weeks in a bioreactor. Bioengineered vessels were then sutured end-to-side to the carotid artery and jugular vein. Graft patency, hemodynamics, and geometry were characterized with serial duplex ultrasonography. Grafts were then retrieved at 1 and 2 months and assessed histologically for matrix remodeling, host response and tissue maturation.
Results: A mean of 444 autologous endothelial progenitor colonies were obtained from each sheep. Transition from CD133+ progenitors to mature endothelial cells expressing vWf, lectin and eNOS was documented using IHC. After 2 weeks of bioreactor pre-conditioning, grafts demonstrated a monolayer of endothelial cells on the lumen surface. Following implantation grafts maintained structural integrity without significant dilation or constrictive remodeling on duplex imaging over time . Healed engineered grafts tolerated needle cannulation three times per week for one month prior to explantation. Histological analysis demonstrated modest repopulation of the vascular scaffold with host cells expressing smooth muscle markers at one month but significant repopulation of the engineered graft wall at 2 months.
Conclusions: Through tissue engineering we have created autologous blood vessels as a viable substitute for prosthetic grafts in a preclinical model of hemodialysis access. These vessels maintained structural integrity and geometry in a high flow environment with repeated percutaneous needle access out to two months. The use of bioengineered blood vessels as an arteriovenous bypass graft is indeed feasible and studies to explore long term patency of these grafts are underway.