Tissue engineered aortic valves based on a knitted scaffold

The aortic valve is situated at the origin of the aorta. The valve can present pathological features which can lead to life-threatening valvular disease. A common treatment for aortic valve failure is replacement by a valve prosthesis. Currently used aortic valve prostheses can derive either from human sources, animal tissues, or be synthetic in nature. Each type of prosthesis has its own limitations but all lack the ability to grow, repair, and remodel. The concept of tissue engineering offers a solution to this problem; in theory, it yields valves that are completely autologous after the scaffold they were cultured on has been biodegraded. The procedure of tissue engineering, as it is described in this thesis, involves donation of saphenous vein cells by a potential valve recipient, that are cultured in vitro until a sufficient amount of cells is obtained to seed onto a valvular scaffold. The cell-seeded scaffold is cultured in vitro for four weeks, after which it has developed into a valve that can be implanted into the recipient. After the scaffold has completely been degraded, the valve is autologous and theoretically not rejected by its host. Currently tissue engineered valves, based on synthetic scaffolds, have been tested in vivo in the pulmonary artery of animals, where pressure gradients are seven fold lower than in the aorta. The goal of this research is to develop and investigate a strong valvular scaffold for a tissue engineered valve that is able to withstand aortic conditions. This scaffold is made of a knitted poly(e-caprolactone) valve, covered with fibrin gel. The degradation of poly(e-caprolactone), (PCL), has been found to occur in at least two stages. In the first stage bulk PCL is hydrolyzed non-enzymatically. This is followed by fragmentation of the polymer and phagocytosis of the fragments by macrophages. The final metabolite of PCL is removed from the body. The PCL that was used in this study, was estimated to biodegrade within 27 months of implantation and has been reported to only evoke a mild inflammatory response after implantation in animals. Culture of cells, encapsulated in fibrin gel, provides an efficient cell seeding procedure and yields homogeneous tissue with a mature extracellular matrix. Furthermore, fibrin gel plugs the relatively large pores in the knitted valvular scaffold and enables cells to overgrow these pores. Tests in a newly developed loading device, showed that the knitted scaffold remained intact after ten million loading cycles. Not only is the knitted scaffold strong, the fibrin gel and the poly(e-caprolactone) are both biocompatible to human myofibroblasts. When the knitted, fibrin covered polycaprolactone scaffold was placed inside a physiologic pulse duplicator, it showed proper opening and closing behavior, but considerable leakage under aortic circumstances when not seeded with cells. When the knitted PCL scaffold was compared to an electrospun PCL scaffold in a tissue engineering process, the pores in the knitted scaffold were shown to be fully2 overgrown by cells, but were too large to retain all the cells during the seeding process, resulting in cell spoilage. By contrast, the pores in the electrospun scaffold were too small for cellular ingrowth and its performance inside the physiologic pulse duplicator was poor. Indeed the electrospun valve ruptured within six hours. The knitted scaffold remained intact. To enhance cellular attachment and retainment of the knitted structure, the valve was consequently knitted out of a more fluffy yarn and treated with a sodium hydroxide solution. A new bioreactor was developed that enables mechanically stimulated stentless valve culture. Two groups of valves, that were based on the more hydrophilic knitted structure, were cultured in this bioreactor. The first group of valves was strained in diastolic position, while visual recordings were made of the valve bellies and the pressure difference over the valve was determined. The second group of valves was only perfused with culture medium. From the images made of the valve bellies, strains were calculated to be large at a low pressure difference over the valve. Furthermore, controlling the magnitude of the imposed strains was very difficult. Perfusion-based cultured valves showed better tissue formation and performed better under aortic flow and pressure circumstances than their strain-based counterparts. Eight valves, based on fibrin coated, polycaprolactone knitted structures, were cultured under perfusion. After four weeks all valves were completely overgrown by cells and all valves remained intact during in vitro flow behavior testing. Seven of these valves opened and closed properly under aortic circumstances although their closing volume was larger than that of the reference valve, a porcine stentless prosthesis. Seven of the valves did not show leakage and five valves had a regurgitant fraction under 10%, the ISO defined acceptable level for non-active aortic valve substitutes. It is possible to make in vitro functional tissue engineered aortic valves, based on a knitted scaffold. These valves withstand aortic flow- and pressure circumstances. What the in vivo behavior of these valves will be and how the valvular tissue will have developed once the scaffold has been degraded, remain areas to be investigated.