Hollow fiber implants: Creating space within living tissue for diagnosis and therapy

The mechanical design of cell encapsulation devices that utilize hollow fiber membranes is fundamental to implant performance, yet previous work in this field has shown that the design of these devices is one of the fundamental barriers to clinical acceptance of this technology. The goal of the studies described in this dissertation was to develop new cell encapsulation devices based on hollow fiber membrane technology that would address deficiencies attributed to previous device designs. The new cell encapsulation devices that were developed during this work addressed the needs for increased mechanical strength, device retrievability and replacement of the encapsulated cell mass. These devices were evaluated by mechanical and bench-top testing, in-vitro studies and in-vivo implantation. The first phase of work provided a design foundation for successive phases. The outcome of this first phase was a device design that was able to protect the relatively fragile membrane from mechanical stresses that were known to cause membrane failure. The design consisted of mechanical structures within the membrane that shielded it from forces due to kinking, buckling and tensile loads. In addition, this design provided a reliable method for filling the device with cells and sealing the end. The design was evaluated in-vivo during the second phase of the study where it was coupled to a membrane that was engineered to provide high flux for relatively large molecules. The device was implanted in a physically demanding location in mice and demonstrated successful delivery of gene products without membrane failure. In addition, we discovered during this in-vivo work that a helical substrate could be used to shuttle a cell mass into and out of a hollow fiber membrane. This motivated the third phase of work the objectives of which were to develop a simplified transcranial access port for a hollow fiber implant which would provide repeated access to the interior of the membrane, thus allowing the transplantation of cells into a previously implanted membrane. This access port was used to implant a hollow fiber membrane into the rat brain. After one year of membrane implantation NIH-3T3 cells were transplanted into the membrane, retrieved after two weeks and shown to be viable. In summary, we developed hollow fiber membrane device designs with improved mechanical strength, simplified cell loading procedures and capability for repeated access to the interior of the membrane.