Effect of beta cell distribution on the performance of a bioartificial pancreas

The insulin response of available prototype hollow fiber bioartificial pancreas devices remains unacceptably slow. Although previous experimental and theoretical investigations provide insight into the factors governing the performance of these devices, there are currently no results on the effects of beta cell distribution on insulin response. The authors have developed a detailed theoretical model for insulin response in a hollow fiber bioartificial pancreas. Model predictions have been shown to be in good agreement with literature data on the insulin release from an in vitro hollow fiber device. In this study, model simulations were used to evaluate the effects of axial and radial variations in beta cell density on the insulin response. For a device with no convective recirculation (i.e., a diffusion controlled device), the radical distribution of cells in the matrix and shell plays a critical role in determining device response, but the insulin response is essentially independent of the axial beta cell distribution. In contrast, for a device with substantial recirculation, there is a very strong dependence on the axial beta cell distribution, with a weaker dependence upon the radial distribution. These results clearly demonstrate the potential importance of beta cell distribution in the analysis of in vivo and in vitro experimental data, and in the design of effective clinical devices.