Study of the effects of tissue reactions on the function of implanted glucose sensors

The relationship between tissue reactions to a subcutaneously implanted glucose sensor and the function of the sensor was evaluated over a period of 4-weeks using tubular, porous polyvinyl alcohol (PVA) sponges implanted subcutaneously in rats. The PVA sponges were used as scaffolds in which the foreign body response could develop. Coil-type glucose sensors were then placed in the center of the PVA sponges and tested on day 3, and weekly thereafter. In the first approach, the sensors were placed in the sponges still implanted in the rats and tested. In vivo glucose sensor sensitivity peaked at day 7 and steadily decreased until day 35. In the second approach, the sensors were placed in the explanted sponges and then tested. This test showed no sensor function after day 7, indicating that functional blood vessels are critical in maintaining any function whatsoever. In both cases the sensors themselves were never implanted to eliminate any potential in vivo degradation of the sensors that could have affected the outcome of this study. Sensors were then tested in absence of sponges and found to be working properly with no change from preimplantation sensitivity. Once sensor testing was concluded, the PVA sponge/tissue samples were prepared for quantitative histological analysis. It was determined that the increase in collagen deposition within the sponge correlated with the decrease in sensor sensitivity. It was also observed that natural angiogenesis (peak at day 14) did not overcome the barrier to glucose diffusion created by the fibrous capsule. (C) 2007 Wiley Periodicals, Inc.