BACKGROUND:Lowering blood glucose concentration slows or prevents the development of complications in diabetes. One of the tools to control glucose levels is continuous glucose measurements. A promising technique involves measurements from glucose sensors implanted directly in skin/subcutis. However, in vivo bioinstability and drift in sensor signals have been reported after implantation, suggestively caused by the infiltration of inflammatory cells and adhesion of proteins to sensor membranes. The aim of this study was to evaluate the in vivo biocompatibility of two electrochemical glucose sensors after implantation in the skin of pigs. METHODS:In vivo biocompatibility of in-house fabricated electrochemical glucose sensors and a commercially available continuous glucose monitoring system (CGMS, Medtronic MiniMed, Northridge, CA) implanted 1 h, 2 h, 24 h, 3 days, or 7 days was examined by histological and immunohistochemical techniques. RESULTS:The extent of inflammation increased significantly as a function of time. The inflammation ranged from an acute focal fibrinous/suppurative dermatitis to a chronic fibrinous and granulating foreign body dermatitis 7 days after implantation. Immunohistochemical stainings showed that heterophilic granulocytes, macrophages, and fibrinogen/fibrinogen fragments D and E were consistent findings. Infiltration of CD3epsilon-positive T-cells was primarily confined to day 7 of implantation. In addition, the pro-inflammatory cytokines interleukin-1 and tumor necrosis factor-alpha played a role in the reaction to sensors. CONCLUSION:The reported in vivo bioinstability of sensors is likely to be caused by protein and cellular biofouling on the sensor membrane. Furthermore, the consistent finding of fibrinogen and fibrinogen fragments D and E at the sensor-tissue interface seems to play an important role in the pathogenesis as it possibly maintains the inflammation by promoting the recruitment of inflammatory cells to the implantation site.