Biocompatibility of an enzyme-based, electrochemical glucose sensor for short-term implantation in the subcutis.

BACKGROUND Continuous glucose measurements provide improved glycemic control and may prevent hypoglycemia and long-term complications of diabetes. One of the most promising techniques is the short-term implantation of electrochemical glucose sensors in subcutis. However, the inflammatory reaction to these sensors may lead to bioinstability of sensor measurements. The purpose of the present investigation was to examine factors contributing to the observed subcutaneous inflammatory reaction to an enzyme-based electrochemical glucose sensor for continuous glucose measurements. The sensor biocompatibility was assessed in vitro and in vivo. METHODS A toxicological assessment was performed on sensor materials and leachables, and the endotoxin content of sensors was determined by a Limulus amoebocyte lysate (LAL) test. Moreover, as a consequence of permanent penetration of the skin by the sensor the role of bacterial migration to the tissue was investigated. In vivo biocompatibility was investigated through histological examination of implanted sensor membranes for 3 days in pigs. Additionally, the effect of needle size and type (normal vs. inserter needle) on tissue trauma at sensor insertion was evaluated, and the healing of subcutis was assessed histologically from 3 to 14 days after removal of sensors. RESULTS The toxicological assessment and the LAL test showed no concerns in a 3-day implantation scenario, and bacterial migration to the subcutis could not be detected. The histological examination showed that a reduction in needle size reduced the extent of inflammation to very low levels, and that the different sensor membranes showed similar extent and type of inflammation. Additionally, the extent of subcutaneous tissue reaction after removal of sensors declined gradually over time and returned to near-normal levels after 2 weeks. CONCLUSION The electrochemical enzyme-based glucose sensor for continuous glucose measurements in subcutis is acceptable from a biocompatibility point of view. Reducing the inserter needle in size reduces the trauma induced at sensor implantation to neglible levels. Furthermore, the tissue reaction to the sensor returns to near-normal 2 weeks after the sensor has been removed following a 3-day implantation period.