A commercial whole blood glucose biosensor with a low sensitivity to hematocrit based on an impregnated porous carbon electrode.

Erythrocytes (red blood cells) are a major source of response variation in biosensor electrodes expected to operate in whole blood. Such a blood-to-plasma difference (hematocrit effect) must be minimized for those sensors directed towards the hospital market where wide variations in hematocrit can be seen. Typically, many current glucose sensors demonstrate a decreasing response to the analyte in the presence of increasing hematocrit levels. A sensor electrode for glucose is described which displays a reduced sensitivity to changes in hematocrit. The working electrode comprises a base porous conducting carbon layer, which is impregnated with a mixture including glucose oxidase and a ferrocene redox mediator. The base carbon layer has a void volume of 50%, an average pore diameter of less than 0.1 microm and a thickness of about 20 microm. The interior void volume of the base carbon layer is filled entirely with a substantial proportion of the impregnating mixture such that very little remains on the exterior. The resulting impregnated porous electrode excludes erythrocytes and is consequently capable of operating acceptably in venous, capillary, arterial and neonatal blood over a wide hematocrit range of 20-70%.