Stabilized needle electrode system for in vivo glucose monitoring based on open flow microperfusion

Preliminary in vitro studies and in vivo performance of amperometric glucose needle enzyme electrodes incorporating an open microflow technique, in which the sensor surface in subjected to a flow of fluid, are reported. Initially using a slow flow (60 microliters h-1) of isotonic phosphate buffer over the enzyme electrode tip, an interface was created which reduced cellular/protein fouling for electrode measurements in whole blood. Here a minor reduction in electrode response (apparent only at high glucose concentration) occurred which was not cumulative and therefore not associated with fouling. The protection afforded by the moving aqueous film was independent of fluid composition; the use of isotonic/hypertonic buffer, addition of anticoagulant (1% m/v heparin) or enhanced fluid viscosity (addition of 1% v/v glycerol) did not affect the system. Implantation of the electrode and its microflow cannula into subcutaneous tissue in rats was associated with a decreased buffer flow (30 microliters h(-1)) and had the effect of (i) reducing electrode stabilization (30 min), (ii) accelerating 'pick up' of tissue glucose changes after intravenous glucose (1-2 min lag) or insulin (3-7 min lag) and (iii) achieving a correlation between tissue and blood glucose values under dynamic conditions (r2 = 0.98, y = 0.99x + 0.23). Reassessment of the electrode response in vitro, following a 4 h monitoring period, provided a sensor response within 3% of the original electrode sensitivity, indicating little or no surface fouling and avoiding the requirement for repeated in vivo calibrations at least over the initial implantation period.