Assessment of subcutaneous glucose concentration: validation of the wick technique as a reference for implanted electrochemical sensors in normal and diabetic dogs.

Employing saline-impregnated cotton threads, an implanted-wick technique was adopted in dogs to obtain specimen from the subcutaneous interstitial compartment in order to estimate its glucose concentration. By measuring the protein, potassium and haemoglobin contents, the centrifuged wick fluid was shown to contain the interstitial concentration of solutes after an equilibration time of approximately 15 min. In normal and in diabetic animals the steady state subcutaneous glucose concentration was almost identical to the circulating glucose level when ranged between 2 and 25 mmol/l. Slow alterations in the circulating glucose profile such as those which appear during an oral glucose tolerance test are closely mirrored by the respective levels in the wick fluid. Fast alterations, however, show deviations. The wick-based glucose levels are well paralleled by the current of Clark type glucose oxidase sensors implanted at the same site. Since, on the basis of in vitro calibrations the sensor outputs have only indicated apparent tissue glucose concentrations of between 70 and 90% of glycaemia, another reference is needed for calibration. Under steady state conditions, the wick method, and on this basis in routine measurements the blood glucose concentration, may be recommended as a reference of implanted sensors which can otherwise not be calibrated in situ.