OBJECTIVE To compare the subcutaneous glucose sensor measurements with two reference methods. Previous studies provide conflicting findings about the real glucose concentrations in subcutaneous tissue. Some suggest substantially lower concentration, whereas others measure proportionally higher glucose concentrations compared with the blood compartment. Before these results can be taken seriously as an expression of the real glucose concentration in the extracellular space, the measurements must be validated by an independent method.

RESEARCH DESIGN AND METHODS We applied a microdialysis-based enzyme sensor to measure glucose concentration in subcutaneous tissue. We also developed two reference methods: subcutaneous filtrate collection and an equilibration method using ultrafiltration membranes to support the earlier findings. We provided an anatomical model to explain the results.

RESULTS The mean overall intercellular filtrate glucose concentration, sampled with the filtrate collector and taken after a 6-h stabilization time, including the values during the glucose clamp period, was 46 ± 9%. The mean subcutaneous glucose concentration measured with the glucose sensor, calibrated in vitro, was 44 ± 8% of the mean venous blood glucose concentration. Mean overall intercellular equilibrate glucose concentration, i.e., the mean glucose concentration in the subcutaneous extracellular space, taken after a 4-h stabilization time, was 46 ± 15% of the mean venous blood glucose concentration.

CONCLUSIONS The close agreement between the mean values of subcutaneous glucose concentrations, obtained with three independent methods— filtration, equilibration, and dialysis (sensor)— shows the real glucose concentration in subcutaneous interstitial fluid is — 50% the blood glucose value in normal humans. Our results clarify some of the conflicting evidence presented in previous studies.