Purpose: Reverse iontophoresis extracts glucose across the skin in the GlucoWatch Biographer, a device to monitor glycemia in diabetes. However, the device must first be calibrated with an invasive "fingerstick" and this has been perceived as a disadvantage. Here, urea, a neutral "internal standard" is extracted simultaneously in an attempt to render the technique completely non-invasive.
Methods: In a 5-h experiment in human volunteers, reverse iontophoretic fluxes of glucose and urea (J(glu) and J(urea), respectively) were measured periodically and correlated with the corresponding blood levels. In the case of glucose, a finger-tip blood sample was taken at the beginning of each collection interval; for urea, three blood samples were assayed: one before, one during, and one at the end of iontophoresis.
Results: The ratio J(glu)/J(urea) divided by the ratio of the systemic concentrations (C(glu)/C(urea)) yielded an extraction coefficient (K) that could be compared between subjects. Though J(glucose) tracked C(glu) faithfully when the volunteers were challenged with an oral glucose load, J(urea) remained quite stable reflecting the fact that C(urea) did not change appreciably during the experiment. However, whereas the variability (expressed as the coefficient of variation) in the normalized extraction flux of urea (J(urea)/C(urea)) was on the order of 25%, that for glucose was greater (>45%), with the result that the values of K (0.45 +/- 0.25) were less constant than anticipated.
Conclusions: Although urea performed quite reasonably as an internal standard, in that its extraction flux and systemic concentration both remained quite constant, the normalized, transdermal, iontophoretic flux of glucose showed interindividual variability due to mechanisms that were not entirely governed by electrotransport. That is, despite good qualitative tracking to blood levels, there appear to be other (biochemical, metabolic, contamination?) factors that impact upon the quantitative results obtained.