Abstract:
For many diabetics, checking their blood sugar level several times daily, for years and decades, has become an unavoidable part of their therapy. Subject of this study was to analyse whether taking blood samples from the abdominal skin offers a more precise, more practical and less painful alternative to the established blood sampling from the finger pulp.
On five consecutive days, 63 diabetics and 16 healthy volunteers each sampled their blood sugar 5 times a day by taking parallel blood samples from both the finger pulp and the abdominal skin. For reference purposes, a lab control check was performed on a blood sample simultaneously drawn from the finger pulp. In more than 1,900 tests, the resultant readings for blood samples taken from the abdominal skin correlated closely with the self-administered readings from the finger pulp (r = 0.95). This was true for a wide range of readings that captured all clinically relevant blood sugar concentrations between values of 30 mg/dl and 350 mg/dl. The evaluation of the differences in measured value according to the method defined by Bland and Altman revealed that more than 95% of the paired values are found in the measuring range of +/- 2 standard deviations from the mean value of the differences. During the error grid analysis - important for the clinical aspects - 99.1% of the values were found in zones A and B. Comparison to the reference method led to matching therapeutic decisions in 99.7% of the cases.
Similarly, practical aspects (secondary bleeding, number of required punctures, consumption of test strips) manifested no differences. Nevertheless, most of the tested diabetics considered self-administered sampling from the abdominal skin just a temporary alternative meant to give the finger pulp a chance to recover.
Thus, the self-administered blood sugar check using the abdominal skin must principally be rated as equivalent method compared to established blood sampling from the finger pulp. The new method proved to be simple, easy to handle, and precise. The high level of statistical correspondence with the conventional method is, however, limited to stable metabolic situations, because the chosen test setup was unable - along the lines of the so-called alternative site testing phenomenon - to capture differences between abdominal skin and finger pulp readings whenever the blood sugar concentration was subject to rapid change.