ABSTRACT
The transient difference between the blood and the subcutaneous glucose concentrations after insulin injection is most substantial (Figure 2). Nevertheless, when this difference is modeled and corrected through an appropriate algorithm, the blood and subcutaneous fluid concentrations are well correlated [14]. In the simplest correction algorithm, it is assumed that after glucose injection the lag is of 3 min and after insulin injections it is of 9 min. Even with these simple assumptions, 93% of the subcutaneous and intravenous sensor-measured glucose concentrations are clinically accurate and the remaining 7% are clinically acceptable in experiments where the glycemia is forced to change very rapidly in the normal rat [11]. In the brittle diabetic chimpanzee, there is also a significant difference between the blood and subcutaneous glucose concentrations when the glucose concentration declines rapidly following insulin injection (at a rate >1.8 mg dL−1 min−1). This difference is, however, not clinically significant when the concentration changes at a rate smaller than 1 mg dL−1 min−1 [15].
