Glucose Sensors and the Alternate Site Testing-Like Phenomenon: Relationship Between Rapid Blood Glucose Changes and Glucose Sensor Signals

Abstract

In the last few years blood glucose meters have been developed allowing glucose measurements in capillary blood samples collected at sites other than the fingertips. The main reason for establishing this so-called alternate site testing (AST) was to sample blood from locations with reduced pain perception. It is well known that capillary blood glucose is closely correlated to systemic (i.e., arterial) glucose levels and that under steady-state conditions, glucose values measured in blood samples collected from alternate sites are virtually identical to those collected from the fingertip. However, during rapid changes in blood glucose levels, glucose concentrations in capillary blood samples from the fingertips can differ considerably in both domains (time and concentration) from those determined in capillary blood from alternate sites (i.e., the so-called AST phenomenon). Such differences can have serious clinical consequences (e.g., risky delays in hypoglycemia detection). There is evidence that all skin sites exhibiting a reduced blood flow (in comparison with the fingertip) within the superficial skin layers are prone to this AST phenomenon. Nearly all glucose sensors having been developed so far or being currently under development measure glucose levels at alternate sites and also in another compartment [e.g., interstitial fluid (ISF)] than blood. So, in principle they might be prone to an AST-like phenomenon (i.e., rapid changes in systemic blood glucose levels may also result in delayed ISF glucose readings). Our knowledge about the impact of an AST-like phenomenon on the performance of glucose monitoring systems is presently very limited. Glucose kinetics in the different compartments during dynamic systemic blood glucose changes have not been fully elucidated yet. If an AST-like phenomenon plays a role with glucose sensors should therefore be studied. Depending on the measurement technology used for the individual type of glucose monitoring system probably this phenomenon has a variable impact on the results obtained.