Modeling interstitial glucose kinetics in subjects with type I diabetes during physiological conditions

The relationship between interstitial glucose (IG) and plasma glucose (PG) is not fully understood. We investigated this relationship during physiological conditions in subjects with type I diabetes. Nine subjects were studied over 12 h. IG was measured every 30 and PG every 15 min. Nine alternative compartment models were formulated assessing one or more effects including insulin and saturability effects. Model parameters were estimated from IG using PG as the forcing function. Acceptable precision (CV<100%) was achieved for most parameter estimates. Model selection process was based on the Akaike criterion. The model which best explained the experimental data included both the effect of insulin on glucose transfer from plasma to interstitial fluid (IF) and the presence of zero-order removal of glucose from IF.