Modeling glucose disposal in diabetic dogs fed mixed meals.

We have utilized a previously described mathematical model to study glucose disposal in fed, conscious, ambulatory, diabetic dogs. The model was applied to estimate the daily disposition of ingested glucose in the periphery, liver, and urine following a regular mixed meal containing 130 g of carbohydrate. Experimental data was obtained from 11 pancreatectomized animals. Both the portal and peripheral routes were used for intravenous insulin infusion and the daily profiles of peripheral plasma glucose and insulin concentrations measured. Total calories in mixed meals derived from carbohydrates (37%), fat (30%), and protein (30%). When judged according to the root-mean-square differences, agreement was excellent between model-predicted and experimentally observed glucose as well as insulin concentrations. This agreement occurred whether or not, in addition to basal insulin, meal insulin was also given. Using the model, we then predicted in detail the rates of glucose uptake in peripheral tissue, liver, and kidneys. With portally infused insulin resulting in diurnal glycemic normalization, the net daily hepatic glucose balance was physiological, being close to zero. Remarkably, with peripheral insulin infusions there was an unphysiological net negative hepatic glucose balance of 10 g/day.

Glucose and insulin dynamics in the anaesthetized dog: a mathematical modeling study.

A combined circulation and organs model of glucose and insulin dynamics is presented. The model is based on physiological parameters, incorporating blood and plasma flow rates, circulatory paths, intra-and extra-vascular glucose and insulin spaces, as well as the specific organs and tissues involved both with insulin disappearance and with glucose production or uptake. Its simulations readily lend themselves to physiological interpretation. To explore its validity, the model was assigned parameters typical of a 12 kg dog and was arranged to accept known glucose and insulin infusions from four different experiments on such diabetic animals. It predicted the observed glucose and insulin concentrations as well as total uptake rates for both moieties. This confirmed the ability of the model to predict with consistency the group mean outcomes of these four experiments when differing routes (portal or peripheral) of infusion were applied. Excellent agreement for most studies was achieved while the need for including more sophisticated dynamics of glucose transport in the liver or into erythrocytes was identified. The model isolates glucose uptake in the periphery, the liver, the brain and the gut and allows a direct comparison of glucose disposal along various routes. Thus the total amount of glucose uptake by peripheral, insulin-dependent tissues is directly calculated to be 22-28% of an intravenous glucose load, regardless of its route of infusion.