Home / Resources / Clinical Gems / International Textbook of Diabetes Mellitus, 4th Ed., Excerpt #85: Measuring Insulin Action In Vivo

International Textbook of Diabetes Mellitus, 4th Ed., Excerpt #85: Measuring Insulin Action In Vivo

Aug 8, 2017

Insulin action dynamics

Insulin sensitivity from the clamp is expressed in terms of the steady-state value reached after a period of hyperinsulinemia. A dose–response can be constructed and sensitivity can be expressed as the ED50, that is, the concentration for half-maximal stimulation of Rd as a function of dose. Alternatively, sensitivity can be expressed (as it usually is) as the Rd at a specific dose of insulin, which may be low, intermediate, or a maximum dose. It is often assumed that steady-state rate of glucose uptake is reached by 180 min after onset of the insulin infusion. In fact, steady-state glucose uptake is not reached at 3h [64,65]. Glucose uptake rate at 3h is only 2/3 the “true” steady-state, which is not achieved until 6h (Figure 15.5). Certainly, shorter periods (e.g. 120 min) are inadequate to reflect steady-state rates of uptake [43,63,66].


Rd at “steady-state” does not incorporate the events that account for the slow time course of activation of glucose uptake [43,63,67]. Events that may determine the time course of glucose uptake in the face of hyperinsulinemia include dynamics of plasma insulin itself [66], access of insulin to the insulin-sensitive tissues (primarily skeletal muscle, liver, and adipose tissue) [68,69], binding of insulin to receptors, and the downstream signaling pathways and mobilization of glucose transporters (primarily glucose transporter 4 (GLUT4; [70–77]). It has been of interest to examine the in vivo pathway of insulin action during clamps to identify the rate-limiting step which determines insulin dynamics in vivo. It is possible that changes in access to insulin-sensitive cells could contribute to a “dynamic” insulin resistance, because insulin action does not generally reach the hyperinsulinemic steady-state value assessed during euglycemic clamp studies.

Access to sensitive cells includes the actual process of move- ment of insulin across the endothelial cells as well as any effect of insulin to access so-called “nutritive” capillaries (i.e., those that perfuse insulin-sensitive and glucose-utilizing or glucose-storing tissues). By measuring insulin in plasma and lymph simultaneously, we were able to determine that it was the movement of insulin between the plasma and the interstitial fluid compartment that was the determinant of insulin action dynamics [68,78].

It is clear that cellular sensitivity to the hormone is not the only action of insulin on carbohydrate metabolism which results in insulin resistance. The ability of the hormone to access sensitive tissues is reduced because of reduced transendothelial transport as well as failure to increase flow through nutritive capillaries. The relative importance of these dynamic components of insulin action under different conditions remains to be established. Given the various dynamic features that contribute to overall insulin action, it is reasonable to utilize the dynamics of insulin action, rather than quasi steady-state alone to represent in vivo insulin sensitivity. One approach that utilizes insulin action dynamics is the “minimal model” approach.

Dynamic insulin action: the minimal model

In the normal 24-h day, the insulin-sensitive tissues are never exposed to steady-state conditions. After a meal, glucose and insulin are changing, and the rate of glucose uptake lags in time behind the time courses of glucose and insulin. It may be preferable to measure insulin action from a dynamic relationship between glucose, the primary nutritive carbohydrate, insulin, the primary anabolic hormone, and the rate of glucose disposition. One approach to such a measurement is the frequently sampled intravenous glucose tolerance test (FSIGT) coupled with the “minimal model.” The underlying concept behind the minimal model approach will be outlined herein, but the reader should refer to more extensive presentations for added information and detail [43,60,79–83].


As discussed, intervening to “open the loop” to perform a glucose clamp requires resources in dollars and labor. Using the power of the digital computer and modeling obviates the need to maintain constant plasma glucose and insulin to obtain a sensitivity measure. To measure dynamic sensitivity the insulin/glucose regulating system must be perturbed. While the administration of oral glucoseChapter15Fig15.6 might be considered the simplest approach given the acceptance of the OGTT as a diagnostic procedure, oral glucose has substantial limitations due to lack of control over the rate of glucose appearance in the extracellular fluid [84,85] (see later). Therefore, intravenous administration was the protocol of choice. The explicit detailed dynamic relationship between glucose and insulin following intravenous glucose is measured by collecting frequent blood samples (Figure 15.6).

While many hormonal and neural signals can impact plasma dynamics, it has been shown that in the main, a few factors dominate the time course of glucose and insulin following glucose injection. These are (1) endogenous glucose production (by liver predominantly and also kidney); (2) insulin secretion and insulin clearance; (3) effect of insulin to alter the rate of glucose uptake by insulin-dependent tissues (mostly skeletal muscle) by a variety of mechanisms including transporter mobilization, enzyme activation, changes in perfusion of nutritive tissues, endothelial transport; and (4) glucose uptake by tissues independent of insulin (brain, gut, red blood cells).

Click here to view all Chapter 15 references.