Chronic hyperglycemia as a cause of insulin Resistance
Insulin resistance in patients with type 1 diabetes—a consequence of glucose toxicity?
Insulin resistance both precedes and predicts type 2 diabetes and therefore is not merely due to hyperglycemia . In the case of type 1 diabetes, it is clear that insulin resistance is an acquired and reversible phenomenon since insulin sensitivity is completely normalized during remission of the disease . In these patients, insulin resistance of glucose utilization is predominantly localized to skeletal muscle . Since the peripheral (although not the portal) insulin concentration is usually similar to that in normal subjects in insulin-treated patients , insulin deficiency cannot explain insulin resistance in skeletal muscle.
Induction of in vivo insulin resistance by hyperglycemia
In type 1 diabetic patients lacking endogenous insulin secretion, effects of hyperglycemia per se on insulin action can be examined. In such a study, insulin-stimulated glucose uptake was measured on two occasions, after 24 hours of hyperglycemia Induced by an intravenous glucose infusion (mean diurnal blood glucose 16 mmol L−1) and after 24 hours of normoglycemia (6 mmol L−1), in the face of maintaining identical diurnal serum insulin profiles by continuous subcutaneous insulin infusion [6,15] (Figure 27.2). After 24 hours of hyperglycemia, the rate of whole-body glucose uptake was consistently lower than after normoglycemia, demonstrating that short-term hyperglycemia can cause insulin resistance in humans [6,15] (Figure 27.2). These observations were subsequently confirmed [17–19]. Despite the strong suggestion that glucose per se was inducing insulin resistance in patients with type 1 diabetes, the complexity of metabolic regulation still left open the possibility that the insulin resistance might result from as yet unidentified factors. Animal models, however, gave clearer proof of the primacy of glucose in this phenomenon. In the rat, for example, removal of 90% of the pancreas causes insulin deficiency, hyperglycemia, and a 30% reduction in insulin-stimulated glucose utilization in skeletal muscle . Selective correction of hyperglycemia with phlorizin, which induces glucosuria via inhibition of glucose reabsorption in the proximal tubule, normalizes glycemia without changing serum insulin concentrations (Figure 27.3). These studies together with the human data provided the first evidence of the ability of the blood glucose concentration itself to regulate insulin sensitivity, and the phenomenon began to be referred to as “glucose toxicity” [7,20].
Hyperglycemia as a mediator of insulin resistance in type 2 diabetes
In type 2 diabetes, a large number of studies are consistent with the glucose toxicity concept although interpretation of these studies is sometimes not as simple as the studies of type 1 diabetes because of the complex pathophysiology of type 2 diabetes. For example, it is known that hyperinsulinemia can cause insulin resistance [21,22], and type 2 diabetic patients early in the course of the disease are often hyperinsulinemic. It could be argued, therefore, that any treatment that makes a subject more insulin sensitive will ultimately lower insulin levels, making them more insulin sensitive through that mechanism alone. Likewise, weight loss improves both glycemia and insulin resistance  but it is difficult to attribute the causality for the improved insulin sensitivity to glycemia alone in such studies. Nevertheless, the data available are at least consistent with the concept of glucose toxicity. Garvey and coworkers, for example, demonstrated that 3 weeks of intensive insulin therapy significantly improved the maximum glucose disposal rate, endogenous glucose output, and insulin secretory capacity , and similar findings were also obtained in other laboratories . The fact that basal insulin levels were unchanged in the weight loss study  and increased after intensive insulin therapy  suggest that insulinemia per se was not the driving force for the changes in insulin sensitivity. Furthermore, other therapeutic interventions such as sulfonylureas also improve insulin sensitivity, consistent with the concept that glucose was a principal driver of the insulin resistance prior to therapy.
Since that time, additional support for this concept has been obtained from a wide variety of in vitro and in vivo models, a few examples of which follow. The in vitro studies are particularly compelling because of the ability to isolate glucose as a dependent variable. Insulin-stimulated glucose transport in muscle strips of hyperglycemic type 2 diabetic patients was lower than that of normoglycemic subjects but prolonged exposure to normoglycemia completely reversed that defect . Isolated adipocytes also respond to high concentrations of glucose by desensitizing their glucose transport system . Numerous other seminal studies in animals that support the concept of glucose toxicity as a major determinant of insulin sensitivity have been reviewed in the past and will not all be recapitulated here [7,20].
Physiologic basis of glucose-induced insulin resistance in vivo
In insulin-treated patients with type 1 diabetes, in whom glucose toxicity appears to be the major cause of insulin resistance, direct quantitation of tissue glucose uptake during insulin stimulation using positron emission tomography has demonstrated that skeletal muscle is the predominant tissue responsible for the defect in insulin-stimulated glucose utilization . Since a substantial fraction of whole-body glucose oxidation is independent of insulin, rates of whole-body glucose oxidation have to be corrected for the estimated contribution of this component of glucose oxidation . When this is done in patients with type 1  or type 2  diabetes, the fractions of glucose oxidized and disposed of nonoxidatively (the sum of glycogen synthesis and nonoxidative glycolysis) are similar to those in normal subjects. These data suggest that regardless of the primary cellular process causing insulin resistance, the ultimate gate-keeper for cellular glucose uptake is located at the level of glucose transport or phosphorylation. In support of this, muscle glucose-6-phosphate concentrations are similar in insulin-resistant patients with type 1 diabetes under conditions where glucose flux is acutely normalized by hyperglycemia under hyperinsulinemic conditions as under conditions of normoglycemic hyperinsulinemia . Since both GLUT4-mediated glucose transport  and glucose
phosphorylation by hexokinase II  are regulated by insulin, either transport or phosphorylation could be rate-limiting for glucose disposal. Another hypothetical possibility is that glucose delivery could be rate-limiting for glucose disposal. Insulin, at physiologic concentrations, increases blood flow, depending on factors such as limb muscularity and physical fitness from −10 to 80% (mean in 75 studies around 20%) . During a similar time period and at similar insulin concentrations, glucose extraction, as determined from the AV-glucose difference across a limb, increased 1000 to 2000% within 30–60 minutes. Although defects in blood flow may be observed at supraphysiologic insulin concentrations in patients with type 1 or type 2 diabetes, studies performed at physiologic insulin concentrations locate impaired insulin action exclusively to glucose extraction . Taken together these data localize hyperglycemia-induced insulin resistance of glucose utilization to early steps in glucose uptake in skeletal muscle. This defect is accompanied by similar relative reductions in glucose oxidation and storage.