Several studies, culminating in the large and definitive Diabetes Control and Complications Trial and the United Kingdom Prospective Diabetes Study [1,2] have established that hyperglycemia is the major risk factor for microvascular diabetic complications. Chronic hyperglycemia also seems to be a significant and independent, albeit weaker risk factor for macrovascular disease . Other adverse consequences of hyperglycemia include an increased susceptibility to infections . These complications affect tissues wherein glucose utilization is predominantly non-insulin-dependent. In addition, high concentrations or flux rates of glucose can exert adverse metabolic consequences on insulin-dependent tissues that regulate glucose disposal. High concentrations of glucose can by themselves cause two of the principal hallmarks of type 2 diabetes, namely deficiencies of insulin secretion and insulin action, both in animals [4,5] and in humans .
The term “glucose toxicity” was first applied to these phenomena by Rossetti and DeFronzo in 1990 , although the term had been used earlier to describe other adverse effects of hyperglycemia. In this review, we will restrict the use of the term to the ability of excess glucose to alter normal glucose homeostasis itself, and further restrict the discussion to those effects that are mediated by metabolism of glucose through its normal pathways.Thus, we will exclude effects of hyperglycemia that are, for example, mediated by nonenzymatic glycation or hyperosmolarity, even though these are of unquestioned importance in diabetes. We also acknowledge that the use of the term “toxicity” may not be the most appropriate for at least two reasons: Firstly and most obviously, glucose does not fit most definitions of “toxin,” and secondly, some of the “toxic” manifestations of glucose such as insulin resistance can also be seen as adaptive mechanisms that actually protect the organism from other, potentially even worse consequences of high nutrient fluxes. We also emphasize that glucose toxicity is but one of many pathogenetic mechanisms operative in type 2 diabetes. There are other routes to insulin resistance and β-cell failure such as inflammation, lipotoxicity, and others that operate in concert with glucose toxicity to create and maintain the full diabetic phenotype.
In this review, after a brief summary of the pathophysiology of hyperglycemia, we will review the evidence that hyperglycemia can cause insulin resistance and impaired insulin secretion. We will then consider the underlying molecular mechanisms for, and finally the clinical relevance of, glucose toxicity.
Hyperglycemia and glucose utilization
The mass-action effect of glucose
In virtually all tissues except the brain, glucose, at a fixed insulin concentration, promotes its own utilization in a concentration-dependent manner (Figure 27.1) . In insulin sensitive tissues, the glucose-induced increase in glucose utilization is dependent upon the insulin concentration. For example, an increase in the blood glucose concentration from a fasting concentration from 5 to a peak postprandial concentration of 8.9mmol L−1 (from 90 to 160 mg dL−1) increases the rate of whole-body glucose utilization two- to threefold more in the presence of postprandial (serum insulin 50–160mUL−1) than fasting (insulin 10–20mUL−1) serum insulin concentrations . This mass-action effect of glucose is quite significant if one compares it to the rate of whole-body glucose utilization which averages ∼2 mg kg−1 min−1 after an overnight fast, and increases two- to threefold to ∼5–7 mg kg−1 min−1 after a meal . As discussed later, the mass-action effect of glucose has important implications for the understanding of the mechanisms underlying glucose toxicity.
Consequences of the acute stimulatory effect of hyperglycemia on postprandial glucose disposal in patients with type 1 and 2 diabetes
Whole-body glucose uptake can be accurately quantitated, using the insulin clamp technique during maintenance of similar glucose and insulin concentrations in normal subjects and diabetic patients . Under such conditions, insulin resistance is observed, that is, the rate of glucose uptake is, on average, reduced in both patients with type 1 and 2 diabetes . If,
however, the rate of glucose uptake is measured in patients with type 1  or 2  diabetes under conditions simulating the actual postprandial glucose and insulin concentrations found in these patients, the rate of absolute insulin-stimulated glucose uptake is normal. This is explained by the ability of hyperglycemia, via glucose mass-action, to compensate for the reduction in glucose utilization caused by insulin resistance. In the absence of any insulin resistance, the mass-action effect of glucose would lead to overutilization of glucose not only in non-insulin-dependent but also insulin-dependent tissues. In non-insulin-dependent tissues not protected by the blood–brain barrier such as peripheral nerves, the kidney, and retina, the rate of glucose utilization is chronically elevated under hyperglycemic conditions, and is a key pathophysiologic abnormality leading to diabetic microvascular complications [1,2]. In contrast, the normal glucose flux to insulin-sensitive tissues, such as skeletal muscle, spares them from hyperglycemia-induced damage.