Impaired Glucose Tolerance

D of the category of impaired glucose tolerance by the National Diabetes Data Group (NDDG) and by the World Health Organization (WHO) has served to highlight the dearth of knowledge about the significance of glucose tolerance levels that fall short of being diagnostic of diabetes, but that are above normal if these are defined as glucose levels above the mean plus two standard deviations in healthy young adults with no family history of diabetes. As pointed out in the accompanying editorial, in some populations glucose concentrations, fasting or following an oral glucose load, are bimodally distributed, and when the logarithms of the glucose concentrations are used, the frequency distributions conform well to a model of overlapping gaussian distributions in which those with diabetes constitute the hyperglycemic component of distribution. What, then, does the lower component represent? One view might be that the lower component, whose upper boundaries extend well beyond those defined as normal in young healthy adults, is a reflection that glucose tolerance tends to diminish with increasing age, obesity, or physical inactivity and that in the general population the distribution is skewed and both the mean and the variance of glucose levels are greater. If so, glucose tolerance levels above the range seen in young healthy adults which now fall into the range labeled "impaired glucose tolerance" might still be considered to be of no particular significance. However, several observations appear to make this viewpoint untenable. As Stern and his colleagues points out, impaired glucose tolerance is associated with an increased risk for the development of diabetes, as well as an increased risk for the development of cardiovascular disease. Whether the latter can be "explained" on the basis of other risk factors and the glucose intolerance is merely an innocent bystander, or whether impaired glucose tolerance has a more direct, pathogenetic role has been a matter of considerable debate and controversy. However, the fact that diabetics have more coronary heart disease than nondiabetics, even after other risk factors such as age, smoking, cholesterol levels, and hypertension have been accounted for in the statistical sense,' certainly leaves some questions about the pathogenetic significance of impaired glucose tolerance in atherosclerosis. Many studies have shown that diabetes is much more likely to develop in persons with impaired glucose tolerance." This risk increases markedly as a function of the degree of the glucose intolerance, even when other risk factors, such as obesity or a family history of noninsulin-dependent diabetes, are taken into account." In general, the degree of glucose intolerance and family history of diabetes have been found to be the strongest predictors of the development of diabetes. The degree of obesity also has often, but not always, remained as a significant predictive factor, and in some studies other factors, such as very low density lipoprotein (VLDL) or increased serum cholesterol levels and high blood pressure, have been shown to be independent determinants. Impaired glucose tolerance is accompanied by hyperinsulinemia. Typically, after an oral glucose load such persons have an exaggerated and attenuated insulin response, which is indicative of impaired insulin action. Indeed, in subjects without diabetes the fasting or postload glucose and insulin levels are generally proportional to the degree of insulin resistance. Up to a certain point the pancreas attempts to compensate by increasing insulin output. Thus, persons with impaired