Insulin: in search of a syndrome

The insulin resistance syndrome has been defined as the clustering of hypertension, dyslipidaemia and impaired glucose tolerance in subjects with high fasting plasma insulin concentrations. The latter are usually taken as a surrogate measure of insulin resistance of whole-body glucose disposal. Although hyperinsulinaemia and insulin resistance are reciprocally related to one another, the association is not very strong. In the data-pooling project of the European Group for the study of Insulin Resistance (EGIR), clamp-derived insulin sensitivity (as the M value) and fasting plasma insulin concentrations were available in 1308 non-diabetic subjects with a wide range of age and body mass index. In this cohort, hyperinsulinaemia (as the upper quartile of fasting plasma insulin distribution in the non-obese segment of the population) and insulin resistance (as the bottom quartile of M value in the same subgroup) were each present in approximately 40% of the whole population, but identified only partially overlapping (60%) subsets of individuals. When the subjects with insulin resistance but without hyperinsulinaemia (n = 198) were compared with the subjects with hyperinsulinaemia but without insulin resistance (n = 267), significant differences emerged in the respective clinical phenotypes. Thus, subjects with 'pure' insulin resistance had a more central fat distribution and presented evidence of excessive lipolysis and endogenous glucose production. In contrast, subjects with 'pure' hyperinsulinaemia had suppressed lipolysis, endogenous glucose production and insulin clearance, higher values of systolic blood pressure and lower values of serum HDL-cholesterol concentrations. The only abnormality common to both phenotypes was the presence of raised serum triglycerides concentrations. This analysis indicates that hyperinsulinaemia and insulin resistance identify partially different subgroups of individuals in a non-diabetic population, and suggests that hyperinsulinaemia and insulin resistance carry distinct pathogenic potential in terms of the components of the insulin resistance syndrome.