Relations between serum lipids, lipoproteins, blood pressure, and insulin metabolism were investigated in 158 healthy men aged 19-77 years and with body mass indexes (BMIs) of 19-41 kg • m" 2 . Mathematical modeling analysis of glucose, insulin, and C-peptide concentrations during an intravenous glucose tolerance test was used to measure parameters of insulin metabolism. In univariate analysis, both fasting and postglucose insulin concentrations showed significant positive associations with fasting serum triglyceride levels (r=0-33 and 0.38, respectively) and systolic (r=022 and 0.26) and diastollc (r=0.21 and 0.24) blood pressure and negative associations with high density lipoprotein subtraction 2 cholesterol (HDL 2 ; r=-0.21 and -0.25). In multivariate analysis, the associations between insulin and HDL 2 cholesterol concentrations were found to depend on triglyceride levels. Insulin resistance and basal pancreatic insulin secretion showed significant positive associations with serum triglycerides, which were independent of the effects of age, BMI, and fat distribution. Hepatic insulin throughput was independently associated with HDL 2 cholesterol. Associations of insulin-related variables with blood pressure were generally dependent on age and BMI. These results underline the importance of insulin sensitivity and insulin concentrations as determinants of triglyceride metabolism. They also indicate a close relation between hepatic insulin handling and HDL] concentration that is independent of triglyceride metabolism. has lent support to the possibility that elevated insulin concentrations may have a direct role in atherogenesis. 4 Furthermore, elevated insulin concentrations are associated with raised blood pressure 5 and serum triglyceride concentrations 6 and with reduced serum high density lipoprotein (HDL) cholesterol concentrations, 7 factors that in themselves may increase coronary heart disease risk.Resistance of glucose elimination to insulin action (insulin resistance) is a major determinant of the plasma insulin concentration 8 and may have a stronger relation with cardiovascular disease risk factors than does plasma insulin concentration.9 -11 Determinants of insulin concentrations, such as insulin resistance, may therefore be more directly involved in the relations between insulin and cardiovascular disease risk than are the insulin concentrations themselves. Computer modeling of plasma glucose, insulin, and C-peptide concentrations during an intravenous glucose tolerance test (FVGTT) allows quantification of the determinants of plasma insulin concentrations, which may be otherwise difficult to measure. We have used this technique in a large group of healthy men to investigate the associations of serum lipid and lipoprotein concentrations and blood pressure with insulinrelated variables. A wide range of variation in age and adiposity in this group provided the necessary variation in insulin metabolism against which these associations could be examined.
Methods
SubjectsWe studied 158 apparently heal...