Relative contribution of insulin sensitivity and [beta ]-cell function to plasma glucose and insulin concentrations during the oral glucose tolerance test

Abstract

Plasma glucose and insulin concentrations have been used in genetic studies as quantitative phenotypic traits and also as surrogates for insulin sensitivity and [beta ]-cell function. However, the significance of these traits in relation to insulin sensitivity and [beta ]-cell function was unknown. We examined how insulin sensitivity and [beta ]-cell function affected plasma glucose and insulin concentrations during the oral glucose tolerance test (OGTT). This is a cross-sectional study enrolling 105 glucose-tolerant subjects (64 females; age, 18 to 40 years; body mass index, 17.58 to 37.57 kg/m2; waist-to-hip ratio, 0.649 to 1.033 cm/cm). They participated in both OGTTs and hyperglycemic clamps. The relationship between plasma glucose and insulin concentrations and indices of insulin sensitivity and [beta ]-cell function was examined. Univariate analyses showed that insulin sensitivity index (ISI) had some influence on plasma insulin concentrations (r2 = .2623 to .3814) during the OGTT; however, it had only modest impacts on plasma glucose levels at 60, 90, and 120 minutes (r2 = .0537 to .1300). Neither first phase (1stIR) nor second phase insulin response (2ndIR) affected plasma glucose concentrations. Multivariate analyses showed an independent impact (all P [lt ] .0001) of ISI on plasma glucose concentrations at 60, 90, and 120 minutes and on plasma insulin concentrations at every time point except at 30 minutes. Except for plasma insulin concentration at 30 minutes, of which 24% of the variation can be explained by 1stIR, [beta ]-cell function (either 1stIR or 2ndIR) only had a very modest impact on 30-, 60-, 90- and 120-minute plasma glucose concentrations and on plasma insulin concentration at 60 minutes. In glucose-tolerant subjects, ISI plays an important role in determining postchallenged plasma glucose concentrations at 60, 90, and 120 minutes, as well as plasma insulin concentrations at fasting, 60, 90, and 120 minutes. However, [beta ]-cell function is only reflected in plasma insulin concentration at 30 minutes through 1stIR. Therefore, we conclude that it is essential to measure [beta ]-cell function in vivo if one plans to study the genetic influence of [beta ]-cell dysfunction.