Insulin sensitivity by oral glucose minimal models: validation against clamp

Measuring insulin sensitivity in the presence of physiological changes in glucose and insulin concentrations, e. g., during a meal or OGTT, is important to better understand insulin resistance in a variety of metabolic conditions. Recently, two oral minimal models have been proposed to measure overall insulin sensitivity (S-I) and its selective effect on glucose disposal (S-I(*)) from oral tests. S-I and S-I(*) have been successfully validated against multiple tracer meal estimates, but validation against euglycemic hyperinsulinemic clamp estimates is lacking. Here, we do so in 21 subjects who underwent both a multiple-tracer OGTT and a labeled euglycemic hyperinsulinemic clamp. Correlation between minimal-model S-I, S-I(*) and corresponding clamp estimates S-I(*clamp), S-I(*clamp) was satisfactory, respectively r = 0.81, P < 0.001, and r = 0.71, P < 0.001. SI was significantly lower than S I clamp (8.08 +/- 0.89 vs. 13.66 +/- 1.69 10(-4) dl.kg(-1).min(-1) per mu U/ml, P = 0.0002), whereas S-I(*) and S-I(*clamp) were very similar (8.17 +/- 1.59 vs. 8.84 +/- 1.39 10(-4) dl.kg(-1).min(-1) per mu U/ml, P = 0.52). These results add credibility to the oral minimal-model method as a simple and reliable physiological tool to estimate S-I and S-I(*), also in large-scale clinical trials.