Evaluation of the quantitative insulin sensitivity check index as an estimate of insulin sensitivity in humans.

The goal of this study was to compare estimates of insulin resistance generated by the quantitative insulin sensitivity check index (QUICKI) with a direct measure of insulin-mediated glucose disposal in healthy, nondiabetic volunteers. For this purpose, the results of measurements of plasma glucose and insulin concentrations in 490 nondiabetic, healthy subjects were used to compute several surrogate estimates of insulin resistance, and these values were compared with a direct measure of insulin-mediated glucose disposal. The results of this analysis showed that estimates of insulin resistance derived from use of QUICKI were significantly correlated (r = -.60, P <.001) with direct measures of insulin-mediated glucose in the 490 subjects studied. It was also noted that QUICKI estimates of insulin resistance were highly correlated with fasting insulin concentrations (r = -.98) and the homeostasis model assessment for insulin resistance (HOMA-IR, r = -.99). On the other hand, the correlation between all 3 of the surrogate methods for estimating insulin resistance and the direct assessment of insulin-mediated glucose disposal was relatively weak, i.e., r =.61, r =.64, and r = -.60) for fasting insulin concentration, HOMA-IR, and QUICKI, respectively. The results of these comparisons do not provide support for the superiority of QUICKI over other commonly used surrogate measures of insulin resistance based upon use of fasting insulin concentration or equations utilizing fasting insulin and glucose concentration. Furthermore, none of the 3 surrogate estimates can account for more than approximately 40% of the variability of the difference in insulin-mediated glucose disposal measured directly in 490 healthy, nondiabetic volunteers. Copyright 2002 by W.B. Saunders Company