We thank Karne et al. for their comments (1) on our article (2). The ideal method to evaluate insulin resistance and insulin sensitivity, outside of the euglycemic-hyperinsulinemic clamp, has not yet been fully established. However, the minimal model, extensively used during the past decades, is considered a valuable surrogate of the clamp (3–5). The genetic factors involved in insulin sensitivity are better determined with the minimal model approach than with indirect methods based on fasting glucose and insulin values (6). Thus, the minimal model, as used in our study, provides more complete information on both, insulin resistance and insulin sensitivity than that given by indirect indexes (7–9). Although, as stated by Karne et al., the minimal model is not a direct measure of insulin sensitivity, the evidence in the literature supports its use as a valid surrogate of the clamp.
Minimal model results have been compared with those obtained by indirect indexes, establishing a good correlation. However, indexes based solely on fasting blood glucose and insulin could not always reliably estimate insulin resistance, since it is possible to have insulin resistance without hyperinsulinemia and conversely hyperinsulinemia without insulin resistance (10). A modified version of QUICKI recently published (11) provides a significantly better correlation with minimal model results than that obtained with QUICKI and homeostasis model assessment for insulin resistance. Interestingly, as observed by us, the values obtained with these latter two indexes were similar.
Karne’s comments on hypertensive subjects and estimation of insulin resistance with minimal model are not substantiated by the results published by other authors (11).
We consider that the minimal model provides reliable data on insulin sensitivity and insulin resistance. Indirect methods, such as those included in our study, are useful in the study of large number of subjects. It is possible that the predictive value of indirect indexes bears relationship with the evolutionary moment of the insulin resistance syndrome and the onset of metabolic abnormalities. In our study, all indexes correlated significantly with the minimal model results. The McAuley index and the clinical parameters of the metabolic syndrome were the best indicators of insulin resistance. The best indirect method is still to be defined. It is possible that the new modified version of QUICKI will provide some advantage.