Jichi Medical School Tochigi, Japan 329-0498
In the August 1998 issue of Journal of Clinical Endocrinology and Metabolism, Legro et al. (1) showed a possibility that a fasting glucose to insulin ratio could be a potential parameter of insulin resistance in women with polycystic ovary syndrome (PCOS). The subjects had almost normal fasting glucose concentrations, but their insulin levels were widely scattered between 10 and 85 µU/ml (1). According to this marked difference in fasting insulin levels, this ratio could serve as an estimate of insulin sensitivity in the individual subjects.
A product of fasting glucose and insulin has also been proposed as a simple measure of insulin resistance (2, 3, 4). This measure has already been used, especially in large epidemiological studies (5, 6). As Legro et al. kindly commented, we have shown that the product of glucose and insulin could be a potential predictor of troglitazone efficacy in type 2 diabetic subjects (7). The product showed a good correlation with the decrease in HbA1c during the 3 months troglitazone therapy (r=-0.638, P<0.0001), suggesting that the product reflects insulin resistance fairly well in those subjects. However, in the same studied subjects, the ratio of fasting glucose to insulin was poorly associated with the decrease in HbA1c (r=0.351, P=0.0245). Theoretically, the difference between these two indices of insulin sensitivity becomes apparent when both fasting glucose and insulin increase. For example, if both glucose and insulin double in parallel, the ratio of glucose to insulin remains unchanged, whereas the product is 4-fold greater than the initial estimate. Such a parallel increase in fasting glucose and insulin is often observed in type 2 diabetic subjects. Thus, the confounded estimation of insulin resistance in diabetic subjects having both hyperglycemia and hyperinsulinemia might result in the poorer association between the ratio of glucose to insulin and the troglitazone efficacy in our previous study.
We agree with the conclusion of Legro et al. (1) that the ratio of fasting glucose to insulin could be a marker of insulin resistance in PCOS subjects with normal fasting glucose. However, it is rather complicated to use these two indices according to fasting glucose levels: the ratio for subjects with normal fasting glucose and the product for those with hyperglycemia. Alternatively, we speculate that the product could also be an estimate of insulin sensitivity in PCOS subjects with normal fasting glucose levels. We would like to know if our assumption is true.
Footnotes
Received October 5, 1998. Address all correspondence to: Shoichiro Nagasaka, Division of Endocrinology and Metabolism, Department of Medicine, Jichi Medical School, 3311-1 Yakushiji, Minamikawachi, Tochigi 329-0498, Japan.
References