A lipid-parameter-based index for estimating insulin sensitivity and identifying insulin resistance in a healthy population.

ABSTRACT

AIM: Insulin resistance needs to be identified as early as possible in its development to allow targeted prevention programmes. Therefore, we compared various fasting surrogate indices for insulin sensitivity using the euglycaemic insulin clamp in an attempt to develop the most appropriate method for assessing insulin resistance in a healthy population. METHODS: Glucose, insulin, proinsulin, glucagon, glucose tolerance, fasting lipids, liver enzymes, blood pressure, anthropometric parameters and insulin sensitivity (Mffm/I) using the euglycaemic insulin clamp were obtained for 70 normoglycaemic non-obese individuals. Spearman's rank correlations were used to examine the association between Mffm/I and various fasting surrogate indices of insulin sensitivity. A regression model was used to determine the weighting for each variable and to derive a formula for estimating insulin resistance. The clinical value of the surrogate indices and the new formula for identifying insulin-resistant individuals was evaluated by the use of receiver operating characteristic (ROC) curves. RESULTS: The variables that best predicted insulin sensitivity were the HDL-to-total cholesterol ratio, the fasting NEFA and fasting insulin. The use of the lipid-parameter-based formula Mffm/I=12x[2.5x(HDL-c/total cholesterol)-NEFA] - fasting insulin appeared to have high clinical value in predicting insulin resistance. The correlation coefficient between Mffm/I and the new fasting index was higher than those with the most commonly used fasting surrogate indices for insulin sensitivity. CONCLUSION: A lipid-parameter-based index using fasting samples provides a simple means of screening for insulin resistance in the healthy population.