[Detection of risk of insulin resistance in the population]

To assess the validity of hyperinsulinemia as a marker of insulin resistance (IR) and to find other common risk factors (RF) associated with IR.

Sample of 91 healthy volunteers with normal glucose tolerance (NGT) was selected from the adult Pilsen population. Following examinations according to the standard protocol were done: medical history, physical examination, body mass index-BMI (kg/m2), waist to hip ratio (WHR), casual blood pressure (BP), plasma lipids, 7-point oral glucose tolerance test (oGTT), IRI (immunoreactive insulin) curve, insulin suppression test (IST) for the evaluation of IR. Specificity and sensitivity of IRI levels as markers of IR were tested. Association between IR measured by IST and fasting and postload IRI was evaluated by single correlation. Multiple logistic regression (MLR) was applied for the calculation of IR prediction. Fasting and postload hyperinsulinemia (mU/l) have very good specificity (about 90%) for the discovery of IR, but sensitivity was different: fasting IRI > 20 = 19%, IRI in the 2nd h of oGTT > 90 = 30%, IRI sum (fasting IRI + IRI in the 1st and 2nd h of oGTT) > 150 = 51%. Simple correlation between IRI and IR was better in IRI sum (0.56 p < 0.001) than fasting IRI (0.28, p < 0.05). Using MLR HDL-ch, TG and IRI sum were selected as the significant and independent factors for the prediction of IR in the subjects from the population (prediction accuracy about 68%).

Fasting IRI is not too good marker for IR assessment in the NGT subjects from the Pilsen population. IRI sum was the best marker for IR and together with HDL-ch and TG level were selected as the significant predictive factors for IR.