Insulin resistance in hypertension quantified by oral glucose tolerance test: comparison of methods.

Four methods reported in the literature for evaluation of insulin sensitivity indexes from oral glucose tolerance tests (OGTTs) were analyzed and compared in order to test their ability to discriminate the insulin-resistant state in hypertension. To this aim, 15 normoglycemic subjects, not affected by metabolic syndrome, underwent a 22-sample, 300-minute OGTT. Eight subjects were normotensive (mean age, 47.0 +/- 4.2 years) and 7 were hypertensive (mean age, 53.6 +/- 1.6 years). The following insulin sensitivity indexes were computed and compared: (a) 2 indexes, ISIE22/300 and ISIE8/180, provided by an integral equation (IE) method applied to the full OGTT and to a reduced 8-sample, 180-minute data subset, respectively; (b) 2 indexes, ISOGIS180 and ISOGIS120, computed by the oral glucose, insulin sensitivity (OGIS) method, which only requires 3 blood samples taken within 180 and 120 minutes, respectively; (c) an index, ISISI, which considers fasting and mean insulinemia and glycemia measured during a 5-sample, 120-minute OGTT; and (d) an index, ISMCR, which considers body mass index and requires 2 blood samples taken within 120 minutes. Except the ISOGIS180, all other indexes were able to detect a significant reduction (unpaired Student t test, P < .05) of insulin sensitivity in our hypertensive group compared with the normotensive group. Failure of ISOGIS180 was explained by the fact that this index did not capture the information portrayed by the peak of insulinemia in hypertensive patients, which occurred around the 90th minute. Intraclass correlation coefficients higher than 0.89 demonstrated a substantial agreement between ISIE22/300 and ISIE8/180 indexes. These are the only indexes characterized by units of measure consistent with the definition of insulin sensitivity as the ability of insulin to enhance glucose effectiveness.