SIMILARITY OF RELATIVE GLYCEMIC BUT NOT RELATIVE INSULINEMIC RESPONSES IN NORMAL, IGT AND DIABETIC SUBJECTS

To determine the validity of measuring relative plasma glucose and ins ulin responses of foods, we studied 4 subject groups: 10 lean normal, 9 obese normal, 9 with impaired glucose tolerance (IGT) and 8 with mil d diabetes (DM). Overnight fasted subjects consumed either 75g oral gl ucose or 50g carbohydrate from a standard test meal. Each subject repe ated each test 4 times on separate days using a randomized block desig n. The mean of the incremental plasma glucose and insulin response are as for each subject's 4 standard test meals were expressed as a % of t he respective mean for the 4 glucose tests and the resulting values te rmed relative responses. Relative glucose responses did not differ sig nificantly in lean (51+/-6), obese (41+/-4), IGT (52+/-4) and DM (54+/ -2) subjects. However, there was heterogeneity of mean relative insuli n responses (p=0.0003) with the values in obese normal (50+/-5), IGT ( 69+/-15) and DM (84+/-7) subjects being significantly different from e ach other. The mean in lean normal subjects, 61+/-3, was significantly less than DM, but no different from obese and IGT. These results supp ort the concept and clinical utility of the glycaemic index as a measu re of the blood glucose raising potential of foods that is valid in no rmal and glucose intolerant subjects. However, they suggest that the i nsulinaemic index is less clinically useful because it varies in diffe rent groups of subjects. (C) 1998 Elsevier Science Inc.