A High-Dose Intravenous Glucose Tolerance Test

Abstract

Previous studies suggested that the rapid intravenous glucose tolerance test has been an insensitive indicator of abnormal carbohydrate metabolism because the dose of glucose usually used (+ 25 g) is too small. Accordingly, in this study a larger dose (50 g/1.73 m2) was used. Tests were performed on 61 healthy nonobese subjects with no diabetic relatives, to determine the normal range of k values. Then the intravenous test was compared to the oral and cortisone oral tests in 72 subjects who were expected to have a high incidence of slightly abnormal glucose tolerance. In healthy subjects less than 50 years old there was no age effect The frequency distribution of k in this group was positively skewed (range 1.0-5.0, mode 1.5-1.9, mean 2.6). Because 95% of k values were greater than 1.5, this was chosen as the lower limit of normal. The intravenous glucose tolerance test was abnormal in 20 of 21 subjects with an abnormal oral test. It was also abnormal in 15 subjects with a normal oral test, in 13 of whom there were other reasons to suspect impaired carbohydrate metabolism. The intravenous test was abnormal in only 11 of 20 subjects with a normal oral but an abnormal cortisone oral test. It is concluded that the sensitivity of the high-dose intravenous glucose tolerance test is between that of the oral and the cortisone oral tests, and that it could become a useful alternative to the oral test for routine use.