Simplifying the intravenous glucose tolerance test.

When used for classifying pregnant patients as normal or abnormal in terms of carbohydrate intolerance, the k value computed from a two-sample intravenous glucose tolerance test (IVGTT) (10 and 60 minutes) is reliable except in a zone of +/- 0.10 k points from the appropriate lower limit of normal. Retesting with six samples obtained in one hour may be required for patients with values within the doubtful range, +/- 0.10 of the lower limit of normal. Considerable savings can be realized nevertheless with the use of this test as compared with doing a full, seven-sample test on each subject. Our results suggest that the abbreviated IVGTT be used more widely as a routine screening procedure for diabetes rather than being restricted to use on suspect populations. One can make a direct conversion of blood sugar readings into k with the use of a table presented in the paper.