To determine if an effective method existed for distinguishing the physiologic hypoglycemia of fasting from pathologic hypoglycemia, 72-hour fasts were conducted in 60 women and 20 men of normal weight, in 16 obese subjects, and in six of 11 patients with insulinomas. Only the pattern of change of the immunoreactive-insulin-to-glucose ratio (the I/G ratio), calculated at major time intervals of the fast, provided a clear-cut distinction between these groups; plasma glucose values alone could not make this distinction; The mean fasting I/G ratio was calculated for each subject from that subject's I/G ratios at 12-hour intervals during the fasting period. In no single case did the mean I/G ratio during fasting for an individual of normal weight equal or exceed the control I/G. I/G ratios increased dramatically during fasting in each patient with an insulinoma. Normal obese patients (15 per cent >ideal body weight) did not provide a diagnostic problem, since, regardless of sex, glucose values of less than 55 mg./dl. did not occur. Although the pattern of change of the I/G ratio was extremely useful, the basal I/G ratio alone was potentially misleading; this was due to overlap of basal I/G ratios between subjects with simple obesity and patients with insulinomas. In addition, absolute values for the I/G ratio varied with the technique employed for measuring glucose and insulin. Change of the I/G ratio, however, was independent of the techniques used for measuring glucose and insulin.

This content is only available via PDF.