INSULIN SURROGATES IN INSULINOMA

Universally accepted criteria for relative hyperinsulinemia have not been established for the diagnosis of insulinoma. Therefore, we sought measures of insulin action which might act as surrogates for insulin measurements and thereby contribute to the assessment of hyperinsulinemia. Since insulin is antilipolytic, antiketogenic, and glycogenic we measured plasma beta-hydroxybutyrate, FFA, and the response of plasma glucose to iv glucagon at the end of the prolonged fast in 40 patients, later confirmed histologically to have insulinoma and 25 normal persons. Plasma beta-hydroxybutyrate and FFA concentrations were significantly lower in the patients with insulinoma (median, range), (0.3, 0.1-2.7 vs. 4.5, 1.2-7.0 mmol, P < 0.0001, and 1.03, 0.17-1.75 vs. 1.79, 1.17-3.12 mmol, P < 0.0001, respectively), whereas the responses of plasma glucose to glucagon were significantly greater (3.0, 1.4-5.4 vs. 0.7, 0.0-1.3 mmol, P < 0.0001) than in the normals. For patients with insulinoma (20/40) and normal subjects (13/25) with plasma glucose less than or equal to 3.3 mmol and plasma insulin and C-peptide concentrations in the normal overnight fasting range, conditions in which hyperinsulinemia is most difficult to assess, a clear distinction was provided by plasma glucose response to glucagon and plasma beta-hydroxybutyrate but not by plasma FFA, plasma insulin, nor plasma C-peptide. We conclude that plasma glucose response to iv glucagon greater than or equal to 1.4 mmol and plasma beta-hydroxybutyrate less than or equal to 2.7 mmol, at the end of the prolonged fast are indicative of hyperinsulinemia of insulinoma when the plasma glucose is less than or equal to 3.3 mmol. In this plasma glucose range these insulin surrogates provide better diagnostic accuracy than plasma insulin and C-peptide.