Assessment of hyperinsulinaemia at the termination of the prolonged fast

Background: Traditional criteria to diagnose hyperinsulinaemic hypoglycaemia are based on insulin measurements by unspecific insulin assays. This study was performed to test whether these traditional criteria can be applied when insulin is measured by specific immunoassays. Methods: 29 consecutive patients undergoing a prolonged fast were included; 11 patients with insulinoma and 18 healthy individuals. We determined plasma glucose, insulin, C-peptide, proinsulin, and beta-hydroxybutyrate concentrations at the termination of the fast. Insulin was measured by an unspecific radioimmunoassay (RIA) and a specific enzyme-linked immunosorbent assay (ELISA). Results: In 11 insulinoma patients, insulin concentrations at median plasma glucose concentration of 2.1 (range 1.3-2.5) mmol/l were 170 (76-340) pmol/l measured by RIA and 61 (11-156) pmol/l by ELISA. Insulin concentrations measured by RIA confirmed hyperinsulinaemia (i.e., > 36 pmol/l, the proposed cut-off value for traditional insulin assays) in all insulinoma patients, whereas insulin concentrations measured by ELISA were < 36 mol/l in four patients. In three insulinoma patients, insulin concentrations measured by ELISA were < 18 pmol/l, a proposed cut-off level to diagnose hyperinsulinaemia for specific insulin assays. Conclusion: When insulin concentrations are measured by specific immunoassays in patients evaluated for fasting hypoglycaemia, traditional reference values cannot be applied. (C) 2004 Elsevier B.V. All rights reserved.