Modified hyperinsulinaemic, eu- and hypoglycaemic clamp technique using lispro-insulin for insulinoma diagnostic

Characterization of metabolically inadequate insulin secretion is essential for insulinoma diagnostics. Hyperinsulinaemic, eu- and hypoglycaemic clamp procedures have been used to suppress endogenous insulin secretion in healthy subjects. The use of exogenous insulin precluded the use of insulin as a parameter to be measured. We now suggest to use exogenous insulin lispro and an insulin-specific ELISA not cross reacting with insulin lispro. Thus, determination of insulin by ELISA in this experimental setting reflects endogenous insulin. A 39-year-old man with a surgically confirmed pancreatic insulinoma was studied under hyperinsulinaemic [lispro insulin 40 mU x m(-2) body surface x min(-1)] clamp conditions. Euglycaemia was achieved (3.8 +/- 0.5 mmol/L) for 1 h and hypoglycaemia (2.36 +/- 0.49 mmol/L) was achieved for another 30 min. Insulin was evaluated by ELISA (cross-reaction with lispro insulin < 0.006%, C-peptide < 0.01%, proinsulin < 0.001%) and by a nonselective RIA (cross-reaction with proinsulin 40%). In control subjects the euglycaemic hyperinsulinaemia suppressed C-peptide to 0.36 +/- 0.03 ng/ml and hypoglycaemic hyperinsulinaemia to 0.29 +/- 0.03 ng/ml. Endogenous insulin was suppressed to 2.8 +/- 0.03 mU/L under euglycaemia and to 2.6 +/- 0.03 mU/L under hypoglycaemia in control subjects. In the insulinoma patient apparently irregular but small changes in both C-peptide (1.43 +/- 0.1 ng/ml) and more pronounced changes in endogenous insulin concentrations 4.41 +/- 0.1 mU/l under euglycaemia and 5.35 +/- 0.3 mU/l under hypoglycaemic conditions, were observed. The basal level of insulin (ELISA insulin 4.6 mU/L) and C-peptide (1.7 ng/ml) were not markedly elevated. Determination of insulin allowed better characterization of irregular pulses because of the shorter half-life of insulin relative to C-peptide. The new modification of sequential eu- and hypoglycaemic clamp procedures should also be useful in pharmacological studies of insulinotropic substances. Direct measurement of peripheral insulin may be more sensitive than C-peptide to detect low levels of autonomous insulin secretion in small insulinomas.