Efficacy of insulin and C-peptide suppression test using a rapid-acting insulin analog to induce hypoglycemia in the diagnosis of insulinoma: A comparison to the supervised prolonged fast test

Abstract

Background: The efficacy of insulin and C-peptide suppression (ICPS) test using a rapid-acting insulin analog to induce hypoglycemia in the diagnosis of insulinoma has never been studied. Objective: To compare the efficacy of using plasma C-peptide (PCP) and plasma insulin (PI) responses to the ICPS test using insulin aspart and the supervised prolonged fast (SPF) test in the diagnosis of insulinoma. Methods: The ICPS test was performed in 15 patients with insulinoma (IN) and 6 patients with non-insulinoma causes of hypoglycemia (non-IN) by intravenous infusion of insulin aspart to induce hypoglycemia. Plasma glucose (PG), C-peptide (PCP), and insulin (PI) levels were measured before and at the end of the test (end-ICPS) when the patients had hypoglycemia, defined by the presence of either PG ≤50 mg/dL with hypoglycemic symptoms or PG ≤40 mg/dL regardless of hypoglycemic symptoms. PCP and PI were measured by an immunoassay system that does not cross-react to insulin aspart (Cobas Modular Analytics e801). The SPF test was also performed in IN. Results: IN had a higher median end-ICPS PI (34.77 versus 0.58 μIU/mL, p < 0.001), lower magnitude of PI suppression (−24.28 % ± 35.5 % versus −93.67 % ± 2.7 %, p < 0.001), higher mean end-ICPS PCP (4.62 ± 3.02 versus 0.49 ± 0.21 ng/mL, p < 0.001), and lower magnitude of PCP suppression (−25.51 % ± 22.2 % versus −70.28 % ± 19.4 %, p < 0.001) than non-IN. In IN, end-ICPS PI was significantly correlated to end-ICPS PCP (r = 0.724, p = 0.002). There were high correlations between end-ICPS PCP and end-SPF PCP (r = 0.882, p < 0.001) and between end-ICPS PI and end-SPF PI (r = 0.794, p < 0.001). The ICPS had a sensitivity and specificity of 93 % and 83 %, respectively, when using an end-ICPS PCP cut-off level of 0.6 ng/mL and 93 % and 100 %, respectively, when using an end-ICPS PI cut-off level of 3 μIU/mL. The ICPS test was terminated in a shorter time than the SPF test (1.01 ± 0.58 versus 7.80 ± 4.10 h, p < 0.001). Conclusion: In the diagnosis of insulinoma, the ICPS test using insulin aspart is practical, safe, less time-consuming, and as effective as the SPF test. The responses of PI to hypoglycemia are more obvious and consistent, without overlap, than the responses of PCP in IN and non-IN. The use of end-ICPS PI is better than end-ICPS PCP in the evaluation of the ICPS test. The ICPS test using a rapid-acting insulin analogue such as insulin aspart can be used instead of the conventional CPS test using recombinant human insulin and should be considered an alternative first-line test to the SPF test.