RESUMO
PURPOSE: To describe the potential and limitations of selective arterial calcium stimulation and hepatic venous sampling (ASVS) in the evaluation of patients with hyperinsulinemic hypoglycemia. MATERIALS AND METHODS: Twenty-seven consecutive adult patients with documented hyperinsulinemic hypoglycemia undergoing an ASVS test and histologically confirmed diagnosis were included. RESULTS: By histology, 24 patients were found to have an insulinoma, two had islet cell hyperplasia, and one had insulin-secreting neuroendocrine carcinoma of the liver. ASVS correctly predicted the source of excessive insulin secretion in 26 of 27 (sensitivity, 96%) patients. In two patients with an insulinoma, ASVS had to be repeated for correct diagnosis; in one patient with a glucose-sensitive insulinoma, ASVS was falsely negative. In one patient, a falsely negative ASVS test was caused by diazoxide treatment; in another patient a falsely negative ASVS test occurred because of extraordinarily high baseline insulin levels when the calcium stimulation was performed. CT and/or MR imaging correctly predicted the source of excessive insulin secretion in 59% of patients, the sensitivity of the intraoperative exploration was 88%. CONCLUSIONS: In the authors' experience, the ASVS test is superior to CT/MR imaging and even the intraoperative exploration in identifying the source of excessive insulin secretion in patients with hyperinsulinemic hypoglycemia. To prevent false negative ASVS tests, medications with an influence on insulin secretion have to be discontinued prior to ASVS and two baseline blood samples should be obtained before the calcium stimulation is performed.