RESUMEN
BACKGROUND AND OBJECTIVE: The aim of this study was to assess the utility of arterial calcium stimulation with hepatic venous sampling (ASVS) in the localization of tumors in patients with endogenous hyperinsulinism not detected with other methods. PATIENTS AND METHODS: We performed a retrospective study of 26 patients admitted to our hospital for hypoglycemia who underwent ASVS because the source of hyperinsulinism was not clearly identified by other imaging techniques. The histopathological result in patients who underwent a surgical procedure was considered the reference for statistical study of the accuracy of this technique. Statistical analysis was performed by comparing proportions with the chi-squared test with Yates' correction for contingency tables, and Cohen's kappa coefficient as a measure of interrater agreement between two observations. RESULTS: Surgery was performed in 17 patients, 13 with positive ASVS and the remaining four with negative results. An insulinoma was removed in 12 patients, and 10 of these were detected in the ASVS. A total of 76.9 % of positive ASVS tests corresponded to a histological diagnosis of insulinoma, and 83% of these insulinomas were positive in ASVS. This association was statistically significant (chi cuadrado=7.340; p=0.012). Two of three patients with nesidioblastosis had a positive response in the ASVS. A good and statistically significant agreement was obtained between histopathologic diagnosis and ASVS results (kappa=0.556, p = 0.007). CONCLUSIONS: ASVS is a useful procedure in the localization diagnosis of endogenous hyperinsulinism not detected by other imaging tests. This technique allows tumors in the pancreatic gland to be identified and may be useful in the choice of the surgical technique to be used.