This study aimed to evaluate the usefulness of the 48-hour fasting test and insulin surrogates followed by a glucagon stimulatory test (GST) for the diagnosis of insulinoma.
Thirty-five patients with suspected insulinoma who underwent 48-hour fasting test and GST were retrospectively included in our study: 15 patients with surgically proven insulinomas and 20 patients in whom insulinoma was clinically ruled out. We determined the duration of the fasting test, plasma glucose levels, serum levels of immunoreactive insulin and C-peptide, and insulin surrogates (serum levels of β-hydroxybutyrate, free fatty acid, and response of plasma glucose to intravenous glucagon [ΔPG]) at the end of the fast.
The sensitivity and specificity of the 48-hour fasting test were 100.0% and 80.0%, respectively, for the diagnosis of insulinoma. When the 48-hour fasting test and immunoreactive insulin, C-peptide, or insulin surrogates were combined, the combination with GST showed the best results. The sensitivity, specificity, and accuracy rate were 93.3%, 95.0%, and 94.3%, respectively, with 1 false-negative case and 1 false-positive case occurring.
A more accurate and less invasive diagnosis of insulinoma was possible by combining the 48-hour fasting test with the GST, compared with the existing method.
From the *Department of Medicine and Bioregulatory Science, Graduate School of Medical Sciences, Kyushu University; †Department of Gastroenterology, National Hospital Organization, Kyushu Medical Center; ‡Department of Anatomic Pathology, Graduate School of Medical Sciences, Kyushu University, Fukuoka, Japan; and §Digestive Diseases Branch, National Institutes of Diabetes, Digestive, and Kidney Diseases, National Institutes of Health, Bethesda, MD.
Received for publication April 14, 2016; accepted October 6, 2016.
Address correspondence to: Tetsuhide Ito, MD, PhD, Department of Medicine and Bioregulatory Science, Graduate School of Medical Sciences, Kyushu University, 3-1-1 Maidashi, Higashi-ku, Fukuoka 812-8582, Japan (e-mail: firstname.lastname@example.org).
The authors declare no conflict of interest.