Cortisol cut-points for the glucagon stimulation test in the evaluation of hypothalamic pituitary adrenal axis

Diagnosis of adrenal insufficiency requires evaluation by dynamic stimulation tests. The insulin tolerance test (ITT) is accepted as the gold-standard test for the evaluation of hypothalamo-pituitary-adrenal (HPA) axis but the test is unpleasant and dangerous. Although it takes more time, glucagon stimulation test (GST) is a good alternative to ITT. The primary aim of this study was to compare the ITT and GSTs in the evaluation of HPA axe in patients with pituitary disorders. We conducted a prospective study in which ITT and GST were performed within 7 days in 81 patients. Serum cortisol was measured. We divided our population in Group 1 (G1): Adrenal Insufficiency (Peak cortisol under ITT <200 ng/mL) and Group 2 (G2): normal response (Peak cortisol under ITT >200 ng/mL). Receiver-operating characteristic (ROC) analysis was performed to identify the thresholds for GST. The mean peak of cortisol under GST was not significantly different from that obtained after ITT in the whole cohort (182.67 ± 89.07 ng/mL vs. 179.75 ± 79.01 ng/mL), and it was significantly reduced in patients of G1 (p < 10^-3). ROC curve analysis showed that the best diagnostic accuracy was obtained with a peak cortisol cut-off to GST of 167 ng/mL (sensitivity, 89%; specificity, 79%). Using this cut-off, 86.4% of the patients were correctly classified. In our prospective series, GST is a potential accurate and safe alternative test for the assessment HPA. Test-specific cut-offs should be applied to avoid misinterpretation.