Screening Tests for Hypercortisolism in Patients With Adrenal Incidentaloma

Background: To compare the diagnostic performance of different first-line screening tests for subclinical hypercortisolism (SH) in patients with adrenal incidentaloma (AI). Methods: We studied a series of patients with AI, with no clinical evidence of hormonal hypersecretion. For screening for SH, all patients performed 1-mg dexamethasone suppression test (1-mg DST), late night salivary cortisol (LNSC) and 24-h urinary free cortisol (UFC). A control group of patients with confirmed Cushing’s syndrome (CS) was used to calculate the diagnostic performance of the screening tests. Results: In the 83 patients with AI, morning cortisol after 1-mg DST was ? 1.8 ?g/dL in 69.9%, 1.9 to 5 ?g/dL in 26.5% and > 5 ?g/dL in 3.6%. LNSC was elevated in 20.5% and all patients had normal UFC levels. In the control group, composed of 50 patients with confirmed CS, all patients who underwent 1-mg DST had cortisol levels > 1.8 ?g/dL (1.9 to 5 ?g/dL in 16.2% and > 5 ?g/dL in 83.3%); LNSC was elevated in 93.8% and the UFC was increased in 85.4% of patients tested. Overall, for the screening of SH, the 1-mg DST presented a sensitivity and specificity of 100% and 69.9% with its lowest threshold (? 1.8 ?g/dL) and 83.3% and 96.4% with its highest threshold (< 5 ?g/dL). LNSC showed a sensitivity and specificity of 93.8% and 79.5% and the UFC of 85.4% and 100%, respectively. Conclusions: The 1-mg DST at its lowest threshold presented the highest sensitivity in identifying SH, but its low specificity encourages us to consider UFC levels, to reduce false-positive test results. J Endocrinol Metab. 2018;8(4):62-68 doi: https://doi.org/10.14740/jem510w