Objective: There is no endocrine test which is completely reliable for the confirmation of Cushing's syndrome and in separation of the various aetiologies. We have tested the hypothesis that overnight dexamethasone pre-treatment should result in a better performance of the lysinevasopressin (LVP) test in the diagnosis of Cushing's syndrome. Study Design and Patients: We studied 61 subjects, including 25 pituitary-dependent and 9 pituitary independent Cushing's (7 adrenal tumours and 2 ectopic ACTH syndromes), 18 euadrenal controls, 4 depressed subjects, and 5 cushingoid patients. The subjects received 1 mg of dexamethasone orally at 2300 h and the following morning they were given 10 IU of lysine-vasopressin im. Measurements: Plasma cortisol (RIA) was measured at times -15, 0, 15, 30, 45, 60, 75, 90 and 120 minutes. Results: The dexamethasone-modified LVP (Dx/LVP) test resulted in four patterns of cortisol response. The dexa sensitive pattern (positive suppression and negative response