Twenty-six consecutive patients with ACTH-dependent Cushing syndrome were subjected to simultaneous, bilateral inferior petrosal sinus sampling for ACTH assay before and after ACTH-releasing hormone (CRH) stimulation. The baseline ACTH inferior petrosal sinus/periphery (IPS/P) ratio was > or = 2 in 12 of 26 patients (46%), whereas the CRH-stimulated IPS/P ratio was > or = 3 in 19 of 26 patients (73%). A pituitary adenoma, ACTH-secreting at immunostaining, was surgically proved in all of the 19 patients who had an ACTH IPS/P ratio > or = 2 basally or > or = 3 after the CRH test but also in three other patients who did not have such ratios. The value of the basal IPS/P ratio and the complete lack of ACTH increase after CRH led to the diagnosis of an ectopic ACTH syndrome in four patients: a bronchial carcinoid was found in three patients, and the site of the tumor was still unknown in the other. In conclusion, the CRH test improved the diagnostic accuracy of inferior petrosal sinus sampling from 61.5% (12 pituitary, 4 ectopic) to 92.0% (19 pituitary, 4 ectopic). Thus it should be performed during the diagnostic process.