We hypothesized that in nonectopic Cushing syndrome there is an insufficient activity of type II (renal) 11beta-hydroxysteroid dehydrogenase (11beta-HSD2) that is related to cortisol excess, rather than to corticotropin (adrenocorticotropic hormone [ACTH]) levels. We measured plasma ACTH and urinary-free cortisol (UFF), urinary-free cortisone (UFE), tetrahydrocortisol (UTHF), and tetrahydrocortisone (UTHE) in 24-h urine samples of 24 healthy subjects and 15 patients diagnosed with nonectopic Cushing syndrome. Then, in the group of patients, a new 24-h urine sample was collected after treatment with 800 mg daily of ketoconazole. The UFF/UFE and UTHF/UTHE ratios were calculated as an estimation of 11beta-HSD2 activity. The patients had an increase in both the UFF/UFE (19.95 +/- 10.3 vs 5.78 +/- 4.72 nmol/24 h; p < 0.0001) and UTHF/UTHE ratios (5.36 +/- 5.23 vs 1.39 +/- 0.95 nmol/24 h; p < 0.001). Both UFF/UFE and UTHF/UTHE ratios decreased after ketoconazole treatment (19.95 +/- 10.3 vs 12.2 +/- 6.9 nmol/24 h; p < 0.005; and 5.36 +/- 5.23 vs 1.62 vs 1.21 nmol/24 h; p < 0.001, respectively). The control subjects had a significant relationship between UFF and UFE (r = 0.70, p < 0.0001), and between UTHF and UTHE (r = 0.75, p < 0.0001) that did not exist in the patient group. After ketoconazole treatment, the decrease in cortisol excretion in the patient group allowed a positive and significant relation between UFF and UFE (r = 0.64, p < 0.01) and between UTHF and UTHE (r = 0.56, p < 0.05) to appear. There was not any significant relationship between either UFF/UFE or UTHF/UTHE ratios and plasma levels of ACTH.