Loperamide, an opiate agonist of high specificity for u-receptors, was recently reported to suppress ACTH and cortisol levels in normal subjects, but not in patients with proven ACTH-dependent Cushing's disease. However, there is little information on the site of action of loperamide in the hypothalamo-pituitary-adrenal axis of man. We investigated the effect of loperamide on pituitary hormone secretion in vivo and in vitro. In seven normal subjects, basal ACTH plasma levels were significantly suppressed 3 h after loperamide administration (16 mg, orally) from 5 +/- 1 to 2 +/- 0 pmol/L (P < 0.0001). After the combined pituitary stimulation test (100-mu-g human CRH, 100-mu-g GnRH, 100-mu-g GH-releasing hormone, and 200-mu-g TRH), the ACTH peak (maximum increase at 30 min) was significantly blunted by loperamide from 9 +/-1 to 4 +/- 1 pmol/L (P < 0.001) and the area under the curve of ACTH from 0-120 min was reduced from 35 +/- 5 to 23 +/- 4 pmol/L.2 h (P < 0.05). In the insulin-hypoglycemia test (0.15 IU/kg BW), neither the ACTH peak nor the area under the curve of ACTH was affected by loperamide. In six patients with Cushing's disease and one patient with secondary adrenal insufficency due to hypothalamic failure, neither basal ACTH and cortisol levels nor CRH-stimulated levels were influenced by loperamide. In four cultured human corticotropic adenomas, loperamide was not able to reduce basal and CRH-induced ACTH secretion. In summary, loperamide is able to reduce basal and CRH-induced ACTH and cortisol levels in normal subjects, but not in patients with Cushing's disease or secondary adrenal failure of hypothalamic origin. Loperamide has no significant effect on insulin-hypoglycemia-induced ACTH and cortisol levels and, therefore, no effect on stress-induced elevation of cortisol levels. Loperamide might act at a suprapituitary site in man in vivo, but, nevertheless, a pituitary site cannot be excluded.