Corticotrophin releasing factor: responses in normal subjects and patients with disorders of the hypothalamus and pituitary.

Synthetic CRF-41 has been given to 43 patients with hypothalamic, pituitary or adrenal diseases and contrasted with the responses in 20 normal subjects. In the normal subjects the mean increment in serum cortisol (+/- SE) was 276 +/- 38 nmol/l; the increments showed a significant negative correlation with the basal serum cortisol levels (r = -0.56; P less than 0.02). The mean peak serum cortisol was 662 +/- 34 nmol/l and the mean peak corticosterone was 28.6 +/- 3.8 nmol/l. There was a significant positive correlation between the peak serum corticosterone and cortisol concentrations (r = 0.84; P less than 0.0001). Dexamethasone pretreatment abolished the rise in cortisol in response to CRF-41. The peak serum cortisol following CRF-41 was not significantly different between the normal subjects and those patients with pituitary disease who had normal cortisol responses to insulin-induced hypoglycaemia. However, in individual patients the peak cortisol levels induced by hypoglycaemia were greater than, but significantly correlated with, those induced by 100 micrograms of CRF-41. Seven patients were ACTH deficient in response to hypoglycaemia, and of these six responded normally to CRF-41. Only one of these patients had a lesion clearly originating in the hypothalamus; four had pituitary tumours with suprasellar extensions and the remaining patient had idiopathic GH and ACTH deficiency. Our data suggest that these patients have a functional defect of ACTH secretion due to the failure of CRF to reach the corticotroph. Of the four patients with pituitary-dependent Cushing's disease who were on no treatment at the time of testing, three showed an exaggerated and one a normal response to CRF-41. These normal or enhanced responses of hypercortisolaemic patients with Cushing's syndrome contrast with the complete inhibition of the responses to CRF-41 in normal subjects given dexamethasone. In the treated patients with Cushing's syndrome and normal serum cortisol levels, those with pituitary-dependent disease showed an enhanced ACTH response to CRF-41 as compared with the ectopic ACTH group, but there was some overlap between the two groups. Acromegalic patients did not show a GH response to CRF-41. We conclude that administration of CRF-41 is a safe new method for investigating disorders of the hypothalamo-pituitary axis.