ENHANCED ADRENOCORTICOTROPIC HORMONE AND CORTISOL RESPONSES TO CORTICOTROPIN-RELEASING HORMONE IN CENTRAL IDIOPATHIC DIABETES-INSIPIDUS

It is well known that arginine vasopressin (AVP) exerts a stimulatory effect on adrenocorticotrophic hormone (ACTH) secretion. Moreover, the re is consistent evidence that the hypothalamic AVP-secreting neurons are involved in the neuroregulation of ACTH secretion. With the aim to throw further light on the interaction between AVP and corticotrophin -releasing hormone (CRH) in the neuroregulation of ACTH secretion, in this study we compared the ACTH and cortisol responses to human CRH (1 00 mu g iv as a bolus) in 18 normal subjects (15 females and three mal es, age 22-35 years) and seven patients with central isolated diabetes insipidus (six females and one male, age 16-40 years). Two patients w ere newly diagnosed and five had discontinued substitution therapy wit h desamino-D-AVP 24 h before testing. All had free access to water bef ore and during the test period. The ACTH and cortisol responses to CRH were higher in subjects with diabetes insipidus than in controls, eit her when evaluated as peak values (ACTH, mean+/-SEM: 17.0+/-1.2 vs 7.7 +/-0.7 pmol/l, p=0.0003; cortisol: 611.3+/-59.4 vs 450.7+/-21.2 nmol/l , p=0.01) or area under curve values (ACTH: 672.5+/-75.7 vs 364.0+/-33 .6 pmol.l(-1).h(-1), p=0.002; cortisol: 29158.0+/-2937.0 vs 23236.7+/- 1052.1 nmol.l(-1).h(-1), p=0.03). These results show that patients wit h diabetes insipidus have an exaggerated pituitary-adrenal response to CRH. This may be due to the fact that in diabetes insipidus AVP secre tion from parvocellular neurons of the paraventricular nucleus in the hypophysial portal system is not impaired. Alternatively, AVP secretio n may be defective in both magnocellular and parvocellular hypothalami c AVP-secreting neurons. In this case, it could be hypothesized that a djustment is made to the feedback regulatory mechanisms of the hypotha lamic-pituitary-adrenal axis, so that the CRH-ACTH axis assumes a main role with respect to the AVP-ACTH axis.