We have previously shown that when tested in the morning, older men and women, pretreated with metyrapone to block endogenous cortisol synthesis, exhibit delayed suppression of plasma ACTH in response to cortisol infusion. To confirm this finding and to determine whether aging-related changes in feedback responsiveness are exaggerated near the time of the circadian nadir in adrenocortical secretion, we performed a similar study in the evening. Healthy young (20 -35 yr, n ϭ 22) and old (Ͼ65 yr, n ϭ 21) men and women were administered metyrapone orally (750 mg) at 1600 and 1900 h, followed by a cortisol infusion of 0.06 mg/kg/h for 150 min. Blood samples were taken at 15-min intervals for 4 h following infusion onset for measurement of plasma ACTH, cortisol, 11-deoxycortisol, and corticosteroid binding globulin. When corrections were made for differences in circulating cortisol concentrations achieved among age and gender subgroups, feedback inhibition of ACTH was found to be significantly greater in young than in old subjects of both genders. Our studies support the hypothesis that glucocorticoid responses to stress in aging individuals are likely to be prolonged due to blunted and delayed inhibition of ACTH secretion, thus increasing the total exposure to glucocorticoids. (J Clin Endocrinol Metab 86: [545][546][547][548][549][550] 2001) A GING IN RODENTS has been found to result in prolonged glucocorticoid responses to stress and impaired feedback inhibition of ACTH secretion by glucocorticoids (1-3). These age-related modifications of hypothalamicpituitary-adrenal (HPA) function have been linked to deficits in learning and memory (4 -6). Until recently, it has not been clear whether similar changes occur in the human HPA axis during aging. Several groups have reported age-associated increases in resting cortisol levels, especially near the nadir of the cortisol circadian rhythm (7-16). The magnitude of pituitary-adrenal responses to ovine corticotropin-releasing factor (CRF) has generally not been found to change with aging in humans (17, 18). However, when human CRF has been used to provoke the HPA axis, greater ACTH and cortisol responses have been found in older individuals (19,20), both with and without concurrent administration of arginine vasopressin (19).Impairment of glucocorticoid feedback inhibition of ACTH secretion, one of the primary age-related changes in HPA function reported in animal studies, has been supported by some, but not all, human studies. Pavlov et al. (17) concluded that the finding of a trend toward increased ovine CRF-stimulated levels of plasma ACTH and cortisol in older subjects in the face of age-related increases in resting cortisol was "consistent with the hypothesis that aging is associated with decreased pituitary sensitivity to negative feedback regulation by glucocorticoids."Studies of age-related changes in feedback inhibition using dexamethasone have reported age-related impairments (11, 20 -26) or no change (18,(27)(28)(29). In these studies, dexamethason...