Effects of the glucocorticoid antagonist RU 486 on pituitary-adrenal function in patients with anorexia nervosa and healthy volunteers: Enhancement of plasma ACTH and cortisol secretion in underweight patients

To further explore whether the hypercortisolism of anorexia nervosa reflects an alteration in the set point for corticotropin-releasing hormone (CRH) secretion or is a manifestation of glucocorticoid resistance, we examined plasma ACTH and cortisol responses to the competitive glucocorticoid antagonist RU 486 (10 mg/kg, p.o. at 8.00 h) versus placebo (PBO) in 7 healthy female volunteers and 8 patients with DSM-III-R anorexia nervosa, all of whom were studied while underweight [64.3 ± 2.1% average body weight (ABW), mean ± SE] and 5 of whom were restudied longitudinally following refeeding (≥ 85% ABW, mean 87.4 ± 0.4% ABW). Blood samples were obtained from 16.00 to 16.30 h and from 4.00 to 8.00 h following dosing. Underweight anorexics were significantly hypercortisolemic by 24 h urinary free cortisol excretion compared with controls (239 ± 37 vs. 119 ± 12 nmol/day, p < 0.01). Both controls and underweight anorexics had robust early morning (4.00-8.00 h) plasma cortisol responses to RU 486 (465 ± 61 and 719 ± 49 nmol/1) compared with PBO β70 ± 52 and 451 ± 31 nmol/1; p < 0.02 and p < 0.01, respectively). The underweight anorexics showed a significant mean early morning plasma ACTH response to RU compared with placebo β.28 ± 0.63 vs. 2.01 ± 0.24pmol/l, p < 0.05), while the controls showed a trend toward an increase in mean plasma ACTH after RU (3.11 ± 0.36 pmol/1) compared with PBO (2.31 ± 0.41 pmol/1, p < 0.13); plasma ACTH means were greater on the RU day than the placebo day at 20 of 25 sampling points (p < 0.001). However, the increment in ACTH on the RU day compared to the placebo day was greater in the underweight anorexics at the first 20 of 25 consecutive time points of the early morning sampling period (p < 0.001). Moreover, underweight anorexics showed a significant plasma ACTH and cortisol response to RU 486 at 16.00-16.30 h (8-8.5 h following administration), while the controls showed no significant response of plasma ACTH or cortisol at this time. When restudied following weight recovery, anorexic patients showed reductions in 24-hour urinary free cortisol excretion (to 191 ±40 nmol/dav) which were no longer significantly elevated compared with control values. During this phase the anorexic patients did not show a significant plasma ACTH increment following RU 486 administration (2.21 ± 0.33 vs. 2.19 ± 0.31 pmol/1), but did show a significant plasma cortisol response to this agent (523 ± 39 vs. 418 ± 35 nmol/l, p < 0.01). The plasma cortisol response to RU 486 was relatively greater than the plasma ACTH response in the underweight anorexics compared to controls, and tended towards normal at weight recovery. Plasma RU 486 levels did not differ between groups. These data suggest that hypercortisolism in underweight anorexics reflects hypersecretion of hypothalamic CRH rather than primary glucocorticoid resistance, and are compatible with our previous findings showing adrenal hyperresponsiveness to ACTH in underweight anorexics, with persistent alterations in HPA function following short-term weight restoration.