The Effect of Dexamethasone on the 24-Hour Profiles of Adrenocorticotropin and Cortisol in Cushing's Syndrome*

ACTH and cortisol are normally secreted episodically rather than continuously. This characteristic of episodic secretion is preserved in patients with Cushing's syndrome. To determine whether exogenous glucocorticoids modulate this pulsatility and to study its possible etiological implications, we obtained 24-h plasma cortisol profiles in seven patients with Cushing's syndrome (five Cushing's disease, one adrenal adenoma, and one bilateral adrenal cortical macronodular hyperpla-sia) before and during suppression with various doses of dexa-methasone low (0.5 mg, every 6 h), high (2 mg, every 6 h), and very high (4 mg, every 6 h). Simultaneous 24-h plasma ACTH profiles were obtained in two patients with Cushing's disease. Blood was drawn at 30-min intervals for 25 h. Individual profiles were analyzed to determine the 24-h mean level, the presence of a circadian component and its amplitude, and the number and magnitude of significant secretory pulses over the 24-h span. The concordance between significant ACTH and cortisol pulses also was quantified. Baseline values in patients were compared o t those in seven normal subjects. Under basal conditions, the 24-h mean cortisol level was 3-t o 4-fold higher than normal in all patients with Cushing's syndrome. In contrast, the basal 24-h mean ACTH level was normal in one, and slightly elevated in the other of the two patients with Cushing's disease in whom plasma ACTH concentrations were measured. However, in contrast to the normal subjects, all ACTH values were above 10 pg/ml even during the period of minimal secretion. Basal circadian variation in adre-nocortical activity, albeit of reduced amplitude, was found in four of five patients with Cushing's disease; it was absent in the patient with adrenal adenoma. Low dose dexamethasone reduced the 24-h mean cortisol level and increased the amplitude of the circadian rhythm, unmasking a diurnal rhythm in the single patient with Cushing's disease in whom no significant circadian periodicity was present in the basal condition. This effect was further increased with the high dose of dexamethasone, which concomitantly reduced the number and increments of the secretory pulses. A lesser effect was found in the patient with bilateral nodular hyperplasia, and no effect was seen in the patient with adrenal adenoma. ACTH pulsatility, but not diurnal rhythm, also was dampened by dexamethasone. Reduction in the magnitude, but not the number, of ACTH secretory pulses by dexamethasone produced a reduced concordance ratio of ACTH with cortisol pulses of 0.39, compared to 0.64 in the basal state. The concordance ratios of cortisol with ACTH pulses remained unchanged (0.89 and 0.88) during the basal state and the administration of high doses of dexamethasone, respectively. These data indicate that during dexamethasone suppression, fewer of the low amplitude ACTH pulses elicit a significant cortisol response and that episodic cortisol secretion is ACTH induced. These data do not support the concept of a defect in the neural clock generating the 24-h periodicity in Cushing's disease. They suggest that the expression of the diurnal rhythm is masked, since it could be magnified and even uncovered by dexamethasone, in association with the reduction in the episodic pulsatility. In adrenal cortical macronodular hyperplasia, the low but persistent circadian variation in cortisol, obliterated by high dose dexamethasone, is in agreement with the postulated dual pituitary and adrenal etiology of this condition.