A study has examined the rates of adrenal crises in patients treated with pituitary or adrenal surgery. Rates were substantial (approximately 9 per 100 patient years), perhaps representing suppression of corticotrope ACTH secretion and deprivation of normal corticotrope number postoperatively. Hormone withdrawal syndrome may have contributed to the rates of apparent adrenal crises given the definition used. Higher rates were seen in patients given relatively high dose glucocorticoids postoperatively in one of the two centres where patients were treated - perhaps some of the patients in the high dose centre had longer periods of corticotrope suppression from exogenous glucocorticoids, increasing the risk period for adrenal crises. The question of optimal glucocorticoid dose and weaning rate after cure of Cushing's syndrome remains a balance between weaning at a rate sufficiently rapid to allow resumption of normal corticotrope function thereby preventing adrenal crises and providing sufficient glucocorticoid support to avoid hormone withdrawal syndrome or even precipitating an adrenal crisis, in the vulnerable 4-6 month period after successful surgery. There is likely to be considerable inter-individual variability in optimum glucocorticoid dose and weaning rate so that close clinical and biochemical monitoring is currently a practical approach.