The differential diagnosis of Cushing's syndrome

The diagnosis of Gushing's syndrome remains one of the most challenging tas ks in clinical neuroendocrinology. The diagnostic procedure can be divided into two distinct steps: diagnosis of the neuroendocrine disorder and diffe rential diagnosis of the precise aetiology. The goat of the first laborator y tests is to obtain biochemical proof of Gushing's syndrome. Patients with Gushing's syndrome are relatively insensitive to glucocorticoid feedback a nd exhibit an oversecretion of cortisol devoid of a circadian cycle. In our experience, a low-dose dexamethasone suppression test provides the most re liable confirmation of steroid resistance, a cortisol level of < 50 nmol/l at 9 a.m. having 98 % sensitivity. A cortisol level below 50 nmol/l at midn ight rules out active Gushing's syndrome with, in our experience, 100 % sen sitivity and a specificity depending on numerous other variables. A very hi gh level of free urinary corticol can be a useful sign. After having establ ished the diagnosis of Gushing's syndrome, a persistently low level of ACTH (< 10 pg/ml), or preferentially an undetectable revel unresponsive to CRH (100 mug iv), is suggestive of an ACTH-independent disorder, and consequent ly of primary adrenal disease. The precise location of the lesion can ident ified with CT or MRI imaging, generally prior to surgical cure. if the ACTH level is detectable, patients with pituitary Gushing's syndrome, or Gushin g's disease, should be differentiated from those with ectopic ACTH secretio n. The secreting tumour may be difficult to localise and diagnosis is never 100 % sure with dynamic tests. Catheterisation of the petrosal sinus with CRH stimulation provides the best sensitivity for differentiating the two a etiologies. We consider a central to peripheral gradient of > 3 to confirm the pituitary origin of the disorder with a 98 % sensitivity. Chest or abdo minal CT can be helpful to identify an ectopic tumour but very small tumour s may go undetected. MRI can detect 60 or 70 % of all pituitary adenomas bu t is virtually non-contributive to the diagnosis of Gushing's disease in ch ildren.