Cost-effectiveness and accuracy of the tests used in the differential diagnosis of Cushing's syndrome.

AIM

Establish the minimal biochemical and radiological examinations necessary and their cost-effectiveness to accurately diagnose the etiology of Cushing's syndrome (CS).

MATERIAL AND METHODS

In 71 patients with CS followed between 1982 and 1997 biochemical studies (basal ACTH, 8 mg dexamethasone suppression test-HDST-, metyrapone stimulation test-MST-, or inferior petrosal sinus catheterization-IPSC-) and radiological investigations (abdominal CT scan, pituitary CT scan or MRI) were performed. Once pathology confirmed the diagnosis (48 pituitary Cushing's disease-CD, 17 adrenal neoplasms, 2 bilateral macronodular hyperplasia-BMH-, and 4 ectopic ACTH syndrome-ES-), the sensitivity, specificity, positive and negative predictive value of the different studies was calculated to establish the most accurate and cost-effective diagnostic protocol.

RESULTS

In ACTH-independent CS (ACTH < or = 9 pg/ml; normal 9 to 54) a unilateral tumor was identified on abdominal CT scanning in 17, and BMH in 1; the other BMH had detectable ACTH (43.2 pg/ml). In ACTH-dependent CS, ACTH was > 9 pg/ml and IPSC (performed in 22) correctly identified 20 patients with CD and differentiated them from 2 with an ES (100% specificity and sensitivity). Pituitary MRI or CT did not disclose an adenoma in 41.7% of patients with CD, and was reported to exhibit a microadenoma in 2 of the 4 patients with ES. HDST and MST were of no additional use in the differentiation between CD and ES.

CONCLUSIONS

Once CS is diagnosed low ACTH and an abdominal CT scan correctly identified all patients of adrenal origin. In ACTH-dependent CS IPSC was the best predictive test to differentiate CD from ES. BMH may behave as ACTH-dependent or independent. The other biochemical and radiological studies performed are not cost-effective and may even be misleading, and should not be routinely performed.