The dexamethasone-suppressed corticotropin-releasing hormone stimulation test and the desmopressin test to distinguish Cushing's syndrome from pseudo-Cushing's states

Abstract

Objective: Cushing's syndrome (CS), when fully expressed, is easily diagnosed. Mild cases, however, may require careful distinction from pseudo-Cushing's states as may occur in depression, alcoholism, polycystic ovary disease and visceral obesity. The aim of the present study is a reappraisal of the diagnostic accuracy of the two tests most commonly used to differentiate CS from pseudo-Cushing's: corticotropin-releasing hormone (CRH) stimulation after low dose dexamethasone administration and desmopressin stimulation.

Design: The study population comprised 32 patients with CS and 23 with pseudo-Cushing's evaluated retrospectively.

Methods: Urinary free cortisol (UFC), serum cortisol at midnight and after low dose dexamethasone (1 mg overnight and 2 mg over two days) were measured. Further, patients were tested with dexamethasone + CRH and desmopressin and the diagnostic performances of the two tests were compared in the entire series as well as in patients with mild hypercortisolism only (i.e. UFC < 690 nmol/24 h).

Results: As expected, measurement of UFC, assessment of cortisol rhythmicity and inhibition after 1 mg/2 mg dexamethasone failed to clearly classify patients with pseudo-Cushing's. Administration of CRH following 2-mg dexamethasone achieved 100% sensitivity but 62.5% specificity. Conversely, desmopressin testing correctly classified all but two patients with pseudo-Cushing's (90% specificity) with 81.5% sensitivity. Diagnostic accuracy was comparable in the subgroup with mild hypercortisolism (21 CS, all 23 pseudo-Cushing's patients). Desmopressin offered an incremental diagnostic effectiveness of 35.8/million inhabitants compared with dexamethasone + CRH as a second-line test.

Conclusions: Neither of the two tests guarantees absolute diagnostic accuracy. The specificity of dexamethasone + CRH is less brilliant than previously reported and appears to be inferior to desmopressin stimulation. The greatest diagnostic effectiveness results from the low-dose dexamethasone test combined with the desmopressin test. Skilful use of dynamic testing and balanced clinical judgement are necessary to distinguish between Cushing's syndrome and pseudo-Cushing's.