Second-line tests in the differential diagnosis of ACTH-dependent Cushing's syndrome.

ABSTRACT

INTRODUCTION: Diagnosing Cushing's syndrome (CS) can be a challenge, especially in ACTH-dependent CS, when it comes to detecting the origin of ACTH secretion. MATERIALS AND METHODS: Retrospective data were collected on 170 patients with ACTH-dependent CS (149 CD, 21 EAS) referring to two endocrinology units, focusing on three non-invasive tests dexamethasone 8 mg overnight challenge (HDDST); corticotrophin-releasing hormone (CRH) assay and the desmopressin (DDAVP) test. RESULTS: Patients with EAS were slightly older and had higher ACTH, serum and urinary cortisol levels than patients with CD (p < 0.01). CD patients had a stronger ACTH and cortisol response after CRH injection (p < 0.0001), and a more pronounced reduction in cortisol levels after HDDST (p < 0.0001). A threshold percentage ACTH increase after CRH stimulation of 72.4 % was able to identify CD with a sensitivity (SE) of 76 % (95 % CI 68-83) and a specificity (SP) of 100 % (95 % CI 83-100). As for HDDST, a cortisol suppression >52.7 % below the basal level suggested a pituitary origin with a SE of 88 % (95 % CI 81-93) and a SP of 90 % (95 % CI 68-99). There were no cases of EAS with positive responses to both these tests. Increases in ACTH and cortisol levels after the DDAVP test were also higher in CD than in EAS (p < 0.01), though the SE and SP were lower. CONCLUSIONS: Patients with CD showed a stronger response to HDDST and CRH, and the adopted cut-offs showed a good SE and SP in discriminating them from patients with EAS. Concordant tests indicated CD when positive, whereas no response to either test was highly suggestive of EAS. The DDAVP test was of limited utility in the diagnostic phase. In conclusion, the choice of tests may play an important part in the differential diagnosis of ACTH-dependent CS.