Endocrine Abstracts

Background: Distinguishing pseudo-Cushing syndrome (PCS) from mild forms of Cushing’s disease (CD) is a challenge. The dexamethasone-Corticotropin Releasing Hormone test (dex-CRH) assumes that cortisol response to CRH is preserved in CD while suppressed in PCS after the low dose dexamethasone suppression test (LDDST).

Methods: Persons with a suspicion of CD based on mild clinical features and a positive first-line screening test were included. They underwent a dex-CRH test between 2010 and 2020. Adrenal causes of hypercortisolism and women using estrogens were excluded. Those with a positive dex-CRH underwent neurosurgery. Those considered as PCS needed to be followed for at least 2 years and not develop new Cushingoid features.

Results: Dex-CRH was performed in 36 cases (F: 24) with a median age of 50 years (IQR; 32-56). Thirteen cases were diagnosed with CD (F: 11), 22 with PCS (F: 13) and 1 had an indeterminate result. The following Cushingoid features were present in cases diagnosed with CD vs PCS: moon face (15 vs 14%), buffalo hump (38 vs 27%), striae (0 vs 9%), myopathy (64 vs 19%), hypertension (54 vs 45%), type 2 diabetes (23 vs 23%) and BMI > 30 kg/m² (46 vs 77%). 7/13 CD cases had a biochemical mild form (UFC <2xULN). UFC was mildly elevated in 16/22 of the PCS group and clearly elevated (≥ 2x ULN) in 4/22 cases. Using a cut-off value of 38 nmol/l, LDDST was positive in 12/13 cases with CD and in 0/22 with PCS and the ensuing CRH-stimulation was positive in respectively 13/13 vs 1/22 cases. Mean serum cortisol level after LDDST in the CD group was 226.5 nmol/l compared to 16.8 nmol/l in the PCS group. After dex-CRH test mean serum cortisol was 433.1 nmol/l compared to 18.8 nmol/l, respectively. Increasing the cut-off value to 87 nmol/l led to exclusion of al PCS cases but also misdiagnosis of 1 CD case.

Conclusions: The prevalence of various Cushingoid features between mild CD and PCS was quite similar except for myopathy being more prevalent in CD. The majority of cases with PCS were obese (77%). LDDST had a good diagnostic accuracy (sensitivity 92%, specificity 100%, PPV 100%, NPV 96%). Additional CRH stimulation improved the sensitivity (100%) but had a negative effect on the specificity (95%). In conclusion, Dex-CRH test with a cut-off value of 38 nmol/l is a good test to differentiate between mild forms of CD and PCS.