ECE2019 Guided Posters Cushing's (12 abstracts)
University Hospital of Bordeaux, Bordeaux, France.
Background: According to the Endocrine Societys guidelines, the screening and diagnosis of Cushings syndrome (CS) are based on the results of the overnight dexamethasone suppression test (DST), midnight salivary cortisol and 24-hour urinary free cortisol (UFC) measurements. These 3 tests reflect cortisol levels at the time of sampling without providing retrospective information. Hair cortisol concentration is a non-invasive way to measure cortisol exposure over longer periods of time (weeks and months) but only few studies report its usefulness for the diagnosis of CS. Most of studies were performed with cortisol immunoassays.
Aim: To assess the diagnostic power of the measurements of cortisol and cortisone in hair using liquid chromatography/mass spectrometry method. We provide reference and pathologic values in several populations: no-obese healthy subjects (H-group) and patients with Cs (Cs-Group).
Methods: After a 2-step extraction (incubation in a dithiothreitol solution followed by a dichloromethane extraction), concentrations of cortisol (HairF) and cortisone (HairE) were measured in the proximal 3 cm of scalp hair by using a home-developed multiplex mass spectrometry assay. Hairs were cut from the posterior vertex of 112 volunteers admitted in the Endocrinology Department of the University Hospital of Bordeaux: 76 (62F/14M, 38 [2183] yrs, median [min-max]) in H-group (44 in a H<50 subgroup were <50 year, 32 in a H>50 subgroup were >50yr), 36 (29F/7M, 53 [1978] year) in the Cs-group. Based on clinical signs and biological markers (UFC, DST, serum or salivary cortisol at midnight), we included in the Cs-group patients either with mild and patent Cs.
Results: HairF and HairE reference values were similar in H<50 and H>50 subgroups. As a whole, the H-group reference values were 9.6 [45.0] (median [95th percentile]) and 14.5 [27.64] pg/mg hair, HairF and HairE respectively. In the Cs-group, both HairF and HairE were increased compared to the H-group: 24.9 [2.0 738.8] (median [min - max]) and 37.5 [7.0 192.4] pg/mg hair (P<0.001 for HairF and HairE). HairE ROC curve was significantly larger than HairF curve (areas 0.871 vs 0.733) to distinguish H-group from Cs-group (P<0.001).
Conclusion: HairF and HairE assays are useful for the diagnosis of Cs. Importantly, HairE seems to be a better reflect the excess of a long-term cortisol exposure than HairF. The determination of HairF and HairE using mass spectrometry assay may improve the diagnostic accuracy of this specific investigation.