SUN-305 Hair Cortisol Measurement: An Innovative Method for Diagnosis and Follow-Up in Patients with Cushing’s Disease

Abstract

Diagnosis of endogenous Cushing’s syndrome entails corticotropic autonomy, lack of circadian rhythm and/or hypercortisolism, evaluated through 24h urinary free cortisol (UFC). Hair cortisol measurement (HCM) has been described as an alternative marker of cortisol exposure over the preceding three months.OBJECTIVES To evaluate HCM in Cushing’s disease (CD). To analyze the correlation between HCM and UFC. To compare HCM values in CD vs controls.PATIENTS AND METHODS 3 cm hair from posterior vertex in CD and in controls age- and gender-matched between May 2017 and May 2019. Controls were low level stressed individuals (Holmes-Rahe’s scale) without adrenal disease. Normal reference interval of HCM was defined (40-128 pg/mg hair). Measurement: Siemens Immulite 2000 (Gwynedd, UK) automated chemoluminiscent immunoassay (CLIA)UFC values within the 3 months previous to hair collection were considered. Controlled CD defined as UFC ≤1 upper normal limit (UNL) with or without treatment, remission as UFC ≤1 without pituitary lesion. Results are presented as median (m) and range. Kruskal-Wallis ANOVA used for median difference evaluation and Kappa index for concordance determination. Chi2 test for comparison of recategorized UFC and HCM. Statistical analysis performed with SPSS 23.0RESULTS23 CD patients recruited, median age 42 ± 11 years; 91% (n=21) female; 10 samples collected at diagnosis and 13 during follow-up. Control group composed of 50 individuals45% (n=10) had controlled CD (mUFC 0.42 UNL, range 0.1-0.9) and a mHCM of 134.5 pg/mg (62-334) and 55% (n=12) did not have control (mUFC 2.2, 1.1-6) and a mHCM of 150.5 (75-459). After recategorization of UFC (> o ≤ 1 UNL) and HCM (> o ≤ 128 pg/mg), determinations were associated (Chi2, p= 0.18), however, the concordance was acceptable (Kappa index = 0.276).After dividing CD patients according to HCM, 35% (n=8) had normal HCM: mHCM 113.5 (62-126) and mUFC 0.45 (0.1- 4.4). Among them, 63% (n=5) had controlled CD (mHCM 110, 62-121; mUFC 0.39, 0.1-0.85); 25% (n=2) had active CD (mUFC 2.7, 1.1-4.4; mHCM 121, 75-126). 65% had high HCM (n=15): mHCM 167 (132-459) and mUFC 1.36 (0.1-6). Most of them had active CD (n=11, 73%): mHCM 160 (132-459) and mUFC 2.2 (1.1-6). Four patients with elevated HCM (m 248, 148-334) had normal UFC (m 0.61, 0.12-0.92): 2 were in remission, 1 had normal postsurgical UFC with active disease in the follow-up and 1 had normal UFC under medical treatment.Controls (n=50) had mHCM 62.5 (40-126), significantly different from CD.CONCLUSIONSWe evaluated HCM in CD, proposing this method as an additional diagnostic test for hypercortisolism. The acceptable concordance between UFC and HCM is possibly due to the different duration of the evaluated periods.The difference in the HCM values observed between controlled or active CD patients and controls permits the consideration of the method as an alternative in the diagnosis and/or follow-up of CD.