Endocrine Abstracts (2017) 49 EP13 | DOI: 10.1530/endoabs.49.EP13

Hair cortisol in patients with primary aldosteronism

Aaron Hodes1,3, Maya Lodish1, Amit Tirosh1,3, Jerrold Meyer5, Elena Belyavskaya1, Charalampos Lyssikatos1, Kendra Rosenberg5, Andrew Demidowich1, Jeremy Swan1, Nichole Jonas1, Constantine Stratakis1 & Mihail Zilbermint1,2


1Section on Endocrinology and Genetics, Eunice Kennedy Shriver National Institute of Child Health and Human Development, National Institutes of Health, Bethesda, Maryland, USA; 2Johns Hopkins University School of Medicine, Division of Endocrinology, Diabetes, and Metabolism, Baltimore, Maryland, USA; 3Sackler Faculty of Medicine, Tel Aviv University, Tel Aviv, Israel; 4Albert Einstein College of Medicine, Department of Radiology, Jacobi Medical Center, Bronx, New York, USA; 5Department of Psychological and Brain Sciences, University of Massachusetts, Amherst, Massachusetts, USA.


Context: Primary aldosteronism (PA) is a common cause of secondary hypertension, and some PA adrenal tumors co-secrete glucocorticoids, causing subclinical or overt Cushing syndrome (CS). We recently reported correlations between hair cortisol concentration (HCC) and serum and urinary levels of cortisol in patients with CS.

Objective: To determine correlations of segmental hair cortisol and biochemical markers of a referred cohort of patients for the evaluation of PA and to compare to patients with normal cortisol secretion. This pilot study was conducted at the National Institutes of Health Clinical Center.

Methods: Hair samples were collected from 12 study subjects (four with PA, two with PA and cortisol co-secretion, and six controls), with mean age 40.1±21.4 years, mean BMI 28.1±5.3 kg/m2. Diurnal serum cortisol and ACTH measurements, 24-h-urinary free cortisol corrected by body surface area (UFC/BSA) and 17-hydroxysteroids, corrected for creatinine (17OHS/Cr) were measured. Patients underwent step-wise diagnosis, with measurement of serum aldosterone and plasma renin activity followed by saline suppression and/or oral salt loading tests. Patients without PA or CS were excluded as controls, if midnight serum cortisol >1.8 μg/dl. Segmental hair samples from each patient were processed and analysed for cortisol according to the methods described by Meyer et al. (2014).

Results: Age, blood pressure, urinary and midnight cortisol levels were higher (P<0.05) in the PA group compared to the controls. Average hair cortisol values in the groups with PA and controls were 30.0±31.0 pg cortisol/mg hair (median: 16.3 pg/mg; interquartile range: 12.6–37.4 pg/mg) and 22.9±27.6 pg cortisol/mg hair (median: 11.3 pg/mg; interquartile range 7.9–26.6 pg/mg), respectively (P>0.05). No correlation between hair cortisol and serum and urinary markers of hypercortisolemia was seen in the six patients with PA.

Discussion: We found that no statistically significant difference in hair cortisol between groups and no significant correlation of hair cortisol with urinary or serum cortisol evaluations. We speculate that the lack of correlation may be due to insufficient power. More research is needed on the use of hair cortisol in patients with PA.