Endocrine Abstracts

Objective: This study aims to assess the utility of salivary cortisol (SC) in diagnosing adrenal insufficiency (AI) and to analyze the integrity of the hypothalamic-pituitary-adrenal axis. This involves evaluating cortisol rhythm, its correlation with plasma levels, and the impact of exogenous hydrocortisone (HC) substitution on SC determination.

Methods: A prospective study was conducted on all patients with suspected AI undergoing adrenocorticotropic hormone (ACTH) testing (ACTH-t) between 2014 and 2022 at the Endocrinology Department in an University Hospital. Ambulatory salivary cortisol rhythm was determined, and clinical and analytical variables were collected. Diagnostic performance of SC vs basal plasma cortisol (BC) was assessed using the area under the ROC curve (AUC), with ACTH-t results as the ‘gold standard.’ Correlation between SC and BC values was also determined.

Results: Sixty subjects (60.7% women, 42.9% evaluated for pituitary pathology) with a mean age of 50.7±15.1 years were evaluated. Mean SC, BC, and ACTH levels were 0.378±0.242 µg/dl, 11.3±5.2 µg/dl, and 30.5±45.8 pg/dl, respectively. 6.6% were excluded due to preanalytical contamination in SC determination. A moderate positive correlation (r=0.446, P<0.001) was found between SC and BC levels. The diagnostic yield of SC at 0800 hours was superior to BC (AUC=0.804, P<0.01 and AUC=0.739, P<0.05, respectively). CS values <0.0975 µg/dl were diagnostic of AI, while values >0.708 µg/dl ruled out AI, avoiding 21.4% of ACTH-t. However, SC rhythm determination at 13:00, 18:00, and 24:00 did not allow AI diagnosis (AUC=0.202, 0.280, and 0.283, respectively). Determining SC in patients on HC treatment vs those without treatment demonstrated HC overdosage values at 13:00 (1.089±1.120 vs 0.207±0.227 µg/dl, P<0.01), 18:00 (0.752±0.971 vs 0.151±0.130 µg/dl, P<0.01), and 24:00 (1.069±2.531 vs 0.094±0.071 µg/dl, P<0.01), except at 8:00 (0.274±0.205 vs 0.409±0.246 µg/dl, P=0.058). Conclusions: Determination of SC at 0800 hours is a valid alternative for AI screening, showing good correlation with BC. Only 6.6% of the sample inadequately collected SC. In our population, values <0.0975 µg/dl and >0.708 µg/dl for SC confirmed and ruled out AI, respectively, preventing 21.7% of ACTH-t. SC determination at 13:00, 18:00, and 24:00 did not adequately identify patients with AI and showed overdose values in those on HC treatment.