Endocrine Abstracts

The diagnosis of subclinical hypercortisolism (SH) is a challenging issue, especially as the high prevalence of adrenal incidentalomas (AI) is increasingly recognized. The diagnosis of subclinical hypercortisolism relies on a combination of tests that include overnight (ODST) or low dose dexamethasone suppression (LDDST), plasma ACTH, and 24-h urinary free cortisol (UFC). Late night salivary cortisol (LNSC) has been successfully used in the diagnosis of Cushing’s syndrome (CS) and is now considered a first-line screening test. Nevertheless, its use in SH has been evaluated only in a limited number of reports. Therefore, we tested whether measurement of LNSC can be used to screen patients with AI for SH.

Thirty five patients (21 females) were studied; six patients had CS (group A); and 29 patients had AI. Based on published diagnostic criteria for SH (post-LDDST cortisol levels >1.8 μg/dl and at least one of the following: ACTH levels <10 pg/ml and high UFC >120 μg/24 h, corresponding to the upper normal limit of our assay patients with AI were divided in two groups: patients with SH (group B, n=10) and patients without SH (group C, n=19). Results were compared to normal controls (group D, n=14). LNSC was measured at 23.00 on the day of the study.

LNSC levels (μg/dl) were higher (P<0.05) in groups A (0.79±0.36) and B (0.21±0.06) compared to groups C (0.06±0.01) and D (0.05±0.01). ROC curve analysis revealed that SH can be diagnosed with the sole measurement of late-night salivary cortisol with 100% sensitivity but 50% specificity, with a cut-off of 0.045 μg/dl (AUC=0.9 (95% CI: 0.79–1.03, P=0.0008). A higher cut-off of 0.055 μg/dl diagnosed SH with 90% sensitivity and 71.5% specificity.

These data illustrate that late-night salivary cortisol is of value in the diagnosis of SH in patients with AI. Taking into consideration the ease of sampling, we suggest that salivary cortisol may be used as an additional screening test in this patient group.