Endocrine Abstracts (2008) 16 P29

Late night salivary cortisol in the diagnosis of Cushing's syndrome

Jeyaraman Kanakamani, Gupta Nandita & AC Ammini


All India Institute of Mediacl Sciences, New Delhi, India.


Introduction: Late night salivary cortisol appears to be promising as a simple stress-free screening test for Cushing’s syndrome (CS). But there is no consensus on the cut-off value because of the lack of availability larger studies and of standardization of assays. The normal reference ranges are assay-dependent and should be validated for each laboratory.

Objective: The purpose of this study was to evaluate the usefulness of late-night salivary cortisol in the diagnosis of CS.

Methods: We studied 49 normal subjects and 52 patients with suspected hypercortisolism. Salivary cortisol was collected at 0800 h and 2300 h using a commercially-available salivette and measured by an automated chemiluminescence assay. The study was approved by the Institute’s ethical committee.

Results: Out of 52 patients 20 were confirmed to have CS and in the remaining CS was excluded by dexamethasone suppression test. The 2300-h salivary cortisol of patients with CS was significantly higher (1.24±0.86 μg/dl), as compared with patients in whom CS was excluded (0.109±0.076 μg/dl). The 0800-h cortisol was also significantly higher in CS (1.70±1.40 vs 0.240±0.132 μg/dl). The loss of circadian rhythm highlighted by 0800-h to 2300-h cortisol of < 2 was found in 17 of 20 CS patients. The upper limit of the reference range for 2300 h salivary cortisol was calculated nonparametrically as 0.311 μg/dl, from the data on normal subjects. Two of 20 patients with proven CS had 2300-h salivary cortisol less than the calculated upper limit of the reference range, yielding a sensitivity of 90%; one of these 2 patients had intermittent hypercortisolism.

Conclusion: Late-night salivary cortisol measurement is a simple and reliable screening test for Cushing’s syndrome. A cut-off value of 0.311 μg/dl gives a sensitivity of 90% and specificity of 96.8%. Larger studies and wider availability of validated commercial assays are needed before it could be used as first line screening test.

Article tools

My recent searches

No recent searches.