Raff · The Journal of clinical endocrinology and metabolism 1998 · diagnostic cross-sectional study · n=151

Late-night salivary cortisol as a screening test for Cushing's syndrome.

Cited 312 times in the scientific literature.

Level 3 - non-randomized controlled study

Diagnostic cross-sectional study comparing index test to reference standard in referred patients and healthy controls.

PubMed 9709931 · doi:10.1210/jcem.83.8.4936 · record verified 2026-08-31

What was done

The study evaluated 2300 h and 0700 h salivary cortisol measurements for diagnosing Cushing's syndrome using a commercial collection device and modified radioimmunoassay. The cohort included 73 normal subjects and 78 referred patients (39 with confirmed Cushing's syndrome and 39 in whom Cushing's was excluded or not firmly established). Results were evaluated against reference thresholds and 24-hour urine free cortisol measured within one month.

What was found

Patients with confirmed Cushing's syndrome had significantly elevated 2300 h salivary cortisol (24.0 +/- 4.5 nmol/L) compared to normal subjects (1.2 +/- 0.1 nmol/L) and referred non-Cushing's patients (1.6 +/- 0.2 nmol/L). At an upper reference limit of 3.6 nmol/L, 2300 h salivary cortisol demonstrated 92% sensitivity (36 of 39 detected). Combining an elevated 2300 h salivary cortisol or elevated urine free cortisol detected all 39 cases (100% sensitivity). Morning (0700 h) salivary cortisol showed substantial overlap between groups (23.0 +/- 4.2 nmol/L in Cushing's vs 14.5 +/- 0.8 nmol/L in normals and 15.3 +/- 1.5 nmol/L in non-Cushing's referrals).

Why it matters

Late-night salivary cortisol sampling offers a convenient, non-invasive screening tool to identify circadian rhythm disruption in patients evaluated for hypercortisolism.

Limits

The sample size of confirmed Cushing's cases was small (n = 39). The non-Cushing's referral group included patients with unconfirmed rather than strictly ruled-out diagnoses. Specificity values and confidence intervals were not reported in the abstract.

Cited by