SUN-474 Time-Point Intra-Individual Variability in Salivary Cortisol and Cortisone: A Cross-Sectional Study

Abstract Disclosure: S. Ahmadian: None. C.A. Villavicencio Torres: None. V.M. Fell: None. K.L. Moehnke: None. Y. Chai: None. R.L. Taylor: None. S.K. Grebe: None. A.D. Maus: None. I. Bancos: None. Background: Salivary cortisol (SalF), and cortisone (SalE) interpretation can be influenced by intra-individual variability arising from cortisol secretion cyclicity, assay performance, and sampling techniques. No studies to date have examined the impact of sampling on intra-individual variability. Methods: We conducted a single-center cross-sectional study of adults evaluated with total serum cortisol measurements as a part of the overnight dexamethasone suppression test (DST) or cosyntropin stimulation test (CST), November 2024- January 2025. Participants were provided with two cotton swabs (salivettes) and asked to collect two consecutive saliva samples immediately after the venipuncture for total cortisol. Each cotton swab was kept in the mouth for 2-2.5 minutes. Saliva was tested for SalF and SalE by mass spectrometry. Variability was examined based on cortisol tertiles. Results: A total of 78 paired saliva samples were obtained from 60 participants (45 [75%] women; median age 55 years [IQR 39-66]), who were undergoing either DST (42 [70%]) or CST (18 [30%]). A total of 10 (17%) saliva samples did not contain a sufficient sample for analysis during the first or second saliva collection (7 [12%] from DST group, and 3 [5%] from CST group).Among 53 patients undergoing DST, the median serum total cortisol was 1.3 mcg/dL [IQR 0.9-3], the median first SalF was 20 ng/dL [IQR 7-28], the median second SalF was 12 ng/dL [IQR 6-23], and the median variability was 11% [IQR 3-22%]. The median first SalE was 69 ng/dL [IQR 54-185], while the median second SalE was 63 ng/dL [IQR 56-178], with a median variability of 6% [IQR 4-10].In patients undergoing CST, peak median serum total cortisol was 20 mcg/dL [IQR 18-24], median first SalF was 809 ng/dL [IQR 536-1227], and median second SalF was 1063 ng/dL [IQR 667-1503], with the median variability of 14% [IQR 8-22]. The median first SalE was 2507 ng/dL [IQR1697-2966], while the median second SalE was 2485 ng/dL [IQR 1530-2976], with a median variability of 9% [IQR 3-12].When examining intra-individual variability of SalF, the variability was lowest in the middle tertile, corresponding to serum cortisol of 1.8-8.9 mcg/dL, at 9% [IQR 5-16]. Variability was higher in the lowest tertile (14% [IQR 0-52]) and the highest tertile (12% [IQR 6-16]). Overall, the variability of SalE was lower than that of SalF: 7% [IQR 4-13] in the lowest tertile, 5% [IQR 2-9] in the middle tertile, and 10% [IQR 4-12] in the highest tertile. Conclusion: We show that saliva sampling may impact the results, with 17% patients providing insufficient sample for analysis, and variability of 9-14% for SalF and 5-10% for SalE. Same time-point sampling-related variability should be considered in interpretation of SalF and SalE measurements. Presentation: Sunday, July 13, 2025