Establishment of reference intervals for the salivary cortisol circadian cycle, by electrochemiluminescence (ECLIA), in healthy adults.

OBJECTIVES

To determine salivary cortisol reference intervals in a healthy adult population, at 6 different time points during a 24-hour (h) period.

METHODS

In a prospective study, salivary cortisol concentrations were measured upon waking, one-hour post-waking and at specific times of the day: at 12 h00, 16 h00, 20 h00 and midnight. Samples were analyzed by the first and second-generation electrochemiluminescence assays (ECLIA) from Roche Cobas Cortisol®.

RESULTS

Salivary cortisol values were obtained from 134 healthy volunteers. Reference intervals for the first-generation assay were 6.14-33.19 nmol/L (95% prediction interval) at waking, 5.42-28.06 nmol/L one-hour post-waking, 3.62-16.23 nmol/L at 12 h00, 2.78-15.27 nmol/L at 16 h00, 2.08-14.90 nmol/L at 20 h00 and 2.09-16.92 nmol/L at midnight. Mean salivary cortisol values were 14.63 nmol/L at waking and 6.44 nmol/L at midnight. Reference intervals for the second-generation assay were 1.50-22.02 nmol/L (2.5th to 97.5th percentiles) at waking, 1.50-20.87 nmol/L one-hour post-waking, 1.50-12.51 nmol/L at 12 h00, 1.50-13.03 nmol/L at 16 h00, 1.50-9.52 nmol/L at 20 h00 and 1.50-6.28 nmol/L at midnight. Values for the second-generation assay at all 6 different time points were almost half of the first-generation assay. The second-generation assay showed a better correlation with LC-MS/MS (r = 0,97).

CONCLUSION

Our study confirms that reference intervals for salivary cortisol are not comparable across first and second-generation Roche Cobas Cortisol® assays. Furthermore, the second-generation assay has a better correlation with LC-MS/MS and a better analytical performance (accuracy and precision).