Salivary cortisol as an alternative for serum cortisol in the low-dose adrenocorticotropic hormone stimulation test?

Salivary cortisol is unaffected by cortisol binding globulin and reflects free serum cortisol as compared to total serum cortisol.

The aim of the present study was to compare the salivary cortisol response with the serum cortisol response in a low-dose (1-microg) ACTH test in a clinical setting and to determine the optimal cut-off value of salivary cortisol as an alternative to serum cortisol.

We measured serum and salivary cortisol responses to iv administration of 1-microg ACTH in 51 patients (17 males) referred to the Department of Clinical Chemistry for ACTH-testing. Serum cortisol was assessed before, 20, and 30 min after ACTH-administration, and salivary cortisol was assessed before and 30 min after ACTH administration.

Mean cortisol at baseline, 20, and 30 min were 0.44 micromol/l (SD: 0.22), 0.64 micromol/l (SD: 0.24), and 0.70 micromol/l (SD: 0.25), respectively. Median basal salivary cortisol was 8.4 nmol/l [interquartile range (IQR): 3.8-14.2]. Salivary cortisol at 30 min equaled 35.9 nmol/l (IQR: 21.1-46.2). Basal salivary cortisol was significantly correlated with salivary cortisol at 30 min (r=0.53; p<0.001). Salivary cortisol at 30 min of 23.5 nmol/l had a sensitivity and specificity of 78.1% and 70.0%, respectively as compared to the serum cortisol cut-off values of >0.50 micromol/l.

The salivary low-dose ACTH-test yields more dynamic responses than serum cortisol. However, the sensitivity and specificity of salivary cortisol are too low to be adequate as an alternative to the serum cortisol measurements. In women on estrogen therapy, however, the use of salivary cortisol might be superior to serum cortisol.