A: Yes, when compared to the amount of unbound cortisol in serum or plasma samples. A number of studies have revealed correlations between the two specimen of r > 0.90. However, the correlation between the total cortisol levels in blood and salivary cortisol is usually weaker due to different amounts of CBG found in blood (e.g., high levels of CBG due to oral contraceptive use results in high total cortisol levels; free cortisol levels are usually unaltered in states of increased CBG).

A: No. Studies looking at salivary cortisol levels obtained under minimal and maximal flow rate clearly showed that the cortisol levels in saliva do not depend on flow rate.

A: The level of salivary cortisol is influenced by drugs such as prednisone, dexamethasone and other steroids administered orally or i.v. While prednisone usually crossreacts with the antiserum used for assaying cortisol (leading to false high values), dexamethasone will significantly suppress the HPA axis (resulting in low cortisol levels). Furthermore, if the subjects deliver saliva samples with a low pH (e.g., following stimulation of saliva flow by crystals of vitamin C, consumption of fruit juices immediately before sampling etc.) most immunoassay will produce false high cortisol levels.

A: No. Studies have shown that salivary cortisol is stable at room temperature for a period of 2 weeks. This means, that even after two weeks left without refrigeration a saliva sample will give the same cortisol level as with immediate determination or with the samples frozen until assay. Note however, that sampling devices like the Salivette may begin to mold within 4-7 days at room temperature. This produces a very bad smelling sample. It does not appear to affect the cortisol levels, but it sure affects the mood of the technicians assaying the samples! So whenever possible, saliva samples should be stored at -20 degrees Celsius or lower.

A: No. There is no refrigeration required for the transport between the subject's / patient's home and the lab (or sending a larger batch of samples from lab to lab). However, if it takes more than 4 days before the samples are delivered to the lab, dry ice will prevent the samples from molding.

A: No.

A: There are different options available today. First, several diagnostic companies now offer commercial assay kits for cortisol determination in saliva. Four of the better assays are sold by

These assays appear to give correct values with acceptable sample volumes required. If you do not have a lab or the time to run the assays yourself, you can also send your samples to a lab which provides assay service. Besides our own lab in Dresden, there are other research labs as well as commercial labs which offer custom assays.

A: One way to check this out is to split a saliva sample into several aliquots and do so with 3-10 samples of (possibly) different cortisol concentrations. Send 3-5 aliquots of each sample to the lab which you want to check (you have to give the different aliquots different I.D.s, of course). Divide the standard deviation by the arithmetic mean (from the different aliquots) for each sample and multiply the result by 100%. This will give you the intraassay coefficient of variation (c.v.) for the different samples. The c.v. should not excess 12-15% (the lower the cortisol concentration, the higher the c.v. usually - rule of thumb).

A: If you decide to purchase a kit and do the assays in your own lab, the price differs widely and depends on the conditions you obtain from the manufacturer. As a rule of thumb, expect to pay 3-5 USD per duplicate analysis (including standards and controls) for the reagents. If you send your samples to a lab which offers assays service, expect to pay between 2.50 - 15 USD for the same analysis.