About Saliva Testing

Saliva as sample type offers several advantages when compared to blood, CSF and urine tests. Collection of saliva samples is minimally invasive, safe, quick, pain-free, and can be performed with minimal training. This makes saliva testing ideal for psychology and sports science studies, and particularly suitable for ambulatory studies and when sampling from vulnerable groups such as infants, children, the elderly and animals.

Biomarkers available

Aldosterone
Alpha-Amylase
Androstenedione
C-Reactive Protein
Cortisol
DHEA
DHEA-S
Estradiol
Estriol
Estrone
IL-6
IL-1β
Melatonin
Progesterone
17a-OH-progesterone
Secretory Immunoglobulin A (SIgA)
Testosterone
Transferrin/ Blood Contamination
Uric Acid

Time points

For certain salivary analytes, including cortisol, timing is everything. There are two aspects of this that are separate, but both are important:

Biorhythms – many analytes follow a diurnal rhythm, and others, most notably the female sex hormones, can also vary over a longer period. It is often important that your sampling scheme takes these variations into consideration. Some analyses, such as the Cortisol Awakening Response and Area Under the Curve measures, are even based on the diurnal pattern.

Response times – hormone signalling tends to be a fairly slow process, giving a slow, but long-lasting response when compared to nerve impulses. If performing a same-day treatment response study, such as an acute stress study, the lag time between the treatment administered and expected peak salivary analyte levels in response to treatment should be considered for the analyte that you are interested in. This is different for different analytes and reflects the time taken for their secretion as well as the method by which they enter saliva.

Cortisol time points

For more information on cortisol as an analyte, as well as the timing of your cortisol sampling, both in relation to the time of stimulus and the time of day, the below references should be helpful:

Nicolson, N.A. (2008) Measurement of Cortisol. Handbook of Physiological Research Methods in Health Psychology, p.37.

Kirschbaum, C. & Hellhammer, D.H. (2000). Salivary cortisol. Encyclopedia of stress, 3(379–383).

Dickerson, S.S. & Kemeny, M.E. (2004). Acute stressors and cortisol responses: a theoretical integration and synthesis of laboratory research. Psychological bulletin, 130(3), p.355. (PMID: 15122924).

Correlations

Saliva is a suitable medium for the measurement of many analytes, and there are even a few analytes for which saliva is the best available medium, such as alpha-amylase and secretory IgA. However, not all salivary measures are the equivalent of their serum counterparts, and they cannot always be used to give the same information that studying the same analytes in serum would provide.

Examples of this are IL-1ß and IL-6. Salivary concentrations of IL-1ß and IL-6 follow a different diurnal pattern to serum levels, and the serum and saliva levels of these cytokines do not generally correlate in otherwise healthy individuals. C-Reactive Protein shows better correlation with serum in a number of studies, and you may instead want to consider whether this marker of inflammation would be suitable for your project.

Salimetrics currently believe salivary IL-6 and IL-1B levels to be largely independent of serum levels and to most often represent the degree of inflammation in the mouth, making these kits most suitable for dental research. The concentration range measured by Salimetrics IL-6 kit is also significantly different to that of other kits, so if you are thinking of testing for IL-6 please contact us to discuss your plans.

Similarly, most Salimetrics kits cover the majority of the expected physiological range of analyte concentration in their intended subjects, but not all kits will give you reliable results for all subjects or all sampling timepoints or testing conditions. This can depend strongly on the age or sex of your subjects as well as the time of day at which you sample.

For example, kits for any of the sex hormones would not be expected to give quantifiable results for a large proportion of children, and kits for the female sex hormones will also produce a majority of readings that are undetectable/ not reliably quantifiable in adult men.