Question 1

Is saliva testing effective for sublingually administered hormones?

Saliva testing is a terrific tool for monitoring endogenously produced hormones as well as for many supplementation scenarios. Using saliva testing is, however, not a good idea for sublingually administered hormones. Keep in mind that if a patient supplements with 1mg of testosterone sublingually, it is 1 billion picograms (pg) of testosterone. The reference range for testosterone in a female is 20-50pg/ml in saliva, so there is great potential for contamination. Most of the supplement gets into the body and some filters back into the saliva; however, if just one hundredth of one percent stays in the mouth and contaminates the oral mucosa, the saliva will be falsely elevated by a whopping 1 million pg.

If a physician decides to test a patient's saliva after sublingual administration of a hormone, ZRT can give an accurate value. This value is usually not of clinical relevance and the value obtained by way of blood spot testing would be much more clinically relevant. ZRT recommends blood spot testing for hormones administered sublingually.

Additionally, if any hormone is taken sublingually, the patient should wait at least 36 hours before testing ANY hormones in saliva. If hormone levels of, for example, progesterone are over 1 million (which they likely will be if the patient does not wait 36 hours), the levels can be so high that the other hormone tests can be falsely elevated due to the cross-reactivity from the sublingual hormone.

ZRT's saliva/blood spot combination kits are ideally suited for individuals taking hormones sublingually and by other routes of administration.

Question 2

This line of questioning is quite complicated and easily confused. Here are a few background facts to guide our thinking:

• Without supplementation, studies have shown great correlation between salivary and serum values if the saliva measurements are accurate.
• For hormones like progesterone, a physiologic dose (15-30mg per day) of topical hormone gives salivary values that are greatly elevated above the physiologic range, while serum values often remain unchanged. Some have interpreted this large increase in progesterone, following topical supplementation, as a bioaccumulation of this fat soluble steroid. If this was accurate, we would not expect a first time user to show these greatly elevated values. Observe the following study performed  on a male at ZRT who had never supplemented with progesterone. 15mg of progesterone was administered to the lower back, and serum was taken from both vein and finger stick every two hours while saliva was taken every hour. Saliva levels increased from less than 50 to more than 3,000pg/ml in six hours. The blood taken from the vein (on its way back to the heart) shows no increase in progesterone while the blood taken from capillary blood (finger stick) shows some increase in progesterone. It would seem that serum is grossly underestimating the amount of progesterone getting into the patient.

• When Australian researchers (Waddell and O'Leary, 2002) applied topical progesterone to rats, the saliva gland better represented the amount of progesterone in organs such as the uterus than did the plasma, which showed more progesterone metabolites than progesterone.
• In a 1995 study, Chang found that breast tissue levels of progesterone went up 100-fold following administration of 20mg of topical progesterone.
• In an independent study, saliva levels were shown to analogously rise about 100-fold, while serum levels less than 2-fold. Saliva levels may reflect the systemic absorption of progesterone much better than serum, which appears to be a dangerous underestimation.

• Topical supplementation of progesterone seems to result in little change in serum results. On the other hand, ZRT's database has shown that with increasing dosages of topical progesterone, reasonably linear, dose-dependent results are seen.

Given the available information, it seems reasonable to conclude that testing serum following topically administered hormones seems potentially misleading. Saliva results follow an expected trend, but the non-supplementing reference ranges seem to be likewise misleading. The best solution from the information available is to reset the reference ranges for various supplementation scenarios using the vast information from the ZRT database. The following chart can be effectively used following progesterone supplementation. If, for example, the patient is supplementing with 20mg of topical progesterone and collects a morning saliva sample 24 hours later, the average expected value (50th % or median) is approximately 450pg/ml. This information can be used to gauge the patient's status relative to what is expected.

Question 3

While this would be very simple to do if it was advantageous, there are two very good reasons why this approach is detrimental to the clinical utility of salivary hormone testing.

Reason #1: There is a circadian rhythm for testosterone and DHEA.

Even labs that pool their saliva samples agree that this is inappropriate for hormones with a circadian rhythm (cortisol has a classic circadian rhythm dropping more than 80% throughout the day). The debate as to whether or not hormones, like testosterone, do have a circadian rhythm, actually isn’t much of a debate. Dozens of studies can be sited from serum or saliva testing showing a significant circadian rhythm for both testosterone and DHEA. Labs choosing to pool samples are ignoring the research on this topic. Observe the following quote from a salivary testosterone study from 1980. “In normal male saliva samples, morning concentrations were significantly higher than evening samples. The circadian rhythm was confirmed by COSINOR analysis.” Many such studies can be seen in the literature as proof that such hormone levels drop throughout the day by about 40-50%. That is not insignificant.

Reason #2: Introduced variation from supplementation.

When an individual takes hormonal supplementation in the morning, hormone levels will be the most variable around the time of the noon collection (just a few hours after administration). For this very reason, we measure only cortisol from this noon sample. If this sample was pooled for analysis, the reliability of the results would be compromised. We don’t want to shoot at a moving target, and trying to measure a hormone at noon that was taken in the morning is just that – a moving target. This is not true of every hormone supplementation scenario. A transdermal patch, for example, is a continuous slow release and pooling the samples would not be a problem. However, for oral, topical, and other routes of administration this is a significant problem. Levels can change dramatically within an hour, therefore if testing is done in the first few hours, this change makes the results too variable for clinical utility.

It is tempting for a laboratory to pool samples throughout the days, as it allows the lab to require less saliva in each collection. This may make the saliva collecting experience less daunting, but it comes at a high price – results are simply less reliable and supplementation reference ranges (if they are even provided) will become much less useful due to the higher variation in the results.

The only benefit to pooling saliva samples throughout the day is that it creates a better “average” of hormone levels as they are prone to sudden, slight variations. Collecting four small samples throughout the day has, however, not been proven to be any better than simply spending a little more time collecting a larger volume (thus a longer amount of time for an “average”) in the morning sample. We advise patients to collect as much saliva as is reasonable for the morning sample, although we need only about 1/3 of the tube for analysis.