Radioimmunoassay for "free" testosterone in human saliva

Measurement of testosterone in human sexuality research: methodological considerations

Testosterone (T) and other androgens are incorporated into an increasingly wide array of human sexuality research, but there are a number of issues that can affect or confound research outcomes. This review addresses various methodological issues relevant to research design in human studies with T; unaddressed, these issues may introduce unwanted noise, error, or conceptual barriers to interpreting results. Topics covered are (1) social and demographic factors (gender and sex; sexual orientations and sexual diversity; social/familial connections and processes; social location variables), (2) biological rhythms (diurnal variation; seasonality; menstrual cycles; aging and menopause), (3) sample collection, handling, and storage (saliva vs. blood; sialogogues, saliva, and tubes; sampling frequency, timing, and context; shipping samples), (4) health, medical issues, and the body (hormonal contraceptives; medications and nicotine; health conditions and stress; body composition, weight, and exercise), and (5) incorporating multiple hormones. Detailing a comprehensive set of important issues and relevant empirical evidence, this review provides a starting point for best practices in human sexuality research with T and other androgens that may be especially useful for those new to hormone research.

Validation of salivary testosterone as a screening test for male hypogonadism

BACKGROUND

Saliva collection is an easy, non-invasive method to measure hormones.

METHODS

Two studies were performed. In the first, a convenience sample of 1,454 males who had submitted saliva for salivary testosterone measurements were studied. In the second study, we intensively studied symptoms and measurements of total testosterone, free testosterone symptoms and measurements of total testosterone, free testosterone and bioavailable testosterone in relationship to salivary testosterone in 127 men. A secondary endpoint was to examine the relationship of salivary testosterone to hypogonadal symptoms in the ADAM and AMS questionnaires.

RESULTS

In the first study, we have shown that salivary testosterone, measured in 1,454 males aged 20 to 89 years, declines by 47% over the lifespan. In the second study, salivary testosterone was strongly correlated with bioavailable testosterone (p < 0.000001) calculated free testosterone (p < 0.00001) and total testosterone (p < 0.002). Salivary testosterone was significantly related to hypogonadal symptoms on the St. Louis University ADAM questionnaire and the Aging Male Survey.

CONCLUSIONS

These studies support the use of salivary testosterone as an acceptable assay for screening for hypogonadism. Salivary testosterone is not a better assay than other measures to diagnose hypogonadism.

Salivary testosterone in children with and without learning disabilities

Previous studies have indicated that the sex steroids have organizational effects upon neural tissue and that abnormal secretion during development may lead to functional anomalies. In this study, we explore the possibility of prepubertal steroid hormone involvement in the etiology of learning disabilities. Salivary testosterone levels in 264 children without learning disabilities (133 males, 131 females) were measured and compared to that in 32 children with learning disabilities (25 males, 7 females). The presence of learning disabilities was significantly associated with higher salivary testosterone. Data from equivalent samples of learning-disabled and control subjects also were compared separately because of disparities in sample size and variable distribution in the total group analysis. A 32-member sample of nonlearning-disabled children was created by randomly selecting individuals who exactly matched the age, race, and sex characteristics of the learning-disabled group. The matched analysis further substantiated the association between testosterone secretion and learning disabilities. Thus, it is possible that some learning disabilities may be associated in part with abnormal testosterone levels.

Metabolism of progesterone and testosterone in human parotid and submandibular salivary glands in vitro

Twenty-three parotid and twelve submandibular human salivary glands were used for analyzing in vitro metabolism of progesterone (P) and/or testosterone (T), in order to find out their possible contribution to the concentrations of salivary P or T. Glands were excised from 21 women and 14 men because of a stone, tumor or sialoadenitis, and the healthy tissue was used for incubation. The tissues were homogenized and incubated with 5 to 7 nmol/l of [14C]P or [14C]T for 2 h, extracted with methylacetate and subjected to two dimensional thin-layer chromatography. The thin-layer plates were autoradiographed, and the radioactivity of the different metabolites were counted to assess the amounts of metabolites formed. No lipoidal or conjugated steroids were formed. All of the radioactivity was associated with the free fraction of steroids. Both P and T were metabolized significantly more actively in male submandibular gland compared to male parotid gland (P < 0.05-0.01). No significant differences were found between female and male parotid glands, nor between the parotid glands of hormonally medicated (oral contraceptives or postmenopausal estrogen treatment) and non-medicated women. The submandibular glands more actively metabolized the steroids studied than the parotids (P < 0.01). The main metabolites were 20 alpha-hydroxy-4-pregnen-3-one for P, and androstenedione for T. In conclusion, the present results bring further evidence that the P or T levels in saliva may not be identical with the unbound steroid fraction in circulation, but a part of the steroids are metabolized in salivary glands. However, due to the rapid passage of steroids from blood to saliva, the metabolism demonstrated probably does not form an important source of error in salivary P and T measurements.

Salivary free testosterone in hirsutism

A sensitive, specific and accurate direct radioimmunoassay of testosterone in human saliva is described. A single salivary testosterone result is shown to be of greater diagnostic use in hirsutism than any of the currently used serum androgen assays. Thus, of 50 hirsute patients, salivary testosterone (Sa-T) was elevated in 34 patients, sex hormone-binding globulin (SHBG) was decreased in 30 women, serum testosterone (S-T) elevated in 13, dehydroepiandrosterone sulphate (DHEA-S) was elevated in 14, and androstenedione in three of the investigated group.

Salivary testosterone levels and the progress of puberty in the normal boy

Salivary testosterone (ST) levels were measured in 84 boys aged 7.3-16.2 from the Edinburgh Growth Study. The correlation coefficient between matched plasma/saliva samples was 0.88. Six samples were collected over the course of one day from 0900 to 2100 h each month in the majority of the children for 4 consecutive months. Mean daily ST levels showed a significant rise between each pubertal stage (genital (G) and pubic hair (PH]. The rise in ST became more rapid once a mean testicular volume (MTV) of 10 ml had been reached. The diurnal rhythm was assessed by individual curve fitting on the log scale and by cosinor analysis. A rhythm was present prepubertally and developed into a pattern similar to that of the adult rhythm by stage G3. The monthly rate of rise of ST was greatest at stage G4. A significant rise in ST levels was detectable immediately prior to an increase in MTV to 3 ml. This allowed earlier recognition of the clinical onset of puberty at testicular volume of 3 ml, which in this group occurred at 10.9 (SD 0.9) years. ST is a non-invasive and sensitive method for the serial monitoring of gonadal function in the prepubertal and adolescent boy.

Treatment of complicated benign prostatic hyperplasia with LHRH-analogues in aged patients

The effect of administration of an LHRH-analogue (LHRH-a) was evaluated in 11 patients with benign prostatic hyperplasia (BPH) in whom there were contraindications for surgery. These patients, who already had impaired potency due to age or serious illness, were given 1500 micrograms LHRH-a in the first week and 1200 micrograms from the eighth day onwards. They all improved significantly (P less than 0.001) with regard to their urological symptoms on day 60 of treatment. Our results demonstrate the influence of androgen in maintaining established BPH, the effectiveness of LHRH-a therapy in selected patients with BPH and the usefulness of a salivary testosterone assay for the follow-up of treatments expected to lower the bioavailability of androgen to target tissues.

Direct assay for testosterone in saliva: relationship with a direct serum free testosterone assay

A direct non-extraction radioimmunoassay for salivary testosterone is described using a modified commercial kit procedure that is in use for total serum testosterone (T). Serum free testosterone was also measured by direct radioimmunoassay. A significant correlation (r = 0.83, p less than 0.01, n = 194) was obtained between salivary and serum free testosterone in matched serum and saliva samples over a wide range of concentrations. Within- and between-batch precision for the salivary testosterone method was 11% and 18%, respectively at a concentration of 170 pmol/l. Recovery of added T was 89% +/- 15% (mean +/- 2 SD) dilution of high samples showed parallelism. Salivary testosterone measured by direct radioimmunoassay offers a simple cheaper alternative to serum free testosterone measurement with the additional advantages of a stress-free non-invasive sampling procedure.

Saliva testosterone time-course response to hCG in adult normal men. Comparison with plasma levels

Testosterone time-course response to 5000 IU hCG was studied simultaneously in the saliva and the plasma of 13 adult normal men. Baseline levels in saliva and plasma were: 93 +/- 9 pg/ml (mean +/- SEM) and 4.9 +/- 0.3 ng/ml respectively. After hCG the same biphasic pattern was observed in both fluids with a similar early response but the delayed peak at 72 h was relatively higher in saliva than in plasma. Thus it was suggested to collect saliva instead of plasma for the evaluation of testicular secretion of testosterone after hCG administration.

Testosterone assays: guidelines for the provision of a clinical biochemistry service

Approximately one year after the devolution of testosterone assays from the SAS, the Analytical Methods Working Party of the Association of Clinical Biochemists set up a working party to investigate the performance of the assays, to survey the available methodology and to give guidance on the factors that influence the assay. This document represents a summary of the deliberations of the group and forms one of a series of similar reports.

Gas chromatography-mass spectrometric analysis of salivary testosterone with reference to diethylstilboestrol-treated prostatic cancer patients

Testosterone has been quantified in saliva by gas chromatography-mass spectrometry with selected ion monitoring of the t-butyldimethylsilyl ether (TBDMS) of the oxime derivative. High specificity of instrumental analysis has been complemented by immunoadsorption for extraction of the analyte. The sensitivity of the detection is approx. 4 pmol/l and the precision of quantification is 3.4% CV for a value of 160 pmol/l and 11% for a value of 25 pmol/l. Female subjects had salivary concentrations of testosterone of 14-52 pmol/l, while testosterone was not detected in the saliva of prostatic cancer patients treated with diethylstilboestrol (DES).

Hormones in saliva

Since the collection of saliva is noninvasive, nonstressful and usually very convenient there have been many recent studies examining the clinical relevance of measuring various hormones in saliva. It now appears that the measurement of most unconjugated steroids in saliva will provide clinically useful data whereas the measurement of conjugated steroids, thyroid hormones, and protein hormones is unlikely to be clinically relevant. The key factors determining whether the salivary concentration of a hormone or drug is likely to be clinically relevant are the mechanisms by which the material enters the saliva; the "free to protein bound" ratio for the material; and the structure of the material, i.e., its molecular weight, polarity and the presence of ionizable groups.

Comparisons of the intra- and inter-individual variability in sex hormone levels of men

Six blood samples were taken from each of 33 healthy young men during the course of 2 weeks. From these, the testosterone, dihydrotestosterone, and oestradiol levels were determined. In addition, free testosterone was measured using saliva samples obtained from 23 of the subjects. Both coefficients of variation and variance analyses showed that the inter-individual variability is greater than the intra-individual for all of the hormones. No influence of the day of the week could be ascertained. However, it was discovered that the serum hormone levels dropped with age (weakly significant for oestradiol, insignificant for the androgens) in spite of the subjects being in or near their third decade. The pooled intra-individual correlations between the hormones all fell within the moderate to middle-high positive domain, whereas the inter-individual correlations ranged between highly positive (testosterone to dihydrotestosterone), middle highly positive (free testosterone to total testosterone), and slightly negative (free testosterone to oestradiol).

Basal and LHRH-stimulated gonadotropin levels and the circadian rhythm of testosterone and the effect of exogenous testosterone thereon

In seven eugonadal men, aged 20-26 years, a fall in plasma and saliva testosterone (T) levels between 8.00 and 16.00 h of the day was observed, but plasma oestradiol-17 beta levels did not show a significant variation. These findings substantiate the existence of a circadian rhythm in T levels. Concurrent with the decrease of T levels over the day, a small but significant rise in basal LH, but not in LHRH-stimulated LH levels were observed. Then the fall of plasma and saliva T levels over the day was prevented by the administration of 80 mg testosterone undecanoate (Andriol, Organon) by mouth at 8.00 h. A rise in plasma T and even more in saliva T levels was measured, which persisted till at least 16.00 h. At this hour basal LH, but not LHRH-stimulated LH levels appeared to be slightly, though significantly depressed. From our data we conclude that fluctuations of T levels of the magnitude of 25% around the baseline values, affect slightly basal LH levels, but not LHRH-stimulated LH levels.

Salivary cortisol: a better measure of adrenal cortical function than serum cortisol

Salivary cortisol concentration was found to be directly proportional to the serum unbound cortisol concentration both in normal men and women and in women with elevated cortisol-binding globulin (CBG). The correlation was excellent in dynamic tests of adrenal function (dexamethasone suppression, ACTH stimulation), in normals and patients with adrenal insufficiency, in tests of circadian variation and randomly collected samples. Women in the third trimester of normal pregnancy exhibited elevated salivary cortisol throughout the day. The relationship between salivary and serum total cortisol concentration was markedly non-linear with a more rapid increase in salivary concentration once the serum CBG was saturated. The rate of equilibrium of cortisol between blood and saliva was very fast, being much less than 5 minutes. These data, combined with a simple, stress-free, non-invasive collection procedure, lead us to suggest that salivary cortisol is a more appropriate measure for the clinical assessment of adrenocortical function than is serum cortisol.

Investigation of hirsutism: testosterone is not enough

In a consecutive series of 41 hirsute women clinically classified as benign androgen excess, only 34% were found to have elevated plasma 'total' testosterone (T), 22% having subnormal sex hormone binding globulin (SHBG). When expressed as the ratio T/SHBG ('free androgen index'), 85% of the patients had values above the normal range. It is concluded that this index is more reliable than total testosterone in assessing androgen status in female patients.

Testosterone concentrations in human seminal plasma and saliva and its correlation with non-protein-bound and total testosterone levels in serum

A sensitive, specific and precise non-chromatographic method for the radioimmunoassay of testosterone in human seminal plasma and saliva from adult and pubertal males is described, and the values compared to total and non-protein-bound testosterone levels in serum. There was a significant correlation between salivary and serum-free levels of testosterone (r = 0.75, P less than 0.001, n = 67) whilst the correlation of serum levels of total testosterone with free as well as with salivary testosterone levels was weaker (r = 0.63 and 0.64, respectively). The salivary and serum levels of free testosterone showed better correlation with the stage of puberty than did the serum levels of total testosterone. Further evidence for a correlation between salivary and serum levels of free testosterone was obtained following oral administration of testosterone undecanoate, as this treatment increased the mean concentration of serum total testosterone after 3 h by 82%, but increased salivary and serum levels of free testosterone by only 30% and 20%, respectively. The coefficient of correlation between serum levels of total testosterone and seminal plasma testosterone was 0.73 (P less than 0.001), whilst the correlation between levels of serum-free testosterone with both salivary and seminal plasma levels of testosterone was statistically non-significant. Our observations on salivary testosterone are in accordance with the diffusion of non-protein-bound steroids into peripheral tissues, and consequently into their secretions. This model, however, does not appear to be applicable to the sex accessory glands.

Testosterone and delta 4-androstenedione in the saliva of patients with Klinefelter's syndrome

This study compares salivary and total plasma levels of testosterone and androstenedione in healthy controls and chromatin positive patients with Klinefelter's syndrome. The mean plasma testosterone level in fifteen Klinefelter patients was significantly lower than in ten control subjects, but the mean androstenedione levels were similar. Saliva of the normal men and patients with Klinefelter's syndrome did not differ significantly in binding potency for testosterone or androstenedione. The mean salivary testosterone level in the Klinefelter patients was significantly lower than in the controls but again salivary androstenedione levels were similar. Although testosterone as well as androstenedione showed a fair correlation between the salivary and plasma concentrations, preliminary data suggested that salivary testosterone levels better characterize the clinical state of androgenicity than do plasma levels. When salivary testosterone and androstenedione were expressed as a percentage of total steroid levels in plasma, the ratios for both steroids were similar in Klinefelter patients and healthy controls and closely approximated to the reported percentages of free steroid levels in plasma. The absolute salivary testosterone concentrations also were almost identical to the reported free plasma hormone levels. Together the data provide indirect evidence that in Klinefelter patients testosterone and androstenedione binding in plasma and saliva does not differ from that in normal men, and measurement of salivary steroids, by reflecting their free hormone concentration in plasma, may be useful in evaluating endocrine function in both health and disease.

Applicability of salivary testosterone measurements for the follow-up of therapy of idiopathic hirsutism

Preliminary data on the applicability of salivary evaluations of testosterone (T), in comparison to plasma evaluations, in monitoring the effects of prednisone treatment in hirsutism are shown by the authors. 7.5 mg daily were administered p.o. to 4 volunteers affected by idiopathic hirsutism and to a fifth case in whom later surgery demonstrated the presence of adrenal virilizing tumour. While in the latter no modification either in plasma or salivary levels of the hormone was shown, in the four cases of idiopathic hirsutism a striking decrease of T levels was observed both in plasma (P less than 0.02) and in saliva (P less than 0.01). The two variations were moreover highly correlated (r = 0.80; P less than 0.001). Data obtained, although from a limited number of cases, seem then to confirm the validity of salivary T determination in this clinical condition, in whom the need of repeated evaluations, if performed on plasma, may become particularly stressful for the patient.

Salivary testosterone: relationship to unbound plasma testosterone in normal and hyperandrogenic women

A sensitive radioimmunoassay (RIA) was used to measure salivary testosterone levels in normal women, in patients with polycystic ovaries (PCO), and in women with hirsutism. There was a highly significant correlation (r = 0.79, P less than 0.001) between the concentration of testosterone in saliva [12.3 +/- 7.8 (SD) pg/ml] and the concentration of unbound testosterone in plasma (5.2 +/- 3.1 pg/ml) in matched samples collected from 56 women including normals, patients with clinical signs of hyperandrogenism, and women treated with a combination of cyproterone acetate (CA) and ethinyl oestradiol (EE). The unbound plasma testosterone was measured in the dialysate directly using a sensitive RIA. Salivary and plasma testosterone levels in patients with PCO (20.6 +/- 8.5 and 626 +/- 187 pg/ml respectively, n = 14) and in those with hirsutism (13.9 +/- 5.6 and 421 +/- 170 pg/ml, n = 30) were significantly higher (P less than 0.001) than levels in normal women (7.7 +/- 2.6 and 196 +/- 68 pg/ml, n = 36). Treatment for 3 months with CA and EE resulted in a decrease (mean 68%) in salivary testosterone levels in all patients studied (n = 15), but the suppression of plasma testosterone (mean 34%) was not observed in all cases. It is concluded that measurement of salivary testosterone gives a useful indication of levels of biologically available androgen in hyperandrogenic women, before and during CA/EE therapy.

Salivary cortisol determination: adaptation of a commercial serum cortisol kit

The determination of cortisol levels in saliva offers a number of significant advantages as compared with plasma or serum measurements, and radioimmunoassay is the technique of choice because of the greater sensitivity required. To date, the assay of salivary cortisol has been limited to only a few centres employing 'in-house' reagents. We describe some simple modifications of a commercial kit, designed for the assay of cortisol in serum and urine, which allow direct and rapid determination of the steroid in saliva. These modifications enable any laboratory with access to a gamma counter to perform large numbers of salivary cortisol assays at relatively low cost.

Steroid hormone content in the submandibular gland of normal and pregnant rats

Using radioimmunoassay, the levels of steroid hormones were determined in the submandibular gland homogenates from sexually immature, mature male and female and pregnant rats. High progestagen content was established in the whole gland of maturing and mature females and males. The highest level of progestagens was in the last pregnancy trimester. High androgen content was found in maturing and mature males only. Androgens in females and oestrogens in both males and females were present in traces only.

Histochemistry of the 3 beta-hydroxysteroid, 17 beta-hydroxysteroid and 3 alpha-hydroxysteroid dehydrogenases in human salivary glands

Human parotid and submandibular glands showed no 3 alpha-hydroxysteroid dehydrogenase (3 alpha-HSD) activity. The 3 beta-hydroxysteroid dehydrogenase (3 beta-HSD) and the 17 beta-hydroxysteroid dehydrogenase (17 beta-HSD) appeared intensely reactive in the duct epithelia of the male and female glands and weakly reactive in the acinar cells of the female ones. The failure to demonstrate 3 alpha-HSD activity indicates that in-vivo androgen activation, if present at all, is not so marked as in target organs. The different distribution of the 3 beta-HSD and 17 beta-HSD in the two sexes can be related not only to the oxidation of androgens but also to the metabolism of the female hormones. Glucose-6-phosphate dehydrogenase (G6PD) and 6-phosphogluconate dehydrogenase (6PGD) do not seem to be specifically influenced by the sex hormones as their pattern of distribution showed no sex differences.