Free Testosterone Index Research Articles

Hair growth and androgen responses in hirsute women treated with continuous cyproterone acetate and cyclical ethinyl oestradiol.

Eighteen hirsute women (8 with polycystic ovarian syndrome, 10 with idiopathic hirsutism) were treated for up to 12 months with cyproterone acetate, 150 mg daily, and ethinyl oestradiol, 50 micrograms on days 5-25 of the menstrual cycle. Hair growth rate and density were measured from standardized serial photographs of a shaved skin area. A significant reduction was seen in mean hair growth rate, total plasma testosterone, free testosterone index, plasma dehydroepiandrosterone, and plasma androstenedione. LH and FSH also fell and sex hormone binding globulin level increased. No significant changes occurred in hair density or in serum PRL levels. A significant correlation was observed between hair growth rate and total plasma testosterone for the pooled results (r = 0.35, P less than 0.005). No significant correlations were seen between hair density and the endocrine parameters studied.

Effects of Cyproterone-Containing Oral Contraceptive on Hormonal Levels in Polycystic Ovarian Disease

The management of polycystic ovarian disease in women not desiring pregnancy is controversial. To avoid the progressive androgenic effects on peripheral target organs, some advocate the use of oral contraceptives. This study reports the effects of a preparation with 50 micrograms ethinyl estradiol and 2 mg cyproterone acetate on gonadotropins, prolactin, testosterone, sex hormone binding globulin (SHBG), androstenedione, and calculated free testosterone index before and after six months of treatment. Gonadotropins, testosterone, and androstenedione levels decreased, prolactin did not change, and sex hormone binding globulin increased as the result of the treatment. This led to a net decrease in the calculated free testosterone.

Salivary Testosterone: A Good Measure to Reflect Free Testosterone

To the Editor. —Salivary testosterone— sampling regimens have the advantage of frequent, easy collection by a noninvasive stress-free technique, and its measurement can provide an index of free testosterone in plasma or serum. 1-3 Several situations, such as hirsutism, pregnancy, and hyperthyroidism, are known in which there is a discordance between total testosterone levels in serum and clinical signs of androgenicity. 3,4 Thus, in Graves' disease testosterone binding is generally increased by thyroid hormone levels and, therefore, levels for total testosterone are increased, whereas levels for free testosterone are normal. 5 We measured total serum and salivary testosterone with a radioimmunoassay kit (Farmos Diagnostica, Oulu, Finland) in a group of 59 normal women and 25 women with Graves' disease and who did not undergo treatment (age, 20 to 56 years). Our data show that the normal levels for total testosterone in serum were 53 ±21 ng/dL and 2.9±1 ng/dL

Salivary androgens in hirsutism: are they of use in routine evaluation?

In the biochemical evaluation of hirsutism, 50% or less of patients have an elevated total serum testosterone. Recent work has suggested that measuring salivary testosterone or a derived serum 'free testosterone index' may be of use in the evaluation of hyperandrogenism. We have measured serum total, derived serum free indices and salivary concentrations of testosterone and 5 alpha-dihydrotestosterone in an unselected group of hirsute patients in order to assess their value in the routine evaluation of hirsutism. The assays were performed using a novel oxidation procedure to overcome the need for chromatographic separation. The 'free testosterone index' gave the best discrimination. Salivary androgen concentrations were comparatively poor and cannot be recommended for routine use.

Impotence in scleroderma.

Hormonal, neurologic, and vascular factors affecting potency were evaluated in 10 men with scleroderma and in 10 age-matched men with rheumatoid arthritis. Impotence was reported by 6 of the patients with scleroderma and none with rheumatoid arthritis. Studies of serum testosterone, free testosterone index, follicle-stimulating hormone, luteinizing hormone, prolactin, estradiol, thyroxine, and thyrotropin did not show a hormonal basis for impotence in any patient. Neurologic causes were not found on physical examination. Penile blood pressures were markedly abnormal in 4 impotent patients, intermediate in 2 impotent and 3 potent patients, and normal in 11 potent patients. A history of claudication and diminished ankle blood pressures indicated large vessel disease in 2 impotent patients; the remaining 4 impotent men had normal ankle pressures, suggesting that their poor penile blood pressures and impotence were due to small vessel disease, perhaps the small artery lesions of scleroderma.

Free Testosterone Index : Comparison with Plasma Free Testosterone

Blood-free testosterone indices were measured among 28 normal men (age; 24-48 yrs.), 20 normal women (20-36 yrs.), 18 pregnant women (22-31 yrs.), 17 males with hypogonadism (23-56 yrs.), 17 males with chronic hepatitis (20-42 yrs.), 24 males with liver cirrhosis (29-68 yrs.), 34 males with hyperthyroidism (20-42 yrs.) and 7 hirsute women (18-31 yrs.), and these were compared with the plasma concentrations of free testosterone. The testosterone index was obtained by multiplying the plasma concentration of testosterone by the percent of sex hormone-binding globulin (SHBG), non-bound testosterone precipitated by dextran-coated charcoal. A significant increase of plasma testosterone was observed in patients with chronic hepatitis (p less than 0.001) and hyperthyroidism (p less than 0.001) as compared with normal men and was also observed in pregnant (p less than 0.01) and hirsute women (p less than 0.01) as compared with normal women. The close negative correlation between plasma levels of testosterone and the percent of SHBG non-bound testosterone (r = -0.87, n = 79, p less than 0.001) was observed among normal men, male patients with chronic hepatitis and hyperthyroidism. The sex hormone binding capacity was increased from two to three fold in patients with chronic hepatitis and hyperthyroidism. The patients with compensated liver cirrhosis had increased plasma testosterone and a decreased percent of SHBG non-bound testosterone, and those with decompensated liver cirrhosis had decreased plasma testosterone and a normal percent of SHBG non-bound testosterone. The plasma concentration of free testosterone was normal in patients with chronic hepatitis and hyperthyroidism. It decreased in pregnancy (p less than 0.01) and increased in hirsute women (p less than 0.01). The blood free testosterone index was slightly high in one third of the patients with chronic hepatitis and hyperthyroidism as compared with that in normal men. However, a close correlation of the percent of SHBG non-bound testosterone and fractional free testosterone (%) measured by equilibrium dialysis (gamma = 0.82, p less than 0.001) was obtained in all subjects (n = 170). These data suggest that the blood free testosterone index parallels the plasma concentration of free testosterone and is useful to evaluate the status of androgenicity.

The Relationship between Sex Steroids and Sex-Hormone-Binding Globulin in Plasma in Physiological and Pathological Conditions

Physiological and many pathological changes in plasma sex-hormone-binding globulin (SHBG) levels have been attributed to the opposing effects of androgens which lower, and oestrogens which elevate, levels. We examined four clinical situations in which changes in SHBG levels may not be explained by sex steroid alterations. (1) Dexamethasone caused an increase in SHBG levels in hyperandrogenaemic hirsute women whether or not androgens were suppressed. (2) In male patients with untreated isolated gonadotrophin deficiency there was a highly significant correlation between SHBG levels and age, but there was no relationship between the levels of SHBG and those of plasma testosterone, androstenedione or DHEAS. (3) Two 46-XY siblings, phenotypic female subjects with complete androgen insensitivity, demonstrated a marked decline in SHBG levels between the ages of 9-13 and 12-16 years. (4) SHBG was suppressed in obese oligomenorrhoeic women while plasma concentrations of testosterone, androstenedione and oestradiol were normal and that of oestrone was elevated; however, the testosterone:SHBG ratio, an index of free testosterone, was elevated. These observations indicate that the decline in SHBG levels which normally occurs in men during the second decade of life is independent of androgen activity and is under the influence of as yet unidentified factors. Glucocorticoids in small doses under the influence of as yet unidentified factors. Glucocorticoids in small doses increase SHBG levels independently of sex steroid alterations while elevated free testosterone concentration may contribute to suppression of SHBG in obesity.

Role of Dehydroepiandrosterone Sulfate as a Prehormone for Ovarian Steroidogenesis

The endocrine effects of induction of ovulation with menotropins were studied in 43 patients: 11 with hypothalamic amenorrhea and 32 with the polycystic ovary syndrome. Patients with polycystic ovary syndrome had higher base-line values of serum follicle-stimulating hormone (FSH), luteinizing hormone (LH), 17 beta-estradiol, dehydroepiandrosterone sulfate, testosterone, and a higher testosterone-free index than those with hypothalamic amenorrhea. During treatment with menotropins, patients with polycystic ovary syndrome had higher values of serum LH, prolactin, dehydroepiandrosterone sulfate, testosterone, percent free testosterone, testosterone-free index, and body weight than those with hypothalamic amenorrhea; serum FSH, dose of menotropins per kilogram body weight, and total follicular volume were higher in patients with hypothalamic amenorrhea than in those with polycystic ovary syndrome. Multiple linear regression after log transformation demonstrated that the testosterone-free index was predicted statistically by total ovarian volume and dehydroepiandrosterone sulfate and that serum 17 beta-estradiol was predicted statistically by total ovarian volume and testosterone-free index. Adding dexamethasone to menotropins in six patients with polycystic ovary syndrome produced significant decreases in 17 beta-estradiol, dehydroepiandrosterone sulfate, testosterone, and testosterone-free index. Higher concentrations of endogenous serum LH and dehydroepiandrosterone sulfate in patients with polycystic ovary syndrome in comparison with those with hypothalamic amenorrhea were associated with higher concentrations of serum testosterone, a lower total follicular volume, and an effective response to menotropins at a lower serum FSH and a lower dose of menotropins per kilogram body weight. These data suggest that serum dehydroepiandrosterone sulfate may be a precursor for ovarian steroidogenesis.

Hidradenitis suppurativa—an androgen—dependent condition

Hidradenitis suppurativa (HS) is a chronic inflammatory condition of apocrine glands: poral occlusion and added bacterial infection are important aetiological factors but the underlying cause remains unknown. Published animal work shows the apocrine gland to be hormone dependent. A similar mechanism is suggested in man by the failure of HS to appear before puberty and its frequent association with acne vulgaris and other androgen‐dependent conditions. Forty‐two patients with HS have been clinically and biochemically assessed for signs of increased androgen activity. Plasma total testosterone (T) sex hormone binding globulin (SHBG), dehydroepiandrosterone sulphate (DHEAS), prolactin, FSH, LH and 17‐β‐oestradiol were measured and compared with groups of normal female controls (n = 25) and subjects with idiopathic hirsutes (n = 37). Nineteen patients with no medical contra‐indications were randomly allocated to a double‐blind within‐patient crossover trial comparing ethinyloestradiol (50 μg) and cyproterone acetate (CPA) (50 mg in a reverse sequential regimen) with ethinyloestradiol (50 mg and levonorgestrel (500 μg) (EU/50). Treatments were sequential for 12 months with crossover at 6 months. Patients were assessed by clinical grading. Results demonstrate total testosterone and T/SHBG (index of free testosterone) to be significantly higher than normal controls in the HS group (P < 0·01) as well as the hirsutes group (P < 0·01). Treatment with CPA demonstrated significant improvement as compared to EU/50. Seven of the 19 patients were ultimately clear of the disease. We propose clinical criteria, including the presence of comedones, associated acne and pilonidal sinus, for establishing the diagnosis of HS. Early recognition is important for satisfactory control of the disease.

The effect of estrogen treatment on plasma concentrations of steroid hormones, gonadotropins, prolactin and sex hormone-binding globulin in post-menopausal women

Twenty-one post-menopausal women on no other medication were treated with a low dose (0.625 mg/day) of conjugated equine estrogen (CEE) for a mean (±SEM) period of 2.6±0.2 mth (range 1.75-4.75). Blood samples were collected before and at the completion of therapy, and alterations in the levels of prolactin (PRL), follicle-stimulating hormone (FSH), luteinizing hormone (LH), sex hormone-binding globulin (SHBG) and certain steroid hormones, including the free testosterone (T) index (T/SHBG) were studied. Following treatment, a significant increase in SHBG levels produced a significant decrease in the free T index ( P < 0.005). As expected, no changes were observed in the levels of PRL and steroid hormones other than estrone (E 1) and estradiol-17-β (E 2). Our observations indicate that treatment of post-menopausal women with low-dose estrogen lowers the unbound T.

Effects of two progestins with different androgenic properties on hepatic endothelial lipase and high density lipoprotein 2

Thirty postmenopausal women were randomly treated with desogestrel (DG) or levonorgestrel (LN) 125 μg/day for 3 weeks. Desogestrel reduced the serum total and free (non-protein bound) testosterone concentrations. It caused a small decrease in the sex hormone binding globulin capacity (SHBG) but did not influence the free testosterone index (testosterone/SHBG ratio). Levonorgestrel, on the other hand, did not influence the free testosterone concentration, but caused a significant increase in the free testosterone index. Levonorgestrel reduced the HDL and particularly the HDL 2 cholesterol concentrations (mean change from 1.75 to 1.45 mmol/l for HDL and from 0.73 to 0.50 mmol/l for HDL 2, P < 0.001). It also caused a reduction in the VLDL triglyceride ( P < 0.05) but not the total serum triglyceride concentration. Desogestrel did not cause any significant changes in HDL or HDL 2 cholesterol concentrations, but it reduced the VLDL triglyceride ( P < 0.01) and total serum ( P < 0.05) triglyceride concentrations. Neither of the two progestins influenced the postheparin plasma lipoprotein lipase (LPL) activity or the serum cholesterol esterification rate by lecithin: cholesterol acyltransferase (LCAT). It is therefore possible that both steroids decreased the hepatic output of triglycerides, which may be clinically important since both progestins are used in combination with ethinylestradiol (EE) which increases the hepatic TG synthesis. The failure of desogestrel to change HDL levels is consistent with earlier data on the lack of effects on HDL by non-androgenic progestins. Levonorgestrel increased the mean activity of postheparin plasma hepatic lipase (HL) from 23.3 to 28.0 pmol · h −1 · ml −1 ( P < 0.05). In contrast, this activity was not influenced by desogestrel. The magnitude of the changes in postheparin plasma HL activity and the free testosterone index (testosterone/SHBG ratio) showed significant positive correlation (+0.41, P < 0.05). On the other hand, the changes in the HDL 2 cholesterol and the postheparin plasma HL activity were inversely interrelated ( r= −0.52, P < 0.01). These relationships are consistent with the idea that the effects of different progestins on the HDL cholesterol are mediated by the sex steroid sensitive hepatic endothelial lipase.

Daily variations of plasma sex hormone-binding globulin binding capacity, testosterone and luteinizing hormone concentrations in healthy rested adult males.

In this study the daily variations of plasma sex hormone-binding globulin (SHBG) binding capacity were measured together with plasma testosterone and luteinizing hormone (LH) concentrations in 7 healthy rested adult males. Plasma SHBG-binding capacity demonstrated a significant circadian rhythm (acrophase = 2.06 p.m.; mesor = 0.35 +/- 0.6 ng testosterone bound/100 ml; amplitude = 17% of the mesor). Plasma testosterone also showed a circadian rhythm (acrophase = 7.02 a.m.; mesor = 4.38 +/- 0.67 ng/ml; amplitude = 18% of the mesor). The free testosterone index (or the ratio between plasma testosterone and SHBG-binding capacity) was not correlated with plasma LH levels. In our hands this last parameter did not vary according to a circadian pattern. These data are discussed in terms of a feedback mechanism controlling the pituitary-testis axis regulation.

Salivary testosterone levels in normal and testosterone treated monkeys

Testosterone concentrations in saliva and matched serum samples were measured by radio-immunoassay, in normal and testosterone treated adult male cynomolgus monkeys ( Macaca fascicularis) and in untreated rhesus monkeys ( Macaca mulatta). Saliva and serum samples were collected under ketamine anaesthesia. In 4 cynomolgus monkeys, the mean basal testosterone levels were 195 ± 22 pmol/1 in saliva and 24 ± 2 nmol/1 in serum ( n = 17). The salivary testosterone corresponded to 0.9 ± 0.1% of the serum testosterone concentration. In 5 rhesus monkeys the mean testosterone concentrations were 147 ± 10 pmol/1 in saliva and 17 ± 2 nmol/1 in matched serum samples with a percent ratio of 1.6 ± 0.3 ( n = 15). Following a single i.m. injection of 50mg testosterone propionate to 4 monkeys, both salivary and serum testosterone levels increased promptly and in parallel. The salivary testosterone levels correlated well with the serum testosterone values. The study suggests that, as in the human, determination of salivary testosterone can be used as an index of free testosterone in this animal. Moreover, the monkey can be used as a model for studies in the human involving monitoring of salivary testosterone.

Male gonadal function in coeliac disease: III. Pituitary regulation.

Pituitary regulation of gonadal function was investigated in 39 consecutive men with treated and untreated coeliac disease and in an intestinal disease control group of 19 men with Crohn's disease of similar age and general nutritional status. Basal serum FSH concentration was increased in 10 of the coeliacs (26%) compared to only two of 19 men with Crohn's disease (11%). This abnormality was observed with equal frequency in both treated and untreated coeliacs, and was not associated with oligospermia. Serum LH concentration was increased in eight of 15 untreated coeliacs (53%) with sub-total villous atrophy, an abnormality which unlike the elevation of serum FSH, appears to return towards normal after gluten withdrawal. Serum LH was high in coeliacs despite marked elevation of the free testosterone index. Exaggerated responses of FSH and LH to LHRH were found in 89% and 45% respectively, of coeliacs with sub-total villous atrophy. However, exaggerated responses of LH alone were found more frequently in coeliacs than in men with Crohn's disease (P less than 0.02) and unlike the exaggerated FSH responses, LH responses were closely related to jejunal morphology. Exaggerated responses of FSH and LH in coeliacs were commonly found when basal gonadotrophin concentrations were normal. The occurrence of exaggerated gonadotrophin responses could not be related to plasma concentration of testosterone, dihydrotestosterone, oestradiol or the free testosterone index. Serum prolactin was modestly raised in 25% of untreated and partially treated coeliacs and in the same proportion of men with Crohn's disease. Elevated serum prolactin concentrations never exceeded 809 mU/l and were not associated with impotence or infertility. This study provides further evidence that in men with coeliac disease there is a derangement of pituitary regulation of gonadal function. This would seem to be part of a wider disturbance of central regulatory mechanisms of endocrine function in coeliac disease.

Decrease in serum testosterone concentration during treatment with tetracycline.

Serum concentrations of testosterone and the binding capacity of sex hormone binding globulin (SHBG) were measured on days 2 days immediately preceding tetracycline treatment, on 3 days of treatment and on 2 days immediately after cessation of treatment. On the treatment days serum mean testosterone concentrations were significantly lower than on the control days (17 +/- 0.9 vs 21 +/- 0.8 nmol/l, P less than 0.01). There were no differences in the SHBG. The 'free testosterone index' behaved like the total testosterone.

Serum gonadotropins and testosterone in men occupationally exposed to carbon disulfide.

Serum levels of gonadotropins and testosterone were determined for 69 men exposed to CS2 in viscose rayon production and for 22 nonexposed male controls. Mean ages of the two groups were 40 and 39 yr, respectively. The duration of exposure to CS2 ranged from 1 to 36 yr (mean, 12.5). Follicle stimulating hormone (FSH) levels were significantly higher among the exposed men than among the controls (13.6 versus 10.0 IU/l). The level of sex hormone binding globulin (SHBG) was significantly lower among the exposed men than among the controls (44 versus 53 nmol/l). Luteinizing hormone (LH) values were significantly higher among the 24-31 yr old exposed men than the controls of the same age. Levels of SHBG, of the free testosterone index, and of FSH and LH in men under 39 yr who had been exposed to CS2 for 1-9 yr differed significantly from those of their controls, whereas only FSH differed significantly among those exposed for 10 or more years in the same age group. In men aged 40 yr or more, after at least 10 yr of exposure, only FSH and LH were significantly higher than in the controls. Although the observed alterations do not seem to have any overt clinical implications for the study groups, they indicate that overall levels of CS2 that are well below the current Finnish threshold limit value (30 mg/m3) may affect the hormonal balance in the pituitary-gonadal axis, favoring the view that there may be an increased risk of latent primary gonadal insufficiency.

Male gonadal function in coeliac disease: 2. Sex hormones.

Hypogonadism, infertility, and sexual dysfunction occur in some men with coeliac disease. We have measured plasma testosterone, dihydrotestosterone, sex-hormone binding globulin, oestradiol, and serum luteinising hormone in 41 men with coeliac disease and have related these findings to jejunal morphology, fertility, semen quality, and sexual function. To determine the specificity of these observations in coeliacs we also studied 19 nutritionally-matched men with Crohn's disease, and men with chronic ill-health due to rheumatoid arthritis and Hodgkin's disease. The most striking endocrine findings in untreated coeliacs were increased plasma testosterone and free testosterone index, reduced dihydrotestosterone (testosterone's potent peripheral metabolite), and raised serum luteinising hormone, a pattern of abnormalities indicative of androgen resistance. As jejunal morphology improved hormone levels appeared to return to normal. This specific combination of abnormalities was not present in any of the disease control groups and, to our knowledge, androgen resistance has not been described previously in any other non-endocrine disorder. Plasma oestradiol concentration was modestly raised in 10% of coeliacs and 11% of patients with Crohn's disease. Unlike plasma androgens and serum luteinising hormone in coeliacs, plasma oestradiol was not clearly related to jejunal morphology. Androgen resistance and associated hypothalamic-pituitary dysfunction appear to be relatively specific to coeliac disease and cannot be explained merely in terms of malnutrition or chronic ill-health. In addition, our findings suggest that this endocrine disturbance may be related to sexual dysfunction in coeliac disease but its relationship to disordered spermatogenesis in this condition has not been clearly established.

Role of androgens in menstrual disorders of nonhirsute and hirsute women, and the effect of glucocorticoid therapy on androgen levels in hirsute hyperandrogenic women

The plasma concentrations of total testosterone, free testosterone index, androstenedione, 17β-estradiol, luteinizing hormone, follicle-stimulating hormone, prolactin, and urinary 17-ketosteroid and 17-ketogenic steroid excretion were measured in 48 nonhirsute and 119 hirsute patients. Hormone data were compared within and between groups according to whether the menstrual cycles were eumenorrheic, amenorrheic, or oligomenorrheic. Eleven hirsute women treated with prednisone were followed for 6 months. It was concluded that: (1) androstenedione, testosterone, free testosterone index, and adrenal androgens alone or in combination play a role in the pathogenesis of the hirsutism observed in eumenorrheic women and in the amenorrhea and oligomenorrhea of both hirsute and nonhirsute women; (2) body weight correlated with adrenal adrogens (17-ketosteroids) in nonhirsute women and with androstenedione in hirsute women; (3) prednisone significantly suppressed androstenedione and 17-ketosteroids (p < 0.05), with a decline of testosterone to 65% and luteinizing hormone to 51% of pretreatment values, with favorable clinical effects on the hirsutism, menstrual dysfunction, and infertility; (4) concentrations of 17β-estradiol were lower in amenorrheic than in eumenorrheic and oligomenorrheic women of both groups.

The pituitary-testicular axis of uraemic subjects on haemodialysis and continuous ambulatory peritoneal dialysis.

Pituitary-testicular function was evaluated in 18 patients with chronic renal failure, 9 treated by maintenance haemodialysis (HD) and 9 by continuous ambulatory peritoneal dialysis (CAPD), and compared with a non-uraemic control group. Serum total testosterone and the free testosterone index were significantly low in both dialysis groups. Basal FSH and LH levels were elevated but this reached significance only with regard to LH. The responses of both FSH and LH to the iv administration of LRH were normal. There was no significant difference between the CAPD and HD groups in any of the hormonal parameters estimated. While CAPD may improve control of some metabolic parameters when compared with HD, it does not improve the function of the pituitary-testicular axis.

Testosterone free index correlates best with dehydroepiandrosterone sulfate

Correlation coefficients for dehydroepiandrosterone sulfate (DHEAS) were determined in women on menotropin. DHEAS was significantly correlated with testosterone free index (TFI), 0.78**; percentage free testosterone (%FT), 0.66**; androstenedione (delta 4A), 0.66*; luteinizing hormone (LH), 0.55**; LH/follicle-stimulating hormone (FSH) ratio, 0.55**; 17-OH-progesterone (17-P), 0.55**; testosterone (T), 0.53**; weight (WT), 0.40**, urinary estriol glucuronide (E3G), 0.33*; and free cortisol index (FFI), 0.32*, with 43 df but not with prolactin (PRL), 0.25. Normal male DHEAS (3.5 +/- 1.2, 25) (microgram/ml; mean +/- standard deviation, n) was higher than normal female DHEAS (2.4 +/- 1.1, 27), P less than 0.01 and DHEAS in women on oral contraceptives (1.9 +/- 1.1, 17) was slightly lower than in normal females, P greater than 0.2. In the combined population (male, female, and females on oral contraceptives) DHEAS was correlated with TFI (0.56**), T (0.54**), %FT (0.52**), delta 4A (0.40**), and age (-0.40**) with 66 df and 17-P (0.30*) with 54 df. TFI appears to be one determinant of plasma DHEAS, **P less than 0.01. *P less than 0.05.