help button home button Endocrine Society JCEM
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Journal of Clinical Endocrinology & Metabolism, doi:10.1210/jc.2006-1847
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
92/7/2696    most recent
Author Manuscript (PDF)
Right arrow Submit a related Letter to the Editor
Right arrow Purchase Article
Right arrow View Shopping Cart
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Copyright Permission
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Nielsen, T. L.
Right arrow Articles by Andersen, M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Nielsen, T. L.
Right arrow Articles by Andersen, M.
Related Collections
Right arrow Male Endocrinology
Right arrow Metabolism
The Journal of Clinical Endocrinology & Metabolism Vol. 92, No. 7 2696-2705
Copyright © 2007 by The Endocrine Society

Visceral and Subcutaneous Adipose Tissue Assessed by Magnetic Resonance Imaging in Relation to Circulating Androgens, Sex Hormone-Binding Globulin, and Luteinizing Hormone in Young Men

Torben Leo Nielsen, Claus Hagen, Kristian Wraae, Kim Brixen, Per Hyltoft Petersen, Egil Haug, Rasmus Larsen and Marianne Andersen

Department of Endocrinology (T.L.N., C.H., K.W., K.B., M.A.), Odense University Hospital, 5250 Odense SV, Denmark; Norwegian Quality Improvement of Primary Care Laboratories (P.H.P.), Division for General Practice, University of Bergen, 5020 Bergen, Norway; Hormone Laboratory (E.H.), Aker University Hospital, 0514 Oslo, Norway; and Department of Informatics and Mathematical Modeling (R.L.), Technical University of Denmark, Copenhagen, 2800 Kongens Lyngby, Denmark

Address all correspondence and requests for reprints to: Torben Leo Nielsen, Dyrupgardvaenget 76, 5250 Odense SV, Denmark. E-mail: Torben{at}dsa-net.dk.


   Abstract
Top
 Abstract
Introduction
 Subjects and Methods
 Results
 Discussion
 References
 
Context: No large studies of young men have examined circulating sex hormones in relation to visceral and sc adipose tissues.

Objective: The aim of this study was to investigate the role of visceral adipose tissue and sc adipose tissue on circulating sex hormones and the impact of obesity on sex hormone reference intervals.

Design, Setting, and Participants: Population-based study of 783 Danish 20- to 29-yr-old men was performed using dual-energy x-ray absorptiometry in all men and magnetic resonance imaging in 406 men.

Main Outcome Measures: Total, bioavailable, and free testosterone, dihydrotestosterone (DHT), total and bioavailable estradiol, SHBG, and LH were measured.

Results: In multiple regressions, visceral adipose tissue was an independent, inverse correlate of bioavailable and free testosterone. Subcutaneous adipose tissue correlated negatively with SHBG and positively with bioavailable estradiol adjusted for total testosterone. Both visceral adipose tissue and sc adipose tissue correlated inversely with total testosterone and DHT. Adjusting for SHBG, only visceral adipose tissue remained significantly correlated. Low total testosterone in viscerally obese men was not accompanied by increased LH. The androgen reference intervals were significantly displaced toward lower limits in obese vs. nonobese men (total testosterone: 8.5–29.3 vs. 12.5–37.6 nmol/liter; bioavailable testosterone: 6.1–16.9 vs. 7.6–20.7 nmol/liter; free testosterone: 0.23–0.67 vs. 0.29–0.78 nmol/liter; and DHT: 0.63–2.5 vs. 0.85–3.2 nmol/liter), whereas total estradiol (36.5–166 pmol/liter) and bioavailable estradiol (23.4–120 pmol/liter) reference intervals were not. In obese men, 22.9% had total testosterone less than 12.5 nmol/liter.

Conclusions: Visceral adipose tissues correlate independently with bioavailable and free testosterone in young men. The inverse relationship between total testosterone and sc adipose tissue seems to be accounted for by variations in SHBG. The reference intervals for total testosterone, bioavailable testosterone, free testosterone, and DHT are displaced toward lower limits in obese men.


   Introduction
Top
 Abstract
 Introduction
Subjects and Methods
 Results
 Discussion
 References
 
INVERSE RELATIONS BETWEEN serum total testosterone and fat mass have been shown in both observational and interventional studies (1, 2, 3, 4, 5, 6, 7). Some authors observed an inverse, linear relationship between free and bioavailable testosterone (7), others reported reduced free or bioavailable testosterone in severely obese men (2), and some found no significant relationships (8, 9). Some studies indicate a disruption of the hypothalamo-pituitary-gonadal axis with attenuated LH pulse amplitude in morbidly obese men (2, 10). The few existing studies on the type of adiposity related with low androgen levels are inconsistent (1, 4, 11).

The diagnosis of androgen deficiency is uncomplicated in severe cases, but reference intervals are necessary due to unspecific symptoms (12) and the growing use of androgen replacement therapy (13). Reference intervals may lack statistical power and can be biased due to: 1) unintentional inclusion of hypogonadal subjects, which can be limited by testicular examination and a medical history focusing on testicular, pituitary, and chronic diseases (14), medication (15, 16, 17, 18, 19), excessive alcohol intake (20), and anabolic steroid abuse; 2) failure to recruit a population-based sample not mirroring the general population; 3) inappropriate timing of blood sampling; and 4) use of inaccurate testosterone assays. Commercially available direct immunoassays for assessment of total testosterone are widely used, but these assays suffer from a lack of accuracy (21, 22, 23, 24). An impact of obesity on androgen reference intervals has not been reported, but the increasing prevalence of obesity and the variation in prevalence between regions may contribute to different reference intervals.

We examined a population-based cohort of 783 Caucasian males, aged 20–29 yr, to investigate the impact of obesity on reference intervals for serum total testosterone, bioavailable testosterone, free testosterone, androstenedione ({Delta}4AD), dihydrotestosterone (DHT), estradiol (E2), and bioavailable E2. In addition, we addressed the following questions: 1) Are free testosterone, bioavailable testosterone, {Delta}4AD, DHT, E2, and bioavailable E2 related with adiposity in young men?; 2) Are variations in sex hormones primarily attributable to central fat mass (CFM)?; 3) If so, are sc adipose tissue, visceral adipose tissue, or both involved?; and 4) Does LH correlate with specific types of adipose tissues?


   Subjects and Methods
Top
 Abstract
 Introduction
 Subjects and Methods
Results
 Discussion
 References
 
Subjects

The Odense Androgen Study is a population-based study of 783 Caucasian men aged 20–29 yr, living in Funen County, Denmark. The study population is described in detail elsewhere (25). In brief, questionnaires were mailed to 3000 men, randomly drawn from the Danish Central Personal Registry (Fig. 1Go). The 2042 respondents were invited, of whom 784 were included. One man dropped out. A medical history of hospitalization, chronic disease, medication, alcohol intake, and anabolic steroid abuse was obtained by questionnaire, interview, and from electronic hospital records. A systematic physical examination was performed, including testicular volume estimation using an orchidometer. Waist circumference (WC) was measured between the iliac crest and lower ribs during expiration. A cutoff of 102 cm was used to define obesity (26). Body mass index (BMI) was computed from body weight and height, and a BMI ≥ 30 kg/m2 was used as a second definition of obesity. The intraobserver coefficients of variation (CVs) for BMI and WC were 1.2 and 1.7%, respectively. The 783 men matched the county population regarding BMI, chronic disease, medication, physical activity, tobacco exposure, and sociodemography (25). The examinations started in March 2002 and terminated in May 2003. The local Ethic Review Board approved the study (No. 20010198), which was conducted according to the Declaration of Helsinki.


Figure 1
View larger version (25K):
[in this window]
[in a new window]

  		FIG. 1. Inclusion of the study population, establishment of the low-risk reference population, and division into healthy nonobese and obese men.

 
Low-risk reference population

We attempted to select a young, male reference population at low risk for secondary androgen deficiency (n = 685) by ruling out subjects meeting the following prespecified criteria: history of bilateral cryptorchidism or congenital hydrocele, small testes (both < 9 ml), chronic disease, average alcohol intake 6 U/d or greater, anabolic steroid abuse, hypogonadotropism (serum LH < 1.0 U/liter), and subclinical hypothyroidism (serum TSH > 6.0 mU/liter). Chronic disease was subdivided into no medication or continuous systemic medication, including high doses of inhaled corticosteroids (budesonide ≥ 800 µg/d and fluticasone ≥ 1000 µg/d). The rule out was performed before sex hormone assessments. The low-risk reference population was partitioned into obese and nonobese men.

Biological within-subject variation

The biological within-subject variation was determined for all serum measurements in 20 men participating twice within 3–6 wk. Their serum samples were analyzed in the same assay. Considering the analytical variation (CVINTRAASSAY), the biological within-subject variation (CVBIOLOGIC) was derived from the total within-subject variation (CVTOTAL-WITHIN) by the formula: (CVBIOLOGIC)2 = (CVTOTAL-WITHIN)2 – (CVINTRAASSAY)2.

Biochemistry

Fasting, venous blood drawn between 0800 and 1000 h was centrifuged, and serum aliquots were stored at –80 C. Total testosterone, {Delta}4AD, and DHT were measured using an in-house RIA (27, 28) using extraction and chromatography (intraassay CVs: 8.2, 9.4, and 9.1%; interassay CVs: 13.8, 11.4, and 11.0%; CVBIOLOGIC: 15.8, 11.9, and 12.1%, respectively). The accuracy of the total testosterone assay is continuously monitored in an external quality assessment program (German Society of Clinical Chemistry). The mean bias estimated from 20 independent control samples was +5.4% [95% confidence interval (CI) +1.2 to +9.7%]. E2 was measured using an in-house RIA (27, 28) after extraction and chromatography (intraassay, interassay, and CVBIOLOGIC: 7.4, 10.5, and 16.8%, respectively). In 48 men, E2 assessments below the limit of detection (40 pmol/liter) were substituted by values 1 U below the detection limit (29, 30).

SHBG was measured by an immunoluminometric assay (Immulite 2000; DPC, Los Angeles, CA; intraassay, interassay, and CVBIOLOGIC: 3.0, 5.0, and 8.8%, respectively) and albumin by the bromine-cresol-green method (Roche/Hitachi-917; Roche Diagnostics, Indianapolis, IN; intraassay, interassay, and CVBIOLOGIC: 0.7, 2.0, and 2.6%, respectively). LH and TSH were assessed by immunofluorometric assays (DELFIA, Wallac, Finland). Intraassay, interassay, and CVBIOLOGIC were 5, 6, 26.4%, and 3.0, 5.0, 27.9%, respectively.

Bioavailable testosterone (BT), free testosterone (FT), and bioavailable E2 (BE2) were calculated from validated formulas (31):

Formula
where A = kSHBG · (kalbumin · albumin + 1), B = kSHBG · (SHBG – TT) + kalbumin · albumin + 1, and C = –TT. BT was calculated by: BT = (FT + kalbumin · albumin · FT)/(1 + kalbumin · FT).

The association constants were: kSHBG: 1.0 · 109 liter/mol, kalbumin: 3.6 · 104 liter/mol (32). Bioavailable E2 was calculated substituting total testosterone (TT), FT, and BT in the formulas with E2, free E2, and bioavailable E2 and association constants for kSHBG,E2 of 0.68 · 109 liter/mol and kalbumin,E2 of 6.0 · 104 liter/mol (33).

Clinical and paraclinical fat parameters

CFM and lower extremity fat mass (LEFM) were measured by dual-energy x-ray absorptiometry (DXA) using a Hologic4500A densitometer (Waltham, MA). The CVTOTAL-WITHIN for CFM and LEFM between two scans were 5.8 and 4.0%, respectively. Magnetic resonance imaging (MRI) was performed in the first 406 consecutively included subjects with an open, low-field (0.2 Tesla) MR unit (Magnetom Open Viva; Siemens AG, Erlangen, Germany). Three abdominal slices (10-mm thick, 20-mm apart, lower slice at the dorsal, intervertebral space of L4/L5) were recorded using an axial, T1-weighted gradient-echo sequence (repetition time: 450 msec, echo time: 15 msec, acquisition matrix: 512 x 288, field of view: 400 mm). A bias correction algorithm was developed to ensure uniform pixel intensities of adipose tissue throughout all images (34). After subtraction of perivertebral and bone marrow fat, the area of total abdominal adipose tissue was assessed from bimodal histograms discriminating between adipose and nonadipose tissue (Adobe Photoshop 7.0; Adobe Systems, Inc., San Jose, CA). The visceral compartment was demarcated, and the visceral adipose tissue area was quantified. Subcutaneous adipose tissue was computed subtracting visceral adipose tissue from total abdominal adipose tissue. Finally, the areas of the three slices were integrated into volumes. The measurements were initiated when the intraobserver CV for visceral adipose tissue in repeated pilot determinations of 51 images was less than 10%. The intraobserver CVs for total abdominal adipose tissue, sc adipose tissue, and visceral adipose tissue were 3.4, 1.7, and 7.2%, respectively. The correlation coefficient between total abdominal adipose tissue and CFM was 0.964.

Data analysis

Natural logarithm transformations were performed to obtain Gaussian distributions of the serum parameters, and the DXA and MRI parameters. In bar charts, LEFM, CFM, visceral adipose tissue, and sc adipose tissue were categorized by means of SD scores (SDS): less than –1.0; –1.0, –0.5; –0.5, 0.0; 0.0, 0.5; 0.5, 1.0; and ≥1.0 (Table 1Go). Linear regression analysis was used to test for univariate, linear trends between the SDS and hormonal concentrations in the low-risk reference population. The variation in androgen levels (dependent variables) in relation to LEFM vs. CFM was analyzed, including all men (except three anabolic steroid abusers) using multiple regression. These analyses were performed using continuous SDS with adjustment for chronic disease/medication and were repeated for sc adipose tissue vs. visceral adipose tissue. If both independents were significantly correlated with total testosterone, DHT, or E2, additional analyses were performed adjusting for SHBG.


View this table:
[in this window]
[in a new window]

  		TABLE 1. Categorization of CFM, LEFM, visceral adipose tissue (VAT), and sc adipose tissue (SAT)

 
The reference intervals (defined by the 2.5 and 97.5 percentiles) were established by rankit plots (35). The ordinate of the cumulative standard normal distribution was transformed from percentiles to rankits. For each reference interval, the method proposed by Lahti et al. (36) was used to interpret, whether a partitioning of the reference population into obese and nonobese men was advisable. The androgen levels of the ruled-out men were compared with the low-risk reference population using multiple regression analysis to adjust for WC. The odds ratios of having a total testosterone level below the lower reference limit of nonobese men were calculated for the obese men and the ruled-out men by logistic regression. The level of significance was set at P < 0.05. Data were analyzed using Stata Statistics/Data Analysis software version 8.2 (StataCorp, College Station, TX).


   Results
Top
 Abstract
 Introduction
 Subjects and Methods
 Results
Discussion
 References
 
The low-risk reference population comprised 685 men in whom the serum total testosterone reference interval was 11.7–37.9 nmol/liter (95% CIs 11.2–12.1 and 36.3–39.2 nmol/liter, respectively). The reference intervals for total testosterone, bioavailable testosterone, free testosterone, DHT, and {Delta}4AD are shown in Table 2Go. Defining obesity as WC ≥ 102 cm and BMI ≥ 30 kg/m2, the prevalence of obesity in the low-risk reference population was 10.2% (n = 70) and 8.0% (n = 55), respectively. The median WC and BMI in the 685 men was 88.0 cm (range 68.5–131.0 cm) and 24.2 kg/m2 (range 16.0–39.8 kg/m2), respectively. Within the low-risk reference population, the distributions of all androgens (except {Delta}4AD) were significantly displaced to the left in obese men (WC ≥ 102 cm), while E2 and bioavailable E2 were not (Table 2Go). Partitioning was suggested for total testosterone, bioavailable testosterone, free testosterone, and DHT (partitioning is reasonable, when the decision for one limit is definite and the other ambiguous) (36). Almost identical reference intervals were established defining obesity by BMI (data not shown). There were 16 obese men in the low-risk reference population (22.9%) who had total testosterone levels below the lower limit (12.5 nmol/liter) of nonobese men (odds ratio 10.4; 95% CI 4.8–22.5). Only seven of these 16 men (43.8%) had bioavailable testosterone levels below the lower limit for bioavailable testosterone.


View this table:
[in this window]
[in a new window]

  		TABLE 2. Reference (Ref) intervals before and upon partitioning of the low-risk reference population (pop) into obese and nonobese men

 
A total of 98 subjects were ruled out of the low-risk reference population. Table 3Go shows the serum concentrations of total testosterone, bioavailable testosterone, DHT, E2, and LH in the low-risk reference population and ruled-out subjects. Total testosterone was significantly decreased in all ruled-out groups, except for bilateral cryptorchidism, high alcohol intake, and subclinical hypothyroidism. The unaltered total testosterone in cryptorchidism was accompanied by significantly elevated LH concentrations. Total testosterone concentrations less than 12.5 nmol/liter were seen in four men with small testes (odds ratio 8.5; 95% CI 2.3–31.5) and six men treated with prednisolone, tramadol, valproic acid (two men), lamotrigine, and cimetidine (odds ratio 4.1; 95% CI 1.4–11.7). Extremely low total testosterone, bioavailable testosterone, DHT, and LH were found in two abusers of anabolic steroids. Adjusting for WC, the 98 ruled-out men had significantly reduced total testosterone, bioavailable testosterone, DHT, and E2 compared with the low-risk reference population, and 15 of the 98 men (15.3%) had total testosterone levels less than 12.5 nmol/liter (odds ratio 3.6; 95% CI 1.8–7.4).


View this table:
[in this window]
[in a new window]

  		TABLE 3. Androgen, E2 and LH levels in 20- to 29-yr-old men of the low-risk reference (ref) population, the low-risk nonobese and obese, and in the ruled-out men

 
Within the low-risk reference population, SHBG, total testosterone, bioavailable testosterone, and DHT were inversely correlated with CFM and LEFM in univariate analyses (Fig. 2Go). We found a nonsignificant inverse trend between E2 and CFM and LEFM, but after adjustment for total testosterone, both E2 and bioavailable E2 correlated positively with CFM and bioavailable E2, also with LEFM (Fig. 3Go). In multiple regression analyses, CFM was independently correlated with reduced SHBG, total testosterone, bioavailable testosterone, free testosterone, and DHT, while the adjusted measure of bioavailable E2 correlated positively with CFM. LEFM correlated positively with SHBG (Table 4Go).


Figure 2
View larger version (54K):
[in this window]
[in a new window]

  		FIG. 2. Serum SHBG (A), total testosterone (B), bioavailable testosterone (C), DHT (D), and LH (E) in 20- to 29-yr-old healthy men in relation to four fat depots: CFM assessed by DXA [column 1 (n = 685)]; LEFM assessed by DXA [column 2 (n = 685)]; visceral adipose tissue assessed by MRI [column 3 (n = 363)]; and sc adipose tissue assessed by MRI [column 4 (n = 363)]. P values in bar charts are results from univariate linear regression analyses. {ddagger}, Omitting men with SDS ≥1.0 from analysis. *, P < 0.05; **, P < 0.01; ***, P < 0.001; and ****, P < 0.0001 vs. men with SDS less than –1.0 (Student’s t test). §, P = 0.006 vs. the preceding group (Student’s t test). Whiskers represent SEM.

 

Figure 3
View larger version (37K):
[in this window]
[in a new window]

  		FIG. 3. Serum E2 (A), E2 adjusted for total testosterone (B), bioavailable E2 (C), and bioavailable E2 adjusted for total testosterone (D) in 20- to 29-yr-old healthy men in relation to four fat depots: CFM assessed by DXA [column 1 (n = 685)]; LEFM assessed by DXA [column 2 (n = 685)]; visceral adipose tissue assessed by MRI [column 3 (n = 363)]; and sc adipose tissue assessed by MRI [column 4 (n = 363)]. P values in bar charts are results from univariate linear regression analyses. *, P < 0.05; and **, P < 0.01 vs. men with SDS less than –1.0 (Student’s t test). Whiskers represent SEM.

 

View this table:
[in this window]
[in a new window]

  		TABLE 4. CFM and LEFM (upper panel) and visceral adipose tissue (VAT) and sc adipose tissue (SAT) (lower panel) as independent contributors to the variation in SHBG and sex hormone levels in multiple linear regression analyses

 
The 406 men examined by MRI comprised 363 subjects within the low-risk reference population. In univariate analyses, inverse relationships were found between SHBG, total testosterone, bioavailable testosterone, DHT, E2, and the volumes of both visceral adipose tissue and sc adipose tissue (Figs. 2Go and 3Go). Multiple regressions (Table 4Go) demonstrated that: visceral adipose tissue was independently related with reduced bioavailable testosterone and free testosterone, while sc adipose tissue was not; sc adipose tissue was independently related with reduced SHBG, while visceral adipose tissue was not; both visceral adipose tissue and sc adipose tissue were negatively correlated with total testosterone and DHT, but only visceral adipose tissue attained significance when adjusted for SHBG. Finally, no significant relationships were observed with E2 and bioavailable E2, but sc adipose tissue was an independent correlate of the adjusted measure of bioavailable E2.

LH increased linearly with increasing visceral adipose tissue (P = 0.004) when omitting the most viscerally obese men (Fig. 3Go), and was significantly elevated in the group of men with the second largest visceral adipose tissue volumes compared with the leanest (P = 0.02) and to the utmost viscerally obese men (P = 0.006). There were no relations between LH and sc adipose tissue, CFM, or LEFM.

The study population was included over 1 yr: total testosterone and bioavailable testosterone were 1.7 and 1.0 nmol/liter higher from April-September vs. October-March (P < 0.01). Bioavailable testosterone declined by an average 0.17 nmol/liter/yr in these 20- to 29-yr-old men (P < 0.0001), whereas visceral adipose tissue and sc adipose tissue increased by 2.4 and 3.5 cm2/yr (mean of the three slices, P < 106 and P = 0.017, respectively). These findings did not change any relationships reported previously. The hour of blood sampling did not significantly influence the concentrations.


   Discussion
Top
 Abstract
 Introduction
 Subjects and Methods
 Results
 Discussion
References
 
Our population-based data obtained in healthy young men demonstrate a linear, inverse relationship between CFM and total testosterone, and also between CFM and bioavailable testosterone, free testosterone, and DHT. The use of abdominal MRI in a large number of consecutively included subjects showed that the decreased bioavailable testosterone and free testosterone were attributable to visceral adipose tissue. This is in line with the observations by Seidell et al. (4) in 23 healthy 25- to 50-yr-old men, whereas no relation between free testosterone and visceral adipose tissue was found in a study of 17 obese men aged 25–69 yr (11). We are not aware of any large, population-based studies of young men that examined these issues with actual measures of visceral adipose tissue and sc adipose tissue. Couillard et al. (1) used computerized tomography to examine the relationship between visceral adipose tissue, sc adipose tissue, and androgens in 217 men aged 17–64 yr. They did not include measures of free testosterone or bioavailable testosterone but found that total testosterone correlated inversely with sc adipose tissue, but not with visceral adipose tissue.

Our data obtained in relatively lean young men [BMI ≥ 35 kg/m2 (n = 6) and BMI ≥ 40 kg/m2 (none)] demonstrate that reduced bioavailable testosterone and free testosterone are not restricted to massively obese men. Thus, reduced total testosterone in obese men is not uniquely secondary to declines in SHBG levels. However, we did find that part of the decline in total testosterone could be accounted for by declining SHBG levels observed with increasing sc adipose tissue. This is in agreement with the study by Couillard et al. (1). In studies of 40- to 70-yr-old men (37, 38), testosterone as well as SHBG has been linked with the metabolic syndrome and excess visceral adipose tissue. The increase in visceral adipose tissue by age (1) may result in diverging relationships in younger vs. older cohorts. We demonstrated that this increase in visceral adipose tissue was highly significant and well under way from age 20–29 yr. The positive correlations between CFM and the total testosterone adjusted measures of E2 and bioavailable E2 are anticipated signs of increased aromatase activity in obese men. The independent relationship between sc adipose tissue and bioavailable E2 adjusted for total testosterone is in line with previous findings (39). The lack of relationship with the unadjusted measures may be explained by reduced levels of substrate (testosterone) in the obese men and a lower aromatase activity in young compared with older men (39), in whom E2 increases with increasing adiposity (40, 41). Besides variations in substrate availability and aromatase activity, the decline in SHBG in obesity may also subject a higher proportion of non-SHBG bound androgens to metabolism. Schneider et al. (41) found an increased hormonal clearance of androgens to estrogens in obese men. However, the extraglandular formation of E2 by aromatization from total testosterone constitutes less than 1% (42) of the approximately 5- to 6-mg total testosterone produced daily by the testes in normal men. Therefore, it is less likely that increased metabolic clearance rates in the presence of normal production rates can account for the inverse relation between adiposity and androgen levels.

Our observation of increased LH and concomitant declines in total testosterone and bioavailable testosterone with increasing visceral adipose tissue may suggest an intact function of the hypothalamic-pituitary feedback system in response to declining Leydig cell function. The abrupt drop in LH levels in the most viscerally obese subjects may reflect a sudden incapacity of the pituitary to keep up its LH secretion. Our observation is in line with previous studies reporting an attenuation of LH pulse amplitude in severely obese men (2, 10). The identification of a possible link between visceral obesity and decreased LH levels would be a significant discovery. Free fatty acids have been linked with inhibition of another anterior pituitary hormone, GH (43, 44). The enrollment of more massively obese men could possibly have provided further evidence for a biphasic relationship, but the population-based design of the study opposed such a selection. At the other end, only three men were underweight (BMI < 18.5 kg/m2), so the parabolic relationship was not due to enrollment of underweight or anorectic subjects, in whom hypogonadotropic hypogonadism is common (45). In addition to attenuation of LH pulse amplitudes and aromatization of androgens to estrogens, Isidori et al. (46) have proposed that increased leptin levels in obese men may inhibit Leydig cell function.

Our partitioning of the reference population into obese and nonobese subjects produced significantly different reference intervals for total testosterone, bioavailable testosterone, free testosterone, and DHT. This has not been shown previously. We suggest that reference intervals for these androgens should be established in nonobese men because nearly one in every four obese men had total testosterone levels below the reference limit of nonobese men. However, the diagnosis of androgen deficiency in obese men necessitates additional measures of either free testosterone or bioavailable testosterone because only every second obese man with low total testosterone had bioavailable testosterone levels below the lower limit of bioavailable testosterone in nonobese men. This study is the first to present statistical reference intervals for total testosterone, bioavailable testosterone, free testosterone, and DHT in obese, but otherwise healthy, young men. We found that total testosterone, bioavailable testosterone, and free testosterone levels as low as 8.5, 6.1, and 0.23 nmol/liter, respectively, are within the statistically expected range in these men. These limits may be useful to discriminate between severely hypogonadal obese men and men with subnormal levels secondary to the obesity itself. A reversibility of the latter condition has been documented by Niskanen et al. (3), who showed that weight loss and subsequent long-term weight maintenance in obese men normalized hypogonadal total testosterone and bioavailable testosterone levels in 70 and 50%, respectively. However, obese men with low androgen levels may benefit from androgen replacement therapy because the relation between androgen levels and obesity is probably bidirectional (6, 47). Woodhouse et al. (6) found a linear relationship between adiposity and the dose of testosterone administered to young subjects who had their endogenous androgen production blocked. The subjects receiving the lowest dose experienced a drop in total testosterone from mid-normal to hypogonadal levels with a concomitant increase in total, central, and peripheral fat mass. A similar pattern was observed in a parallel study of older men (48). Moreover, testosterone and DHT inhibit the differentiation of mesenchymal pluripotent stem cells toward the adipogenic lineage and stimulate the commitment into the myogenic lineage via the androgen receptor. In the recently published clinical guidelines from The Endocrine Society (12), increased body fat is listed only among the less specific signs associated with androgen deficiency. So far, there is not sufficient evidence to support androgen replacement therapy in obese but otherwise asymptomatic men with low total testosterone.

Future studies combining clinical and paraclinical signs and symptoms of androgen deficiency with androgen measurements may evaluate whether the reference limits of obese men in our study are applicable as decision limits for androgen replacement therapy in obese men.

Of the ruled-out subjects, a high risk of low total testosterone was seen in men with small testes, the medically treated men, and in anabolic steroid abusers. The difference in androgen levels between medically treated and untreated chronic diseases may be explained by more serious disease in men requiring medication but may also be related to the treatment (15, 19, 49, 50, 51). The majority of the ruled-out groups had significantly decreased total testosterone, indicating that the rule-out procedure was reasonable, overall. The concept of establishing a reference interval in a "low risk" population has previously been used by Jorgensen et al. (52).

The assays used in this study were optimized to ensure complete dissociation of the sex hormones from binding proteins (by extraction) and to avoid cross reactivity with other steroids (by celite chromatography). To focus on the relation between androgen levels and adiposity, a young population with a narrow age interval was chosen to minimize the effects of age on androgen levels and binding proteins. Possible influences of comorbidity are limited in younger men, and efforts were taken to rule out conditions that could possibly influence both androgens and body composition (testicular pathology, chronic diseases, abuse of anabolic steroids, etc.). We have previously shown (25) that our cohort matched the background population of young Danish men. Several imaging techniques and anthropometric measures are used to study the apparent negative and positive metabolic consequences of CFM and LEFM, respectively (53, 54). We used DXA to measure CFM and LEFM, and found that reduced SHBG and androgen levels with increasing adiposity were attributable to CFM. When analyzed for independent relations with the fat types actually constituting the CFM, namely visceral adipose tissue and sc adipose tissue, we found that SHBG was related with sc adipose tissue, while bioavailable testosterone and free testosterone were related with visceral adipose tissue. Accordingly, assessment of CFM using DXA (and probably anthropometrics) is not necessarily valuable as a measure of visceral adiposity.

In conclusion, total testosterone, bioavailable testosterone, free testosterone, DHT, and SHBG decline linearly with increasing CFM in young men. Visceral adipose tissues independently account for the decline in bioavailable testosterone and free testosterone, while the decline in total testosterone and DHT with increased sc adipose tissue was secondary to decreased SHBG levels. E2 adjusted for total testosterone is increased with increasing CFM, but increased metabolic clearance can hardly explain the inverse relation between androgen levels and adiposity, which is possibly related with decreased secretion of androgens of either primary or secondary origin. The reference intervals for total testosterone, bioavailable testosterone, free testosterone, and DHT were displaced toward lower limits in obese men, suggesting that reference intervals for these androgens should be established in low-risk nonobese men.


   Footnotes
 
Disclosure Summary: T.L.N., K.W., P.H.P., E.H., and R.L. have nothing to declare. K.B. consults for Eli Lilly, Servier, Nycomed, and Merck Sharp & Dohme. C.H., K.B., and M.A. have received lecture fees less than U.S. $10,000 in the past 2 yr. C.H. received industrial entities for this publication by Novo Nordic and Pfizer. K.B. did not receive entities for this publication but has received a grant from Merck Sharp & Dohme for other studies.

First Published Online April 10, 2007

Abbreviations: {Delta}4AD, Androstenedione; BMI, body mass index; CFM, central fat mass; CI, confidence interval; CV, coefficient of variation; DHT, dihydrotestosterone; DXA, dual-energy x-ray absorptiometry; E2, estradiol; LEFM, lower extremity fat mass; MRI, magnetic resonance imaging; SDS, SD score; WC, waist circumference.

Received August 22, 2006.

Accepted April 2, 2007.


   References
Top
 Abstract
 Introduction
 Subjects and Methods
 Results
 Discussion
 References
 

  1. Couillard C, Gagnon J, Bergeron J, Leon AS, Rao DC, Skinner JS, Wilmore JH, Despres JP, Bouchard C 2000 Contribution of body fatness and adipose tissue distribution to the age variation in plasma steroid hormone concentrations in men: the HERITAGE Family Study. J Clin Endocrinol Metab 85:1026–1031[Abstract/Free Full Text]
  2. Giagulli VA, Kaufman JM, Vermeulen A 1994 Pathogenesis of the decreased androgen levels in obese men. J Clin Endocrinol Metab 79:997–1000[Abstract]
  3. Niskanen L, Laaksonen DE, Punnonen K, Mustajoki P, Kaukua J, Rissanen A 2004 Changes in sex hormone-binding globulin and testosterone during weight loss and weight maintenance in abdominally obese men with the metabolic syndrome. Diabetes Obes Metab 6:208–215[CrossRef][Medline]
  4. Seidell JC, Bjorntorp P, Sjostrom L, Kvist H, Sannerstedt R 1990 Visceral fat accumulation in men is positively associated with insulin, glucose, and C-peptide levels, but negatively with testosterone levels. Metabolism 39:897–901[CrossRef][Medline]
  5. Simon D, Charles MA, Nahoul K, Orssaud G, Kremski J, Hully V, Joubert E, Papoz L, Eschwege E 1997 Association between plasma total testosterone and cardiovascular risk factors in healthy adult men: The Telecom Study. J Clin Endocrinol Metab 82:682–685[Abstract/Free Full Text]
  6. Woodhouse LJ, Gupta N, Bhasin M, Singh AB, Ross R, Phillips J, Bhasin S 2004 Dose-dependent effects of testosterone on regional adipose tissue distribution in healthy young men. J Clin Endocrinol Metab 89:718–726[Abstract/Free Full Text]
  7. Zumoff B, Strain GW, Miller LK, Rosner W, Senie R, Seres DS, Rosenfeld RS 1990 Plasma free and non-sex-hormone-binding-globulin-bound testosterone are decreased in obese men in proportion to their degree of obesity. J Clin Endocrinol Metab 71:929–931[Abstract]
  8. Glass AR, Swerdloff RS, Bray GA, Dahms WT, Atkinson RL 1977 Low serum testosterone and sex-hormone-binding-globulin in massively obese men. J Clin Endocrinol Metab 45:1211–1219[Medline]
  9. Stanik S, Dornfeld LP, Maxwell MH, Viosca SP, Korenman SG 1981 The effect of weight loss on reproductive hormones in obese men. J Clin Endocrinol Metab 53:828–832[Abstract]
  10. Vermeulen A, Kaufman JM, Deslypere JP, Thomas G 1993 Attenuated luteinizing hormone (LH) pulse amplitude but normal LH pulse frequency, and its relation to plasma androgens in hypogonadism of obese men. J Clin Endocrinol Metab 76:1140–1146[Abstract]
  11. Rissanen J, Hudson R, Ross R 1994 Visceral adiposity, androgens, and plasma lipids in obese men. Metabolism 43:1318–1323[CrossRef][Medline]
  12. Bhasin S, Cunningham GR, Hayes FJ, Matsumoto AM, Snyder PJ, Swerdloff RS, Montori VM 2006 Testosterone therapy in adult men with androgen deficiency syndromes: an Endocrine Society clinical practice guideline. J Clin Endocrinol Metab 91:1995–2010[Abstract/Free Full Text]
  13. Barrett-Connor E, Bhasin S 2004 Time for (more research on) testosterone. J Clin Endocrinol Metab 89:501–502[Free Full Text]
  14. Handelsman DJ 1994 Testicular dysfunction in systemic disease. Endocrinol Metab Clin North Am 23:839–856[Medline]
  15. Brunet M, Rodamilans M, Martinezosaba MJ, Santamaria J, Tofigueras J, Torra M, Corbella J, Rivera F 1995 Effects of long-term antiepileptic therapy on the catabolism of testosterone. Pharmacol Toxicol 76:371–375[Medline]
  16. Lindquist M, Edwards IR 1992 Endocrine adverse-effects of omeprazole. BMJ 305:451–452[Medline]
  17. Rajfer J, Sikka SC, Rivera F, Handelsman DJ 1986 Mechanism of inhibition of human testicular steroidogenesis by oral ketoconazole. J Clin Endocrinol Metab 63:1193–1198[Abstract]
  18. Loriaux DL, Menard R, Taylor A, Pita JC, Santen R 1976 Spironolactone and endocrine dysfunction. Ann Intern Med 85:630–636[Medline]
  19. Vanderpump MP, Tunbridge WM 1993 The effects of drugs on endocrine function. Clin Endocrinol (Oxf) 39:389–397[Medline]
  20. Cicero TJ 1982 Alcohol-induced deficits in the hypothalamic-pituitary-luteinizing hormone axis in the male. Alcohol Clin Exp Res 6:207–215[Medline]
  21. Taieb J, Mathian B, Millot F, Patricot MC, Mathieu E, Queyrel N, Lacroix I, Somma-Delpero C, Boudou P 2003 Testosterone measured by 10 immunoassays and by isotope-dilution gas chromatography-mass spectrometry in sera from 116 men, women, and children. Clin Chem 49:1381–1395[Abstract/Free Full Text]
  22. Matsumoto AM, Bremner WJ 2004 Serum testosterone assays–accuracy matters. J Clin Endocrinol Metab 89:520–524[Free Full Text]
  23. Wang C, Catlin DH, Demers LM, Starcevic B, Swerdloff RS 2004 Measurement of total serum testosterone in adult men: comparison of current laboratory methods versus liquid chromatography-tandem mass spectrometry. J Clin Endocrinol Metab 89:534–543[Abstract/Free Full Text]
  24. Herold DA, Fitzgerald RL 2003 Immunoassays for testosterone in women: better than a guess? Clin Chem 49:1250–1251[Free Full Text]
  25. Nielsen TL, Wraae K, Brixen K, Hermann AP, Andersen M, Hagen C 2006 Prevalence of overweight, obesity and physical inactivity in 20- to 29-year-old, Danish men. Relation to sociodemography, physical dysfunction and low socioeconomic status: the Odense Androgen Study. Int J Obes (Lond) 30:805–815[CrossRef][Medline]
  26. Lean ME, Han TS, Morrison CE 1995 Waist circumference as a measure for indicating need for weight management. BMJ 311:158–161[Abstract/Free Full Text]
  27. Parker Jr CR, Ellegood JO, Mahesh VB 1975 Methods for multiple steroid radioimmunoassay. J Steroid Biochem 6:1–8[Medline]
  28. Lykkesfeldt G, Bennett P, Lykkesfeldt AE, Micic S, Moller S, Svenstrup B 1985 Abnormal androgen and oestrogen metabolism in men with steroid sulphatase deficiency and recessive X-linked ichthyosis. Clin Endocrinol (Oxf) 23:385–393[Medline]
  29. Greendale GA, Edelstein S, Barrett-Connor E 1997 Endogenous sex steroids and bone mineral density in older women and men: the Rancho Bernardo Study. J Bone Miner Res 12:1833–1843[CrossRef][Medline]
  30. Greendale GA, Palla SL, Ursin G, Laughlin GA, Crandall C, Pike MC, Reboussin BA 2005 The association of endogenous sex steroids and sex steroid binding proteins with mammographic density: results from the Postmenopausal Estrogen/Progestin Interventions Mammographic Density Study. Am J Epidemiol 162:826–834[Abstract/Free Full Text]
  31. Vermeulen A, Verdonck L, Kaufman JM 1999 A critical evaluation of simple methods for the estimation of free testosterone in serum. J Clin Endocrinol Metab 84:3666–3672[Abstract/Free Full Text]
  32. Moll Jr GW, Rosenfield RL, Helke JH 1981 Estradiol-testosterone binding interactions and free plasma estradiol under physiological conditions. J Clin Endocrinol Metab 52:868–874[Medline]
  33. Endogenous Hormones and Breast Cancer Collaborative Group 2003 Free estradiol and breast cancer risk in postmenopausal women: comparison of measured and calculated values. Cancer Epidemiol Biomarkers Prev 12:1457–1461[Abstract/Free Full Text]
  34. Engholm R, Dubinskiy A, Larsen R, Hanson LG, Christoffersen BO, An adipose segmentation and quantification scheme for the intra abdominal region on minipigs. Proc International Symposium on Medical Imaging by The International Society for Optical Engineering (SPIE), San Diego, CA, 2006, 6144 II (Abstract)
  35. Petersen PH, Blaabjerg O, Andersen M, Jorgensen LGM, Schousboe K, Jensen E 2004 Graphical interpretation of confidence curves in rankit plots. Clin Chem Lab Med 42:715–724[CrossRef][Medline]
  36. Lahti A, Hyltoft PP, Boyd JC, Fraser CG, Jorgensen N 2002 Objective criteria for partitioning Gaussian-distributed reference values into subgroups. Clin Chem 48:338–352[Abstract/Free Full Text]
  37. Kupelian V, Page ST, Araujo AB, Travison TG, Bremner WJ, McKinlay JB 2006 Low sex hormone-binding globulin, total testosterone, and symptomatic androgen deficiency are associated with development of the metabolic syndrome in nonobese men. J Clin Endocrinol Metab 91:843–850[Abstract/Free Full Text]
  38. Laaksonen DE, Niskanen L, Punnonen K, Nyyssonen K, Tuomainen TP, Salonen R, Rauramaa R, Salonen JT 2003 Sex hormones, inflammation and the metabolic syndrome: a population-based study. Eur J Endocrinol 149:601–608[Abstract]
  39. Vermeulen A, Kaufman JM, Goemaere S, van Pottelberg I 2002 Estradiol in elderly men. Aging Male 5:98–102[Medline]
  40. Vermeulen A, Kaufman JM, Giagulli VA 1996 Influence of some biological indexes on sex hormone-binding globulin and androgen levels in aging or obese males. J Clin Endocrinol Metab 81:1821–1826[Abstract]
  41. Schneider G, Kirschner MA, Berkowitz R, Ertel NH 1979 Increased estrogen production in obese men. J Clin Endocrinol Metab 48:633–638[Abstract]
  42. MacDonald PC, Madden JD, Brenner PF, Wilson JD, Siiteri PK 1979 Origin of estrogen in normal men and in women with testicular feminization. J Clin Endocrinol Metab 49:905–916[Abstract]
  43. Lipman RL, Taylor AL, Schenk A, Mintz DH 1972 Inhibition of sleep related growth hormone release by elevated free fatty acids. J Clin Endocrinol Metab 35:592–594[Medline]
  44. Tsushima T, Sakuma M, Irie M 1970 Effect of changes in plasma free fatty acids level on secretion of human growth hormone. Endocrinol Jpn 17:369–377[Medline]
  45. Wheeler MJ, Crisp AH, Hsu LK, Chen CN 1983 Reproductive hormone changes during weight gain in male anorectics. Clin Endocrinol (Oxf) 18:423–429[Medline]
  46. Isidori AM, Caprio M, Strollo F, Moretti C, Frajese G, Isidori A, Fabbri A 1999 Leptin and androgens in male obesity: evidence for leptin contribution to reduced androgen levels. J Clin Endocrinol Metab 84:3673–3680[Abstract/Free Full Text]
  47. Kaufman JM, Vermeulen A 2005 The decline of androgen levels in elderly men and its clinical and therapeutic implications. Endocr Rev 26:833–876[Abstract/Free Full Text]
  48. Bhasin S, Woodhouse L, Casaburi R, Singh AB, Mac RP, Lee M, Yarasheski KE, Sinha-Hikim I, Dzekov C, Dzekov J, Magliano L, Storer TW 2005 Older men are as responsive as young men to the anabolic effects of graded doses of testosterone on the skeletal muscle. J Clin Endocrinol Metab 90:678–688[Abstract/Free Full Text]
  49. Kamischke A, Kemper DE, Castel MA, Luthke M, Rolf C, Behre HM, Magnussen H, Nieschlag E 1998 Testosterone levels in men with chronic obstructive pulmonary disease with or without glucocorticoid therapy. Eur Respir J 11:41–45[Abstract/Free Full Text]
  50. Reid IR, Ibbertson HK, France JT, Pybus J 1985 Plasma testosterone concentrations in asthmatic men treated with glucocorticoids. Br Med J (Clin Res Ed) 291:574
  51. Van Thiel DH, Gavaler JS, Smith Jr WI, Paul G 1979 Hypothalamic-pituitary-gonadal dysfunction in men using cimetidine. N Engl J Med 300:1012–1015[Abstract]
  52. Jorgensen LG, Stahl M, Brandslund I, Hyltoft PP, Borch-Johnsen K, de Fine ON 2001 Plasma glucose reference interval in a low-risk population. Impact of the new WHO and ADA recommendations on the diagnosis of diabetes mellitus. Scand J Clin Lab Invest 61:181–190[CrossRef][Medline]
  53. Buemann B, Sorensen TI, Pedersen O, Black E, Holst C, Toubro S, Echwald S, Holst JJ, Rasmussen C, Astrup A 2005 Lower-body fat mass as an independent marker of insulin sensitivity–the role of adiponectin. Int J Obes (Lond) 29:624–631[CrossRef][Medline]
  54. Tanko LB, Bruun JM, Alexandersen P, Bagger YZ, Richelsen B, Christiansen C, Larsen PJ 2004 Novel associations between bioavailable estradiol and adipokines in elderly women with different phenotypes of obesity: implications for atherogenesis. Circulation 110:2246–2252[Abstract/Free Full Text]




This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
92/7/2696    most recent
Author Manuscript (PDF)
Right arrow Submit a related Letter to the Editor
Right arrow Purchase Article
Right arrow View Shopping Cart
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Copyright Permission
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Nielsen, T. L.
Right arrow Articles by Andersen, M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Nielsen, T. L.
Right arrow Articles by Andersen, M.
Related Collections
Right arrow Male Endocrinology
Right arrow Metabolism

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Endocrinology Endocrine Reviews J. Clin. End. & Metab.
Molecular Endocrinology Recent Prog. Horm. Res. All Endocrine Journals