Abstract
Purpose
To evaluate the association of body size – captured via whole body dual-energy x-ray absorptiometry (DXA) and physical measurement – with serum sex steroid hormones and sex hormone binding globulin (SHBG) we utilized cross-sectional data and serum samples from the National Health and Nutrition Examination Survey (NHANES; 1999-2004).
Methods
Testosterone, androstanediol glucuronide (3-alpha-diol-G), estradiol and SHBG were measured via immunoassay in serum samples from a total of 898 adult men (ages 20-90) participating in the morning examination. As part of the NHANES data collection DXA scans and measurements of weight, height and waist circumference were performed by trained staff. Linear regression was used to estimate associations between body size and hormone levels adjusted for potential confounders and NHANES sampling procedures.
Results
Total bone area (cm2) was inversely associated with total testosterone (ng/mL) [beta=-0.12; p-value<0.01], while bone mineral density (g/cm2) was inversely associated with SHBG (nmol/L) [beta=-17.16; p-value=0.01]. Increased percent body fat was associated with lower concentrations of total testosterone [beta=-0.16; p-value<0.01] and SHBG [beta=-1.11; p-value<0.01] and higher concentrations of free estradiol (fg/mL) [beta=12.52; p-value<0.01].
Conclusions
Clinical measures of body fat (measured via DXA scan) and anthropometric measures of body fat (BMI and waist circumference) provided similar inferences regarding the association between increased body fat and hormone levels in men. Increased body fat was associated with lower circulating levels of testosterone (total and free) and SHBG and higher circulating levels of free estradiol in men, while decreased bone mineral density was associated with higher circulating levels of SHBG.
Keywords: dual-energy X-ray absorptiometry, DXA, estradiol, testosterone, androstanediol glucuronide, sex hormone binding globulin, National Health and Nutrition Examination Survey, NHANES, men
Overweight and obesity are associated with increased risk of some cancers [1;2]. It is hypothesized that some of this risk may be partially explained by obesity’s influence on sex steroid hormone concentrations, in men [3]. Further, the association between body mass index (BMI) and sex steroid hormones has been characterized in a number of studies [4-8]. Bone mineral density, another component of body size, has also been correlated with sex steroid concentrations [9;10]; however, the associations between hormone concentrations and bone mineral content, bone area, as well as the components of body size, including total mass, lean mass, fat mass and adult stature, are not well understood.
Dual-energy x-ray absorptiometry (DXA) has become one of the most widely accepted methods of measuring soft tissue and bone composition. In 1999 whole-body DXA scans were added to the National Health and Nutrition Examination Survey (NHANES). Data on bone mineral content, bone mineral density, bone area, total mass, fat mass, lean mass and percent body fat are provided by the DXA scan. The aim of this study was to determine whether clinical and anthropometric body size measurements were associated with serum hormone (testosterone, estradiol and androstanediol glucuronide (3-alpha-diol-G)) and sex-hormone binding globulin (SHBG) levels within a representative sample of US men aged 20-99 years. SHBG is the major carrier of testosterone and estradiol in circulation and was used to calculate levels of free testosterone and free estradiol; 3-alpha-diol-G, a major metabolite of dihydroxytestosterone, is a correlate of tissue androgen levels [11].
SUBJECTS AND METHODS
Study population
NHANES is a series of studies conducted by the National Center for Health Statistics, Centers for Disease Control and Prevention to assess health in the US population through an interview in the household followed by an examination in a mobile examination center [12]. Beginning in 1999, NHANES became a continuous survey (reported in 2-year data cycles). In order to select a representative sample of the civilian, non-institutionalized US population, participant sampling followed a multi-stage, stratified, probability-cluster design. The survey design oversampled adolescents, older Americans, Mexican Americans and non-Hispanic blacks to ensure adequate sample sizes for sub-analysis. The examination component consisted of medical, dental and physiological measurements administered by medical personnel. Blood samples were also collected during the examination. A limited amount of surplus serum, urine, and plasma samples are available from NHANES participants who consented for their specimens to be used for future studies. Access to these surplus samples are allowed to investigators through a formal research proposal submission and approval process. The study presented here, utilizing surplus sera for hormone measurements, underwent scientific and institutional review and was approved by the National Center for Health Statistics Institutional Review Board.
The current study utilizes survey data and sera from study years 1999-2004. During this time there were 15,184 male participants, and these men were randomly assigned to morning or afternoon examinations. Our analysis focused on men aged 20 years and older participating in the morning examination (n=3,281). We focused on the morning examination to control for possible effects of diurnal variation in sex hormone production. Hormone and SHBG levels were tested using serum that had been previously selected for inclusion into the morning organochlorine subsample so the data analysis could take advantage of the sampling and weights already developed for this subsample. A total of 1,085 men who attended the morning exam were in this subsample, and 985 had surplus sera available for hormone measurement. We excluded 76 men because they had ‘0’ values for the sampling weights for the morning fast, indicating that their data would not contribute to the analysis, and 11 individuals because their DXA data were highly variable and therefore not reliable. The resulting analytic sample included data from 898 men.
Hormone and SHBG levels
Serum testosterone, estradiol, 3-alpha-diol-G, and SHBG levels were measured at the Children’s Hospital, Boston, MA. Testosterone, estradiol, and SHBG concentrations were measured using the Elecsys 2010 system (Roche Diagnostics, Laval, QC, Canada). 3-alpha-diol-G was measured by enzyme immunoassay using the Direct 3-alpha-Diol-G ELISA kit (ALPCO Diagnostics, Salem, NH). Laboratory methods and quality control statistics were reported previously [13;13]. A subset of 21 samples was repeated to assess reproducibility; the coefficients of variation (CVs) were 3.4% for testosterone, 4.7% for SHBG, 5.2% for 3-alpha-diol-G, and 15.8% for estradiol.
Using methods described by Vermeulen et al.[14], free testosterone was calculated from testosterone and SHBG values while free estradiol was calculated from estradiol and SHBG values.
Body size measurements
As part of the mobile examination participants underwent whole body DXA using a Hologic QDR 4500A fanbeam densitometer (Hologic Inc, Bedford MA). The scan for each participant was analyzed by the University of California-San Francisco, Department of Radiology, using standard radiologic techniques and study specific protocols developed for NHANES [15]. Our analyses used the five NHANES DXA multiple imputation data files released by the National Center for Health Statistics for study years 1999-2004. Multiple imputation was conducted, in part, because data were not missing completely at random, the details of which are described elsewhere [16].
Anthropometric body size measurements, standing height, weight and waist circumference, were measured by trained study personnel during the examination. BMI was calculated as the ratio of weight (kg) to height squared (m2).
Statistical analysis
Pearson correlation coefficients were calculated to evaluate correlation between the different DXA measurements. Multiple linear regression models were used to estimate adjusted change in hormone concentration associated with increases in body size as measured by 8 body size measurements from the DXA scan, as well as height, BMI, and waist circumference based on physical measurement. Individual models were run for each hormone-body size association adjusting for age, race/ethnicity, smoking status, alcohol consumption and physical activity, as these factors may influence both hormone concentrations and body size.
Effect modification by age and race/ethnicity were evaluated and results were reported overall and stratified by age, as race/ethnicity was not a significant effect modifier. We also examined the hormone-body size association while simultaneously adjusting for the other hormones measured (e.g. total testosterone-BMI association adjusted for SHBG, 3-alpha-diol-G and total estradiol), which facilitated the evaluation of the association with one hormone while holding constant the concentrations of the other hormones.
To account for the complex survey design (including oversampling), survey nonresponse, and post-stratification in NHANES and to obtain estimates that are representative of the civilian non-institutionalized U.S. population, all statistical analyses employed appropriate survey sampling weights [12]. DXA analyses further utilized the five imputed datasets [16]. Statistical significance was determined based on a two-sided p-value less than 0.05. Analyses were conducted using SAS callable SUDAAN v10.0 (RTI International, Research Triangle Park, NC) as implemented in SAS v9.2 (SAS, Cary, NC) to account for the multi-stage sampling procedure utilized in designing NHANES.
RESULTS
Selected demographic and biologic characteristics of the study population weighted by the sampling fraction are shown in Table 1. The weighted median age for the study population was 41.8 years. The race/ethnicity distribution of the study population was 71.6% non-Hispanic white, 10.4% non-Hispanic black, 8.1% Mexican American, and 9.9% men of other races/ethnicities. On average the study participants were 176.6 cm tall, had a BMI of 27.9 kg/m2 and a waist circumference of 99.6 cm. Based on DXA scan measurements the older men had higher total percent body fat than younger men (31.0% vs. 26.4%) and lower total bone mineral content than younger men (2500.5 grams vs. 2756.4 grams). The mean hormone concentrations were 5.1 ng/mL for total testosterone, 105.9 pg/mL for free testosterone, 37.1 nmol/L for SHBG, 8.3 ng/mL for 3-alpha-diol-G, 33.7 pg/mL for total estradiol and 896.9 fg/mL for free estradiol.
Table 1.
Selected demographic and biologic characteristics of adult men in the NHANES survey, 1999-2004.
| All N = 898 |
Age at interview
|
|||||||
|---|---|---|---|---|---|---|---|---|
| 20-44 n = 409 |
45-69 n = 336 |
70 + n = 153 |
||||||
|
| ||||||||
| Median (IQR) | Median (IQR) | Median (IQR) | Median (IQR) | |||||
| Age (years) | 41.8 (30.9-54.3) | 31.8 (25.1-37.1) | 53.3 (48.4-59.3) | 75.2 (71.6-80.5) | ||||
| Race/ethnicity | n | % | n | % | n | % | n | % |
| Non-Hispanic White | 470 | 71.6 | 178 | 65.6 | 186 | 77.2 | 106 | 86.3 |
| Non-Hispanic Black | 161 | 10.4 | 87 | 12.3 | 56 | 8.4 | 18 | 7.1 |
| Mexican American | 200 | 8.1 | 107 | 11.0 | 68 | 4.9 | 25 | 3.1 |
| Other | 67 | 9.9 | 37 | 11.2 | 26 | 9.4 | 4 | 3.6 |
| Cigarette smoking | ||||||||
| Never | 385 | 44.5 | 223 | 54.0 | 112 | 33.5 | 50 | 31.9 |
| Former | 275 | 26.9 | 57 | 15.4 | 127 | 36.8 | 91 | 59.3 |
| Current | 238 | 28.5 | 129 | 30.5 | 97 | 29.7 | 12 | 8.8 |
| Number of alcohol drinks (per week) | ||||||||
| 0 | 262 | 27.7 | 82 | 21.1 | 114 | 34.3 | 66.0 | 40.7 |
| less than 1 | 199 | 22.6 | 94 | 23.5 | 68 | 20.1 | 37.0 | 28.8 |
| 1 to less than 7 | 230 | 28.8 | 137 | 37.0 | 71 | 20.6 | 22.0 | 13.8 |
| 7 + | 171 | 20.8 | 72 | 18.4 | 75 | 25.0 | 24.0 | 16.8 |
| Physical activity (MET-hours per week of moderate to vigorous activity) | ||||||||
| Inactive | 363 | 33.8 | 155 | 33.2 | 122 | 31.4 | 86 | 49.4 |
| < 3 Met-hours/week | 306 | 36.3 | 134 | 35.0 | 126 | 38.8 | 46 | 33.4 |
| 3 + Met-hours/week | 229 | 29.9 | 120 | 31.8 | 88 | 29.9 | 21 | 17.2 |
| Body size measurements | Mean | (SE) | Mean | (SE) | Mean | (SE) | Mean | (SE) |
| From DXA scan | ||||||||
| Total bone area (cm2) | 2278.6 | (9.4) | 2283.2 | (11.8) | 2289.5 | (13.6) | 2192.6 | (17.1) |
| Total bone mineral content (BMC) (g) | 2717.8 | (21.5) | 2756.4 | (27.3) | 2706.7 | (29.6) | 2500.5 | (37.8) |
| Total bone mineral density (g/cm2) | 1.2 | (0.0) | 1.2 | (0.0) | 1.2 | (0.0) | 1.1 | (0.0) |
| Fat mass (g) | 25384.6 | (433.2) | 23843.4 | (615.8) | 27545.2 | (644.2) | 25566.0 | (650.5) |
| Lean mass excluding BMC (g) | 59869.5 | (472.5) | 60223.6 | (614.6) | 60743.4 | (645.2) | 53070.7 | (617.1) |
| Lean mass including BMC (g) | 62587.2 | (489.2) | 62980.0 | (635.7) | 63450.1 | (663.8) | 55571.2 | (643.7) |
| Total mass (lean+fat) (g) | 87971.8 | (869.5) | 86823.4 | (1158.3) | 90995.3 | (1251.2) | 81137.2 | (1098.5) |
| Total percent body fat (%) | 28.0 | (0.2) | 26.4 | (0.4) | 29.5 | (0.3) | 31.0 | (0.5) |
| From physical measurement | ||||||||
| Height (cm) | 176.6 | (0.3) | 176.9 | (0.4) | 176.9 | (0.5) | 172.9 | (0.7) |
| Body mass index (BMI) (kg/m2) | 27.9 | (0.3) | 27.4 | (0.3) | 28.8 | (0.4) | 27.0 | (0.4) |
| Waist circumference (cm) | 99.6 | (0.6) | 96.0 | (0.9) | 104.1 | (0.9) | 102.5 | (1.0) |
| Sex steroid hormone and SHBG concentrations | ||||||||
| Testosterone (ng/mL) | 5.1 | (0.1) | 5.6 | (0.1) | 4.6 | (0.1) | 4.1 | (0.1) |
| Free testosterone (pg/mL) | 105.9 | (2.2) | 126.9 | (2.4) | 86.3 | (2.7) | 55.8 | (1.7) |
| SHBG (nmol/L) | 37.1 | (0.9) | 30.4 | (1.1) | 41.5 | (1.5) | 62.6 | (2.8) |
| 3-alpha-diol-G (ng/mL) | 8.3 | (0.3) | 8.4 | (0.2) | 8.3 | (0.5) | 7.1 | (0.5) |
| Estradiol (pg/mL) | 33.7 | (1.0) | 35.1 | (1.3) | 32.9 | (1.2) | 27.9 | (1.7) |
| Free estradiol (fg/mL) | 896.9 | (31.1) | 971.5 | (39.3) | 845.4 | (39.1) | 629.0 | (48.7) |
IQR = Interquartile range
DXA = Dual X-ray absorptiometry
Percents, means and standard errors have been weighted for NHANES sampling and represent characteristics in the U.S. general population.
The DXA body size measurement correlations are presented in Table 2. Total bone area was positively correlated with total bone mineral content and total bone mineral density (correlation coefficient = 0.89 and 0.64, respectively). As expected, total mass and its components, fat mass and lean mass, were highly correlated (correlation coefficient = 0.91 and 0.92, respectively). Lean mass including bone mineral content and lean mass excluding bone mineral content were perfectly correlated (correlation coefficient=1.00), therefore we only present results for lean mass excluding bone mineral content.
Table 2.
Pearson correlation coefficientsa for the dual-energy X-ray absorptiometry (DXA) measurements, NHANES 1999-2004.
| Total bone area | Total bone mineral content | Total bone mineral density | Fat mass | Lean mass excluding BMC | Lean mass including BMC | Total mass (lean+fat) | Total percent body fat | |
|---|---|---|---|---|---|---|---|---|
| Total bone area (cm2) | 1.00 | |||||||
| Total bone mineral content (BMC) (g) | 0.89 | 1.00 | ||||||
| Total bone mineral density (g/cm2) | 0.64 | 0.91 | 1.00 | |||||
| Fat mass (g) | 0.43 | 0.29 | 0.12 | 1.00 | ||||
| Lean mass excluding BMC (g) | 0.79 | 0.70 | 0.49 | 0.68 | 1.00 | |||
| Lean mass including BMC (g) | 0.80 | 0.72 | 0.52 | 0.68 | 1.00 | 1.00 | ||
| Total mass (lean+fat) (g) | 0.68 | 0.56 | 0.35 | 0.91 | 0.92 | 0.92 | 1.00 | |
| Total percent body fat (%) | 0.09 | -0.04 | -0.15 | 0.87 | 0.29 | 0.28 | 0.62 | 1.00 |
all p-values < 0.001
Hormone associations with DXA measurements
Results from the multiple linear regression models evaluating the association between body size measurements and hormone concentrations are presented in Table 3. Total bone area (in cm2) was inversely associated with total testosterone (ng/mL) [beta (estimated change in testosterone for a 100 cm2 increase in total bone area) = -0.12; p-value < 0.01] and free testosterone (ng/mL) concentrations [beta = -2.81; p-value < 0.01]. Total bone area was not associated with SHBG, 3-alpha-diol-G, total estradiol or free estradiol. Higher total bone mineral content (grams) was associated with lower concentrations of testosterone [beta (estimated change in testosterone for a 100 gram increase in bone mineral content) = -0.04; p-value < 0.01], free testosterone [beta = -0.83; p-value = 0.02] and SHBG [beta = -0.35; p-value = 0.02]. Total bone mineral density (grams/cm2) was inversely associated with SHBG [beta (estimated change in SHBG for a 1 unit increase in bone mineral density) = -17.16; p-value = 0.01], but not associated with the hormones measured.
Table 3.
Adjusted associations between sex steroid hormones, SHBG and body size measurements in the NHANES survey, 1999-2004.
| Total bone area (cm2) per 100 cm2 increase | Total bone mineral content (BMC) (g) per 100 g increase | Total bone mineral density (g/cm2) | Fat mass (g) per 1 kg increase | Lean mass excluding BMC (g) per 1 kg increase | |||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Betaa | SE | p | Betaa | SE | p | Betaa | SE | p | Betaa | SE | p | Betaa | SE | p | |
| Testosterone (ng/mL) | -0.12 | 0.03 | < 0.01 | -0.04 | 0.01 | < 0.01 | -0.97 | 0.65 | 0.14 | -0.09 | 0.01 | < 0.01 | -0.06 | 0.01 | < 0.01 |
| Free testosterone (pg/mL) | -2.81 | 0.57 | < 0.01 | -0.83 | 0.33 | 0.02 | -8.18 | 15.63 | 0.60 | -1.14 | 0.16 | < 0.01 | -0.77 | 0.13 | < 0.01 |
| SHBG (nmol/L) | -0.43 | 0.35 | 0.23 | -0.35 | 0.14 | 0.02 | -17.16 | 6.00 | 0.01 | -0.59 | 0.06 | < 0.01 | -0.47 | 0.06 | < 0.01 |
| 3-alpha-diol-G (ng/mL) | 0.20 | 0.12 | 0.11 | 0.10 | 0.06 | 0.09 | 4.06 | 2.30 | 0.09 | 0.05 | 0.03 | 0.06 | 0.06 | 0.02 | 0.03 |
| Estradiol (pg/mL) | 0.16 | 0.27 | 0.56 | 0.14 | 0.13 | 0.28 | 7.72 | 5.46 | 0.17 | 0.13 | 0.08 | 0.13 | 0.11 | 0.08 | 0.14 |
| Free estradiol (fg/mL) | 4.00 | 8.16 | 0.63 | 4.52 | 3.98 | 0.26 | 265.71 | 171.32 | 0.13 | 7.55 | 2.58 | 0.01 | 6.08 | 2.40 | 0.02 |
| Total mass (lean mass + fat mass) (g) per 1 kg increase | Percent body fat (%) | Standing Height (cm) per 10 cm increase | Body mass index (BMI) (kg/m2) | Waist circumference (cm) | |||||||||||
| Betaa | SE | p | Betaa | SE | p | Betaa | SE | p | Betaa | SE | p | Betaa | SE | p | |
|
| |||||||||||||||
| Testosterone (ng/mL) | -0.04 | <0.01 | < 0.01 | -0.16 | 0.01 | < 0.01 | -0.06 | 0.13 | 0.65 | -0.15 | 0.02 | < 0.01 | -0.06 | 0.01 | < 0.01 |
| Free testosterone (pg/mL) | -0.53 | 0.08 | < 0.01 | -1.96 | 0.29 | < 0.01 | -4.32 | 2.44 | 0.08 | -1.76 | 0.30 | < 0.01 | -0.77 | 0.12 | < 0.01 |
| SHBG (nmol/L) | -0.29 | 0.03 | < 0.01 | -1.11 | 0.11 | < 0.01 | 1.78 | 1.09 | 0.11 | -1.20 | 0.12 | < 0.01 | -0.48 | 0.04 | < 0.01 |
| 3-alpha-diol-G (ng/mL) | 0.03 | 0.01 | 0.03 | 0.07 | 0.04 | 0.11 | 0.23 | 0.23 | 0.32 | 0.10 | 0.05 | 0.04 | 0.05 | 0.02 | 0.03 |
| Estradiol (pg/mL) | 0.07 | 0.04 | 0.12 | 0.17 | 0.13 | 0.18 | 0.76 | 1.03 | 0.47 | 0.22 | 0.15 | 0.17 | 0.10 | 0.07 | 0.13 |
| Free estradiol (fg/mL) | 3.75 | 1.34 | 0.01 | 12.52 | 3.75 | < 0.01 | 8.40 | 28.80 | 0.77 | 13.57 | 4.93 | 0.01 | 5.99 | 2.13 | 0.01 |
adjusted for age, race/ethnicity, smoking status, alcohol consumption and physical activity
adjusted for age, race/ethnicity, smoking status, alcohol consumption and physical activity
Increased total mass in grams (lean mass and fat mass combined) was associated with lower concentrations of total testosterone [beta (estimated change in testosterone in ng/mL for a 1 kilogram increase in total mass) = -0.04; p-value < 0.01], free testosterone [beta = -0.53; p-value < 0.01], and SHBG [beta = -0.29; p-value < 0.01] and higher concentrations of 3-alpha-diol-G [beta = 0.03; p-value = 0.03] and free estradiol [beta = 3.75; p-value = 0.01]. Total mass was not associated with total estradiol concentration. The beta-coefficients and standard errors for hormone concentrations in men and the components of total mass – fat mass and lean mass excluding bone mineral content – were similar to the values presented for total mass. Percent body fat was associated with lower concentrations of total testosterone [beta (estimated change in testosterone for a 1% increase in percent body fat) = -0.16; p-value < 0.01], free testosterone [beta = -1.96; p-value < 0.01], and SHBG [beta = -1.11; p-value < 0.01] and higher concentrations of free estradiol [beta = 12.52; p-value < 0.01] but not 3-alpha-diol-G or total estradiol.
Hormone associations with anthropometric body size measurements
BMI was inversely associated with total testosterone [beta (estimated change in testosterone for a 1 kg/m2 increase in BMI) = -0.15; p-value < 0.01], free testosterone [beta = -1.76; p-value < 0.01] and SHBG [beta = -1.20; p-value < 0.01], and positively associated with 3-alpha-diol-G [beta = 0.10; p-value = 0.04] and free estradiol [beta = 13.57; p-value = 0.01] (Table 3). Similarly, waist circumference was inversely associated with total testosterone [beta (estimated change in testosterone for a 1 cm increase in waist circumference) = -0.06; p-value < 0.01], free testosterone [beta = -0.77; p-value < 0.01] and SHBG [beta = -0.48; p-value < 0.01], and positively associated with 3-alpha-diol-G [beta = 0.05; p-value = 0.03] and free estradiol [beta = 5.99; p-value = 0.01]. BMI and waist circumference were not associated with total estradiol. Standing height was not associated with any of the hormone concentrations measured [all p-values > 0.07].
The associations between body size measurements and hormones with mutual adjustment for other hormones measured are presented in Supplemental Table 1. The inverse associations between the body size measurements and testosterone remained after adjusting for SHBG, 3-alpha-diol-G and total estradiol. In the mutually adjusted models, the inverse associations between body size measurements and SHBG were no longer significant, whereas positive associations between body size measurements and total estradiol became significant. The association between percent body fat and total estradiol when adjusting for covariates alone was null; however, when the model was also adjusted for total testosterone, SHBG and 3-alpha-diol-G, percent body fat was positively associated with total estradiol [beta = 0.79; p-value < 0.01]. After mutual adjustment for total testosterone, SHBG and total estradiol the significant associations between body size measurements and 3-alpha-diol-G remained. Further, percent body fat was positively associated with 3-alpha-diol-G [beta = 0.12; p-value = 0.03] in the model that adjusted for covariates as well as the other hormones measured.
Stratification by age group
Associations of body size measurements and hormone concentration among young adult (men aged 20-44 years), middle-aged (men aged 45-69 years) and older men (men aged 70 years or older) are shown in Table 4. The results described in the remainder of this section were all significantly modified by age-group (p-value interactions ≤0.05). The inverse associations of total bone area, total bone content and total bone mineral density with total testosterone were limited to the youngest age group. Similarly, the inverse associations between total bone area, total bone mineral content and free testosterone were limited to the youngest age group. Among the oldest age group, total bone area, total bone mineral content and total bone mineral density were all positively associated with free testosterone with the numerically largest association between total bone mineral density and free testosterone [beta = 59.22; p-value < 0.01]. For SHBG, the inverse associations with total bone area, total bone mineral content and total bone mineral density were limited to the youngest and oldest age groups with the strongest inverse association in the oldest age group [beta = -77.91; p-value < 0.01].
Table 4.
Adjusted associations sex steroid hormones, SHBG and body size measurements stratified by age groups in the NHANES survey, 1999-2004.
| Age groups | Total bone area (cm2) per 100 cm2 increase | Total bone mineral content (BMC) (g) per 100 g increase | Total bone mineral density (g/cm2) | Fat mass (g) per 1 kg increase | Lean mass excluding BMC (g) per 1 kg increase | |||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Betaa | SE | p | p-int b | Betaa | SE | p | p-int b | Betaa | SE | p | p-int b | Betaa | SE | p | p-int b | Betaa | SE | p | p-int b | |
| Testosterone (ng/mL) | ||||||||||||||||||||
| 20-44 | -0.21 | 0.05 | < 0.01 | < 0.01 | -0.09 | 0.02 | < 0.01 | < 0.01 | -2.94 | 0.98 | < 0.01 | 0.02 | -0.10 | 0.01 | < 0.01 | 0.01 | -0.08 | 0.01 | < 0.01 | < 0.01 |
| 45-69 | -0.04 | 0.05 | 0.43 | 0.01 | 0.03 | 0.98 | 0.68 | 1.22 | 0.58 | -0.06 | 0.01 | < 0.01 | -0.04 | 0.01 | < 0.01 | |||||
| 70+ | 0.06 | 0.08 | 0.45 | 0.03 | 0.04 | 0.46 | 1.11 | 1.61 | 0.49 | -0.07 | 0.03 | 0.01 | -0.01 | 0.02 | 0.77 | |||||
| Free testosterone (pg/mL) | ||||||||||||||||||||
| 20-44 | -3.93 | 0.96 | < 0.01 | < 0.01 | -1.57 | 0.55 | 0.01 | < 0.01 | -41.36 | 24.95 | 0.11 | 0.01 | -1.40 | 0.22 | < 0.01 | 0.01 | -1.13 | 0.22 | < 0.01 | < 0.01 |
| 45-69 | -2.33 | 1.07 | 0.04 | -0.33 | 0.53 | 0.53 | 15.42 | 23.98 | 0.52 | -0.82 | 0.18 | < 0.01 | -0.46 | 0.16 | 0.01 | |||||
| 70+ | 2.23 | 1.00 | 0.03 | 1.53 | 0.52 | 0.01 | 59.22 | 18.49 | < 0.01 | -0.55 | 0.31 | 0.08 | 0.44 | 0.25 | 0.08 | |||||
| SHBG (nmol/L) | ||||||||||||||||||||
| 20-44 | -0.78 | 0.36 | 0.04 | 0.05 | -0.47 | 0.15 | < 0.01 | 0.01 | -19.43 | 6.22 | < 0.01 | 0.01 | -0.59 | 0.07 | < 0.01 | 0.61 | -0.49 | 0.07 | < 0.01 | 0.43 |
| 45-69 | 0.88 | 0.72 | 0.23 | 0.16 | 0.27 | 0.56 | -3.26 | 11.95 | 0.79 | -0.48 | 0.13 | < 0.01 | -0.34 | 0.10 | < 0.01 | |||||
| 70+ | -2.74 | 1.27 | 0.04 | -1.92 | 0.57 | < 0.01 | -77.91 | 20.86 | < 0.01 | -0.96 | 0.37 | 0.01 | -0.97 | 0.31 | < 0.01 | |||||
| 3-alpha-diol-G (ng/mL) | ||||||||||||||||||||
| 20-44 | -0.06 | 0.09 | 0.48 | 0.01 | -0.01 | 0.04 | 0.87 | 0.06 | 0.83 | 1.78 | 0.64 | 0.28 | 0.02 | 0.02 | 0.29 | 0.01 | 0.02 | 0.02 | 0.42 | 0.05 |
| 45-69 | 0.52 | 0.26 | 0.06 | 0.24 | 0.14 | 0.08 | 7.97 | 4.82 | 0.11 | 0.07 | 0.05 | 0.19 | 0.10 | 0.05 | 0.04 | |||||
| 70+ | 0.58 | 0.25 | 0.02 | 0.23 | 0.14 | 0.11 | 6.55 | 5.94 | 0.28 | 0.23 | 0.07 | < 0.01 | 0.11 | 0.06 | 0.06 | |||||
| Estradiol (pg/mL) | ||||||||||||||||||||
| 20-44 | 0.05 | 0.47 | 0.92 | 0.12 | 0.01 | 0.23 | 0.98 | 0.22 | -0.26 | 8.98 | 0.98 | 0.30 | 0.07 | 0.13 | 0.58 | 0.61 | 0.04 | 0.13 | 0.77 | 0.41 |
| 45-69 | -0.27 | 0.65 | 0.69 | 0.10 | 0.28 | 0.72 | 12.14 | 10.44 | 0.25 | 0.16 | 0.09 | 0.09 | 0.15 | 0.11 | 0.20 | |||||
| 70+ | 2.09 | 0.73 | 0.01 | 1.09 | 0.41 | 0.01 | 38.17 | 15.20 | 0.02 | 0.25 | 0.18 | 0.16 | 0.47 | 0.17 | 0.01 | |||||
| Free estradiol (fg/mL) | ||||||||||||||||||||
| 20-44 | 6.46 | 14.88 | 0.67 | 0.20 | 3.62 | 7.52 | 0.63 | 0.25 | 142.35 | 284.68 | 0.62 | 0.36 | 6.93 | 4.11 | 0.10 | 0.93 | 4.89 | 3.95 | 0.22 | 0.82 |
| 45-69 | -15.05 | 21.13 | 0.48 | -0.24 | 8.60 | 0.98 | 260.13 | 324.99 | 0.43 | 7.20 | 3.02 | 0.02 | 6.14 | 3.41 | 0.08 | |||||
| 70+ | 50.22 | 18.12 | 0.01 | 30.05 | 10.18 | 0.01 | 1124.78 | 362.52 | < 0.01 | 8.50 | 4.05 | 0.04 | 12.50 | 4.45 | 0.01 | |||||
| Age groups | Total mass (lean mass + fat mass) (g) per 1 kg increase | Percent body fat (%) | Standing Height (cm) per 10 cm increase | Body mass index (BMI) (kg/m2) | Waist circumference (cm) | |||||||||||||||
| Betaa | SE | p | p-int b | Betaa | SE | p | p-int b | Betaa | SE | p | p-int b | Betaa | SE | p | p-int b | Betaa | SE | p | p-int b | |
|
| ||||||||||||||||||||
| Testosterone (ng/mL) | ||||||||||||||||||||
| 20-44 | -0.05 | 0.01 | < 0.01 | < 0.01 | -0.17 | 0.02 | < 0.01 | 0.34 | -0.12 | 0.17 | 0.49 | 0.13 | -0.19 | 0.02 | < 0.01 | < 0.01 | -0.07 | 0.01 | < 0.01 | < 0.01 |
| 45-69 | -0.03 | 0.01 | < 0.01 | -0.14 | 0.02 | < 0.01 | -0.04 | 0.17 | 0.83 | -0.10 | 0.02 | < 0.01 | -0.05 | 0.01 | < 0.01 | |||||
| 70+ | -0.02 | 0.01 | 0.10 | -0.13 | 0.04 | < 0.01 | 0.30 | 0.20 | 0.15 | -0.12 | 0.05 | 0.02 | -0.05 | 0.02 | 0.01 | |||||
| Free testosterone (pg/mL) | ||||||||||||||||||||
| 20-44 | -0.70 | 0.11 | < 0.01 | < 0.01 | -2.17 | 0.38 | < 0.01 | 0.16 | -2.80 | 3.37 | 0.41 | < 0.01 | -2.53 | 0.43 | < 0.01 | < 0.01 | -1.02 | 0.16 | < 0.01 | < 0.01 |
| 45-69 | -0.35 | 0.09 | < 0.01 | -1.62 | 0.43 | < 0.01 | -9.11 | 3.60 | 0.02 | -0.90 | 0.28 | < 0.01 | -0.47 | 0.16 | < 0.01 | |||||
| 70+ | -0.02 | 0.16 | 0.92 | -1.27 | 0.44 | 0.01 | 6.16 | 2.66 | 0.03 | -0.62 | 0.59 | 0.30 | -0.20 | 0.19 | 0.29 | |||||
| SHBG (nmol/L) | ||||||||||||||||||||
| 20-44 | -0.30 | 0.04 | < 0.01 | 0.35 | -1.04 | 0.11 | < 0.01 | 0.92 | 0.33 | 1.10 | 0.76 | 0.04 | -1.19 | 0.13 | < 0.01 | 0.58 | -0.44 | 0.05 | < 0.01 | 0.30 |
| 45-69 | -0.22 | 0.06 | < 0.01 | -1.12 | 0.25 | < 0.01 | 6.25 | 2.30 | 0.01 | -1.02 | 0.27 | < 0.01 | -0.42 | 0.08 | < 0.01 | |||||
| 70+ | -0.59 | 0.18 | < 0.01 | -0.99 | 0.58 | 0.10 | -1.80 | 2.57 | 0.49 | -2.14 | 0.55 | < 0.01 | -0.96 | 0.19 | < 0.01 | |||||
| 3-alpha-diol-G (ng/mL) | ||||||||||||||||||||
| 20-44 | 0.01 | 0.01 | 0.33 | 0.02 | 0.05 | 0.04 | 0.17 | 0.02 | -0.07 | 0.26 | 0.79 | 0.34 | 0.05 | 0.04 | 0.24 | 0.07 | 0.02 | 0.02 | 0.11 | 0.03 |
| 45-69 | 0.05 | 0.03 | 0.08 | 0.06 | 0.10 | 0.57 | 0.79 | 0.45 | 0.08 | 0.14 | 0.09 | 0.13 | 0.08 | 0.05 | 0.11 | |||||
| 70+ | 0.10 | 0.03 | < 0.01 | 0.31 | 0.08 | < 0.01 | 0.21 | 0.57 | 0.72 | 0.32 | 0.13 | 0.02 | 0.09 | 0.04 | 0.02 | |||||
| Estradiol (pg/mL) | ||||||||||||||||||||
| 20-44 | 0.03 | 0.07 | 0.66 | 0.39 | 0.10 | 0.18 | 0.58 | 0.73 | 1.72 | 1.22 | 0.17 | 0.29 | 0.04 | 0.23 | 0.87 | 0.39 | 0.06 | 0.10 | 0.55 | 0.59 |
| 45-69 | 0.08 | 0.05 | 0.12 | 0.31 | 0.16 | 0.06 | -2.18 | 2.43 | 0.38 | 0.38 | 0.17 | 0.03 | 0.15 | 0.07 | 0.03 | |||||
| 70+ | 0.23 | 0.09 | 0.02 | 0.07 | 0.29 | 0.80 | 3.72 | 2.43 | 0.13 | 0.47 | 0.38 | 0.23 | 0.12 | 0.14 | 0.38 | |||||
| Free estradiol (fg/mL) | ||||||||||||||||||||
| 20-44 | 3.27 | 2.18 | 0.14 | 0.82 | 11.10 | 5.19 | 0.04 | 0.83 | 46.10 | 36.64 | 0.21 | 0.23 | 10.58 | 7.13 | 0.15 | 0.71 | 5.27 | 3.02 | 0.09 | 0.80 |
| 45-69 | 3.60 | 1.68 | 0.04 | 15.30 | 4.96 | < 0.01 | -94.61 | 76.05 | 0.22 | 16.54 | 6.04 | 0.01 | 6.55 | 2.02 | < 0.01 | |||||
| 70+ | 6.57 | 2.24 | 0.01 | 6.38 | 6.60 | 0.34 | 95.76 | 64.03 | 0.14 | 15.36 | 9.59 | 0.12 | 5.36 | 3.45 | 0.13 | |||||
adjusted for age, race/ethnicity, smoking status, alcohol consumption and physical activity
p-int = p-value for likelihood ratio test of interaction.
adjusted for age, race/ethnicity, smoking status, alcohol consumption and physical activity
p-int = p-value for likelihood ratio test of interaction.
The inverse associations between total mass and total testosterone as well as total mass and free testosterone were apparent for the young and middle-age groups with the strongest inverse associations in the youngest age group for both comparisons. The positive association between total mass and 3-alpha-diol-G was limited to the oldest age group. The effect modification by age-group was consistent for the other components of mass, fat mass and lean mass excluding bone mineral content. The association between percent body fat and 3-alpha-diol-G was also modified by age; the positive association was limited to the oldest age group [beta = 0.31; p-value < 0.01].
BMI was inversely associated with total testosterone for all age groups; the magnitude of the association was numerically largest in the youngest age group. The same was true for the association between waist circumference and total testosterone. The association between waist circumference and 3-alpha-diol-G was modified by age; the positive association with 3-alpha-diol-G was apparent only in the oldest age group.
DISCUSSION
In this large, representative sample of US men, total bone mineral content was inversely associated with testosterone (free and total) and SHBG but not associated with the other sex hormones measured, while total bone mineral density was only associated with lower SHBG. Higher total mass and lean mass were associated with lower concentrations of testosterone (free and total) and SHBG and higher concentrations of 3-alpha-diol-G and free estradiol, while higher fat mass and percent body fat were associated with lower concentrations of testosterone (free and total) and SHBG and higher concentrations of free estradiol. Higher BMI and waist circumference were both associated with lower concentrations of total testosterone, free testosterone and SHBG and higher concentrations of 3-alpha-diol-G and free estradiol. Standing height was not associated with either hormone or SHBG concentrations.
There is accumulating evidence that bone mineral density is associated with hormones in men. Estradiol has been positively associated with bone mineral density, such that low estradiol correlates with low bone mineral density as well as fracture risk, in the majority of cross sectional [17-22] and longitudinal studies [10;23;24] to date. Free and total estradiol were not associated with bone mineral density in our study, the null association between bone mineral density and total estradiol has been reported in other studies [25-29] and is consistent with an NHANES III study utilizing data from 1988-1991 and DXA scans of the femur [30].
Associations between bone mineral density and testosterone have been less consistent, [10;18;20-23;25;26;29;31] however, some studies support a positive association between bone mineral density and free/bioavailable [32] testosterone [18;23;25;26;31] or total testosterone [18;22;31]. Age significantly modified the association between bone mineral density and testosterone in our study. Specifically, the inverse association between bone mineral density and testosterone was limited to the youngest age group. We also observed a strong positive association between bone mineral density and free testosterone in older men, which is consistent with the literature [18;23;26;31]. The results from our analysis support the findings of other studies that bone mineral density is inversely associated with SHBG [17;18;21;27;29;30].
In a prior publication using NHANES III data from 1988-1994, Rohrmann and colleagues reported inverse associations between percent body fat, measured by bioelectric impedance, and testosterone, free testosterone and SHBG and positive associations between percent body fat and 3-alpha-diol-G and estradiol (total and free) [8]. Our results using data from 1999-2004 and percent body fat as measured from the whole body DXA scan also show inverse associations between percent body fat and testosterone (total and free) and SHBG, and a positive association with free estradiol. We did not however, replicate the positive association between percent body fat and 3-alpha-diol-G or total estradiol.
BMI and waist circumference have been consistently associated with decreased testosterone, free testosterone and SHBG [4-6;8]. Specifically these associations were reported in a previous study from NHANES III (n=1265) [8], and in large population-based studies from Oxford, England (n=750),[4] the Netherlands (n=400) [5], and Norway(n=1563) [6]. In a study of Greek men (n=112, age 40+) BMI was inversely associated with serum testosterone although the result did not achieve statistical significance [7]. Associations with 3-alpha-diol-G and estradiol (total and free) have either been less consistent across study populations or not reported. While we reported positive associations between BMI, waist circumference and 3-alpha-diol-G, Rohrmann and colleagues reported a null association between BMI and 3-alpha-diol-G and a positive association between waist circumference and 3-alpha-diol-G [8]. Conversely, Allen and colleagues reported a positive association between BMI and 3-alpha-diol-G and a null association between waist circumference and 3-alpha-diol-G [4]. Other studies evaluating body size and hormones did not evaluate 3-alpha-diol-G [5-7]. Overall, associations between adiposity-related measurements using DXA (percent body fat, total mass, fat mass and lean mass excluding bone mineral content) and hormone levels were similar to the associations for anthropometric measures (BMI and waist circumference).
The inverse association between body size and testosterone and direct association between body size and 3-alpha-diol-G are consistent with the observed associations with prostate cancer. Specifically, studies have reported increased risk of high-grade prostate cancer associated with low testosterone levels [33] as well as increased risk of aggressive prostate cancer with obese BMI [34]. Further, 3-alpha-diol-G is used as a surrogate marker of steroid 5-alpha-reductase enzymatic activity, or more generally of intra-prostatic androgencity, and may also be associated with increased prostate cancer risk [35]. It is likely, however, that the association needs to be evaluated separately for low-grade and high-grade tumors, given the different associations with obese BMI. The positive association between body size and free estradiol is consistent with increased risk of obesity with colon cancer. Specifically, it is hypothesized that estrogens are associated with colon cancer through multiple mechanisms, including independent effects on risk as well as by strong modifying effect on the association of BMI and colon cancer [36]. Specifically it is hypothesized that estrogen may act indirectly on colon cancer risk through its regulatory action on insulin and the insulin-like growth factor (IGF) system [37].
This study has several strengths including the large sample size and well-characterized study population with a nationally representative sample. Other strengths include the standardized measures of clinical and anthropometric body size performed by trained study personnel, precise measurements of hormones, and availability of adjustment factors that are modifiable correlates of both hormones and body size. Study limitations include the cross-sectional study design, which does not provide direct evidence of cause and effect, and that hormone measurements were based on single serum sample. Serum samples were, however, selected from fasting samples drawn in the morning, to minimize effects of diurnal variation in total and free testosterone. An additional limitation is the use of total bone mineral density, which does not permit evaluation using cortical or trabecular bone compartments.
Our study provides further support that increased body fat is associated with lower circulating levels of testosterone (total and free) and SHBG and higher circulating levels of free estradiol. Clinical measures of body fat (percent body fat, total mass and fat mass as measured via DXA scan) and anthropometric measures of body fat (BMI and waist circumference) provided similar inferences regarding the association between increased body fat and hormone levels in men; however, the associations with DXA measurements need to be replicated in other large population-based samples.
Supplementary Material
Acknowledgments
This work was supported by the Intramural Research Program of the National Institutes of Health, Division of Cancer Epidemiology and Genetics, National Cancer Institute, Department of Health and Human Services.
Footnotes
The authors declare that they have no conflict of interest.
Reference List
- 1.Flegal KM, Graubard BI, Williamson DF, Gail MH. Excess deaths associated with underweight, overweight, and obesity. JAMA. 2005;293(15):1861–1867. doi: 10.1001/jama.293.15.1861. [DOI] [PubMed] [Google Scholar]
- 2.Flegal KM, Graubard BI. Estimates of excess deaths associated with body mass index and other anthropometric variables. Am J Clin Nutr. 2009;89(4):1213–1219. doi: 10.3945/ajcn.2008.26698. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Calle EE, Kaaks R. Overweight, obesity and cancer: epidemiological evidence and proposed mechanisms. Nat Rev Cancer. 2004;4(8):579–591. doi: 10.1038/nrc1408. [DOI] [PubMed] [Google Scholar]
- 4.Allen NE, Appleby PN, Davey GK, Key TJ. Lifestyle and nutritional determinants of bioavailable androgens and related hormones in British men. Cancer Causes Control. 2002;13(4):353–363. doi: 10.1023/a:1015238102830. [DOI] [PubMed] [Google Scholar]
- 5.Muller M, den T I, Thijssen JH, Grobbee DE, van der Schouw YT. Endogenous sex hormones in men aged 40-80 years. Eur J Endocrinol. 2003;149(6):583–589. doi: 10.1530/eje.0.1490583. [DOI] [PubMed] [Google Scholar]
- 6.Svartberg J, Midtby M, Bonaa KH, Sundsfjord J, Joakimsen RM, Jorde R. The associations of age, lifestyle factors and chronic disease with testosterone in men: the Tromso Study. Eur J Endocrinol. 2003;149(2):145–152. doi: 10.1530/eje.0.1490145. [DOI] [PubMed] [Google Scholar]
- 7.Tamimi R, Mucci LA, Spanos E, Lagiou A, Benetou V, Trichopoulos D. Testosterone and oestradiol in relation to tobacco smoking, body mass index, energy consumption and nutrient intake among adult men. Eur J Cancer Prev. 2001;10(3):275–280. doi: 10.1097/00008469-200106000-00012. [DOI] [PubMed] [Google Scholar]
- 8.Rohrmann S, Shiels MS, Lopez DS, et al. Body fatness and sex steroid hormone concentrations in US men: results from NHANES III. Cancer Causes Control. 2011;22(8):1141–1151. doi: 10.1007/s10552-011-9790-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Falahati-Nini A, Riggs BL, Atkinson EJ, O’Fallon WM, Eastell R, Khosla S. Relative contributions of testosterone and estrogen in regulating bone resorption and formation in normal elderly men. J Clin Invest. 2000;106(12):1553–1560. doi: 10.1172/JCI10942. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Barrett-Connor E, Mueller JE, von Muhlen DG, Laughlin GA, Schneider DL, Sartoris DJ. Low levels of estradiol are associated with vertebral fractures in older men, but not women: the Rancho Bernardo Study. J Clin Endocrinol Metab. 2000;85(1):219–223. doi: 10.1210/jcem.85.1.6327. [DOI] [PubMed] [Google Scholar]
- 11.Duffy DM, Legro RS, Chang L, Stanczyk FZ, Lobo RA. Metabolism of dihydrotestosterone to 5 alpha-androstane-3 alpha, 17 beta-diol glucuronide is greater in the peripheral compartment than in the splanchnic compartment. Fertil Steril. 1995;64(4):736–739. doi: 10.1016/s0015-0282(16)57848-0. [DOI] [PubMed] [Google Scholar]
- 12.Centers for Disease Control and Prevention (CDC) National Center for Health Statistics (NCHS). National Health and Nutrition Examination Survey Data. Hyattsville, MD: U.S. Department of Health and Human Services, Centers for Disease Control and Prevention; 2011. [Google Scholar]
- 13.Nyante SJ, Graubard BI, Li Y, et al. Trends in sex hormone concentrations in US males 1988-1991 to 1999-2004. Int J Androl. 2011 doi: 10.1111/j.1365-2605.2011.01230.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Vermeulen A, Verdonck L, Kaufman JM. A critical evaluation of simple methods for the estimation of free testosterone in serum. J Clin Endocrinol Metab. 1999;84(10):3666–3672. doi: 10.1210/jcem.84.10.6079. [DOI] [PubMed] [Google Scholar]
- 15.National Center for Health Statistics. National Health and Nutrition Examination Survey body composition procedures manual. Hyattsville, MD: Division of Health and Nutrition Examiniation Surveys, National Center for Health Statistics, Centers for Disease Control and Prevention, U.S. Department of Health and Human Services; 2011. [Google Scholar]
- 16.Schenker N, Borrud LG, Burt VL, et al. Multiple imputation of missing dual-energy X-ray absorptiometry data in the National Health and Nutrition Examination Survey. Stat Med. 2011;30(3):260–276. doi: 10.1002/sim.4080. [DOI] [PubMed] [Google Scholar]
- 17.Slemenda CW, Longcope C, Zhou L, Hui SL, Peacock M, Johnston CC. Sex steroids and bone mass in older men. Positive associations with serum estrogens and negative associations with androgens. J Clin Invest. 1997;100(7):1755–1759. doi: 10.1172/JCI119701. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.van den Beld AW, de Jong FH, Grobbee DE, Pols HA, Lamberts SW. Measures of bioavailable serum testosterone and estradiol and their relationships with muscle strength, bone density, and body composition in elderly men. J Clin Endocrinol Metab. 2000;85(9):3276–3282. doi: 10.1210/jcem.85.9.6825. [DOI] [PubMed] [Google Scholar]
- 19.Khosla S, Melton LJ, III, Atkinson EJ, O’Fallon WM. Relationship of serum sex steroid levels to longitudinal changes in bone density in young versus elderly men. J Clin Endocrinol Metab. 2001;86(8):3555–3561. doi: 10.1210/jcem.86.8.7736. [DOI] [PubMed] [Google Scholar]
- 20.Gennari L, Merlotti D, Martini G, et al. Longitudinal association between sex hormone levels, bone loss, and bone turnover in elderly men. J Clin Endocrinol Metab. 2003;88(11):5327–5333. doi: 10.1210/jc.2003-030736. [DOI] [PubMed] [Google Scholar]
- 21.Goderie-Plomp HW, van der Klift M, de RW, Hofman A, de Jong FH, Pols HA. Endogenous sex hormones, sex hormone-binding globulin, and the risk of incident vertebral fractures in elderly men and women: the Rotterdam Study. J Clin Endocrinol Metab. 2004;89(7):3261–3269. doi: 10.1210/jc.2002-022041. [DOI] [PubMed] [Google Scholar]
- 22.Fink HA, Ewing SK, Ensrud KE, et al. Association of testosterone and estradiol deficiency with osteoporosis and rapid bone loss in older men. J Clin Endocrinol Metab. 2006;91(10):3908–3915. doi: 10.1210/jc.2006-0173. [DOI] [PubMed] [Google Scholar]
- 23.Greendale GA, Edelstein S, Barrett-Connor E. Endogenous sex steroids and bone mineral density in older women and men: the Rancho Bernardo Study. J Bone Miner Res. 1997;12(11):1833–1843. doi: 10.1359/jbmr.1997.12.11.1833. [DOI] [PubMed] [Google Scholar]
- 24.Amin S, Zhang Y, Sawin CT, et al. Association of hypogonadism and estradiol levels with bone mineral density in elderly men from the Framingham study. Ann Intern Med. 2000;133(12):951–963. doi: 10.7326/0003-4819-133-12-200012190-00010. [DOI] [PubMed] [Google Scholar]
- 25.Keles I, Aydin G, Basar MM, et al. Endogenous sex steroids and bone mineral density in healthy men. Joint Bone Spine. 2006;73(1):80–85. doi: 10.1016/j.jbspin.2005.04.003. [DOI] [PubMed] [Google Scholar]
- 26.Mellstrom D, Johnell O, Ljunggren O, et al. Free testosterone is an independent predictor of BMD and prevalent fractures in elderly men: MrOS Sweden. J Bone Miner Res. 2006;21(4):529–535. doi: 10.1359/jbmr.060110. [DOI] [PubMed] [Google Scholar]
- 27.Bjornerem A, Emaus N, Berntsen GK, et al. Circulating sex steroids, sex hormone-binding globulin, and longitudinal changes in forearm bone mineral density in postmenopausal women and men: the Tromso study. Calcif Tissue Int. 2007;81(2):65–72. doi: 10.1007/s00223-007-9035-z. [DOI] [PubMed] [Google Scholar]
- 28.Khosla S, Melton LJ, III, Atkinson EJ, O’Fallon WM, Klee GG, Riggs BL. Relationship of serum sex steroid levels and bone turnover markers with bone mineral density in men and women: a key role for bioavailable estrogen. J Clin Endocrinol Metab. 1998;83(7):2266–2274. doi: 10.1210/jcem.83.7.4924. [DOI] [PubMed] [Google Scholar]
- 29.Legrand E, Hedde C, Gallois Y, et al. Osteoporosis in men: a potential role for the sex hormone binding globulin. Bone. 2001;29(1):90–95. doi: 10.1016/s8756-3282(01)00478-1. [DOI] [PubMed] [Google Scholar]
- 30.Paller CJ, Shiels MS, Rohrmann S, et al. Relationship of sex steroid hormones with bone mineral density (BMD) in a nationally representative sample of men. Clin Endocrinol (Oxf) 2009;70(1):26–34. doi: 10.1111/j.1365-2265.2008.03300.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Murphy S, Khaw KT, Cassidy A, Compston JE. Sex hormones and bone mineral density in elderly men. Bone Miner. 1993;20(2):133–140. doi: 10.1016/s0169-6009(08)80022-0. [DOI] [PubMed] [Google Scholar]
- 32.Mellstrom D, Vandenput L, Mallmin H, et al. Older men with low serum estradiol and high serum SHBG have an increased risk of fractures. J Bone Miner Res. 2008;23(10):1552–1560. doi: 10.1359/jbmr.080518. [DOI] [PubMed] [Google Scholar]
- 33.Schatzl G, Madersbacher S, Thurridl T, et al. High-grade prostate cancer is associated with low serum testosterone levels. Prostate. 2001;47(1):52–58. doi: 10.1002/pros.1046. [DOI] [PubMed] [Google Scholar]
- 34.Hsing AW, Sakoda LC, Chua S., Jr Obesity, metabolic syndrome, and prostate cancer. Am J Clin Nutr. 2007;86(3):s843–s857. doi: 10.1093/ajcn/86.3.843S. [DOI] [PubMed] [Google Scholar]
- 35.Hsing AW. Hormones and prostate cancer: where do we go from here? J Natl Cancer Inst. 1996;88(16):1093–1095. doi: 10.1093/jnci/88.16.1093. [DOI] [PubMed] [Google Scholar]
- 36.Slattery ML, Ballard-Barbash R, Edwards S, Caan BJ, Potter JD. Body mass index and colon cancer: an evaluation of the modifying effects of estrogen (United States) Cancer Causes Control. 2003;14(1):75–84. doi: 10.1023/a:1022545017867. [DOI] [PubMed] [Google Scholar]
- 37.Polderman KH, Gooren LJ, Asscheman H, Bakker A, Heine RJ. Induction of insulin resistance by androgens and estrogens. J Clin Endocrinol Metab. 1994;79(1):265–271. doi: 10.1210/jcem.79.1.8027240. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.