Nutrologia.med.brThe Effects of Serum Testosterone, Estradiol, and Sex
Hormone Binding Globulin Levels on Fracture Risk in
Erin S. LeBlanc, Carrie M. Nielson, Lynn M. Marshall, Jodi A. Lapidus,Elizabeth Barrett-Connor, Kristine E. Ensrud, Andrew R. Hoffman, Gail Laughlin,Claes Ohlsson, and Eric S. Orwoll, for the Osteoporotic Fractures in Men StudyGroup Bone and Mineral Unit (E.S.L., C.M.N., L.M.M., J.A.L., E.S.O.), Oregon Health and Science University,Portland, Oregon 97239; Department of Family and Preventive Medicine (E.B.-C., G.L.), University ofCalifornia, San Diego, San Diego, California 92093; Departments of Medicine and Epidemiology &Community Health (K.E.E.), University of Minnesota and Department of Medicine (K.E.E.), VeteransAffairs Medical Center, Minneapolis, Minnesota 55417; Department of Medicine (A.R.H.), StanfordUniversity, Palo Alto, California 94305; and Department of Internal Medicine (C.O.), Center for BoneResearch at the Sahlgrenska Academy, SE-416 85 Go¨teborg, Sweden Context: The relationship between sex steroids and fracture is poorly understood.
Objective: The objective of the study was to examine associations between nonvertebral fracture
risk and bioavailable estradiol (bioE2), bioavailable testosterone (bioT), and SHBG.
Design: This was a case-cohort study.
Setting: The Osteoporotic Fractures in Men Study (MrOS) was conducted in a prospective U.S.
cohort in 5995 community-dwelling men 65 yr old or older.
Participants: Participants included a subcohort of 1436 randomly chosen white men plus all 446
minorities and all those with incident hip and other nonvertebral fractures.
Main Outcome Measures: Baseline testosterone and estradiol were measured by mass spectrom-
etry (MS) and SHBG by RIA.
Results: Men with the lowest bioE2 (⬍11.4 pg/ml) or highest SHBG (⬎59.1 nM) had greater risk of
all nonvertebral fractures [adjusted hazard ratio (HR) [95% confidence interval]: 1.5 (1.2–1.9) and
1.4 (1.1–21.8), respectively]. Men with the lowest bioT (⬍163.5 ng/dl) had no increased fracture risk
after adjustment for bioE2 [adjusted HR 1.16 (0.90 –1.49)]. A significant interaction between SHBG
and bioT (P ⫽ 0.03) resulted in men with low bioT and high SHBG having higher fracture risk [HR
2.1 (1.4 –3.2)]. Men with low bioE2, low bioT, and high SHBG were at highest risk [HR 3.4 (2.2–5.3)].
Conclusions: Older men with low bioE2 or high SHBG levels are at increased risk of nonvertebral
fracture. When SHBG levels are high, men with low bioT levels have higher risk. The strongest
association occurred when all measures were considered in combination. (J Clin Endocrinol Metab
94: 3337–3346, 2009)
Ithasbeenspeculatedthatsexsteroidscontributetofrac- gens and estrogens have in vitro and in vivo bone effects
ture risk in older men (1). With aging, sex steroid con- and trophic effects on skeletal development (6). Estradiol centrations decline (2, 3), fracture rate increases (4), and has been consistently associated with skeletal character- testosterone therapy improves bone density (5). Andro- istics (6 –11), but whether testosterone has independent ISSN Print 0021-972X ISSN Online 1945-7197 Abbreviations: bioE2, Bioavailable estradiol; bioT, bioavailable testosterone; BMD, bone Printed in U.S.A.
mineral density; BMI, body mass index; CI, confidence interval; CV, coefficient of variation; Copyright 2009 by The Endocrine Society HR, hazard ratio; MrOS, Osteoporotic Fractures in Men Study.
doi: 10.1210/jc.2009-0206 Received February 13, 2009. Accepted June 25, 2009.
First Published Online July 7, 2009 J Clin Endocrinol Metab, September 2009, 94(9):3337–3346
LeBlanc et al. Sex Steroids and Fracture in Men J Clin Endocrinol Metab, September 2009, 94(9):3337–3346 effects on bone density, structure, or biochemical indi-ces is uncertain (12). Testosterone may affect variousextraskeletal functions relevant to fracture, includingmuscle strength, physical activity, cognition, and fallrate (13–18).
High SHBG has been independently associated with fracture risk (19 –25). By binding to testosterone andestradiol, SHBG reduces circulating sex steroid concen-trations and thereby their cellular actions. SHBG mayhave independent effects via a receptor mediated mech-anism or affect sex steroid interaction with cellular re-ceptors (26 –28).
Although several publications suggest lower estra- diol and/or testosterone or higher SHBG are linked tohigher fracture rates (11, 19 –21, 29), few studies haveadequate power to assess independent and/or interde-pendent effects of estradiol, testosterone, and SHBG.
Most previous studies measured sex steroids using RIAtechniques, which are susceptible to artifact, particu-larly at low concentrations (30, 31).
FIG. 1. Case-cohort design for the MrOS sex steroids and fracture
We report associations between fracture risk and sex study. *, Subcohort consisted of 1436 randomly selected non-Hispanic steroids in a large cohort of older men. Sex steroid levels white men and all 446 minority men. Weighting was used in analysesto account for stratified sampling by race.
were measured using liquid chromatography/massspectrometry, a method with high accuracy (32, 33). We subcohort and 342 incident fracture cases (102 from subcohort; examined interactions between sex steroids, fracture 240 outside subcohort).
risk, and other variables including bone mineral density(BMD), age, body composition, physical activity, and physical performance. We assessed the SHBG-fracture Race/ethnicity, education level, smoking and alcohol con- sumption, occurrence of fracture after age 50 yr, medical history, association, both independently and in combination and previous 12-month fall occurrence were determined by ques- with sex steroids.
tionnaire at baseline. Current medications were recorded. Phys-ical activity was assessed with the Physical Activity Score for theElderly (36). Height (centimeters) and weight (kilograms) weremeasured using standard protocols. Grip strength (kilograms), Subjects and Methods
lower extremity power, time to complete a narrow walk (6 m ⫻20 cm), and ability to rise from a chair without arms were as- sessed (34).
The Osteoporotic Fractures in Men Study (MrOS) study en- rolled 5995 participants from March 2000 through April 2002 Sex steroid measurements
as previously described (34, 35). Community-based recruitment Baseline fasting morning blood was collected. Serum was pre- occurred at six U.S. academic medical centers in Birmingham, pared immediately after phlebotomy and stored at ⫺70 C. Total AL; Minneapolis, MN; Palo Alto, CA; Pittsburgh, PA; Portland, serum testosterone and estradiol were measured using a com- OR; and San Diego, CA. Eligible participants were at least 65 yr bined gas chromatographic-negative ionization tandem mass old, could walk without assistance, and had not had bilateral hip spectrometry and liquid chromatographic electrospray tandem replacement surgery. The institutional review board at each cen- mass spectrometry bioanalytical method (Taylor Technology, ter approved the study protocol. All participants gave written Princeton, NJ). A 1/(concentration)2 weighted least squares re- informed consent. We used a case-cohort design: a random sub- gression procedure was used to fit a linear function to the cali- sample of the original cohort (subcohort) was selected indepen- bration data. The lower limit of detection for estradiol is 0.625 dently of fracture cases, and all cases outside the subsample were pg/ml (2.29 pmol/liter), and for testosterone is 25.0 pg/ml (0.09 selected (Fig. 1). We selected 2048 men for steroid measurements nmol/liter). Duplicate aliquots from each participant's serum (subcohort). A total of 1436 were randomly chosen plus all 446 were assayed and results averaged. Testosterone intraassay co- minorities were included. They were followed for 4.7 (⫾.9) yr.
efficient of variation (CV) was 2.5% and interassay CV, 6.0%; Measures were also obtained in men (n ⫽ 3865) who experienced the estradiol intraassay CV was 6.4% and interassay CV, 10.1%.
an incident nonvertebral fracture between enrollment and July Serum SHBG concentrations were measured using an Immulite 2006. Therefore, fracture cases could arise from either the sub- analyzer with chemiluminescent substrate (Diagnostic Products cohort (n ⫽ 126) or the remainder of the cohort (n ⫽ 299). After Corp., Los Angeles, CA). The standard curve ranged from 0.2 to exclusions, the final study population included 1738 men in the 180 nm/liter. The SHBG intraassay CV was 4.4% and interassay J Clin Endocrinol Metab, September 2009, 94(9):3337–3346 CV, 6.0%. Albumin values for free hormone calculations were (40) and were considered statistically significant if P ⬍ 0.10. We obtained from baseline serum using routine colorimetric then categorized men into eight mutually exclusive categories.
methods (interassay CV 2.0%). Calculation of bioavailable The reference category (lowest risk) contained men with bioT fractions of testosterone and estradiol was by the method of and bioE2 in the highest three quartiles and SHBG in the lowest Sodergard et al. (37). Using this method, the Spearman cor- three quartiles. The eighth category (highest risk) contained men relation coefficient for bioavailable testosterone and free tes- with bioT and bioE2 in the lowest quartile and SHBG in the tosterone was 0.98 and for bioavailable estradiol and free highest quartile. Each intermediate category contained men who estradiol was 0.98, both P ⬍ 0.0001.
were in one or more high-risk quartiles of bioT, bioE2, or SHBG.
All Cox proportional hazard models were fit using the weighting method of Barlow et al. (41) for case-cohort analysis.
Areal proximal femur BMD was measured using dual-energy Age, race, and body mass index (BMI) were included as covari- x-ray absorptiometry (QDR 4500W; Hologic Inc., Bedford, ates in all models. Additional potential confounders were added, MA). Participants were scanned according to standardized pro- and if addition changed the HR for the sex steroid variable by cedures and scanners were calibrated at baseline. Whole body, more than 10%, it was retained in the model. Primary analyses spine, hip, and linearity phantoms were measured at all sites at were of each sex steroid individually. Subsequently models were baseline, and spine and hip phantoms were scanned throughout adjusted for other sex steroids. For example, the model evalu- the study to monitor longitudinal changes. Daily quality control ating bioE2 was also adjusted for the dichotomous bioT and scans showed no shifts in scanner performance at any site during SHBG variables to determine whether this altered the HR for To estimate the proportion of fracture cases that would be attributable to low bioE2, low bioT, and high SHBG, we con- Ascertainment of incident fractures
ducted an exploratory attributable fraction analysis. The av- We contacted 99% of participants every 4 months by mail or erage attributable fraction method (42) was used to obtain telephone to ask about recent fractures. All reported nonspine attributable fraction estimates for each sex steroid and SHBG fractures were adjudicated by physician review of radiology re- and adjust for the other sex steroid/SHBG measures and for ports or x-rays if radiology reports were unavailable. Fracture age, BMI, and BMD. To conduct this exploration with readily follow-up was 99%. Using a group of investigators, fractures available statistical code (43), we assumed a simple case-con- were adjudicated as traumatic if circumstances leading to the trol design and estimated odds ratios using multivariable lo- fracture would likely have resulted in a fracture in a normal gistic regression.
To determine the robustness of our findings, we performed sensitivity analyses. To evaluate whether models were robust to potentially influential observations, we calculated Df␤ for each Cox proportional hazards models, with weighting to accom- of the sex steroid variables in the final models, with and without modate the stratified sampling and case-cohort design, were used interaction terms. Using a cutoff of the absolute value of 2/冑n, no to evaluate associations between sex steroids and time to incident points were considered influential. However, plots of each Df␤ by identification number allowed us to identify those observa- Three methods were used to evaluate associations between tions with relatively more influence than others. When these sex steroids and time to first fracture. We first created quartiles were excluded (n ⫽ 3 for full model without interaction term, n ⫽ of sex steroid variables based on distributions in the subcohort.
15 for full model with interaction term), there were no changes Because men in second, third, and fourth quartiles had similar in tests of the null hypothesis (i.e. no term gained or lost statistical risks of fracture, we created dichotomous variables; for testos- significance), and only the adjusted HR for bioE2 was attenuated terone and estradiol, the lowest quartile was compared with the (by 0.1%). The HRs for other terms were unchanged or strength- other three quartiles; for SHBG, the highest quartile was com- ened by the exclusion of observations with relatively larger ab- pared with the lowest three quartiles. Second, we used restricted solute values of Df␤.
cubic spline Cox proportional hazard models to examine sexsteroid variables as continuous and to test whether associationswith incident fracture were nonlinear (38). Third, we performed exploratory cut point analysis. We dichotomized sex steroids atvarious quantiles using log likelihoods of Cox proportional haz- Most nonvertebral fractures were judged as nontrau- ard models. The cut point at which the sex steroid variable was matic (nontraumatic n ⫽ 280, traumatic n ⫽ 62). There dichotomized to produce the highest profile log likelihood was were few traumatic hip fractures (n ⫽ 2), and their ex- considered the best value for further dichotomizing (39). The clusion did not affect analyses. The subcohort and frac- cubic spline and cut point analyses supported use of the firstquartile as a cut point.
ture case characteristics are shown in Table 1. Corre- We evaluated interactions among bioavailable testosterone lations between serum levels of sex steroids and SHBG (bioT), bioavailable estradiol (bioE2), and SHBG. We stratified [bioE2 and SHBG: r ⫽ ⫺0.13 (P ⬍ 0.0001), bioT and each dichotomous sex steroid variable (dichotomized at lowest SHBG: r ⫽ 0.27 (P ⬍ 0.0001)] and between bioE2 and quartile for bioE2 and bioT and highest quartile for SHBG) and bioT [r ⫽ 0.37 (P ⬍ 0.0001)] were moderate. Age was evaluated adjusted hazard ratios (HRs) for remaining sex steroidvariables in each stratum. For example, we tested the association negatively associated with bioT and bioE2 (r ⫽ ⫺0.19 between bioT and fracture in each stratum of SHBG. Additive to ⫺0.09) and positively associated with SHBG (r ⫽ interactions were tested in Cox proportional hazards models 0.24) (P ⬍ 0.0001). BMI was negatively associated with LeBlanc et al. Sex Steroids and Fracture in Men J Clin Endocrinol Metab, September 2009, 94(9):3337–3346 TABLE 1. Selected characteristics of men in the MrOS sex steroid case-cohort study
Subcohort (n ⴝ 1738)a fracture cases (n ⴝ 1636)
cases (n ⴝ 342)b
Mean ⴞ SD or %
Mean ⴞ SD or %
Mean ⴞ SD or %
Self-reported health Fair/poor/very poor Cigarette smoking Current alcohol consumption Greater than zero and less than seven drinks Seven or more drinks per week Physical performance Narrow walk (m/sec) Leg power (100 W) History of falls reported at baseline Previous nontrauma fracture after age 50 yr Total testosterone (ng/dl) Total estradiol (pg/ml) BioT (ng/dl)c BioE2 (pg/ml)c a Subcohort consisted of 1436 randomly selected non-Hispanic white men and all 446 minority men. It includes 102 incident fracture cases (Figure1). Minorities were oversampled in the subcohort; b fracture cases include 102 incident fracture cases inside the subcohort and 240 incidentfracture cases outside the cohort (Figure 1). Of the nonvertebral fractures, 74 (21.6%) were hip fractures; c to convert bioE2 to picomoles per liter,the conversion factor is 3.671; to convert bioT to nanomoles per liter, the conversion factor is 0.0347.
bioT and SHBG (r ⫽ ⫺0.31 to ⫺0.30) and positively BMI, the HR for all nonspine fracture in those in the lowest associated with bioE2 (r ⫽ 0.17) (P ⬍ 0.0001). Weight bioE2 quartile vs. the highest three quartiles was 1.48 decreased by 0.36% per year during follow-up. Corre- [95% confidence interval (CI) 1.18 –1.86; Table 2 and Fig.
lations between bioT, bioE2, and SHBG and BMD were 2A]. The association was similar after adjustment for bioT between ⫺0.05 and ⫺0.2 (P ⬍ 0.0001).
and SHBG but was somewhat attenuated after adjustmentfor total hip BMD (HR 1.29; 95% CI 1.01–1.64). A sim- Fracture risk and sex steroids
ilar association was present between bioE2 and hip frac- Men with lower levels of bioE2 were at higher risk of ture risk (HR 1.57; 95% CI 0.95–2.59; Table 3). Total nonvertebral fracture. After adjustment for age, race, and estradiol was not significantly associated with nonverte- TABLE 2. Hazard ratios (95% CI) for association between nonvertebral fractures and sex steroids
1.49 (1.19 –1.87) 1.39 (1.10 –1.75) 1.63 (1.30 –2.04) Adjusted for age, race, BMI 1.48 (1.18 –1.86) 1.28 (1.00 –1.64) 1.44 (1.14 –1.82) Adjusted for bioE2c 1.16 (0.90 –1.49) 1.42 (1.12–1.80) Adjusted for bioTc 1.42 (1.12–1.80) 1.48 (1.17–1.88) Adjusted for SHBGc 1.46 (1.16 –1.83) 1.33 (1.04 –1.70) Full model including bioE2, bioT, and SHBGc 1.39 (1.09 –1.76) 1.20 (0.93–1.56) 1.45 (1.14 –1.84) Full model additionally adjusted for BMDc 1.29 (1.01–1.64) 1.24 (0.96 –1.59) 1.36 (1.07–1.72) a HR is for lowest quartile vs. highest three; for bioE2 lowest quartile was less than 11.4 pg/ml (⬍41.8 pmol/liter); for bioT lowest quartile was lessthan 163.5 ng/dl (⬍5.67 nmol/liter); b HR is for highest quartile (SHBG ⱖ59.1 nM) vs. lowest three; c also adjusted for age, race, and BMI; BMDrefers to total hip BMD.
J Clin Endocrinol Metab, September 2009, 94(9):3337–3346 no longer significant after adjustment for bioE2 (HR 1.16; 95% CI 0.90 –1.49). When only nontraumatic fractureswere considered, the association between bioT and frac-ture risk was stronger (HR 1.45; 95% CI 1.12–1.89) and remained significant after adjustment for bioE2 (HR 1.31; 95% CI 1.00 –1.72). Inclusion of BMD in the model did not significantly affect the association, regardless oftrauma status. The HRs for the relationship between bioTand hip fracture risk were similar (Table 3). Total testos-terone levels were not associated with nonvertebral (HR1.02; 95% CI 0.79 –1.32) or hip fracture risk (HR 0.93; < 11.4 11.4-14.0 14.1-17.0 > 17.1 95% CI 0.51–1.71).
Quartile of bioavailable estradiol (pg/ml) Fracture risk and SHBG
After adjustment for age, race, and BMI, men with the highest quartile of SHBG were at increased risk of non-vertebral fracture compared with those in the lowest three quartiles (HR 1.44; 95% CI 1.14 –1.82; Table 2 and Fig.
2C). The association remained consistent after adjustment for sex steroids but was slightly attenuated after adjust-ment for BMD. Associations between SHBG level andfracture risk were slightly stronger but not substantivelyaltered when only nontraumatic fractures were considered(HR 1.57; 95% CI 1.22–2.03). Hip fracture risk was ap- < 163.5 163.5-202.3 202.4-242.8 proximately doubled in men with high SHBG (HR 2.17; Quartile of bioavailable testosterone (ng/dl) 95% CI 1.31–3.59; Table 3) and was not influenced by further adjustment for sex steroids or BMD.
Consideration of covariates
The associations between fracture risk and sex ste- roids and SHBG were not substantively altered by se-quential adjustment for other potential confounders,including physical activity, physical performance, andprevious falls. Limiting analyses to non-Hispanic whiteparticipants and excluding hip fractures did not alterthe findings.
< 35.3 35.3-45.9 46.0-59.0 > 59.1 Quartile of SHBG (nM) FIG. 2. HRs and 95% CIs for risk of nonvertebral fractures by quartiles
of sex steroids (adjusted for age, race, BMI). A, Bioavailable estradiol.
Spline analyses showed a nonlinear association be- B, Bioavailable testosterone. C, SHBG. To convert bioavailable estradiol tween serum bioE2 and nonvertebral fracture (P for non- to picomoles per liter, the conversion factor is 3.671; to convertbioavailable testosterone to nanomoles per liter, the conversion factor linearity ⫽ 0.045; Fig. 3A). Log likelihood cut point anal- ysis showed that dichotomizing bioE2 at 12.5 pg/ml (45.9pmol/liter) maximized model fit for nonvertebral frac- bral (HR 1.09; 95% CI 0.86 –1.39) or hip fracture risk tures. This threshold concentration was similar to that (HR 1.52; 95% CI 0.91–2.52). These associations were associated with increased fracture risk in the lowest quar- essentially unchanged when only nontraumatic fractures tile of bioE2 [⬍11.4 pg/ml (41.8 pmol/liter)]. Spline anal- were considered.
ysis did not reveal nonlinearity in the associations between After adjustment for age, race, and BMI, men with bioT fracture risk and bioT or SHBG (Fig. 3, B and C).
in the lowest quartile had a higher risk of nonvertebralfracture than those in the highest three quartiles (HR 1.28; Interaction between bioT, SHBG, and fracture risk
95% CI 1.00 –1.64; Table 2 and Fig. 2B). The association We observed a significant additive interaction between was slightly stronger after adjustment for SHBG but was bioT and SHBG (P ⫽ 0.03). Nonvertebral fracture risk for LeBlanc et al. Sex Steroids and Fracture in Men J Clin Endocrinol Metab, September 2009, 94(9):3337–3346 TABLE 3. Hazard ratios (95% CI) for association between hip fractures and sex steroids
1.56 (0.96 –2.54) 1.74 (1.05–2.86) 3.53 (2.20 –5.68) Adjusted for age, race, BMI 1.57 (0.95–2.59) 1.33 (0.79 –2.25) 2.17 (1.31–3.59) Adjusted for bioE2c 1.18 (0.68 –2.04) 2.14 (1.30 –3.54) Adjusted for bioTc 1.50 (0.89 –2.54) 2.23 (1.35–3.69) Adjusted for SHBGc 1.54 (0.93–2.53) 1.42 (0.84 –2.40) Full model including bioE2, bioT, and SHBGc 1.43 (0.84 –2.43) 1.26 (0.73–2.20) 2.18 (1.31–3.61) Full model additionally adjusted for BMDc 1.00 (0.56 –1.77) 1.59 (0.90 –2.81) 2.09 (1.23–3.56) a HR is for lowest quartile vs. highest three; for bioE2 lowest quartile was less than 11.4 pg/ml (⬍41.8 pmol/liter); for bioT lowest quartile was lessthan 163.5 ng/dl (⬍5.67 nmol/liter); b HR is for highest quartile (SHBG ⱖ59.1 nM) vs. lowest three; c also adjusted for age, race, and BMI; BMDrefers to total hip BMD.
the lowest quartile of bioT was greater among men with SHBG. For hip fracture risk, the fraction attributed to low SHBG in the highest quartile (HR 2.10; 95% CI 1.39 – bioE2 was 0.1%, to low bioT was 2.7%, and to high 3.17; Fig. 4A) than in the lowest three SHBG quartiles (HR SHBG was 14.6%.
0.99; 95% CI 0.73–1.35; Fig. 4A). These associations re-mained after adjustment for bioE2 and did not appear tobe from a shift in the SHBG distribution; median SHBG levels did not differ between low and high bioT groups(69.0 vs. 70.4 nM, respectively, P ⫽ 0.7), and adjustment In this large prospective study of older men, those with the of the models with an SHBG2 term did not affect the in- lowest bioE2 or the highest SHBG had higher risks of teraction. We evaluated whether the stronger association nonvertebral fracture. BioT had a weak association with of bioT with fracture risk in the highest SHBG quartile nonvertebral fracture that disappeared after adjustment could have been due to particularly low levels of bioT in for bioE2. The association between bioT and nontrau- the high SHBG group. The median bioT levels within the matic fracture risk was stronger and remained after ad- lowest bioT quartile were slightly lower in the highest justment for bioE. When high SHBG levels are present, SHBG quartile compared with the lower quartiles [127.7 low bioT was associated with a substantially increased vs. 138.7 ng/dl (4.43 vs. 4.81 nmol/liter, respectively,) P ⫽ fracture risk even with bioE2 adjustment. The associations 0.03]. However, in age-, race-, and BMI-adjusted models, were similar, perhaps slightly stronger, for hip fracture.
even very low levels of bioT were not associated with in- Total sex steroids were not associated with fracture. These creases in fracture risk (Fig. 2B). These results indicate that results have important implications for understanding low concentrations of bioT impart particular risk in the how sex steroids and SHBG affect fracture risk and for presence of high SHBG.
determining the clinical role of these measurements.
Our finding that low bioE2 was independently associ- Combinatorial effects of estradiol, testosterone,
ated with increased fracture risk extends earlier reports of and SHBG on fracture risk
estrogen's importance for men's skeletal health (11, 19, When the combined effects of sex steroid or SHBG lev- 29). Previous studies evaluating the sex steroid-fracture els were examined, the associations with fracture risk were association have been inconsistent and limited by cross- strengthened. The highest nonvertebral fracture risk was sectional design, low participant and fracture numbers, in men (n ⫽ 74, 3.7%) in the lowest quartiles of bioT and and/or RIA-based sex steroid measurements (11, 20, 22, bioE2 and highest quartile of SHBG (HR 3.39; 95% CI 23, 25, 29, 44). Two recent studies used mass spectrom- 2.19 –5.27; Fig. 4B). Risk estimates were similar or stron- etry to more accurately measure testosterone and estra- ger when only nontraumatic fractures were included in the diol. In the Dubbo cohort, total testosterone had a strong analyses; in the lowest quartiles of bioT and bioE2 and and estradiol a weak association with osteoporotic frac- highest quartile of SHBG, the HR was 4.02 (95% CI 2.54 – ture risk, (21), but independent effects were not assessed.
6.37). The effects of combining high-risk categories were Another large, prospective study (MrOS Sweden) (19) also evident for hip fracture; men with low bioE2 and bioT found that lower free and total estradiol were associated and high SHBG levels had a 3.8-fold higher risk of hip with nonvertebral and vertebral fracture risk. Their results fracture (95% CI 1.48 –9.92).
are very similar to ours and together provide compellingevidence for estradiol's effects on fracture risk. Attenua- tion of bioE2's association with fracture by adjustment for The fraction of nonspine fracture risk attributable to BMD suggests that estradiol's positive effects on fracture low bioE2 was 5.7%, 1.5% to low bioT, and 7.7% to high risk may be due, in part, to an effect on bone density (7, 9, J Clin Endocrinol Metab, September 2009, 94(9):3337–3346 (log stioa r 1.0 High bioE2 Low bioE2 High bioE2 Low bioE2 High bioT
FIG. 4. Combinations of sex steroids and SHBG and risk of
nonvertebral fracture. A, bioT and SHBG. B, bioT, bioE2, and SHBG.
There were 1079 men in the high bioT, low SHBG category; 397 menin the high bioT, high SHBG category; 392 men in the low bioT, lowSHBG category; and 110 men in the low bioT, high SHBG category.
䡺, Low SHBG; F, high SHBG.
identified similar thresholds of bioE2 below which frac-ture risk was increased [11.4 –12.5 pg/ml (41.8 – 45.9pmol/liter); free estradiol: 0.4 – 0.5 pg/ml (1.47–1.84pmol/liter)]. MrOS Sweden found a similar fracture riskthreshold level [free estradiol: 0.3 pg/ml (1.10 pmol/liter)] (19). Together these results support the hypoth-esis that a threshold range of bioE2 is necessary forskeletal health (45).
High SHBG levels were associated with increased non- FIG. 3. Spline models for the detection of any nonlinear relationships
vertebral fracture risk, independent of sex steroids and between sex steroids or SHBG and nonvertebral fracture risk. A, bioE2.
BMD. SHBG has been associated with bone density (22, B, bioT. C, SHBG. To convert bioavailable estradiol to picomoles perliter, the conversion factor is 3.671; to convert bioavailable 46), bone turnover markers (22, 46), proximal femur ex- testosterone to nanomoles per liter, the conversion factor is 0.0347.
pansion and bending resistance (47), and fracture risk inmen (19, 22, 46) and women (24). SHBG may directly 11). However, the association remained significant after influence intracellular signaling via a membrane receptor BMD adjustment, suggesting additional effects.
that requires SHBG-sex steroid interactions (26, 27) or a We found a nonlinear association between estradiol megalin-mediated endocytic pathway that involves un- and fracture risk. Evaluations using quartile analysis, bound SHBG (26, 28). Through these pathways, SHBG spline analysis, and log likelihood cut point analysis could amplify the effects of sex steroid sufficiency or de- LeBlanc et al. Sex Steroids and Fracture in Men J Clin Endocrinol Metab, September 2009, 94(9):3337–3346 ficiency (26). However, SHBG could also be a marker for Our results have potential clinical implications. They nonskeletal factors affecting fracture risk. Lower insulin affirm the robust and independent effects of bioE2 and or IGF-I levels could increase SHBG, resulting in the SHBG in fracture prediction. Moreover, we provide fur- SHBG-fracture risk association. SHBG increases with age ther evidence for a threshold level of bioE2, below which but decreases with obesity. It is affected by frailty and fracture risk is increased. Hence, estradiol and SHBG mea- nutritional status. In our study adjustment for age, leg surements should be valuable in clinical situations. Al- power, physical activity, BMI, and previous falls did not though estradiol and SHBG levels are not commonly mea- alter the association between SHBG and fracture risk.
sured when assessing skeletal health or fracture risk in Despite strong cellular and animal data suggesting an- men, our results and those of MrOS Sweden (19) suggest drogens have positive bone effects, clinical studies offer no revision of these practices (49). Second, our results sup- clear evidence of an independent androgen effect on bone port previous findings that bioavailable or free levels of mass or fracture (11, 20, 22, 23, 25, 44). Consistent with sex steroids are more robustly associated with fracture risk previous reports (19, 21), we found men with low bioT than are total sex steroid concentrations. Although some had higher fracture risk, but the association weakened investigators argue that total T and total E are biologically when adjusted for bioavailable estradiol. The association more relevant than bioT or bioE2, our results suggest that was more robust when only nontraumatic fractures were bioavailable, not total, levels are associated with fracture considered, suggesting a stronger link with osteoporotic risk. It remains common to measure total sex steroid levels fractures. This could be a reflection of low testosterone's in clinical situations; however, bioavailable or free levels effects on fall risk (48), potentially mediated through ex- may be more appropriate as predictive tools. Given the traskeletal functions including muscle strength, physical limitations of the analog free testosterone assays, clinical activity, and cognition (13–18). Indeed, the association application of these findings would require more accurate between low bioT and fracture risk was not attenuated by and standardized assay methods and development of con- BMD adjustment, suggesting non-BMD-related factors sensus concerning assay result use in clinical decision mak- are important.
ing. Third, the associations we observed were most ap- We found novel evidence of a bioT-SHBG interaction.
parent when sex steroids and SHBG were considered in Men with low bioT and high SHBG were at substantially combination. Men with low bioT and bioE2 and high higher risk of nonvertebral and hip fracture even after SHBG levels are at highest risk. If validated, approaches adjustment for bioE2. Men with low bioT and bioE2 and that incorporate all three measures into clinical algorithms high SHBG had even greater risk of nonvertebral (HR 3.4) should be developed.
and hip fracture (HR 3.8), especially when only nontrau- This study has several limitations. We did not measure matic fractures were considered. Thus, bioT, bioE2, and changes in sex steroids and SHBG over time so cannot SHBG each play a role in fracture determination, but the determine how hormonal changes associate with fracture cumulative effects of sex steroid, and SHBG levels may be risk. Use of dichotomous cutoffs for sex steroid levels were most important. Although the findings in MrOS Sweden based on observed associations with fracture and could (19) are similar to ours, combinatorial effects of sex ste- have overestimated the associations. The cohort was rel- roids and SHBG have rarely been reported. Given these atively healthy and primarily Caucasian and although results, combinatorial effects should be evaluated in ad- similar to more representative populations such as Na- ditional studies and with other endpoints (e.g. bone loss, tional Health and Nutrition Examination Survey, caution body composition changes, cardiovascular events, mor- should be used in generalizing our results to other groups tality). However, these results should be interpreted with of men. The number of hip fractures that occurred during caution because delineating each hormone's independent follow-up was relatively small, but nevertheless, the asso- effect on fracture risk by statistical methods is challenging ciations between sex steroid and SHBG levels and hip frac- in the presence of complex interrelationships among ture risk were robust. Our findings need to be validated in bioE2, bioT, and SHBG. This is a particular issue in our other cohorts of older men.
study because bioavailable levels were derived from mass This study also has considerable strengths. It is one of action equations that included SHBG. Nevertheless, sev- the largest to address the association between sex steroids eral analytical approaches (see Results) provided consis- and fracture risk in elderly men. Fractures were carefully tent evidence of a nonartifactual interaction between ascertained and verified, potentially important confound- SHBG and bioT. Our findings cannot be considered proof ing variables were evaluated, and sex steroid measure- of independent molecular effects of bioavailable sex ste- ments were performed using gas chromatography/mass roids and SHBG, but they are consistent with that spectrometry to avoid inaccuracy at low concentrations (30, 31). Many participants are over age 80 yr, a segment J Clin Endocrinol Metab, September 2009, 94(9):3337–3346 of the population that is expanding and is at high fracture Segre GV, Crowley Jr WF 1989 Increases in bone density during
risk but has not been well studied.
treatment of men with idiopathic hypogonadotropic hypogonad-ism. J Clin Endocrinol Metab 69:776 –783 In summary, men with low bioE2 levels and high SHBG 6. Khosla S 2004 Role of hormonal changes in the pathogenesis of
levels had increased rates of incident fractures. Low bioT osteoporosis in men. Calcif Tissue Int 75:110 –113 was associated with an increased risk of nontraumatic 7. Khosla S, Melton III LJ, Robb RA, Camp JJ, Atkinson EJ, Oberg AL,
Rouleau PA, Riggs BL 2005 Relationship of volumetric BMD and
fractures and there was an interaction between SHBG and structural parameters at different skeletal sites to sex steroid levels bioT; men with low bioT were at higher risk in the pres- in men. J Bone Miner Res 20:730 –740 ence of high SHBG levels. Men who were in the highest- 8. Khosla S, Melton III LJ, Atkinson EJ, O'Fallon WM 2001 Relation-
risk quartiles for bioT, bioE2, and SHBG had a markedly ship of serum sex steroid levels to longitudinal changes in bonedensity in young versus elderly men. J Clin Endocrinol Metab 86: increased fracture risk. Our results suggest that bioavail- able sex steroid and SHBG measurements may be useful in 9. Szulc P, Munoz F, Claustrat B, Garnero P, Marchand F, Duboeuf F,
the clinical assessment of fracture risk in older men and Delmas PD 2001 Bioavailable estradiol may be an important deter-
minant of osteoporosis in men: the MINOS study. J Clin Endocrinol
that the physiological implications of hypogonadism Metab 86:192–199 should be considered in light of possible interactions 10. Slemenda CW, Longcope C, Zhou L, Hui SL, Peacock M, Johnston
among sex steroids and SHBG.
CC 1997 Sex steroids and bone mass in older men. Positive associ-
ations with serum estrogens and negative associations with andro-
gens. J Clin Invest 100:1755–1759
11. Amin S, Zhang Y, Sawin CT, Evans SR, Hannan MT, Kiel DP,
Wilson PW, Felson DT 2000 Association of hypogonadism and
estradiol levels with bone mineral density in elderly men from the
Framingham study. Ann Intern Med 133:951–963
We thank Lori Lambert for her statistical work on previous ver- 12. Beck TJ, Oreskovic TL, Stone KL, Ruff CB, Ensrud K, Nevitt MC,
sions of this manuscript.
Genant HK, Cummings SR 2001 Structural adaptation to changing
skeletal load in the progression toward hip fragility: the study of
Address all correspondence and requests for reprints to: Eric osteoporotic fractures. J Bone Miner Res 16:1108 –1119 Orwoll, M.D., Bone and Mineral Unit (CR 113), Oregon Health 13. Rudman D, Drinka PJ, Wilson CR, Mattson DE, Scherman F, Cui-
and Science University, 3181 SW Sam Jackson Park Road, Port- sinier MC, Schultz S 1994 Relations of endogenous anabolic hor-
land Oregon 97239. E-mail: firstname.lastname@example.org.
mones and physical activity to bone mineral density and lean body This work was supported by the Osteoporotic Fractures in mass in elderly men. Clin Endocrinol (Oxf) 40:653– 661 Men by National Institutes of Health (NIH) funding. The fol- 14. Szulc P, Claustrat B, Marchand F, Delmas PD 2003 Increased risk
of falls and increased bone resorption in elderly men with partial lowing institutes provided support: the National Institute of Ar- androgen deficiency: the MINOS study. J Clin Endocrinol Metab thritis and Musculoskeletal and Skin Diseases, the National In- stitute on Aging, the National Center for Research Resources, 15. Szulc P, Duboeuf F, Marchand F, Delmas PD 2004 Hormonal and
and NIH Roadmap for Medical Research under the following lifestyle determinants of appendicular skeletal muscle mass in men: grant numbers: U01 AR45580, U01 AR45614, U01 AR45632, the MINOS study. Am J Clin Nutr 80:496 –503 U01 AR45647, U01 AR45654, U01 AR45583, U01 AG18197, 16. Roy TA, Blackman MR, Harman SM, Tobin JD, Schrager M, Metter
U01-AG027810, and UL1 RR024140. Additional support for EJ 2002 Interrelationships of serum testosterone and free testosterone
these analyses was provided by Merck & Co., Eli Lilly, and index with FFM and strength in aging men. Am J Physiol Endocrinol Amgen and NIH Grant AR049828.
Metab 283:E284–E294 Disclosure Statement: E.S.L., C.M.N., L.M.M., J.A.L., 17. Barrett-Connor E, Goodman-Gruen D, Patay B 1999 Endogenous
K.E.E., A.R.H., G.L., C.O., and E.S.O. have nothing to disclose.
sex hormones and cognitive function in older men. J Clin EndocrinolMetab 84:3681–3685 E.B.-C. has received grant support and/or consulting fees from 18. Moffat SD, Zonderman AB, Metter EJ, Blackman MR, Harman
the National Institutes of Health; Amgen; Eli Lilly and Co.; SM, Resnick SM 2002 Longitudinal assessment of serum free tes-
Merck & Co., Inc.; Pfizer Pharmaceuticals; Proctor & Gamble; tosterone concentration predicts memory performance and cogni- Roche; and Amylin. This financial support does not represent a tive status in elderly men. J Clin Endocrinol Metab 87:5001–5007 conflict of interest.
19. Mellstrom D, Vandenput L, Mallmin H, Holmberg AH, Lorentzon
M, Oden A, Johansson H, Orwoll ES, Labrie F, Karlsson MK, Ljung-
gren O, Ohlsson C 2008 Older men with low serum estradiol and
high serum SHBG have an increased risk of fractures. J Bone Miner
Res 23:1552–1560 20. Center JR, Nguyen TV, Sambrook PN, Eisman JA 2000 Hormonal
1. Orwoll ES 2003 Men, bone and estrogen: unresolved issues. Osteo-
and biochemical parameters and osteoporotic fractures in elderly poros Int 14:93–98 men. J Bone Miner Res 15:1405–1411 2. Davidson JM, Chen JJ, Crapo L, Gray GD, Greenleaf WJ, Catania
21. Meier C, Nguyen TV, Handelsman DJ, Schindler C, Kushnir MM,
JA 1983 Hormonal changes and sexual function in aging men. J Clin
Rockwood AL, Meikle AW, Center JR, Eisman JA, Seibel MJ 2008
Endocrinol Metab 57:71–77 Endogenous sex hormones and incident fracture risk in older men: 3. Orwoll E, Lambert LC, Marshall LM, Phipps K, Blank J, Barrett-
the Dubbo Osteoporosis Epidemiology Study. Arch Intern Med 168: Connor E, Cauley J, Ensrud K, Cummings S 2006 Testosterone and
estradiol among older men. J Clin Endocrinol Metab 91:1336 –1344 22. Legrand E, Hedde C, Gallois Y, Degasne I, Boux de CF, Mathieu E,
4. Cooper C, Campion G, Melton III LJ 1992 Hip fractures in the
Basle MF, Chappard D, Audran M 2001 Osteoporosis in men: a
elderly: a world-wide projection. Osteoporos Int 2:285–289 potential role for the sex hormone binding globulin. Bone 29:90 –95 5. Finkelstein JS, Klibanski A, Neer RM, Doppelt SH, Rosenthal DI,
23. Bjornerem A, Ahmed LA, Joakimsen RM, Berntsen GK, Fonnebo V,
LeBlanc et al. Sex Steroids and Fracture in Men J Clin Endocrinol Metab, September 2009, 94(9):3337–3346 Jorgensen L, Oian P, Seeman E, Straume B 2007 A prospective study
study—a large observational study of the determinants of fracture in of sex steroids, sex hormone-binding globulin, and non-vertebral older men. Contemp Clin Trials 26:569–585 fractures in women and men: the Tromso Study. Eur J Endocrinol 35. Blank JB, Cawthon PM, Carrion-Petersen ML, Harper L, Johnson
JP, Mitson E, Delay RR 2005 Overview of recruitment for the os-
24. Cummings SR, Browner WS, Bauer D, Stone K, Ensrud K, Jamal S,
teoporotic fractures in men study (MrOS). Contemp Clin Trials Ettinger B 1998 Endogenous hormones and the risk of hip and
vertebral fractures among older women. Study of Osteoporotic 36. Washburn RA, Smith KW, Jette AM, Janney CA 1993 The Physical
Fractures Research Group. N Engl J Med 339:733–738 Activity Scale for the Elderly (PASE): development and evaluation.
25. Goderie-Plomp HW, van der Klift M, de Ronde W, Hofman A, de
J Clin Epidemiol 46:153–162 Jong FH, Pols HA 2004 Endogenous sex hormones, sex hormone-
37. Sodergard R, Backstrom T, Shanbhag V, Carstensen H 1982 Cal-
binding globulin, and the risk of incident vertebral fractures in el- culation of free and bound fractions of testosterone and estradiol- derly men and women: the Rotterdam Study. J Clin Endocrinol 17␤ to human plasma proteins at body temperature. J Steroid Bio- Metab 89:3261–3269 chem 16:801– 810 26. Kahn SM, Hryb DJ, Nakhla AM, Romas NA, Rosner W 2002 Sex
38. Heinzl H, Kaider A 1997 Gaining more flexibility in Cox propor-
hormone-binding globulin is synthesized in target cells. J Endocrinol tional hazards regression models with cubic spline functions. Com- put Methods Programs Biomed 54:201–208 27. Rosner W, Hryb DJ, Khan MS, Nakhla AM, Romas NA 1999 An-
39. Tableman M, Kim JS 2003 Survival analysis using S: analysis of
drogen and estrogen signaling at the cell membrane via G-proteins time-to-event data. Boca Raton, FL: CRC Press; 172–175 and cyclic adenosine monophosphate. Steroids 64:100 –106 40. Li R, Chambless L 2007 Test for additive interaction in proportional
hazards models. Ann Epidemiol 17:227–236 28. Hammes A, Andreassen TK, Spoelgen R, Raila J, Hubner N, Schulz
41. Barlow WE, Ichikawa L, Rosner D, Izumi S 1999 Analysis of case-
H, Metzger J, Schweigert FJ, Luppa PB, Nykjaer A, Willnow TE
cohort designs. J Clin Epidemiol 52:1165–1172 2005 Role of endocytosis in cellular uptake of sex steroids. Cell 42. Eide GE, Gefeller O 1995 Sequential and average attributable frac-
tions as aids in the selection of preventive strategies. J Clin Epidemiol 29. Barrett-Connor E, Mueller JE, von Muhlen DG, Laughlin GA,
Schneider DL, Sartoris DJ 2000 Low levels of estradiol are associ-
43. Ruckinger S, von Kries R, Toschke AM 2009 An illustration of and
ated with vertebral fractures in older men, but not women: the Ran- programs estimating attributable fractions in large scale surveys cho Bernardo Study. J Clin Endocrinol Metab 85:219 –223 considering multiple risk factors. BMC Med Res Methodol 9:7 30. Taieb J, Mathian B, Millot F, Patricot MC, Mathieu E, Queyrel N,
44. Nyquist F, Gardsell P, Sernbo I, Jeppsson JO, Johnell O 1998 As-
Lacroix I, Somma-Delpero C, Boudou P 2003 Testosterone mea-
sessment of sex hormones and bone mineral density in relation to sured by 10 immunoassays and by isotope-dilution gas chromatog- occurrence of fracture in men: a prospective population-based raphy-mass spectrometry in sera from 116 men, women, and chil- study. Bone 22:147–151 dren. Clin Chem 49:1381–1395 45. Khosla S 2008 Estrogen and bone: insights from estrogen-resistant,
31. Stanczyk FZ, Cho MM, Endres DB, Morrison JL, Patel S, Paulson
aromatase-deficient, and normal men. Bone 43:414 – 417 RJ 2003 Limitations of direct estradiol and testosterone immuno-
46. Lormeau C, Soudan B, d'Herbomez M, Pigny P, Duquesnoy B, Cor-
assay kits. Steroids 68:1173–1178 tet B 2004 Sex hormone-binding globulin, estradiol, and bone turn-
32. Siekmann L 1979 Determination of steroid hormones by the use of
over markers in male osteoporosis. Bone 34:933–939 isotope dilution-mass spectrometry: a definitive method in clinical 47. Kaptoge S, Dalzell N, Folkerd E, Doody D, Khaw KT, Beck TJ, Lov-
chemistry. J Steroid Biochem 11:117–123 eridge N, Mawer EB, Berry JL, Shearer MJ, Dowsett M, Reeve J 2007
33. Lawson AM, Gaskell SJ, Hjelm M 1985 International Federation of
Sex hormone status may modulate rate of expansion of proximal femur Clinical Chemistry (IFCC), Office for Reference Methods and Ma- diameter in older women alongside other skeletal regulators. J Clin terials (ORMM). Methodological aspects on quantitative mass Endocrinol Metab 92:304–313 spectrometry used for accuracy control in clinical chemistry. J Clin 48. Orwoll E, Lambert LC, Marshall LM, Blank J, Barrett-Connor E,
Chem Clin Biochem 23:433– 441 Cauley J, Ensrud K, Cummings SR 2006 Endogenous testosterone
34. Orwoll E, Blank JB, Barrett-Connor E, Cauley J, Cummings S,
levels, physical performance, and fall risk in older men. Arch Intern Ensrud K, Lewis C, Cawthon PM, Marcus R, Marshall LM, McGowan
Med 166:2124 –2131 J, Phipps K, Sherman S, Stefanick ML, Stone K 2005 Design and base-
49. Gennari L, Khosla S, Bilezikian JP 2008 Estrogen and fracture risk
line characteristics of the osteoporotic fractures in men (MrOS) in men. J Bone Miner Res 23:1548 –1551
© Alturos Ltd 2006-2013 Applying Process Improvement in an NHS Pharmacy to Reduce Cost, Increase Productivity and Create a Better Working Environment In July 2010, the Pharmacy at one site of one of London's largest NHS Foundation Trust's, started applying Lean working methods. Since that time, a number of projects have developed. This article
the abortion pill by W. David Hager, M.D. A positive pregnancy test is one of the most life-changing moments for a woman. Never is it more important to base your decisions on accurate information. Try to think beyond the pressures you face right now and consider the long-term impact of your choices. You may have considered —or someone around you may have suggested—having an abortion.