| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Original Studies |
Endocrine Research Unit (S.K., B.L.R.); Division of Endocrinology, Metabolism, and Nutrition; Departments of Internal Medicine, Health Sciences Research (L.J.M., E.J.A., W.M.O.), and Laboratory Medicine and Pathology (G.G.K.), Mayo Clinic and Mayo Foundation, Rochester, Minnesota 55905
Address all correspondence and requests for reprints to: Sundeep Khosla, M.D., Mayo Clinic, 200 First Street SW, 5–164 West Joseph, Rochester, Minnesota 55905. E-mail: khosla.sundeep{at}mayo.edu
 |
Abstract |
|---|
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Estrogen (E) deficiency has clearly been identified as a major risk factor for osteoporosis in women, and the effects of E on the skeleton have been the subject of intensive investigation. Recent evidence indicates that E deficiency may be responsible not only for the early postmenopausal phase of rapid bone loss, but also for the late slow phase of bone loss associated with aging. The latter had been attributed at least in part to an age-related increase in bone turnover, due largely to increases in serum PTH levels with aging (5, 6, 7). E therapy of elderly postmenopausal women, however, appears to prevent the secondary hyperparathyroidism and increased bone turnover associated with aging (7, 8). This has led to the hypothesis that the extraskeletal effects of E, namely on intestinal calcium absorption (9, 10), renal calcium handling (11, 12), and perhaps direct effects on PTH secretion (13), may be responsible for preventing the age-related increase in serum PTH and in bone turnover in late postmenopausal women treated with E (14).
Despite the fact that men do not have the equivalent of menopause and that serum total testosterone (T) levels decrease only marginally with age (15, 16, 17), men have substantial age-related decreases in bone mineral density (BMD) in both cross-sectional (18) and longitudinal (19) studies. Moreover, previous epidemiological studies assessing the relationship between serum total T levels and BMD have generally found either no relationship (17, 20) or even a negative association between total T levels and BMD in aging men (21). Thus, the absence of substantial decreases in serum total T levels in aging men has led to the belief that T does not play a major role in bone loss in aging men.
Recent studies have suggested instead the possibility that, as in women, E may play a key role in regulating bone turnover in men. Smith et al. (22) described a male with homozygous mutations in the E receptor gene who, even in the presence of normal T and free T levels, had unfused epiphyses and marked osteopenia, along with elevated indexes of bone turnover. Subsequently, Morishima et al. (23) and Carani et al. (24) reported clinical findings in two males with homozygous mutations of the gene that codes for the enzyme, aromatase, which is responsible for the conversion of androgens to E. In both instances, BMD was significantly reduced, and bone turnover markers were markedly elevated despite normal T levels. Treatment with T did not significantly affect bone metabolism in one patient (24), whereas treatment with E markedly increased BMD in both patients (24, 25). Finally, a recent epidemiological study by Slemenda et al. (21) found that serum estradiol (E2), but not T, levels were positively associated with BMD in men over age 65 yr.
Despite these findings, a major conceptual problem with attributing age-related bone loss in men to E deficiency is that, as in the case of total T levels, serum total E levels decline only marginally with age in men (15). However, there are several important limitations of previous studies attempting to define the relationship of T or E to BMD in men. First, most have studied subjects in a narrow age range, such as elderly men (20, 21, 26). Second, they have failed to measure levels of circulating bioavailable sex steroids. The bioavailable sex steroids comprise the fractions that are free or associated with albumin in the circulation (27, 28, 29, 30), and it is these fractions that have rapid access to target tissues (31, 32). In contrast, the fraction bound to sex hormone-binding globulin (SHBG) does not have free access to target tissues. As SHBG levels increase with age in men (33, 34), measurement of total T or E levels does not accurately reflect the actual levels of these steroids available to tissues. Moreover, although several studies have measured circulating free T and E levels (21, 26), the free fraction constitutes only 1–3% of the total circulating sex steroids (30), and failure to account for the 35–55% of the circulating steroids bound to albumin vastly underestimates the proportion available to target tissues.
In the present study, we addressed these limitations in several ways. First, we defined the age-related changes in circulating bioavailable T and E levels in a population-based, age-stratified sample of men, aged 23–90 yr (n = 346). Next, we assessed age-related changes in serum PTH and in markers of bone turnover in these men and related these to BMD and the sex steroids. We also assessed the relative importance of bioavailable T vs. E levels in determining BMD in men by multivariate analyses. Finally, we performed similar studies in a parallel cohort of 304 women so that these relationships in aging men vs. women could be compared.
|
Subjects and Methods |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Subjects were recruited from an age-stratified random sample of Rochester, MN, men and women that were selected using the medical records linkage system of the Rochester Epidemiology Project (35). Over half of the Rochester population is identified annually in this system, and the majority are seen in any 3-yr period. Thus, the enumerated population approximates the underlying population of the community, including both free-living and institutionalized individuals. Altogether 1138 men and 938 women aged 20 yr and over were approached, but 239 men and 126 women were ineligible (among the men, 109 were demented and could not give informed consent, 13 were radiation workers, 91 died before contact, 25 were debilitated due to terminal cancer, and 1 was unable to participate due to pending legal action; among the women, 89 were demented, 11 were pregnant, 9 were radiation workers, 8 were participants in an ongoing clinical trial of osteoporosis prophylaxis, and 9 died before they could be contacted). Of the 899 eligible men, 348 participated and provided full study data, although 2 were excluded from analysis because 1 was receiving T therapy and 1 had inexplicably high (into the range of premenopausal women) E2 and estrone (E1) levels. Of the 812 eligible women, 351 participated and provided full study data, although 47 of the 213 postmenopausal women were receiving E replacement therapy and were excluded from this analysis. Thus, the total number of subjects included in this analysis was 650 (346 men and 304 women). All but 13 men and 3 women were Caucasian, reflecting the ethnic composition of the population (96% white in 1990). The men ranged in age from 23–90 yr, and the women ranged in age from 21–94 yr.
As this sample of subjects was population based, the overall results are applicable to the general population of men and women in the community. However, we also performed subset analyses in normal subjects, excluding those with rheumatoid arthritis (4 men and 5 women), Paget’s disease (5 men), gastrointestinal resection (11 men and 1 woman), significant renal insufficiency (defined as serum creatinine >2 mg/dL, 6 men), chronic obstructive pulmonary disease (14 men and 3 women), prostate cancer or bilateral orchidectomies (12 men), premature menopause (at <35 yr of age; 1 woman), or current therapy with corticosteroids (7 men and 7 women), thiazide diuretics (21 men and 33 women), anticonvulsants (2 men and 2 women), or oral contraceptives (28 premenopausal women). After all of these exclusions, the subset of normal men and women consisted of 280 and 231 subjects, respectively.
Study protocol
BMD (grams per cm2) was determined for the total body, spine (L2–L4), proximal femur (total), and middistal radius using dual energy x-ray absorptiometry with the Hologic QDR2000 instrument (Hologic, Waltham, MA) using software version 5.40. As we did not specifically exclude subjects with spinal osteoarthritis or aortic calcification, which can confound the BMD measurement (36), we assessed the midlateral instead of the antero-posterior spine, which largely excludes these confounders from the scanning field. The coefficients of variation (CVs) for the total body, lateral spine, femur, and radius were 0.8%, 2.1%, 1.8%, and 1.7%, respectively.
Fasting state serum samples were obtained between 0800–0900 h, and a 24-h urine collection was completed. All samples were stored at -70 C until analyzed.
Laboratory methods
Fasting serum samples were assayed by RIA for total T (Diagnostic Products Corp., Los Angeles, CA; interassay CV, 11%), E2 (Diagnostic Systems Laboratories, Webster, TX; interassay CV, 11%), E1 (Diagnostic Systems Laboratories; interassay CV, 9%), and SHBG (Wien Laboratories, Succasunna, NJ; interassay CV, 7%). In addition, the non-SHBG-bound (bioavailable) fractions of total T, E2, and E1 were measured using a modification of the techniques of O’Connor et al. (27) and Tremblay et al. (28). Briefly, tracer amounts of 3H-labeled T, E2, or E1 were added to serum aliquots. An equal volume of saturated solution of ammonium sulfate (final concentration, 50%) was added to precipitate SHBG with its bound steroid. Separation of the SHBG fraction was performed by centrifugation at 1100 x g for 30 min at 4 C. The percentage of labeled steroid remaining in the supernatant (the free and albumin-bound fractions) was then calculated. The bioavailable steroid concentration was obtained by multiplying the total steroid concentration, as determined by RIA, by the fraction that was non-SHBG bound. Free T was measured by RIA using a T analog with low affinity for SHBG and albumin (Diagnostic Systems Laboratories; interassay CV, 10%). Serum dehydroepiandrosterone sulfate (DHEAS) was measured by RIA (Diagnostic Products Corp.; interassay CV, <8%). Serum LH and FSH were measured by immunoradiometric assays (Diagnostic Products Corp.; interassay CV, 14% for LH and 11% for FSH).
Serum intact PTH was measured by immunochemiluminometric assay (37) (interassay CV, 14%). Bone formation was assessed by serum osteocalcin, measured by RIA using antibody G12 (interassay CV, <6%) (5), as well as by serum bone alkaline phosphatase (BAP), measured by enzyme-linked immunosorbent assay (5) (ELISA; interassay CV, <11%), and serum carboxyl-terminal propeptide of type I collagen (PICP), also measured by ELISA (Prolagen-C, Metra Biosystems, Mountain View, CA; interassay CV, <7%). Bone resorption was evaluated by measurement of 24-h urinary levels of the N-telopeptide of type I collagen (NTx) and free deoxypyridinoline (f-Dpd), both assessed as nanomoles per L glomerular filtrate. NTx and f-Dpd were measured by ELISA kits (Osteomark, Ostex, Seattle, WA; interassay CV, 10%; and Pyrilinks-D, Metra Biosystems, Mountain View, CA; interassay CV, 14%). The glomerular filtration rate was assessed by creatinine clearance.
Statistical analysis
Pearson correlations were used to summarize relationships between the various continuous variables. Log transformations were used on highly skewed variables. When appropriate, Spearman correlations were used. The percent change from age 25–85 yr was estimated from predictions using a loess model in S-plus (38), using the following formula: percent change = [(value at age 85 yr - value at age 25 yr)/value at age 25 yr] x 100. The smoother function was also used as a means to visually explore the data in the various plots. Stepwise model selection was used to determine the relationship of sex steroid variables and bone turnover markers with BMD. Higher ordered terms and interactions were checked, but were only included if they made a meaningful change to the model. Finally, model assumptions were checked by examination of the model residuals.
|
Results |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Table 1
shows the correlation
coefficients between BMD and age in the men and women. In this
cross-sectional analysis, BMD declined significantly with age in both
sexes at all sites, and Table 1 also indicates the percent decrease
from age 25–85 yr in BMD at the various sites. Thus, at all sites, the
percent decrease in BMD was approximately twice as great in the women
as it was in the men.
|
also shows the age-related changes in serum sex steroid levels
in the men and women. Between age 25–85 yr, the decline in
bioavailable T was almost twice that in total T in both sexes, and
bioavailable T decreased by approximately 64% and 28% in the men and
women, respectively (Table 1 and Fig. 1, A and B). In the men, total E (E2 plus
E1) declined only marginally with age, whereas
bioavailable E decreased by 47% (Table 1 and Fig. 2, A and B). In contrast, total and
bioavailable E levels decreased similarly in the women. For simplicity,
we have presented the data for total and bioavailable
E2 plus E1, although
E2 and E1 individually
showed similar changes. As expected, there was a precipitous drop in
serum total and bioavailable E levels in the women around the time of
the menopause, whereas there was a gradual, age-related decrease in
bioavailable E levels in the men (Fig. 2, A and B). There was also a
marked difference between sexes in changes in SHBG with age; SHBG
increased by more than 2-fold over the life span in the men, whereas it
was virtually unchanged in the women (Table 1 and Fig. 3). Thus, the greater decrease in
bioavailable T and E in the men was explained in large part by the
age-related increase in SHBG levels. Finally, serum DHEAS levels also
decreased significantly with age in both sexes (Table 1).
|
|
|
). Thus, the
age-related decreases in the molar ratios of T to SHBG and E to SHBG
were comparable to the decreases in directly measured bioavailable T
and E, respectively. In addition, the free T level, which represents
the T fraction not bound to either SHBG or albumin, decreased with age
in a manner similar to bioavailable T in both sexes.
The age-related decreases in sex steroids were accompanied by parallel
increases in serum LH and FSH levels (Table 1). Of note, serum LH in
the men did not correlate with serum total T or E levels (r = 0.03
and -0.04, respectively), but was inversely correlated with serum
bioavailable T and E levels (r = -0.35 and r = -0.29,
respectively; both P < 0.001). Similarly, serum FSH in
the men was only weakly inversely correlated with serum total T and E
levels (r = -0.11; P = 0.05; and r = -0.15;
P = 0.01, respectively), but showed a strong negative
association with serum bioavailable T and E levels (r = -0.46 and
r = -0.35, respectively; both P < 0.001). In
contrast, in the women, serum LH and FSH were both inversely associated
with serum total and bioavailable E levels (r = -0.19 and r
= -0.21, respectively; both P < 0.001 for total and
bioavailable E vs. LH; and r = -0.54 and -0.55,
respectively; both P < 0.001 for total and
bioavailable E vs. FSH). However, neither serum LH nor FSH
was correlated with T levels in the women (data not shown).
Age-related changes in serum PTH and bone turnover markers
Figure 4, A–C, shows the changes in
serum PTH, serum osteocalcin, and urinary NTx excretion as a function
of age in the men and women. In these cross-sectional data, serum PTH
increased similarly with age in both sexes; over the life span, the
percent increase in serum PTH was 84% in the men and 64% in the
women. Serum osteocalcin and urinary NTx excretion decreased with age
in the men up to approximately age 50 yr and increased with age
thereafter. Similarly, serum osteocalcin and urinary NTx decreased with
age in the premenopausal women and increased with age in the
postmenopausal women. Table 2 shows the
correlation coefficients between all of the measured bone turnover
markers and age in the men and women. Serum BAP had an age-related
pattern similar to that of serum osteocalcin, although none of the
individual correlation coefficients was statistically significant.
Serum PICP decreased with age in the men up to age 50 yr and in the
premenopausal women, and then showed little or no change with age in
the men over age 50 yr or in the postmenopausal women. Urinary f-Dpd
excretion showed age-related changes similar to those for urinary NTx
excretion.
|
|
Table 3 shows the results of the
univariate analyses relating BMD at the various skeletal sites to the
sex steroid and DHEAS levels in the men and women. In the men, serum
total T levels did not correlate with BMD at any site, except weakly at
the radius. In contrast, serum bioavailable T levels were significantly
correlated with BMD at all sites in the men. In the women, serum total
and bioavailable T levels were equally correlated with BMD at the
various sites. Serum total and bioavailable E levels were correlated
with BMD at the various sites in the men, although, as for T, the
correlations in the men were considerably stronger for bioavailable as
opposed to total E. In contrast, serum total and bioavailable E levels
were equally correlated with BMD in the women. Serum DHEAS levels were
also correlated with BMD in both sexes.
|
shows the results of the
univariate analyses relating BMD at the various sites to serum PTH and
the bone turnover markers. Serum PTH was inversely associated with BMD
at all sites in men and women, except for total body BMD in the men.
All of the bone turnover markers, except for serum PICP, were generally
inversely correlated with BMD at the various sites, although the
associations between the turnover markers and BMD were much stronger in
the women than in the men. Finally, the sex steroids, particularly the
bioavailable fractions, were inversely correlated with bone resorption
markers (r = -0.20 and r = -0.19, respectively; both
P < 0.001 for bioavailable T vs. urinary
NTx in the men and women; and r = -0.12; P = 0.02
and r = -0.29; P < 0.001 for serum bioavailable
E vs. urinary NTx in the men and women, respectively). Serum
osteocalcin was not correlated with bioavailable T in men or women, but
did correlate inversely with bioavailable E in the women (r =
-0.26; P < 0.001).
|
, serum total T was negatively
correlated with proximal femur BMD in the men under age 50 yr. Serum E,
particularly bioavailable E levels, were positively correlated with BMD
in the younger and in the older men. Serum total and bioavailable T
levels were correlated with proximal femur BMD in the pre- and
postmenopausal women. Neither serum total nor bioavailable E levels
correlated with proximal femur BMD in the premenopausal women, but were
strongly associated with BMD in the postmenopausal women. Proximal
femur BMD was inversely correlated with bone turnover markers in the
postmenopausal women and in the men over age 50 yr, but was only weakly
related or had no relationship to the turnover markers in the
premenopausal women or men under age 50 yr.
|
shows the results of
the analysis at the proximal femur. The results of the analysis were
similar in the men under and over age 50 yr and are therefore presented
for the group as a whole. However, the relationships were significantly
different in the pre- vs. postmenopausal women and are
therefore presented separately for both groups. In the men, serum
bioavailable E was the most important sex steroid predicting proximal
femur BMD. After this had entered the model, neither serum T nor DHEAS
was an independent predictor. Of note, once bioavailable E had entered
the model, nonbioavailable E was negatively associated with proximal
femur BMD. In the postmenopausal women, serum bioavailable E was also
the most important sex steroid predicting BMD, although DHEAS also
entered the model after bioavailable E. In contrast, in the
premenopausal women, serum bioavailable T, rather than E, was an
independent predictor of BMD.
|
), and which was picked in the multivariate model
depended on whether the bone turnover markers were included in the
model; in the absence of the bone turnover markers, bioavailable T was
picked as the more important predictor of radius BMD, whereas if the
bone turnover markers were included, bioavailable E was selected.
Finally, as the above analyses were performed in the entire
population-based sample of men and women, we repeated the analyses in
the strictly defined normal subjects (see Materials and
Methods) with the same results in terms of predictors of BMD as
those shown in Table 6 for the entire cohort.
|
Discussion |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Our data for serum bioavailable E and T levels may help to resolve these issues. By studying subjects over the broad age range of 23–90 yr and by directly measuring bioavailable T and E levels, we were able to demonstrate that elderly men have marked age-related decreases in both serum bioavailable T and E levels. Moreover, although both serum bioavailable T and E levels were correlated with BMD, serum bioavailable E was found to be the consistent independent predictor of BMD by multivariate analysis. Taken together, our data are consistent with the hypothesis that age-related decreases in E availability could at least in part account for the decrease in BMD with age in men. Our findings are also consistent with those of Greendale et al. (40), who recently reported that in elderly men and women, serum bioavailable E was more strongly associated with BMD than was bioavailable T. However, as they only studied elderly subjects, the relationship of sex steroids to age-related decreases in BMD were unclear. Despite these findings, however, further longitudinal and direct intervention studies are clearly needed to quantify the relative contributions of T vs. E in determining age-related bone loss in men.
In contrast to women, who have a precipitous decrease in serum E levels around the time of the menopause, the age-related decrease in serum bioavailable T and E levels is much more gradual in aging men. This would suggest that the dramatic decrease in serum E levels at the time of the menopause in women triggers a rapid phase of bone loss, which is absent in men. Indeed, this probably accounts for our observation that the percent decrease in BMD over the life span at all sites was approximately twice as great in women as it was in men. However, men made acutely hypogonadal by orchidectomy also have a rapid phase of bone loss (39); as both T and E levels fall to extremely low levels after orchidectomy, these studies do not address the issue of the relative contributions of T vs. E in mediating postorchidectomy bone loss. Nonetheless, the similar relationships between bioavailable E and BMD noted in this study in both men and postmenopausal women suggest a fundamentally similar role for E in determining BMD in both sexes. We also found, however, that E levels did not predict BMD in premenopausal women, suggesting that these women all had E levels above some threshold such that variations in E levels were no longer associated with BMD.
These studies also demonstrate that whereas serum total and bioavailable E levels decrease in women principally because of a decrease in ovarian E production, they decrease in men principally because of an age-related increase in SHBG levels. Indeed, although serum SHBG levels increased markedly in the men, they changed little over the life span in the women. Similar gender differences in SHBG levels with aging have been reported by Goodman-Gruen et al. (41) in a study of men and women, aged 50–82 yr. The reasons for this difference in changes in serum SHBG with age between men and women are unclear. However, the biological relevance of the decrease in bioavailable sex steroids in the men is supported by the much stronger inverse correlations in these subjects between bioavailable T and E levels and LH and FSH levels compared with the relationship between total T and E levels and gonadotropin levels.
Our findings are also consistent with previous observations in E receptor-negative (22) and aromatase-deficient males (23, 24, 25), which had suggested an important role for E in skeletal metabolism in men. Moreover, in a study of aged male rats, Vanderschueren et al. (42) found no differences in the effects of orchidectomy or treatment with the aromatase inhibitor, vorazole, on the decrease in bone density, suggesting that the aromatization of androgens to estrogens was playing a major role in skeletal maintenance in the male rats.
We also assessed serum PTH levels and biochemical markers of bone turnover and found, as we (5, 6, 7) and others (43) have previously reported, that there is a significant age-related increase in serum PTH levels in both men and women. Our findings also indicate that bone formation and resorption markers decreased in men between 20 and 50 yr of age and in premenopausal women, probably reflecting higher bone turnover in young individuals in the third decade who are completing the phase of skeletal consolidation. After age 50 yr in the men and in postmenopausal women, however, the bone resorption markers (urinary NTx and f-Dpd) increased significantly with age, as we have shown previously for women (44). Of the bone formation markers, only serum osteocalcin showed a consistent increase with age in both sexes. In general, bone turnover markers showed inverse correlations with BMD in the men and the women, although the relationships were stronger in the women. However, even adjusting for effects of bone turnover, serum bioavailable E levels remained significant predictors of BMD in the men and the postmenopausal women. The persistent relationship between BMD and serum bioavailable E (even after adjusting for the effects of bone turnover) noted in our study may be due to less variability in the bioavailable E measurement than in the measurement of the bone turnover markers. Alternatively, these findings could be due to the fact that E may be affecting both bone resorption and formation, whereas markers assessing these processes separately, such as osteocalcin and NTx, may not have as strong a predictive value. Finally, our data show that circulating total and bioavailable E levels are approximately twice as high in men as those in postmenopausal women. In recent studies, we have demonstrated that reduction of the low levels of serum E in a group of late (mean age, 69 yr) postmenopausal women to near undetectable levels by the administration of letrozole, an aromatase inhibitor that blocks the conversion of weak androgens to E in adipose and other peripheral tissues, increased bone resorption by about 15% (45). Thus, it is likely that the circulating levels of E in aging men have significant effects on bone turnover, although similar studies using aromatase inhibitors are needed to establish a causal relationship between E and bone turnover in aging men as well as the relative contributions of E vs. T in determining rates of bone turnover in men.
Our findings may also have practical, clinical implications for the prevention and treatment of osteoporosis in aging men. As noted earlier, we (7, 8) and others (46) have shown that E treatment of elderly women can prevent or reverse the age-related increase in bone turnover and prevent bone loss (47). As we found that serum bioavailable E levels decreased markedly in aging men and also correlated with BMD, it is plausible that treatment of elderly men with E or selective E receptor modulators, such as raloxifene, may reduce bone turnover and prevent bone loss in aging men. Indeed, a preliminary study by Taxel et al. (48) found that E treatment of elderly men (2 mg/day micronized 17ß-estradiol) for 9 weeks significantly reduced markers of bone resorption by 11 to 27%. Moreover, as T therapy, in fact, represents combined therapy with T and E (due to the aromatization of T to E), T therapy of aging men may well have significant beneficial effects on BMD and bone turnover, as suggested by several preliminary studies (49, 50, 51). Indeed, one recent study of T therapy of eugonadal men with spinal osteoporosis found that the increase in spine BMD after T therapy correlated with the T-induced increase in serum E2 levels, but not with increases in the serum T levels themselves (51).
Although our data indicate an important role for E in the regulation of skeletal metabolism in men, T clearly also has significant skeletal effects. Osteoblasts contain androgen receptors (52), and T is almost certainly responsible for the sexual dimorphism of the skeleton that develops after puberty and probably also stimulates periosteal growth of cortical bone (53). It is, therefore, of interest that at the distal radius site in the men, bioavailable T and E levels were approximately equally correlated with BMD, and which was picked in the multivariate model depended on whether the bone turnover markers were included in the model. This would suggest that at this predominantly cortical site, bioavailable T may have a greater effect on BMD in the men than at the other sites assessed, although more work is clearly needed to address this issue. Finally, serum bioavailable T was a significant independent predictor of BMD in the premenopausal women, suggesting that when T levels are low, as in this group, variations in these levels are significantly related to BMD. Conversely, when both T and E levels are low (as in the case of postmenopausal women), or when T levels are relatively high but E levels are low (as in the case of men), the variations in E levels are more important predictors of BMD.
In summary, these data indicate that aging men have significant decreases in bioavailable sex steroid levels, which correlate with BMD. Moreover, serum bioavailable E levels predict BMD in men and in postmenopausal women, suggesting fundamentally similar roles for E in skeletal metabolism in both sexes. We also demonstrate very similar changes in serum PTH and bone turnover in aging men and women. Taken together with the recent data from E receptor-negative and aromatase-deficient males (22, 23, 24, 25), our findings suggest the need to reevaluate the traditional belief that the effects of sex steroids on the male skeleton are almost entirely due to T. Additional longitudinal studies should be made to compare the effects of T and E on bone and calcium homeostasis in men and to test the hypothesis that E deficiency is a major cause of bone loss in aging men.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Received January 13, 1998.
Revised March 16, 1998.
Accepted March 24, 1998.
|
References |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  C. Meier, T. V. Nguyen, D. J. Handelsman, C. Schindler, M. M. Kushnir, A. L. Rockwood, A. W. Meikle, J. R. Center, J. A. Eisman, and M. J. Seibel Endogenous Sex Hormones and Incident Fracture Risk in Older Men: The Dubbo Osteoporosis Epidemiology Study Arch Intern Med, January 14, 2008; 168(1): 47 - 54. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Khosla Estrogen and the Death of Osteoclasts: A Fascinating Story IBMS BoneKEy, October 1, 2007; 4(10): 267 - 272. [Full Text] [PDF]
|
|
|
|
 |
|
 I. Titon, A. Cailleux-Bounacer, J. P. Basuyau, H. Lefebvre, A. Savoure, and J. M. Kuhn Evaluation of a standardized short-time calcium suppression test in healthy subjects: interest for the diagnosis of primary hyperparathyroidism Eur. J. Endocrinol., September 1, 2007; 157(3): 351 - 357. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 V. J. Armstrong, M. Muzylak, A. Sunters, G. Zaman, L. K. Saxon, J. S. Price, and L. E. Lanyon Wnt/beta-Catenin Signaling Is a Component of Osteoblastic Bone Cell Early Responses to Load-bearing and Requires Estrogen Receptor {alpha} J. Biol. Chem., July 13, 2007; 282(28): 20715 - 20727. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Bjornerem, L. A. Ahmed, R. M. Joakimsen, G. K R. Berntsen, V. Fonnebo, L. Jorgensen, P. Oian, E. Seeman, and B. Straume A prospective study of sex steroids, sex hormone-binding globulin, and non-vertebral fractures in women and men: the Tromso Study Eur. J. Endocrinol., July 1, 2007; 157(1): 119 - 125. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. A. Wolf, P. D. Terry, J. D. Potter, and R. M. Bostick Do Factors Related to Endogenous and Exogenous Estrogens Modify the Relationship between Obesity and Risk of Colorectal Adenomas in Women? Cancer Epidemiol. Biomarkers Prev., April 1, 2007; 16(4): 676 - 683. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. A. Riancho, C. Valero, A. Naranjo, D. J. Morales, C. Sanudo, and M. T. Zarrabeitia Identification of an Aromatase Haplotype That Is Associated with Gene Expression and Postmenopausal Osteoporosis J. Clin. Endocrinol. Metab., February 1, 2007; 92(2): 660 - 665. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. B. Rapuri, J. C. Gallagher, and L. M. Smith Smoking Is a Risk Factor for Decreased Physical Performance in Elderly Women J. Gerontol. A Biol. Sci. Med. Sci., January 1, 2007; 62(1): 93 - 99. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Khosla Sex Hormone Binding Globulin: Inhibitor or Facilitator (or Both) of Sex Steroid Action? J. Clin. Endocrinol. Metab., December 1, 2006; 91(12): 4764 - 4766. [Full Text] [PDF]
|
|
|
|
 |
|
 A. J. Chien and P. E. Goss Aromatase Inhibitors and Bone Health in Women With Breast Cancer J. Clin. Oncol., November 20, 2006; 24(33): 5305 - 5312. [Full Text] [PDF]
|
|
|
|
 |
|
 K. S. Nair, R. A. Rizza, P. O'Brien, K. Dhatariya, K. R. Short, A. Nehra, J. L. Vittone, G. G. Klee, A. Basu, R. Basu, et al. DHEA in Elderly Women and DHEA or Testosterone in Elderly Men N. Engl. J. Med., October 19, 2006; 355(16): 1647 - 1659. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. R. Smith Treatment-related osteoporosis in men with prostate cancer. Clin. Cancer Res., October 15, 2006; 12(20): 6315s - 6319s. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. A. Fink, S. K. Ewing, K. E. Ensrud, E. Barrett-Connor, B. C. Taylor, J. A. Cauley, E. S. Orwoll, and for the Osteoporotic Fractures in Men Study Group Association of Testosterone and Estradiol Deficiency with Osteoporosis and Rapid Bone Loss in Older Men J. Clin. Endocrinol. Metab., October 1, 2006; 91(10): 3908 - 3915. [Abstract] [Full Text] [PDF]
|
|
