HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Ravaglia, G. Articles by Cavalli, G. Search for Related Content PubMed PubMed Citation Articles by Ravaglia, G. Articles by Cavalli, G.

Body Composition, Sex Steroids, IGF-1, and Bone Mineral Status in Aging Men

^a Department of Internal Medicine, Cardioangiology, and Hepatology
^b Division of Geriatric Medicine, S. Orsola-Malpighi Hospital, University of Bologna, Italy
^c Laboratory of Radioimmunology, Orthopedic Institutes "Rizzoli," University of Bologna, Italy

Giovanni Ravaglia, Dipartimento di Medicina Interna, Cardioangiologia, Epatologia, Policlinico S. Orsola, Via Massarenti 9, 40138 Bologna, Italy E-mail: ravaglia{at}almadns.unibo.it.

William B. Ershler, MD

	Abstract

Top Abstract Methods Results Discussion References

 
Background. Bone loss in elderly men is associated with changes in body composition and reduced secretion of endogenous anabolizing hormones. The independent influences of body composition and endocrine factors on male bone metabolism, however, are unclear.

Methods. Bone mass density (BMD) (bone mass content [BMC, g]/projected bone area [BA, cm²]) at different skeletal sites, skeletal muscle, and body fat mass were measured by dual-energy X-ray absorptiometry in 129 men aged 20 to 95 years. Free testosterone, 17-ß-estradiol, dehydroepiandrosterone-sulfate, and insulin-like growth factor 1 (IGF-1) serum concentrations were measured. Because BMD may fail to control for differences in skeletal size, the associations of bone mass with body composition and hormones were studied by comparing BMD regression models incorporating age and knee height only with BMC regression models also incorporating BA.

Results. Skeletal muscle had close associations ( p at least < .01) with BMD and BMC at almost all skeletal sites, but the strength of these associations was generally reduced in BMC with respect to BMD models. Weak associations ( p < .05) were found in both models for fatness with femoral bone and for 17-ß-estradiol with total body and femoral bone. The association of 17-ß-estradiol with spinal bone was significant ( p < .05) in the BMD but not in the BMC model. No association of BMC or BMD with androgens and IGF-1 reached significancy.

Conclusions. Skeletal muscle may be more important than fatness and anabolizing hormones in preserving bone mass in elderly men. In contrast to traditional belief, estrogens may be more important than androgens and IGF-1 in male bone metabolism.

ALTHOUGH hip fractures in elderly men account for one third of all hip fractures and yield a higher mortality rate than that of women, determinants of male skeleton are still poorly defined ⁽¹⁾. Many experimental studies have shown that in men, the age-related bone and muscle losses are associated ^{(2)(3)(4)(5)(6)}. It is not clear whether this relationship depends on the decreased mechanical loading of the muscle on the skeleton or only reflects the declining secretion of androgens and insulin-like growth factor 1 (IGF-1), which are supposed to have anabolic effects on both muscle and bone tissues ⁽⁷⁾. Previous studies of the effects of the age-related decline in gonadal ^{(8)(9)(10)(11)} and adrenal androgens ^(12)(13) and in circulating IGF-1 levels ^(14)(15) on bone mass of elderly men, moreover, have given conflicting results. A possible explanation for these differences is the lack of statistical control for confounding variables such as age ⁽⁸⁾, health condition ⁽¹⁰⁾, or body and skeletal size ^{(9)(11)(12)(13)(14)(15)(16)}.

Another controversial issue is whether, in elderly men, fatness has a protective role against bone loss as it has in postmenopausal women. In women, this relationship involves both the increased mechanical stress on the axial skeleton and the increased peripheral aromatization of androgens to estrogens in adipose tissue ⁽¹⁷⁾. With regard to men, reports of a positive association between adiposity and bone status are far from uniform ^(2)(4)(18), and only recently, estrogens have been suggested to be involved in male skeletal metabolism ^(19)(20).

This study was designed to evaluate the independent relationships of soft body tissues, IGF-1, androgens, and estrogens with the male skeleton. A special effort was made to eliminate all known medical and behavioral factors affecting bone metabolism and all age- and size-related statistical artifacts.

	Methods

Top Abstract Methods Results Discussion References

 
Subjects
One hundred and fifty independently living male volunteers were recruited by advertisement by the medical students and staff at the Department of Internal Medicine, University of Bologna, and by visit to a senior citizen center in Bologna. Subjects were excluded if they had diseases that might have been relevant to consideration of bone mass, body composition, and the endocrine system (diagnosis of osteoporosis, renal or hepatic disease, malignancy, cardiovascular disease, or endocrine diseases). Previous history of back pain or fractures, obesity, unusual diets or physical activities, smoking, and taking prescription medications for acute or chronic inflammatory diseases were also exclusion criteria. One hundred and twenty-nine subjects (age range, 20–95 years) met the study criteria. Physical examination, routine blood studies, and urine analysis were made to confirm their good health. All subjects provided informed consent, and all procedures were approved by the local institutional review board.

Bone and Body Composition Measurements
Anthropometry.-- With the subjects barefoot, wearing only light indoor clothes, and instructed to hold themselves upright, body weight (BW) was measured to the nearest 0.1 kg and height (H) to the nearest 0.5 cm by a beam scale with stadiometer (Seca, Hamburg, Germany). Body mass index (BMI) was calculated for all subjects as BW (kg) divided by H (m). Knee height was measured by a sliding caliper as described by Chumlea and coworkers ⁽²¹⁾.

Dual-energy x-ray absorptiometry (DXA).-- Regional and whole-body scans were performed by a model QDR 2000 DXA densitometer, software version 7.10 (Hologic, Inc., Waltham, MA), which uses a switched, pulsed, stable dual-energy radiation with two peak kilovoltages (kVp) of 70 and 140 kVp. Regional bone mass density (BMD), obtained from the regional bone mass content (BMC, g) divided by the scanned bone area (BA, cm²), was estimated for lumbar spine (L2–L4 vertebrae) and proximal femur (neck, Ward's triangle, greater trochanter, and intertrochanteric region) by standard scan protocols.

Analysis of whole-body scans was done using the enhanced analysis protocol version 5.64 (Hologic, Inc.), which provides measurements for bone as well as for the lean and fat components of body weight. The precision of measurements was 1.6% for lean mass, 1.9% for fat mass, and 1.8% for whole-body BMC (coefficient of variation calculated from 10 paired measurements).

Lean mass as measured by DXA includes skeletal muscle as well as organs and other nonmuscle soft tissues that may have no biological relation to the age-related loss of bone. Because our attention focused on the association between bone and skeletal muscle, analyses were made using appendicular skeletal muscle (ASM), derived as the sum of the fat-free soft tissue masses of the arms and the legs, as described by Heymsfield and coworkers ⁽²²⁾.

Laboratory
Blood samples were taken in the early morning after an overnight fast, put on ice, and processed within 1 hour. Serum was separated by centrifugation (3000 x g, 30 min, 4°C), and aliquots were appropriately stored at -70°C and analyzed within 6 months.

Circulating levels of 17-ß-estradiol (E2) were assayed in serum by fluoroimmunoenzymatic assay (Eurogenetics, Tessenderlo, Belgium; intraassay and interassay coefficients of variation were 5% and 8%). Free testosterone (Diagnostic Systems Laboratories, Webster, TX; intraassay and interassay coefficients of variation were 4% and 8%) and dehydroepiandrosterone sulfate (DHEAS; Radim, Rome, Italy; intraassay and interassay coefficients of variation were 6% and 9%, respectively) were assayed in serum by radioimmunoassay kits. IGF-1 was assayed in serum by a commercial immunoradiometric assay kit using a modified version of the standard acid-ethanol extraction procedure to separate binding proteins (Diagnostic Systems Laboratories; intraassay and interassay coefficients of variation 2% and 4%, respectively). All samples were run in duplicate.

Statistical Analysis
Data were reported as mean ± standard deviation. Simple correlation analysis (Pearson product moment significant at p < .05) was used to assess the univariate relations between the measured variables. The relationship of each variable with bone mineral status was tested by a forward stepwise-fitting algorithm for multiple linear regression analysis. Prentice and coworkers ⁽¹⁶⁾ suggest two different strategies for skeletal and body size adjustment when analyzing absorptiometric bone data by multiple regression models. Although the expression of bone data as BMD provides, by itself, a degree of standardization for differences in bone size, when BMD is used as the dependent variable in a multiple regression model, further correction for body size effects can be obtained by forcing weight and height along with the other independent variables of interest. The expression of bone data as BMD, however, implies the theoretical assumption, not always true in practice, that BMC is directly proportional to BA at all skeletal regions. Direct proportionality between two variables means that a 1% increase in one is matched by a 1% increase in the other. When using a regression model of log-transformed variables, this situation would be represented by a power coefficient equal to 1. In our data set, however, the power coefficients for regressions of the logarithms of the BMCs on the logarithms of the corresponding BAs were significantly greater than 1 ( p < .05) for total body (coefficient ± SE, 1.753 ± 0.065), lumbar spine (1.395 ± 0.065), trochanter (1.291 ± 0.111), and Ward's triangle (1.729 ± 0.252). These results suggested that the division of BMC by BA could not completely adjust for size variation among our study subjects for these regions.

To avoid forcing bone data to fit a predefined relationship, Prentice and associates ⁽¹⁶⁾ suggest using BMC as the dependent variable instead of BMD and to force BA in all regression models as an independent variable along with weight and height.

In this study, data were analyzed by both the BMD and the BMC model suggested by Prentice and associates ⁽¹⁶⁾. Due to the specific features of our study, however, the following alterations were made to the original approach: (a) as fat mass and ASM were included as independent variables in all regression models, BMD and BMC were not adjusted for body weight because the sum of the fat and lean masses equals weight, thus not allowing the specific effects of individual body compartments and the generic effect of body weight on bone to be disentangled by multivariate analysis ⁽⁴⁾; (b) knee height was used instead of stature, because the latter is affected by age-related changes of the spine that may distort comparisons between persons of different ages for skeletal length ⁽²¹⁾; and (c) age was forced in all stepwise procedures to control for any age-related covariance.

Before performing multivariate analysis, all the variables were converted to natural logarithms, so that the final regression coefficients would provide information in proportional terms about the influence of each independent variable on bone mass. For example, performing the regression model for whole-body data in Table 4 without transforming the data to natural logarithms, we found that a 1-pg/ml difference in E2 corresponds to a difference in whole-body BMC of approximately 0.6 g. Knowing the absolute values of both variables and their expected ranges in the population is required to evaluate the importance of the association. According to the same model with logarithms, by contrast, a 1% change in E2 is associated with a 0.02% change in BMC, a result that can be more directly and easily interpreted.

	Results

Top Abstract Methods Results Discussion References

 
The characteristics of the study population are presented in Table 1 . As shown by the univariate correlations reported in Table 2 , age was negatively associated with BMC at all skeletal sites, except for the spine and the intertrochanteric region, and with BMD at all skeletal sites. ASM was associated with BMC and BMD at all sites, whereas FM was weakly associated with intertrochanteric BMC only. DHEAS, E2, and IGF-1 were associated with total body, femoral neck, and Ward's triangle BMC. Free testosterone, DHEAS, E2, and IGF-1 were associated with BMD at all sites except for lumbar spine.

The final multiple-regression equations for associations of body composition and hormonal variables with BMD (adjusted for age and knee height) and BMC (adjusted for BA, age and knee height) are given in Table 3 and Table 4 , respectively.

Fat mass entered the multivariate models for both BMD and BMC at all femoral regions except for the trochanter. E2 entered the multivariate models for BMD at all skeletal sites. Similar associations were found when using the BMC regression models except for the spine and the trochanter. E2 was the only variable associated with total body BMC.

In both the BMD and BMC multivariate models, the statistically significant associations of free testosterone, DHEAS, and IGF-1 with bone mineral variables found in univariate analysis were not confirmed.

	Discussion

Top Abstract Methods Results Discussion References

 
In this carefully selected, healthy, male population, aging was associated with a steady decrease in bone mass parameters (BMD and BMC) at all the measured sites except for the spinal region, whose apparent stability may be due to the increasing presence with age of osteoarthritis and dystrophic calcification. Besides lower bone mass values, older subjects from this population also exhibited reduced amounts of skeletal muscle, lower circulating levels of androgens and IGF-1, and unchanged levels of E2. Fat mass showed a positive association with age that was reversed after about the seventh decade. Because of the cross-sectional structure of our study, a possible explanation for this feature could be a selective mortality of overweight subjects.

A multivariate analysis taking into account age- and size-related covariance identified skeletal muscle and fat mass as significantly associated with bone mineral status at different skeletal regions. If the association between bone mass and soft tissues was determined by a mass-related (gravitational) effect only, both muscle and fat mass should be related with the same strength to bone mineral status at weight-bearing sites, because skeletal responses to gravitational forces from similar amounts of muscular and adipose tissue would be the same. In our study, however, the association between bone and muscle mass was very strong, whereas the association between bone and fat mass was weak and might not be found at all weight-bearing skeletal sites, suggesting that the stress imposed on the skeleton by physical (muscular) activity is more important for bone mass than the simple gravitational effect.

Among postmenopausal women, adiposity is suggested to prevent bone loss not only because of the increased loading on the axial skeleton, but also because of the beneficial effects resulting from the increased peripheral aromatization of androgens to estrogens ⁽¹⁷⁾. In the male population evaluated in this study, E2 was associated with total body bone mineral and with bone mineral at almost all femoral sites. These results are in agreement with recently published data of significant positive associations between estrogens and bone mass in young ^(18)(19) and elderly healthy men ^{(13)(23)(24)(25)}.

Androgens, by contrast, were not associated to bone mass of men. Because testosterone supplementation therapy, however, has a real potential to improve bone mass in elderly male subjects, a plausible hypothesis is that the aromatization of androgens to estrogens, rather than their direct action on bone, plays the major role in regulating bone metabolism in men ⁽²⁴⁾.

The lack of positive associations between bone and IGF-1 levels is in contrast with previous studies in adult men ⁽¹⁴⁾, but in agreement with two recent studies focused on elderly people ^(26)(27) and showing that IGF-1 was associated with BMD in older women but not in men. The etiology of this gender difference is still unknown, but it has been hypothesized that most of the anabolic effects on bone mass commonly ascribed to the somatotrophic axis might be eventually caused by IGF-1-mediated changes in sex steroids bioavailability ⁽²⁸⁾.

The results of the present study also support the argument of Prentice and coworkers ⁽¹⁶⁾ that although BMD is a useful predictor of fracture risk in clinical practice, it is not always a satisfactory adjustment for bone size at all skeletal sites. In our study, in fact, the use of BMD as a dependent variable resulted in artifacts and inflated associations with muscle mass for all the skeletal regions whose BMC was not in a one-to-one relationship with the corresponding BA.

Although this issue has been mainly of concern in the clinical interpretation of bone mineral data in growing children ⁽²⁹⁾, our study confirms previous findings that BMD may overestimate the associations between bone mineral and other size-related variables in adult and elderly subjects ⁽⁴⁾, and this may be of importance in epidemiological research.

In conclusion, this study suggests that muscle mass, through mechanical loading forces, may be more important than fatness and hormonal status in skeletal maintenance in men. In contrast to traditional belief, moreover, estrogens may be more important than androgens and IGF-1 in male bone metabolism. The magnitude of the associations found in this study, however, was very small: at the femoral neck, for example, each 1% change in muscle mass corresponded to a mean variation in BMC of only 0.6%, and the contributions of E2 and fat mass were less than 0.1%. It is possible that other factors than those measured play a more fundamental role in male bone regulation, but due to the limitations of the cross-sectional design used in this study, we cannot exclude that bone mass of older men may be correlated less strongly with the point-in-time body composition and endocrine function than with those of preceding years, particularly at the time of attainment of bone mass peak.

	Acknowledgments

 
This work was partially supported by a grant from the Ministero dell'Università e della Ricerca Scientifica e Tecnologica (MURST, ex-60% fund).

Received August 25, 1998

Accepted October 7, 1999

	References

Top Abstract Methods Results Discussion References

 

1. Seeman E, 1997. Osteoporosis in men. Baillieres Clin Rheumatol 11:613-629. [Medline]
2. Reid IA, Lindsay DP, Evans MC, 1992. Fat mass is an important determinant of whole body bone density in premenopausal women but not in men. J Clin Endocrinol Metab 75:779-782. [Abstract]
3. Hughes VA, Frontera WR, Dallal GE, Lutz KJ, Fisher EC, Evans WJ, 1995. Muscle strength and body composition: associations with bone density in older subjects. Med Sci Sports Exerc 27:967-974. [Medline]
4. Baumgartner RN, Stauber PM, Koehler K, Romero L, Garry PJ, 1996. Associations of fat and muscle masses with bone mineral in elderly men and women. Am J Clin Nutr 63:365-372. [Abstract/Free Full Text]
5. Bevier WC, Wiswell RA, Pyka G, Kozak KC, Newhall KM, Marcus R, 1989. Relationship of body composition, muscle strength, and aerobic capacity to bone mineral density in older men and women. J Bone Miner Res 4:421-432. [Medline]
6. Nuti R, Matini G, Gennari C, 1995. Age-related changes in whole skeleton and body composition in healthy men. Calcif Tissue Int. 57:336-339. [Medline]
7. Slemenda CW, 1995. Body composition and skeletal density. Mechanical loading or something more?. J Clin Endocrinol Metab. 80:1761-1763. [Medline]
8. Foresta C, Ruzza G, Mioni R, et al. 1984. Osteoporosis and decline of gonadal function in the elderly male. Horm Res 19:18-22. [Medline]
9. Meier DE, Orwoll ES, Keenan EJ, Fagestrom RM, 1987. Marked decline in trabecular bone mineral content in healthy men with age: lack of association with sex steroid levels. J Am Geriatr Soc 35:189-197. [Medline]
10. Stanley HL, Schmitt BP, Poses RM, Deiss WP, 1991. Does hypogonadism contribute to the occurrence of a minimal trauma hip fracture in elderly men?. J Am Geriatr Soc. 39:766-771. [Medline]
11. Rudman D, Drinka PJ, Wilson CR, et al. 1994. Relations of endogenous anabolic hormones and physical activity to bone mineral density and lean body mass. Clin Endocrinol. 40:653-661. [Medline]
12. Szathmari M, Szucs J, Feher T, Hollo I, 1994. Dehydroepiandrosterone sulphate and bone mineral density. Osteoporosis Int 4:84-88. [Medline]
13. Greendale GA, Edelstein S, Barrett-Connor E, 1997. Endogenous sex steroids and bone mineral density in older women and men: the Rancho Bernardo study. J Bone Miner Res 12:1833-1843. [Medline]
14. Johansson AG, Forslund A, Hambraeus L, Blum WF, Ljunghall S, 1994. Growth hormone-dependent insulin-like growth factor binding protein is a major determinant of bone mineral density in healthy men. J Bone Miner Res 9:915-921. [Medline]
15. Toogood AA, Adams JE, O'Neill PA, Shalet SM, 1997. Elderly patients with adult-onset growth hormone deficiency are not osteopenic. J Clin Endocriniol Metab 82:1462-1466. [Abstract/Free Full Text]
16. Prentice A, Parson TJ, Cole TJ, 1994. Uncritical use of bone mineral density in absorptiometry may lead to size related artifacts in the identification of bone mineral determinants. Am J Clin Nutr 60:837-842. [Abstract/Free Full Text]
17. Glauber HS, Vollmer WM, Nevitt MC, Ensrud KE, Orwoll ES, 1995. Body weight versus body fat distribution, adiposity, and frame size as predictors of bone density. J Clin Endocrinol Metab 80:1118-1123. [Abstract]
18. Edelstein SL, Barrett-Connor E, 1993. Relation between body size and bone mineral density in elderly men and women. Am J Epidemiol 138:160-169. [Abstract/Free Full Text]
19. Smith EP, Body J, Frank GR, et al. 1994. Estrogen resistance caused by a mutation in the estrogen receptor gene in a man. N Engl J Med. 331:1056-1061. [Abstract/Free Full Text]
20. Morishima A, Grumbach MM, Simpson ER, Fisher C, Qin K, 1995. Aromatase deficiency in male and female siblings caused by a novel mutation and the physiological role of estrogens. J Clin Endocrinol Metab 80:3689-3698. [Abstract]
21. Chumlea WC, Roche AF, Steinbaugh ML, 1985. Estimating stature from knee height for persons 60 to 90 years of age. J Am Geriatr Soc 33:116-121. [Medline]
22. Heymsfield SB, Smith R, Aulet M, et al. 1990. Appendicular skeletal muscle mass: measurement by dual photon absoprtiometry. Am J Clin Nutr 52:214-218. [Abstract/Free Full Text]
23. Slemenda CW, Longcope C, Zhou L, Hiu SL, Peacock M, Johnston CC, 1997. Sex steroids and bone mass in older men. J Clin Invest 100:1755-1759. [Medline]
24. Anderson FH, Francis RM, Peaston RT, Wastell HJ, 1997. Androgen supplementation in eugonadal men with osteoporosis: effects of six months' treatment on markers of bone formation and resorption. J Bone Miner Res. 12:472-478. [Medline]
25. Khosla S, Melton LJ, III Atkinson EJ, O'Fallon WM, Klee GG, Riggs BL, 1998. 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. 83:2266-2274. [Abstract/Free Full Text]
26. Barrett-Connor E, Goodman-Gruen D, 1998. Gender differences in insulin-like growth factor and bone mineral density association in old age: the Rancho Bernardo Study. J Bone Miner Res. 13:1343-1349. [Medline]
27. Langlois JA, Rosen CJ, Visser M, et al. 1998. Association between insulin-like growth factor I and bone mineral density in older women and men: the Framingham Heart Study. J Clin Endocrinol Metab. 83:4257-4262. [Abstract/Free Full Text]
28. Pfeilschifter J, Scheidt-Nave C, Leidig-Bruckner G, et al. 1996. Relationship between circulating insulin-like growth factor components and sex hormones in a population-based sample of 50- to 80-year-old men and women. J Clin Endocrinol Metab 81:2534-2540. [Abstract]
29. Warner JT, Cowan FJ, Dunstan FD, Evans WD, Webb DK, Gregory JW, 1998. Measured and predicted bone mineral content in healthy boys and girls aged 6–18 years: adjustment for body size and puberty. Acta Paediatr 87:244-249. [Medline]

This article has been cited by other articles:

				J. E. Morley Editorial: Citations, Impact Factor, and the Journal J. Gerontol. A Biol. Sci. Med. Sci., December 1, 2002; 57(12): M765 - 769. [Full Text] [PDF]

				J. E. Morley, H. M. Perry III, and D. K. Miller Editorial: Something About Frailty J. Gerontol. A Biol. Sci. Med. Sci., November 1, 2002; 57(11): M698 - 704. [Full Text] [PDF]

				A. Fisher and J. E. Morley Editorial: Antiaging Medicine: The Good, the Bad, and the Ugly J. Gerontol. A Biol. Sci. Med. Sci., October 1, 2002; 57(10): M636 - 639. [Full Text] [PDF]

				R. A. Wiswell, S. A. Hawkins, H. C. Dreyer, and S. V. Jaque Maintenance of BMD in Older Male Runners Is Independent of Changes in Training Volume or V{middle dot}O2peak J. Gerontol. A Biol. Sci. Med. Sci., April 1, 2002; 57(4): M203 - 208. [Abstract] [Full Text] [PDF]

				A. M. Matsumoto Andropause: Clinical Implications of the Decline in Serum Testosterone Levels With Aging in Men J. Gerontol. A Biol. Sci. Med. Sci., February 1, 2002; 57(2): M76 - 99. [Full Text]

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Ravaglia, G. Articles by Cavalli, G. Search for Related Content PubMed PubMed Citation Articles by Ravaglia, G. Articles by Cavalli, G.