0021-972X/01/$03.00/0
5
The Journal of Clinical Endocrinology & Metabolism
Copyright © 2001 by The Endocrine Society
Serum Leptin Levels Are Associated with Bone Mass in
Nonobese Women*
JULIE A. PASCO, MARGARET J. HENRY, MARK A. KOTOWICZ,
GREGORY R. COLLIER, MADELEINE J. BALL, ANTONY M. UGONI,
GEOFFREY C. NICHOLSON
The University of Melbourne, Department of Clinical and Biomedical Sciences, Barwon Health (J.A.P.,
M.J.H., M.A.K., G.C.N.), and Department of General Practice and Public Health (A.M.U.), Metabolic
Research Unit, School of Health Sciences (G.R.C.), Victoria 3220, Australia; and School of
Biomedical Sciences (M. J. B.), University of Tasmania, Tasmania 7250, Australia
ABSTRACT
Both serum leptin and bone mineral density are positively corre-
lated with body fat, generating the hypothesis that leptin may be a
systemic and/or local regulator of bone mass. We investigated 214
healthy, nonobese Australian women aged 20 –91 yr. Bone mineral
content, projected bone area, and body fat mass were measured by
dual energy x-ray absorptiometry and fasting serum leptin levels by
RIA. Associations between bone mineral content (adjusted for age,
body weight, body fat mass, and bone area) and the natural logarithm
of serum leptin concentrations were analyzed by multiple regression
techniques. There was a significant positive association at the lateral
spine, two proximal femur sites (Ward’s triangle and trochanter), and
L EPTIN IS A hormone with a diverse range of local and
systemic physiological functions. In addition to its role
as a satiety factor and involvement in regulating energy
balance, leptin modulates the reproductive (1, 2), hemato-
poietic, and immune systems (3–5); promotes angiogenesis
(6, 7); and is involved in brain development (8) and regula-
tion of carbohydrate metabolism (9 –11). In vitro studies have
demonstrated a direct effect of leptin on human marrow
stromal cells, stimulating osteoblast differentiation and min-
eralization of bone matrix (12). A role for leptin in fetal bone
metabolism has been suggested (13). Leptin has also been
implicated in the development of the periosteal envelope in
growing bone (1), suggesting an anabolic effect on the
skeleton.
Both serum leptin (14 –18) and bone mass (19, 20) are
positively correlated with body fat. Mechanical loading on
the skeleton (21) and/or the actions of a mediator between
adipose tissue and bone may contribute to the association
between body fat and bone mass (19). We hypothesized that
leptin may be such a mediator and, therefore, we have eval-
uated the relationship between serum leptin concentrations
and bone mass in women.
Received August 9, 2000. Revision received December 7, 2000. Ac-
cepted December 8, 2000.
Address all correspondence and requests for reprints to: Dr. J. A.
Pasco, The University of Melbourne, Department of Clinical and Bio-
medical Sciences, Barwon Health, P.O. Box 281, Geelong 3220, Australia.
E-mail: juliep@barwonhealth.org.au.
* The project was supported by the Victorian Health Promotion
Foundation.
1884
Vol. 86, No.
Printed in U.S.A.
AND
Downloaded from https://academic.oup.com/jcem/article/86/5/1884/2847967 by guest on 17 March 2023
whole body (partial r2 ⫽ 0.019 to 0.036; all P ⬍ 0.05). Similar trends
were observed at the femoral neck and posterior-anterior-spine. With
bone mineral density the dependent variable (adjusted for age, body
weight, and body fat mass), the association with the natural logarithm
of leptin remained significant at the lateral spine (partial r2 ⫽ 0.030;
P ⫽ 0.011), was of borderline significance at the proximal femur sites
(partial r2 ⫽ 0.012 to 0.017; P ⫽ 0.058 to 0.120), and was not signif-
icant at the other sites. Our results demonstrate an association be-
tween serum leptin levels and bone mass consistent with the hypoth-
esis that circulating leptin may play a role in regulating bone mass.
(J Clin Endocrinol Metab 86: 1884 –1887, 2001)
Subjects
Materials and Methods
Subjects were from a large age-stratified sample of women drawn at
random from electoral rolls spanning the Barwon Statistical Division in
southeastern Australia for participation in the Geelong Osteoporosis
Study (22, 23). Serum leptin levels were determined for a subgroup
encompassing a wide range of body mass index (BMI) for involvement
in other studies (24, 25). As the exponential relationship between body
fat mass and circulating leptin levels in nonobese subjects diminishes
among the obese (14, 16, 25), we excluded obese women [BMI ⬎ 30.0
(26)]. Three hundred sixty-five women were eligible to participate in our
study. Subjects were also excluded if they were currently exposed to
glucocorticoids (n ⫽ 4) or the oral contraceptive pill (n ⫽ 73); were
breast-feeding (n ⫽ 18); had a fasting plasma glucose ⬎ 7.0 mmol/liter
(n ⫽ 4), incomplete sets of scans (n ⫽ 23); or had prostheses (n ⫽ 7),
pacemakers (n ⫽ 1), silicon implants (n ⫽ 2), or nonremovable jewelry
(n ⫽ 29) that would affect scan interpretation. None of the subjects was
pregnant or using hormone replacement therapy. Of the 214 nonobese
subjects included in the study (aged 20 –91 yr), 133 were premenopausal,
67 were postmenopausal, and 14 had indeterminate menopausal status.
All were free from drugs and diseases known to affect bone metabolism.
The study was approved by the Barwon Health Research and Ethics
Advisory Committee, and informed consent was obtained from all
participants.
Measurements
Body weight and height were measured to the nearest 0.1 kg and 0.1
cm, respectively, and BMI calculated as weight/height2 (kg/m2). Dual-
energy x-ray absorptiometry was performed using a Lunar Corp. (Mad-
ison, WI). DPX-L densitometer and analyzed with Lunar Corp. DPX-L
software version 1.31. Bone mineral content (BMC) (g), areal bone min-
eral density (BMD) (g/cm2), and projected bone area (cm2) were mea-
sured at the spine in the posterior-anterior (PA, L2– 4) and lateral pro-
jections (L3), proximal femur (femoral neck, Ward’s triangle, trochanter),
whole body, ultradistal (UD), and midforearm sites. In vivo short-term
 SERUM LEPTIN AND BONE MASS IN NONOBESE WOMEN 1885

precision for BMC, BMD, and projected area, respectively, was 1.4%, TABLE 2. Correlations between leptin (ln) and BMC or BMD
0.6%, 1.5% for PA-spine; 2.9%, 3.4%, 4.0% for lateral spine; 2.9%, 1.6%,
2.3% for the femoral neck; 3.0%, 2.1%, 2.8% for Ward’s triangle; 4.3%, BMC BMD
1.6%, 3.9% for the trochanter; 0.6%, 0.4%, 0.9% for the whole body; 1.6%, r P r P
2.1%, 1.7% for UD-forearm; and 0.8%, 1.1%, 0.9% for the midforearm.
Body fat mass (g) was determined from whole-body scans, with a pre- All
cision of 3.8%. Venous blood samples were collected following an over- Femoral neck 0.040 0.557 ⫺0.011 0.877
night fast, separated by centrifugation and stored at ⫺80 C until analysis. Ward’s triangle 0.057 0.408 ⫺0.030 0.665
Serum leptin concentrations were determined by a commercial RIA Trochanter 0.161 0.018 0.102 0.137
(Linco Research, Inc., St. Louis, MO). The interassay coefficient of vari- PA-spine ⫺0.025 0.713 ⫺0.009 0.898
ation ranged from 4.1– 8.2%, and the intraassay coefficient of variation Lateral spine ⫺0.074 0.281 ⫺0.103 0.135
was 5%. Whole body 0.094 0.170 0.087 0.207
UD-forearm ⫺0.092 0.179 ⫺0.072 0.291
Midforearm ⫺0.083 0.227 ⫺0.172 0.012
Statistics
Premenopausal

Downloaded from https://academic.oup.com/jcem/article/86/5/1884/2847967 by guest on 17 March 2023

Serum leptin concentrations were transformed to the natural loga- Femoral neck 0.074 0.397 0.045 0.603
rithm (ln) to normalize the data before analysis. Regression techniques Ward’s triangle 0.053 0.546 0.005 0.951
(27) were used to develop equations for predicting BMC and BMD at Trochanter 0.138 0.113 0.079 0.367
each site. Higher than linear adjustments for age, centered about the PA-spine 0.079 0.364 0.046 0.596
mean to reduce collinearity (27), were included for the forearm sites. Lateral spine 0.085 0.329 0.081 0.355
Linear adjustments were made for age at the other sites and for body Whole body 0.208 0.016 0.207 0.017
weight and body fat mass at all sites. BMC was also adjusted for pro- UD-forearm 0.059 0.497 0.039 0.655
jected bone area to correct for differences in areas scanned. All between- Midforearm ⫺0.037 0.669 ⫺0.084 0.336
predictor correlations were ⬍0.9, as required for valid regression anal-
ysis (27). Menopause status was not included in the models for the entire Postmenopausal
sample because its contribution was negligible. Partial r2 values of the Femoral neck 0.142 0.250 0.106 0.392
predictors (27) were calculated for each predictor using site-specific Ward’s triangle 0.187 0.130 0.123 0.321
models. Significance was set at P ⬍ 0.05 and all statistical analyses were Trochanter 0.104 0.400 0.237 0.053
performed using Minitab (release 12) software package. PA-spine 0.020 0.874 0.141 0.255
Lateral spine ⫺0.032 0.796 0.016 0.899
Whole body 0.167 0.178 0.218 0.076
Results
UD-forearm ⫺0.102 0.414 ⫺0.016 0.895
Table 1 lists subject characteristics. Median serum leptin Midforearm 0.028 0.824 ⫺0.047 0.708
(range) was 11.3 ng/ml (2.1– 89.3). Univariate analysis indi-
cated that leptin (ln) was correlated with all indices of body
fatness: r ⫽ 0.60 for body weight; r ⫽ 0.80 for body fat mass; association of leptin (ln) with BMC (Table 3) adjusted for age,
r ⫽ 0.83 for per cent body fat and r ⫽ 0.74 for BMI (all P ⬍ body weight, body fat mass and projected bone area (partial
0.001). Correlations between leptin (ln) and BMC or BMD in r2 ⫽ 0.019 to 0.036; all P ⬍ 0.05). Similar trends were observed
the entire sample and according to menopausal status dis- at the femoral neck and PA-spine (partial r2 ⫽ 0.013, 0.011;
played no consistent pattern (Table 2). P ⫽ 0.103, 0.137; respectively). At the whole body, body
At the lateral spine and two proximal femur sites (Ward’s weight and body fat mass were not significant predictors of
triangle and trochanter), there was an independent positive BMC after adjusting for age and bone area. Whether these
variables were forced into the regression equation, leptin (ln)
remained a significant predictor of BMC (P ⬍ 0.05). No
TABLE 1. Subject characteristics (mean and SD)
association was detected between leptin (ln) and BMC at the
Mean SD UD-forearm (P ⫽ 0.825), and a trend toward a negative
Age (yr) 44.5 13.1 association was observed at the midforearm (P ⫽ 0.158).
Body weight (kg) 62.6 8.6 When BMD (adjusted for age, body weight, and body fat
Height (cm) 162.5 6.4 mass) was analyzed (Table 4), the effect of leptin (ln) re-
BMI (kg/m2) 23.7 2.9 mained significant at the lateral spine (partial r2 ⫽ 0.030, P ⫽
Body fat mass (g) 21,462 6594
Leptin (ln) 2.40 0.72 0.011). The association was of borderline significance at the
BMC (g) proximal femur sites (partial r2 ⫽ 0.012 to 0.017, P ⫽ 0.058 to
Femoral neck 4.34 0.77 0.120) and was not significant at the other sites (P ⬎ 0.05).
Ward’s triangle 1.99 0.52 Among postmenopausal women, significant positive asso-
Trochanter 9.52 2.54
PA-spine 49.99 9.79
ciations between leptin (ln) and BMC were observed at the
Lateral spine 2.29 0.75 PA-spine and whole body (P ⫽ 0.011 and 0.016, respectively);
Whole body 2506 387 associations at other sites, and using BMD as the dependent
UD-forearm 1.77 0.32 variable, were not significant (P ⫽ 0.082– 0.443). No significant
Midforearm 3.06 0.40 associations were observed among premenopausal women.
BMD (g/cm2)
Femoral neck 0.930 0.141
Ward’s triangle 0.806 0.164 Discussion
Trochanter 0.786 0.124
PA-spine 1.188 0.169
To our knowledge, this is the first reported association
Lateral spine 0.669 0.184 between bone mass and serum leptin concentrations, inde-
Whole body 1.132 0.091 pendent of body weight, and body fat mass. The regression
Ultradistal-forearm 0.322 0.054 models we used to test the association between BMC and
Midforearm 0.688 0.076 leptin included adjustments for body weight, to account for
1886 PASCO ET AL. JCE & M • 2001
Vol. 86 • No. 5

TABLE 3. Multiple regression analysis with BMC (g) as the
dependent variable and age (yr), body weight (kg), body fat mass
(g), bone area (cm2), and leptin (ng/mL) (ln) as independent
variables in the models for the proximal femur sites (femoral neck,
Ward’s triangle, and trochanter), spine (PA and lateral), and whole
body for the whole cohort
TABLE 4. Multiple regression analysis with areal BMD (g/cm2)
as the dependent variable and age (yr), body weight (kg), body fat
mass (g), and leptin (ng/ml) (ln) as independent variables in the
models for the proximal femur sites (femoral neck, Ward’s
triangle, and trochanter), spine (PA and lateral), and whole body
for the whole cohort

Variable ␤ coefficient (slope) P value Partial r2 Variable ␤ coefficient (slope) P value Partial r2
Femoral neck Femoral neck
Age ⫺0.019 0.000 0.139 Age ⫺0.004 0.000 0.151
Body weight 0.060 0.000 0.114 Body weight 0.012 0.000 0.112
Body fat mass ⫺0.000 0.001 0.052 Body fat mass ⫺0.000 0.001 0.047
Bone area 0.828 0.000 0.257 Leptin (ln) 0.034 0.115 0.012
Leptin (ln) 0.161 0.103 0.013 Ward’s triangle

Downloaded from https://academic.oup.com/jcem/article/86/5/1884/2847967 by guest on 17 March 2023

Ward’s triangle Age ⫺0.006 0.000 0.200
Age ⫺0.014 0.000 0.193 Body weight 0.012 0.000 0.088
Body weight 0.031 0.000 0.089 Body fat mass ⫺0.000 0.002 0.043
Body fat mass ⫺0.000 0.002 0.047 Leptin (ln) 0.046 0.058 0.017
Bone area 0.786 0.000 0.398 Trochanter
Leptin (ln) 0.118 0.047 0.019 Age ⫺0.002 0.006 0.036
Trochanter Body weight 0.012 0.000 0.126
Age ⫺0.034 0.000 0.066 Body fat mass ⫺0.000 0.002 0.045
Body weight 0.113 0.000 0.067 Leptin (ln) 0.031 0.120 0.012
Body fat mass ⫺0.000 0.008 0.033 PA-spine
Bone area 0.931 0.000 0.620 Age ⫺0.005 0.000 0.153
Leptin (ln) 0.552 0.025 0.024 Body weight 0.015 0.000 0.131
PA-spine Body fat mass ⫺0.000 0.002 0.046
Age ⫺0.193 0.000 0.130 Leptin (ln) 0.029 0.253 0.006
Body weight 0.494 0.000 0.063 Lateral spine
Body fat mass ⫺0.001 0.011 0.031 Age ⫺0.010 0.000 0.493
Bone area 1.466 0.000 0.386 Body weight 0.008 0.001 0.051
Leptin (ln) 1.573 0.137 0.011 Body fat mass ⫺0.000 0.011 0.031
Lateral spine Leptin (ln) 0.055 0.011 0.030
Age ⫺0.034 0.000 0.469 Whole body
Body weight 0.028 0.001 0.048 Age ⫺0.003 0.000 0.235
Body fat mass ⫺0.000 0.006 0.036 Body weight 0.007 0.000 0.126
Bone area 0.622 0.000 0.409 Body fat mass ⫺0.000 0.031 0.022
Leptin (ln) 0.210 0.006 0.036 Leptin (ln) 0.015 0.216 0.007
Whole body
Age ⫺4.589 0.000 0.161
Body weight 3.314 0.429 0.002 levels and BMC was no longer significant after adjusting for
Body fat mass ⫺0.005 0.350 0.004
Bone area 1.498 0.000 0.560
body fat mass (17). The inclusion of obese subjects may have
Leptin (ln) 63.80 0.018 0.030 diminished the association. There is a different relationship
between body fat mass and circulating leptin levels among the
obese, who display a wide range of serum leptin concentrations
the mechanical loading on the skeleton caused by gravita- that overlap those observed in lean subjects (14, 15). The find-
tional forces (21), and bone area, to correct for differences in ings in the present study are reported for nonobese women
areas scanned. We also adjusted for body fat mass to allow alone for whom there is an exponential relationship between
for the potential influence of other humoral factors associ- body fat mass and serum leptin. In accordance with results from
ated with adipose tissue (28). Using these models, significant our study, another study of 94 adult women (18) reported no
positive associations were demonstrated at two proximal relationship between leptin and BMD or bone geometry at the
femur sites (Ward’s triangle and trochanter), the lateral distal radius. The reasons for the lack of association at non-
spine, and whole body. Leptin may have contributed to the weight-bearing sites remain unclear.
variance attributed to body fat mass; however, leptin alone In cross-sectional studies (17, 18), no correlation was ob-
explained a small proportion (1– 4%) of the variance in ad- served between circulating leptin and markers of bone turn-
justed BMC. Significant associations between leptin and ad- over. Because the subjects were likely to be in a steady state
justed BMC were observed at the PA-spine and whole body with bone formation coupled to resorption (29), it would
among postmenopausal women. Smaller numbers in the sep- seem unlikely that any association would be observed.
arate analyses of premenopausal and postmenopausal Changes in leptin concentrations and bone turnover at the
women may have limited our ability to detect significant individual bone remodeling units might not produce mea-
associations at other sites. surable systemic changes. However, the hypothesis that
The high correlations between circulating leptin concentra- there is a relationship between serum leptin and bone turn-
tions and indices of adiposity have been reported previously over could be tested by producing pharmacological alter-
(14 –18). In a study of 54 postmenopausal women (BMI 15.8 – ations in serum leptin and measuring turnover response.
42.9 kg/m2), the positive association between plasma leptin Unlike a recent study (30), earlier studies in mice (31) and
 SERUM LEPTIN AND BONE MASS IN NONOBESE WOMEN
pubertal girls (1) suggested that the effect of leptin on the skel-
eton occurs in cortical bone, whereas leptin-treated ob/ob mice
were shown to gain both trabecular and cortical bone (32). Our
data indicate the associations between bone mass and serum
leptin were not as strong when BMD rather than BMC was used
as the dependent variable. BMD is influenced by bone size (33).
The association between leptin and BMD (i.e. the ratio of BMC/
bone area) is conceptually different from the association with
BMC after adjusting for bone area, which partially compensates
for the confounding influence of bone size. Furthermore, if
leptin promotes periosteal bone apposition, the amount of min-
eral at the bone-soft tissue interface might increase, resulting in
an increase in apparent bone area. Thus, BMD may remain
relatively unchanged because any increase in BMC would be
offset by increases in bone size. Further studies on the associ-
ation among leptin concentration, cortical thickness, and med-
ullary width may indicate whether an anabolic effect occurs at
the endosteal or periosteal surface of bone.
Weight loss or gain in adult women is associated with
corresponding changes in circulating leptin levels (14, 34)
and bone mass (35). Furthermore, reduced body fat (36),
reduced leptin levels (25), and reduced bone mass (37) have
been observed among smokers. These patterns may suggest
that changes in body fat may, in part, be translated into
changes in bone mass through fluctuations in circulating
leptin levels and/or other mediators of adipose tissue origin.
Raised peripheral leptin levels may favor bone formation
while suppressing adipogenesis. At a local level, bone marrow
adipocytes produce leptin, which may enhance osteogenic ac-
tivity and inhibit adipogenic activity (12). Failure to identify
leptin receptors or leptin effects on osteoblasts in primary os-
teoblast cultures from calvaria (30) may reflect changes asso-
ciated with osteoblast differentiation. Recent data suggest that
leptin may also inhibit fetal bone resorption (13).
This cross-sectional study of the relationship between bone
mass and circulating leptin levels does not address the question
as to whether leptin is involved in skeletal growth and the
development of peak bone mass. However, the results suggest
that serum leptin may play a role in regulating skeletal mass in
nonobese adult women, and these findings need to be explored
in men. Further studies on the association between circulating
leptin levels and bone geometry may provide insight into the
mechanism of leptin’s effect on bone.
Acknowledgment
Our thanks are extended to S. Panahi and B. Skoric for their help
scanning subjects, A. de Silva and M. S. Solin for measuring serum leptin
concentrations, and B. M. Burgess for assistance in collating data.
References
1. Matkovic V, Ilich JZ, Skugor M, et al. 1997 Leptin is inversely related to age
at menarche in human females. J Clin Endocrinol Metab. 82:3239 –3245.
2. Cunningham MJ, Clifton DK, Steiner RA. 1999 Leptin’s actions on the re-
productive axis: perspectives and mechanisms. Biol Reprod. 60:216 –222.
3. Gainsford T, Willson TA, Metcalf D, et al. 1996 Leptin can induce prolifer-
ation, differentiation, and functional activation of hemopoietic cells. Proc Natl
Acad Sci USA. 93:14564 –14568.
4. Lord GM, Matarese G, Howard JK, Baker RJ, Bloom SR, Lechler RI. 1998
Leptin modulates the T-cell immune response and reverses starvation-induced
immunosuppression. Nature. 394:897–901.
5. Gainsford T, Alexander WS. 1999 A role for leptin in hemopoieses? Mol
Biotechnol. 11:149 –158.
1887
6. Bouloumie A, Drexler HC, Lafontan M, Busse R. 1998 Leptin, the product of
Ob gene, promotes angiogenesis. Circ Res. 83:1059 –1066.
7. Sierra-Honigmann MR, Nath AK, Murakami C, et al. 1998 Biological action
of leptin as an angiogenic factor. Science. 281:1683–1686.
8. Steppan CM, Swick AG. 1999 A role for leptin in brain development. Biochem
Biophys Res Commun. 256:600 – 602.
9. Kamohara S, Burcelin R, Halaas JL, Friedman JM, Charron MJ. 1997 Acute
stimulation of glucose metabolism in mice by leptin treatment. Nature.
389:374 –377.
10. Emilsson V, Liu YL, Cawthorne MA, Morton NM, Davenport M. 1997 Ex-
pression of the functional leptin receptor mRNA in pancreatic islets and direct
inhibitory action of leptin on insulin secretion. Diabetes. 46:313–316.
11. Lostao MP, Urdaneta E, Martinez-Anso E, Barber A, Martinez JA. 1998
Presence of leptin receptors in rat small intestine and leptin effect on sugar
absorption. FEBS Lett. 423:302–306.
12. Thomas T, Gori F, Khosla S, Jensen MD, Burguera B, Riggs BL. 1999 Leptin
Downloaded from https://academic.oup.com/jcem/article/86/5/1884/2847967 by guest on 17 March 2023
acts on human marrow stromal cells to enhance differentiation to osteoblasts
and to inhibit differentiation to adipocytes. Endocrinology. 140:1630 –1638.
13. Ogueh O, Sooranna S, Nicolaides KH, Johnson MR. 2000 The relationship
between leptin concentration and bone metabolism in the human fetus. J Clin
Endocrinol Metab. 85:1997–1999.
14. Maffei M, Halaas J, Ravussin E, et al. 1995 Leptin levels in human and rodent:
measurement of plasma leptin and ob RNA in obese and weight-reduced
subjects. Nat Med. 1:1155–1161.
15. Ostlund Jr RE, Yang JW, Klein S, Gingerich R. 1996 Relation between plasma
leptin concentration and body fat, gender, diet, age and metabolic covariates.
J Clin Endocrinol Metab. 81:3909 –3913.
16. Considine RV, Sinha MK, Heiman ML, et al. 1996 Serum immunoreactive-
leptin concentrations in normal-weight and obese humans. N Engl J Med.
334:292–295.
17. Goulding A, Taylor RW. 1998 Plasma leptin values in relation to bone mass
and density and to dynamic biochemical markers of bone resorption and
formation in postmenopausal women. Calcif Tissue Int. 63:456 – 458.
18. Rauch F, Blum WF, Klein K, Allolio B, Schonau E. 1998 Does leptin have an
effect on bone in adult women? Calcif Tissue Int. 63:453– 455.
19. Reid IR, Ames R, Evans MC, et al. 1992 Determinants of total body and
regional bone mineral density in normal postmenopausal women—a key role
for fat mass. J Clin Endocrinol Metab. 75:45–51.
20. Khosla S, Atkinson EJ, Riggs BL, Melton III LJ. 1996 Relationship between
body composition and bone mass in women. J Bone Miner Res. 11:857– 863.
21. 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.
22. Pasco JA, Sanders KM, Henry MJ, Nicholson GC, Seeman E, Kotowicz MA.
2000 Calcium intakes among Australian women: Geelong Osteoporosis Study.
Aust N Z J Med. 30:21–27.
23. Henry MJ, Pasco JA, Nicholson GC, Seeman E, Kotowicz MA. 2000 Preva-
lence of osteoporosis in Australian women: Geelong Osteoporosis Study. J Clin
Densitom. 3:261–268.
24. Solin MS, Ball MJ, Robertson I, et al. 1997 Relationship of serum leptin to total
and truncal body fat. Clin Sci (Colch). 93:581–584.
25. de Silva A, de Courten M, Zimmet P, et al. 1998 Lifestyle factors fail to explain
the variation in plasma leptin concentrations in women. Nutrition. 14:653– 657.
26. Bray GA. 1978 Definition, measurement, and classification of the syndromes
of obesity. Int J Obes. 2:99 –112.
27. Kleinbaum DG, Kupper LL, Muller KE. 1988 Applied regression analysis and
other multivariable methods, ed. 2. Boston: PWS-Kent Publishing Co.
28. Ribot C, Tremollieres F, Pouilles JM. 1994 The effect of obesity on postmeno-
pausal bone loss and the risk of osteoporosis. Adv Nutr Res. 9:257–271.
29. Frost H. 1973 Bone remodelling and its relationship to metabolic bone diseases.
Springfield: C.C. Thomas.
30. Ducy P, Amling M, Takeda S, et. al. 2000 Leptin inhibits bone formation
through a hypothalamic relay: a central control of bone mass. Cell. 100:197–207.
31. Liu C, Grossmann A, Bain S, et. al. 1997 Leptin stimulates cortical bone
formation in obese (ob/ob) mice. J Bone Miner Res. 12:S115. (Abstract).
32. Steppan CM, Crawford DT, Chidsey-Frink KL, Ke H, Swick AG. 2000 Leptin
is a potent stimulator of bone growth in ob/ob mice. Regul Pept. 92:73–78.
33. Seeman E. 1997 From density to structure: growing up and growing old on the
surfaces of bone. J Bone Miner Res. 12:509 –521.
34. Scholz GH, Englaro P, Thiele I, et. al. 1996 Dissociation of serum leptin
concentration and body fat content during long term dietary intervention in
obese individuals. Horm Metab Res. 28:718 –723.
35. Compston JE, Laskey MA, Croucher PI, Coxon A, Kreitzman S. 1992 Effect
of diet-induced weight loss on total body bone mass. Clin Sci (Colch).
82:429 – 432.
36. Samaras K, Kelly PJ, Spector TD, Chiano MN, Campbell LV. 1998 Tobacco
smoking and oestrogen replacement are associated with lower total and central
fat in monozygotic twins. Int J Obes Relat Metab Disord. 22:149 –156.
37. Hopper JL, Seeman E. 1994 The bone density of female twins discordant for
tobacco use. N Engl J Med. 330:387–392.