The authors have no conflict of interest.
Structural and Biomechanical Basis of Sexual Dimorphism in Femoral Neck Fragility Has Its Origins in Growth and Aging†
Version of Record online: 1 OCT 2003
Copyright © 2003 ASBMR
Journal of Bone and Mineral Research
Volume 18, Issue 10, pages 1766–1774, October 2003
How to Cite
Duan, Y., Beck, T. J., Wang, X.-F. and Seeman, E. (2003), Structural and Biomechanical Basis of Sexual Dimorphism in Femoral Neck Fragility Has Its Origins in Growth and Aging. J Bone Miner Res, 18: 1766–1774. doi: 10.1359/jbmr.2003.18.10.1766
- Issue online: 2 DEC 2009
- Version of Record online: 1 OCT 2003
- Manuscript Accepted: 21 MAY 2003
- Manuscript Revised: 8 MAY 2003
- Manuscript Received: 9 JAN 2003
- bending strength;
- femoral neck;
- hip fractures;
The structural basis for sex differences in femoral neck (FN) fragility was studied in 1196 subjects and 307 patients with hip fracture. The absolute and relative patterns of modeling and remodeling on the periosteal and endocortical envelopes during growth and aging produce changes in FN geometry and structure that results in FN fragility in both sexes and sexual dimorphism in hip fracture risk in old age.
Introduction: Femoral neck (FN) fragility in old age is usually attributed to age-related bone loss, while the sex differences in hip fracture rate are attributed to less bone loss in men than in women. The purpose of this study was to define the structural and biomechanical basis underlying the increase in FN fragility in elderly men and women and the structural basis of sex differences in hip fracture incidence in old age.
Materials and Methods: We measured FN dimensions and areal bone mineral density in 1196 healthy subjects (801 females) 18–92 years of age and 307 patients (180 females) with hip fracture using DXA. We then used the DXA-derived FN areal bone mineral density (BMD) and measured periosteal diameter to estimate endocortical diameter, cortical thickness, section modulus (a measure of bending strength), and buckling ratio (indices for structural stability).
Results: Neither FN cortical thickness nor volumetric density differed in young adult women and men after height and weight adjustment. The sex differences in geometry were confined to the further displacement of the cortex from the FN neutral axis in young men, which produced 13.4% greater bending strength than in young women. Aging amplified this geometric difference; widening of the periosteal and endocortical diameters continued in both sexes but was greater in men, shifting the cortex even further from the neutral axis maintaining bending strength in men, not in women. In both sexes, less age-related periosteal than endocortical widening produced cortical thinning increasing the risk for structural failure by local buckling of the enlarged thin walled FN. Relative to age-matched controls, women and men with hip fractures had reduced cortical thickness, but FN periosteal diameter was increased in women and reduced in men, differences are likely to be originated in growth.
Conclusions: The absolute and relative patterns of modeling and remodeling on the periosteal and endocortical envelopes during growth and aging produce changes in FN diameters, cortical thickness, and geometry that results in FN fragility in both sexes and sexual dimorphism in hip fracture risk in old age.
The mineralized skeleton is defined externally by its periosteal envelope and internally by the endocortical, trabecular, and intracortical components of the endosteal envelope (Fig. 1). As tubular bones such as the femur increase in length during growth, periosteal bone formation widens the bone's periosteal diameter, accounting for 95% of peak cortical thickness achieved during growth in males and 85% in females.(1–3) Concurrently, endocortical bone resorption excavates a medullary canal until puberty. Sex differences in long bone morphology emerge largely during the peripubertal period when periosteal apposition increases bone's diameter and cortical thickness in males while periosteal apposition decreases in females, establishing the sex differences in periosteal diameter; changes that are likely to be sex hormone-dependent based on animal studies.(4–7) Endocortical apposition in females contributes the remaining 15% of peak cortical thickness so that cortical thickness and volumetric density are no different by sex.(1–3,8) The sex difference resides in the further placement of the cortex from the neutral axis in young adult males than females at the completion of linear growth.(2,3,9)
Aging is accompanied by bone loss. As most studies of aging are carried out using bone densitometry, information regarding the structural basis of bone loss in each sex and the structural basis for sex differences in bone loss is lacking. This information is needed because bending and compressive strength in old age is determined by the cross-sectional area of the bone and the distribution of this area in relation to the neutral axis, not just by its bone mineral mass.(10–12) Just as during growth, the absolute and relative movement of the periosteal and endocortical surfaces during aging establish the cross-sectional area and its distribution in old age. The diminution in femoral neck bone mineral mass during advancing age in women and men reported using densitometry provides no information regarding the relative contributions of periosteal apposition and endocortical resorption to net bone loss during aging, and no information regarding the spatial distribution of the diminishing bone mineral mass. For example, the greater diminution in vertebral bone mineral density (BMD) in women than men fails to convey that this is the net result of greater periosteal bone formation in men rather than greater net endocortical bone resorption in women.(13–15)
Femoral neck fragility in old age is usually attributed to age-related bone loss, whereas the sex differences in hip fracture rate are attributed to less bone loss in men than in women. The purpose of this study was to define the structural and biomechanical basis underlying both the increase in femoral neck fragility in elderly men and women and the structural basis of sex differences in hip fracture incidence in old age. We measured the age- and sex-specific differences in the cross-sectional dimensions, mass, and geometry of the femoral neck in healthy women and men and in women and men with hip fractures. We asked (1) what are the age-related changes in periosteal diameter, cortical thickness, bending strength, and the risk of local buckling in men and women, and (2) is the structural and biomechanical basis of hip fracture different in women and men?
MATERIALS AND METHODS
We recruited 127 men 45–95 years of age and 180 postmenopausal women 46–94 years of age with hip fracture sustained by a fall from no greater than the standing position. These patients were recruited from the major hospitals in the Northern Corridor of Melbourne from admission and discharge records and then contacted by letters. The diagnosis was confirmed by X-ray and surgical reports. Approximately 80% of patients were recruited within 12 months of fracture (mean, 3.9 months; range, 0.1–11.9 months); 20% were recruited within 1–6 years of fracture (mean, 2 years). We excluded patients taking corticosteroids. Patients receiving treatment for osteoporosis such as bisphosphonates or hormone replacement therapy for more than 12 months before hip scans were also excluded. Five patients (3 males) receiving bisphosphonates for less than 1 year were not excluded. Several observations in men with hip fractures have been reported.(16) In addition, 801 healthy women 18–92 years of age and 395 healthy men 18–91 years of age with no history of spine and hip fractures were studied. Subjects taking medication and having illness known to affect bone were excluded. These subjects were recruited from the community by posted flyers or word of mouth as part of ongoing research in the Department. All cases and controls were white.
Among the 1196 healthy subjects, there were 275 premenopausal women and 104 men 18–43 years of age used to compare the sex differences in geometry in young adulthood and to determine the T-scores for fracture cases and elderly controls. A total of 429 postmenopausal women and 282 men 45–92 years of age (similar age range to fracture cases) were used to calculate Z-score (see Statistical analysis). In addition, 187 postmenopausal women were closely matched by age with the women with hip fracture, and 134 healthy men were matched by age with the men with hip fractures. Informed consent was obtained from all the participants. The study was approved by the Institutional Human Research Ethics Committee.
Femoral neck dimensions and hip structural parameters
Bone mineral content (BMC), areal BMD, and the dimensions of the femoral neck were measured using DXA (DPX-L; Lunar Corp., Madison, WI, USA). Femoral neck volumetric BMD was calculated as BMC divided by volume of the scanned region.(17) Femoral neck axis length (FNAL) was the length of a line following the central long axis extending from the cortical rim of the femoral head to the lateral aspect of the femur below the greater trochanter. Femoral neck periosteal diameter at the mid-point of the FNAL was measured using the DXA ruler option.(16) All measurements were made by one investigator (YD). Measurements were performed on the unfractured femoral neck in patients with hip fracture. The CV for femoral neck BMC, areal and volumetric BMD, and dimensions ranged from 1.6% to 3.1%.(16)
We used the DXA-derived areal BMD and measured periosteal diameter to estimate femoral neck mineralized bone tissue cross-sectional area (CSA, excluded bone marrow space), cross-sectional moment of inertia (CSMI), section modulus (a measure of bending strength), endocortical diameter, cortical thickness, and buckling ratio. The CSA is computed as:
where ρm is the effective density of bone mineral in fully mineralized bone tissue (1.05 g/cm3).(11)W is the femoral neck periosteal diameter at the middle point of FNAL.
The section modulus (Z) is:
and p is the trabecular porosity. ED is the estimated endocortical diameter:
where fc is the assumed proportion of cortical mass in the femoral neck (0.6 was used here). The buckling ratio (BR) is used to estimate conditions in osteoporotic bones where the cortices have become structurally unstable.(18) CT is the estimated cortical thickness from the annulus model.
The data were expressed in absolute terms, as SD scores relative to the young normal mean (T-scores), or as age-predicted mean (Z-scores) adjusted for height and weight and age.(19) One-sample t-tests were used to determine whether the T- or Z-scores differed from zero. Because the dimensions of bone and structural parameters are body-size dependent, analysis of covariance (ANCOVA) was used to determine the significance of any trait difference between groups adjusting for height, weight, and age. Linear regression analyses were used to define the age-related changes in body height and weight adjusted bone dimensions and structural parameters (adjusted values were created by summing individual's residuals from the regression on height and weight, with the population mean within gender). Results were regarded as statistically significant at the value of p < 0.05 (two tailed).
The results in the table are shown in absolute terms unadjusted for height and weight and as T- and Z-scores. Using these unadjusted values, young adult men had a wider femoral neck periosteal diameter (13.3%), thicker cortex (10.2%), and greater section modulus (43.2%), but not buckling ratio, than women (Table 1). After adjustment for sex differences in body height and weight (the remaining results presented below were all height and weight adjusted), sex differences remained in periosteal diameter (4.2%, p < 0.01), endocortical diameter (4.9%, p < 0.01), and section modulus (13.4%, p < 0.01); the 4.0% difference in cortical thickness between men and women did not reach statistical significance (p = 0.13).
As age advanced, femoral neck diameters widened; periosteal diameter was 11.0% wider in elderly than young men and 3.7% wider in elderly than young women (respective adjusted slopes 0.0080 versus 0.0028 cm/year, p < 0.01). Endocortical diameter was 15.2% wider in elderly men and 7.3% wider in elderly women than corresponding young persons (respective adjusted slopes 0.0095 versus 0.0046 cm/year, p < 0.01). Thus, men had an almost 3-fold greater periosteal expansion and a 2-fold greater endocortical expansion than women (Figs. 2 and 3).
In each sex, periosteal expansion across age was less than endocortical expansion as reflected in the greater slopes of endocortical than periosteal diameters (women, 0.0046 versus 0.0028 cm/year, respectively; men, 0.0095 versus 0.0080 cm/year, respectively). Thus, the surfaces approximated producing cortical thinning that occurred to a similar extent in men (by −17.4%) and in women (by −18.3%) (respective adjusted slopes, −0.0008 versus −0.0009 cm/year, not significant; Table 1; Figs. 2 and 3). Therefore, the thinned cortex was placed further from the neutral axis in elderly men and elderly women than it was in young men and women.
In addition, the displacement of the thinned cortex in elderly men relative to elderly women was 0.395 cm, three times greater than it was in young men and women (0.136 cm). The more distant geometric displacement of the thinner cortex from the femoral neck neutral axis maintains the section modulus in men. In women, the modest periosteal expansion and the reduced cortical thickness resulted in a fall in section modulus by 6.9%. The risk of structural failure by local buckling (buckling ratio) increased with age by 28.6% in women and by 30.0% in men due mainly to enlargement of the femoral neck periosteal diameter with comparable decline in cortical thickness (Table 1; Fig. 4). Volumetric BMD was also not significantly different between men and women with hip fracture and was similarly reduced relative to young (both −2.40 SD) and age-matched controls (−1.17 versus −0.90 SD; Table 1).
Compared with age-matched controls, women with hip fracture had a wider femoral neck periosteal diameter (6.9%) and endocortical diameter (8.7%), reduced cortical thickness (−14.9%) and normal section modulus, and increased buckling ratio (34.0%). Men with hip fractures had reduced femoral neck periosteal diameter (−2.4%), normal endocortical diameter, reduced cortical thickness (−18.9%) and section modulus (−21.0%), and increased buckling ratio (32.3%). The difference in periosteal diameter between men and women with hip fractures was 0.34 cm, one-half that in controls (0.64 cm). The buckling ratio was similar in men and women with hip fractures (13.88 versus 13.44, respectively) and was similarly elevated relative to age-matched controls (34.0% versus 32.3%; Table 1, Figs. 2 and 4).
The main finding of this study is that the geometry of the femoral neck establishes bone strength in each sex, and sex differences in geometry account for sexual dimorphism in bone fragility. The periosteal diameter of the bone and the proximity of the endocortical surface to the periosteal surface establishes cortical thickness and the distance the cortex is placed from the neutral axis of the bone, as well as the relationship between the periosteal diameter of the femoral neck and its cortical thickness; both are important biomechanical determinants of bone strength. Differences in the distance the cortex is placed from the neutral axis of the bone, rather than the cortical thickness itself, accounts for sex differences in bone strength in old age. This structural organization of the femoral neck in old age is established by the absolute and relative modeling and remodeling that takes place on these surfaces throughout the whole of life-during growth as well as during aging. It is with this perspective that we discuss the results of this study.
Young healthy adult men had a wider femoral neck periosteal diameter, a wider endocortical diameter, and a thicker cortex than women before adjusting for sex differences in height and weight. However, after accounting for sex differences in stature by adjusting for sex differences in height and weight, both periosteal and endocortical diameters, but not cortical thickness, remained wider in men. This more distant placement of the cortex from the femoral neck neutral axis in young adult men than in young adult women is the result of sex differences in the growth patterns of the periosteal and endosteal surfaces during infancy and childhood that are likely to be, in part, sex hormone dependent.
During prepubertal growth, the extent of periosteal apposition and net endocortical resorption is similar in males and females so that the dimensions of long bone cross-sections, its periosteal diameter and cortical thickness do not differ(2,3) (Fig. 5). Because periosteal apposition is greater than net endocortical resorption in both sexes, the enlarging bone develops a thicker cortex in both sexes, but there are still little sex differences in these dimensions until puberty. During puberty, cortical thickening in males occurs by further periosteal apposition, which also enlarges the femoral neck periosteal diameter. However, in females, periosteal apposition decreases at puberty limiting femoral neck periosteal diameter while net endocortical apposition occurs thickening the cortex. There is, therefore, no sex difference in cortical thickness or volumetric BMD at the completion of puberty; what differs by sex is the spatial distribution of the cortex. The greater peripubertal periosteal apposition and net endocortical resorption in males than females shifts the whole cortex more distant from the neutral axis in males than in females. Therefore, the greater bending strength in young adult males than females is the result of the sexual dimorphism in bone geometry produced during puberty; there is no sex difference in cortical thickness or volumetric BMD in young adulthood reported here or in other studies.(20)
The morphological changes that occur during adulthood are an extension of this pattern of growth and amplify the sex difference in bone geometry established during growth. Widening of the femoral neck periosteal diameter (by periosteal apposition) and endocortical diameter (by net bone resorption) continues in both sexes, but both surfaces (as in peripubertal growth) widen more in men than in women, so that the sex difference in the position of the cortex established during growth increases even further. However, in sharp contrast to growth, periosteal widening becomes less than endocortical widening in both sexes, so that the slowly enlarging bone develops a thinner and thinner cortex (whereas during growth the rapidly enlarging bone developed a thicker and thicker cortex).
As the degree of approximation of the periosteal and endocortical surfaces in men and women remained the same, the cortical thickness and volumetric density produced was no different in elderly men and women in old age, although both are reduced relative to their young normal counterparts. Thus, as reported during growth, it is the more distant placement of the thinned cortex in men than in women, not its thickness or volumetric BMD in old age that results in maintenance of bending strength across age in men (despite the decline in cortical thickness). In women, the modest widening of the femoral neck fails to offset cortical thinning so that bending strength declines with age.
Thus, periosteal apposition plays a most important role in establishing sex differences in the bending strength in young adulthood and in maintaining or amplifying these sex differences in the face of bone loss during advancing age. The reason for this is that bending strength of a unit of bone mineral is proportional to the fourth power of the distance from the neutral axis.(11) Therefore, during growth, even a small amount of bone deposited on the periosteal surface, distant from the neutral or long axis of the shaft, contributes more to bending strength than the same amount of bone deposited on the endocortical surface. Likewise, during aging, even a small amount of bone deposited on the periosteal surface offsets the loss of strength incurred by a greater amount of bone resorbed from the endocortical surface in men, but only partially does so in women.
The strength of bone does not only reside in its resistance to bending. Structural stability is also determined by the resistance to structural failure by local buckling (collapse) during loading, and this is in turn, is determined by the relationship of the periosteal diameter of the bone to its cortical thickness.(18,21) Thus, in both sexes, the enlarged bone had a thinner cortex, resulting in structural instability caused by a tendency for structural failure by local buckling (collapse). This increased risk for local buckling was captured in the buckling ratio, which increased with age similarly in men and women and was elevated to a similar degree in men and women with hip fractures. Therefore, men are at risk of hip fracture as well as women and may have a similar threshold for risk of structural failure by local buckling. However, fewer men than women may fracture their hip because men maintain their bending strength throughout life while women do not.
Women with hip fractures had a wider femoral neck periosteal diameter and endocortical diameter and reduced cortical thickness relative to age-matched controls. Men with hip fractures had a narrower femoral neck periosteal diameter but normal endocortical diameter and reduced cortical thickness, relative to age-matched controls. These structural abnormalities can be explained by considering the differing behavior of the periosteal and endocortical surfaces during growth and aging; the observations cannot be explained by “excessive bone loss” during aging alone.(17) In females, it is well established, at least in animal studies, that estrogen production reduces periosteal apposition and increases endocortical apposition.(7,9)
Therefore, we propose that relative estrogen deficiency during growth may remove the inhibitory effect of estrogen on periosteal apposition, resulting in the wider femoral neck periosteal diameter, while loss of the modest endocortical apposition by relative estrogen deficiency will result in the wider endocortical diameter, as reported in rats(4,7) and in this study. Relative testosterone deficiency reduces periosteal apposition in male rats, producing the smaller femoral neck periosteal diameter with little effect on the endocortical diameter(7); this was also observed in this study. In rats, the greater periosteal diameter in females and reduced periosteal diameter in males after prepubertal gonadectomy reduces the sex differences in periosteal diameter.
In this study, the difference in periosteal diameter between men and women with hip fractures was halved relative to that in controls of the same age. Moreover, the deficit in cortical thickness in men with hip fractures relative to their controls was −1.34 SD. This was twice the deficit in women relative to their controls (−0.72 SD) perhaps because relative testosterone deficiency reduces periosteal apposition, while relative estrogen deficiency reduces endocortical apposition that is compensated for, in part, by greater periosteal apposition.
These results suggest that the structural abnormalities found in women and men with hip fractures have their origins, at least in part, in growth. Bending strength was not reduced in women with hip fractures because they enter menopause with increased femoral neck periosteal diameter, but structural failure by local buckling may follow as estrogen deficiency erodes the cortex of a larger bone producing structural instability. A growth-related origin of these abnormalities is supported by finding FNAL (which does not increase after epiphyseal closure) was greater in women with hip fractures. In addition, femoral neck periosteal diameter is increased in premenopausal daughters of women with hip fractures. If the larger femoral neck periosteal diameter was the result of greater age-related periosteal apposition, their daughters would not have increased femoral neck periosteal diameter.(17) The offspring of men with fractures have reduced bone size,(22) so that men with hip fractures probably enter adulthood with a narrow femoral neck periosteal diameter that results in a reduced bending strength. Finding the smaller femoral periosteal diameter is a risk factor for stress fractures among military recruits (19 years old) in young men but not in young women (both had lower bending strength) may also support this notion.(23) The risk of buckling emerges as endocortical resorption thins the cortex.
This study has several limitations. Being cross-sectional, secular changes may influence the results. Most studies suggest periosteal apposition is greater in men than women.(24–28) However, others suggest it occurs only in men,(29–32) occurs to a similar extent or no increase in each sex,(11,33) or occurs more greatly in women.(34) Several reasons may account for these disparate findings. All of these studies are cross-sectional and subject to secular trends in body dimensions that may be population- and sex-specific.(35) However, secular trends produce increases in height. If this is correct, then the changes we have observed are likely to be conservative; earlier born cohorts (forming the elderly in a cross-sectional sample) are likely to have had a shorter stature in youth than later born persons (forming the young normal group). Secular effects will not influence comparison of hip fracture cases with age-matched controls. All of the hip fracture cases were scanned after fracture, so some of the observed differences between cases and controls may be caused by postfracture bone loss caused by decreased ambulation, use of walking devices, and decreased nutrition.
In summary, sex differences in the diameter of the long bone, the thickness of the cortex, and the distance the cortical shell is placed from the neutral axis of the femoral neck are determined by the absolute and relative movements of the periosteal and endocortical surfaces during growth and during aging.(36,37) In men, greater periosteal apposition than in women offset their greater endocortical bone loss, maintaining the bending strength of bone. Smaller FN diameter in men with hip fractures suggests that reduced periosteal apposition during growth or aging may contribute to the pathogenesis of femoral neck fragility. In women, hip fracture and a higher FN diameter in young adulthood confers as a structural disadvantage as endocortical resorption erodes the cortex in which modest periosteal apposition fails to compensate, increasing the fracture risk by increasing the risk of buckling. Thus, periosteal bone formation throughout life plays a most important role in determining bone strength in old age within a population of the same sex, between individuals of the opposite sex, and perhaps between individuals differing by racial or ethnic origin. The differing extent of periosteal apposition and net endocortical resorption in men and women during both growth and aging largely explains the sexual dimorphism in bone geometry and bone fragility.
We thank Research Nurse Jan Edmonds for assistance with subject recruitment during this study. We also thank the senior technologists, Alison Evans and Patricia D'Souza, for their technical assistance. This work was supported partly by a grant from Australia National Health and Medical Research Council (G145820).
- 11970 The Earlier Gain and Later Loss of Cortical Bone. CC Thomas Publishers, Springfield, IL, USA.
- 21999 The differing tempo of growth in bone size, mass, and density in girls is region-specific. J Clin Invest 104:795–804., , , , ,
- 32000 Heterogeneity in the growth of the axial and appendicular skeleton in boys: Implications for the pathogenesis of bone fragility in men. J Bone Miner Res 15:1871–1878., , , , ,
- 41989 Different effects of gonadal function on bone histomorphometry in male and female rats. J Bone Miner Res 4:557–563., , , ,
- 51995 Early induction of alterations in cancellous and cortical bone histology after orchiectomy in mature rats. J Bone Miner Res 10:1735–1744.,
- 61999 The effect of gonadectomy on bone size, mass and volumetric density in growing rats are gender-, site-, and growth hormone-specific. J Bone Miner Res 14:802–809., , , ,
- 72003 The structural and hormonal basis of sex differences in peak appendicular bone strength in rats. J Bone Miner Res 17:150–155., , , , , ,
- 81997 Differential effect of gender on the sizes of the bones in the axial and appendicular skeletons. J Clin Endocrinol Metab 82:1603–1607., , , , ,
- 92001 The development of bone strength at the proximal radius during childhood and adolescence. J Clin Endocrinol Metab 86:613–618., , ,
- 102001 Sexual dimorphism in skeletal size, density, and strength. J Clin Endocrinol Metab 86:4576–4584.
- 112000 Structural trends in the aging femoral neck and proximal shaft: Analysis of the third national health and nutrition examination survey dual-energy x-ray absorptiometry data. J Bone Miner Res 15:2297–2304., , , ,
- 121991 Toward a cure for osteoporosis: Reversal of excessive bone fragility. Osteoporos Int 2:12–19.
- 131990 Sex differences in age-related changes in vertebral body size, density and biomechanical competence in normal individuals. Bone 11:67–73.,
- 142001 Sexual dimorphism in vertebral fragility is more the results of gender differences in bone gain than bone loss. J Bone Miner Res 16:2267–2275., , ,
- 152001 The biomechanical basis of vertebral body fragility in men and women. J Bone Miner Res 16:2276–2283., ,
- 162001 Fracture site-specific deficits in bone size and volumetric density in men with spine or hip fractures. J Bone Miner Res 16:120–127., , ,
- 172001 The contribution of reduced peak accrual of bone and age-related bone loss to osteoporosis at the spine and hip: Insights from the daughters of women with vertebral or hip fractures. J Bone Miner Res 16:1101–1107., , ,
- 182001 Structural adaptation to changing skeletal load in the progression toward hip fragility: The study of osteoporotic fractures. J Bone Miner Res 16:1108–1119., , , , , , ,
- 191989 Reduced bone mass in the daughters of women with osteoporosis. N Engl J Med 320:554–558., , , , ,
- 202001 Does body size account for gender differences in femur bone density and geometry? J Bone Miner Res 16:1291–1307., ,
- 212001 Effects of current and discontinued hormone replacement therapy on hip structural geometry: The study of osteoporotic fractures. J Bone Miner Res 16:2103–2110., , , , ,
- 222003 Deficient acquisition of bone during maturation underlies idiopathic osteoporosis in men: Evidence from a three-generation family study. J Bone Miner Res 18:303–311., , , ,
- 231996 Dual-energy X-ray absorptiometry derived structural geometry for stress fracture prediction in Male U.S. marine corps recruits. J Bone Miner Res 11:645–653., , , , , , ,
- 241977 Age and sex-related changes in the structure and strength of the human femoral shaft. J Biomech 10:223–231.,
- 251988 Sex differences in age-related remodeling of the femur and tibia. J Orthop Res 6:886–896.,
- 261992 Continuing bone expansion and increasing bone loss over a two-decade period in men and women from a total community sample. Am J Human Biol 4:57–67., , , ,
- 271983 The effects of composition, structure and age on the torsional properties of the human radius. J Biomech 16:603–608.,
- 282000 Regional variations in cortical modeling in the femoral mid-shaft: Sex and age differences. Am J Phys Anthropol 112:191–205., , ,
- 291997 Bone densitometry of proximal femur in Chinese subjects: Gender differences in bone mass and bone areas. Bone 20:365–369., , , , ,
- 302002 Young-old differences in vertebral and proximal femoral bone density and size: The Mr Os Study. J Bone Miner Res 17:S1.;S179, , , , ,
- 312002 Young-old differences in proximal femoral BMD and size in women. J Bone Miner Res 17:S1.;S265, , , , ,
- 322002 Geometry of the hip in relation to age and sex-a cross-sectional study in healthy adults. J Bone Miner Res 17:S1.;S265, , , , ,
- 331994 Geometric structure of the femoral neck measured using dual-energy X-ray absorptiometry. J Bone Miner Res 9:1053–1064., , , , , , ,
- 341972 Confirmation of the sex difference in continuing subperiosteal apposition. Am J Phys Anthropol 36:377–380., , ,
- 351964 Secular increase in height: Is the end in sight? Lancet 5:1195–1196.,
- 361992 Bone resorption and formation on the periosteal envelope of the ilium: A histomorphometric study in healthy women. J Bone Miner Res 7:1475–1482., ,
- 372002 Parathyroid hormone and periosteal bone expansion. J Bone Miner Res 17:1741–1743.