| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Original Studies |
Departments of Pediatrics (L.K.B., M.-C.W.), Statistics (T.H., B.N.), Health Research and Policy (T.H.), and Medicine (R.M.), Stanford University School of Medicine, and the Musculoskeletal Research Laboratory, Geriatric Research, Education, and Clinical Center, Veterans Affairs Medical Center (R.M.), Palo Alto, California 94305
Address all correspondence and requests for reprints to: Laura K. Bachrach, M.D., Room S302, Stanford Medical Center, Stanford, California 94305-5208. E-mail: lkbach{at}leland.stanford.edu
 |
Abstract |
|---|
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Despite the proliferation of pediatric research in the past decade (2, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24), several aspects of bone mineral acquisition remain controversial. The timing of peak bone mass is disputed; some studies conclude that bone mass is maximal by age 20 yr (4, 5, 6, 7, 8, 14, 15, 16, 17), whereas other reports indicate measurable bone mineral gains into the third decade (9, 18). In part, these differences may reflect small sample sizes or cross-sectional (10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 23) vs. longitudinal design (4, 5, 6, 7, 8, 9). The tempo of bone mineral accrual has been shown to vary by skeletal site, with gains continuing longer at whole body than at hip or spine (6, 7, 9). Ethnic differences in bone mass have been observed in some (20, 22, 24), but not all (11, 12), studies. Similarly, there are discrepancies concerning the magnitude of gender differences in bone mass. These controversies have persisted in part because of a relative paucity of data from males and nonwhites. Variability in reported ethnic or gender differences also results when different noninvasive methods are used to assess bone mineral. Finally, the influence of exercise, diet, and other lifestyle variables on bone mineral acquisition is debatable (2, 8, 16). In some studies, activity and calcium intake have been estimated to make a 5–10% difference in bone mass (5, 8, 9), whereas others have found no significant correlation (4). These divergent findings may be explained by the lack of standardized instruments to quantitate diet and activity and a paucity of longitudinal bone accrual studies.
Perhaps the most debated issue in pediatric bone research is the optimal technique for determining bone mass in growing children. Dual energy x-ray absorptiometry (DXA) is widely accepted as the preferred method for assessing bone mass in children because of its speed, precision, availability, and modest radiation exposure (25). Despite these advantages, DXA is limited because it does not measure volumetric bone density. By convention, bone mass is reported as bone mineral content (BMC; grams) or areal bone density (BMD; grams per cm2), terms that are strongly influenced by bone size (26, 27). As a result, DXA may overestimate true bone mass in larger individuals and underestimate bone mass in smaller individuals. Several investigators have derived models to estimate volumetric bone density to reduce the influence of bone size on DXA measurements (4, 5, 13, 27). Although the validity of these models remains controversial (26, 28, 29), they provide a reasonable approach to adjusting for bone size. Quantitative computed tomography (QCT) circumvents the measurement artifacts associated with DXA by measuring volumetric bone mineral density directly. However, QCT is not widely available and requires considerably higher radiation exposure than DXA (25). Furthermore, published pediatric norms for QCT (20, 30) are more limited than those for DXA. Quantitative ultrasound is emerging as an attractive method to evaluate pediatric bone mineral because it is portable, inexpensive, and requires no ionizing radiation (25). Quantitative ultrasound has not supplanted DXA, however, because this technique is less precise (25) and pediatric reference data are limited (31, 32).
This longitudinal study was designed to examine the influence of gender and ethnicity on bone mineral acquisition in healthy youths and young adults (aged 9–25 yr). A total of 423 black, Asian, Hispanic, and white males and females had measurements of bone mass made by DXA at study entry and yearly thereafter for up to 4 yr. Growth, pubertal development, diet, and exercise were monitored during this period as well. Normative curves for areal (BMD) and volumetric bone mineral apparent density (BMAD) at the spine, proximal hip, and whole body have been modeled from these longitudinal data.
|
Subjects and Methods |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
A convenience sample of healthy youth was recruited from the community through advertisements and personal contact (21, 22). Individuals with a history of medical conditions or use of medications affecting bone mineral were excluded. Subjects were encouraged to return annually for a total of four visits or until they had reached age 26 yr. Recruitment occurred between May 1992 and February 1996; data collection ended in February 1997. The cohort at entry included 103 non-Hispanic whites, 103 Hispanics, 103 Asians, and 114 non-Hispanic blacks, aged 8.8–25.9 yr; 230 females and 193 males were enrolled as previously reported (22). For simplicity, ethnicity and race will be used as interchangeable terms, and the groups will be referred to as white, Hispanic, Asian, and black. A total of 280 subjects completed 2 visits; 189 were studied 3 times, and 113 were evaluated 4 times. Subjects who completed fewer than 4 visits included those who refused, relocated, or reached age 26 yr during the study period; in addition, subjects who were recruited late in the study did not complete all visits because funding had terminated.
Clinical assessment
All assessments performed at study entry were repeated at each annual follow-up visit (21, 22). In brief, height was measured using a wall-mounted stadiometer, and weight was determined with a balance beam using a standardized protocol. Pubertal stage was estimated with a self-assessment instrument using Tanner classifications of breast development in females and genital development in males (33). Calcium/protein ratio was assessed using the NCI Food Questionnaire (34). Habitual physical activity was determined from a structured interview directed at recall of recreational exercise during the previous 12 months.
Bone densitometry
Bone mineral was measured by DXA (QDR 1000W, software version 6.10, Hologic, Inc., Waltham, MA) in the pencil beam mode. Regions of interest included the lumbar spine (L2–L4), left proximal femur (femoral neck and total hip), and whole body. A standardized protocol was employed for analysis of the proximal hip to reduce measurement inconsistencies that may result from changes in femoral neck geometry during growth (22). BMC (grams) and BMD (grams per cm2) were determined by Hologic, Inc., software. BMAD (grams per cm3), an estimate of volumetric bone density, was calculated as previously described (4, 27). Spine BMAD was derived using the formula BMC ÷ (area)1.5; femoral neck BMAD was calculated from the expression BMC ÷ (area)2. The expression whole body BMC ÷ height (cm) was calculated to adjust for whole body bone size. In our laboratory, the in vivo coefficient of variation for replicate measurements of BMD in adolescents and young adults is 0.6% at all skeletal sites studied.
The study protocol was approved by the Stanford University administrative panel on human subjects in medical research. Written consent was obtained from all participants and from the parents of subjects under age 18 yr.
Statistical models and methods
We used several biostatistical techniques to model the bone mineral percentile curves. A mixed effects model was employed to allow inclusion of all data from each subject despite irregularly spaced measurements or missing values (35). The model also adjusts for measurement errors and other random fluctuations. Semiparametric regression models were used to permit flexible modeling of the curves using splines (36). After inspection of the data, we selected knots (ages) where shifts in the bone acquisition curves were greatest. A piecewise cubic polynomial model was applied between adjacent pairs of knots, with constraints built in to force the adjacent cubic polynomial to join up smoothly. Robust nonparametric smoothing techniques were used to provide estimates of the SD as a function of age (37). A detailed description of these modeling techniques is provided in the references (35, 36, 37).
Fitting the models
The models were fit by maximum likelihood estimation that incorporates estimates of random error between and within individuals in a single model. This method adjusts for highly correlated repeated measurements from single individuals, thus allowing examination of fixed effects (such as ethnicity). We used the linear mixed effects software in SPlus (version 4.3) for this modeling and fit separate models for males and females (38).
Age-adjusted mean and SD curves were produced for areal BMD of the spine, femoral neck, total hip, and whole body. Curves for spine and femoral neck BMAD and whole body BMC/height ratio were also generated as estimates of volumetric bone density. For the bone measurements at each site, we tested whether models varied by ethnic group. Where significant ethnic differences in mean bone mass were observed, separate curves were created. If differences between two or more racial groups were not significant, data were pooled to maximize the number of subjects in the reference curves.
Modeling the tempo of bone acquisition
To examine gender and ethnic differences in the tempo of bone mineral acquisition, we applied a simpler statistical model than that used to create the SD curves. Analysis was designed to estimate the age at which bone accrual reached a plateau, and a simple linear model of bone mineral acquisition rates was used between age 9 yr and the age at which bone acquisition leveled off. We used a grid-search technique to localize the age at plateau and then tested the validity of that assumption using a bootstrap technique (39). For each ethnicity and gender, we determined the age at which bone acquisition plateaus, the mean bone mineral achieved by that age, and the rate of bone acquisition (the slope of the curve between age 9 yr and the age at plateau).
All measures of bone mass increased with age except femoral neck BMAD, which appeared to decline transiently during midpuberty. To explore the reason for this change, we plotted both femoral neck BMC and area curves, obtained by smoothing the observed data. These curves were scaled to fit on the same axes, and the curves and their derivatives were compared.
|
Results |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Entry data for the age, height, weight, pubertal stage, dietary intake, and physical activity of the study subjects have been published previously (21, 22). Two changes were made in the original cohort after the initial report; 1 black male subject was subsequently excluded because of a skeletal contracture, and 1 Hispanic female was added to the cohort. Subjects were asked to return annually for up to 4 visits or until age 26 yr. Of 423 subjects studied at baseline, 280 subjects completed 2 visits, 189 were studied 3 times, and 113 were evaluated 4 times.
Tables 1
and 2 summarize the number of subjects
studied at each age and the mean anthropometric and bone mineral data.
Data from all visits for each subject are included. Bone mineral
results for each subject are also superimposed on the age-adjusted
curves (Figs. 1 and 2) to show the ethnic distribution by age
and to display the individual variability in bone mineral acquisition
rates. Participants ranged in age from 9–25 yr at entry. Of the 230
females, 17% were prepubertal or in early puberty, 35% were at
midpuberty, and 48% were pubertally mature. Among the 193 males, 28%
were in pre- or early puberty, 33% were midpubertal, and 39% had
reached sexual maturity. The mean age for each pubertal group did not
differ by ethnicity. There were also no ethnic differences in mean
weight, height, or body mass index (BMI) for males or females until
late puberty. Among the mature females, Hispanics were shorter than
whites; black and Hispanic females had greater BMI than Asian and white
females. Asian males weighed less than white males at maturity, but did
not have significant differences in BMI. A few ethnic differences in
diet were also observed. The mean calcium/protein and calcium/energy
ratios were greater in white than in Asian or black females; white
males consumed more calcium per g protein than Asian males in early and
late puberty (22). There were no significant differences in habitual
weight-bearing physical activity between ethnic groups.
|
|
|
|
Mean and SD curves for spine bone density (BMD) by age
are shown for females (Fig. 1
) and males (Fig. 2). Mean BMD for black
females was approximately 10% greater than mean BMD in nonblacks.
Differences among white, Hispanic, and Asian females were not
significant, and data from these three groups were combined. Like the
females, black males had greater areal bone density at the spine than
all nonblack subjects; however, white and Asian males had greater mean
spine BMD than Hispanics. The difference between blacks and whites was
approximately 3% at the spine (Fig. 2).
Age-adjusted mean and SD curves for BMD and BMAD for all other skeletal sites are provided at the study web site (http://www-stat.stanford.edu/pediatric-bones). This site has been designed to provide a standard deviation or z-score for BMD or BMAD if age, gender, race/ethnicity, and densitometry data (BMC, BMD, and bone area) are entered.
Figure 3 summarizes ethnic differences in
areal and volumetric bone mineral densities at all skeletal sites. For
each measure of BMD or BMAD, the analysis determined whether there was
a shift upward or downward in the mean levels for Asians, blacks, or
Hispanics compared with the mean level for the white cohort. The
dashed line represents the mean level of age-adjusted BMD or
BMAD for the white cohort, and the horizontal bars indicate
the mean level (±SE) for males and females in
each ethnic group. Among females (right column), blacks had
significantly greater mean values for all measures of bone mineral at
all skeletal sites. As estimated volumetric (BMAD) as well as areal
(BMD) bone mineral densities were significantly greater in black
females, the observed differences could not be attributed simply to
racial differences in bone size. Differences in the mean bone mineral
levels between the nonblack groups at some sites reached statistical
significance but were more modest than black-white differences. Asian
females had significantly lower femoral neck BMD, whole body BMD, and
whole body BMC/height ratio than Hispanic and white subjects. Femoral
neck BMAD in Asian and white females was lower than that in
Hispanics.
|
). Some ethnic differences in mean bone mineral
curves were also found among the nonblack groups. Spine BMD was greater
in whites and Asians than in Hispanics (Fig. 2); whites also had
greater mean total hip BMD, whole body BMD, and whole body BMC/height
ratio than Asian and Hispanic males.
The magnitude of change in areal bone density during puberty was
greater than the change in estimated volumetric bone density. In
females, BMD increased approximately 50% at the spine (Fig. 1), 25%
at the femoral neck, 30% at the total hip, and 30% for the whole body
between ages 9–18 yr. Males showed gains of similar or slightly
greater magnitude from 9–18 yr (Fig. 2). Spine BMAD increased by
approximately 20% in females and males, and no net increase in femoral
neck BMAD occurred during adolescence.
The tempo of bone mineral acquisition
All measures of bone mineral increased with age except for femoral neck BMAD, which varied in midadolescence but did not increase significantly between early and late adolescence. Femoral BMAD values declined transiently between ages 10–13 yr in females and between ages 13–15 yr in males. To explore the reason for this change, we compared the changes in femoral neck BMC and area curves. The transient decrease in femoral BMAD coincided with more rapid gains in femoral neck area than in BMC, resulting in a transient decline in the ratio of BMC to (area).2 The subsequent rise in femoral BMAD occurred as gains in femoral neck area leveled off and BMC continued to increase (data not shown).
To examine the effects of gender and ethnicity on the tempo of bone
mineral acquisition, we used a simpler statistical model as described
in Materials and Methods. For males and females in each
racial group, we estimated the inflection point at which gains in BMD
reached a plateau, the mean BMD value at that age, and the rate (slope)
of bone mineral acquisition from early puberty until the plateau was
reached. Data for spine BMD are shown in Fig. 4. For the entire female cohort, total
hip, spine, and whole body BMD reached a plateau at ages 14.1, 15.7,
and 16.4 yr, respectively. Black and Asian females tended to reach this
plateau earlier than white and Hispanic females. For males, the plateau
in total hip BMD was observed at 15.7 yr, whereas spine and whole body
BMD leveled off at age 17.6 yr. Asian males tended to reach maximal BMD
earlier than the mean at total hip and spine.
|
). Blacks had significantly greater areal bone densities than
nonblacks at the spine, total hip, and whole body. For black females,
the mean spine and hip bone density was approximately 0.06
gm/cm2 greater than the mean for all females; for
whole body, the difference was 0.05 gm/cm2. Black
males achieved a mean BMD that was approximately 0.025, 0.03, and 0.05
gm/cm2 greater for spine, whole body, and total
hip, respectively, than the mean for all males (data not shown). |
Discussion |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
The combination of statistical methods used to generate the bone mineral curves optimized the information to be gained from our longitudinal data. The mixed effects model allowed the inclusion of data from all subjects regardless of the number of study visits and the variable intervals between visits. The model made appropriate adjustments for measurement errors and for inclusion of multiple data points from the same individual. Cubic regression splines permitted flexible modeling of the bone acquisition curves through adolescence, when the velocity of gains vary considerably. We used robust nonparametric smoothing to estimate the magnitude of the SD. We found that the SD from the mean was not fixed but, rather, increased with age. Maximum likelihood estimations allowed us to examine the effect of ethnicity by adjusting for the repeated measurements taken from the same subject.
We found that blacks had significantly greater areal and volumetric bone density than nonblacks (Hispanic, Asian, and white) at all skeletal sites. For example, mean spine BMD for age was approximately 10% greater in black females and 3% greater in black males. Several previous studies from our group (22) and others (11, 20, 24, 40, 41, 42) observed that black American youths have 5–23% more bone mass than whites depending upon skeletal site, age, and gender. Significant differences have not been found between black and white South African youths (12). The differences we observed in bone mass between blacks and whites could not be attributed entirely to racial differences in bone size, because blacks also had greater estimated volumetric bone density (BMAD). Gilsanz et al. also found that QCT measurements of volumetric bone density at the spine were greater in blacks than whites (20, 42). By contrast, they observed no racial difference in femoral shaft cortical bone density, although blacks had greater cross-sectional area in that region (20, 42). Data from cross-sectional studies indicated that racial differences in bone mass increase during puberty because black adolescents gain more bone mineral in late adolescence (20, 22, 42). However, one longitudinal study found that bone mass increased more in black than in white children, aged 8–10 yr (24). We were unable to determine whether there were racial differences in the rates of bone accrual at differing ages because the statistical approach used assumed the shape of the bone mineral curves to be similar for all groups.
We observed fewer significant differences in bone mineral among Asian, Hispanic, and white subjects, and the magnitude of these differences was more modest than that of the black vs. nonblack differences. Asian females had lower mean femoral neck BMD, whole body BMD, and whole body BMC/height ratio than whites and Hispanics; femoral neck BMAD in Asian and white females was lower than that in Hispanics. Among males, spine BMD was lower in Hispanics than in Asian or whites. Total hip BMD, whole body BMD, and whole body BMC/height ratio were also greater in whites than in Hispanic or Asian males. The few differences we observed between Hispanic and white cohorts contrast with data of McCormick et al. (11), showing no significant differences in spine BMD (corrected for body weight) between Hispanic and white children (11). Once again, the discrepant findings may reflect differences in cohort size, cross-sectional vs. longitudinal study design, or the statistical tests applied. In our study, Asian youths had lower measures of bone mineral at some sites than their white American counterparts, similar to findings from several adult studies (43). Some of the apparent differences in BMD between Asians and Caucasians can be explained by racial differences in bone size and body weight (21, 22, 43). We speculate that the Westernized diet and activity patterns of our largely American-born Asian cohort attenuated ethnic differences in bone size and geometry (44), thus reducing differences in areal bone density between Asians and whites.
We found more ethnic differences in bone mineral in this longitudinal study than we found at baseline in this cohort (21, 22). At study entry, differences between blacks and whites were not observed at all skeletal sites and pubertal stages (22). Mean BMD and BMAD for Asian, Hispanic, and white subjects were similar until late puberty, when Asian males had lower whole body BMD and BMC/height ratio than whites (22). Several factors may explain the apparent increase in ethnic differences in the current study. Firstly, the dataset used for longitudinal analyses was larger, increasing our power to detect small differences. Secondly, we applied a mixed effects model for this study compared with the regression models used for the cross-sectional data analysis. Finally, the bone mineral data presented here are adjusted only for age, whereas our earlier studies controlled for several confounding variables, such as height, weight, and pubertal stage. Although the latter approach corrected for appropriate genetic and lifestyle factors, this reduced our power to detect small ethnic differences.
We found that gender, but not ethnicity, had a significant influence on the timing of bone mineral acquisition. Gains in bone mineral plateaued earlier in females than in males. There was a tendency for Asian youth to reach a plateau earlier than blacks, whites, and Hispanics at some sites, but these differences were not significant. Our conclusions about the tempo of bone mineral acquisition are generally consistent with findings from studies in predominantly white youth in the U.S. and abroad (4, 5, 6, 7, 8, 9, 10, 11, 13, 14, 15, 16, 17, 23). In a longitudinal study of Canadian youth, Bailey et al. found that peak gains in spine, femoral neck, and whole body BMC occurred at 13 yr in females and 14.4 yr in males, approximately 1 yr after peak height velocity (45). Lu et al. found that spine BMD plateaued at ages 15.7 and 17.4 yr in Australian females and males, respectively, similar to our estimates of 15.7 and 17.6 yr (7). Most studies have concluded that females and males attain 95–100% of their peak bone mass by the late teen years (6, 14, 15, 16, 17, 19). However, Recker et al. found that women gained 5% and 12% in spine and whole body BMD, respectively, during the decade between ages 20–29 yr (9). We had insufficient numbers of older subjects to address the timing of peak bone mass.
Our data support the conclusion that pubertal gains in areal bone density largely reflect increases in bone size rather than changes in true volumetric bone mineral density (4, 13, 26). Between early and late adolescence, BMD increased dramatically at all sites. By contrast, gains in estimated BMAD at the spine were more modest, and there was no net increase in femoral neck BMAD. These findings support prior studies showing that estimated volumetric bone density increases at the spine, but not at the femoral neck or shaft, during adolescence (4, 5, 13). The changes we observed in BMAD are also consistent with QCT data, indicating pubertal gains in both size and volumetric density of vertebrae, whereas the volumetric bone density of the femoral shaft does not change (26, 30). The transient decrease that we observed in femoral neck BMAD during early puberty has not been reported previously, perhaps because earlier studies were cross-sectional or because mixed effect modeling was not used to analyze the data.
We acknowledge several limitations of this study. The cohort studied was a convenience sample, which may not be representative of all American youth. Despite a reasonably large cohort size at study entry, the number of subjects in each age, ethnicity, and gender group at baseline was limited because of the spectrum of age and ethnicity. Furthermore, the cohort size was diminished by attrition in subsequent years, particularly among nonwhite participants. Attrition rates of 43–51% have been observed in studies of black adults (46, 47). Retaining teens in a clinical study is even more challenging because of their mobility, dependence upon parental cooperation, and competing social and academic interests. Finally, it was beyond the scope of this paper to address the influence of lifestyle variables on longitudinal gains in bone mineral. Studies in predominantly white cohorts indicate that calcium intake or physical activity may account for as much as 5–10% of the variance in bone mineral accrual among children (1, 8, 16) or young adults (9). Furthermore, common lifestyle patterns may contribute to genetic factors to produce ethnic differences in bone mass. In a study of black and white adults, Ettinger et al. found that correcting for body size, composition, diet, physical activity, endocrine function, and markers of bone metabolism reduced, but did not eliminate, racial differences in peak bone mass (48). We found few ethnic differences in lifestyle in this cohort at study entry, and these variables were weak predictors of bone mineral (23). However, we intend to extend our analyses of these data to explore the correlation between diet and activity and longitudinal changes in bone mass.
Despite these weaknesses, we believe that the bone mineral curves from this study provide valuable reference norms. To date, DXA manufacturers have included little or no pediatric data in their software, making it difficult to interpret scans in younger patients. There are several published studies in predominantly white cohorts (6, 10, 13, 14, 15, 16, 17, 18, 19, 23), but few reports of Hispanic or black youth (11, 20). The only bone mineral data from Asian American youth to date were collected from our study group (21, 22). The sample size of 423 subjects makes this one of the larger studies and the only reported longitudinal study to include ethnically diverse youth. The statistical approach used to model the bone mineral curves allowed inclusion of both cross-sectional and longitudinal data. We provide gender- and ethnic-specific curves for most skeletal sites, including total hip. The use of gender-specific reference data is particularly important, because adolescent differences in the tempo of bone gain may lead to the overdiagnosis of osteopenia in males (6, 49). It is equally important to use ethnic-specific reference standards, because mean differences in bone mineral between blacks and nonblacks approximate 5–10%, the equivalent of 0.5–1.0 SD. Bone mineral data are presented in terms of both areal (BMD) and estimated volumetric (BMAD) mineral density to allow adjustments for bone size. The reference data presented here can be used without modification when interpreting bone scans obtained on the Hologic, Inc., 1000W in the pencil beam mode. When interpreting studies collected using other densitometers or software versions, it will be necessary to correct for systematic differences (50).
There is continued debate about the best method to adjust for body size and pubertal stage when interpreting pediatric DXA scans (23, 51). Our bone mineral data are presented by chronological age rather than pubertal stage to facilitate their use when pubertal stage is not available. Furthermore, the statistical model used to create the bone mineral curves required a continuous variable. However, we recognize that it may be appropriate to adjust for pubertal stage or skeletal age rather than chronological age when evaluating data from youths with premature or delayed maturation (1). For children with significant growth retardation, it may also be helpful to compare estimates of volumetric bone density (BMAD) to reduce the influence of bone size (5, 13, 27). Other investigators have suggested that BMD be adjusted for weight (11), lean body mass (52), or height (23) to correct for differences in body size. Molgaard et al. proposed that whole body BMD be replaced by measurements of bone area corrected for height and of BMC corrected for bone area and height (23). Regardless of the method used to adjust for body size, our data suggest that gender and ethnic differences in bone mass are sufficient to justify the use of sex- and race-specific reference data when interpreting pediatric bone mineral.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Received April 5, 1999.
Revised August 4, 1999.
Accepted August 19, 1999.
|
References |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  M. D. Walker, R. Novotny, J. P. Bilezikian, and C. M. Weaver Race and Diet Interactions in the Acquisition, Maintenance, and Loss of Bone J. Nutr., June 1, 2008; 138(6): 1256S - 1260S. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. J. Kalkwarf, B. S. Zemel, V. Gilsanz, J. M. Lappe, M. Horlick, S. Oberfield, S. Mahboubi, B. Fan, M. M. Frederick, K. Winer, et al. The Bone Mineral Density in Childhood Study: Bone Mineral Content and Density According to Age, Sex, and Race J. Clin. Endocrinol. Metab., June 1, 2007; 92(6): 2087 - 2099. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. Nordstrom, M. Neovius, and A. Nordstrom Early and Rapid Bone Mineral Density Loss of the Proximal Femur in Men J. Clin. Endocrinol. Metab., May 1, 2007; 92(5): 1902 - 1908. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Radovick and S. DiVall Approach to the Growth Hormone-Deficient Child during Transition to Adulthood J. Clin. Endocrinol. Metab., April 1, 2007; 92(4): 1195 - 1200. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
O. Ali, M. Shim, E. Fowler, M. Greenberg, D. Perkins, W. Oppenheim, and P. Cohen Growth Hormone Therapy Improves Bone Mineral Density in Children with Cerebral Palsy: A Preliminary Pilot Study J. Clin. Endocrinol. Metab., March 1, 2007; 92(3): 932 - 937. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. H. Tobias, C. D. Steer, C. Vilarino-Guell, and M. A. Brown Estrogen Receptor {alpha} Regulates Area-Adjusted Bone Mineral Content in Late Pubertal Girls J. Clin. Endocrinol. Metab., February 1, 2007; 92(2): 641 - 647. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. A Ward, R. L Ashby, S. A Roberts, J. E Adams, and M Zulf Mughal UK reference data for the Hologic QDR Discovery dual-energy x ray absorptiometry scanner in healthy children and young adults aged 6-17 years Arch. Dis. Child., January 1, 2007; 92(1): 53 - 59. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. J. Loud and C. M. Gordon Adolescent bone health. Arch Pediatr Adolesc Med, October 1, 2006; 160(10): 1026 - 1032. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. I. Gafni, R. Hazra, J. C. Reynolds, F. Maldarelli, A. N. Tullio, E. DeCarlo, C. J. Worrell, J. F. Flaherty, K. Yale, B. P. Kearney, et al. Tenofovir Disoproxil Fumarate and an Optimized Background Regimen of Antiretroviral Agents as Salvage Therapy: Impact on Bone Mineral Density in HIV-Infected Children Pediatrics, September 1, 2006; 118(3): e711 - e718. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Hazra, R. I. Gafni, F. Maldarelli, F. M. Balis, A. N. Tullio, E. DeCarlo, C. J. Worrell, S. M. Steinberg, J. Flaherty, K. Yale, et al. Tenofovir Disoproxil Fumarate and an Optimized Background Regimen of Antiretroviral Agents as Salvage Therapy for Pediatric HIV Infection Pediatrics, December 1, 2005; 116(6): e846 - e854. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. M. Buison, D. A. Kawchak, J. I. Schall, K. Ohene-Frempong, V. A. Stallings, M. B. Leonard, and B. S. Zemel Bone Area and Bone Mineral Content Deficits in Children With Sickle Cell Disease Pediatrics, October 1, 2005; 116(4): 943 - 949. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 V. Matkovic, P. K Goel, N. E Badenhop-Stevens, J. D Landoll, B. Li, J. Z Ilich, M. Skugor, L. A Nagode, S. L Mobley, E.-J. Ha, et al. Calcium supplementation and bone mineral density in females from childhood to young adulthood: a randomized controlled trial Am. J. Clinical Nutrition, January 1, 2005; 81(1): 175 - 188. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Horlick, J. Wang, R. N. Pierson Jr., and J. C. Thornton Prediction Models for Evaluation of Total-Body Bone Mass With Dual-Energy X-Ray Absorptiometry Among Children and Adolescents Pediatrics, September 1, 2004; 114(3): e337 - e345. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. Gronowitz, D. Mellstrom, and B. Strandvik Normal Annual Increase of Bone Mineral Density During Two Years in Patients With Cystic Fibrosis Pediatrics, August 1, 2004; 114(2): 435 - 442. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. L. Gourlay and S. A. Brown Clinical Considerations in Premenopausal Osteoporosis Arch Intern Med, March 22, 2004; 164(6): 603 - 614. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
|
