| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
ORIGINAL RESEARCH COMMUNICATION |
1 From the Department of Pediatrics, The Children’s Hospital of Philadelphia (MBL, BAW, AMT, and BSZ), and the Department of Epidemiology and Biostatistics (MBL and JS), University of Pennsylvania School of Medicine, Philadelphia
2 Supported by NIH grants K08-DK02523 and 1-R03-DK058200 (both to MBL), the General Clinical Research Center (M01RR00240), and the Nutrition Center, The Children’s Hospital of Philadelphia, University of Pennsylvania School of Medicine. 3 Address reprint requests to MB Leonard, The Children’s Hospital of Philadelphia, 34th Street and Civic Center Boulevard, Philadelphia, PA 19104. E-mail: leonard{at}email.chop.edu.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONAPPENDIX AREFERENCES |
|---|
Objective: The objective of this study was to determine the effect of childhood obesity on skeletal mass and dimensions relative to height, body composition, and maturation in males and females.
Design: In 132 nonobese (body mass index < 85th percentile) and 103 obese (body mass index 
95th percentile) subjects aged 4–20 y, whole-body and vertebral bone mineral content (BMC) was determined by using dual-energy X-ray absorptiometry, and bone area, areal bone mineral density (BMD), and fat and lean masses were measured. Vertebral volumetric BMD was estimated as BMC/area1.5.
Results: Obesity was associated with greater height-for-age, advanced maturation for age, and greater lean mass for height (all P < 0.001). Sex-specific multivariate regressions with adjustment for maturation showed that obesity was associated with greater vertebral areal BMD for height, greater volumetric BMD, and greater vertebral BMC for bone area (all P < 0.05). After adjustment for maturation and lean mass, obesity was associated with significantly greater whole-body bone area and BMC for age and for height (all P < 0.001).
Conclusions: In contrast with the results of prior studies, obesity during childhood and adolescence was associated with increased vertebral bone density and increased whole-body bone dimensions and mass. These differences persisted after adjustment for obesity-related increases in height, maturation, and lean mass. Future studies are needed to determine the effect of these differences on fracture risk.
Key Words: Adolescent • child • obesity • body composition • bone mineral density • bone mineral content • dual-energy X-ray absorptiometry
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONAPPENDIX AREFERENCES |
|---|
Despite the epidemic of childhood obesity, the effect of obesity on bone mineral accrual during growth is poorly understood. Studies using dual-energy X-ray absorptiometry (DXA) to examine the effect of obesity on bone mass in children have yielded conflicting results, largely related to differing approaches to the assessment of two-dimensional projected DXA bone measures relative to age, bone size, and body size. Some studies reported normal or increased bone mineral content (BMC) in obese children (5–8), whereas others concluded that obese children have decreased bone mass relative to bone size and body weight (9, 10). The need for accurate assessment of the structural effects of obesity on bone mineral accretion is underscored by conflicting reports of the effects of increased fat mass and body mass index (BMI) on fracture risk in childhood (11–14). The objective of the present study was to determine the effect of childhood obesity on skeletal mass and dimensions relative to height, body composition, and maturation in a large multiethnic cohort of children and adolescents.
|
SUBJECTS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONAPPENDIX AREFERENCES |
|---|
Second, additional obese subjects were identified through a systematic review of all clinical DXA scans performed during a 3-y interval in subjects referred from the CHOP Weight Management Program. All children and adolescents seen in the Weight Management Program were routinely referred for lumbar spine and whole-body DXA scans at the time of the initial evaluation. Approval to analyze the DXA scans and review the medical records was obtained from the CHOP Institutional Review Board. Subjects were excluded if they had chronic medical conditions (other than obesity) or were taking medications that could affect growth and development. Seventy-five eligible subjects were identified.
Anthropometry and Tanner staging
All height and weight measurements were performed in the Nutrition and Growth Laboratory at the time of the DXA scans. Weight (in kg) was measured by using a digital electronic scale on which the subjects stood, and height (in cm) was measured by using a wall-mounted stadiometer. Pubertal status was determined by physical examination and classified according to the method of Tanner (15). Tanner staging of the 181 healthy control subjects was performed by trained research staff, and Tanner staging of the additional obese subjects was performed by the Medical Director of the Weight Management Program (AMT).
Age- and sex-specific z scores (SD scores) for height, weight, and BMI were calculated by using year 2000 growth data from the National Center for Health Statistics, Centers for Disease Control and Prevention (16). The BMI percentiles were used to classify subjects as follows: healthy weight, 5th–85th BMI percentiles; overweight, 85th–94th BMI percentiles; or obese, 95th BMI percentile (17). For the purpose of these analyses, the study was limited to the comparison between the 132 subjects having a healthy weight and the 103 obese children.
Lumbar spine and whole-body DXA scans
Bone mass in the anterior-posterior lumbar spine (L1–4) and in the whole body was measured by using DXA (QDR 2000; Hologic, Waltham, MA) with a fan beam in the array mode. All subjects were measured on the same machine. The measurements were performed by using standard positioning techniques. Quality-control scans were performed daily by using a simulated L1–4 lumbar spine made of hydroxyapatite encased in epoxy resin. In our institution, the in vitro CV was <0.6%, and the in vivo CV in adults was <1%.
The DXA scans were analyzed to generate measures of vertebral projected bone area (in cm2), BMC (in g), and areal bone mineral density (BMD, in g/cm2). Because the standard software for the analysis of the lumbar spine frequently fails to detect a complete bone map in children, all lumbar spine scans were analyzed with the low-density software as previously described (18). Lumbar spine bone mineral apparent density (BMAD, in g/cm3), which was calculated as lumbar spine BMC/lumbar spine bone area1.5, was used as an estimate of volumetric BMD (19).
Whole-body scans were analyzed to generate whole-body projected bone area and BMC. Because pediatric whole-body data may be confounded by variability in relative skull size (20), all whole-body DXA results presented in this study represent the bone area and BMC excluding the skull. Estimates of lean mass (in kg), fat mass (in kg), and percentage body fat were also obtained from the whole-body DXA scan excluding the skull. The lean mass measure excluded bone mass. A single operator (BAW) reanalyzed all DXA scans to standardize the analyses.
Statistics
Analyses were conducted by using STATA 7.0 (Stata Corporation, College Station, TX). Two-sided tests of hypotheses were used, and a P value < 0.05 was considered statistically significant. Initial analyses were descriptive and included the calculation of means with SDs for subject characteristics for the nonobese and obese subjects. Mean differences in anthropometric and body-composition variables between the nonobese and obese subjects were assessed by using the t test. The sex, race, and Tanner stage distributions were compared by using the chi-square test. DXA bone measures were compared by using the Wilcoxon rank-sum test.
The relations between bone measures, age, height, and body composition were explored graphically. Natural log transformations of the outcomes or explanatory variables were used to improve the fit of models. Body weight, height, fat mass, lean mass, and all bone measures (BMC, bone area, areal BMD, and BMAD) were log transformed. The assumptions of the regression models (linearity, normality of residuals, constant variance) were assessed via graphical checks and the following tests: the Shapiro-Wilk test of normality of the residuals, the Ramsey omitted variable test, and the Cook-Weisburg test for heteroskedasticity.
The log-transformed bone and body-composition measures in the obese subjects were compared with those in the nonobese subjects. Bone measures were first assessed relative to age. Because of the differences in height z scores between the obese and the nonobese subjects, the models were subsequently adjusted for height. The log-linear relation between spine BMC and projected bone area is well established, according to the power law relation: BMC 
(Area)p (19, 21, 22). Therefore, spine BMC was assessed by using a log-transformed model adjusted for bone area. For the whole body, bone area and BMC were assessed by using log-transformed models adjusted for height. A prediction model in which this approach was used for whole-body BMC relative to height has been described in children (23). A recent pediatric study showed that a log-transformed model of whole-body BMC adjusted for height was highly correlated with cortical bone strength (r = 0.63) as measured with the use of peripheral quantitative computed tomography (pQCT), whereas BMC adjusted for bone area was not correlated with strength (24). Whole-body bone area relative to height was assessed as a measure of bone dimensions or breadth. The recent pediatric pQCT study showed that increased bone area relative to height was correlated with increased cortical cross-sectional area in the mid-shaft of the tibia diaphysis (24). The fit of all final models was assessed via the adjusted R2 value.
Models were adjusted for Tanner stage (with stage 1 as the referent group) and race (African American compared with all others). Sex differences in bone mineral accretion during growth and maturation are well documented. Therefore, each model was tested for sex interactions. The simplest models assessed DXA bone results relative to age and showed significant interactions between sex and age for lumbar spine BMD and whole-body BMD. In the multivariate models, significant interactions between height and sex or between Tanner stage and sex were detected for lumbar spine BMC, BMD, and BMAD and whole-body BMD. Given the complexity of the sex interactions in these models and the known sex differences in bone mineral accretion during growth, all bone results are presented stratified by sex. It was beyond the scope of this study to address each observed sex interaction independently of obesity effects. However, all models were tested for obesity-by-sex interactions, and the results are presented here.
Because the outcome measures were log transformed, the effect of obesity in each multivariate model is presented as the adjusted ratio of the outcome measure in the obese subjects to the outcome measure in the nonobese control subjects along with the 95% CI. The adjusted ratio and 95% CI were calculated as the exponentiated estimate and the 95% confidence limits, respectively, of the regression parameter for obesity. This approach provided an estimate of the difference in bone measures between the obese and the nonobese subjects across the entire range of values.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONAPPENDIX AREFERENCES |
|---|
. The age, sex, and racial distributions were not significantly different between the obese and nonobese subjects. The Tanner stage distributions and the proportion of children that were prepubertal also did not differ significantly between the 2 groups; however, the obese children were significantly younger at each Tanner stage (Tanner stages 2–5) than were the nonobese children. Height z scores and lean and fat masses were significantly greater in the obese subjects than in the nonobese control subjects.
|
Comparison of bone measures between obese and nonobese subjects
The DXA results are summarized in Table 2. All lumbar spine and whole-body DXA results were significantly greater in the obese subjects than in the nonobese subjects.
|
and its legend, and the adjusted R2 values for each of the models presented in Figure 1 are summarized in Table 3.
|
|
Two approaches were used to examine lumbar spine volumetric BMD: BMAD (19) and BMC relative to bone area (25). The effect of obesity on BMAD in the females is shown in Figure 2. Comparable results were seen in the males. In the males and the females, BMAD was significantly greater in the obese children than in the control subjects after adjustment for Tanner stage and race. In the males and the females, lumbar spine BMC relative to bone area was significantly greater in the obese subjects than in the normal-weight control subjects after adjustment for Tanner stage and race. Both of these approaches used bone area to adjust for differences in skeletal size; therefore, including height as a covariate did not improve the fit of the model, and height was not independently associated with either estimate of volumetric BMD. Including age as a covariate did not improve the fit of any of the multivariate models described above.
|
and its legend, and the adjusted R2 values for the models are summarized in Table 3
.
|
and 5, respectively.
|
|
0.93), and the obese subjects had significantly greater lean mass relative to height than did the nonobese subjects, as described above. To determine whether the greater whole-body BMC and bone area for height observed in the obese children was due to greater lean mass for height (ie, greater bone loading by muscle forces), lean mass was included in the multivariate models for whole-body BMC for height and whole-body bone area for height. Whole-body lean mass, Tanner stage, and obesity each were positively and independently associated with greater whole-body bone area for height and whole-body BMC for height. Whole-body BMC for bone area was significantly lower in the obese subjects than in the normal-weight subjects after adjustment for Tanner stage and race in both the males and the females. Including age as a covariate did not improve the fit of any of the multivariate models described above.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONAPPENDIX AREFERENCES |
|---|
A multistaged approach was needed because of the limitations of DXA and the nature of the differences between obese and nonobese children. DXA is a two-dimensional approach that provides an estimate of density expressed as grams per projected area (areal BMD, in g/cm2). This estimate is not a measure of volumetric density (g/cm3) because it provides no information about bone depth. Bones in taller persons are longer, wider, and deeper; however, DXA bone area only captures the bone length and width and not the greater depth. Therefore, areal BMD inherently overestimates the volumetric BMD of tall persons (26). Consequently, assessment of areal BMD for age is biased by increased stature in obese children. BMAD (19) and regression of BMC against bone area (21) were used to estimate vertebral volumetric BMD in the present study. Both strategies showed significantly greater vertebral BMD in the obese subjects than in the nonobese subjects.
The differences in body size, body composition, and maturation between the obese and the nonobese subjects dictated the need for multivariate models that account for these potential confounders. Although our results showed greater vertebral BMC for bone area in the obese subjects, Goulding et al (9) recently reported significantly lower vertebral BMC for bone area in 18 obese children than in healthy-weight control subjects after adjustment for height, weight, and Tanner stage. These contradictory conclusions regarding the effect of obesity are explained by the fact that the authors included both weight and height in their regression models and failed to consider the joint distribution of weight, height, and obesity status in interpreting their findings.
Our interpretation takes the joint distribution of lean mass and height into account in the sense that obesity status is assessed relative to the values of these covariates. In our model, the coefficient for obesity status represents the difference in expected outcome between obese and nonobese subjects of the same height, Tanner stage, race, and lean mass. This interpretation is possible because 2 subjects of the same height, Tanner stage, race, and lean mass may differ in their obesity status. However, if our model also included weight, then this interpretation of the regression coefficient for obesity status would not be possible, because, by definition, both of 2 subjects with the same height and weight would be either obese or nonobese. In a model that includes both height and weight, the effect of obesity, therefore, cannot be assessed solely by the regression coefficient for obesity status; the regression coefficients for weight and height must be considered. A mathematical illustration of these effects in which the model reported by Goulding et al (9) is used is provided in Appendix A.
The interpretation of whole-body DXA measures requires similar considerations. Cortical bone constitutes 80% of whole-body bone mass. We recently compared varied analytic strategies for whole-body DXA results with pQCT measures of cortical cross-sectional area and strength in 150 healthy children (24). DXA whole-body bone area for height and BMC for height were both strongly and positively associated with pQCT cross-sectional area and strength, which suggests broader bones with less resistance to bending. DXA bone area for weight, BMC for weight, and BMC for bone area were poor predictors of bone cross-sectional area and strength.
As summarized in Figure 3, the obese subjects in the present study had markedly greater whole-body bone area and BMC relative to age and height than did the nonobese subjects. In contrast with our findings, other investigators have reported normal (8) or decreased (10) whole-body BMC and bone area relative to body size. Again, this relates to the choice of body size covariates. Manzoni et al (8) reported no differences in whole-body bone measures between obese children and nonobese control subjects after adjustment for height, weight, lean mass, and fat mass. As detailed above, it is not appropriate to fix height and weight in the comparison between obese and nonobese subjects. Goulding et al (9) reported lower whole-body bone area and BMC relative to body weight in obese children. We were able to reproduce this result within our data; however, we believe this is inappropriate because it results in the comparison of children of different stature. An obese child will be substantially shorter than a nonobese child of the same weight, and this difference in BMC and bone area will reflect differences in bone length rather than width.
Prior investigators have proposed assessing whole-body BMC relative to bone area as an approach to adjust for differences in skeletal size (27). Although this approach may be valid in the lumbar spine, application to the whole body is problematic, as shown by our comparison between DXA and pQCT (24). The skeleton is a composite of many tube-like (eg, limbs) and broad (eg, pelvis) structures that vary in depth and thickness. Therefore, bone area is a poor measure of the volume of bone over which the BMC is distributed. Second, DXA bone area is a function of subject height and bone width. In a child with broad bones for height, assessment of BMC for bone area will result in the comparison of that child with a significantly taller child of comparable bone area, and this difference in BMC and bone area will reflect differences in bone length rather than width. Therefore, the lower whole-body BMC for bone area observed in obese children may not represent a deficit in bone mass.
A final technical limitation of DXA relates to the potential for projection error in the assessment of bone measures in obese subjects. DXA manufacturers have largely converted to fan beams that introduce magnification errors in measures of bone area and BMC. In the case of Hologic scanners, as the bone is elevated a greater distance off the table (as occurs with obesity), bone geometry and BMC are underestimated (28). However, in the present study, whole-body bone area and BMC were increased for age and height in childhood obesity. These findings cannot be attributed to projection error because such an error would result in an underestimation of BMC and bone area. Rather, our observation of increased whole-body bone area and BMC likely underestimates the true magnitude of the increases seen in obesity. Note that the lower vertebral bone area for height in the obese females than in the nonobese females may have been due to magnification error. Sex differences in fat distribution (subcutaneous or visceral) (29) may have differing effects on regional elevation of the skeleton.
There are several possible mechanisms for increased bone mass in childhood obesity. Hormonal influences, such as increased conversion of androstenedione to estrogen or increased circulating leptin concentrations, may play a role (30). Leptin acts as a growth factor on the chondrocytes of skeletal growth centers via insulin-like growth factor I receptor expression and thereby potentially contributes to the increased linear growth and skeletal mass observed in childhood obesity (31). During puberty, estrogen promotes accrual of bone mass on the cortical endosteal surface and in trabecular bone (32).
Increased biomechanical loading due to increased body weight and increased lean muscle mechanical forces may also have contributed to the increased bone dimensions and mass observed in the obese subjects. Increased loading of long bones produces the greatest mechanical stresses on the subperiosteal surface and stimulates bone formation by subperiosteal expansion (33). For example, loading in the playing arm in racquet sports induces significant increases in bone dimensions and mass (34). The effect of weight gain on bone loading has been examined in adults. A longitudinal study in postmenopausal women showed that women who gained weight experienced a significant increase in hip cortical section modulus through periosteal expansion (35). A QCT study in healthy children suggested that weight bearing and mechanical stresses are important determinants of cortical bone mass, whereas trabecular bone density is influenced by hormonal factors associated with sexual development (36).
A study of bone biomechanics in adult male rats with dietarily induced obesity showed significantly greater bone strength in the obese rats than in the controls (37). The cross-sectional geometry and ultimate fracture load of the femur were higher in the obese rats than in the controls.
Ultimately, the clinical significance of altered bone mineral accrual lies in the occurrence of fractures. Studies on the effect of fat mass and BMI on fracture risk have yielded conflicting results (11–14). We propose that the increased BMC in childhood obesity results in increased bone strength but that this increase is not sufficient to overcome the significantly greater forces generated when an obese child falls on an outstretched arm. Although childhood obesity may result in an increased risk of childhood forearm fractures, the effect of obesity on life-long fracture risk is unknown.
Future studies using 3-dimensional imaging techniques such as QCT are needed to determine the effect of obesity on bone mass and dimensions, to determine whether the onset of obesity before completion of puberty results in greater bone mineral accretion than does obesity later in life, to determine whether the increases in spine and whole-body BMC are sustained into adulthood, and to further evaluate the mechanisms for increased bone mass relative to pubertal stage and muscle mass in obese children and adolescents.
|
APPENDIX A |
|---|
|
TOPABSTRACTINTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONAPPENDIX AREFERENCES |
|---|
 | (A1) |
 | (A2) |
40% more than do the nonobese subjects of a given height. In the present study, the obese subjects weighed, on average, 46% more than did the nonobese subjects of the same height, which is consistent with the positive effect of obesity on vertebral BMC for bone area. Note that the regressions for BMC showed that obesity was negatively associated with BMC for height and that weight was positively associated with BMC for height, so that the results of the analysis differ if only the regression coefficient for obesity is assessed. In contrast, in our analyses, both lean mass and obesity were positively associated with BMC for height, which simplified the interpretation of the regressions.
|
ACKNOWLEDGMENTS |
|---|
MBL and BSZ contributed to the study design, data collection, data analysis, and the writing of the manuscript. JS contributed to data analysis and the writing of the manuscript. BAW contributed to data collection and data analysis. AMT contributed to the study design, data collection, and the writing of the manuscript. None of the authors had any financial or personal interest in any company or organization sponsoring the research.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONAPPENDIX AREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  G. Ducher, S. L Bass, G. A Naughton, P. Eser, R. D Telford, and R. M Daly Overweight children have a greater proportion of fat mass relative to muscle mass in the upper limbs than in the lower limbs: implications for bone strength at the distal forearm Am. J. Clinical Nutrition, October 1, 2009; 90(4): 1104 - 1111. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. M Lenders, H. A Feldman, E. Von Scheven, A. Merewood, C. Sweeney, D. M Wilson, P. D. Lee, S. H Abrams, S. E Gitelman, M. S Wertz, et al. Relation of body fat indexes to vitamin D status and deficiency among obese adolescents Am. J. Clinical Nutrition, September 1, 2009; 90(3): 459 - 467. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. J. Howie, D. M. Sloboda, T. Kamal, and M. H. Vickers Maternal nutritional history predicts obesity in adult offspring independent of postnatal diet J. Physiol., February 15, 2009; 587(4): 905 - 915. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. Mardones, L. Villarroel, L. Karzulovic, S. Barja, P. Arnaiz, M. Taibo, and F. Mardones-Restat Association of perinatal factors and obesity in 6- to 8-year-old Chilean children Int. J. Epidemiol., August 1, 2008; 37(4): 902 - 910. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. S. Jones, S. C. Kaste, W. Liu, C. Cheng, W. Yang, K. G. Tantisira, C.-H. Pui, and M. V. Relling CRHR1 Polymorphisms Predict Bone Density in Survivors of Acute Lymphoblastic Leukemia J. Clin. Oncol., June 20, 2008; 26(18): 3031 - 3037. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. K Pollock, E. M Laing, C. A Baile, M. W Hamrick, D. B Hall, and R. D Lewis Is adiposity advantageous for bone strength? A peripheral quantitative computed tomography study in late adolescent females Am. J. Clinical Nutrition, November 1, 2007; 86(5): 1530 - 1538. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. B. Leonard Glucocorticoid-Induced Osteoporosis in Children: Impact of the Underlying Disease Pediatrics, March 1, 2007; 119(Supplement_2): S166 - S174. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. L. Burr, C. G. Taylor, and H. A. Weiler Dietary Conjugated Linoleic Acid Does Not Adversely Affect Bone Mass in Obese fa/fa or Lean Zucker Rats. Experimental Biology and Medicine, November 1, 2006; 231(10): 1602 - 1609. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Leger, D. Marinovic, C. Alberti, S. Dorgeret, D. Chevenne, C. L. Marchal, N. Tubiana-Rufi, G. Sebag, and P. Czernichow Lower Bone Mineral Content in Children with Type 1 Diabetes Mellitus Is Linked to Female Sex, Low Insulin-Like Growth Factor Type I Levels, and High Insulin Requirement J. Clin. Endocrinol. Metab., October 1, 2006; 91(10): 3947 - 3953. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. D. Taylor, K. R. Theim, M. C. Mirch, S. Ghorbani, M. Tanofsky-Kraff, D. C. Adler-Wailes, S. Brady, J. C. Reynolds, K. A. Calis, and J. A. Yanovski Orthopedic complications of overweight in children and adolescents. Pediatrics, June 1, 2006; 117(6): 2167 - 2174. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
U. Alexy, T. Remer, F. Manz, C. M Neu, and E. Schoenau Long-term protein intake and dietary potential renal acid load are associated with bone modeling and remodeling at the proximal radius in healthy children Am. J. Clinical Nutrition, November 1, 2005; 82(5): 1107 - 1114. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. M Burnham, J. Shults, E. Semeao, B. J Foster, B. S Zemel, V. A Stallings, and M. B Leonard Body-composition alterations consistent with cachexia in children and young adults with Crohn disease Am. J. Clinical Nutrition, August 1, 2005; 82(2): 413 - 420. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. J Foster, J. Shults, B. S Zemel, and M. B Leonard Interactions between growth and body composition in children treated with high-dose chronic glucocorticoids Am. J. Clinical Nutrition, November 1, 2004; 80(5): 1334 - 1341. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. B. Leonard, H. I. Feldman, J. Shults, B. S. Zemel, B. J. Foster, and V. A. Stallings Long-Term, High-Dose Glucocorticoids and Bone Mineral Content in Childhood Glucocorticoid-Sensitive Nephrotic Syndrome N. Engl. J. Med., August 26, 2004; 351(9): 868 - 875. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
) for the ratio of a given lumbar spine bone measure in the obese subjects to that in the nonobese control subjects. The horizontal line at the ratio of 1.0 represents no difference between the obese and the nonobese subjects. *,**,***Significant effect of obesity: *P < 0.05, **P < 0.01, ***P < 0.001. Adjusted R2 values for each model are presented in Table 3
