. H4 NH“ ‘liﬁuuﬁ. 1‘11“.
A g. 3% a3
i V

i... 0

$11.. . o
g”?

mm wk“;
5:.” .ﬂr—W
fi‘v ,

..

1:5
. 3W. . .
ﬁawumwunwmw

.epdqprvﬁhirﬁtn ”
. *3... . Rum:
an 4..." .
d

an»...
vﬂnWHnuur -
d...

 

. ~ 3.3.
d u .- .
awn. .m. .

V. .
a. MW»...

 

}!\.. I...

unfivuleht

i ‘ ‘ 3y! I.
.51.! l...

 

 

LIBRARY *
2 Michigan State
SIDE) University

 

 

This is to certify that the
dissertation entitled

EXAMINATION OF DIET, PHYSICAL ACTIVITY,
BIOMARKERS OF BONE MINERALIZATION, BONE
MINERAL CONTENT AND BODY COMPOSITION IN

CHILDREN BETWEEN 5 YEARS OF AGE AND PUBERTY

presented by

MARCIA KELLY SCOTT

has been accepted towards fulfillment
of the requirements for the

Ph.D. degree in Food Science and Human
Nutrition

 

 

ﬂ.)

.4 ' -Jr'
\J' (VIP/L \'I- [267} a
ajor Professor’s Signature

/I a f/

411
/

4 .1
v j

Date

MSU is an afﬁnnative-action, equal-opportunity employer

_....-.-_-u-o-

 

 

PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

5/08 K‘IProj/Accapres/CIRClDateDue.rndd

EXAMINATION OF DIET, PHYSICAL ACTIVITY, BIOMARKERS OF BONE
MINERALIZATION, BONE MINERAL CONTENT AND BODY COMPOSITION IN
CHILDREN BETWEEN 5 YEARS OF AGE AND PUBERTY
By

Marcia Kelly Scott

A DISSERTATION

Submitted to
Michigan State University
in partial fulﬁllment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY
Department of Food Science and Human Nutrition

2008

ABSTRACT
EXAMINATION OF DIET, PHYSICAL ACTIVITY, BIOMARKERS OF BONE
MINERALIZATION, BONE MINERAL CONTENT AND BODY COMPOSITION IN
CHILDREN BETWEEN 5 YEARS OF AGE AND PUBERTY
By

Marcia Kelly Scott

Osteoporosis is predicted to become a disease of epidemic proportions worldwide.
Prevention of osteoporosis begins in childhood, recognizing that attainment of higher
bone mineral content during the ﬁrst two decades of life decreases the risk of developing
fractures later in life. Body composition, bone mineral density (BMD), bone mineral
content (BMC), diet composition, bone metabolism biomarkers, and physical activity
levels were measured in healthy, prepubertal children (n=52), mean age of 7.6 years.
BMD, BMC and body composition were determined by dual energy x-ray absorptiometry
(DXA). BMD Z-scores ranged from —-1.2 to +1.8 with 31% of subjects having Z-scores
below negative 0.2. BMD and BMC correlated positively with percent body fat within a
healthy range (r = 0.535 for BMD; r = 0.646 for BMC) and with total daily energy
expenditure (DTEE) above basal energy expenditure (BEE) (r = 0.459 for BMD; r =
0.591 for BMC). BMD and BMC correlated negatively with protein intake (r = -0.508
for BMD;‘r = -0.564 for BMC), energy intake (r = -0.510 for BMD; r = -0.578 for BMC),
calcium intake (r = -0.277 for BMD), phosphorus intake (r = -0.378 for BMD; r = -0.348
for BMC) and with serum osteocalcin (OC) (r = -0.507 for BMD; r = -0.499 for BMC).

Four prediction models for BMC and BMD were developed. Total BMC can be
predicted by percent body fat, fruit and vegetable intake, calcium, phosphorus, energy

and magnesium intakes, along with DTEE above BEE. This model explains 72.9% of the

variability in BMC. Alternately, BMC can be predicted by serum OC, urinary
deoxypyridinium (DPD), DTEE above BEE, and percent body fat, with this model
explaining 64.9% of the variability in BMC. BMD can be predicted by percent body fat,
fruit and vegetable intake, calcium, phosphorus, energy, and magnesium intakes. This
model explains 58.8% of the variability in BMD. The second prediction model for BMD
includes percent body fat, serum OC, urinary DPD, serum 25(OH) Vitamin D3, and
DTEE above BEE, explaining 58.9% of the variability in BMD.

This study is unique in looking at many modiﬁable environmental influences of
bone mineralization in a group of exclusively pre-pubertal subjects. With one third of
these children having negative BMD Z-scores before they reach puberty, these data
suggest concerns about inadequate progress toward attaimnent of peak bone mass. The
negative correlation of protein to bone mass warrants further examination given the
protein rich diets consumed by children in the US, with these subjects consuming over 3
times the recommended amount. Physical activity and fruit and vegetable intake appear
to be strongly, positively associated to bone mass, supporting public health efforts to
increase both physical activity and intake of fruits and vegetables. Biomarkers of bone
metabolism, serum OC (marker of bone formation and turnover), and urinary DPD
(marker of bone resorption) may merit further consideration for their potential use in
monitoring bone growth in young children. Diet and lifestyle habits are forming for a
lifetime in these early childhood years so knowing where to focus suggestions and
interventions toward bone-building lifestyles at this age may decrease the risk of

developing osteoporosis later in life as well as decreasing fracture risk throughout life.

Copyright by
MARCIA KELLY SCOTT
2008

DEDICATION

This work is dedicated to Taylor, Robinson, Alexandra, and Christian Scott
My family and my best friends

And to Jack and Janet Kelly and R. Taylor Scott, IV
Parents and ﬁrst teachers

And to Jenny Taylor Bond
Teacher, mentor, friend, sister

ACKNOWLEDGEMENTS

I am most grateful to my committee members Dr. Wanda Chenoweth, Dr.
Gretchen Hill, Dr. Mike Orth, and Dr. Jenny Bond for their invaluable contributions to
this project and to my graduate education. They are educators extraordinaire and they
make the world a better place through their contributions to science, to education and to
students. I thank each of them from the bottom of my heart.

I am grateful for the contributions of the staff and faculty of the Department of Food
Science and Human Nutrition including their support through the College of Human Ecology

Marie Dye Doctoral, the John Harvey Kellogg Nutrition Fellowship, the College of Human
Ecology Beth and Holly Fryer Scholarship, the College of Human Ecology H.A.M.M.
Scholarship, the College of Human Ecology Jeanette Lee Scholarship, American Dietetic
Association Foundation Scholarships, and graduate assistantships.

I would also like to acknowledge the very special contributions of Dr. Jerry Cash
to this project. Jerry Cash, along with his wife Stella, is committed to the health and well
being of children and to the education students and I thank them for their support.

Special acknowledgement of support goes John Bond, Maria Nnyepi and Zalilah
Mohd-Shan'ff as well as to my community of friends and colleagues for their many

gestures of support and all of the smiles along the journey.

vi

TABLE OF CONTENTS

LIST OF TABLES ........................................................................ iv
LIST OF FIGURES ....................................................................... vi
KEY TO ABBREVIATIONS ........................................................... vii
INTRODUCTION ........................................................................ 1
CHAPTER 1 ................................................................................ 2
REVIEW OF LITERATURE
Background Information ................................................. 3
Bone and Bone Growth .................................................. 5
Inﬂuence of Gender and Race on Bone ............................... 6
Role of Biomarkers in Assessment of Bone Growth ............... 8
Relationship of Body Composition to Bone .......................... 12
Bone Growth in Special POpulations .................................. 15
Inﬂuence of Physical Activity on Bone ............................... 16
Relationship of Diet to Bone ........................................... 19
Background Information Summary .................................... 26
Study Objective and Hypothesis ....................................... 27
Additional Study Hypotheses and Questions ........................ 27
CHAPTER 2 ................................................................................ 28

METHODS — HEALTHY KIDS NUTRITION STUDY (HKNS)

Study Summary ........................................................... 29
Project Design ............................................................. 29
Subjects and Subject Recruitment
Study Stafﬁng
Measurement and Analysis of Variables ............................. 33
Diet Analysis
Anthropometrics
Measurement of Physical Activity
Biomarker Assays
Vitamin D Measurement
Dual X-ray Absorptiometry
Sample Size and Power Calculation

vii

CHAPTER 3 ........

Data Analysis

........................................................................ 45

BONE MINERALIZATION IN PREPUBERTAL CHILDREN: ASSOCIATION
OF DIET, BODY COMPOSITION, AND PHYSICAL ACTIVITY

Abstract ...................................................................... 46
Introduction ................................................................. 48
Subjects and Methods ..................................................... 50
Results ....................................................................... 54
Discussion ................................................................... 68
CHAPTER 4 ................................................................................ 77

ASSOCIATION OF BODY COMPOSITION, DIET, PHYSICAL ACTIVITY,
AND BONE BIOMARKERS WITH MEASURES OF BONE
MINERALIZATION IN PREPUBERTAL CHILDREN

Abstract ....................................................................... 78
Introduction .................................................................. 80
Subjects and Methods ...................................................... 82
Results ........................................................................ 85
Discussion .................................................................... 91
CHAPTER 5 ................................................................................. 104

STRENGTHS, LIMITATIONS, AND IMPLICATIONS

APPENDICES ......

......................................................................... 109

APPENDICES 1 through 8: Study Forms ........................................ 110

Appendix 1
Appendix 2
Appendix 3
Appendix 4
Appendix 5
Appendix 6
Appendix 7
Appendix 8

UCRIHS Approvals

Consent to participate in a research study

HKNS Recruitment ﬂyer

Authorization for Disclosure of Health Information
HKNS Health History Questionnaire

HKN S Debrieﬁng Protocol

Physician order for DEXA

Study protocol checklist

APPENDICES 9 through 20: Supplemental Data Tables ...................... 123
Appendix 9 Nutrient intake as measured by usual, one-day dietary recall by age
Appendix 10 Nutrient intake as measured by food frequency questionnaire

with and without supplements by age

Appendix 11 Dietary intakes according to 2005 US. Dietary Guidelines by age

viii

Appendix 12
Appendix 13

Appendix 14
Appendix 15
Appendix 16
Appendix 17
Appendix 18

Appendix 19

Appendix 20

REFERENCES ......

Site-Speciﬁc measures of bone mineralization by gender

Energy expenditure above BBB and activity counts at 3 levels of
intensity by age

Energy expenditure above BBB and activity counts at 3 levels of
intensity by gender

Correlations of BMD and BMC with variables inﬂuencing bone
mineralization by gender

Correlations of BMD and BMC with variables inﬂuencing bone
mineralization by age

Correlations of BMD and BMC by height with variables
inﬂuencing bone mineralization by gender

Correlations of BMD and BMC by height with variables
inﬂuencing bone mineralization by age

Multiple regression model predicting BMC by height as a function
of serum OC by height, DEE at light intensity above BBB, and
DEE at moderate intensity above BEE

Crosstabs of dependent variable categories with independent
variable categories

........................................................................ 147

ix

LIST OF TABLES

Table 2.1 .................................................................................... 44

Comparison of BMD and BMC averages by Age Group and Gender and Average of
within group Standard Deviation. Lunar and HKNS Data

Table 3.1 .................................................................................... 55

Physical characteristics, measures of bone mineralization, energy expenditure, and serum
25(OI-I) Vitamin D3 by gender and in total

Table 3.2 .................................................................................... 56
Daily dietary intakes according to 2005 US. Dietary Guidelines by gender and in total
Table 3.3 ..................................................................................... 58
Daily nutrient intake as measured by usual, one-day dietary recall by gender and in total
Table 3.4 ...................................................................................... 59

Daily nutrient intake as measured by food frequency questionnaire with and without
supplements by gender and in total

Table 3.5 .................................................................................... 64
Correlations of BMD and BMC with variables inﬂuencing bone mineralization
Table 3.6 .................................................................................... 66

Multiple regression analysis of BMD by % body fat and intakes of kcals by weight, Ca
by height, P by height, Mg, and fruits and vegetables

Table 3.7 ...................................................................................... 67

Multiple regression analysis of BMC by % body fat, DTEE above BEE, intakes of kcals
by weight, Ca by height, P by height, Mg, and fruits and vegetables

Table 4.1 .................................................................................... 87

Physical characteristics, measures of bone mineralization, and biomarkers

List of Tables continued

Table 4.2 .................................................................................... 88
Intake of key bone-building nutrients per day measured by usual, one-day dietary recall
Table 4.3 .................................................................................... 90
Correlations of BMD and BMC with variables inﬂuencing bone mineralization

Table 4.4 ..................................................................................... 92

Multiple regression model predicting BMD as a function of % body fat, serum DC by
height, urinary DPD by height, serum 25(OH) Vitamin D3, and DTEE above BEE

Table 4.5 .................................................................................... 93

Multiple regression model predicting BMC as a function of % body fat, serum OC by
height, urinary DPD by height, and DTEE above BEE

xi

LIST OF FIGURES

Figure 2.1 .................................................................................... 33
Healthy Kids Nutrition Study Protocol
Figure 3.1 .................................................................................... 62

Mean Nutrient Adequacy Ratios for Select Nutrients as Measured by Recall and by F FQ
without supplements

Figure 3.2 .................................................................................... 69

Correlation between actual BMD and predicted value of BMD based on % body fat,
intakes of kcals by weight, Ca by height, P by height, Mg, and fruits and vegetables

Figure 3.3 ..................................................................................... 70
Correlation between actual BMC and predicted value of BMC based on % body fat,
DTEE above BEE, intakes of kcals by weight, Ca by height, P by height, Mg, and fruits
and vegetables

Figure 4.1 ..................................................................................... 95
Correlation between actual BMD and predicted value of BMD based on % body fat,
serum OC by height, urinary DPD by height, serum 25(OH) Vitamin D3 and DTEE above
BEE

Figure 4.2 .................................................................................... 96

Correlation between actual BMC and predicted value of BMC based on % body fat,
serum DC by height, urinary DPD by height, and DTEE above BEE

xii

KEY TO COMMONLY USED ABBREVIATIONS
BEE basal energy expenditure

BMC bone mineral content

BMD bone mineral density

BMI body mass index

Ca calcium

cm centimeters

(1 day

DEE daily total energy expenditure
DPD deoxypyridinoline

DRI Dietary Reference Intake
DXA dual energy x-ray absorptiometry
HKNS Healthy Kids Nutrition Study
K potassium

kcal kilocalories

kg kilograms

LBM lean body mass

Mg magnesium

mg milligrams

mL milliliters

mmol millimoles

ng nanograms

OC osteocalcin

P phosphorus

xiii

INTRODUCTION

The World Health Organization predicts that osteoporosis will eventually become
a disease of epidemic proportions worldwide. In the US, according to the USDHHS
Healthy People 2010 report, one in ten people in the US. have osteoporosis and at least
one third of them will experience an osteoporotic-related fracture after 50 years of age
(U SDHHS, 2000). Twenty four percent of people over 50 years of age who suffer hip
fractures die within one year of the fracture and a majority of the remaining peOple never
returns to their pre-fracture level of function. Often considered a geriatric disease,
osteoporosis may also be considered a pediatric disease with geriatric consequences.
Prevention of osteoporosis should likely begin during childhood, recognizing that
attainment of higher bone mineral content during the ﬁrst two decades of life decreases
the risk of developing fractures later in life (Faulkner and Bailey, 2007).

This study, the Healthy Kids Nutrition Study, examined the diet, physical activity,
and body composition of a group of children in the US. Midwest and looked at the
relationship of these variables to the children’s bone mass with the goal of identifying
lifestyle factors that impact bone health. Diet and lifestyle habits are forming for a
lifetime in these early childhood years so knowing where to focus suggestions and
interventions toward bone-building lifestyles at this age may decrease the risk of

developing osteoporosis later in life as well as decrease fracture risk throughout life.

CHAPTER ONE

CHAPTER ONE
REVIEW OF THE LITERATURE
Background Information

Osteoporosis is increasingly recognized as one of the major public health
problems facing aging individuals of both genders, worldwide. It is predicted to become
a disease of epidemic proportions within the next several decades (Riggs and Melton,
1995; WHO, 2003). AS a disease, osteoporosis is broadly deﬁned as diminished bone
mass and reduced bone mineral density (BMD) at the level of 2.5 standard deviations
below the referent BMD of young adults (Riggs and Melton, 1995; USDHHS, 2000;
WHO, 2003). Osteoporosis presents clinically as fractures, both traumatic and
nontraumatic in nature, resulting in signiﬁcant morbidity, mortality, and health care costs
(WHO, 2003). The two most important ways to reduce risk of osteoporosis are to attain
peak bone mass in early life and to have a low rate of bone loss in later life.

Peak bone mass has genetic, hormonal, nutritional, and behavioral determinants
(Soyka et al., 2000). What processes and factors inﬂuence bone growth in the important
ﬁrst two decades of life? Research over the past 20 years has begun to clarify the genetic
and environmental contributors to bone growth in children. Advances in the ﬁelds of
bone imaging and biomarker assays have contributed to an understanding of the
importance of attainment of maximal bone mass during the growth years. The role of
physical activity in a bone building lifestyle is now well recognized (WHO, 2003).
Recognition of the critical importance of appropriate intake of the bone building nutrients

has inspired renewed efforts to determine optimal intake levels as well as optimal total

diet composition in order to attain maximumbone mineral density and decrease the risk
of bone disease.

In considering ways to decrease both the incidence and the severity of
osteoporosis, approaches to interventions for this disease target methods of prevention
and abatement of osteoporosis not just in adulthood but also during infancy, childhood,
and adolescence. There is a consensus in the literature that attainment of higher bone
mineral content during these ﬁrst two decades of life decreases the risk of developing
fractures later in life (Heaney et al., 2000). Though osteoporosis is generally considered
a geriatric disease, osteoporosis may actually be a pediatric disease with geriatric
consequences. Investigation of the pediatric origins of osteoporosis is centered on the
underlying processes and factors playing a role in the evolution of reduced bone mass and
reduced bone mineral density.

Bone mass at all ages exists on a continuum, determined by genetic factors and
modiﬁed in either direction by environmental factors (Matkovic et al., 1998; Heaney et
al., 2000). From birth through young adulthood, bone mass increases at a steady rate
(Glastre et al., 1990; Southard et al., 1991; Heaney et al., 2000). Bone mass also tends to
track throughout life, meaning that infants, children and adolescents that have higher
bone mass tend to be the adults who have high bone mass (Ferretti et al., 1998; Heaney et
al., 2000). This association introduces the possibility that “at risk” individuals may be
identiﬁed early in life such that risk can be minimized to the extent that is possible.
Approximately 26% of adult calcium is laid down during the 2 years of peak skeletal
growth that occurs in adolescence (Bailey et al., 2000). Statistics on lifetime fracture

incidence show that between the ages of approximately 5 and 15 years for boys and girls

and then after the age of 50 for women, the incidence of limb fractures is highest (Heaney
et al., 2000). Low bone mineral content (BMC) in children is considered to put children
at increased risk of fracture (Whiting, 2002). Attainment of peak bone mass may be an
important pediatric health goal in order to minimize bone fractures through all stages of
life.

In addition to osteoporosis, other public health concerns in children are associated
with bone health. Calcium intake, physical activity, and body composition appear to be
linked to bone growth (Soyka et al., 2000; Heaney et al., 2000; Greer and Krebs, 2006).
Current trends in diet intake in children show increased consumption of soft drinks
replacing other traditional “kid drinks” such as juice, milk, and water, trends that are
contributing to an increasing incidence of overweight children as well as increasing
concerns about the intake of nutrients important to bone health (Greer and Krebs, 2006).
The prevalence of pediatric overweight is increasing at an alarming rate in the US,
having doubled over the past two decades (AAP/CON, 2003). The National Center for
Health Statistics (N CHS) deﬁnes overweight in the pediatric population as body mass
index (BMI) at or above the 95th percentile for age and gender with “at risk for
overweight” being deﬁned by a BMI between the 85th and 95th percentile (CDC/NCHS,
2003). Currently 15.3% of 6- to 11- year old children are at or above the 95th percentile
for BMI on the standard growth charts as developed by the Centers for Disease Control
and Prevention, National Center for Health Statistics (AAP/CON, 2003). This increase in
percent overweight is accompanied by a decrease in physical activity along with an
increase in the pediatric incidence of many of the comorbid conditions seen with adult

obesity such as cardiovascular, endocrinologic, orthopedic, pulmonary and psychological

problems (AAP/CON, 2003). Because of the overlapping etiologies of chronic health
conditions that have roots in childhood, decreasing the risk of these conditions in the
pediatric population becomes even more important to pubic health in the long term.
Bone and Bone Growth

The human skeleton contains 206 individual bones. They are composed of two
types, cortical and trabecular bone. Cortical bone is the dense, compact tissue on the
outer surface of bones, includes the long bones, and makes up 75-80 % of bone mass.
Trabecular bone is more mesh-like than cortical bone, is present in ﬂat bones, ends of
long bones, and vertebrae and makes up 20-25 % of bone mass. All bone tissue is in a
constant state of ﬂux, involved in one or more of three processes: growth, modeling and
remodeling. Growth is under the control of the endocrine system and encompasses
overall skeletal growth. Modeling involves the laying down of bone onto bone surfaces
and results in a net gain of bone. Remodeling is the dominant process in adults and is
basically a recycling process. Older bone tissue is resorbed followed by new bone
formation at the same site, preserving bone’s mechanical integrity (Bailey et al., 1996).
Bone is largely mineral (70-90%); however the remaining matter is organic material,
most of which is collagen. Collagen plays a critical role in the structure and function of
bone tissue hence many of the biomarkers of bone metabolism are related to the structure
and formation of collagen, also referred to as bone matrix proteins (Young, 2003).

The newest frontiers in bone growth and health are in the areas of determining
patterns of bone growth and in identiﬁcation of osteoporosis-susceptibility genes. The
tempo of growth, direction of growth, region of growth and rate of growth in differing

sites even on a single bone appear to vary with age and exposure to all of the many

modiﬁers of bone growth. Bass et al., 1999, determined that by seven years of age, bone
had reached 80% of its maturational peak but only 40% of its peak mineral content. They
suggest that because growth provides ﬁrst a bigger skeleton and then second a denser
Skeleton, negative inﬂuences on bone growth in childhood are of particular importance in
the eventual prevention of bone fragility with age. Many potentially important candidate
genes for osteoporosis have been identiﬁed (Cusack and Cashman, 2003). Many of these
genes involve the regulatory proteins that inﬂuence bone growth while others of these so-
called osteoporosis-susceptibility genes interact with nutritional factors that inﬂuence
bone health. AS progress is made in identifying these genetic pathways, genotype-
speciﬁc bone health recommendations may evolve (Cusack and Cashman, 2003).
Inﬂuence of Gender and Race on Bone

As adults, males tend to have greater bone mass than do women (Nieves et al.,
2005) and blacks tend to have higher bone mass than whites or Asians (Pothiwala et al.,
2006), but that is not necessarily the case for pre-pubertal children. Assessment of bone
growth in children has progressed from the earlier work using bone age to single photon
absorptiometry (SPA) to the currently-used dual x-ray absorptiometry (DEXA) and
quantitative computerized tomography (Gluer etal., 1998; Shore and Poznanski, 1996).
Initially, normative data for BMD in children was determined, measuring BMC at several
sites with SPA. Geusens et a1. (1991) did not see gender differences in BMC or BMD in
prepubertal children. Patel et a1. (1992) investigated differences in BMC of black versus
white boys and girls, 6 to 20 years of age. They found that after adjusting for height,
there were no race or sex differences in BMC. Gilsanz et a1. (1991) saw no difference in

vertebral bone density in black versus white prepubertal girls, although their later

response to puberty did differ. Russell et al. (2001) saw increased bone age, considered a
correlate of BMC, in African American children ﬁve to twelve years of age over the bone
age of Caucasian children of the same chronological age but they attributed the difference
to the greater adiposity of the African American children.

In contrast, Specker et al. (1987) reported detectable gender differences in BMC
after 4 years of age in a cohort of predominantly Caucasian children. Li et al. (1989)
observed higher BMC in black children and in males. Bell et al. (1991) also observed
gender and racial differences in the BMD of prepubertal children. Nowack et al. (1995)
compared Hawaiian, Filipino, Asian, and Caucasian children and found that Hawaiian
children had a higher BMC than Asian or Caucasian children. F erretti et al., (1998)
report that BMC follows lean body mass (LBIVD, fat mass, and height in prepubertal,
Argentine boys and girls. Several studies using dual energy x-ray absorptiometry (DXA)
report that, in general, BMD increases with age, height, weight and Tanner stage (Glastre
et al., 1990; Southard et al., 1991; Del-Rio et al., 1994; Zanchetta et al., 1995).

Maynard et al. (1998) used Fels Longitudinal Study data to summarize BMC and
BMD in 8 tol 8 year old white children. They found no gender differences in the eight
and nine year old children, a ﬁnding conﬁrmed by Fassler and Bonjour (1995) and by
Nguyen et al. (2001). In an attempt to better understand both gender and ethnic
inﬂuences on bone in prepubertal children, Horlick et al. (2000) measured BMD and
BMC in white, black, and Asian children. They found BMC differed by gender but not
ethnicity and BMD differed for blacks versus non-blacks but not gender. They concluded
that both measures vary primarily as a function of bone area and body size. Clearly,

certain nonmodiﬁable factors such as race/ethnicity and gender do play a role in

determining a child’s bone growth toward attainment of peak bone density; however it
remains to be elucidated which of these factors, if any, inﬂuence bone growth leading to
osteoporosis risk later in life.
Role of Biomarkers in Assessment of Bone Growth

In addition to direct measurement of BMD with DXA, biomarkers of bone grth
may be useful in predicting bone activity. Much progress is being made with the use of
biomarkers of bone turnover, bone resorption and bone formation in adult women
receiving treatment for low BMD. Biomarkers Shown to have a positive correlation to
bone formation include oseteocalcin (0C), bone alkaline phosphatase (BALP) and
procollagen type I C-terminal propeptide (PICP). Biomarkers shown to have a
correlation with bone resorption include hydroxproline, pyridinium cross-links
(pyridinium (PYD) and deoxypyridinium (DPD)), galactosyl-hydroxylysine, and cross-
linked C-terminal telopeptides of type I collagen (ICTP) (Eyre, 1996). Use of biomarkers
in children has been less extensively evaluated, but some studies indicate that there may
be a use for biomarkers in monitoring growth and therapeutic intervention in some
populations of children (Crofton, 1998). In cross-sectional studies of children over a
wide range of ages, bone biomarker concentrations tend to mirror the grth curve, a
reﬂection of the active bone formation and resorption process inherent to bone growth
(Crofton, 1998). Validation of biomarkers, methods of assay of biomarkers,
understanding of biological variability and potential clinical application of biomarkers in
the pediatric population are areas of study that are in their infancy.

Osteocalcin, also known as serum bone gla protein, is a frequently measured

biomarker, considered to represent bone turnover (Lester, 1995). Osteocalcin is a 49

amino acid protein found in bone and synthesized predominantly by osteoblasts (Delmas,
1995). Osteocalcin represents up to 25% of the noncollagenous protein in bone. Post-
translational vitamin K-dependent carboxylations of its three glutamic acid residues give
OC hydroxyapatite-binding properties (Price et al., 1980). The majority of secreted 0C
is incorporated into the bone matrix but a fraction of newly synthesized OC is released
into circulation where it can be detected by immunoassay. Intact 0C is susceptible to
proteolytic degradation in serum prior to its renal clearance. Immunoreactivity of the
degraded forms of de novo OC may be Similar to fragments released during osteoclastic
degradation of bone matrix (N oonan et al., 1997). Osteocalcin is generally regarded as a
marker of bone formation (Delmas, 1995).

Increased serum concentrations of OC are observed among postmenopausal
women relative to premenopausal women and are associated with rapid bone loss
(Gomez et al.,l994; Johansen et al.,1988; Ross and Knowlton, 1998). Increased 0C is
also seen in a number of conditions characterized by excessive bone turnover including
osteoporosis, Paget's disease, hyperparathyroidism, thyrotoxicosis, and metastatic cancer
(Delmas, 1995; Price et al., 1980; Gomez et al., 1994; Clarke et al., 1995). Osteocalcin
concentrations decrease following antiresorptive therapy in a dose-dependent manner
(Johansen et al., 1988; Gamero et al.,1994). These Short—term changes are inversely
correlated with long-term changes in BMD (Garnero et al., 1994). Osteocalcin
concentrations are decreased in hypothyroidism, hypoparathyroidism, and in Cushing's
syndrome caused by pharmacological glucocorticoid excess (Delmas, 1995; Clarke et al.,

1995).

10

Evaluation of the literature in the area of biomarkers in children is inﬂuenced by
the need to recognize that much of the available data mix pre-pubertal and pubertal
populations of children. Studies have measured OC in children with congenital adrenal
hyperplasia (Lisa et al., 1995) and in normal children of increasing age (Tommasi et al.,
1996), and serum bone gla protein in normal children that correlated weakly with BMD
(Glastre et al., 1990). Hillman et al. (1996) reported decreases in OC in children with
PKU. Slemenda et al. (1997) documented changes in OC concentrations in prepubertal
and pubertal subjects with OC concentrations peaking at Tanner I then beginning to
decline. Fares et al. (2003) noted an increase in OC in boys and girls that peaked at
pubertal Tanner III then began a decline. On the other hand, Mora et al. (1999) observed
an inverse association between 0C and BMD in children at varying stages of puberty.
Slemenda et al. (1997) also noticed an inverse relationship between OC and BMD in a
group of Ca supplemented children. Once the Ca supplementation ended, BMD and OC
concentrations returned to concentrations similar to their unsupplemented twin controls.
Van Coeverden et al. (2002) documented signiﬁcant positive correlations between OC
and BMC at several sites in a group of pen-pubertal Dutch children. These observations
support OC’S potential role in the assessment of bone growth in children. In normal,
healthy children, OC concentrations can be expected to increase with age and body Size
until puberty, then gradually decline to low adult levels (Tommasi et al., 1996;
Bonoﬁglio et al., 2000).

‘ Deoxypyridinoline is considered a biomarker of bone resorption. The organic
matrix of bone consists of approximately 90% type I collagen (Seyedin and Rosen,

1990). Trifunctional pyridinium crosslinks, PYD or DPD, form between hydroxylysine

ll

or lysine residues at the C- and N-telopeptide ends of one collagen molecule and the
helical portion of a neighboring molecule during collagen maturation (Seibel et al.,

1992). This cross-linking provides the ﬂexibility necessary for structural integrity to the
collagen ﬁbril. Osteoclastic degradation of bone collagen releases the crosslinks into
circulation, and they are excreted in urine. Though the crosslinks are present in a number
of tissues, the molar ratio of PYD and DPD in urine is very similar to that in bone,
indicating that urinary concentrations of both are derived mainly from bone.
Deoxypyridimium in particular, with a more limited tissue distribution than PYD, is
derived almost exclusively from bone (Seibel et al., 1992; Robins etal., 1994; Delmas,
1995; Robins, 1995; James et al., 1996). As products of collagen maturation, they cannot
be reused in new collagen synthesis, nor are they further metabolized (Robins et al.,
1994; Delmas, 1995; Robins, 1995). Of the total pool of urinary DPD, approximately 40—
45% is free and the remainder is bound to peptides (Seibel et al., 1992; Robins et al.,
1994; Robins, 1995).

Increased urinary concentrations of DPD are observed among postmenopausal
women compared to premenopausal women and are associated with rapid bone loss
(Hesley et al., 1998; Ross and Knowlton, 1998) and an increased risk of hip fracture
(Garnero et al., 1996; Daele et al., 1996). Increased DPD excretion is seen in a number
of conditions characterized by excessive bone resorption, including osteoporosis, Paget's
disease, hyperparathyroidism,.thyrotoxicosis, malignant hypercalcemia, and metastatic
cancer (Seibel et al., 1992; Delmas, 1995; Robins et al., 1994; Robins, 1995).
Deoxypyridimium concentrations decrease rapidly following antiresorptive therapy in a

dose-dependent manner (Delmas, 1995; Ross and Knowlton, 1998; Garnero et al., 1994).

12

These short-term changes are inversely correlated with long-term changes in bone
mineral density (ROSS and Knowlton, 1998; Garnero et al., 1994).

A handful of studies have looked at DPD as a marker of bone mineralization in
children. Initial efforts to establish reference concentrations have been reported (Lieuw-
A-Fa et al., 1995; Rauch et al., 1994). Rauch et al. (1994) as well as Bollen and Eyre
(1994) report that DPD concentrations are highly correlated with growth velocity in
normal children. Mora et al. (1999) looked speciﬁcally at DPD in relation to Tanner
stage of development and found DPD peaking at Tanner 11 then declining with a positive
relationship to bone volume rather than to bone density per se. Van Coeverden et al.
(2002) documented signiﬁcantly positive correlations between DPD and BMC at several
sites in a group of peri-pubertal Dutch children. In general, DPD concentrations will
increase with age until puberty and then begin to decline to a sustained low adult level
(Rauch et al., 1994; Acil et al., 1996).

Relationship of Body Composition to Bone

The relationship of body composition to bone growth in children has not been as
extensively studied aS it has been in adults. Based on descriptive studies primarily in
adult populations, it is considered that body weight over all weight ranges and bone
density are positively related, with body weight being one of the best predictors of BMD
(Whiting, 2002). Ilich et al. (1998) found that both lean body mass and body fat
predicted bone mass in premenarchal girls from eight and thirteen years age. Pietrobelli
et al. (2002) found that lean mass was the best predictor of BMC in 133 children and
adolescents studied. Other recent pediatric studies suggest that overweight children have

lower than predicted bone mineral content (Goulding et al., 2000). It is important to note

13

that most of the studies of overweight children include children between four and twenty
years of age. Children are often grouped by age with limited regard to pubertal status
(Zamboni et al., 1988; Rauch et al., 1994; VandenBergh et al., 1995; Molgaard et al.,
1997; Carter etal., 2001; Iuliano-Bums et al., 2003), which limits the ability to generalize
the ﬁndings due to the signiﬁcant role of pubertal hormones in bone growth as well as the
variations in initiation and duration of puberty for children. Few studies look exclusively
at prepubertal children (De Simone et al., 1995; Manzoni et al., 1996).

Over the past decade, a few studies have provided insight into BMD in children of
varying body composition. DeSchepper et al. (1995) studied the BMD of the lumbar
spine of 59 obese children (24 of whom were prepubertal). While they did not see
differences in spine BMD, they did see trends in BMD when they compared children
based on duration and severity of obesity once the data were corrected for age and
pubertal status. Children who were obese for more than seven years had a higher mean
BMD than children obese for less than four years. Children with severe obesity had
higher BMDs than children with moderate or mild obesity. In this study, only the lumbar
spine was measured. Though this site does include trabecular bone, which is more
sensitive to metabolic change than cortical bone, this single site measurement is not ideal
for following BMD since whole body scans are available. Manzoni et al. (1996) looked
at total and regional bone mineral content in 65 obese, prepubertal Italian children as
compared to 50 lean counterparts. Lean mass correlated best to BMC in this population.
No difference in total body BMC was found when corrected for age, sex or body
composition of these children. However they observed signiﬁcantly different BMC in

the arms, legs, and trunk, when comparing the obese and lean children, suggesting that

14

BMC must be considered relative to body size. Looking at the inﬂuence of weight, age,
and puberty on bone size and BMC, Molgaard et al. (1997) found that skeletal size is
determined by body Size while BMD is determined by age and pubertal status as opposed
to weight. These ﬁndings are important to note because BMC depends on both the size
and density of bone.

In a rat study, Foldes et al. (1992), found signiﬁcant differences in the bones of
lean versus obese juvenile rats. The obese juvenile ratS’ bones were lighter and smaller
than their lean counterparts. In an earlier human study with Sixteen obese, prepubertal
children, Zamboni et al. (1988) found that BMC and BMC/bone width at the radius were
lower in obese children as compared to non-obese controls. Differences in diet and
several hormone concentrations were also seen. De Simone et al. (1995) looked
speciﬁcally at growth velocity and bone growth over a four-year period in 1250 obese
children between four and eighteen years or age with careful attention to Tanner stage of
development. Growth velocity and Skeletal maturation was accelerated in prepubertal,
obese children who had a less dramatic pubertal growth spurt as compared to normal
children. The obese children made progress toward their adult size at a younger age. The
differences seen in bone age and chronological age in these obese. children raise the
question of the appropriateness of the use of reference standards based solely on age in
the assessment of bone growth as well as in the determination of the need for bone
building nutrients.

Fischer et al. (2000) noted that obese children had larger bones hence more total
body BMC than their lean counterparts. However, this study combined subjects, aged

ﬁve to thirteen and at various Tanner stages, basing the comparative analysis solely on

15

whether the child was lean or obese. Hasanoglu et al. (2000) also found higher BMD in
37 obese children compared to non-obese children. However when they divided the
children by Tanner staging instead of age, they found that BMD was predicted by Tanner
stage rather than body composition. Goulding et al. (2000) also grouped obese children
by chronological age but expressed BMC and bone area as a function of body weight.
They reported a mismatch between body weight and bone growth in overweight children.
Bone mass and bone area were lower relative to body weight, a ﬁnding supported in
obese rats (Foldes et al., 1992). Goulding et al. (2001) followed fracture cases in a case-
control study of boys between three and nineteen years of age and found that not only did
boys with the highest BMI have lower BMD and BMC, they also had a signiﬁcantly
higher risk of distal forearm fracture, adding an orthopedic risk to the already
acknowledged risks associated with poor bone mineralization.
Bone Growth in Special Populations

Investigation of special populations of children, having conditions such as inborn
errors of metabolism or eating disorders, also lend insight into the factors inﬂuencing
bone grth in children. Using wrist radiographs to look at differences in diet and bone
status, Baer et al. (1997) compared ambulatory and nonambulatory children. A high
percentage of the nonambulatory children had low Ca and Vitamin D intake and
ambulatory children had signiﬁcantly higher bone area. Tsukahara et al. (1992) observed
reduced BMD, as measured by DXA, in a small population of Japanese children with a
variety of chronic diseases. Both Allen et al. (1994) as well as Hillman et al. (1996)
reported decreased bone mineralization in children with phenylketonuria (PKU) as

determined by DXA. In the children with PKU studied by Allen et al. (1994), BMD was

16

lower than in the control subjects in Spite of a higher intake of Ca for the PKU children.
Chaturvedi et al. (1993) saw a signiﬁcant reduction in BMD in malnourished Indian
children. Signiﬁcant deviations in BMD compared to normal, healthy children have been
documented in children with bone diseases, renal disease, and endocrine disorders (Shore
and Poznanski, 1996) as well as in premature infants (Steichen et al., 1988; Specker et al.,
2001). Soyka et al., 1999, reported that anorectic, adolescent girls as young as twelve
years of age had signiﬁcantly decreased BMD. Poor mineral accrual persisted in these
girls during the ﬁrst year of their recovery (Soyka et al., 2002). Dibba et al. (2000)
measured BMD and BMC in primarily pre-pubertal children in rural Gambia whose Ca
intake was approximately 350 mg per day in a randomized, double-blind, placebo-control
study. The Ca supplemented group consumed at least 700 mg Ca per day. Measures of
bone mineralization increased in the supplemented group but the children remained
lighter, Shorter, and less mature than reference children of the same age. It is evident that
optimal Skeletal growth in this group of Gambian children needed more than calcium
supplementation.
Influence of Physical Activity on Bone

Physical activity inﬂuences bone growth and turnover both mechanically and
metabolically. Mechanical inﬂuence occurs when bone is exposed to force of varying
magnitude and duration. Metabolic inﬂuence occurs when bone is exposed to a hormonal
and nutritional milieu that is itself inﬂuenced by physical activity as well as many other
factors. Bone responds to mechanical strain in an adaptive manner explained by the
mechanostat theory (Bailey et al., 1996; Murphy and Carroll, 2003). This theory posits

that there are four mechanical usage windows of varying and proportional effective strain

17

levels, each of which stimulates a different bone response, varying from remodeling to
modeling to repair. Varying levels of physical activity change the strain on bone and
result in bone responses that maximize bone strength during and beyond the growth years
(Bailey et al., 1996; Murphy and Carroll, 2003). Animal models have conﬁrmed that
dynamic strain of abnormal distribution, in other words the on-again, off-again plus
twisting and pounding action that reﬂect normal variations of movement, is more
osteogenic than stress alone and more osteogenic than repetitive strain on bone (Murphy
and Carroll, 2003).

Because bone responds to mechanical strain, the inﬂuence of physical activity on
bone mineral accretion during childhood is of great interest. This response is confounded
by the changing hormones of puberty. Therefore, it is important that data reﬂecting bone
growth for prepubertal children must be separate from that of peripubertal or pubertal
children. Slemenda et al. (1994) found that weight-bearing exercise is associated with
increased BMD in prepubertal and peripubertal children, but the greatest inﬂuence of
physical activity was in prepubertal children. Kroger et al. (1993) did not see a
relationship between BMD and physical activity, monitored over a one year period in
seven to twenty year old children, a subpopulation of which were prepubertal children.
Bailey et al. (1999), following children for Six years, a period which included some
prepubertal years, documented much greater total body BMC for boys and girls who were
physically active. Three exercise intervention studies of pre- and peri-pubertal children
also showed a positive effect of weight bearing activity on bone growth (Morris, et al.,

1997; Bradney et al., 1998; McKay et al., 2000).

18

Strong evidence that physical activity contributes to bone mineral accrual can be
seen in the so-called unilateral control studies. Such studies compare limbs that receive
different mechanical loading from physical activity. Any differences seen can be
attributed to differences in the mechanical load because both limbs have the same
genetic, metabolic, and nutritional inﬂuences. Though these studies did not look
exclusively at prepubertal children, a consistent association between limb use and BMD
prevails. The dominant arms of Little League baseball players (Watson, 1974), and of
tennis and squash players (Haapasalo et al., 1994; Kannus et al., 1995) had greater BMD
than the nondominant arm. Even comparison of dominant and nondominant arms of
children involved in routine daily activities Showed increased BMD in the dominant arm
(Faulkner et al., 1993; Bailey et al., 1996).

Simple comparisons of active versus inactive populations support the positive
association between physical activity and BMD. Iuliano-Bums et al. (2003) saw a 3%
increase in BMC in their group of exercised versus non-exercised group of pre- and
early-pubertal girls. The bone building effect of exercise was further enhanced by Ca
supplementation. Specker and Binkley (2003) observed a positive effect of physical
activity in children three to ﬁve years old only in children with high Ca intake (1354
gm/day versus the placebo of 940 gm/day). On the other hand, Molgaard et al. (2001)
saw no signiﬁcant association with physical activity level and BMC in their study of five
to nineteen year old children, nor did VandenBergh et al. (1995) ﬁnd a relationship
between BMC and physical ﬁtness in seven to eleven year old subjects. Zanker et al.
(2003) compared two small groups of children seven to eight years old, comparing a

group of gymnasts with non-gymnasts, ﬁnding signiﬁcantly higher BMD in the female

19

gymnasts, but not in male gymnasts. Bailey et al. (1999) observed that in the year after
attainment of peak BMC velocity, physically active boys and girls had 9% to 17% greater
bone mass than their inactive peers.

Metabolic consequences of physical activity are suggested to be due, in part, to an
endocrine effect. Physical exercise is known to stimulate growth hormone (GH) release
from the pituitary with varying responses observed due to varying duration and intensity
of physical activity (Murphy and Carroll, 2003). This subsequent stimulation of insulin-
like growth factors from the liver further inﬂuences skeletal growth. In addition,
assuming the development of exercise-induced metabolic acidosis, the subsequent
decrease in serum Ca is sufﬁcient to stimulate increased parathyroid hormone (PTH)
secretion. Intermittent PTH administration has an anabolic effect on skeletal tissue
(Murphy and Carroll, 2003).

Relationship of Diet to Bone

Calcium intake is known to be associated with development of peak bone mass
(Weaver, 2000). During attainment of peak bone mass, intakes of less than 1 gm/d of
calcium are associated with lower bone density (Weaver, 2000). When conditions related
to children’s Ca intakes have been examined, differences in the BMD relative to
differences in Ca intake are reported. In reviewing the literature for the inﬂuence of
calcium intake on bone mass, it is important to note that results are often reported for
groups of children who are peri—pubertal instead of clearly delineating pubertal status.
The onset of puberty initiates a hormonal milieu that dramatically changes bone growth,

making pubertal status an important variable.

20

Over 99% of Ca in the body is found in teeth and bones which comprise up to 2
percent of adult body weight (FNB/IOM, 1997). In bone, Ca exists primarily as
hydroxyapatite (Ca10(PO4)6(OH)2) and is absorbed across the intestinal mucosa by active
transport (predominantly with lower intakes) as well as by passive diffusion
(predominantly with higher intakes) (FNB/IOM, 1997). The active transport process is
dependent on 1, 25 dihydroxy Vitamin D3. Fractional Ca absorption varies inversely
with dietary Ca intake and varies throughout the life span. In infancy, fractional
absorption of Ca from an adequate diet is estimated at 60%, adjusting to 28% in pre-
pubertal children, 34% in early puberty, 25% in late puberty and young adulthood and
declining by 0.21% per year with aging (FNB/IOM, 1997). Data for children twelve
months to nine years of age are limited. In addition to a handful of balance studies, the
adequate intake (AI) was estimated by correlating Ca intake and with bone mineral. The
A1 for Ca for children four to eight years of age was set at 800 mg/day, but the need for
more data on Ca balance, Ca accretion and bone mineralization speciﬁc to pre-pubertal
children was acknowledged (FNB/IOM, 1997).

In a carefully controlled study with pre-pubertal subjects, Lee at al. (1993)
reported a positive relationship between higher Ca intake and higher BMC in children of
both genders who were followed ﬁom birth to ﬁve years of age. More speciﬁcally, Ca
intake during the second year of life had the strongest correlation to BMC at the age of 5
years. Henderson and Hayes (1994) also saw increased BMD (measured by DXA) in
children and adolescents of both genders with a milk allergy who consumed higher
amounts of Ca. Ilich et al. (1998) reported that, in a large group of preadolescent

females, BMD of the whole body and BMD of the radius shaft was positively inﬂuenced

21

by lean body mass, body fat, skeletal age and dietary Ca intake. Barr et al. (2001)
estimated habitual Ca intake in nine to twelve year old girls and found that it was
positively associated with total body BMC over a two year peripubertal period, the ﬁrst
measurement being premenstrual but not necessarily prepubertal. Calcium intake was
found to explain 1.6% to 5.3% of the variance in a two-year change in BMC. In a
longitudinal study, Fisher et al. (2004), showed that the Ca intake of girls between ﬁve
and nine years of age positively predicted BMD at nine years of age. Chan (1991)
established differences in BMD between peripubertal girls receiving more than 1000 mg
of Ca/day compared to those receiving less than 1000 mg/day. He reported that BMD is
associated with age, weight and height as well as Ca intake. In a retrospective study,
Stracke et al. (1993) measured BMD in adult men and women and saw a relationship
between low BMD and recollection of low intake of milk and milk products as a child
and adolescent. Stallings et al. (1994) compared the BMD of nineteen children on a low
lactose diet, matched in age to children from Chan’s study, and found that a low lactose
diet resulted in a low calcium intake and lower BMD scores. Cadogan et al. (1997), who
looked at Ca intake as milk in schoolgirls, some of who were pre-pubertal and pooled
relative to Tanner stages, found that for some of the analyses, the girls who received their
Ca via milk had more signiﬁcant gains in BMD. In a randomized, controlled trial of milk
supplementation of male and female school children beginning when the children were
seven to nine years of age, F ehily et al. (1992) followed the children for fourteen years.
At 20 to 23 years of age, BMD and BMC tended to be higher in the milk-supplemented
group with strong positive associations of BMC with adult body weight and sports

activity during adolescence. Using National Health and Nutrition Examination Survey III

22

(NHANES III) data, Optowsky and Bilezikian (2003) looked for an association between
childhood milk consumption and BMD in young adult women and postmenopausal
women. Early milk consumption was positively related to BMD for white women but
not for black women and was positively related to adult milk consumption in both
groups.

In spite of evidence of the importance of Ca in attainment of bone mass in
prepubertal children, the inﬂuence and long term beneﬁt of supplemental Ca remains
unclear. Johnston et al. (1992) using a co-twin study model, saw increased BMD in
prepubertal children who received Ca supplements. A follow-up study on these children,
however, indicated that this beneﬁt was not sustained as the BMD of the supplemented
and unsupplemented groups were Similar six years later (Slemenda et al., 1997). Bonjour
et al. (1997) supplemented prepubertal girls’ diets with Ca for one year in a double blind,
placebo controlled study to conﬁrm that a Ca enriched diet increased bone mass accrual.
A follow-up study Showed this increased bone mass was sustained beyond the end of
supplementation (Bonjour et al., 2001). On the other hand, Lee et al. (1996) saw the
beneﬁt of Ca supplementation disappear eighteen months after treatment was withdrawn
from this group of Chinese children who had received 300 mg of elemental calcium for
eighteen months. Overall, evidence related to the long-term effects of Ca
supplementation in childhood indicates that gains are not generally sustained (Abrams,
2005) suggesting bone mineralization depends on a constellation of factors, only one of
which is Ca intake.

In addition to Ca, several other dietary factors are considered of importance in

adequate bone mineralization, including vitamin D, phosphorus (P), and protein.

23

Phosphorus is one of the most abundant elements on earth and has a close
interrelationship with calcium in the human body. Most of the P in the body is found in
bone where the constant Ca to P ratio of 2:1 is maintained (Anderson et al., 2006).
Phosphorus homeostasis is maintained by PTH and 1,25 dihydroxy Vitamin D.
Phosphorus is abundant in the US. diet, found in high levels in soft drinks as well as in
meat and food additives (Anderson et al., 2006). The Estimated Average Requirement
(EAR) for P in the four to eight year old child is 405 mg/day (FNB/IOM, 1997). The
median intake is estimated at 420 mg with the upper 50% of children consuming 420 to
1100 mg (Anderson et al., 2006). These intake estimates are for P contributed to the diet
by food sources with the exclusion of food additives, as the P in food additives is not
considered in the nutrient databases used for dietary analysis. The concern with high P
intake is that chronic high intake may impair the adaptive mechanism needed for
adequate Ca absorption and optimal bone accretion through interruption of the 1, 25
dihydroxy Vitamin D and PTH homeostasis system (Anderson et al., 2006), resulting in
an adverse effect on bone.

Vitamin D also plays a critical role in bone health. The active form of Vitamin D,
1, 25 dihydroxy Vitamin D3, is considered to be a steroid hormone. Vitamin D is a
regulator of Ca homeostasis and has roles in a wide variety of cell differentiation and
proliferation processes (Norman and Henry, 2006). The current A1 for Vitamin D for
children is ﬁve micrograms/day (FNB/IOM, 1997). Measuring circulating concentrations
of the precursor to active Vitamin D, serum 25-hydroxy Vitamin D3, best assesses
Vitamin D status (Molgaard and Michaelsen, 2003). Concentrations below 10 ng/mL are

considered at risk for deﬁciency (Norman and Henry, 2006) though there is no general

24

consensus on how to deﬁne optimal Vitamin D status (Molgaard and Michaelsen, 2003).
Fortiﬁed milk and sunlight are the primary sources of Vitamin D intake for US. children.
Concern over Vitamin D status is increasing due to decreased exposure to sunlight and
decreased consumption of ﬂuid milk (Molgaard and Michaelsen, 2003; Norman and
Henry, 2006).

Protein’s relationship to bone health is paradoxical. Protein is considered to have
a bone health promoting effect in the elderly, but at higher levels of intake, many
consider protein to be deleterious to bone health (Ginty, 2003). The positive inﬂuences
of protein on bone are related to observed decreases in fracture risk in the elderly as
protein intake increases over a baseline deﬁcient level (Ginty, 2003). The inﬂuence of
protein on attainment of peak bone mass during growth is not as clear-cut. An adequate
intake of protein is necessary for growth and insulin-like growth factor (IGF) is an
acknowledged mediator of anabolic processes whose production is up regulated by
dietary protein (Ginty, 2003). Calcium supplementation via one pint of milk daily
increased plasma IGF-1 and bone mineralization in their group of pen-pubertal girls,
hypothesizing an association between the increase in dietary protein and rise in IGF
(Cadogan et al., 1997). Later, Ginty et al. (2004) saw an increase in IGF-I and bone
mineral gain secondary to calcium supplementation rather than protein increase in their
study, raising the possibility that IGF-I responds to Ca as well as protein. How the
anabolic action of IGF-I is regulated is not clear, especially in the area of the
interrelationship of level of protein and Ca intake and bone mineralization in prepubertal

children.

25

Protein, especially that of animal origin, is hypothesized to have a negative
relationship to bone due to the reduction in blood pH resulting from hepatic oxidation of
the sulfur-containing amino acids methionine and cysteine to H2804 as well as
ammonium ion production resulting in increased bone resorption and increased urinary
Ca losses (Ginty, 2003; Massey, 2003). In addition to meat protein, dairy and many
legume and grain products have high potential renal acid loads (Massey, 2003; New,
2003). The resulting chronic, diet-induced, low-grade, metabolic acidosis is thought to
reduce bone mineral content over time as bone mineral is slowly used to buffer the net
acid load of the diet (F rassetto et al., 1998; Remer et al., 2003). Recent work has
established the likely protective relationship of fruit and vegetable and potassium (K)
intake on indices of bone health in populations of children and adults (New, 2003).
Fruits and vegetables contribute to a dietary alkali load (K, sodium, Ca, and Mg generate
salts of bicarbonate and citrate), which contribute to neutralizing the pH-lowering effects
of protein (F rassetto et al., 1998; Remer et al., 2003; Ginty, 2003). Because bone health
is inﬂuenced by many other factors including, other nutrients, physical activity, and
hormonal status, in addition to pH, the Food and Nutrition Board of the Institute of
Medicine has concluded in the DRI for protein issued in 2002, relative to consideration of
intake of protein on bone health, that “the potential for implications of high dietary
protein are not sufﬁciently unambiguous at present to make recommendations”
(FNB/IOM, 2002).

Other nutrients involved in bone health include magnesitun (Mg), ﬂuoride (F),
and Vitamin K. Magnesium is a required cofactor for more than 300 enzymes and 50%

to 60% of the Mg in the body resides in bone. Though Mg deﬁciency is considered a risk

26

factor for osteoporosis (Volpe, 2006), the role of Mg in attainment of peak bone mass in
children remains to be determined. Fluoride’s role in dental health is well known and 99
percent of the body’s F is found in calciﬁed tissues (FNB/IOM, 1997). Through the
years there have been a few reports to link F to BMD (FNB/IOM, 1997) but, like Mg, the
role of F in attainment of peak bone mass in children is unknown. The role of Vitamin K
in blood coagulation is fairly well understood, but more recently Vitamin K is receiving
more attention for its role in bone health (Bugel, 2003). In both of these roles, Vitamin K
is essential for the post-translational conversion of glutamyl residues to y-
carboxyglutamyl (Gla) residues in a class of Vitamin K-dependent proteins, including
prothrombin and osteocalcin. Initial intervention studies indicate that perhaps Vitamin K,
when supplemented along with Vitamin D, may increase BMD in adults (Bugel, 2003).
In pioneering work in children, Kalkwarf et a1. (2004) measured Vitamin K intake and
several markers of bone turnover in girls between three and sixteen years of age. They
were able to establish an association between Vitamin K status and reduced bone
turnover in these pen-pubertal children.
Background Information Summary

The literature on factors contributing to the attainment of peak bone mass in pre-
pubertal children is growing but much remains to be explained relative to prevention of
osteoporosis risk later in life. Clearly there are nutritional modiﬁers such as Ca and
protein intake, body composition modiﬁers such as percent body fat and physical activity
modiﬁers. It is very important to understand the role of these modiﬁable inﬂuences of
bone growth independent of the inﬂuence of the hormones of puberty. Much of the

literature to date groups pre-pubertal children with those who are in puberty for analysis,

27

confounding the conclusions and generalizations that can be made about bone growth in
those populations. Investigations of pre-pubertal children cannot be based on age but
must instead be based on Tanner staging to allow consideration of physiologic age. Diets
need to be carefully considered using validated tools and trained assessors. In addition,
physical activity and body composition must also be precisely and accurately measured.
It is only in understanding the associations among all of these factors that an
understanding of bone growth in children can evolve toward recommendations that will
impact the incidence of osteoporosis in these children’s later years.

Study Objective and Hypothesis

This study addresses the question: Is there an association between diet, biomarkers of
bone turnover, physical activity, and body composition in their effect on bone
mineralization in healthy, Caucasian children between the ages of ﬁve years and puberty?

Additional Study Hypotheses and Questions

Children who have a higher percent of lean tissue (relative to total body weight) will have
higher measures of bone mineralization.

Children who are more physically active have higher measures of bone mineralization.

Children who consume 67% or more of the Al for calcium will have measures of bone
mineralization within normal limits.

Identify how study variables compare to reference values for this population of children.
Describe the sources of bone building nutrients in this population of healthy children.

Is there an association between levels of OC and DPD and BMD in prepubertal children?
Identify values for serum OC and urinary DPD in a group of healthy children.

The levels of OC, considered to reﬂect bone formation, will increase as bone mineral
status decreases.

The levels of DPD, considered a marker of bone resorption, will increase as bone
mineralization increases.

28

CHAPTER TWO

29

CHAPTER TWO
METHODS FOR THE HEALTHY KIDS NUTRITION STUDY

Study Summary

The Healthy Kids Nutrition Study (HKNS) was an observational study of a
convenience sampling of children between the ages of ﬁve years and puberty. Bone
mineral density and body composition of the children was determined by dual energy x—
ray absorptiometry (DXA). Diets of the children were analyzed using a food frequency
questionnaire (FFQ) and a usual, 24-hour diet recall administered by a registered
dietitian. Two biomarkers of bone metabolism, serum osteocalcin (OC) and urinary
deoxypyridinoline (DPD) were determined. Serum Vitamin D was also determined.
Heights and weights of the children were collected. Physical activity for three days was
measured using accelerometer technology. This study measured modiﬁable determinants
of bone mineralization to examine associations among those variables and bone mineral
density and bone mineral content in prepubertal children.
Project Design
Subjects and subject recruitment

This study was approved by full review of the University Committee on Research
Involving Human Subjects (UCRIHS) at Michigan State University (MSU), was assigned
the Institution Review Board number 03-1019 (Appendix 1; Appendix 2), and was
renewed annually. Subjects were recruited from the greater Lansing, MI, area through
word of mouth and ﬂiers (Appendix 3). The recruitment target was 50 subjects who met
the study inclusion criteria; 52 subjects were enrolled. Subjects were generally well,

free-living children with no chronic disease conditions or birth defects other than possible

30

allergies or food sensitivities. Pubertal status was determined by parent/child interview
to ascertain Tanner Stage 1 status. Children with minor conditions, such as ADHD or
mild allergies, which have no direct relationship to bone growth, were included after
review by the study physician (R.T. Scott, DO). Due to the pre-pubertal age of these
subjects and the relatively small sample size, all races and both genders were recruited.
The following criteria were used for subject selection. Inclusion criteria were well
children, both genders, weight by NCHS standards of _>_ 5th percentile BMI, age ﬁve years
to puberty (puberty being deﬁned by >Tanner I) at the beginning of the study, at least 37
weeks gestational age at birth, and all races. Exclusion criteria were any chronic disease
or birth defect, condition or medication with potential impact on bone growth, chronic
steroid use, pubertal stage of Tanner Stage 2 to 5, weight by NCHS standards of 5 5th
percentile BMI. Data were collected between February 2005 and June 2006.

Client conﬁdentiality and HIPAA compliance were considered (Appendix 4).
Subjects were not informed of the exact nutrient or food group being investigated; rather
the consent stressed investigation of overall nutrition and bone growth. A health
questionnaire (Healthy Kids Nutrition Study Health Questionnaire) was administered to
conﬁrm each child’s qualiﬁcation to participate and to provide a general family history
relative to bone health (Appendix 5). The health questionnaire covered inclusion and
exclusion criteria as well as family history of osteoporosis and obesity, and the child’s
history of any illness or infection, which may have inﬂuenced growth. Once enrolled in
the study, random numeric coding of all subjects’ records and samples assured

conﬁdentiality.

31

Study procedures were scheduled at the caregiver’s convenience, including
evening and weekend contacts as necessary. Over a period of two to ﬁve weeks, subjects
and parents came to three appointments for data collection, a process outlined in Figure
2.1 (with additional detail in Appendix 8). Parking vouchers were provided and study
personnel assisted with pick up and delivery of study supplies and samples as needed.
The total incentive for participation was $100.00, provided as a gift card for Target or
Meijer. Each caregiver will be provided with a report of the study results for their child,
per the study debrieﬁng protocol (Appendix 6). Research personnel remain available for
questions or concerns post-participation. The children’s caregivers were instructed to
report any abnormal ﬁndings, i.e. BMD Z-scores below —0.2, to their provider of choice
for appropriate medical follow-up.

Study Stafﬁng

A team of researchers and staff with the speciﬁc expertise required to complete all
aspects of this study collaborated, developed and completed this study. The project
manager, a Registered Dietitian, conducted all subject interviews and instructed the
subjects and parents on study protocol and procedures. She obtained anthropometrics,
conducted the diet recall interview, and provided instruction on completing the FF Q,
wearing the activity monitor, and obtaining the ﬁrst morning urine sample. She coded
the FFQ for analysis and entered the 24-hour recall dietary information for analysis.
Entries were double checked a second time by either a research assistant or the project
manager. The nutrition database program within the MyPyramid Tracker (Center for
Nutrition Policy and Promotion, 2005) was used for diet analysis of the recalls. The F FQ

was sent to the Harvard Charming Laboratory for scanning. A phlebotomist who had

32

 

 

  

388$ seam aoaaaz BE Assam
2 2:3

33

experience in drawing blood from children collected one ﬁve ml blood sample from each
subject for serum Vitamin D and serum OC. For Vitamin D analysis, the samples were
sent to the Regional Laboratory of Ingham Regional Medical Center, Lansing, MI. The
laboratory of Dr. Michael Orth, Department of Animal Science, MSU, conducted
biomarker assays, OC and DPD. A radiology technician at Fiechtner Research under the
supervision of Justus J. Fiechtner, MD, MPH, conducted the DXA procedures. The US
Food and Drug Administration (FDA) requires a physician order for DXA measurement,
thus a standing order from the study physician was placed on ﬁle in each subject’s data
ﬁle (Appendix 7). The study was coordinated out of the Michigan State University
Department of Food Science and Human Nutrition with an ofﬁce in the College of
Osteopathic Medicine’s Department of Family and Community Medicine. The project
manager and two research assistants conducted data entry for statistical analysis. All
entries were cross-checked by the data entry team. Statistical consultation was used as
described in the data analysis section.
Measurement and Analysis of Variables
Diet analysis

Individual diet information was collected as a one day, usual diet recall and
through a food frequency questionnaire (FFQ), the Youth and Adolescent Food
Frequency Questionnaire developed by Harvard Charming Laboratory. The nutrient
content of diets of the children was estimated from both the recall and the FFQ. The
nutrition database program within the CNPP/USDA/DHHS MyPyramid Tracker, based

on the USDA food database, was used for analysis of the diet recalls (CNPP, 2005). All

34

records, if submitted incompletely, received individual follow-up by study personnel with
the parent or guardian.

The diet assessment tools and analysis program provide estimates for over sixteen
nutrients. The recall analysis also estimated intake according to the 2005 US Dietary
Guideline’s food groups (CNPP, 2005). The FFQ chosen has been validated for use in
estimating Ca intakes of young children (Stein et al., 1992; Rockett et al., 1997; Rockett
and Colditz, 1997). Its validity in determining density of other nutrients in the diet is not
as strong. Combining the FFQ with a careful diet recall improves the results to give a
better estimate of intake. The food frequency used, the Harvard Semiquantitative Youth
and Adolescent Food Frequency Questionnaire, was validated at Harvard University,
School of Public Health, Boston, MA, USA (Thompson and Byers, 1994; Rockett et al.,
1997). This food frequency can be customized via individual coding by the investigator
to include speciﬁc foods of interest. Due to the prevalence of consumption of calcium-
fortiﬁed juice, this single food was added to the FF Q. Supplement use was measured in
the F FQ but not in the diet recall. The FFQ asks whether or not vitamins are used, how
often, and for how long.

In order to explore the potential relationship of dietary protein on measures of
bone mineralization, net endogenous acid production of the subjects’ diets was estimated
using the renal net acid excretion (RNAE) equation developed by Frassetto et al. (1998).
This method of estimation uses dietary protein intake and dietary potassium intake such

that RNAE = 62 (gm protein / mEq K) —17.9.

35

Anthropometrics

Height was measured to the nearest centimeter with a stadiometer (Invicta
Plastics, Oadby, Leicester, England). Weight was determined to the nearest 0.1 kg on a
digital scale (Seca, Model 882, Hanover, Maryland, USA). Standard pediatric
measurement protocols to assure precision and accuracy were employed (Gibson, 1990;
Lee and Nieman, 1996) and followed the National Center for Health Statistics guidelines
for standard measurement (NCHS, 1996).
Measurement of Physical Activity

The Actical accelerometer (MiniMitter, Bend, OR) was used to measure physical
activity. Actical is a compact, battery-operated physical activity monitor with physical
characteristics similar to a small wristwatch. The monitor consists of the activity monitor
itself and a waist or wrist/ankle band. For this study, the accelerometer was held ‘at the
waist via a belt, available in 2 sizes and adjustable with Velcro-type closures. Actical
utilizes a piezoelectric accelerometer to monitor the occurrence and degree of motion.
The sensor is an omni directional accelerometer, resulting in sensitivity to motion in all
directions (MiniMitter Corp, 2003). It has been validated for use in estimating energy
expenditure in young children (Pfeiffer et al., 2006). The sensor and associated signal
processing considers both the degree and intensity of motion to produce an electrical
current that varies in magnitude. An increased degree of speed and motion produces an
increase in voltage. Actical stores this information as activity counts. The sampling
frequency was 32 Hz and an epoch measurement of 15 seconds was chosen. Subjects
wore the monitor on the waist continually for three, consecutive days (one weekend day

and two weekdays).

36

The three-day record of physical activity data from the monitor record was
downloaded using the Actical Reader, companion hardware for use with the monitor,
providing database-ready statistics on physical activity. The data were collected and
reported as activity counts, energy expenditure, and duration of expenditure. The height
and weight of each subject was uploaded to the Actical prior to its wearing. From this
information, basal energy expenditure (BEE), using the Harris and Benedict equation
(Harris and Benedict, 1919), was estimated by the Actical program so that daily total
energy expenditure (DTEE) from physical activity was reported as kcals above estimated
BEE. In addition, a brief physical activity questionnaire was completed that included two
questions about physical activity from the CDC Youth Risk Behavior Surveillance
Survey (CDC-MMWR. 2006) to allow comparison to national data. These data were
collected for future use with the potential to validate the survey tool.

Biomarker Assays

Serum intact osteocalcin (OC) was measured with the MetraTM Osteocalcin assay
(Quidel Corporation, Special Products Group, Santa Clara, CA, USA), an enzyme
immunoassay in a microtiter stripwell format utilizing a murine monoclonal anti-0C
antibody (Gomez et al., 1994). The OC in the sample competes for antibody binding
sites with 0C coated on the stripwell. A rabbit anti-mouse IgG antibody conjugated to
alkaline phosphatase is added and the reaction is detected with the substrate, p-
nitrophenyl phosphate. Color developed during the incubation of captured enzyme
conjugate and substrate is measured at 405 nm in a microtiter plate reader. The OC

values of unknown Specimens are calculated from a calibration curve ﬁt with a 4-

37

parameter logistic equation. Values are expressed in ng/mL. Twenty-ﬁve uL of serum
per well (assayed in duplicate) were used for determination of OC.

The detection limit of the OC assay was 0.45 ng/mL. Samples up to 32 ng/mL
could be read without dilution. Intra-assay precision coefﬁcient of variations (CVS) were
4.8-10.0%. Interassay precision CVs were 48-98%. The manufacturer has established
reference intervals for healthy men (3.4-9.1 ng/mL) and women (3.7-10.0 ng/mL). No
reference standards were provided by the manufacturer for children. For this assay, a ﬁve
m1 venous blood sample was drawn via antiseptic techniques with universal biohazard
precautions. Immediately after being drawn, the blood was allowed to clot. The serum
was harvested after the sample was centrifuged for 20 minutes at 1500xG. All samples
were immediately frozen at -20°C and then transferred to -80°C storage within 4 hours
of being drawn. All stored samples were analyzed in a single batch within sixteen
months of being drawn.

Urinary free deoxypyridinoline (DPD) was measured with the MetraTM DPD
assay (Quidel Corporation). The assay is highly speciﬁc for the DPD molecule and does
not cross-react signiﬁcantly with pyridinoline, other collagen crosslinks, or collagen
peptides (Robins et al., 1994). Metra DPD is a competitive enzyme immunoassay in a
micro assay stripwell format utilizing a monoclonal anti-DPD antibody coated on the
strip to capture DPD. The DPD in the sample competes with conjugated DPD-alkaline
phosphatase for the antibody and the reaction is detected with the substrate, p-nitrophenyl
phosphate. Color developed during the incubation of captured enzyme conjugate and
substrate is measured at 405 nm in a 96-well micro assay plate reader. The DPD values

of unknown specimens are calculated from a calibration curve ﬁt with a 4-parameter

38

logistic equation. The DPD values are expressed in nmol/L. Urine (ﬁfty uL) was diluted
1:10 in Assay Buffer per well (assayed in duplicate) as required for determination of
DPD.

The detection limit of this assay is 1.1 nmol/L. Samples up to 300 nmol/L could
be read without additional dilution. Intra-assay precision CVS were 4.3-8.4%. Interassay
precision CVs are 3.1-4.8%. Due to the diurnal variation of DPD excretion, the reference
and study samples were determined from ﬁrst morning urine samples collected prior to
10 AM. Parents and subjects were instructed on the urine sample collection and given
sterile supplies. The DPD concentrations were corrected for differences in urine
concentration and output by dividing by creatinine as measured in each urine sample
(Quidel Corporation). The ﬁnal creatinine-corrected DPD results were expressed as
nmol/mmol. The manufacturer has established reference intervals for healthy men (2.3-
5.4 nmol/mmol) and women (3.0-7.4 nmol/mmol); none are provided for children.

First-moming urine samples were collected without preservative. Parents were
instructed to put the urine specimen tubes in the home freezer until transport to the
project ofﬁce, after which the samples were transferred to -80°C storage and stored for
less than 12 months. The DPD is stable for at least 21 months when stored at 340°C.
Modeling studies suggest DPD will be stable for at least 20 years when stored at -20°C
(Gerrits et al., 1995). Samples may be frozen and thawed up to 5 times. Prolonged
exposure to light, especially sunlight, should be avoided, but routine processing is not
affected by normal, artiﬁcial laboratory lighting. All samples were analyzed within

sixteen months of being stored.

39

Vitamin D measurement

Serum 25(OH) Vitamin D3 was measured one time during the study, concurrent
with the DXA measurement. The 5 ml sample was drawn in a nonfasting state, clotted,
and then spun for serum. The serum was stored at -20°C and transported to the Regional
Laboratory of Ingham Regional Medical Center (IRMC), Lansing, MI, within 24 hours of
being drawn. Samples were stored at IRMC at 320°C, batched, then sent to the
accredited laboratory, LabCorp, Dublin, OH, for analysis of 25(OH) Vitamin D3 via an
immunocherniluminometric (ICMA) assay (personal communication, LabCorpS
Analytical Laboratories, Dublin, OH). All samples were analyzed within four months of
being drawn using standardized storage and handling protocols. The reference standards
provided by the analytical lab vary with age and season as 25(OH) Vitamin D3 exhibits
seasonal variation related to sun exposure although recommendations for optimal levels
in children are not season-speciﬁc (Molgaard and Michaelson, 2003; Norman and Henry,
2006)
Dual Energy X-ray Absorptiometry

Whole and regional body BMC, BMD and body composition of lean mass and fat
mass were determined by DXA. Whole body measures were taken using a GE LUNAR
Prodigy machine at F iechtner Research with pediatric software Speciﬁc to the machine
(software version 6.81; General Electric Lunar Corporation, Madison, WI).
Measurement protocols speciﬁc to the GE-LUNAR Prodigy densitometer were
employed, including daily calibration with a standard calibration block and calibration
tri-weekly with a body phantom. The CV of this machine is reported to be 0.15%

(personal communication, Fiechtner Research). The LUNAR densitometer is FDA

40

approved for investigational pediatric use in the USA and a physician’s order is required
for use (Appendix 7).

Bone mineral content as measured by DXA is the amount of mineral per length of
the bone scanned, expressed in g/cm. Dividing the BMC by bone area gives bone
mineral density in g/cmz. Both measures are projectional area densities as opposed to
true volumetric measures. Projectional measures, as opposed to cross-sectional
measures, allow for the assessment of both trabecular as well as cortical components of
the bone being measured. True volume measures of bone density are only measurable by
using quantitative computed tomography (QCT), a technology not currently available to
the investigators and which exposes children to Signiﬁcantly more radiation than does
DXA (Shore and Poznanski, 1996). Because of growth, it is important to measure more
than one Site of the skeleton via DXA to get the best overall assessment of bone
mineralization in children; thus a whole body scan was conducted. Results can be
expressed as BMC, bone area and BMD. The GE Lunar Prodigy machine measured body
composition (fat and lean) at the same time BMD was scanned, hence Lunar’s deﬁnition
of this measure as a “total body scan.” The full body scan took approximately two
minutes and exposed each child to 0.04 mRem of radiation, a very low dose exposure. In
comparison, a television emits 10.0 mRem of radiation over the course of a year while a
typical air ﬂight in North America exposes the passenger to a 40.0 mRem dose of
naturally occurring radiation. A standard medical x-ray carries a 40.0 mRem dose (J.

Downs, GE Lunar, personal communication, 2004; Njeh et al., 1997).

41

Sample Size and Power Calculation

Reference data for the bone density of healthy, age-matched controls is included in the
LUNAR pediatric software. These data are age and gender speciﬁc though at many age
groups, data for prepubertal and pubertal children are pooled. The primary research
question involves examination of the relationship of modiﬁable inﬂuences of bone
growth within the study population rather than in comparison to the reference data,
though the reference data is analogous to information that would be obtained by a control
group of subjects. The BMD reference values from the GE Lunar DXA program were
used in power calculations. According to Lunar’s data (Wacker and Barden, unpublished
monograph, 2001), the reference data are based on 1494 children between ﬁve and 19
years of age, only 471 of whom were between ﬁve and ten years of age. The average
standard deviation (SD) for total body BMD in both males and females between the ages
of 5 and 19 years is reported by Lunar to be 0.07 (Table 2.1). The SD within each age
group is not reported. Given these data, the mean BMD for the seven year old male was
used as the mean BMD of the population and a SD of 0.07 was used to calculate power,
resulting in a minimum sample size of 16 to detect differences at a 95% Conﬁdence
interval. In consideration of realistic expectations of subject recruitment and retention
over a reasonable time period as well as the limitations of the information available to
estimate power, 52 subjects were recruited, allowing for subgroups by age or gender for
post hoc categorization. Differences in measures of bone mineralization have been seen
in studies of children with as few as fourteen to sixteen subjects in each group (Zamboni
et al., 1988, Volek et al., 2003). The design of this study allowed for addition of subjects

at any time during the study, addition of subjects after completion of this initial study,

42

addition of more study groups, and further follow-up of these same children over time.
In the end, 52 subjects went though the study protocol with 51 subjects providing blood
samples and all 52 participating in all other aspects of the study.
Data Analysis

Descriptive statistics were employed to describe the study population with
data being presented as means i SDS. The primary dependent variables are BMD, BMC,
the Lunar DXA reported Z-score and BMC expressed relative to height in centimeters.
The independent variables are select components of the dietary intake closely associated
with bone growth as assessed by recall and by F P Q, diet as assessed by the USDA
Dietary Guidelines, serum 25(OH) Vitamin D, serum OC, urinary DPD, gender, age,
body composition expressed as percent fat and lean tissue, and physical activity
expressed as energy expenditure. A two-tailed t-test was employed to determine
signiﬁcant differences between gender and age categories in all variables under study.
Diets were expressed as nutrition adequacy ratios (NARS) for bone building nutrients.
Relationships between variables were explored with 2-tailed Pearson’s correlations.
Multiple variable regression with stepwise enter method was used to determine variables
that may be signiﬁcant predictors of BMD and BMC. Data were also grouped to allow
further analyses by cross tabulation. Data was analyzed using SPSS, version 13, Base
System (SPSS, Inc., Chicago, IL, USA). The signiﬁcance level for all tests was

established as p<0.05 and all tests are bi-directional.

43

 

 

beam cost; Z 83— .3:on 2: 5 38.33 .8 828 :82 £225
38828888 .8883 .58 5.82 GB. £215 ”82:86:00 .8150? baoom Reece—:2:— 05 8 coup—omni .UEm use 035
22.9. one been :38 22:8 can 28. Low 88 02882 2533; .m: 5me .3 Loo—235v 855 m0 .3 88 8:208. 8 3235 828 :82 853.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

. . . . . . . . so
8 w: 8 8 mm B: 8 o 8 :N 8 o 8 on 8 8 am swag...
B
So
R R 8 mm _ a
. . . . . . . . 20
8 8: 8 o 8 8 8: 8 o _ 8 8: o; o 8 8 e : 8” o 8 8% 585
8.82 88 a. o 8.82 88 mm 3%: 88 4 2089A EH
8.8: 88 8 8.2: 88 N 8.8. 88 8 mass 28 m 2388 2:2
8.8: 88 8 8:: 88 m 8.88 88 8 2.8 88 m 228.388
8.8 88 mm 8.8 88 m 8.8 28 a. 8.8 88 m 238% 85m
8.8 88 8 2.8 88 A 88R 58 8 :88 88 A 238A 8
8.8 Ed 2 8.5 28 a 8.8 88 t 8.8 28 4 238A of
A858 A858 A808 A833
w w w w
A 535 88 z A 52m 88 z A 52m 92m 2 A 535 85 2 AA w
83 m8: 83 mzm: as a a <
882 moEEom

 

 

Baa Amzmm 8a 884

A8838 88853...“ 95% 55:3 CO 0888 use dovcow use 95% own ,3 8888 035 use 035 we :88.qu0

mm 2931

44

CHAPTER THREE

45

ABSTRACT

BONE MINERALIZATION IN PREPUBERTAL CHILDREN: ASSOCIATION
OF DIET, BODY COMPOSITION, AND PHYSICAL ACTIVITY

by

Marcia Kelly Scott
Background: Osteoporosis is predicted to become a disease of epidemic proportions
within the next several decades. Attainment of higher bone mineral content during the
ﬁrst two decades of life decreases the risk of developing osteoporotic fractures later in
life. Prevention of osteoporosis begins in childhood with attainment of peak bone mass.
Objective: To further clarify associations among determinants of bone growth toward
attainment of peak bone mass including body composition, diet composition, and
physical activity with bone mineral density (BMD) and bone mineral content (BMC) in
prepubertal children, mean age of 7.6 years.
Design: BMD, BMC, and body fat were measured via DEXA in 52 subjects. Daily total
energy expenditure (DTEE) was measured via accelerometer. Diet was assessed with a
usual, one—day recall interview and a food frequency questionnaire. Relationships
between variables were explored with two-tailed t-tests, two-tailed Pearson’s correlations
and stepwise multiple regression analysis (pS0.05).
Results: In these children who were all within the normal, healthy range of body weight,
bone mineralization correlates positively with percent body fat (r=0.535 for BMD and
r=0.646 for BMC) and with DTEE (r=0.460 for BMD and r=0.591 for BMC). Protein
intake had signiﬁcant negative relationships to both BMD, r=-O.508, and BMC, r=-0.564.
Energy intake also had signiﬁcant negative relationships to both BMD, r=-O.510, and

BMC, r=-0.578. A signiﬁcant negative association was found between Ca intake and

46

BMD (F-0.277, p.<_0.05). In the ﬁnal regression models, energy intake and P intake by
height were signiﬁcant negative predictors of BMD and BMC while DTEE, percent body
fat, Ca intake by height, Mg intake, and fruit and vegetable intake were signiﬁcant
positive predictors. The relationships of these variables explain 58.8% of the variability
in BMD and 72.9% of the variability in BMC.

Conclusion: This study supports the interrelationships of diet, physical activity, and
body composition in contributing to development of peak bone mass. Increased bone
mineralization was associated with higher body fat, more physical activity, and higher
fruit and vegetable intake as well as moderate energy, Ca, Mg and P intake, supporting
consideration of these variables in assessing bone health in children. These associations
suggest that children at risk of poor bone mineralization can be identiﬁed in a
noninvasive manner such that early intervention and prevention can be addressed at a

young age.

Chapter Three formatting is consistent with the guidelines for authors for the American

Journal of Clinical Nutrition.

47

CHAPTER THREE
BONE MINERALIZATION IN PREPUBERTAL CHILDREN: ASSOCIATION
OF DIET, BODY COMPOSITION AND PHYSICAL ACTIVITY
INTRODUCTION

Osteoporosis, recognized as one of the major public health problems facing aging
individuals of both genders, worldwide, is predicted to become a disease of epidemic
proportions within the next several decades (Riggs and Melton, 1995 ; WHO, 2003).
Broadly deﬁned as diminished bone mass and reduced bone mineral density (BMD) at
the level of 2.5 standard deviations below the referent BMD of young adults (Riggs and
Melton, 1995; USDHHS, 2000; WHO, 2003), osteoporosis presents clinically as
fractures, both traumatic and nontraumatic in nature, resulting in signiﬁcant morbidity,
mortality, and health care costs (WHO, 2003). The two most important ways to reduce
risk of osteoporosis are to attain peak bone mass in early life and to have a low rate of
bone loss in later life; hence, approaches to interventions for this disease target methods
of prevention and abatement of osteoporosis not just in adulthood but also during infancy,
childhood, and adolescence. Though osteoporosis is generally considered a geriatric
disease, osteoporosis may actually be a pediatric disease with geriatric consequences
(Faulkner and Bailey, 2007).

Many believe that attainment of higher bone mineral content during the ﬁrst two
decades of life decreases the risk of developing fractures later in life (Heaney et al., 2000;
Faulkner and Bailey, 2007). Bone mass tends to track throughout life, meaning that
infants, children and adolescents that have higher bone mass tend to be the adults who

have high bone mass (Ferretti et al., 1998; Heaney et al, 2000). This association

48

introduces the possibility that “at risk” individuals may be identiﬁed early in life. Peak
bone mass has genetic, hormonal, nutritional, and behavioral determinants (Soyka et al.,
2000). Some of these determinants are modiﬁable and interventions targeted at
individuals at risk should focus on these modiﬁable determinants of bone health.
Generalizing the ﬁndings of bone growth studies in children is challenging due to
the variations in initiation and duration of puberty for children. Studies of bone growth in
early childhood should consider pubertal status as well as modiﬁable inﬂuences on bone
growth including body composition, physical activity, and diet due to the potential
interaction of all of these determinants of bone health. Body composition may inﬂuence
bone mineralization, especially in overweight children, who represent an increasing
proportion of the pediatric population (Goulding et al., 2000; Goulding et al., 2001).
Children with higher levels of physical activity have higher measures of bone mineral
mineralization (Slemeda et al., 1994; Bailey et al., 1999; Iuliano-Burns et al., 2003;
Specker and Binkley, 2003; Zanker et al., 2003). Calcium intake is acknowledged to be
associated with development of peak bone mass (Weaver, 2000). During attainment of
peak bone mass, intakes of less than 1 g/d of Ca are associated with lower bone density
(Weaver, 2000). In spite of the importance of Ca in bone mineralization, evidence
related to the long-term effects of Ca supplementation in childhood suggests that bone
density gains associated with supplementation are not generally sustained (Abrams,
2005). Other diet components including phosphorus, Vitamin D and protein play
important roles in bone mineralization (Ginty, 2003; Anderson et al., 2006; Norman and
Henry, 2006). High protein intake is hypothesized to have a negative impact in bone

mineralization (Frassetto et al., 1998). Thus, dietary intake of several key bone-building

49

nutrients should be considered when investigating the interrelationships of determinants
of bone mineralization in early childhood.

This study, the Healthy Kids Nutrition Study (HKNS), was undertaken to further
clarify associations among diet, physical activity, and body composition with bone
growth and mineralization in a group of children between ﬁve-years of age and puberty.
The HKN S is unique in its simultaneous consideration of key variables of bone
mineralization. Understanding these associations early in life may contribute to the
ability to identify children who are “at-risk” in terms of progress toward attaining peak
bone mass. Early intervention could decrease their risk of developing osteoporosis later
in life.

SUBJECTS AND METHODS
Subjects

Fifty-two subjects from the greater Lansing, MI, area were enrolled in the study.
The children were between ﬁve-years of age and puberty (<Tanner l), > the 5th percentile
BMI by NCHS standards, and 2 37 weeks gestational age at birth. They were free living,
included both genders and were generally well, without any chronic disease conditions or
birth defects other than possible allergies or food sensitivities. Children with minor
conditions that have no direct relationship to bone growth were included after review by
the study physician. Pubertal status was determined by a thorough parent/child interview
to ascertain Tanner Stage I status. Although recruitment by race was not targeted, all
children were at least 50% Caucasian. The study was approved by full review of the
University Committee on Research Involving Human Subjects at Michigan State

University and met conﬁdentiality and HIPAA requirements.

50

Diet analysis

Diet information was collected in two ways, with a diet recall and with a food
frequency questionnaire (FFQ). An experienced Registered Dietitian obtained a usual,
one-day diet recall via a consensus interview with a parent and the subject. The parent,
along with the child, completed the Youth and Adolescent Food Frequency Questionnaire
developed by Harvard Channing Laboratory, (Harvard University, School of Public
Health, Boston, MA, US). Ca-fortiﬁed juice was added to this FFQ via individual coding
by the investigator. The nutrient content of diets of the children was assessed by both the
recall and the FFQ, providing estimates of over sixteen nutrients. The recall analysis also
estimated intake according to the Dietary Guideline’s food groups (CNPP, 2005). The
same Registered Dietitian conducted the diet interviews, coded the FFQ for analysis, and
entered the diet recall information for analysis using the nutrition database program
within the MyPyramid Tracker, based on the USDA food database (CNPP, 2005). The
coded FF Q was sent to Harvard Charming Laboratory for analysis. All records, if
submitted incompletely, received individual follow-up by study personnel. Net
endogenous acid production of the subjects’ diets was estimated using the renal net acid
excretion (RNAE) equation developed by Frassetto et al. (1998).
Anthropometrics

Height was measured to the nearest centimeter with a stadiometer (Invicta
Plastics, Oadby, Leicester, England). Weight was determined to the nearest 0.1 kg on a
digital scale (Seca, Model 882, Hanover, Maryland, USA). Standard pediatric

measurement protocols to assure precision and accuracy were employed (Gibson, 1990;

51

Lee and Nieman, 1996) and followed the National Center for Health Statistics guidelines
for standard measurement (N CHS, 1996).
Measurement of Physical Activity

The Actical accelerometer (MiniMitter, Bend, OR) was used to measure physical
activity. Actical, a compact, battery-operated physical activity monitor with physical
characteristics similar to a small wristwatch, was worn at the waist on a neoprene belt.
This accelerometer has been validated for use in estimating energy expenditure in young
children (Pfeiffer etal., 2006). The sampling frequency was 32 Hz and an epoch
measurement of 15 seconds was chosen. Subjects wore the monitor continuously for
three consecutive days (one weekend day and two weekdays). The Actical program
provided the estimate of physical activity as kcals daily total energy expenditure (DTEE)
above basal energy expenditure (BEE).
Serum Vitamin D measurement

Because of the critical role of Vitamin D in bone mineralization and Ca
metabolism, serum Vitamin D was assessed in all subjects. A phlebotomist with
experience in drawing blood from children collected one, nonfasting, ﬁve-ml blood
sample for serum 25(OH) Vitamin D3. This single serum 25(OH) Vitamin D3 sample per
subject was measured at the accredited analytical laboratory, LabCorp, (Dublin, OH) via
an immunochemiluminometric assay (ICMA). The reference standards provided by the
laboratory vary with age and season due to seasonal variability of sun exposure (range

7.0 ng/mL to 48.0 ng/mL).

52

DXA

Whole body measures were taken using a GE LUNAR Prodigy machine at
Fiechtner Research (Lansing, MI) with pediatric soﬁware speciﬁc to the machine
(software version 6.81; General Electric Lunar Corporation, Madison, WI).
Measurement protocols speciﬁc to the GE-LUNAR Prodigy densitometer were
employed, including daily calibration with a standard calibration block and calibration
tri-weekly with a body phantom. The CV of this machine is reported to be < 1.0%
(personal communication, Fiechtner Research). Measures obtained included body
composition (kg fat mass and kg lean mass), BMD (gm/cmz) and BMC (gm). The full
body scan took approximately two minutes and exposed each child to 0.04 mRem of
radiation. Motion artifacts were not a problem in this population.
Statistical Analysis

The BMD reference values from GE Lunar DXA program were used in power
calculations (Wacker and Barden, 2001). The mean BMD for the seven-year-old male
was used as the mean BMD of the population and a SD of 0.07 was used to calculate
power. Recruitment of 52 subjects assured the sample size necessary to detect
differences at a 95% Conﬁdence Interval.

Descriptive statistics were employed to describe the study population with values
reported as means i SDS. The primary dependent variables are BMD, BMC, and the
Lunar DXA reported Z-score. The independent variables are intake of nutrients in the
diet closely associated with bone growth as assessed by recall and by FFQ, diet as

assessed by the USDA Dietary Guidelines, serum 25(OH) Vitamin D3, measures of body

53

composition, and physical activity expressed as DTEE above BEE. A two-tailed t-test
was employed to determine signiﬁcant differences between genders in all variables.
Relationships between variables were explored with 2-tailed Pearson’s correlations.
Multiple regressions with stepwise variable entry were used to determine variables that
may be signiﬁcant predictors of BMD and BMC. Data were analyzed using SPSS,
version 13, Base System (SPSS, Inc., Chicago, IL, USA). The signiﬁcance level for all
tests was established as p<0.05 and all tests are bi-directional.
RESULTS

Descriptive characteristics of the study subjects are summarized by age in Table
3.1 by gender and in total. The mean age of the subjects was 7.6 years (range 5.1 — 11.9).
Twenty—seven of the subjects were male; twenty-ﬁve were female. BMIs ranged from
13.7 to 23.9 with a mean BMI of 16.8. Z-scores for the BMD by Lunar DXA ranged
from negative 1.2 to positive 1.8. The mean Z-score was 0.14 with 30.8 percent of these
subjects having Z-scores at or below negative 0.2. There were no gender differences in
mean fat mass, lean mass, BMC or BMD (t-test, p S 0.05). Mean serum 25(OH) Vitamin
D3 was 30.6 ng/mL i 10.2 (range 10.8 — 62.3). No individual had a Vitamin D3 outside
of the normal range.
Diet

Dietary intakes according to the 2005 US Dietary Guidelines are summarized in
Table 3.2 by gender and for the total. Mean daily grain intake was 8.3 i 3.7-ounce
equivalents: Mean milk intake was 3.4 i 1.8-cup equivalents. Mean meat and bean

intake was 3.7 i 1.9-ounce equivalents. Fruit and vegetable intakes were combined for

analysis. Combined mean intake of fruits and vegetables was 2.2 i 1.3 cup-equivalents.

54

Table 3.1

Physical characteristics, measures of bone mineralization, energy expenditure,
and serum 25(OH) Vitamin D3 by gender and in total

 

 

 

. Gender

Varrable

Male (n=27) Female (n=25) Total (n=52)

Age Mean 3. so 7.4 a: 1.4 7.8 a 1.9 7.6 a 1.7

Range 5.6-11.1 5.1-11.9 5.1-11.9

BMI Mean a so 17.3 a 2.5 16.3 a 1.9 16.8 a 2.3

Range 14.7 - 23.9 13.7 - 20.6 13.7 - 23.9

Weight (kg) Mean 3 so 26.3 :1: 5.9 27.5 a 7.6 26.9 a 6.8

Range 17.5 - 43.7 18.2 - 47.0 17.5 - 47.0

Height (cm) Mean 3 so 128.4 3: 13.5 124.8 a 9.9 126.5 a 11.8

Range 105.2 - 168.5 107.25 - 141.0 105.2 - 168.5

Fat mass (kg) Mean a so 5.95 a 4.78 6.00 :1: 3.41 5.97 a 4.14

Range 2.09 - 18.77 2.17 - 16.42 2.09 - 18.77

DTEE above Mean 3: so 609 3. 225 512 a 200 562 a 217

BEE (kcals) 2 Range 323 — 1120 281 - 973 281 — 1120

BMD (g/emz) Mean 4 so 0.885 a 0.062 0.837 a 0.060 0.847 3: 0.061

Range 0.764 - 1.003 0.751 - 0.980 0.751 - 1.003

Z-seore ‘ Mean 1 so 0.31 :1: 0.64 -004 a 0.59 0.14 a 0.64

(Lunar) Range -O.60 - 1.80 -120 - 1.20 -120 - 1.80

BMC (8) Mean a so 960 a 262 933 a 229 947 :1: 245

Range 618—1657 567- 1514 567—1657

Serum 25-OH, Mean :1: so 30.7 a 9.4 30.3 a 11.3 30.6 a 10.2

V“ D (“g/m” 3 Range 13.8 - 57.5 10.8 - 62.3 10.8 - 62.3

 

‘ Signiﬁcant difference between male and female subjects per t-test (p_<_0.05)
2 For this variable, n=26 (males), n=25 (females), n=51 (total)
3 For this variable, n=26 (males), n=23 (females), n=49 (total)
BMI: Body Mass Index
DTEE above BEE: Daily Total Energy Expenditure above Basal Energy Expenditure
BMD: Bone Mineral Density
BMC: Bone Mineral Content

55

Table 3.2

Daily dietary intakes according to 2005 US. Dietary Guidelines by gender1

 

 

 

Variable Gender
Male (n=27) Female (n=25) Total (n=52)
Grain intake2 Mean 4 so 8.9 a 4.1 7.7 a 3.1 8.3 a 3.7
Range 2.8 - 22.7 0.8 - 14.1 0.8 - 22.7
Milk intake3 Mean :1: so 3.2 a 1.9 3.6 a 1.7 3.4 a 1.8
Range 0.2 - 6.4 1.0 - 7.0 0.2 - 7.0
Meat and bean intake2 Mean a so 3.9 a 2.0 3.4 a 1.8 3.7 a: 1.9
Range 0.7 - 9.4 0.1 - 6.4 0.1 - 9.4
Fruit and Vegetable Mean :t SD 2.3 :t 1.4 2.00 d: 1.3 2.2 :1: 1.3
intaké Range 0.5 - 5.5 0.1 - 4.1 0.1 - 5.5

 

12005 US. Dietary Guidelines recommendations for children ages 2 to 8 years:

Grain, 6 02.; Milk, 2 cups; Meat and beans, 5 02.; Fruit, 1.5 cups; and Vegetable 2.5 cups.
2 In oz. equivalents
3 In cup equivalents

56

There were no differences in intake of fruits and vegetables by gender. Meat and bean
intake, at 3.7-ounce equivalents, was lower than the recommended 5-ounce equivalents;
however, there was a very broad range of intakes between 0.1-ounce equivalents and 9.4-
ounce equivalents. Total intake of fruits and vegetables in this group of children, at 2.2
cup-equivalents for both fruits and vegetables combined, was lower than the
recommended amount of 1.5 cups of fruits and 2.5 cups of vegetables (a total of 4 cup-
equivalents of fruits and vegetables per day).

Nutrient intakes as measured by usual, one-day recall are summarized by gender
and for the total in Table 3.3. Intakes according to recall do not include supplements.
There was a signiﬁcant difference between males and females only for ﬁber (t-test, p S
0.05) with males consuming more ﬁber. Mean caloric intake by recall was 2180 i 545
(range 1123 — 3152). Mean protein intake by recall was 79 g i 21 (range 37 — 124),
representing a mean protein intake of 2.94 g/kg body weight. Total mean intake of Ca by
recall was 1382.4 mg i 580.9 (range 252.9 — 2715.0).

Nutrient intakes as measured by FF Q are summarized in Table 3.4. The FFQ
reported intakes both with and without supplements. The FFQ asks whether or not
vitamins are used, how often, and for how long. There were no signiﬁcant differences in
intake of any nutrient by gender as measured by FFQ. Fifty-six percent of the subjects
reported supplement use. Of those 56%, 62 °/o report taking supplements for less than 4
years and fewer than 5 times a week. Mean caloric intake by FFQ was 2281 i 474 (range
1265 - 3353). Mean protein intake by FFQ was 95 g i 21 (range 37 - 137), representing

a mean protein intake of 3.5 g/kg body weight. Total mean intake of Ca without

57

Table 3.3

Daily nutrient intake2 as measured by usual, one-day dietary recall by gender and in total

 

 

 

Variable Gender
Male (n=27) Female (n=25) Total (n=52)
Total kcals Mean3so 22193507 2,1383 591 2.1803545
Range 1123-3152 1167-3115 1123-3152
Protein (g) Mean 3 so 79 3 22 79 3 20 79 3 21
Range 37 - 124 48 - 113 37 -124
Cholesterol Mean 3 SD 215 3 136 201 3 113 208 3 124
(mg) Range 25 - 607 64 - 539 25 - 607
Carbohydrate Mean :1: SD 308 i 72 294 :1: 98 301 :t 85
(g) Range 132-432 102-511 102-511
Fiber (g)1 Mean3 so 2038 153 5 1837
Range 6-41 2-28 2-41
Fat (g) Mean 3 so 79.2 3 28.2 75.9 3 25.4 77.6 3 267
Range 31.5 - 171.3 34.3 - 127.4 31.5 -1713
Vit A (meg Mean 3 so 966 3 491 1039 3 790 1001 3 647
RAE) Range 271 - 1753 152 - 4359 152 - 4359
v11 (3 (mg) Mean 3 so 115 3 86 93 3 62 104 3 76
Range 15-351 4-265 4-351
Vit E (mg) Mean 3 so 7.0 3 3.7 5.7 3 3.3 6.4 3 3.6
Range 2.0 -21.1 2.1 — 17.2 2 - 21.1
Ca (mg) Mean 3 so 1344 3 668 1424 3 479 1382 3 581
Range 253 - 2715 790 - 2505 253 — 2715
P (mg) Mean 3 so 1584 3 505 1565 3 442 1575 3 471
Range 645 - 2766 1001 - 2531 645 — 2766
Mg (mg) Mean 3 so 327.4 3 97.4 296.1 3 85.0 312.4 3 92.1
Range 104.5 - 539.2 158.4 - 486.2 104.5 - 539.2
Fe (mg) Mean 3 so 18.8 3 8.9 16.7 3 6.8 17.2 3 8.0
Range 5.8 - 48.6 5.1 - 32.1 5.1 - 48.6
Zn (mg) Mean3 so 13.1352 11.6333 12434.4
Range 3.9 - 28.7 6.8 - 18.4 3.9 - 28.7
K (mg) Mean 3 so 2865 3 934 2,705 3 911 2788 3 917
Range 1087 - 5199 1558 - 5083 1087 - 5199
Se (meg) Mean 3 so 105.5 3 40.9 95.0 3 30.6 100.4 3 36.3
Range 36.3 - 215.0 47.8 - 188.8 36.3 - 215.0

 

1 Signiﬁcant difference between male and female subjects per t-test (p50.05)
2 Vitamin D not reported via one-day recall (not available in database)

58

Table 3.4

Daily nutrient intake2 as measured by food frequency questionnaire1

with and without supplements by gender and in total

 

 

 

Variable Gender
Male (n=27) Female (n=25) Total (n=52)
Total kcals Mean 3 so 2250 3 484 2315 3 472 2281 3 474
Range 1265 - 3353 1374 - 3327 1265 - 3353
Protein (g) Mean i SD 93 d: 22 97 :h 20 95 i 21
Range 37 - 126 45 - 137 37 -137
Animal protein in Mean 3 SD 62 :t 17 67 3 17 64 3 17
diet (8) Range 25 - 98 25 - 106 25 - 106
Non-animal Mean 3 so 30 3 11 30 3 8 30 3 10
protein in diet (g) Range 12 - 64 14 — 47 12 - 64
Cholesterol (g) Mean 3 SD 240 d: 55 257 a: 85 248 i 71
Range 144 - 353 91 - 388 94 - 388
Carbohydrate (g) Mean 3 SD 309 :t 82 314 3 68 311 3 75
Range 157 - 460 185 - 461 157 - 461
Fiber (g) Mean 3 so 20.3 3 7.3 19.5 3 6.0 19.9 3 6.6
Range 7.7 - 36.9 9.5 - 35.5 7.7 - 36.9
Fat (g) Mean 3 so 73.4 3 15.8 78.2 3 20.3 75.7 3 18.0
Range 44.0 - 118.8 44.7 - 110.2 44.0 - 118.8
K(mg) Mean3so 31273927 32193701 31713819
Range 1060 - 4785 1411 - 4864 1060 - 4864
Vit A With Mean 3 so 8399 3 5179 7909 3 3038 8163 3 4251
<ng supp Range 2045 - 24575 3578 - 13120 2045 - 24575
RAE) Without Mean3 so 10243465 10183353 10213411
supp Range 246 - 2246 571 - 1960 246 - 2246
VitC With Mean3so 130371 129348 129360
(mg) supp Range 35 - 322 32 — 223 32 - 313
Without Mean3so 114360 113341 113351
supp Range 35 - 262 26 - 183 26 - 262
Vito With Mean3 so 3923189 413 3156 4023172
(1U) suop Range 45 - 708 48 - 863 45 - 863
Without Mean3so 3103159 3323116 3203139
supp Range 45 - 568 48 - 490 45 - 568
Vit E With Mean 3 so 8.4 3 3.5 8.4 3 2.5 8.4 3 3.0
(mg) suop Range 3.3 - 16.3 3.8 - 14.0 3.3 - 16.3
Without Mean :1: SD 6.5 i 1.7 6.6 i 1.7 6.6 :1: 1.7
Shop Range 3.3 - 10.0 3.2 - 10.0 3.2 - 10.0

Table continued - next page

59

Table 3.4 - continued

Ca (mg) With Mean 3 so 1497 3 500 1503 3 327 1500 3 422
supp Range 462 — 2577 790 - 1978 426 - 2577
Without Mean 3 so 1464 3 496 1471 3 325 1467 3 419
sopp Range 426 - 2,532 722 - 1,978 426 - 2532
P (mg) With Mean 3 so 1742 3 443 1823 3 374 1781 3 409
supp Range 725 - 2559 829 - 2576 725 - 2576
Without Mean d: SD 1742 i 443 1823 i 374 1781 3: 409
SUpp Range 725 - 2559 828 - 2576 725 - 2576
Mg With Mean 3 so 329.2 3 91.3 338.2 3 74.2 333.5 3 82.8
(mg) supp Range 125.4 - 476.1 186.6 - 531.1 125.4 - 531.1
Without Mean 3 so 321.0 3 86.0 330.4 3 73.4 325.5 3 79.5
SUpp Range 125.4 - 473.3 166.6 - 511.1 125.4 - 511.1
Fe (mg) With Mean 3 so 20.2 3 8.3 20.7 3 9.0 20.4 3 8.5
sopp Range 7.2 — 38.3 10.2 - 54.7 7.2 - 54.7
Without Mean 3 so 16.1 3 4.9 16.6 3 6.5 16.3 3 5.7
supp Range 7.2 - 29.9 10.2 - 44.7 7.2 - 44.7
Zn (mg) With Mean 3 so 16.7 3 6.1 16.8 3 5.0 16.7 3 5.5
supp Range 6.7 - 30.0 8.2 - 29.9 6.7 - 30.0
Without Mean3 so 13.4338 13.6327 13533.3
supp Range 6.6 - 26.1 8.2 - 18.9 6.6 - 26.1

 

' Youth and Adolescent Food Frequency Questionnaire, Harvard Channing Laboratory, Harvard University,

School of Public Health, Boston, MA, USA
2 Vitamin K and selenium not reported (not available in database).

60

supplements by F FQ was 1467 mg i 419 (range 426 — 2532). The Adequate Intake (A1)
of Ca is 800 mg/day (FNB/IOM, 1997).

Analysis of dietary intake by FFQ also provides insight into the food sources of
nutrients in the diets. Single item dairy foods (frozen dairy products, cheese, yogurt,
ﬂuid milks) made up 21 % of total caloric intake, 29.7 % of the total contribution to
protein intake, 61.8 % of the contribution to Ca intake, 44.1 % of the contribution to P
intake and 82.3 % of the contribution to Vitamin D intake (excluding supplement use).
After dairy foods, the next highest contributors to caloric intake were peanut butter and
jelly sandwiches (5.4%), chicken (4.4%), and cold cereal (3.9%). After dairy foods, the
next highest contributors to protein intake were chicken (12.4%), peanut butter and jelly
sandwiches (4.2%), and beef (3.9%). The next highest contributors to P intake, after
dairy foods, were chicken (5.33%), peanut butter and jelly sandwiches (3.1 1%), and cold
cereal (3.02%).

Nutrient intake can be expressed by nutrient adequacy ratios (NAR), a calculation
which is an index of adequacy of a speciﬁc nutrient in the diet relative to an established
standard need (Gibson, 1990). In this research, the most current US Dietary Reference
Intakes (DRI) (FNB/IOM, 2005; FNB/IOM, 1997) were used as the standard. In order to
evaluate dietary intakes as measured by recall and intakes as measured by F FQ without
supplements, NAR values were calculated for nutrients that have been shown to have a
strong association with bone growth (Figure 3.1). At a calculated NAR of greater than or
equal to 1.0, intake can be considered to be adequate (Gibson, 1990). Both measures of
diet assessment result in nutrient adequacy ratios of greater than 1.5 for protein, Vitamin

A, Vitamin C, Ca, P, Mg, iron and zinc. The mean NAR for Vitamin E are 0.91 by recall

61

 

 

M 3552.53 o? 0.: 3 ”22113383 3 «<2 E

 

3:05: Z

 

<0 m => 0 =>

 

a.

  

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

$56233 3233 Our,— 3 28 :88 .3 @2338 mm 3:053: 828 m8 922V «.038 momsvova 305:: 502

2 Ram;

62

and 0.94 by PF Q. The mean NAR for potassium (K) are 0.73 per recall and 0.83 per
FFQ. The US DRI for protein in early childhood (ages 4 to 13 years) is 0.95 g/kg body
weight (FNB/IOM, 2005). Protein intake in this subject population was 3.1 times greater
than the DR] per recall and 3.7 times greater than the DRI per FFQ. Food frequencies
tend to result in higher estimates of intake than do recalls, with both methods tending to
result in over reporting in populations with lower intakes, populations that would include
children (Thompson and Byers, 1994). Given that NAR for 10 nutrients in Figure 3.1
were similar for both FFQ and recall, and given that both methods of assessment tend to
over report intake, the intake as measured by diet recall was used in subsequent data
analysis because the recall provided slightly lower estimates of intake.
Physical Activity

In exploring the relationships of the measured variables inﬂuencing bone
mineralization with the dependent variables of BMD and BMC, several signiﬁcant
correlations were found as summarized in Table 3.5. Daily total energy expenditure
above calculated basal energy expenditure (BEE) (Harris and Benedict, 1919) was 562 i
217 kcals (range 281 — 1 120) (Table 3.1). Daily total energy expenditure above BEE had
a strong, significant positive relationship to both BMD (r=0.460, p30.01) and BMC
(r=0.591, pS0.01). Likewise, percent body fat had a strong, signiﬁcant positive
relationship to both BMD (r=0.535, pS0.01) and BMC (r=0.646, pS0.01). Both protein
and kilocalories, expressed relative to body weight, had strong, signiﬁcant negative
relationships to both BMD and BMC (for protein and BMD, r=-0.508, p500] , 2-tailed,

for protein and BMC r=-0.564, p500], 2-tailed, for kcals and BMD, r=-0.510, ps0.01, 2-

tailed, for kcals and BMC r=-0.578, pS0.01, 2-tailed). In order to evaluate the wide range

63

Correlations of BMD and BMC with variables inﬂuencing bone mineralization

Table 3.5

 

 

Pearson Correlation (n=52) BMD (g/cmz) BMC (g)
% body fat 0,535M 0.646”
DTEE above BEEl (kcals) 0.460** 0591”
Protein by weight2 (g/kg) -O.508** -0.564**
kcals by weight2 (kcals/kg) -0510” -0,578**
Ca by height2 [(mg/d)/cm] -()_277* -0260
p by height2 [(mg/d)/cm] 4,378.... -0.348*
Mg (mg)2 -0135 -O.126
K (mg)2 -0137 -0.069
Fruit and Vegetable intaltez-3 0,265 0,256
RNAE 0.196 0.247

 

** Pearson correlation signiﬁcant at the 0.01 level (2-tailed).

*Pearson correlation signiﬁcant at the 0.05 level (2-tailed).

' For this variable, n = 51

2 All nutrient intakes as assessed by recall.

3 Intake per 2005 US Dietary Guidelines in cup equivalents.

BMD: Bone Mineral Density

BMC: Bone Mineral Content

DTEE above BEE: Daily total energy expenditure above basal energy
expenditure per Harris-Benedict equation estimate (Harris JA, Benedict FG. A
biometric study of basal metabolism in man. (Publication No. 279) Washington,
DC: Carnegie Institute of Washington; 1919).

RNAE: Renal Net Acid Excretion as estimated by the method of Frassetto LA et
a1. Estimation of net endogenous noncarbonic acid production in humans from
diet potassium and protein contents. AJCN 1998; 68:576-83.

64

of Ca intakes independent of age or body size, both of which also ranged widely, intake
was expressed by height (mg Ca intake per day/cm height). A signiﬁcant negative
association was found between Ca by height and BMD (r=-O.277, pS0.05, 2-tailed).
There was no correlation between Ca intake by height and BMC. Phosphorus intake by
height (mg per day/cm height) was also negatively correlated to BMD (r=-0.3 78, pS0.0S,
2-tailed) and to BMC (r=-0.348, pS0.05, 2-tailed). No signiﬁcant correlations were
found with other nutrients or food groups or with total Ca intake not expressed by height.
Due to the strong negative correlation of protein to BMD and BMC, renal net acid
excretion was estimated using the method of Frassetto et al. (1998). No signiﬁcant
correlation was found between RNAE, dietary potassium or fruit and vegetable
consumption and BMD or BMC (Table 3.5).

Table 3.6 and Table 3.7 report ﬁnal multiple regression analysis predicting BMD
and BMC, respectively, from measured variables inﬂuencing bone growth in this
population of prepubertal children. Independent variables found to have a signiﬁcant
correlation and those reported to have a strong association to bone grth in the literature
(DTEE, % body fat, protein intake /kg body weight, kcals intake/kg body weight, Ca
intake/cm height, P intake/cm height, Mg intake, K intake, ﬁ'uit and vegetable intake)
were entered in a stepwise manner. The ﬁnal model for predicting BMD (Table 3.6)
includes percent body fat, dietary intake of kcal by body weight, Ca and P by height, Mg
intake and fruit and vegetable intakes. In this model, caloric intake relative to weight and
P relative to height were the only signiﬁcant negative predictors of BMD. The associated
relationships of all of these variables explain 58.8 % of the variability in BMD. The ﬁnal

model for predicting BMC (Table 3.7) includes DTEE above BEE, percent body fat, kcal

65

Table 3.6

Multiple regression analysis of BMD by % body fat and intakes of kcals by weight,
Ca by height, P by height, Mg, and fruits and vegetables

 

 

 

BMD
B 3 SEE Sig.

Constant 0.824 :1: 0.035 0.000
% body fat 0.003 :1: 0.001 0.006
kcals by weight‘ -0.001 :t 0.000 0.003
Ca by height (mg/em)l 0.009 3 0.003 0.006
p by height (g/em)l -0017 3 0.005 0.002
Mg (mg)1 0.000 3 0.000 0.000
Fruit and vegetable intake1‘2 0.008 3 0.005 0.093

R2 = 58.80%
ANOVA 10,448 0.000

 

' All nutrient intakes as assessed by recall
2 Intake per 2005 US Dietary Guidelines in cup equiv.
BMD: Bone Mineral Density

Variables entered into the regression model (stepwise): protein intake by weight, calcium
intake by height, DTEE above BEE, % body fat, P intake by height, daily total Mg, fruit and
vegetable intake, kcal intake by weight

66

Table 3.7

Multiple regression analysis of BMC by % body fat, DTEE above BEE, intakes of kcals
by weight, Ca by height, P by height, Mg, and fruits and vegetables

 

 

 

BMC
B 3 SEE Sig.
Constant 704.8913 125.332 0.000
% body fat 9.389 3 3.104 0.004
DTEE above BEE 0.249 3 0.109 0.027
kcals by weight1 -4.235 :t 1.306 0.002
Ca by height (mg/em)1 26.973 3 10.854 0.017
P by height (g/em)l -47.822 3 17.744 0.010
Mg (mg)1 1.603 3 0.412 0.000
Fruit and vegetable 34.133 3 15.675 0.035
intake"2
R2 = 72.90%
ANOVA 16,521 0.000

 

' All nutrient intakes as assessed by recall

2 Intake per 2005 US Dietary Guidelines in cup equiv.

BMC: Bone Mineral Content

DTEE above BEE: Daily Total Energy Expenditure above Basal Energy
Expenditure per Harris-Benedict equation estimate (Harris JA, Benedict PC. A
biometric study of basal metabolism in man. (Publication No. 279) Washington,
DC: Carnegie Institute of Washington; 1919).

Variables entered into the regression model (stepwise): protein intake by weight,

calcium intake by height, DTEE above BEE, % body fat, P intake by height,
daily total Mg, fruit and vegetable intake, kcal intake by weight

67

intake by weight, Ca and P intake by height, Mg intake and fruit and vegetable intake,
with the association among these variables explaining 72.9% of the variability in BMC.
Caloric intake relative to weight and P intake relative to height were negative predictors
of BMC. In that the B coefﬁcient is indicative of the relative importance of an
independent variable in contributing to the dependent variable in a given prediction
model, Ca intake by height (B=0.009 i 0.003, p=0.006), P intake by height (B=-0.017 i
0.005, p=0.002), and fruit and vegetable intake (B=0.008 i 0.005, p=0.093) are the
strongest predictors of BMD. Calcium intake by height (B=26.973 i 10.854, p=0.017), P
intake by height (B=-47.822 i 17.744, p=0.010), and fruit and vegetable intake
(B=34.133 i 15.675, p=0.035) are also the strongest predictors of BMC. The correlation
between BMC and BMD and their unstandardized values as predicted by each model is
illustrated in Figure 3.2 and Figure 3.3. Gender was entered into each model and did not
inﬂuence the outcome of the ﬁnal model.
DISCUSSION

Associations between diet, physical activity, body composition, and bone growth
and mineralization are well documented for special populations such as the elderly and
adolescents but studies looking exclusively at early childhood are limited. This study
conﬁrms these associations in a population of healthy, prepubertal, Caucasian children.
Much of the variability in bone mineralization in this group of prepubertal children
between 5 to 11 years of age can be explained by modiﬁable inﬂuences on bone
mineralization. This group of children all had BMD Z-scores above minus 1.2. However

nearly a third of the children had Z-scores below minus 0.2. By deﬁnition. BMD Z-

68

Figure 3.2

Correlation between actual BMD and predicted value of BMD based on % body fat,
intakes of kcals by weight, Ca by height, P by height, Mg, and fruits and vegetables

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

1.05 ~
1 - O 3 g___
0
0.95 m o
. I
. O z/
0 . 1”
0_9 a. -. -_ 9 O ’/’ __ #_
9 ,’ 0
g x’ t 2
co ° x . R = 0.4361
0 , O
035 9 ‘ 1’; ________3-3. , 37‘, ..
/ v
v’li . O
o O
. 1" O 3
’ O
0.8 . 1’ 034 ° _ ,_ __
’4’ o 9
If 0 .
’ O
/ O .
I O . 1
0.75 T—~—-—~ ~ 9 —-.-.._... . -_ --. .
0.7 a , r , , ,
0.7 0.75 0.8 0.85 0.9 0.95 I

Predicted Value

 

 

69

 

 

 

 

Figure 3.3

Correlation between actual BMC and predicted value of BMC based on % body fat,

DTEE above BEE, intakes of kcals by weight, Ca by height, P by height,

 

 

 

 

 

 

 

 

 

 

 

 

 

Mg, and ﬁ'uits and vegetables
1700 _ g ,
O
1500«~———- *3" ~ — ——— ’ “—3
I” O
O O [I]
1300 ._ 0 ’1 -
0 II]
o o a.
z 1100 — 33 A 9 3,-9 —’--—~-’————- -— 3
c: /
o 3’ 2 _
,, [gar .. R —0.6829
0 /’ ’
900 — — 17,33;— .__ .393— 3 — m — —— —- 3
,8”
3’ 9 9
700 ._ _ _ .._... /,_f, --_-._ 3-3.; _ _ 3.3—
. ,z’ 0 ° 0
O ’9’, .
/// O
500 . . . . .
500 600 700 800 900 1000 l 100 l 200 l 300 1400 l 500

Predicted Value

 

70

scores represent the central range of BMDs in a healthy reference population. The
assumption is that, in the absence of disease, this central range is a desirable range.
Because BMD tends to track throughout life (Ferretti et a1, 1998; Heaney et al., 2000),
the third of these children whose Z-scores are approaching negative 1.0 in early
childhood, may already be at increased risk of fractures and osteoporosis later in life. In
other words, progress toward accrual of peak bone mass may already be slowed in
children with negative Z-scores for BMD during their early childhood years.

Mean intakes of the 16 nutrients measured in this group of children met or
exceeded the US DRI. The recommendations for intakes of nutrients in children are
based on age rather than physiologic maturity or body size. Growth rates in children
occur in four general stages, made apparent by the varying slopes of pediatric growth
curves (CDC, 2003) and deﬁned chronologically. Growth rates are highest in periods of
birth to 3 years (infancy and toddlerhood) and then again in the period of puberty and
adolescence (ages 9 to 18 years, where on an individual basis there is tremendous
variation). The intermediate years, or early childhood years of 4 to 8 years, are a time of
slower growth. On the other hand, physiologic grth and maturation in adolescence is
followed through Tanner staging, a classiﬁcation of stages of sexual development that is
aligned with skeletal growth rates (Tanner, 195 5). Nutrient intake recommendations have
been set based on these general categories of pediatric growth as well (FNB/IOM, 1997;
FNB/IOM, 2005). Likewise, standards for bone mineral measurements are based on
chronological age (Wacker and Barden, 2001). As such, both nutrient requirements and
bone growth standards have a chronological rather than a physiological basis. By

limiting the HKNS subjects to Tanner stage 1, any associations conﬁrmed in this study

71

are more likely due to physiologic processes other than puberty regardless of age or body
size.

The subjects in this study ranged in age between 5 and 11 years of age, ranged in
weight between 17 and 47 kg and ranged in height between 105 and 169 cm. These 52
children had great variation in body size, and hence bone size, at any given age. Given
the large range in body sizes, variables related to bone mineralization could be expressed
relative to height allowing for consideration of variability of body size in children at any
given age. The Ca and P intake did not show a relationship to BMD or BMC until they
were adjusted for height, after which signiﬁcant correlations were found. Differences
attributable to gender were not seen in this group of subjects though gender differences
have been found in some studies of children in this age range (Specker et al., 1987; Bell
et al., 1991; Bounds et al, 2005) but not in others (Fassler and Bonjour, 1995; Maynard et
al., 1998; Horlick et al., 2000; Nguyen et al., 2001).

This group of children can be considered well-nourished with mean intakes of
most nutrients having a NAR of greater than one. Calcium intake in this group was at or
above the Adequate Intake (A1) of 800 mg/day for most of the subjects, with 69% of
subjects consuming greater than 1050 mg Ca/day by recall or greater than 1300 mg
Ca/day by FFQ. Only eight children consumed less than 800 mg Ca/day. Children in the
HKN S differ from national samples that predict less than 40% of children in the US
between 6 and 11 years of age achieve the A1 for Ca (Greer and Krebs, 2006). Protein
intake was 3 to 4 times higher than the DRI of 0.95 g/kg body weight, and Vitamin C was
4.5 times the DRI (FNB, IOM, 1997). The high protein intake assessed by recall was

inconsistent with the low meat and bean group intake, assessed by the Dietary Guidelines

72

as lower than recommended (CNPP, USDA-USDHHS, 2007). This inconsistency could
have been due to the large overall range of intakes and the contribution of grains and
dairy foods to the total protein intake. Given the adequacy of these diets on a nutrient-
by-nutrient basis, the inﬂuence that diet has on bone appears to relate to the relationship
of these nutrients to each other when considered as a group as well as their subsequent
interaction with bone mineralization and with other modiﬁable inﬂuences such as
physical activity. On an individual basis, intake of protein, kcals, Ca, P, Mg, and K had
an inverse relationship to BMD and BMC. However, when the association of these
nutrients, was expressed relative to body and bone size, using multivariate regression
analysis, only intakes of kcals, Ca, P, and Mg remained in the model as signiﬁcant
dietary predictors of BMD and BMC. In the prediction model, intakes of kcals/kg body
weight, Ca/cm height, and P/cm height have an inverse relationship to BMD and BMC
while Mg intake had a positive relationship.

Protein’s negative correlation to bone mineralization has been hypothesized to be
related to the effect of a chronic, low grade metabolic acidosis from the high levels of
sulfur containing amino acids and ammonium from protein metabolism which place a
hydrogen burden on the system and result in an increase in urinary Ca (Frassetto et al.,
1998; Ginty, 2003; Massey, 2003; New, 2003). Grains have high concentrations of these
amino acids as well and this group of subjects consumed an average of over 8 cups of
grain foods per day. To explore this association with protein further, renal net acid
excretion was estimated (Frassetto et al., 1998) but did not correlate to BMD or BMC in
these subjects. Potassium intake (primarily from foods in the fruit and vegetable group)

may offset the hypercalciuric effect of protein through contribution of base excess

73

(Massey, 2003; New, 2003). Though neither K intake per day nor fruit and vegetable
intake per day correlated with BMD or BMC, fruit and vegetable intake did have a
signiﬁcant association with BMD and BMC in the ﬁnal prediction models. This is
especially interesting because protein was not a signiﬁcant predictor in the ﬁnal model,
suggesting that other dietary factors may have had a mitigating inﬂuence on the negative
effect attributed to protein (New, 2003; Tylavsky et al., 2004; Vatanparast et al., 2005).

In addition to interrelationships among nutrients, body composition as percent
body fat was also associated with BMC and BMD, remaining in the ﬁnal model as a
signiﬁcant predictor of BMD and BMC. Percent body fat draws a more meaningful
relationship to measures of bone mineralization because some measures of lean tissue can
include bone and are less modiﬁable, in a practical sense, than is body fat. A higher
percent body fat predicts a higher BMD. However, the range of percent body fat within
which this association was seen is, itself, within a normal, healthy range of body
composition. In contrast, for this model, a low percent body fat contributes to predicting
a lower BMD and BMC. It appears that in this group of young children, all of whom
were well nourished, higher body fat within a normal range was beneﬁcial to bone
mineralization.

Physical activity is acknowledged to be beneﬁcial to bone health (Murphy and
Carroll, 2003). However the role physical activity plays in the accrual of peak bone mass
and its interactions with the other factors inﬂuencing bone growth in the early childhood
years is less clear. In general, children are less physically active now than in the past.
This decrease is estimated as a decline of approximately 600 kcals/day over the past 50

years (Boreham and Riddoch, 2001). The mean DTEE above BEE of 562 kcals (Table

74

3.1) is 27 % of mean daily total caloric intake (Table 3.3) in this population and was
signiﬁcantly correlated to BMD and BMC (Table 3.5). In adults as well as in animal
models, vigorous activity is an important contributor to bone mineralization (McMurray,
1995; Boreham and Riddoch, 2001). When energy expenditure was considered along
with other factors in bone mineralization in the prediction model, DTEE remained
signiﬁcant as a positive predictor of BMC but not of BMD in these subjects. Physical
activity must be considered in understanding the interrelationships among factors
predicting bone mineralization of children of this age given the strong positive
relationship between bone mineralization and energy expenditure through activity.

This group of children was selected to reﬂect a group of generally healthy
children living in the US Midwest. Predicting their BMD or BMC would involve, in part,
measurement of those modiﬁable inﬂuences of bone growth retained by the models
developed. Thirty percent of these children had Z-scores that identify them as having
BMD and BMC in what is technically deﬁned as the lower range of normal and at risk of
poor progress toward attainment of peak bone mass already in this early childhood
period. Based on the increasing incidence and prevalence of osteoporosis in older adults,
what is thought to be normal bone mineralization may actually be suboptimal. Bone
growth and mineralization exist on a continuum which, when shifted, changes the end
point of that continuum. In these early childhood years diet and lifestyle habits are
forming for a lifetime so knowing where to focus suggestions and interventions toward
bone-building lifestyles at this age may decrease the risk of developing osteoporosis later

in life as well as decrease fracture risk throughout life.

75

This study conﬁrmed several associations that support the importance of
considering as many of the inﬂuences on bone mineralization as possible in order to
assess adequacy of bone mineralization in prepubertal children. Characteristics of
children who tend to have lower measures of bone mineralization include high protein
intake, low percent body fat (within normal weight subjects), low levels of physical
activity, high calcium intake, and low intake of fruits and vegetables. These associations
suggest lifestyle adjustments that can inﬂuence BMD and BMC toward attainment of
peak bone mass. Health care professionals can provide meaningful suggestions for
identiﬁed “at risk” children that contribute to increasing bone mass and decreasing risk of

osteoporosis later in life.

76

CHAPTER FOUR

77

ff

 

CHAPTER FOUR
ABSTRACT
ASSOCIATION OF BODY COMPOSITION, DIET, PHYSICAL ACTIVITY, AND
BONE BIOMARKERS WITH MEASURES OF BONE MINERALIZATION IN
PREPUBERTAL CHILDREN
by
Marcia Kelly Scott

MICROABSTRACT: Prevention of osteoporosis begins in childhood with
attainment of peak bone mass. Body composition, diet, activity, and biomarkers of bone
metabolism were measured in 52 children. Better bone mass was associated with body
fat, physical activity, serum OC and Vitamin D, and urinary DPD, supporting their
consideration in assessing bone health in children.

INTRODUCTION: Osteoporosis is predicted to become a disease of epidemic
proportions within the next several decades. Attainment of higher bone mineral content
during the ﬁrst two decades of life decreases the risk of developing fractures later in life.
To ﬁirther clarify associations among determinants of bone growth, body composition,
bone mineral density (BMD) and bone mineral content (BMC), diet composition, bone
metabolism biomarkers, and physical activity levels were measured in prepubertal
children, mean age of 7.6 years.

METHODS: BMD, BMC, and body fat were measured via DEXA in 52 subjects.
Daily total energy expenditure (DTEE) was measured via accelerometer. Diet was

assessed with a usual, one-day recall interview. Serum OC, serum 25(OH) Vitamin D3

and urinary DPD were measured. Relationships between variables were explored with

78

two-tailed t-tests, two-tailed Pearson’s correlations and stepwise multiple regression
analysis (pS0.0I).

RESULTS: Bone mineralization correlates positively with percent body fat
(r=0.535 for BMD and r=0.646 for BMC) and with DTEE (r=0.460 for BMD and.
r=0.591 for BMC). Protein intake had signiﬁcant negative relationships to both BMD,
r=-0.508, and BMC, r=-0.564. A signiﬁcant negative association was found between Ca
intake and BMD (r=-0.277, pS0.05). The ﬁnal regression models for predicting BMD
and BMC include percent body fat, serum OC, urinary DPD, serum 25(OH) Vitamin D3,
and DTEE above BEE. In these models, OC and DPD were signiﬁcant negative
predictors of BMD and BMC while DTEE and percent body fat were positive predictors.
The relationships of these variables explain 58.9 % of the variability in BMD and 64.9%
of the variability in BMC.

CONCLUSIONS: This study afﬁrms the interrelationships of diet, physical
activity and body composition in contributing to development of peak bone mass. These
associations, along with biomarkers of bone metabolism, introduce the possibility that

children at risk of poor bone mineralization may be identiﬁed and treated early in life.

EXERCISE

BONE TURNOVER MARKERS
BODY COMPOSITION
CLINICAL/PEDIATRICS
NUTRITION

79

 

CHAPTER FOUR
ASSOCIATION OF BODY COMPOSITION, DIET, PHYSICAL ACTIVITY, AND
BONE BIOMARKERS WITH MEASURES OF BONE MINERALIZATION IN
PREPUBERTAL CHILDREN
INTRODUCTION

Osteoporosis is increasingly recognized as one of the major public health
problems facing aging individuals of both genders, worldwide. It is predicted to become
a disease of epidemic proportions within the next several decades (Riggs and Melton,
1995; WHO, 2003). Osteoporosis is broadly deﬁned as diminished bone mass and
reduced bone mineral density (BMD) at the level of 2.5 standard deviations below the
referent BMD of young adults (Riggs and Melton, 1995; USDHHS, 2000; WHO, 2003).
Osteoporosis presents clinically as fractures, both traumatic and nontraumatic in nature,
resulting in signiﬁcant morbidity, mortality, and health care costs (WHO, 2003). The
two most important ways to reduce risk of osteoporosis are to optimize bone mass while
maturing and minimize bone loss while aging.

In considering ways to decrease both the incidence and the severity of
osteoporosis, interventions for this disease target methods of prevention, not just in
adulthood but also during infancy, childhood, and adolescence. Though osteoporosis is
generally considered a geriatric disease, osteoporosis may actually be a pediatric disease
with geriatric consequences (Faulkner and Bailey, 2007). Attainment of higher bone
mineral content (BMC) during the ﬁrst two decades of life decreases the risk of
developing fractures later in life (Heaney et al., 2000). From infancy through to
adulthood, bone mass tends to track throughout life, meaning that infants, children, and

adolescents who have higher bone mass tend to be the adults who have high bone mass

80

(F erretti et al., 1998; Heaney et al, 2000). This association introduces the possibility that
“at risk” individuals may be identiﬁed early in life. Peak bone mass has genetic,
hormonal, nutritional, and behavioral determinants (Soyka et al., 2000). Some of these
determinants can be modiﬁed in “at risk” individuals.

In many studies of bone growth, children are grouped primarily by age with
secondary consideration of pubertal status, if at all. Grouping by age limits the ability to
generalize the results due to variation in initiation and duration of puberty, which

accelerates bone development. In addition, physical activity must also be considered. l

 

Children with higher levels of physical activity have higher measures of bone mineral Ei-
mineralization (Bailey et al., 1999; Iuliano-Burns et al., 2003; Slemeda et al., 1994;
Specker and Binkley, 2003; Zanker et al., 2003). Body composition itself may inﬂuence
bone mineralization, especially in overweight children (Goulding et al., 2000, 2001).
Dietary inﬂuences on bone mineralization in early childhood have been focused primarily
on calcium (Ca) intake. Calcium intake is associated with development of peak bone
mass (Weaver, 2000). During attainment of peak bone mass, Ca intakes of less than 1
g/day are associated with lower bone density (Weaver, 2000). However, the evidence
related to the long term effects of Ca supplementation in childhood indicates that gains
are not generally sustained (Abrams, 2005), indicating bone mineralization depends on
multiple factors, only one of which is Ca intake. Thus, intake of Ca as well as other key
bone-building nutrients should be considered in conjunction with other factors when
investigating determinants of bone mineralization in early childhood.

This study was undertaken to further clarify associations among diet, physical

activity, and body composition with bone growth and mineralization in a group of

81

children between 5-years of age and puberty. In addition, the relationships of biomarkers
of bone mineralization to prepubertal bone grth were explored. Understanding these
associations early in life may contribute toward an eventual ability to identify children
who are “at-risk” in terms of progress in attaining peak bone mass such that early
intervention can decrease their risk of developing osteoporosis later in life.
SUBJECTS AND METHODS

The Healthy Kids Nutrition Study (HKN S) enrolled 52 subjects of both genders
from the greater Lansing, Michigan area. The children were between 5-years of age and
puberty (<fanner 1), greater than the 5‘“ percentile BMI by NCHS standards, and 2 37
weeks gestational age at birth. They were generally healthy, without any chronic disease
conditions or birth defects other than possible allergies or food sensitivities. Children
with minor conditions that have no direct relationship to bone growth were included after
review by the study physician. Pubertal status was determined by a thorough parent/child
interview to ascertain Tanner Stage I status. Although recruitment by race was not
targeted, all children were at least 5 0% Caucasian. The study was approved by full
review of the University Committee on Research Involving Human Subjects (U CRIHS)
at Michigan State University (MSU) and met conﬁdentiality and HIPAA requirements.

Diet information was collected as a usual, one day diet recall obtained via a
consensus interview with a parent and the subject by a Registered Dietitian. The same
interviewer entered the diets for analyses into the nutrition database program within the
MyPyramid Tracker, based on the USDA food database (CNPP, 2005). This diet
assessment tool and analysis program provided intake estimates for 28 nutrients (CNPP,

2005), ﬁve of which are reported in this paper.

82

 

Standard pediatric measurement protocols to assure precision and accuracy were
employed (Gibson, 1990; Lee and Nieman, 1996) and followed the National Center for
Health Statistics guidelines for standard measurement (N CHS, 1996). Height was
measured to the nearest centimeter with a stadiometer (Invicta Plastics). Weight was
determined to the nearest 0.1 kg on a digital scale (Seca, Model 882).

The Actical accelerometer (MiniMitter) was used to measure physical activity.
The accelerometer was worn at the waist on a neoprene belt. This accelerometer has

been validated for use in estimating energy expenditure in young children (Pfeiffer et al.,

2006). The sampling frequency was 32 Hz and an epoch measurement of 15 seconds was 1.
chosen. Subjects wore the monitor continuously for three, consecutive days (one
weekend day and two weekdays).

The height and weight of each subject were uploaded to the Actical from which
basal energy expenditure (BEE), using the Harris and Benedict equation (Harris and
Benedict, 1919), was estimated by the Actical program so that daily total energy
expenditure (DTEE) from physical activity was reported as kcals above estimated BEE.

Serum intact osteocalcin (OC) was measured with the MetraTM Osteocalcin assay
(Quidel Corporation, Special Products Group, Santa Clara, CA, USA), an enzyme
immunoassay. Values are expressed in ng/mL. Twenty-ﬁve uL of serum per well
(assayed in duplicate) were used for determination of OC. The manufacturer has
established reference intervals for healthy men (3.4-9.1 ng/mL) and women (3.7-10.0
ng/mL). No reference standards were provided for children.

Urinary free deoxypyridinoline (DPD) was measured with the MetraTM DPD

assay (Quidel Corporation), a competitive enzyme. DPD values, expressed in nmol/L,

83

 

were corrected for differences in urine concentration and output by dividing by creatinine
(Quidel Corporation). The ﬁnal creatinine-corrected results were expressed as
nmol/mmol. The manufacturer has established reference intervals for healthy men (2.3-
5.4 nmol/mmol) and women (3.0-7.4 nmol/mmol); none are provided for children. Due
to the diurnal variation of DPD excretion, the ﬁrst morning urine sample was used.
Parents and subjects were instructed on the urine collection and given sterile supplies for
the ﬁrst morning urine sample collection.

Serum 25(OH) Vitamin D3 was measured once during the study at the analytical
lab of LabCorp (Dublin, OH) via an immunochemiluminometric (ICMA) assay. The
reference standards provided by the lab vary with age and season due to the seasonal
variability of sun exposure.

Whole body measures were made using a GE LUNAR Prodigy machine at
F iechtner Research (Lansing, MI) with pediatric software speciﬁc to the machine
(software version 6.81; General Electric Lunar Corporation). Measurement protocols
speciﬁc to the GE-LUNAR Prodigy densitometer were employed, including daily
calibration with a standard calibration block and calibration tri-weekly with a body
phantom. The CV of this machine is reported to be < 1.0% as in other papers) (personal
communication, Fiechtner Research). Measures obtained included body composition (kg
fat mass and kg lean mass), BMD (gm/cmz) and BMC (gm). The full body scan took
approximately 2 minutes and exposed each child to less than 0.04 mRem of radiation.
Motion artifacts were not a problem in this population.

The BMD reference values from the GE Lunar DXA program were used in power

calculations (Wacker and Barden, 2001). The mean BMD for the 7-year-old male was

84

used as the mean BMD of the population and a SD of 0.07 was used to calculate power.
Recruitment of 52 subjects assured the sample size necessary to detect differences at a
95% Conﬁdence Interval.

Descriptive statistics were employed to describe the study population with values
reported as means i SDs. The primary dependent variables are BMD, BMC, and the
Lunar DXA-reported Z-score. The independent variables are select components of the
dietary intake as assessed by recall, serum 25 (OH) Vitamin D3, serum OC, urinary DPD,
measures of body composition, and physical activity expressed as daily total energy
expenditure in kilocalories. A two-tailed t-test was employed to determine signiﬁcant
differences between genders in all variables. Relationships between variables were
explored with 2-tailed Pearson’s correlations. Multiple regressions with stepwise
variable entry were used to determine variables that may be signiﬁcant predictors of
BMD and BMC. Data were analyzed using SPSS, version 13, Base System (SPSS, Inc.).
The signiﬁcance level for all tests was established as p<0.05 and all tests were bi-
directional.

RESULTS

Descriptive and body composition characteristics of the study subjects are
summarized in Table 4.1, by gender and for all children. The mean age of the subjects
was 7.6 years (range 5.1 to 11.9). Twenty-seven of the subjects were male; twenty-ﬁve
were female. BMIs ranged from 13.7 to 23.9 with a mean BMI of 16.8. Z-scores for the
BMDs by Lunar DXA ranged from negative 1.2 to positive 1.8. The mean Z-score was
0.14 with 30.8 percent of these subjects having Z-scores at or below negative 0.2. There

were no differences in body composition measures between genders.

85

Biomarker measures are also summarized in Table 4.1. Mean serum 25(OH)
Vitamin D3 was 30.0 ng/mL i 10.2 (range 10.8 to 62.3). No subject had a Vitamin D3
outside of the normal range as deﬁned by the analytical laboratory. Mean serum OC
concentration was 25.4 ng/mL : 4.65 (range 14.71 to 35.57). Mean urinary DPD
concentration was 20.38 i 5.96 nmol/mmol creatinine (range 9.02 to 41.80). There was a
signiﬁcant difference between males and females only for serum OC (t-test, p S 0.05)

with females having higher serum OC concentrations.

Nutrient intakes are summarized by gender and for all children in Table 4.2.

13"

There were no differences by gender. Mean caloric intake was 2180 i 545 (range 1123
to 3152). Mean protein intake was 79 g i 21 (range 37 to 124), representing a mean
protein intake of 2.94 g/kg body weight. Total mean intake of Ca was 1382 mg i 581
(range 253 to 2715).

Based on the activity counts as measured by the activity monitor, energy
expenditure above calculated basal energy expenditure (BEE) (Harris and Benedict,
1919) is estimated as smnmarized in Table 4.1. Total daily energy expenditure (DTEE)
above BEE for these subjects was 562 i- 217 kcals (range 281 — 1120). There was no
difference in DTEE above BEE by gender.

In exploring the relationships of the measured variables inﬂuencing bone
mineralization with the dependent variables of BMD and BMC, several signiﬁcant
correlations were found as summarized in Table 4.3. Percent body fat had a strong,
signiﬁcant positive relationship to both BMD (r=0.535, p50.01) and BMC (r=0.646,
pS0.01). OC by height showed a signiﬁcant negative correlation to BMD (r=-0.507,

pS0.01) and to BMC(r=-0.499, p_<_0.01). Neither serum 25(OH) Vitamin D3 nor urinary

86

 

Table 4.1

Physical characteristics, measures of bone mineralization, and biomarkers

 

 

 

Variable Gender
Male (n=27) Female (n=25) Total (n=52)
Age Mean 3 so 7.4 3 1.4 7.8 3 1.9 7.6 3 1.7
Range 5.6-11.1 5.1-11.9 5.1-11.9
BMI Mean 3 so 17.3 3 2.5 16.3 3 1.9 16.8 3 2.3
Range 14.7 - 23.9 13.7 - 20.6 13.7 - 23.9
Weight (kg) Mean 3 so 26.3 3 5.9 27.5 3 7.6 26.9 3 6.8
Range 17.5 - 43.7 18.2 - 47.0 17.5 - 47.0
Height (cm) Mean 3 so 128.4 3 13.5 124.8 3 9.9 126.5 3 11.8
Range 105.2 - 168.5 107.25 -1410 105.2 -168.5
Fat mass (kg) Mean 3 so 6.0 3 4.8 6.0 3 3.4 6.0 3 4.1
Range 2.1 -18.8 2.8 - 16.4 2.1-18.8
DTEE above Mean 3 so 609 3 225 512 3 200 562 3 217
BEE (kcals) 1 Range 323 - 1120 281 - 973 281 — 1120
BMD (g/emz) Mean 3 so 0.885 3 0.062 0.837 3 0.060 0.847 3 0.061
Range 0.764 -1.003 0.751 - 0.980 0.751 - 1.003
Z-score 2 Mean 3 so 0.31 3 0.64 -004 3 0.59 0.14 3 0.64
(Lunar) Range -0.60 - 1.80 -120 - 1.20 -120 - 1.80
BMC (8) Mean 3 so 960 3 262 933 3 229 947 3 245
Range 618 - 1,657 567 - 1,514 567 - 1,657
Serum 0C 2‘3 Mean 3 so 23.82 3 4.39 26.98 3 4.43 25.40 3 4.65
(“g/ml“) Range 14.71 - 35.57 17.42 - 33.41 14.71 - 35.57
Urinary DPD Mean 3 so 20.00 3 5.83 20.79 3 6.20 20.38 3 5.96
(nmol/mmol
creatinine) Range 10.81 - 41.80 9.02 - 35.83 9.02 - 41.80
Serum 25(OH) Mean 3 so 30.4 3 9.4 29.5 3 11.3 30.0 3 10.2
Vitamin D3
(ng/mL) 4 Range 13.8 - 57.5 10.8 - 62.3 10.8 - 62.3

 

I For this variable, n=26 (males), n=25 (females), n=51 (total)
2 Signiﬁcant difference between male and female subjects per t-test (pS0.05)
3 For this variable, n=25 (males), n=25 (females), n=50 (total)
4 For this variable, n=26 (males), n=23 (females), n=49 (total)
BMI: Body Mass Index BMD: Bone Mineral Density BMC: Bone Mineral Content
OC: Serum osteocalcin DPD: Urinary deoxypyridinoline
DTEE above BEE: Daily Total Energy Expenditure above Basal Energy Expenditure (BEE estimate per

Harris IA, Benedict PO. A biometric study of basal metabolism in man. (Publication No. 279) Washington,

DC: Carnegie Institute of Washington; 1919).

87

Table 4.2

Intake of key bone-building nutrients per day measured by usual, one-day dietary recalll

 

 

 

Variable Gender
Male (n=27) Female (n=25) Total (n=52)
Mean3 so 2.2193507 2.1383591 2.1803545
Total kcals
Range 1,123 - 3,152 1,167 - 3.115 1,123 - 3,152
Protein (g) Mean 3 so 79 3 22 79 3 20 79 3 21
Range 37 -124 48 -113 37 -124
Ca (mg) Mean 3 so 1.343 3 669 1.424 3 479 1,382 3 581
Range 253 - 2.715 790 - 2,505 253 - 2,715
P (mg) Mean 3 so 1,584 3 505 1,565 3 442 1.575 3 471
Range 645 - 2,766 1.001 - 2,531 645 - 2.766
Mg (mg) Mean 3 so 327.4 3 97.4 296.1 3 85.0 312.4 3 92.1
Range ' 104.5 - 539.2 158.4 - 486.2 104.5 - 539.2

 

I Intakes by recall do not include supplements.

88

DPD by height showed any correlation to BMD or BMC. Daily total energy expenditure
had a strong, signiﬁcant positive correlation to both BMD (F0459, pS0.01) and BMC
(r=0.59l , pS0.01). Protein, expressed relative to body weight, had strong, signiﬁcant
(p50.01) negative relationships to both BMD, r=-0.508, and BMC, r=-0.564. Due to the
wide range of subjects’ body sizes and reported intakes, intake of Ca was expressed as
mg/day/cm height. A signiﬁcant negative association was found between Ca by height
and BMD (r=-0.277, pS0.05). There was no correlation between Ca intake/cm height
and BMC. Signiﬁcant correlations (p50.01) were also found for kcal intake per kg body
weight (for BMD, r=-0.510, and for BMC r=-0.578) and for mg phosphorus intake per
day by cm height (for BMD, r=-0.378, and for BMC, r=-0.348). No signiﬁcant
correlation to BMD or BMC was found with other nutrients (data not shown).

Table 4.4 and Table 4.5 report ﬁnal multiple regression analyses predicting BMD
and BMC from measured variables inﬂuencing bone growth in this population of
prepubertal children. Independent variables found to have a signiﬁcant correlation and
those with a strong association to bone grth from the literature were entered in a
stepwise manner (DTEE, % body fat, protein per kilogram body weight, Ca per
centimeter height) as were the biomarkers serum OC, urinary DPD and serum 25(OH)
Vitamin D3. Though four nutrients had signiﬁcant correlations to BMD and BMC,
protein and kilocalories as well as Ca and phosphorus have collinear relationships that
would inﬂuence the regression analyses given the limited nutrients included. Both
protein and Ca have been shown previously in the literature to have a strong association

to bone mineralization thus only protein and Ca were entered in the regression analyses.

89

 

Table 4.3

Correlations of BMD and BMC with variables inﬂuencing bone mineralization

 

 

Pearson Correlation BMD (g/cmz) BMC (g/cm)
% body fat 0.535” 0.646**
(N = 52)
Protein intake by weight‘ (g/kg) -0.508** -0.564**
(N = 52)
Ca intake by height' [(mg/d)/cm] -0.277* -0.260
(N = 52)
DTEE above BEE 0.459“ 0.591 **
(N = 51)
OC by height ((ng/mL)/cm) -0.507** -0.499**
(N = 50)
DPD by height ((nmol/mmol Cr)/cm) -0.245 -0.204
(N = 52)
Serum 25(OH) Vitamin D3 (ng/mL) 0.028 -0.066
(N = 49)

 

** Pearson correlation signiﬁcant at the 0.01 level (2-tailed).

* Pearson correlation signiﬁcant at the 0.05 level (2-tailed).

1 All nutrient intakes as assessed by recall

BMD: Bone Mineral Density

BMC: Bone Mineral Content

OC: Serum osteocalcin

DPD: Urinary deoxypyridinoline

Cr: Creatinine

DTEE above BEE: daily total energy expenditure above basal energy expenditure per
Harris-Benedict equation estimate (Harris JA, Benedict FG. A biometric study of basal
metabolism in man. (Publication No. 279) Washington, DC: Carnegie Institute of
Washington; 1919).

90

The ﬁnal model for predicting BMD includes percent body fat, serum OC, urinary DPD,
serum 25(OH) Vitamin D3, and DTEE above BEE. In this model, the biomarkers OC and
DPD relative to height were signiﬁcant negative predictors of BMD. The associated
relationships of all of these variables explain 58.9 % of the variability in BMD. The ﬁnal
model for predicting BMC includes percent body fat, serum OC, urinary DPD and DTEE
above BEE, with the association among these variables explaining 64.9% of the
variability in BMC. The B coefﬁcient is indicative of the relative importance of an

independent variable in contributing to the dependent variable in a given prediction

ll-‘”“"--'ﬂl

model. Serum OC by height (B=-0.435 i 0.165, p=0.012), and urinary DPD by height
(B=-0.262 i 0.131, p=0.054) are the strongest, albeit negative, predictors of BMD.
Serum DC by height (B=-1528.178 i 620.868, p=0.018), and urinary DPD by height
(B=-936.158 i 492.864, p=0.064) are also the strongest predictors of BMC. The
correlation between actual BMC and actual BMD and their values as predicted by each
model are illustrated in Figure 4.1 and Figure 4.2. The prediction models show good
agreement with the actual data, R2 for the BMD model being 0.5295 and R2 for the BMC
model being 0.6181.
DISCUSSION

Studies of bone growth in early childhood consider the combined effects of
modiﬁable inﬂuences on bone grth including body composition, physical activity, and
diet due to the potential interaction of all of these determinants of bone health. Much of

the available literature investigating bone growth in childhood is difﬁcult to interpret

91

 

Table 4.4

Multiple regression model predicting BMD as a ﬁinction of % body fat, serum OC by
height, urinary DPD by height, serum 25(OH) Vitamin D3, and DTEE above BEE

 

 

 

 

Predictors B
% body fat 0.003T
DC by height ((ng/mL)/cm) -0.4352
DPD by height ((nmol/mmol Cr)/cm) -0.2623
Serum 25(OH) Vitamin D3 (ng/mL) 0.0014
DTEE above BEE (kcals) 0.000071342
Constant 0.8431
R2 = 58.9%
' p < 0.01
2p<0.05
3 p = 0.054
4 p = 0.072

BMD: Bone Mineral Density

OC: Serum osteocalcin

DPD: Urinary deoxypyridinoline

Cr: Creatinine

DTEE above BEE: Daily Total Energy Expenditure above Basal
Energy Expenditure per Harris-Benedict equation estimate (Harris
IA, Benedict FG. A biometric study of basal metabolism in man.
(Publication No. 279) Washington, DC: Carnegie Institute of
Washington; 1919).

Variables entered into the regression model (stepwise): protein
intake by weight, calcium intake by height, % body fat, serum OC
by height, urinary DPD by height, serum 25-OH Vitamin D3,
DTEE above BEE.

92

Table 4.5

Multiple regression model predicting BMC as a function of % body fat, serum OC by
height, urinary DPD by height, and DTEE above BEE

 

 

 

 

Predictors B
% body fat 12.054I
DC by height ((ng/mL)/cm) -1 528.1782
DPD by height ((nmol/mmol Cr)/cm) -936.1583
DTEE Above BEE (kcals) 0.377l
Constant 948.689l
R2 = 64.9%
‘ p < 0.01
2p<0.05
3p = 0.064

BMC: Bone Mineral Content

OC: Serum osteocalcin

DPD: Urinary deoxypyridinoline

Cr: Creatinine

DTEE above BEE: Daily Total Energy Expenditure above Basal
Energy Expenditure per Harris-Benedict equation estimate (Harris
JA, Benedict FG. A biometric study of basal metabolism in man.
(Publication No. 279) Washington, DC: Carnegie Institute of
Washington; 1919).

Variables entered into the regression model (stepwise): protein
intake by weight, calcium intake by height, % body fat, serum OC
by height, urinary DPD by height, serum 25(OH) Vitamin D3,
DTEE above BEE.

93

because study populations often include both prepubertal and pubertal children (Zamboni
et al., 1988; Rauch et al., 1994; VandenBergh et al., 1995; Molgaard et al., 1997; Carter
et al., 2001; Iuliano-Burns et al., 2003). The aim of the present study was to examine the
association of BMD and BMC with protein and Ca intake, body composition, physical
activity, and biomarkers of bone metabolism in order to better understand their
relationship to and inﬂuence on bone growth during childhood, speciﬁcally the period
between 5 years of age and puberty.

In these prepubertal subj ects, gender differences were not apparent for any of the
variables except serum OC by cm height and DXA Z-score. The literature reports
inconsistencies related to gender differences in bone mineralization in prepubertal
children. Some studies have seen gender differences (Specker et al., 1987; Bell et al.,
1991; Horlick et al., 2000). Many studies found no gender differences (Geusens et al.,
1991; Patal et al., 1992; Fassler and Bonjour, 1995; Maynard et al., 1998; Nguyen et al.,
2001). As in this study, most of these studies adjusted for height or some other measure
that accounted for body or bone size. Because of the similarities between the genders of
the HKNS subjects for other bone-related variables measured, gender was not entered
into the prediction models.

All of the HKNS children had Z-scores for BMD within one standard deviation of
the mean. Reference norms for BMD and BMC contribute to determination of Z-scores
speciﬁc to the equipment on which BMD is measured (Wacker and Barden, 2001). This
measure places children in reference to other children in the population. The referent
populations of children with whom these subjects are compared are drawn from the same

larger population that is, itself, the population for whom osteoporosis is thought will

94

 

 

Figure 4.1

Correlation between actual BMD and predicted value of BMD based on
% body fat, serum OC by height, urinary DPD by height,
serum 25(OH) Vitamin D3 and DTEE above BEE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

1.05 —- _- 3-. 3 3 3.3 3..-- -
1.00 o g
0
0.95 a 9 /
O
9 o
o R2=0.5295
0.90 °
9 o
./
O
O
0.85 9o e
O
.9
O
o o 8 o’
O
0.80 H2 - a
' o 9 I
O
0
O O .
0.75 3
0.70 1 1' T I I 1 :
0.70 0.75 0.80 0.85 0.90 0.95 1.00 ‘
Predicted Value of BMD '

 

95

 

 

Figure 4.2

Correlation between actual BMC and predicted value of BMC based on
% body fat, serum OC by height, urinary DPD by height, and DTEE above BEE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

1700.0 — ,
0
1500.0 9
. O /
1300.0 °
0
o , o R2=O.6181
o. o ,
1100.0 9
o
0 § ° l
0
900.0 . V . o -»- ~33 -. 3
' o
o 9 1
o . 9
700.0 . v 4 9 3-3 - - ______
/ . . ° . . i
0
500.0 a . . . . 4 1
500.0 700.0 900.0 1100.0 1300.0 1500.0 17000 !
l
Predicted Value of BMC i

 

96

 

reach epidemic proportions. Thus, Z-scores do not necessarily indicate that optimal
progress toward peak bone mass is occurring. If osteoporosis is a pediatric disease with
geriatric consequences, lower Z-scores may indicate increased risk. Given that 31% of
the children had Z-scores below negative 0.2, some children may not be making
satisfactory progress toward attaining peak bone mass within their genetic potential.
Body composition in children has been positively related to measures of bone
mineralization. Fehily et al. (1992) and Specker et al. (2001) saw a positive association

with total body weight. Pietrobelli et al. (2002) and Ferretti et al. (1998) saw a positive

12"“—'"11

relationship with lean mass. Ilich et a1. (1998) found body fat to be a positive predictor
of bone mass. In adults, in general, increased body weight predicts increased bone
density (Pietrobelli et a1, 2002). These 52 subjects had a wide range of heights and
weights with some of the younger children being bigger than some of the older children;
however, all subjects were within the range of normal grth for age falling within the
5th and 95th percentiles on the NCHS growth charts (N CHS, 1996). Lean mass could
have been used to explore the relationship of body composition to bone mineralization as
lean mass is considered to reﬂect muscle mass. Through the mechanical load of muscle’s
action on the skeleton, muscle can inﬂuence greater bone mineralization. Both muscle
and organ tissues are included in the lean mass estimate provided via whole body DXA.
Height, weight, and lean mass have strong collinear relationships to bone mass and to
each other, as do weight and fat mass, complicating the choice of a single body
composition measure to use to predict bone mineralization. Percent body fat had strong,
positive correlations to BMD and BMC, was least complicated by collinear relationships.

and, from a clinical perspective, can be estimated through simple anthropometric

97

 

measures and is modiﬁable through lifestyle habits. For these reasons, percent body fat
was used in the models to predict BMD and BMC. Body fat appeared to have a
beneﬁcial effect on bone mineralization in prepubertal children in that a higher percent
body fat, within the range of normal, predicts a higher BMD, whereas, in this model, a
low percent body fat, one still within the normal range, contributes to predicting a lower
BMD and BMC.

This group of children was a well-nourished group of subjects with mean intakes
of most nutrients at or above the US RDA (FNB/IOM, 1997; 2005). Mean energy intake
was 8l-kcals/kg mean body weight, which is within the range of normal for healthy
children of this age (FNB/IOM, 2005). The US RDA for protein in early childhood is
0.95 grams/kilogram body weight (FNB/IOM, 2005). Protein intake in this subject
population was 3.1 times greater than the DRI recommendation. The strong negative
correlation of this protein intake with BMD and BMC differs from ﬁndings of recent
studies in children within this age range (Ilich et al., 1998; Bounds et al. 2005) and is, to
our knowledge, the only such negative correlation seen to date in a group of children.
Given the high protein intake of these children, the strong negative relationship of bone
mineralization to dietary protein merits further exploration.

Ca intake in this group was at or above the US Adequate Intake (A1) of 800
mg/day for most of the subjects. Only 8 children consumed less than 800 mg Ca/day;
none had excessive intake. These children differ from national samples that predict less
than 40% of children in the US between 6 and 11 years of age achieve the Al for Ca
(Greer and Krebs, 2006). The high mean Ca intake of the HKNS subjects, 1300 mg/day.

is in the range of total Ca intake that is generally provided in supplementation trials that,

98

on the whole, do increase bone mass during the period of supplementation (Abrams,
2005). However, as with protein, Ca intake relative to height had a signiﬁcant negative
correlation with BMD, though not with BMC. This negative correlation was likely
related to the collinear relationship of protein and Ca in that the highest source of protein
and the highest source of Ca were single item dairy foods such as milk, chocolate milk,
yogurt and cheese (unpublished data).

On an individual nutrient basis, protein and Ca were more than adequate in the
diets but both protein and Ca had an inverse relationship to BMD and BMC. When,
however, the association of these nutrients, expressed relative to body and bone size, was
examined through the multivariate regression process, they did not remain in the model
as signiﬁcant dietary predictors of BMD and BMC.

Evaluation of the literature in the area of biomarkers in children is inﬂuenced by
the need to recognize that much of the available data mixes pre-pubertal and pubertal
population of children. OC concentrations in the HKNS subjects represent OC values for
a healthy population of prepubertal children. In normal, healthy children, OC
concentrations can be expected to increase with age and body size until puberty, then
gradually decline to low adult levels (Tommasi et al., 1996; Bonoﬁglio et al., 2000).
Several studies have measured OC concentrations in normal children of varying and
increasing ages and found OC correlated weakly with BMD (Glastre et al., 1990;
Tommasi et al., 1996). Slemenda et al. (1997) documented changes in OC concentrations
in prepubertal and pubertal subjects with DC peaking at Tanner I then beginning a
decline. Fares et al. (2003) noted an increase in OC in boys and girls that peaked at

pubertal Tanner 111 then began a decline. Van Coeverden et al. (2002) documented

99

signiﬁcant positive correlations between OC and BMC at several sites in a group of
peripubertal Dutch children. Conversely, consistent with the prepubertal HKNS subj ects,
Mora et al. (1999) observed an inverse association between OC and BMD in children at
varying stages of puberty. Slemenda et al. (1997) also noticed an inverse relationship
between OC and BMD in a group of Ca-supplemented children. Collectively, though
these observations are not always consistent, the potential role of OC in the assessment of
bone growth in children as a marker for bone mineralization is supported.

In association with other independent variables, urinary DPD was a signiﬁcant
predictor of BMD and BMC. A few studies have looked at DPD as a marker of bone
mineralization in children. Initial efforts to establish reference levels have been reported
(Rauch et al., 1994; Lieuw-A-Fa et al., 1995). Rauch et al. (1994) as well as Bollen and
Eyre (1994) report that DPD levels are highly correlated with grth velocity in normal
children. Mora et a1. (1999) looked speciﬁcally at DPD in relation to Tanner stage of
development. They found DPD peaking at Tanner 11 then declining with a positive
relationship to bone volume rather than to bone density per se. Van Coeverden et al.
(2002) documented signiﬁcantly positive correlations between DPD and BMC at several
sites in a group of peripubertal Dutch children. In general, DPD concentrations will
increase with age until puberty and then will begin to decline to a sustained low adult
concentration (Rauch et al., 1994; Acil et al., 1996). In the HKNS subjects, urinary DPD,
expressed per cm height, did not show a signiﬁcant correlation to BMD or BMC on its
own but was entered into the regression model to predict BMD and BMC. In association

with other independent variables, urinary DPD per cm height was a signiﬁcant predictor

100

 

of both BMD and BMC and may have potential as a biomarker for bone mineralization in
children.

Public health concerns about adequate Vitamin D status have been raised in
various populations including the elderly, postmenopausal women, and children in
northern climes (Norman and Henry, 2006). Docio et al. (1998) and Lips (2004) suggest
that a minimum desirable level of serum 25(OH) Vitamin D3 in children should be 20
ng/mL or higher with slightly lower levels acceptable when Ca intake is high. The mean
serum level of 25(OH) Vitamin D3 in the HKN S subjects was 30.0 ng/mL. Seven
subjects, or 16% of the group, had levels at or below 20 ng/mL. Like DPD, serum
Vitamin D did not show a signiﬁcant correlation to BMD or BMC on its own but was
entered into the regression model due to its potential for an association through
interactions with other variables as well as its overall importance in bone metabolism. In
association with the other variables, 25(OH) Vitamin D3 appears to have a signiﬁcant
positive relationship to BMD in these prepubertal children.

Physical activity is acknowledged to be beneﬁcial to bone health. However, the
role physical activity plays in the accrual of peak bone mass and its interactions with
other factors inﬂuencing bone grth in the early childhood years is less clear. In
general, children are less physically active now than in the past. This decrease is
estimated as a decline of approximately 600 kcals/day over the past 50 years (Boreham
and Riddoch, 2001). Previous studies in prepubertal children documented increases in
bone mineralization in response to physical activity (Slemenda et al., 1994; VandenBergh
et al., 1995). In a group of even younger children than these in the HKNS, children age

3- to 5- years, Specker and Binkley (2003) observed that physical activity was beneﬁcial

101

to bone mineralization only at higher levels of Ca intake. Molgaard et al. (2001) and
Iuliano-Burns et al. (2003) saw increased BMC in groups of girls who exercised and had
higher Ca intakes, though only some of the girls were prepubertal. When energy
expenditure was considered along with other factors in bone mineralization in the
prediction model, DTEE remained signiﬁcant as a positive predictor of BMC and BMD.
Physical activity must be considered in understanding the interrelationships among
factors predicting bone mineralization of children of this age. This relationship is
especially pertinent given the downward trend in children’s physical activity over the past
50 years.

Based on the relationships of measured variables to BMD and BMC, two
prediction models were developed. Expressing the variables OC, DPD, and Ca relative to
bone size is most appropriate in that the independent variables measured are being
examined for their effect on bone metabolism. These prediction models, developed from
a small cross-section of children in the Midwest. suggest that risk for lower BMD can be
predicted from an anthropometric measure (percent body fat), two blood tests (OC and 25
(OH) Vitamin D3), urinary DPD, and an estimate of physical activity. Low BMC can be
predicted from serum OC, urinary DPD, an anthropometric measure of body fat, and an
estimate of physical activity. DPD can also be measured in the serum so potentially a
urine sample would not be needed.

Thirty percent of the HKNS children had Z-scores that identify them as having
BMDs and BMCs in the lower range of normal and possibly already at risk of poor
progress toward attainment of peak bone mass. Public health efforts with children

currently promote healthy body weights through diet and physical activity in order to

102

decrease the risk of overweight and of weight-related chronic diseases. The physical
activity focus will also promote bone health, which our study supports. Initial screening
and assessment of bone status may be possible through an equation that considers key
variables. Decreasing the risk of osteoporosis should begin in childhood with assessment

and intervention that promotes optimal bone mass.

103

 

CHAPTER FIVE

104

 

CHAPTER FIVE

Strengths, Limitations, and Implications

The Healthy Kids Nutrition Study is unique in its simultaneous consideration of
many of the variables that inﬂuence bone mineralization, joining a small body of research
conducted exclusively on prepubertal children of elementary school age. Many factors
are known to inﬂuence bone mineralization. Because of potential interactions among
these factors it is important to study as many of them together as possible. This study
was designed to measure these variables as accurately and precisely as possible. The
subjects were carefully screened for health and prepubertal status. DXA determined
body composition as well as bone mineral measures. Physical activity was measured
over several days with a validated activity monitor. An experienced, registered dietitian
assessed diets using careful interview methods and a validated F FQ. Experienced,
qualiﬁed laboratory personnel conducted the serum and urinary measures. The design of
the study allows for additional data to be gathered either with additional children or with
these same children over time.

Though sufﬁcient numbers of subjects were recruited to allow us to detect
differences in bone mineralization, this was a convenience sample of Caucasian children
from the US. Midwest thus limiting the generalizability of the results. As a cross
sectional study, these data can drawn associations but does not suggest causality. Most of
the children appeared to be of similar middle-class socioeconomic status and there were
several groups who, through word of mouth recruitment, attended the same school and

even the same class. Regional similarities in food markets, recreation opportunities, and

105

weather also contribute to the homogeneity of the group. Differences within the subjects
could be detected when the continuous variables examined but once groups were divided
through post hoc categorization, some of the groups were too small to allow for the
groups to be compared, for example low versus high calcium intake.

This work contributed to the literature examining variables inﬂuencing bone
mineralization. It supports the possibility that “at risk” children can be identiﬁed in a
public health or clinical setting. Lifestyle factors that can be promoted to improve bone
health have also been suggested. Protein intake in excess of the DRI appears to
negatively impact bone in these children. Calcium intake, considered by height instead
of chronological age, predicts bone mineralization differently than just total daily calcium
intake, suggesting that height may need to be considered in assessing adequacy of
calcium intake. Vitamin D concentrations were within the range of normal but higher
levels did predict higher bone mass, and concentrations were lower than what some
recent studies suggest is optimal, indicating a need to look further into Vitamin D status
in Midwestern children. These children were all of normal weight. Even within this
range of normal weights, controlling for height, the children with a higher percent body
fat tended to have higher bone mass suggesting that body composition within the normal
range can inﬂuence bone health. Finally, this study reports serum OC and urinary DPD
levels for a group of healthy, prepubertal children. Osteocalcin showed a strong
association with BMD raising the potential for its use as a biomarker to follow bone
mineralization, especially in high-risk children.

The ﬁndings of this study related to relationships between diet and bone

mineralization suggest that perhaps requirements for bone building nutrients in children

106

should be based body size rather than on age. If requirements are based solely on age,
nutrition recommendations and interventions for conditions such as obesity may result in
compromised linear grth or compromised bone mineralization. Caloric limitations
usually suggested to treat overweight children could increase the risk of poor bone
mineralization if there is a concurrent, proportional decrease in intake of bone building
nutrients. When bone mineral content and BMD are calculated and then compared to age
referenced normals, the larger body and bone size of the overweight child may not be

properly assessed given that the reference values are age based. Likewise, children at the

.h-‘"_ —H-

lower end of the growth curve may also be poorly assessed when compared to age
referenced normals. In addition, the need to look at a variety of bone sites in order to get
the best picture of bone growth in the preadolescent is borne out by the previous studies,
each of which observed that use of overall BMC/age or BMD/age does not allow
appropriate assessment of bone growth due to the differences in body composition and
differences in the effect of physical activity relative to body composition on bone growth
parameters (Fischer et al., 2000; Goulding et al., 2000; Hasanoglu et al., 2000; Molgaard
et al.. 2001.)

Future research in this area could include following this group of children in to
puberty to see if their bone mineralization tracks into adolescence as has been suggested
in the literature. The strong negative relationship to dietary protein merits further
exploration. The low intake of fruits and vegetables along with the high protein intake in
these children points the investigation in the area of dietary and metabolic acid base
balance. In addition, this pilot group could be extended to include sufﬁcient children to

have at least 16 children in each post hoc group so additional relationships can be

107

 

explored. Similar studies with children of other races would be important as well as a
comparison of groups with high and low intakes of protein, calcium and fruits and
vegetables. Additional cross sectional as well as longitudinal studies are needed in this
population in order to understand better how to keep children off the trajectory toward

osteoporosis in their later years.

108

:__fl

{-
l

 

APPENDICES

109

 

OFFICE OF
REGULATORY
AFFAIRS

Human Research
Protection Programs

BIOMEDICAL I HEALTH
INSTITUTIONAL REVIEW
BOARD (BIBS)

COMMUNITY RESEARCH
INSTITUTIONAL REVIEW
BOARD (came)

SOCIAL SCIENCE]
BEHAVIORAL I EDUCATION
INSTITUTIONAL REVIEW
BOARD (sine)

202 Olds Hal

East Lansing, Michigan
48824-1046
517-355-2180

Fax: 517432—4503

wwwhurnanresearchmsuedu
IRBO msuedu

 

USU 1‘: an 407ml” action
equal-opportunity Uu‘lllulloll

Appendix 1

MICHIGAN STATE Renewal
U N I V E R S I T Y App'ication
Approval

 

January 18. 2008

To; Jenny BOND
204 Trout FSHN Bldg.

Re: lRBll 03-1019 Category. EXPEDITED 2-8
Renewal Approval Date: January 14. 2008
Project Expiration Date: January 13. 2009

Title: EXAMINATION OF DIET. PHYSICAL ACTIVITY. BIOMARKERS OF BONE MINERALIZATION. BONE
MINERAL CONTENT AND BODY COMPOSITION IN CHILDREN BETWEEN 5 YEARS OF AGE AND
PUBERTY

The Institutional Review Board has completed their review of your project. I am pleased to advise you that the
renewal has been approved.

This letter notes approval for data analysis only (contact with subjects and data collection is complete).
Any further recruitment. data collection or contact with subjects will require IRB review and approval via
a revision before implementation.

The review. by the committee has found that your renewal is consistent with the continued protection of the rights
and welfare of human subjects. and meets the requirements of MSU's Federal Wide Assurance and the Federal
Guidelines (45 CFR 46 and 21 CFR Part 50). The protection of human subjects in research is a partnership

between the IRB and the investigators. We look forward to working with you as we both fulﬁll our responsibilities.

Renewals: IRB approval is valid until the expiration date listed above. If you are continuing your project. you
must submit an Application for Renewal application at least one month before expiration. If the project is
completed. please submit an Application for Permanent Closure.

Revisions: The IRB must review any changes in the project. prior to initiation Of the change. Please submit an
Application for Revision to have your changes reviewed. It changes are made at the time of renewal. piease
include an Application for Revlslon with the renewal application.

Problems: If issues should arise during the conduct of the research. such as unanticipated problems. adverse
events. or any problem that may increase the risk to the human subjects. notify the IRB ofﬁce promptly. Forms
are available to report these issues.

Please use the IRB number listed above on any forms submitted which relate to this project. or on any
correspondence with the IRB ofﬁce._

Good luck in your research. If we can be of further assistance. please contact us at 517-355—2180 or via email
at lRB@msu.edu. Thank you for your cooperation.

Sincerely.

ﬂA/AEE .

Peter Vasilenko. PhD.
BIRB Chair

0: R. Taylor SCOTT
82090 W. Fee

110

Appendix 2

Examination of diet, physical activity, biomarkers of bone mineralization, bone
mineral content and body composition in children between 5 years of age and

puberty
Consent to participate in a research study.

Please initial each section to indicate that you have read that section and have had your questions
answered.

This research is investigating the diet and nutritional needs of children relative
to their body composition, bone development and physical activity patterns.
Parents/legal guardians will be closely involved in all aspects of the study. This study is
being conducted at Michigan State University in the Department of Food Science and
Human Nutrition. The study experience will involve two (2) or three (3) appointments of
1-2 hours duration on the campus of Michigan State University and one appointment at
Fiechtner Research on Jolly Road in the SE part of Lansing. Enrollment is for a six (6)
month time frame.

Participants in this study must meet three (3) conditions.
1. Participants must be 5 years of age or older.
2. Participants must be prepubertal
(as determined by physician review of parental questionnaire).
3. Participants can be any size or weight.

You are being asked to involve your child in the following activities and
procedures: '

Complete a brief family health history.
Health questions will be limited to issues related to the research question.
Participants can decline to answer any question asked throughout the study on
any topic.

Engage in an interview called a diet history with the study nutritionist.
Complete a survey on usual food intake over time.
This survey will be entered for computer analysis of nutrient intake and results
will be given to the parent/guardian.

Complete a survey on usual physical activities.
This survey will be entered for computer summary of usual physical activities.

Have height and weight measured.
Equipment used is similar to that used in routine health care.

111

Wear a small physical activity monitor on a waist belt for 72 hours.
The monitor is the size of a wristwatch and functions similarly to a car’s
acceleration detector and a pedometer that measures walking steps. It is worn
under the clothes and the child will have the feeling of wearing a belt. The
pedometer is water resistant and can be worn all 72 hours but may be removed
for bathing if desired. The data recorded by the electronic monitor will be
processed and decoded by computer. The results will be reported to the
parent/guardian.

Provide one blood sample and one urine sample to measure Vitamin D
and two compounds related to bone growth.
Approximately one teaspoon of blood will be taken from the child’s arm by a
health care worker experienced in drawing blood from children using safe and
sterile methods and equipment. The blood sample will be analyzed for Vitamin D
and one measure of bone growth. The child will feel the discomfort of the needle
when the blood is drawn. There may be some temporary bruising at the site of
the blood draw. About 1/4 cup of a ﬁrst morning urine sample. obtained by the
child in the privacy of their home then kept cold until the study appointment. will
be analyzed for the other marker of bone growth. The tests of bone growth are
approved only for research use in children because they are newly developed
and research is needed to begin to determine how to interpret these tests in
children. Vitamin D levels will be reported to the parent/guardian.

Measurement of body fat, lean tissue and bone size on equipment known
as a DEXA machine.
The DEXA machine is familiar to many in its use to check the bone density of
older women and men. This machine can measure bone and also fat tissue and
lean tissue. Tests are done in a “child friendly" manner under the supervision of
a specialist physician and a certiﬁed technician who check the equipment daily
for proper functioning. The full body scan will take 2-4 minutes and will expose
your child to 0.04 mRem of radiation, a very low dose exposure. In comparison.
a television puts out 10.0 mRem of radiation over the course of a year while a
typical air flight in North America exposes the passenger to a 40.0 mRem dose.
A standard medical x-ray carries a 40.0 mRem dose. A full report of the body
scan results will be provided to the parent/guardian.

Protection of your child’s privacy
All of the information collected will be confidential. To the maximum extent
allowed by law, no information will be released without your consent. Once
collected, the information about your child will be given an anonymous research
code and then combined with the rest of the data. In case of an urgent need to
contact the parent/guardian about any results, a coded ID card will be kept in a
secure place. separate from the results and available only to the primary
investigators and project manager.

112

Costs of participation

Your child’s rights

Parents/guardians will need to transport children to two (2) study appointments
and to one (1) appointment without the child. They may need to take time off of
work to do so and may choose to hire a babysitter for siblings during that time.
Every effort will be made to schedule appointments at the convenience Of the
family. Your child will receive a set of tests conducted at no cost to the family.
Your participation in the study activities and procedures listed above will not
involve any additional costs to you or your health care insurer.

Beneﬁts and risks of participation to the parent/guardian and to the child
Each child will receive:
A thorough diet analysis.
A summary of physical activity patterns.
Body composition measures (bone size. height. weight, % lean mass. % body fat)
Results of whole body scan. bone scan and Vitamin D blood test
Compensation of $100 divided to allow some compensation at each study visit

($25 at appointments one and two. $50 at appointment three).

Health information and/or referrals for area health programs.

TFIT“ ‘
‘ 4}

We currently base diet recommendations on a child's age. It is possible that we
should consider more than just age in determining the nutritional needs of
children so that we can improve the health of this special group of children
between 5 years and puberty. Your child’s participation in this project will help us
in determining these special needs.

As speciﬁed in the sections detailing activities and procedures. risks of
participation include mild discomfort of the needle when blood is drawn and
bruising at the site of the bloOd draw. very low dose radiation exposure with the
DEXA procedure and inconvenience to the family for appointments. ﬁlling out
surveys and collecting un'ne.

Participation in this study is entirely voluntary. Refusal to participate will involve
no penalty or loss of beneﬁts to which the child is otherwise entitled. The child
may discontinue participation at any time without penalty or loss of beneﬁts to
which the child is otherwise entitled. including the partial compensation received
through the time of withdrawal. NO further compensation from the study will be
awarded once participation is discontinued. Any results or information that are
obtained which pose an immediate health concern will be brought to the
parent/guardian’s attention for referral to the child’s primary care provider.

 

113

If your child is injured as a result of his/her participation in this research project. Michigan State
University will assist you in obtaining emergency medical care. if necessary. for research related
injuries. If your child has insurance for medical care. your child’s insurance carrier will be billed in
the ordinary manner. As with any medical insurance. any costs that are not covered or in excess
of what are paid by your child’s insurance. including deductibles. will be your responsibility. The
University’s policy is not to provide ﬁnancial compensation for lost wages. disability. pain or
comfort. unless required by law to do so. This does not mean that you are giving up any legal
rights your child may have.

For questions or concerns about this study or to report a research-related injury, contact
Jenny Bond. Ph.D.. R.D.. Primary Investigator. 517—676—2676. jbond@msu.edu. For questions
or concerns about research subjects’ rights or a research-related injury, contact Peter
Vasilenko. Ph.D.. Director of Human Research Protecions. 202 Olds Hall. MSU, East Lansing, Mi.
48824-1047, 517-355-2180. fax 517-432—4503. e-mail irb@msu.edu.

Your signature below indicates that you have read this informed consent and your
questions, if any. have been answered by the researchers such that you agree to have
your child participate in all described aspects of this research study.

 

 

Signature of Parent/Guardian Date

 

 

Signature of Child (if age appropriate) Date

 

aiild’s verbal assent obtained - conﬁrmation by parent

 

 

Print Name of Child who is participating in the study

UCRIHS Approval

114

 

Appendix 3

Are you a parent
of a child over 4 years of age?

Are you interested in helping your child keep his or her
diet, physical activity and growth in a healthy
balance?

Michigan State University
Department of Food Science and Human Nutrition
is conducting a study of healthy children, looking at relationships
between nutrition, body composition, bone growth and physical
activity.

WHO?
Children who are 5 years old and older and who have not yet entered puberty. Siblings welcome.

WHAT?

Measurements of the body such as height and weight will be taken. A whole body scan for bone
growth and body composition will be done and there will be two blood samples taken.
Physical activity patterns will be followed. All tests and assessments will be paid for by the
study.

WHEN?
If qualiﬁed, children will be seen three or four times over 6 months for tests and data collection.
Children can enter the study at any time during the enrollment period.

WHERE?
On and near the MSU campus.

BENEFITS?

Families will receive complete reports and explanations of all tests that will include physical
activity details, results of the body scans and body measurements and complete diet nutrient
analysis. Compensation of up to $100 is offered. All information is conﬁdential.

INTERESTED?
If you have questions or would like to get more information:
Leave your name and number at (517)353-3037
OR
Send us an e-mail so we can contact you (msuhkns@hotmail.com)

Healthy Kids Nutrition Study

B 205 E West Fee Hall

Department of Family and Community Medicine

and Department of Food Science and Human Nutrition
Michigan State University

East Lansing, MI 48824

517-353-3037 leave a message

or e-mail msuhkns @hotmail.com

115

Appendix 4

 

PATIENT AUTHORIZATION FOR DISCLOSURE
OF HEALTH INFORMATION FOR RESEARCH

 

 

 

Patient Name:
Address:
Date of Birth: SS#:_not required

 

 

 

 

 

I AUTHORIZE THE DISCLOSURE OF MY HEALTH INFORMATION

 

 

 

F ROM:_Dr. Justus Fiechtner, MD TO:_Healthy Kids Nutrition Study
Name of hospital or health care system or provider Name of researcher or research group
3394 Jolly Road Dept of Food Science and Human Nutrition
Address Address
Lansing, MI 48910 _Michigan State University, East Lansing, MI 48824_
517-272-9727 Study speciﬁc phone number added here_
Phone/Fax Number Phone/Fax Number

DESCRIPTION OF INFORMATION TO BE DISCLOSED (select one):

__ALL information contained in my medical record.
.0_R
_X_ ONLY disclose the following information: Results of DEXA whole body scan

RESEARCH STUDY FOR THIS DISCLOSURE:

Title of Study: Examination of diet. physical activity. biomarkers of bone
mineralization, bone mineral content and body composition in children between 5
Bars of age and puberty

 

Name of Research Leader: Dr. Jenny T. Bond, PhD, RD and Dr. R. Taylor Scott, DO
Afﬁliation of Researcher: Michigan State University
IRB# 03-1019 Name of IRB same as project title

 

 

EXPIRATION (ﬁll in one of the following):

Your Authorization to disclose the above information expires on__9 months after enrollment in the study
(see consent date)_; or

Has no expiration date ; or

Expires at the end of the research study ; or

Expires six months from the date signed

116

Appendix 4 continued

REVOCATION, REFUSAL, REDISCLOSURE:

You may revoke this Authorization in writing at any time by contacting the ofﬁce of Dr. F iechtner (see
above) (e.g., the healthcare system or provider or hospital named above), but it will not affect any
information already released to the researcher(s).

You may refuse to sign this authorization and your refusal will not affect your ability to obtain treatment,
however, it may affect your ability to participate in this research study.

Your information that is disclosed to the researcher(s) may no longer be protected by Federal privacy
regulations if the researcher(s) is not a health care provider covered by the regulations, however the
researcher(s) agrees to protect your information as required by law.

 

 

Signature of Patient or Personal Representative Date

 

 

Name of Personal Representative and Relationship to Patient (or description of authority to act on behalf of
the patient)

PROVIDE COPY TO PATIENT

117

Appendix 5

Healthy Kids Nutrition Study Health History Questionnaire

Child’s Name:

 

Parent/Guardian providing this information:

 

A child’s growth is determined by many factors including the child’s health history as well as
the health history of the child’s family. Your answers to these questions about your child
provide additional insight into your child’s growth pattern.

The ﬁrst page of questions conﬁrm that your child is eligible to participate in this study based 9“
on his/her health history. The second set of questions provides information about your child’s

health and growth history as well as his/her family’s health history. All information is

conﬁdential. Your child’s name will be removed from this questionnaire and replaced with an

anonymous identiﬁer. Your answers to the ﬁrst page of questions are required for participation

in this study. Your answers to the second set of questions provide very important and helpful

information but completion is not required for participation.

Child’s Health History

Your child’s date of birth:

 

Your child’s birth weight:

 

How many weeks or days from the predicted birth due date was your child born?

 

Has your child been previously diagnosed with any medical condition?

 

Has your child been hospitalized for any medical condition?

 

Please describe and list approximate dates of diagnosis and treatment.

Has your child been seen by a doctor, nurse, nutritionist or any other health care professional for
any kind of growth or nutritional problem?

 

Please describe and list the approximate dates of diagnosis and treatment.

118

 

Appendix 5 continued

Your child’s pattern of growth
Circle the closest description for your child's growth from birth to today:

When your child was

0-12 months old small for age average for age large for age

12-24 months old

2 - 3 years old
3 - 4 years old
4 - 5 years old
5 — 6 years old
6 — 7 years old
7 — 8 years old

9 — 10 years old

small for age
small for age
small for age
small for age
small for age
small for age
small for age

small for age

average for age
average for age
average for age
average for age
average for age
average for age
average for age

average for age

large for age
large for age
large for age
large for age
large for age
large for age
large for age

large for age

10 — 1 1 years old small for age average for age large for age

11 — 12 years old small for age average for age large for age

Family Health History - questions speciﬁcally related to growth and nutrition

Please list the child’s family members who you consider to be overweight or underweight. You
need not list names, rather list the relationship to the child, for example maternal
grandmother/underweight.

Does anyone in the child’s family have a history of weak bones? List their relationship and
what you know about the bone problem. Also note the age of the family member when he/she
developed the bone problem.

Does anyone in the child’s family have a history of any kind of nutritional problems? List their
relationship and what you know about the problem. Also note the age of the family member
when he/she developed the nutritional problem.

Can you think of any other health issues related to hormones or metabolism, body size or bone

growth in the child’s family? Please list the family relationship and the health issue and age at
which the health issue began, if known.

119

Appendix 6

Debrieﬁng Protocol
Study Exit Interview Checklist

Study investigators will meet with each child and parent! guardian at the completion of
the child’s study enrollment. This meeting will be conducted in a conﬁdential
environment, allowing at least one-hour time. Written results and summaries will be
provided in duplicate for the convenience of the parent! guardian in sharing with their
child’s primary health care provider. This interview will be conducted by a health care
professional experienced in pediatric health counseling. At the meeting, the following
topics will be covered:

1. Review of all dietary information obtained with smnmary and explanation of
the nutrients assessed, giving special attention to inadequate and excessive
intakes as compared to current dietary recommendations.

 

2. Review physical activity data, comparing activity records obtained with
pediatric health guidelines for physical activity.

3. Review blood test results and interpret the results for the parent/ guardian and
child.

4. Review DEXA scan results and interpret the results for the parent/ guardian
and child.

5. Review and explain the anthropometric measurements obtained (height and
weight). Measurements will be recorded on a pediatric growth chart provided
for the parent/guardian’s records.

6. Provide the parent/ guardian with information on any health concerns and with
referrals/contacts for health programs in which they may be interested.
Special attention will be paid to addressing the parent] guardian and/or child’s
concerns, if any, related to any negative aspects of their body size, as well as
concerns about growth, health or development and chronic disease risk or
health concern of any kind.

7. Final study compensation payment will be dispersed as well as an additional
small, age appropriate gift for the child.

120

 

Appendix 7

SUBJECT ID

 

Standing Order

DEXA Total Body Scan (pediatric)
At the ofﬁce of Justus Fiechtner, MD
3394 Jolly Road
Lansing, MI 48910
517-272-9727

Scan to be conducted in conjunction with the Healthy Kids Nutrition Study, MSU

Date:

 

Study PI:

 

R. Taylor Scott, DO

121

I‘m—_ﬁ

Appendix 8
Study Protocol Checklist

Date/initial each item as completed
Preliminary recruitment contact:

Initial qualifying contact:

Mail map/parking pass/health form:
First Study Appointment at Fee:
Subject contact information:

Review health questionnaire:

Review Tanner stages:

Child DOES meet study criteria
Child DOES NOT meet study criteria

Read/review/sign consent:

 

Read/review/sign HIPAA:
Obtain weight: /
Obtain height: /

 

Obtain usual food intake:

Provide parents to do list:

Instruct Food, Fitness, Fun time survey:
Instruct Eating survey (record #):
Instruct urine collection:

Make Fiechtner Research appointment:

Date/time:

 

Map to F iechtner Research:
Activity monitor instructions:
Date for second appointment:

Date/time:

 

Give gift certiﬁcate #1 ($25):
Target or Meijer (circle)

122

Subject #

 

Second Study Appointment at Fee:
Drop off urine for DPD and creatinine:
Check cap seals and labeling of urine:
(Urine) To storage in Anthony:
Collect Eating survey:
Collect Fitness, Fun and Free time survey:
Review surveys for completeness:

FF Q:

Activity:
Conﬁrm Fiechtner appointment:
Collect activity monitor:
Review monitor experience:
Give gift certiﬁcate #2 and #3 ($50):
After second appointment:

Download monitor data:

Sterilize monitor belt:

Third study appointment:
At F iechtner Research

DEXA:
Serum for Vitamin D:
Serum for CC:

Gift certiﬁcate #3 ($25):

After tl_|ird study appointment:

 

Vitamin D and CC to storage:

Fourth Study Appointment:
(OPTIONAL)

Debrieﬁng per protocol:

Appendix 9

Nutrient intake as measured by usual, one-day dietary recall by age

 

 

 

Variable Age
< 7.6 yrs (n=28) > 7.6 yrs (n=24) Total (n=52)
Total kcals Mean a SD 2,060 a: 595 2,319 a 451 2108 a 544
Range 1.1233152 1.4733115 1,123-3,152
Protein (g) Mean :h SD 74 :l: 20 84 a 19 79 a 20
Range 37 -124 50 -119 37 -124
Cholesterol Mean a SD 189 a 116 230 a 131 208 :1: 124
(mg) Range 25 - 539 87 - 607 25 - 607
Carbohydrate Mean :1: SD 289 d: 100 315 :t 60 301 i 84
(8) Range 102-511 211-444 102-511
Fiber (g) Mean 3: SD 18.2 a 8.7 17.3 a 4.6 17. a 7.1
Range 2.0 - 41.0 10.0 - 28.0 2.0 - 41.0
Fat (g) Mean 3: SD 71.5 a 26.3 84.7 :1: 25.9 77.6 :1: 26.7
Range 31.5-171.3 34.3-131.5 31.5-171.3
Vit A (mcg Mean a SD 941 a 458 1,070 a: 819 1,001 a 646
RAE) Range 271 - 1,753 152 - 4,359 152 - 4,359
Vit C' (mg) Mean a SD 85.0 a 70.3 126.6 a 76.8 104.2 a 75.6
Range 4.0 - 327.8 38.0 - 350.5 4.0 - 350.5
Vit E (mg) Mean a SD 6.7 a 4.4 6.0 a 2.3 6.4 a 3.6
Range 2.1- 21.1 2.0 - 10.6 2.0 - 21.1
Ca (mg) Mean 3: SD 1,309.4 a 555.7 1,467.5 a 609.5 1,382.4 a 580.9
Range 259.9 - 2,505.0 661.0 - 2,715.0 252.9 - 2,715.0
P (mg) Mean a SD 1,509.7 :1: 448.7 1,651.2 a 493.9 1,575.0 a 470.9
Range 645.4 - 2,306.4 1,110.5 - 2,766.0 645.4 - 2,766.0
Mg (mg) Mean a SD 306.3a 104.3 319.5 a 77.1 312.4 a 92.1
Range 104.5 - 539.2 219.6 - 486.2 104.5 - 539.2
Fe (mg) Mean :t SD 16.2 d: 7.0 19.6 :t 8.8 17.7 i 8.0
Range 5.1 - 32.1 8.1 - 48.6 5.1- 48.6
Zn (mg) Meana SD 11.8dz3.4 13.1153 12.4a4.4
Range 3.9 - 17.0 6.4 - 28.7 3.9 - 28.7
K (mg) Mean 3: so 2,702 a 922 2,888 a 920 2,788 a 917
Range 1,087 - 4,232 1,307 - 5,199 1,087 - 5,199
Se (meg) Mean a SD 98.2 a 41.1 103.1 a 30.6 100.4 a 36.3
Range 36.3 - 215.0 52.7 - 188.8 36.3 - 215.0

 

1 Signiﬁcant difference between younger versus older subjects per t-test (p_<_0.05)

123

Appendix 10

Nutrient intake as measured by food frequency questionnairel
with and without supplements by age

 

 

 

Variable Age
< 7.6 yrs (n=28) > 7.6 yrs (n=24) Total (n=52)
Total kcals Mean 3 SD 2313 a 438 2244 :1: 521 2281 a 474
Range 1385 - 3353 1265 - 3327 1265 - 3353
Protein (g) Mean :h SD 97 i 17 92 :l: 24 95 i: 21
Range 57 -126 37 -137 37 -137
Animal protein in Mean a SD 66 a 16 62 a 18 64 a 17
diet (a) Range 38 - 98 25 - 106 25 - 106
Non-animal protein Mean 2t SD 31 d: 10 29.42 i: 9 30 :t 10
in diet (8) Range 14 - 64 12 - 46 12 - 64
Cholesterol (g) Mean 1 SD 246 :t 62 251 :l: 81 248 a: 71
Range 94 - 325 111 - 388 94 - 388
Carbohydrate (g) Mean :t SD 322 d: 72 299 :l: 78 311 i 75
Range 165-460 157-461 157-461
Fiber (g) Mean 3: SD 20.5 a 6.5 19.2 a 6.9 19.9 :1: 6.6
Range 9.2 - 35.5 7.7 - 36.9 7.7 - 36.9
Fat (g) Mean a SD 74.5 a 16.2 77.1 a 20.3 75.723: 18.0
Range 49.5 - 118.8 44.0 - 110.2 44.0 - 118.8
K (mg) Mean a so 3287 a 755 3036 a 886 3171 :1: 820
Range 1299 - 4785 1060 - 4864 1060 - 4864
Vit A With Mean a SD 8403 a 4187 7883 a 4398 8163 a: 4251
(meg SUPP Range 2194 - 24348 2045 - 24575 2045 - 24575
RAE) Without Mean a SD 1009.2 a 393.2 1035.7 a 439.1 1021.4 a 411.1
SUPP Range 246.4 - 2204.2 317.2 - 2245.8 246.4 - 2245.8
Vit C With Mean a SD 133.8 a 54.6 124.5 a 67. 129.5 a: 60.3
(mg) SUPP Range 31.8 - 271.1 34.9 - 312.5 31.8 - 312.5
Without Mean a SD 114.3 a 46.8 111.9 a 56.4 113.2 :1- 50.9
SUPP Range 26.2 - 210.7 34.9 - 262.1 26.2 - 262.1
Vit D (iu) With Mean a SD 436 a 168 363 :1: 173 402 a 172
SUPP Range 45 - 863 48 - 604 45 - 863
Without Mean a SD 337 a 135 300 a- 143 320 a 139
SUPP Range 45 - 543 48 - 568 45 - 568
VitE With Meana SD 8.63: 3.1 8.1a3.0 8.4a 3.0
(mg) SUPP Range 3.8 - 15.8 3.3 - 16.3 3.3 - 16.3
Without Mean a SD 6.1 a 1.4 6.7 a 2.0 6.6 a 1.7
SUPP Range 3.2 - 10.0 3.3 - 10.0 3.2 - 10.0

124

Appendix 10 continued

Ca (mg)

P (mg)

Mg (mg)

Fe (mg)

Zn (mg)

VVﬁh
supp
Without
supp
Vth
supp
VVﬁhout
supp
VVﬁh
supp
Without
supp
VVﬁh
supp
Without
supp
VVﬁh
supp
VVﬁhout

.supp

hdean3:S[)
Range
hdewn3:SI)
Range
h4eand:S[)
Range
hdeand:S[)
Range
hAean3:SI)
Range
lAean3:S[)
Range
h4ewnd:S[)
Range
hAeand:S[)
Range
hAeand:SI)
Range
h4ewn3:SI)
Range

1579.9 3 433.0
426.2 - 2,577.2
1540.7 :1: 429.1
426.2 - 2531.6
1842.0 a 361.9
770.3 - 2559.2
1842.0 3: 361.9
770.3 - 2,559.2
345.7 a 77.3
139.6 - 531.1
336.2 :E 73.6
139.6 - 511.1
21.96 :1: 9.8
8.6 - 54.7

17.0 :1: 6.9

8.6 - 44.7

17.7 a 5.9

8.1 - 29.9

13.8 d: 3.6

8.1 - 26.1

1406.2 1 396.5
610.1 - 1,964.0
1381.0 :l: 397.6
610.1 - 1,918.4
1709.6 1 455.4
725.2 - 2,576.2
1709.6 :L' 455.4
725.2 - 2,576.2
319.3 a: 88.4
125.4 - 481.0
313.0 a 85.8
125.4 - 469.6
18.8 :1: 6.7

7.2 - 38.3

15.6 a 3.9

7.2 - 23.7

15.6 a 5.0

6.7 - 30.0

13.0 a 2.9

6.6 - 18.9

1499.7 :1: 421.7
426.2 - 2,577.2
1467.0 a 418.6
426.2 - 2531.6
1780.9 :t 409.0
725.2 - 2,576.2
1780.9 3 409.0
725.2 - 2,576.2
333.5 :1: 82.8
125.4 - 531.1
325.5 a 79.5
125.4 - 511.1
20.4 a 8.5

7.2 - 54.7

16.3 a 5.7

7.2 - 44.7
16.72 :t 5.5

6.7 - 30.0

13.5 a 3.3

6.6 - 26.1

 

I Youth and Adolescent Food Frequency Questionnaire, Harvard Charming Laboratory, Harvard University,

School of Public Health, Boston, MA, USA

125

 

tr:-
i

 

Appendix 11
Dietary intakes according to 2005 US. Dietary Guidelines by age1

 

 

 

Variable Age
< 7.6 yrs (n=28) > 7.6 yrs (n=24) Total (n=52)
Grain intakez Mean 3 SD 7.6 a 4.2 9.2 a 2.8 8.3 a 3.7
Range 0.8 - 22.7 3.3 - 14.9 0.8 - 22.7
Milk intake3 Mean 3 SD 3.2 a 1.8 3.5 a 1.8 3.4 a 1.8
Range 0.2 - 7.0 0.8 - 7.0 0.2 - 7.0
Meat and bean intake2 Mean 3; SD 3.5 a 2.1 3.9 a 1.8 3.7 a 1.9
Range 0.1 - 9.4 0.9 - 6.2 0.1 - 9.4
Fruit and Vegetable Mean i SD 1.9 :l: 1.3 2.4 :1: 1.3 2.2 :t 1.3
intake3 Range 0.1 - 5.5 0.5 - 4.9 0.1 - 5.5

 

12005 US. Dietary Guidelines recommendations for children ages 2 to 8 years: Grain, 6 02.; Milk, 2 cups;
Meat and beans, 5 02.; Fruit, 1.5 cups; and Vegetable 2.5 cups.

2 In oz. equivalents

3 In cup equivalents

126

Appendix 12
Site-Speciﬁc measures of bone mineralization by gender

 

 

 

Variable Gender
Male Female Total
BMD (g/cmz) Mean a SD 0.885 a: 0.062 0.837 a 0.060 0.847 a 0.061
Range 0.764 - 1.003 0.751 - 0.980 0.751 -1.003
N 27 25 52
Z-Scorel (Lunar) Mean 3: SD 0.31 a 0.64 -004 a 0.59 0.14 a 0.64
Range 060 - 1.80 -120 - 1.20 -120 - 1.80
N 27 25 52
BMC (g) Mean a SD 960 a 262 933 a 229 . 947 a 245
Range 618 - 1,657 567 -1,514 567 - 1,657
N 27 25 52
Arm 13MD Mean 3 SD 0.601 a 0.058 0.610 a 0.049 0.605 :e 0.054
(ii/cm ) Range 0.486 - 0.711 0.533 - 0.734 0.486 - 0.734
N 27 24 51
Arm BMC (g) Mean a SD 84 a 28 89 a 32 86 a 30
Range 46 - 154 38 - 190 38 - 192
N 27 24 51
($321)“) Mean :1: SD 0.780 a 0.117 0.785 a 0.107 0.783 a 0.111
Range 0.571 - 1.052 0.639 - 1.067 0.571 - 1.067
N 27 24 51
Leg BMC(g) MeaniSD 3013: 127 3023113 301a119
Range 136 - 593 134 - 598 134 - 598
N 27 24 51
Egg-DMD Mean :1: SD 0.754 a 0.097 0.750 a 0.081 0.752 a 0.089
Range 0.599 - 1.028 0.600 - 0.932 0.599 - 1.028
N 27 24 51
Pelvis BMC (8) Mean a SD 97 a 34 95 a 28 96 a 31
Range 51 -192 46 - 155 46 -192
N 27 24 51

’ Signiﬁcant difference between male and female subjects per t—test (pS0.05)
BMD: Bone Mineral Density
BMC: Bone Mineral Content

127

Appendix 12 continued

Arm BMC (g)

Leg BMD
(g/cm’)

Leg BMC (g)

Pelvis BMD
(g/cm’)

Pelvis BMC (g)

OC1 (ng/mL)

DPD
(nmol/mmol
Creatinine)

Serum 25-OH,
Vit D (ng/mL)

Mean :1: SD
Range

N

Mean i SD
Range

N

Mean :E SD
Range

N

Mean :t SD
Range

N

Mean i SD
Range

N

Mean i SD
Range

N

Mean 1 SD
Range

N

Mean i SD
Range

N

84 a: 28

46 - 154

27

0.780 :t 0.117
0.571 - 1.052
27

301 d: 127
136 - 593

27

0.754 t 0.097
0.599 - 1.028
27

97 i 34

51 - 192

27

23.82 i 4.39
14.71 - 35.57
25

20.00 :1: 5.83
10.81 - 41.80
27

30.40 i 9.35
13.80 - 57.50
25

89 a 32

38 - 190

24

0.785 a: 0.107
0.639 - 1.067
24

302 :1: 113
134 - 598

24

0.750 d: 0.081
0.600 - 0.932
24

95 :l: 28

46 - 155

24

26.98 a: 4.43
17.42 - 33.41
25

20.79 a 6.20
9.02 - 35.83
25

29.47 a 11.29
10.80 - 62.30
21

86 :1: 3O

38 - 192

51

0.783 :1: 0.111
0.571 - 1.067
51

301 :t 119
134 - 598

51

0.752 d: 0.089
0.599 - 1.028
51

96 :l: 31

46 - 192

51

25.40 :h 4.65
14.71 - 35.57
49

20.38 i 5.96
9.02 - 41.80
52

29.98 i 10.17
10.80 - 62.30
46

 

1 Signiﬁcant difference between male and female subjects per t-test (pS0.05)
BMI: Body Mass Index

BMD: Bone Mineral Density
BMC: Bone Mineral Content
0C: Serum osteocalcin

DPD: Urinary deoxypyridinoline

128

Appendix 13

Energy expenditure above BBB and activity counts at 3 levels of intensityl by age

 

 

 

 

Variable Age
< 7.6 yrs (n=28) > 7.6 yrs (n=24) Total (n=52)
v 2 Meand: SD 493 :1: 172 639a238 562a217
E Total
:g Range 281- 973 328 - 1,120 281-1,120
g. Li ht, Meani SD 106a26 145a38 124337
0 g Range 71 - 164 94-234 71 —243
g Moderate Meand: SD 202a51 237:: 110 2185:84
5 Range 119-292 131-549 119—549
€ . Mean3: SD 189:1:142 2353119 2103133
D Vigorous
Range 51 - 576 48 - 525 48 - 576
Mean 3: SD 502914 3 270079 449995 3: 137417 478572 a 218843
Total
Range 194342 - 1120443 192744 - 698463 192744 - 1120443
E MeanaSD 4165a 1028 4411 i880 4278i961
8 Sedentary
g Range 2945 - 5892 1452 - 5617 1452 - 5892
.2.
g Light Mean a SD 44813 a 30491 37450 a 7388 41426 a 23056
:3 Range 25024 - 189367 21841 - 51880 21841 -189367
>~.
T21 Meani SD 140410436704 120627i35616 131310a37196
Q Moderate
Range 70565 — 235209 86386 - 206384 70565 - 235209
. Mean 3; SD 318256 a 256267 278737 a 125181 300078 a 205617
Vigorous
Range 71705 - 904875 58684 - 517570 56684 - 904875

 

IMeasured over 72 hours with an Actical accelerometer (MiniMitter Corp, Bend, OR)

2 Signiﬁcant difference between younger versus older subjects per t-test (pS0.05)
BEE: Basal Energy Expenditure per Harris-Benedict equation estimate (Harris JA, Benedict FG. A
biometric study of basal metabolism in man. (Publication No. 279) Washington, DC: Carnegie Institute of

Washington; 1919).

129

Appendix 14

Energy expenditure above BEE and activity counts at 3 levels of intensity1 by gender

 

 

 

 

Variable Gender
Male (n=27) Female (n=25) Total (n=52)
D Meana SD 609a225 512a200 561:1:217
Total
,5 Range 323 - 1120 281 - 973 281 - 1120
'U
5 , MeaniSD 1243:39 124:1:36 124:1:37
E- Light
a Range 77 - 243 71 - 209 71 - 243
5.5 Meani SD 2373:101 199a61 2183:84
3 Moderate
o Range 119-549 129-351 119-549
>5
7:; _ Meani SD 2261-127 195a139 2103133
Q Vlgorous
Range 73 - 252 48 - 576 48 - 576
502530 3: 454618 3 478572 a
T t 1 Mean * SD 197950 239583 218843
0 a
Ran 6 275700 - 192744 - 192744 -
E g 1071662 1 120443 1 120443
8 Mean a SD 4207 a 1019 4,349 a 914 4,278 a 961
Sedentary
3;: Range 1452 - 5790 2,495 - 5,892 1,452 - 5,892
:3 Li ht Mean 3: SD 37213 a 6888 45639 a 31636 41426 a 23056
g g Range 21841 - 47808 25024 -189367 21841 - 189367
3 Means: SD 141381 a38866 121239433209 131310337196
g; Moderate
3 Range 87126 - 235209 70565 - 188864 70565 - 235209
318685 a 281471 :1: 300077 :1;
Vigorous Mean i SD 201836 211800 205617
Range 91863 - 904875 58684 - 893740 58684 - 904875

 

rMeasured over 72 hours with an Actical accelerometer (MiniMitter Corp, Bend, OR)

BEE: Basal Energy Expenditure per Harris-Benedict equation estimate (Harris JA, Benedict FG. A
biometric study of basal metabolism in man. (Publication No. 279) Washington, DC: Carnegie Institute of

Washington; 1919).

130

Appendix 15

Correlations of BMD and BMC with variables inﬂuencing bone mineralization by

 

 

 

 

 

 

 

 

 

 

 

 

gender

By Gender BMD (g/cmz) BMC (g)
r Slg. r Slg.

% body fat Total (n=51) 0.535 0000** 0.646 0000**
Male (n=27) 0.582 0001** 0.714 0000**

Female (n=25) 0.545 0005** 0.578 0002**

DTEE above BEE Total (n=51) 0.459 0000** 0.591 0.000**
(kcals) Male (n=26) 0.465 0017* 0.645 0.000**
Female (n=25) 0.423 0035* 0.527 0.007“

Protein by weight Total (n=52) 0.508 0000** -O.564 0000**
(gfkg)' Male (n=27) 0.476 0012* 0.523 0.005**
Female (n=25) 0.542 0005** 0616 0001**

kcals by weight Total (n=52) -0.509 0.000** -0.578 0.000“
(kcals/kg)' Male (n=27) -0.670 0.000** 0.680 0000**
Female (n=25) -0.349 0.087 -0.460 0.021 *

Ca by height Total (n=51) 0.277 0047* 0.260 0.062
[(mg/d)/cm]' Male (n=27) 0.015 0.940 0.009 0.963
Female (n=25) -0.614 0.001“ -O.625 0.001“

P by height (g/om)T Total (n=52) 0.378 0006** 0.348 0011*
Male (n=27) -0.206 0.301 0.182 0.365

Female (n=25) 0.604 0001** 0.573 0.003**

Mg (mg)I Total (n=52) 0.135 0.341 0.126 0.372
Male (n=27) 0.259 0.192 0.233 0.241

Female (n=25) 0.042 0.8412 0.003 0.990

K (mg)I Total (n=52) 0.137 0.331 0.069 0.625
Male (n=27) 0.055 0.786 0.018 0.930

Female (n=25) 0.265 0.200 0.148 0.481

Fruit and Total (n=52) 0.265 0.058 0.256 0.067
Vet-Fetable2 Male (n=27) 0.131 0.514 0.161 0.422
Female (n=25) 0.402 0047* 0.376 0.064

RNAE Total (n=52) 0.196 0.163 0.247 0.078
Male (n=27) 0.174 0.387 0.308 0.118

Female (n=25) 0.275 0.183 0.153 0.464

 

** Pearson correlation signiﬁcant at the 0.01 level (2-tailed).

* Pearson correlation signiﬁcant at the 0.05 level (2-tailed).
’ All nutrient intakes as assessed by recall
2 Intake per 2005 US Dietary Guidelines in cup equivalents
BMD: Bone Mineral Density
BMC: Bone Mineral Content

131

Appendix 15 continued

DTEE above BEE: Daily Total Energy Expenditure above Basal Energy Expenditure per Harris-Benedict
equation estimate (Harris JA, Benedict FG. A biometric study of basal metabolism in man. (Publication
No. 279) Washington, DC: Carnegie Institute of Washington; 1919).

RNAE: Renal Net Acid Excretion as estimated by the method of Frassetto LA et al. Estimation of net
endogenous noncarbonic acid production in humans from diet potassium and protein contents. AJCN
1998; 68:576-83.

132

Appendix 16
Correlations of BMD and BMC with variables inﬂuencing bone mineralization by age

 

 

 

 

 

 

 

 

 

 

 

BMD (g/om’) BMC (g)
By Age r Sig. r Ski
% body fat Total (n=52) 0.535 0000** 0.646 0000**
< 7.6 yrs (n=28) 0.373 0050* 0.540 0.003**

> 7.6 yrs (n=24) 0.594 0002** 0.748 0.000**

DTEE above BEE Total (n=51) 0.459 0000** 0.591 0000**
(kcals) < 7.6 yrs (n=27) 0.033 0.869 0.231 0.246
> 7.6 yrs (n=24) 0.591 0002** 0.670 0000**

Protein by weight Total (n=52) -0.508 0.000" -O.564 0.000“
(g/kg)l < 7.6 yrs (n=28) 0.440 0019* 0.572 0001**
> 7.6 yrs (n=24) 0.541 0.006** 0.607 0002**

kcals by weight Total (n=52) -0.509 0.000M -O.578 0.000“
(kcals/kg) < 7.6 yrs (n=28) 0.440 0019* 0.579 0.001**
> 7.6 yrs (n=24) 0.587 0003** 0.663 0000**

Ca by height Total (n=51) 0.277 0047* 0.260 0.062
[(mg/d)/cm]' < 7.6 yrs (n=28) 0.303 0.117 0.329 0.087
> 7.6 yrs (n=24) 0.292 0.166 0.307 0.144

P by height (g/om)1 Total (n=52) 0.378 0.006** 0.348 0011*
< 7.6 yrs (n=28) 0.400 0035* 0.428 0023*

> 7.6 yrs (n=24) 0.383 0.065 0.366 0.079

Mg (mg)I Total (n=52) 0.135 0.341 0.126 0.372
< 7.6 yrs (n=27) 0.284 0.143 0.348 0.07

> 7.6 yrs (n=24) 0.074 0.73 0.084 0.697

K (mg)1 Total (n=52) 0.137 0.331 0.069 0.625
< 7.6 yrs (n=28) -0.076 0.701 0.134 0.497

> 7.6 yrs (n=24) 0.285 0.177 0.185 0.386

Fruit and Total (n=52) 0.265 0.058 0.256 0.067
Vegetable2 < 7.6 yrs (n=28) 0.354 0.065 0.203 0.301
> 7.6 yrs (n=24) 0.099 0.647 0.164 0.444

RNAE Total (n=52) 0.196 0.163 0.247 0.078
< 7.6 yrs (n=28) 0.125 0.525 0.282 0.145

> 7.6 yrs (n=24) 0.185 0.386 0.137 0.525

 

** Pearson correlation signiﬁcant at the 0.01 level (2-tailed).
* Pearson correlation signiﬁcant at the 0.05 level (2-tailed).

’ All nutrient intakes as assesed by recall
2 Intake per 2005 US Dietary Guidelines in cup equivs
BMD: Bone Mineral Density
BMC: Bone Mineral Content

133

 

Appendix 16 continued

DTEE above BEE: Daily Total Energy Expenditure above Basal Energy Expenditure per Harris-Benedict
equation estimate (Harris JA, Benedict F G. A biometric study of basal metabolism in man. (Publication No.
279) Washington, DC: Carnegie Institute of Washington; 1919).

RNAE: Renal Net Acid Excretion as estimated by the method of Frassetto LA et al. Estimation of net
endogenous noncarbonic acid production in humans from diet potassium and protein contents. AJCN

1998; 68:576-83.

134

 

 

Appendix 17
Correlations of BMD and BMC by height with variables inﬂuencing bone mineralization

 

 

 

 

 

 

 

 

 

 

 

 

by gender

BMD (g/cm2) BMC/ht (g/cm)

By Gender Pearson Pearson .
Correlation lg' Correlation Slg'
% body fat Total (n=51) 0.535 0.000** 0.646 0.000**
Male (n=27) 0.582 0.001** 0.729 0.000**
Female (n=25) 0.545 0.005** 0.591 0002**
Protein by weight Total (n=52) 0.508 0.000** 0.603 0.000**
(g/kg) Male (n=27) 0.476 0012* 0.553 0.003**
Female (n=25) 0.542 0.005** 0.685 0.000**
Ca by height‘ Total (n=52) 0.277 0047* 0.265 0.058
[(mg/d)/cm] Male (n=27) 0.015 0.94 0.041 0.838
Female (n=25) 0.614 0.001** 0.624 0.001**
DEE light above Total (n=50) 0.626 0.000** 0.762 0.000**
BEE (kcals) Male (n=25) 0.711 0.000** 0.855 0.000**
Female (n=25) 0.541 0005** 0.659 0.000**
DEE moderate Total (n=50) 0.567 - 0.000** 0.694 0.000**
above BEE (kcals) Male (n=25) 0.605 0.001** 0.738 0.000**
Female (n=25) 0.514 0.009** 0.583 0.002**
DEE vigorous above Total (n=50) 0.191 0.183 0.271 0.057
BEE (kcals) Male (n=25) 0.138 0.51 0.286 0.165
Female (n=25) 0.220 0.291 0.237 0.254
oc by height Total (n=50) 0.507 0.000** 0.480 0.000**
((ng/mL)/cm) Male (n=25) 0.419 0037* 0.313 0.128
Female (n=25) 0.568 0.003** 0.653 0.000**
DPD by height Total (n=52) 0.245 0.079 0.180 0.201
((nmol/mmol Male (n=27) 0.006 0.978 0.121 0.548
Cr)/cm) Female (n=25) 0.450 0024* 0.493 0012*
Serum 25-OH, Vit D Total (n=49) 0.028 0.848 0.028 0.849
(ng/mL) Male (n=26) -0.116 0.573 0.008 0.969
Female (n=23) 0.165 0.452 0.063 0.774
RNAE Total (n=52) 0.196 0.163 0.247 0.078
Male (n=27) 0.174 0.387 0.287 0.147
Female (n=25) 0.275 0.183 0.240 0.249

 

** Pearson correlation signiﬁcant at the 0.01 level (2-tailed).
* Pearson correlation signiﬁcant at the 0.05 level (2-tailed).

' All nutrient intakes as assessed by recall
BMD: Bone Mineral Density

BMC: Bone Mineral Content

135

Appendix 17 continued

DEE Light/ModerateNigorous above BEE: Daily Energy Expenditure Light/Moderate/Vigorous above
Basal Energy Expenditure per Harris-Benedict equation estimate (Harris JA, Benedict FG. A biometric
study of basal metabolism in man. (Publication No. 279) Washington, DC: Carnegie Institute of
Washington; 1919).

0C: Serum osteocalcin

DPD: Urinary deoxypyridinoline

Cr: Creatinine

RNAE: Renal Net Acid Excretion as estimated by the method of Frassetto LA et al. Estimation of net
endogenous noncarbonic acid production in humans from diet potassium and protein contents. AJCN
1998; 68:576-83.

136

Appendix 18
Correlations of BMD and BMC by height with variables inﬂuencing bone mineralization

 

 

 

 

 

 

 

 

 

 

 

by age
BMD (g/cmz) BMC/ht (g/cm)

B A e Pearson . Pearson .
y g Correlation Slg' Correlation Slg'
% body fat Total (n=52) 0.535 0.000** 0.646 0.000**
< 7.6 yrs (n=28) 0.373 0050* 0.508 0006**
> 7.6 yrs (n=24) 0.594 0.002** 0.726 0.000**
Protein by weight Total (n=52) 0.508 0.000** 0.603 0.000**
(g/kg) < 7.6 yrs (n=28) 0.440 0019* 0.594 0.001**
> 7.6 yrs (n=24) 0.541 0006** 0.623 0.001**
Ca by height1 Total (n=52) 0.277 0047* 0.265 0.058
[(mg/d)/crn] < 7.6 yrs (n=28) 0.303 0.117 0.333 0.084
> 7.6 yrs (n=24) 0.292 0.166 0.267 0.207
DEE light above BEE Total (n=52) 0.626 0.000** 0.762 0.000**
(kcals) < 7.6 yrs (n=27) 0.462 0015* 0.647 0.000**
> 7.6 yrs (n=23) 0.575 0.004** 0.702 0.000**
DEE moderate above Total (n=52) 0.567 0.000** 0.694 0.000**
BEE (kcals) < 7.6 yrs (n=27) 0.242 0.223 0.385 0047*
> 7.6 yrs (n=23) 0.647 0.001** 0.822 0.000**
DEE vigorous above Total (n=52) 0.191 0.183 0.271 0.057
BEE (kcals) < 7.6 yrs (n=27) 0.230 0.248 0.083 0.679
> 7.6 yrs (n=23) 0.466 0025* 0.517 0011*
QC by height Total (n=52) 0.507 0.000** 0.480 0.000**
((ng/mL)/cm) < 7.6 yrs (n=26) 0.381 0.054 0.419 0033*
> 7.6 yrs (n=24) 0.561 0.004** 0.447 0029*
DPD by height Total (n=52) 0.245 0.079 -0.180 0.201
((nmol/mmol Cr)/cm) < 7.6 yrs (n=28) 0.261 0.18 0.239 0.22
> 7.6 yrs (n=24) 0.112 0.602 0.049 0.821
Serum 25-01—1, Vit D Total (n=49) 0.028 0.848 0.028 0.849
(ng/mL) < 7.6 yrs (n=27) 0.062 0.759 0.215 0.281
> 7.6 yrs (n=22) 0.143 0.527 0.183 0.415
RNAE Total (n=52) 0.196 0.163 0.247 0.078
< 7.6 yrs (n=27) 0.125 0.525 0.255 0.191
> 7.6 yrs (n=23) 0.185 0.386 0.186 0.384

 

 

** Pearson correlation signiﬁcant at the 0.0] level (2-tailed).
* Pearson correlation signiﬁcant at the 0.05 level (2-tailed).

' All nutrient intakes as assesed by recall

BMD: Bone Mineral Density

BMC: Bone Mineral Content

137

Appendix 18 continued

DEE Light/Moderate/Vigorous above BEE: Daily Light/ModerateNigorous Energy Expenditure above
Basal Energy Expenditure per Harris-Benedict equation estimate (Harris JA, Benedict FG. A biometric
study of basal metabolism in man. (Publication No. 279) Washington, DC: Carnegie Institute of
Washington; 1919).

0C: Serum osteocalcin

DPD: Urinary deoxypyridinoline

Cr: Creatinine

RNAE: Renal Net Acid Excretion as estimated by the method of F rassetto LA et al. Estimation of net
endogenous noncarbonic acid production in humans from diet potassium and protein contents. AJCN
1998; 68:576-83.

138

 

Appendix 19
Multiple regression model predicting BMC by height as a function of serum DC by
height, DEE at light intensity above BBB, and DEE at moderate intensity above BEE

 

 

 

Predictors B
OC by height ((ng/mL)/cm) -9.057]
DEE Light Above BEE 0.017l
DEE Moderate Above BEE 0.0061

R2 = 73.40%
‘ p < 0.01

0C: Serum osteocalcin

DEE Light/Moderate above BEE: Daily Energy Expenditure at
light/moderate intensity level above Basal Energy Expenditure per
Harris-Benedict equation estimate (Harris JA, Benedict FG. A
biometric study of basal metabolism in man. (Publication No. 279)
Washington, DC: Carnegie Institute of Washington; 1919).
Variables entered into the regression model (stepwise): protein
intake by weight, calcium intake by weight, % body fat, serum 0C
by height, urinary DPD by height, serum 25-OH Vitamin D, DEE
above BEE at light intensity, DEE above BEE at moderate
intensity, DEE above BEE at vigorous intensity.

139

Appendix 20

Crosstabs of dependent variable categories with independent variable categories

20a: Crosstabs of bone mineral density categories with variables inﬂuencing bone

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

mineralization
% Fat mass by percentile BMD by categories Total
Low BMD Medium BMD High BMD
Less than the 15th percentile 5 0 2 7
Between 15th percentile to 11 11 15 37
less than the 85th percentile
Equal to or greater than the 0 1 7 8
85th percentile
Total 16 12 24 52
Ca by height by categories BMD by categories Total
Low BMD Medium BMD High BMD
Low Ca by height 4 3 9 16
Medium Ca by height 3 4 5 12
High Ca by height 9 5 10 24
Total 16 12 24 52
P by height by categories BMD by categories Total
Low BMD Medium BMD High BMD
Low P by height 2 4 10 16
Medium P by height 3 3 6 12
High P by height 11 5 8 24
Tgal 16 12 24 52
Mg by categories BMD by categories Total
Low BMD Medium BMD High BMD
Low Mg 4 6 6 16
Medium Mg 2 1 9 12
High Mg 10 5 9 24
Total 16 12 24 52
Fruit and Vegetable intake by BMD by catgories Total
categories Low BMD Medium BMD High BMD
Less than 2 cup equiv. 11 7 10 28
Between 2 and 3 cup equiv. 3 3 4 10
More than 3 cup equiv. 2 2 10 14
Total 16 12 24 52

 

140

 

Appendix 20a continued

 

 

 

Kcals by weight by categories BMD by categories Total
Low BMD Medium BMD High BMD

Low kcals by weight 2 4 10 16

Medium kcals by weight 2 3 7 12

High kcals by weight 12 5 7 24

Total 16 12 24 52

 

141

 

 

20b: Crosstabs of bone mineral content categories with variables inﬂuencing bone

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

mineralization
% Fat mass by percentile BMC by categories Total
Low BMC Medium High BMC
BMC
Less than the 15th percentile 4 2 1 7
Between 15th percentile to less 12 9 16 37
than the 85th percentile
Equal to or greater than the 85th 0 1 7 8
percentile
Total 16 12 24 52
Ca by height by categories BMC by categories Total
Low BMC Medium High BMC
BMC
Low Ca by height 3 3 10 16
Medium Ca by height 1 5 6 12
High Ca by height 12 4 8 24
Total 16 12 24 52
P by height by categories BMC by categories Total
Low BMC Medium High BMC
BMC
Low P by height 1 5 10 16
Medium P by height 2 4 6 12
High P by height 13 3 8 24
Total 16 12 24 52
Mg by categories BMC by categories Total
Low BMC Medium High BMC
BMC
Low Mg 3 6 7 16
Medimn Mg 3 1 8 12
High Mg 10 5 9 24
Total 16 12 24 52
Fruit and Vegetable intake by BMC by categories Total
categories Low BMC Medium High BMC
BMC
Less than 2 cup equiv. 10 7 11 28
Between 2 and 3 cup equiv. 3 3 4 10
More than 3 cup equiv. 3 2 9 14
Total 16 12 24 52

 

142

Appendix 20b continued

 

 

 

DTEE above BEE by categories BMC by categories Total
Low BMC Medium High BMC
BMC

Low daily energy expenditure 8 7 2 17
Medium daily energy 5 3 8 16
expenditure

High daily energy expenditure 3 2 13 18
Total 16 12 23 51

 

 

143

 

20c: Crosstabs of bone mineral density categories with variables related to bone

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

mineralization
% Fat mass by percentile BMD by categories Total
Low BMD Medium High BMD

BMD
Less than the 15th percentile 5 0 2 7
Between 15th percentile to less 11 11 15 37
than the 85th percentile
Equal to or greater than the 85th 0 1 7 8
percentile
Total 16 12 24 52
DC by height by categories BMD by categories Total

Low BMD Medium High BMD

BMD
Low DC by height 3 1 1 1 15
Medium OC by height 2 4 10
High OC by height 11 7 7 25
Total 16 12 22 50
Urinary DPD by height by BMD by categories Total
categories Low BMD Medium High BMD

BMD
Low Urinary DPD by height 4 5 7 16
Medium Urinary DPD by height 1 3 8 12
High Urinary DPD by height 11 4 9 24
Total 16 12 24 52
Serum Vitamin D, 25-OH by BMD by categories Total
categories Low BMD Medium High BMD

BMD
Low Serum Vitamin D 3 6 5 14
Medium Serum Vitamin D 4 1 5 10
High Serum Vitamin D 8 5 12 25
Total 15 12 22 49

 

144

 

Appendix 200 continued

 

 

 

DEE at moderate level above BMD by categories Total
BEE by categories Low BMD Medium High BMD

BMD
Low DEE at moderate activity 7 6 2 15
level
Medium DEE at moderate 3 5 8 16
activity level
High DEE at moderate activity 6 1 12 19
level
Total 16 12 22 50

 

 

145

20d: Crosstabs of bone mineral content by height categories with variables related to bone

 

 

 

 

 

 

 

 

 

 

 

mineralization
OC by height by categories BMC by height by categories Total
Low BMC Medium High BMC
per cm BMC per cm per cm
Low DC by height 3 3 9 15
Medium OC by height 2 2 6 10
High OC by height 11 6 8 25
Total 16 11 23 50
DEE at light activity level above BMC by height by categories Total
BEE by categories Low BMC Medium High BMC
per cm BMC per cm per cm
Low DEE at light activity level 1 1 4 1 16
Medium DEE at light activity 5 4 6 15
level
High DEE at light activity level 0 4 15 19
Total 16 12 22 50
DEE at moderate level above BMC byheight by categories Total
BEE by categories Low BMC Medium High BMC
per cm BMC per cm per cm
Low DEE at moderate activity 7 4 4 15
level
Medium DEE at moderate 3 7 6 16
activity level
High DEE at moderate activity 6 1 12 19
level
Total 16 12 22 50

 

146

 

 

REFERENCES

 

147

REFERENCES

Abrams SA. Calcium supplementation during childhood: long-term effects on bone
mineralization. Nutrition Reviews 2005;63:251-5.

Abrams SA, Copeland KC, Gunn SK, Gundberg CM, Oerter Klein K, Ellis KJ. Calcium
absorption, bone mass accumulation, and kinetics increase during early pubertal
development in girls. Journal of Clinical Endocrinology and Metabolism
2000;85:1805-9.

Acil Y, Brinckmann J, Hotbohm H, Muller PK. Changes with age in the urinary excretion
of hydroxylysylpyridinium (HP) and lysypyridinium (LP). Scandanavian Journal
of Clinical Laboratory Investigation 1996;56:275-83.

Allen JR, Humphries IR, Waters DL, Roberts DC, Lipson AH, Howman-Giles RG,
Gaston KJ. Decreased bone mineral density in children with phenylketonuria.
American Journal of Clinical Nutrition 1994;59:419-22.

American Academy of Pediatlics, Committee on Nutrition. Policy Statement: Prevention
of Pediatric Overweight and Obesity. Journal of Pediatrics 2003;112:424-30.

Anderson JJB, Klemmer PJ, Watts MLS, Garner SC, Clavo MS. Phosphorus. In:
Bowman BA, Russell RM, eds. Present Knowledge in Nutrition, Ninth Edition.
Washington DC: ILSI Press, 2006:383-99.

Anderson LF, Lande B, Arsky GH, Trygg K. Validation of a semi-quantitative food-
frequency questionnaire used among 12-month-old Norwegian infants. European
Journal of Clinical Nutrition 2003;57:881-8.

Baer MT, Kozlowski BW, Bliley EM, Trahms CM, Taylor ML, Hogan MP. Vitamin D,
calcium, and bone status in children with developmental delay in relation to
anticonvulsant use and ambulatory status. American Journal of Clinical Nutrition
1997;65:1042-51.

Bailey DA, Faulkner RA, McKay HA. Growth, Physical Activity, and Bone Mineral
Acquisition. Exercise and Sport Sciences Reviews 1996;24:233-66.

Bailey DA, Martin AD, McKay HA, Whiting S, Mirwald R. Calcium accretion in girls
and boys during puberty: a longitudinal analysis. Journal of Bone and Mineral
Research 2000;15:2245-50.

Bailey DA, McKay HA, Mirwald RL, Crocker PRE, Faulkner RA. A six-year
longitudinal study of the relationship of physical activity to bone mineral accrual
in growing children: the University of Saskatchewan bone mineral accrual study.
Journal of Bone and Mineral Research 1999;14:1672-9.

148

 

 

 

 

 

Barger-Lux MJ and Heaney RP. The role of calcium in preventing bone fragility,
hypertension, and certain cancers. Journal of Nutrition 1994;124:14068-1 1S.

Barker ME and Blumsohn A. Is Vitamin A consumption a risk factor for osteoportic
fracture? Proceedings of the Nutrition Society 2003;62:845-50.

Barr SI, Petit MA, Vigna YM, Prior JC. Eating attitudes and habitual calcium intake in
peripubertal girls are associated with initial bone mineral content and its change
over 2 years. Journal of Bone and Mineral Research 2001 ;l6:940-7.

Bass S, Delmas PD, Pearce G, Hendrich E, Tabensky A, Seeman E. The differing tempo
of growth in bone size, mass, and density in girls is region speciﬁc. Journal of
Clinical Investigation 1999;104:795-804.

Bell NH, Shary J, Stevens J, Garza M, Gordon L, Edwards J. Demonstration that bone
mass is greater in black than white children. Journal of Bone and Mineral
Research 1991;62719-23.

 

Blum RE, Wei EK, Rockett HRH, Langeliers JD, Leppert J, Gardner JD, Colditz GA.
Validation of a food frequency questionnaire in Native American and Caucasion
children 1 to 5 years of age. Maternal and Child Health Journal 1999;3z167-72.

Bollen A-M, Eyre DR. Bone resorption rates in children monitored by the urinary assay
of collagen type I cross-linked peptides. Bone 1994;15:31-34.

Bonjour JP, Carrie AL, F errare S, Clavien H, Slosman D, Theintz G, Rizzoli R.
Calcium-enriched foods and bone mass grth in prepubertal girls: a randomized,
double-blind, placebo-controlled trial. Journal of Clinical Investigation
1997;99:1287-94.

Bonjour JP, Chevalley T, Ammann P, Slosman D, Rizzoli R. Gain in bone mineral mass
in pre-pubertal girls 3.5 years after discontinuation of calcium supplementation; a
follow-up study. Lancet 2001;358:1208-12.

Bonoﬁglio D, Magglioni M, Catalano S, Marisco S, Aquila S, Giomo A, Ando S.
Parathyroid hormone is elevated but bone markers and density are normal in young

female subjects who consume inadequate dietary calcium. British Journal of
Nutrition 2000;84:1 1 1-6.

 

Boreham C, Riddoch C. The physical activity, ﬁtness and health of children. Journal of
Sports Sciences 2001;19:915-29.

Bounds W, Skinner J, Carruth BR, Ziegler P. The relationship of dietary and lifestyle

factors to bone mineral indexes in children. Journal of the American Dietetic
Association 2005;105:735-41.

149

Bradney M, Pearce G, Naughton G, Sullivan C, Bass S, Beck T, Carlson J, Seaman E.
Moderate exercise during growth in pre-pubertal boys: changes in bone mass, size,
volumetric density and bone strength: a controlled prospective study. Journal of
Bone and Mineral Research 1998;13:1814-21.

Bryant RJ, Wastney ME, Martin BR, Wood 0, McCabe GP, Morshidi M, Smith DL,
Peacock M, Weaver CM. Racial differences in bone turnover and calcium
metabolism in adolescent females. Journal of Clinical Endocrinology and
Metabolism 2003;88: 1043-47.

Bugel S. Vitamin K and bone health. Proceedings of the Nutrition Society 2003;62:830-
43.

Cadogan J, Eastell R, Jones N, Barker ME. Milk intake and bone mineral acquisition in
adolescent girls: randomized, controlled intervention trial. British Medical Journal
1997;315:1255-60.

Carter LM, Whiting SJ, Drinkwater DT, Zello GA, Faulkner RA, Bailey DA. Self-
reported calcium intake and bone mineral content in children and adolescents.
Journal of the American College of Nutrition 2001;20:502-9.

Centers for Disease Control, Morbidity and Mortality Weekly Report. Youth Risk
Behavior Surveillance — United States 2005. MMWR Surveillance Summaries,
June 9, 2006; 55(SS05):1-108.

Centers for Disease Control, National Center for Health Statistics. CDC Growth Charts
for Children in the United States, 2000. http://www.cdc.gov/nchs. Accessed July
2003.

Center for Nutrition Policy and Promotion, US Department of Agriculture and US
Department of Health and Human Services. MyPyramid Tracker; an online
dietary and physical activity assessment tool. USDA and USDHHS;2005.
http://www.mvpvramidtracker.gov/. Accessed 23 April 2007.

Center for Nutrition Policy and Promotion, US Department of Agriculture and US
Department of Health and Human Services Dietary Guidelines for Americans
2005. USDA and USDHHS;2005.
http://www.cnpp.usda.gov/DietarvGuidelineshtm. Accessed 23 April 2007.

Chan GM. Dietary calcium and bone mineral status of children and adolescents.
American Journal of Diseases of Children 1991;145:631-4.

Chan GM, Hoffman K, McMurry M. Effects of dairy products on bone and body
composition in pubertal girls. Journal of Pediatrics 1995;126:551-6.

150

 

 

Chaturvedi A, Garg OP, Choudhary B, Garg P. Bone mineral content in normal and
malnourished children. Indian Pediatrics 1993;30:489-94.

Chee WSS, Suriah AR, Zaitun Y, Chan SP, Yap SL, Chan YM. Dietary calcium intake
in postmenopausal Malaysian women: comparison between the food frequency
questionnaire and three-day food records. Asia Paciﬁc Journal of Clinical
Nutrition 2002;11:142-6.

Clarke BL, Muhs JM, O'Connell MJ, McCarthy JT, O'Fallon WM, Riggs BL.
Assessment of bone resorption in metabolic bone disorders using a new enzyme
immunoassay for urinary free pyridinoline: comparison with standard methods of
assessment of bone formation. Endocrine Practice 1995;1:248-56.

Crofton PM. New aspects of biochemical markers of bone turnover in paediatrics. In:
Schonau E, Matkovic V, eds. Paediatric Osteology: Prevention of Osteoporosis - a
Paediatric Task. Singapore: Elsevier Science, 199823-14.

 

Cusack S and Cashman KD. Impact of genetic variation on metabolic response of bone
to diet. Proceedings of the Nutrition Society 2003;62:901-12.

Daele PLAv, Seibel MJ, Burger H, Hofrnan A, Grobbee DE, van Leeuwen JPTM,
Birkenhager J C, Pols HAP. Case-control analysis of bone resorption markers,
disability, and hip fracture risk: the Rotterdam study. British Medical Journal
1996;312:482-3.

Delmas PD. Biochemical markers for the assessment of bone turnover. In: Riggs BL,
Melton LJ, eds. Osteoporosis: Etiology, Diagnosis, and Management, 2" Edition.
Philadelphia: Lippincott-Raven Publishers, 1995 :3 19-3 3.

Del-Rio L, Carrascosa A, Pons F, Gusinye M, Yeste D, Domenech FM. Bone mineral
density of the lumbar spine in white Mediterranean Spanish children and
adolescents: changes related to age, sex, and puberty. Pediatric Research
1994;35:362-6.

De Schepper J, Van den Broeck M, J onckheer MH. Study of lumbar spine bone mineral
density in obese children. Acta Paediatrics 1995;84:313-5.

De Simone M, Farello G, Palumbo M, Gentile T, Ciuffreda M, Olioso P, Cinque M, De
Matteis F. Growth charts, grth velocity and bone development in childhood
obesity. International Journal of Obesity 1995;19:851-7.

Dibba B, Prentice A, Ceesay M, Striling DM, Cole TJ, Poskitt EME. Effect of calcium

supplementation on bone mineral accretion in Gambian children accustomed to a
low-calcium diet. American Journal of Clinical Nutrition 2000;71:544-9.

151

 

Docio S, Rjancho JA, Perez A, Olmos J M, Amado JA, Gonzalez-Macias J. Seasonal
deﬁciency of Vitamin D in children: a potential target for osteoporosis-preventing
strategies? Journal of Bone and Mineral Research 1998;13:544-48.

Eastell R, Colwell A, Hampton L, Reeve J. Biochemical markers of bone resorption
compared with estimates of bone resorption from radiotracer kinetic studies in
osteoporosis. Journal of Bone and Mineral Research 1997;12:59-65.

Eyre DR. Biochemical markers of bone turnover. In: Favus MJ, ed. Primer on the
Metabolic Bone Diseases and Disorders of Mineral Metabolism, Third Edition.
Philadelphia: Lippincott-Raven, 1996:] 14-18.

Fares JE, Choucair M, Nabulsi M, Salamon M, Shahine CH, Fuleihan GE. Effect of
gender, puberty, and Vitamin D status on biochemical markers of bone
remodeling. Bone 2003;33:242-7.

F assler ALC, Bonjour JP. Osteoporosis as a pediatric problem. Pediatric Clinics of ‘
North America 1995;42:811-824. 2..

Faulkner RA, Bailey DA. Osteoporosis: a pediatric concern? Medicine and Sport
Science 2007;51:1-12.

Faulkner RA, Houston CS, Bailey DA, Drinkwater DT, McKay HA, Wilkinson AA.
Comparison of bone mineral content and bone mineral density between dominant

and non-dominant limbs in children 8-16 years of age. American Journal of
Human Biology 1993;52491-9.

F ehily AM, Coles RJ, Evans WD, Elwood PC. Factors affecting bone density in young
adults. American Journal of Clinical Nutrition 1992;56:578-86.

F erretti JL, Capozza RF, Cointry GR, Garcia SL, Plotkin H, Alvarez Filgueira AL,
Zanchetta JR. Gender-related differences in the relationship between densitometric

values of whole-body bone mineral content and lean body mass in humans 7
between 2 and 87 years of age. Bone 1998;22:683-90.
Feskanich D, Rockett HRH, Colditz GA. Modifying the Healthy Eating Index to assess 1 d

diet quality in children and adolescents. Journal of the American Dietetic
Association 2004;104:1375-83.

Fewtrell MS, Prentice A, Jones SC, Bishop NJ, Stirling D, Buffenstein R, Lunt M, Cole
TJ, Lucas A. Bone mineralization and turnover in preterm infants at 8-12 years of
age: the effect of early diet. Journal of Bone and Mineral Research 1999;14:810-
20.

152

Field AE, Gillrnan MW, Rosner B, Rockett HR, Colditz GA. Association between fruit
and vegetable intake and change in body mass index among a large sample of
children and adolescents in the United States. International Journal of Obesity
2003;27:821-26.

Fischer S, Milinarsky A, Giadrosich V, Dib G, Arriagada M, Arinoviche R. X-ray
absorptiometry of bone in obese and eutrophic children from Valparaiso, Chile.
Journal of Rheumatology 2000;27:1294-6.

Fisher J 0, Mitchell DC, Smikilas-Wright H, Mannino ML, Birch LL. Meeting calcium
recommendations during middle childhood reﬂects mother-daughter beverage
choices and predicts bone mineral status. American Journal of Clinical Nutrition
2004;79:698-706.

Flynn A. The role of dietary calcium in bone health. Proceedings of the Nutrition
Society 2003;62:851-58.

F oldes J, Shlh MS, Levy J. Bone structure and calcium metabolism in obese Zucker rats.
International Journal of Obesity 1992;16:95-102.

Food and Nutrition Board, Institute of Medicine. Dietary reference intakes for calcium,
phosphorus, magnesium, Vitamin D and ﬂuoride. Washington DC: National
Academy Press, 1997.

Food and Nutrition Board, Institute of Medicine. Dietary reference intakes for energy,
carbohydrates, ﬁber, fat, protein, and amino acids (macronutrients). Washington
DC: National Academy Press, 2005.

Frassetto LA, Todd KM, Morris RC, Sebastian A. Estimation of net endogenous
noncarbonic acid production in humans from diet potassium and protein contents.
American Journal of Clinical Nutrition 1998;68:576-83.

Garner SC, Anderson JJB, Ambrose WW. Skeletal tissues and mineralizaton. In:
Anderson JJB, Garner SC, eds. Calcium and Phosphorus in Health and Disease.
Boca Raton: CRC Press, 1996:97-117.

Garnero P, Hausherr E, Chapuy M-C, Marcelli C, Grandjean H, Muller C, Corrnier C,
Bréart G, Meunier PJ, Delmas PD. Markers of bone resorption predict hip fracture
in elderly women: The EPIDOS prospective study. Journal of Bone and Mineral
Research 1996;11:1531-8.

Garnero P, Shih WJ, Gineyts E, Karpf DB, Delmas PD. Comparison of new biochemical
markers of bone turnover in late postmenopausal osteoporotic women in response

to alendronate treatment. Journal of Clinical Endocrinology and Metabolism
1994;79:1693-1700.

153

 

 

Genits MI, Thijssen J HH, Van Rijn HJ M. Determination of pyridinoline and
deoxypyridinoline in urine, with special attention to retaining their stability.
Clinical Chemistry 1995; 41:571-574.

Geusens P, Cantatore F, Nijs J, Proesmans W, Emma F, Dequeker J. Heterogeneity of
grth of bone in children at the spine, radius and total skeleton. Growth,
Development and Aging 1991;55:249-56.

Gibson RS. Principles of Nutritional Assessment. New York: Oxford University Press,
1990.

Gilsanz V, Roe TF, Mora S, Costin G, Goodman WG. Changes in vertebral bone density
in black girls and white girls during childhood and puberty. New England Journal
of Medicine 1991;325:1597-600.

Ginty F. Dietary protein and bone health. Proceedings of the Nutrition Society
2003;62:867-76.

Ginty F, Prentice A, Laidlaw A, McKenna L, Jones S, Stear S, Cole TJ. Calcium
carbonate supplementation is associated with higher plasma IGF-I in 16- to 18-
year-old boys and girls. In: Burckhart P, Dawson-Hughes B, Heaney RP, eds.
Nutritional Aspects of Osteoporosis, Second Edition. USA: Elsevier Science,
2004; 45-57.

Glastre C, Braillon P, David L, Cochat P, Meunier PJ, Delmas PD. Measurement of bone
mineral content of the lumbar spine by dual energy x-ray absorptiometry in normal
children: correlations with growth parameters. Journal of Clinical Endocrinology
and Metabolism 1990;70:1330-3.

Gluer CC, Barkmann R, Kuhn B, Starnpa B. New aspects of bone densitometry. In:
Schonau E, Matkovic V, eds. Paediatric Osteology: Prevention of Osteoporosis - a
Paediatric Task. Singapore: Elsevier Science, 1998:15-21.

Gomez B, Bailey CA, Jenkins DK, Kelm RJ, Seyedin S. An enzyme immunoassay for
intact, newly synthesized osteocalcin: A marker of bone formation (Abstract).
International Conference on Progress in Bone and Mineral Research. Vienna,
Austria, 1994.

Goulding A, Jones IE, Taylor RW, Williams SW, Manning PJ. Bone mineral density and
body composition in boys with distal forearm fractures: a dual-energy x-ray
absorptiometry study. Journal of Pediatrics 2001;139:509-15.

Goulding A, Taylor RW, Jones IE, McAuley KA, Manning PJ, Williams SM.

Overweight and obese children have low bone mass and area for their weight.
International Journal of Obesity 2000;24:627-32.

154

!--L.

 

 

 

Greer FR, Krebs NF. Optimizing bone health and calcium of infants, children, and
adolescents. Pediatrics 2006;1 171578-85.

Green JH, Booth CL, Bunning RLW. Assessment of a rapid method for assessing
adequacy of calcium intake. Asia Paciﬁc Journal of Clinical Nutrition
2002;11:147-50.

Haapasalo H, Kannus P, Sievanen A, Heinonen A, Oja P, Vuori I. Long-terrn unilateral
loading and bone mineral density and content in female squash players. Calciﬁed
Tissue International 1994;54:249-55.

Harris JA, Benedict F G. A biometric study of basal metabolism in man. (Publication No.
279) Washington, DC: Carnegie Institute of Washington, 1919.

Hasanoglu A, Bideci A, Cinaz P, Turner L, Unal S. Bone mineral density in childhood
obesity. Journal of Pediatirc Endocrinology and Metabolism 2000;13:307-11.

Heaney RP. Calcium, dairy products and osteoporosis. Journal of the American College
of Nutrition 2000;19:838-998.

Heaney RP, Abrams S, Dawon-Hughes B, Looker A, Marcus R, Matkovic V, Weaver C.
Peak bone mass. Osteoporosis International 2000;11:985-1009.

Heaney RP, Michael Davies K, Barger-Lux J. Calcium and weightzclinical studies.
Journal of the American College of Nutrition 2002;21:1528-58.

Henderson RC, Hayes PR. Bone mineralization in children and adolescents with milk
' allergy. Bone and Mineral 1994;27:1-12.

Hesley RP, Shepard JA, Jenkins DK, Riggs BL. Monitoring estrogen replacement
therapy and identifying rapid bone losers with an immunoassay for
deoxypyridioline. Osteoporosis International l998;8:159-64.

Hillman L, Schlotzhauer C, Lee D, et al. Decreased bone mineralization in children with
phenylketonuria under treatment. European Journal of Pediatrics 1996;155:8148-
52.

Horlick M, Thornton J, Wang J, Levine LS, Fedun B. Pierson RN. Bone mineral in
prepubertal children: gender and ethnicity. Journal of Bone and Mineral Research
2000;15:1393-7.

Huybrechts 1, De Bacquer D, Matthys C, De Backer G, De Henauw S. Validity and

reproducibility of a semi-quantitative food-frequency questionnaire for estimating
intake in Belgian preschool children. British Journal of Nutrition 2006;95:802-16.

155

 

 

Ilich JZ, Skugor M, Hangartner T, Baoshe A, Matkovic V. Relation of nutrition, body
composition and physical activity to skeletal development: a cross-sectional study
in preadolescent females. Journal of the American College of Nutrition

1998;17:136-47.

Iuliano-Bums S, Saxon L, Naughton G, Gibbons K, Bass, SL. Regional speciﬁcity of
exercise and calcium during skeletal growth in girls: a randomized controlled trial.
Journal of Bone and Mineral Research 2003;18:156-62.

James IT, Walne AJ, Perrett D. The measurement of pyridinium crosslinks: a
methodological overview. Annals of Clinical Biochemistry 1996;33:397-420.

Johansen J S, Riis BJ, Delmas PD, Christiansen C. Plasma BGP: An indicator of
spontaneous bone loss and the effect of oestrogen treatment in postmenopausal
women. European Journal of Clinical Investigation 1988;18:191-195.

Johnston CC, Miller JZ, Slemenda CW, Reister TK, Hui S, Christian JC, Peacock M.
Calcium supplementation and increases in bone mineral density in children. New
England Journal of Medicine 1992;327:82-7.

Kalkwarf HJ, Khoury J C, Bean J, Elliot JG. Vitamin K, bone turnover, and bone mass in
girls. American Journal of Clinical Nutrition 2004;80:1075-80.

Kannus P, Haapasalo H, Sankelo M, Sievanen H, Pasanen M, Heinonen A, Oja P, Vuori
I. Effect of starting age of physical activity on bone mass in the dominant arm of
tennis and squash players. Annals of Internal Medicine 1995;123:27-31.

Kroger H, Kotaniemi A, Kroger L, Alhava E. Development of bone mass and bone
density of the spine and femoral neck — a prospective study of 65 children and
adolescents. Bone and Mineral 1993;23:171-82.

Lee RD, Neiman DC. Nutritional Assessment. Second edition. St. Louis: Mosby, 1996.

Lee WTK, Leung SSF, Leung DMY, Tsang HSY, Lau J, Cheng JCY. A randomized
double-blind controlled calcium supplementation trial, and bone and height
acquisition in children. British Journal of Nutrition 1995;74:125-39.

Lee WTK, Leung SSF, Leung DMY Cheng JCY. A follow-up study on the effects of
calcium-supplement withdrawal and puberty on bone acquisition of children.

American Journal of Clinical Nutrition 1996;64:71-7.

Lee WT, Leung SS, Leung DM. Bone mineral acquisition in low calcium intake children
following a withdrawal of calcium supplement. Acta Paediatrics 1997;86:570-76.

156

 

 

 

 

Lee WTK, Leung SSF, Lui SSH, Lau J. Relationship between long-term calcium intake
and bone mineral content of children aged from birth to 5 years. British Journal of
Nutrition 1993;70:235-48.

Lee WTK, Leung SSF, Wang S-H, Xu S-H, Zeng W-P, Lau J, Oppenheimer SJ, Cheng J.
Double-blind controlled calcium supplementation and bone mineral accretion in
children accustomed to low calcium diet. American Journal of Clinical Nutrition
1994;60:744-752.

Lester GE. Serum and urine markers of bone turnover. In: Anderson JJ B, Garner SC,
eds. Calcium and Phosphorus in Health and Disease. Boca Raton: CRC Press,
1995: 147-54.

Li J Y, Specker BL, Ho ML, Tsang RC. Bone mineral content in black and white children
1 to 6 years of age. Early appearance of race and sex differences. American
Journal of Diseases of Children 1989;143:1346-9.

Lieuw-A-F a M, Sierra RI, Specker BL. Carboxy-terminal propeptide of human type I
collagen and pyridinium cross-links as markers of bone grth in infants 1 to 18
months of age. Journal of Bone and Mineral Research 1995;10:849-53.

Lips P. Which circulating level of 25-hydroxyvitamin D is appropriate? Journal of
Steroid Biochemistry and Molecular Biology 2004;89-90:611-4.

Lisa L, Neradilova M, Tomasova N, Soutorova M, Zimak J. Osteocalcin in congenital
adrenal hyperplasia. Bone 1995;16:57-9.

Lloyd T, Andon MB, Rollings N, Martel JK, Landis JR, Demers LM, Eggli DF,
Kieselhorst K, Kulin HE. Calcium supplementation and bone mineral density in
adolescent girls. Journal of the American Medical Association 1993;270:841-4.

Lloyd T, Petit MA, Lin HM, Beck TJ. Lifestyle factors and the development of bone
mass and bone strength in young women. Journal of Pediatrics 2004;144:776-82.

MacKelvie KJ, McKay HA, Khan KM, Crocker PRE. A school-based exercise
intervention augments bone mineral accrual in early pubertal girls. J oumal of
Pediatrics 2001;139:501-8.

Manzoni P, Brambilla P, Pietrobelli A, Beccaria L, Bianchessi A, Mora S, Chiumello G.
Inﬂuence of body composition on bone mineral content in children and

adolescents. American Journal of Clinical Nutrition 1996;64:603-7.

Massey, LK. Dietary animal and plant protein and human bone health: a whole foods
approach. Journal of Nutrition 2003;133:8628-58.

157

 

Matkovic V, Badenhop NE, Landoll JD, Ilich JZ, Rosen CJ, Buell JL. Skeletal growth in
a nutrition perspective: genetic and endocrine interaction. In: Schonau E,
Matkovic V, eds. Paediatric Osteology: Prevention of Osteoporosis - a Paediatric
Task. Singapore: Elsevier Science, 1998:53-71.

Matkovic V, Goel PK, Badenhop-Stevens NE, Landoll JD, Li B, Ilich JZ, Skugor M,
Nagode LA, Mobley SL, Ha EJ, Hangartner TN, Clairrnont A. Calcium
supplementation and bone mineral density in females from childhood to young
adulthood: a randomized controled study. American Journal of Clinical Nutrition
2005;81:175-88.

Maynard LM, Guo SS, Chumlea WC, Roche AF, Wiseman WA, Zeller CM, Towne B,
Siervogel RM. Total-body and regional bone mineral content and areal bone
mineral density in children aged 8-18y: the Fels longitudinal study. American
Journal of Clinical Nutrition 1998;68:1111-7.

McGowan M, Gibney MJ. Calcium intakes in individuals on diets for the management of
cow’s milk allergy; a case control study. European Journal of Clinical Nutrition
1993;47:609-16.

McKay HA, Petit MA, Schutz RW, Prior JC, Bass SI, Khan KM. Augmented
trochanteric bone mineral density after modiﬁed physical education classes: A

randomized school-based exercise intervention study in prepubescent and early
pubescent children. Journal of Pediatrics 2000;136:156-62.

McMurray RG. Effects of physical activity on bone. In: Anderson, JJB and Garner SC,
eds. Calcium and Phosphorus in Health and Disease. Boca Raton, FL: CRC Press
1995;301-18.

Miller GD, Weaver CM. Required versus optimal intakes: a look at calcium. Journal of
Nutrition 1994;124:1404S-5S.

MiniMitter Corporation. Actical product technical bulletin and product literature. Bend,
Oregon, 2003. http://wwwminimittencom. Accessed 23 April 2007.

 

Molgaard C, Michaelsen KF. Vitamin D and bone health in early life. Proceedings of
the Nutrition Society 2003;62:823-8.

Molgaard C, Thomsen BL, Prentice A, Cole TJ, Michaelsen KF. Whole body bone
mineral content in healthy children and adolescents. Archives of Diseases of
Children 1997;76:9-15.

Molgaard C, Thomsen BL, Michealsen KF. The inﬂuence of calcium intake and physical

activity on bone mineral content and bone size in healthy children and adolescents.
Osteoporosis International 2001;12:887-94.

158

 

 

 

 

Montomoli M, Gonnelli S, Giacchi M, Mattei R, Cuda C, Rossi S, Gennari C. Validation
of a food frequency questionnaire for nutritional calcium inatke assessment in
Italian women. European Journal of Clinical Nutrition 2002;56:21-30.

Mora S, Pitukcheewanont P, Kaufman FR, Nelson J C, Gilsanz V. Biochemical markers
of bone turnover and the volume and the density of bone in children and different
stages of sexual development. Journal of Bone and Mineral Research
1999;14:1664-71.

Morris FL, Naughton GA, Gibbs JL Carlson J S Wark JD. Prospective 10-month exercise
intervention in pre-menarcheal girls: positive effects of bone and lean mass.
Journal of Bone and Mineral Research 1997;12:1453-62.

Munday K. Vitamin C and bone biomarkers: investigations in a Gambian population.
Proceedings of the Nutrition Society 2003;62:429-36.

Murphy NM, Carroll P. The effect of physical activity and its interaction with nutrition
on bone health. Proceedings of the Nutrition Society 2003;62:829-38.

National Center for Health Statistics. NHANES Anthropometric Video. 1996.
http://www.cdc.gov/nchs/about/maior/nhanes/avideo.htrn (accessed 04Apri12007).

New SA. Intake of fruit and vegetables: implications for bone health. Proceedings of the
Nutrition Society 2003;62:889-99.

Nguyen TV, Maynard LM, Towne B, Roche AF, Wisemandle W, Li J, Guo SS, Chumlea
WC, Siervogel RM. Sex differences in bone mass acquisition during growth: the
F els Longitudinal Study. Journal of Clinical Densitometry 2001 ;4: 147-157.

Nieves JW, Fromica C, Rufflng J, Zion M, Garrett P, Lindsay R, Cosman F. Males have
larger skeletal size and bone mass than females, despite comparable body size.
Journal of Bone and Mineral Research 2005;20:529-35.

Njeh CF, Samat SB, Nightingale A, McNeil EA, Boivin CM. Radiation dose and in vitro
precision in paediatric bone mineral density measurement using dual x-ray
absorptiometry. British Journal of Radiology 1997;70:719-27.

Noonan K, Kalu ME, Holownia P, Burrin J M. Effect of different storage temperatures,
sample collection procedures and immunoasay methods on osteocalcin
measurement. European Journal of Clinical Chemistry and Clinical Biochemisrty
1997;34:841-4.

Norman AW, Henry HH. Vitamin D. In: Bowman BA, Russell RM, eds. Present

Knowledge in Nutrition, Ninth Edition. Washington DC: ILSI Press, 2006: 198-
210.

159

Nowack MK, Brizzolara S, Lally DA. Bone mineral content in Hawaiian, Asian, and
Filipino children. Hawaii Medical Journal 1995;54:388-9.

Optowsky AR and Bilezikian JP. Racial differences in the effect of early milk
consumption on peak and postmenopausal bone mineral density. Journal of Bone
and Mineral Research. 2003;18:1978-88.

Owen DB. Handbook of Statistical Tables. Reading, MA: Addison-Wesley Publishing,
1962.

Patel DN, Pettifor JM, Becker PJ, Greive C, l K, Leschner. The effect of ethnic group on
appendicular bone mass in children. Journal of Bone and Mineral Research
1992;7:263-72.

Pfeiffer KA, Mciver KL, Dowda M, Almeida MJCA, Pate RR. Validation and
calibration of the Actical accelerometer in preschool children. Medicine and
Science in Sports and Exercise 2006;38:152-7.

Pietrobelli A, Faith MS, Wang J, Brambillo P, Ciumello G, Heymsﬁeld SB. Association
of lean tissue and fat mass with bone mineral content in children and adolescents.
Obesity Research 2002; 10:56-60.

Pocock NA, Elsman JA, Hopper J L, Yeates MG, Sarnbrook PN, Ebert S. Genetic
determinants of bone mass in adults; a twin study. Journal of Clinical
Investigation 1987;80:716-10.

Pothiwala P, Evans EM, Chapman-Novakofski KM. Ethnic variation in risk for
osteoporosis among women: a review of biological and behavioral factors. Journal
of Women’s Health 2006;15:709-15.

Price PA, Parthemore JG, Deftos LJ, Nishimoto SK. New biochemical marker for bone
metabolism. Measurement by radioimmunoassay of bone gla protein in the plasma
of normal subjects and patients with bone disease. Journal of Clinical
Investigation 1980;66:878-83.

Prynne CJ, Ginty F, Paul AA, Bolton—Smith C, Stear SJ, Jones SC, Prentice A. Dietary
acid-base balance and intake of bone-related nutrients in Cambridge teenagers.
European Journal of Clinical Nutrition 2004;58:1462-71.

Rauch F, Schoenau E. Changes in bone density during childhood and adolescence: an
approach based on bone’s biological organization. Journal of Bone and Mineral
Research 2001;16:597-604.

Rauch F, Schoenau E, Woitge H, Remer T, Seibel M. Urinary excretion of hydroxy-

pyridinium cross-links of collagen reﬂects skeletal growth velocity in normal
children. Experimental and Clinical Endocrinology 1994;102:94-7.

160

ll

Remer T, Dimitriou T, Manz F. Dietary potential renal acid load and renal net acid
excretion in healthy, free-living children and adolescents. American Journal of
Clinical Nutrition 2003;77:1255-60.

Riggs BL, LJ Melton. The worldwide problem of osteoporosis: insights afforded by
epidemiology. Bone 1995;17:5058-118.

Robins SP. Collagen crosslinks in metabolic bone disease. Acta Orthopaedics
Scandanavia 1995;66:171-5.

Robins SP, Woitge H, Hesley R, Ju J, Seyedin S, Seibel MJ. Direct, enzyme-linked
immunoassay for urinary deoxypyridinoline as a speciﬁc marker for measuring
bone resorption. Journal of Bone and Mineral Research 1994;9z1643-9.

Rockett HRH, Breitenbach M, Frazier AL, Witschi J, Wolf AM, Field AE, Colditz GA.
Validation of a youth/adolescent food frequency questionnaire. Preventive
Medicine 1997;26:808-16.

Rockett HRH , Colditz GA. Assessing diets of children and adolescents. American
Journal of Clinical Nutrition 1997;65:11168-228.

Ross PD, Knowlton W. Rapid bone loss is associated with increased levels of
biochemical markers. Journal of Bone and Mineral Research 1998;13:297-302.

Roux JP, Arlot ME, Gineyts E, Meunier PJ, Delmas PD. Automatic-interactive
measurement of resorption cavities in transiliac bone biopsies and correlation with
deoxypyridinoline. Bone 1995; 17:153-156.

Russell DL, Keil MF, Bonat SH, Uwaifo GI, Nicholson JC, McDufﬁe JR, Hill SC,
Yanovski JA. The relationship between skeletal maturation and adiposity in
African American and Caucasian children. Journal of Pediatrics 2001;139:844-8.

Seibel MJ, Robins SP, Bilezikian JP. Urinary pyridinium crosslinks of collagen. Speciﬁc
markers of bone resorption in metabolic bone disease. Trends in Endocrinology
and Metabolism 1992;3 :263-70.

Seibel MJ, Woitge HW. Basic principles and clinical applications of biochemical
markers of bone metabolism: biochemical and technical aspects. Journal of

Clinical Densitometry 1999; 2 :299-321 .

Seyedin SM, Rosen DM. Matrix proteins of the skeleton. Current Opinions in Cell
Biology 1990;22914-9.

161

Shore RM, Poznanski AK. Radiologic evaluation of bone mineral in children. In: Favus
MJ, ed. Primer on the Metabolic Bone Diseases and Disorders of Mineral
Metabolism, Third Edition. Philadelphia: Lippincott-Raven, 1996:119-34.

Slemenda CW, Peacock M, Hui S, Zhou L, Johnston CC. Reduced rates of skeletal
remodeling are associated with increased bone mineral density during the
development of peak skeletal mass. Journal of Bone and Mineral Research
1997;12:676-82.

Slemenda CW, Reister TK, Hui SL, Miller JZ, Christian JC, Johnston CC. Inﬂuences on
skeletal mineralization in children and adolescents: evidence for varying effects of
sexual maturation and physical activity. Journal of Pediatrics 1994;125:201-7.

 

Southard RN, Morris JD, Mahan JD, et al. Bone mass in healthy children: measurement
with quantitative DXA. Radiology 1991;179:735-8.

Soyka LA, Fairﬁeld WP, Klibanski A. Hormonal determinants and disorders of peak
bone mass in children. Journal of Clinical Endocrinology and Metabolism
2000;85:3951-63.

Soyka LA, Grinspoon S, Levitsky LL, Herzog DB, Klibanski A. The effects of anorexia
nervosa on bone metabolism in female adolescents. Journal of Clinical
Endocrinology and Metabolism 1999;84:4489-96.

Soyka LA, Misra M, Frenchman A, Miller KK, Grinspoon S, Schoenfeld DA, Klibanski
A. Abnormal bone mineral accrual in adolescent girls with anorexia nervosa.
Journal of Clinical Endocrinology and Metabolism 2002;87:4177-85.

 

Specker B, Binkley T. Randomized trial of physical activity and calcium supplementation
on bone mineral content in 3- to 5-year old children. Journal of Bone and Mineral
Research 2003;18:885-92.

Specker BL, Brazerol W, Tsang RC, Levin R, Searcy J, Steichen J. Bone mineral content
in children 1 to 6 years of age. American Journal of Diseases of Children
1987;141:343-4.

Specker BL, J ohannsen N, Binkley T, Finn K. Total body bone mineral content and
tibial cortical bone measures in preschool children. Journal of Bone and Mineral
Research 2001 :16:2298-2305.

Specker B, Mulligan L, Ho M. Longitudinal study of calcium intake, physical activity,

and bone mineral content in infants 6-18 months of age. Journal of Bone and
Mineral Research 1999;14:569-76.

162

Stallings VA, Oddleifson NW, Negrini BY, Zemel BS, Wellens R. Bone mineral content
and dietary calcium intake in children prescribed a low-lactose diet. Journal of
Pediatric Gastroenterology and Nutrition 1994;18:440-5.

Steichen JJ, Steichen PA, Tsang RC. Bone mineral content measurement in small infants
by single photon absorptiometry: Current methodological concerns. Journal of
Pediatrics 1988;113:181-8.

Stein AD, Shea S, Basch CE, Contendo IR, Zybert P. Consistency of the Willett
semiquantitative food frequency questionnaire and 24-hour dietary recalls in
estimating nutrient intakes of preschool children. American Journal of
Epidemiology 1992;13 5 :667-677.

Stracke H, Renner E, Knie G, Leidig G, Minnie H, Federlin K. Osteoporosis and bone
metabolic parameters in dependence upon calcium intake through milk and milk
products. European Journal of Clinical Nutrition 1993;47:617-22.

Tanner JM. Growth at adolescence. Springﬁeld ILzCC Thomas, 1955.

Teucher B, Fairweather-Tait S. Dietary sodium as a risk factor for osteoporosis: where is
the evidence? Proceedings of the Nutrition Society 2003;62:859-66.

Thompson FE, Byers T. Dietary Assessment Resource Manual. Journal of Nutrition
1994;124:22458-2317S.

Tommasi M, Bacciottini L, Benucci A, et al. Serum biochemical markers of bone
turnover in healthy infants and children. International Journal of Biological
Markers l996;1 1 :159-64.

Tsukahara H, Sudo M, Umezaki M, Hiraoka M, Yamamoto K, Ishii Y, Haruki S. Dual-
energy X-ray absorptiometry in the lumbar spine, proximal femur and distal radius
in children. Pediatric Radiology 1992;22:560-2.

Tylavsky F A. Nutrition inﬂuences on bone growth in children. Journal of Nutrition
2004;134:6898-908.

Tylavsky FA, Holliday K, Danish R, Womack C, Norwood J, Carbone L. Fruit and
vegetable intakes are an independent predictor of bone size in early pubertal
children. American Journal of Clinical Nutrition 2004;79:311-7.

USDHHS. United States Department of Health and Human Services (U SDHHS), Ofﬁce
of Disease Prevention and Promotion, Healthy People 2010: Arthritis,
Osteoporosis and Chronic Back Conditions. November 2000.
http://www.healthvpeople.gov/document/HTML/Volume1/02Arthritis.htm.
Accessed 23 April 2007.

163

Van Coeverden SCCM, Netelenbos JC, de Ridder CM, Roos JC Popp-Snijders C,
Delemarre-van de Waal HA. Bone metabolism markers in healthy pubertal boys
and girls. Clinical Endocrinology 2002;57:107-16.

VadenBergh MF Q, DeMan SA, Wittemen J CM, Hofrnan A, Trouerbach WT, Grobbee
DE. Physical activity, calcium intake, and bone mineral content in children in the
Netherlands. Journal of Epidemiology and Community Health 1995;49:299-304.

Volek J S, Gomez AL, Scheet TP, Sharman MJ, French DN, Rubin MR, Ratarness NA,
McGuigan MM, Kraemer WJ. Increasing ﬂuid milk favorably affects bone
mineral density responses to resistance training in adolescent boys. Journal of the
American Dietetics Association 2003;103:1353-1356.

Vople SL. Magnesium. In: Bowman BA, Russell RM, eds. Present Knowledge in
Nutrition, Ninth Edition. Washington DC: ILSI Press 2006:400-8.

Wacker W, Barden HS. Pediatric reference data for male and female total body and
spine BMD and BMC. Presented at the International Society of Clinical
Densitometry: Dallas, TX; March 2001. (personal communication)

Watson R. Bone growth and physical activity in young males. In: Mazess R, ed.
International Conference on Bone Mineral Measurements. NIH 75-883,
Washington, DHEW Publication; 1 974:380-5.

Weaver, C. Calcium requirements of physically active people. American Journal of
Clinical Nutrition 2000;72(Suppl):579S-84S.

Weisnier RL, Krumdieck CL. Dairy foods and bone health: examination of the evidence.
American Journal of Clinical Nutrition 2000;72:681-9.

Whiting SJ. Obesity is not protective for bones in childhood and adolescence. Nutrition
Reviews 2002;60:27-36.

Whiting SJ, Vatanparast H, Baxter-Joens A, Faulkner RA, Mirwald R, Bailey DA.
Factors that affect bone mineral accrual in the adolescent growth spurt. Journal of
Nutrition 2004;134:6968-7008.

World Health Organization, Scientiﬁc Group on the Prevention and Management of
Osteoporosis. Prevention and management of osteoporosis: WHO Technical

Report Series 921. Geneva, Switzerland: WHO, 2003.

Young MF. Bone matrix proteins: their function, regulation, and relationship to
osteoporosis. Osteoporosis International 2003;14(Suppl 3):S35-S42.

Zamboni G, Sofﬁati M, Giavarina D, Tato L. Mineral metabolism in obese children. Acta
Paediatrics Scandanavia 1988;77:741-6.

164

Zanchetta JR, Plotkin H, Alvarez-Filgueira ML. Bone mass in children: normative values
for the 2-20-year-old population. Bone 1995;16:393S-9S.

Zanker CL, Gannon L, Cooke CB, Gee KL, Oldroyd B, Truscott JG. Differences in bone
density, body composition, physical activity, and diet between child gymnasts and
untrained children 7-8 years of age. Journal of Bone and Mineral Research
2003;28:1043-50.

Zemel MB. Regulation of adiposity and obesity risk by dietary calcium: mechanisms and
implications. Journal of the American College of Nutrition 2002;21:1468-518.

Zemel MB, Shi H, Greer B, Dirienzo D, Zemel PC. Regulation of adiposity by dietary
calcium. FASEB Journal 2000;14:1132-8.

165

 

11111113111111leLilli