Jtlmwr 73:.
t. 1.3

 

h.“
.
c

.15.”... . . . ‘ it.
afiflm . . 4 V gum
‘ .

i... .1:
. 7.

 

In
, 339.. 4.3.:
. 4.. if: ($3641..
.0 S!!.wl§

V1331: .
3)....

. 31.... 9;: 4.
i 2...! fin”

 

 

 

|\)SI ‘5...
(tailli. .

 

:3: Q

L 5):}: 7.. . . .

,|l. l!‘WHt||I.Ir|IIIuI.I!

 

THESiS

x}

                                                                                               

l‘JJlllHlHi

 

‘2 302048 8460
Saw
LIBRARY
Michigan State
' University

 

 

 

This is to certify that the

dissertation entitled

EAK OXYGEN CONSUMPTION
IN YOUNG DISTANCE RUNNERS

presented by

Joey C. Eisenmann

has been accepted towards fulfillment
of the requirements for

Ph . D . degree in W9)!

Major professor

Date 6 3 ° '0 °

MS U i: an Affirmative Action/ Equal Opportunity Institution 0-12771

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

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

11/00 WWW-p.14

BLOOD LIPIDS AND PEAK OXYGEN CONSUMPTION
IN YOUNG DISTANCE RUNNERS

By

Joey C. Eisenmann

A DISSERTATION
Submitted to

Michigan State University
in partial fulfillment of the requirements

for the degree of
DOCTOR OF PHILOSOPHY

Deparunent of Kinesiology
2000

ABSTRACT

BLOOD LIPIDS AND PEAK OXYGEN CONSUMPTION
IN YOUNG DISTANCE RUNNERS

By

Joey C. Eisenmann

This dissertation includes a series of papers on blood lipids and peak Vo2 in
young male and female distance runners, 9 to 19 yrs of age. Two independent samples -
a mixed-longitudinal cohort of 27 males and 27 females (Young Runners Study I [YRS
l]), and a cross-sectional cohort of 48 males and 22 females (Young Runners Study 11
[YRS Ill) - were used in the analysis. '

The development of blood lipids in young distance runners appears to be similar
to the general population - total cholesterol (TC) and low-density lipoprotein (LDL)
remain stable, high-density lipoprotein (HDL) declines during adolescence (especially in
males), and triglycerides (TG) increases with age. The lack of the attenuation in HDL
may lend to the robustness of normal growth and maturation, including genes, hormones,
and fat distribution, on the development of HDL in males regardless of exercise training.
A superior blood lipid profile was not observed in young distance runners compared to
age- and sex-specific reference values for United States youth, except for higher HDL
prior to age14 yrs. In contrast to mean values, there was considerable variability in blood
lipids including dyslipidernic values.

Heterogeneity in blood lipids among young distance runners was also considered.
Determinants included training volume (TV, km per wk), peak oxygen consumption

(peak V02, ml'kg’1min"‘), and body fatness (sum of six skinfolds, SSF; trunk-to-extremity

ratio, TER). Increased weekly running distance was not related to blood lipids in young
distance runners. However, TV may be indirectly related with HDL through its
relationship with peak Vo2 in males. A unique finding was the differential relationships
between TV and HDL when the entire sample was grouped according to modified
clinical cut-points. Partial correlations indicate that the association between peak Vo2
and HDL remained significant after controlling for the concomitant variation in SSF and
explained 9% of the variance in HDL. The association between SSF and HDL did not
remain significant after controlling for the concomitant variation in peak V0,. The role of
genes, peak V0,, and body fatness in the modulation of elevated blood lipid levels has
also been indicated.

As expected, an age-related increase in absolute peak Vo2 occurred in both sexes
with sex differences emerging during adolescence. When expressed per unit body mass,
peak Vo2 (ml'kg"‘min"‘) remains stable until age 15 when it increases in boys, and

decreases in girls. In contrast, relative peak Vo2 (ml°kg"°'75

min”) increases throughout
the age range in boys and increases in girls until agelS yrs, and peak Vo2 adjusted for
body mass (ml'min") increases with age in boys and girls. Allometric scaling factors
varied by analytical methods. The overall mean cross-sectional scaling factor was

1.01 $0.03 (SE) in boys and 0.85:0.05 (SE) in girls. Mean ontogenetic allometric scaling
factors were 0.81 and 0.61 in males and females, respectively. Thus, it was concluded
that the interpretation of growth-related changes in peak Vo2 of young distance runners

was dependent upon the manner of expressing peak V0;2 relative to body size and/or the

statistical technique employed.

ACKNOWLEDGEMENTS

I would like to briefly acknowledge many individuals who have contributed to

this dissertation and/or my scholastic progress.

This dissertation was supported in part by the William Wohlgamuth Memorial
Fellowship and the Institute for the Study of Youth Sports.

Dr. Robert Malina, my major professor and dissertation director, for teaching me
human growth and maturation, human variability, and most importantly, that physical
activity is not the answer to all the world's (health-related) problems, but rather a
small portion of the phenotypic variance that includes genes, environmental factors,
and the genotype-environmental interaction.

Dr. James M. Pivarnik for directing and supervising activities of the Human Energy
Research Laboratory, and excellent mentoring in pediatric exercise physiology and
epidemiology.

Dr. Chris J. Womack for insightful conversations on physical activity and
atherosclerosis.

Dr. Mathew J. Reeves and other members of the Department of Epidemiology for
introducing me to the academic discipline of epidemiology and pushing me to "think
like an epidemiologist".

Members of the Human Energy Research Laboratory, especially Karin Allor, John
Zubek, and Candace Perkins, for intellectual and technical support.

JoAnn lanes for excellent administrative support.

iv

Vern Seefeldt, John Haubenstricker, and other faculty and staff who participated in
the Young Runners Study 1, 1982-1986.

Former college baseball teammates and friends for support, friendship, and great
memories that provided vitality during my graduate studies.

Layton and Donna Eisenmann and Gil and Elna Herbel for being great parents.
My wife, Beth Herbel-Eisenmann, for understanding the trials and tribulations of
graduate education, and for simply being a great friend.

My son, Kaleb Herbel—Eisenmann, for allowing me to re-discover the joys of

childhood, particularly. free play and curiosity.

TABLE OF CONTENTS

LIST OF TABLES
LIST OF FIGURES

CHAPTER 1 INTRODUCTION
References

CHAPTER 2
REVIEW OF LITERATURE: PART I
Introduction
Blood lipids and coronary heart disease
Early origins of atherosclerosis
Development of blood lipids in the general pediatric population
Blood lipids in child and adolescent athletes
Physical activity and blood lipids in children and adolescents
Dose-response issues
Training volume and blood lipids in distance runners
Aerobic fitness and blood lipids
Body fatness and blood lipids
Physical activity and blood lipids: mechanisms
Summary
References

CHAPTER 3
MIXED-LONGITUDINAL ANALYSIS OF BLOOD LIPIDS IN YOUNG
DISTANCE RUNNERS
~ Abstract

Introduction

Methods

Results

Discussion

Acknowledgements

References

CHAPTER 4
BLOOD LIPIDS OF YOUNG DISTANCE RUNNERS: DISTRIBUTION
AND COMPARISON TO REFERENCE VALUES AND CLINICAL
CUT-POINT S

Abstract

Introduction

Methods

Results

vi

ix

$8383$92

Discussion 87

Acknowledgements 93

References 94
CHAPTER 5

BLOOD LIPIDS OF YOUNG DISTANCE RUNNERS: INT ER-REIATION SHIPS
AMONG TRAINING VOLUME, PEAK OXYGEN CONSUMPTION, BODY

FATNESS, AND BLOOD LIPIDS 104
Abstract 105
Introduction 107
Methods 109
Results 114
Discussion 115
Acknowledgements 125
References 126

CHAPTER 6

REVIEW OF LITERATURE: PART II - 138
Introduction 139
Age- and sex-associated variation in peak Vo2 140
Maturity-associated variation in peak Vo2 143
Peak Vo2 in young athletes 145
Basic principles and historical background of scaling 147
Allometric scaling, body size, and peak Vo2 150
Scaling peak V02: implications for endurance performance and
health-related fitness 153
Summary 154
References 156

CHAPTER 7

SCALING PEAK V02 T0 BODY MASS IN YOUNG DISTANCE RUNNERS 165
Abstract 166
Introduction 168
Methods 169
Results 176
Discussion 179
Acknowledgements 189
References 190

CHAPTER 8

SUMMARY AND RECOMMENDATIONS 200

vii

APPENDIX A
SELF-REPORTED TRAINING VOLUME 208

APPENDIX B
CONSENT FORM 210

viii

LIST OF TABLES
Table 2.1. Guidelines for interpreting blood lipid values in children and
adolescents
Table 2.2. Summary of blood lipid studies in child and adolescent athletes

Table 3.1. Age-associated variation in blood lipids of young male distance
runners

Table 3.2. Age-associated variation in blood lipids of young female
distance runners

Table 3.3. Means and standard deviations for blood lipids at baseline and
at last visit in longitudinal cohorts of male and female distance runners

Table 3 4. Comparison of blood lipids of adolescent and adult distance
runners and non-athletes

Table 4.1. Characteristics of young distance runners
Table 4.2. Distribution of subjects by pubertal status

Table 4.3. Blood lipids of young distance runners compared to reference
means and medians

Table 4.4. Distribution of subjects by clinical cut-points
Table 5.1. Characteristics of young distance runners
Table 5.2. Blood lipids of young distance runners
Table 5.3. Partial correlations, controlling for chronological age and
pubertal stage, for training volume, peak V02, body fatness, and blood

lipids in young distance runners

Table 6.1. Mean peak oxygen consumption in cross-sectional studies of child
and adolescent endurance athletes

Table 6.2. Longitudinal studies of physiological capacity in child and
adolescent endurance athletes

Table 6.3. Summary of scaling factors for peak Vo2 in children and
adolescents

ix

49

69

7O

71

72

ER

132

133

134

159

160

161

Table 6.4. A comparison of two elite distance runners of differing body
mass and submaximal and peak Vo2 based on the simple ratio standard and
allometric scaling

Table 7.1. Age-specific values for body size and peak Vo2 in male distance
runners

Table 7.2. Age-specific values for body size and peak Vo2 in female
distance runners

Table 7.3 Age- and sex-specific proportionality coefficients (a) and
allometric scaling factors (b) in young distance runners

Table 7.4. Diagnostic criteria for the relationship between peak Vo2 and
body size

Table 7.5. Summary of allometric ontogenetic scaling factors in children
and adolescents

162

194

195

196

197

198

LIST OF FIGURES

Figure 2.1. Mean values by sex and age for cholesterol
Figure 2.2. Mean triglyceride values by sex and age

Figure 3.1a. Total cholesterol in male distance runners compared to
age-specific reference values

Figure 3.1b. Hi gh-density lipoprotein cholesterol in male distance runners
compared to age-specific reference values

Figure 3.1c. Low-density lipoprotein cholesterol in male distance runners
compared to age-specific reference values

Figure 3.1d. Triglycerides in male distance runners compared to
age-specific reference values

Figure 3.2a. Total lipoprotein cholesterol in female distance runners
compared to age-specific reference values

Figure 3.2b. Hi gh-density lipoprotein cholesterol in male distance runners
compared to age-specific reference values

Figure 3.1c. Low-density lipoprotein cholesterol in female distance runners
compared to age-specific reference values

Figure 3.1d. Triglycerides in female distance runners compared to
age-specific reference values

Figure 4.1a. Total cholesterol in young distance runners

Figure 4.1b. High-density lipoprotein cholesterol in young distance runners
Figure 4.1c. Low-density lipoprotein cholesterol in young distance runners
Figure 4.1d. Triglycerides in young distance runners

Figure 5.1a. Inter-relationships among training volume, sum of six skinfolds,
peak oxygen consumption, and high-density lipoprotein in 45 young male
distance runners 10 to 19 yrs of age

Figure 5.1b. Inter-relationships among training volume, sum of six skinfolds,
peak oxygen consumption, and low-density lipoprotein in 45 young male

distance runners 10 to 19 yrs of age

xi

51

52

73

74

75

76

78

79

100

101

102

103

135

136

Figure 5.10. Inter-relationships among training volume, sum of six skinfolds,
peak oxygen consumption, and triglycerides in 45 young male

distance runners 10 to 19 yrs of age 137
Figure 6.1. Longitudinal studies of absolute peak Vo2 in male athletes 163
Figure 6.2. Longitudinal studies of relative peak Vo2 in young athletes 164
Figure 7.1. Longitudinal studies of absolute peak Vo2 in male athletes 199

Figure 7.2. Longitudinal studies of relative peak Vo2 in young athletes 199

xii

CHAPTERI

INTRODUCTION

It is a basic assumption that physical activity has an essential influence on
biological growth and maturation (Malina, 1969; Rarick, 1960), although the specific
amount of physical activity that is needed has not been established. Physical activity
occurs across a broad spectrum, from very light activities to intensive training for sport.
Both extremes of the spectrum have received attention due to the potential influences of
physical activity on biological and psychological processes, and the health of the human
organism. Possible negative consequences of intensive training during childhood and
adolescence are often a concern among physicians, coaches, and parents. The potential
physiological benefits of intensive u'aining during growth and maturation have received
corresponding attention by the scientific community.

This dissertation considers the influence of biological growth, maturation, and
intensive endurance exercise training on blood lipids and peak oxygen consumption
(peak V02) and contributes to the disciplines of human growth, cardiovascular disease
epidemiology, and pediatric exercise physiology. The timing of this work is relevant to
the progress of pediatric exercise science given current emphasis on the health outcomes
of exercise in children related to the prevention of chronic disease, and the expression of
physiological variables relative to variation in body size of growing children and
adolescents.

This dissertation is divided into two parts, each consisting of a literature review
and a series of papers that adds to our understanding of the child and adolescent athlete.
Each paper is in manuscript form (i.e., abstract, introduction, methods, results, and 1

discussion). Some aspects of each paper, particularly the methods sections, are repetitive,

but lend to the readability of the paper. The final chapter provides a summary and
recommendations for future research.

Part I focuses on blood lipids of competitive young distance runners and
encompasses Chapters 2 through 5. Chapter 2 introduces the reader to the importance of
the study of physical activity and blood lipids during childhood and adolescence, and
reviews several aspects of the topic including: blood lipids and coronary heart disease,
evidence for the early origins of atherosclerosis, the development of blood lipids in the
general pediatric population and in young athletes, and the relationship between physical
activity and blood lipids in youth. Chapters 3 and 4 describe age-associated variation and
distribution of blood lipids in young distance runners compared to the general population.
Chapter 5 examines the heterogeneity of blood lipid phenotypes in young distance
runners by considering the influence of training volume, peak V0,, and body fatness.

Part II focuses on growth-related changes in peak Vo2 of competitive young
distance runners and encompasses Chapters 6 and 7. Chapter 6 provides an introduction
to the problem of interpreting growth-related changes in peak Vo2 and reviews age-, sex-,
and maturity-associated variation in peak Vo2 in the general population of youth and in
youth athletes, and the concept of allometric scaling. Chapter 7 compares the use of
traditional and allometric scaling techniques in the interpretation of growth-related
changes in peak Vo2 of young distance runners.

Two independent samples of young distance runners from the mid-Michigan area
are included in the analysis. An earlier mixed-longitudinal study of 27 male and 27
female distance runners (Young Runners Study, YRS I) aged 8-18 years was used to

examine the age-related changes in blood lipids and peak Vo2 across the adolescent age

range. The study was conducted by the Institute for the Study of Youth Sports at
Michigan State University as part of an interdisciplinary investigation of the influence of
intensive endurance training and competition on biological and psychological outcomes
from 1982 and 1986 (Seefeldt, 1986). During the 1999-2000 academic year, a cross-
sectional study was conducted to examine the dose-response relationship between
training volume (i.e., running mileage) and blood lipids in young distance runners
(Young Runners Study II, YRS 11). Body size and peak Vo2 were also measured in this

study.

REFERENCES
Malina RM (1969) Exercise as an influence upon growth. Clin Pediar 8: 16-26.

Rarick GL (1960) Exercise and growth. In WR Johnson (ed.): Science and Medicine of
Exercise and Sport. New York: Harper and Brothers, pp. 440-465.

Seefeldt VD (1986) Elite young runners: an interdisciplinary perspective. In M Weiss
and D Gould (eds.): Sport for Children and Youths. Charnpaign, IL: Human Kinetics, pp.
213-284.

CHAPTER 2
REVIEW OF LITERATURE

PART I

INTRODUCTION

There is a considerable current interest in the physical activity and health-related
physical fitness of youth (Malina, 1997), particularly related to risk factors for coronary
heart disease (CHD) (Casperson et al., 1998; Despres et al., 1990a). Although the
clinical manifestations of atherosclerosis are not evident until adulthood, childhood
precursors of the disease are clearly evident (Berenson et al., 1998; Mahoney et al.,
1996). As a result, preventive strategies for CHD during childhood and adolescence have
been recommended.

A unique approach to understanding the association between a causal factor (i.e.,
physical activity) and a health outcome (i.e., blood lipids) is the study of a special
exposure cohort (Rothman and Greenland, 1998). Although this approach is limited due
to selection bias, special exposure cohorts permit the study of a range of exposures and
outcomes that may not be common in the general population, thus providing a
comprehensive model of the health effects of a given exposure.

Since regular physical activity, a relatively high aerobic fitness level, and a low
amount of adiposity have beneficial effects on CHD risk factors, morbidity, and mortality
in adults (Blair et al., 1989; Paffenbarger and Lee, 1996), many investigators have been
interested in the blood lipid profile of well-trained endurance athletes (Haskell, 1984). In
general, adult endurance-trained athletes have superior blood lipid profiles compared to
the general population (Haskell, 1984). However, limited information is available on the

blood lipid profile of young athletes.

The purpose of this review is to examine the early origins of atherosclerosis, the
development of plasma triglycerides and lipoproteins, and the associations between
physical activity, aerobic fitness, body fatness and blood lipids with special reference to

the child and adolescent athlete.

BLOOD LIPIDS AND CORONARY HEART DISEASE

Atherosclerosis is generally described as a slowly progressing disease in which
there are focal lesions in the large arteries that rarely produces symptoms until middle age
and that often go undiagnosed until the time of the first myocardial infarction (Woolf,
1999). Several risk factors or biomarkers have been identified for CHD (Hopkins and
Williams, 1981). Since Part I of this dissertation focuses on the blood lipids of young
distance runners, lipoprotein metabolism, the etiology of blood lipids in atherosclerosis,
and the causal relationship between blood lipids and CHD will be briefly considered here.
For a complete discussion of lipoproteins in health and disease, the reader is referred
elsewhere (Betteridge et al., 1999).

Lipoprotein metabolism is a dynamic and complex system designed to transport
lipids between the intestine, liver, and peripheral tissues via the plasma and interstitial
fluid. In the blood circulation, blood lipids and lipoproteins undergo a complex series of
modifications that alter their structure, composition, and function. After binding to a
receptor, lipoproteins are internalized by cells and used for energy production or storage,
membrane biogenesis, or sex steroid synthesis. Given the dynamic nature of this system,
a single cholesterol or triglyceride measurement is at best a snap shot of a moving

picture.

Cholesterol is carried in the bloodstream by protein-lipid combinations known as
lipoproteins. Four basic classes of lipoproteins have been categorized according to their
gravitational density: chylomicrons, very-low density lipoprotein (VLDL), low-density
lipoprotein (LDL), and hi gh-density lipoprotein (HDL). Other lipoprotein subfractions
include intermediate-density lipoprotein (IDL) and lipoprotein (a) [Lp (a)]. Most (60-
75%) of the cholesterol in humans is carried by LDL. The cholesterol-rich LDL particles
infiltrate the arterial intima and form a major part of the buildup in the artery wall to form
atherosclerotic plaque. In contrast, HDL carries cholesterol away from the arterial wall
to the liver for catabolism and excretion, thus providing a protective effect against
atherosclerosis. Any disruption of the delivery of cholesterol to peripheral tissues
(referred to as the LDL receptor pathway or the lipolytic cascade) and/or the return of
cholesterol to the liver (referred to as reverse cholesterol transport or the HDL cascade)
can cause changes in the plasma lipoprotein profile and result in CHD.

In 1984, the National Heart, Lung, and Blood Institute (NHLBI) and the National

Institutes of Health Office of Medical Applications of Research conducted a Consensus
Development Conference on Lowering Blood Cholesterol to Prevent Heart Disease to
address the causal relationship between blood cholesterol levels and CHD (NIH
Consensus Statement, 1985). Based on the genetic, experimental, and epidemiological
data, the panel of experts concluded as follows:
”Elevation of blood cholesterol levels is a major cause of coronary artery disease. It has
been established beyond a reasonable doubt that lowering definitely elevated blood
cholesterol levels (specifically, blood levels of low-density lipoprotein cholesterol) will
reduce the risk of heart attacks caused by coronary heart disease (p. 2080).”

In a more recent consensus, the importance of HDL and triglycerides (TG) in the

pathogenesis of atherosclerosis was explored (NIH Consensus Development Panel on

9

Triglyceride, 1993). The panel confirmed findings of four major studies that a l mg/dl
increase in HDL results in a 2 to 3% decrease in CHD risk after adjustment for other risk
factors. However, the independent contribution of TG on CHD risk remains

inconclusive.

EARLY ORIGINS OF ATHEROSCLEROSIS

Even though the clinical manifestations of CHI) occur in adulthood, various
studies have shown that atherosclerosis has its origins early in life (Berenson et al., 1998;
Enos et al., 1955; Mahoney et al., 1996; McNamara et al., 1971; Strong et al., 1997).
This rationale stems from autopsy reports of coronary atherosclerotic lesions in youth, the
prevalence of CHD risk factors in youth, the tracking of CHD risk factors from
childhood to adolescence to young adulthood, and the predictability of adult heart disease
from childhood and adolescent CHD risk factors.

Early autopsy studies of soldiers. Fatty streak and plaque formation of the
coronary arteries and aorta has been documented by autopsy studies of American soldiers
from the Korean and Vietnam Wars. The first report indicated that 232 of 300 (773%)
young (mean age, 22.1 yrs) American soldiers of the Korean War demonstrated some
gross evidence of CHD (Enos et al., 1955). In a study of US. soldiers (mean age, 22.1
yrs, range 18-37 yrs) of the Vietnam War, 79 of 105 (75.2%) cases demonstrated some
gross evidence of CHD (McNamara et al., 1971). In a more detailed analysis, 47 (44.8%)
exhibited some degree of atherosclerosis, 27 (25.7%) had involvement of two or more

vessels, and 5 (4.8%) had severe evidence of atherosclerosis.

10

The Bogolusa and Muscatine Studies. In the 19703, two major epidemiological

studies, the Bogolusa Heart Study and the Muscatine Study, began to investigate the

development of CHD risk factors in children and adolescents. The Bogolusa Heart Study
initially began during the 1973-74 school year in Washington Parish, Louisiana, 3
political ward consisting of a bi-racial population of approximately 22,000 (63% White,
37% Black). The center was designated as a Specialized Center of Research for
Atherosclerosis by the National Heart, Lung, and Blood Institute. Since the initial cross-
sectional survey, several follow-up surveys have been conducted (1976-77, 1978-79,
1981-82, 1984-85, 1987-1988, 1993-94).

The Muscatine Study began during the 1971-72 and 1972-73 academic years and
thereafter included biennial surveys in school-aged children and adolescents and a
follow-up survey between the ages of 20 and 34 yrs (Muscatine Young Adult Follow-Up
Survey). The initial study population included school children of Muscatine, Iowa. Like
the Bogolusa site, Muscatine was chosen due to the relative stability of the population
and its proximity to the medical examining team (University of Iowa, Iowa City). In
contrast to the Bogolusa Heart Study, the sample in the Muscatine Study consisted of a
majority of Caucasians (about 96%).

The first published reports from both studies consisted of the descriptive
epidemiology of CHD risk factors in preschool children, school-aged children and
adolescents (Berenson etal., 1978; Frerichs et al., 1976; Lauer et al., 1975; Srinivasan et
al., 1976). The age-, sex-, and race-associated variation for lipids will not be described
here but rather as part of the section of the review on the development of blood lipids (see

below). During subsequent follow-up studies, the tracking of CHD risk factors has been

11

determined over various time periods (i.e., 3-8 yr intervals) (Bao et al., 1994; Bao et al.,
1995b; Clarke etal., 1978; Shear et al., 1986). Along with the critical issue of the
persistence of CHD risk factors and the predictability of adult CHD, another major
component in establishing evidence in the early origins of atherosclerosis has been to
determine the inter-relationships of CHD risk factors, environmental factors, and family
history. An important finding from such analyses has been that CHD risk factors cluster
in obese children and adolescents (Smoak et al., 1987). Furthermore, 61% of those
subjects in the upper quartile of multiple risk factors during childhood remained in the
upper quartile as young adulthood (Bao et al., 1994).

Familial studies have indicated that adverse CHD risk factors are more common
in offspring of parents and relatives with hypertension, diabetes, obesity, hyperlipidernia,
and history of myocardial infarction (Bao et al., 1997; Bao et al., 1995a; Muhonen et al.,
1994). Besides the associations between familial history of CHD and CHD risk factors in
offspring, genetic studies have also been conducted to determine candidate genes for
'CHD risk factors in these two studies (Amos et al., 1989; Anderson et al., 1989; Bucher
et al., 1982; Srinivasan et al., 1996).

Recent autopsy reports from the Bogolusa Heart Study have furthered
understanding of the relationship between antemortem childhood CHD risk factors and
the extent of fatty streaks and fibrous plaques in the aorta and coronary arteries (Berenson
et al., 1998). In 204 autopsy cases, 2-39 yrs of age, 93 cases bad data on antemortem risk
factors. Results indicate that the body mass index (BMI), blood pressure (BP), TC, TG,
LDL, and HDL, as a group, are strongly associated with the extent of lesions (canonical

correlation, r=0.70). The effect of multiple risk factors, or clustering, on the percent of

12

 

intimal surface covered with fatty streaks indicate that risk factors during childhood or
adolescence are strongly related to the extent of lesions in the aorta and coronary arteries
of young adults, and that as the number of risk factors increases so does the severity of
asymptomatic coronary and aortic atherosclerosis in young people.

Using a noninvasive method known as computerized beam tomography to detect
the presence of atherosclerotic plaque in asymptomatic subjects with CHD, The
Muscatine Study has also shown the relationship between childhood CHD risk factors
and coronary calcification in young adulthood (29 to 37 yrs, mean age = 33 yrs)
(Mahoney et al., 1996). Results indicate that 31% of men and 10% of women have
coronary artery calcification. Although CHD risk factors measured during the previous
and most recent visits showed the strongest associations with coronary artery
calcification, childhood (8—18-yrs, mean age=15 yrs) body weight was a significant
predictor of adult coronary artery calcification [Odds ratio (OR) = 3.0]. SBP, DBP, TC,
and TG measured during childhood were not significantly different (p>0.05) between
young adults showing the presence and absence of coronary artery calcification. Body
weight, the BMI and triceps skinfold thickness were larger (p<0.01) in males, but not
females, during childhood among those who displayed coronary calcification.
Unfortunately, HDL, LDL, and apolipoprotein levels were not screened during childhood
in this sample.

The Bogolusa Heart Study and the Muscatine Study have provided valuable
information regarding the development and persistence of CHD risk factors from

childhood through adolescence into young adulthood. As these studies continue, they

13

will provide additional information on the long-term persistence of CHD risk factors into
rnid- and late-adulthood.

Pathobiological Determinants of Atherosclerosis in Youth (PDAY). In 1984, a
multi-institutional (9 centers) study was organized to document the pathobiology of
lesion development and its association with postmortem CHD risk factors in 15 to 34 yr
olds who died from external causes (i.e., homicide, accident, or suicide) (Strong et al.,
1998). Between June 1, 1987, and August 31, 1994, arteries and other tissues from about
3000 autopsy cases were examined for lesions and surrogates or markers for preexisting
risk factors. Post-mortem risk factors included serum lipoprotein cholesterol, serum
thiocyanate (smoking), wall thickness of the small renal arteries (BP), glycohemoglobin
(impaired glucose tolerance and diabetes), the BMI, panniculus thickness (adiposity),
adipose tissue fatty acids (diet), and apolipoprotein polymorphisms. The degree of
atherosclerosis was measured as follows: collagen and cholesterol content in lesion-prone
and lesion-resistant areas of the aorta; macrophage, smooth muscle cell, T-lymphocytes,
and mast cell counts in various lesion types; and apolipoprotein B, Lp (a) and oxidized
LDL in the arterial wall measured chemically and morphometrically by pathologists and
by computer image analysis. Results indicate that fatty streaks covering 15-25% of the
intimal surface are well established by age 15-19 yrs and progress until 30-34 yrs of age.
Among the postmortem risk factors, VIDL, LDL, glycohemoglobin, and thiocyanate are
positively, and HDL negatively related to fatty streak involvement. The prevalence of
raised lesions (5% or greater of intimal surface) 2-fold greater in hypertensive (classified
by renal index) men 15-24 yrs of age. The BMI was associated with more extensive fatty

streaks and raised lesions in right coronary artery but not in the aortas of men. No

14

association between the BMI and lesions were found in women. Thickness of panniculus
adiposus was associated with more extensive lesions in the right coronary artery in both
men and women.

Summary. Historically, insight into the early origins of atherosclerosis were
gained from autopsy studies of children and American soldiers. Since the early autopsy
studies, three studies - Bogolusa, Muscatine, and PDAY - have provided greater insights
into the development of CHD risk factors, the natural history of CHD, and the association
of risk factors with the development of CHD. The findings indicate that atherosclerosis
is a progressive disease that has its origins in childhood and adolescence, and highlight

the importance of the prevention of adult CHD during childhood and adolescence.

DEVELOPMENT OF BLOOD LIPIDS IN THE GENERAL PEDIATRIC
POPULATION

With the realization of the early origins of atherosclerosis, pediatricians and other
health professionals recognized the need to begin examining the CHD risk factors of
children and adolescents as a means of establishing primary prevention. Besides the
Bogolusa Heart Study and the Muscatine Study, the major population study of blood lipid
distributions in US. children and adolescents is the Lipid Research Clinics Prevalence
Study (The Lipid Research Clinics, 1980). This study was initiated in 1971 and included
13, 665 children and adolescents 6 to 18 yrs old. Data from this large population study
has been used to establish reference values for clinical medicine and the descriptive
epidemiology of blood lipid distributions in children and adolescents. The Third

National Health and Nutrition Examination Survey (NHANES III) conducted between

15

1988 and 1994 provides current data on the blood lipid distributions of U.S. children and
adolescents (Hickman et al., 1998). Age-, sex-, maturity-, race-, and time-associated
variation in blood lipid distributions in children and adolescents are subsequently
described.

Age-associated variation. Age-associated variation in TC, HDL, LDL, and TG
during childhood and adolescence are shown in Figures 2.1 and 2.2. Following the first
year of life, median values for TC, HDL, and LDL are somewhat stable throughout the
first two decades of life, whereas TG increase throughout this period. The pattern of
development for TC and LDL shows relatively stable levels until adolescence (about 160-
165 mg/dl and 90 mg/dl, respectively), a decline during adolescence, and an increase in
TC and LDL during late adolescence. The pattern of development for HDL shows
relatively stable levels until adolescence (about 50 mg/dl) and a decline during
adolescence (especially in males, see below), which remain at relatively stable levels into
young adulthood. The decrease in TC is mainly due to the decrease in HDL as the
decrease in LDL is modest. During late adolescence, the increase in TC is a result of the
increase in LDL. Depending upon how the data are reported, the age-related trend for
TG is quite variable. Some reports show a relatively stable level prior to adolescence
with an increase during and following adolescence (Cresanta et al., 1984), whereas others
indicate an increase from 6 to 19 yrs of age (T amir et al., 1981). The discrepancy
between studies appears to be due to the age groups of the subjects (i.e., 1 year vs.
multiple year grouping).

Sex-associated variation. Sex differences in TC, HDL, LDL, and TG during

childhood and adolescence are also shown in Figures 2.1 and 2.2. Results from

16

NHANES III indicate that females have slightly greater mean levels of TC (167 v. 163
mgldl) and LDL (99 vs. 91 mgldl) compared to males (Hickman etal., 1998). This
observation is consistent across age and ethnic groups. Prior to adolescence, HDL is
slightly greater in boys, but the sharp decline during puberty results in higher levels of
HDL in girls that persists into adulthood. TG increases during this period with a marked
sex difference. Boys exhibit an increase from about 50 mg/dl at age 6 to 90 mg/dl at age
18, while the increase in girls during this same age span is from 65 to 80 mg/dl (T arnir et
al., 1981).

Maturity-associated variation. Biological maturation refers to the timing and
tempo of progress toward a mature (adult) state (Malina and Bouchard, 1991). Biological
maturation is associated with dynamic changes in body size and composition, sex
hormones, and various physiological parameters. Therefore, it is important to consider
biological maturation in pediatric studies, particularly during adolescence, as a distinct
measure from chronological age.

Biological maturation can be assessed by skeletal, sexual, or somatic maturity
characteristics. Typically, sexual maturation has been used in studies examining blood
lipids during adolescence (Armstrong et al., 1992; Berenson et al., 1981; Morrison et al.,
1979; Tell, 1985). The assessment of sexual maturation is based on the development of
the secondary sex characteristics - breasts, gentitals, and pubic hair - in both sexes as
described by Tanner (1962). When reviewing studies, it is important to consider the
presentation of sexual maturity status. A specific stage should be characterized for each
indicator (i.e., pubic hair stage 2), rather than an "average" stage or simply ”Tanner stage

2".

l7

As with many other biological variables, there are dynamic changes in blood
lipids during pubescence before adult patterns are established. In general, the pattern of
development of blood lipids during pubescence follows that for age-related changes -
TC, HDL, and LDL decline and TG increases (Armstrong et al., 1992; Berenson et al.,
1981; Morrison et al., 1979; Tell, 1985). However, blood lipids are more closely related
to sexual maturity status than to chronological age (Tell, 1985), and greater changes are
observed when plotted by maturity status than chronological age (Siervogel et al., 1989;
Tell, 1985). Additionally, early maturers have lower HDL compared to late maturers
when assessed by sexual maturity (Tell et al., 1985), but not by skeletal or somatic
maturity (Siervogel et al., 1989).

Maturity-associated changes in blood lipids have led to investigations of the inter-
relationship of pubertal changes in sex hormones, body size, and blood lipids. In the
Princeton Maturation Study, estradiol, testesterone, the Quetelet index (BMI), and their
interactions explained 47%, 76%, 87%, and 56% of the variance in HDL, LDL,
LDL:HDL, and TG, respectively in 30 adolescent boys (Laskarzewski et al., 1983).
Complex interactions between blood lipids and testosterone at varying levels of estradiol
and the Quetelet index were also demonstrated. This study highlights the significant
contributions and complex interactions of sex hormones and body size on blood lipids
during adolescence in boys.

Race-associated variation. Given the bi-racial sample of the Bogolusa Heart
Study, the first reports of racial differences in blood lipids were demonstrated in early
publications from this study. In general, Black children have higher mean levels of TC

and HDL, and lower TG than White children (Frerichs et al., 1976). Recent data from

18

NHANES III confirm these findings and also allow comparison to Mexican-American
children (Hickman et al., 1998). Non-Hispanic Blacks had the highest mean TC and
HDL, whereas values were similar between non-Hispanic Whites and Mexican
Americans for TC. LDL in 12 to 19 yr olds indicated similar values between White and
Black females and greater values in Black males. Similar values for TG were also
observed between Mexican American and White adolescents of both sexes.

International comparisons of a given variable indicate the biological variability
among diverse populations. Geographic variation is present among the available data
representing world populations. Labarthe et al. (1994) concluded that there is no unique
population, but rather a continuous distribution of values and no distinct outliers. Among
world populations, African youth have the lowest TC values (about 130 mgldl) and
Finnish youth have the highest (about 195 mgldl).

Secular trends in lipid distributions of children and adolescents. Based on the
data from three national representative samples (NHES III, NHANES I, and NHANES
III), there appears to be a downward trend in mean TC among sex- and race-groups 12 to
17 yrs (Hickman et al., 1998). Mean declines in TC among White males and females and
Black males and females are 8, 7, 5, and 4 mgldl, respectively. The mean decline in TC
is 7 mg/dl when the total sample is considered. Secular changes in lipoprotein
subfractions and TG were not considered. Compared to the LRC data, it appears that TG
has increased and HDL has decreased in White males and females, while LDL has
declined in White males and increased in White females. Cautioned is urged in the

comparisons as age-group classifications vary between surveys.

19

Prevalence of hypercholesterolemia. The National Cholesterol Education
Program (NCEP), Report of the Expert Panel on Blood Cholesterol Levels in Children
and Adolescents recommends that an acceptable TC value for youth 2-19 yr of age be
<170mg/dl (4.4mmol/1), which corresponds with the 75th percentile of the US population
and abnormally elevated levels be determined by a TC >200mg/dl (5.17mmol/l) which
corresponds to the 95th percentile of the U.S. population (National Cholesterol Education
Program, 1992). Recommended clinical cut-points for other lipoproteins are found in
Table 2.1.

Despite a detailed account of the developmental pattern of blood lipid levels in
youth, the prevalence of hypercholesterolemia in the pediatric population is often not
reported. Results from 6500 White middle-class children (mean age, 6.4 yrs) seen at
private pediatrician offices for well-child visits indicate that 19% had a TC exceeding
185 mg/dl and 8.5% had a TC exceeding 200 mg/dl (Garcia and Moodie, 1989). The
estimation of the prevalence of hypercholesterolemia in this study may be questioned due

to the representiveness of clinical samples.

BLOOD LIPIDS IN CHILD AND ADOLESCENT ATHLETES

Cross-sectional comparisons of individuals engaged in endurance training and the
general population have been used to demonstrate differences between trained and
untrained individuals and by inference the influence of physical activity or exercise
training on the blood lipid profile. As previously mentioned, comparisons of athletes and
controls introduce selection bias. However, establishing an understanding of the

influence of intensive training on the blood lipid profile requires the recruitment of

20

individuals engaged in endurance training since the general population does not regularly
engage in such energy expenditure or vigorous physical activity. Despite the possible
genetic pre-disposition of the athletes, some of the variation in a biological variable can
be explained by environmental factors and the genotype-environmental interaction in an
athlete.

Compared to adult athletes, little attention has been given to the study of the
blood lipid profile in child and adolescent athletes. The child and adolescent athlete is
often ill-defined, so that the evaluation and comparability of studies can be difficult.
Table 2.2 provides a summary of serum lipoproteins in seven studies of child and
adolescent athletes. Based on these reports, Rowland (1993) concluded that prepubertal
athletes possess a favorable lipoprotein profile. The appropriateness of this conclusion
can be questioned given that the comparisons were made to control subjects. The blood
lipid levels of the control subjects should have also been compared to reference values
since the control subjects were a convenient sample rather than a random sample in most,
if not all, instances. If the control values were different than the reference medians (i.e.,
TG lower, etc.) the interpretation of results from the individual studies and the general
conclusions may be questioned.

Aside from selection bias, other limitations also exist in studies of blood lipids in
young athletes. Sample sizes are generally small and even smaller when considered by
specific age groups (i.e., 707.99). It is thus difficult to establish age-associated
variation for blood lipids of young athletes across the pediatric age range. The normal
pattern of development of blood lipids in this group remains to be established by a

longitudinal study. Some studies also combine sexes, thus not alldwing for the

21

consideration of sex-associated variation. Limited and inconsistent information on
training history is generally provided. When training information beyond frequency
(days per week, months per year, etc.) is provided, it is often reported as hours per week.
Many activities may occur during exercise training such as warm-up, flexibility
exercises, etc., which actually do not increase energy expenditure at levels near those
obtained during vigorous exercise training. By observing activity patterns during practice
and competition, it has been determined that about 40% of the total time spent during
sport activities (basketball, soccer) was occupied by sitting and standing (Katzmarzyk
and Malina, 1997). More than likely, the percentage of time spent in inactivity for
distance running or swimming is minimal given the nature of the sports. Finally, as with
cross-sectional studies in the general population, studies of young athletes have failed to
consider other biological factors that influence blood lipids (i.e., biological maturity, .
fatness, diet, family history, etc.). 7
Since it is generally assumed that child and adolescent athletes have a superior
lipid profile given their levels of habitual physical activity and intensive training, little
information is available on the prevalence of hyperlipidernia in young athletes. Recent
attention has been given to pre-participation screening examinations, including blood
pressure and cholesterol, of young athletes due to sudden cardiac death among some
young athletes (Gutgessel et al., 1997). In a 1988 survey of 777 student athletes (454
males, 323 females) 11 to 15 yrs, 15.4% and 13.6% of boys and girls, respectively, had
TC levels above 185 mg/dl (>90th percentile of LRC) (Kyle et al., 1991). Mean TC in
boys and girls were 155 and 158 mgldl, respectively, and the range for the entire sample

was 65 to 274 mgldl. Of the 114 student athletes with an initial TC >185 mgldl, 74

22

(response rate = 65%) were returned for a second cholesterol test. Total cholesterol
remained above normal in 38 (51%) of those re-tested. Unfortunately, the sample was
not stratified by sport and information on the training experience or other confounding
variables (family history, sexual maturity status, etc.) were not considered. Thus, despite
participation in youth sports, young athletes may display hypercholesterolenria. Future
research should provide descriptive information and establish prevalence rates of
dyslipidemia among youth athletic groups (i.e., endurance, strength/power) and in

specific sports (cross-country, basketball, football, etc.).

PHYSICAL ACTIVITY AND BLOOD LIPIDS IN CHILDREN AND
ADOLESCENTS

Current interest in the prevention of CHD and influence of exercise intervention
on blood lipids in childhood. and adolescence has gained attention among researchers in
clinical pediatric cardiology, cardiovascular disease epidemiology and pediatric exercise
science. The relationship between physical activity and blood lipids in children and
adolescence has been reviewed extensively (Armstrong and Simons-, 1994; Casperson et
al., 1998; Despres et al., 1990a; Tolfrey et al., 2000). Complete summary tables of cross-
sectional, prospective cohort, and experimental studies conducted within various samples
of children and adolescents are provided in these reviews. Only major large-population
cross-sectional studies are reviewed here, along with a limited number of prospective and
retrospective cohort and experimental studies available. No case-control studies have

been apparently been reported in the literature.

C ross-sectional studies.

Young Hearts Study. The Young Hearts Study began in 1990 and is an ongoing
study of CHD risk factors in youth from Northern Ireland (Boreham et al., 1997). The
study population consisted of 1015 schoolchildren (251 12 yr old boys, 258 12 yr old
girls, 252 15 yr old boys, 254 15 yr old girls) randomly selected from 16 schools.
Physical activity was assessed by an interviewer-administered questionnaire and a
physical activity score was computed. Sports participation was based on the number of
sports or other physical activity sessions reported aside from school-related physical
activity. Blood lipids were assayed according to World Health Organization (WHO)
standards. Results of stepwise multiple linear regression controlling for dietary intake,
cigarette smoking, social class, school type, sexual maturity, and body size indicated that
physical activity was associated with TC:HDL in 15 yr old boys. In boys, a 20%
difference in physical activity was associated with a 1.54 -fold increase in the probablilty
of being in the high risk group for TC:HDL (>4.0). No significant relationships between
physical activity, sports participation, and TC:HDL were found in girls.

- Oslo Heart Study: In a sample of 431 boys and 397 girls 10-15 yrs from six Oslo
schools, self-reported physical activity was related to TG in girls (r: -0.13) but not boys.
When grouped by activity status, TG was found to be similar in boys participating in little
to moderately frequent physical activity and significantly lower in girls reporting
engaging in physical activity at least 2-3 times per week compared to infrequently (54.8
mg/dl v. 65.3 mgldl) (Tell and Vellar, 1988). TC, HDL and TC:HDL were not

significantly associated with physical activity. Physical activity frequency and intensity

was based on the response to how often the subject exercised (for at least half an hour) so
that they were out of breath and sweating.

Maximal aerobic power was positively associated with HDL and HDL:TC and
negatively associated with TC and TG in both boys and girls. However, correlations
were low (0.05>r<0.25). When subjects were grouped into quartiles Of peak V02, there
was a significant linear trend for HDL and TG in girls. TC was lower in more aerobically
fit boys and girls, but there was no significant trend. These data suggest a dose-response
relationship between aerobic fitness and blood lipids in youth.

Singapore Youth Coronary Risk and Physical Activity Study. This sample
included 1579 schoolchildren 6—18 yr of age from 12 schools in six geographic regions of
Singapore (Schmidt et al., 1998). Physical activity was assessed by self-report. In boys,
physical activity was significantly correlated with TC (r = -0.13) and TG (r = -0.18),
while no significant relationships were found in girls. When boys and girls were grouped
by an arbitrary cut-point of physical activity status (inactive to vigorous physical
activity), the only significant difference was between those with relatively no activity and
those with vigorous physical activity (TC, 159.7 v. 136.7 mgldl, respectively). The other
three groups possessed a mean TC of 147 mgldl.

Cardiovascular Risk in Young Finns Study. The Young Finns Study is a
multicenter study of athersclerotic precursors in Finnish children and young adults
(Raitakari et al., 1997). The initial cohort in 1980 included 3596 children and young
adults aged 3, 6, 9, 12, 15, and 18 yrs of age. Two follow-up studies were conducted in
1983 and 1986. Physical activity was assessed by questionnaire including frequency and

intensity. An index of physical activity was calculated from the product of frequency,

25

intensity, and duration. Subjects were then grouped into high, moderate, and low levels
of physical activity based on arbitrary cut-points of the activity index. Blood lipid

measurements included TC, HDL, HDL2, HDL3, LDL, Apolipoprotein B (Apo B), and
serum lecithinzcholesterol acyltransferase (LCAT).

Although not significant, TC and LDL were lower in boys in the high physical
activity group. A linear trend for an increase in HDL and HDL? and a decrease in Apo B
with increasing physical activity level was found in boys, whereas a decrease in T6 was
associated with increasing physical activity in both sexes. No significant association was
found between physical activity and LCAT.

Summary: In general, correlations between physical activity and blood lipids are
low and often non-significant. When grouped by physical activity or aerobic fitness
level, youth in the upper extreme display a better blood lipid profile. This difference is
apparent moreso in boys than girls. However, the data are inconsistent. Methodological
shortcomings limit previous cross-sectional studies in this area (Armstrong and Simons-,
1994; Casperson et al., 1998). Specifically, the measurement of habitual physical activity

and the lack of control of confounding variables may distort the findings from cross-
sectional studies.

Prospective and retrospective cohort studies. A limited number of prospective
and retrospective cohort studies have been conducted to examine the influence of
physical activity on blood lipids in children and adolescents. Results from two
prospective cohort studies suggest that baseline physical activity level is inversely related
with TG levels in 3-4 yr olds in a1 yr follow-up (DuRant et al., 1993), and 12, 15, and 18-

yr old youth in the Young Finns Study in a 3 yr follow-up (Porkka et al., 1994)

26

In cohort studies, sports participation or physical education interventions have
been used as a proxy for habitual physical activity in youth. Using this approach to
determine high school activity status in middle-aged men, a retrospective cohort study
was conducted by researchers from the Institute for Aerobics Research. The results
indicated that TG and TC did not differ between former high school or college athletes
and non-athletes (Brill et al., 1989). In a prospective study of the long-term effects of
increased physical education on adult health outcomes, the blood lipid profile did not
differ between experimental and control men or women (Trudeau et al., 2000). Thus,
prior athleticism or physical education, which are assumed to be associated with higher
energy expenditure, have little apparent impact on adult health outcomes, including blood
lipids.

Experimental studies. Few well-designed experimental studies have examined. the
influence of increased physical activity or exercise training on blood lipids. In general,
exercise training studies have failed to have a significant impact upon the lipoprotein
' profile of children and adolescents (Casperson et al., 1998). Tolfrey et al. (1998b) noted
that several deficiencies in the available exercise training studies and attempted to
overcome such limitations. Twenty-eight (14 boys, 14 girls) ”prepubertal" children
(mean age 10.7 + 0.7 years) completed a 12 week exercise training program consisting of
stationary cycling 3 times per week at 80% of maximum heart rate for 30 minutes. To
control for possible confounding, alterations in the pre- and post-intervention values for
peak V02, habitual physical activity, and percentage body fat were included in the I
analyses in an effort to identify the independent effects of the exercise training program.

Age- and sexual maturity-matched controls were used. Sexual maturity status was

27

assessed pre- and post-intervention and although not reported in this paper, a companion
paper (T olfrey et al., 1998a) examining the training effect on peak V02 indicated that
sexual maturity status did not change. However, a few comments regarding the sexual
maturity classification and status reported by the authors should be considered. First, the
authors described the inclusion of subjects by a ”sexual maturity status no more than two
according to the criteria of Tanner for breast, pubic hair, and genital development in girls
and boys, respectively”. According to the criteria, Tanner stage 1 is the prepubertal state;
so the authors misidentif y the subjects as prepubertal in this study. Second, even though
subjects did not advance in sexual maturity stage as indicated by the criteria of Tanner, it
is important to note that the processes of biological maturation vary in tinting and tempo
and ”prepubertal” subjects vary in skeletal maturation (Malina and Bouchard, 1991).
Nonetheless, the exercise training group demonstrated a significant increase in HDL-C
(9.3%, 1.08 to 1.18 mmol/L), decrease in LDL-C (10.2%, 2.94 to 2.64 mmol/L), decrease
in TC/HDL-C (11.6%, 4.13 to 3.65), and decrease in LDL-C/HDL-C (17.2%, 2.85 to
2.36 mmol/L). Although not significant, TC also declined (4.2%, 4.33 to 4.15 mmol/L).
All significant differences involved an interaction for group and time except for HDL-C
that showed only a significant main effect.

Summary . In general, TC is not associated with physical activity in children and
adolescents. Some studies suggest that HDL-C and TG are higher and lower,
respectively, in more active than inactive youth, and LDL-C may also be lower in more
active youth. Armstrong and Simons- (1994) conclude that the evidence for these
conclusions are not as compelling as that for adults. It has also been noted that the results

are confounded in part by methodological difficulties in estimating habitual physical

28

activity in children and adolescents (i.e., misclassification), body fatness, variation in
maturity status and progress, and the interaction among these variables and blood lipids.
Why do some cross-sectional studies show no association between physical
activity and blood lipids, and some exercise training studies fail to show an improvement
in the blood lipid profile? First, the subjects in these studies probably display a normal
metabolic profile (Despres et al., 1990a). In the case where blood lipids did change
following exercise training, it is important to consider baseline values of the blood lipids
as increased physical activity often improves a hyperlipidemic profile (Tolfrey et al.,
1998b) . Second, confounding variables such as body composition, chronological age,
biological maturity status, diet, and/or aerobic fitness are not considered. Third, the
appropriate exercise training protocol including training frequency, intensity, and
duration for altering blood lipids in adolescents has not yet been established. Armstrong
and Simons-Morton (1994) suggest that blood lipids may not vary greatly in children and
adolescents since there is relatively less variance in both physical activity and blood lipid

levels in children and adolescents compared to adults.

DOSE-RESPONSE ISSUES

A major question in the physical activity epidemiology literature is the optimal
amount of physical activity required to alter health outcomes, risk factors, morbidity, and
mortality (Blair and Connelly, 1996; Lee and Paffenbarger, 1996; Morris, 1996). Haskell
(1994) points out that minimal and adequate amounts are also important to determine. To
establish recommendations for an appropriate exercise prescription, or a general

statement regarding physical activity and health, requires empirical data to support some

29

minimal, adequate, or optimal level of physical activity that has a favorable effect on a
selected biological outcome. Currently, there is no clear answer to how long (duration)
or how intense physical activity needs to be to produce favorable results. Despite some
uncertainty about appropriate levels of physical activity, the Centers for Disease Control
(CDC) and American College of Sports Medicine (ACSM) have recommended that
individuals accumulate 30 minutes or more of moderate physical activity most days of
the week (Pate et al., 1995). It is quite probable that the minimal, adequate, and optimal
amount of physical activity varies by individual, especially in the context of emerging
knowledge on genotype-environmental interactions.

Recommendations for appropriate levels of physical activity in youth are
currently based upon the adult model and expert opinion. In their review, Armstrong and
Simons-Morton (1994) concluded that the empirical dose-response data relating the effect
of physical activity to blood lipids in adolescents are nonexistent. However, more recent
studies provide some evidence to support a dose-response relationship. Among quintiles
of physical activity groups, the most active group (highest quintile) showed a lower TC in
Singapore youth (Schmidt et al., 1998). In the males participating in the Young Finns
Study, TC (p <0.15), IDL (p<0. l9), and TG (p<0.0003) tended to be lower and HDL
(p<0.04) and HDL:TC (p<0.02) tended to be higher in the active group (Raitakari et al.,
1997). Further study is warranted to establish the dose-response relationship in the

general population of youth.

30

TRAINING VOLUME AND BLOOD LIPIDS IN DISTANCE RUNNERS
Related to the issue of dose-response is the idea that health benefits accrue at
levels greater than the current recommendations. It has been shown that blood lipids are
superior in endurance trained adults, but whether the benefits are related to training
volume (i.e., distance run per week) has received limited attention. Observation of a
special exposure group allows for a unique opportunity to examine a comprehensive
model of the relationship between an exposure and outcome.

In an early report of 90 middle-aged male runners, distance run per week was
positively correlated with HDL (r=0.50) and remained significant after adjustment for
percentage body fat (r=0.40) (Rotkis et al., 1982). When runners were grouped by
mileage (low mileage, 10 to 19 miles per week; intermediate mileage, 20 to 39 miles per
week; and high mileage, 40 or more miles per week), HDL increased across groups (47
mgldl, 53 mgldl, and 60 mgldl, respectively), whereas TC (217 mgldl, 211 mgldl, 203
mgldl, respectively), non-HDL cholesterol (170 mgldl, 158 mgldl, 143 mgldl,
respectively), and TC:HDL (4.52, 3.99, and 3.31, respectively) decreased. However,

values across training groups were not adjusted for differences in age or body fatness. In
contrast, there was no relationship between TV and HDL (r=0.05) in a smaller sample of
33 middle-aged men (Williams, 1990). In both reports, training volume was estimated by
averaging self-reported weekly training mileage of the preceding six months.
Recently, Williams (1996, 1997) demonstrated that the benefits of exercise accrue in
a dose-response manner at levels of physical activity exceeding the current minimal
guidelines. Subjects included 8,283 male and 1,837 female adult recreational distance

runners participating in the National Runners' Health Study. Data were compiled from a

31

questionaire distributed at races and to subscribers of Runner' World and clinical records.
Training volume was calculated as the average distance run per week based on the
preceding 5 years. Significant linear trends were reported for HDL and TC:HDL in both
sexes and TG in men. No significant trend was evident for LDL.

Based on the aforementioned studies, it has been suggested that an exercise level
equivalent to jogging 10-15 miles/wk is necessary to significantly alter blood lipids
(Superko, 1991; \Vrlliams, 1994). Armstrong and Simons-Morton (1994) extrapolated
this recommendation to establish physical activity guidelines for youth, The authors
suggested that an adolescent would need to jog at a speed of 8 km/hr for approximately 2
hours per week, which from their own experiences was equivalent to about 80% of
maximal heart rate with young adolescents and about 75% of maximal heart rate for
young adults. They, therefore, recommended that four 30 minute exercise sessions per
week at 75-80% of maximum heart rate may be an appropriate prescription. Recall that
Tolfrey et al. (1998b) showed changes in the blood lipid profile of prepubertal subjects
following an 8 week training program consisting of three (rather than four) 30 minute
exercise sessions per week at 80% of maximal heart rate. However, it still remains to be
shown whether a dose-response or threshold effect of physical activity on blood lipids
exists in children and adolescents at or greater than the recommended exercise volume.
The study of adolescent distance runners provides a unique opportunity to evaluate the

current recommendations.

32

AEROBIC FITNESS AND BLOOD LIPIDS

A low level of aerobic fitness is an independent predictor of an increased risk for
cardiovascular disease in adults (Blair et al., 1989). In youth, more aerobically fit youth
generally have higher HDL and lower TG, although not all studies show such results
(Armstrong and Simons-, 1994; Despres et al., 1990a; Tolfrey et al., 2000). However,
this may be related to body composition as highly fit individuals are generally leaner than
unfit individuals. Chronological age may also influence the relationship, as HDL
declines during adolescence in males. Therefore, it is important to control for body
composition, age, and biological maturity when examining the relationship between
aerobic fitness and blood lipids. In studies that have controlled for some index of body
size and/or composition, the relationship between peak Vo2 and blood lipids no longer
remains significant (Al-Hazzaa et al., 1994; Armstrong et al., 1991; Hager et al., 1995;
Sallis et al., 1988). For example, Sallis et a1 (1988) reported significant correlations
between predicted peak Vo2 and HDL of 0.18 and 0.29 in 5th and 6th grade males and
females, respectively. Partial correlations, controlling for the BMI, reduced the
coefficients to 0.01 and 0.04 in males and females, respectively.

'Few studies have examined the relationship between peak vo2 and blood lipids in
athletic populations. In a previous mixed-sample (males and females) of 8 to 15 yr old
mid-Michigan distance runners, peak Vo2 was related to HDL (r=0.39) (Smith et al.,
1986). Atomi et al. (1986) and Macek et al. (1989) both reported a significant
relationship between peak Vo2 and TG and HDL in mixed-samples that included youth
athletes. Valimaki et al. (1980) also reported a significant relationships between total

work per unit body weight and HDL in 20 boys (r=0.53) and TG in 11 girls 02 -0.83).

33

Unfortunately, these studies did not conduct separate analyses for athletes. In a study of
national level adult athletes, peak V0, explained 25% of the variance in HDL (Berg and
Keul, 1985), and was significantly related to HDL (r=0.26) in Olympic athletes

(T sopanakis et al., 1986).

Genetic factors may also play a role in the modulating the relationship between
aerobic fitness and blood lipids. In a study of healthy, untrained adult men and women,
associations between peak V02 and blood lipids varied according to apolipoprotein (apo)
E phenotype (St.-Amand et al., 1999). There were significant relationships in men and
women between peak V0, and TG in carriers of the apo E2 isoforrn and apoFB
homozygotes. Peak V0, was significantly related to HDL in male Apo E3 homozygotes,
and to all blood lipids (TC, HDL, LDL, TG) in female Apo E3 homozygotes. There were
no significant relationships among male apo E4 carriers, and only _HDL was related to
peak V0, in female apo E4 carriers. When fat mass and glucose tolerance were
controlled, only HDI, remained significantly correlated with peak V0, in men and
women. It appears that a high level of aerobic fitness is associated with a favorable blood
lipid profile in individuals homogenoues for the apo F3 isoforrn and with reduced TG in
individuals who are apo E2 carriers. However, these associations were largely attributed

to the covariance of body fatness and glucose tolerance.

BODY FATNESS AND BLOOD LIPIDS
Various measures of body fatness and fat distribution (e.g., BMI, waist
circumference, waist-to-hip ratio, skinfold thickness, estimated percent body fatness) are

positively associated with atherogenic blood lipids and negatively associated with HDL

34

across the lifespan (Guo et al., 1994). Children and adolescents with excess adiposity
generally have a poorer blood lipid profile compared to leaner youth (Fripp et al., 1985;
Johnston, 1985; Smoak et al., 1987; Williams et al., 1992).

Few studies have examined the relationship between adiposity and blood lipids in
athletes. Athletes are generally characterized by relative leanness. However, some
variability does exist in measures of adiposity within this group. In a study of national
level athletes from a broad spectrum of sports (i.e., sprinters, cyclists, hammer throwers,
etc.), relative body weight (kg/(cm-100)) explained about 7% of the variance in TC and
TG and 20% of the variance in LDL (Berg and Keul, 1985). Unfortunately, discipline-
specific relationships were not explored. In Olympic athletes, relative body weight was
significantly related to HDL (r: -0.22), LDL (r: 0.18), and VLDL (m 0.17) (T sopanakis
et al., 1986). In middle-age distance runners, percentage body fat and HDL (m -0.36),
TC (r=0.38) and non-HDL cholesterol (r=0.48) were significantly related (Rotkis et al.,
1982). In contrast, correlations between various measures of body size (e.g., BMI,
relative weight, percentage body fat) and HDL were low in another sample of middle-
aged distance runners (r =0.05- 0.08) (Williams, 1990). The reason for the discrepancy in
the relationship of percentage body fat and HDL between the two studies of middle-aged
distance runners is not known.

Only two studies have examined the relationship between adiposity and blood
lipids in young athletes. Atomi et al. (1986) reported a significant relationship between
percentage body fat and TC and HDL in a mixed-sample of 10-12 yr old Japanese boys
and girls that included soccer players. Unfortunately, the correlation coefficients were

not reported. In a study of young (mean age = 12 yrs) female athletes (22 gymnasts, 20

35

swimmers), relationships between the sum of four skinfolds and estimated fat mass and
blood lipids were low to moderate (r <0.45) (Valimaki et al., 1980). Surprisingly, the

. relationship between adiposity and HDL was positive (i.e., increased fatness associated
with higher HDL).

Besides total body fatness, relative fat distribution, and specifically a truncal
and/0r visceral fat patterning, has been strongly linked to an adverse blood lipid profile in
youth and adults (Baumgartner et al., 1989; Despres eta1., 1990b; Freedman et al., 1989).
During adolescence, an increase in subcutaneous abdominal adipose tissue and
redistribution of fatness to the trunk results in an increase in the trunk-to-extremity
skinfold ratio in boys (Malina et al., 1999). The redistribution of adipose tissue during
male adolescence is associated with a decrease in HDL. (Baumgartner et al., 1989; van
Lenthe et al., 1998). This relationship may also be augmented by hormonal changes
during male puberty (Laskarzewski et al., 1983; Roemmich and Rogol, 1999; Srinivasan
et al., 1985).

Only one study has apparently examined the contribution of relative fat

distribution to blood lipids in athletes. Williams (1990) reported a low correlation (r=-
0.13) between the ratio of abdominal girth and bi-iliac diameter and HDL in middle-aged
male distance runners. No study has apparently examined the relationship between fat
distribution and blood lipids in young athletes.

The role of genes and body fatness in the modulation of elevated blood lipid
levels has also been indicated (Katzrnarzyk et al., 1999). Maximal heritability estimates
of abdominal visceral fatness measured by computerized tomography approximate 50-

55% with a major gene associated with total fat mass either directly or indirectly

36

affecting abdominal fatness. Polymorhpisms in lipoprotein genes (ApoA-II MspI,
HindlII, and apoB-lOO EcoRl) may also influence the relationship between abdominal

visceral fatness and blood lipid levels.

PHYSICAL ACTIVITY AND BLOOD LIPIDS: MECHANISMS

It is clear from the literature in adults that regular aerobic physical activity
attenuates the natural progression of atherosclerosis by favorably altering the blood lipid
profile (Haskell, 1984). What mechanisms are responsible for the alteration of
lipoproteins by increased physical activity? Several studies have found that increased
clearance of TG, increased lipoprotein lipase (LPL) activity, decreased hepatic lipase
activity, and increased lecithin:cholesterol acyltransferase (LCAT) activity may be
considered as possible mechanisms for the observed changes in blood lipids in humans
(Haskell, 1984; Stefanik and Wood, 1994). S

The relationships between physical activity, peak V0,, body fatness and blood
‘ lipids have been discussed previously. What remains to be distinguished is whether the
blood lipid profile is influenced primarily by increased physical activity, increased
aerobic fitness, changes in body composition, or a combination of these variables
(Krauss, 1989; Thompson, 1990; Williams, 1993). Some authors suggest that weight loss
is critical to an exercise effect on blood lipids, particularly HDL, while others have
shown that changes in blood lipids may occur without weight loss (Thompson, 1990).
However, weight gain is expected during childhood and adolesence, thus the applicability
of the adult model has limitations. On the other hand, peak V0, is thought to be related to

HDL (Tikkanen et al., 1991) through its association with the percentage of slow-twitch

37

muscle fibers and oxidative capacity of skeletal muscle (Bergh et al., 1978). The
possibility that leaner and more aerobically fit individuals are more likely to engage in

higher training levels also cannot be dismissed (Williams et al., 1982).

SUMMARY

Several areas related to physical activity, aerobic fitness, body fatness, and blood
lipids in children and adolescents have been reviewed. The evidence of early
atherosclerosis in youth is suggestive of the need to begin preventive measures during
childhood and adolescence. However, definitive data showing the association of physical
activity and aerobic fitness on the blood lipid profile of children and adolescents is
lacking, and thus the role of physical activity and aerobic fitness during childhoodpn the
etiology of atherosclerosis remains to be established. The association of body fatness and
blood lipids during childhood and adolescence is clear, particularly among obese children
and adolescents. Changes in truncal fatness and sex hormones during adolescence also
contribute to the decline in HDL during male adolescence.

Several methodological issues, particularly the quantification of physical activity,
need to be addressed in cross-sectional and prospective studies examining the association
between physical activity and blood lipids (and other health-related outcomes) in youth.
Perhaps the simple observation that a relatively small variance in both physical activity
and blood lipids in children and adolescents compared to adults may explain the lack of,
or low associations reported in cross-sectional studies. Likewise, the lack of a training
effect in experimental studies has been explained by the eulipidemic profile, failure to

account for confounding variables (pre- and post-test), and/or an inadequate exercise

38

training volume. The recent study of Tolfrey et al. (1998b) was a well-designed
experimental trial that showed changes in the blood lipid profile of prepubertal subjects
and may serve as a model for future exercise training studies in youth. Genetic factors
such as the apo E4 phenotype may also explain the variability in blood lipids and their
response to physical activity and should be considered in future studies.

Clearly, the blood lipid profile of young athletes has not been fully described.
The available cross-sectional data suggest that young athletes possess a superior blood
lipid profile compared to the general population. However, hypercholesterolemia does
exist in young athletes (Kyle et al., 1991) and requires further exploration. Likewise, the
heterogeneity of blood lipids in young athletes remains to be investigated. The study of
young athletes, as a special exposure group, would also allow the Opportunity to examine
the influence of physical activity levels equal to and greater than the current
recommendations (4 days/wk, 30 min/session, 75—80% maxHR) on the blood lipid
profile. It remains to be established whether a dose-response relationship, or a threshold
effect, exists between physical activity and blood lipids in the pediatric population.
Longitudinal study of physically active pubertal boys would allow study of the potential
attenuation of the decline in HDL during puberty by regular endurance exercise.

The determinants of blood lipids during adolescence are multi-factorial and
involve a complex interaction of genetic and environmental factors. It is the aim of Part I
of this dissertation to exarrrine the contribution of age, sexual maturity, physical activity,

aerobic fitness, and body fatness t0 the blood lipid profile of young distance runners.

39

REFERENCES

Al-Hazzaa HM, Sulaiman MA, Al-Matar AJ, Al-Mobaireek KF (1994) Cardiorespiratory
fitness, physical activity patterns and coronary risk factors in preadolescent boys. Int J
Sports Med 15:267-272.

Amos CI, Cohen JC, Srinivasan SR, Freedman DS, Elston RC, Berenson GS (1989)
Polymorphism in the 5' -flanking region of the insulin gene and its potential relation to
cardiovascular disease risk: observations in a biracial community. The Bogolusa Heart
Study. Atherosclerosis 79:51-57.

Anderson RA, Burns TL, Lee J, Swenson D, Bristow JL (1989) Restriction fragment

length polymorphisms associated with abnormal lipid levels in an adolescent population.
Atherosclerosis 77:227-237.

Armstrong N, Balding J, Gentle P, Kirby B (1992) Serum lipids and blood pressure in
relation to age and sexual maturity. Ann Hum Biol 19:477-487.

Armstrong N, Simons- B (1994) Physical activity and blood lipids in adolescents. Pediatr
Exerc Sci 6:381-405.

Armstrong N, Williams J, Balding J, Gentle P, Kirby B (1991) Cardiorespiratory fitness,
physical activity patterns, and selected coronary artery risk factor variables in 11- to 16-
year-olds. Pediatr Exerc Sci 3:219-228.

Atomi Y, Kuroda Y, Asarrri T, Kawahara T ( 1986) HDL-cholesterol of children (10 to 12
years of age) related to Vo2max, body fat, and sex. In J Rutenfranz, R Mocellin and F
Klimt (eds.): Children and Exercise XII. Champaign, IL: Human Kinetics, pp. 167- 172.

Bao W, Srinivasan S, Wattigney W, Berenson G (1994) Persistence of multiple
cardiovascular risk clustering related to syndrome X from childhood to young adulthood.
Arch Intern Med 154: 1842-1847.

Bao W, Srinivasan SR, Valdez R, Greenlund KJ, Wattigney WA, Berenson GS (1997)
Longitudinal changes in cardiovascular risk from childhood to young adulthood in
offspring of parents with coronary artery disease: the Bogolusa Heart Study. JAMA
278: 1749-1754.

Bao W, Srinivasan SR, Wattigney WA, Berenson GS (1995a) The relation of parental
cardiovascular disease to risk factors in children and young adults: the Bogolusa Heart
Study. Circulation 91 :365-371 .

Bao W, Threefoot SA, Srinivasan SR, Berenson GS (1995b) Essential hypertension

predicted by tracking of elevated blood pressure from childhood to adulthood: the
Bogolusa Heart Study. Am J Hypertens 8:657-665.

40

Baumgartner RN, Siervogel RM, Chumlea WC, Roche AF (1989) Associations between
plasma lipoprotein cholesterols, adiposity and adipose tissue distribution during
adolescence. Int J Obes 13:31-41.

Berenson GS, Foster TA, Frank GC, Frerichs RR, Srinivasan SR, Voors AW, Webber LS
(1978) Cardiovascular disease risk factor variables at the preschool age: The Bogolusa
Heart Study. Circulation 57:603-612.

Berenson GS, Srinivasan SR, Bao W, Newman WP, Tracy RE, Wattigney WA (1998)
Association between multiple cardiovascular risk factors and atherosclerosis in children
and young adults. N Engl J Med 338: 1650-1656.

Berenson GS, Srinivasan SR, Cresanta JL, Foster TA, Webber LS (1981) Dynamic
changes of serum lipoproteins in children during adolescence and sexual maturation. Am
J Epidemiol 113:157-170.

Berg A, Keul J (1985) Influence of maximum aerobic capacity and relative body weight
on the lipoprotein profile in athletes. Atherosclerosis 55:225-231.

Bergh U, Thorstensson A, Sjodin B, Hulten B, Piehl K, Karlsson J (1978) Maximal
oxygen uptake and muscle fiber types in trained and untrained humans. Med Sci Sports
10: 151-154.

Betteridge DJ, Illingworth DR, Shepherd J, eds. (1999) Lipoproteins in Health and
Disease. London: Arnold.

Blair SN, Connelly JC (1996) How much physical activity should we do?: the case for
moderate amounts and intensities of physical activity. Res Q Exerc Sport 67: 193-205.

Blair SN, Kohl HW, Paffenbarger RS, Clark DG, Cooper KH, Gibbons LW (1989)
Physical fitness and all-cause mortality: a prospective study of healthy men and women.
‘JAMA 17:2395—2401.

Boreham CA, Twisk J, Savage MJ, Cran GW, Strain JJ (1997) Physical activity, sports
participation, and risk factors in adolescents. Med Sci Sport Exerc 29:788-793.

Brill PA, Burkhalter HE, Kohl HW, Goodyear NN, Blair SN (1989) The impact of
previous athleticism on exercise habits, physical fitness, and coronary heart disease risk
factors in nriddle-aged men. Res Q Exerc Sport 60:209-215.

Bucher KD, Schrott HG, Clarke WR, Lauer RM (1982) The Muscatine Cholesterol
Study: distribution of cholesterol levels within families of probands with high, low, and
middle cholesterol levels. J Chron Dis 35:385-400.

Casperson CJ, Nixon PA, DuRant RH (1998) Physical activity epidemiology applied to
children and adolescents. Exerc Sci Sports Rev 26:341-403.

41

Clarke WR, Schrott HG, Leaverton PE, Connor WE, Lauer RM (1978) Tracking of blood
lipids and blood pressure in school age children: the Muscatine Study. Circulation
58:626-634.

Cresanta JL, Srinivasan SR, Webber LS, Bereneson GS (1984) Serum lipid and
lipoprotein cholesterol grids for cardiovascular risk screening of children. Am J Dis Child
138:379-387.

Despres J-P, Bouchard C, Malina RM (1990a) Physical activity and coronary heart
disease risk factors during childhood and adolescents. Exerc Sport Sci Rev 18:243-261.

Despres J-P, Moorjani S, Lupien P, Trarnblay A, nadeau A, Bouchard C (1990b)
Regional distribution of body fat, plasma lipoproteins, and cardiovascular disease.
Atherosclerosis 10:497-511.

DuRant R, Baranowski T, Rhodes T, Gutin B, Thompson W, Carroll R, Puhl J, Greaves
K (1993) Association among serum lipid and lipoprotein concentrations and physical
activity, physical fitness, and body composition in young children. J Pediatr 123: 185-192.

Enos WF, Beyer JC, Holmes RH (1955) Pathogenesis of coronary disease in American
soldiers killed in Korea. JAMA 158:912-914.

Freedman DS, Srinivasan SR, Harsha DW, Webber LS, Berenson GS (1989) Relation .of
body fat patterning to lipid and lipoprotein concentrations in children and adolescents: the
Bogolusa Heart Study. Am J Clin Nutr 50:930-939.

Frerichs RR, Srinivasan SR, Webber LS, Berenson GS (1976) Serum cholesterol and
triglyceride levels in 3,446 children from a biracial community: The Bogolusa Heart
' Study. Circulation 54:302-308.

Fripp R, Hodgson J, Kwiterovich P, Werner J, Schuler H, Whitman V (1985) Aerobic
capacity, obesity, and atherosclerotic risk factors in male adolescents. Pediatr 75:813-
818.

Garcia R, Moodie D (1989) Routine cholesterol surveillance in childhood. Pediatr
84:751-755.

Guo S, Salisbury S, Roche AF, Chumlea WC, Siervogel RM (1994) Cardiovascular
disease risk factors and body composition: a review. Nutr Rev 14: 1721-1777.

Gutgessel HP, Atkins DL, Day RW (1997) Common cardiovascular problems in the
young: Part II. Hypertension, hypercholesterolemia and preparticipation screening of
athletes. Am Fam Physician 56:1993-1998.

Hager RI.., Tucker LA, Seljaas GT (1995) Aerobic fitness, blood lipids and body fat in
children. Am J Public Health 85: 1702-1706.

42

Haskell WL (1984) The influence of exercise on the concentrations of triglyceride and
cholesterol in human plasma. Exerc Sport Sci Rev 12:205-244.

Haskell WL (1994) Health consequences of physical activity: understanding and
challenges regarding dose-response. Med. Sci. Sports Exerc. 26:649-660.

Hickman TB, Briefel RR, Carroll MD, Rifltind BM, Cleeman JI, Maurer KR, Johnson
CL (1998) Distributions and trends of serum lipid levels among United States children
and adolescents ages 4—19 years: data from the Third National Health and Nutrition
Examination Survey. Prev Med 27:879-890.

Hopkins PN, Williams R (1981) A survey of 246 suggested coronary risk factors.
Atherosclerosis 40: 1-52.

Johnston F (1985) Health implications of childhood obesity. Ann Intern Med 103:1068-
1072.

Katzrnarzyk PT, Malina RM (1997) The contribution of participation in organized youth
sports to daily energy expenditure in children and youth. East Lansing, MI: Institute for
the Study of Youth Sports.

Katzmarzyk PT, Perusse L, Bouchard C (1999) Genetics of abdominal visceral fat levels.
Am J Hum Biol 11:225-235. -

Krauss RM (1989) Exercise, lipoproteins, and coronary artery disease. Circulation
79:1143-1145. '

Kyle JM, Walker RB, Riales RR, Petty G], Thomas JA, Roberts MD (1991) Student
athletes cholesterol screening during routine precompetition examination. J Fam Pract
33: 172-176.

Laskarzewski P, Morrison J, Gutai J, Orchard T, Khoury P, Glueck C (1983) High and
low density lipoprotein cholesterols in adolescent boys: relationships with endogenous
testosterone, estradiol, and quetelet index. Metabolism 32:262-271.

Lauer R, Connor W, Leaverton P, Reiter M, Clarke W (1975) Coronary heart disease risk
factors in school children: The Muscatine Study. J Pediatr 86:697-706.

Lee I-M, Paffenbarger RS (1996) How much physical activity is optimal for health?:
methodological considerations. Res Q Exerc Sport 67:206-208.

Macek M, Bell D, Rutenfranz J, Vavra J, Masopust J, Neidhart B, Schmidt K-H (1989) A

comparison of coronary risk factors in groups of trained and untrained adolescents. Eur J
Appl Physiol 58:577-582.

43

Mahoney LT, Burns TL, Stanford W, Thompson BH, Witt JD, Rost CA, Lauer RM
( 1996) Coronary risk factors measured in childhood and young adult life are associated

with coronary artery calcification in young adults: the Muscatine Study. J Am Coll
Cardiol 27:277-284.

Malina RM (1997) Activity and fitness of youth: are they related? do they track? In K
Froberg, O Lammert, H St. Hanson and CJR Blirnkie (eds.): Exercise and Fitness-
Benefits and Risks: Children & Exercise XVIII: Odense University Press, pp. 161-171.

Malina RM, Bouchard C (1991) Growth, Maturation, and Physical Activity.Charnpaign,
IL: Human Kinetics.

Malina RM, Koziel S, Bielicki T ( 1999) Variation in subcutaneous adipose tissue
distribution associated with age, sex, and maturation. Am J Hum Biol 11:189-200.

McNamara JJ, Molot MA, Stremple JF, Cutting RT (1971) Coronary artery disease in
combat casualties in Vietnam. JAMA 216: 1 185-1 187.

Morris JN (1996) Exercise versus heart attack: questioning the consensus? Res Q Exerc
Sport 67:216-220.

Morrison JA, Iaskarzewski PM, Rauh JL, Brookman R, Mellies M, Frazer M, Khoury P,
deGroot I, Kelly K, Glueck CJ (1979) Lipids, lipoproteins, and sexual maturation during
adolescence: the Princeton Maturation Study. Metabolism 28:641-649.

Muhonen LE, Burns TL, Nelson RP, Lauer RM (1994) Coronary risk factors in
adolescents related to their knowledge of familial coronary heart disease and
hypercholesterolemia: the Muscatine Study. Pediatr 93:444-451.

National Cholesterol Education Program (1992) Report of the expert panel on blood
cholesterol levels in children and adolescents. Pediatr 89 (Suppl 3): Part 2:525-584.

NIH Consensus Development Panel on Triglyceride HDL, and Coronary Heart Disease,
(1993) Triglyceride, high density lipoprotein, and coronary heart disease. JAMA
269:505-510.

NIH Consensus Statement (1985) Lowering blood cholesterol to prevent heart disease.
JAMA 253:2080-2086.

Paffenbarger RS, Lee I-M (1996) Physical activity and fitness for health and longevity.
Res Q Exerc Sport 67 (Suppl. 3):11-28.

Pate RR, Pratt M, Blair SN, Haskell WL, Macera CA, Bouchard C, Buchner D, Ettinger

W, Heath GW, King AC, Kriska A, Leon AS, Marcus BH, Morris J, Paffenbarger RS,
Patrick K, Pollock ML, Rippe JM, Sallis J, Wilmore JH (1995) Physical activity and

44

WWW?

health: A recommendation from the Centers of Disease Control and Prevention and the
American College of Sports Medicine. JAMA 273:402-407.

Porkka KVK, Viikari JSA, Taimela S, Dahl M, Akerblom HK (1994) Tracking and
predictiveness of serum lipid and lipoprotein measurements in childhood: a l-yr follow-
up. Am J Epidemiol 140: 196-1 110.

Raitakari O, Taimela S, Porkka K, Telama R, Valimaki I, Akerblom H, Viikari J (1997)
Associations between physical activity and risk factors for coronary artery disease: The
Cardiovascular Risk in Young Finns Study. Med Sci Sports Exerc 29: 1055-1061.

Roemmich J, Rogol A (1999) Hormonal changes during puberty and their relationship to
fat distribution. Am J Hum Biol 11:209-224.

Rothman KJ, Greenland S (1998) Modern Epidemiology. 2nd, Philadelphia: Lippincott-
Raven.

Rotkis TC, Cote R, Coyle E, Wilmore JH (1982) Relationship between high density
lipoprotein cholesterol and weekly running mileage. J Cardiac Rehab 2: 109-1 12.

Rowland TW (1993) The physiological impact of intensive training on the prepubertal
athlete. In BR Cahill and AJ Pearl (eds.): Intensive Participation in Children's Sports.
Champaign, IL: Human Kinetics, pp. 167-193.

Sallis JF, Patterson TL, Buono MJ, Nader PR (1988) Relation of cardiovascular fitness
and physical activity to cardiovascular disease risk factors in children and adults. Am J
Epidemiol 127:933-941.

Schmidt G, Walkuski J, Stensel D (1998) The Singapore Youth Coronary Risk and
Physical Activity Study. Med Sci Sports Exerc 30:105-113.

Shear CL, Burke GL, Freedman DS, Berenson GS (1986) Value of childhood blood
pressure measurements and family history in predicting future blood pressure status:
results from 8 years of follow-up in the Bogolusa Heart Study. Pediatr 77:862-869.

Siervogel RM, Baumgartner RN, Roche AF, Chumlea WC, Glueck CJ (1989) Maturity
and its relationship to plasma lipid and lipoprotein levels in adolescents: the Fels
Longitudinal Study. Am J Hum Biol 1:217-226.

Smith BW, Metheny WP, Sparrow AW (1986) Serum lipid and lipoprotein profiles in
age-group runners. In MR Weiss and D Gould (eds.): Sport for Children and Youths.
Champaign, 11.: Human Kinetics, pp. 269-273.

Smoak CG, Burke GL, Webber LS, Harsha DW, Srinivasan S, Berenson GS (1987)
Relation of obesity to clustering of cardiovascular disease risk factors in chidren and
young adults. Am J Epidemiol 125:364-372.

45

Srinivasan SR, Ehnholm C, Wattigney WA, Bao W, Berenson GS (1996) The relation of
apolipoprotein E polymorphism to multiple cardiovascular risk in children: the Bogolusa
Heart Study. Atherosclerosis 123:33-42.

Srinivasan SR, Frerichs RR, Webber LS, Berenson GS (1976) Serum lipoprotein profile
' in children from a biracial community: The Bogolusa Heart Study. Circulation 54:309-
3 18.

Srinivasan SR, Sundaram GS, Williamson GD, Webber LS, Berenson GS (1985) Serum
lipoproteins and endogenous sex hormones in early life: observations in children with
different lipoprotein profiles. Metabolism 34:861-867.

St.-Amand J, Prud'homme D, Moorjani S, Nadeau A, Tremblay A, Bouchard C, Lupien
PJ, Despres J -P (1999) Apolipoprotein E polymorphism and the relationships of physical
fitness to plasma lipoprotein-lipid levels in men and women. Med Sci Sports Exerc

3 1:692-697.

Stefanik ML, Wood PD (1994) Physical activity, lipid and lipoprotein metabolism, and
lipid transport. In C Bouchard, S RJ. and T Stephens (eds.): Physical Activity, Fitness,
and Health. Champaign, IL: Human Kinetics, pp. 417-431.

Strong JP, Malcom GT, McMahon CA, Tracy RE, Newman 111 WP, Herderick EE,
Corhill JF (1998) Prevalence and extent of atherosclerosis in adolescents and young
adults. JAMA 281:727-735.

Strong JP, Malcom GT, Oalmann MC, Wissler RW (1997) The PDAY Study: natural
history, risk factors, and pathobiology. Ann NY Acad Sci 811:226—235.

Superko HR (1991) Exercise training, serum lipids, and lipoprotein particle: is there a
change threshold? Med Sci Sports Exerc 23:677-685.

' Tarnir I, Heiss G, Glueck CJ, Christensen B, Kwiterovich P, Rifltind BM (1981) Lipid
and lipoprotein distributions in White children ages 6-19 yr: the Lipid Research Clinics
Program Prevalence Study. J Chron Dis 34:27-39.

Tanner JM (1962) Growth at Adolescence. 2nd Edition, Oxford: Blackwell.

Tell GS (1985) Cardiovascular disease risk factors related to sexual maturation: the Oslo
Youth Study. J Chron Dis 38:633-642.

Tell GS, Mittelmark MB, Vellar OD (1985) Cholesterol, high density lipoprotein
cholesterol and triglycerides during puberty: the Oslo Youth Study. Am J Epidenriol
122:750-761.

Tell GS, Vellar OD (1988) Physical fitness, physical activity, and cardiovascular disease
risk factors in adolescents: The Oslo Youth Study. Prev Med 17:12-24.

46

The Lipid Research Clinics (1980) Population Studies Data Book: Aggregrate
distributions of lipids, lipoproteins and selected variables in 11 North American
populations. Washington, DC: U.S. Department of Health and Human Services.

Thompson PD (1990) What do muscles have to do with lipoproteins? Circulation
81: 1428-1430.

Tikkanen HO, Harkcnen M, Naveri H, Hamalainen E, Hovainio R, Sama S, Heikki Frick
M (1991) Relationship of skeletal muscle fiber type to serum high density lipoprotein
cholesterol and apolipoprotein A-I levels. Atherosclerosis 90:49-57.

Tolfrey K, Campbell IG, Batterham AM (1998a) Aerobic trainability of prepubertal boys
and girls. Pediatr Exerc Sci 10:248-263.

Tolfrey K, Campbell IG, Batterham AM (1998b) Exercise training induced alterations in
prepubertal children's lipid-lipoprotein profile. Med Sci Sports Exerc 30: 1684-1692.

Tolfrey K, Jones AM, Campbell [G (2000) The effects of aerobic exercise training on the
lipid-lipoprotein profile of children and adolescents. Sports Med 29:99-112.

Trudeau F, Espindola R, laurencelle L, Dulac F, Rajic M, Shephard RJ (2000) Follow-up
of participants in the Trois-Riviéres Growth and Development Study: exarning their
health-related fitness and risk factors as adults. Am J Hum Biol 12:207-213.

Tsopanakis C, Kotsarellis D, Tsopanakis AD (1986) Lipoprotein and lipid profiles of
elite athletes in Olympic sports. Int J Sports Med 7:316—321.

Valimaki I, Hursti M-L, Pihlakoski L, Viikari J (1980) Exercise performance and serum
‘ lipids in relation to physical activity in schoolchildren. Int J Sports Med 1:132-136.

van Lenthe FJ, van Mechelen W, Kemper HCG, Twisk JWR (1998) Association of a
central pattern of body fat with blood pressure and lipoproteins from adolescence into
adulthood. Am J Epiderrricl 147:686-693.

Williams DP, Going SB, Lohman TG, Harsha DW, Srinivasan S, Webber LS, Berenson
GS (1992) Body fatness and risk for elevated blood pressure, total cholesterol, and serum
lipoproteins ratios in children and adolescents. Am. J. Public Health 82:358-363.

Williams PT (1990) Weight set-point theory predicts HDL-cholesterol levels in
previously obese long-distance runners. Int J Obes 14:421-427.

Williams PT (1993) Hi gh-density lipoproteins and lipase activity in runners.
Atherosclerosis 98:251-254.

Williams PT (1994) Lipoproteins and adiposity show improvement at substantially higher
exercise levels than those currently recommended. Circulation 90:1-471 (Abstract).

47

Williams PT (1996) High-density lipoprotein cholesterol and other risk factors for
coronary heart disease in female runners. N Engl J Med 334: 1298-1303.

Williams PT (1997) Relationship of distance run per week to coronary heart disease risk
factors in 8283 male runners. Arch Intern Med 157: 191-198.

Williams PT, Wood PD, Haskell WL, Vranizan K (1982) The effects of waning rrrileage
and duration on plasma lipoprotein levels. JAMA 247:2674-2679.

Woolf N (1999) Pathology of atherosclerosis. In DJ Betteridge, DR Illingworth and J
Shepherd (eds.): Lipoproteins in Health and Disease. London: Arnold, pp. 533-540.

Table 2.1. Guidelines for interpreting blood lipid values in children and adolescents.

 

Categon Level (mg/d1)
TC
High _>_200
Borderline high 170-199
Desirable _<_170
LDL
High _>_130
Borderline high 110-129
Desirable $110
HDL 2-9 yrs of age 10-19 yrs of age
Low 540 _<_35
Borderline low 40-45 35-45
Desirable 245 _>_45
TG
High 2100 _>_l30
Borderline high 75-99 90-129
Desirable _<_75 _<_9O

 

TC, total cholesterol; LDL, low density lipoprotein cholesterol; HDL, high density
lipoprotein cholesterol; TG, triglycerides.

Adapted from Kwiterovich (1989).

49

dezo NN, mes—=03. 2 v.09. .5... «25.8 .= car. 2... 2.0.082: 95.089

 

 

mqu< Pena—=0: moo... Hangmi 2 max >ho QB. 40 ICE For 4.0
<w..3er. a. $3.02. an...” amp—.2. 330.258. 0 Z . .- .w .3 am we
9.. Como. mwmfiae...” 20.3.5 ..0~ 2 Eu. :9 2m.
.8...— . v... .0104 8 £23.
4 .“ .qw ow mm
DA. .0. .0.
2.85.85?» 10.8... wee: 8.60. 8: Au 3.: .8109. .84 um Z. m xu.A.m .8 mm 8 0m
.23 Ex. floor on cum—88a: 0.. Go 2.. Ana. .0. ANN. Bu.
cream—=8: ..m.= £3.88 3.53 .m .0.
208. 9.22.8 3.5.5. .05
.56. Em: .56. «8..
95.... a. a... 3.95»: 0.358 20. 30..an um Z. .u .o..m .8 a. 6. cm 0. am a.
:88 353.5 Cu 7.. rush .0...
$3
9...... a. w.. 3.2.5»: wig—5.3m w. .08. m 30¢... .m 7... .1. 0-5 .8 em G. 8 3. q. .0.
:80. . .0 Z. 3..
. c 3
>83. 2 a... 4.0.3.0 moooon 33.0.0935 .5 .8084 N. 7.. 5.5 .3 0m 6. go AA.
Come. ace ..0.. Vw <2“ u .5; G...
0...; 58.5.5... 90 v...
Zeno.” o. w... 0898.05?» m£.33.=m u 53>. 0.8 5.8.8.... No 7. .0. .m .o. mu .0. q. 8. MA
Come. 2... £5.05 003.85.9— 2... ANN.
$22.8 we 3
No .u .3 :8 A. 5 me
E: .6.
Win a. c... (<8. 55....» .. ... AMA 7. ..-.m .mm
200..
um... .u .mm

 

7.2:. EU. 3.0% 0902.8 :92. A» @8595. 2403.
Econ. :0...“ $58 Ed axe—.88.. .= 3m}...

Cholesterol (mgldl)

175 -

125-

LDL

 

 

 

 

ABC (ym)

Figure 2.1. Mean values by age and sex for cholesterol. Data
from Lipids Research Clinics Prevalence Study (1980).

51

mu

5’
l

Triglycerides (mgldl)
i

”-1

 

 

 

Ase (m)

Figure 2.2. Mean triglyceride values by age and sex. Data
from Lipids Research Clinics Prevalence Study (1980).

52

‘i‘ W? 1

CHAPTER3

MIXED-LONGITUDINAL ANALYSIS OF BLOOD LIPIDS

IN YOUNG DISTANCE RUNNERS

53

ABSTRACT

Limited information is available on age-and sex-associated variation in blood
lipids among young athletes. A mixed-longitudinal design was used to examine the
development of blood lipids of competitiveiyoung distance runners followed from 1982
to 1985. Total cholesterol (TC), hi gh-density lipoprotein (HDL), low-density lipoprotein
(LDL), and triglycerides (T G) were determined by standard procedures. Serial data
included 99 annual measurements for 27 males and 84 annual measurements for 27
females aged 9 to 18 yrs. In general, TC and LDL remained stable, and HDL declined
with age, especially in males. TG increased with age. Age-related trends were
statistically significant for HDL and TG in boys only (p<0.05). TC and LDL were
slightly greater in boys at all ages except 11, 15, and 17 yrs (p>0.05). HDL was similar
between the sexes until 13 yrs when values became greater in girls (3.2 to 13.8 mgldl)
(p<0.05 in 17+ yrs). No clear pattern of sex differences emerged for TG. Compared to
the general population, blood lipids of young distance runners showed the following
trends: 1) TC was above reference medians, 2) LDL tended to approximate or to be
slightly above reference medians, 3) TG fluctuated about the reference medians, and 4)
HDL was higher in distance runners compared to the reference medians prior to age 14
yrs, but in the older age groups, especially males, HDL either approximated or fell
slightly below the reference medians. There was considerable variability in blood lipid
levels among the runners. In 21 males and 18 females with serial data for 3 to 5 yrs,
HDL declined 22.4 and 18.3 mg/dl (p<0.05) whereas TG increased 18.0 and 14.0 mg/dl
(p<0.05 in females only) in males and females, respectively. Tracking coefficients over

intervals of 3 to 5 yrs were moderate to high (0.48—0.90), except for TC in males (0.08).

54

INTRODUCTION

Although clinical manifestations of coronary heart disease (CHD) are not evident
until adulthood, the origins of CHD me well established in childhood and adolescence
(Berenson et al., 1998; Enos et al., 1955; Mahoney et al., 1996; Strong et al., 1997).
Given evidence on the pediatric origins of atherosclerosis, it has been Suggested that
preventive strategies, including increased physical activity, begin during childhood and
adolescence. Epidemiological studies in adults indicate that regular physical activity and
relatively high aerobic fitness have beneficial effects on CHI) risk factors, morbidity, and
mortality (Blair et al., 1989; Paffenbarger and Lee, 1996). However, the roles of physical
activity and aerobic fitness during childhood and adolescence on the etiology of
atherosclerosis are complex and remain to be established.

Age- and sex-associated variation of blood lipids in children and adolescents are
well described in the general population (Hetzel and Berenson, 1987). In contrast, little
information is available on age— and sex-associated variation of blood lipids in young
athletes, a subgroup of the population that is generally exposed to high levels of physical
activity on a regular basis and possesses high levels of aerobic fitness. Studies of the
blood lipid profiles of young athletes are limited (Atomi et al., 1986; Kyle et al., 1991;
Macek et al., 1989; Nizankowska-Blaz and Abramowicz, 1983; Smith et al., 1985;
Valimaki etal., 1980; Zonderland etal., 1984). Many of these cross-sectional studies
have methodological shortcomings including small sample sizes, narrow age ranges,
mixed samples (i.e., sexes combined), inconsistent and/or lack of information on training
history, and failure to consider confounding variables such as maturity status, dietary

intake, adiposity, smoking, and family history of vascular disease.

55

Cross-sectional comparisons of endurance-trained individuals and control subjects
have been used to demonstrate the influence of physical activity or exercise training on
5 various biological variables. Comparisons of athletes and control subjects have a
selection bias that limits generalizations. However, to establish an understanding of the
influence of intensive training on the blood lipid profile requires the recruitment of
individuals regularly engaged in endurance training since the general population does not
participate in high levels of vigorous physical activity. Although there may be genetic
pre-disposition among athletes, part of the variation in a biological variable in elite
athletes may also be explained by environmental factors and genotype-environmental
interactions. Thus, the phenotypic expression of a biological variable in athletes also
expresses the influence of an environmental factor such as regular physical activity.

Unlike cross-sectional approaches, longitudinal study of young distance runners
provides an opportunity to examine the potential intervention of regular endurance
training on the modification of the blood lipid profile during the early years of life when
the atherosclerotic process begins. Although such a study has been proposed (Macek et
al., 1989), apparently it has not been conducted. As a result, the pattern of development
of blood lipids in young athletes is incomplete. This paper describes age- and sex-
associated variation of blood lipids in a mixed-longitudinal sample of male and female
distance runners, 9 to 18 yrs of age. Specific questions include the following: (1) How
does the blood lipid profile of young distance runners compare to the general pediatric
population?, and (2) Is the decline in high-density lipoprotein during male adolescence
attenuated by regular endurance exercise training? It was hypothesized that the blood

lipid profile of young distance runners will be superior to that in the general population as

56

observed in well-trained adult endurance athletes (Haskell, 1984), and the decline in
hi gh-density lipoprotein in males during adolescence would be attenuated in young

distance runners.

METHODS

Subjects. Runners between the ages of 8 and 15 yrs who consistently placed within the
top five finishers of road races 10 km or more by age and sex were identified and
contacted for the study. Race results were obtained from a statewide running publication,
Michigan Runner, between May and August 1981. Of the runners contacted (response
rate unknown), 27 male and 27 female distance runners participated in the study.
Subjects entered the study between 8.0 to 15.7 yrs of age. Of the total sample, 21 males
and 18 females were followed at approximately annual intervals for 3 to 5 yrs. The
remainder of the subjects (6 boys and 9 girls) participated in 1 or 2 annual visits. Overall,
99 and 84 annual observations were available for males and females, respectively. In a
sub-sample of subjects ( 16 boys, 19 girls), mean (1SD) reported weekly training volumes
were 1503:920 and 18651790 km per year in males and females. respectively. Parental
consent and child assent was obtained prior to the study. The study was approved by the

Michigan State University Committee on Research Involving Human Subjects.

Blood lipids. A venous blood sample was drawn from an antecubital vein after a 12 hour
fast. Lipid analysis was performed according to the procedures described by the Lipid
Research Clinics (Lipid Research Clinics Manual of Laboratory Operations, 1974). The

ratios TC:HDL and 1.13th were derived from the respective concentrations.

57

Statistical analysis. Subjects were divided into whole year age groups (i.e., 11.0 to
11.99), except for the youngest age group that consisted of subjects 9.0 to 10.99 yrs.
Each age and sex group included only one observation per subject, and subjects were
regarded as independent in each age group. Descriptivestatistics were calculated for
each sex within each age group. A one-sample t-test was conducted to examine the age-
and sex-specific means between the runners and United States reference values
(Christensen et al., 1980). This approach was taken instead of comparison to a control
group since control groups are often convenient samples and are not a random
representative sample. Linear regression was used to determine age-related trends for
each blood lipid. Age-specific sex differences were determined by a series of
independent t-tests. Paired t-tests were used to examine changes in blood lipids between
baseline and last visit in the 21 males and 18 females with serial data for 3 to 5 yrs .
Tracking coefficients were determined by Pearson correlations controlling for age at
baseline. An alpha level of 0.05 was used for significance and adjusted according to the

Bonferroni procedure for multiple comparisons.

RESULTS

Age- and sex-specific values for blood lipids are shown in Tables 3.1 and 3.2 and
Figures 3.1a-d and 3.2a-d. The variability in blood lipids among individuals should be
noted.
Males. In boys, mean TC remains between 170-180 mg/dl from 9 tol4 yrs, decreases to

164 mg/dl at 15 and 16 yrs, and increases to 180 mg/dl in the oldest age group. The

58

pattern for HDL shows an age-related decline in boys. Mean HDL is stable at 60 mgldl
in 9 to 12 year olds and progressively declines to 40.9 mgldl in the oldest age group.
Mean LDL remains fairly constant across age groups between 96.2 and 108.1 mgldl.
Mean TG fluctuates irregularly between 56.3 and 79.3 mgldl from 9 and 15 yrs and
increases in the oldest age groups. The mean TC:HDL ratio remains relatively stable
between 2.98 and 3.17 until 13 yrs and increases consistently thereafter reaching a high
value of 4.64 in the oldest age group. The same pattern emerges for the mean LDL.:HDL
ratio with values between 1.73 and 1.92 until 13 yrs and a peak value of 2.82 in the oldest
age group. Only HDL and TG show significant (p<0.05) age-related trends in males.
Females. In girls, mean values of TC varies between 165 and 180 mg/dl with the
exception of a value of 194 mg/dl at 11 yrs and a low value of 159 mgldl at 16 yrs. Mean
HDL is highest in the youngest age group and remains stable at about 60 mgldl from 11
to 15 yrs before declining in the oldest age groups. There is no clear age-related pattern
for LDL with values fluctuating between 92 and 119 mgldl. Mean TG levels are lower
between 9 and 13 yrs (51.5 to 65.8 mgldl) than between 14 and 18 yrs (69.5 to 82.8
mgldl). Both the mean TC:HDL and LDth ratios are lowest in the youngest and
highest in oldest age groups. After a modest increase, both the TC:HDL and LDLzHDL
ratios progressively decline from age 11 to 14 yrs. No significant age-related trends are
evident in females.

Sex differences. TC and LDL are greater in male runners at all ages except 11, 15 and 17
yrs. HDL is similar between sexes until age 13 yrs, after which values remain greater in

girls (3.2 to 13.8 mgldl) (p<0.05 in 17+ yrs). No clear pattern of sex differences emerges

59

for TG, although TG values are significantly greater in the oldest age group of boys
(p<0.05).

Comparison to reference values. Compared to reference medians for United States youth
(Figures 3.1a-d and 3.2a-d), TC is above reference medians in both sexes across the age
range studied. TC is 7.1-25.1 mgldl greater in males (p<0.05 at 10 and 14 yrs) and 3.0-
30.9mg/dl greater in females (p<0.05 at 11 and 14 yrs). In boys, HDL is slightly above
(1.5-7.5 mgldl; p <0.05,14 yrs) the reference medians until 15 yrs when values decline to
the median. In girls, HDL is above (2.7-15.1 mgldl; p <0.05, at 10 and 14 yrs) reference
medians at all ages except 16 yrs when values are equal to reference medians. LDL is
slightly greater (1.2- 13.5 mgldl; p<0.05, at 14 yrs) than reference medians in males and
fluctuates (-2 to 24.5 mgldl; p<0.05, at 11 yrs) about reference medians in females. TG
fluctuates (1.3 to 24.3 mgldl; p <0.05, at 13 yrs) about reference medians in males with
the exception of a large increase in the oldest age group. In females, TC is less than the
reference medians from 9 to 13 yrs but increases with age so that by 15 yrs values were
above the reference.

Longitudinal analyses. Means and standard deviations for blood lipids at baseline and
the last visit are shown in Table 3.3. The mean duration of follow-up is 3.29:0.57 and
3.21:0.78 yrs in males and females, respectively. TC and HDL decreases significantly
(p<0.05) in both sexes and TG increases in both sexes (p<0.05 in females only).
Correlations are significant between baseline and follow-up values for all blood lipids

(0.48-0.90), except TG in males (0.07).

DISCUSSION

This study provides information on the development of blood lipids in a mixed-
longitudinal sample of young distance runners. A major strength of this study is the
mixed-longitudinal design which permits analysis of serial observations during
adolescence in young athletes.

The results indicate that the development of blood lipids in young distance
runners is similar to youth in the general population - TC and LDL remain stable, I-IDL
declines during adolescence (especially in males), and TG increases with age. In contrast
to observations in adult endurance athletes, the young distance runners do not possess a
superior blood lipid profile except for HDL in the younger age groups. Compared to age-
and sex-specific reference values for United States youth (Christenson et al., 1980), TC
remains above the medians, LDL tends to approximate or to be slightly above the
medians, and TG fluctuates about the medians. HDL values are higher in male and
female runners prior to age 14 yrs. In the older age groups, especially in males, HDL
either approximates or falls below the reference medians. Sex differences in young

distance runners are similar to those observed in the general population except for the
development of LDL, which is higher in males (Christenson et al., 1980).

Few studies have examined blood lipids in young endurance athletes. In a review,
Rowland (1993) concluded that prepubertal athletes possess a superior lipoprotein profile
compared to the general pediatric population. The appropriateness of this conclusion can
be questioned given that comparisons were made with controls and not representative
reference values. If the control and reference values differ (i.e., TG lower than 50th

percentile, etc.), the interpretation of results may be questioned. Reference values from

61

the Lipids Research Clinics Prevalence Study were used as the comparison in the present
study since they were based on a large, random representative sample from the United
States during approximately the same time period as the study of young distance runners.

The variability of blood lipids among runners in this sample is of interest. The
results are apparently not related to dietary composition. Compared to a small control
sample at the time of the first visit, the young runners did not significantly differ in
dietary intake of energy, protein, or fats among the runners (Schemmel et al., 1986). No
subjects indicated frequent smoking.

Some runners displayed blood lipid levels that would be considered borderline or
dyslipidemic based on clinical guidelines. On the other hand, most of the runners have
blood lipid levels that are within desirable clinical guidelines, particularly HDL and TG.
Nevertheless, despite regular participation in an endurance sport, some young athletes.
may display undesirable levels of cholesterol. Kyle et al. (1991) reported the variability
in TC (65 to 274 mgldl) in young athletes, but did not address it in the discussion. On the
' other hand, the prevalence of hypercholesterolemia in the young athletes was 15.4% and
13.6% in boys and girls, respectively [TC levels above 185 mgldl (>90th percentile of
reference values)]. Future research should recognize dylipidemias in young athletes and
establish prevalence rates of dyslipidemia among youth athletic groups in general (i.e.,
endurance, strength/power, speed) and in specific sports (cross-country, basketball,
football, etc.).

The rationale for the hypotheses of this study was based on the exposure of young
distance runners to high levels of vigorous physical activity. Age-specific means

indicated that HDL in runners less than 14 yrs was the only blood lipid that appeared to

62

be enhanced compared to the general population. The absence of lower IDL levels in
young distance runners may be masked by the effect of exercise training on the
biologically important LDL subfractions. Trained adult runners show lower levels of
small, dense LDL particles compared to non-runners (“Williams et al., 1986). The
inability to demonstrate a superior blood lipid profile in adolescent distance runners may
be due to the presence of desirable values in both runners and the general population of
youth. Compared to the blood lipid profile of the adult endurance athlete, young distance
runners have lower levels of TC, LDL, and TG (Table 3.4). This comparison would
suggest that compared to adult endurance athletes, young endurance athletes, and youth
in general, possess a more favorable blood lipid profile. Thus, the prevention of
atherosclerosis may have the greatest impact during the transition from adolescence into
adulthood.

In contrast to the finding of greater HDL in runners prior to 14 yrs, the age-related
decline in HDL during adolescence in boys was not attenuated, suggesting that regular
endurance training during adolescence does not attenuate the decline in HDL in boys.
Longitudinal analysis also indicated a decline in HDL in female runners. The age-related
decline. in HDL during adolescence in boys has been explained by corresponding changes
in androgens and body fat distribution (Baumgartner et al., 1989; Laskarzewski et al.,
1983). Previous studies indicate that HDL remains stable in girls during adolescence
(Baumgartner et al., 1989; Christenson et al., 1980). The initial values are significantly
higher than the general population, and although values decrease with age, they remain

slightly above those observed in the general population.

The representativeness of the sample of young runners limits generalizations of
the results of this study. Subjects were not chosen randomly and the response rate was
unknown. A representative sample of 'elite' young athletes from various sports would
provide greater insights into the development of blood lipids among active youth.

In conclusion, the results of this study do not support the hypothesis that young
distance runners possess a superior blood lipid profile compared to the general population
during adolescence, except for HDL in the younger age groups. Likewise, the attenuation
of the decline in HDL during male adolescence was not observed in young distance
runners. During adolescence, it appears that blood lipids show a similar pattern of
development in well-trained endurance athletes and the general population. In contrast to
the comparison of mean values with reference values, most young distance runners
possess a favorable blood lipid profile when compared to clinical values. However,
dyslipidemic values are present. Further longitudinal study of the influence of regular

exercise on the blood lipid profile during adolescence is warranted.

ACKNOWLEDGEMENTS
Special thanks to Vern D. Seefeldt and other faculty and staff of the Institute for
the Study of Youth Sports who contributed in the data collection of this study. The

dedication of parents and participants involved in this study is also appreciated.

65

REFERENCES

Atomi Y, Kuroda Y, Asami T, Kawahara T (1986) HDL-cholesterol of children (10 to 12
years of age) related to Vozmax, body fat, and sex. In J Rutenfranz, R Mocellin and F
Klimt (eds.): Children and Exercise XII. Champaign, IL: Human Kinetics, pp. 167-172.

Baumgartner RN, Siervogel RM, Chumlea WC, Roche AF (1989) Associations between
plasma lipoprotein cholesterols, adiposity and adipose tissue distribution during
adolescence. Int J Obes 13:31-41.

Berenson GS, Srinivasan SR, Bao W, Newman WP, Tracy RE, Wattigney WA (1998)
Association between multiple cardiovascular risk factors and atherosclerosis in children
and young adults. N Engl J Med 338: 1650-1656.

Blair SN, Kohl HW, Paffenbarger RS, Clark DG, Cooper KH, Gibbons LW (1989)
Physical fitness and all—cause mortality: a prospective study of healthy men and women.
JAMA 17:2395-2401.

Christenson B, Glueck C, Kwiterovich P, Degroot I, Chase G, Heiss G, Mowery R,
Tamir I, Rifkind B (1980) Plasma cholesterol and triglyceride distributions in 13,665
children and adolescents: the Prevalence Study of the Lipid Research Clinics Program.
Pediatr Res 14:194-202.

Enos WF, Beyer JC, Holmes RH (1955) Pathogenesis of coronary disease in American
soldiers killed in Korea. JAMA 158:912-914.

Haskell WL (1984) The influence of exercise on the concentrations of triglyceride and
cholesterol in human plasma. Exerc Sport Sci Rev 12:205-244.

Hetzel BS, Berenson GS, eds. (1987) Cardiovascular Risk Factors in Childhood:
Epidemiology and Prevention. Amsterdam: Elsevier.

Hickman TB, Briefel RR, Carroll MD, Rifkind BM, Cleeman JI, Maurer KR, Johnson
CL (1998) Distributions and trends of serum lipid levels among United States children
and adolescents ages 4-19 years: data from the Third National Health and Nutrition
Examination Survey. Prev Med 27:879-890.

Kyle JM, Walker RB, Riales RR, Petty G], Thomas JA, Roberts MD (1991) Student
athletes cholesterol screening during routine precompetition examination. J Fam Pract
33:172-176.

Labarthe DR, O'Brien B, Dunn K (1991) International comparisons of plasma cholesterol
and lipoproteins. Ann NY Acad Sci 301:108-119.

Laskarzewski P, Morrison J, Gutai J, Orchard T, Khoury P, Glueck C (1983) High and
low density lipoprotein cholesterols in adolescent boys: relationships with endogenous
testosterone, estradiol, and quetelet index. Metabolism 32:262-271.

Lipid Research Clinics Manual of Laboratory Operations (1974) Lipid and Lipoprotein
Analysis. Washington, DC: United States Department of Health, Education, and Welfare.

Macek M, Bell D, Rutenfranz J, Vavra J, Masopust J, Neidhart B, Schmidt K-H (1989) A
comparison of coronary risk factors in groups of trained and untrained adolescents. Eur J
Appl Physiol 58:577-582.

Mahoney LT, Burns TL, Stanford W, Thompson BH, Witt JD, Rost CA, Lauer RM
(1996) Coronary risk factors measured in childhood and young adult life are associated

with coronary artery calcification in young adults: the Muscatine Study. J Am Coll
Cardiol 27:277-284.

Nizankowska-Blaz T, Abramowicz T (1983) Effects of intensive physical training on
serum lipids and lipoproteins. Acta Paediatr Scand 72:357-359.

Paffenbarger RS, Lee I-M (1996) Physical activity and fitness for health and longevity.
Res Q Exerc Sport 67 (Suppl. 3):]1-28.

Rowland TW (1993) The physiological impact of intensive training on the prepubertal,
athlete. In BR Cahill and AJ Pearl (eds.): Intensive Participation in Children's Sports.-
Champaign, IL: Human Kinetics, pp. 167-193.

Schemmel RA, Stone M, Conn C (1986) Comparison of dietary habits and nutrient intake
of competitive runners with age- and gender-matched controls. In MR Weiss and D

' Gould (eds.): Sport for Children and Youths. Champaign, IL: Human Kinetics, pp. 230-
236.

Smith BW, Metheny WP, Van Huss WD, Seefeldt VD, Sparrow AW (1983) Serum lipids
and lipoprotein profiles in elite age-group endurance runners. Circulation :III-191
(Abstract).

Smith BW, Sparrow AW, Heusner W, Van Huss WD, Conn C (1985) Serum lipid
profiles of pre-teenage swimmers. Med Sci Sports Exerc 17:220 (Abstract).

Strong JP, Malcom GT, Oalmann MC, Wissler RW (1997) The PDAY Study: natural
history, risk factors, and pathobiology. Ann NY Acad Sci 811:226-235.

Valimaki I, Hursti M-L, Pihlakoski L, Viikari J (1980) Exercise performance and serum
lipids inrelation to physical activity in schoolchildren. Int J Sports Med 1: 132-136.

Williams PT, Krauss RM, Wood PD, Lindgren FT, Giotas C, Vranizan KM (1986)
Lipoprotein subfractions of runners and sedentary men. Metabolism 35:45-52.

67

Zonderland ML, Erich WBM, Peltenburg AL, Havekes L, Bernink MJE, Huisveld IA
(1984) Apolipoprotein and lipid profiles in young female athletes. Int J Sports Med 5:78-
82.

Hugo w... 5.8-8.8982. $58.05 .= 6.00.. .65» 0.. «055 as... 8.358 2.50:...

 

 

a an ID? PUP HQ ACID—t FUCIUP

o- .c .N .mob ANmb. mob CMN. .omb ANA.A. qmm Gm. .. w. 3 8am. ..mm AOQA.
.wNN .m mum. Am. 54 N4- .oN ..m.u.A.ma o.om.N.mA

. . .w 30.. ANAb. a. .m Cub. 8N Nah. mob 09.0. New 3.8. .me 8.2.
.meom Nm.mw MN- .Nm NM. .3. . .a©-m.mN o.mN-w .mq

.N .N 3mm Awmb. m. .. CON. .8... 09m. mam Sway. m. .N 3.0m. .bN 8.8.
.chwN we. .8 mm. .mm N. -mN .bAimNo charubq

.w .4 God Nu... MGM Cwb. OWN AN..M. do... 0A6. w. 3 86m. ..®A 863
.NTNOm A. .mQ am. 50 Am. .40 N. .NAuq 0.8-? .o

.A .q _ .13.. ANNA. MMM Cum. 5.8m COM. qu Cub. ubm APmN. Nam Ache.
.mMNOA mmhm um. .mm am. am N.Am.A.mo ..Nq-w.ma

.m .w 39m New. Am.m Cob. .on Cub. emu ANNb. 9A.. 8.8. N. 5 80A.
. .QNNM N404 q . - 3. we. .OA Newman. ..AN-w.ma

.m m SAN Cub. Am... G. .. .8... Nab. 3.. CNN. Pam Ache. N.Nq 86m.
.NqNAq mum... ac- .MA mm-N.w NwmAbA ..8-w.om

3+ 4 .mob $9M. 3.0 3.0. gm.— QAN. .8... 3m. .. AbA CA3 NmN C. .N.
.wwNm. NWMA do. 30 quNq Name“... quAbu

 

man 8.: ~04 0.068.328?

>.. 558m 30 axe—.88.. .a am}...
5.58 Ed .525 EU. 8:. Bane.

69

Hugo uN. >ma-mmmoo.m.2. $53.0: .a 6.02. :03... ococam ..oaao 3.0.8:? 2553.

 

 

a 4.0 EDP ..UF 4.0 ACID? FUFHIUF

0- .o m .490 390. mm. 3.0. 0m.@ :06. m..m Cab. N60 894. ..MA 3mm.
.kN-NNw k.4-44 am. .wA NNba NNN-w ha . .ON-N .m

. . .. .090 090. EN ANON. . .0.m Gmh. amm N. .m. wit. 2. .0. N. .0 2.8.
.N4-NAO N4. . . 9- .mm N0. .om ..4N-m.4o ohm-A. .m

.N .w .4w.m $90. m0.m 20b. .oNN 30.9 mmh 20.3 m. .o APmN. ..mm 3.44.
. .4-NwN wu- .ON aw- .A4 u9ma ..4rbo o.am.w.om

.w .m .amu ANNA. m0.4 ANcb. 00... NM... $0 Gob. N00 3.0N. ..4A 8.8.
.uN-NNm w .-.o0 2-30 N4-.40 ..mA.MNO o.4m.w.m.

.A .N .4m.m Gab. 3.0 Cub. 0m... Ath. 40b 00b. NOA 86m. ..am 80a.
.NaNAw $0”. 4&- .MA 8. .ma . .00.... .0 o.0o-Nmm

.m .N .406 EmN. m0. Emu. .90 64.0. mNm 34h. PNw 2.0m. ..mo 8.4m.
.N4-N4A N4-00 aa- .44 w0N3 N. 5.946 0.40-0.8

.a m 50.0 20.3 mNo Ga. 0NA Cub. 49a Cub. w.O0 8.... ..mo Ban.
:5. .4. fixmm 44-. . . MN- .uw NMN-PM4 ..ww-NNN

.4... m .mmb GNa. 3.4 Cu... . .m.0 390. ¢0.m GA... w.m4 2.04. Nwo 3.0a.
..0-N4. uu-4m am. .0. N0. .m N. 3.9% rah-PA.

 

man 8.: ..2. mac—divas?

>= $5.2 30 30332. .= 3m}...
<m.=om Ed .525 EU. 2:. 35m».

70

Table 3.3. Means and standard deviations for blood lipids at baseline and at last visit in
longitudinal cohorts of male and female distance runners.

 

{ Variable Males (n=21) Females (n=18)
TC (mg/d1)
Baseline 177.6 (23.0) 180.2 (36.1)
Last visit 163.1 (29.6)* 170.0 (34.3)*
HDL (mg/d1)
Baseline 66.7 (9.2) 71.2 (16.8)
Last visit 44.3 (9.2)* 52.9 (13.0)*
LDL (mg/d1)
Baseline 97.1 (24.3) 97.8 (31.7)
Last visit 101.5 (26.3) 103.3 (31.9)
TG (mg/d1)
Baseline 68.5 (39.1) 54.4 (26.9)
Last visit 86.5 (45.5) 68.4 (29.9)*

 

* significantly different from baseline (p<0.05).

71

Table 3.4. Comparison of blood lipid profiles of adolescent and adult distance runners
and non-athletes. Data from present study, Christenson et al., (1980), and Haskell

 

 

( 1984).
13 yr old 13 yr old 16 yr old 16 yr old Adult Sedentary

runner reference runner reference runner adult
Males
TC 171 160 164 152 200 212
HDL 56 56 46 46 64 43
LDL 98 95 101 93 125 139
TG 79 62 87 68 70 146
Females .
TC 165 162 159 160 193 209
HDL . 60 53 52 53 75 56
LDL 93 93 92 95 1 13 124
TO 64 72 74 68 56 123» '

 

See text for abbreviations.

All values are expressed in mgldl.

72

'l‘C (mgldl)

225-

200‘

175 -‘

1504

125-1

 

MALES

 

 

 

 

 

 

 

 

10 11 12 13 14 15 16 17 18
Axum)
Figure 3. la. Total cholesterol in male distance runners

compared to age-specific reference values (Christenson et al.,
1980).

73

HDL(mg/dl)

 

 

 

 

 

 

 

 

9101112131415161718
Axum)

figure 3.1b. High-density lipoprotein (HDL) cholesterol in
male distance runners compared to age-specific reference
values (Christenson et al., 1980).

74

LDL (mg/d1)

 

 

 

 

 

 

 

 

 

 

 

 

 

MALES
1..
_ /' P95
125-./ “r “r
o, .0.. ’30-.“ I
100 "1 . a. ,o “.0.“ T ---- .0 ------ O
75" J. "- .11. J
v P5
50 ' I ' l 1 I F l 1
9 10 11 12 13 14 15 16 17 13

Age (yrs)
Figure 3.1c. Low-density lipoprotein (LDL) cholesterol in

male distance runners compared to age-specific reference
values (Christenson et al., 1980).

75

TG (mgldl)

MALES _
175 "

1504
P95

 

P50

 

 

 

 

 

 

 

9101112131415161718
A8606)

Figure 3.1d. Triglycerides (T G) in male distance runners

compared to age-specific reference values (Christenson et
al., 1980

76

TC (mg/d1)

 

 

 

 

 

 

 

100 r 1 r 1 1 1 T I 1
9 10 ll 12 l3 14 15 16 17 18

A80 (Y's)

Figure 3.2a. Total cholesterol in female distance runners
compared to age-specific reference values (Christenson et al.,

1980

HDL (mg/d1)

 

 

 

 

 

 

 

 

FEMALES
80" ..- .,. T
‘ P95
70~\T\ i
or... I
so- 1 "°"~-o---<>‘ Ono
‘\ \T - .0
50.4 , 1 P50
40‘ J. 1.. ._
\A_ 4" P5
30-
20 l l l l I l l I l
9 10 11 12 13 14 15 16 17 18
Axum)

Figure 3.2b. High-density lipoprotein (HDL) cholesterol in
female distance runners compared to age-specific reference
values (Christenson et al., 1980).

LDL (mgldl)

 

 

 

 

 

 

175 "
FEMALES
150 ‘
w P95
125-J
m u-
1 P50
75 ‘
P5
w F T I r I T T I I
9 10 ll 12 l3 14 15 16 17 18
Age (yrs)

Figure 3.2c. Low-density lipoprotein (LDL) cholesterol in
female distance runners compared to age-specific reference
values (Christenson et al., 1980).

79

TG (mgldl)

150 -
FEMALES _

125‘ —— P95

 

100- 1r

 

 

 

 

 

 

 

 

 

 

754
P50
g)-
‘ EPS
ab
25 I T F 1 l T 1 r 1
9 10 11 12 13 14 15 16 17 18

Age (m)

Figure 3.2d. Triglycerides (T G) in female distance runners
compared to age-specific reference values (Christenson et

al.. 1980

CHAPTER4

BLOOD LIPIDS OF YOUNG DISTANCE RUNNERS:

DISTRIBUTION AND COMPARISON

TO REFERENCE VALUES AND CLINICAL CUT-POINTS

81

ABSTRACT

This report describes the distribution of blood lipids in a sample of 48 male and
22 female young distance runners,lO-19 yrs of age. A fasting blood sample was obtained
by fingerprick. Total cholesterol (TC), high-density lipoprotein (HDL), and triglycerides
(T G) were analyzed by a portable cholesterol analyzer (Cholestech LDX). Low-density
lipoprotein (LDL) was estimated by the Friedewald equation. Comparisons were made
between sexes, and to a current reference sample (NHANES 111, 1988-1994) and clinical
cut-points. LDL was significantly greater in male compared to female runners. Males
also had a greater likelihood of having "undesirable" levels of blood lipids (i.e., above
desirable levels based on clinical cut-points) than females. Compared to current
reference medians, mean values of TC (p=0.07) and LDL (p<0.005) were higher, while
HDL (p=0.24) was slightly higher in male distance runners. Blood lipids in female
distance runners were comparable to reference medians. Compared to reference means,
TC, LDL, and TG in female runners are significantly lower (p<0.05). Although some
‘ subjects had dyslipidemic values, most possessed desirable levels of blood lipids. Thus,
blood lipids of young male distance runners from the mid-Michigan area are not, on
average, superior to the general population of U.S. youth. In females, results depend on
the reference of comparison (i.e., means or medians). Young distance runners show

considerable heterogeneity in blood lipid phenotypes, including dyslipidemic values.

INTRODUCTION

Adult long-distance runners generally display a superior blood lipid profile (i.e.,
elevated hi gh-density lipoprotein, lower low-density lipoprotein and triglycerides)
compared to the general population (Haskell, 1984). Relatively few studies have
examined the blood lipid profile of young athletes (Atomi et al., 1986; Kyle et al., 1991;
Macek et al., 1989; Nizankowska—Blaz and Abramowicz, 1983; Smith et al., 1985;
Valimaki et al., 1980; Zonderland et al., 1984). Results of these studies suggest that
young athletes also have a superior blood lipid profile compared to non-athletic control
samples. The studies are limited, however, by subject selection, sample size, mixed
samples (either by sex or sport), and inappropriate control groups. Further, the inter-
individual variability of blood lipids among young athletes is not ordinarily considered.

This brief report provides descriptive data on the distribution of blood lipids in a
current sample of young distance runners. Comparisons are made to a current reference

sample, clinical cut-points, and previous studies of young athletes.

METHODS

Subjects. Males and females 10 tol9 years of age participating on mid-Michigan junior
or senior high school cross-country and track teams during Fall 1999 and Spring 2000
were invited to participate in the current study. Subjects were also recruited by
advertisements in the local newspaper and at local road races. Exclusion criteria included
current smokers, excessive alcohol intake, current use of blood cholesterol lowering and
anti-hypertensive medication, anabOlic steroid use, hepatic, renal, and thyroid disease, or

training less than 30—40 weeks per year or the past three consecutive months. The total

83

number of eligible subjects in the mid-Michigan area during recruitment is difficult to
determine given the exclusion criteria. Forty-eight males and 22 females volunteered to
participate in the study. Parental consent and subject assent were obtained prior to
testing. The study was approved by the Michigan State University Committee on

Research Involving Human Subjects.

Blood lipids. Data collection occurred between the hours of 7:00 a.m.—12:00 pm. A
fasting blood sample was obtained by fingerprick after the subject had been seated for 10
minutes. Blood was collected in a 35 micro-liter capillary tube. Upon collection,
samples were analyzed for total cholesterol (TC), high-density lipoprotein (HDL), and
triglycerides (T G) within 5 minutes by a portable cholesterol analyzer according to the
protocol of the manufacturer (Cholestech LDX System, Hayward, CA). The lower limit
of analytic capacity of the Cholestech LDX for TG is 45 mgldl. 1n the present study,
eight individual values were recorded at the lower limit (e.g., <45 mgldl). Low-density
lipoprotein (LDL) cholesterol was estimated by the Friedewald equation (Friedewald et
al., 1972).

The total error of measurement of the Cholestech LDX analyzer has been
determined as 12.7%, 18.8%, and 19.7% for TC, HDL, and TG, respectively. The
reference laboratory measurement error is 8.1%,12.9%, and 5.1% for TC, HDL, and TG,
respectively (Bard et al., 1997). The total error for HDL met the standard set forth by
National Cholesterol Education Program (<22%), but TC and TG were slightly higher

than the respective standards (8.9 and 15%).

Within-day reliability was determined prior to the onset of the study by five
consecutive measurements with Cholestech Level 1 and 2 liquid control reagents. Day-
to-day reliability was determined by daily calibration prior to each testing session
throughout the study. Within-subject reliability was determined by duplicate measures of
l of every 5 male and female subjects. Coefficients of variation were used to express the
precision of the within-day and day-to-day trials and compared to national standards. The
coefficients of variation (CV) for within-day and day-to-day trials were less than 2% and
4%, respectively, for standard controls of TC, HDL, and TG. Within-subject precision of

blood lipid measurements was high (0.996, TC; 0.943, HDL; 0.970, TG).

Anthropometry. Stature and body mass were measured according to the procedures of the
International Biology Program (Weiner and Lourie, 1969). Stature was measured with
the subject standing erect, without shoes, and with weight distributed evenly between
both feet, heels together, arms relaxed at the sides, and the head in the Frankfort
horizontal plane with a fixed stadiometer. Body mass was measured on a beam scale

with the subject attired in running shorts and T—shirt without shoes. The stadiometer and

scale were calibrated throughout the study.

Sexual maturity status. Given the difficulty in direct assessment of sexual maturity status
in a non-medical environment, a self- assessment of sexual maturity status was used in
the current study. Self—assessment was conducted in a separate, private station following
an explanation of the purpose of the assessment. Subjects rated their stage of sexual

development relative to sex-appropriate sets of drawings/photos and verbal descriptions

85

(Van Wieringen et al., 1971) based on the criteria of Tanner (1962). In a study of 174
female and 178 male Brazilian youth age 6-26 years (Matsudo and Matsudo, 1994), the
concordance between self and physician assessments of secondary sex characteristics was
reasonably high (60—71.3%). Test-retest concordance (i.e., reproducibility) was also

similar between self- and physician-assessments.

Statistical analysis. Descriptive statistics were calculated for anthropometric and blood
lipid values. Sex differences were examined by a non-parametric Mann-Whitney U test.
A one-sample independent t-test was used to examine differences between group means
for distance runners and the general population. Reference values (means and medians)
for blood lipids from the Third National Health and Nutrition Examination Survey
(NHANES III) were used for comparison (Hickman et al., 1998). The sample was also
grouped by age (12-15 yrs and 16-19 yrs) for comparative purposes. Individual values
for each blood lipid were plotted by age relative to lines of identity for clinical cut-points

(see Table 4.4). An alpha level. of p<0.05 was used for statistical significance.

RESULTS

Subject characteristics are given in Table 4.1. Some of the inter—individual
variability in body size can be accounted for by chronological age and sexual maturity
status. Only 2 subjects are younger than 12.0 yrs of age. Both subjects are pre-
pubescent, whereas the remainder are late- or post- pubescent (genital/breast stages 4 or
5) (Table 4.2). Due to the lack of variation in pubertal status, maturity-associated

variation of blood lipids was not examined in this sample.

86

One subject was identified as an outlier and therefore eliminated from the analysis
(TC=309). The subject was referred to a physician for further diagnosis. Blood lipids of
young male and female distance runners and available reference medians are shown in
Table 43. LDL is significantly greater in male compared to female runners (p<0.05).
Mean values of TC (p=0.07) and LDL (p<0.005) are higher in male runners compared to
reference medians. HDL is also slightly higher in male distance runners (p=0.24). Mean
values in female distance runners are comparable to reference medians, but compared to
reference means, TC, LDL, and TG in female runners are significantly lower (p<0.05).

Scatterplots of individual values in relation to clinical cut-points are shown in
Figures la—d. Although some subjects have dyslipidemic values, most possess desirable
blood lipid levels (Table 4.4). When borderline and high (or low for HDL) values are
grouped together to represent "undesirable” levels, chi-square analysis indicates that _
males have a greater likelihood of being classified as undesirable for LDL and TG
compared to females (p<0.05). There is a trend for a greater likelihood of males being

‘ classified as undesirable for TC (p=0.20) and HDL (p=0.12).

DISCUSSION

This is perhaps the largest sample of young male and female athletes which
fasting blood lipids (TC, HDL, LDL, TG) are reported . A previous report of 777 (454
males and 323 females) athletes in West Virginia included only TC (Kyle et al., 1991).
The findings from this brief report indicate that mean values for blood lipids of young
male distance runners from the mid-Michigan area are not superior to the general

population of U.S. youth. TC, LDL, and T6 are lower in young female distance runners

87

compared to the general population. There is considerable inter-individual variability in
blood lipids, including dyslipidemic values, among adolescent distance runners.
Nevertheless, most subjects displayed desirable levels compared to clinical cut-points
(Table 4.4).

Previous studies comparing the blood lipids of young athletes and controls have
reported similar levels of TC, lower levels of TG and LDL, and higher levels of HDL
(Atomi et al., 1986; Kyle et al., 1991; Macek et al., 1989; Nizankowska—Blaz and
Abramowicz, 1983; Smith et al., 1985; Valimaki et al., 1980; Zonderland et al., 1984).
Mean differences among samples for TC are generally within 1:10 mgldl while mean
differences for LDL and TG range from 5-25 mgldl and 20-35 mgldl, respectively. Two
studies reported no appreciable differences in TO (Atomi et al., 1986; Smith et al., 1985).
Mean differences among previous studies for HDL range from 4—17 mgldl higher in
adolescent athletes. This study is apparently the first to report higher levels of LDL in
adolescent athletes. The findings in the present sample generally contrast previous
studies that have shown that child and adult male endurance athletes possess a superior
blood lipid profile compared to the general population. The reasons for this observation
are considered subsequently.

Previous cross-sectional studies have several methodological limitations that need
to be addressed. First, the main purpose of cross-sectional studies of athletes is to
provide information on the status of a sample. To truly provide an accurate description of
a group (i.e., adolescent distance runners), a random representative sample should be
selected. An additional question is,“Who is the adolescent distance runner?” Athletes

may compete at several levels (i.e., international, national, regional, state, local,

88

recreational), and often compete in more than one sport. In most studies, inadequate
information is provided on the sample and subject selection process. No study has
included a random representative sample of young distance runners or athletes.
Likewise, no study has included a random sample of young athletes at different
competitive levels.

Response rate is another measure of subject recruitment that has been neglected.
It is also important to consider the selection of control subjects. Only one study has used
randomly selected controls (Macek et al., 1989). Others are based on convenient
samples. Therefore, blood lipids in control subjects may not be representative of the
general population. Since most of these studies were conducted outside of the U.S., the
availability of national reference data is unknown. To avoid such complications,
reference medians from NHANES 111 (1988-1994) were used in the current study.
Additional methodological issues include the analytic and biological variability that
influence blood lipids.

Compared to previous studies of young athletes, mean values of TC and LDL are
comparable. In contrast, levels of HDL and TG are lower and higher, respectively. The
range of previously reported values of HDL and TG are 52-75 mgldl and 54—71 mgldl,
respectively. However, comparison of the results to international samples warrants
caution due to known population differences in blood lipids (Labarthe et al., 1991).
Subjects in previous studies were also generally younger than in the present study. The
age-related decrease in HDL and increase in TG is well documented, particularly in
males (T amir et al., 1981), and may contribute to differences between samples of young

athletes and in the comparison with reference values.

89

A finding of potential interest in the present study is the heterogeneity in blood
lipid phenotypes among young distance runners. Few studies examine variability in the
phenotypic expression of a given biological variable of athletic subgroups (Eisenmann
and Malina, in press). Valimiki et al. (1980) showed variability of HDL plotted against
physical work capacity in a small sample of boys. Extrapolation from the figure shows
values ranging from 55 to 85 mgldl in 9 trained boys. The range of HDL (and other
blood lipids) is much greater in the present sample of boys (28 to 88 mgldl) and includes
dyslipidemic values (Table 4.4). Only one study has specifically examined the
prevalence of dyslipidemia in young athletes (Kyle et al., 1991). In the sample of 777
West Virginia junior high school athletes, 114 (15%) were classified as having elevated
TC (>185 mgldl). Sex-specific prevalence rates were 15.4% and 13.6% for boys and
girls, respectively. Of these individuals, 8% (11:60) had values greater than 200 mgldl.
In comparison, 6 of the 69 (9%) runners in the original sample had values greater than
200 mgldl. Overall, TC ranged from 65-274 mgldl in the sample of Kyle et al. (1991)
compared to 100—241 in the present study. Unfortunately, athletic subgroups were not
identified in the West Virginia sample. In adult male runners, the prevalence of clinically
diagnosed low HDL (<35 mgldl) and high LDL (>160 mgldl) was associated with
distance run per week (Williams, 1997). These findings along with those of the present
study show that even distance runners involved in high levels of training may possess
dyslipidemic values. The results also provide implications for the pre-season screening of
youth athletes for dyslipidemia, particularly if a family history of cardiovascular disease

is present (American Academy of Pediatrics, 1992).

“VT" ' '

 

Variability in blood lipids may also result from analytical or biological factors.
As shown by Kyle et al. (1991), a single measurement warrants consideration of daily
variation and regression to the mean. Upon follow—up testing, 38 of 74 (51%) of the
West Virginia athletes had values that remained above 185 mgldl. Due to feasibility
issues, repeat measurements were not taken on subjects who displayed initial
dyslipidemic values in the present study. However, the day-to-day variation suggests that
multiple measures may provide a better indication of individual values. Based on the
single measurement obtained in this study, the coefficients of variation was low (2-4%)
and similar to other reports using the same portable analyzer. Rogers et al. (1993)
reported within-day and day-to-day precision between 1.5-1.8% and 2.8-3.4%,
respectively for TC. These results provide evidence of the reliability and reproducibility
of blood measurements in this study. lnforrnation on the precision of blood lipid
measurements is not provided in previous studies of young athletes.

Biological factors that need to be considered include chronological age, seasonal
variation, dietary and alcohol intake, acute and chronic exercise, family history,

. genotype, psychological stress, biological maturity, and body fatness. Other factors
(secondary dyslipoproteinemias, medication, trauma and acute infections, pregnancy,
blood collection conditions (e.g., fasted vs. non—fasted state, position, storage, etc.) (Naito
and Kwak, 1992) were considered in the design of this study. Each of these analytical
and biological factors should be considered when comparing blood lipid studies.

In conclusion, the results of the study suggest that young male distance runners
from the mid-Michigan area do not possess a superior blood lipid profile compared to

reference values for United States youth and contrast previous studies of young athletes.

91

Results for females depend on the reference comparison (i.e., means or medians).
However, several methodological limitations need to be recognized, specifically subject
selection, biological confounders, analytical errors, and comparison methods. The
considerable heterogeneity in blood lipid phenotypes and dyslipidemia of young athletes .
warrants exploration. The contribution of training volume, peak oxygen consumption,
body fatness, and family history of vascular diseases on the blood lipid profile in this

sample is considered in a separate analysis.

ACKNOWLEDGMENTS

This study was supported in part by the William Wohlgamuth Memorial
Fellowship and The Institute for the Study of Youth Sports. Special thanks is given to
members of the Human Energy Research Laboratory for excellent technical assistance
during data collection and cross-country coaches and athletes who participated in this

study.

REFERENCES

American Academy of Pediatrics (1992) National Cholesterol Education Program Report
of the expert panel on blood cholesterol levels in children and adolescence. Pediatr
89:525-577.

Atomi Y, Kuroda Y, Asami T, Kawahara T. (1986) HDL-cholesterol of children (10 to 12
years of age) related to Vo2max, body fat, and sex. In J Rutenfranz, R Mocellin and F
Klimt (eds.): Children and Exercise XII. Champaign, IL: Human Kinetics, pp. 167-172.

Bard RL, Kaminsky LA, Whaley MH, Zajakowski S (1997) Evaluation of lipid profile
measurements obtained from the Cholestech LDX analyzer. J Cardiopul Rehab 17:413-
418.

Eisenmann JC, Malina RM (in press) Body size and endurance performance. In RJ
Shephard (ed.): Endurance in Sport. Oxford: Blackwell Science.

Friedewald WT, Levy RI, Fredrickson DS (1972) Estimation of the concentration of low
density lipoprotein cholesterol in plasma without use of the preparative ultracentrifuge.
Clin Chem 18:499-502.

Haskell WL (1984) The influence of exercise on the concentrations of triglyceride and
cholesterol in human plasma. Exerc Sport Sci Rev 12:205—244.

Hickman TB, Briefel RR, Carroll MD, Rifkind BM, Cleeman JI, Maurer KR, Johnson
CL (1998) Distributions and trends of serum lipid levels among United States children
and adolescents ages 4—19 years: data from the Third National Health and Nutrition
Examination Survey. Prev Med 27:879-890.

Kyle JM, Walker RB, Riales RR, Petty G], Thomas JA, Roberts MD (1991) Student
athletes cholesterol screening during routine precompetition examination. J Fam Pract
33:172-176.

Labarthe DR, O'Brien B, Dunn K (1991) lntemational comparisons of plasma cholesterol
and lipoproteins. Ann NY Acad Sci 301:108-119.

Macek M, Bell D, Rutenfranz J, Vavra J, Masopust J, Neidhart B, Schmidt K-H (1989) A
comparison of coronary risk factors in groups of trained and untrained adolescents. Eur J
Appl Physiol 58:577-582.

Matsudo SMM, Matsudo VKR (1994) Self-assessment and physician assessment of
sexual maturation in Brazilian boys and girls: concordance and reproducibility. Am J
Hum Biol 6:451-455.

Naito HK, Kwak Y-S (1992) Accurate measurement of serum total cholesterol: the need
for standardization. J Am Coll Nutr 11 (Suppl.):8S-15S.

94

Nizankowska-Blaz T, Abramowicz T (1983) Effects of intensive physical training on
serum lipids and lipoproteins. Acta Paediatr Scand 72:357—359.

Rogers FJ, Misner L, Ockene 1S, Nicolosi RJ (1993) Evaluation of seven Cholestech
LDX analyzers for total cholesterol determinations. Clin Chem 39:860—864.

Smith BW, Sparrow AW, Heusner W, Van Huss WD, Conn C (1985) Serum lipid
profiles of pre-teenage swimmers. Med Sci Sports Exerc 17:220 (Abstract).

Tamir I, Heiss G, Glueck CJ, Christensen B, Kwiterovich P, Rifkind BM (1981) Lipid
and lipoprotein distributions in White children ages 6-19 yr: the Lipid Research Clinics
Program Prevalence Study. J Chron Dis 34:27—39.

Tanner JM (1962) Growth at Adolescence. 2nd edition, Oxford: Blackwell.

Tikkanen HO, Harkcnen M, Naveri H, Hamalainen E, Elovainio R, Sarna S, Heikki Frick
M (1991) Relationship of skeletal muscle fiber type to serum high density lipoprotein
cholesterol and apolipoprotein A-l levels. Atherosclerosis 90:49-57.

Valimaki I, Hursti M-L, Pihlakoski L, Viikari J (1980) Exercise performance and serum
lipids inrelation to physical activity in schoolchildren. Int J Sports Med 1: 132- 136.

Van Wieringen JC, Wafelbakker F, Verbrugge HP, de Haas JH (1971) Growth Diagrams
1965 Netherlands: Second National Survey on 0-24-year-oldsGroningen: Wolters-
Noorhoff Publishing.

Weiner JS, Lourie JA (1969) Human Biology: A Guide to Field Methods. Oxford:
Blackwell Science.

Williams PT (1997) Relationship of distance run per week to coronary heart disease risk
factors in 8283 male runners. Arch Intern Med 157:191-198.

Williams PT, Wood PD, Haskell WL, Vranizan K (1982) The effects of running mileage
and duration on plasma lipoprotein levels. JAMA 247:2674—2679.

Zonderland ML, Erich WBM, Peltenburg AL, Havekes L, Bernink MJE, Huisveld IA
(1984) Apolipoprotein and lipid profiles in young female athletes. Int J Sports Med 5:78-
82.

95

Table 4.1. Characteristics of young distance runners.

 

Males (n=47) Females (n=22)
Age (yrs) 16.7 (2.0) 15.5 (2.6)
’ 104193 9.9-18.4
Ht (cm) 174.0 (7.3)* 161.1 (9.8)
149.2- 185.1 132.7-178.1
Wt (kg) 62.1 (8.1)* 50.1 (10.4)
39.9-85.9 27.0-71.1

 

Values are mean (SD) and range.
*p<0.05 between sexes.

Table 4.2. Distribution of subjects by pubertal status.

“.15. A‘Ifii‘iiimfir

 

 

M F M F
G/Bl 1 l PHI 1 1
G/B2 2 l PH2 l l
G/B3 1 l PH3 l l
G/B4 20 9 PH4 8 3
G/BS 23 10 PHS 36 16

 

G, genital (males), B, breast (females); PH, pubic hair. The corresponding number
represents the stage (e.g., PHI, stage 1 - pubic hair).

Table 4.3. Blood lipids of young distance runners compared to reference means and medians (Hickman e1

 

al., 1998).
Males Females
Runners Reference Reference Runners Reference Reference
Mean Median Mean Median
TC
Total 165.8 (4.4)‘ 158 (1.2) 156 157.6 (3.8)” 167 ( 1.3) 161
12-15yrs 148.5 (4.7) 158 (1.6) 157 150.7(10.5) 164 (1.9) 159
16—19 yrs 170.1 (5.4) 158 (1.8) 155 161.0 (3.7) 171 (2.3) 163
HDL
Total 48.2 (1.9) 47 (0.6) 46 51.7 (2.4) 52 (0.5) 51
12-15 yrs 51.8 (4.3) 48 (0.7) 46 45.2 (2.5) 51 (0.8) 50
16-19 yrs 46.9 (2.2) 46 (0.9) 45 53.9 (3.1) 52 (0.7) 52
LDL
Total 101.6 (4.0)“ 91 (2.1) 88 91.0 (3.6)” 99 (2.4) 92
12-15 yrs 79.6 (5.3) 88 (2.4) 83 89.5 (8.9) 94 (2.8) 90
16-19 yrs 107.1 (5.4) 94 (3.8) 89 92.6 (4.0) 103 (4.4) 94
TC
Total 83.0 (5.6) ' 91 (4.0) 74 75.1 (21.8)" 96 (3.9) 77
12-15 yrs 74.1 (10.2) 87 (7.0) 72 82.2 (14.4) 96 (5.6) 81
16-19 yrs 86.3 (6.6) 94 (6. 1) 79 72.4 (4.0) 96 (5.9) 76

 

Values are mean (SE).
Sample sizes for males are; total=47, 12-15 yrs=10, 16-19 yrs=36.
Sample sizes for females are; total=22, 12-15 yrs=6, 16-19 yrs=15.
' significant sex difference (p<0.05).
" significantly different from reference mean (p<0.05).

° significantly different from reference median (p<0.05).

le 4.4.

ist 'bution o sub'ects b clinical cut- ints.

 

Category Level (mgldl) Total Males Females
TC
High _>_200 6 (8.7) 6 (12.8) 0 (0.0)
Borderline high 170-199 17 (24.6) 12 (25.5) 5 (22.7)
Desirable 5170 46 (66.7) 29 (61.7) 17 (67.3)
LDL
High 2130 8 (12.3) 7 (163) 1 (4.5)
Borderline high “0129 7 (10.8) 6 (13.9) 1 (4.5)
Desirable 5110 50 (76.9) 30 (69.8) 20 (90.0)
HDL
Low _<_35 8 (11.6) 7 (14.9) 1 (4.5)
Borderline low 35—45 20 (29.0) 15 (31.9) 5 (22.7)
Desirable 245 41 (59.4) 25 (53.2) 16 (72.7)
TG
High 2 130 8 (1 1.6) 7 (14.9) 1 (4.5)
Borderline high 90—129 12 (17.4) 10 (21.3) 2 (9.1)
Desirable 590 49 (71.0) 30 (63.8) 19 (86.4)

 

TC, total cholesterol; LDL, low density lipoprotein cholesterol; HDL, high density

lipoprotein cholesterol; TG, triglycerides.

Values represent number of subjects (%).

TC (mg/d1)

 

 

 

 

m-
00
225- O
O O
200 (65-_—
O o O % o
175‘ O 0 Ma]
fib— es
0 0:50 O 0800%
150- o % O 613% 000 0 Females
O O
125- o 00 °
0 0
11X)“ 0
75 1 1 1 I r I 1 r 1 r 1
9 10 11 12 13 14 15 16 17 I8 19 20

Age (3113)

Frgure 4.1a. Total cholesterol (TC) in young distance runners.
Lines of identity represent clinical cut-points (see Table 4.4)

100

HDL(mg/d1)

95 -

 

 

 

 

o
85-
O O o
75‘
o
°o
65-
0000 o
oo 0
55- O
0 0 0° é‘uo
0 359100
45 O % U
35 A 6) 313028800 0 Males
vg 8 Females
25 I I I T T I I I I I I
9 10 11 12 13 14 15 l6 17 18 19 20

A8¢ (yrs)

Figure 4.1b. Hi gh-density lipoprotein (HDL) in young distance runners.
Lines of identity represent clinical cut-points (see Table 4.4).

101

1Dl.(rng/dl)

 

 

 

150-
140- 2°
130 :0—
- O
110 a 0
° “6‘8
100‘
0
Q0
00 00
80- O o
0 0° % 9 0 Males
70- ° 00
O 0 Females
60 r r 1 T F1 1 Tu 1 1

 

910111213141516171819 20

Age (yrs)

Figure 4.1c. Low-density lipoprotein (LDL) in young distance runners.
Lines of identity represent clinical cut-points (see Table 4.4).

102

TG (mgldl)

190 -
180‘
170 -'
160 '-
150 -
140 ‘

 

130
120 "
llO "1
100 '1

 

90
80—
70!
60—
50

 

#120030

0
O
0
e————
0
U
$0“ $3 0 Males
002088 0 Females
O

 

40

 

9

rrr
181920

Figure 4. 1d. Triglycerides (TG) in young distance runners.
Lines of identity represent clinical cut-points (see Table 4.4).

'CHAPTER 5

INTER-RELATIONSHIPS AMONG TRAINING VOLUME,

PEAK OXYGEN CONSUMPTION, BODY FATNESS, AND BLOOD LIPIDS

IN YOUNG DISTANCE RUNNERS

104

ABSTRACT

The inter-relationships among training volume (km per wk), peak oxygen
consumption (peak V02, nrl'kg"‘nrin"‘), body fatness, and blood lipids were examined in
45 males, 22 female young distance runners, 10 to 19 yrs of age. Training volume (TV)
was estimated as the average of self -reported running habits during the past 3 months.
Peak Vo2 was measured by indirect calorimetry during an incremental treadmill test to
exhaustion. Skinfold thicknesses were measured with a Lange caliper at six sites (SSF).
The trunk-to-extremity ratio (TER) was calculated and used as an index of relative
subcutaneous fat distribution. Blood lipids were measured in a fasted state by a portable
cholesterol analyzer (Cholestech LDX). Relationships were assessed by partial
correlations controlling for age and self -assessed stage of pubertal development. Overall,
correlations are low to moderate. TV is weakly correlated with blood lipids, and the
relationships are in the expected direction for HDL and TC in males but not in females.
TV may be indirectly related with HDL through its relationship with peak Vo2 in males
(1:032). There are differential relationships between TV and HDL when the entire
sample was grouped according to modified clinical cut-points. TV is significantly related
to I-IDL in subjects with HDL<45 mgldl (r=0.40, p<0.05). The TER is not related to
blood lipids in males, with partial correlation coefficients near zero. In females,
correlations between SSF, TER and blood lipids are similarly low (m 0.16 to -0.27), with
the exception of a moderately high correlation between TER and HDL (r: -0.60). The
most consistent inter-relationships exist among TV, peak V02, SSF, and HDL. Partial
correlations range from low (0.10, TV) to moderate (0.37, SSF; 0.41, peak V0,). The

correlation between peak Vo2 and HDL remain significant after controlling for age and

105

SSF, while the partial correlation between SSF and HDL controlling for age and peak
Vo2 is reduced and not significant (r=-0. 19, p=0.20). Similar relationships were found in
females. When age and SSF are controlled, the correlation between peak Vo2 and HDL is
0.32, whereas the partial correlation between SSF and HDL, controlling for age and peak
Vo2 , is 0.00. The results highlight the complex inter-relationships among training
volume, peak V02, body fatness and HDL, and indicate the unique contribution of peak

V02 as an important predictor of HDL in young distance runners.

106

WRESP‘“

INTRODUCTION
Relationships among physical activity, physical fitness, and health are topics of
considerable interest and research in pediatric exercise epidemiology (Casperson et al.,
1998). In particular, the prevention of atherosclerosis by modification of the blood lipid
profile with exercise has been a major focus (Despres et al., 1990; Gutin and Owens,
1996; Riopel et al., 1986). A potential outcome is the establishment of recommendations
for physical activity in the adolescent population (Sallis et al., 1994). Based on currently
available data, it has been recommended that four 30 minute exercise sessions per week
at 75-80% of maximum heart rate may be an appropriate prescription for adolescents 12-
21 years of age (Armstrong and Simons-Morton, 1994). This recommendation is derived
from adult data suggesting that an exercise level equivalent to jogging 10—15 miles/wk is
necessary to significantly alter or favorably maintain blood lipid levels (Superko, 1991;
Williams, 1994).
A question related to the relationships between physical activity and blood lipids

’in children and adolescents is the following: Do health benefits accrue in youth at levels
of exercise that exceed current recommendations? A possible methodological approach
to this question is the use of a special exposure group (i.e., distance runners), which
would allow for the examination of an exposure (i.e., high levels of physical activity) that
is generally not observed in the general population. Recently, Williams (1996, 1997,
1998) has used this approach to challenge adult guidelines for physical activity (Pate et
al., 1995). Results suggest that increased levels of training in adult recreational runners

offers further health benefits, including an improved blood lipid profile. No study has

107

apparently examined the influence of training volume (distance run per week) on blood
lipids in young endurance athletes.

Maximal aerobic fitness and body fatness are also correlates of blood lipids
during childhood and adolescence. More aerobically fit youth generally have higher
high-density lipoprotein (HDL) and lower triglycerides (T G) (Despres et al., 1990;
Malina, 1990). Various measures of body fatness are positively associated with
atherogenic blood lipids and negatively associated with HDL (Guo et al., 1994). Relative
body fat distribution, and specifically a truncal and/or visceral fat patterning, is also
strongly linked to an adverse blood lipid profile in youth and adults (Baumgartner et al.,
1989; Despres, 1997; Freedman et al., 1989). The influence of whole-body aerobic
fitness and fatness are thought to be mediated by skeletal muscle and adipose tissue
lipoprotein lipase, and recent interest has centered on the determination of the
independent contribution of aerobic fitness and adiposity to lipoprotein metabolism,
particularly HDL (Krauss, 1989; Thompson, 1990; Williams, 1993). Studies of adult
athletes (Berg and Keul, 1985; Tsopanakis et al., 1986) and youth (Smith et al., 1986;
Valimaki et al., 1980) athletes have mainly examined the bivariate relationships among
aerobic'fitness, body size, and blood lipids. Few studies have examined the independent
contribution of aerobic fitness and body fatness to the blood lipid profile during
childhood and adolescence (Suter and Hawes, 1993; Tolfrey et al., 1999), and no study
has apparently examined these factors as independent determinants of blood lipids in
young endurance athletes.

This study considers the heterogeneity of blood lipids previously described in a

sample of well-trained, young distance runners in the context of two specific objectives.

108

 

First, the existence of a dose-response relationship between levels of physical activity
above the current recommended guidelines and blood lipids was evaluated in young
distance runners. Second, the relationships among peak oxygen consumption (peak V02),
subcutaneous fatness and blood lipids in young distance runners with and without
controlling for the concomitant variation of each predictor variable were examined in an
attempt to assess independent contributions of peak Vo2 and fatness to HDL. It was
hypothesized that training volume and peak Vo2 would be favorably associated with
blood lipids, and body fatness, specifically truncal fatness, would be adversely related to

blood lipids in this sample.

METHODS

Subjects. Males and females, 10 tol9 years of age, participating on mid-Michigan junior
or senior high school cross-country and track teams during the Fall 1999 and Spring 2000
were invited to participate in the current study. Subjects were also recruited by an
advertisement in the local newspaper and at local road races. Exclusion criteria included
current smokers, excessive alcohol intake, current use of blood cholesterol lowering and
anti-hypertensive medication, anabolic steroid use, hepatic, renal, and thyroid disease, or
training less than 30—40 weeks per year or the past three consecutive months. The total
number of eligible subjects in the mid-Michigan area during recruitment is difficult to
determine given the exclusion criteria. Parental consent and subject assent were obtained
prior to testing. The study was approved by the Michigan State University Committee on

Research Involving Human Subjects.

109

Blood lipids. Data collection occurred between the hours of 7 a.m.-l2:00 pm. A fasting
blood sample was obtained by fingerprick after the subject had been seated for 10
minutes. Blood was collected in a 35 micro-liter capillary tube. Upon collection,
samples were analyzed according to the manufacturer for TC, HDL, and TG within 5
minutes by a portable cholesterol analyzer (Cholestech LDX System, Hayward, CA).
Low density lipoprotein cholesterol was estimated by the Friedew ald equation
(Friedewald et al., 1972).

The total error of measurement of the Cholestech LDX analyzer has been
determined as 12.7%, 18.8%, and 19.7% (coefficients of variation, 1.4—4.1%, 3.5-5.6%,
and 4.6-5.8%) for TC, HDL, and TG, respectively. Reference laboratory measurement
errors are 8.1%,12.9%, and 5.1% for TC, HDL, and TG, respectively (Bard et al., 1997).
The total error for HDL met the standard set forth by National Cholesterol Education
Program (<22%), but TC and TG were slightly higher than the respective standards (8.9
and 15%).

Within-day reliability was determined prior to the. onset of the study by five
consecutive measurements with Cholestech Level 1 and 2 liquid control reagents. Day-
to-day reliability was determined by daily calibration prior to each testing session
throughout the study. Within-subject reliability was determined by duplicate measures of
l of every 5 male and female subjects. The coefficients of variation was used to express
the precision of the within-day and day-to—day trials and compared to national standards.
The coefficients of variation (CV) for within-day and day-to-day trials were less than 2%
and 4%, respectively for standard controls of TC, HDL, and TG. Within-subject

precision of blood lipid measurements were high (0.996, TC; 0.943, HDL; 0.970, TG).

110

 

Training volume (TV). Information regarding training practices was collected using
personal training records or a standard training invoice. An interview was conducted
when necessary to obtain or clarify information. If the subject maintained a regular
training log, they were asked for the contents for purposes of the study. Subjects who did
not maintain a regular training record were asked to complete a standard training
inventory (Appendix A), or in some cases, training history was obtained from coaches.
Training volume was estimated by averaging the reported weekly distance over the

preceding 3 months and recorded as km per week.

Peak oxygen consumption (peak V02). A maximal exercise test was conducted on a
motorized treadmill to exhaustion in an air-conditioned laboratory (20—22 degrees C,
relative humidity 45-60%). The treadmill protocol was determined by the subject's
estimated 5km race pace. Subjects walked/jogged at a speed of 3 mph and 4.5 mph for 1
min each. This initial warm-up period was followed by 4 minute stages at 6, 7.5, and 8
mph (depending on estimated %km race pace) and then an increase in grade of 2.5%

, every minute until exhaustion or test termination. Expired gases were collected for the
measurement of oxygen consumption (V02), carbon dioxide production (VCoz), and
minute ventilation. Expired gases were continually sampled and averaged every 20
seconds via the open circuit method using a metabolic cart (Gould 2900; Dayton, OH).
Expired gas volumes were measured with a flow probe anemometer and expired gas
concentrations were measured by electronic analyzers. Prior to testing, expired gas
volumes were calibrated with a 3-L syringe and gas concentrations were calibrated with

standard gases of known concentrations. Heart rate was continually monitored by pulse

111

telemetry (Polar Advantage). End of test criteria were established by volitional
exhaustion, HR 290% of age-predicted maximum, respiratory exchange ratio >10, and a
plateau in Vo2 (defined by an increase in Vo2 of <20 mlkg" min"1 with increasing
workload). Two of the latter three criteria must have been met for a subject to be

included in the analysis.

Body fatness. Skinfold thicknesses were measured by standard procedures in duplicate
with a Lange calipers as a double fold of skin underlying soft tissue at six anatomical
sites on the right side of the body (Malina, 1995). The following skinfolds were
measured to the nearest 0.5 mm: triceps, biceps, subscapular, suprailiac, abdominal, and
medial calf. If the measurements varied by more than 1 mm, additional measurements
were taken until the difference was less than 1 mm. Total subcutaneous skinfold
thickness was expressed as the sum of the six skinfolds (SSF). The individual skinfold
measurements were reproducible with intraclass correlations 20.96. The intra-observer
technical errors of measurement ranged from 0.20 mm for the biceps to 3.0 mm for the
suprailiac skinfolds. The ratio of the sum of trunk skinfolds (subscapular, suprailliac, and
abdominal) to the sum of extremity skinfolds (triceps, biceps, and medial calf) (TER) was
used an index of the relative subcutaneous fat distribution. The TER was then regressed
on SSF, and the residuals were retained to represent an index of relative subcutaneous fat

distribution independent of overall subcutaneous fatness.

Sexual maturity status. Given the difficulty in direct assessment of sexual maturity status

in a non-medical environment, a self - assessment of sexual maturity status was used in

112

the current study. Self-assessment was conducted in a separate, private station following
an explanation of the purpose of the assessment. Subjects rated their stage of sexual
development relative to sex-appropriate sets of drawings/photos and verbal descriptions
(Van Wieringen et al., 1971) based on the criteria of Tanner (1962). In a study of 174
female and 178 male Brazilian youth age 6-26 years (Matsudo and Matsudo, 1994), the
concordance between self and physician assessments of secondary sex characteristics was
reasonably high (60-7l.3%). Test-retest concordance (i.e., reproducibility) was similar

between self— and physician-assessment.

Statistical analysis. Exploratory data analysis was conducted to examine the distribution
of data and detect any outliers. In males, two outliers (+2 SD) were identified for fatness
and were not considered in .the analysis. T0 was not normally distributed and
logarithmically transformed (logTG). Descriptive statistics (means, SD, and range) were
computed for all variables. Partial correlations were calculated for each of the predictor
variables and blood lipids. Chronological age and stage of genital or breast development
were controlled in the analyses since both are associated with changes in blood lipids
during adolescence (Morrison et al., 1979; Tel], 1985). Furthermore, it is important to
control for both variables given the inter-individual differences in the timing and tempo
of sexual maturation (Malina and Bouchard, 1991). Based on preliminary findings, inter-
relationships between peak V02, SSF, and HDL were further explored. Partial
correlations were computed between HDL and (a) peak Vo2 controlling for age and SSF,

and (b) SSF, controlling for age and peak V02 to evaluate the independent contributions

113

of peak V02 and SSF to HDL. An alpha level of p<.05 was used in all analyses which

were executed with the SPSS package.

RESULTS

Tables 5.1 and 5.2 provide the descriptive statistics for chronological age, body
size, peak V02, TV, and blood lipids. Males are taller, heavier, and have less SSF but a
greater TER than females. Males also have higher values for peak Vo2 and TV, but
considerable heterogeneity exists in the sample. Some of heterogeneity in body size can
be attributed to chronological age and sexual maturation. Therefore, age and stage of
pubertal development (genital in boys, and breasts in girls) were controlled in
correlational analyses. Compared to recent United States reference medians (Hickman et
al., 1998), mean values of TC, LDL, and TG are higher in male runners. HDL is also
higher in male distance runners. In females, mean values are comparable to reference
medians. In contrast to mean values, most runners possess desirable blood lipid levels
compared to clinical cut-points (Kwiterovich, 1989).

Table 5.3 shows the partial correlations, controlling for chronological age and
stage of pubertal development, among TV, peak V02, SSF, TER, and blood lipids in
young distance runners. Overall, correlations are low to moderate. TV is weakly
correlated with blood lipids with relationships in the expected direction for HDL and TG
in males but not females. Peak Vo2 is significantly related to HDL and LDL in males
‘ (p<0.05). SSF is negatively related to HDL in both sexes (p<0.05 in males). The
correlations between SSF and LDL and TG are comparable in males and females, but the

directions of the relationship differ. The TER is not related to blood lipids in males

114

(p>0.05), with partial correlation coefficients near zero. Correlations between TER and
blood lipids are similar to those for SSF in females, with the highest coefficient between
TER and HDL (r: -0.60, p<0.05).

Figures 5.1a shows the inter-relationships among TV, peak V02, SSF, and HDL
in males. Partial correlations range from 0.10 to 0.41. The only non-significant
relationships are between TV and HDL (r=0.10, p=0.49) and TV and SSF (r=-0.l l,
p=0.56). Although TV is not directly related to HDL, it may be indirectly related through
its relationship with peak Vo2 (@032, p<0.05). Given the comparable relationships
among peak V02, SSF, and HDL (r=0.37-0.41, p<0.05), the independent contributions of
each predictor variable on HDL were further explored. When age and SSF are
controlled, the correlation between peak Vo2 and HDL remains significant (r=0.31,
p=0.04), whereas the partial correlation between SSF and HDL, controlling for age and
peak Vo2 , declines and is not significant (m-0.19, p=0.20). Similar relationships occur in
females. When age and SSF are controlled, the correlation between peak Vo2 and HDL is
0.32, whereas the partial correlation between SSF and HDL, controlling for age and peak
Vo2 , is 0.00. Figures 5.1b and 5.10 are provided to illustrate the inter—relationships

among TV, peak V02, SSF, and LDL and TG, respectively, in males.

DISCUSSION

Training volume and blood lipids. A purpose of this study was to address the role
of physical activity on blood lipids in children and adolescents who have habitual
physical activity levels equal to or greater than the current recommendations. It has been

suggested that an exercise level equivalent to jogging 10—15 miles/wk is necessary for to

115

significantly alter blood lipids (Superko, 1991; Williams, 1994). Armstrong and Simons-
Morton (1994) extrapolated this recommendation to the adolescent population and
suggested that an adolescent would need to jog at a speed of 8 km/hr for approximately 2
hours per week, which from their own experience is equivalent to about 80% of maximal
heart rate with young adolescents and about 75% of maximal heart rate for young adults.
The authors, therefore, recommended that four 30 minute exercise sessions per week at
75-80% of maximum heart rate may be an appropriate prescription. Since few
adolescents engage in such activity, the use of a special exposure group such as young
distance runners would allow for the testing of these recommendations. Therefore, young
distance runners were studied to determine if a dose-response or threshold effect of
physical activity on blood lipids exists in children and adolescents with habitual physical
activity levels greater than the recommended exercise volume.

It was hypothesized that training volume would be positively related to a
favorable blood lipid profile among adolescent distance runners as in adult endurance
athletes. Correlations between TV and HDL and TG were favorable in males, but low.
However, TV may indirectly influence HDL in males through its relationship with peak
Vo2 (1:032) and to a lesser extent body fatness (Figure 5.1). These inter-relationships
suggest the importance in considering the confounding effects on body fatness and peak
Vo2 when assessing the influence of TV on HDL. It is also possible that leaner and more
aerobically fit individuals are more likely to engage in higher training levels (Williams et
al., 1982).

In an early report on 90 middle-aged male runners, distance run per week was

positively correlated with HDL (r=0.50) and remained significant after adjustment for

116

percentage body fat (r=0.40) (Rotkis et al., 1982). Recently, Williams (1996, 1997)
demonstrated that the benefits of exercise continued to accrue in a dose-response manner
at levels of physical activity exceeding the current minimal guideline in large samples of
non-smoking, recreational male (11: 8283) and female (n=l837) adult distance runners.
Significant linear trends were reported for HDL and TC:HDL in both sexes and TG in
men. No significant trend was reported for LDL. In a smaller sample of 33 middle-aged
men (mean age = 45 yrs), Williams (1990) reported no relationship between TV and
HDL (r=0.05). The results of the current study of young distance runners may thus be
due to sample size.

An alternative explanation may be that increased levels of training do not
influence lipoprotein metabolism when blood lipids are already at desirable levels during
childhood and adolescence. Cross-sectional studies of habitual physical activity and
blood lipids generally show low associations (Armstrong and Simons—Morton, 1994; .
Tolfrey etal., 2000). In contrast to cross-sectional studies, prospective studies suggest
that pre-training levels of blood lipids influence the response to exercise training (Tolfrey
et al., 2000). Likewise, Barr et al. (1991) found that increasing TV in collegiate male
swimmers from 22,000 m/wk to 44,000 m/wk over a six week period did not alter HDL
or TG, and lowered LDL only slightly relative to baseline levels. It is possible that when
blood lipids are already desirable in youth and/or physical activity levels are relatively
high, increased levels of training do not influence their metabolism. This observation has
been referred to as a 'ceiling' or 'fioor' effect (Tolfrey et al., 2000). To explore this
hypothesis, the sample was divided into sub-groups based on clinical cut-points. The cut-

points were modified since the number of subjects with blood lipid levels greater than the

117

clinical cut-point of dyslipidemia was limited. Therefore, individual values classified as
borderline dyslipidemic or dyslipidemic were grouped together as "undesirable" (HDL <
45 mgldl). Results indicated that HDL was positively related to TV (r=0.40) in the
undesirable group, but was unrelated and in the opposite direction as expected in the
desirable group (HDL 2 45 mgldl, r: -0.23). Perhaps, the reason that LDL or TG values
were not modulated by levels according to the clinical cut—point is that the values were
not excessively dyslipidemic compared to the HDL values. In other words, low levels of
HDL may be more sensitive to exercise training than moderately elevated levels of TG
and LDL. This preliminary finding suggests that relationship between TV and HDL may
be modulated by level according to clinical cut-point. Additional cross-sectional and
prospective training studies are required to further examine this suggestion.

Other factors that were not considered in the present study, but may influence the
relationship between TV and. blood lipids, include: the intensity of exercise training
(Williams, 1998), genotype and genotype-environmental interactions (Taimela et al.,
1996), dietary intake and composition (Brown and Cox, 1998; Leddy et al., 1997;
Lukaski et al., 1984; Thompson et al., 1984). Additional factors that are known to
influence blood lipids were controlled in the design of this study. No subject reported
regular alcohol intake, smoking, anabolic steroid use, nor medication or metabolic
disorders (e.g., liver, kidney, or thyroid disease), which may adversely influence
lipoprotein metabolism.

The estimation of TV may also influence the results. TV was determined as the
average distance run per week in the preceding 3 months. In the National Runners'

Health Study, TV was estimated by averaging yearly distances for the preceding 5 years

118

and the reliability was 0.89 (W illiarns, 1997). The intensity of exercise training was not
addressed in the current study and may influence the results. Future studies should
establish the reliability and validity of self-reported training volume in young athletes,
while the continuous measurement of heart rate during training sessions would provide
valuable information with regards to the role of exercise intensity and training volume on
blood lipids in endurance athletes.

Peak V02 and blood lipids. The positive association between peak Vo2 and HDL
is consistent with previous studies of youth (Al-Hazzaa et al., 1994; Armstrong et al.,
1991; Macek et al., 1989; Sallis et al., 1988; Suter and Hawes, 1993; Tell and Vellar,
I988; Valimaki et al., 1980). The association between peak Vo2 and I-IDL(=0.41) is
similar to previous studies of well-trained athletes as well. In a previous mixed-sample
(males and females) of mid-Michigan distance runners, peak Vo2 was related to HDL
(1:039) (Smith et al., 1986). In a study of national level adult athletes, peak Vo2
explained 25% of the variance in HDL (Berg and Keul, 1985) and was significantly
related to HDL(1=0.26) in Olympic athletes (Tsopanakis et al., 1986). The proposed
mechanism for this relationship involves properties of skeletal muscle and adipose tissue
that influence peak Vo2 and lipid metabolism (Tikkanen et al., 1991). Skeletal muscle
and adipose tissue lipoprotein lipase (LPL) activity is higher in trained versus untrained
individuals (Nikkila et al., 1978), and the percentage of slow oxidative muscle fibers is
positively correlated with HDL (Tikkanen et al., 1991). In combination, a higher
proportion of slow oxidative fibers favors fatty acid metabolism and increases the

likelihood that LPL activity will be increased with exercise (Stefanik and Wood, 1994).

119

The role of genes and aerobic fitness levels in the modulation of blood lipid levels
has also been indicated (Katzmarzyk et al., 1999a; St.-Amand et al., 1999). Heritability
estimates for peak Vo2 per unit kg approximate 25-40% (Bouchard et al., 1997).
Polymorphisms in lipoprotein genes (apolipoprotein B) may also influence the
relationship between aerobic fitness and blood lipid levels (St.-Amand et al., 1999). It is
possible that blood lipids in adolescent distance runners are moderated by the genetic
contribution of aerobic fitness or the pleiotropy (shared genes) of aerobic fitness and
blood lipids.

The positive relationship between peak Vo2 and LDL is a unique finding that
lacks an explanation. Although other studies of youth found a similar positive
relationship, the relationship was not as strong (Suter and Hawes, 1993; Tolfrey et al.,
1999).

Body fatness and blood lipids. The relationships between SSF and HDL and TG
are consistent with the literature, but once again the relationship with LDL lacks an
explanation. In general, indicators of body fatness are positively related to TG, TC, and
LDL, and negatively to HDL throughout the lifespan (Guo et al., 1994). No previous
study of young male athletes has examined the relationship of body size or adiposity with
blood lipids. Relative body weight (kg/(cm—100)) explained about 7% of the variance in
TC and TG, and 20% of the variance in LDL in national level adult athletes (sprinters,
hammer throwers, distance runners, etc.) (Berg and Keul, 1985). In Olympic athletes,
relative body weight was significantly related to HDL (r: -0.22), LDL(1=0.18), and
VLDL (r=0.17) (Tsopanakis et al., 1986). Rotkis et al. (1982) found significant

relationships between percentage body fat and HDL(1= -0.36), TC (r=0.38) and non-

120

HDL cholesterol (r=0.48) in middle-aged distance runners. In contrast, correlations
between various measures of body size (e.g., BMI, relative weight, percentage body fat)
. and HDL were low in 33 middle-aged distance runners (r =0.05- 0.08) (Williams, 1990).
The relationship between body fatness and HDL can be partly explained by adipose
tissue lipoprotein lipase (LPL) activity (Nikkila et al., 1978),

The findings for TER in males are inconsistent with previous findings in youth
that show that a central fat patterning, or truncal/visceral fatness, is associated with
adverse blood lipids (Baumgartner et al., 1989; Freedman et al., 1989; van Lenthe et al.,
1998). During adolescence, an increase in subcutaneous abdominal adipose tissue and
redistribution of body fatness to the trunk results in an increase in the TER in boys
(Malina et al., 1999). The relationships between TER and blood lipids were actually more
pronounced in females, especially for HDL. The use of the TER as an appropriate index
of relative subcutaneous fat distribution in young distance runners has not been
considered. Since young distance runners possess a low level of whole-body fatness and
a relatively low level of periperal' (i.e., extremity) fatness, even a slightly greater sum of
trunk skinfolds results in a higher TER. The relatively low amount of peripheral fatness
probably reflects morphological properties necessary for success in endurance sport. The
relationship could also be attributed to the independent effects of somatotype, namely
ectomorphy, on blood lipids (Katzmarzyk et al., 1999b). Similarly, a low correlation (1:-
0.13) between the ratio of abdominal girth and bi-iliac diameter and HDL in middle-aged
male distance runners (Williams, 1990).

The role of genes and body fatness in the modulation of elevated blood lipid

levels has also been indicated (Katzmarzyk et al., 1999c). Heritability estimates of

121

abdominal visceral fatness measured by computerized tomography approximate 50-55%
with a major gene associated with total fat mass either directly or indirectly affecting
abdominal fatness. Furthermore, polymorhpisms in lipoprotein genes (ApoA-II Mspl,
HindIII, and apoB-100 EcoRI) may influence the relationship between abdominal
visceral fatness and blood lipid levels. The correlation between parental BMI and
measures of fatness in adolescent distance runners were positive and ranged from 0.01
(paternal BMI and TER) to 0.36 (maternal BMI and SSF). Parental BMI was calculated
from self-reported heights and weights of the parents. Correlations are stronger between
maternal BMI and offspring fatness. A possible cross-trait familial resemblance for body
fat and bloods lipids has also been hypothesized (Perusse et al., 1997). This hypothesis
suggests that trait I in a parent (e.g., body fat) is linked with trait 2 in an offspring (e.g.,
blood lipids) and provides an indication about the contribution of shared genes and/or .
environmental factors. In the present study, low correlations (r=0.12-0.23) existed
between paternal BMI and offspring blood lipids. Therefore, it is possible that blood
lipids in adolescent distance runners are moderated by the genetic contribution of body
fatness and/or the pleiotropy (shared genes) of body fatness and blood lipids. It is also
possible that shared environmental factors (i.e., dietary intake) could contribute to the
relationship between parental fatness, offspring fatness and blood lipids.

This is apparently the first study to examine multiple determinants of blood lipids
in adolescent distance runners. Previous reports examining the univariate relationships
between relative body size, peak Vo2 and blood lipid levels have been discussed.
However, the unique contribution of peak Vo2 and adiposity were not established in these

studies. One study of male master level athletes using multivariate linear regression

122

showed that percentage body fat explained 29 and 41% of the variance in HDL and TG,
and peak Vo2 accounted for an increase of 6 and 2% of the variance in HDL and TG,
respectively (Yataco et al., 1997). In the present study, partial correlations were
computed to separate the independent contributions of peak Vo2 and SSF on HDL.
Results indicate that the association between peak Vo2 and HDL remained significant
after controlling for the concomitant variation in SSF and explained 9% of the variance in
HDL. The association between SSF and HDL did not remain significant after controlling
for the concomitant variation in peak V02. This finding would suggest that skeletal
muscle properties are an important factor in determining HDL in young well-trained
distance runners, although the influence of adipose tissue cannot be dismissed.
Conclusions. In conclusion, it appears that health benefits (i.e., blood lipid levels)
do not accrue in young distance runners at levels exceeding the current minimal
guidelines. However, this result may be due to the small sample size that lacks the
statistical power to test for incremental changes across a range of high levels of training
(Williams, 1997). In general, this study contributes to understanding the heterogeneity of
blood lipids in young distance runners. The inter-relationships among TV, peak Voz,
body fatness, and blood lipids appear to be complex. Although TV was not directly
related to HDL, it may be related through its association with peak Vo2 and body fatness.
The relationship between TV and HDL may also be modulated by the level of HDL
according to clinical cut-points. Finally, peak Vo2 is an important predictor of HDL in
young distance runners independent of body fatness. It is possible that the inter-
relationships peak Vo2, body fatness'and HDL is mediated through shared genes.

Further study is warranted to explore the contribution of growth, maturation, genetics,

123

 

exercise training, and skeletal muscle and adipose tissue properties on lipoprotein

metabolism during adolescence.

124

 

ACKNOWLEDGEMENTS
This study was supported in part by the William Wohlgamuth Fellowship for the
Study of Youth Sports and the Institute for the Study of Youth Sports. Special thanks is

given to members of the Human Energy Research Laboratory who assisted in data

collection.

125

REFERENCES

Al-Hazzaa HM, Sulaiman MA, Al-Matar AJ, Al-Mobaireek KF (1994) Cardiorespiratory
fitness, physical activity patterns and coronary risk factors in preadolescent boys. Int J
Sports Med 15:267-272.

Armstrong N, Simons-Morton B (1994) Physical activity and blood lipids in adolescents.
Pediatr Exerc Sci 6:381-405.

Armstrong N, Williams J, Balding J, Gentle P, Kirby B (1991) Cardiorespiratory fitness,
physical activity patterns, and selected coronary artery risk factor variables in 11- to 16-
year-olds. Pediatr Exerc Sci 32219-228.

Bard RL, Kaminsky LA, Whaley MH, Zajakowski S (1997) Evaluation of lipid profile
measurements obtained from the Cholestech LDX analyzer. J Cardiopul Rehab 17:413-
418.

Barr SI, Costill DL, Fink W], Thomas R (1991) Effect of increased training volume on
blood lipids and lipoproteins in male collegiate swimmers. Med Sci Sports Exerc
23:795-800.

Baumgartner RN, Siervogel RM, Chumlea WC, Roche AF (1989) Associations between
plasma lipoprotein cholesterols, adiposity and adipose tissue distribution during
adolescence. Int J Obes 13:31-41.

Berg A, Keul J (1985) Influence of maximum aerobic capacity and relative body weight
on the lipoprotein profile in athletes. Atherosclerosis 55:225-231.

Bouchard C, Malina RM, Peruss'e L (1997) Genetics of Fitness and Physical
Perforrnance.Champaign, IL: Human Kinetics.

Brown RC, Cox CM (1998) Effects of high fat versus high carbohydrate diets on plasma
lipids and lipoproteins in endurance athletes. Med Sci Sports Exerc 30: 1677-1683.

Casperson CJ, Nixon PA, DuRant RH (1998) Physical activity epidemiology applied to
children and adolescents. Exerc Sci Sports Rev 26:341-403.

Despres J-P (1997) Visceral obesity, insulin resistance, and dyslipidemia: contribution of
endurance exercise training to the treatment of the plurimetabolic syndrome. Exerc Sport
Sci Rev 25:271-300.

Despres J-P, Bouchard C, Malina RM (1990) Physical activity and coronary heart disease
risk factors during childhood and adolescents. Exerc Sport Sci Rev 18:243-261.

126

Freedman DS, Srinivasan SR, Harsha DW, Webber LS, Berenson GS (1989) Relation of
body fat patterning to lipid and lipoprotein concentrations in childem and adolescents: the
Bogolusa Heart Study. Am J Clin Nutr 50:930-939.

Friedewald WT, Levy RI, Fredrickson DS (1972) Estimation of the concentration of low
density lipOprotein cholesterol in plasma without use of the preparative ultracentrifuge.
Clin Chem 18:499-502.

Guo S, Salisbury S, Roche AF, Chumlea WC, Siervogel RM (1994) Cardiovascular
disease risk factors and body composition: a review. Nutr Rev 14:1721—1777.

Gutin B, Owens S (1996) Is there a scientific rationale supporting the value of exercise
for the present and future cardiovascular health of children? the pro argument. Pediatr
Exerc Sci 8:294-302.

Hickman TB, Briefel RR, Carroll MD, Rifkind BM, Cleeman JI, Maurer KR, Johnson
CL (1998) Distributions and trends of serum lipid levels among United States children
and adolescents ages 4—19 years: data from the Third National Health and Nutrition
Examination Survey. Prev Med 27:879-890. -

Katzrnarzyk PT, Malina RM, Bouchard C (1999a) Physical activity, physical fitness, and
coronary heart disease risk factors in youth: The Quebec Family Study. Prev Med
29:555-562.

Katzmarzyk PT, Malina RM, Song TMK, Bouchard C (1999b) Physique, subcutaneous
fat, adipose tissue distribution, and risk factors in the Quebec Family Study. Int J Obes
23:476-484.

Katzmarzyk PT, Pérusse L, Bouchard C (1999c) Genetics of abdominal visceral fat
levels. Am J Hum Biol 11:225-235.

Krauss RM (1989) Exercise, lipoproteins, and coronary artery disease. Circulation
79: 1 143-1 145.

Kwiterovich PO (1989) Beyond Cholesterol: the John Hopkins Complete Guide for
Avoiding Heart DiseaseBaltimore, MD: The John Hopkins Press.

Leddy J, Horvath P, Rowland J, Pendergast D (1997) Effect of a high or a low fat diet on
cardiovascular risk factors in male and female runners. Med Sci Sports Exerc 29:17-25.

Lukaski HC, Bolonchuck WW, Klevay LM, Mahalko JR, Milne DB, Sandstead HH
(1984) Influence of type and amount of dietary lipid on plasma lipid concentrations in
endurance athletes. Am J Clin Nutr 39:35-44.

127

run-r:— as:
'I

Macek M, Bell D, Rutenfranz J, Vavra J, Masopust J , Neidhart B, Schmidt K-H (1989) A
comparison of coronary risk factors in groups of trained and untrained adolescents. Eur J
Appl Physiol 58:577-582.

Malina RM (1990) Growth, exercise, fitness, and later outcomes. In C Bouchard, RJ
Shephard, T Stephens, JR Sutton and BD McPherson (eds.): Exercise, Fitness, and
Health: A Consensus of Current Knowledge. Champaign, IL: Human Kinetics, pp. 637-
653.

 

Malina RM (1995) Anthropometry. In PJ Maud and C Foster (eds.): Physiological
Assessment of Human Fitness. Champaign, IL: Human Kinetics, pp. 205-219.

Malina RM, Bouchard C (1991) Growth, Maturation, and Physical Activity.Charnpaign,
IL: Human Kinetics.

Malina RM, Koziel S, Bielicki T (1999) Variation in subcutaneous adipose tissue
distribution associated with age, sex, and maturation. Am J Hum Biol 11:189-200.

Matsudo SMM, Matsudo VKR (1994) Self-assessment and physician assessment of

sexual maturation in Brazilian boys and girls: concordance and reproducibility. Am J
Hum Biol 62451-455.

Morrison JA, Laskarzewski PM, Rauh JL, Brookman R, Mellies M, Frazer M, Khoury P,
deGroot 1, Kelly K, Glueck CJ (1979) lipids, lipoproteins, and sexual maturation during
adolescence: the Princeton Maturation Study. Metabolism 28:641-649.

Nikkila EA, Taskinen M-R, Rehunen S, Harkcnen M (1978) Lipoprotein lipase activity
in adipose tissue and skeletal muscle of runners: relation to serum lipoproteins.
Metabolism 27: 1661-1671.

Pate RR, Pratt M, Blair SN, Haskell WL, Macera CA, Bouchard C, Buchner D, Ettinger
W, Heath GW, King AC, Kriska A, Leon AS, Marcus BH, Morris J, Paffenbarger RS,
Patrick K, Pollock ML, Rippe JM, Sallis J, Wilmore JH (1995) Physical activity and
health: A recommendation from the Centers of Disease Control and Prevention and the
American College of Sports Medicine. JAMA 273:402—407.

Pérusse L, Rice T, Despres JP, Rao DC, Bouchard C (1997) Cross-trait familial
resemblance for body fat and blood lipids: familial correlations in the Quebec Family
Study. Arterioscler Throm Vasc Biol 17:3270-327'7.

Riopel DA, Boerth RC, Coates TJ, Hennekens CH, Miller WM, Weidman WH (1986)
Coronary risk factor modification in children: exercise - A statement for physicians by
the Committee on Atherosclerosis and Hypertension in Childhood of the Council on

Cardiovascular Disease in the Young, American Heart Association. Circulation
74: 1 189A-1 191A.

128

Rotkis TC, Cote R, Coyle E, Wilmore JH (1982) Relationship between high density
lipoprotein cholesterol and weekly waning mileage. J Cardiac Rehab 2:109-112.

Sallis JF, Patrick K, Long BJ (1994) Overview of the International Consensus
Conference on Physical Activity Guidelines for Adolescents. Pediatr Exerc Sci 6:299-
301.

Sallis JF, Patterson TL, Buono MJ, Nader PR (1988) Relation of cardiovascular fitness
and physical activity to cardiovascular disease isk factors in children and adults. Am J
Epidemiol 1272933—941.

Smith BW, Metheny WP, Sparrow AW (1986) Serum lipid and lipoprotein profiles in
age-group runners. In MR Weiss and D Gould (eds.): Sport for Children and Youths.
Champaign, IL: Human Kinetics, pp. 269-273.

St.-Amand J, Prud'homme D, Moorjani S, Nadeau A, Tremblay A, Bouchard C, Lupien
PJ, Després J-P (1999) Apolipoprotein E polymorphism and the relationships of physical
fitness to plasma lipoprotein-lipid levels in men and women. Med Sci Sports Exerc

3 1:692-697.

Stefanik ML, Wood PD (1994) Physical activity, lipid and lipoprotein metabolism, and
lipid transport. In C Bouchard, S R]. and T Stephens (eds.): Physical Activity, Fitness,
and Health. Champaign, IL: Human Kinetics, pp. 417-431.

Superko HR (1991) Exercise training, serum lipids, and lipoprotein particle: is there a
change threshold? Med Sci Sports Exerc 232677-685.

Suter E, Hawes MR (1993) Relationship of physical activity, body fat, diet, and blood
lipid profile in youths 10-15 yr. Med Sci Sports Exerc 25:748-754.

Taimela S, Lehtimaki T, Porkka KV, Rasanen L, Viikari J S (1996) The effect of physical
activity on serum total and low-density lipoprotein cholesterol concentrations varies with
apolipoprotein E phenotype in male children and young adults: The Cardiovascular Risk

in. Young Finns Study. Metabolism 45:797-803.

Tanner JM (1962) Growth at Adolescence, 2nd edition. Oxford: Blackwell.

Tell GS (1985) Cardiovascular disease risk factors related to sexual maturation: the Oslo
Youth Study. J Chron Dis 382633-642.

Tell GS, Vellar OD ( 1988) Physical fitness, physical activity, and cardiovascular disease
risk factors in adolescents: The Oslo Youth Study. Prev Med 17:12-24.

129

Thompson PD, Cullinane EM, Eshleman R, Kantor MA, Herbert PN (1984) The effects
of high-carbohydrate and high-fat diets on the serum lipid and lipoprotein concentrations
of endurance athletes. Am J Clin Nutr 33: 1003-1010.

Tikkanen HO, Harkcnen M, Naveri H, Hamalainen E, Elovainio R, Sarna S, Heikki Frick
M (1991) Relationship of skeletal muscle fiber type to serum high density lipoprotein
cholesterol and apolipoprotein A-I levels. Atherosclerosis 90:49-57.

Tolfrey K, Campbell IG, Jones AM (1999) Selected predictor variables and the lipid-
lipoprotein profile of prepubertal girls and boys. Med Sci Sports Exerc 31: 1550-1557.

Tolfrey K, Jones AM, Campbell IG (2000) The effects of aerobic exercise training on the
lipid—lipoprotein profile of children and adolescents. Sports Med 29:99-112.

Tsopanakis C, Kotsarellis D, Tsopanakis AD (1986) Lipoprotein and lipid profiles of
elite athletes in Olympic sports. Int J Sports Med 7:316—321.

Valimaki I, Hursti M-L, Pihlakoski L, Viikari J (1980) Exercise performance and serum
lipids inrelation to physical activity in schoolchildren. Int J Sports Med 1:132-136.

van Lenthe FJ, van Mechelen W, Kemper HCG, Twisk JWR (1998) Association of a
central pattern of body fat with blood pressure and lipoproteins from adolescence into
adulthood. Am J Epidemiol 147:686-693.

Van Wieringen JC, Wafelbakker F, Verbrugge HP, de Haas JH (1971) Growth Diagrams
1965 Netherlands: Second National Survey on 0-2A—year-oldsGroningen: Wolters-
Noorhoff Publishing.

Williams PT (1990) Weight set-point theory predicts HDL-cholesterol levels in
previously obese long-distance runners. Int J Obes 14:421-427.

Williams PT (1993) High-density lipoproteins and lipase activity in runners.
Atherosclerosis 98:25] -254.

Williams PT (1994) Lipoproteins and adiposity show improvement at substantially higher
exercise levels than those currently recommended. Circulation 90:1-471 (Abstract).

Williams PT (1996) Hi gh-density lipoprotein cholesterol and other risk factors for
coronary heart disease in female runners. N Engl J Med 334:1298-1303.

Williams PT (1997) Relationship of distance run per week to coronary heart disease risk
factors in 8283 male runners. Arch Intern Med 157:191-198.

Williams PT (1998) Relationships of heart disease risk factors to exercise quantity and
intensity. Arch Intern Med 158:237-245.

130

Williams PT, Wood PD, Haskell WL, Vranizan K (1982) The effects of running mileage
and duration on plasma lipoprotein levels. JAMA 247:2674-2679.

Yataco A, Busby-Whitehead J, Drinkwater D, Katzel L (1997) Relationship of body

composition and cardiovascular fitness to lipoprotein lipid profiles in master athletes and
sedentary men. Aging 9:88-94.

I31

Table 5.1. Characteristics youngdistance runners.

C

812411.812.)
Age (yrs) 16.9 (2.0)
Ht (cm) 173.9 (7.3)
Wt (kg) 61.7 (7.4)
SSF (mm) 40.7 (7.6)
SUM 3T (mm) 24.8 (5.8)
SUM 3E (mm) 15.8 (3.4)
TER (mm/mm) 1.60 (0.37)
Peak Vo2 4225.6 (597.2)
(ml'min")
Peak Vo2 66.9 (5.8)
(nd'kg"'min")
TV (km'wk") 47.7 (22.8)

n=4

Bans:
10.4—19.4

1492-1851
39.9-76.3
25.0-55.5
14.0-40.0
8.0-27 .0
0.70-2.48

2262-5526

54.8-77.0

15-88

females ( n=22)
Meanfill) m
15.5 (2.6) 9.6- 18.4
161.1 (9.8) 132.7-178.1
50.1 (10.4) 27.0-71.1
58.3 (17.7) 32.5-99.0
29.5 (9.4) 14.5-50.0
18.8 (8.8) 16.0-49.0
1.05 (0.13) 0.85-1.29
2869.8 (540.3) 1709-3725
56.8 (5.3) 48.0-63.3
35.2 (13.8) 15-60

 

See text for abbreviations.

132

Table 5.2. Blood lipids of youngdistance runners.

 

Males (n=45)
Mean (SD) Range
TC 167.2 (30.2) 100-241
HDL 48.6 (13.7) 28-89 '
LDL 102.9 (26.4) 55-174
TG $.2 (38.8) 45-188

Females (n=22)

Me_a_n_l.S.D.1
157.6 (18.0)

51.7 (11.1)
91.0 (17.0)
75.1 (21.8)

m
117-188
35-78
58-131
50-149

 

See text for abbreviations.
Values expressed in mgldl.

133

Table 5.3. Partial correlations, controlling for chronological age and pubertal stage, for
traimflvolumemeak V02, body fatness, and blood lipids in youngdistance runners.

 

 

 

Males

‘ Blood Lipids TV Peak V02 SSF TER
HDL .10 .41“ -34* -.01
LDL .27 .36* -. 19 .07
logTG -.l l .01 .22 -.04
Females
Blood Lipids TV Peak V02 SSF TER
HDL -.27 .41 -.27 -.60*
LDL .17 -.15 .18 -.18
logTG .09 .30 -.21 . l6

 

TV, training volume; peak V02, peak oxygen consumption; SSF, sum of six skinfolds;

TER, trunk-to-extremity ratio.

*p<0.05

134

 

 

 

 

 

 

 

 

 

 

 

 

O. 10
-0.1 1 032
HDL
-0.34‘ 041'
SSF -0.37' Peak V02

 

 

 

 

 

 

Figure 5.1a. Inter-relationships among training volume (TV), sum of six skinfolds

(SSF), peak oxygen consumption (peak V0,), and high-density lipoprotein (HDL) in
45 young male distance runners 10 to 19 years of age. Values are partial correlation
coefficients controlling for age and genital stage of pubertal development. ° p<0.05.

135

 

 

 

 

 

 

 

 

 

 

0.27
.011 1 032
LDL
-019 0.36'
SSF 037‘ Peak Va,

 

 

 

 

 

 

 

 

Figure 5.1b. Inter-relationships among training volume (TV), sum of six skinfolds
(SSF), peak oxygen consumption (peak V0,), and low-density lipoprotein (LDL) in 45
young male distance runners 10 to 19 years of age. Values are partial correlation
coefficients controlling for age and genital stage of pubertal development. ' p<0.05.

136

 

 

 

 

 

 

 

 

 

 

.0.1 1
-0.1 1
TO
0.22
[ SSF -0.37‘
J

 

 

 

 

 

Peak v0,

 

Figure 5.1c. Inter-relationships among training volume (TV), sum of six skinfolds

(SSF), peak oxygen consumption (peak V0,), and triglycerides (TG) in 45 young male

distance runners 10 to 19 years of age. Values are partial correlation coefficients
controlling for age and genital stage of pubertal development. ' p<0.05.

137

 

CHAPTER6

REVIEW OF LITERATURE

PART II

138

INTRODUCTION

Peak oxygen consumption (V02) can be defined as the maximal ability of the
g organism to uptake (via pulmonary respiration), deliver (via the cardiovascular system),
and utilize (via the oxidative capacity of skeletal muscle) oxygen. Together, these
processes govern the flow of oxygen from ambient air to the mitochondria through a
series of structural barriers that can be considered the oxygen transport system (Weibel,
1984). Peak Vo2 can be easily measured by a progressive, incremental exercise test that
commences when additional power output fails to elicit any further increase in whole-
body V02.

The importance of peak Vo2 can be viewed from auxological, physical
performance, and health perspectives. From an auxological viewpoint, growth-related
changes in peak Vo2 may serve to indicate the structural (or quantitative) and functional
(or qualitative) changes in the oxygen transport system. In terms of physical performance
and health, peak Vo2 is related to endurance performance (Coyle, 1995) and various
chronic diseases (Blair et al., 1989), at least in adults.

Despite the relative ease of measurement of peak V02, there is some controversy
in the analysis and interpretation of growth-related changes in physiological capacity and
performance. To provide a clear understanding of the growth-related changes in peak
V02, the confounding factor of growth-related changes in body dimensions must be
appropriately controlled. The use of allometric scaling has recently received considerable
attention within the field of pediatric exercise science (Armstrong and Welsman, 1994;
Winter, 1996). The use of correctly scaled data has important implications regarding the

understanding of growth-related changes in peak V02, physiological differences between

139

adults and children, evaluation of youth athletes, and health-related fitness during
childhood and adolescence.

Age-, sex-, and maturity-associated variation in absolute peak Vo2 (L'min") and
peak Vo2 expressed per kg of body weight (ml'kg"'min") in the general population is
briefly reviewed. Extensive reviews have been published elsewhere (Armstrong and
Welsman, 1994; Krahenbuhl et al., 1985; Malina and Bouchard, 1991; Rowland, 1996).
Lirrrited information on the age-, sex-, and maturity-associated variation in peak Vo2 of
young athletes is also critically examined. The basic principles of allometric scaling are
presented to provide the reader with background of this concept prior to reviewing the

growth-related changes in peak Vo2 from an allometric perspective.

AGE-AND SEX-ASSOCIATED VARIATION IN PEAK Vo2

Age- and sex-associated variation in peak Vo2 are considered together since sex
differences are apparent during the transition into puberty. Absolute
peak Vo2 generally increases with chronological age during the first two decades of life
(Armstrong and Welsman, 1994; Krahenbuhl et al., 1985). The correlation between
chronological age and absolute peak Vo2 is moderately high from 8-16 yrs of age, r=0.75
and 0.53 in boys and girls, respectively (Armstrong and Welsman, 1994). The difference
in the correlations between sexes can be attributed to the leveling 'of absolute peak Vo2 in
females during puberty and continued increase in boys. Data compiled by Krahenbuhl et
al. (1985) indicate that average values for peak Vo2 increase from about 1.0 L'min" at age
6 yrs to about 2.0 Lmin" at 12 yrs; thereafter, sex differences become more apparent. By

the age of 15 yrs, peak Vo2 is 2.8 L'min" in boys and 2.0 L'min'l in girls. Between 15 and

140

18 yrs of age, peak Vo2 generally increases in boys and plateaus in girls. Mirwald and
Bailey (1986) reported an average yearly increase of 11% in 8-16 yr old boys (n=75) and
8-13 yr old girls (n=22) followed longitudinally. The largest absolute increases occur
between 13 and 14 yrs (332 admin") in boys and 11 and 12 yrs (271 ml'min") in girls.

When expressed per kg of body weight, peak Vo2 remains stable in boys and
declines in girls between 6-18 yrs (Armstrong and Welsman, 1994; Krahenbuhl et al.,
1985). The average value in boys is approximately 52 ml'kg"'min". The average value
for an 8 yr old girl (50 nrl'kg"'min") is similar to that of an 8 yr old boy, but declines to
45 1'rrl'kg"'min‘l by age 12 yrs and 40 ml'kg"'rrrin" by age 16 yrs.

Why does absolute peak Vo2 increase with chronological age? Why do sex
differences emerge, particularly during puberty? Reasons for the age- and sex-associated
variation in peak Vo2 have not been as extensively investigated. Peak Vo2 is ultimately
determined by the flow of oxygen from the ambient air across a series of structural
resistors, which include: ventilation, pulmonary gas diffusion across the alveolar-
capillary membrane, the binding of oxygen to hemoglobin, cardiac and circulatory
function, gas diffusion across the capillary-myocyte barrier, oxidative metabolism within
skeletal. muscle, and finally, skeletal muscle contraction. Structural and/or functional
changes or differences at any of these sites could explain the age- and sex-associated
variation in peak V02. Although age-associated changes in physiological function with
increasing body size and organ system development are evident, the specific adaptations
remain poorly understood. Rowland (1990) addresses several important questions related
to this dilemma: (I) Do age-related changes in peak Vo2 occur as a continuum throughout

the pediatric years, or only at critical points (such as puberty)? (2) How great is the inter-

141

individual variability in the rate of growth-related changes in peak V02? (3) How
important is the influence of physical activity or exercise training on growth-related
changes in peak Vo,?, (4) How significant are the contributions of functional changes
compared to changes in body size? The last question is very difficult to address given
ethical and methodological constraints in pediatric research.

Regardless of limitations in research design, the available information indicates
that body size, specifically fat-free mass, accounts for a major portion of the variance in
peak Vo2 during childhood and adolescence (Eisenmann and Malina, in press; Rowland,
1996). Rowland (1996) suggests that because relative peak Vo2 (nrl'kg“‘rr1in") remains
stable in boys during the pediatric years, the increase in absolute peak Vo2 during this
same time period can be explained solely on the basis of dimensional (quantitative)
changes in the oxygen transport system and peripheral muscle mass, without size-
dependent functional (qualitative) changes. This conclusion is also based on the
comparative analysis of the increase in absolute peak V02, pulmonary (lung weight and
vital capacity), and cardiac (left ventricular volume) parameters; all show about a 50%
increase between 8 and 16 yrs. This hypothesis is supported by the early study of
Asmussen and Heeboll-Nielson (1955). Although the authors hypothesized that both
quantitative and qualitative changes occur during growth, the actual exponent for peak
Vo2 was very close to being proportional to the third power of the linear dimension
(stature) or body mass (see section on allometric scaling). However, longitudinal studies
that examine both structural and functional parameters of the oxygen transport system

areneeded.

142

W?Dl“ -

Sex differences in peak Vo2 throughout childhood and adolescence have been
primarily explained by differences in body composition (Armstrong and Welsman, 1994;
Rowland, 1996). When peak Vozis expressed per kg of fat-free mass, the sex difference
is greatly reduced. However, a small difference still cannot be accounted; therefore other
biological or socio-environmental factors apparently contribute to sex-associated
variation in the phenotypic expression of peak V02. During adolescence, hemoglobin
concentration and habitual physical activity may also contribute to the sex difference in
oxygen transport capacity. Although physical activity levels during adolescence decline
moreso in females, Armstrong and Welsman (1994) argue that children and adolescents

rarely experience levels of physical activity necessary to alter peak V02.

MATURITY-ASSOCIATED VARIATION IN PEAK V0,

Variation in a biological variable can be considered relative to maturity status
(skeletal age, pubertal stage), or relative to the timing of a given pubertal event (peak
height velocity [PHV]), age at menarche). Subjects can also be grouped by timing, i.e.,
early, average, or late maturing. Using PHV requires longitudinal data throughout the
adolescent growth spurt, whereas the other two approaches can be obtained by either a
cross-sectional or longitudinal observations. Relatively limited information is available
on the maturity-associated variation in peak Voz. In general, skeletal age and absolute
peak Vo2 are highly correlated (r=0.89) over a broad age range, 8-18 yrs, but are lower
(r=0.40-0.60) within narrower age ranges. In contrast, relative peak Vo2 is not

significantly related to skeletal age (Malina and Bouchard, 1991).

143

Although some studies have used secondary sex characteristics to group subjects
as “prepubertal”, “pubertal”, and/or “post-pubertal”, few have considered peak Vo2
within each maturity stage. Armstrong et a1 (1991) grouped subjects into sexual maturity
stages, but did not account for variation in chronological age within and between maturity
groups. This is important since even though youth may be in the same pubertal stage,
chronological age per se can influence biological functions. In other words, a 12 yr old
and 14 yr old in stage 3 of pubic hair are different. Another methodological limitation
was the use of an average maturation score combining pubic hair and genital in boys, and
pubic hair and breast, in girls. Allowing for there methodological limitations, absolute
peak Vo2 increased in both sexes whereas relative peak Vo2 was relatively stable in both
sexes across puberty. However, this study provided little insight into the understanding
of the influence of maturation per se on peak V02. In a subsequent study, Armstrong and
colleagues (1998) enrolled sixth grade children in a longitudinal study. The design
allowed for a sample within a narrow chronological age range (1220.4 yrs). Therefore,
when subjects were grouped into maturity groups (pubic hair only), the effect of

chronological age is limited. Results confirmed previous findings that absolute peak Vo2
increased and relative peak Vo2 remained relatively stable in both sexes across maturity
groups within this narrow age range.

Among the longitudinal studies including peak V0,, several report increments but
few use smoothing techniques or graphic or algebraic fitting procedures (Beunen and
Malina, 1988). Armstrong and Welsman (1994) noted that the use of mathematical
models to fit growth curves for peak Vo2 should be interpreted with caution since a

limited number of annual observations were considered in these studies. The available

144

data indicate that absolute peak Vo2 shows a clear adolescent spurt in both sexes at the
age of PHV (Beunen and Malina, 1988). Estimated peak velocities of absolute Vo2 from
the Saskatchewan Growth Study are 412 ml'min'l and 284 rnl'rnin'l in boys and girls,
respectively (Mirwald and Bailey, 1986). There is no clear change in relative peak Vo2
during adolescence, but body size may confound this observation.

Maturity-associated variation can also be considered when adolescents are
grouped as early, average, and late maturing based on an indicator of maturity status. In
general, absolute peak Vo2 is greater in early maturers and relative peak Vo2 is greater in
late maturers (Kemper and Verschuur, 1987; Malina et al., 1997). The differences in
absolute peak Vo2 reflects the greater body size of early maturers; in contrast, the smaller

body size of late maturers may account for the greater relative peak V02.

PEAK V02 IN YOUNG ATHLETES

Comparisons between young athletes and the general population have been used
' to draw inferences about the influence of physical activity or exercise training on peak
V02. Allowing for potential genetic pre-disposition of athletes, some of the variation in
, peak Vo2 can be explained by environmental factors and genotype-environmental
interactions. It is also important to remember that child and adolescent athletes are often
ill-defined so that comparability among studies may be questioned.

Table 6.1 provides a summary of cross-sectional studies of peak Vo2 in child and
adolescent athletes, while Figures 6.1 and 6.2 show age-associated variation in peak Vo2
of young athletes followed longitudinally. In general, child and adolescent endurance

athletes possess a superior peak Vo2 compared to the general population.

145

Cross-sectional studies indicate the physiological profile of young athletes.
Longitudinal studies of young athletes have mainly been conducted to determine the
influence of intensive training on physiological measurements associated with endurance
performance (Table 6.2). These studies ordinarily rely either on bi-annual or annual
visits and continue from 2-8 years in 10-20 yr old youth. Only one longitudinal study
included girls (Baxter-Jones et al., 1993). As in the general population, absolute peak
Vo2 increases with age, but results for relative peak Vo2 are equivocal. Some studies
show a stable pattern of development (Daniels et al., 1978) while others report an age-
related increase in relative peak Vo2 of young athletes (Murase et al., 1981; Paterson et
al., 1987). The latter finding raises questions pertinent to the present study. Does the
lack of an increase in weight-specific peak Vo2 suggest that intensive training does not
influence the developmental plasticity of peak Voz?, or is this observation confounded by
the manner in which peak Vo2 is expressed relative to body size?

Another relevant question is the following: Are the growth increments of peak
Vo2 greater in young athletes compared to the general population? Such a comparison
would provide insight into the influence of intensive training during childhood and
adolescence. However, genetic regulation of growth of the components of the oxygen
transport system still cannot be dismissed. Only two longitudinal studies have reported
the growth velocities of peak Vo2 (Maingourd et al., 1994; Paterson et al., 1987).
Maingourd et al. (1994) found that peak Vo2 increased 309 nrl'min"°kg‘l prior to PHV and
433 ml'min"'kg" after PHV in 11-15 yr old hockey players. Growth increments
extrapolated from a figure in the study by Paterson et al. (1987) indicate that peak Vo2

increased between 250-350 rrrl'min"‘yr'l prior to PHV and between 450-500 ml'min"'yr" at

146

PHV. Yearly increments estimated from the age-specific mean values of peak Vo2 in
Daniels et al. (1978) also suggest an increase of about 500 ml'min"'yr“ during the
adolescent growth spurt. Compared to the 320 ml'min’l increase in the general population
of boys (Mirwald and Bailey, 1986), it appears that growth increments in peak Vo2 of
young athletes are greater during adolescence. However, increments have a disadvantage
since the continuity of the developmental process is ignored and methodological
inconsistencies may be introduced (Beunen and Malina, 1988). Future studies need to
systematically address this issue.

Only the Training of Young Athletes (T OYA) study has examined maturity-
associated variation in peak Vo2 of young athletes grouped by sexual maturity status
(Baxter-Jones et al., 1993). Subjects were grouped in pre- ( stage 1), mid- ( stages 2 and
3), and late- (stages 4 and 5) pubertal groups based on development of breasts in girls and
genitalia in males. Absolute peak V02 increased with advanced maturity status and
relative peak Vo2 remained stable in girls and increased in male swimmers and soccer

players across maturity groups.

BASIC PRINCIPLES AND HISTORICAL BACKGROUND OF SCALING
The concept of scaling is a fundamental principle of engineering and a basic
concept in the zoological sciences, particularly among comparative mammalian
physiologists. Scaling is a cornerstone in the search for unifying principles among
animals of differing body rrrass and shape (White, 1987). Engineers have long

recognized that when the size of a structure is increased, three parameters could possibly

147

“1'. fix"! I!.l'2:l0t

change - the dimensions, the materials, or the design of the structure. In animals, linear
dimensions and body mass can easily be measured.

Principles of scaling are based on dimensionality theory, where surface area is
proportional to the volume raised to the 213 power. Thus, as the volume of a body
increases, its surface area does not increase in the same proportion, but rather as the 213
power of the volume. This argument holds only for an isometric body. Animals (and
particularly growing animals), however, are not isometric, since certain proportions
change in a regular fashion. Non-isometric scaling is referred to as allometric scaling
(Schmidt-Nielsen, 1984).

A fundamental question related to scaling is, ‘How should differences in body
size be partitioned mathematically or statistically?’ Several authors have addressed this
question (Smith, 1984; Winter, 1996). Traditionally, exercise physiologists have
expressed physiological measurements as a ratio standard (i.e., Vo2 as ml'kg"‘min“).
Fifty years ago, Tanner (1949) addressed the theoretically fallacious and misleading
practice of expressing physiological measurements per unit of body mass or per unit of
surface area. Although acknowledged periodically thereafter, until recently the issue of
scaling of Vo2 has not been systematically addressed in humans. Nevertheless, the use of
ratio standards remains common in exercise science.

How differences in body size can be partitioned mathematically or statistically
may best be answered by the following three points of Packard and Boardman (1987) in a
critical review of the use and misuse of ratios in ecological physiology:

1) Biologists were influenced by competent biometricians to use ratio

standards to scale physiological data,

148

2) Ratios are easy to compute, and
3) It is hard to imagine that ratios can be misleading, because they
are so easy to compute and comprehend.

Due to the theoretical and statistical lirrritations of the ratio standard, other
statistical methods including linear regression, analysis of covariance (ANCOVA),
allometric models, power functions, and multilevel modeling have recently gained
attention in the exercise sciences (Armstrong and Welsman, 1994; Winter, 1996). The
most widely used of these models is allometry. The allometric, or power function,
equation has the general form y = ax” which describes the curvilinear relationship
between a biological variables. In biological problems related to body size, the
independent variable (x) is body mass (Mb) so that the allometric equation is expressed
as, y = a M, b. In the present context, peak Vo2 represents y, or the dependent variable.

Early studies of animals representing a wide range of animals from rodents to

elephants indicated that a scaling factor of approximately 0.74 best describes the

relationship between body size and resting metabolic rate (Brody, 1945; Kleiber, 1932).

Taylor and Weibel (1981) used allometric scaling to compare the size of various
structures in the oxygen transport system with peak Vo2 in wild African and domestic
mammals ranging from a 0.5 kg dwarf mongoose to a 263 kg Zebu cow. Findngs from
this study were consistent with the theoretical models that peak Vo2 scales to M,, “7’.

When should allometry be used? Calder (1987) suggests that as useful as

allometry may be, there is lack of a general consensus on principles of its application. To

clarify the application of allometry, the following points have been emphasized by

Schmidt-Nielson (1984).

149

firm—rm er. 3:»; ‘

W"“mm-W‘m

° Allometric equations are descriptive; they are not biological laws.

' Allometric equations are useful for showing how a variable quantity is related to
body size, all other things being equal.

° Allometric equations are valuable tools because they may reveal principles and
connections that otherwise remain obscure.

' Allometric equations are useful as a basis for comparisons and can reveal
deviations from a general pattern.

' Allometric equations are useful in estimating the expected magnitude of some
variable for a given body size.

' Allometric equations cannot be used to extrapolate beyond the range of the data

on which they are based.

ALLOMETRIC SCALING, BODY SIZE, AND PEAK VO,

Peak Vo2 increases as a function of body size throughout the animal kingdom. As
‘ noted above, peak Vo2 in wild and domestic mammals scales approximately pr0portional
to the theoretical value of Mb 0‘75 . In humans, the peak Vo2 -body mass relationship also
exists, particularly during growth and maturation. Given variation and change in body
size and peak Vo2 associated with growth and maturation, much of the attention on
scaling peak Vo2 for body size has been directed at children and adolescents. Table 6.3
provides a summary of reported scaling factors for peak Vo2 in children and adolescents.
Mean scaling factors for peak Vo2 range from 0.27 to 1.09. An argument can be made
that just as many scaling factors approximate Mb "0 as those that approximate the

theoretical values of 0.67 and 0.75. This observation has led to the recommendation that

150

peak Vozbe expressed as a simple ratio standard in children and adolescents (Bar-Or,
19$). Nevill (1994) has explained this observation based upon the findings of
Alexander et al. (1981), who demonstrated that larger mammals have a greater proportion
of segmental muscle mass in relation to their total body mass, i.e., leg muscle mass is
proportional to body mass”. From an ontogenetic perspective, children may exhibit a
disproportionate increase in muscle mass and, therefore, violate the assumption of the
allometric model.

Until this point, the growth-related changes in peak Vo2 have been considered in
absolute terms and expressed as a ratio standard. Armstong and Welsman (1994) argue
against the expression of peak Vozas a ratio standard for growth-related comparisons,
stating that it clouds the understanding of growth and maturational changes in the oxygen
transport system. As noted earlier, peak Vo2 expressed per unit body mass remains
relatively stable during childhood and adolescence in boys and decreases with age in
girls, particularly during adolescence. To explore the growth-related changes in peak V02,
Armstrong and colleagues (1994) have used various allometric scaling techniques.
Adjusted means (ANCOVA controlling for body mass) for peak Vo2 were similar
between 10 and 15 year old boys (2.21 and 2.30 L'min", respectively). In a subsequent
paper, a significant increase in peak Vo2 from prepubertal to pubertal and adult males,
and between prepubertal and pubertal girls was noted when data were fitted by linear and
log-linear allometric models adjusted for body mass (Welsman et al., 1996). Adjusted
means were similar between pubertal and adult females, suggesting that peak Vo2 remains
constant during this period. These results are intriguing, given past assumptions about

growth-related changes in peak V02. Recently, Armstrong et al. (1998) also

151

demonstrated a maturity-associated increase in peak Vo2 in 12 year old boys and girls.
Log-linear adjusted means increased from 2.01 to 2.30 L‘min" in boys and 1.78 to 1.99
L'min'l in girls grouped by stage of pubic hair development.

The preceding results are based on cross-sectional analyses. A longitudinal study
of 11-14 yr old youth active in sport (track, wrestling, basketball) found that the peak Vo2
-body size relationship varied with maturity status and sex (Beunen et al., 1997). In early
and average maturing boys, peak Vo2 increased at a slightly higher rate than expected
from the increase in body mass, whereas in later maturing boys the increase was smaller
than expected. In contrast, the increase of peak Vo2 in girls active in sport (track, rowing)
was generally unrelated to the increase in body mass or stature. There was also
considerable inter-individual variation in scaling coefficients during early and mid-
adolescence. These results, though limited to early and mid-adolescence, suggest sex-
and maturity-associated variation in growth of the oxygen transport system.

A different interpretation of growth-related changes in peak Vo2 of young athletes
is also evident when peak Vozis expressed per kgms (Sjodin and Svedenhag, 1992), or
when body size is controlled in multi-level modeling (Baxter-Jones et al., 1993).
Although peak Vo2 remained stable during adolescence in young distance runners when
expressed as a ratio standard, peak Vo2 expressed by the exponent 0.75 showed an
increase from 161 to 186 ml'kgms'rnin'l (Sjodin and Svedenhag, 1992) . The increase in
peak Vo2 appeared to occur from 3 yrs before PHV to 6 months after PHV with a plateau
occurring thereafter. In the TOYA study, peak Vo2 increased after statistically
controlling for age, stature, and body mass in males across pubertal status, and in girls

from pre- to mid-puberty with no further increase between rnid- and late-puberty (Baxter-

152

“simmer-er 12.-9

“F .fl' ’9‘ hm . 1

Jones et al., 1993). However, despite acclaimed usefulness in the interpretation of
longitudinal data, the biological significance of the results derived from the multilevel

modeling approach is difficult to interpret.

SCALING PEAK V0,: IMPLICATIONS FOR ENDURANCE PERFORMANCE
AND HEALTH-RELATED FITNESS

The importance of scaling peak Vo2 for differences in body size in the
interpretation of growth-related changes in peak Vo2 has been considered. An additional
question is "What are the implications for endurance performance and health-related
fitness?" The application of scaled peak Vo2 to the interpretation of endurance
performance and health-related fitness has received limited attention. Based on the
various analyses and interpretations, the utility of allometric scaling may need to be
considered in the context of the problem.

Svedenhag (1995) recently reviewed the implications of scaling peak Vo2 and
submaximal V02 (V02 mhm) for evaluations of the endurance athlete. The author
suggested that whether peak Vo2 and Vo2 mm is scaled to Mb "0 or Mb 0'75 may influence
the evaluation and the selection of a training program for an endurance athlete (Table
6.4). In this example, Runners A and B have an equivalent fractional utilization of V02
(%V02) and similar performance levels. Based on the traditional ratio standard for
Vo2 m and peak Vo2 (ml'kg"'rrrin“), Runner A has a better running economy but a
lower peak V02, whereas Runner B has a poorer running economy and a higher peak V02.
This may lead a coach or athlete to manipulate training in an attempt to improve upon the

poorer functional capacity. In contrast, if values are expressed per unit kg 0'75 , the

153

runners have similar values, or perhaps results contrary to the initial analysis. Thus,
scaling Vo2 mm and peak Vo2 may influence the findings at evaluation and resultant

. training programs for endurance athletes. In contrast, Nevill ( 1997) suggested that the
ratio standard provides the best predictor of weight-bearing athletic performance. Using
a multiple log-linear regression, the best predictor of 5 km race performance in adults
was almost exactly proportional to the ratio standard (Nevill et al., 1992).

When considering such physiological variables as peak Vo2 in the context of risk
factors, Nevill (1997) also noted the importance of partitioning the confounding effects of
chronological age, body size, and biological maturity status. However, no study could be
identified that specifically addressed the implication of scaled peak Vo2 on the health-

related fitness of either children or adults.

SUMMARY

Absolute peak Vo2 is related to chronological age, biological age, and body size
during childhood and adolescenbe, and increases by about 11% annually with the largest
increase occurring near the time of PHV. Although limited by methodological and
ethical constraints, it appears that quantitative increases in overall body size and the
components of the oxygen transport system rather than qualitative changes in the
functional properties of the components of the oxygen transport system explain growth-
related changes in peak V02. Sex-associated variation occurs due to differences in body
composition, hematological, and perhaps, physical activity patterns. Sex differences are

small prior to puberty, but increase greatly during adolescence. Future longitudinal

154

studies of children and adolescents, both athletes and non-athletes, need to consider both
quantitative and qualitative changes in the oxygen transport system.

When expressed as a ratio standard, peak Vo2 remains stable in boys and
decreases in girls, particularly during adolescence. However, exercise scientists have
been criticized for not recognizing the imperfections of ratio standards and for being
unaware of alternative methods for partitioning the effects of body size in human studies
(Winter, 1996). Therefore, current studies are exploring the use of allometric scaling in
expressing peak Vo2 during childhood and adolescence. However, it remains to be
demonstrated if allometric scaling within a small magnitude of differences in body size
warrants such statistical manipulation. According to Calder (1987), a small size range
within a species obscures overall trends, patterns, and constraints of body size. Thus,
scaling differences in body size within a small range of body size to understand variation
in biological function may be of limited value. In contrast, others argue that scaling body
size helps us to understand the growth and maturation of the oxygen transport system, its
response to submaximal and maximal exercise (Armstrong and Welsman, 1994), its
relationship to risk factors (Nevill, 1997), and the evaluation of endurance athletes
(Svedenhag, 1995). The application of allometric scaling to the age-, sex-, and maturity-
associated variation in peak Vo2 of well-trained pediatric endurance athletes is limited.

Part II of this dissertation examines this issue.

155

REFERENCES

Alexander RM, Jayes AS, Maloiy GMO, Wathuta EM (1981) Allometry of the leg
muscles of mammals. J Zool 194:539-552.

Armstrong N, Welsman JR (1994) Assessment and interpretation of aerobic fitness in
children and adolescents. Exerc Sport Sci Rev 22:435-476.

Armstrong N, Welsman JR, Kirby BJ (1998) Peak oxygen uptake and maturation in l2-yr
olds. Med Sci Sports Exerc 30:165-169.

Armstrong N, Williams J, Balding J, Gentle P, Kirby B (1991) The peak oxygen uptake
of British children with reference to age, sex and sexual maturity. Eur J Appl Physiol
62:369-375.

Asmussen E, Heebol-Nielsen K (1955) Dimensional analysis of physical performance
and growth in boys. J Appl Physiol 7:593-603.

Bar-Or O (19$) Pediatric Sports Medicine for the Practitioner. New York: Springer-
Verlag.

Baxter-Jones A, Goldstein H, Helms P (1993) The development of aerobic power in
young athletes. J Appl Physiol 75: 1 160-1 167.

Beunen GP, Malina RM (1988) Growth and physical performance relative to the timing
of the adolescent spurt. Exerc Sci Sports Rev 16:503-540.

Beunen GP, Rogers DM, Woynarowska B, Malina RM (1997) Longitudinal study of
ontogenetic allometry of oxygen uptake in boys and girls grouped by maturity status. Ann
Hum Biol 24:33-43.

Blair SN, Kohl HW, Paffenbarger RS, Clark DG, Cooper KH, Gibbons LW (1989)

Physical fitness and all-cause mortality: a prospective study of healthy men and women.
JAMA 17:2395-2401.

Brody S (1945) Bioenergetics and Growth, with Special Reference to the Efficiency
Complex in Domestic Animals.New York: Reinhold.

Calder WA (1987) Scaling energetics of homeothemric vertebrates: an operational
allometry. Ann Rev Physiol 49:107-120.

Coyle E (1995) Integration of the physiological factors determining endurance
performance ability. Exerc Sport Sci Rev 23:25-63.

Daniels J, Oldridge N, Nagle F, White B (1978) Differences and changes in V02 among
young runners 10 to 18 years of age. Med Sci Sports 10:200—203.

156

Daniels J, Oldridge N, N agle F, White B (1978) Differences and changes in V02 among
young runners 1010 18 years of age. Med Sci Sports 10:200-203.

Eisenmann JC, Malina RM (in press) Body size and endurance performance. In R]
Shephard (ed.): Endurance in Sport. Oxford: Blackwell Science.

Kemper HCG, Verschuur R (1987) Longitudinal study of maximal aerobic power in
teenagers. Ann Hum Biol 14:435-444.

Kleiber M (1932) Body size and metabolism. Hilgardia 6:315-353.

Krahenbuhl GS, Skinner J S, Kohrt WM (1985) Developmental aspects of maximal
aerobic power in children. Exerc Sport Sci Rev 13:503-538.

Maingourd Y, Libert JP, Bach V, Jullien H, Tanguy C, Freville M (1994) Aerobic
capacity of competitive ice hockey players 10-15 years old. Jap J Physiol 44:255-270.

Malina RM, Beunen G, Lefevre J, Woynarowska B (1997) Maturity-associated variation
in peak oxygen uptake in active adolescent boys and girls. Ann Hum Biol 24: 19-31.

Malina RM, Bouchard C (1991) Growth, Maturation, and Physical Activity. Champaign,
IL: Human Kinetics.

Mirwald RL, Bailey DA (1986) Maximal Aerobic Power: A Longitudinal Analysis.
London, Ontario: Sports Dynamics.

Murase Y, Kobayashi K, Kamei S, Matsui H (1981) Longitudinal study of aerobic power
in superior junior athletes. Med Sci Sports Exerc 13:180-184.

Nevill AM (1994) The need to scale for differences in body size and mass: an
explanation of Kleiber's 0.75 mass exponent. J Appl Physiol 65:110-117.

Nevill AM (1997) The appropriate use of scaling techniques in exercise physiology.
Pediatr Exerc Sci 9:295-298.

Nevill AM, Ramsbottom R, Williams C (1992) Scaling physiological measurements for
individuals of different body size. Eur J Appl Physiol 65: 1 10-1 17.

Packard G, Boardman T (1987) The misuse of ratios to scale physiological data that vary
allometrically with body size. In M Feder, A Bennet, W Burggren and R Huey (eds.):
New Directions in Ecological Physiology. Cambridge: Cambridge University Press, pp.
216—236.

157

Rowland TW (1990) Developmental aspects of physiological function relating to aerobic
exercise in children. Sports Med 10:255-266.

Rowland TW (1996) Developmental Exercise Physiology. Champaign, IL: Human
Kinetics.

Schmidt-Nielsen K (1984) Scaling: Why is animal size so important? Cambridge:
Cambridge University Press. ’

Sjodin B, Svedenhag J (1992) Oxygen uptake during running as related to body mass in
circumpubertal boys: a longitudinal study. Eur J Appl Physiol 65:150-157.

Smith R (1984) Allometric scaling in comparative biology: problems of concept and
method. Am J Physiol 246:R152-R160.

Svedenhag J (1995) Maximal and submaximal oxygen uptake during running: how
should body mass be accounted for? Scand J Med Sci Sports 5: 175-180.

Tanner JM (1949) Fallacy of per-weight and-per-surface area standards, and their relation
to spurious correlation. J Appl Physiol 2: 1-15.

Taylor CR, Maloiy GMO, Weibel EW, Langman VA, Kamau JMZ, Seeherrnan HJ,
Heglund NC (1981) III. Scaling maximum aerobic capacity to body mass: wild and
domestic mammals. Respir Physiol 44:3-37.

Weibel E (1984) The Pathway for Oxygen: Structure and Function in the Mammalian
Respiratory System. Cambridge: Harvard Press.

Welsman JR, Armstrong N, Nevill AM, Winter EM, Kirby BJ (1996) Scaling peak V02
for differences in body size. Med Sci Sports Exerc 28:259-265.

White F (1987) Scaling and structure-function relationships. Ann Rev Physiol 49:105-
106.

Winter EM (1996) Importance and principles of scaling for size differences. In 0 Bar-Or
(ed.): The Child and Adolescent Athlete. Oxford: Blackwell Science, pp. 673-679.

158

Table 6.1. Mean peak oxygen consumption (peak V02) reported in cross-sectional studies
of child and adolescent endurance athletes.

 

Study N Age (yrs) Sex Sport Vom,
(mlwkglmin-l)
Dill and Adams 6 17 M Running 72.0
(1971)
Vaccaro and Clarke 15 9-11 M Swimming 55.4
(1978)
Kobayashi et al. 4 17 M Running 73.9
(1978)
Mayers and Gutin 8 8-11 M Running 56.6
(1979)
Lehmann et al. 8 12 M Running 60.3
(1981)
Sundberg and 12 12 M Running 59.3
Elovaino (1982)
12 16 66.4
Fario et al. (1989) 15 15-19 M Cycling 75.5
Nudel et al. (1989) 16 8-17 M, F Running 61.0
Cunningham ( 1989) 20 15 F Running 62.1
Cunningham (1990) 12 16 F Running 66. l
12 16 M 74.6
Rowland et al. 10 . 1 1-13 M Running 61.2
(1991)

 

- Adapted from Rowland (1996).

159

was; 11"“ r ~;

 

 

Hue—u 9N. _Loammzamau. 3:38 6». EGGS—6m?»— uumuumc. 3 uEE 33 36—8qu uses—.38 unauaum.

 

meg, meguunm 73369.. 32: al.—35mm
Uu3u_m u” a. No 3»? Baa—u 3338 as.» u6=uu8a u<uQ =6 urgmu 3 8.33 eumr <6»“
23$ 332.8 mmuu _oLm v.8 A338- a 36:25 n2. ~-m auunumuu 3 2338» <6» £5. emu
55:33»: v.3
3:38 u. a. Z auvueuuu 3u_u 38a; 33 ~33»— <mmz «on uh mauaummu 3 4u_m:<u ~6qu <6»
Com: _6=m 2238 £333 3 Eu a... «an m 635 4.8268 3m 8 am 3.xm.-_33.-_v
033363332 umuw 3.»— VB 5.33m 3.84 _m «a
6*. emu

3.836: u. a. _m cove 3322. 3 $6... :36
Gem: €u~u 3 Eu =23. ocean—u 6*.
36.8 nuupua 3 .3" emu 2.3 «a

macaw: 3a m 3333 23 A 3:332. comm“
m<u=au=rum emu 5.8
:88

mama—£63m up mm 3&u 23 we 3338

a. :83 u€m33uau umum 5-3 v.8
A3mxuq-_6=mm8&=m:

Zaamoea u” I muu ecu—8v. 2368“ umum _o-

m... 283 _m v.3

33.»— 5mm" .2. m
v.8

u<uQ a 3633.. 3..
m <3

33:». 52. m2. w
<8.“ 3=_:-_u<u_
38155
333— <mu= ~64 e
v.8

maunuuuu 3 nu. <6»as

3— .6 am 31.53.33.233 <4. GA
8 AN 3_.xm.-_33.-_x <._<<6»as
mSEu“ €33 ezmaua 8 m:<.
magnum" mauquuuu 3 <6»? ouuza
a. 3.2. firm—u <4 mSEu

wee: 3 nu_um<u <6»? a 363—;
3.84 2.2 3 338 mace?
auuauuuu 3 «.338» <6»“ mzuauumu
3 tum—n <6» uxEummua muq wmed
mum—n <6» 39.3.8; :32. emu 33
66% umnu sunu 833:2.“

vuuw <6» 33 <.—. “.6296;
8.8—».an $3. 3.2“ 322338
8383 3.63 «a 8 k4

 

<6» . 6xv~mu= 8333.363 <4. <u=Eu8Q 3336—9 vI<. mum—n ram—z <26qu

160

Table 6.3. Summary of scalingfactors for peak V& in children and adolescents.

 

 

Study N Sex 13g; (yrs) Mode Mb Exponent
Cross-sectional
Astrand (1952) 68 M, F 6-17 T 095"
Cooper et al. 58 M 6—17 C 1.09
(1984)
51 F 0.83
109 M,F 1.01
Rogers et a1. 25 M,F 7-10 T 0.47
( 1995)
27 M,F 1 l— 14 0.62
Welsman et al. 32 M, F 9-10 T 0.52
(1997)
Armstrong et 212 M,F 12 T 0.65
al. ( 1998)
Longitudinal
Astrand et al. 30 F 12-16 0.97:?
(1963)
Daniels et a1. 14 M 10-15 T 1.07#
( 1972)
Klissouras 50* 7-13 0.88#
(1972) '
Bailey et al. 51 M 8-15 T 0.82A
(1978) ,
Paterson et al. 18 M 1 1-15 T 1.02
( 1987) 1.19**
Sjodin & 8 M 11-15 T 1.01
Svedenhag (trained)
(1992)
4 M 0.78
. (untrained)
Beunen et al. 47 M 11-14 C 0.80 (early + avg.
(1997) maturers)
0.57 (late maturers)“
31 F 0.27 (early + avg.
maturers)

0.42 (late maturers)M

 

Mb, body mass; T, treadmill; C, cycle ergometer.

Table adopted from Rowland (1996).

#As estimated by McMiken (1976).

*25 sets of twins including 15 pairs of identical twins and 10 pairs of fraternal twins.
"Results of ontogenetic allometric scaling.

"As reported by Rowland (1996).

161

flea—u ab. > 83318.. 6.. 2.6 uznu E338 3383 6.. 65.3.5 66% Beam e3. 2.638333
e36 Rm.» <6 cemuq 6.. £3 _u 336 upeeaea e3. e=63u3u wueze .

 

<6» 2.638. vuer <6»

Zemmawmv 1: 3m. 323.33.. 3.55333. 3_.wm._.3m..-_ 3___.m.°.d.3m...-_

 

”332 > mo um mmu .8 dub NN—

weeeua w mo um a. .m an mwb Em

 

>66..an 3.63 m<uau==em 28$.
1C. 3.826.3— 32.8263 26» 2.638568.» <6»v.

162

V02 (ml/min)

4250 '-

4000-1 I

 

 

I 0"
.’ .‘v’
’ 5
I

3750 - 9 ’, I:

I i"

:' I
3500 ‘ I II'd‘

I, 0'

I f

I. ’ '
3250 - ,’
9' x
.: I .
.3 I ;
3000 ‘ j. I’p
3" I, I.
.3 l I
2750 .. 1' .' -------- 0mm" Cunningham et al., 1987
:0” 0"
:é ." ""0"" Daniels et al., 1978
2500 - ,’ p
i I '0
$.31; ----fi--- Murase et al., 1981
I"
2250 - 33
..I'E - - '3'- - - Baxter-Jones et al., 1993
U..."
2000 - R5
1750 I I l T r l
10 12 l4 16 I8 20

Age (yrs)

Figure 6.]. Longitudinal studies of absolute peak V02 in male athletes.

163

peak V02 (ml/kg/min)

 

 

 

70‘
,9
I°‘~~~ v”6
65" I’ ‘9,—
o’. v
E a”; ' -a’l‘ia
60- ‘m ,’ B
I
wv ‘m
55‘
D.-. ---m---
B- ""Ii
50'-
-.-....-.
--+--
45-
--v--
-.q..-
40 I I I I I n
8 10 12 l4 l6 18 20
Age (yrs)

Daniels et al (1978) - boys
Cunningham et al.(l989) - boys
Sjodin and Svedenhag (1992) - boys
Baxter-Jones et al (1993) - boys

Baxter-Jones et al (1993) -girls

Figure 4. Longitudinal studies of relative peak V02 in young athletes.
Solid lines represent age-related changes in the general population.

164

CHAPTER7

SCALING PEAK V02 T0 BODY MASS

IN YOUNG MALE AND FEMALE DISTANCE RUNNERS

165

ABSTRACT

This study examined age- and sex-associated variation in peak Vo2 of young male
and female distance runners from an allometric scaling perspective. Subjects were from
two separate studies of 9—19 yr old distance runners from the mid-Michigan area, one
conducted between1982-1986 (Young Runners Study 1, YRS I) and the other in 1999-
2000 (Young Runners Study 11, YRS 11). Data from 27 males and 27 females from YRS
I and 48 males and 22 females from the YRS II were included, and a total of 139 and 108
measurements of body size and peak Vo2 in males and females, respectively, were
available. Subjects were divided into whole year age groups. A 2x9 (sex x age group)
ANOVA was used to examine differences in peak V02. Intra-individual ontogenetic
allometric scaling was determined in 20 males and 17 females measured annually for 3-5
years. Allometric scaling factors were calculated using linear regression of log-
transformed data. Results indicated that 1) absolute peak Vo2 increases with age in boys
and girls, 2) relative peak Vo2 (ml'kg"‘min"‘) remains stable until age 15 when it increases
in boys and decreases in girls, 3) relative peak Vo2 (rril'kg"°‘75 min"') increases throughout
the age range in boys and increases in girls until age 15 yrs, and 4) peak Vo2 adjusted for
body mass (ml'min"') increases with age in boys and girls. The overall mean cross-
sectional scaling factor was 1.01:0.03 (SE) in boys and 0.85:0.05 (SE) in girls.
Significant age x sex interactions and significant scaling factors between sexes identifies
the progressive divergence of peak Vo2 between adolescent male and female distance
runners. Mean ontogenetic allometric scaling factors were 0.81 (0.71-0.92, 95% CI) and
0.61 (0.50-0.72, 95% CI) in males and females, respectively (p=0.002). There was

considerable variation in individual scaling factors (0.51-1.31 and 0.28-0.90 in males and

166

females, respectively). The results suggest that the interpretation of growth-related
changes in peak Vo2 of young distance runners is dependent upon the manner of

expressing peak Vo2 relative to body size and/or the statistical technique employed.

167

INTRODUCTION

During growth and maturation, absolute peak Vo2 (nrl'min"‘) increases as a
function of body size (Armstrong and Welsman, 1994; Krahenbuhl et al., 1985). A
major question related to this observation is, “Are the growth-related improvements in
physiological capacity a function of increasing body size or qualitative changes in the
structural and functional capacity independent of body size or both?” (Rowland, 1990;
Winter, 1996). To provide answers to this question, the potentially confounding effect of
variation in body size must be appropriately partitioned.

Age- and sex-associated variation in peak Vo2 has been extensively studied in the
general population (Armstrong and Welsman,-1994; Krahenbuhl et al., 1985). Several
cross-sectional studies have characterized the physiological profile of young endurance
athletes, but longitudinal studies of the development of peak Vo2 in young athletes,
especially females, are rather limited (Baxter-Jones et al., 1993; Beunen et al., 1997;
Daniels et al., 1978; Maingourd et al., 1994; Murase et al., 1981; Paterson et al., 1987;
Sjodin and Svedenhag, 1992). These studies generally include small sample sizes and are
limited to a narrow age range (i.e., 11-15 yrs), and therefore, do not describe the growth-
related changes in peak Vo2 across the entire adolescent period. Longitudinal studies are
important to identify individual and population growth patterns. Thus, there is a need for
analyses of longitudinal data examining the age- and sex-associated variation of peak Vo2
in young athletes from various sports.

Traditionally and conventionally, peak V0,, is expressed as a ratio standard, or per
kg of body mass (ml'kg"'rnin"). When expressed as the simple ratio standard, peak Vo2

remains stable in boys and declines in females during adolescence (Krahenbuhl et al.,

168

1985). By expressing peak Vo2 in this manner, it is assumed that peak Vo2 is
“normalized” and the influence of body mass is removed. However, the theoretical and
. statistical limitations of the ratio standard have been widely addressed, yet largely
ignored (Tanner, 1949; Winter, 1996). Therefore, alternate statistical models, including
analysis of covariance (ANCOVA), allometric scaling, and multilevel modeling have
been used to create a ”size-free" expression of peak V0,. The mathematical model that is
widely used to create a “size-free” variable is allometry. Besides the calculation of cross-
sectional allometric scaling factors, intra-individual, or ontogenetic, scaling factors can
be calculated from longitudinal records. Ontogenetic allometry refers to differential
growth in the individual growth process (Gould, 1966). Few studies have employed
ontogenetic allometry to examine growth-related changes of peak Vo2 in young athletes
(Beunen et al., 1997; Paterson et al., 1987; Sjodin and Svedenhag, 1992).

The purpose of this study is to examine age- and sex-associated variation of peak

Vo2 in competitive young distance runners from an allometric scaling perspective.

NIETI-IODS
Design

Subjects are from two separate studies of young distance runners from the mid-
Michigan area conducted at the Institute for the Study of Youth Sports at Michigan State
University. The first study (Young Runners Study 1, YRS I) was an inter-disciplinary,
mixed-longitudinal assessment of intensive training and competition on "elite" young
distance runners between 1982 and 1986 (Seefeldt, 1986). The second study (Young

Runners Study 11, YRS II) was a cross-sectional study of the association between training

169

volume and blood cholesterol that included the measurement of peak V02. Data sets were
pooled for the cross-sectional analysis to create a larger sample for age group
comparisons. Differences in subject inclusion criteria, treadmill protocol, and exercise
testing systems between the two studies are recognized. However, subjects from both
studies were highly trained based on training history, race performance, and peak V02.
Previous studies have also shown that minimal differences in peak Vo2 occur due to
treadmill protocol (speed and incline, continuous versus discontinuous) (Paterson et al.,
1981; Skinner et al., 1971) and exercise testing systems (automated versus non-
automated) (Jones, 1984). The former has specifically been addressed in adolescent
distance runners (Rivera-Brown et al., 1994). Only data from the YRS I was used for the

ontogenetic allometric analysis.

Subjects

YRS I. Runners between the ages of 8 and 15 years, who consistently placed
”within the top five finishers of road races of 10 km or more by age and sex, were
identified and contacted for the study. Race results were obtained from a statewide
running publication, Michigan Runner, between May and August 1981 . Of the runners
contacted (response rate unknown), 27 males and 27 females agreed to participate in the
study. Subjects entered the study at 8.0 to 15.7 years of age and were followed annually.
Twenty males and 17 females were followed at approximately annual intervals for 3 to 5
years. The remaining subjects (7 boys and 10 girls) participated in either 1 or 2 annual
visits. Each age and sex group included only one observation per subject, thus the

subjects were treated as independent in each age group. A total of 99 and 84 annual

170

measurements were available for males and females, respectively. Parental consent and
child assent was obtained prior to the study. The study was approved by the Michigan
State University Committee for Research Involving Human Subjects.

YRS II. Forty-eight males and 22 females, 10-19 years of age, agreed to
participate in the study. Eligible subjects participating on local Michigan junior or senior
high school cross—country teams or local track clubs during Fall,1999 and Spring, 2000
were invited to participate. Administrators of the various organizations (i.e., camps,
schools) were contacted and given information about the study. A brief description of the
study was provided to athletes participating on interscholastic teams and track clubs.
Information about the study was also provided at local road races and running facilities,
and in the local newspaper. Due to the primary objective of the study being the
investigation of blood cholesterol the following exclusion criteria were indicated: current
smoking, excessive alcohol intake, current use of blood cholesterol lowering and anti-
hypertensive medication, anabolic steroid use, hepatic, renal, and thyroid disease. In
addition, subjects who had trained less than 30-40 weeks per year or non-consecutively
during the past 3 consecutive months were also excluded to ensure a sample engaged in
regular participation in long distance running. Parental consent and child assent was
obtained prior to the study. The study was approved by the Michigan State University

Committee for Research Involving Human Subjects.

Anthropometry.
YRS I. Chronological age was calculated as the difference between observation

date and birthdate, and expressed as a decimal age. Anthropometry was conducted by

171

two experienced anthropometrists according to standard procedures (Weiner and Lourie,
1969). Stature was measured with a fixed stadiometer. The subject stood erect, without
shoes and with weight distributed evenly between both feet, heels together, arms relaxed
at the sides, and the head in the Frankfort horizontal plane . Body mass was measured
with the subject attired in gym shorts and T-shirt without shoes on a balance beam scale.
Measurements were conduced between early morning and mid-afternoon. Intra— and/or
inter-observer reliabilities were not reported.

YRS II. Chronological age was calculated as the difference between observation
date and birthdate, and expressed as a decimal age. Stature and body mass were
measured according to the procedures of the International Biology Program (Weiner and
Lourie, 1969). Stature was measured with a fixed stadiometer. The subject stood erect,
without shoes and with weight distributed evenly between both feet, heels together, arms
relaxed at the sides, and the head in the Frankfort horizontal plane. Body mass was
measured with the subject attired in gym shorts and T-shirt without shoes on a balance
beam scale. The stadiometer and scale were calibrated periodically during the study.
Intra-observer reliability was conducted on a small sub-sample by the principal
investigator (ICE). The intra—class correlation coefficient was 0.99 for both stature and
body mass, whereas the intra-observer technical errors of measurement were 0.42 cm for

stature and 0.08 kg for body mass.

Measurement of maximal oxygen consumption.
YRS I. An intermittent progressive treadmill protocol consisting of 3 minute work

intervals and 3 minute rest intervals until volitional exhaustion was used to determine

172

peak V0,. The protocol began with a warm-up at 6 mph and 0% grade. Foolowing the
warm-up, the grade was increased to 5%. Speed increased 1 mph and grade increased
1% in each subsequent stage until volitional exhaustion. Expired gases were collected
using the Douglas bag method. Gas concentrations were analyzed with Beckrnan oxygen
and carbon dioxide analyzers within two minutes after collection. Gas volumes were
measured with a Parkinson-Cowan CD2 dry gas meter. Prior to testing, expired gas
volumes were calibrated with a 3-L syringe and gas concentrations were calibrated with
standard gases of known concentrations. Heart rate was monitored using a commercial
electrocardiogram. End of test criteria were established by volitional exhaustion, HR
290% of age-predicted maximum, respiratory'exchange ratio >10, and a plateau in Vo2
(defined by an increase in Vo2 of <20 rnl'kg"l min"I with increasing workload). Two of
the latter three criteria must have been met for a subject to be included in the analysis.
YRS II. A maximal exercise test was conducted on a motorized treadmill to
exhaustion in an air-conditioned laboratory (20-22 degrees C, relative humidity 4S-60%).
The treadmill protocol was determined by the subject's estimated 5 km race pace.
Subjects walked/jogged at a speed of 3 mph and 4.5 mph for 1 min each. This initial
warm-up period was followed by 4 minute stages at 6, 7.5, and 8 mph (depending on
estimated 5 km race pace) and then increased in grade of 2.5% every minute until
exhaustion or test termination. Expired gases were collected for the measurement of
oxygen consumption (V02), carbon dioxide production (VCoz), and minute ventilation.
Expired gases were continually sampled and averaged every 20 seconds via the open
circuit method using a metabolic cart (Gould 2900; Dayton, OH). Expired gas volumes

were measured with a flow probe anemometer and expired gas concentrations were

173

measured by electronic analyzers. Prior to testing, expired gas volumes were calibrated
with a 3-L syringe and gas concentrations were calibrated with standard gases of known
concentrations. Heart rate was continually monitored by pulse telemetry (Polar
Advantage). End of test criteria were established by volitional exhaustion, HR _>_90% of
age-predicted maximum, respiratory exchange ratio >1.0, and a plateau in Vo2 (defined
by an increase in Vo2 of <20 rnl'kg"l min"1 with increasing workload). Two of the latter

three criteria must have been met to be included in the analysis.

Statistical analysis.

Subjects were divided into whole year age groups (i.e., 11.0 to 11.99), except for
the youngest age group in both sexes, which consisted of subjects 9.0 to 10.99 years, and
the oldest age group in girls that consisted of subjects 17.0 to 19.49 years. Descriptive
statistics were calculated by age and sex groups for absolute peak Vo2 and relative peak
Vo2 expressed per kg "0 and per kg 0‘75. The exponent 0.75 is common in the allometric
literature and is based on both theoretical and statistical evidence. A 2x9 (sex x age
group) ANOVA was used to examine differences in peak V02. Paired post hoc
differences were examined by the Scheffé test. The allometric analysis was applied to
the entire group for each sex (i.e., scaling factor for all boys and all girls) and to each
age- and sex-specific group (i.e., scaling factor for 14 yr old girls, etc.).

Allometric scaling. Prior to allometric analysis, the relationship between body
mass and peak Vo2 was initially checked for linearity after Tanner (1949). In this
procedure, the Pearson correlation coefficient (r) between body mass and absolute peak

Vo2 was compared with the ratio of the coefficient of variation (CV) for the two variables

174

((SDJXx)/SD’/Xy). If r is approximately equal to the CV, a linear relationship is
indicated and the simple ratio standard (ml'kg“'min") is appropriate. Conversely, if these
two terms are not similar a linear relationship does not exist and the simple ratio standard
is inappropriate.

The allometric relationship between body size and peak Vozis based on the
general allometric equation,

y = ax" (Equation 1)

where Y = absolute peak V02; x = body mass, b = scaling factor; and a =
proportionality constant. The statistical approach to allometry is to use a logarithmic
transformation as follows:

log Y = b ‘ log Mb + log a (Equation 2)

where b is the slope of the linear regression line on a double logarithmic plot.
The slope is calculated by regression analysis, where b in the regression output is equal to
the scaling factor and the inverse log of log a is equivalent to the constant (a) in Equation
'1. Analysis of covariance (ANCOVA) of log-transformed data was used to confirm the
allometric analysis and generate adjusted means for age- and sex—specific groups.

Ontogenetic allometry. Individual (ontogenetic) scaling factors were calculated
for individual longitudinal records for subjects who were assessed annually for 3 to 5
years. Of the 27 males and 27 females enrolled in YRS I, 20 males and 17 females were
considered in the present analysis. A least squares linear regression was carried out for
the records of each subject on the double logarithmic transformations of peak Vo2 and
body mass. Individual regressions were checked for goodness of fit by examining the

multiple r value and the p value from the ANOVA. Sex-specific means and standard

175

deviations of the ontogenetic allometric scaling factors were calculated. The difference
was examined by an independent t-test.

Regression diagnostics. Residuals (predicted - observed peak V02) were
converted to absolute values and correlated with the predictor variable (log body mass) to
examine the data for heteroscedasticity. Pearson correlations were also calculated
between the simple ratio standard and the common power function ratio (ml'kg'°'75min"‘)
as a diagnostic test. In this case, if the influence of body size has been removed, the

correlation should not be different from zero (Batterham et al., 1997)

RESULTS

Age- and sex-specific anthropometric and peak Vo2 values are reported in Tables
7.1 and 7.2. Stature reaches a plateau at 17 yrs in boys and 15 yrs in girls. Body mass
progressively increases across age in both sexes. Prior to 14 yrs, girls are taller and
heavier than boys, thereafter, boys are taller and heavier than girls. Mean statures for
both males and females approximate the medians of U.S. reference values (Hamill et al.,
1977) and mean body mass for both males and females is somewhat below the reference
medians. Stature and mass also maintain their position relative to the reference values
across age (Eisenmann et al., 1999).

Means for absolute peak Vo2 (ml‘min") increase with age in both sexes (p<0.05).
Absolute differences between the sexes are small (134—186 ml'min") prior to 14 yrs,
when the differences increase sharply in each age group and reach a mean difference of

1000—1500 mlmin" in the oldest age groups (p<0.05).

176

There is no significant age-related trend for peak Vo2 expressed as the simple
ratio standard (ml'kg"‘min"‘) (p>0.05). Means of relative peak Vo2 remain stable in boys
between 9-15 yrs (61-63 ml'kg"‘min"'), but increase in the older age groups (65-67 ml'kg"
1min"). In girls, means for relative peak V62 remain stable between 9-15 yrs of age (55-
58 ml°kg"'min"‘) and decrease in the oldest age groups (52-53 ml'kg"min"‘). Sex
differences vary between 5—7 rnlkg"rnin"l prior to 16 yrs and increase to 12-15 ml‘kg"
’min"1 in the oldest age groups (p<0.005). When peak Vozis expressed to the theoretical
value of body mass 0.75, it increases significantly with age (p<0.05). A plateau is
evident in males between 14-16 yrs, and there are two instances of a decline in age-
specific means in females, one at 13 yrs and the other in the oldest age groups. Similar to
absolute values, sex differences are small prior tolS yrs, and then increase (p<0.05 at all
age groups).

Peak Vo2 adjusted for body mass also shows a significant age-related increase
(p<0.05). The largest differences in adjusted means occur in the youngest and oldest age
groups (600-750 ml'min”). Mean differences between 12—15 yrs of age are 410~475
admin”, and there is a significant age group x sex interaction in adjusted means
(p=0.001).

Results of the cross-sectional allometric analysis are shown in Table 7.3. Overall,
body mass exponents are 1.01:0.03 (SE) and 0.85:0.05 (SE) in boys and girls,
respectively. The adjusted 7’ is 0.89 in boys and 0.75 in girls. Age-specific scaling
factors are closer to the theoretical values of 0.67 and 0.75 in boys, but do not fit the
model closely and in two age groups are not significantly different from zero. In girls,

three of the eight age-specific models are not significantly different from zero. The

177

significant models have scaling factors between 0.53 and 0.89. In general, the age-
specific models fit better in males than females. The cumulative effect of multiple age
groups on the overall scaling factor is also shown in Table 7.3. Although age-specific
scaling factors differ from those calculated for the entire sample, this may be due to small
age-specific sample sizes and a lack of variation in body mass and peak Vo2 within age-
specific groups. Scaling factors begin to approximate the overall sex-specific scalin g
factor when multiple age groups are considered.

The computation of Tanner's "special circumstance" (Tanner, 1949) and other
diagnostic results are reported in Table 7.4. Body mass is significantly related to absolute
peak Vo2 in males (m 0.95) and females (r=0;87). As a group, there is a similarity
between r (body mass and absolute peak V02) and CV for boys. A ge—specific
calculations produce divergent ratios, especially in girls, suggesting a non-linear
relationship. As a group, the correlations between the simple ratio standard and body
mass are 0.07 and -0.41 in boys and girls, respectively. Correlations between scaled peak
Vo2 and body mass are 0.71 and 0.03 in boys and girls, respectively. Correlations
between absolute residuals and log body mass are 0.07 and -0.11 in boys and girls,
respectively. Age-specific correlations vary between the sexes with coefficients
approaching zero in some age groups when peak Vo2 is expressed per unit body mass
0.75. Correlations do not approach zero in any age group in females.

In general, the intra- individual (ontogenetic) linear regression shows a better fit
in boys than girls. In boys, 4 of 20 scaling factors are not significantly different from
zero (p>0.10). Logarithmically transformed peak Vo2 and mass are highly related (r

>0.85) in all but one male subject. In contrast, scaling factors are significantly different

178

from zero in 6 of 17 females. The relationship between logarithmically transformed peak
Vo2 and mass is high (r >0.85) in 8 females, and moderate (0.40-0.85) in 7 others. Based
_ on a combination of the correlation coefficients and least squares regression model, one
male and two female subjects were eliminated from the analysis.

Ontogenetic scaling factors show considerable variation (range, 0.51-1.31 and
0.29-0.90 in males and females, respectively). Five males exhibit scaling factors 20.99.
The mean (95% confidence interval) ontogenetic scaling factors are 0.81 (0.71-0.92) and

0.61 (0.50-0.72) in males and females, respectively (p=0.002 between group differences).

DISCUSSION

This study examined age- and sex—associated variation in peak Vo2 of 9-19 yr old
distance runners and provides unique information from three perspectives. First,
previous studies are generally limited to a relatively narrow age range (i.e., 11-15 yrs),
and therefore, do not describe growth-related changes in peak Vo2 across the entire
adolescent period. Second, only one longitudinal study (Baxter-Jones et al., 1993) has

included females across a broad age range in childhood and adolescence. No study has

included young distance runners of both sexes 9 to 19 years. Third, this study used
allometric scaling techniques to interpret the age- and sex-associated variation in peak
Vo2 of young distance runners.

The observed values for absolute and relative peak Vo2 expressed per unit body
mass in this sample of young distance runners are similar to those previously reported in
longitudinal studies of young endurance athletes (Figures 7.1 and 7.2). Relative peak Vo2

in females is somewhat lower than the cross-sectional data of MasSachusetts cross-

179

country runners (mean = 66 ml'kg"‘min"‘) (Cunningham, 1990), but higher than the
maturity-grouped values of swimmers ( mean = 51-52 ml'kg"min"‘) in the Training of
Young Athletes (T OYA) mixed-longitudinal study (Baxter-Jones et al., 1993).

The pattern of development of relative peak Vo2 in males is similar to that
reported by Daniels et al. (1978) until the age of 16 yrs, when a divergent pattern occurs
between studies. In the present sample, there is an increase in relative peak Vo2 whereas
there is a decrease in Daniels et al. (1978). Previous studies have also reported an age-
related increase in relative peak Vo2 of young athletes (Murase et al., 1981; Paterson et
al., 1987). The typical pattern of development in the general population of normal,
healthy boys is a steady value of approximately 52 ml‘kg"'rnin"l (Krahenbuhl et al.,
1985). The age-related increase has led some investigators to suggest an influence of
exercise training on peak Vo2 during growth and maturation (Murase et al., 1981).

Less information is available on the age-related trend in female athletes. In the
general population of normal, healthy females, relative peak Vo2 decreases during
' adolescence (Krahenbuhl et al., 1985). In the only study that reported age (maturity)-
specific values, relative peak Vo2 remains stable at about 52 rnl'kg"‘rnin"l in pre-, mid-,
and late-pubertal swimmers (Baxter-Jones et al., 1993). Results from the present study
show a relatively stable pattern between the ages of 9-15 yrs of age at 55-58 mljkg’h’nin"1
before decreasing in the oldest age groups to 52-53 ml‘kg"‘min"'. More evidence is
needed to establish if the age-related decline of peak Vo2 in adolescent females is
attenuated with exercise training.

Many authors have argued the interpretation of the growth-related changes in

peak Vo2 on the basis of theoretical and statistical limitations of the simple ratio standard

180

(Armstrong and Welsman, 1994; Baxter-Jones et al., 1993; Sjodin and Svedenhag, 1992;
Winter, 1996). Therefore, alternate statistical models, including allometric scaling,
ANOVA, and multilevel modeling, have been used in an attempt to create a "size-free”
expression of peak V02. The use of alternate models has resulted in different
interpretations of growth-related changes in peak V02. For example, Sjodin and
Svedenhag (1992) showed an increase in peak Vo2 expressed per body mass°‘75 from 3
yrs before peak height velocity until 1 yr after peak height velocity in 8 male distance
runners. Others have shown an increase in sealed peak Vo2 in the general population of
normal, healthy boys (Kemper and Verschuur, 1987 ; Rogers et al., 1995; Rowland et al.,
1997; Welsman et al., 1996). Armstrong and colleagues (Armstrong and Welsman, 1994;
Armstrong et al., 1998; Welsman et al., 1996) have used adjusted means produced from
ANCOVA (controlling for body mass) to explore age- and growth-related changes in
peak Vo2 of normal, healthy children and adolescents. The results generally indicate an
increase in adjusted means across age- and maturity-groups in males, but an increase in
adjusted means from prepuberty to puberty and similar values between pubertal and
young adulthood in females. The results suggest that peak Vo2 remains constant from
late adolescence into young adulthood in females.

Recently, multilevel modeling has been applied to investigate the growth-,
maturity-, and training-related changes in peak Vo._. (Baxter-Jones et al., 1993; Winter,
1996). Multilevel modeling attempts to partition the independent and multiplicative
effects of age, body size and composition, pubertal status, and exercise training on a
dependent variable (e.g., peak V02). Studies using multilevel modeling have

demonstrated size-independent effects of sex and maturity on peak Vo2 (Armstrong et al.,

181

1999; Baxter-Jones et al., 1993). Results from the TOYA study indicate that peak Voz,
controlling for age and body size, increases with pubertal status in male and female
athletes, although an increase between mid- and post-pubescent groups in males is not
evident in females (Baxter-Jones et al., 1993). The results are intriguing, given past
assumptions about growth-related changes in peak V02. However, despite acclaimed
usefulness in the interpretation of longitudinal data, the biological significance of the
results derived from the multilevel modeling approach is difficult to interpret.

Sex differences in peak Vo2 during growth and maturation are well documented in
the general population of normal, healthy children and adolescents (Armstrong and
Welsman, 1994; Krahenbuhl et al., 1985). Less information is available on age-specific
differences of young athletes due to the lack of longitudinal studies of female athletes and
the narrow age ranges reported in cross-sectional studies. A significant age x sex group
interaction in the present study indicates a progressive divergence in peak V02. A ge-
specific differences in absolute peak Vo2 are slight (134—186 ml‘min"') prior to age 14 yrs
when the difference increases sharply in each age group until reaching a mean difference
of 1000-1500 ml'min"'in the oldest age groups. Similar to absolute values, sex
differences are reduced prior to 15 yrs, and then increase sharply when peak Vo2 is
expressed per unit body massLo or body mass”. The mean ontogenetic scaling factor
was significantly different between male and female adolescent distance runners, which
is consistent with the literature. The difference in scaling factors probably reflects
variation in the rate of change in peak Vo2 with body mass.

Mean cross-sectional scaling factors are 1.01 in males and 0.85 in females. These

values are similar to those reported for body mass and peak Vo2 in cross-sectional

182

analyses of longitudinal data of other male athletes (McMiken, 1976; Paterson et al.,
1987) and cross-sectional analysis of 617 yr old males and females (Cooper et al., 1984).
However, mean scaling factors reported in the literature show considerable variability
(Eisenmann and Malina, in press). Age-specific scaling factors in this study show
considerable disparity with estimates for the total sample (Table 7.4). In boys, age-
specific scaling factors range from 0.52-0.90 and most conform to the theoretical values
of 0.67 and 0.75. In girls, age-specific scaling factors range from -0.09 to 1.41. In both
sexes, age-specific scaling models do not represent a good fit as indicated by adjusted r2
values and non-significant log-linear regression models. This observation probably
reflects the small range of body size within an age group (Calder, 1987), small age-
specific sample sizes, confounding influences of biological maturity status (Beunen et al.,
1997), and differences in body composition, especially among females. Indeed, when
multiple age groups were considered, scaling factors began to approximate the overall
sex-specific scaling factor.

Table 7.5 provides a summary of longitudinal studies utilizing ontogenetic
scaling. The mean ontogenetic scaling factor of 0.81 in males is considerably less than
previous studies of highly trained adolescent athletes (Paterson et al., 1987; Sjodin and
Svedenhag, 1992). In contrast, similar results have been obtained for active boys in the
Saskatchewan Growth Study (Beunen et al., in review) and early and late maturing boys
training in Polish sports schools (track, wrestling, or basketball) (Beunen et al., 1997).
The mean scaling factor in the present study is actually higher than that in late maturing
boys from the Polish sports schools. The subjects in the study by Rowland et al. (1997)

were described as "physically active and inclined towards sports participation" (p. 264)

183

based on a parental description. However, only one was engaged in regular aerobic
training (swimming). An explanation for such a high ontogenetic scaling factor (1.10) in
, this sample is unknown. Interestingly, the average cross-sectional exponent was only
0.53.

The mean ontogenetic scaling factor in female distance runners is higher than
maturity-grouped girls from Polish sports schools (track or rowing) (Beunen et al., 1997)
and lower than recreational sport participants (Rowland et al., 1997). Ontogenetic
scaling factors in 10 of 16 female distance runners are not significantly different from
zero, indicating that the growth of peak Vo2 is not related to growth in body mass. The
lack of fit in female runners also reflects a plateau or decline in peak Vo2 with age
(Beunen et al., 1997) as typically observed in female adolescents. Therefore, the higher
scaling factor found by Rowland et al. (1997) may be due to age—associated variation, as
the mean age at entry in their study was 9.2 yrs whereas most of the female subjects in
the present study entered at 12-14 yrs of age.

Previous studies also show considerable variability in individual scaling factors
(Table 7.5). The range in male distance runners (0.51-1.31) is similar with that reported
in the Saskatchewan Growth Study (0.56-1.18) (Beunen et al., in review). Sjodin and
Svedenhag (1992) show a range from approximately 0.85-1.20 in young male distance
runners aligned to peak height velocity. The range in female distance runners (0.29-0.90)
is similar to that reported for active girls (0.18-1.11) (Rowland et al., 1997).

It has been suggested that variability in scaling exponents is due to factors other
than body mass including: individual variation in geometric similarity, changes in the

ratio of leg muscle mass to body mass, differences in physical activity and/or training

184

level, and individual differences in rates of development of size-independent factors such
as skeletal muscle oxidative enzyme capacity or myocardial contractility (Rowland et al.,
1997). The last mentioned factors would suggest that qualitative changes in the
functional capacity of specific sub-components of the oxygen transport system also
contribute to the growth-related changes in peak V02. Genotype may also contribute to
the variability in the phenotypic expression of peak V02. Familial aggregation for peak
Vo2 in the sedentary state and the peak Vo2 response to exercise training has been
demonstrated in the HERTIAGE Study (Bouchard et al., 1998; Bouchard et al., 1999).
The identification of genes and mutations responsible for the heterogeneity of peak Vo2
are currently under investigation. Early evidence suggests that muscle-specific creatine
kinase (Rivera et al., 1997; Rivera et al., 1999), Na"-I(*-ATPase (Rankinen et al., 2000),
and angiotensin converting enzyme (Gayagay et al., 1998) are possible candidate genes,
Future cross-sectional and longitudinal studies examining allometric relationships
between body mass and peak V02 may benefit from the identification of specific
Candidate genes associated with peak V0,.

The observed variability in the ontogenetic scaling factors may be related to
maturity-associated variation in body mass and peak V02. Given the individuality of
timing and tempo of maturation, year-to-year changes in body mass and peak Vo2 may
have been masked by maturity effects. Maturity-associated variation in peak Vo2 has
been recently estimated using various statistical models (Armstrong etal., 1998; Baxter-
Jones et al., 1993; Beunen et al., in review; Beunen et al., 1997). Peak Vo2 increases at a
slightly higher rate in early and average maturing males than expected from the increase

in body mass (Beunen et al., in review; Beunen et al., 1997). In one study, the increase is

185

smaller than expected in later maturing boys (Beunen et al., 1997). In the present samme
of distance runners, differences in biological maturity were evident as determined by
skeletal age estimated from the hand-wrist x-ray obtained on the first visit. The mean
difference between chronological age and skeletal age was -0.52 in 12 males and -0.57 in
10 females. Unfortunately, an insufficient number of subjects were available for the
analysis of skeletal maturity. Future studies should consider maturity-associated
variation in peak V02.

A scaling factor less than unity indicates that the conventional simple ratio
standard (mlkg"min"') is erroneous. This tenet has theoretical, statistical, and empirical
grounds. Theoretically, dimensionality theory predicts that metabolic rate should relate
to body mass by a scaling exponent of 0.67. McMahon ( 1973) has proposed the 'elastic
similarity' model suggesting that elastic criteria impose limits on biological proportions
and metabolic rates. Based. on this model the theoretical value is 0.75. Early studies on a
wide range of animals from rodent to elephant indicated that a scaling factor of
approximately 0.74 best describes the relationship between body size and resting
metabolic rate (Brody, 1945; Kleiber, 1932). Taylor and colleagues (1981) found that
peak V'o2 scaled approximately to 0.75in wild African and domestic mammals ranging
from 0.5 kg (dwarf mongoose) to 263 kg (Zebu cattle). Statistically, Tanner (1949)
addressed the fallacious and misleading practice of expressing physiological
measurements per unit of body mass or per unit of surface area.

-0.75

When expressed per kg a different interpretation of the growth-related changes
in peak Vo2 of young athletes is evident. Although peak Vo2 remains relatively stable

during adolescence in young male distance runners when expressed as a ratio standard, it

186

increases when expressed per kgm’. This finding confirms previous results across a
narrower age range (Sjodin and Svedenhag, 1992). In females, peak V02 expressed per
kg‘175 increases from 9 to 15 yrs and then shows a decline.

Most important to this study is the identification of an appropriate model to
interpret growth-related changes in peak Vo2 of young distance runners, and children and

adolescents in general. Several authors argue that peak Vo2 should be expressed in

0.67. l

accordance with theoretical values according to dimensionality theory (i.e, ml'kg min '
or ml‘kg °'75'min " ) (Armstrong and Welsman, 1994; Heil, 1997 ; Katch, 1973; Nevill,
1994; Rogers et al., 1995; Svedenhag, 1995). The first step in the investigation of
appropriate scaling procedures should involve the calculation of Tanner's "special
circumstances (Batterham et al., 1997). If r is equal, or approximately equal, to the ratio
of the coefficient of variations, a linear relationship is evident and the simple ratio
standard (ml‘kg'l'min") is appropriate. Conversely, if these two terms are not similar, a
linear relationship does not exist and the appropriate power function ratio should be
calculated. Other regression diagnostics used in this study (i.e., correlations between
residuals, simple and power function ratios, and body mass) were used to examine if the
influence of body mass was removed (i.e., the correlation should not be different from
zero if the influence of body mass has been removed) (Batterham et al., 1997 ; Welsman
et al., 1996). Based on these criteria, the simple ratio standard could be empirically
justified in males, while the power function ratio could be empirically justified in females

(Table 7.3). Other authors (Bar-Or, 1983; Batterham et al., 1999) have also concluded

that the mass exponent for peak Vo2 is close to unity.

187

In conclusion, the results of this study suggest that the interpretation of growth-
related changes in peak Vo2 of young distance runners is dependent upon the expression
of peak Vo2 relative to body size and/or the statistical technique employed. Considerable
variability in individual growth patterns in scaled peak Vo2 points to the fact that
determining a single scaling factor is difficult and may actually be problematic given the
genetic, environmental, and genetic-environmental interactions that influence peak V02.
The most appropriate means of normalizing peak Vo2 for body size still remains
problematic (Rowland, 1998; Rowland et al., 1997). Exercise scientists have been
criticized for not recognizing the imperfections of ratio standards and being unaware of
alternative methods for partitioning the effects of body size in human studies (Winter,
1996). However, it remains to be demonstrated that allometric scaling among a small
magnitude of variation in body size warrants such statistical manipulation. According to
Calder (1987), small size ranges within a species obscure overall trends, patterns, and
constraints of size. Thus, scaling differences in body size among a small range of body
size to understand variation in biological function may be of limited value. In contrast,
others argue that scaling body size helps us to understand the growth and maturation of
the oxygen transport system, and its response to submaximal and maximal exercise
(Armstrong and Welsman, 1994). To solve the problem of the structural and functional
consequences of changes in size or scale among growing and maturing children and
adolescents, pediatric exercise scientists should perhaps collaborate with comparative
mammalian physiologists for whom the statistical tool of allometry has been central for

many years.

188

ACKNOWLEDGEMENTS

This study was supported in part by the William Wohlgamuth Memorial

Fellowship and the Institute for the Study of Youth Sports at Michigan State University.

Special thanks to Vern Seefeldt, Wayne Van Huss, Bill Heusner, and other members of

the Human Energy Research Laboratory for data collection in YRS I and YRS II.

189

”TI-Hm.

REFERENCES

Armstrong N, Welsman JR (1994) Assessment and interpretation of aerobic fitness in
children and adolescents. Exerc Sport Sci Rev 22:435-476.

Armstrong N, Welsman JR, Kirby BJ (1998) Peak oxygen uptake and maturation in 12-yr
olds. Med Sci Sports Exerc 30:165-169.

Armstrong N, Welsman JR, Nevill AM, Kirby BJ (1999) Modeling growth and
maturation changes in peak oxygen uptake in 11-13 yr olds. J Appl Physiol 8722230-
2236.

Bar-Or O (1983) Pediatric Sports Medicine for the Practitioner.New York: Springer-
Verlag.

Batterham AM, George KP, Mullineaux DR (1997) Allometric scaling of left ventrcular
mass by body dimensions in males and females. Med Sci Sports Exerc 29:181-186.

Batterham AM, Vanderburgh PM, Mahar MT, Jackson AS (1999) Modeling the
influence of body size on Vozpeak: effects of model choice and body composition. J
Appl Physiol 87: 1317-1325.

Baxter-Jones A, Goldstein H, Helms P (1993) The development of aerobic power in -
young athletes. J Appl Physiol 75:1160-1167.

Beunen G, Baxter-Jones A, Mirwald RL, Thomis M, Lefevre J, Malina RM, Bailey DA
(in review) Intra-individual allometric development of aerobic power in 8 to 16 year-old
, boys: association with maturity level and activity level.

Beunen GP, Rogers DM, Woynarowska B, Malina RM (1997) Longitudinal study of
ontogenetic allometry of oxygen uptake in boys and girls grouped by maturity status. Ann
Hum Biol 24:33-43.

Bouchard C, Daw EW, Rice T, Pérusse L, Gagnon J, Province MA, Leon AS, Rao DC,
Skinner JS, Wilmore JH (1998) Familial resemblance for Vozmax in the sedentary state:

the HERITAGE family study. Med Sci Sports Exerc 30:252-258.

Bouchard C, Ping A, Rice T, Skinner J S, Wilmore JH, Gagnon J, Pe’russe L, Leon AS,
Rao DC (1999) Familial aggregation of Vozmax response to exercise training: results

from the HERITAGE Family Study. J Appl Physiol 87:1003-1008.

Brody S (1945) Bioenergetics and Growth, with Special Reference to the Efficiency
Complex in Domestic Animals. New York: Reinhold.

190

Cooper DM, Weiler-Ravell D, Whipp BJ, Wasserrnan K (1984) Aerobic parameters of
exercise as a function of body size during growth in children. J Appl Physiol 56:628-634.

Cunningham LN (1990) Physiologic comparison of adolescent female and male cross-
country runners. Pedatr Exerc Sci 2:313-321.

Daniels J, Oldridge N, Nagle F, White B (1978) Differences and changes in V02 among
young runners 10 to 18 years of age. Med Sci Sports 10:200-203.

Eisenmann JC, Haubenstricker JL, Seefeldt VD, Malina RM (1999) Growth status and
estimated growth rates of competitive young distance runners. Med Sci Sports Exerc 31
(Suppl.):S 169 (Abstract).

Eisenmann JC, Malina RM (in press) Body size and endurance performance. In RJ
Shephard (ed.): Endurance in Sport. Oxford: Blackwell Science.

Gayagay G, Yu B, Hambly B, Boston T, Hahn A, Celerrnajer DS, Trent R (1998) Elite
endurance athletes and the ACE 1 allele - the role of genes in athletic performance. Hum
Genet 103:48-50.

Gould SI (1966) Allometry and size in ontogeny and phylogeny. Biol Rev 41:587-640.

Hamill PVV, Johnson CL, Reed RB, Roche AF (1977) NCHS growth curves for children
birth-18 years, United States: Vital and Health Statistics.

Heil DP (1997) Body mass'scaling of peak oxygen uptake in 20- to 79-yr-old adults. Med
Sci Sports Exerc 29: 1602- 1608.

Jones NL (1984) Evaluation of a microprocessor-controlled exercise testing system. J
Appl Physiol 57: 1312-1318.

Katch VL (1973) Use of the oxygen/body weight ratio in correlational analyses: spurious
correlations and statistical considerations. Med Sci Sports 5:253-257.

Kemper HCG, Verschuur R (1987) Longitudinal study of maximal aerobic power in
teenagers. Ann Hum Biol 14:435—444.

Kleiber M (1932) Body size and metabolism. Hi1 gardia 6:315—353.

Krahenbuhl GS, Skinner JS, Kohrt WM (1985) Developmental aspects of maximal
aerobic power in children. Exerc Sport Sci Rev 13:503-538.

Maingourd Y, Libert JP, Bach V, Jullien H, Tanguy C, Freville M (1994) Aerobic
capacity of competitive ice hockey players 10-15 years old. Jap J Physiol 44:255-270.

McMahon T (1973) Size and shape in biology. Science 179:1201-1204.

191

McMiken DF (1976) Maximum aerobic power and physical dimensions of children. Ann
Hum Biol 3: 141-147.

Murase Y, Kobayashi K, Kamei S, Matsui H (1981) Longitudinal study of aerobic power
in superior junior athletes. Med Sci Sports Exerc 13:180-184.

Nevill AM (1994) The need to scale for differences in body size and mass: an
explanation of Kleiber's 0.75 mass exponent. J Appl Physiol 65:110-117.

Paterson DH, Cunningham DA, Donner A (1981) The effect of different treadmill speeds
on the variability of V°2max in children. Eur J Appl Physiol 47: 1 13-122.

Paterson DH, McLellan TM, Stella RS, Cunningham DA (1987) Longitudinal study of
ventilation threshold and maximal 02 uptake in athletic boys. J Appl Physiol 62:2051-
2057.

Rankinen T, Pérusse L, Borecki I, Chagnon YC, Gagnon J, Leon A, Skinner JS, Wilmore
JH, Rao DC, Bouchard C (2000) The Na+-K+-ATPase alpha2 gene and trainability of
cardiorespiratory endurance: the HERITAGE Family Study. J Appl Physi0188:346-351.

Rivera MA, Dionne FT, Simoneau J-A, Pérusse L, Chagnon M, Chagnon Y, Gagnon J ,
Leon AS, Rao DC, Skinner JS, Wilmore JH, Bouchard C (1997) Muscle-specific creatine
kinase gene polymorphism and Vozmax in he Heritage Family Study. Med Sci Sports
Exerc 29:1311-1317.

Rivera MA, Pérusse L, Simoneau J-A, Gagnon J, Dionne FT, Leon AS, Skinner JS,

Wilmore JH, Province M, Rao DC, Bouchard C (1999) Linkage between a muscle-
specific CK gene marker and Vozmax in the HERITAGE Family Study. Med Sci Sports

Exerc 31:698-701.

Rivera-Brown AM, Rivera MA, Frontera WR (1994) Achievement of V°2max in
adolescent runners: effects of testing protocol. Pediatr Exerc Sci 6:236-245.

Rogers DM, Olson BL, Wilmore JH (1995) Scaling for the VoZ-to-body size relationship
among children and adults. J Appl Physiol 79:958-967.

Rowland TW (1990) Developmental aspects of physiological function relating to aerobic
exercise in children. Sports Med 10:255-266.

Rowland TW (1998) The case of the elusive denominator. Pediatr Exerc Sci 10:1-5.

Rowland TW, Vanderburgh PM, Cunningham L (1997) Body size and the growth of
maximal aerobic power in children: a longitudinal analysis. Pediatr Exerc Sci 9:262-274.

192

Seefeldt VD (1986) Elite young runners: an interdisciplinary perspective. In M Weiss
and D Gould (eds.): Sport for Children and Youths. Champaign, IL: Human Kinetics, pp.
213-284.

Sjodin B, Svedenhag J (1992) Oxygen uptake during running as related to body mass in
circumpubertal boys: a longitudinal study. Eur J Appl Physiol 65:150-157.

Skinner JS, Bar Or 0, Bergsteinova V, Bell CW, Royer D, Buskirk ER (1971)
Comparison of continuous and intermittent tests for determining maximal oxygen intake
in children. Acta Paediatr Scand 217:24-28.

Svedenhag J (1995) Maximal and submaximal oxygen uptake during running: how
should body mass be accounted for? Scand J Med Sci Sports 5:175-180.

Tanner JM (1949) Fallacy of per-weight and per-surface area standards, and their relation
to spurious correlation. J Appl Physiol 2: 1-15.

Taylor CR, Maloiy GMO, Weibel EW, Langman VA, Kamau JMZ, Seeherrnan HJ,
Heglund NC (1981) III. Scaling maximum aerobic capacity to body mass: wild and
domestic mammals. Respir Physiol 44:3-37.

Weiner J S, Lourie JA (1969) Human Biology: A Guide to Field Methods. Oxford:
Blackwell Science.

Welsman JR, Armstrong N, Nevill AM, Winter EM, Kirby BJ (1996) Scaling peak V02
for differences in body size. Med Sci Sports Exerc 28:259-265.

Winter EM (1996) Importance and principles of scaling for size differences. In 0 Bar-Or
(ed.): The Child and Adolescent Athlete. Oxford: Blackwell Science, pp. 673-679.

193

Hezu q... >mu-m_»uu=..u <35... 3.. 963. «Eu 3.. vuew <6N 3 3e_u 33.58 3.32.? Zoe... usage... nuiezo...
e3. ..eamu e..u @3138 m9. e: 3.328 2qu ”Esau.” 38.5 ..6.. week <6».

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

> mu .. I. 839 (5 Awmv _uuew <6» >&. .58.. muer <6» muew <6»
m3...» A3339 vuew <6» A3_.wm._.33._v A3_._nm_d3:.._v
A333,;
we...
0. _o a 5.8m 3.9 w _ .N AAAV _kob ANNANV N486 aNQ a. 5 Add A 5.9
39¢. 59— Nmbhwwb _AMOINNae MPMQNM 5mm. Sub
_ _ _N _Aw.m 3.9 wwh AAAV 3 5b A899 Macaw mud 3. C Gnu :99
_u _ .O. _ mab NA.M.u©.N 5.3-8.5 mwbhwh _NAb. 39%
_N —A Gob 3.3 web 3.9 NAGP— AAANQV §.A mmb 3.9 _quA _ANV
_uAQ- _mAM Nmb-mw.o _8omamo MNOQOQ Gob- 5mm
5 a 39% 3.9 AVA 3.9 Nawmh 38.9 waob 00.x 3.9 _mwA 399
50%. 30.0 Nmemb _Sobqeo Am.m..: .o 59m. _omb
_A No 3N— thv Amd 20.9 “sq—I). ammbv mon— .m cub 3.9 _mah :Auv
_Awu- _qu w _ .O.Gw.N _cakomo mo. 7th Cam. EMA
_m _m Sch 3.9 man. :39 waANh 33.9 m SOQ .ONQ 3.9 3mm 20.9
Gob- _ww.m wmbfimb NNoolAaww mm bud.— _Ao.c-n_ _.m
3 HA 3A; 3.9 8; 3.9 wmmAb AAAm.9 uNumb OPW 3.9 30% :N.9
59o. 36b Amunuw .A woqummw mm .MQNU _Aqu - Bad
3 No sub 3.9 am: And ANAm.o QQNDV qumb aflm 3.9 _dw AN. .3
_8b. _mAQ MMAIE .m waq _ -MONN mm. Tum; _AA. TN _ Nb
5 _A Gab 3.: mum 3.9 Ammad Ammmbv wAOAd mum 8.9 _2 .O ANNQV
39? _mm._ a .ohab quo-MAmN AA. _QMM _NOUIN 3%
4.68— 50

 

 

.320 SN. >mo-mvoo.mo 5.38 3.. .63. .50 E... can.” <oN ... 38.30 .5858 2:52.? 2.25. 3.53:. 32.30...
E... Rama ..8 3352. 3.. a: $53.0... axon... 3.5.8.. 325... 3.. .5...” <0».

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

>mo .. I. Ana. «5 Arm. .uomw <oN >a.=m§. 1am.» <o~ 12..» <0N
«8.... 013...... wow—n <0N AB..wm._3.=._v A..._,..m..d.3.=._.
A 3.43:...
0......
O. .0 O .AA.AA¢.3 wN.MAA.3 .moch .WN.3 N . 0M.M M9”. 3.9 .3. .m A . N.M.
.wN3- .MN.A NM.9 wmu .A8uNOM0 K3.0-3M3 . .Mb- .MAb
.. . . .Amb A93 wwh 3.9 .93.“. Awa. .3 NN03.. M90 AM.3 .wmb A .N.Mv
53.9—M33 NN.c-AM.M . .30-N530 AflM- 39m . . ..m- .MM.A
.N .M .Mw.m 3.9 A0. 3.3 NN8.0 3.33.3 Nuqob M13. . 3.3 33.0 A .N.9
.AA.M- .8.A Amb-Awmb .moa-wao Awe-$0 .NN.M- .GMG
.w .3 .Mmb 3.: AM.0 3.3 NAA..N 33.9 3200 MA.m 3.3 .A..M A.w.3
.AOA» .300 woM-Mflq NOWAYquo ANA-3M3 . 3.0-3. .3
.A .A .3..w 333 A33 AA.3 N3mm3 33.3 N330 M90 A “AV .M. .. AN. .9
.Mw 3- .3M. A0.a-MA.u NONO-w0Ao wow-$3 .03.“..- . mu .m
.M . . .333 3.3 M0.A AM.AV gmb ANQA.3 N3NM.A M3.N 3.9 .M. .0 A .M.3
.M. .A- .30.. Aw.M-3. .0 NwMN-wwg AM.M-3m.. .NMb- ..uaA
.3 .N .3Mb AWAV MAG A93 NwMA.m 38.3 NMA. .w MAD A93 .A. .. 20.9
.Mab- . m. .N 3.33M.— NAm0-wA30 u0.M-30.0 . .0.N- .Mm.0
.3- .w .3 .awN AWAV MM.M 3.3 N®©A3 Ammobv NaAQb M. .m 3.3 .wON 23.3
.MN.N- .mN. Ago-33 N .00-waM A. .0390 . 5.0- .33.0
HOS. .0®

 

 

195

Table 7.3. Age- and sex-specific proportionality coefficients (a) and allometric scaling
factors (b) in young distance runners.

 

 

Age group a b Multiple r Adjusted r2 b'
Boys
9-10 5.36 (0.65) 0.64 (0.19) .71 .46 -
l l 5.41 (0.83) 0.64 (0.23) .65 .36 0.67 (0.09)
12 5.11 (0.52) 0.73 (0.14) .84 .68 0.78 (0.08)
13 4.89 (0.69) 0.79 (0.18) .75 .53 0.80 (0.07)
14 4.54 (0.34) 0.90 (0.09) .93 .86 0.77 (0.15)
15 4.67 (0.60) 0.87 (0.15) .85 .70 0.92 (0.09)
16 5.07 (0.84) 0.78 (0.20) .77 .55 0.94 (0.04)
17 6.18 (1.05) 0.52 (0.25)* .46 .16 0.99 (0.03)
18 5.36 (1.48) 0.72 (0.34)* .51 .20 -
Total 4.12 (0.12) 1.01 (0.03) .95 .89

(95 CI, 0.96-1.03)
Girls
9-10 6.01 (0.68) 0.43 (0.19)* .64 .32 -
l l 4.46 (0.14) 0.88 (0.14) .89 .79 0.76 (0.12)
12 5.18 (0.71) 0.69 (0.19) .70 .45 0.82 (0.08)
13 6.46 (0.57) 0.35 (0.15) .53 .23 0.75 (0.06)
14 8.26 (1.29) -0.09 (0.32)* .09 -0.08 0.77 (0.06)
15 6.91 (1.09) 0.26 (0.28)* .30 -0.01 0.78 (0.06)
16 7.36 (1.28) 0.14 (0.32)* .19 -0.12 0.78 (0.05)
17+ 2.40 (1.15) 1.41 (0.28) .72 .50 -
Total 4.58 (0.18) 0.85 (0.05) 87 .75

(95 CI, 0.76-0.94)

 

b', cumulative scaling factor when an additional age group is considered (i.e., 9-10 yr
plus 11 yr; 9-10 yr, 11 yr plus 12 yr, etc.).
Values are mean (SE). *p>0.05

196

Table 7.4. Diagnostic criteria for the relationships

 

 

between peak Voland body size.

Age group CV R1 R2 R3
Boys

9—10 1.22 0.73 -0.51 —0.21
l 1 0.97 0.63 -0.44 -0. 14
12 1.02 0.83 -0.45 0.04
13 0.87 0.71 -0.26 0.17
14 0.98 0.92 -0.25 0.41
15 1.14 0.81 —0.28 0.55
16 1.01 0.78 -0.34 0.00
17 0.85 0.48 -0.42 -0.04
18 0.72 0.54 -0.21 0.00
Total 0.90 0.93 0.07 0.71
Girls

9— 10 1.47 0.62 -0.74 -0.53
1 1 1.1 1 0.88 -0.33 .021
12 1.09 0.75 -0.40 -0.09
13 1.57 0.51 -0.74 -0.57
14 0.97 -0.09 -0.75 -0.68
15 1.1 1 0.35 -0.65 -0.64
16 1.24 0.24 -0.74 -0.47
17+ 0.53 0.73 03 l 0.43
Total 1.08 0.87 -0.41 0.03

 

CV, ratio of coefficient of variation; R1, correlation

coefficient between body mass and absolute peak
V02; R2, , correlation coefficient between body
mass and relative peak V02; R3, , correlation

coefficient between body mass and scaled peak V02.

197

flaw—n NM. 9.3an 0m flan—«So oiomaaozo mow—Emannoa 3 2:58: man moo—88:8.

 

 

macaw wagons” 2.3: mow—Em @084 ”mama
macaw: m3 gang—Sm m "852— ammsgo «5:53 _0— 0.mm._.~0
200$ A 5:852. 0.qm 0.00.0.mm
09830: on a 3003 E San enigma 0. _0
macaw: 2 a. 2003 mun—Egoama Susi—i Baum 0.m0
A=H—$
58 Bushman Baum gum C 0.!
mgfiméama Ems—«mam «cam—am 0.3
gnu:
.58 32:15 «nae—015100 0.3
woe—8: a. E. a: 3525 4.08— ?uflwv 0.00 000;. _m
me; 33:32 9". 5 00m
963m... Bug—dam €nt 0.0m
~88 53:88 Ann—0V 0.00
52:76 Ann—N0 0.00
><a8mo 0.83 0.00
>35 Ann—Av 0.3
”Sign a. a. 2003 2 Big _._0 0.49:3
0 moan—8 000 0. 5.:—
deoa .430. N0 Baa 036:8 2583 0.9 0.2.70.
3 3:5? ammnmaon 35.88 0.0. 0.8-0.00

 

198

V02 (ml/min)

V02max (ml/kg/min)

4500 -'

 

 

Present study
Cunningham et al. (I987)
Daniels et al. (1978)
Murase et al. (1981)

Baxter-Jones et al. (1993)

 

 

 

 

 

Age (yrs)

Daniels et al (1978) — boys
Cunningham et al.(l989) - boys
Sjodin and Svedenhag (1992) - boys
Baxter-Jones ct al (1993) - boys

Baxter-Jones et al (1993) -girls

4250 -‘
4000 -'
3750 q
3500 -'
3250 -
3000 -‘
2750 d
2500 d
2250 4
2000 -'
1750 I T I l I
8 10 12 14 16 18 20
Age (yrs)
Figure 7.]. Longitudinal studies of absolute peak V02 in male athletes.
70 -
65 -
60 - ' ' '5' " '
-.-..-.-.
55 - ---o---
.. .. V— -
50 -*
_._a..-
45 -‘ —0— Present study - boys
'"""*"'"“ Present study - girls
40 I I I I I I
8 10 12 14 16 18 20

Figure 7.2. Longitudinal studies of relative peak Vo2 in young athletes.
Solid lines represent age-related changes in the general population.

199

CHAPTER8

SUMMARY AND RECOMMENDATIONS

200

This dissertation focuses on blood lipids and peak Vo2 in young male and female
distance runners. The studies are based on two independent samples of young distance
runners from the mid-Michigan area. Both studies were supported in part by the Institute
for the Study of Youth Sports. The mixed-longitudinal cohort of the Young Runners
Study I (YRS I) (1982-1986) included 27 males and 27 females. Twenty males and 18
females were followed annually for 3 to 5 years, and the remaining subjects were
followed for 1 or 2 years. A total of 99 and 84 observations were available for analysis.
This study formed the basis for age-specific analyses of blood lipids and peak V02.
During 1999-2000, a cross-sectional study (Young Runners Study 11, YRS II) was
undertaken to specifically address the dose-response relationship between training
volume and blood lipids, specifically hi gh-density lipoprotein (HDL), in young distance
runners. This sample included 48 males and 22 females.

Age- and sex-associated variation in blood lipids of 9 to 18 yr olds from the YRS
I were initially described (Chapter 3). The development of blood lipids in young distance
runners appears to be similar to the general population - TC and LDL remain stable, HDL
declines during adolescence (especially in males), and TG increases with age. One of
the major study hypothesis was that the decline in HDL during male adolescence would
be attenuated in young distance runners given the high levels of exercise training. The
lack of the attenuation may lend to the robustness of normal growth and maturation,
including genes, hormones, and fat distribution, on the development of HDL in males
regardless of exercise training. This sample also did not display a superior blood lipid
profile compared to age- and sex-specific reference values for United States youth,

except for higher HDL in male and female runners prior to agel4 yrs. Unlike

201

anthropometric and functional capacity variables, there was considerable variability in
blood lipids, including dyslipidemic values. These results were also supported by the
findings from the YRS 11 (Chapter 4).

Heterogeneity in blood lipids among young distance runners was considered in
Chapter 5. Determinants included training volume (km per wk), peak oxygen
consumption (peak V02, ml'kg’lmin"), and body fatness. YRS II was originally designed
to address the dose-response relationship between training volume and blood lipids.
Young distance runners were identified as a "special exposure group" to test the
hypothesis that physical activity at levels greater than the current recommendations
would result in accrued health benefits as shown in adults. Results did not indicate that
increased weekly running distance was related to blood lipids in young distance runners.
TV may be indirectly related with HDL through its relationship with peak Vo2 in males.
A unique finding was the differential relationships between TV and HDL when the entire
sample was grouped according to modified clinical cut-points. Specifically, TV was
significantly related to HDL in subjects with HDL<45 mgldl (1:040, p<0.05). This
finding supportsjthe notion that increased levels of training do not influence lipoprotein
metabolism when blood lipids are already desirable during childhood and adolescence.

The complex inter-relationships between body fatness, peak V02, and HDL were
further explored. Partial correlations indicate that the association between peak Vo2 and
HDL remained significant after controlling for the concomitant variation in SSF and
explained 9% of the variance in HDL. The association between SSF and HDL did not
remain significant after controlling for the concomitant variation in peak V02. This

finding would suggest that skeletal muscle properties are an important factor in

202

determining HDL in young well-trained distance runners, although the influence of
adipose tissue cannot be dismissed.

The role of genes, peak V02, and body fatness in the modulation of elevated blood
lipid levels has also been indicated. The correlation between parental BMI and measures
of fatness and blood lipids in adolescent distance runners were positive. Therefore, it is
possible that blood lipids in adolescent distance runners are moderated by the genetic
contribution of body fatness and/or the pleiotropy (shared genes) of body fatness and
blood lipids. It is also possible that shared environmental factors (i.e., dietary intake)
could contribute to the relationship between parental fatness, offspring fatness and blood
lipids.

The results from Part I indicate that the phenotypic expression of blood lipids in
young distance runners is influenced by multiple factors. The contribution of growth,
maturation, genetics, and skeletal muscle and adipose tissue properties on lipoprotein
metabolism during adolescence are definitely as important, if not more important, than
exercise training.

Part 11 examined the usefulness of allometric scaling as a tool in the interpretation
of age- and growth-related changes in peak V02. As expected, an age-related increase in
absolute peak Vo2 occurred in both sexes with sex differences emerging during
adolescence. However, the ability to "standardize" peak Vo2 for comparative purposes is
important to understanding the growth of the oxygen transport system, and its
relationship with health-related fitness variables and endurance performance. When
expressed per unit body mass, peak Vo2 (ntl'kg"‘rrtin"‘) remains stable until age 15 when
.4175

it increases in boys, and decreases in girls. In contrast, relative peak Vo2 (ml'kg min"')

203

increases throughout the age range in boys and increases in girls until age15 yrs, and peak
Vo2 adjusted for body mass (ml'min") increases with age in boys and girls. Allometric

‘ scaling factors varied by analytical methods. The overall mean cross-sectional scaling
factor was 1.01:0.03 (SE) in boys and 0.85:0.05 (SE) in girls. Mean ontogenetic
allometric scaling factors were 0.81 and 0.61 in males and females, respectively. Thus, it
was concluded that the interpretation of growth-related changes in peak Vo2 of young
distance runners was dependent upon the manner of expressing peak Vo2 relative to body
size and/or the statistical technique employed.

The results from Part 11 do not answer the question, ”What is the most appropriate
means of normalizing peak Vo2 for body size?". Rather, the results point out the fact that
the problem still remains. Pediatric exercise physiologists are confronted with the
dilemma of accounting for differences in body size and functional capacity during growth
and maturation. To understand the development of the oxygen transport system (and
other functional capacities) first requires an understanding of human growth and
maturation. In order to understand the functional consequences of a change in body size
also requires an understanding of allometry and more recently, advanced statistical
techniques such as multilevel modeling. Although exercise scientists have been criticized
for not recognizing the imperfections of ratio standards and being unaware of alternative
methods for partitioning the effects of body size in human studies, it remains to be
demonstrated that allometric scaling among a small magnitude of variation in body size
warrants such statistical manipulation. Thus, scaling differences in body size among a
small range of body size to understand variation in biological function may be of limited

value. To solve the problem of the structural and functional consequences of changes in

204

size or scale among growing and maturing children and adolescents, pediatric exercise

scientists should collaborate with comparative mammalian physiologists for whom the

statistical tool of allometry has been central for many years.

RECOMMENDATIONS FOR FUTURE RESEARCH

An exciting part of science is the continuing cycle of research. Although this

dissertation has added insight into the influence of growth, maturation, and exercise

training on blood lipids and peak Vo2 in young distance runners, it has also served as a

catalyst for future studies.

In general, future research should examine the complex inter-relationships
between growth, maturation, genetics, exercise training, skeletal muscle and
adipose tissue properties and lipoprotein metabolism.

A ”pure” longitudinal or mixed-longitudinal study of blood lipids is warranted.
Such a study would re-examine if exercise training attenuates the decline in HDL
during male adolescence, and if it does not - why? Measurements of sexual
maturity, sex hormones, and body fatness (including visceral fatness) should be
included to identify possible biological mechanisms.

Familial resemblance of aerobic fitness, body fatness, and blood lipids may prove
to also explain some of the variation in the blood lipid phenotype. The
measurement of these variables on parents of young distance runners would be
valuable. Furthermore, the identification of apolipoprotein E phenotypes may
also prove beneficial in examining the inter-relationships between exercise

training, peak V02, body fatness, and blood lipids.

205

What is the time course of the development of high levels of HDL in adult
endurance athletes? A longitudinal study of collegiate distance runners may add
insight into this question.

The blood lipid profile in large samples and elite samples of sport-specific youth
have not been explored, nor has the prevalence of dyslipidemia in youth athletes
been adequately addressed. In young distance runners, it would be interesting to
study national competitors.

Investigations of the HDL and LDL sub—classes and apolipoproteins would add to
our understanding of lipoproteins in young athletes.

Measurement issues, such as day-to-day variation, need to be addressed. Perhaps,
2-3 measurements over a short time period should be averaged to represent the
"true" blood lipid phenotype.

The measurement of training volume also requires attention. Perhaps, estimates
based on longer time periods are more appropriate. Training intensity should also
be included in the derivation of training volume. Dietary information should also
be included in future studies.

The maturity-associated variation in blood lipids has yet to be established in
highly trained youth endurance athletes.

Prospective training studies examining the exercise training response in young
athletes who possess dyslipidemic values are necessary.

It is important for biostatisticians to communicate the biological relevance of
multi-level modeling in the interpretation of growth- and training-related changes

in peak Vo2 and other physiological parameters.

206

' Appropriate animal models need to be developed to explore the growth- and
training-related changes in the oxygen transport system, including the structure-

function relationships in lung, heart, blood, and skeletal muscle.

207

APPENDIX A

SELF -REPORTED TRAINING VOLUME

208

HEEL—2Q EmfiO—wd

2:32

 

£3. :5 game—=8 em m 02.2... «53:5. 3&2. 88? E88 SE88 05.: "33:5 EmBQ .858.

 

K98: >3.— .Sww .38 .37. >=m met. 09

 

«$5. £8 :8
983m...

:5: .84 Q
U><m an!
saw—n :5.
ko: 5:83
£25. £8 :8
368%

:5: can 2
Erma 3w”
inn—n 9w.
1%: 556%
$05. .88.:
cm «9:
555.0 88
an. s

<55. 88:.
2 05:4
5555 $8
Sax-«.380
«$5. .882:
0.. <9:
"33:5 :8
anti»

 

 

 

 

 

 

 

 

 

 

 

 

 

 

$825. com, ME: 030

 

209

 

APPENDIX C

CONSENT FORM

210

Informed Consent Form

Influence of intensive training on blood cholesterol
in young distance runners

Heart disease occurs in adults, but some children have high levels of blood pressure and
cholesterol that may cause heart disease. Regular physical activity has beneficial effects on
heart disease in adults. However, the effects of physical activity on cholesterol in children
is not clear. Since sports are important to daily physical activity in kids, this study will
show how important it is to be active in youth sports.

We are studying the effects of running mileage and cardiovascular fitness on blood
cholesterol levels in young distance runners 10-18 yrs of age. To be in the study, runners
must have been training for the previous 3 months or more. Participation in the study will
include measurement of body size, blood pressure, blood cholesterol, cardiovascular
fitness, self-assessed sexual maturity status, medical history, and physical activity. Body
size measurements Ml] include height, weight, and fatness. Body fatness will be measured
on the arm, back, stomach, and leg by a small pinch of the skin. Blood pressure will be
measured with a cuff around the arm. Blood cholesterol will be measured by a small
fingerprick. Sexual maturity will be assessed by yourself in a private setting by choosing
illustrations that look most like you- Cardiovascular fitness, a good predictor of endurance
performance, will be measured by a best effort on a treadmill. Testing will be conducted in
the morning following a period in which you cannot eat for 12 hours. Food will be
provided immediately following the measurement of blood cholesterol. We ask that the
parent or guardian assist in completing the medical history and physical activity
questionnaires. All information will be confidential and shared with the subject. Testing
will take about 1 1/2 hours in the Human Energy Research Laboratory at Michigan State
University.

There are minor risks involved in the testing. The risks of participation are no greater than
completing a medical exam or physical fitness test. I may experience slight physical
discomfort during the measurement of body fatness, blood pressure, and blood cholesterol.
I may feel uneasy about assessing my own sexual maturity. During the exercise test, I may
become tired and faint. However, the exercise test will be like running during practice or a
race- I understand that participation is voluntary and I may choose not to participate at all,
refuse to participate in certain tests, or discontinue participation at any time without penalty.
Every effort will be made to protect me from having harmful problems during the testing.
Pregnant women should not participate in the study. I understand that if I am injured as a
result of my participation in this research project, Michigan State University will provide
emergency medical care if necessary. I further understand that if my injury is not caused
by negligence of MSU I am personally responsible for the expense of this emergency care
and any other medical expenses incurred as a result of my injury.

The benefits of my participation in this study. include receiving information about my body
size, blood pressure, blood cholesterol, and cardiovascular fitness the day of testing and a
complete copy of my information in the mail. I will also be mailed a summary of the
results of this study upon completion.

211

If you or your parents have any questions about the study, you may contact the following
individuals:

Joey C. Eisenmann Robert M. Malina, PhD.

Research Assistant Director

Institute for the Study of Youth Sports institute for the Study of Youth Sports
Michigan State University Michigan State University

East Lansing, Ml 48824 ‘ East Lansing, Ml 48824

(517) 432-1416 (517) 355-7620
eisenma3@pilot.msu.edu rmalina@pilot.rnsu.edu

or

David Wright, Ph.D., Chair

University Committee on Research Involving Human Subjects
246 Administration Building

Michigan State University

(517) 355-2180

UCRIHS @ pilot.msu.edu

I have been provided information about this study and have had the opportunity to ask
questions about this study. I agree to participate in this study by signing my name on the line
below.

Parent/Guardian Date

Subject Date

 

Note: This form must be signed by BOTH the athlete and the parent/guardian to give consent
for the athlete to participate in the study. Please remember to bring this form on the date of
testing.

212

 
   

 
 

STRTE UNIV. L IBRQRIES
ll lll NW Ill llll Ill! N Ill Illll ll ”Ill II I ll
3020488460

E!

 

n I CH 1 can
lllllllmlml
3'I2