A Age-related changes in cardiovascular disease risk

factors of hypercholesterolemic children

Andrew M. Tershakovec, MD, Abbas F. Jawad, PhD, Virginia A. Stallings, MD, Jean A. Cortner, MD,
Babette S. Zemel, PhD, and Barbara M. Shannon, PhD

pared the interrelationships among

Objective: To describe the age-related changes in cardiovascular disease risk these risk factors in hypercholes-
factors in young, hypercholesterolemic (HC) children. terolemic children. Differences between
Methods: Hypercholesterolemic (n = 227) and nonhypercholesterolemic these risk factors in distinct popula-
(NHC) (n = 80) children between the ages of 4 and 10 years were identified. tions, such as obese children and HC
Height, weight, skin-fold and blood pressure measurements, and total choles- children, may provide insight into the
mechanisms promoting the expression
terol levels were measured. The HC group also had insulin levels evaluated.
of CVD risk factors.
The groups were compared by analysis of variance. Simple Spearman correla-
tions evaluated the associations between factors within each group.
See editorial, p. 383.
Results: The HC and NHC groups had similar mean ages, heights, and
weights, both contained 51% girls, and all were white subjects. Percent weight- The relationships between the risk
for-height median, and biceps, triceps, suprailiac and subscapular skin-fold factors for CVD in children seem to be
measurements were all larger for the HC group. A significant age interaction age related—the clustering of the risk
demonstrated that the HC group’s larger suprailiac and sum of skin-fold mea- factors is stronger in older children and
sures were expressed in the 8.0- to 9.9-year-old children, but not the 4.0- to 5.9- adults than in younger children.1,17
year-olds. For both groups, systolic blood pressure was associated with the However, most studies have evaluated
these factors among older children, ado-
measures of adiposity. For the HC group, insulin levels were also associated
lescents, and adults, limiting the under-
with adiposity.
standing of these associations in young
Conclusions: These results suggest that: (1) children with HC have greater children. Evaluating risk factors such as
body fat, (2) the expression of the hypercholesterolemia precedes the expres- elevated blood pressure, lipid levels, and
sion of increased body fat, (3) body fat increases with age, and (4) altered in- insulin levels in younger children and in
sulin and blood pressure levels are expressed in association with the increased a cohort with hyperlipidemia may pro-
body fat in children with HC. Confirmation with longitudinal data is necessary. vide insight into the processes that influ-
(J Pediatr 1998;132:414-20) ence the development of these risk fac-
tors.

Risk factors for cardiovascular disease out specific recognized risk before CVD Cardiovascular disease
HC Hypercholesterolemic
frequently cluster within individuals in screening).1,4,7-22 It has been postulated
NHC Nonhypercholesterolemic
both children and adults.1-6 Most of the that the presence of excess adiposity in- WHM Percent weight-for-height median
studies evaluating these factors in chil- fluences metabolism in such a way as to
dren have enrolled obese cohorts or induce increases in blood pressure, lipid As part of a study designed to evalu-
studied these relationships in the gener- levels, and insulin levels.5,6,23,24 Howev- ate the efficacy of nutrition education
al population (e.g., in individuals with- er, few studies have assessed and com- programs to lower elevated cholesterol
levels in children, a cohort of HC, 4- to
From the Division of Gastroenterology and Nutrition and Division of Biostatistics, Children’s Hospital of Philadelphia, and
the Department of Nutrition, Pennsylvania State University, University Park, Pennsylvania.
10-year-old children was identified.25-28
Supported by the National Heart, Lung, and Blood Institute (grant No. HL43880-03), the Howard Heinz In an effort to define the age-related
Endowment, National Institutes of Health grant No. DK-19525, and the University of Pennsylvania Re- changes in the relationship between
search Foundation. Critikon, Inc. donated the Dinamap model 8100 monitor used in this study. risk factors for CVD in HC children,
Submitted for publication Mar. 13, 1997; accepted Sept. 18, 1997.
the associations between anthropomet-
Reprint requests: Andrew M. Tershakovec, MD, Division of Gastroenterology and Nutrition,
The Children’s Hospital of Philadelphia, 324 S. 34th St., Philadelphia, PA 19104-4399. ric measures, blood pressure, insulin
Copyright © 1998 by Mosby, Inc. levels, and cholesterol levels were eval-
0022-3476/98/$5.00 + 0 9/21/86308 uated. Assessment of such a young,

414
THE JOURNAL OF PEDIATRICS TERSHAKOVEC ET AL.
VOLUME 132, NUMBER 3, PART 1

HC cohort offers a unique view of the

development of the relationships be-
tween these CVD risk factors.

METHODS
The Children’s Health Project was an
education-demonstration project de-
signed to evaluate two nutrition educa-
tion approaches to lowering low-density
lipoprotein cholesterol levels of HC 4- to
10-year-old children.25-28 A cholesterol
screening program was conducted in
nine suburban pediatric practice offices
to identify “at-risk” 3.9- to 9.9-year-old
children (i.e., those with nonfasting
plasma total cholesterol levels above
4.55 mmol/L (176 mg/dl, 75th per-
centile). At-risk children were invited to
provide two fasting venous blood sam-
ples (on separate days within 2 weeks of
each other) for lipid profile analysis. The
lipoprotein levels derived from these two
analyses were averaged. Children whose
average low-density lipoprotein choles-
terol was between the 80th and 98th
percentiles (2.77 to 4.24 mmol/L [107 to Figure. Sum of skin folds (geometric means and 95% confidene intervals) as a function of age for
HC and NHC groups.
164 mg/dl] for boys, and 2.90 to 4.24
mmol/L [112 to 164 mg/dl] for girls)
were considered HC. From those chil-
dren whose screening total cholesterol boards of The Children’s Hospital of cal Chemistry slides (Eastman Kodak
level was less than the 60th percentile Philadelphia, the Pennsylvania State Co.). Low-density lipoprotein choles-
(4.21 mmol/L [163 mg/dl] for boys and University, and Abington Memorial terol levels were calculated according to
4.37 mmol/L [168 mg/dl] for girls), a Hospital. the Friedwald equation.31 The Chil-
non-HC control group was randomly se- dren’s Hospital Clinical Chemistry Lab-
lected. An adaptive randomization pro- Lipid Analysis oratory participates in the College of
cedure and a sequential treatment as- Capillary blood samples for total American Pathologists proficiency test-
signment method were used to balance cholesterol screening were analyzed ing program.
the groups by age, gender, and season of with a Kodak Ektachem DT-60 ana-
involvement. To be eligible to partici- lyzer (Eastman Kodak Co., Rochester, Insulin Assessment
pate, children had to be free of sec- N.Y.) in each participating practice. Plasma samples for the HC children
ondary causes of hypercholesterolemia, As previously reported,30 external and were stored at –20° C and analyzed by
and to be at least 85% but not greater internal quality assurance programs using a commercially available double-
than 130% of ideal body weight.29 were implemented in compliance with antibody radioimmunoassay (ICN) at
For both groups of children, height, National Cholesterol Education Pro- the University of Pennsylvania Diabetes
weight, skin-fold thickness measure- gram guidelines. Research Center’s Radioimmunoassay
ments, screening total cholesterol, and For the low-density lipoprotein cho- Core Facility.
blood pressure were assessed before lesterol determinations, venous blood
initiating the nutrition education in- samples were refrigerated, centrifuged, Anthropometric Assessment
tervention, whereas lipid profiles and and the plasma transported to the Clini- For anthropometric evaluation, chil-
insulin levels were only assessed for cal Chemistry Laboratory at The Chil- dren wore underpants and lightweight
the HC children. dren’s Hospital of Philadelphia. Total hospital gowns. Trained anthro-
A parent of each child consented to cholesterol, triglycerides, and high-den- pometrists obtained each child’s weight
participation. The protocol was ap- sity lipoprotein cholesterol levels were (accurate to 0.1 kg) using a digital
proved by the institutional review measured using Kodak Ektachem Clini- floor scale (Seca Alpha model 770,

415
 TERSHAKOVEC ET AL.
Table 1.Characteristics of HC and NHC groups of children
Age (yr)
Gender distribution (% female)
Screening total cholesterol*
Height (cm)*,†
Weight (kg)†
WHM %*,†
Systolic blood pressure (mm Hg)*,†
Diastolic blood pressure (mm Hg)*,†
Sample size
Total group
4.0–5.9 years
6.0–7.9 years
8.0–9.9 years
Date presented as mean ± SE.
*Significant difference between groups (p < 0.05).
†Least square mean.
Table II. Skin-fold thickness measurements of HC and NHC children
Hypercholesterolemic
Biceps*,† 5.4 (5.2, 5.6)
Triceps*,† 9.5 (9.2, 9.8)
Subscapular*,† 6.2 (6.0, 6.4)
Suprailiac*,† 6.1 (5.8, 6.4)
Sum of skin folds*,† 27.2 (26.1, 28.3)
Skin-fold measurements in millimeters.
*Data presented as geometric mean (95% confidence interval of geometric mean).
†Significant difference between groups (p < 0.05).
Seca, Columbia, Md.) and height (accu-
rate to 0.1 cm) using a wall-mounted sta-
diometer (Holtain Ltd., Crymych, Eng-
land). Measurements were taken in
duplicate and the average used in the
analyses. The percent weight-for-height
median was calculated with the Centers
for Disease Control Anthropometric
Software Package.29 Skin-fold thickness
was measured at the biceps, triceps,
suprailiac, and subscapular sites with
Holtain skinfold calipers according to
standardized methods.32 Skin-fold mea-
surements were completed in triplicate
and averaged. Interobserver and in-
416
THE JOURNAL OF PEDIATRICS
MARCH 1998
Hypercholesterolemic Nonhypercholesterolemic
6.3 ± 0.1 6.4 ± 0.2
51 51
197.1 ± 0.9 140.0 ± 2.2
118.9 ± 0.4 120.6 ± 0.6
23.6 ± 0.3 23.6 ± 0.4
104.6 ± 0.7 102.1 ± +1.1
103.1 ± 0.6 99.6 ± 1.0
51.7 ± 0.4 50.2 ± 0.7
227 80
105 33
72 27
50 20
Statistical Analysis
Analysis of variance with three inter-
action terms (age group, gender, choles-
Nonhypercholesterolemic terol level) was completed for the an-
4.8 (4.5, 5.1) thropometric assessments. For the
8.7 (8.2 , 9.3) purposes of analysis, the children were
5.4 (5.1, 5.8) divided into three age groups (4.0- to
5.2 (4.8, 5.6) 5.9-year-old children, 6.0- to 7.9-year-
24.1 (22.6, 25.8) old children, and 8.0- to 9.9-year-old
children) and two cholesterol level
groups (HC and NHC) (Table I). Sum
of skin folds was calculated by adding
the biceps, triceps, suprailiac, and sub-
scapular measurements for each child.
To meet the analysis of variance normal-
ity assumptions and requirement for ho-
traobserver reliability coefficients for mogeneity of variance, skin-fold mea-
the anthropometrists were consistently surements were analyzed using log
above 0.99 for all measurements. transformation. Geometric means (anti-
log transformations of log-transformed
Blood Pressure Assessment data) are presented in Table II and the
Four blood pressure measurements Figure. Because of the limited number
were obtained with the children at rest, of nonwhite children in the program and
after the children had been seated sever- the reported racial differences in CVD
al minutes by use of a Dinamap model risk factors,1,14 these analyses were lim-
8100 monitor (Critikon, Inc., Tampa, ited to white children only. When the
Fla.). The four measurements were ob- overall analysis of variance indicated a
tained at 1-minute intervals. The second, significant difference between groups
third, and fourth measurements of dias- (p < 0.05), subgroup comparisons were
tolic and systolic blood pressure were completed with a Bonferroni correction
averaged for the analyses. for multiple comparisons.
THE JOURNAL OF PEDIATRICS
VOLUME 132, NUMBER 3, PART 1
Simple Spearman correlations were
used to evaluate the relationship be-
tween anthropometric assessments,
blood pressure, and cholesterol level in
all subjects, and insulin level for the HC
children. The percent WHM median (as
a measure of relative weight), suprailiac
skin-fold thickness (as a proxy of trun-
cal obesity), and sum of skin folds (as a
proxy for total body adiposity) were se-
lected for this analysis. The correlation
matrix was examined for potential inter-
relationships between variables. When
such interrelationships were possible,
the partial correlations for the potential
covariates were evaluated. All analyses
were completed with SAS software.33
RESULTS
From October 1990 to December
1992, 227 HC and 80 NHC children
were identified and enrolled. Insulin lev-
els were available for 185 of the HC chil-
dren. The mean (±SD) for selected char-
acteristics of the two groups are listed in
Table I. The geometric means (and 95%
confidence intervals) for the skin-fold
thickness measures are listed in Table II.
The HC group was slightly shorter, had
a greater percent WHM median, thicker
skin-fold thickness measurements, and
higher systolic blood pressure measure-
ments than the NHC group.
A significant age group × cholesterol
level interaction term was observed for
the suprailiac skin-fold and sum of skin-
fold measurements. Further evaluation
demonstrated that this difference was
related to the large difference observed
between the HC and NHC children in
the 8.0- to 9.9-year-old group. This is
demonstrated for the sum of skin folds
in the Figure. It is worth noting that all
of the skin-fold and blood pressure mea-
surements demonstrated this age trend;
however, it was only statistically signifi-
cant with the suprailiac and sum of skin-
fold measurements. Girls had greater
skin-fold measurements than boys.
However, none of the gender interaction
terms were significant (e.g., the gender
differences did not vary by age or cho-
lesterol level).
Table III. Correlation of variables and potential covariates of HC and NHC children
Hypercholesterolemic WHM
group
Systolic blood pressure (mm Hg) 0.16*
Partial out insulin 0.16*
Partial out cholesterol 0.16*
Cholesterol NS
Partial out insulin — —
Insulin 0.18*
Partial out cholesterol 0.18*
Noncholesterolemic
group WHM
Systolic blood pressure (mm Hg) 0.32§
“Partial out” refers to the correlation between the two variables adjusted for the influence of the third
variable.
NS, p > 0.10.
*p ≤ 0.10.
†p ≤ 0.05.
‡p ≤ 0.01.
§p ≤ 0.005.
To evaluate interrelationships between
anthropometric variables (percent
WHM median, suprailiac skin fold, sum
of skin folds) and potential risk factors
for CVD (diastolic blood pressure, sys-
tolic blood pressure, total cholesterol
and, for the HC group only, insulin
level), a correlation matrix was generat-
ed and evaluated. The influence of po-
tential covariates was also evaluated. As
shown in Table III for the HC group,
WHM was significantly associated with
systolic blood pressure, whereas the as-
sociations with sum of skin folds and
suprailiac skin fold were of borderline
significance. Accounting for the associa-
tion between insulin level and the an-
thropometric measures or the association
between cholesterol level and the anthro-
pometric measures did not change the
magnitude or significance of the associa-
tions between systolic blood pressure
and the anthropometric measures. Total
cholesterol and insulin level were also
significantly associated with the sum of
skin-fold measures. WHM and supraili-
ac skin-fold thickness were significantly
associated with insulin level. Accounting
for the influence of total cholesterol had
little effect on the association between in-
sulin level and the anthropometric mea-
sures; similarly, accounting for the influ-
ence of insulin level had little effect on
TERSHAKOVEC ET AL.
Suprailiac Sum of skin folds
0.14† 0.14†
0.13† 0.14†
0.13† 0.14†
NS 0.16*
0.19‡
0.27§ 0.25§
0.25§ 0.24§
Suprailliac Sum of skinfolds
NS NS
the association between total cholesterol
and sum of skin folds. For the NHC chil-
dren, only the association between
WHM and systolic blood pressure levels
was significant.
Insulin levels could not be measured
in 42 HC children. To ensure that the
HC children without available insulin
levels were not different with regard to
important variables, such as age, gender
distribution, weight z score, height z
score, WHM, biceps, triceps, subscapu-
lar, or suprailiac skin fold, sum of skin
folds, total cholesterol level, systolic or
diastolic blood pressure, the HC chil-
dren with available insulin levels (n =
185) were compared with those without
available insulin levels. There were no
significant differences between the
groups in any of the variables evaluated.
DISCUSSION
We report the evaluations of some im-
portant risk factors for CVD in a cohort
of HC children. The results suggest that
among children with elevated choles-
terol levels there exists an age-related
expression of increased body fat with a
tendency for a central distribution (e.g.,
suprailiac, as opposed to biceps or tri-
ceps). Increased blood pressure and in-
417
 TERSHAKOVEC ET AL.
sulin levels were also associated with the
increased body fat and relative weight.
The relationships between measures of
adiposity and cholesterol level, and mea-
sures of adiposity and insulin levels were
independent of each other in this sam-
ple. These observations suggest that HC
children express an increase in body fat
as they grow older, and that this increase
in body fat is associated with increases
in blood pressure and insulin levels. In
other words, the hypercholesterolemia
precedes the development of the in-
creased body fat and increases in blood
pressure and insulin levels.
Most of the other published studies
evaluating CVD risk factors in children
have focused on the clustering of these
risk factors in samples from the general
population without predefined risk fac-
tors1-4,8,14-18,21,34-40 or evaluated the
prevalence of such risk factors in obese
children.9-13,20,41 Webber et al.1 de-
scribed a greater than expected cluster-
ing of CVD risk factors, including in-
creased cholesterol levels, increased
blood pressure, and weight-height index,
in 5- to 14-year-old children but not in
preschool children. Boulton and John-
ston17 observed a similar lack of cluster-
ing of risk factors in 4-year-old children.
Although the degree of association be-
tween anthropometric measures and
serum cholesterol level in the older chil-
dren was small, children at the higher lev-
els of blood pressure and relative weight
did have higher cholesterol levels.1,14 In
addition, insulin levels have been shown
to track from childhood into early adult-
hood, and higher insulin levels cluster
with higher relative weights, blood lipid
levels, and blood pressure.15,17
In obese children, blood lipid levels,
insulin levels, and blood pressure have
been associated with measures of obe-
sity. In addition, a central versus pe-
ripheral pattern of fat has also been
positively correlated with these CVD
risk factors. 9,12,14,20,42-44 Evaluating
the time course of metabolic changes
in childhood obesity, Le Stunff and
Bougneres 22 demonstrated that early
in the course of obesity, lipid oxidation
was altered; glucose oxidation differ-
ences were also observed, but later
after the onset of obesity.
418
Evaluating our results in light of
these past observations raises some in-
teresting questions. It has been sug-
gested that altered lipid oxidation and
free fatty acid levels associated with
obesity may influence insulin resistance
and other factors (e.g., blood
pressure*). Furthermore, it has been
suggested that obesity (and perhaps
truncal obesity) induce the initial meta-
bolic derangements, which then initiate
the other observed changes. Our data
suggest that in children with hypercho-
lesterolemia, the lipid abnormality pre-
cedes the development of the increased
body fat. The increased body fat then is
associated with increased blood pres-
sure and insulin levels. Because most of
the previously published studies de-
scribing the clustering of risk factors in
children have evaluated children older
than our cohort, it is possible that these
other studies were identifying at least
some HC children who had already de-
veloped increased body fat, blood pres-
sure, and insulin levels when evaluated.
The age-related increase in the strength
of these interrelationships is supported
by the lack of such relationships previ-
ously observed in preschool chil-
dren1,17 and the previously described
time-related differences in glucose and
lipid metabolism in obese children.22
Little is known about the clustering
of risk factors for CVD in hyperlipi-
demic children. We evaluated the asso-
ciation between relative weight and hy-
perlipidemia in a different group of
children with familial combined hyper-
lipidemia.51 An association between age
and expression of the familial combined
hyperlipidemia was observed, as well as
an independent relationship between
relative weight and the expression of
familial combined hyperlipidemia.
These observations are consistent with
our current data. It is possible that the
associations we describe in this report
are limited to children with familial
combined hyperlipidemia. We do not
have enough family medical informa-
tion to define which children in our co-
hort have familial combined hyperlipi-
demia, but the lack of insulin resistance
*References 7, 11, 17, 19, 20, 23, 24, 45-50.
THE JOURNAL OF PEDIATRICS
MARCH 1998
reported in adults with heterozygous
familial hypercholesterolemia and non-
familial isolated hypercholesterolemia52
suggests that all lipid disorders are not
the same in this regard.
Care must be taken when evaluating
our data. Our study was cross-section-
al; longitudinal observations of these
relationships within a child over time
would be more informative. Further-
more, the observed associations do not
define cause and effect. Finally, the
data described are limited to white chil-
dren who were well nourished and who
had a relatively normal body weight.
Given the well-described racial differ-
ences in CVD risk factors,† evaluating
these associations within different
racial groups in a prospective longitu-
dinal manner is necessary.
†References 1, 14, 34, 35, 38, 47, 53.
We acknowledge the invaluable support of the
participating practices in subject recruitment, the
help of Deirdre MacLeod in preparing the manu-
script, and the editorial assistance of Drs. Paul
Coates and Daniel Rader.
REFERENCES
1. Webber LS, Voors AW, Srinivasan SR,
Frerichs RR, Berenson GS. Occurrence
in children of multiple risk factors for
coronary artery disease: the Bogalusa
Heart Study. Prev Med 1979;8:407-18.
2. Wilmore JH, McNamara JJ. Preva-
lence of coronary heart disease risk fac-
tors in boys, 8 to 12 years of age. J Pedi-
atr 1974;84:527-33.
3. Lauer RM, Connor WE, Leaverton PE,
Reiter MA, Clarke WR. Coronary heart
disease risk factors in school children:
the Muscatine study. J Pediatr 1975:
697-706.
4. Raitakari OT, Porkka KVK, Ronnemaa
T, Knip M, Uhari M, Akerblom HK, et
al. The role of insulin in clustering of
serum lipids and blood pressure in chil-
dren and adolescents: the Cardiovascu-
lar Risk in Young Finns Study. Dia-
betologia 1995;38:1042-50.
5. Despres J-P. Obesity and lipid metabo-
lism: relevance of body fat distribution.
Curr Opin Lipid 1991;2:5-15.
6. Guo S, Salisbury S, Roche AF, Chumlea
WC, Siervogel RM. Cardiovascular dis-
ease risk factors and body composition:
a review. Nutr Res 1994;11:1721-77.
7. Smoak CG, Burke GL, Webber LS,
Harsha DW, Srinivasan SR, Berenson
GS. Relation of obesity to clustering of
THE JOURNAL OF PEDIATRICS
VOLUME 132, NUMBER 3, PART 1
cardiovascular disease risk factors in
children and young adults: the Bogalusa
Heart Study. Am J Epidemiol 1987;
125:364-72.
8. Jiang X, Srinivasan SR, Urbina E,
Berenson GS. Hyperdynamic circula-
tion and cardiovascular risk in children:
the Bogalusa Heart Study. Circulation
1995;91:1101-6.
9. Brambilla P, Manzoni P, Sironi S, Si-
mone P, Del Maschio A, di Natale B, et
al. Peripheral and abdominal adiposity
in childhood obesity. Int J Obes
1994;18:795-800.
10. Caprio S, Hyman LD, McCarthy S,
Lange R, Bronson M, Tamborlane WV.
Fat distribution and cardiovascular risk
factors in obese adolescent girls: impor-
tance of the intrabdominal fat depot. Am
J Clin Nutr 1996;64:12-7.
11. Monti LD, Brambilla P, Stefani I,
Caumo A, Magni F, Poma R, et al. In-
sulin regulation of glucose turnover and
lipid levels in obese children with fasting
normoinsulinaemia. Diabetologia 1995;
38:739-47.
12. Waliu Islam AHM, Yamashita S, Kotani
K, Nakamura T, Tokunaga K, Arai T, et
al. Fasting plasma insulin level is an im-
portant risk factor for the development
of complications in Japanese obese chil-
dren—results from a cross-sectional and
a longitudinal study. Metabolism
1995;4:478-85.
13. Agostoni C, Riva E, Bellu R, Vincenzo
SS, Grazia BM, Giovanni M. Relation-
ships between the fatty acid status and
insulinemic indexes in obese children.
Prostaglandins Leukot Essent Fatty
Acids 1994;51:317-21.
14. Frerichs RR, Webber LS, Srinivasan
SR, Berenson GS. Relation of serum
lipids and lipoproteins to obesity and
sexual maturity in white and black chil-
dren. Am J Epidemiol 1978;108:486-96.
15. Burke GL, Webber LS, Srinivasan SR,
Radhakrishnamurthy B, Freedman DS,
Berenson GS. Fasting plasma glucose
and insulin levels and their relationship
to cardiovascular risk factors in chil-
dren: Bogalusa Heart Study. Metabo-
lism 1986;35:441-6.
16. Jiang X, Srinivasan SR, Webber LS,
Wattigney WA, Berenson GS. Associa-
tion of fasting insulin level with serum
lipid and lipoprotein levels in children,
adolescents, and young adults: the Bo-
galusa Heart Study. Arch Intern Med
1995;155:190-6.
17. Boulton TJC, Johnston O. A coronary
risk-factor profile of 4 year olds. II.
Inter-relationships, clustering, and
tracking of blood pressure, serum
lipoproteins, and skinfold thickness.
Aust Paediatr J 1978;14:278-82.
18. Shear CL, Freedman DS, Burke GL,
Harsha DW, Berenson GS. Body fat
patterning and blood pressure in chil-
dren and young adults: the Bogalusa
Heart Study. Hypertension 1987;9:236-
44.
19. Kikuchi DA, Srinivasan SR, Harsha
DW, Webber LS, Sellers TA, Berenson
GS. Relation of serum lipoprotein lipids
and apolipoproteins to obesity in chil-
dren: the Bogalusa Heart Study. Prev
Med 1992;21:177-90.
20. Zwiauer KFM, Pakosta R, Mueller T,
Widhalm K. Cardiovascular risk factors
in obese children in relation to weight
and body fat distribution. J Am Coll
Nutr 1992;11:41S-50S.
21. Ronnemaa T, Knip M, Lautala P, Viikari
J, Uhari M, Leino A, et al. Serum in-
sulin and other cardiovascular risk indi-
cators in children, adolescents and
young adults. Ann Med 1991;23:67-72.
22. Le Stunff C, Bougneres P-F. Time
course of increased lipid and decreased
glucose oxidation during early phase of
childhood obesity. Diabetes 1993;
42:1010-6.
23. Walker M. Obesity, insulin resistance,
and its link to non-insulin-dependent dia-
betes mellitus. Metabolism 1995;44:18-20.
24. Kissebah AH, Peiris AN. Biology of re-
gional body fat distribution: relation-
ship to non-insulin-dependent diabetes
mellitus. Diabetes Metab Rev 1989;
15:83-109.
25. Shannon BM, Greene G, Stallings VA,
Achterberg C, Krotan-Berman M, Gre-
goire J, et al. A dietary education pro-
gram for hypercholesterolemic children
and their parents. J Am Diet Assoc
1991;91:208-12.
26. Stallings VA, Shannon BM, Greene
GW, Collins S, Berman M, Wellock A,
et al. Preliminary report of a home-based
education program for dietary treatment
of hypercholesterolemia in children. Am
J Health Promotions 1993;8:106-8.
27. Shannon BM, Tershakovec AM, Martel
JK, Achterberg CL, Cortner JA,
Smiciklas-Wright H, et al. Reduction of
elevated LDL-cholesterol levels of 4-10
year old children through home-based
dietary education. Pediatrics 1994;
94:923-7.
28. Tershakovec AM, Shannon BM, Achte-
rberg CL, McKenzie JM, Martel JK,
Smiciklas-Wright H, et al. A one year
follow-up of nutrition education for hy-
percholesterolemic children. Am J Pub-
lic Health. 1998;88:258-61.
29. Anthropometric Software Package, ver-
sion 1.01. Atlanta (GA): Centers for
Disease Control and Prevention; 1987.
30. Bennett MJ, Tershakovec AM, Cortner
JA, Shannon BM. A quality assurance
program for the measurement of capil-
lary blood cholesterol levels in private
pediatric practice: the Children’s Health
Project. Am J Dis Child 1993;147:340-5.
TERSHAKOVEC ET AL.
31. Freidwald WR, Levy RI, Frederickson
DS. Estimation of the concentration of
low density lipoprotein cholesterol
without the use of the preparative ul-
tracentrifuge. Clin Chem 1972;18:
499-502.
32. Harrison G, Burkirk E, Carter J, John-
ston F, Lohman T, Pollock M, et al.
Skinfold thicknesses and measurement
technique. In: Lohman TG, Roche AF,
Martorell R, editors. Anthropometric
standardization reference manual.
Champaign (IL): Human Kinetics
Books; 1988. p. 55-70.
33. SAS/STAT Software, version 6.11. Cary
(NC): SAS Institute; 1995.
34. Arslanian S, Suprasongsin C. Differ-
ences in the in vivo insulin secretion and
sensitivity of healthy black versus white
adolescents. J Pediatr 1996;129:440-3.
35. Daniels SR, Obarzanek E, Barton BA,
Kimm SYS, Similo SL, Morrison JA.
Sexual maturation and racial differences
in blood pressure in girls: the National
Heart, Lung, and Blood Institute
Growth and Health Study. J Pediatr
1996;129:208-13.
36. Craig SB, Bandini LG, Lichtenstein
AH, Schaefer EJ, Dietz WH. The im-
pact of physical activity on lipids,
lipoproteins, and blood pressure in
preadolescent girls. Pediatrics 1996;
98:389-95.
37. Travers SH, Jeffers BW, Bloch CA, Hill
JO, Eckel RH. Gender and Tanner
stage differences in body composition
and insulin sensitivity in early pubertal
children. J Clin Endocrinol Metab
1995;80:172-8.
38. Jiang X, Srinivasan SR, Radhakrishna-
murthy B, Dalferes ER Jr, Berenson
GS. Racial (black-white) differences in
insulin secretion and clearance in adoles-
cents: the Bogalusa Heart Study. Pedi-
atrics 1996;97:357-60.
39. Bao W, Srinivasan SR, Berenson GS.
Persistent elevation of plasma insulin
levels is associated with increased car-
diovascular risk in children and young
adults: the Bogalusa Heart Study. Circu-
lation 1996;93:54-9.
40. Braun B, Zimmermann MB, Kretchmer
N, Sparago RM, Smith RM, Gracey M.
Risk factors for diabetes and cardiovas-
cular disease in young Australian aborig-
ines: a 5-year follow-up study. Diabetes
Care 1996;19:472-9.
41. Hoffman RP, Stumbo PJE, Janz KF,
Nielsen DH. Altered insulin resistance
is associated with increased dietary
weight loss in obese children. Horm Res
1995;44:17-22.
42. Gillum RF. The association of the ratio
of waist to hip girth with blood pressure,
serum cholesterol and serum uric acid in
children and youths aged 6-17 years. J
Chron Dis 1987;40:413-20.
419
 TERSHAKOVEC ET AL.
43. Freedman DS, Srinivasan SR, Harsha
DW, Webber LS, Berenson GS. Rela-
tion of body fat patterning to lipid and
lipoprotein concentrations in children
and adolescents: the Bogalusa Heart
Study. Am J Clin Nutr 1989;50:930-9.
44. Shear CL, Freedman DS, Burke GL,
Harsha DW, Berenson GS. Body fat pat-
terning and blood pressure in children
and young adults: the Bogalusa Heart
Study. Hypertension 1987;9:236-44.
45. Reaven GM. Role of insulin resistance
in human disease. Diabetes 1988;37:
1595-607.
46. Bonora E, Zenere M, Branzi P. Influ-
ence of body fat and its regional localiza-
tion on risk factors for atherosclerosis in
young men. Am J Epidemiol 1992;135:
1271-8.
420
47. Voors AW, Radhakrishnamurthy B,
Srinivasan SR, Webber LS, Berenson
GS. Plasma glucose level related to
blood pressure in 272 children, ages 7-
15 years, sampled from a total biracial
population. Am J Epidemiol 1981;
113:347-56.
48. Reaven GM, Lithell H, Landsberg L.
Hypertension and associated metabolic
abnormalities—the role of insulin resis-
tance and the sympathoadrenal system.
N Engl J Med 1996;334:374-81.
49. Kissebah AH, Peiris AN. Biology of re-
gional body fat distribution: relationship
to non-insulin-dependent diabetes melli-
tus. Diabetes Metab Rev 1989;5:83-109.
50. Despres J-P, Moorjani S, Lupien PJ,
Tremblay A, Nadeau A, Bouchard C.
Regional distribution of body fat, plas-
THE JOURNAL OF PEDIATRICS
MARCH 1998
ma lipoproteins, and cardiovascular dis-
ease. Arteriosclerosis 1990;10:497-511.
51. Shamir R, Tershakovec AM, Gallagher
PR, Liacouras CA, Hayman LA, Cort-
ner JA. The influence of age and relative
weight on the presentation of familial
combined hyperlipidemia in childhood.
Atherosclerosis 1996;121:85-91.
52. Karhapaa P, Voutilainen E, Kovanen PT,
Laakso M. Insulin resistance in familial
and nonfamilial hypercholesterolemia.
Arterioscler Thromb 1993;13:41-7.
53. Berenson GS, Radhakrishnamurthy B,
Srinivasan SR, Voors AW, Foster TA,
Dalferes ER, et al. Plasma glucose and
insulin levels in relation to cardiovascu-
lar risk factors in children from a bira-
cial population—the Bogalusa Heart
Study. J Chron Dis 1981;34:379-91.