Professional Documents
Culture Documents
KEY WORDS
obesity, adolescent health, cardiovascular risk factors,
lipoproteins, preventive medicine
ABBREVIATIONS
LDLlow-density lipoprotein
HDLhigh-density lipoprotein
apoapolipoprotein
IRinsulin resistance
HOMAhomeostasis model assessment
SR-BIscavenger receptor class B type I
www.pediatrics.org/cgi/doi/10.1542/peds.2010-1395
doi:10.1542/peds.2010-1395
Accepted for publication Feb 8, 2011
Address correspondence to Carlos Posadas-Romero, MD, Juan
Badiano No 1 Col Seccin XVI, Tlalpan, CP 14080.
E-mail: cposadasr@yahoo.com.
PEDIATRICS (ISSN Numbers: Print, 0031-4005; Online, 1098-4275).
Copyright 2011 by the American Academy of Pediatrics
FINANCIAL DISCLOSURE: The authors have indicated that they
have no personal nancial relationships relevant to this article
to disclose.
abstract
OBJECTIVE: This study aimed to evaluate high-density lipoprotein functionality and the cardiovascular risk factor prole in the overweight
pediatric population. We hypothesized that overweight adolescents
with low high-density lipoprotein cholesterol and elevated triglyceride
plasma levels have metabolic abnormalities and dysfunctional highdensity lipoprotein particles, similar to those reported in adults.
PATIENTS AND METHODS: Overweight adolescents with (group 1 [n
21]) and without (group 2 [n 36]) atherogenic dyslipidemia (highdensity lipoprotein cholesterol: 40 mg/dL and triglycerides: 150
mg/dL) and normal-weight normolipidemic subjects, as a reference
(group 3 [n 36]), were included. The cardiovascular risk factor
prole (lipids, lipoproteins, high-sensitivity C-reactive protein, and insulin), high-density lipoprotein subclass distribution, composition, and
cholesterol efux capacity were studied.
RESULTS: Group 1 adolescents showed abnormalities in high-density
lipoprotein subclass distribution and high-density lipoprotein chemical composition, as well as a signicantly lower capacity to promote
cholesterol efux (14.8 2.8, 16.5 3.8, 20.4 3.5, for groups 1, 2 and
3, respectively). High-density lipoprotein2a (R2 0.212, 0.472, P
.0001) and the Tanner score (R2 0.054, 0.253, P .02) were the
independent predictors of cholesterol efux. Group 1 also showed a
higher degree of cardiovascular abnormalities (an adverse lipoprotein
prole, greater insulin resistance and systemic inammation; and
lower low-density lipoprotein size) than group 2, even after BMI and
Tanner score adjustment.
CONCLUSIONS: This study suggests that atherogenic dyslipidemia
identies overweight adolescents with quantitative, qualitative, and
functional high-density lipoprotein abnormalities. Atherogenic dyslipidemia seems to be a marker of an increased risk for developing cardiovascular disease and indicates that those adolescents should be a
target of aggressive prevention programs and lipid management
guidelines. Pediatrics 2011;127:e1521e1527
e1521
Although clinical atherosclerotic coronary artery disease does not occur until
the fourth or fth decade of life, it is
known that atherosclerosis already is
present in youth,1 and its initial stages
are associated with dyslipidemia.2 An association between increased low-density
lipoprotein (LDL) cholesterol and cardiovascular risk is well established.3 Epidemiologic studies also have consistently
shown that low plasma levels of highdensity lipoprotein (HDL) cholesterol is
an important independent risk factor for
coronary disease.4 However, it has been
reported that some HDL characteristics
may be more important than HDL cholesterol levels in predicting future coronary
disease.5 HDL is a heterogeneous group
of large (HDL2b and HDL2a) and small
(HDL3a, HDL3b, and HDL3c) lipoprotein
particles6 that differ in size, lipid composition, density, and function.7 Under normal conditions, the antiatherogenic effect of both large and small HDL particles
is likely attributable to its central role in
reverse cholesterol transport and to
their antioxidant, anti-inamatory, and
anti-thrombotic actions that may contribute to the prevention of atherosclerosis development and progression.8
Childhood obesity is one of the most
serious and urgent health problems in
both developed and emerging economies.9 Atherogenic dyslipidemia,
which is characterized by high triglycerides, low HDL cholesterol concentrations, and small dense LDL particles, is
commonly seen in obese children and
adolescents.10 In addition, abnormalities in the HDL subclass distribution
and chemical composition have both
been shown in adults11,12 and children13,14 with obesity, insulin resistance, and type 2 diabetes. In adults,
lower levels of large particles and increased levels of small HDL are associated with increased coronary disease
prevalence and increased recurrence
of coronary events.5 In children, adiposity and its adverse metabolic proe1522
MEDINA-URRUTIA et al
METHODS
A cross-sectional survey to investigate
cardiovascular risk factors in adolescents was conducted in 510 adolescents attending a junior high school in
Mexico City.13 Weight, height, waist circumference, and blood pressure were
measured according to standard procedures.13 BMI was calculated as
ARTICLES
Boys/Girls, %
Tanner, %, 2/34/5
Age, means SD, y
BMI, means SD, kg/m2
Waist, means SD, cm
Systolic blood pressure, means SD,
mm Hg
Diastolic blood pressure, means SD,
mm Hg
Total cholesterol, means SD, mg/dL
LDL cholesterol, means SD, mg/dL
HDL cholesterol, means SD, mg/dL
Triglycerides, means SD, mg/dL
nonHDL cholesterol, means SD
apoAI, means SD, mg/dL
ApoB, means SD, mg/dL
ApoB-to-apoAI ratio, means SD
Total cholesteroltoHDL cholesterol ratio,
means SD
LDL size, means SD, nm
C-reactive protein, means SD, mg/L
Glucose, means SD, mg/dL
Insulin, means SD, mU/L
HOMA-IR, means SD
Dyslipidemic
Overweight,
Group 1, n 21
Normolipidemic
Overweight,
Group 2, n 36
Normolipidemic
Normal Weight,
Group 3, n 36
52/48
0/79/21
13.3 0.9
29.8 3.8a,b
90.0 9.4a
111.3 8.4a
47/53
0/77/23
13.1 0.8
27.0 2.9a
84.8 10.6a
109.4 8.1a
53/47
6/84/10
13.1 1.0
19.7 0.7
66.5 3.8
101.6 8.5
69.4 5.6a,b
64.3 5.5
62.6 4.8
151.9 24.0b
90.4 19.3a,b
31.1 5.5a,b
189.7 40.4a,b
120.7 21.3a,b
109.3 19.7a,b
78.5 13.7a,b
0.74 0.1a,b
4.9 0.6a,b
141.7 23.3
78.0 18.7
46.0 5.4a
110.4 39.2a
95.7 20.7
134.0 19.0
63.5 13.2
0.48 0.1
3.1 0.6
147.5 17.7
80.9 14.4
51.6 1.6
81.5 22.4
93.9 15.6
137.8 22.4
57.8 14.3
0.42 0.08
2.8 0.4
26.6 0.6a,b
0.8 1.1
85.0 6.2
19.8 8.2a,b
4.3 2.0a,b
27.2 0.6
0.9 1.1
88.6 6.0
14.4 8.0a
3.1 1.9a
27.4 0.6
0.9 1.2
85.9 7.0
9.0 3.6
1.9 0.8
Characterization of HDL
Total protein, total cholesterol, free
cholesterol, phospholipid, and triglyceride contents of isolated HDL were determined using commercially available enzymatic assays in a Hitachi 902
analyzer. Cholesteryl esters were calculated by multiplying the difference
between total and free cholesterol by
1.67.22 Total lipoprotein mass was calculated as the sum of total protein,
cholesteryl esters, free cholesterol,
phospholipids, and triglycerides. Apo
content (apoAI, apoAIV, and apoE) was
evaluated semiquantitatively by sodium dodecyl sulfate polyacrylamide
gel electrophoresis.23
Cholesterol Efux Assay
Cellular cholesterol efux was determined using Fu5AH rat hepatoma cells
according to the procedure described
by de la Llera Moya et al,24 with slight
modications. Briey, Fu5AH cells
were maintained in minimal essential
PEDIATRICS Volume 127, Number 6, June 2011
medium containing 5% of bovine serum, 250 000 cells per well were
placed on 24-well plates, and 24 hours
after plating radiolabeled cholesterol
(1,2-3H cholesterol; American Radiolabeled Chemicals, Inc, St Louis, MO) was
added (0.5 Ci per well). Cells were
then washed and incubated for 18
hours in minimal essential medium
with 0.5% of bovine serum albumin. Finally, cells were washed and incubated at 37C for 4 hours with 2.0%
diluted serum or isolated total HDL
fractions (20 g HDL-protein/mL), prepared by sequential ultracentrifugation, as previously described. Radioactivity was measured in medium and
cells, and the percentage of cholesterol efux was calculated. All determinations were made in triplicate. As an
internal control, a total serum and isolated total HDL were included in each
plate, obtaining coefcients of variation of 3.3% and 5.3%, respectively.
Statistical Methods
Statistical comparisons were performed using SPSS for Windows (version 9.0; SPSS, Chicago, IL). Results are
expressed as means SD for continuous variables, and skewed variables
were previously log transformed. P
values less than 0.05 were considered
signicant. Between-group differences
in continuous variables were analyzed
by analysis of covariance. Pearson correlation coefcients were calculated
to evaluate relationships between variables, and multivariate stepwise linear
regression analysis was performed to
determine the relative contribution of
the variables studied to the cholesterol efux.
RESULTS
Age, gender, and Tanner stage distribution were similar across groups (Table 1). Group 1 showed a statistically
e1523
Dyslipidemic
Overweight,
Group 1, n 21
Normolipidemic
Overweight,
Group 2, n 36
Normolipidemic
Normal Weight,
Group 3, n 36
9.8 2.5a,b
16.6 2.7a,b
27.8 1.6a
25.7 2.4a
20.0 2.7a,b
8.6 0.1a,b
11.0 2.4a
19.8 2.6a
28.3 1.6a
24.8 2.3a
16.1 2.6
8.7 0.1a
14.1 2.7
22.7 2.5
26.7 2.1
22.2 2.0
14.2 3.3
8.8 0.2
50.3 3.0a,b
25.6 2.1
5.1 0.9a,b
16.1 2.0a,b
2.9 0.5a,b
45.4 16.8a,b
1.7 0.9a,b
47.2 2.9
26.9 2.1
3.3 0.9
19.2 1.9
3.4 0.5a
58.5 16.4
2.8 0.9
45.1 5.2
27.4 2.7
3.1 1.0
20.6 2.2
3.8 0.5
62.0 19.2
3.0 0.7
HDL subclasses are expressed as the relative proportion of total HDL protein SD. HDL composition was estimated as the
relative proportion of HDL total mass, and HDL-phospholipid to apoAI and HDL-phospholipid to HDL cholesterol were
calculated as molar ratios.
a P .01 vs group 3 (analysis of covariance).
b P .05 vs group 2 (analysis of covariance); P values of the difference between groups 1 and 2 were adjusted for BMI,
Tanner score, and HOMA-RI.
MEDINA-URRUTIA et al
FIGURE 1
Cholesterol efux assays were carried out using a monolayer of cultured cells (Fu5AH rat
hepatoma cells) and whole serum as cholesterol acceptor. P values of the difference between
groups 1 and 2 were adjusted for BMI, Tanner score, and HOMA-RI. Analysis of covariance was
used.
ARTICLES
TABLE 3 Simple and Multivariate Correlation Analysis of Biochemical Variables and Cholesterol
Efux in All the Adolescents Studied
Univariate analysis
Tanner
Triglycerides
HDL cholesterol
Total cholesterol to HDL cholesterol
Insulin
HOMA-IR
HDL2b, %
HDL2a, %
HDL3a , %
HDL3b, %
HDL3c, %
HDL size, nm
HDLtotal protein, %
HDL-phospholipids, %
HDL-triglycerides, %
HDLcholesteryl ester, %
HDLfree cholesterol, %
HDL-phospholipidtoapoAI ratio
HDL-phospholipidtoHDL cholesterol ratio
Multivariate analysis
2
0.258
0.313
0.452
0.315
0.283
0.301
0.397
0.465
0.163
0.421
0.372
0.426
0.285
0.293
0.115
0.216
0.350
0.348
0.304
.03
.01
.0001
.01
.005
.003
.0001
.0001
NS
.0001
.0001
.0001
.005
.05
NS
.05
.001
.001
.003
0.054a
0.253
0.248b
0.509
.0001
0.212a
0.472
.0001
P
.02
R indicates the Pearson correlation coefcient; R2 indicates the regression coefcient in stepwise correlation analysis. NS
indicates not signicant.
a Excludes HDL cholesterol.
b Includes all the variables, with P .05 at the unvaried analysis.
DISCUSSION
Cholesterol in peripheral tissues is derived from local synthesis and direct
uptake of circulating lipoproteins.
Most peripheral cells lack the capacity
to catabolize cholesterol and can only
dispose of it by efux to extracellular
acceptors such as HDL. Reverse cholesterol transport is a complex process by which cellular cholesterol
excess in peripheral tissues is transported by HDL to the liver for excretion
in the bile and feces, and it is thought
to be the major mechanism by which
HDL protects against atherosclerotic
cardiovascular disease.7,8,25 Cholesterol
efux, the rst step of the reverse cholesterol transport process, is mediated
by a variety of lipid transporter molecules, including adenosine triphosphate binding cassette transporter A
and G and scavenger receptor class B
type I (SR-BI).25 Adenosine triphosphate
binding cassette transporter A1 mediates the efux to small lipid poor HDL,
whereas SR-BImediates the efux to mature phospholipid rich HDL.26
In adults, Brites et al18 reported that,
compared with control subjects, adult
subjects with low HDL cholesterol and
high triglycerides showed a reduced
capacity to promote cholesterol efux
from Fu5AH cells using either whole
serum or isolated HDL fractions. Another study conducted in adults27
found that cholesterol efux from
Fu5AH system to serum was preserved
in hypertriglyceridemic subjects with
normal HDL cholesterol; however, in
those with high triglycerides and low
HDL cholesterol a signicantly reduced
capacity of serum to promote cholesterol efux was observed. The reduced
cholesterol efux in our group of overweight adolescents with atherogenic
dyslipidemia is in agreement with
these studies in adults.18,27 The remarkable similarity between our results
and those reported in adults demonstrates that the abnormal HDL efux
capacity already is present in dyslipidemic overweight adolescents. Together, these results indicate that HDL
from adolescents and adults with
atherogenic dyslipidemia are not only
low but also dysfunctional. The relevance of our ndings is highlighted by
a study17 suggesting that cholesterol
efux capacity may be an independent
predictor of cardiovascular risk, even
after adjusting for HDL cholesterol and
apoAI values. The prognostic implications of the HDL abnormalities found in
our group of overweight dyslipidemic
subjects are not known but may contribute to a higher cardiovascular risk.
The ability of HDL to promote cholesterol efux may be affected by HDL size,
e1525
CONCLUSIONS
It is known that lipoprotein levels track
strongly from childhood and adolescence to adulthood, and abnormal lipoprotein concentrations in early life
may induce arterial changes that contribute to adult atherosclerosis. Our
results show that the simple measurement of HDL cholesterol and triglycerides allows for the identication of
overweight adolescents with a global
adverse atherogenic prole, including
physicochemical and functional abnormalities of HDL that may increase their
risk of future development of diabetes
and cardiovascular disease.
ACKNOWLEDGMENTS
We are grateful to the school authorities and to the adolescents and their
parents for making this study possible.
The authors also thank Dr Margarita
de la Llera-Moya and Dr George H.
Rothblat for providing the Fu-5AH cells
and for their assistance in the preparation of the cholesterol efux assays.
We also appreciate the kind collaboration of Maria del Carmen Gonzlez
Salazar for collecting anthropometric
measurements.
REFERENCES
1. McGill HC Jr, McMahan CA, Zieske AW, et al.
Association of coronary heart disease risk
factors with microscopic qualities of coronary atherosclerosis in youth. Circulation.
2000;102(4):374 379
2. Berenson GS, Wattigney WA, Tracy RE, et al.
Atherosclerosis of the aorta and coronary
arteries and cardiovascular risk factors in
persons aged 6 to 30 years and studied at
e1526
MEDINA-URRUTIA et al
tor against coronary heart disease: the Framingham Study. Am J Med. 1977;62(5):
707714
5. Asztalos BF, Collins D, Cupples LA, et al.
Value of high-density lipoprotein (HDL) subpopulations in predicting recurrent cardiovascular events in the Veterans Affairs HDL
Intervention Trial. Arterioscler Thromb
Vasc Biol. 2005;25(10):21852191
ARTICLES
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
e1527
References
Subspecialty Collections
Reprints
The online version of this article, along with updated information and services, is
located on the World Wide Web at:
http://pediatrics.aappublications.org/content/127/6/e1521.full.html