Variability
in Response
to a Low-Fat,
Low-Cholesterol
Diet
in Children
with
Elevated
Low-Density
Lipoprotein
Cholesterol
Levels
Eric S. Quivers, MD*; David
J.
Driscoll, MDC; Colleen D. Garvey;Ann M. Harrist; Jay Harrison, MSII; Diane M. Huse, MS;
Paul Murtaugh, PhDII; and William H. Weidman, MD*
ABSTRACT. The reduction of dietary cholesterol and fat lowers low-density lipoprotein cholesterol (LDL-C) and reduces risk of coronary heart disease in adults. The
purpose of this study was to determine the individual variability of response of serum lipid and lipoprotein levels to a low-fat, low-cholesterol diet in children with elevated LDL-C levels. Thirty-two children (2 to 16 years of age) enrolled in a diet modification program, who had LDL-C levels of at least 110 mg/dL but normal triglyc-eride levels for their ages, were studied. Lipid levels and dietary nutrients were analyzed at the time of admission, and final assessments were made at least 3 months after entry. There was a significant correlation, for the group as a whole, between change in LDL-C concentration and change in grams of dietary saturated fat; however, there was marked individual variability in LDL-C response. There were no significant correlations between changes in LDL-C levels and changes in either total fat, polyun-saturated fat, or cholesterol intake. It is concluded that modest decreases in dietary saturated fat coincide with a lowering of LDL-C concentration, over a short term, in many children, but the degree of lowering varies consid-erably from one child to another. This variability is consistent with the concept that response of serum lipid levels to dietary changes is modified by genetic, meta-bolic, and other, as of yet, undefined variables. Pediatrics
1992;89:925-929; low-density lipoprotein cholesterol, die-tary saturated fat, variable response.
ABBREVIATIONS. LDL-C, low-density lipoprotein cholesterol; HDL-C, high-density lipoprotein cholesterol; NS, not significant.
Atherosclerosis, or its precursors, begins in
child-hood.’3 There is evidence that increased levels of
serum cholesterol, particularly low-density lipopro-tein cholesterol (LDL-C), are responsible, in a large part, for the development of atherosclerosis.45 Chil-dren in the United States have higher levels of blood
cholesterol than do children in countries where the
incidence of coronary heart disease in adults is low.6’7
In addition, children with high serum cholesterol
levels have an increased risk of having high levels
subsequently as adults.8 Evidence strongly suggests that lowering blood cholesterol levels in children will
* From the Section of Pediatric Cardiology, Cardiovascular Health Clinic for the Young, §Department of Nutrition, and IlSection of Medical Statistics,
Mayo Clinic, Rochester, MN. Dr Quivers is a fellow of the Mayo Clinic.
Received for publication Jun 22, 1990; accepted Aug 26, 1991.
Reprint requests to (D.J.D.) Pediatric Cardiology, Mayo Clinic, Rochester,
MN 55905.
PEDIATRICS (lSSN 0031 4005). Copyright ‘i) 1992 by the American Acad-emy of Pediatrics.
reduce their risk of coronary heart disease during
adulthood. The first strategy to lower lipid levels in American children is the development of nutritional habits that include the limitation of dietary fat to 30%
of calories, with no more than 10% from saturated
fat, up to 10% from polyunsaturated fat, the
remain-der from monosaturated fat, and less than 300 mg of
cholesterol per day (American Heart Association and
National Cholesterol Education Program Step 1 Diet). It is important to know what effect can be expected.
The response of plasma lipid levels to change in
dietary nutrients has been studied mostly in adults, with fewer studies in children. Almost all published
studies suffer from the same problems. Most studies
of the response of serum lipid levels, in adults, to
dietary nutrients have focused on average effects and
present only mean Other reports have been
of small groups studied under tightly controlled
en-vironments.12’4 Jacobs et al’5 studied 58 mentally
retarded, institutionalized males. The actual response of serum cholesterol level to diets low in saturated fat
was related to the expected response based on the
Keys-Minnesota equation. Eighty-two percent
re-sponded within 50% of predicted, 9% were
hypores-ponders, and another 9% were hyperresponders.
McGandy et al’6 and Stein et al’7 reported the
re-sponse of serum cholesterol levels, in adolescent
males, to diets high in polyunsaturated fat. The
de-crease in serum cholesterol concentration was slightly
less in those with admission levels equal to or less
than 199 mg/dL than in those with levels equal to or
exceeding 200 mg/dL. However, as in many of the
adult studies, only mean values were given, individ-ual baseline and response levels were not presented, and the possibility that regression to the mean could have been partly responsible for these differences
was not addressed. Other reports concerning dietary
changes and effect on serum cholesterol and LDL-C
levels in children have also presented only mean
values.’82#{176} In two studies21’22 only mean lipid level
changes were reported. The authors stated that
dif-ferences in baseline dietary fat and cholesterol2’ or
baseline levels of total and LDL-C22 effected response. Neither study presented a statistical analysis of
in-dependent factors effecting response of blood lipid
levels.
Individual baseline blood levels and the
corre-sponding response levels are rarely presented, and
there are no frequency distributions of the responses.
im-Dietary Variables Average calories Total fat, g
%Calories from total
TABLE 1. Nutrition and Blood Lipid Assessment
SD 363.54 15.91 5.73 CV* 0.25 0.32 0.20 Average (Range) Mean (Range) 1475.47 (841-2437) 48.43 (16-86) 29.35 (12.3-40.2) 14.84 (4.0-28.0) 9.03 (3.4-14.2) 6.74 (1.6-19.0) 4.24 (1.4-11.7) 132.46 (19-234) 201.84 (162-329) 147.69 (114.8-284.2) 44.81 (33-59) 59.71 (24-115) 5,39 0.36 2.34 0.26 3.63 0.54 2.52 0.59 63.78 0.48 31.18 32.83 7.79 23.60 0.15 0.22 0.17 0.40
portant in the evaluation of the effect of dietary
changes on a population as is the average response.
When investigators at the Mayo Clinic opened the
Cardiovascular Health Clinic for the Young in 1987,
a protocol was developed to allow evaluation of the
nutritional management of children and adolescents
with hyperlipidemia. The present study was done to
evaluate the short-term (3 to 40 months) effectiveness
of dietary management, using the Step 1 Diet, in
children with serum LDL-C levels above 1 10 mg/dL
(75th percentile) and normal total triglyceride levels (less than the 95th percentile for age).
MATERIALS AND METHODS
From the 259 patients enrolled in the Cardiovascular Health Clinic for the Young between 1986 and 1989, we selected a subgroup of 32 patients (2 to 16 years of age) having (1) two initial LDL-C values of at least 1 10 mg/dL; (2) normal triglyceride levels for their age groups (initial triglyceride levels <100 mg/dL for 2-through 9-year-old children, <129 mg/dL for 10- through 14-year-olds, and <152 mg/dL for 15- to 16-year-olds); (3) at least 90 days of participation in the study; (4) at least three lipid and lipoprotein measurements; and (5) at least two nutritional assessments. The majority of the patients had lipid levels measured because of a family history of hyperlipidemia and/or premature vascular dis-ease. Of the admission lipid measurements, the one closest to the diet evaluation was used; there was an average of 1 1 days between this lipid measurement and the first nutritional assessment, and 5 days between the last lipid measurement and last nutritional as-sessment (Table 1). Lipid measurements preceded nutritional as-sessments. These 32 patients (16 female, 16 male) were enrolled between March 26, 1987, and September 7, 1989, and had a median
time on study of 401 days (range, 137 to 1195 days). The rationale for selecting this subgroup was to limit the cohort to one with
LDL-Celevation only so that very-low-density lipoprotein abnormalities
in some, but not others, would not complicate interpretation of
dietary response of LDL-C.
The admission and follow-up lipid and lipoprotein levels and dietary nutrients were compared with Wilcoxon signed-rank tests. Relationships between the changes in LDL-C levels and changes in dietary variables were examined with Spearman rank
correla-tions.
The evaluation and management of these patients was per-formed in a uniform fashion by a team consisting of two pediatric cardiologists, three pediatric registered dietitians, and a clinic co-ordinator. During the initial visit, a complete medical history and a detailed family history with particular reference to hyperlipidemia, obesity, diabetes, hypertension, stroke, and myocardial infarction were obtained. All patients had a physical examination; remea-surement of serum levels of total cholesterol, triglycerides, high-density lipoprotein cholesterol (HDL-C), thyroxine, sodium, potas-sium, serum protein, albumin, bilirubin, aspartate aminotrans-ferase, and creatinine; and urinalysis. Total cholesterol, total tn-glyceride, sodium, potassium, protein, albumin, bilirubin, aspartate aminotransferase, and creatinine levels were measured enzymati-cally using the Hitachi 717 analyzer. High-density lipoprotein cholesterol was measured by selective precipitation with dextran
No. of lipid measurements 4 (2-8)
Days between first and last lipid 444 (137-1195)
measurements
No. of nutritional assessments 3 (2-6)
Days between first and last nutn- 437 (1 19-1176) tional assessments
Days between first laboratory visit I 1 (0-3 1) and first nutritional assessment
Days between last lipid measure- 5 (0-1 1) .
ment and last nutritional
assess-ment
sulfate and calcium chloride. Low-density lipoprotein cholesterol was calculated using the Friedewald formula (LDL-C = total cho-lesterol - HDL-C - total triglycerides -‘- 5). The patient and family
were instructed to maintain a prospective 3- to 5-day dietary history (including only one weekend day). Three weeks after the initial visit, the results of laboratory tests and the dietary history were reviewed with the family and initial dietary instruction was
pro-vided. Dietary nutrients were estimated using the Nutritionist III Computer Program. Depending on the patient’s age, several aids to dietary modification were employed to simplify adherence to dietary alterations. Subsequently, dietary counseling was per-formed 6, 12, 18, 24, 30, 36, 42, 48, and 52 weeks after the initial
visit and subsequently every 3 to 6 months. Serum lipid and
lipoprotein levels, height, weight, blood pressure, and hip and waist measurements were obtained about 12, 30, and 48 weeks after the initial visit and every 6 months thereafter. Prospective 3-to 5-day diet diaries also were obtained at the time of the lipid measurements (Table 1).
The goal of the dietary modification was to provide a Step I Diet: 30% of calories from fat, less than 10% from saturated fat, no more than 10% from polyunsaturated fat, and the remainder from monosaturated fat; dietary cholesterol less than 300 mg/d; and adequate caloric intake for growth, physical activity, and maintenance of normal weight. The primary tool in teaching the concept of the fat/cholesterol content of foods was a multicolored graphic display giving the relative fat content of a variety of common and popular foods. This diagram illustrated how the fat content of food can change by method of food preparation and gave ideas for substituting lower for higher fat foods. The child was encouraged to use this diagram daily for help with food selection. The child was not told that he or she must avoid any food (including foods containing saturated fat and cholesterol). He or she was told that all foods were acceptable but their frequency of use and amount eaten were the primary issues. Dietary changes were tailored to each child’s individual needs and desires and usually only one change was made in the diet at a time,
The evaluation of compliance to diet in a cohort of free-living subjects is difficult. Compliance was evaluated in several ways: 5-day diet diaries were completed by the patient and parents at the time of admission, at 18 weeks, 36 weeks, 52 weeks, and every 6 months thereafter following admission; these diaries were reviewed by the dietitian and the entries were reviewed with the patient and parents. Worksheets were completed by the patient and parents
every 6 weeks to assess the patients’ and parents’ understanding of the nutritional recommendations; these were discussed in
meet-ings of the dietitian and family. We believe compliance was
ac-ceptable for the purposes of the study.
RESULTS
Table 2 presents descriptive statistics for initial values of the dietary and serum variables. At the time
TABLE 2. Admission Evaluation
fat Saturated fat, g
% Calories from sat-urated fat Polyunsaturated fat,
g
% Calories from pol-yunsaturated fat Dietary cholesterol,
mg/d
Serum variables, mg/ dL
Total cholesterol LDLt cholesterol HDL cholesterol Triglycerides
* Coefficient of variation = standard deviation/mean.
t Low-density lipoprotein.
of admission, the average of the daily mean total
calories was 1475.5; the average of the daily mean
percentage of calories from total fat was 29.4%; from
saturated fat, 9.0%, and from polyunsaturated fat,
4.2%. The mean intake of saturated fat was 14.8 g.
The mean daily cholesterol intake was 132.5 mg. The
mean serum total cholesterol level at entry was 201.8
mg/dL; LDL-C, 147.7 mg/dL; HDL-C, 44.8 mg/dL;
and triglycerides, 59.7 mg/dL.
Table 3 summarizes variable values at the times of
the last measurements, and Table 4 shows the
changes in dietary and serum variables (final minus
initial). At the final dietary assessment, the daily mean total calorie intake increased to 1544.3 (not significant [NSJ); the daily mean percentage of calories from total
fat decreased to 29.2 (P .016), and the mean
per-centage of calories from saturated fat decreased to 7.7
(P = <.001). The mean intake of saturated fat
de-TABLE 3. Final Values
Dietary Variables Mean (Range) SD CV* Average calories 1544.31 (892-2706) 417.73 0.27 Total fat, g 49.04 (25-90) 16.58 0.34
%Calories from total 29.22 (19.6-81.1) 10.68 0.37 fat
Saturated fat, g 13.18 (6.2-28.2) 4.99 0.38
% Calories from satu- 7.68(4.7-13.2) 1.95 0.25 rated fat
Polyunsaturated fat, 7.25 (3.0-24.0) 4.16 0.57
g
% Calories from pol- 4.29 (1.7-9.9) 2.06 0.48 yunsaturated fat
Dietary cholesterol, 110.18 (47-214) 33.52 0.30
mg/d
Serum variables, mgi
dL
Total cholesterol 190.88 (149-312) 34.20 0.18 LDLt cholesterol 132.48 (85.7-276.0) 37.54 0.28 HDL cholesterol 45.06 (26-64) 9.02 0.20
Triglycerides 83.38 (28-242) 50.41 0.60
* Coefficient of variation = standard deviation/mean. t Low-density lipoprotein.
High-density lipoprotein.
TABLE 4. Differe nces Between Admission and Fina 1 Evaluation
Dietary Variables Mean (Range) SD P Value*
Average calories 68.84 (-716-1081) 396.09 .430 Total fat, g 0.62 (-49-51) 20.94 .906
% Calories from to- -0.13 (-14.0-53.9) 11.84 .016 tal fat
Saturated fat, g -1.66 (-15.7-9.2) 5.04 .115
% Calories from sat- -1.35 (-5.4-2.1) 1.93 <.001 urated fat
Polyunsaturated fat, 0.51 (-11.5-15.7) 5.75 .570
g
% Calories from pol- 0.05 (-8.8-7.0) 3.43 .546
yunsaturated fat
Dietary cholesterol, -22.28 (-145-109) 64.20 .101 mg/d
Serum variables, mg/dL
Total cholesterol -10.97 (-38-45) 20.99 .003 LDLt cholesterol -15.20 (-41.7-36.8) 18.98 <.001 HDI4 cholesterol 0.25 (-12-24) 6.37 .960 Triglycerides 25.29 (-47-203) 52.10 .034
* P values are for two-sided signed rank tests of whether the
changes are significantly different from zero. t Low-density lipoprotein.
High-density lipoprotein.
creased to 13.2 g (NS). However, these values varied
widely among patients, with final percentage of
cal-ories from total fat ranging from 19.6 to 81 .1, per-centage of calories from saturated fat from 4.7 to 13.2,
and grams of saturated fat from 6.2 to 28.2. The
percentage of calories from polyunsaturated fat
in-creased to an average of 7.25 and dietary cholesterol decreased to an average of 1 10.2 mg; neither of these differences was statistically significant.
Among the lipid measurements, significant
de-creases were observed in LDL-C (mean 132.5 mg/dL,
P < .001) and total serum cholesterol (mean 190.9 mg/dL, P = .003); see Tables 3 and 4. Triglyceride
levels increased significantly (P = .034), to a mean
value of 83.4 mg/dL. There was no significant change
in HDL-C level. The range for final LDL-C
measure-ments was 86 to 276 mg/dL; for HDL-C, 26 to 64
mg/dL; for total cholesterol, 149 to 312 mg/dL; and
for triglycerides, 28 to 242 mg/dL.
Spearman rank correlation coefficients were used
to examine relationships between changes in LDL-C
values and changes in dietary variables (Table 5).
There was a significant correlation between LDL-C
changes and change in grams of dietary saturated fat
(P = .025). There were no significant correlations
between change in LDL-C and changes in total fat,
polyunsaturated fat, and dietary cholesterol. We also
analyzed the rates of change of LDL-C (magnitude of
change divided by time between initial and final
measurements, in milligrams per deciliter per day).
The rate of change of LDL-C was significantly
cone-lated with change in grams of saturated fat (P = .004)
but not with any of the other dietary variables.
The average decrease in LDL-C was 10.4% for the
group. Eighteen of the 32 had at least a 10%
reduc-TABLE 5. Statistical Analysis
Dependent Variable : Change i n LDL (Final - Initial) Independent Variable Pearson P Spearman P
(Change = Final - Ini- Come- Value Cone- Value
hal) lation lation
Total fat, g .261 .148 .275 .128
% Calories from total fat .003 .986 .000 .998
Saturated fat, g .296 .100 .396 .025
% Calories from satu- .015 .936 .014 .938
rated fat
Polyunsaturated fat, g .297 .099 .161 .380
% Calories from polyun- .135 .463 .070 .703
saturated fat
Dietary cholesterol, mg/ .142 .437 .123 .501 d
Dependent Variable: Rate of Change of LDL* (U/d)
Independent Variable
(Change = Final - Ini-tial) Pearson Come-lation P Value Spearman Come-lation P Value
Total fat, g
% Calories from total fat
.164 -.095 .370 .606 .303 .022 .092 .903
Saturated fat, g
% Calories from
satu-.314 .048 .080 .793 .494 .100 .004 .586 rated fat
Polyunsaturated fat, g
%Calories from
polyun-saturated fat .437 .274 .013 .130 .224 .126 .219 .492
Dietary cholesterol, mg/ d
.035 .850 .091 .619
0
r=0.296 (P-0.1)
I I i I I I I I i#{149} I
20
0)
E ____________________________ _______________
, -20
-15 -10 -5 -3 -2 .1 0 1 2 3 5
I. ReduCtiOn ‘I. Increase ‘I
A Sat. fat, 9
Fig 1. Individual change in grams of dietary saturated fat plotted against change in milligrams per deciliter of calculated low-density lipoprotein (LDL) cholesterol.
C
0
Lu.
0
Change in LDL, %
Fig 2. Frequency of the percentage of decrease or increase in low-density lipoprotein (LDL) cholesterol between first and last lipid measurements.
.40 -30 -20 -10 10 20 30 40
30
0
c O
190 110 150 03
E 130
-I
0 110 -I
90 10
100 2 m 400 500 600
0 100 200 0 400 500 600
Days from entry
22 20 0) 18
i 16
14
c 12
10
220 200
#{149}0 03 180 . 160 140 120 100
Fig 3. Upper panel demonstrates change in grams of dietary satu-rated fat in six individuals and corresponding low-density lipopro-tein (LDL) cholesterol levels in the lower panel (diet “responders”).
0 100 200 4153 500 600 700 800 Days from entry
Fig 4. Upper panel demonstrates change in grams of dietary satu-rated fat in six individuals and corresponding low-density lipopro-tein (LDL) cholesterol levels in the lower panel (diet “nonrespond-em’).
tion. Although there was an overall correlation of
change in the amount of saturated fat in the diet with
change in LDL-C, there was considerable individual
variability in response of LDL-C to change in
satu-rated fat (Figs 1 and 2). Some individuals had large
decreases in LDL-C with decreases in dietary
satu-rated fat, whereas others demonstrated a lesser
de-crease or even an increase of LDL-C with similar
dietary changes. In Fig 3, dietary and LDL-C changes
are demonstrated in six patients in whom there was
an appropriate LDL-C response to change in dietary
saturated fat (responders). In Fig 4, dietary saturated
fat amounts and corresponding LDL-C levels are
demonstrated in six patients who did not have an
appropriate LDL-C response to change in dietary
saturated fat (nonresponders).
DISCUSSION
Interpretation of the data was made difficult
be-cause of certain confounding factors.
All children in this study were referred to the
Health Clinic because they came from families in
which an adult member had hyperlipidemia or
pre-mature cardiovascular disease, and all had LDL-C
levels of 1 1 0 mg/dL or higher. As a result, the cohort
could have been enriched with children in whom the
genetic effect on lipid levels is stronger than in the general population.
In addition, many of the families already had mod-ified the family diet and some of the children at the time of admission were eating diets with total fat less
than 30% of calories and cholesterol less than 300
mg/d. Dietary counseling involved not only decreas-ing total fat, saturated fat, and cholesterol intake in
10 some children, but also increasing dietary fat in five
children in whom dietary total fat calories were less than 20%. The goal of the diet program was a gradual
movement of both groups of children toward a Step
1 Diet. The fact that the initial diet of some of the children was already low in total fat, saturated fat, and cholesterol suggests that the results, after change in diet, may understate results that could occur if the initial diet had been higher in total fat, saturated fat, and cholesterol.
The low (4.2% of calories) dietary polyunsaturated
fat at the time of admission and our inability to
significantly increase this percentage in most children
was surprising. The major reason for this was that
the children in the study used very little
oleomargar-me and preferred to increase their fat intake with
foods high in monosaturated fat such as peanut
but-ter.
Although, for the group as a whole, there was not
correlated with changes in LDL-C concentration. This correlation did not exist for change in the percentage of total calories provided by saturated fat. This
sug-gests that the actual amount of saturated fat in the
diet and not its contribution to total calories affected LDL-C level.
The results of this study indicate that a short-term reduction of the grams of dietary saturated fat coin-cides with a reduction in serum total cholesterol and
LDL-C values in most children with elevated LDL-C,
without adversely affecting serum HDL-C level and
with only a slight increase in total triglyceride value.
However, there is a marked individual variability in
response of LDL-C.
It has been reported that a maximum reduction of
about 10% to 15% of serum LDL-C can be expected
through dietary modification alone. In this study, 18
of the 32 patients reduced their LDL-C levels by at
least 10% following the initiation of diet modification.
The average reduction was 10.4%, but some children
had large changes in LDL-C level associated with
modest dietary changes, whereas others had only
small changes in LDL-C level. It was recognized that
changes in serum lipid and lipoprotein levels could, in part, be the result of regression to the mean. One
might expect that, if regression to the mean were
operating, those patients with the highest initial levels
of LDL-C would show the largest decreases in
LDL-C
over the course of their involvement in the program.A simple correlation of change in LDL-C and initial
value of LDL-C is not enough to resolve this question,
however, because of the automatic negative
correla-tions between the change (final minus initial) and the initial value. This “spurious’ correlation, which
con-founds other reported comparisons of changes in
serum lipids and initial values, has hindered our
assessment of the role of regression to the mean in
explaining our results, and the issue remains
unre-solved.
Elevated LDL-C is the result of both environmental (mainly diet) and genetic factors. It is not possible at present to predict, in an individual child with elevated
LDL-C, what the relative effects of environment and
genetics may be in causing the elevation. Those with
marked elevations of LDL-C (>200 mg/dL) are likely to have a monogenic form of hypercholesterolemia
and will respond the least to dietary changes. Most
of the children, however, will have levels between
1 1 0 mg/dL and 1 60 mg/dL and are likely to have a
polygenic form of hypercholesterolemia. It is in this
group that one can expect an average reduction in
LDL-C of 10%; if this reduction were maintained into
adulthood, it would result in about a 20% reduction in the incidence of coronary heart disease. However, it is in this group that response to diet may be most
unpredictable, in spite of dietary compliance. It is
important to monitor lipid levels during dietary man-agement and, as the patient approaches adolescence, if the LDL-C level remains significantly elevated, drug treatment has to be considered.
REFERENCES
I. Holman RL. Atherosclerosis: a pediatric nutrition problem. Am I Clin
Nutr. 1961;9:565
2. Strong JP, McGill HC Jr. The natural history of coronary atherosclerosis. Am IPathol. 1962;40:37-49
3. Strong JP, McGill HC Jr. The pediatric aspects of atherosclerosis. / Atheroscler Res. 1969;9:251-266
4. Newman WP, Freedman DS, Voors AW, et al. Serum lipoproteins and
systolic blood pressure are related to atherosclerosis in early life: the
Bogalusa Heart Study. N Engi /Med. 1986;314:138-143
5. Freedman DS, Newman WP III, Tracy RE, et al. Pathophysiology and
natural history of atherosclerosis: black-white differences in aortic streaks in adolescence and early adulthood: the Bogalusa Heart Study.
Circulation. 1988;77:856-864
6. Knuiman JT, Hermus RJ, Haulvast JG. Serum total and high density
lipoprotein (HDL) cholesterol concentration in rural and urban boys from 16 countries. Atherosclerosis. 1980;36:529-537
7. Blackburn H. Epidemiologic evidence and public health implications: Proceedings of The International Symposium on Epidemiology and Prevention of Atherosclerotic Disease. Prey Med. 1983;12:1-.10
8. Lauer RKM, Lee J, Clarke WR. Factors affecting the relationship between
childhood and adult cholesterol levels: the Muscatine Study. Pediatrics.
1988;82:309-31 8
9. Dreon D, Vranizan K, Krauss R, et al. The effects of polyunsaturated
fat versus monosaturated fat on plasma lipoproteins. /AMA. 1990;
263:2462-2466
10. Budow L, Beseler L, Arden M, et al. The effect of dietary therapy on
growth in hyperlipidemic children and adolescents: a prospective clinical trial. Pediatr Res. 1989;25:22A
I I. Mattson F, Erickson B, Kligman A. Effect of dietary cholesterol on serum
cholesterol in man. Am /Clin Nutr. 1972;25:589-594
12. Grundy S. Comparison of monosaturated fatty acids and carbohydrates
for lowering plasma cholesterol. N EngI /Med. 1986;314:745-748
1 3. Quintav E, Grundy M, Ahrens E. Effects of dietary cholesterol on the
regulation of total body cholesterol in man. /Lipid Res. 1971;12:233 14. Denke M, Breslow J. Effects of a low fat diet with and without
intermit-tent saturated fat and cholesterol ingestion on plasma lipid, lipoprotein, and apoproteins in normal volunteers. /Lipid Res. 1988;29:963-969
15. Jacobs D, Anderson J, Hannen P. et al. Variability in individual serum
cholesterol response to change in diet. Arteriosclerosis. 1983;3:349-356 16. McGandy R, Hall B, Ford C, et al. Dietary regulation of blood cholesterol
in adolescent males: a pilot study. Am /Clin Nutr. 1972;25:61-66
I 7. Stein E, Mendelsohn D, Fleming M, et al. Lowering of plasma cholesterol
levels in free-living adolescent males: use of natural and synthetic
polyunsaturated foods to provide balanced fat diets. Am / Clin Nutr. 1975;28:1204-1216
I 8. Walter H. Modification of blood cholesterol levels in a school-based
population. In: Primary prevention of atherosclerosis in childhood: the
role of lipids. Proceedings from a videoconference; March 28, 1985. New York, NY
19. Kwiterovich P. Bachorik P. Franklin F, et aI. Effect of dietary treatment
on the plasma levels of lipids, lipoprotein cholesterol and LDL B protein
in children with type II hyperlipidemia. Prog Clin Biol Res. 1985;188:123-137
20. Widhalm K. Dietary treatment of hypercholesterolemia in children: recent aspects. Prog Clin Biol Res. 1988;255:219-230
21. McMurray M, Hopkins P. Gould R, et al. Family-oriented nutrition
intervention for a lipid clinic population. /Am Diet Assoc. 1991;91:57-65
22. Glassman M, Spars A, Berezin 5, et al. Treatment of type ha
hyperlipi-demia in childhood by a simplified American Heart Association diet and
