Effect of 7-Year Infancy-Onset Dietary Intervention on Serum Lipoproteins and Lipoprotein Subclasses in Healthy Children in the Prospective, Randomized Special Turku Coronary Risk Factor Intervention Project for Children (STRIP) Study
Background— We previously showed that low-saturated-fat dietary intervention from infancy until 5 years of age safely and effectively reduced serum cholesterol concentration. We now report how such intervention influenced serum lipids, LDL particle size, and HDL subfractions in children when they reached the age of 7 years.
Methods and Results— Healthy 7-month-old infants (n=1062) were randomized to the intervention (n=540) and control (n=522) groups. Each year, two individualized counseling sessions were organized to the intervention families. Serum lipid values were measured annually. The intervention boys had 0.20 to 0.39 mmol/L lower serum cholesterol values than the control boys throughout the follow-up (always P<0.05), but the values of the intervention and control girls did not differ. The LDL particle sizes and HDL subfractions were determined in a random subgroup of 96 intervention and 101 control children at the age of 7 years. The mean particle diameter of major LDL peak was 262.6 Å in the intervention boys and 258.5 Å in the control boys (P=0.05), and 259.2 Å in the intervention girls and 261.3 Å in the control girls (P=0.30). HDL2 and HDL3 cholesterol concentrations did not differ between the intervention and control children or between the two genders.
Conclusions— The 7-year intervention favorably influenced not only the serum total and LDL cholesterol concentrations but also the LDL particle size in boys. LDL particle size remained unchanged in girls, as did HDL2 and HDL3 concentrations in both genders.
Received September 17, 2002; de novo received January 30, 2003; revision received May 15, 2003; accepted May 16, 2003.
High LDL cholesterol concentration is an important risk factor for coronary heart disease (CHD).1 The size of LDL particles may determine their atherogenicity. The predominance of small, dense LDL (sdLDL) or LDL subclass pattern B (diameter of most LDL particles <255 Å) may be particularly atherogenic.2,3 This may be due to their low binding affinity to the LDL receptor, low resistance to oxidative stress, and prolonged plasma half-life.3,4 Small particle size may promote penetration into the intima and LDL cholesterol accumulation in the arterial wall, accelerating the development of atherosclerosis.3,4 Both genetic and environmental factors can influence LDL particle size.5–7
HDL particles can be divided into HDL2 and HDL3 fractions according to their density. The lipid-rich HDL2 may be more important in the protection from atherosclerosis.8
We recently showed that individualized, repeatedly given dietary counseling aiming at a low–saturated-fat, low-cholesterol diet from the age of 7 months markedly lowered total and non-HDL cholesterol and apolipoprotein B concentrations up to the age of 5 years.9,10 We have extended the study to 7 years of age, and report now, in addition to the follow-up data, for the first time the effects of the intervention on serum LDL and HDL fractions.
Study Design and Subjects
The study design of the on-going prospective, randomized STRIP study (Special Turku Coronary Risk Factor Intervention Project for Children), which began in 1990, has been published.9 Briefly, 1062 healthy infants were recruited and randomized to an intervention group (n=540) or a control group (n=522).
The intervention group received twice a year dietary and lifestyle counseling given by a team consisting of physicians and dietitians. The families recorded child’s food consumption for four days twice a year as described.9 The purpose of the counseling was to minimize children’s exposure to known environmental atherosclerosis risk factors. A dietitian checked the food records and suggested appropriate changes to diet. The diet was designed to meet the Nordic Dietary Recommendations.11 Since the age of 3 years, the recommended intakes comprised protein 10 to 15 E% (percentage of energy), fat 30 E% (saturated fat ≤10 E%), and carbohydrate 55 to 60 E%.
The control children have received the basic health education given at Finnish well-baby clinics. During their regular STRIP visits, they received no detailed dietary counseling.
Data used in the longitudinal analyses of serum lipid values comprised all those children whose blood samples had been successfully obtained at the ages of 7 months (baseline) and 7 years (n=511). The HDL and LDL subfractions were determined in a random subsample of 197 consecutive children at their 7-year STRIP visit (later called the lipoprotein subgroup). Of them, 51 girls and 45 boys were from the intervention group and 50 girls and 51 boys from the control group. LDL particle determination was successfully completed in 176 samples.
The study was approved by the Joint Commission on Ethics of the Turku University and the Turku University Central Hospital. Informed consent was obtained from all parents.
Serum total and HDL cholesterol, apoA-I, apoB, and triglyceride values were measured as described.9,12 Nonfasting blood samples were drawn when children’s age was less than 5 years, whereas fasting samples were taken at the ages of 5 and 7 years. The Friedewald formula13 was used to calculate serum LDL cholesterol values. Differential precipitation with heparin-manganese and low–molecular weight dextran sulfate14 was used to determine the proportions of the HDL fractions HDL2 and HDL3.
LDL particle sizes were determined by nondenaturing polyacrylamide gel electrophoresis using a described method.15,16 The serum samples were stored at −70°C, and commercial 7.5% Tris-HCl gels (BIO-RAD) were used. The gels were stained with Coomassie Brilliant Blue R-250 and scanned with green filter (520 to 570 nm) with a computer-assisted scanning densitometer (Model GS-700 Imaging Densitometer, BIO-RAD). The particle diameter of each LDL peak was determined by comparing the mobility of the sample lipoproteins with the mobility of a calibrated reference LDL preparations run on each gel. The particle diameters of the reference LDL preparations were evaluated using electron microscopy. Coefficients of variation for intragel and intergel precisions for the control sample used were 0.8% and 1.7%, respectively.
Differences in serum lipid values at the age of 7 years between intervention and control groups were tested with a two-sample t test, and the results are presented as mean±SD with 95% confidence interval for the mean (95% CI). Comparison of the distributions of major LDL particle size between intervention and control groups was tested with Wilcoxon two-sample test. Triglyceride values were skewly distributed and therefore log-transformed for the analysis (CIs not presented). Pearson’s correlation coefficients or Spearman’s rank test (triglycerides) were used to evaluate the associations between various variables. The overall intervention effect was evaluated by analyzing the longitudinal data (7 months−7 years) using the analysis of covariance for unbalanced repeated measures with the 7-month value as covariate. The same analysis was also used in evaluating the differences of serum LDL cholesterol and triglyceride concentrations between the groups at the ages of 5 and 7 years. Genders were analyzed separately. In all tests, P≤0.05 was considered significant.
Serum Lipids and Lipoproteins During the 7-Year Intervention
The baseline-adjusted means of serum total and non-HDL cholesterol and apoB concentrations were lower in the intervention boys than in the control boys throughout the 7-year period (Table 1). HDL cholesterol and apoA-I concentrations were lower in the intervention boys until 36 months, but the difference then disappeared during the follow-up.
The intervention and control girls did not differ in the baseline-adjusted means of serum cholesterol, non-HDL cholesterol, HDL/total cholesterol ratio, or apoB concentrations during the 7-year intervention (Table 2). The control girls had higher adjusted means of HDL cholesterol concentration than the intervention girls at the age of 13 months (P=0.004). ApoA-I concentrations were lower in the intervention girls up to 3 years (P=0.008).
The fasting LDL cholesterol and triglyceride concentrations at the ages of 5 and 7 are presented in Figure 1. From 5 to 7 years, triglycerides tended to decrease in the intervention boys and increase in control boys (difference in change between groups, P=0.042). As expected, the LDL cholesterol values were lower in the intervention boys than in the control boys (P<0.0001, 95% CI −0.333 to −0.150 mmol/L). In the intervention girls and control girls, the triglyceride and LDL cholesterol values were similar (P=0.75 and P=0.18, respectively), and the values also remained rather stable from 5 to 7 years of age.
Energy Nutrient Intakes and Serum Lipid Values at 7 Years of Age
At the age of 7 years, the mean daily intakes of energy, total fat, and carbohydrates were 6818 kJ, 30.7 E%, and 52.8 E% in the intervention boys and 6955kJ, 31.6 E% and 52.8 E% in the control boys, respectively (always P>0.20). The intake of saturated fat was 11.5 E% in the intervention boys and 13.6 E% the control boys (P<0.001). The (polyunsaturated+ monounsaturated fat)/saturated fat [(P+M)/S] ratios in the two groups of boys were 1.5 and 1.2 (P=0.002), respectively.
The mean daily intakes of energy, total fat, and carbohydrates were 6256 kJ, 30.4 E%, and 53.2 E% in the intervention girls and 6340 kJ, 31.8 E%, and 52.2 E% in the control girls, respectively (for the differences, P=0.47, P=0.005, and P=0.06). The intake of saturated fat of the intervention girls was 11.8 E% and that of the control girls 13.7 E% (P<0.001). The [(P+M)/S] ratios in the two groups of girls were 1.4 and 1.2 (P<0.001), respectively.
In 7-year-olds, the mean total and LDL cholesterol concentrations were 5% and 7% lower in the intervention boys than in the control boys, respectively (Table 3). The HDL/total cholesterol ratio was 5% higher in the intervention boys than in the control boys. Serum apoA-1 concentration showed no difference, but apoB concentration was 9% lower in the intervention boys than in the control boys.
The 7-year-old intervention and control girls showed no differences in any of the measured lipid, lipoprotein, or apolipoprotein values.
The dietary (P+M)/S ratio correlated inversely with total and LDL cholesterol and apoB concentrations in boys (r=−0.14, r=−0.16, and r=−0.18, respectively; always P<0.02), but not in girls. There was no correlation between carbohydrate intake and any of the lipid variables in either gender.
The heights or weights at the age of 7 years of the intervention and control children and of the two genders were not different (data not shown). No correlation was found between parameters of adiposity (weight, relative weight, and body mass index) and lipid or lipoprotein concentrations except a weak positive correlation between adiposity variables and serum triglycerides in boys (r=0.15, r=0.12, and r=0.15, respectively; always P<0.025).
LDL and HDL Subclasses in 7-Year-Old Children (the Lipoprotein Subgroup)
The mean average LDL particle diameters were 260.3 Å and 260.2 Å in all boys (n=81) and in all girls (n=95), respectively. The intervention boys but not girls had larger major LDL particle diameter than their controls (Table 4). The median of major LDL particle size was 263 Å in intervention boys and 257 Å in control boys, and the distribution curves differed significantly between the groups (P=0.048; Figure 2).
To examine the possible relationship between serum LDL concentration and particle size, we divided the children into two groups using their median LDL cholesterol (2.79 mmol/L) as the cut-off point. The intervention boys with low LDL concentration had large LDL particles, but the control boys with low LDL concentration had small LDL particles (average particle sizes, 264.4 Å and 255.9 Å; P=0.006). The intervention and control boys with LDL cholesterol concentration above the median had closely similar LDL particle sizes (average particle sizes, 261.4 Å and 263.6 Å; P=0.51). In girls, the LDL concentrations and LDL particle sizes were randomly distributed in both the intervention and control groups. Serum triglyceride concentration and LDL particle size correlated poorly in all groups of children (always r<0.20; P=NS).
HDL2 and HDL3 cholesterol concentrations were similar between the intervention and control boys and the intervention and control girls (Table 4).
This study confirms our former findings that dietary counseling given repeatedly starting at the age of 7 months aiming at permanently low intake of saturated fat decreases lipid risk factors of CHD in the children.9,10 Our study demonstrates that serum total cholesterol, HDL and non-HDL cholesterol, apoA-I, and apoB concentrations are persistently lower in the intervention boys than in the control boys when they have reached the age of 7 years. At 5 to 7 years old, children usually go through great changes in their daily life when daycare, preschool, and school usually begin. Despite this, the intervention effect remains essentially similar as in younger children. We also show that LDL particles in 7-year-old intervention boys are larger in size than those in control boys, indicating that the intervention has a favorable effect on the LDL particle composition.
Fasting serum LDL cholesterol and triglycerides were for the first time determined when the children were 5 years old.10 Now, we present these values for the 7-year-old children. LDL cholesterol was lower in the intervention boys than in the control boys at both time points. Serum triglycerides tended to decrease in the intervention boys and to increase in the control boys.
The LDL particles were larger in the intervention boys than in the control boys. This difference was confined to those who had LDL cholesterol values below the median, suggesting that the dietary intervention had a favorable influence on both serum LDL cholesterol concentration and composition of the LDL particles.
In adults, LDL particle size is inversely associated with serum triglycerides.2,17 We did not find such a correlation and nor was it found in a previous study on children under 10 years of age.18 In the Bogalusa heart study, the correlation between triglyceride concentration and LDL particle size in 10- to 17-year-old children was weak (r=−0.21), and many of these subjects were already in puberty.19 Arisaka et al5 observed that triglycerides were higher in 7- to 13-year-old children with LDL pattern B than those with pattern A, but correlation was not reported. Therefore, it is likely that the correlation between triglyceride concentration and LDL particle size strengthens with age.
Changing from high-fat to low-fat, high-carbohydrate diet may produce decreases in LDL cholesterol and HDL cholesterol concentrations and increases in triglycerides.20 LDL subclass pattern contributes to lipoprotein response to change in diet. Dreon et al21 in their study of the effects of short-term diet on LDL subclasses in normolipidemic men found that men with LDL pattern B achieved greater reductions in LDL cholesterol and apoB concentrations and number of small LDL particles than men with LDL pattern A on a low-fat, high-carbohydrate diet. Thus, pattern B subjects seem to benefit the most from a low-fat diet. In a study on younger subjects (mean age 14 years, n=50), a 10-day extremely low-fat (10%), high-carbohydrate diet converted LDL pattern A to pattern B in 6 genetically predisposed children.22 However, the findings of these studies were not explained by changes in serum triglyceride concentration.21,22 Thus, heredity and genetic variation in LDL subclass patterns strongly contribute to LDL particle’s responses to diet. Our study is the first to examine the effect of a long-term dietary intervention on LDL particle distribution in solely prepubertal children. As we measured LDL particle size only once after 7 years of intervention, we are unable to judge whether the effect of intervention depended on the LDL particle size at the beginning of the intervention.
Despite the fact that the intervention reduced saturated fat intake as much in the intervention girls than in the intervention boys, it had no effect on any lipid or lipoprotein variable in the girls at the age of 7 years. In agreement, LDL cholesterol response to sitostanol ester margarine use is greater in boys than in girls.23 Possible causes for inefficient intervention impact on serum lipid values in girls include gender differences in body composition, serum sex hormone concentrations, and exercise habits.
We conclude that the 7-year low–saturated-fat dietary intervention had a favorable influence on serum total and LDL cholesterol concentration, as well as LDL particle size in boys. LDL concentration and particle size remained unchanged in girls, as did the HDL2 and HDL3 concentrations in both genders.
This study was supported by grants from the Mannerheim League for Child Welfare, Finnish Cardiac Research Foundation, Foundation for Pediatric Research, Finland, Academy of Finland, Yrjö Jahnsson Foundation, Sigrid Juselius Foundation, Turku University Foundation, Juho Vainio Foundation, Finnish Cultural Foundation, City of Turku, Raisio Group Research Foundation, and Van den Bergh Foods Company.
National Cholesterol Education Program. Report of the expert panel on blood cholesterol levels in children and adolescents. Pediatrics. 1992; 89 (suppl 3): 2.
Williams PT, Krauss RM, Vranizan KM, et al. Changes in lipoprotein subfractions during diet-induced and exercise-induced weight loss in moderately overweight men. Circulation. 1990; 81: 1293–1304.
Austin MA, Brunzell JD, Fitch WL, et al. Inheritance of low density lipoprotein subclass patterns in familial combined hyperlipidemia. Arteriosclerosis. 1990; 10: 520–530.
Salonen JT, Salonen R, Seppänen K, et al. HDL, HDL2, and HDL3 subfractions, and the risk of acute myocardial infarction: a prospective population study in eastern Finnish men. Circulation. 1991; 84: 129–139.
Niinikoski H, Viikari J, Rönnemaa T, et al. Prospective randomized trial of low-saturated-fat, low cholesterol diet during the first 3 years of life: the STRIP baby project. Circulation. 1996; 94: 1386–1393.
Rask-Nissilä L, Jokinen E, Rönnemaa T, et al. Prospective, randomized, infancy-onset trial of the effects of a low-saturated-fat, low-cholesterol diet on serum lipids and lipoproteins before school age: the Special Turku Coronary Risk Factor Intervention Project (STRIP). Circulation. 2000; 102: 1477–1483.
NNR, Nordic nutrition recommendations 1996. Nordic working group on diet and nutrition. Scand J Nutr. 1996; 40: 161–165.
Friedewald WT, Levy RI, Fredrickson DS. Estimation of the concentration of low-density lipoprotein cholesterol in plasma, without use of the preparative ultracentrifuge. Clin Chem. 1972; 18: 499–502.
Gidez LI, Miller GJ, Burstein M, et al. Separation and quantitation of subclasses of human plasma high density lipoproteins by a simple precipitation procedure. J Lipid Res. 1982; 23: 1206–1223.
Nichols AV, Krauss RM, Musliner TA. Nondenaturing polyacrylamide gel electrophoresis. In: Segrest JP, Albers JJ, eds. Methods in Enzymology: Plasma Lipoproteins. London, UK: Academic Press; 1986; 128: 417.
Lahdenperä S, Tilly-Kiesi M, Vuorinen-Markkola H, et al. Effects of gemfibrozil on low-density lipoprotein particle size, density distribution, and composition in patients with type II diabetes. Diabetes Care. 1993; 16: 584–592.
McNamara JR, Campos H, Ordovas JM, et al. Effect of gender, age, and lipid status on low density lipoprotein subfraction distribution: results of the Framingham Offspring Study. Arteriosclerosis. 1987; 7: 483–490.
Mensink RP, Katan MB. Effect of dietary fatty acids on serum lipids and lipoproteins: a meta-analysis of 27 trials. Arterioscler Thromb. 1992; 12: 911–919.
Dreon DM, Fernstrom HA, Miller B, et al. Low-density lipoprotein subclass patterns and lipoprotein response to a reduced-fat diet in men. FASEB J. 1994; 8: 121–126.
Dreon DM, Fernstrom HA, Williams PT, et al. Reduced LDL particle size in children consuming a very-low-fat diet is related to parental LDL-subclass patterns. Am J Clin Nutr. 2000; 71: 1611–1616.