- Email: [email protected]
Fatty acid composition of serum cholesteryl esters in relation to serum lipids and apolipoproteins in 3–18-year-old finnish children and adolescents
Arherosclerosls, 59 (1986) 113-119 Elsevier Scientific Publishers Ireland,
113 Ltd.
ATH 03721
Fatty Acid Composition of Serum Cholesteryl Esters in Relation to Serum Lipids and Apolipoproteins in 3-l&year-old Finnish Children and Adolescents T. Moilanen’,
T. Solakivi-Jaakkola’, J. Viikari 2, L. R&s;inen 3, H.K. Akerblom M. Uhari 5, M. Pasanen 6 and T. Nikkari ’
4,
’Deportment of Biomedical Scrences and ’Department of Publtc Health, Unioersity of Tumpere, Tampere. .’Department oJMedtcrne, University of Turku, Turku, -’Department of Nutntton and the 4 Children’s Hospital, Unioersity of Helsinkt, Helstnkt, and ’Department of Pediatrtcs. Uniuerstty of Oulu. Oulu (Finland) (Received 26 March, 1985) (Revised, received 26 August, 1985) (Accepted 29 August, 19851
Summary Serum cholesteryl ester (CE) fatty acids, serum lipids and apolipoproteins were analysed from 1348 Finnish children aged from 3 to 18 years. The study was part of a comprehensive survey of coronary heart disease risk factors and their determinants in Finnish children and adolescents. The percentage of CE linoleate (18 : 2) had significant negative correlations with the serum concentrations of total and LDL cholesterol, triglycerides as well as apo B. The associations of HDL to CE fatty acids were weak, although there was a significant positive correlation of CE-18 : 2 to HDL cholesterol and apo A-I. The ratio of HDL to total cholesterol and the ratio of apo A-I to apo B increased strongly in quintiles of increasing percentage of CE-18 : 2. In conclusion, the present results show significant associations between serum CE fatty acid composition and the different components of serum lipoproteins among free-living Finnish children and imply a relation between the quality of dietary fats and serum lipoproteins within this population.
Key words:
Apolipoproteins lipids
- Children - Cholesteryl
The study was supported by grants from the Operation Red Heart Fund of Lions International, Finland, the Research Council for Medicine. Academy of Finland and the Juho Vainio Foundation. Address for correspondence: Teemu Moilanen, M.D., Department of Biomedical Sciences, University of Tampere, P.O. Box 607, SF-33101 Tampere 10, Finland. 0165-0327/86/$03.50
‘n 1986 Elsevier Science Publishers
esters - Linoleic acid - Serum fatty acids - Serum
Introduction Saturated animal fat and cholesterol in the diet increase the total and LDL cholesterol as well as apo B concentrations in serum, while polyunsaturated fat has an opposite effect [1,2]. Reports
B.V. (Biomedical
Division)
114 on the influence of dietary fat on HDL cholesterol and apo A-I are more controversial. In some studies an increase in the ratio of polyunsaturated to saturated (P/S) fatty acids in the diet brought about an elevation of serum HDL [3,4], while no influence [5] or the opposite effect was obtained in others [6,7]. Although dietary interventions generally produce clear-cut effects on serum lipids, cross-sectional studies relating the diet to serum lipid concentrations within a population have usually shown only weak associations [8,9]. The fatty acid composition of serum lipids reflects the quality of dietary fat and is a useful estimate of the type of fat used by an individual [lo-121. Subjects with coronary heart disease (CHD) have reduced proportions of linoleic acid in serum and adipose tissue [13,14] and it has been suggested that the composition of serum fatty acids is also an independent risk factor of CHD [15]. Since the rate of CHD in Finland is high [16], the levels of various CHD risk factors in Finnish children and adolescents were surveyed in a comprehensive multicentre study in 1980 [17]. The study included among others a dietary interview and determinations of serum total and HDL cholesterol, triglycerides, apolipoproteins A-I and B, as well as an analysis of fatty acids in serum cholesteryl esters (CE). The results on correlations between serum fatty acids and dietary data have been published elsewhere [18]. This report gives the relationships of serum fatty acid compositions to serum lipids and apolipoproteins in 1348 Finnish children that were a part of the multicentre study. Subjects and Methods The study was carried out in October-November 1980 in 5 university cities and their surrounding rural communities in Finland. An invitation with a description of the study was sent to the parents, whereafter 83% of the invited children participated. The study included a total of 3,596 boys and girls aged 3, 6, 9, 12, 15 and 18 years. A detailed description of the population sample is given in ref. [17]. The fatty acid composition of serum CEs was analysed in 1348 children representing all age groups and areas and both sexes. Venous blood was drawn after an overnight
fast, and serum was separated after 2 h clotting at room temperature. Serum aliquots were stored at -6O’C for l-14 months until analysed. CE fatty acid composition remains stable at this temperature for at least one year [19]. Serum cholesterol concentrations were measured using an enzymatic Boehringer CHOD-PAP kit, triglycerides with a Boehringer fully enzymatic system and an automatic analyzer, and HDL cholesterol after precipitation of very low and low density lipoproteins with Mg2+ and dextran sulphate 500 [20]. LDL cholesterol was calculated from the equation of Friedewald et al. [21]. Apolipoproteins A-I and B were determined with radial immunodiffusion [22]. The inter-assay coefficients of variation for apo A-I and apo B were 5.1 and 4.0%, respectively. The procedures followed in fatty acid analyses have been described elsewhere [19]. In short, lipids were extracted from 0.5-ml aliquots of serum with chloroform/methanol (1 : 1, v/v). The extract was fractionated with thin-layer chromatography, and the fraction corresponding to CE was saponified with KOH/ethanol. Methyl esters of fatty acids were analysed in a Varian 2100 gas chromatograph using two 0.3 cm (i.d.) x 180 cm glass columns packed with 3% EGSS-X on loo-120 mesh Gas Chrom Q (Applied Science Lab.). The temperature was programmed from 140 to 190°C and all peaks corresponding to fatty acids from 14: 0 to 22 : 6 were quantified with a flame ionisation detector and Hewlett-Packard model 5880 A integrator. The reproducibility of the fatty acid analysis has been reported earlier [23]. The coefficient of variation “was 0.4% for linoleic acid (18 : 206) and ranged from 0.6 to 4.1% for most fatty acids. All statistical calculations were done at the Computer Centre of the University of Tampere. Linear correlation coefficients were calculated from the total sample of 1348 children. Age- and sexadjusted quintiles of CE-18: 2 percentage were formed by dividing each age and sex class into corresponding quintiles and combining these classifications. Results Fatty acid composition of serum CEs in the total sample of 1348 children and adolescents is shown in Table 1.
115 TABLE
saturated and 03 polyunsaturated fatty acids (Table 2), but correlated only weakly with y-linolenic acid (18 : 3~6). dihomo-y-linolenic acid (20 : 3~6) arachidonic acid (20 : 4~6) and docosahexaenoic acid (22 : 603). The correlations between HDL cholesterol and serum CE fatty acids were generally low. However, there was a positive correlation between HDL cholesterol and CE-18: 2 and a strong negative association with CE-16: 0. The ratio of HDL to total cholesterol had stronger correlations with major CE fatty acids than either one alone. The correlations between triglycerides and fatty acids were generally parallel to those between cholesterol and fatty acids. The concentrations of serum lipids in age- and sex-adjusted quintiles of CE-18: 2 are shown in Fig. 1. Serum total cholesterol, LDL cholesterol and triglycerides decreased linearly as the perHDL cholesterol centage of CE-18 : 2 increased. did not show any significant trends between the 18 : 2-quintiles, but the ratio of HDL to total cholesterol increased linearly with increasing percentage of CE-18 : 2.
1
PERCENTAGE ACIDS
IN
1348
COMPOSITION
OF
FINNISH
CHILDREN
SERUM
CE
AGED
FATTY FROM
3-IX-YEARS Mean
SD
14:o
0.97
0.27
16:0
10.43
0.66
Fatty acid
16: 1
4.20
1.08
17:l
0.36
0.11
0.90
0.17
18:0 18: 1
21.36
2.14
18:2w6
50.96
4.92 0.30
18:3w6
0.85
18:3w3
1.13
0.23
20: 3~6
0.73
0.15
20:4w6
5.64
0.85
20:5w3
1.20
0.37
22:6w3
0.66
0.17
Serum
lipoproteins
Both total and LDL cholesterol concentrations had negative correlation with CE-18: 2 and positive correlations with most saturated, monoun-
TABLE
2
LINEAR
CORRELATION
Fatty
COEFFICIENTS
BETWEEN
SERUM
CE FATTY
ACIDS
AND
SERUM
LIPIDS
Serum lipids
acid
Total
LDL a
HDL
HDL/Tot.
Tri-
chol.
chol.
chol.
chol.
g’yc
14:o
0.195
0.172
0.038
-0.116
16:0
0.066
0.111
- 0.147
-0.171
0.069
16:l
0.133
0.122
- 0.032
-0.136
0.201
17: 1
0.178
0.177
- 0.008
-0.147
0.113
18:O
0.104
0.137
-0.122
-0.181
0.109
18: 1
0.100
0.132
- 0.101
18:2w6
0.157
- 0.184
-0.168
0.098
18:3w6
0.060
0.033
~ o.OQ1
- 0.050
18:3w3 20: 3~6
0.165
0.164
_ 0.014
-0.013
- 0.140
0.119
0.016
-0.112
- 0.098
0.042
0.162
0.07 1
0.210
-0.120 0.195
20:4w6
0.033
0.067
- 0.056
- 0.074
- 0.077
20:5w3
0.187
0.184
0.059
- 0.097
- 0.014
22:6w3
0.042
0.075
Saturated
0.128
0.160
Monounsat.
0.120
0.141
Total
w6
Total
w3
-0.155
-0.177
0.213
Statistical significance * Calculated
of the correlation
with the equation
-0.112
coefficient: et al.
0.060
- 0.121
-0.197
0.121
- 0.086
-0.172
0.089
0.221
of Friedewald
-0.132
0.000 I = 0.071,
P < 0.01;
0.117
0.199
-0.126
-0.169 r = 0.089.
P i 0.001;
0.060 n = 1.348
116 TOTAL
CHOLESTEROL
LDL-CHOLESTEROL
HDL/TOTAL
HDL-CHOLESTEROL
APO .B
CHOLESTEROL
APO A-l/APO
TRIGLYCERIDES
Y&s----3 Quintiles
of CE-18:2
4
%
Fig. 2. Concentrations sex-adjusted quintiles i 95% confidence
A-l
LDL-CHOLESTEROL/APO
6
Ye----
high
Ouintiles
Fig. 1. Concentrations of serum lipids in age- and sex-adjusted quintiles of CE-18 : 2 in 1348 children (means& 95% confidence limits).
TABLE
8
HDL-CHOLESTEROL/APO
of CE-18:2
3
4
of serum apolipoproteins in age- and of CE-18: 2 in 1348 children (means
limits).
3
LINEAR
CORRELATION
COEFFICIENTS
BETWEEN
SERUM
CE FATTY
ACIDS
Serum apolipoproteins
Fatty acid
Apo A-I
Apo B
Apo A-I
AND SERUM
APOLIPOPROTEINS
HDL chol
LDL chol
Apo A-I
Apo B
Apo B 14:o 16:0 16 : 1 17:l 18:0 18:l 18:206 18:3w6 18:3w3 20:3w6 20~4~6 20: 5~3 22:6w3
0.145 0.073 0.122 0.212 0.128 0.170 -0.194 0.051 0.177 - 0.082 0.044 0.172 0.097
- 0.095 - 0.086 -0.125 - 0.212 - 0.088 -0.170 0.195 - 0.033 -0.166 0.021 -0.053 -0.159 - 0.109
0.017 - 0.089 0.032 0.110 -0.102 - 0.026 0.016 - 0.045 0.068 -0.133 - 0.058 0.140 - 0.074
- 0.072 -0.100 0.108 - 0.087
0.122 0.170 - 0.194 0.225
-0.110 -0.172 0.193 -0.215
- 0.080 - 0.007 0.000 0.103
-
Saturated Monounsat. Total w6 Total w3 Statistical
0.037 0.099 0.065 0.103 0.052 0.103 0.105 0.042 0.073 0.020 0.018 0.047 0.074
-
significance
of the correlation
coefficient:
r = 0.071, P < 0.01; r = 0.089,
high
%
P -G0.001;
-
-
0.052 0.066 0.012 0.040 0.024 0.043 0.002 0.035 0.006 0.159 0.043 0.037 0.024
0.070 - 0.030 0.008 0.015 n = 1,348
117 Serum apolipoproteirrs
Apo B had statistically significant correlations with most fatty acids in CE: negative with 18 : 2 and positive with saturated, monounsaturated and w3 polyunsaturated fatty acids (Table 3). The correlations between apo A-I and CE fatty acids were generally low, although there was a positive correlation between apo A-I and CE-18 : 2 and also between apo A-I and total 06 fatty acids in CE. The ratios of HDL cholesterol to apo A-I and LDL cholesterol to apo B had only low correlations with the major CE fatty acids. There was a linear decrease in concentration of apo B in quintiles of increasing CE-18 : 2 percentage (Fig. 2). The changes in apo A-I in these quintiles were less linear, although the concentration of apo A-I in the highest quintile of CE-18 : 2 was significantly higher than in the lowest quintile. The ratio of apo A-I to apo B increased strongly with the increasing quintile of CE-18: 2, but no differences were seen in the ratio of LDL cholesterol to apo B or in the ratio of HDL cholesterol to apo A-I between the quintiles of CE-18 : 2. Discussion Evaluation of the diet of an individual with the dietary interview methods used in most epidemiological studies gives relatively rough dietary intake data because of the large intra-individual variation in dietary habits. This is considered to be the major reason for the failure of most cross-sectional studies to show the relations between dietary intake data and serum lipoprotein concentrations within a free-living population [8,9]. The fatty acid composition of serum lipid fractions reflects the experimental changes in the quality of dietary fat [lo-12,24,25] and correlates well with the dietary data assessed with a dietary interview [23,26]. In the present sample of 1348 children the correlation coefficient between CE-18 : 2 and the P/S ratio of the diet estimated by the 48-h recall survey was 0.57 [18]. The observed correlation coefficients together with the trends in concentrations of serum total and LDL cholesterol in the quintiles of CE-18 : 2 show a definite negative relationship between the percentage of CE-18 : 2 and the concentration of
total or LDL cholesterol in serum. This confirms the similar [23], or parallel but statistically weaker [26] findings in our previous studies carried out with smaller number of subjects. It is also in good agreement with the literature on the influence of change in dietary P/S ratio on serum lipoproteins in dietary intervention studies [27.28]. The associations between serum apo B and CE-fatty acids closely paralleled those between LDL cholesterol and CE-fatty acids. There is no earlier data on correlations between serum fatty acids and apolipoproteins. The present results are in accordance with the reports on the decreasing effect on apo B concentration of an increase in dietary P/S ratio [2,29,30]. The exact biochemical mechanisms by which the LDL cholesterol and apo B are lowered are not yet known. but there is evidence that diet affects rather the synthesis or catabolism of LDL particles than their composition [30,31]. Our finding of an unchanged LDL cholesterol to apo B ratio in increasing quintiles of CE-18 : 2 is in agreement with these results. The relations of serum fatty acids to HDL cholesterol and apo A-I were weak, although there were significant positive correlations of CE-18 : 2 to both. The majority of data on the relationship between the quality of dietary fats and serum HDL has been obtained from intervention experiments. where an increase in dietary P/S ratio led to a reduction of both LDL and HDL cholesterol [32]. The P/S ratio of the diets high in polyunsaturated fat in most intervention studies has varied from 1 to 5, and been thus appreciably higher than the P/S ratios in the present sample: the mean dietary P/S ratios in the lowest and highest quintile of CE-18 : 2 were 0.16 and 0.36, respectively. It is possible that a modest increase in dietary P/S ratio can result in a decrease of total and LDL cholesterol without a simultaneous decrease in HDL. The observed associations of CE fatty acids and serum lipoproteins imply a corresponding relationship between the quality of dietary fats and serum lipoproteins within free-living Finnish children. Acknowledgements The skillful technical assistance of Miss P&vi Koivumaki is gratefully acknowledged.
118 References 1 Keys. A.. Anderson, J.T. and Grande. F., Serum cholesterol responses to changes in the diet. Metabolism. 14 (1965) 747. 2 Becker, N., Illingworth, D.R.. Alaupovic, P., Connor. W.E. and Sundberg, E.E., Effects of saturated, monounsaturated, and w-6-polyunsaturated fatty acids on plasma lipids, lipoproteins. and apolipoproteins in humans, Amer. J. Clin. Nutr., 37 (1983) 355. 3 HJermann, I.. Enger, SC.. Helgeland. A.. Holme. I., Leren, P. and Trygg, K., The effect of dietary changes on high density lipoprotein cholesterol, Amer. J. Med.. 66 (1979) 105. 4 Blaton, V., De Buyzere, M., Declercq, B., Pracetyo. A., Vanderkelen. G., Delanghe, J. and Spincematlle. J., Effect of polyunsaturated isocaloric fat diets on plasma lipids, apolipoproteins and fatty acids, Atherosclerosis. 53 (1984) 9. 5 Wolfram, G., Adam, 0. and Zollner, N.. Der Einfluss van Menge und Art des Nahrungsfettes auf die Lipide in den HDL des Serums beim Menschen, Verh. Dtsch. Ges. Inn. Med.. 86 (1980) 902. 6 Shepherd, J., Packard, C.J., Patsch, J.R., Gotto, Jr., A.M. and Taunton. O.D., Effects of dietary polyunsaturated and saturated fat on the properties of high density lipoproteins and the metabolism of apolipoprotein A-I, J. Clin. Invest., 61 (1978) 1582. 7 Ehnholm, C., Huttunen. J.K., Pietinen. P., Leino, U., Mutanen. M., Kostiainen, E., Iacano, J.M.. Dougherty. R. and Puska, P., Effect of a diet low in saturated fatty acids on plasma lipids, lipoproteins, and HDL subfractions, Arteriosclerosis, 4 (1984) 265. 8 Liu, K., Stamler, J., Dyer, A., McKeever, J. and McKeever, P., Statistical methods to assess and minimize the role of intra-individual variability in obscuring the relationship between dietary lipids and serum cholesterol, J. Chron. Dis., 31 (1978) 399. 9 Jacobs, Jr., D.R., Anderson, J.T. and Blackburn, H., Diet and serum cholesterol - Do zero correlations negate the relationship?, Amer. J. Epidemiol., 110 (1979) 77. 10 Arvidson, G. and Malmros, H., Studies on the influence of dietary fat on the fatty acid composition of serum lipids in humans. Z. Ernahr. Wiss., 11 (1972) 105. 11 Vessby, B., Lithell, H., Gustafsson, I.-B. and Boberg, J.. Changes in the fatty acid composition of the plasma lipid esters during lipid-lowering treatment with diet, clofibrate and nicenitrol, Atherosclerosis, 35 (1980) 51. 12 Nikkari. T. and Sale. M.. Serum fatty acids in Finnish population groups, In: R. Marcuse (Ed.). Proceedings of the 12th Scandinavian Symposium on Lipids, Lipidforum, Goteborg, Sweden, 1984, pp. 80-86. K.J., Brett, C.. Stovold, R., Chapman, A., 13 Kingsbury, Anderson, J. and Morgan, D.M., Abnormal fatty acid composition and human atherosclerosis, Postgrad. Med. J., 50 (1974)425. 14 Lang, P.D., Degott, M., Heuck. C.C., Opherk. D. and Vollmar, J.. Fatty acid composition of adipose tissue, blood lipids, and glucose tolerance in patients with different degrees of angiographically documented coronary arteriosclerosis, Res. Exp. Med. (Berl.), 180 (1982) 161.
15 Miettinen, T.A., Naukkarinen. V., Huttunen, J.K., Mattila, S. and Kumlin, T., Fatty-acid composition of serum lipids predicts myocardial infarction, Brit. Med. J.. 285 (1982) 993. 16 Keys, A. (Ed.), Coronary heart disease in seven countries, Circulation, 41 (Suppl. 1) (1970) 1. 17 Akerblom. H.K.. Viikari, J., Uhari, M., Rasanen, L., Byckling, T., Louhivuori, K., Pesonen, E.. Suoninen, P.. Pietikainen, M., Lahde, P.-L., Dahl, M., Aromaa. A.. Sarna. S. and Pyorala, K., Atherosclerosis precursors in Finnish children and adolescents, Part 1 (A general description of the cross-sectional study of 1980, and an account on the children’s and families’ state of health). Acta Paediat. Stand.. Suppl., 318 (1985) 49. 18 Moilanen, T.. Rlsanen, L., Viikari. J., Akerblom, H.K., Ahola, M.. Uhari, M.. Pasanen, M. and Nikkari, T.. Fatty acid composition of serum cholesteryl esters in 3 to l&yearold Finnish children and its relation to diet, Amer. J. Clin. Nutr.. 42 (1985) 708. 19 Moilanen, T. and Nikkari, T., The effect of storage on the fatty acid composition of human serum. Clin. Chim. Acta, 114 (1981) 111. 20 Finley, P.R., Schifman, R.B.. Williams, J. and Liehti, D.A., Cholesterol in high density lipoprotein - Use of Mg2+/ dextran sulfate in its enzymic measurement, Clin. Chem., 24 (1978) 931. W.T., Levy, R.I. and Fredrickson, D.S., Esti21 Friedewald. mation of the concentration of low-density-lipoprotein cholesterol in plasma without use of the preparative ultracentrifuge, Clin. Chem., 18 (1972) 499. T.. Nikkari, T., Viikari, J., Akerblom, 22 Solakivi-Jaakkola, H.K., Rasanen, L., Uhari, M., Dahl, M., Lahde. P.-L., Pesonen, E. and Suoninen, P., Atherosclerosis precursors in Finnish children and adolescents, Part 6 (Serum apolipoproteins A-I and B), Acta Paediat. Stand., Suppl., 318 (1985) 119. 23 Nikkari, T., Rasanen, L., Viikari, J.. Akerblom. H.K., Vuori, I., Pyorala, K., Uhari, M., Dahl, M., Lahde, P.-L., Pesonen, E. and Suoninen. P., Serum fatty acids in 8-year-old Finnish boys ~ Correlations with qualitative dietary data and other serum lipids, Amer. J. Clin. Nutr., 37 (1983) 848. H.J., De Boer. J. and De Iongh, H., Influence 24 Thomasson, of dietary fats on plasma lipids, Path. Microbial., 30 (1967) 629. I.-B.. Boberg, J.. Karlstrom, B., 25 Vessby, B., Gustafsson. Lithell, H. and Werner, 1.. Substituting polyunsaturated for saturated fats as a single change in a Swedish diet - Effects on serum lipoprotein metabolism and glucose tolerance in patients with hyperlipoproteinemia, Europ. J. Clin. Invest., 10 (1980) 193. L., Viikari, J.. Akerb26 Moilanen, T., Nikkari, T., Raslnen. lam, H.K., Ahola, M., Dahl, M., Lahde, P.-L., Pesonen. E.. Pietikainen, M.. Sepp&nen, A., Suoninen, P. and Uhari, M., Plasma cholesteryl ester fatty acids in 3- and 12-year-old Finnish children, Atherosclerosis, 48 (1983) 49. 21 Truswell, A.S., Diet and plasma lipids - A reappraisal, Amer. J. Clin. Nutr., 31 (1978) 977. Jr., S.H., Harris, W.S., Connor. W.E. and II28 Goodnight, lingworth, D.R., Polyunsaturated fatty acids, hyperlipidemia, and thrombosis, Arteriosclerosis, 2 (1982) 87.
119 29 Ehnholm. C.. Huttunen. J.K.. Pietinen. P., Leino. U.. Mutanen. M.. Kostiainen, E.. Pikkarainen. J., Dougherty, R.. Iacano. J. and Puska. P., Effect of diet on serum lipoprotetns in a population with a high risk of coronary heart disease, N. Engl. J. Med., 307 (1982) 850. 30 Vega, G.L.. Groszek. E., Wolf, R. and Grundy. SC., Influence of polyunsaturated fats on composition of plasma lipoprotetns and apoltpoproteins, J. Lipid Res., 23 (1982) 811.
31 Schonfeld. G., Patsch, W.. Rudel, L.L., Nelson. C.. Epstein. M. and Olson, R.E.. Effects of dietary cholesterol and fatty acids on plasma lipoproteins, J. Clin. Invest.. 69 (1982) 1072. 32 Vesshy. B.. Ltthell, H. and Boherg. J.. Reduction of Ioh, density and high density lipoprotein cholesterol by fatmodified diets ~ A survey of recent findings, Human Nutrition: Clinical Nutrition. 36C (19X2) 203
113 Ltd.
ATH 03721
Fatty Acid Composition of Serum Cholesteryl Esters in Relation to Serum Lipids and Apolipoproteins in 3-l&year-old Finnish Children and Adolescents T. Moilanen’,
T. Solakivi-Jaakkola’, J. Viikari 2, L. R&s;inen 3, H.K. Akerblom M. Uhari 5, M. Pasanen 6 and T. Nikkari ’
4,
’Deportment of Biomedical Scrences and ’Department of Publtc Health, Unioersity of Tumpere, Tampere. .’Department oJMedtcrne, University of Turku, Turku, -’Department of Nutntton and the 4 Children’s Hospital, Unioersity of Helsinkt, Helstnkt, and ’Department of Pediatrtcs. Uniuerstty of Oulu. Oulu (Finland) (Received 26 March, 1985) (Revised, received 26 August, 1985) (Accepted 29 August, 19851
Summary Serum cholesteryl ester (CE) fatty acids, serum lipids and apolipoproteins were analysed from 1348 Finnish children aged from 3 to 18 years. The study was part of a comprehensive survey of coronary heart disease risk factors and their determinants in Finnish children and adolescents. The percentage of CE linoleate (18 : 2) had significant negative correlations with the serum concentrations of total and LDL cholesterol, triglycerides as well as apo B. The associations of HDL to CE fatty acids were weak, although there was a significant positive correlation of CE-18 : 2 to HDL cholesterol and apo A-I. The ratio of HDL to total cholesterol and the ratio of apo A-I to apo B increased strongly in quintiles of increasing percentage of CE-18 : 2. In conclusion, the present results show significant associations between serum CE fatty acid composition and the different components of serum lipoproteins among free-living Finnish children and imply a relation between the quality of dietary fats and serum lipoproteins within this population.
Key words:
Apolipoproteins lipids
- Children - Cholesteryl
The study was supported by grants from the Operation Red Heart Fund of Lions International, Finland, the Research Council for Medicine. Academy of Finland and the Juho Vainio Foundation. Address for correspondence: Teemu Moilanen, M.D., Department of Biomedical Sciences, University of Tampere, P.O. Box 607, SF-33101 Tampere 10, Finland. 0165-0327/86/$03.50
‘n 1986 Elsevier Science Publishers
esters - Linoleic acid - Serum fatty acids - Serum
Introduction Saturated animal fat and cholesterol in the diet increase the total and LDL cholesterol as well as apo B concentrations in serum, while polyunsaturated fat has an opposite effect [1,2]. Reports
B.V. (Biomedical
Division)
114 on the influence of dietary fat on HDL cholesterol and apo A-I are more controversial. In some studies an increase in the ratio of polyunsaturated to saturated (P/S) fatty acids in the diet brought about an elevation of serum HDL [3,4], while no influence [5] or the opposite effect was obtained in others [6,7]. Although dietary interventions generally produce clear-cut effects on serum lipids, cross-sectional studies relating the diet to serum lipid concentrations within a population have usually shown only weak associations [8,9]. The fatty acid composition of serum lipids reflects the quality of dietary fat and is a useful estimate of the type of fat used by an individual [lo-121. Subjects with coronary heart disease (CHD) have reduced proportions of linoleic acid in serum and adipose tissue [13,14] and it has been suggested that the composition of serum fatty acids is also an independent risk factor of CHD [15]. Since the rate of CHD in Finland is high [16], the levels of various CHD risk factors in Finnish children and adolescents were surveyed in a comprehensive multicentre study in 1980 [17]. The study included among others a dietary interview and determinations of serum total and HDL cholesterol, triglycerides, apolipoproteins A-I and B, as well as an analysis of fatty acids in serum cholesteryl esters (CE). The results on correlations between serum fatty acids and dietary data have been published elsewhere [18]. This report gives the relationships of serum fatty acid compositions to serum lipids and apolipoproteins in 1348 Finnish children that were a part of the multicentre study. Subjects and Methods The study was carried out in October-November 1980 in 5 university cities and their surrounding rural communities in Finland. An invitation with a description of the study was sent to the parents, whereafter 83% of the invited children participated. The study included a total of 3,596 boys and girls aged 3, 6, 9, 12, 15 and 18 years. A detailed description of the population sample is given in ref. [17]. The fatty acid composition of serum CEs was analysed in 1348 children representing all age groups and areas and both sexes. Venous blood was drawn after an overnight
fast, and serum was separated after 2 h clotting at room temperature. Serum aliquots were stored at -6O’C for l-14 months until analysed. CE fatty acid composition remains stable at this temperature for at least one year [19]. Serum cholesterol concentrations were measured using an enzymatic Boehringer CHOD-PAP kit, triglycerides with a Boehringer fully enzymatic system and an automatic analyzer, and HDL cholesterol after precipitation of very low and low density lipoproteins with Mg2+ and dextran sulphate 500 [20]. LDL cholesterol was calculated from the equation of Friedewald et al. [21]. Apolipoproteins A-I and B were determined with radial immunodiffusion [22]. The inter-assay coefficients of variation for apo A-I and apo B were 5.1 and 4.0%, respectively. The procedures followed in fatty acid analyses have been described elsewhere [19]. In short, lipids were extracted from 0.5-ml aliquots of serum with chloroform/methanol (1 : 1, v/v). The extract was fractionated with thin-layer chromatography, and the fraction corresponding to CE was saponified with KOH/ethanol. Methyl esters of fatty acids were analysed in a Varian 2100 gas chromatograph using two 0.3 cm (i.d.) x 180 cm glass columns packed with 3% EGSS-X on loo-120 mesh Gas Chrom Q (Applied Science Lab.). The temperature was programmed from 140 to 190°C and all peaks corresponding to fatty acids from 14: 0 to 22 : 6 were quantified with a flame ionisation detector and Hewlett-Packard model 5880 A integrator. The reproducibility of the fatty acid analysis has been reported earlier [23]. The coefficient of variation “was 0.4% for linoleic acid (18 : 206) and ranged from 0.6 to 4.1% for most fatty acids. All statistical calculations were done at the Computer Centre of the University of Tampere. Linear correlation coefficients were calculated from the total sample of 1348 children. Age- and sexadjusted quintiles of CE-18: 2 percentage were formed by dividing each age and sex class into corresponding quintiles and combining these classifications. Results Fatty acid composition of serum CEs in the total sample of 1348 children and adolescents is shown in Table 1.
115 TABLE
saturated and 03 polyunsaturated fatty acids (Table 2), but correlated only weakly with y-linolenic acid (18 : 3~6). dihomo-y-linolenic acid (20 : 3~6) arachidonic acid (20 : 4~6) and docosahexaenoic acid (22 : 603). The correlations between HDL cholesterol and serum CE fatty acids were generally low. However, there was a positive correlation between HDL cholesterol and CE-18: 2 and a strong negative association with CE-16: 0. The ratio of HDL to total cholesterol had stronger correlations with major CE fatty acids than either one alone. The correlations between triglycerides and fatty acids were generally parallel to those between cholesterol and fatty acids. The concentrations of serum lipids in age- and sex-adjusted quintiles of CE-18: 2 are shown in Fig. 1. Serum total cholesterol, LDL cholesterol and triglycerides decreased linearly as the perHDL cholesterol centage of CE-18 : 2 increased. did not show any significant trends between the 18 : 2-quintiles, but the ratio of HDL to total cholesterol increased linearly with increasing percentage of CE-18 : 2.
1
PERCENTAGE ACIDS
IN
1348
COMPOSITION
OF
FINNISH
CHILDREN
SERUM
CE
AGED
FATTY FROM
3-IX-YEARS Mean
SD
14:o
0.97
0.27
16:0
10.43
0.66
Fatty acid
16: 1
4.20
1.08
17:l
0.36
0.11
0.90
0.17
18:0 18: 1
21.36
2.14
18:2w6
50.96
4.92 0.30
18:3w6
0.85
18:3w3
1.13
0.23
20: 3~6
0.73
0.15
20:4w6
5.64
0.85
20:5w3
1.20
0.37
22:6w3
0.66
0.17
Serum
lipoproteins
Both total and LDL cholesterol concentrations had negative correlation with CE-18: 2 and positive correlations with most saturated, monoun-
TABLE
2
LINEAR
CORRELATION
Fatty
COEFFICIENTS
BETWEEN
SERUM
CE FATTY
ACIDS
AND
SERUM
LIPIDS
Serum lipids
acid
Total
LDL a
HDL
HDL/Tot.
Tri-
chol.
chol.
chol.
chol.
g’yc
14:o
0.195
0.172
0.038
-0.116
16:0
0.066
0.111
- 0.147
-0.171
0.069
16:l
0.133
0.122
- 0.032
-0.136
0.201
17: 1
0.178
0.177
- 0.008
-0.147
0.113
18:O
0.104
0.137
-0.122
-0.181
0.109
18: 1
0.100
0.132
- 0.101
18:2w6
0.157
- 0.184
-0.168
0.098
18:3w6
0.060
0.033
~ o.OQ1
- 0.050
18:3w3 20: 3~6
0.165
0.164
_ 0.014
-0.013
- 0.140
0.119
0.016
-0.112
- 0.098
0.042
0.162
0.07 1
0.210
-0.120 0.195
20:4w6
0.033
0.067
- 0.056
- 0.074
- 0.077
20:5w3
0.187
0.184
0.059
- 0.097
- 0.014
22:6w3
0.042
0.075
Saturated
0.128
0.160
Monounsat.
0.120
0.141
Total
w6
Total
w3
-0.155
-0.177
0.213
Statistical significance * Calculated
of the correlation
with the equation
-0.112
coefficient: et al.
0.060
- 0.121
-0.197
0.121
- 0.086
-0.172
0.089
0.221
of Friedewald
-0.132
0.000 I = 0.071,
P < 0.01;
0.117
0.199
-0.126
-0.169 r = 0.089.
P i 0.001;
0.060 n = 1.348
116 TOTAL
CHOLESTEROL
LDL-CHOLESTEROL
HDL/TOTAL
HDL-CHOLESTEROL
APO .B
CHOLESTEROL
APO A-l/APO
TRIGLYCERIDES
Y&s----3 Quintiles
of CE-18:2
4
%
Fig. 2. Concentrations sex-adjusted quintiles i 95% confidence
A-l
LDL-CHOLESTEROL/APO
6
Ye----
high
Ouintiles
Fig. 1. Concentrations of serum lipids in age- and sex-adjusted quintiles of CE-18 : 2 in 1348 children (means& 95% confidence limits).
TABLE
8
HDL-CHOLESTEROL/APO
of CE-18:2
3
4
of serum apolipoproteins in age- and of CE-18: 2 in 1348 children (means
limits).
3
LINEAR
CORRELATION
COEFFICIENTS
BETWEEN
SERUM
CE FATTY
ACIDS
Serum apolipoproteins
Fatty acid
Apo A-I
Apo B
Apo A-I
AND SERUM
APOLIPOPROTEINS
HDL chol
LDL chol
Apo A-I
Apo B
Apo B 14:o 16:0 16 : 1 17:l 18:0 18:l 18:206 18:3w6 18:3w3 20:3w6 20~4~6 20: 5~3 22:6w3
0.145 0.073 0.122 0.212 0.128 0.170 -0.194 0.051 0.177 - 0.082 0.044 0.172 0.097
- 0.095 - 0.086 -0.125 - 0.212 - 0.088 -0.170 0.195 - 0.033 -0.166 0.021 -0.053 -0.159 - 0.109
0.017 - 0.089 0.032 0.110 -0.102 - 0.026 0.016 - 0.045 0.068 -0.133 - 0.058 0.140 - 0.074
- 0.072 -0.100 0.108 - 0.087
0.122 0.170 - 0.194 0.225
-0.110 -0.172 0.193 -0.215
- 0.080 - 0.007 0.000 0.103
-
Saturated Monounsat. Total w6 Total w3 Statistical
0.037 0.099 0.065 0.103 0.052 0.103 0.105 0.042 0.073 0.020 0.018 0.047 0.074
-
significance
of the correlation
coefficient:
r = 0.071, P < 0.01; r = 0.089,
high
%
P -G0.001;
-
-
0.052 0.066 0.012 0.040 0.024 0.043 0.002 0.035 0.006 0.159 0.043 0.037 0.024
0.070 - 0.030 0.008 0.015 n = 1,348
117 Serum apolipoproteirrs
Apo B had statistically significant correlations with most fatty acids in CE: negative with 18 : 2 and positive with saturated, monounsaturated and w3 polyunsaturated fatty acids (Table 3). The correlations between apo A-I and CE fatty acids were generally low, although there was a positive correlation between apo A-I and CE-18 : 2 and also between apo A-I and total 06 fatty acids in CE. The ratios of HDL cholesterol to apo A-I and LDL cholesterol to apo B had only low correlations with the major CE fatty acids. There was a linear decrease in concentration of apo B in quintiles of increasing CE-18 : 2 percentage (Fig. 2). The changes in apo A-I in these quintiles were less linear, although the concentration of apo A-I in the highest quintile of CE-18 : 2 was significantly higher than in the lowest quintile. The ratio of apo A-I to apo B increased strongly with the increasing quintile of CE-18: 2, but no differences were seen in the ratio of LDL cholesterol to apo B or in the ratio of HDL cholesterol to apo A-I between the quintiles of CE-18 : 2. Discussion Evaluation of the diet of an individual with the dietary interview methods used in most epidemiological studies gives relatively rough dietary intake data because of the large intra-individual variation in dietary habits. This is considered to be the major reason for the failure of most cross-sectional studies to show the relations between dietary intake data and serum lipoprotein concentrations within a free-living population [8,9]. The fatty acid composition of serum lipid fractions reflects the experimental changes in the quality of dietary fat [lo-12,24,25] and correlates well with the dietary data assessed with a dietary interview [23,26]. In the present sample of 1348 children the correlation coefficient between CE-18 : 2 and the P/S ratio of the diet estimated by the 48-h recall survey was 0.57 [18]. The observed correlation coefficients together with the trends in concentrations of serum total and LDL cholesterol in the quintiles of CE-18 : 2 show a definite negative relationship between the percentage of CE-18 : 2 and the concentration of
total or LDL cholesterol in serum. This confirms the similar [23], or parallel but statistically weaker [26] findings in our previous studies carried out with smaller number of subjects. It is also in good agreement with the literature on the influence of change in dietary P/S ratio on serum lipoproteins in dietary intervention studies [27.28]. The associations between serum apo B and CE-fatty acids closely paralleled those between LDL cholesterol and CE-fatty acids. There is no earlier data on correlations between serum fatty acids and apolipoproteins. The present results are in accordance with the reports on the decreasing effect on apo B concentration of an increase in dietary P/S ratio [2,29,30]. The exact biochemical mechanisms by which the LDL cholesterol and apo B are lowered are not yet known. but there is evidence that diet affects rather the synthesis or catabolism of LDL particles than their composition [30,31]. Our finding of an unchanged LDL cholesterol to apo B ratio in increasing quintiles of CE-18 : 2 is in agreement with these results. The relations of serum fatty acids to HDL cholesterol and apo A-I were weak, although there were significant positive correlations of CE-18 : 2 to both. The majority of data on the relationship between the quality of dietary fats and serum HDL has been obtained from intervention experiments. where an increase in dietary P/S ratio led to a reduction of both LDL and HDL cholesterol [32]. The P/S ratio of the diets high in polyunsaturated fat in most intervention studies has varied from 1 to 5, and been thus appreciably higher than the P/S ratios in the present sample: the mean dietary P/S ratios in the lowest and highest quintile of CE-18 : 2 were 0.16 and 0.36, respectively. It is possible that a modest increase in dietary P/S ratio can result in a decrease of total and LDL cholesterol without a simultaneous decrease in HDL. The observed associations of CE fatty acids and serum lipoproteins imply a corresponding relationship between the quality of dietary fats and serum lipoproteins within free-living Finnish children. Acknowledgements The skillful technical assistance of Miss P&vi Koivumaki is gratefully acknowledged.
118 References 1 Keys. A.. Anderson, J.T. and Grande. F., Serum cholesterol responses to changes in the diet. Metabolism. 14 (1965) 747. 2 Becker, N., Illingworth, D.R.. Alaupovic, P., Connor. W.E. and Sundberg, E.E., Effects of saturated, monounsaturated, and w-6-polyunsaturated fatty acids on plasma lipids, lipoproteins. and apolipoproteins in humans, Amer. J. Clin. Nutr., 37 (1983) 355. 3 HJermann, I.. Enger, SC.. Helgeland. A.. Holme. I., Leren, P. and Trygg, K., The effect of dietary changes on high density lipoprotein cholesterol, Amer. J. Med.. 66 (1979) 105. 4 Blaton, V., De Buyzere, M., Declercq, B., Pracetyo. A., Vanderkelen. G., Delanghe, J. and Spincematlle. J., Effect of polyunsaturated isocaloric fat diets on plasma lipids, apolipoproteins and fatty acids, Atherosclerosis. 53 (1984) 9. 5 Wolfram, G., Adam, 0. and Zollner, N.. Der Einfluss van Menge und Art des Nahrungsfettes auf die Lipide in den HDL des Serums beim Menschen, Verh. Dtsch. Ges. Inn. Med.. 86 (1980) 902. 6 Shepherd, J., Packard, C.J., Patsch, J.R., Gotto, Jr., A.M. and Taunton. O.D., Effects of dietary polyunsaturated and saturated fat on the properties of high density lipoproteins and the metabolism of apolipoprotein A-I, J. Clin. Invest., 61 (1978) 1582. 7 Ehnholm, C., Huttunen. J.K., Pietinen. P., Leino, U., Mutanen. M., Kostiainen, E., Iacano, J.M.. Dougherty. R. and Puska, P., Effect of a diet low in saturated fatty acids on plasma lipids, lipoproteins, and HDL subfractions, Arteriosclerosis, 4 (1984) 265. 8 Liu, K., Stamler, J., Dyer, A., McKeever, J. and McKeever, P., Statistical methods to assess and minimize the role of intra-individual variability in obscuring the relationship between dietary lipids and serum cholesterol, J. Chron. Dis., 31 (1978) 399. 9 Jacobs, Jr., D.R., Anderson, J.T. and Blackburn, H., Diet and serum cholesterol - Do zero correlations negate the relationship?, Amer. J. Epidemiol., 110 (1979) 77. 10 Arvidson, G. and Malmros, H., Studies on the influence of dietary fat on the fatty acid composition of serum lipids in humans. Z. Ernahr. Wiss., 11 (1972) 105. 11 Vessby, B., Lithell, H., Gustafsson, I.-B. and Boberg, J.. Changes in the fatty acid composition of the plasma lipid esters during lipid-lowering treatment with diet, clofibrate and nicenitrol, Atherosclerosis, 35 (1980) 51. 12 Nikkari. T. and Sale. M.. Serum fatty acids in Finnish population groups, In: R. Marcuse (Ed.). Proceedings of the 12th Scandinavian Symposium on Lipids, Lipidforum, Goteborg, Sweden, 1984, pp. 80-86. K.J., Brett, C.. Stovold, R., Chapman, A., 13 Kingsbury, Anderson, J. and Morgan, D.M., Abnormal fatty acid composition and human atherosclerosis, Postgrad. Med. J., 50 (1974)425. 14 Lang, P.D., Degott, M., Heuck. C.C., Opherk. D. and Vollmar, J.. Fatty acid composition of adipose tissue, blood lipids, and glucose tolerance in patients with different degrees of angiographically documented coronary arteriosclerosis, Res. Exp. Med. (Berl.), 180 (1982) 161.
15 Miettinen, T.A., Naukkarinen. V., Huttunen, J.K., Mattila, S. and Kumlin, T., Fatty-acid composition of serum lipids predicts myocardial infarction, Brit. Med. J.. 285 (1982) 993. 16 Keys, A. (Ed.), Coronary heart disease in seven countries, Circulation, 41 (Suppl. 1) (1970) 1. 17 Akerblom. H.K.. Viikari, J., Uhari, M., Rasanen, L., Byckling, T., Louhivuori, K., Pesonen, E.. Suoninen, P.. Pietikainen, M., Lahde, P.-L., Dahl, M., Aromaa. A.. Sarna. S. and Pyorala, K., Atherosclerosis precursors in Finnish children and adolescents, Part 1 (A general description of the cross-sectional study of 1980, and an account on the children’s and families’ state of health). Acta Paediat. Stand.. Suppl., 318 (1985) 49. 18 Moilanen, T.. Rlsanen, L., Viikari. J., Akerblom, H.K., Ahola, M.. Uhari, M.. Pasanen, M. and Nikkari, T.. Fatty acid composition of serum cholesteryl esters in 3 to l&yearold Finnish children and its relation to diet, Amer. J. Clin. Nutr.. 42 (1985) 708. 19 Moilanen, T. and Nikkari, T., The effect of storage on the fatty acid composition of human serum. Clin. Chim. Acta, 114 (1981) 111. 20 Finley, P.R., Schifman, R.B.. Williams, J. and Liehti, D.A., Cholesterol in high density lipoprotein - Use of Mg2+/ dextran sulfate in its enzymic measurement, Clin. Chem., 24 (1978) 931. W.T., Levy, R.I. and Fredrickson, D.S., Esti21 Friedewald. mation of the concentration of low-density-lipoprotein cholesterol in plasma without use of the preparative ultracentrifuge, Clin. Chem., 18 (1972) 499. T.. Nikkari, T., Viikari, J., Akerblom, 22 Solakivi-Jaakkola, H.K., Rasanen, L., Uhari, M., Dahl, M., Lahde. P.-L., Pesonen, E. and Suoninen, P., Atherosclerosis precursors in Finnish children and adolescents, Part 6 (Serum apolipoproteins A-I and B), Acta Paediat. Stand., Suppl., 318 (1985) 119. 23 Nikkari, T., Rasanen, L., Viikari, J.. Akerblom. H.K., Vuori, I., Pyorala, K., Uhari, M., Dahl, M., Lahde, P.-L., Pesonen, E. and Suoninen. P., Serum fatty acids in 8-year-old Finnish boys ~ Correlations with qualitative dietary data and other serum lipids, Amer. J. Clin. Nutr., 37 (1983) 848. H.J., De Boer. J. and De Iongh, H., Influence 24 Thomasson, of dietary fats on plasma lipids, Path. Microbial., 30 (1967) 629. I.-B.. Boberg, J.. Karlstrom, B., 25 Vessby, B., Gustafsson. Lithell, H. and Werner, 1.. Substituting polyunsaturated for saturated fats as a single change in a Swedish diet - Effects on serum lipoprotein metabolism and glucose tolerance in patients with hyperlipoproteinemia, Europ. J. Clin. Invest., 10 (1980) 193. L., Viikari, J.. Akerb26 Moilanen, T., Nikkari, T., Raslnen. lam, H.K., Ahola, M., Dahl, M., Lahde, P.-L., Pesonen. E.. Pietikainen, M.. Sepp&nen, A., Suoninen, P. and Uhari, M., Plasma cholesteryl ester fatty acids in 3- and 12-year-old Finnish children, Atherosclerosis, 48 (1983) 49. 21 Truswell, A.S., Diet and plasma lipids - A reappraisal, Amer. J. Clin. Nutr., 31 (1978) 977. Jr., S.H., Harris, W.S., Connor. W.E. and II28 Goodnight, lingworth, D.R., Polyunsaturated fatty acids, hyperlipidemia, and thrombosis, Arteriosclerosis, 2 (1982) 87.
119 29 Ehnholm. C.. Huttunen. J.K.. Pietinen. P., Leino. U.. Mutanen. M.. Kostiainen, E.. Pikkarainen. J., Dougherty, R.. Iacano. J. and Puska. P., Effect of diet on serum lipoprotetns in a population with a high risk of coronary heart disease, N. Engl. J. Med., 307 (1982) 850. 30 Vega, G.L.. Groszek. E., Wolf, R. and Grundy. SC., Influence of polyunsaturated fats on composition of plasma lipoprotetns and apoltpoproteins, J. Lipid Res., 23 (1982) 811.
31 Schonfeld. G., Patsch, W.. Rudel, L.L., Nelson. C.. Epstein. M. and Olson, R.E.. Effects of dietary cholesterol and fatty acids on plasma lipoproteins, J. Clin. Invest.. 69 (1982) 1072. 32 Vesshy. B.. Ltthell, H. and Boherg. J.. Reduction of Ioh, density and high density lipoprotein cholesterol by fatmodified diets ~ A survey of recent findings, Human Nutrition: Clinical Nutrition. 36C (19X2) 203