Introduction

The prevalence of child obesity is increasing rapidly in both developed and developing countries, which to a great extent is linked to inappropriate diet and increased sedentary behavior (1). Because these factors have become difficult to influence (2, 3), it has been suggested that metabolic disturbances might have causal importance for the fast growing problem (4).

Dietary fat, such as the polyunsaturated essential fatty acids (PUFAs),¹ influences the fatty acid (FA) composition of tissue lipids (5). In the general recommendations to substitute saturated for unsaturated fat, there has been a shift in the balance between n-6 and n-3 PUFA in the human diet over the past 30 years, and the n-6-to-n-3 ratio of the Western diet is nowadays often over 20, far higher than the recommendation of a ratio below 5 (6, 7, 8).

The essential FAs are needed for normal growth, reproduction, and cell function (9, 10). They also regulate gene expression (11). The perinatal period has been shown to be sensitive to nutritional stimuli such as the n-6 and n-3 PUFA, which affect the structure of cell membranes, as well as future function and metabolism (12, 13). Recently, we have found that overweight and symptoms related to the metabolic syndrome can be programmed in the adult animal by modulating essential FA in the perinatal period (14, 15). These results from experimental studies support the hypothesis presented by Ailhaud and Guesnet (4), who provide evidence that obesity per se is associated with a disturbed n-6/n-3 PUFA balance in the maternal diet during pregnancy, lactation, and early childhood. Dietary FA composition might have importance for the increasing prevalence of childhood overweight and obesity (16). There are few data about the PUFA status in children and adolescents, especially its correlation to metabolic variables and to the distribution of adipose tissue (AT) (17, 18).

Serum phospholipids (PLs) mirror the dietary intake for recent months and, thereby, also roughly mirror membrane composition in the body. Therefore, PLs can be important substitutes for the more difficult measurements used to analyze tissue membranes when evaluating relationships to metabolism (19). More than 95% of the serum PLs are phosphatidylcholine, one of the major PLs in membranes; therefore, in contrast to other tissues, it is not considered necessary to separate different PLs in serum. Free FAs constitute a minor pool that is very rapidly changed during a day by, for instance, actual metabolism and physical activity. If analyzing total lipids, triglycerides and cholesterol esters also are included. The triglycerides mirror the dietary intake of the last day and can be subject to great variations (20), but cholesterol esters are more reflective of long-term intake (21). PLs seem to be best repeatable despite lower concentration (22). Due to these facts, it has been common to use serum PLs as markers of the FA status of an individual. Although it is important to remember that all tissues can have different patterns of FA, serum (or plasma, which in this context is equivalent) is usually accepted as a general marker of FA status.

Computerized tomography and magnetic resonance imaging (MRI) are accurate imaging techniques for assessing body fat distribution (23), but their availability and costs have limited their use outside research settings. Thus, based on a relationship between visceral AT (VAT) and anthropometric measures, other indirect indices of body fat distribution have been used (24). In adults, the waist-to-hip ratio and waist circumference are often used as markers of VAT (25, 26). In children and adolescents, the correlation between waist measurements and VAT, as measured by imaging techniques, has not been convincing. The role of subcutaneous abdominal AT (SAAT) as a contributor to metabolic dysfunction might be underestimated (27, 28). The MRI technique makes it possible to analyze whether there are differences in the deep and superficial SAAT in childhood in relation to metabolic markers, in accordance with what has been presented as a novel approach in lean and obese adults (29, 30).

The aim of this work was, first, to analyze the relation between AT compartments and metabolic measures to the concentrations of PUFA in serum PLs and, second, to investigate whether the abdominal AT distribution visualized by MRI in children correlated to BMI and to other commonly used anthropometric measurements, such as waist circumference and the waist-to-hip ratio, and to the metabolic markers.

Research Methods and Procedures

Definitions

Obesity was defined according to the standard definition of the International Obesity Task Force (31). Puberty maturation was assessed by an experienced pediatrician (S.M.) according to Tanner (32).

Subjects

Ten adolescents with obesity (five females) were recruited from the Outpatient Clinic at the Queen Silvia Children's Hospital (Göteborg, Sweden). Clinical data of the patients are given in Table 1. The females had a mean puberty state of 4 (range, 2 to 5) and a breast development of 4 (range, 2 to 5); corresponding figures in the males were 2.2 (1 to 5) and for genital development, 2.2 (1 to 5).

Table 1. - Clinical characteristics of the study groups [ median (range)] .

Full table

Fifteen healthy adolescents (nine females) with normal weight were randomly recruited from schools in the Göteborg area. The females had a mean puberty state of 3.4 (2 to 5) and breast development of 3.5 (3 to 5); the mean puberty status of the males was 3 (1 to 5) and for genital development, 2.5 (1 to 4).

The study was conducted according to the Declaration of Helsinki and approved by the Ethics Committee of Göteborg University, and informed consent was obtained from all patients and parents.

Methods

Height, weight, and circumferences of the waist and hip were measured. Waist-to-hip ratio and BMI (kilograms per meter squared) were calculated. BMI-standard deviation scores (BMI-sds) were calculated adjusted for age and gender according to Karlberg et al. (33).

MRI was performed with a 1.5-T scanner (Magnetom Vision Plus; Siemens, Erlangen, Germany) with a 25 mT/m maximum gradient and a phased array body coil. Subjects were studied in supine position, and a cross-sectional image was taken at an abdominal level in line with the umbilicus, superior to the iliac crest (Figure 1). SAAT was defined as the compartment outside and VAT the compartment inside the boundaries of the abdominal wall musculature in continuity with the deep fascia of the paraspinal musculature (23). Using a mouse-driven cursor, the areas of superficial and deep SAAT were defined by tracing the subcutaneous fascia as previously described (29, 30). All images were measured in centimeters squared and independently read by two operators. The mean values of the two measurements were calculated.

Figure 1.

Cross-sectional MRI at the umbilical level. (Arrow) Fascia that divides the subcutaneous AT depot into a superficial and a deep compartment.

Full figure and legend (50K)

Blood samples after a 12-hour fast were analyzed for plasma glucose, serum concentrations of free insulin, triacylglycerol and high-density lipoprotein (HDL)-cholesterol, and serum activities of aspartate aminotransferase and alanine aminotransferase according to routine. Serum low-density lipoprotein (LDL)-cholesterol was calculated using the Friedewald equation (34). To assess the pancreatic -cell function, the homeostasis model analysis (HOMA) was calculated (20 insulin/(glucose - 3.5)), and as an approximation of insulin resistance (IR), the HOMA-IR formula was used (insulin/(ln glucose 22.5)) (35).

Sera were kept frozen (- 70 °C) until analysis of FA composition of PLs. Lipids were extracted and fractionated as previously reported (36). The FA methyl derivatives were separated by capillary gas-liquid chromatography in a Hewlett-Packard 6890 gas chromatograph using helium as a carrier gas. The separation was recorded with HP GC Chem Station software (HP GC, Wilmington, DE). FA 21:1 was used as internal standard, and the FA fractions were identified by comparison with retention times of pure reference substances (Sigma Aldrich Sweden AB, Stockholm, Sweden). The interassay coefficient of variation was 0.6% for linoleic acid (18:2 n-6) and 1.5% for arachidonic acid (AA; 20:4 n-6) (n = 15).

Statistical Analysis

All data are presented as median (range), except the FA data, which are presented as mean (SD). Overall differences between the study groups were analyzed using the Kruskal-Wallis ANOVA and Mann Whitney's U test. Pearson correlation coefficients were calculated for assessment of correlations between FA and AT compartments and biochemical and clinical variables. A value of p < 0.05 was considered statistically significant.

Results

MRI and Anthropometry

All AT areas, except VAT area, were markedly increased in all obese children compared with the non-obese control subjects (Table 1). The VAT area was significantly increased only in males, and the relative amount of VAT to total AT area of the abdomen was reduced both in obese females and males compared with non-obese controls. The waist and hip circumferences were larger in the obese than in the non-obese children, but the waist-to-hip ratio was only increased in obese females (Table 1). AT areas showed strong correlations to BMI, waist and hip, but not to the waist-to-hip ratio (data not shown).

Biochemical Variables

Median serum insulin was highly elevated in obese females and males compared with the controls (Table 2). The HOMA -cell function index was significantly elevated, but HDL-cholesterol was significantly lower only in the obese females. Plasma concentrations of glucose, serum concentrations of triacylglycerol, total and LDL-cholesterol, and HOMA-IR index were not different from controls. Liver enzyme activities were normal. Serum insulin and HOMA -cell function were significantly associated with all AT compartments. Both waist and hip circumferences correlated positively to all fat depots but also strongly to BMI, serum insulin, and the HOMA -cell function. The waist-to-hip ratio showed no correlation to the abdominal AT, total or individual depots, or to biochemical variables.

Table 2. - Biochemical data of the four study groups [ median (range)] .

Full table

FA Analysis

There was no sex difference in FA pattern within the groups of obese and non-obese children. Comparing obese children and controls revealed significant differences in females.

Obese Females vs. Non-obese Females

The mean total serum concentration of n-3 FA was significantly lower in the obese females vs. controls (p = 0.006) (Table 3) as was the concentration of docosahexaenoic acid (DHA; 22:6n-3) (p = 0.01). The AA concentration did not differ, but the AA-to-DHA ratio and the overall n-6-to-n-3 ratio were increased. Nervonic acid, 24:1n-9, was also significantly increased as was the 24:1n-9-to-24:0 ratio (p = 0.02), suggesting an increased 9 desaturase activity. Higher concentrations of the saturated FA (SFA) 20:0 and 22:0 and of total SFA were found in the obese compared with controls. The total concentration of PUFA was decreased, resulting in lower unsaturation index in the obese females.

Table 3. - Serum phospholipid FA in mole percentage [ mean (standard deviation)] .

Full table

Obese Males vs. Non-obese Males

The obese males showed a similar trend as the females, but the differences from the age-matched controls seldom reached significance (Table 3). The total concentration of SFA was significantly higher in the obese, as was the FA 20:0 and the 24:1n-9-to-24:0 ratio (p = 0.02).

PUFA in Relation to Markers of the Metabolic Syndrome

The total amount of n-3 PUFA was inversely correlated to both BMI-sds (r = - 0.48, p = 0.02) and both deep and superficial SAAT, but not to VAT (Table 4). An inverse correlation was also found for the HOMA index of -cell function (r = - 0.44, p = 0.03). DHA was inversely correlated to all fat depots (r = - 0.45, p = 0.03) except VAT and to serum insulin and the HOMA -cell function (r = - 0.50, p = 0.04) (Figure 2). Both individual and total amounts of SFA were strongly associated to all AT depots (r = 0.60, p = 0.002), including VAT. Nervonic acid was similarly strongly correlated to all SAAT compartments except VAT. Both waist and hip circumferences correlated positively to total SFA and nervonic acid. The sum of PUFA showed a similar but inverse correlation, not associated to VAT.

Figure 2.

Relationship between the serum PL molar percentage concentration of DHA (22:6 n-3) and the HOMA -cell function index (percentage). Boys () and girls (O) are indicated with open (controls) and filled (obese) symbols.

Full figure and legend (57K)

Table 4. - Correlation coefficients between fat depots and selected serum FAs and biochemical variables.

Full table

Discussion

This study showed that obese adolescents had a different pattern of FA in serum PL compared with lean controls. The obese patients had significantly higher concentrations of SFA, especially nervonic acid, and lower levels of PUFA. The latter was due mainly to low concentrations of DHA, resulting in increased AA-to-DHA ratio, both DHA level and the ratio being inversely associated with the HOMA -cell function index. Using MRI to define different AT compartments revealed differences from adults; none of the metabolic markers was related to VAT, but to all different subcompartments of SAAT. The results point to the importance of the SAAT in children and suggest that their lower DHA may relate to the metabolic dysfunction.

Only a few studies have reported the FA composition of lipids in serum and tissues in obese children. Higher levels of n-6 FA have been reported in different serum lipid fractions (17, 18), but the relation to the n-3 FA levels has not been discussed. In a recent study of 10 obese children in the Mediterranean area, AA was more often increased in AT of the obese children (37). A reason for the predominance of high n-6 in other studies might be that Sweden generally has a lower intake of n-6 FAs, giving an overall n-6-to-n-3 FA ratio of 5 compared with the rest of Europe, where the ratio is 10 (7). The liability in females corroborates previous studies, which have also found more marked indications of biochemical changes related to the metabolic syndrome in obese females (38). Our study was not designed to investigate food intake; therefore, it cannot be excluded that all changes might be diet induced. On the other hand, the relation to the biochemical markers of glucose homeostasis suggested an association between the FA pattern and the risk for diabetes, whatever the cause of the observed lipid abnormality may be. Furthermore, low DHA concentration was also noticed in some of the controls without influence on the insulin production (Figure 2), suggesting that a metabolic abnormality might explain the variation rather than diet. Such abnormality might be genetically determined or related to the change of fat quality in Western diet during the latest decades (6, 7, 8), through fetal programming, i.e., influence of the essential FAs (EFAs) in the maternal diet during pregnancy and lactation, resulting in a disturbed lipid pattern (14, 15, 16). The FA content of breast milk reflects the diet of the mother (39). EFAs are known to influence gene expression of lipolytic and glycolytic enzymes (40), and programming might give persistent changes if induced during critical periods of early development (41, 42).

The balance between the n-6 and n-3 series of FA and the endogenous FA of the n-7 and n-9 series in transformation to very long PUFA determines the competition between the substrates of the different series for the elongases and desaturases. This activity is indicated by the ratio between the FA in each series, and those did not differ for the EFA between obese and controls in this study. On the other hand, the high levels of nervonic acid and the high ratio of nervonic acid to its saturated analogue indicated a high stearyl-coenzyme A desaturase (SCD) activity (Table 4), which has previously been notified in obesity (5). Because palmitic and stearic acids are considered preferred substrates to this enzyme, it was unexpected not to find indications of increased transformations to palmitoleic and oleic acids, respectively. SCDs are identified in several isoforms, and SCD1 is better characterized than the others (43). The expression of SCDs are influenced by dietary factors, including PUFA and hormonal changes (i.e., insulin), and high SCD activity has been implicated in a wide range of disorders including diabetes, atherosclerosis, and obesity (44, 45).

Our finding of an inverse correlation between the relative amount of PUFA and serum insulin and the HOMA assessment of -cell function (Table 3 and Figure 2) is supported by Agostoni et al. (18), who showed a close relationship between the FA status, n-3 PUFA deficiency in particular, and insulin secretion, indicating that this relationship could be a marker of developing IR.

Visceral adiposity is an established risk factor for metabolic diseases in adults (46) and in children (47). However, a metabolic influence of VAT in the adolescents in this study was not found. Previous studies performed in obese children and adolescents have shown that hyperinsulinemia was related to VAT (47, 48) and to SAAT (28). It has been stated that the precise depot associated with IR was associated with the individual obesity status; in lean children it was subcutaneous, and in obese children it was visceral (27). Despite a very high degree of obesity in the children in our study, hyperinsulinemia was exclusively related to SAAT or general adiposity, not to VAT. SAAT, but not VAT, was also associated to the abnormal FA pattern and HOMA-IR.

The issue of VAT vs. SAAT as a major contributor to metabolic disturbances in adults is debated. In a study on lean and obese adults, Kelley et al. (29) showed that SAAT was associated with IR and hyperglycemia. They used computerized tomography to separate SAAT into deep and superficial layers and showed that the deep layer resembled VAT metabolically. The deep SAAT, therefore, was hypothesized to be responsible for the correlation between SAAT and hyperinsulinemia and hyperglycemia (29, 30). We could not confirm such metabolic difference between the deep and superficial SAAT because serum insulin and HOMA assessment of -cell function correlated both to the deep and superficial layers of and to total SAAT (Table 4). Also, the correlation pattern to the FA composition was similar. The fact that no certain differences could be detected might reflect changes in metabolism of compartments during the development of obesity or to how VAT is measured (27). In addition, the deep layer seems to be relatively larger in adults, constituting 60% of total SAAT in adults (29) as compared with our adolescents in whom it was <50% .

Our study has certain limitations. The small sample size reduces the power of the analysis, but the cost of MRI was a limiting factor. Despite this, the results differed clearly between the groups. A further weakness was that no functional measures of insulin response and IR were performed. Although HOMA-IR has been shown to be a relatively accurate measure of IR (35), using a functional test of insulin response would be preferable.

In conclusion, a higher ratio of n-6 to n-3 PUFA was found in obese adolescents vs. lean subjects and an association to abdominal AT, especially subcutaneous AT, anthropometric data, and metabolic variables. Differences in FA composition between obese and lean subjects were mainly observed in females. DHA correlated inversely to serum insulin and HOMA assessment of -cell function and to all SAAT compartments but not to VAT. The results suggest that the relation between AT compartments and lipid metabolism might differ in children and adults and that especially n-3 FAs are decreased in obese children.

Notes

¹ Nonstandard abbreviations: PUFA, polyunsaturated essential fatty acid; FA, fatty acid; AT, adipose tissue; PL, phospholipid; MRI, magnetic resonance imaging; VAT, visceral AT; SAAT, subcutaneous abdominal AT; BMI-sds, BMI-standard deviation score(s); HDL, high-density lipoprotein; LDL, low-density lipoprotein; HOMA, homeostasis model analysis; IR, insulin resistance; AA, arachidonic acid; DHA, docosahexaenoic acid; SFA, saturated FA; EFA, essential FA; SCD, stearyl-coenzyme A desaturase.

References

1. Ebbeling, C. B., Pawlak, D. B., Ludwig, DS. (2002) Childhood obesity: public-health crisis, common sense cure. Lancet. 360: 473–482. | Article | PubMed | ISI |
2. Summerbell, C. D., Waters, E., Edmunds, L. D., Kelly, S., Brown, T., Campbell, KJ. (2005) Interventions for preventing obesity in children. Cochrane Database Syst Rev. 3: CD001871 | PubMed |
3. Snethen, J. A., Broome, M. E., Cashin, SE. (2006) Effective weight loss for overweight children: a meta-analysis of intervention studies. J Pediatr Nurs. 21: 45–56. | Article | PubMed |
4. Ailhaud, G., Guesnet, P. (2004) Fatty acid composition of fats is an early determinant of childhood obesity: a short review and an opinion. Obes Rev. 5: 21–26. | Article | PubMed | ChemPort |
5. Pan, D. A., Hulbert, A. J., Storlien, LH. (1994) Dietary fats, membrane phospholipids and obesity. J Nutr. 124: 1555–1565. | PubMed | ChemPort |
6. Simopoulos, AP. (1999) Essential fatty acids in health and chronic disease. Am J Clin Nutr. 70: (Suppl 3), 560–569S.
7. Sanders, TAB. (2000) Polyunsaturated fatty acids in the food chain in Europe. Am J Clin Nutr. 71: 176–178S.
8. Kris-Etherton, P. M., Taylor, D. S., Yu-Poth, S., et al (2000) Polyunsaturated fatty acids in the food chain in the United States. Am J Clin Nutr. 71: 179–188S.
9. Innis, SM. (2000) Essential fatty acids in infant nutrition: lessons and limitations from animal studies in relation to studies on infant fatty acid requirements. Am J Clin Nutr. 71: 238–244S.
10. Clandinin, M. T., Jumpsen, J., Suh, M. (1994) Relationship between fatty acid accretion, membrane composition, and biologic functions. J Pediatr. 125: S25–S32. | Article | PubMed | ChemPort |
11. Clarke, S. D., Gasperikova, D., Nelson, C., Lapillonne, A., Heird, WC. (2002) Fatty acid regulation of gene expression: a genomic explanation for the benefits of the Mediterranean diet. Ann NY Acad Sci. 967: 283–298. | PubMed | ChemPort |
12. Clandinin, M. T., van Aerde, J. E., Merkel, K. L., et al (2005) Growth and development of preterm infants fed infant formulas containing docosahexaenoic acid and arachidonic acid. J Pediatr. 146: 461–468. | Article | PubMed | ISI | ChemPort |
13. Innis, SM. (2003) Perinatal biochemistry and physiology of long-chain polyunsaturated fatty acids. J Pediatr. 143: (Suppl 1), 1–8. | PubMed |
14. Korotkova, M., Gabrielsson, B., Holmäng, A., Larsson, B-M, Hanson, LÅ, Strandvik, B. (2005) Gender related long-term effects in adult rats by perinatal dietary ratio of n-6/n-3 fatty acids. Am J Physiol Regul Integr Comp Physiol. 288: R575–R579. | PubMed | ChemPort |
15. Korotkova, M., Ohlsson, C., Gabrielsson, B., Hanson, LÅ, Strandvik, B. (2004) Perinatal essential fatty acid deficiency affects weight and bone growth and mineralization in adult male rats. Br J Nutr. 92: 643–648. | Article | PubMed | ChemPort |
16. Ailhaud, G., Massiera, F., Weill, P., Legrand, P., Alessandri, J-M, Guesnet, P. (2006) Temporal changes in dietary fats: role of n-6 polyunsaturated fatty acids in excessive adipose tissue development and relationship to obesity. Progr Lipid Res. 45: 203–236. | Article | ChemPort |
17. Decsi, T., Molnár, D., Koletzko, B. (1996) Long chain polyunsaturated fatty acids in plasma lipids of obese children. Lipids. 31: 305–311. | Article | PubMed | ChemPort |
18. Agostoni, C., Riva, E., Bellú, R., et al (1994) Relationships between the fatty acid status and insulinemic indexes in obese children. Prostagl Leukotr Ess Fatty Acids. 51: 317–321. | Article | ChemPort |
19. Stubbs, C. D., Smith, AD. (1984) The modification of mammalian membrane polyunsaturated fatty acid composition in relation to membrane fluidity and function. Biochim Biophys Acta. 779: 89–137. | PubMed | ISI | ChemPort |
20. Moore, R. A., Oppert, S., Eaton, P., Mann, JI. (1977) Triglyceride fatty acids confirm a change in dietary fat. Clin Endocrinol (Oxf). 7: 143–149. | Article | PubMed | ChemPort |
21. Zock, P. L., Mensink, R. P., Harryvan, J., de Vries, J. H., Katan, MB. (1997) Fatty acids in serum cholesteryl esters as quantitative biomarkers of dietary intake in humans. Am J Epidemiol. 145: 1114–1122. | PubMed | ISI | ChemPort |
22. Seppanen-Laakso, T., Laakso, I., Vanhanen, H., Kiviranta, K., Lehtimaki, T., Hiltunen, R. (2001) Major human plasma lipid classes determined by quantitative high-performance liquid chromatography, their variation and associations with phospholipid fatty acids. J Chromatogr B Biomed Sci Appl. 754: 437–445. | Article | PubMed | ChemPort |
23. Chowdhury, B., Sjöström, L., Alpsten, M., Kostanty, J., Kvist, H., Löfgren, R. (1994) A multicompartment body composition technique based on computerized tomography. Int J Obes Relat Metab Disord. 18: 219–234. | PubMed | ChemPort |
24. Neovius, M., Linné, Y., Rössner, S. (2005) BMI, waist-circumference and waist-hip-ratio as diagnostic tests for fatness in adolescents. Int J Obes. 29: 163–169. | Article | ISI | ChemPort |
25. Pouliot, M-B, Després, J-P, Lemieux, S., et al (1994) Waist circumference and abdominal sagittal diameter: best simple anthropometric indexes of abdominal visceral adipose tissue accumulation and related cardiovascular risk in men and woman. Am J Cardiol. 73: 460–468. | Article | PubMed | ISI | ChemPort |
26. Després, J-P, Prud'homme, D., Pouliot, M-C, Tremblay, A., Bouchard, C. (1991) Estimation of deep adipose-tissue accumulation from simple anthropometric measurements in men. Am J Clin Nutr. 54: 471–477. | PubMed | ISI | ChemPort |
27. Goran, M. I., Gower, BA. (1999) Relation between visceral fat and disease risk in children and adolescents. Am J Clin Nutr. 70: 149–156S.
28. Gower, B. A., Nagy, T. R., Trowbridge, C. A., Dezenberg, C., Goran, MI. (1998) Fat distribution and insulin response in prepubertal African American and white children. Am J Clin Nutr. 67: 821–827. | PubMed | ChemPort |
29. Kelley, D. E., Thaete, L. F., Troost, F., Huwe, T., Goodpaster, BH. (2000) Subdivisions of subcutaneous abdominal adipose tissue and insulin resistance. Am J Physiol Endocrinol Metabol. 278: E941–E948. | ChemPort |
30. Smith, S. R., Lovejoy, J. C., Greenway, F., Ryan, D., et al (2001) Contributions of total body fat, abdominal subcutaneous adipose tissue compartments, and visceral adipose tissue to the metabolic complications of obesity. Metabolism. 50: 425–435. | Article | PubMed | ISI | ChemPort |
31. Cole, T. J., Bellizzi, M. C., Flegal, K. M., Dietz, WH. (2000) Establishing a standard definition for child overweight and obesity worldwide: international survey. BMJ. 320: 1–6. | Article | PubMed |
32. Tanner, J. M., Whitehouse, RH. (1976) Clinical longitudinal standards for height, weight, height velocity, weight velocity and stages of puberty. Arch Dis Child. 51: 170–179. | PubMed | ISI | ChemPort |
33. Karlberg, J., Luo, Z. C., Albertsson-Wikland, K. (2001) Body mass index reference values (mean and SD) for Swedish children. Acta Paediatr. 90: 1427–1434. | Article | PubMed | ISI | ChemPort |
34. Friedewald, W. T., Levy, R. I., Fredrickson, DS. (1972) Estimation of the concentration of low-density lipoprotein cholesterol in plasma, without use of the preparative ultracentrifuge. Clin Chem. 18: 499–452. | PubMed | ISI | ChemPort |
35. Matthews, D. R., Hosker, J. P., Rudenski, A. S., et al (1985) Homeostasis model assessment: insulin resistance and B-cell function from fasting plasma glucose and insulin concentrations in man. Diabetologia. 28: 412–419. | Article | PubMed | ISI | ChemPort |
36. Korotkova, M., Gabrielsson, B., Hanson, LÅ, Strandvik, B. (2001) Maternal essential fatty acid deficiency depresses serum leptin levels in suckling rat pups. J Lipid Res. 42: 359–365. | PubMed | ChemPort |
37. Savva, S. C., Chadjigeorgiou, B., Hatzid, B., et al (2004) Association of adipose tissue arachidonic acid content with BMI and overweight status in children from Cyprus and Crete. Br J Nutr. 91: 643–649. | Article | PubMed | ChemPort |
38. Murphy, M. J., Metcalf, B. S., Voss, L. D., et al (2004) Girls at five are intrinsically more insulin resistant than boys: the programming hypothesis revisited: the earlybird study (Early Bird 6). J Pediatr. 113: 82–86. | Article |
39. Chen, Z. Y., Kwan, K. Y., Tong, K. K., Ratnayake, W. M. N., Li, H. Q., Leung, SSF. (1997) Breast milk fatty acid composition: a comparative study between Hong Kong and Chongqing Chinese. Lipids. 32: 1061–1067. | Article | PubMed | ChemPort |
40. Jump, D. B., Clarke, S. D., Thelken, A., Liimatta, M. (1994) Coordinate regulation of glycolytic and lipogenic gene expression by polyunsaturated fatty acids. J Lipid Res. 35: 1076–1084. | PubMed | ISI | ChemPort |
41. Barker, DJ. (1990) The fetal and infant origins of adult disease. BMJ. 301: 1111 | PubMed | ChemPort |
42. Lucas, A., Fewtrell, M. S., Cole, TJ. (1999) Fetal origins of adult disease-the hypothesis revisited. BMJ. 319: 245–249. | PubMed | ChemPort |
43. Ntambi, J. M., Miyazaki, M., Dobrzyn, A. (2004) Regulation of stearyl-CoA desaturase expression. Lipids. 39: 1061–1065. | Article | PubMed | ChemPort |
44. Ntambi, J. M., Miyazaki, M. (2004) Regulation of stearyl-desaturases and role in metabolism. Progr Lipid Res. 43: 91–104. | Article | ChemPort |
45. Ntambi, J. M., Miyazaki, M., Stoehr, J. P., et al (2002) Loss of stearoyl-CoA desaturase-1 function protects mice against adiposity. Proc Natl Acad Sci U S A. 99: 11482–11486. | Article | PubMed | ChemPort |
46. Matsuzawa, Y., Shimomura, I., Nakamura, T., Keno, Y., Kotani, K., Tokunaga, K. (1995) Pathophysiology and pathogenesis of visceral fat obesity. Obes Res. 3: (Suppl 2), 187–194.
47. Gower, B. A., Nagy, T. R., Goran, MI. (1999) Visceral fat, insulin sensitivity, and lipids in prepubertal children. Diabetes. 48: 1515–1521. | Article | PubMed | ISI | ChemPort |
48. Caprio, S. (1999) Relationship between abdominal visceral fat and metabolic risk factors in obese adolescents. Am J Hum Biol. 11: 259–266. | Article | PubMed | ISI |

Acknowledgments

The study was supported by grants from the Västra Götaland Region and the Faculty of Medicine, Göteborg University. We gratefully acknowledge the technical assistance of Eva Bergelin, Stig Eriksson, Jacqueline Nel, Lena Strid, and Berit Holmberg and the pediatric nurses Susann Regber and Charlotte Arfwidsson.