点击显示 收起
1 From the Human Nutrition Laboratory, Institute of Food Science and Nutrition, Swiss Federal Institute of Technology, Zurich, Switzerland (IA and MBZ); the Department for Growth and Development, University Children's Hospital, Zurich, Switzerland (LM); the Clinic for Endocrinology and Diabetes, University Hospital Zurich, Zurich, Switzerland (GS and RL); and the Children's Hospital of Eastern Switzerland, St Gallen, Switzerland (Dl'A)
2 Supported by the Swiss Diabetes Foundation (Steinhausen, Switzerland), the Swiss Ministry of Health (Bern, Switzerland), and the Swiss Federal Institute of Technology (Zurich, Switzerland). 3 Reprints not available. Address correspondence to MB Zimmermann, Institute of Food Science and Nutrition, Human Nutrition Laboratory, ETH Zurich, LFV E19, Schmelzbergstrasse 7, 8092 Zürich, Switzerland. E-mail: michael.zimmermann{at}ilw.agrl.ethz.ch.
ABSTRACT
Background: In obese children, subclinical inflammation is often present and is correlated with the metabolic syndrome. Dietary factors, such as fatty acids and antioxidants, potentially modulate the association between adiposity and subclinical inflammation, but few data are available in children.
Objective: The aim of the study was to determine whether dietary fat or antioxidant intakes influence circulating tumor necrosis factor (TNF-), interleukin 6 (IL-6), C-reactive protein (CRP), and leptin concentrations in overweight children.
Design: In a cross-sectional study of 6–14-y-old normal-weight (n = 33), overweight (n = 19), and obese (n = 27) Swiss children, nutritional intakes were assessed from two 24-h dietary recalls and a 1-d dietary record. Percentage body fat from skinfold thicknesses, waist-hip ratio, and blood pressure were measured. Fasting blood samples were collected for the measurement of insulin, glucose, HDL-cholesterol, triacylglycerol, CRP, IL-6, TNF-, and leptin concentrations.
Results: CRP, IL-6, and leptin increased significantly (P < 0.02) with increasing adiposity, independent of age; TNF- did not increase. Total dietary fat and the percentage of energy from fat were significant predictors of CRP concentration, independent of body mass index (P < 0.05). Meat intake was a significant predictor of IL-6 and leptin, independent of body mass index (P < 0.05). Intakes of antioxidant vitamins (vitamins E and C and ß-carotene) were significant predictors of leptin (P < 0.05) but not of CRP, IL-6, or TNF-.
Conclusions: Overweight Swiss children as young as 6 y have elevated concentrations of inflammatory markers. Intakes of total fat and antioxidant vitamins are determinants of subclinical inflammation in this age group.
Key Words: Children • obesity • inflammation • diet • C-reactive protein • interleukin 6 • leptin • tumor necrosis factor • fat • antioxidants
INTRODUCTION
The prevalence of pediatric obesity is increasing globally at an alarming rate (1, 2). Many of the cardiovascular and metabolic complications of excess body fat are already detectable in overweight children (3). Childhood obesity is a significant early risk factor for adult cardiovascular disease and diabetes (4, 5). In obesity, adipocytes produce proinflammatory cytokines, including tumor necrosis factor (TNF-) and interleukin 6 (IL-6) (6, 7). IL-6 stimulates the hepatic synthesis of C-reactive protein (CRP), an inflammatory marker. In addition, leptin, which is elevated in obese subjects, has been shown to be proinflammatory (8). Subclinical inflammation in adults is associated with a greater risk of type 2 diabetes and cardiovascular disease (9-11). In obese older children and adolescents, elevated concentrations of CRP and IL-6 correlate with components of the metabolic syndrome (12), including elevated body mass index (BMI), high triacylglycerol concentrations, low HDL-cholesterol concentrations, elevated systolic blood pressure, and impaired glucose tolerance (12-17). Subclinical inflammation also correlates with markers of oxidative stress (18). In obesity, CRP concentrations predict 8-iso-prostaglandin F2 (8-iso-PGF2) excretion (a marker of lipid peroxidation) independently of insulin and leptin concentrations (19). In adults, increased oxidative stress in adipose tissue may be a mechanism of obesity-associated insulin resistance and the metabolic syndrome (20, 21).
Dietary factors, such as fatty acids (22, 23) and antioxidants (24), potentially modulate the association between adiposity and subclinical inflammation (25). Both the type and total amount of dietary fat can influence plasma and tissue fatty acid composition (26, 27), which, in turn, can modify inflammation (28, 29) and inflammation-mediated disease (30). Overweight adolescents with elevated CRP and IL-6 concentrations have higher plasma concentrations of saturated fatty acids and lower concentrations of n–3 polyunsaturated fatty acids (31). Dietary antioxidants can influence the expression of proinflammatory cytokines (32). Low antioxidant vitamin concentrations and reduced antioxidant capacity are common in obese children (33-37). Therefore, the aim of the present study was to investigate the dietary determinants of subclinical inflammation in overweight young children. Specifically, we examined whether dietary fat or antioxidant intakes influence TNF-, IL-6, CRP, and leptin concentrations in overweight Swiss children.
SUBJECTS AND METHODS
The subjects for this study were 6–14 y-old children (n = 79) living in northern Switzerland. The children were a convenience sample recruited through letters to primary schools and pediatric clinics. Children recruited from the pediatric clinics were otherwise healthy overweight children presenting to the clinic for the first time for weight-loss counseling. Our intent was to enroll 25 normal-weight, 25 overweight, and 25 obese children for the study. Informed written consent was obtained from the parents, and informed oral assent was obtained from the children. Ethical approval for the study was obtained from the ethics committee of the Swiss Federal Institute of Technology in Zürich.
Well-trained female interviewers assessed the dietary intakes of each child using two 24-h dietary recalls and one 1-d weighed food record in the family home. Each child was visited 3 times by the same interviewer within 3 wk. The 24-h dietary recalls were done at the first and second visits, and, at the second visit, the interviewer gave instructions and guidelines for the 1-d weighed food record. At the third visit, the 1-d food record was carefully reviewed, and a questionnaire on usual dietary habits was completed with the child. The questionnaire included queries on food and beverage preferences, including sweets, baked goods, and snacks, and frequency and temporal patterns of snacking.
Volumes and portion sizes for the 24-h dietary recalls were estimated by using measuring cups and spoons, photographs of food portions, and graduated food samples of cheese and bread. The combination of 24-h dietary recalls and a food record provides a good overview of a child's habitual diet; this approach has been validated in children as young as 8 y of age (38). An appointment was scheduled at the hospital clinic, at the parents' convenience, usually within 1–2 wk of the dietary assessment.
The children arrived at the hospital clinic in the morning after a 12-h overnight fast. Twelve milliliters of blood (2 mL into EDTA-containing tubes) was collected by venipuncture. Height was measured to the nearest 0.5 cm and weight to the nearest 100 g with a digital balance (BF 18; Beurer, Ulm, Germany). Pubertal staging was done. Waist and hip circumferences were measured with a nonstretchable measuring tape. Skinfold thicknesses were measured at the triceps and subscapular sites with a Harpenden Skinfold Caliper (HSK-BI; British Indicators, West Sussex, United Kingdom) with a constant spring pressure of 10 g/mm2 and a resolution of 0.2 mm. A single experienced observer (IA) performed all of the measurements of waist and hip circumferences and skinfold thicknesses. After a 15-min rest, supine resting blood pressure was measured by auscultation.
Data analysis
Dietary data obtained from the 3 records were checked carefully and entered by the lead interviewer (IA) into a nutrition software system (EBISpro for Windows 4.0; Dr J Erhardt, University of Hohenheim, Germany). This system translates the amount of food eaten into individual nutrients and assigns consumed foods into 22 food groups. The program is based on the German Food and Nutrition Data Base BLS 2.3 (Federal Health Department, Berlin, Germany), and, for foods specific to Switzerland, incorporates values from the Swiss Food Composition Database (39). Energy and nutrient data were averaged across the 3 d to obtain a mean daily energy and nutrient intake for each child. The reference values for nutrient intakes for Germany, Austria, and Switzerland—the D-A-CH references (40)—were used for comparison of the actual intakes with the recommended intakes for the respective age groups.
The BMIs of the children were calculated as weight (kg)/height squared (m). BMI SD scores (BMI-SDS = individual BMI value – reference mean BMI value ÷ SD, to scale the data for comparison across ages and sex) were calculated and used in the analysis. Age- and sex-specific criteria from the US Centers for Disease Control and Prevention (41) were used to classify children as normal weight, overweight (above the 85th percentile), or obese (above the 95th percentile). These criteria were previously validated in Swiss children at this age (42). Percentage body fat (%BF) was calculated by using the equation of Lohman (43). Fat-free body mass (FFB) was determined as follows:
For boys:
RESULTS
Anthropometric and metabolic data of the children, by weight classification, are shown in Table 1. %BF, waist-hip ratio, fasting insulin, insulin resistance as defined by QUICKI, triacylglycerols, and systolic blood pressure increased significantly with increasing adiposity (P < 0.01), whereas HDL-cholesterol concentrations were lower with increasing adiposity (P < 0.01). Increasing adiposity was associated with significantly higher circulating concentrations of CRP (P < 0.02), IL-6 (P < 0.02), and leptin (P < 0.01) but not of TNF-.
View this table:
TABLE 1. Anthropometric and metabolic data for 6- to 14-y-old Swiss children by weight classification1
Daily intakes of energy, total fat, types of fat, protein, and antioxidant vitamins, by weight classification, are shown in Table 2. The only significant differences were in intakes of meat and protein (P < 0.05) between normal-weight and obese children and in percentages of energy as protein (P < 0.05) between normal-weight and overweight and between normal-weight and obese children; intakes increased with increasing adiposity. To further investigate the relation between dietary factors and BMI as continuous variables, the following correlations were calculated: energy intake (r = 0.053, P = 0.644), fat intake (r = –0.036, P = 0.750), saturated fatty acid intake (r = –0.089, P = 0.435), polyunsaturated fatty acid intake (r = 0.061, P = 0.593), monounsaturated fatty acid intake (r = –0.025, P = 0.825), protein intake (r = 0.326, P = 0.003), and meat intake (r = 0.388, P < 0.001). Again, the only dietary factors significantly correlated with BMI-SDS were protein and meat intakes.
View this table:
TABLE 2. Nutritional intakes of 6- to 14-y-old Swiss children by weight classification1
The results of the univariate regressions of CRP, IL-6, TNF-, and leptin as dependent variables on BMI, %BF, waist-hip ratio, age, and sex as predictors are shown in Table 3. BMI, %BF, and waist-hip ratio were significant predictors of circulating CRP, IL-6, and leptin but not of TNF-. Sex was not a significant predictor, with the exception of leptin, for which values were significantly higher in girls (P < 0.01). Age was a significant predictor of TNF- (P < 0.01), but not of the other inflammatory markers.
View this table:
TABLE 3. Associations of C-reactive protein (CRP), interleukin 6 (IL-6), tumor necrosis factor (TNF-), and leptin with BMI, percentage body fat (%BF), waist-hip ratio, age, and sex in 6- to 14-y-old Swiss children in univariate regression models1
Associations of the 4 inflammatory markers with individual components of the metabolic syndrome, in univariate regressions and after control for BMI-SDS, are shown in Table 4. For CRP, IL-6, and TNF-, no significant associations with components of the metabolic syndrome remained after control for BMI-SDS. In contrast, associations of leptin with fasting insulin, QUICKI, systolic blood pressure, and triacylglycerols remained significant after control for adiposity.
View this table:
TABLE 4. Associations of inflammatory markers with individual components of the metabolic syndrome in 6- to 14-y-old Swiss children in multivariate regression models1
Regressions of dietary factors on CRP, IL-6, and leptin, after control for BMI-SDS and age, are shown in Table 5. Total fat intake, percentage of energy as fat, plant oils, saturated fatty acid, polyunsaturated fatty acid, and monounsaturated fatty acid were all significant predictors of CRP. Vitamin C, vitamin E, and ß-carotene intakes were significant negative predictors of leptin concentrations (P = 0.05), and meat intake was a significant predictor of IL-6 (P = 0.032). None of the dietary factors shown in Table 5 were significant predictors of TNF- after adjustment for BMI-SDS and age (data not shown).
View this table:
TABLE 5. Associations of nutritional factors with C-reactive protein (CRP), interleukin 6 (IL-6), and leptin concentrations in 6- to 14-y-old Swiss children in multivariate regression models1
DISCUSSION
In Swiss children as young as 6 y, measures of adiposity (BMI, %BF, and waist-hip ratio) were significant predictors of 3 inflammatory markers: IL-6, CRP, and leptin. Although data from younger children is scarce, and most studies measured only CRP (12, 13, 15, 16), our findings are consistent with previous studies, which showed increased subclinical inflammation in obese prepubertal children (3). In our sample, no association of serum TNF- with adiposity was observed. Some studies have found elevated circulating TNF- concentrations in obese children (45, 46), whereas others have not (47). It appears that TNF- acts primarily via its transmembrane form in an autocrine-paracrine fashion in adipocytes (6, 48, 49). Thus, although TNF- expression is increased in adipose tissue, circulating TNF- concentrations in obese animals and humans are often not detectable or only mildly elevated (49).
Although dietary studies in obese children have focused mainly on the metabolic syndrome and dyslipidemia (50, 51), diet may also modulate the association between adiposity and subclinical inflammation (22, 24, 25). Differences in total fat intake and specific fatty acids can modulate the inflammatory response (22, 23). High concentrations of plasma free fatty acids (FFAs)—due to high fat intakes, increased lipolysis, or both—are found in overweight adults with the metabolic syndrome (52). High concentrations of FFAs, especially saturated fatty acids, impair glucose and lipid metabolism (52) and induce expression of proinflammatory cytokines in adipocytes (53), but this may not be a generalized effect of all saturated fatty acids. Palmitate, but not laurate or docosahexaenoic acid, induces IL-6 in adipocytes (53), skeletal muscle cells (54), and endothelial cells (55). Palmitate, but not other plasma FFAs, correlates in vivo with IL-6 concentrations, independent of body fat mass (55). Conjugated linoleic acid (18:2; with a conjugated double bond) may also increase production of IL-6 and other proinflammatory cytokines in adipose tissue (56). In contrast, n–3 polyunsaturated fatty acids may ameliorate the inflammatory response (57, 58). Supplementation with n–3 polyunsaturated fatty acids modified phospholipid fatty acid composition and decreased production of TNF- and IL-6 by monocytes (59), when compared with palmitic acid (60). Diets high in -linolenic acid (ALA,18:3n-3) tend to reduce inflammation (61-63). In dyslipidemic adults, daily supplementation with ALA for 3 mo decreased concentrations of CRP and IL-6 by 38% and 11%, respectively (64).
Overweight adolescents with elevated CRP and IL-6 concentrations have higher plasma saturated fatty acid and lower n–3 polyunsaturated fatty acid concentrations than do normal-weight control subjects (31). It has been suggested that high dietary intakes of saturated fatty acids may stimulate IL-6 secretion, and, conversely, high intakes of n–3 polyunsaturated fatty acids may inhibit low-grade inflammation (31, 65). However, in our study, total fat intake and percentage of energy from fat—and not specific types of dietary fat—predicted subclinical inflammation, independent of BMI. Also, intake of food groups high in saturated fats, such as meat and milk products, did not predict subclinical inflammation after control for adiposity, with the exception of a significant association between meat intake and IL-6 and leptin. In adults, a diet pattern high in processed meats correlated with inflammatory markers, including IL-6 (66).
Oxidative stress may play a central role in the pathogenesis of various disease states, including diabetes and insulin resistance (21, 67). Obesity per se may induce systemic oxidative stress (20), which may contribute to the dysregulation of adipocytokines and to the development of the metabolic syndrome (68, 69). In obese adults, subclinical inflammation correlates with markers of oxidative stress (18), and CRP concentrations predict 8-iso- PGF2 excretion independently of insulin and leptin concentrations (19). Obesity in adults may be associated with lower concentrations of -tocopherol and ß-carotene (18, 70).
Decreased antioxidant vitamin concentrations and reduced antioxidant capacity also appear to be characteristic of obese children (33, 34, 36). In prepubertal obese children, vitamin E concentrations are significantly lower and malondialdehyde concentrations are significantly higher than those of normal-weight children, and these abnormalities are improved by a dietary restriction weight-loss program (35). In 6–19-y-old children in the second National Health and Nutrition Examination Survey (NHANES III), obese children had low serum concentrations of -tocopherol and ß-carotene (37). However, in NHANES III, no significant differences in intakes of ß-carotene, -tocopherol, fruit, or vegetables were observed between obese and nonobese children, as assessed by 24-h dietary recall and food-frequency questionnaires (37). Similarly, we found that adiposity did not predict antioxidant vitamin intakes. Intakes of vitamin C, vitamin E, and ß-carotene did not predict subclinical inflammation, as measured by IL-6 and CRP, but did predict leptin.
CRP is significantly associated with adiposity and components of the metabolic syndrome in adults (16, 71-74). CRP also is a predictor of insulin resistance in adults (16, 71-74), but the relation is variably attenuated after adjustment for adiposity (71, 72). In children, data are more limited, but CRP has also been correlated with components of the metabolic syndrome (12-17). However, studies suggest that increases in CRP, independent of adiposity, are associated with dyslipidemia but are not strongly associated with other components of the metabolic syndrome in children. Similarly, in our study, univariate analyses found correlations between IL-6, CRP, leptin, and several components of the metabolic syndrome (Table 4). However, after control for BMI, the associations with CRP and IL-6 were no longer significant. In contrast, leptin remained a significant predictor of fasting insulin, QUICKI, systolic blood pressure, and triacylglycerols, even after control for BMI.
In adults, stronger associations between adiposity and CRP have been reported for women than for with men (75, 76). In children, the data are unclear. In Quebec, CRP concentrations were higher in 9- and 16-y-old girls than in boys of the same age (15). In UK children aged 9–11 y, CRP concentrations were 47% higher in girls than in boys (13). In adolescents in the 1999–2000 US NHANES (14), females had higher CRP concentrations than did males only at 16–19 y of age. In contrast, CRP concentrations were similar by sex in a study in children in Taipei (17). In our sample, no significant age or sex effect on IL-6 or CRP was observed, whereas female sex was a significant predictor of leptin concentrations.
Dietary assessment is challenging, particularly in children. Children's abilities to record or remember their diet, as well as their knowledge of food and food preparation, is limited (77, 78). Different assessment tools have strengths and weaknesses. A dietary record, considered the "gold standard," provides an accurate quantitative account of a person's diet, but it is demanding and time consuming and requires motivated and literate subjects. In contrast, a 24-h dietary recall provides only a snapshot of a person's diet on the day before the interview, but it requires less motivation to perform and the subject does not have to be literate. Because of the wide variation in daily diets, a single day assessment by a 24-h dietary recall is unlikely to yield an accurate picture of a subject's diet (78-80). Lytle et al (38) validated a 24-h dietary recall in children, supplemented with food records done by the children's parents. Because this combination of interviewing both children and their parents showed good results, and because 24-h dietary recalls and a diet record are complementary (38), we chose to combine two 24-h dietary recalls with a 1-d dietary record in the present study. This method aimed to minimize time demands on the children and their families while trying to ensure an accurate overview of their diet.
This study had several limitations. Because it was cross-sectional, the directionality of the reported associations cannot be established. The use of a single fasting measurement of CRP, IL-6, TNF-, or leptin may not necessarily reflect chronic inflammatory status; the biological variability of CRP in adults has been estimated at 22% (81). Finally, the sample size may have been insufficient to detect weak associations. However, studies of the early mechanisms of subclinical inflammation in young children are valuable because they are not confounded by chronic inflammatory conditions, atheromatous disease, smoking, or alcohol consumption.
ACKNOWLEDGMENTS
We thank the participating children and their families. Special thanks go to MT Achermann and S Jacob (University Hospital Zurich); RF Hurrell, K Hotz, and C Zeder (ETH Zurich); I Hutter (Children's Hospital, St Gallen); and the staff of the Children's Hospitals in Biel and Basel.
IA, MBZ, GS, and DA performed the fieldwork and the data collection. IA, MBZ, and GS supervised the laboratory analysis and completed the data analysis. IA, MBZ, and LM conducted the statistical analysis. IA and MBZ wrote the first draft of the manuscript. All authors contributed to the study design and edited the manuscript. None of the authors had a financial or personal conflict of interest in regard to this study.
REFERENCES