High‐sensitivity C‐reactive Protein is a Predictive Factor of Adiposity in Children: Results of the Identification and prevention of Dietary‐ and lifestyle‐induced health Effects in Children and InfantS (IDEFICS) Study
Whereas cross‐sectional studies have shown that obesity is associated with increased C‐reactive protein (
Methods and Results
We assessed cross‐sectionally and longitudinally the association of high‐sensitivity (hs)‐
The role of chronic low‐grade inflammation as a possible link between obesity and its metabolic and cardiovascular sequelae has been convincingly showed during the last years.1
C‐reactive protein (CRP) has long been considered a prototypic biomarker of subclinical inflammation.2 Since analytical standards were defined for its measurement,3 the high‐sensitivity CRP (hs‐CRP) assay is considered a sensitive predictor of atherosclerosis, diabetes and coronary heart diseases also in apparently healthy individuals.4, 5, 6, 7
About 10 years ago, large epidemiological studies unequivocally showed the association of low‐grade inflammation and overweight in children.8, 9, 10 The number of studies exploring different aspects of this association has been steadily increasing in the last years, as recently reviewed by Tam et al.11
The summary evidence from previous studies is that a consistent association exists between obesity and increased CRP levels in children. However, since most of the information derives from cross‐sectional investigations, there is a need to confirm this association in longitudinal studies, aiming to confirm the value of hs‐CRP as an early‐detectable surrogate marker in the natural history of the association between childhood obesity and cardiovascular disease late in adulthood.
The aim of the present study was to assess cross‐sectionally and longitudinally the association of hs‐CRP levels with overweight/obesity and related cardiometabolic risk factors in a large sample of European preschool and primary school children participating in the IDEFICS project. The 2‐year follow‐up of the Identification and prevention of Dietary‐ and lifestyle‐induced health Effects in Children and InfantS (IDEFICS) cohort allowed to test whether higher baseline levels of hs‐CRP were associated with higher body mass and increased incidence of overweight/obesity over time.
Materials and Methods
The study was conducted according to the standards of the Declaration of Helsinki. All applicable institutional and governmental regulations pertaining to the ethical use of human volunteers were followed during this research. Approval by the appropriate ethics committees was obtained by each of the 8 participating centers carrying out the fieldwork. Participants were not subjected to any study procedure before both the children and their parents gave their oral (children) and written (parents) informed consent for examinations, collection of samples, subsequent analysis and storage of personal data and collected samples.
IDEFICS is a large European multicenter study of childhood obesity. Details of the general design, instruments and survey characteristics can be found elsewhere.12
A cohort of 16 224 children aged 2 to 9 years was recruited into a population‐based baseline survey from 8 European countries ranging from the north to the south and from the east to the west (Belgium, Cyprus, Estonia, Germany, Hungary, Italy, Spain, and Sweden). The baseline survey (T0) was the starting point of the cohort study aimed to prospectively evaluate the role of the factors assessed at baseline on the development of overweight/obesity over time, and to assess the feasibility, effectiveness, and sustainability of a community‐oriented intervention program. Comparable intervention and control regions were selected in each country. In the intervention regions, a coherent set of intervention modules were implemented, focusing on diet, physical activity, and stress‐coping capacity. All the materials for the interventions were centrally developed and culturally adapted. A detailed description of the IDEFICS intervention program has been recently published.13
A second survey (T1) reassessed the children 2 years later. The follow‐up survey was synchronized with the baseline to account as much as possible for seasonal variation. In fact, the study protocol, closely followed by each survey center, allowed for a 2‐year period (±1 month) between T0 and T1.
Of the 16 224 children examined at T0, 7187 children were excluded from the present analysis. In particular, 4766 children had no hs‐CRP measured (Boys=50%; age=5.8± 1.8 years; mean±SD) and 2421 children (Boys=53%; age=5.9±1.8 years) were excluded because of the use of medications during the week previous to the visit. Eventually, the cross‐sectional analysis was carried out in 6616 children (Boys=50%; age=6.3±1.7 years). Excluded children were significantly younger than those included (P<0.001), likely due to the lower acceptance of blood drawing in younger children, while anthropometric parameters were not statistically significant different between the 2 groups (data not shown).
Of these 6616 children, 4110 were re‐examined 2 years later (T1) (Boys=51%; age=8.2±1.7 years, control group n=1960, intervention n=2150). The distribution of children in the control and intervention region was balanced across the baseline hs‐CRP categories. Since no statistically significant differences in the anthropometric outcomes were observed between children in the intervention and in the control regions, longitudinal analyses were performed on the whole T1 population sample.
Anthropometry and Blood Pressure
Children underwent a standardized physical examination. Anthropometric data included body weight and height, waist and hip circumferences, and the measurement of skinfold thickness. A detailed description of the anthropometric measurements in the IDEFICS study, including intra‐ and interobserver reliability, has been recently published.14
The measurement of weight was carried out using an electronic scale (Tanita BC 420 SMA, Tanita Europe GmbH) to the nearest 0.1 kg with children wearing light clothes without shoes. Height was measured using a telescopic height‐measuring instrument (Seca 225 stadiometer) to the nearest 0.1 cm. Body mass index (BMI) was calculated as weight (in kg) divided by height squared (in m). For each child, z‐scores of BMI and of BMI variation over the 2‐year follow‐up were determined according to the sex‐ and age‐specific z‐scores by Cole et al.15
Waist circumference was measured using an inelastic tape (Seca 200), precision 0.1 cm, range 0±150 cm at the midpoint between the iliac crest and the lower border or tenth rib with the subject in a standing position and recorded at the nearest 0.1 cm. Both triceps and subscapular skinfold thickness were measured by means of a caliper (Holtain, Holtain Ltd, range 0±40 mm). Measures were taken twice on the right hand side of the body and the mean was calculated.
For the definition of overweight/obesity, children were grouped into 2 categories using the cut‐points defined by Cole et al16, 17: (1) underweight (thinness grade III to I) plus normal weight and (2) overweight plus obese.
Systolic and diastolic blood pressure (BP) was measured after 5 minutes rest in the seated position, at the child right arm with a cuff of appropriate size for arm circumference using an automatic sphygmomanometer (Welch Allyn 4200B‐E2 Inc.). Two measurements were made and the average of the measurements was used for analysis.
Children were asked to participate, on voluntary basis, in blood drawing. A detailed description of sample collection and analytical procedures in the IDEFICS survey has been recently published.18, 19
CRP levels were measured in a central laboratory with a high‐sensitivity assay using latex‐enhanced nephelometry (BN2‐Nephelometer, Siemens), The lower limit of detection of the assay was 0.02 mg/dL.
Blood glucose, total‐ and HDL‐cholesterol, and triglycerides were assessed, in fasting status by the portable Cholestech LDX analyzer at each point‐of‐care center.20 LDL‐cholesterol was calculated according to the formula of Friedwald.21
Serum insulin and glycated hemoglobin (HbA1c) were measured in a central laboratory through enzyme‐linked immunosorbent assay kit (MODULAR E170, Roche Diagnostics) and by high‐pressure liquid chromatography with an automated HLC‐723G7 analyzer respectively. Homeostatic model assessment (HOMA) index was calculated according to the formula: HOMA = [serum insulin (μU/mL) × blood glucose (mmol/L)]/22.5.22 Categories of “insulin‐resistant” children were defined by the 75th percentile of HOMA index for boys (normal, HOMA index≤1.16, high >1.16) and girls (normal, HOMA index≤1.24, high >1.24).
Parents were asked to fill in a questionnaire to collect data on the participating child (date of birth, weight and length at birth, physical activity and lifestyle factors, personal and familial medical history). Parental information was collected as well (age, self‐reported weight and height, education level, and occupation). With regard to physical activity, parents were invited to indicate how many minutes per day their child used to spend in outdoor activities (ie, time spent playing in the yard or street around house or outdoor recreation area [swimming pool, zoo, park] either on week days or weekend days). An additional question investigated whether or not the child would have attended a club to practice sport. The total weekly physical activity time was calculated as the sum of the number of hours spent in each activity category. Educational level of the parents was estimated by International Standard Classification of Education (ISCED) criteria 23 in 6 categories (from 0 to 5). Parents were considered overweight if their self‐reported BMI was ≥25 kg/m2. Breastfeeding was classified as: none; 0 to 3 months; 3 to 6 months; >6 months. Parents were also asked to indicate whether their child had taken any kind of medication, including self‐prescribed drugs, vitamin, and mineral supplements, during the week before blood sample drawing.
Statistical analyses were performed using PASW (Predictive Analytics SoftWare) Statistics (version 18; SPSS Inc.). The analyses were performed separately in boys and girls. Data are presented as mean (95% confidence intervals) or median and range, as indicated. Categorical variables are shown as absolute frequencies. Given that the distribution of hs‐CRP was skewed, log‐transformed values were used for statistical analysis. For ease of interpretation, untransformed values are presented. Because about 50% of children had values less than the minimal detectable concentration, in this analysis hs‐CRP is treated as a categorical rather than as a continuous variable. The following 3 hs‐CRP cut–offs were arbitrarily defined to categorize hs‐CRP levels in our population: (I) hs‐CRP under the detection limit (<0.02 mg/dL); (II) hs‐CRP>0.02 mg/dL and <75th sex‐specific percentile of those with hs‐CRP values over the detection limit (<0.177 mg/dL in boys and 0.180 mg/dL in girls) (III) hs‐CRP>75th sex‐specific percentile of those with hs‐CRP values over the detection limit (>0.177 mg/dL in boys and >0.180 mg/dL in girls). Prevalence of overweight/obesity across hs‐CRP categories was tested by linear by linear association chi‐square test statistics.
Multiple linear regression analysis (GLM) was used to compare mean values of outcome variables across categories of hs‐CRP. P values for linear trend were calculated. All analyses were adjusted for child age, country of origin, breast‐feeding, physical activity, parental overweight/obesity (none or at least one affected parent), and parental education. BP and metabolic variables were also adjusted for BMI.
Longitudinal analyses were performed using the 2‐year variation in the outcome adiposity variables (follow‐up, T1 minus baseline, T0) across the baseline hs‐CRP categories, by multiple linear regression analysis adjusted for the respective baseline value of the outcome variable and for other relevant confounders (age, country of origin, physical activity, parental overweight and obesity, breast‐feeding, parental education, intervention/control study group). P values for linear trend were calculated. Incident cases of overweight/obesity over the 2‐year follow‐up according to hs‐CRP groups at baseline were assessed by logistic regression analysis, taking into account age, country of origin, and intervention/control study group. Odds ratio and their 95% confidence intervals were calculated. A 2‐tailed P value less than 0.05 was considered statistically significant.
Baseline anthropometric and metabolic characteristics of the male and female participants are summarized in Table 1. Hs‐CRP levels ranged between 0.02 mg/dL (detection limit) and 8.50 mg/dL in boys and 9.35 mg/dL in girls, respectively.
|Boys (n=3347)||Girls (n=3269)|
|Sum of skinfolds, mm||17.17±7.58||19.95±8.11|
|Waist circumference, cm||54.92±7.06||54.43±6.93|
|SBP, mm Hg||100.6±9.2||100.4±9.2|
|DBP, mm Hg||62.6±6.6||63.4±6.5|
|Insulin, pmol/L||28.89±24.64 (n=2905)||33.20±27.39 (n=2777)|
|HOMA‐Index||0.90±0.83 (n=2732)||1.00±0.87 (n=2592)|
|Total Cholesterol, mmol/L||4.10±0.76||4.22±0.82|
|hs‐CRP (mg/dL), median (range)||0.03 (0.02 to 8.50)||0.04 (0.02 to 9.35)|
Table 2 shows the anthropometric and metabolic variables at baseline (T0) according to the 3 categories of hs‐CRP in boys and girls. The prevalence of overweight/obesity significantly increased across the categories of hs‐CRP in both sexes (Table 2). At multiple regression analysis, adjusted for age, country of origin, breast‐feeding, physical activity, parental overweight/obesity, and parental education, a linear highly significant increase in BMI z‐score, waist circumference, and sum of skinfolds was observed both in boys and in girls, moving across the categories of hs‐CRP.
|I||II||III||P Value for Trend||I||II||III||P Value for Trend|
|<0.020 mg/dL||>0.020<0.177 mg/dL||>0.177 mg/dL||<0.020 mg/dL||>0.020<0.180 mg/dL||>0.180 mg/dL|
|SBP, mm Hga||100.1±0.2||101.1±0.2||100.7±0.4||0.03||100.3±0.3||100.2±0.2||100.8±0.4||NS|
|DBP, mm Hga||62.3±0.2||62.9±0.2||63.2±0.3||0.004||63.0±0.21||63.5±0.17||64.0±0.30||0.007|
After adjustment for BMI, a significant increase in diastolic blood pressure was observed in both sexes, while higher systolic blood pressure values in the higher hs‐CRP categories were observed only in boys. In both sexes higher values of hs‐CRP were associated with lower total cholesterol and HDL‐cholesterol levels. The ratio between total and HDL‐cholesterol was significantly higher in girls with higher hs‐CRP levels. Indices of glucose status/insulin‐resistance did not vary across hs‐CRP categories in both sexes, but for the lower plasma glucose levels observed in the boys with higher hs‐CRP values. At categorical analysis, the proportion of children with higher insulin resistance (HOMA index > 75th percentile of the distribution for boys and girls respectively) did not differ across the categories of hs‐CRP.
Among the 6616 children examined at T0, 4110 were reexamined 2 years later (T1) (62% of the T0 sample). Table 3 shows the prevalence of overweight/obesity and the adjusted changes in anthropometric variables over the 2‐year follow‐up across the hs‐CRP categories assessed at baseline (T0). In both boys and girls, over the 2‐year follow‐up, participants with higher baseline hs‐CRP levels showed significantly higher increase in BMI z‐score and indices of central adiposity, after adjustment for the respective baseline value and independently of the study group (control versus intervention). The highly significant statistical association between baseline hs‐CRP values and the variation of BMI over the 2‐year follow‐up was indeed not affected by the intervention (P for trend: boys, unadjusted=0.000066; adjusted=0.000062; girls, unadjusted=0.001; adjusted=0.001).
|I||II||III||P Value for Trend||I||II||III||P Value for Trend|
|<0.020 mg/dL||>0.020<0.177 mg/dL||>0.177 mg/dL||<0.020 mg/dL||>0.020<0.180 mg/dL||>0.180 mg/dL|
|Δ BMI z‐score||0.107±0.022||0.196±0.023||0.279±0.041||<0.0001||0.071±0.027||0.188±0.022||0.235±0.039||<0.001|
|Δ SS, mm||2.046±0.166||3.018±0.172||3.058±0.312||0.0001||2.682±0.194||3.260±0.159||3.404±0.281||0.019|
|Δ WC, cm||4.150±0.122||4.614±0.127||4.658±0.227||0.011||4.013±0.154||4.424±0.125||5.006±0.220||0.0003|
Out of 4110 children reexamined at T1, 1951 boys and 1819 girls were not overweight/obese at baseline (T0). After the 2‐year follow‐up, 179 incident cases of overweight/obesity were observed in boys and 137 in girls. At logistic regression analysis−adjusted for age, country of origin, physical activity, parental overweight and obesity, breast‐feeding, parental education, and intervention/control study group−the risk to become overweight/obese, according to the baseline hs‐CRP levels, increased significantly across the hs‐CRP categories (Table 4).
|Boys (n=1952)||Girls (n=1819)|
|OR||(95% CI)||P Value||OR||(95% CI)||P Value|
The current study evaluated the association between hs‐CRP levels, adiposity indices and cardiometabolic abnormalities in the preschool and primary school children participating in the European IDEFICS cohort. The longitudinal evaluation allowed the current study to focus on the predictive value of baseline hs‐CRP levels on the changes in anthropometric parameters and on the development of overweight/obesity over the 2‐year follow‐up. Cross‐sectional analyses revealed that boys and girls with higher plasma concentrations of hs‐CRP at baseline present (1) higher BMI z‐scores and central fat deposition; (2) higher prevalence of overweight/obesity; (3) higher BMI‐adjusted BP and lower BMI‐adjusted HDL‐cholesterol levels, with a significantly worse ratio between total and HDL‐cholesterol in girls. Longitudinal analyses showed that higher plasma concentrations of hs‐CRP at baseline predict (1) a higher increase of BMI z‐scores and central adiposity indexes over the 2‐year follow‐up, and (2) a significantly higher risk of incident overweight/obesity in both boys and girls.
Our data showed a strong correlation between adiposity indices and hs‐CRP in children. This correlation was similar to other studies where plasma hs‐CRP levels were positively associated with adiposity in adults 24, 25, 26 and children.8, 9, 10, 27 In our study, cross‐sectional analyses also showed that higher hs‐CRP values are associated with cardiovascular risk factors, like higher BP and lower HDL‐cholesterol levels. Previous studies already suggested that high CRP levels in children are associated with cardiovascular risk factors including HDL‐cholesterol and BP,8, 28, 29 although in some studies the association did not survive after adjustment for body mass.29, 30 In our study, the low‐grade proinflammatory status consistent with higher hs‐CRP levels is associated to a worse atherogenic profile already in preschool and school children, independently from body mass. The finding is of interest because the low‐grade inflammation has been found to predict the development of CVD 31 and type 2 diabetes.6
With regard to insulin resistance, Shea et al32 and Moran et al33 provided evidence of an association of CRP levels with insulin resistance markers in children, although no longer significant after adjustment for adiposity in the study by Moran.33 In our study we did not observe differences in insulin resistance measures between the different groups of hs‐CRP, probably due to the fact that our cohort is on average too young for detecting some metabolic derangements that might occur more likely later with age.34, 35 In fact, the International Diabetes Federation and the American Heart Association have both proposed a definition of metabolic syndrome in children and adolescents.34, 35 In both guidelines, the cut‐offs for metabolic and BP variables were defined only for children above 10 years of age, due to the high variability of these variables below this age. It is thus possible that the association would be more evident during adolescence rather than childhood. Long‐term prospective studies, as the ongoing follow‐up of the IDEFICS cohort, are needed to answer this question.
A strength of the present study is the use of precisely standardized phenotypic measurements within the 8 European countries participating in the survey. In fact, all measurements were conducted according to detailed standard operation procedures. In particular, subsamples of study subjects were examined repeatedly to calculate the inter‐ and intraobserver reliability of anthropometric measurements.14
The primary limitation is the use of a single hs‐CRP measurement. In adults, repeat hs‐CRP measurements are recommended, as concentrations are affected by factors such as recent infection and inflammatory conditions. The measurement of other inflammatory biomarkers could have provided further insights on the early association between inflammation and body fat accumulation. Other tests for active infection, for example cytokines or white blood cell count, were not performed, although the exclusion from the analysis of children who took medications during the week before the blood drawing may be considered as an adequate proxy. The assessment of the parental smoking and the consequent secondary smoking exposure of children would have been of interest, but unfortunately these questions were not included in the IDEFICS questionnaire. A limitation of the longitudinal analysis is the lower number of children attending the follow‐up examination and the relatively short duration of the follow‐up. However, the longitudinal association of basal hs‐CRP values with adiposity indices and incidence of overweight/obesity clearly emerged even after the relatively short follow‐up period and notwithstanding the expected reduction of the sample size, mainly due to the burden of the extensive examination protocol. The comparability of the core measurements between the children retained and those lost at follow‐up makes us confident that selection bias did not affect the longitudinal results. The planned follow‐up of the IDEFICS cohort in subsequent years will allow further evaluation of the association of hs‐CRP with body mass and metabolic variables during the transition from childhood to adolescence. Finally, the possibility that the intervention program conducted in half of the population could have marginally affected the relationship between the baseline hs‐CRP values and the phenotypes of interest over the follow‐up cannot be excluded. This possibility was taken into account by adjusting for the control/intervention group in the longitudinal analysis. In summary, we confirmed the association between hs‐CRP and higher body mass and overweight/obesity risk in a large population of European children. We also showed that hs‐CRP levels are associated to a worse cardiovascular risk profile. The main and novel finding of the study is that the children with higher baseline level of hs‐CRP had a greater increase in body mass and central adiposity over time and therefore were at higher risk of developing overweight/obesity during growth, as compared with children less exposed to low‐grade inflammation.
We thank the IDEFICS children for their participation in this extensive examination and all the centers that provided data.
Sources of Funding
We gratefully acknowledge the financial support of the European Community within the Sixth RTD Framework Programme Contract No. 0 16181 (FOOD) and the grant support from the EU for the IDEFICS study. The information in this document reflects the author's view and is provided as it is. Dr Sioen is financially supported by the Research Foundation ‐ Flanders (Grant no: 1.2.683.11.N.00).
- 1 Gregor MF, Hotamisligil GS. Inflammatory mechanisms in obesity. Annu Rev Immunol. 2011; 29:415–445.Google Scholar
- 2 Marnell L, Mold C, Du Clos TW. C‐reactive protein: ligands, receptors and role in inflammation. Clin Immunol. 2005; 117:104–111.Google Scholar
- 3 Roberts WL. CDC/AHA workshop on markers of inflammation and cardiovascular disease: application to clinical and public health practice: laboratory tests available to assess inflammation—performance and standardization: a background paper. Circulation. 2004; 110:e572–e576.Google Scholar
- 4 Danesh J, Collins R, Appleby P, Peto R. Association of fibrinogen, c‐reactive protein, albumin, or leukocyte count with coronary heart disease: meta‐analyses of prospective studies. JAMA. 1998; 279:1477–1482.Google Scholar
- 5 Pai JK, Pischon T, Ma J, Manson JE, Hankinson SE, Joshipura K, Curhan GC, Rifai N, Cannuscio CC, Stampfer MJ, Rimm EB. Inflammatory markers and the risk of coronary heart disease in men and women. N Engl J Med. 2004; 351:2599–2610.Google Scholar
- 6 Pradhan AD, Manson JE, Rifai N, Buring JE, Ridker PM. C‐reactive protein, interleukin 6, and risk of developing type 2 diabetes mellitus. JAMA. 2001; 286:327–334.Google Scholar
- 7 Tzoulaki I, Murray GD, Lee AJ, Rumley A, Lowe GD, Fowkes FG. C‐reactive protein, interleukin‐6, and soluble adhesion molecules as predictors of progressive peripheral atherosclerosis in the general population: Edinburgh Artery Study. Circulation. 2005; 112:976–983.Google Scholar
- 8 Cook DG, Mendall MA, Whincup PH, Carey IM, Ballam L, Morris JE, Miller GJ, Strachan DP. C‐reactive protein concentration in children: relationship to adiposity and other cardiovascular risk factors. Atherosclerosis. 2000; 149:139–150.Google Scholar
- 9 Visser M, Bouter LM, McQuillan GM, Wener MH, Harris TB. Low‐grade systemic inflammation in overweight children. Pediatrics. 2001; 107:E13.Google Scholar
- 10 Ford ES. C‐reactive protein concentration and cardiovascular disease risk factors in children: findings from the National Health and Nutrition Examination Survey 1999–2000. Circulation. 2003; 108:1053–1058.Google Scholar
- 11 Tam CS, Clément K, Baur LA, Tordjman J. Obesity and low‐grade inflammation: a paediatric perspective. Obes Rev. 2010; 11:118–126.Google Scholar
- 12 Ahrens W, Bammann K, Siani A, Buchecker K, De Henauw S, Iacoviello L, Hebestreit A, Krogh V, Lissner L, Mårild S, Molnár D, Moreno LA, Pitsiladis YP, Reisch L, Tornaritis M, Veidebaum T, Pigeot I; IDEFICS Consortium . The IDEFICS cohort: design, characteristics and participation in the baseline survey. Int J Obes (Lond). 2011; 35:S3–S15.Google Scholar
- 13 De Henauw S, Verbestel V, Mårild S, Barba G, Bammann K, Eiben G, Hebestreit A, Iacoviello L, Gallois K, Konstabel K, Kovács E, Lissner L, Maes L, Molnár D, Moreno LA, Reisch L, Siani A, Tornaritis M, Williams G, Ahrens W, De Bourdeaudhuij I, Pigeot I; IDEFICS Consortium . The IDEFICS community‐ oriented intervention programme: a new model for childhood obesity prevention in Europe?Int J Obes (Lond). 2011; 35:S16–S23.Google Scholar
- 14 Stomfai S, Ahrens W, Bammann K, Kovács E, Mårild S, Michels N, Moreno LA, Pohlabeln H, Siani A, Tornaritis M, Veidebaum T, Molnár D; IDEFICS Consortium . Intra‐ and inter‐observer reliability in anthropometric measurements in children. Int J Obes (Lond). 2011; 35:S45–S51.Google Scholar
- 15 Cole TJ, Freeman JV, Preece MA. British 1990 growth reference centiles for weight, height, body mass index and head circumference fitted by maximum penalized likelihood. Stat Med. 1998; 17:407–429.Google Scholar
- 16 Cole TJ, Bellizzi MC, Flegal KM, Dietz WH. Establishing a standard definition for child overweight and obesity worldwide: international survey. BMJ. 2000; 320:1–6.Google Scholar
- 17 Cole TJ, Flegal KM, Nicholls D, Jackson AA. Body mass index cut offs to define thinness in children and adolescents: international survey. BMJ. 2007; 335:194.Google Scholar
- 18 Peplies J, Fraterman A, Scott R, Russo P, Bammann K. Quality management for the collection of biological samples in multicentre studies. Eur J Epidemiol. 2010; 25:607–617.Google Scholar
- 19 Peplies J, Günther K, Bammann K, Fraterman A, Russo P, Veidebaum T, Tornaritis M, Vanaelst B, Mårild S, Molnár D, Moreno LA, Ahrens W; IDEFICS Consortium . Influence of sample collection and preanalytical sample processing on the analyses of biological markers in the European multicentre study IDEFICS. Int J Obes (Lond). 2011; 35:S104–S112.Google Scholar
- 20 Shephard MD, Mazzachi BC, Shephard AK. Comparative performance of two point‐of‐care analysers for lipid testing. Clin Lab. 2007; 53:561–566.Google Scholar
- 21 Friedewald WT, Levy RI, Fredrickson DS. Estimation of the concentration of low‐density lipoprotein cholesterol in plasma, without use of the preparative ultracentrifuge. Clin Chem. 1972; 18:499–502.Google Scholar
- 22 Matthews DR, Hosker JP, Rudenski AS, Naylor BA, Treacher DF, Turner RC. Homeostasis model assessment: insulin resistance and beta‐cell function from fasting plasma glucose and insulin concentrations in man. Diabetologia. 1985; 28:412–419.Google Scholar
- 23 ISCEDInternational Standard Classification of Education ISCED 1997. Montreal: UNESCO Institute for Statistics; 2006. Available at: http://www.uis.unesco.org/TEMPLATE/pdf/isced/ISCED_A.pdf. Accessed March 14, 2011.Google Scholar
- 24 Rohde LEP, Hennekens CH, Ridker PM. Survey of C‐reactive protein and cardiovascular risk factors in apparently healthy men. Am J Cardiol. 1999; 84:1018–1022.Google Scholar
- 25 Visser M, Bouter LM, McQuillan GM, Wener MH, Harris TB. Elevated C‐reactive protein levels in overweight and obese adults. JAMA. 1999; 282:2131–2135.Google Scholar
- 26 Ford ES. Body mass index, diabetes, and C‐reactive protein among U.S. adults. Diabetes Care. 1999; 22:1971–1977.Google Scholar
- 27 Ford ES, Galuska DA, Gillespie C, Will JC, Giles WH, Dietz WH. C‐reactive protein and body mass index in children: findings from the Third National Health and Nutrition Examination Survey, 1988–1994. J Pediatr. 2001; 138:486–492.Google Scholar
- 28 Caballero AE, Bousquet‐Santos K, Robles‐Osorio L, Montagnani V, Soodini G, Porramatikul S, Hamdy O, Nobrega AC, Horton ES. Overweight Latino children and adolescents have marked endothelial dysfunction and subclinical vascular inflammation in association with excess body fat and insulin resistance. Diabetes Care. 2008; 31:576–582.Google Scholar
- 29 López‐Jaramillo P, Herrera E, Garcia RG, Camacho PA, Castillo VR. Inter‐relationships between body mass index, C‐reactive protein and blood pressure in a Hispanic pediatric population. Am J Hypertens. 2008; 21:527–532.Google Scholar
- 30 Lambert M, Delvin EE, Paradis G, O'Loughlin J, Hanley JA, Levy E. C‐reactive protein and features of the metabolic syndrome in a population‐based sample of children and adolescents. Clin Chem. 2004; 50:1762–1768.Google Scholar
- 31 Emerging Risk Factors Collaboration ; Kaptoge S, Di Angelantonio E, Pennells L, Wood AM, White IR, Gao P, Walker M, Thompson A, Sarwar N, Caslake M, Butterworth AS, Amouyel P, Assmann G, Bakker SJ, Barr EL, Barrett‐Connor E, Benjamin EJ, Björkelund C, Brenner H, Brunner E, Clarke R, Cooper JA, Cremer P, Cushman M, Dagenais GR, D'Agostino RB, Dankner R, Davey‐Smith G, Deeg D, Dekker JM, Engström G, Folsom AR, Fowkes FG, Gallacher J, Gaziano JM, Giampaoli S, Gillum RF, Hofman A, Howard BV, Ingelsson E, Iso H, Jørgensen T, Kiechl S, Kitamura A, Kiyohara Y, Koenig W, Kromhout D, Kuller LH, Lawlor DA, Meade TW, Nissinen A, Nordestgaard BG, Onat A, Panagiotakos DB, Psaty BM, Rodriguez B, Rosengren A, Salomaa V, Kauhanen J, Salonen JT, Shaffer JA, Shea S, Ford I, Stehouwer CD, Strandberg TE, Tipping RW, Tosetto A, Wassertheil‐Smoller S, Wennberg P, Westendorp RG, Whincup PH, Wilhelmsen L, Woodward M, Lowe GD, Wareham NJ, Khaw KT, Sattar N, Packard CJ, Gudnason V, Ridker PM, Pepys MB, Thompson SG, Danesh J. C‐reactive protein, fibrinogen, and cardiovascular disease prediction. N Engl J Med2012; 367:1310–1320.Google Scholar
- 32 Shea S, Aymong E, Zybert P, Shamoon H, Tracy RP, Deckelbaum RJ, Basch CE. Obesity, fasting plasma insulin, and c‐reactive protein levels in healthy children. Obes Res. 2003; 11:95–103.Google Scholar
- 33 Moran A, Steffen LM, Jacobs DR, Steinberger J, Pankow JS, Hong CP, Tracy RP, Sinaiko AR. Relation of c‐reactive protein to insulin resistance and cardiovascular risk factors in youth. Diabetes Care. 2005; 28:1763–1768.Google Scholar
- 34 Zimmet P, Alberti KG, Kaufman F, Tajima N, Silink M, Arslanian S, Wong G, Bennett P, Shaw J, Caprio S; IDF Consensus Group . The metabolic syndrome in children and adolescents—an IDF consensus report. Pediatr Diabetes. 2007; 8:299–306.Google Scholar
- 35 Goodman E, Daniels SR, Meigs JB, Dolan LM. Instability in the diagnosis of metabolic syndrome in adolescents. Circulation. 2007; 115:2316–2322.Google Scholar