The growing worldwide prevalence of type 2 diabetes mellitus in the young, as underlined by an earlier International Diabetes Federation (IDF) Consensus Statement 1, has highlighted a significant shortfall of data on the epidemiology of the disorder and the identification and treatment of children and adolescents at risk of progression to this disease. Urbanization, unhealthy diets, and increasingly sedentary lifestyles have contributed to increase the prevalence of childhood obesity, particularly in developing countries 2. Current treatment initiatives include school-based programs addressing physical activity and diet, which have been conducted with mixed success in reducing adiposity. There are limited safety data supporting the use of drugs for the treatment of obesity and related conditions such as type 2 diabetes in children and adolescents, and non-compliance in this population suggests that pharmacotherapy is unlikely to be effective long term 1. Although criteria have now been developed for bariatric surgery in teenagers 3, there are few evidence-based data available to support the increasing use of this modality in adolescents. Governments and society in general must be made more aware of the problems associated with obesity and the likelihood of progression to the metabolic syndrome in children and adolescents. Obesity, particularly in the central (abdominal) region, has been determined as a key factor in the etiology of type 2 diabetes 2. The prediction of health risks associated with obesity in youth is improved by the additional inclusion of waist circumference (WC) measure to the body mass index (BMI) percentile 4, 5. Such observations reinforce the importance of including WC in the assessment of childhood obesity to identify those at increased metabolic risk as a result of excess abdominal fat 5. The role of obesity can clearly be demonstrated in Japan, where a parallel increase in type 2 diabetes and obesity in children has occurred over the past few decades 6. Central (abdominal) obesity is also a key component in the IDF definition of metabolic syndrome in adults 2. The link between obesity, metabolic syndrome, and type 2 diabetes has already been characterized in adult populations 2. At present, 50–80% of almost 250 million adults worldwide with diabetes 7 are at risk of death from cardiovascular disease. Those with the metabolic syndrome are also at increased risk being twice as likely to die from, and three times as likely to have, cardiovascular complications as compared with those without the syndrome 8, 9. In addition, adults with the metabolic syndrome have a fivefold greater risk of developing type 2 diabetes 10. Already, one-quarter of the world’s adult population have metabolic syndrome 11, 12, and this condition is appearing with increasing frequency in children and adolescents, driven by the growing obesity epidemic in this young population 13-15. In 2004, the World Health Organization (WHO) reported that an estimated 22 million children younger than 5 yr of age and 10% of school-aged children, between 5 and 17 yr, were overweight or obese 16. WHO predicts that the prevalence of childhood obesity in developed and developing countries will continue to increase as has been seen in recent years. For example, from 1985 to 1997, in young Australians, the prevalence of overweight and obesity combined doubled and that of obesity trebled 17. In Thailand, the prevalence of obesity in those aged 5–12 yr increased from 12.2 to 15.6% in just 2 yr 18. In 2003–2004, 17.1% of children aged 2–19 yr in the USA were obese 19. Obesity is associated with an increase in cardiovascular risk factors (also indicators of metabolic syndrome) 20, and the persistence of these indicators from childhood and adolescence to young adulthood has been shown in several studies, including the Quebec Family Study 21, 22. Recently, the IDF released its guidelines for defining and diagnosing the metabolic syndrome in adults 2. The intention was to rationalize the existing multiple definitions of the syndrome and to avoid the confusion that arose as a result of conflicting opinions on the value of each set of criteria. The use of a single unified definition makes it possible to estimate the global prevalence of metabolic syndrome and make valid comparisons between nations. However, to date, there has not been a unified definition that can be used to assess risk in children and adolescents, and existing adult-based definitions of the metabolic syndrome may not be appropriate to address the problem in this age group. A study of adolescents using modified National Cholesterol Education Program (NCEP) [Adult Treatment Panel III (ATP III)] criteria 23 identified that 12% of the study group had the metabolic syndrome 24. When the ≥95th percentile of BMI was used as a cutoff point in the same study group, 31.3% were identified as having the syndrome, more than double of those previously found to be at risk. Duncan et al. 25 studied 991 adolescents (aged 12–19 yr) from National Health and Nutrition Examination Study (NHANES) 1999–2000 and used the ATP III definition modified for age. The overall prevalence of a metabolic syndrome phenotype among US adolescents increased from 4.2% in NHANES III (1988–1992) to 6.4% in NHANES 1999–2000. Based on population-weighted estimates, they estimated that more than 2 million US adolescents currently have a metabolic syndrome phenotype. In a population-based study of a Canadian Qji-Cree community involving 236 children aged 10–19 yr, Retnakaran et al. reported that 18.6% of the children met the criteria for the metabolic syndrome based on a pediatric metabolic syndrome definition based on the ATP III definition, and they used the ATP III definition modified for age and gender 26. Goodman et al. reported on a school-based, cross-sectional study of 1513 black, white, and Hispanic teenagers 27. Overall, the prevalence of ATP III-defined metabolic syndrome was 4.2% and that of the WHO-defined metabolic syndrome was 8.4%. The metabolic syndrome was found almost exclusively among obese teenagers in whom prevalence of the ATP III-defined metabolic syndrome was 19.5% and prevalence of WHO-defined metabolic syndrome 28 was 38.9%. No race or sex differences were present for ATP III definition. However, non-white teenagers were more likely to have metabolic syndrome by WHO criteria, and it was more common among girls if the WHO definition was used. Chi et al. have recently undertaken a literature review on definitions of the metabolic syndrome in children and adolescents published in the past decade 29. They noted that the prevalence of metabolic syndrome in pre-adolescent girls varies widely because of disagreement among proposed definitions of metabolic syndrome in pediatrics. They called for a consensus definition for the metabolic syndrome in children, which would allow researchers to make better temporal, biological, environmental, and social comparisons between data sets. The American College of Endocrinology definition 30 is not ideal in pediatric subjects as WC is rarely measured in children, and nomograms have only recently become available 31 for some ethnic groups but are not available for all. A recent paper has suggested yet another set of criteria with age- and gender-specific cutoff points 32. The variety of cutoff points used for the different components in this paper underlines the need for a single consistent definition with easily measurable components. Therefore, to date, no formal definition for the diagnosis of the metabolic syndrome in children and adolescents has been developed. The rapid increase in obesity highlights the urgency for a definition that could be used to further understand who is at high risk and to distinguish them from those with ‘simple’ uncomplicated obesity. The metabolic syndrome in adults is defined as a cluster of cardiovascular and diabetes risk factors including abdominal obesity, dyslipidemia, glucose intolerance, and hypertension 2. While the danger associated with clustering of components of the metabolic syndrome has been demonstrated in adults, where the presence of three or more components significantly increases the risk for coronary heart disease death/non-fatal myocardial infarction and the onset of new diabetes 33, few, if any, outcome data in children exist. While one definition, although with gender- and ethnicity-specific cutoff points, is suitable for use in the at-risk adult population 2, transposing a single definition to children and adolescents is problematic. Blood pressure, lipid levels, and anthropometric variables change with age and pubertal development. Puberty impacts on fat distribution and is known to cause a decrease both in insulin sensitivity, of approximately 30% with a complementary increase in insulin secretion 34, and in adiponectin levels 35. Therefore, single cutoff points cannot be used to define abnormalities in children. Instead, values above the 90th, 95th, or 97th percentile for gender and age are used. However, there has not been universal agreement as to which level to use for the criteria for the metabolic syndrome. The importance of the early identification of children at risk of developing the metabolic syndrome and subsequently progressing to type 2 diabetes and cardiovascular disease in later life must not be underestimated. From birth and before, circumstances can predispose a child to conditions such as obesity or dysglycemia. The presence of maternal gestational diabetes 36, low birth weight 37, infant feeding practices 38, early adiposity rebound 39, and genetic factors may all contribute to a child’s future level of risk. Being raised in an ‘obesogenic’ environment can also have a strong impact, as can the influence of socioeconomic factors 40, with weight gain often being observed as a positive correlate to affluence in developing countries. Longitudinal outcome studies and further research on the progression and etiology of the metabolic syndrome are urgently required to ascertain the long-term outcomes of abdominal obesity and clustering of the components of metabolic syndrome in at-risk children and to help improve future definitions of the syndrome. This new IDF definition of metabolic syndrome in children and adolescents was developed during a consensus workshop that brought together experts in the field of the metabolic syndrome and pediatrics. The purpose of the new definition of metabolic syndrome in children and adolescents is to expand on the IDF recommendations for managing type 2 diabetes in the young 1 and to provide a useful and unified tool for identifying those at risk. A clinically accessible diagnostic tool, avoiding measurements that may only be available in research settings, is needed to identify the metabolic syndrome in children and adolescents globally. This need has prompted the IDF to develop a definition that has used the limited data available from existing studies in youth. As with the adult criteria, we look on these new criteria as a starting point. As new information emerges, they can be modified. Inspired, in part, by the IDF worldwide definition of metabolic syndrome in adults 2, this new definition builds on previous studies investigating the prevalence of metabolic syndrome in children and adolescents, which have used modified adult criteria with varying cutoff points 12-14, 41, 42 (Table 1). The wide variety of cutoff points used has emphasized the need for a single consistent set of criteria, which is easily measurable and can be used as the basis for future work 29. Because of the developmental challenges presented by the age-related differences in children and adolescents, the new IDF definition of metabolic syndrome has been divided according to the following age groups: 6 to <10, 10 to <16, and ≥16 yr (Table 2). In all the three age groups, abdominal obesity is the ‘sine qua non’. We suggest that below the age of 10 yr, the metabolic syndrome as an entity is not diagnosed, although a strong message for weight reduction will be made for these children. At the age of 10 yr and more, a diagnosis of metabolic syndrome can be made. It requires the presence of abdominal obesity plus the presence of two or more of the other components (elevated triglycerides, low high-density lipoprotein (HDL)-cholesterol, high blood pressure, and elevated plasma glucose). The IDF adult criteria 2 can be used for adolescents aged ≥16 yr, while a modified version of these criteria will be applied to those aged 10 to <16 yr (use 90th percentile cutoff point for waist and <40 mg/dL of HDL for both sexes). On the basis of emerging new data, these criteria may change in the future. In adults, insulin resistance and abdominal obesity are considered to be significant causative factors in the development of the metabolic syndrome 9, 43, 44. The link between obesity, insulin resistance, and the risk of developing the metabolic syndrome has also been described in children 22, 27. With measurement of insulin resistance considered to be impractical for clinical use, abdominal adiposity was positioned as the ‘sine qua non’ in the IDF definition of metabolic syndrome in adults 2 and is recognized to be an independent risk factor for the development of cardiovascular disease in adults 45. Abdominal obesity can be easily assessed using the simple measure of WC, which is known to correlate more strongly with visceral adipose tissue (VAT) than BMI in adults 46 and is a strong predictor of cardiovascular disease risk factors in children 47. The correlation between WC and VAT has also been more recently demonstrated in children 48, further strengthening the existing evidence that WC is an effective measure of abdominal obesity 49 in the youth population. In children and adolescents, a number of studies have demonstrated a similar link between childhood obesity and elevated cardiovascular risk in later life. The Bogalusa Heart study showed that childhood overweight is related to the development of adverse risk factors (BMI, lipids, insulin, diabetes mellitus, and blood pressure) in adulthood and is attributable to the strong persistence of weight status from childhood to adulthood 50. Of the overweight children in the Bogalusa Heart study (BMI ≥95th percentile), 77% remained obese in adulthood. Furthermore, the Muscatine study demonstrated that in young adults, excess weight was the earliest predictor of coronary artery calcification 51. The ATP III definition, applied to a cohort of individuals aged 12–19 yr (NHANES III), identified that 4% of those studied were found to have the metabolic syndrome, with 80% of those meeting the criteria of being overweight 13. Using a modified version of the ATP III definition, metabolic syndrome in adolescents has also been linked to high levels of C-reactive protein, a pro-inflammatory marker. Of the five components of metabolic syndrome, C-reactive protein was higher only among those with abdominal obesity 41. Waist circumference in children is an independent predictor of insulin resistance, lipid levels, and blood pressure 4, 52-54– all components of metabolic syndrome. Moreover, in obese youth with similar BMI, insulin sensitivity is lower in those with high VAT and high waist/hip ratio 53, 54. Furthermore, insulin sensitivity decreases and insulin levels increase with increasing WC percentiles 3. These data, combined with the unequivocal evidence of the dangers of abdominal obesity in adulthood, support the use of abdominal obesity as the ‘sine qua non’ for the diagnosis of metabolic syndrome in children and adolescents. Percentiles rather than absolute values of WC have been used in the new criteria to compensate for varying degrees of development and ethnicity in the youth population. WC percentile data are becoming increasingly available worldwide 31, 55-58. Children with a WC >90th percentile are more likely to have multiple risk factors than those with a WC below this level 59. Several studies attempting to estimate the prevalence of metabolic syndrome in children and adolescents have already used the 90th percentile as a cutoff point for WC 13, 14, 41. We have also chosen to use the 90th percentile as a cutoff point for WC based on this existing evidence and aim to reassess criteria and cutoff points in 5 yr and modify the guidelines, if necessary, based on the new outcome data. Previous studies investigating the metabolic syndrome in children and adolescents have used a range of cutoff points primarily based on ATP III criteria for categorizing additional components of the syndrome, i.e., triglycerides, HDL-cholesterol, blood pressure, and fasting glucose (Table 1) 12-14, 41, 42. Other definitive sources include the National High Blood Pressure Education Program, which recommends blood pressure cutoff points of >90th or >95th percentile adjusted for height, age, and gender to identify ‘high normal’ blood pressure or prehypertension and high blood pressure or hypertension in children and adolescents 60. Cutoff points for impaired fasting glucose have previously followed recommendations by the American Diabetes Association (ADA) [100–125 mg/dL (≥5.6–6.9 mmol/L)] 61 and the NCEP/ATP III in adults [≥110 mg/dL (6.1 mmol/L)] 23, although the latter has recently changed to the lower ADA recommended levels 62. Criteria for defining lipid (triglyceride and HDL-cholesterol) imbalances are even less consistent in the youth population, with recommendations by the NCEP/ATP III (age specific), NHANES III (age and gender specific), and the National Growth and Health Study (age, gender, and ethnic specific), employing either absolute value or percentile cutoff points. In view of this lack of consistency, we believe that use of the adult levels for the present is wise until further information is available. We recommend the following topics as priorities for future research: Develop a better understanding of the relationship between body fat and its distribution in children and adolescents, e.g., dual energy X-ray absorptiometry (DEXA), WC, BMI, and height and weight percentiles; a) Explore whether early growth patterns predict future adiposity and features of the metabolic syndrome, diabetes, and cardiovascular disease and b) explore whether low birth weight predicts future metabolic syndrome, diabetes, and cardiovascular disease; Perform factor analysis in children and adolescents to establish grouping of metabolic characteristics – adiposity, dyslipidemia, hyperinsulinemia, hypoadiponectinemia, and insulin resistance; Investigate how should obesity in children could be better defined, e.g., weight/height, WC etc.; Develop ethnic-specific normal ranges for WC, ideally based on healthy values; Perform ethnic-specific studies of WC etc. vs. abdominal (truncal) fat based on magnetic resonance imaging and DEXA; Support studies of adiponectin, leptin, etc. in children and adolescents to determine if they may be predictors of metabolic syndrome in adulthood; Initiate long-term studies of multi-ethnic cohorts followed into adulthood to determine the natural history and effectiveness of intervention strategies, particularly lifestyle. In conclusion, to combat any conflict that could arise from these multiple interpretations of the metabolic syndrome in children and adolescents, the IDF consensus group has aimed primarily at developing a simple, easy-to-apply definition to begin using in the clinical setting. In the absence of definitive research findings at this time, the proposed IDF definition of the metabolic syndrome in children and adolescents (Table 2) adheres to the absolute values presented in the adult definition 2, with the exception of WC. As described previously, until such time that outcome data from studies in children and adolescents indicate otherwise, WC percentiles are recommended for use. Early detection, followed by treatment in the form of lifestyle intervention and possibly pharmacotherapy, if its safety has been clearly demonstrated, is vital in halting the progression of this syndrome pathway in the adolescent population. It is likely that this will reduce morbidity and mortality in adulthood, as well as minimize the global socioeconomic burden of cardiovascular disease and type 2 diabetes. The workshop was sponsored by an unrestricted educational grant to the IDF Task Force on Epidemiology and Prevention from sanofi-aventis.