Key points

Excessive adiposity is increasing on both the national and international levels. Adults, adolescents, and children are all affected by the epidemic. Obesity in the child and adolescent population is a growing problem. Rates of obesity have risen dramatically in a short period of time. Between 1999 and 2000, the prevalence of US teenagers aged 12 to 19 years who were identified as overweight or obese was 30% ± 1.4%, and in 2003 to 2004 it increased significantly to 34.3% ± 2.6%. Similar increases were seen in children, with the most dramatic increase seen in the 6-year-old to 11-year-old cohort. According to the 2013 Heart Disease and Stroke Statistics Update, 23.9 million (31.8%) children aged 2 to 19 years are overweight or obese and 12.7 million (16.9%) are obese.

The primary goal of labeling adolescents as overweight or obese should be to identify a population that is at an increased risk for current or future disease related to their excess adiposity. Obesity has a strong association with cardiovascular disease and, specifically, accelerated atherosclerosis. As a result, it is of the utmost importance that clinicians identify and risk stratify overweight and obese individuals in order to institute primary prevention and/or treatment to reduce future cardiovascular morbidity and mortality in this vulnerable population.

This article (1) discusses the definition of child and adolescent obesity and the need for standardization, (2) discusses anthropometric measurements and the potential pitfalls and benefits of using additional measures when assessing the overweight and/or obese child, (3) reviews data to advocate for ethnic-specific cut-points and further research in minority populations, (4) reviews literature about the metabolic syndrome in children and current recommendations, (5) reviews literature concerning the cardiovascular effects of obesity, and (6) summarizes systematic reviews of obesity prevention studies. In conclusion, this article calls for a standardized definition of obesity and the measures of obesity in the child and adolescent; as well as increased awareness and screening of the cardiovascular complications of the obesity epidemic.

Definitions of adolescent obesity

One of the first problems to arise when considering obesity in the pediatric population is how to define and identify it. The 2007 obesity guidelines by the American Academy of Pediatrics recommends using body mass index (BMI), a measure of body weight relative to height. Unlike in adults, in whom absolute BMI cutoff points are used to define obesity, this article recommends the use of percentiles specific for age and gender to categorize children as underweight, normal weight, overweight, or obese. The expert panel advocates the use of 2 specific cutoff points to minimize overdiagnosis and prevent underdiagnosis, using the 85th and 95th percentiles for age and gender. If patients are between the 85th and 94th percentiles, they should be categorized as overweight, and if they are greater than or equal to the 95th percentile they are categorized as obese. This approach represents a change in terminology, but not in cutoff points, from the 1998 expert committee recommendations in which the term obese was avoided.

This recommended cutoff point is in agreement with the US Centers for Disease Control and Prevention (CDC) growth standards of 2000. However, in addition to this recommendation there are several other organizations with varying cutoff points. The World Health Organization (WHO) developed international standards for children 0 to 5 and 5 to 19 years of age, as did the International Obesity Task Force (IOTF). In addition, there are several country-specific references that are used in individual nations.

The lack of agreement about definitions and cutoff points has been attributed to the lack of strong evidence and the absence of a definite correlation between childhood weight and future health outcomes. Although there is clearly no perfect measure or cutoff point, several studies have compared the different growth curves and cutoff points of different organizations. The studies showed disagreement between growth curves and showed that usage of the WHO criteria yielded a higher prevalence of overweight/obesity. The most recent study was conducted in Spain and classified participants as obese, overweight, or normal weight based on CDC, WHO, and the Spanish Reference Criteria. Forty-eight percent were categorized as obese by the WHO criteria, 43% by the CDC criteria, and 16% by the Spanish Reference Criteria. The study also analyzed biochemical variables and vascular parameters in the participants. Of particular interest are the differences noted between the WHO and CDC criteria. Application of the WHO criteria showed that obese children had significantly higher insulin levels, homeostasis model assessment (HOMA) index, and vascular parameters (with the exception of intima media thickness), and lower high-density lipoprotein (HDL) levels than overweight children. Overweight children characterized by the WHO criteria had higher HOMA index and arterial compliance, and lower HDL levels compared with normal-weight individuals. The CDC criteria also identified significant differences between the obese participants and the normal-weight individuals; however, fewer biochemical differences and no vascular differences were found between the overweight and obese groups. These findings suggest that the WHO criteria may be more useful to identify individuals with biochemical and vascular derangements, and that these individuals may, as a result, be at higher risk for future poor health outcomes. Larger studies in a more heterogeneous population are needed to further validate these findings. A Canadian study showed similar findings when comparing the WHO, IOTF, and CDC cut-points with a 2004 prevalence of overweight/obesity of 35% using the WHO criteria, 26% with IOTF criteria, and 28% with CDC data.

A recent review of multiple systematic reviews and clinical guidelines performed in the United Kingdom concluded that, among the various indices available to gauge body fatness in children and adolescents, subjective clinical assessment is the worst measure, and BMI-for-age percentiles using national reference data is the best means to diagnose/define obesity in children and adolescents. The review also discusses the Cole -IOTF reference data, which were intended to define child and adolescent obesity for international comparisons of the prevalence of obesity rather than for the clinical diagnosis of overweight and obesity.

Overall, the literature seems to support the use of BMI for age using national reference data to identify overweight and obese children and adolescents. Consistency in technique of measurement and application of criteria is likely of the utmost importance in clinical practice. It seems that, despite the ease of application of BMI for age, many practitioners are reluctant to use this powerful clinical tool ( Table 1 ).

Definitions of adolescent obesity

One of the first problems to arise when considering obesity in the pediatric population is how to define and identify it. The 2007 obesity guidelines by the American Academy of Pediatrics recommends using body mass index (BMI), a measure of body weight relative to height. Unlike in adults, in whom absolute BMI cutoff points are used to define obesity, this article recommends the use of percentiles specific for age and gender to categorize children as underweight, normal weight, overweight, or obese. The expert panel advocates the use of 2 specific cutoff points to minimize overdiagnosis and prevent underdiagnosis, using the 85th and 95th percentiles for age and gender. If patients are between the 85th and 94th percentiles, they should be categorized as overweight, and if they are greater than or equal to the 95th percentile they are categorized as obese. This approach represents a change in terminology, but not in cutoff points, from the 1998 expert committee recommendations in which the term obese was avoided.

This recommended cutoff point is in agreement with the US Centers for Disease Control and Prevention (CDC) growth standards of 2000. However, in addition to this recommendation there are several other organizations with varying cutoff points. The World Health Organization (WHO) developed international standards for children 0 to 5 and 5 to 19 years of age, as did the International Obesity Task Force (IOTF). In addition, there are several country-specific references that are used in individual nations.

The lack of agreement about definitions and cutoff points has been attributed to the lack of strong evidence and the absence of a definite correlation between childhood weight and future health outcomes. Although there is clearly no perfect measure or cutoff point, several studies have compared the different growth curves and cutoff points of different organizations. The studies showed disagreement between growth curves and showed that usage of the WHO criteria yielded a higher prevalence of overweight/obesity. The most recent study was conducted in Spain and classified participants as obese, overweight, or normal weight based on CDC, WHO, and the Spanish Reference Criteria. Forty-eight percent were categorized as obese by the WHO criteria, 43% by the CDC criteria, and 16% by the Spanish Reference Criteria. The study also analyzed biochemical variables and vascular parameters in the participants. Of particular interest are the differences noted between the WHO and CDC criteria. Application of the WHO criteria showed that obese children had significantly higher insulin levels, homeostasis model assessment (HOMA) index, and vascular parameters (with the exception of intima media thickness), and lower high-density lipoprotein (HDL) levels than overweight children. Overweight children characterized by the WHO criteria had higher HOMA index and arterial compliance, and lower HDL levels compared with normal-weight individuals. The CDC criteria also identified significant differences between the obese participants and the normal-weight individuals; however, fewer biochemical differences and no vascular differences were found between the overweight and obese groups. These findings suggest that the WHO criteria may be more useful to identify individuals with biochemical and vascular derangements, and that these individuals may, as a result, be at higher risk for future poor health outcomes. Larger studies in a more heterogeneous population are needed to further validate these findings. A Canadian study showed similar findings when comparing the WHO, IOTF, and CDC cut-points with a 2004 prevalence of overweight/obesity of 35% using the WHO criteria, 26% with IOTF criteria, and 28% with CDC data.

A recent review of multiple systematic reviews and clinical guidelines performed in the United Kingdom concluded that, among the various indices available to gauge body fatness in children and adolescents, subjective clinical assessment is the worst measure, and BMI-for-age percentiles using national reference data is the best means to diagnose/define obesity in children and adolescents. The review also discusses the Cole -IOTF reference data, which were intended to define child and adolescent obesity for international comparisons of the prevalence of obesity rather than for the clinical diagnosis of overweight and obesity.

Overall, the literature seems to support the use of BMI for age using national reference data to identify overweight and obese children and adolescents. Consistency in technique of measurement and application of criteria is likely of the utmost importance in clinical practice. It seems that, despite the ease of application of BMI for age, many practitioners are reluctant to use this powerful clinical tool ( Table 1 ).

Anthropometric measures of overweight and obesity

Although BMI is the most thoroughly researched and advocated tool to identify children and adolescents who are overweight/obese, it is not the only tool. Other measures of excess adiposity are available and the benefits and drawbacks of these measures have been discussed in the literature. Waist circumference, skinfold thickness, hip circumference, waist/hip ratio, and waist/height ratio are examples of the more commonly used alternative measurements. In addition, measurements that can be derived from those listed earlier include arm muscle area, arm fat area, the Rohrer index, and the conicity index. More direct methods of measuring increased adiposity include dual-energy x-ray absorptiometry (DXA), densitometry, underwater weighing, computed tomography scan, and foot-to-foot bioelectrical impedance analysis; however, these methods are expensive and have not been found to be reliable in individual children.

Studies have noted that skinfold thicknesses may be more strongly associated with body fatness in children, although these studies were performed in specific ethnic groups. Other studies comparing the two have shown that skinfold measurement and BMI are both good measures of adiposity in adolescents. In a recent study when triceps and subscapular skinfold thicknesses and BMI were compared with the reference standard percent body fat by DXA in a large multiethnic population, girls with increased percent body fat by DXA were identified equally well by BMI and skinfold thickness. Boys were identified slightly more accurately by skinfold thickness, albeit not by a large margin. In contrast, skinfold thickness was superior to BMI when identifying children with low body fat. In addition, total fat mass was more strongly correlated with BMI than skinfold thickness. Skinfold thickness is an inexpensive and easy way to measure adiposity, but it is not as accurate or reproducible as other measurements. Although it is likely not of use in isolation, it may add some useful information when following a patient over time with several measurements once the operator has become familiar with the technique.

Waist circumference is an easy and inexpensive method that has been well researched in the adult population and has been shown to predict development of disease and to correlate strongly with body fat in adults. However, the measurement procedure has not been standardized and there are no validated reference data or substantial research in children. There are multiple ways to measure waist circumference including the midpoint between the lowest rib and the iliac crest, immediately below the lowest rib, at the umbilicus, at the narrowest and widest points, and immediately above the iliac crest. Adult studies have shown that the site of measurement does not influence the risk of cardiovascular disease mortality in adults. However, in pediatrics, single cutoff points are not reasonable because of the rapid somatic growth of children, so it may be more important to standardize the measurement as well as reference the measurement to a certain age and gender. A major obstacle in the use of waist circumference in the adolescent population is the lack of standardized cutoff points in the literature. A systematic review by de Moraes and colleagues analyzed the various cutoff points used for waist circumference to evaluate abdominal obesity in the adolescent population and there was no consensus in the literature. A recent study of a large population of multiethnic school-aged children in the United States showed that DXA fat mass was highly correlated with BMI (Spearman rank correlation coefficient [rs] = 0.83) and sum of skinfolds (rs = 0.9), and DXA truncal fat was highly correlated with waist circumference (rs = 0.79). Body fat distribution has been associated with obesity and not just total body fat. As a result, it has been suggested that combining BMI with an index of body fat distribution may increase the physician’s ability to identify children with cardiovascular and metabolic risk factors.

Because of the lack of universally accepted cutoff points for waist circumference, the use of waist/height ratio has been suggested, because it is independent of age and sex. In adults, a waist/height ratio greater than 0.5 has been associated with increased cardiovascular risk and the investigators propose using this cutoff in children and adolescents as well. Other studies have shown that although waist circumference and BMI have strong correlations with percent body fat and a high probability of identifying children and adolescents as obese and overweight, the waist/hip ratio was not well correlated with percent body fat and was not significantly effective at identifying obese and overweight.

In conjunction with BMI, waist circumference may be the most valuable tool in the arsenal of anthropometric measurements because of its ease of performance; however, a standardized method and cutoff points are necessary to make it a more powerful and effective screening tool. Cook and colleagues proposed a series of growth curves for waist circumference that may be useful for clinical application but need to be validated in future longitudinal studies.

Racial/ethnic differences in obesity

In addition to the controversies regarding the definition and standardization of measures of obesity, there are complex differences in rates of obesity and predisposition to obesity among various ethnicities. Many large, longitudinal studies have been performed in predominantly white cohorts. As a result, the current recommendations and data are most representative of the white population and therefore may not consistently identify at-risk youths of other ethnicities. Different ethnicities seem to be prone to the metabolic derangements associated with obesity at different levels. In the most recent National Health and Nutrition Examination Survey (NHANES), from 2009 to 2010, the prevalence of obesity in children and adolescents was 21.2% for Hispanic subjects, 24.3% for non-Hispanic black subjects, and 14% for non-Hispanic white subjects. There was also a racial/ethnic-specific trend identified among non-Hispanic black male subjects with a significant increasing trend of obesity (odds ratio, 1.10; 95% confidence interval, 1.03–1.17). As a result, this cohort needs to be monitored even more carefully. The EPOCH (Evaluating Processes of Care and the Outcomes of Children in Hospital) study of fat distribution, prevalence of obesity, and the metabolic syndrome among a diverse group of Colorado youth found similar results. Hispanic and African American youth had a higher prevalence of obesity and metabolic syndrome as well as more centralized fat distribution and larger abdominal subcutaneous fat deposition than their non-Hispanic white counterparts. In the CATCH (Child and Adolescent Trial for Cardiovascular Health) study, a large, multicenter, multiethnic, school-based intervention study, mean BMI and prevalence of overweight and obesity were higher for African American and Hispanic youth than for white youth. The investigators discussed the importance of the effect of socioeconomic status in the prevalence rates and cautioned that race-specific references are confounded by socioeconomic status.

It has been established in the adult population that people from south Asia have metabolic derangements at lower BMIs and anthropometric cut-points than their white counterparts. These differences have been attributed to differences in body fat distribution and composition. A recent study in children showed that fractional body fat content was significantly greater at any BMI among subjects from south Asia. In addition, ethnicity-specific differences in metabolic derangements were similar to those seen in adults.

These findings suggest that ethnicity-specific BMI and anthropometric cut-points are needed to properly identify high-risk minority children who may not be correctly identified by current standards. Longitudinal data in nonwhite Hispanic, Asian, and Native American populations are also needed to help establish ethnicity-specific recommendations in children.

Identification of the metabolic syndrome in childhood and adolescence

The metabolic syndrome in adults is a cluster of conditions consisting of at least 3 of the following: high waist circumference, increased systolic or diastolic blood pressure, increased serum fasting glucose and/or triglyceride levels, and a low HDL cholesterol level. In adults, the US National Cholesterol Education Program Adult Treatment Panel III and the International Diabetes Federation have established the 2 main sets of criteria and cutoffs for diagnosis.

However, the metabolic syndrome in the pediatric population has been more difficult to define. As with the definition of obesity, there is a lack of consensus regarding the definition of the metabolic syndrome in youth. A review by Ford and Li showed that 40 unique definitions of pediatric metabolic syndrome were used in 27 different publications. Cook and colleagues chose 4 definitions and applied them to a population of adolescents from NHANES from 1999 to 2002 to determine the prevalence of the metabolic syndrome based on definition. The prevalence of the metabolic syndrome varied from 2% to 9.4% of all teenagers in the United States and 12.4% to 44.2% in obese teenagers. The study also showed great variation between genders and ethnicities. Despite the lack of consensus, longitudinal studies have shown that pediatric metabolic syndrome predicts adult metabolic syndrome, diabetes mellitus (DM), and cardiovascular disease. It is important to determine which definition would capture the largest number of at-risk children. Because interventions are both time and resource intensive, it is also important to identify those who are at risk.

Several studies have made efforts to determine which components of the definition are most effective and yield the highest statistical results. Huang and colleagues found that the use of more stringent criteria yielded a higher specificity but sacrificed sensitivity, whereas the opposite was true when using only 1 component of the definition. Using multiple variables in childhood yielded a high positive predictive value and may be the most useful way to prognosticate future risk. Schubert and colleagues analyzed the sensitivity, specificity, and positive and negative predictive values of components of childhood metabolic syndrome for adult metabolic syndrome and type 2 DM using data from 3 longitudinal studies. Multiple components were better at predicting metabolic syndrome in adulthood, as was shown by Huang and colleagues. However, with regard to type 2 DM, the metabolic components were more effective at identifying children not at risk than identifying those who were at risk. The metabolic components seem to be able to screen out children who are not at risk in order to focus on those with the potential to develop disease. The EPOCH study showed that ethnic minorities might have an increased risk for early development of the metabolic syndrome compared with their non-Hispanic white counterparts, beyond their increased obesity risk.

In light of the absence of a consensus definition, the most recent recommendations by the National Lung, Heart, and Blood Institute are (1) to intensify therapy with an emphasis on lifestyle modification in the presence of any combination of multiple risk factors; (2) the presence of obesity should prompt specific evaluation of all other cardiovascular risk factors; and (3) in the setting of obesity with any other major risk factor, the clinician should initiate interventions such as intensive weight reduction and risk factor–specific interventions as well as prompt evaluation for DM, liver function abnormalities, left ventricular hypertrophy, and sleep apnea.

• •
  

  Pediatric metabolic syndrome predicts adult metabolic syndrome, DM, and cardiovascular disease

  
  
• •
  

  Presence of obesity should prompt specific evaluation of all other cardiovascular risk factors

Only gold members can continue reading. Log In or Register to continue

Share this:

• Click to share on Twitter (Opens in new window)
• Click to share on Facebook (Opens in new window)

Related

Related posts:

1. The Asymptomatic Teenager with an Abnormal Electrocardiogram
2. The Teenager with Palpitations
3. Cardiomyopathies Encountered Commonly in the Teenage Years and Their Presentation
4. A Pediatric Approach to Family History of Cardiovascular Disease

Stay updated, free articles. Join our Telegram channel

Join