Adolescent Obesity and Its Risks: How to Screen and When to Refer

With a BMI at the 85th percentile, on initial assessment our patient might be perceived as being at moderate or even low risk for obesity and its associated comorbidities. However, a more careful review has elicited several additional risk factors suggesting more appropriate classification in the high-risk category. First, family history of type 2 diabetes on both sides of his family suggests a degree of genetic predisposition. Second, Hispanic ethnicity is known to be independently associated with insulin resistance, type 2 diabetes, and NAFLD [26]. Moreover, physical exam findings of an elevated waist circumference (90th percentile for age and ethnicity [27]) and acanthosis nigricans are also supportive of insulin resistance. As a result, despite having a BMI at the 85th percentile, this adolescent is at high risk and further evaluation is warranted based on both Expert Committee and ADA guidelines. Detailed discussion of certain risk factors is outlined below.

Pattern of Adipose Tissue Distribution: Utility of BMI and Waist Circumference

BMI is a clinical tool that serves as a surrogate marker of adiposity, but since it does not directly measure body fat it provides a statistical, rather than inherent, description of risk. While it is a relatively specific marker (~95%) with moderately high sensitivity and positive predictive value (~70–80%) at BMI levels > 95th percentile, sensitivity and positive predictive value decrease substantially at lower BMI percentiles (PPV 18% in a sample of overweight children) [28]. Current CDC BMI percentile charts consider age and gender differences but do not take into account sexual maturation level or race/ethnicity, both of which are independently correlated with BMI [29]. That is, children with similar BMIs of the same age and sex may exhibit varying degrees of adiposity and risk attributable to their pubertal stage and/or ethnicity [30]. For example, many studies have demonstrated that at the same BMI percentile, Asian Americans tend to have more adiposity compared with non-Hispanic whites [31], whereas African Americans tend to have more fat-free mass compared with non-Hispanic whites [32]. As a result of these differences, some advocate for adjusting cut-offs for BMI based on ethnicity and/or utilizing alternative measures of adiposity such as waist circumference or waist to hip ratio. However, in order for these latter methods to be useful, standardized methods of measurement and normative reference values must be developed. In summary, though BMI can be a useful screening tool, it is an indirect measure of adiposity and cannot discern adipose distribution. Therefore, it is important to remember that when used alone, BMI may overlook children with high inherent risk for disease.

Abdominal adiposity is associated with increased metabolic risk, including insulin resistance, type 2 diabetes, hypertension, cardiovascular disease, and mortality [33]. Waist circumference, a marker of abdominal/truncal obesity, has been considered as a potential marker in place of or in combination with BMI to identify children with increased metabolic risk. In adults, it is well established that an elevated waist circumference is associated with increased health risk, even among those within a normal-weight BMI category [34], and it is recommended that waist circumference in addition to BMI be used to assess health risk [35]. Many studies have documented similar associations between increased waist circumference and metabolic risk factors in childhood and adolescence [36–38]. Specifically, waist circumference is an independent predictor of both insulin sensitivity and increased visceral adiposity tissue (VAT) in children and adolescents [39]. Waist circumference can provide valuable information beyond BMI alone and may be beneficial in the clinical setting in identifying adolescents at risk for obesity-associated comorbidities.