Utility of the modified ATP III defined metabolic syndrome and severe obesity as predictors of insulin resistance in overweight children and adolescents: a cross-sectional study

Sarita Dhuper¹ , Hillel W Cohen² , Josephine Daniel¹ , Padmasree Gumidyala¹ , Vipin Agarwalla¹ , Rosemarie St Victor¹ and Sunil Dhuper³

¹Department of Pediatrics, Brookdale University Hospital & Medical Center, Brooklyn, NY, USA

²Department of Epidemiology and Population Health, Albert Einstein College of Medicine, Bronx, NY, USA

³Department of Medicine, Coney Island Hospital, Brooklyn, New York, USA

author email corresponding author email

Cardiovascular Diabetology 2007, 6:4doi:10.1186/1475-2840-6-4


Published:	14 February 2007

Abstract

Background

The rising prevalence of obesity and metabolic syndrome (MetS) has received increased attention since both place individuals at risk for Type II diabetes and cardiovascular disease. Insulin resistance (IR) has been implicated in the pathogenesis of obesity and MetS in both children and adults and is a known independent cardiovascular risk factor. However measures of IR are not routinely performed in children while MetS or severe obesity when present, are considered as clinical markers for IR.

Objective

The study was undertaken to assess the utility of ATPIII defined metabolic syndrome (MetS) and severe obesity as predictors of insulin resistance (IR) in a group of 576 overweight children and adolescents attending a pediatric obesity clinic in Brooklyn.

Methods

Inclusion criteria were children ages 3–19, and body mass index > 95th percentile for age. MetS was defined using ATP III criteria, modified for age. IR was defined as upper tertile of homeostasis model assessment (HOMA) within 3 age groups (3–8, n = 122; 9–11, n = 164; 12–19, n = 290). Sensitivity, specificity, positive predictive values and odds ratios (OR) with 95% confidence intervals (CI) were calculated within age groups for predicting IR using MetS and severe obesity respectively.

Results

MetS was present in 45%, 48% and 42% of the respective age groups and significantly predicted IR only in the oldest group (OR = 2.0, 95% CI 1.2, 3.4; p = .006). Sensitivities were <55%; specificities <63% and positive predictive values ≤ 42% in all groups. Severe obesity was significantly associated with IR in both the 9–11 (p = .002) and 12–18 (p = .01) groups but positive predictive values were nonetheless ≤ 51% for all groups.

Conclusion

The expression of IR in overweight children and adolescents is heterogeneous and MetS or severe obesity may not be sufficiently sensitive and specific indicators of insulin resistance. In addition to screening for MetS in overweight children markers for IR should be routinely performed. Further research is needed to establish threshold values of insulin measures in overweight children who may be at greater associated risk of adverse outcomes whether or not MetS is present.