Identifying adolescent metabolic syndrome using body mass index and waist circumference.

INTRODUCTION: Metabolic syndrome is increasing among adolescents. We examined the utility of body mass index (BMI) and waist circumference to identify metabolic syndrome in adolescent girls. METHODS: We conducted a cross-sectional analysis of 185 predominantly African American girls who were a median age of 14 years. Participants were designated as having metabolic syndrome if they met criteria for 3 of 5 variables: 1) high blood pressure, 2) low high-density lipoprotein cholesterol level, 3) high fasting blood glucose level, 4) high waist circumference, and 5) high triglyceride level. We predicted the likelihood of the presence of metabolic syndrome by using previously established cutpoints of BMI and waist circumference. We used stepwise regression analysis to determine whether anthropometric measurements significantly predicted metabolic syndrome. RESULTS: Of total participants, 18% met the criteria for metabolic syndrome. BMI for 118 (64%) participants was above the cutpoint. Of these participants, 25% met the criteria for metabolic syndrome, whereas only 4% of participants with a BMI below the cutpoint met the criteria for metabolic syndrome (P <.001). Girls with a BMI above the cutpoint were more likely than girls with a BMI below the cutpoint to have metabolic syndrome (P = .002). The waist circumference for 104 (56%) participants was above the cutpoint. Of these participants, 28% met the criteria for metabolic syndrome, whereas only 1% of participants with a waist circumference below the cutpoint met the criteria for metabolic syndrome (P <.001). Girls with a waist circumference above the cutpoint were more likely than girls with a waist circumference below the cutpoint to have metabolic syndrome (P = .002). Stepwise regression showed that only waist circumference significantly predicted metabolic syndrome. CONCLUSION: Both anthropometric measures were useful screening tools to identify metabolic syndrome. Waist circumference was a better predictor of metabolic syndrome than was BMI in our study sample of predominantly African American female adolescents living in an urban area.