Cardiorespiratory Fitness Is Strongly Related to the Metabolic Syndrome in Adolescents

RESEARCH DESIGN AND METHODS—

The study sample consisted of 1,561 adolescents, aged 12–19 years, who participated in the National Health and Nutrition Examination Survey (NHANES) 1999–2002, which is a nationally representative cross-sectional survey (10). All participants had a complete fasting (≥12 h) blood sample and measures of waist circumference, blood pressure, and CRF.

Maximal aerobic capacity (Vo_2max) was the objective CRF measure obtained in NHANES. Vo_2max is a health-related component of physical fitness that relates to the ability of the circulatory and respiratory systems to supply oxygen during sustained physical activity (11). Vo_2max was estimated from heart rate measurements obtained during a multistage submaximal test where participants walked or ran on a treadmill to achieve 75–90% of their age-predicted maximal heart rate (12). Estimated Vo_2max was regressed on age within each sex, and the residuals were retained to represent age-adjusted values. The sample was then divided into low, moderate, and high CRF tertiles based on the age-adjusted Vo_2max values.

In adults, the metabolic syndrome criteria established by the National Cholesterol Education Program (NCEP) is consistently used in research and clinical settings (13). According to the NCEP, metabolic syndrome includes having three or more of the following: high waist circumference (≥102 cm in men and ≥88 cm in women), high plasma triglycerides (>1.69 mmol/l), low HDL cholesterol (<1.04 mmol/l in men and <1.29 mmol/l in women), high fasting plasma glucose (≥5.6 mmol/l), and high blood pressure (≥130/85 mmHg). There is no universally accepted metabolic syndrome criterion for adolescents. In this study, we used age- and sex-specific metabolic syndrome thresholds for adolescents that were linked to the NCEP adult thresholds using growth curve modeling (14). The growth curves for each metabolic syndrome component pass through the adult thresholds at 20 years such that adolescent thresholds are linked to the adult thresholds noted above. Thus, metabolic syndrome for adolescents was defined as having three or more of the five metabolic syndrome components using the adolescent thresholds.

Statistical analyses were performed using Stata Intercooled 7 (StataCorp) to account for the complex design and weighting of NHANES. The prevalence of metabolic syndrome was compared across CRF tertiles using χ² analyses. The odds of having metabolic syndrome and its component risk factors were compared across CRF tertiles using logistic regression analyses. Age, ethnicity (non-Hispanic white, non-Hispanic black, Hispanic, and other races including multiracial), poverty-to-income ratio (based on family income and size, divided into quartiles), smoking (never, former, or current), carbohydrate consumption (percentage of calories from carbohydrate, with <40% as low, 40–60% as moderate, and >60% as high), and fat consumption (percentage of calories from fat, with <30% as low, 30–40% as moderate, and >40% as high) were included as confounders in the logistic regression analyses.

RESULTS—

The prevalence of metabolic syndrome was 7.6% in the entire sample. The prevalence of metabolic syndrome decreased across low, moderate, and high CRF tertiles within both male (24.3, 5.0, and 0.1%, respectively) and female (17.3, 5.9, and 0.9%) subjects and also within the non-Hispanic white (30.0, 6.0, and 0.5%, respectively), non-Hispanic black (21.3, 5.0, and 0.5%), and Hispanic (14.4, 6.5, and 0.9%) ethnic groups.

In comparison with the low CRF tertile (referent group, odds ratio [OR] 1.00), the likelihood of having any of the individual metabolic syndrome components was significantly lower in the high CRF tertile (P < 0.05). The ORs (95% CIs) for metabolic syndrome components in the high CRF tertile were as follows: 0.02 (0.01–0.05) for high waist circumference, 0.18 (0.11–0.28) for low HDL cholesterol, 0.24 (0.13–0.42) for high triglycerides, 0.51 (0.27–0.96) for high plasma glucose, and 0.17 (0.09–0.32) for high blood pressure. As shown in Table 1, the ORs for metabolic syndrome decreased in a gradient manner when moving from low to moderate to high CRF tertiles. These results were consistent within all sex and ethnic groups (Table 1). Given the small number of subjects (n = 72), analyses were not performed within the “other” race group.

CONCLUSIONS—

CRF was a strong, independent predictor of metabolic syndrome and its component risk factors in this sample of 12–19 year olds. Interestingly, the prevalence of metabolic syndrome in the lowest adolescent CRF tertile was comparable with the overall prevalence of metabolic syndrome in U.S. adults, (15) while essentially none of the adolescents with a high CRF had the metabolic syndrome. The results were consistent by sex and ethnicity.

The strengths of this study include the large and representative study sample (10), the use of age-appropriate adolescent metabolic syndrome criteria that are linked to health-based adult criteria (14), and the use of an objective CRF measure, which is largely determined by physical activity participation in recent weeks and months (11). The use of strong measures may help explain the strength of the relations observed in this study. Most previous studies in this topic have relied on questionnaire measures of physical activity, the imprecision of which may have biased the results toward a null effect.

The findings presented in this report provide compelling evidence that moderate and high CRF levels are associated with a significantly lower metabolic syndrome prevalence in adolescents. Therefore, interventions aimed at combating metabolic syndrome in this age-group may want to focus on improving CRF.