ABSTRACT
BACKGROUND: To establish the rate of agreement in predicting metabolic syndrome (MS) in different pediatric classifications using percentiles or fixed cut-offs, as well as exploring the influence of cholesterol. SUBJECTS AND METHODS: Cross-sectional study in a tertiary care center. Nine hundred and twenty-three obese children and adolescents were evaluated for metabolic characteristics, cholesterol levels, the agreement rate and prevalence of MS across age subgroups with pediatric National Cholesterol Education Program/ Adult Treatment Panel III (NCEP-ATP III) and International Diabetes Federation (IDF) classifications. RESULTS: The overall prevalence of MS was 36.2% and 56.7% with NCEPATP III and IDF. The overall concordance was fair (k: 0.269), with substantial values observed only in children older than 10 (k: 0.708) and 16 yr (0.694). Concordant subjects for both classifications, ≤6 yr, had higher triglycerides, blood pressure (p<0.05) and lower HDL-cholesterol (p<0.0001), with respect to those found to be discordant. Concordant subjects ranging 6-10 yr had all parameters higher than those discordant for IDF (p<0.01) and insulin resistance (p<0.05) than those discordant for NCEP-ATP III. Concordant subjects ≥10 yr presented more altered parameters than those included only in NCEP-ATP III (p<0.05). Overt glucose alterations were uncommon (7.4%; confidence interval 95% 0.1-14.9%), although glucose was modestly higher in MS subjects (p<0.01). Total and LDL-cholesterol was lower in subjects with MS than in those without (p<0.05), and in concordant rather than discordant subjects (p<0.05). CONCLUSIONS: Classifications of MS do not identify the same pediatric population. Subjects who satisfied any classification were the most compromised. Lipid alterations were precocious in the youngest. Obese youths with MS presented lower total and LDL-cholesterol.
