The pediatric population has various criteria for measuring metabolic syndrome (MetS). The diversity of consensus for diagnosis has led to different non-comparable reported prevalence. Given the increase in its prevalence in pediatric ages, it is necessary to develop efficient methods to encourage early detection. Consequently, early screening for the risk of MetS could favor timely action in preventing associated comorbidities in adulthood. This study aimed to establish the diagnostic capacity of models that use non-invasive (anthropometric) and invasive (serum biomarkers) variables for the early detection of MetS in Chilean children. A cross-sectional study was carried out on 220 children aged 6 to 11. Multivariate logistic regressions and discriminant analyses were applied to determine the diagnostic capacity of invasive and non-invasive variables. Based on these results, four diagnostic models were created and compared: (i) anthropometric, (ii) hormonal (insulin, leptin, and adiponectin), (iii) Lipid A (high-density cholesterol lipoprotein [HDL-c] and triglycerides [TG]) and (iv) Lipid B (TG/HDL-c). The prevalence of MetS was 26.8%. Lipid biomarkers (HDL-c and TG) and their ratio (TG/HDL-c) presented higher diagnostic capacity, above 80%, followed by body mass index (BMI, 0.71-0.88) and waist-to-height ratio (WHtR, 0.70-0.87). The lipid model A was the most accurate (sensitivity [S] = 62.7%, specificity [E] = 96.9%, validity index 87.7%), followed by the anthropometric model (S = 69.5%, E = 88.8% and validity index = 83.6%). In conclusion, detecting MetS was possible through invasive and non-invasive methods tested in overweight and obese children. The proposed models based on anthropometric variables, or serum biomarkers of the lipid model A, presented acceptable validity indices. Moreover, they were higher than those that measured adipokines, leptin, and adiponectin. The anthropometric model was the most cost-effective and easy to apply in different environments.
Keywords: adipokines; anthropometry; early diagnosis; metabolic biomarkers; metabolic syndrome; pediatric obesity.