An Approach to Early Detection of Metabolic Syndrome through Non-Invasive Methods in Obese Children

Background: Metabolic Syndrome (MetS) has a high prevalence in children, and its presence increases in those with a high BMI. This fact confirms the need for early detection to avoid the development of other comorbidities. Non-invasive variables are presented as a cost-effective and easy to apply alternative in any clinical setting. Aim: To propose a non-invasive method for the early diagnosis of MetS in overweight and obese Chilean children. Methods: We conducted a cross-sectional study on 221 children aged 6 to 11 years. We carried out multivariate logistic regressions, receiver operating characteristic curves, and discriminant analysis to determine the predictive capacity of non-invasive variables. The proposed new method for early detection of MetS is based on clinical decision trees. Results: The prevalence of MetS was 26.7%. The area under the curve for the BMI and waist circumference was 0.827 and 0.808, respectively. Two decision trees were calculated: the first included blood pressure (≥104.5/69 mmHg), BMI (≥23.5 Kg/m(2)) and WHtR (≥0.55); the second used BMI (≥23.5 Kg/m(2)) and WHtR (≥0.55), with validity index of 74.7% and 80.5%, respectively. Conclusions: Early detection of MetS is possible through non-invasive methods in overweight and obese children. Two models (Clinical decision trees) based on anthropometric (non-invasive) variables with acceptable validity indexes have been presented. Clinical decision trees can be applied in different clinical and non-clinical settings, adapting to the tools available, being an economical and easy to measurement option. These methods reduce the use of blood tests to those patients who require confirmation.