[Identification of prognostic significant markers of early diagnosis of obesity and metabolic disorders].

Abstract

Obesity, being a global epidemic of the 21st century, increases the likelihood of various diseases and conditions that are associated with increased mortality. Given the risks associated with obesity throughout life, identifying early predictors of its development is a priority task of prevention.

Objective: Assessment of the prognostic significance of laboratory parameters, functional tests and instrumental examination data for the early diagnosis of obesity and associated metabolic disorders.

Material and methods: The studies were conducted at the Scientific and Clinical Center 1 of the B.V. Petrovsky Russian Scientific Center of Surgery with the participation of 77 patients (43 men and 34 women) aged 33 to 65 years. After signing voluntary informed consent for a special study with the processing of their personal data, all patients underwent a comprehensive clinical, laboratory and functional examination. The study design provided for a three-time examination of patients with an interval of 6 and 12 months. To search for predictors of early diagnosis of obesity and metabolic disorders, we used an algorithm for constructing a mathematical model of multiple regression. The dynamics of the body mass index and the insulin resistance index served as the resulting feature.

Results: To build a mathematical predictive model, a matrix of independent variables was formed, including parameters characterizing the biochemical status of patients, the severity of systemic inflammation, body composition, as well as arterial stiffness and vascular condition based on the results of ultrasound duplex scanning of the brachiocephalic arteries. The choice of the most informative set of independent variables that act as predictors provides for achieving the maximum additive effect in explaining the variance of the resulting feature. The construction of the actual mathematical multiple regression model was based on the results of the selection of independent factors carried out using the method of sequential hypothesis testing. As a result of applying the algorithm of sequential hypothesis testing, three independent variables (predictors) were identified to predict the degree of BMI increase and four variables to predict the development of metabolic disorders. The accuracy of the developed model, verified at the final stage of the study, showed its high information content.

Conclusion: Two clusters of predictors were identified: the cluster of predictors for the risk of obesity development includes the systemic inflammatory response index, the triglyceride-glucose index and the blood leptin level. The cluster of biomarkers of metabolic disorders consists of the atherogenicity coefficient, the blood malondialdehyde level, the systemic inflammation index and the body roundness index.