Prediction of course and progression of nonalcoholic fatty liver disease in patients with obesity and pathology of the biliary tract with the help of mathematical modeling

Determination of the most important functional parameters of the liver, lipid peroxidation (LPO) and antioxidant protection (AOP) markers of endogenous intoxication and indicators of connective-tissue metabolism in patients with non-alcoholic fatty liver disease (NAFLD) associated with obesity and pathology of the biliary tract and development of criteria for predicting the likelihood of the transformation of non-alcoholic steatosis (NAHS) to non-alcoholic steatohepatitis (NASH) using mathematical modeling was performed. The data of patients with different clinical stages of NAFLD in combination with obesity and BT pathology were analyzed: the study group consisted of patients with NASH (n=100) and control group – patients with NAHS (n=100). The diagnosis of NAFLD and pathology of BT was established on the basis of anamnesis, clinical and instrumental (ultrasound imaging of the abdomen) examination. The body mass index was determined according to the Quetelet formula. Thymol test, total protein level and protein fractions in blood serum, the concentration of molecules of average weight (MSM) and their fractional composition were investigated. The condition of the LPO system was evaluated by the concentration of malondialdehyde, schiff bases and intermediate LPO products. The AOP was assessed by the activity of superoxide dismutase and catalase in the hemolysate of erythrocytes. The process of fibrosis was evaluated according to the content of free oxyproline, total oxyproline and protein-binding oxyproline, hexosamines. According to the correlation analysis 30 factors that increase the risk of transformation of NAHS in NASH (p<0.05) were selected. ROC-analysis was conducted with the definition of factor levels that provides the maximum difference between the compared groups in terms of sensitivity and specificity. Logistic regression model was built basing on certain criteria for distinguishing between different NAHP stages. Logistic regression model was built with the coefficients В0=-2,610 and В1=0,157 (p<0.001 by Student's t test and Wald), the adequacy of the model – χ2=131.4; p<0.001. Designed logistic regression model allows to predict the progression of non-alcoholic steatosis to steatohepatitis stage in persons with obesity and pathology of the biliary tract with the help of laboratory data with indicators of sensitivity – 83.0%, specificity – 91.0%, forecast accuracy – 87.0%, validity – 76.1%.