Репозиторий Университета

AIM: To develop predictive model for hepatic insufficiency in obstructive jaundice.MATERIAL AND METHODS: Obstructive jaundice was modeled by the author's method on 48 mini pigs, while morpho-functional features of erythrocytes were studied by using of INTEGRA Aura atomic force microscope (NT-MDT, Zelenograd, Russia). Histological specimens were stained with hematoxylin and eosin. Discriminant analysis was used to create predictive model for hepatic insufficiency.RESULTS: Mathematical model of hepatic insufficiency prediction has been developed. Sensitivity and specificity of this model were 94.1% and 74.2% respectively. Total percentage of correct predictions was 81.3%.CONCLUSION: Severe obstructive jaundice contributes erythrocyte's transformation from biconcave to dome-shaped followed by changes of its physical properties. Erythrocyte's volume and activity of cytolysis enzymes are the most informative to predict hepatic insufficiency. Our model allows us to diagnose this complication at early stages and to correct pre-, intra- and postoperative therapy.

© 2018 GEOTAR Media. All rights reserved. The purpose of the study is to develop an accessible method of dynamic control of patients with mechanical jaundice for a practical doctor. Data from 258 medical histories of patients with mechanical jaundice were studied. To solve this problem we used a method for calculating the severity level using a universal statistical program StatSoft Statistica for OS Windows. A mathematical relationship between a degree of possibility of a fatal outcome and the nine significant parameters was developed. As a result of the calculations, such factors as disease duration (days), blood bilirubin, body temperature, blood leucocytosis, pulse rate, patient age, blood creatinine, blood lymphocytes, respiratory rate appeared to be significant parameters in calculating the severity of mechanical jaundice of a particular patient. At the second stage, a multivariate regression analysis was performed, used to assess the degree of influence of various factors on the resulting indicator. A statistical dependence was found on the basis of which a formula combining 9 significant parameters was developed to determine the severity level of a patient based on the degree of influence of each of these factors on the resulting parameter: G = 0,002 (d×b) + 1,2 (t - 36,6) + 0,001 (p×v) + 0,322 (L/1000 - 5) + + 0,22 (16 - Lim) + 0,0085 (K - 60) + 0,165 (Di - 20). The formula took into account the stochastic dependence of the resulting parameter on the factors influencing it. When calculating a correlation analysis was conducted aimed at establishing a stochastic relationship between the parameters and estimating the tightness of the connection between the factors and the resulting index. This method of calculating the severity level can be applied by any surgeon and can accurately monitor the minimum changes in the patient's condition, as well as the effect of individual elements of treatment on the state of the disease.