Research of non-invasive diagnosis model of fibrosis in chronic liver disease

Objective To develop a non-invasive diagnosis model of fibrosis based on clinical,serum markers and ultrasonography checking in chronic liver disease.Methods Two hundred and seventy-one patients with chronic liver disease were randomly divided into an estimation group(190 cases)and a validation group(81 cases).Liver biopsies and hepatic tissue pathology checking were done.Seventeen clinical,serum markers and ultrasonography checking were detected and recorded in the estimation group.The independent hazard factors were screened according to single factor analysis and multiple factor Logistic regression analysis.The non-invasive diagnosis model of fibrosis was established according to the independent hazard factors and applied to the validation group to test its accuracy.Results Among 10 variables associated with liver fibrosis selected by single factor analysis,international normalized ratio of prothrombin time,alkali phosphatase and middle hepatic vein were identified by multiple factor Logistic regression analysis as independent hazard factors of fibrosis.The non-invasive diagnosis model of fibrosis constructed from the above three factors was established.The scores of non-invasive diagnosis model of fibrosis in S0 and S1 were lower than those of S2,S3 and S4(P0.01).There was positive correlation between the score of non-invasive diagnosis model of fibrosis and the stages of fibrosis(γs=0.545,P0.01).In the receiver operating characteristic curve(ROC)analysis,the areas under ROC(AUC)of significant fibrosis,extensive fibrosis and cirrhosis were 0.757,0.748 and 0.903,respectively.The best diagnosis values were 5.47,6.09 and 6.97,respectively.The diagnosis accuracies were 75.8%,70% and 88.9%,respectively.The AUC of significant fibrosis,extensive fibrosis and cirrhosis in validation group were 0.737,0.721 and 0.908,respectively.Conclusion There is better accuracy and reproducible with the score of non-invasive diagnosis model of fibrosis in evaluating the stages of fibrosis of chronic liver disease.Non-invasive diagnosis model of fibrosis could be used to replace liver biopsy to dynamic changes of chronic liver disease.