Non-invasive markers and predictors of severity of hepatic fibrosis in HCV patients at Sharkia Governorate, Egypt

ABSTRACT

This study included 109 patients with detectable hepatitis C virus [HCV] by real time PCR. The patients were classified into three different pathological stages and grades according to the new concept of histopathological staging and grading. The different clinical, biochemical, virological and ultrasonographic parameters were assessed and analyzed and the variables that showed a significant association with the histopathological staging and grading were included in the multivariate logistic regression analysis. The regression model revealed that platelet count, matrix metalloproteinase-9 [MMP-9], portal vein diameter, splenic longitudinal axis, alanine transaminase, aspartate transaminase and viral load added a significance to the model in a decreasing order of significance. From these findings, a new score ranged from 0-9 was generated. The score model was applied to the patients to assess its validity, where it proved to be accurate in discriminating patients with mild inflammation and fibrosis [sensitivity 81.8%, specificity 80.5% and accuracy 80.7%] and more accurate in detecting patients with cirrhosis [specificity 96.6%, sensitivity 80% and accuracy 93.6%], but less accurate in detecting patients with moderate to severe fibrosis [specificity 66.7%, sensitivity 68.7% and accuracy 67.9%]. Also, the results revealed that co-infection with schistosomiasis, old age >/45 years and positive history of blood transfusion as a source of infection were significantly associated with severe hepatic pathology