Prediction of fibrosis in HCV-infected liver transplant recipients with a simple noninvasive index

Authors


Abstract

Recurrent hepatitis C is a frequent event in liver transplantation (LT). Serial liver biopsies remain the best way of monitoring disease progression. Due to the limitations of a liver biopsy, there is an interest in developing noninvasive markers of liver fibrosis. While several models for predicting fibrosis have been constructed in patients who have not undergone transplantation, these are lacking in the transplant population. The aim of this study was to construct one simple model based on routine laboratory data to predict fibrosis in hepatitis C virus (HCV)-infected LT patients. A total of 510 yearly protocol liver biopsies performed in 188 LT patients (67% male; median age 54 years) were divided into 2 groups: training set (n = 414) and validation set (n = 96). Laboratory variables at time of biopsies were recorded. Multivariate analysis identified 4 variables as independent predictors of fibrosis: prothrombin time (PT), albumin/total protein ratio, aspartate aminotransferase (AST), and time since LT. The area under the receiver operating characteristic (ROC) curves (AUCs) were 0.80 and 0.84 for the training and the validation set, respectively. In the training set, using a cutoff of 0.2, the model had a sensitivity, specificity, positive predictive value, and negative predictive value of 74%, 69%, 42%, and 90%, respectively, to differentiate significant (bridging fibrosis and cirrhosis) from mild fibrosis (none or portal). In the validation cohort, these values increased to 87%, 71%, 49%, and 95%, respectively. In conclusion, in the LT setting, a simple fibrosis index is useful to select HCV-infected patients with a very low risk of significant fibrosis in whom protocol liver biopsies may be avoided. (Liver Transpl 2005;11:456–462.)

Ancillary