Validation and comparison of indexes for fibrosis and cirrhosis prediction in chronic hepatitis C patients: proposal for a pragmatic approach classification without liver biopsies

Abstract

Noninvasive indexes have been developed to predict fibrosis staging. The aim of this study was to assess the diagnostic accuracy of these indexes in comparison with liver histology in hepatitis C virus (HCV)-infected patients. A total of 235 consecutive patients with HCV infection from the Fibropaca multicentre independent study were included in this paper. FibroTest (FT), aspartate aminotransferase to platelet ratio index (APRI) and Forns score were assessed in the cohort and compared with liver histology performed on the same day. The main end point was the area under characteristic curves (AUCs) for the diagnosis of significant fibrosis (F2-F4) and cirrhosis (F4) by the METAVIR classification. Mean age was 46 (+/-11) years, 55% were males, 42% (n = 99) had significant fibrosis (F2-F4) and 7% (n = 16) had cirrhosis (F4). For the diagnosis of significant fibrosis, respective AUCs of FT, APRI and Forns score were 0.81 (95% confidence interval: 0.76-0.86), 0.71 (0.67-0.79) and 0.76 (0.70-0.82); for cirrhosis prognosis, AUCs of FT and APRI were 0.82 (0.77-0.87) and 0.81 (0.76-0.86) (AUCs not significantly different). Using each index independently, all patients were classified by FT, 214 (91%) patients were classified by APRI and 129 (55%) by Forns score. There were significantly more cases of discordances between APRI and liver biopsy than between FT or Forns score and liver biopsy (P < 0.05). Performing all scores (FT, Forns and APRI) without liver biopsy allowed fibrosis to be well evaluated in 191 patients (81.3%), including patients with FT failure. Liver biopsy remained mandatory to evaluate fibrosis in 44 patients (18.7%). Our study shows that performing all the tests and liver biopsy improves the diagnostic accuracy for liver fibrosis in chronic hepatitis C patients without patent comorbidities. The combination of all tests with liver biopsy allowed 225/235 (96%) patients to be correctly classified. The combination of all tests without liver biopsy allowed 191/235 (81.3%) patients to be correctly classified; liver biopsy remained mandatory in some patients (18.7%).