Etiology-related determinants of liver stiffness values in chronic viral hepatitis B or C

Abstract

Background & Aims

Transient elastography (TE) has gained popularity for the non-invasive assessment of severity of chronic viral hepatitis, but a comprehensive evaluation of the factors that might account for discrepancy in diagnostic accuracy between TE and the standard of care liver biopsy (LB) is still needed.

Methods

Patients with chronic hepatitis-B (HBV, n = 104) or -C (HCV, n = 453) underwent percutaneous LB concomitantly with TE (FibroScan®; Echosens, Paris, France). Liver cell necroinflammatory activity (A) and fibrosis (F) were assessed by METAVIR. Perisinusoidal fibrosis was rated with a 0–3 score. Determinants of TE results were investigated by a linear regression model whereas discordance between TE and LB results was assessed by logistic regression.

Results

Fibrosis (p <0.0001) and liver cell necroinflammatory activity (p <0.0001) were independently associated with TE results in both HBV and HCV patients, whereas steatosis (p <0.0001) was independently associated with TE in HCV only. Fibrosis overestimation was predicted by severe/moderate necroinflammatory activity in HBV and by older age (41–60 or >60 years vs. <40), >2 UNL AST and >2 UNL GGT, as well as severe/moderate necroinflammatory activity and severe/moderate steatosis in HCV. In the latter patients, however, moderate/severe necroinflammatory activity and steatosis were the only independent predictors of fibrosis overestimation.

Conclusions

Fibrosis and necroinflammatory activity are the main determinants of TE in chronic viral hepatitis. Since TE staging of fibrosis is influenced by necroinflammatory activity and steatosis, a diagnostic LB is deemed necessary for a reliable intra-patient TE monitoring of the course of viral hepatitis.

Introduction

Transient elastography (TE) is a simple, non-invasive approach for predicting liver disease severity, based upon a mechanical wave generated by vibration that provides an estimate of the liver stiffness, which in turn has been demonstrated to correlate with hepatic fibrosis [1]. Within a defined clinical context, non-invasive assessment of liver disease severity with TE is an attractive alternative for many patients with chronic liver disease (CLD), to avoid the risks inherent with invasive procedures like liver biopsy (LB). However, the relative advantages and disadvantages of TE need to be better elucidated, since increased liver stiffness values by TE do not reflect liver fibrosis only, but also the grade of necroinflammatory activity of the liver [2], [3], [4], [5], [6], [7], [8]. Indeed, inflammatory cell infiltrates, interstitial oedema, liver cell swelling and necrosis, cholangiocellular damage, pericellular fibrosis, and steatosis differently contribute to the liver damage caused by various etiologic agents. More recently, extrahepatic cholestasis and an increased venous pressure due to cardiac failure have added to the list of potential TE confounders [9], [10].

Under defined circumstances, TE compared well with LB as the reference standard to accurately identify patients with CLD with significant liver fibrosis or cirrhosis [1], [2], [3], [4], [5], [6], [7], [8], with a satisfactory inter- and intra-observer reproducibility [5]. However, despite an apparently high diagnostic accuracy, several studies reported incomplete concordance between TE and LB in patients with CLD, owing to a substantial overlap between diagnostic categories, especially in patients with pre-cirrhotic changes of the liver [1], [2], [3], [4], [5], [6], [7], [8], [11], [12], [13], [14]. Due to the increasing role of TE in the management of patients with CLD, the variables related to the patient, the reference standard or TE technique that could favor discordant results between TE and LB, need to be comprehensively evaluated. This is particularly true in patients with chronic viral hepatitis, in whom TE is progressively replacing LB not only in the clinical surveillance programs, but also in the treatment decision making process [15]. In this study, 608 patients with CLD due to either hepatitis B (HBV) or hepatitis C (HCV) virus infection were concurrently investigated with TE and LB, with a particular focus on patients showing discordant results with these diagnostic techniques.