EASL-ALEH 2015 Algorithm for the Use of Transient Elastography in Treatment-naive Patients With Hepatitis B

An Independent Validation

Jean Nana; Kristina Skaare; Jean Luc Bosson; Vincent Leroy; Tarik Asselah; Michael Adler; Nathalie Sturm; Jean-Pierre Zarski


J Viral Hepat. 2021;28(8):1169-1176. 

In This Article

Patients and Methods

Design Overview and Participants

This is a serial cross-sectional study on a sample of consecutive naive patients from 3 centres, French (Service universitaire d'hépato-gastroenterologie, Grenoble; Service d'Hépatologie, Hospital Beaujon, Paris) and Belgian (Gastro-Entérologie d'Hépato-Pancréatologie, Hôpital Académique ERASME, Bruxelles). Patients were chronic hepatitis B virus carriers in whom liver biopsy and TE were performed for evaluating hepatic fibrosis. Patients underwent liver biopsy and TE the same day or within a maximum of 90 days, between January 2004 and June 2015. Diagnosis of chronic hepatitis was based on the presence of hepatitis B surface antigen (HBsAg) and positive HBV-DNA level for at least 6 months. Each hepatologist applied their usual criteria to determine whether liver biopsy was indicated according to the European Association for the Study of the Liver (EASL) guidelines, those indications varied overtime. Exclusion criteria included age under 18 years, hepatitis C virus (HCV)-HBV co-infection, co-infection with human immunodeficiency virus, hepatitis delta virus, others causes of liver disease, alcohol consumption over 30 g/day, hepatocellular carcinoma, previous antiviral treatment and previous liver transplantation. Demographic data were recorded at the time of liver biopsy. The study protocol conformed to the ethical guidelines of the 1975 Helsinki Declaration and was approved by our institutional review board, reference number DC-2008-727. Patients were enrolled after giving their written informed consent.

Liver Biopsy

Percutaneous liver biopsy was performed by senior experienced hepatologists operators using a 16G disposable needle (Hepafix, Braun, Melsungen, Germany). Tissue samples were fixed in formalin and embedded in paraffin. All specimens were analysed by a senior pathologist. Specimens shorter than 15 mm or not suitable for fibrosis assessment due to fragmentation were excluded. Liver fibrosis and necrotic and inflammatory activity were evaluated according to the METAVIR scoring system. Significant steatosis was defined at the cut-off of 5%. All sections were retrospectively blindly reviewed by a senior liver pathologist.

Transient Elastography

FibroScan™ (Echosens) was performed by experienced physicians (>100 examinations before the study) using the standard probe (M probe) the day of liver biopsy. The manufacturer's instructions were followed. FibroScan™ examen was repeated until 10 valid measurements were recorded. Liver stiffness results (kilopascals [kPa]) were expressed as the median and the interquartile range (IQR) of all valid measurements. Transient elastography was considered as reliable when the IQR/median was lower than 30% according to quality criteria.[17]

Blood Tests

Fasting blood samples were collected the day of the biopsy and served for immediate dosages; serum HBV-DNA levels were measured using the Cobas® Ampliprep/Cobas TaqMan® assay (Roche Diagnostics), and level was expressed as IU/mL. The definition of normal ALT levels was 30 U/L for men and 19 U/L for women.[18]

Statistical Analysis

Descriptive results were expressed as a median (interquartile range) for continuous variables or as number and proportions for categorical variables. The overall diagnostic value was expressed with area under ROC curves (AUROC) given with 95% confidence intervals (95% CI) for three relevant diagnostic targets: F ≥ 2, F ≥ 3 and F = 4. For each diagnostic target, optimal cut-offs were determined according to the Youden method or for achieving 90% NPV in the prediction of significant fibrosis, severe fibrosis and cirrhosis. For each cut-off, sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV) and percentage of correctly classified patients were calculated. Because of missing data for ALT level (n = 2) and viral load (n = 21), a technique of imputation was used. Missing ALT values were imputed by the median value of patients with the same necrotic and inflammatory activity according to the METAVIR scoring system. Missing viral load was imputed by the median value of patients in the same group of age and fibrosis stage. For HBeAg (n = 9), logistic regression with age and fibrosis stages was used.

The recommendations of the Liver-FibroSTARD for the manuscripts of diagnostic studies on non-invasive tools for liver fibrosis evaluation were taken into consideration.[19]