Original Contribution
Hepatic Elastometry and Glissonian Line in the Assessment of Liver Fibrosis

https://doi.org/10.1016/j.ultrasmedbio.2020.12.015Get rights and content

Abstract

The aim of this study was to identify a method for staging hepatic fibrosis using a non-invasive, rapid and inexpensive technique based on ultrasound morphologic hepatic features. A total of 215 patients with different liver diseases underwent B-mode (2-D brightness mode) ultrasonography, vibration-controlled transient elastography, 2-D shear wave elastography and measurement of the controlled attenuation parameter with transient elastography. B-Mode images of the anterior margin of the left lobe were obtained and processed with automatic Genoa Line Quantification (GLQ) software based on a neural network for staging liver fibrosis. The accuracy of GLQ was 90.6% during model training and 78.9% in 38 different patients with concordant elastometric measures. Receiver operating characteristic curve analysis of GLQ performance using vibration-controlled transient elastography as a reference yielded areas under the curves of 0.851 for F ≥ F1, 0.793 for F ≥ F2, 0.784 for F ≥ F3 and 0.789 for F ≥ F4. GLQ has the potential to be a rapid, easy-to-perform and tolerable method in the staging of liver fibrosis.

Introduction

Liver affections constitute a heterogeneous group of diseases with many different etiologies of which the most common are alcoholic and non-alcoholic fatty liver disease and viral hepatitis (Carbone and Neuberger 2014; Wilkins et al. 2015; Murphy et al. 2017; Liu et al. 2019). Chronic liver disease (CLD) causes constant inflammation that leads to the development of fibrosis and cirrhosis, which may cause liver failure and hepatocellular carcinoma development (Schuppan and Afdhal 2008). Accurate prediction of disease progress is crucial for proper patient management, and the Metavir classification was previously defined to objectively track disease progression (especially fibrosis staging) (Goodman 2007). Fibrosis progression is staged according to Metavir as follows: F0 = no fibrosis, F1 = mild fibrosis, F2 = significant fibrosis, F3 = severe fibrosis and F4 = cirrhosis.

To date, the gold standard technique is liver biopsy although it is an invasive procedure with possible complications such as hemorrhage and pain, but mortality is rare. Moreover, it is affected by sampling bias because of the small size of the hepatic tissue studied (1/50,000th of liver weight) (Van Thiel et al. 1993; Martínez et al. 2011).

Some studies have focused on the use of specific serum biomarkers, but these failed to assess early stages of the disease accurately; therefore, their use in clinical practice was limited (Lurie et al. 2015).

Computed tomography (CT) and magnetic resonance imaging (MRI) are effective for the diagnosis and grading of liver fibrosis, but they have limitations in that they are expensive (Sistrom and McKay 2005) and slightly widespread among peripheral hospitals (Ginde et al. 2008; Economou and Panteli 2019), which makes their use to screen and evaluate the large number of patients with liver diseases impractical in clinical practice. Moreover, the ionizing radiation employed by CT contraindicates its use for long-term follow-up and for women of childbearing age or pregnant women (Schmidt 2012). MRI evaluation, on the other hand, can be limited by metal prostheses, patient claustrophobia and its long duration (Yousaf et al. 2018). Despite these limits, MRI has high sensitivity, specificity and accuracy in detecting and staging hepatic fibrosis (Han et al. 2017; Park et al. 2017; Li et al. 2019).

B-Mode echography is the most widely used method for CLD assessment, although this technique allows only qualitative evaluation of liver echostructure and Glisson capsule regularity. Furthermore, some issues are related to the competence and experience of physicians, which makes the effectiveness of ultrasonography dependent on physicians (Ferral et al. 1992; Kudo et al. 2008).

Ultrasound elastography is an effective non-invasive method for liver staging that provides a quantitative measurement of liver fibrosis by detecting changes in hepatic stiffness caused by pathologic conditions (i.e., inflammation or fibrosis) (Murad Gutiérrez and Romero Enciso 2018). Different elastometric techniques and devices have been developed for this purpose, but the most commonly used and validated techniques are vibration-controlled transient elastography (VCTE), shear wave elastography (SWE), sound touch elastography (STE) and acoustic radiation force impulse (ARFI) elastography.

Vibration-controlled transient elastography has been a pioneering technology in the elastometric diagnosis of cirrhosis (87% and 91% sensitivity and specificity, respectively) and fibrosis (70% and 84% sensitivity and specificity respectively) (Tsochatzis et al. 2011; Bota et al. 2013; Brener 2015; Ragazzo et al. 2017). VCTE has also been proposed as a reference method for studying elastometric techniques other than liver biopsy (Sporea et al. 2014).

Some studies have established that ARFI elastography has a predictive value similar to that of VCTE with high specificity and sensitivity (Tsochatzis et al. 2011; Bota et al. 2013). Gatos et al. (2020) emphasized that VCTE, STE and SWE achieve similar performance using liver biopsy as a reference, and the receiver operating characteristic curve (ROC) yielded areas under the curve (AUROC) of approximately 0.95 in differentiating between different Metavir stages.

An innovative approach is the use of computer neural networks that allow training deep learning algorithms through some medical features and can be used in clinical research, computer-aided diagnosis (CAD) or classification of the severity of some medical phenomena (Handelman et al. 2018). A texture analysis of SWE images with a CAD algorithm has been proposed and reached a sensitivity of 83.3% and specificity of 89.1% with an AUROC of 0.85 in distinguishing healthy patients from those with CLD (Gatos et al. 2016). Similarly, a machine-learning algorithm has been successfully employed in the quantification of stiffness values derived from SWE images and surpassed in accuracy previous studies and expert radiologists in the differentiation of healthy patients from CLD patients (sensitivity = 93.5%, specificity = 81.2% and AUROC = 0.87) (Gatos et al. 2017). The machine learning implementation of SWE analysis has been tested in other studies with satisfying results (Gatos et al. 2019; Durot et al. 2020).

In a previous study (Borro et al. 2018) our group highlighted that Genoa Line Quantification (GLQ) was a valid, low-cost and quick method to stage liver fibrosis. This technique evaluates capsular irregularity at the level of the anterior margin, which is induced by fibrosis progression and is currently the object of investigation from both imaging and histologic points of view (Pickhardt et al. 2016; Xu et al. 2016). Ultrasound evaluation of the liver surface was previously identified as an effective method for staging fibrosis, as it distinguishes cirrhosis from moderate and mild to absent fibrosis (reliable indicator of 0.9524). However, ultrasonography was performed by expert radiologists and, therefore, was an operator-dependent examination (Nishiura et al. 2005). The aim of the present study was to develop and evaluate the effectiveness of a new fibrosis staging method (namely, GLQ) based on computer evaluation of ultrasound images and enhancement with the CAD technique.

Section snippets

Patients and study design

The present pilot study evaluated the effectiveness of evaluation of Glisson's line in patients with CLD of different etiologies. Overall, 215 patients referred to the Gastroenterological Clinic of IRCCS Ospedale San Martino Policlinico from January to June 2019 were included. We collected the most recent liver transaminase results available (i.e., aspartate aminotransferase and alanine aminotransferase). All patients underwent hepatic ultrasound evaluation with different techniques: B-mode

Demographic and clinical characteristics of all patients

The present pilot study included 215 patients (92 women) affected by different CDLs: 53 (24.7%) had NAFLD, 44 hepatitis C virus (HCV), 25 exotoxic hepatitis, 19 autoimmune hepatitis, 11 hepatitis B virus (HBV), 6 primary biliary cholangitis and 37 liver transplant recipients. The outstanding 20 patients had other rare causes of liver dysfunction.

Table 1 summarizes the demographic, ultrasonographic and biochemical characteristics of all 149 patients whose Glisson's line images were used to train

Discussion

The aim of this study was to evaluate the effectiveness of Glisson's line imaging, analyzed with GLQ software, in staging hepatic fibrosis both as an alternative exam and as a complementary method to other fibrosis staging techniques.

In the present study, we determined the efficacy of GLQ in correctly staging hepatic fibrosis after training on Glisson's line images using VCTE and SWE stages as true reference values. GLQ was based on neural network detection and was trained on images from

References (51)

  • V Murad Gutiérrez et al.

    Liver elastography: What it is, how it is done, and how it is interpreted

    Radiologia

    (2018)
  • H Perazzo et al.

    Points to be considered when using transient elastography for diagnosis of portal hypertension according to the Baveno's VI consensus

    J Hepatol

    (2015)
  • TG Ragazzo et al.

    Accuracy of transient elastography-FibroScan®, acoustic radiation force impulse (ARFI) imaging, the enhanced liver fibrosis (ELF) test, APRI, and the FIB-4 index compared with liver biopsy in patients with chronic hepatitis C

    Clinics

    (2017)
  • D Schuppan et al.

    Liver cirrhosis

    Lancet

    (2008)
  • CL Sistrom et al.

    Costs, charges, and revenues for hospital diagnostic imaging procedures: Differences by modality and hospital characteristics

    J Am Coll Radiol

    (2005)
  • I Sporea et al.

    Which are the cut-off values of 2D-shear wave elastography (2D-SWE) liver stiffness measurements predicting different stages of liver fibrosis, considering transient elastography (TE) as the reference method?

    Eur J Radiol

    (2014)
  • EA Tsochatzis et al.

    Elastography for the diagnosis of severity of fibrosis in chronic liver disease: A meta-analysis of diagnostic accuracy

    J Hepatol

    (2011)
  • VWS Wong et al.

    Validity criteria for the diagnosis of fatty liver by M probe-based controlled attenuation parameter

    J Hepatol

    (2017)
  • T Yousaf et al.

    Advances in MRI Methodology

    Int Rev Neurobiol

    (2018)
  • P Borro et al.

    Quantification of ultrasound imaging in the staging of hepatic fibrosis

    Panminerva Med

    (2018)
  • S Bota et al.

    Meta-analysis: ARFI elastography versus transient elastography for the evaluation of liver fibrosis

    Liver Int

    (2013)
  • S. Brener

    Transient elastography for assessment of liver fibrosis and steatosis: An evidence-based analysis

    Ont Health Technol Assess Ser

    (2015)
  • PE Chang et al.

    Clinical applications, limitations and future role of transient elastography in the management of liver disease

    World J Gastrointest Pharmacol Ther

    (2016)
  • KJ Choi et al.

    Development and validation of a deep learning system for staging liver fibrosis by using contrast agent-enhanced CT images in the liver

    Radiology

    (2018)
  • C Economou et al.

    Assessment report: Monitoring and documenting systemic and health effects of health reforms in Greece

    (2019)
  • Cited by (3)

    • The study of hepatic fibrosis: state of art and future perspectives

      2021, Gazzetta Medica Italiana Archivio per le Scienze Mediche
    View full text