Strain elastography for noninvasive assessment of liver fibrosis: A prospective study with histological comparison

Abstract

The aim of this study was to prospectively evaluate the diagnostic performance of strain elastography for the assessment of liver fibrosis in patients with chronic liver disease using Ishak (0-6) histology stage as a reference standard. Ninety-eight consecutive patients with suspected chronic liver disease scheduled for liver biopsy (n = 78) or histologically confirmed cirrhosis (n = 20) were enrolled. Liver fibrosis Index (LF Index) calculated by strain elastography, liver stiffness by transient elastography and serum fibrosis markers (aspartate aminotransferase-to-platelet ratio index and King's Score) were measured. Spearman's correlation coefficient between the LF Index, liver stiffness, serum fibrosis markers and fibrosis stage were calculated and compared using areas under the receiver-operating characteristics (AUROCs) curves. Among 73 patients who underwent strain elastography, there was weak correlation between fibrosis stage and the LF Index (Spearman's: ρ = 0.385 for Ishak score; P = 0.001). Among 52 patients who underwent strain elastography and transient elastography, the AUROC values using LF Index, transient elastography, aspartate aminotransferase-to-platelet ratio index and King's Score for diagnosing significant fibrosis (Ishak score ≥ 3) were 0.79, 0.87, 0.86 and 0.85, respectively (P < 0.0001) and for diagnosing severe fibrosis/cirrhosis (Ishak score ≥ 5) were 0.83, 0.94, 0.92 and 0.92, respectively (P < 0.0001). When comparing the diagnostic performance using LF Index, transient elastography, aspartate aminotransferase-to-platelet ratio index and King's Score, transient elastography shows a significantly higher AUROC value than LF Index in detecting severe fibrosis (P = 0.0149). The diagnostic performance of LF Index calculated by strain elastography was not statistically significantly different to the other noninvasive tests for the assessment of significant liver fibrosis but inferior to transient elastography for the assessment of severe fibrosis/cirrhosis.

Keywords: Liver; liver fibrosis; real-time elastography; strain elastography; ultrasound.

Figure 1.

(a) Real-time tissue elastogram from a 27-year-old male with hepatitis B viral hepatitis which demonstrates relative “soft” liver texture. The histological fibrosis stage is Ishak 1. (b) Real-time tissue elastogram from a 63-year-old female with hepatitis C viral hepatitis which demonstrates relative “hard” liver texture. The histological fibrosis stage is Ishak 6.

Figure 2.

Box and whisker plots of (a) LF Index, (b) TE, (c) APRI, and (d) King’s Score according to different Ishak fibrosis stage. The length of the box represents the interquartile ranges (second and third quartiles) in which 50% of the values are located. Circles or stars represent outliers. The thick line through each box represents the median value. The error bars show the minimum and maximum values (range). Open circles and stars represent outliers.

Figure 3.

Comparison of the receiver operating characteristic curves of noninvasive methods for diagnosis of (a) significant fibrosis (Ishak ≥ stage 3) and (b) severe fibrosis/cirrhosis (Ishak ≥ stage 5) in patients with chronic liver disease in the study population.