Comparison of Transient Elastography and Acoustic Radiation Force Impulse for Non-Invasive Staging of Liver Fibrosis in Patients With Chronic Hepatitis C

Abstract

Transient elastography (TE) is adequate for a diagnosis of cirrhosis, but its accuracy for milder stages of fibrosis is much less satisfactory. The objective of this study was to compare the performance and the discordance rate of acoustic radiation force impulse (ARFI) and TE with liver biopsy in a cohort of chronic hepatitis C (CHC) patients. One hundred thirty-nine consecutive patients with CHC were enrolled in two tertiary centers, and evaluated for histological (Metavir score) and biochemical features. All patients underwent TE and ARFI. TE was unreliable in nine patients (6.5%), while in no cases (0%) were ARFI invalid measurements recorded (P=0.029). By area under receiver operating characteristic curve (AUROC), the best cutoff values for TE and ARFI for significant fibrosis (≥F2) were ≥6.5 kPa (AUROC: 0.78) and ≥1.3 m/s (AUROC: 0.86), respectively. For severe fibrosis (F3-F4), these cutoff values were 8.8 kPa (AUROC: 0.83) for TE and 1.7 m/s (AUROC: 0.94) for ARFI. For cirrhosis, TE had its best cutoff at ≥11 kPa (AUROC: 0.80) and ARFI at ≥2.0 m/s (AUROC: 0.89). By pairwise comparison of AUROC, ARFI was significantly more accurate than TE for a diagnosis of significant and severe fibrosis (P=0.024 and P=0.002, respectively), while this difference was only marginal for cirrhosis (P=0.09). By partial AUROC analysis, ARFI performance results significantly higher for all three stages of fibrosis. The average concordance rates of TE and ARFI vs. liver biopsy were 45.4 and 54.7%, respectively. By multivariate analysis, ARFI was not associated with alanine aminotransferase (ALT), body mass index, Metavir grade, and liver steatosis, while TE was significantly correlated with the ALT value (P=0.027). In a cohort of patients with CHC, ARFI imaging was more accurate than TE for the non-invasive staging of both significant and severe classes of liver fibrosis.