Evaluation of Ultrasound Shear Wave Elastography (SWE) and Attenuation Imaging (ATI) for the Assessment of Hepatic Fibrosis and Steatosis-A Prospective Comparison With MRI Liver Elastography and Proton Density Fat Fraction (PDFF).

INTRODUCTION

To assess competing US techniques in evaluation of chronic liver disease, we performed a study to determine the diagnostic accuracy of 2D-shear wave elastography (2D-SWE) and attenuation imaging (ATI) compared with more established methods of point shear wave elastography (pSWE) and visual B-mode grading of hepatic steatosis. Magnetic resonance elastography (MRE) and proton density fat fraction (MR-PDFF) were used as respective reference standards.

METHODS

Fifty-five patients underwent same-day liver ultrasound (conventional B-mode, pSWE, 2D-SWE, and ATI) and 3 T MRI (MRE and MR-PDFF). Diagnostic performance for steatosis (mild, moderate, severe) and fibrosis (≥ F2, ≥ F3, and F4) grades was assessed using receiver operating characteristic (ROC) curve analysis, with area under the curve (AUC) calculated. Agreement was evaluated using Cohen's kappa for categorical comparisons and intraclass correlation coefficients (ICC) for continuous measurements within the same modality.

RESULTS

For fibrosis, 2D-SWE demonstrated high sensitivity (100%) for advanced fibrosis (≥ F3; AUC 0.87), compared with pSWE (sensitivity of 71.4%). In contrast, pSWE showed higher specificity (92.3%) for significant fibrosis (≥ F2) compared to 2D-SWE (73.9%). Categorical agreement with MRE was higher for pSWE (kappa = 0.66) than 2D-SWE (kappa = 0.45). For steatosis, visual B-mode across all grades had AUC 0.85-0.89 compared to 0.68-0.85 for ATI. ATI performed best in detecting severe steatosis (AUC 0.85).

CONCLUSION

Across the spectrum of clinically relevant assessment of liver fibrosis and steatosis, pSWE and visual B-mode grading showed better correlation with MRI results than 2D-SWE and ATI, respectively. However, they did provide good diagnostic performance in grading liver fibrosis and steatosis in advanced disease. Further research may better define more accurate standardised cut-off scores for the 2D-SWE and ATI techniques.