Diagnostic Performance of 9 Quantitative Ultrasound Parameters for Detection and Classification of Hepatic Steatosis in Nonalcoholic Fatty Liver Disease

Background: Nonalcoholic fatty liver disease (NAFLD) is a leading cause of chronic liver diseaseworldwide. Quantitative ultrasound (QUS) parameters based on radiofrequency raw data show promise in quantifying liver fat. Purpose: The aim of this study was to evaluate the diagnostic performance of 9 QUS parameters compared with magnetic resonance imaging (MRI)-estimated proton density fat fraction (PDFF) in detecting and staging hepatic steatosis in patients with or suspected of NAFLD. Materials and Methods: In this Health Insurance Portability and Accountability Act-compliant institutional review board-approved prospective study, 31 participants with or suspected of NAFLD, without other underlying chronic liver diseases (13 men, 18 women; average age, 52 years [range, 26-90 years]), were examined. The following parameters were obtained: acoustic attenuation coefficient (AC); hepatorenal index (HRI); Nakagami parameter; shear wave elastography measures such as shear wave elasticity, viscosity, and dispersion; and spectroscopy-derived parameters including spectral intercept (SI), spectral slope (SS), and midband fit (MBF). The diagnostic ability (area under the receiver operating characteristic curves and accuracy) of QUS parameters was assessed against different MRIPDFF cutoffs (the reference standard): 6.4%, 17.4%, and 22.1%. Linearity with MRI-PDFF was evaluated with Spearman correlation coefficients (p). Results: The AC, SI, Nakagami, SS, HRI, and MBF strongly correlated with MRI-PDFF (P = 0.89, 0.89, 0.88, -0.87, 0.81, and 0.71, respectively [P < 0.01]), with highest area under the receiver operating characteristic curves (ranging from 0.85 to 1) for identifying hepatic steatosis using 6.4%, 17.4%, and 22.1% MRI-PDFF cutoffs. In contrast, shear wave elasticity, shear wave viscosity, and shear wave dispersion did not strongly correlate to MRI-PDFF (P = 0.45, 0.38, and 0.07, respectively) and had poor diagnostic performance. Conclusion: The AC, Nakagami, SI, SS, MBF, and HRI best correlate with MRI-PDFF and show high diagnostic performance for detecting and classifying hepatic steatosis in our study population.

Automated summary

In this study, we found that AC, Nakagami, SI, SS, MBF, and HRI have the best correlation with MRI-PDFF and show high diagnostic performance for detecting and classifying hepatic steatosis.