Intra-individual Comparisons of the Ultrasound-Guided Attenuation Parameter and the Magnetic Resonance Imaging-Based Proton Density Fat Fraction Using Bias and Precision Statistics.

ABSTRACT

Ultrasound-based techniques using the attenuation coefficient, including the ultrasound-guided attenuation parameter (UGAP), have been developed for the quantification of hepatic steatosis. The magnetic resonance imaging-based proton density fat fraction (MRI-PDFF) is considered to be more accurate than liver biopsy for liver fat quantification. The aim of this study was to perform intra-individual comparisons of UGAP and MRI-PDFF for determining hepatic steatosis grade. The study enrolled 309 patients who underwent UGAP and MRI-PDFF measurements. Bland-Altman analysis was conducted after transforming MRI-PDFF values to a normal distribution and converted to a common set of units using linear regression analysis for differing scales. The expected limits of agreement (LOA) was defined as the square root of the sum of the squares of UGAP and MRI-PDFF precision. A Bland-Altman plot revealed that the bias and upper and lower LOAs (ULOA and LLOA) were -0.0047, 0.1160 and -0.1255, respectively. The percentage difference indicated that the mean, ULOA and LLOA were -1.1434%, 18.1723% and -20.4590%, respectively. The calculated expected LOA was 18.5449%, and 283 of 309 patients (91.6%) had a percentage difference within 18.5449%. Bland-Altman analysis revealed that UGAP and MRI-PDFF were interchangeable within a clinically acceptable range.