Optimal liver segments for hepatic steatosis assessment using ultrasound attenuation imaging with MRI-PDFF as the reference standard

Abstract

This study aims to identify optimal liver segments for hepatic steatosis assessment using UltraSound Attenuation (USAT) imaging, with MRI-proton density fat fraction (PDFF) as the reference standard. The prospective study included 240 patients who received weight loss counseling at our hospital. All participants underwent USAT examination, among whom 112 also underwent MRI-PDFF examination. USAT measurements were taken in liver segments II/III, IV, V, VI, VII, and VIII, with medians calculated for analysis. In this study, a total of 240 patients were evaluated. The success rate of USAT measurements was significantly higher in the right liver lobes compared to the left liver lobes, with liver segment V and VIII demonstrating the highest success rate, which was statistically significantly different from segments II/III and VI ( P <0.001). The USAT measurements in the right liver lobes were close to each other, with little variability and high repeatability. The correlation between the liver IV segment and the whole liver was statistically significant ( Z = 2.286, P = 0.022). When hepatic steatosis grade ≥ 1 was diagnosed, the AUROC of liver segment VIII was the largest, reaching 0.863 (0.757 to 0.968). The AUROC of whole liver reached the maximum when the diagnosis of hepatic steatosis grade ≥ 2 and ≥ 3, reaching 0.914 (0.854–0.974) and 0.921 (0.850–0.992), respectively. For S ≥ 1, the right anterior lobe of the liver provides comparable performance to the entire liver and may be preferred given higher feasibility. For S ≥ 2 and S ≥ 3, the entire liver averaging offers the best diagnostic accuracy, though the right anterior lobe of the liver remains a clinically acceptable alternative when the entire liver acquisition is not feasible.