Diagnostic performance of liver steatosis analysis and ultrasound-guided attenuation parameter in quantifying hepatic steatosis: a comparative evaluation using controlled attenuation parameter as reference

TL;DR

This study compares two new ultrasound techniques for measuring liver fat with a standard method, finding they perform similarly and offer better ease of use.

Contribution

First direct comparison of LiSA and UGAP against CAP for hepatic steatosis quantification with CAP-referenced cut-off values for LiSA.

Findings

LiSA and UGAP showed strong correlation with CAP (r = 0.83 and 0.81, respectively).

Both LiSA and UGAP achieved high technical success rates (99.74%) compared to CAP (96.84%).

LiSA and UGAP had comparable diagnostic performance (AUROCs) for different steatosis grades.

Abstract

Accurate, non-invasive quantification of hepatic steatosis is crucial for clinical management. Recently developed ultrasound-guided attenuation techniques, such as Liver Steatosis Analysis (LiSA) and the Ultrasound-Guided Attenuation Parameter (UGAP), integrate real-time B-mode imaging with quantitative measurement, potentially overcoming the limitation of the lack of real-time guidance in the traditional Controlled Attenuation Parameter (CAP). However, direct comparative evidence for their performance in clinical practice remains insufficient. This study aimed to conduct a head-to-head comparison of the diagnostic performance of LiSA and UGAP, using the widely adopted CAP as a clinical reference standard.

This prospective study ultimately included 357 participants. All participants underwent LiSA, UGAP, and CAP examinations during a single visit. Operations followed standardized protocols with strict quality control (IQR <40 dB/m, IQR/M <30%). CAP-defined steatosis grades (S0-S3) served as the reference. Pearson correlation analysis was used to assess the associations of each parameter with CAP, Body mass index (BMI), etc. Diagnostic performance was evaluated using receiver operating characteristic (ROC) curve analysis, and the area under the ROC curves (AUROCs) of LiSA and UGAP were compared using the DeLong test.

Both LiSA and UGAP demonstrated very high technical success rates (both 99.74%), significantly higher than that of CAP (96.84%, p = 0.002). Their measurements showed strong correlations with CAP values (LiSA: r = 0.83; UGAP: r = 0.81, both p < 0.001). For discriminating different steatosis grades (≥S1, ≥S2, =S3), LiSA achieved AUROCs of 0.96, 0.91, and 0.86, respectively, while UGAP achieved AUROCs of 0.96, 0.91, and 0.85, respectively. The DeLong test indicated no statistically significant difference in diagnostic performance between the two techniques across all grades (all p-values >0.62). This study provides, for the first time, CAP-referenced cut-off value suggestions for LiSA in diagnosing S2 and S3 steatosis (269.00 dB/m and 300.00 dB/m, respectively).

LiSA and UGAP show high agreement with CAP in diagnosing and grading hepatic steatosis, with comparable performance to each other. By virtue of their inherent integration with real-time B-mode imaging, they offer higher technical success rates and operational ease, demonstrating superior clinical feasibility compared to CAP. These findings support LiSA and UGAP as effective and practical alternatives or complements to CAP, enriching the toolkit for point-of-care steatosis assessment.

Figures