Diagnostic Performance of Ultrasound Attenuation Imaging in Detecting Hepatic Steatosis Publisher Pubmed



Naghibi H ; Shakiba M ; Azizi N
Source: Journal of Medical Ultrasonics (Singapore) Published:2026

Abstract

Purpose: Hepatic steatosis is a hallmark of metabolic dysfunction-associated steatotic liver disease, and increased hepatic fat (or steatosis) is associated with liver inflammation, fibrosis, and adverse outcomes. While MRI-derived proton density fat fraction (MRI–PDFF) is an accurate noninvasive reference standard, ultrasound attenuation imaging (UAI) offers an accessible cost-effective alternative. We herein systematically review and meta-analyze the accuracy of UAI in grading hepatic steatosis. Methods: We searched Embase, PubMed/Medline, Scopus, and Web of Science through December 22, 2024, for studies comparing UAI with MRI–PDFF. Data on diagnostic performance were extracted from eligible studies. Pooled diagnostic odds ratios (DOR), sensitivity, specificity, likelihood ratios, and areas under the curve (AUCs) were calculated using random-effects models, with heterogeneity assessed through I2. Results: Twelve studies with a total of 3344 patients were included. For differentiating grade 0 from grade ≥ 1 steatosis, the pooled AUC was 0.82 (95% CI 0.78–0.85) and DOR was 21.10 (I2 = 67.2%), with sensitivity and specificity of 83%. For differentiating grades ≤ 1 from grade ≥ 2, the AUC was 0.80 (95% CI 0.73–0.86), DOR 19.87 (I2 = 89%), sensitivity 84%, and specificity 80%. For differentiating grades ≤ 2 from grade 3, the AUC was 0.78 (95% CI 0.66–0.87), DOR 19.21 (I2 = 85.3%), sensitivity 86%, and specificity 80%. Conclusion: UAI exhibits good diagnostic accuracy for grading hepatic steatosis and offers a practical, noninvasive alternative to MRI–PDFF, being particularly well-suited for screening and patient follow-up. Standardized protocols and further multicenter studies are needed to promote its clinical application. © The Author(s), under exclusive licence to The Japan Society of Ultrasonics in Medicine 2025.