Evaluation of Hepatic Steatosis and Fibrosis in Steatotic Liver Disease Ultrasound-Derived Fat Fraction (UDFF) and Auto pSWE by Using Deep Abdominal Transducer (DAX) and Liver Biopsy Correlation.

Abstract

Objectives: To evaluate the diagnostic performance of ultrasound-derived fat fraction (UDFF) in detecting and grading hepatosteatosis using liver histology as the reference, and to assess the effectiveness of point shear wave elastography (pSWE), UDFF, and auto-pSWE in diagnosing steatohepatitis and detecting fibrosis in metabolic dysfunction-associated steatotic liver disease (MASLD).

Methods: In this prospective study, patients underwent liver biopsy following UDFF and pSWE measurements using deep abdominal transducer (DAX) and conventional abdominal probes by 2 operators. Imaging findings were compared with histopathology to assess diagnostic performance. Associations between nonalcoholic fatty liver disease activity score (NAS), fibrosis stage, and imaging parameters were also evaluated.

Results: A total of 121 patients were included. The median age was 50 years, and 57 (41%) were male. Histology confirmed steatotic liver disease in 43 cases. Excellent interobserver agreement was observed for UDFF (ICC = 0.974), pSWE (ICC = 0.958), and auto-pSWE (ICC = 0.960). UDFF showed a stepwise increase with steatosis grade and was moderately correlated with histological fat content (r = 0.676 for Sonographer 1, r = 0.638 for Sonographer 2; P < .001). For detecting S ≥ 1 steatosis, the optimal UDFF thresholds were ≥8.4% (Area Under the Curve (AUC) = 0.968; Se = 92.5%, Sp = 93.2%) for Sonographer 1 and ≥8.6% (AUC = 0.951; Se = 90.0%, Sp = 86.3%) for Sonographer 2. For moderate steatosis (S ≥ 2), the cutoffs were ≥10.4% (AUC = 0.932; Se = 100%, Sp = 79.6%) and ≥10.6% (AUC = 0.925; Se = 100%, Sp = 76.5%), and for severe steatosis (S = 3), ≥18.3% (AUC = 0.961; Se = 100%, Sp = 77.4%) and ≥18.7% (AUC = 0.949; Se = 100%, Sp = 77.4%) for Sonographer 1 and 2, respectively. UDFF positively correlated with body mass index and subcutaneous fat thickness, and negatively with both pSWE and auto-pSWE. A strong correlation was observed between pSWE and auto-pSWE for both observers. A weak positive correlation was found between NAS and auto-pSWE in MASLD cases. The optimal thresholds to detect fibrosis (≥F1) were 5.05 and 4.95 kPa for Sonographer 1, and 5.05 and 4.85 kPa for Sonographer 2, for pSWE and auto-pSWE measurements, respectively.

Conclusion: DAX-derived UDFF and auto-pSWE are reproducible, noninvasive biomarkers with strong diagnostic value in assessing steatosis and fibrosis in MASLD.