Affiliation:
1. Rocky Vista University Ivins Utah USA
2. Weill Cornell Medicine Cornell University New York New York USA
3. Canon Medical Systems USA Tustin California USA
Abstract
PurposeTo assess diagnostic performance of quantitative ultrasound (QUS) biomarkers in assessing hepatic steatosis.MethodsWe prospectively recruited 125 participants (mean age 54 years) who underwent liver QUS, magnetic resonance imaging (MRI), and laboratory tests within 30 days in this IRB approved study. Based on MRI‐proton density fat fraction (MRI‐PDFF) and MRE, we divided 125 participants into normal liver, nonalcoholic fatty liver (NAFL) and liver fibrosis (≥F1) groups. We examined diagnostic performance of ultrasound attenuation coefficient (AC), normalized local variance (NLV), superb microvascular imaging‐based vascularity index (SMI‐VI), and shear wave velocity (SWV) for determining hepatic steatosis and fibrosis using area under receiver operating characteristic curve (AUC). We also analyzed correlations of QUS biomarkers to MRI using Spearman correlation coefficient.ResultsWe observed significant differences in AC, NLV, and SMI‐VI among the three groups (22 participants with normal liver, 78 with NAFL, and 25 with liver fibrosis). AUC of AC, NLV, and SMI‐VI for determining ≥ mild steatotic livers (MRI‐PDFF ≥5%) was 0.95, 0.90, and 0.92, respectively. AUC of SWV for determining ≥ F1 liver fibrosis was 0.93. The correlation of MRI‐PDFF was positive to AC (r = 0.91) and negative to NLV (r = −0.74), SMI‐VI (r = −0.8) in NAFL group. There was a significant difference in regression slope of AC to MRI‐PDFF in livers with and without ≥F1 (0.84 vs 0.91, P = .02).ConclusionsQUS biomarkers have high sensitivity and specificity to determine and grade hepatic steatosis and detect liver fibrosis. The effect of liver fibrosis on the performance of QUS biomarkers in quantifying liver fat content warrants further investigation.
Funder
Canon Medical Systems USA
Subject
Radiology, Nuclear Medicine and imaging,Radiological and Ultrasound Technology