Contrast-enhanced US of High-Risk Indeterminate Focal Liver Observations Categorized as LR-4 or LR-M at CT/MRI

Background: Indeterminate focal liver observations in patients at risk for hepatocellular carcinoma (HCC) may require invasive biopsy or follow-up, which could lead to delays in definitive categorization and to postponement of treatment. Purpose: To examine clinical effect of contrast-enhanced US (CEUS) in participants with high-risk indeterminate liver observations categorized as Liver Imaging Reporting and Data System (LI-RADS) category LR-4 (probably HCC) or LI-RADS category LR-M (probably or definitely malignant but not HCC specific) at CT or MRI. Materials and Methods: This was a secondary analysis of a prospective international multicenter validation study for CEUS LI-RADS (January 2018 to August 2021). CEUS was performed within 4 weeks of CT or MRI. Tissue histologic and CT or MRI follow-up data were used as reference standards. Clinical effect of CEUS for HCC was evaluated in observations 10 mm or larger categorized as CT/MRI LR-4 and LR-M. Results: Included were 109 participants (mean age, 64.3 years +/- 8.3 [SD]; 68.8% [75 of 109] male participants) with 113 observations (>= 10 mm) categorized as CT/MRI LR-4 (53.1%; 60 of 113) or LR-M (46.9%; 53 of 113). CEUS resulted in management recommendation changes in 33.6% (95% CI: 25, 43; 38 of 113) of observations; among these, 95% (95% CI: 82, 99; 36 of 38) were correct. A total of 30.1% (34 of 113) of CT/MRI LR-4 and LR-M observations were categorized at CEUS as LI-RADS category LR-5 (definite HCC), making biopsy unnecessary; 94% (32 of 34) of these categorizations were correct. Of CT/MRI LR-4 observations, 7% (four of 60) were categorized as CEUS LR-M; subsequent biopsy confirmed non-HCC malignancy in all participants. Clinical impact of CEUS was more substantial for observations 20 mm or larger (n = 68); CEUS helped appropriately categorize both LR-5 and LR-M lesions as HCC and non-HCC malignancies, respectively, and resulted in management recommendation changes in 40% (27 of 68) of observations with 100% accuracy. Conclusion: CEUS resolved some high-risk indeterminate liver observations (categorized as LR-4 and LR-M at CT or MRI), with particularly high clinical impact for observations measuring at least 20 mm.