Additional value of 10-min far-delayed phase in extracellular contrast agent-enhanced MRI for diagnostic performance of hepatocellular carcinoma based on LI-RADS v2018

ObjectivesTo evaluate the added value of a 10-min far-delayed phase (TDP) in extracellular contrast agent (ECA)-enhanced MRI for diagnosis of hepatocellular carcinoma (HCC) based on the Liver Imaging Reporting and Data System (LI-RADS) in patients at high risk for HCC.MethodsWe included 166 patients at high risk for HCC who underwent ECA-enhanced MR imaging in this prospective study. Two board-certificated abdominal radiologists blinded to clinical history and final diagnosis reviewed the images based on LI-RADS v2018 criteria. Imaging features and diagnostic performance for HCC was compared before and after adding TDP. We categorized lesions with targetoid imaging features including rim-APHE, target DWI, centripetal enhancement, and peripheral washout, which were defined as LR-M in LI-RADS. And imaging differences between targetoid HCCs and non-HCC malignancies were also calculated.And we calculated the differences in imaging features between targeted HCC and non-HCC malignancies in ECA-enhanced MRI.MethodsWe included 166 patients at high risk for HCC who underwent ECA-enhanced MR imaging in this prospective study. Two board-certificated abdominal radiologists blinded to clinical history and final diagnosis reviewed the images based on LI-RADS v2018 criteria. Imaging features and diagnostic performance for HCC was compared before and after adding TDP. We categorized lesions with targetoid imaging features including rim-APHE, target DWI, centripetal enhancement, and peripheral washout, which were defined as LR-M in LI-RADS. And imaging differences between targetoid HCCs and non-HCC malignancies were also calculated.And we calculated the differences in imaging features between targeted HCC and non-HCC malignancies in ECA-enhanced MRI.ResultsA total of 190 hepatic observations (151 HCCs, 33iCCA, 6cHCC-CCA) were included. TDP can improve detection of the enhancing capsule (72.8% VS. 80.1%, p = 0.001) and mosaic architecture (26.5% VS. 33.1%, p = 0.002). The group with TDP had superior sensitivity (73.5% VS. 76.8%, p = 0.025) and accuracy (76.8% VS 80.0%, p = 0.025) in HCC diagnosis compared with the group without TDP using LI-RADS v2018 criteria. For LR-M observations, the addition of TDP enabled better identification of enhancing capsule in LR-M HCCs (55.2% VS. 68.9%, p = 0.045) and peripheral washout appearance (15.2% VS. 33.3%, p = 0.014) in non-HCC malignancies.ConclusionAdding a 10-min far-delayed phase in ECA-enhanced MRI protocol may improve HCC detection sensitivity and characterization of targetoid HCCs in patients at risk for HCC.