Liver Imaging Reporting and Data System on CT and gadoxetic acid-enhanced MRI with diffusion-weighted imaging

Objectives: To assess major imaging features of Liver Imaging Reporting and Data System (LI-RADS) on contrast-enhanced CT and gadoxetic acid-enhanced MRI and to estimate whether the combination of signal intensity favouring HCC on hepatobiliary phase (HBP) and diffusion-weighted images (DWI) can act as a major feature in LI-RADS. Methods: Four hundred twenty one patients with 445 observations were included. Major features of LI-RADS on CT and MRI as well as HBP and DWI features were assessed. Diagnostic performances of LR-5 according to LI-RADS v2014 and modified LI-RADS which incorporate combination of HBP and DWI were assessed. Pairwise comparisons of the receiver operating characteristic (ROC) curves were performed. Results: For HCCs, capsule appearance had the highest rate of discordance between CT and MRI (32.7%), followed by washout appearance (22.2%). Specificity (75%) of LR-5 of LI-RADS v2014 was lower than that (77.1–79.2%) of modified LI-RADS. Area under the ROC curve of modified LI-RADS (0.755–0.775) was not significantly different from that of LI-RADS v 2014 (0.709) (p > 0.05). Conclusions: There were substantial discordances between CT and MRI for capsule and washout appearances in hepatic observations, and combination of gadoxetic acid-enhanced MRI and DWI might be able to be incorporated as a major feature of LI-RADS. Key points: • Major imaging features of LI-RADS showed substantial discordances on CT and MRI. • An observation may be categorized differently depending on used imaging exam. • CT and MRI should both be performed for LR-3 and LR-4 observations. • Combination of gadoxetic acid-enhanced MRI and DWI may be a major feature.