Material and Method: Between July 2011 and October 2012, 14 cirrhotic patients with 34 hepatic nodules underwent hepatic Gadoxetic acid-enhanced MR imaging. Two radiologists reviewed in consensus, the two sets of Gadoxetic acid-enhanced MR images in a retrospective fashion. Set 1 of MR images include unenhanced and Gadoxetic acid-enhanced dynamic images, and set 2 of MR images include unenhanced, Gadoxetic acid-enhanced dynamic, and hepatobiliary phase images. The diagnostic performance of the two sets of Gadoxetic acid-enhanced MR imaging in the diagnosis of HCC were assessed. Comparison of sensitivity, specificity, and accuracy between the two imaging sets according to the lesion size were made.

Results: Higher sensitivity (between 95% and 75%) and negative predictive value (between 87.5% and 68.8%) were demonstrated on set 2 of MR images (addition of hepatobiliary phase images). No difference of accuracy (76.5%) between the two imaging sets were found. Excellent sensitivities (100%), specificities (100%), and accuracies (100%) in both imaging sets were shown in lesions more than 2 cm in diameter. Higher sensitivity (between 91.7% and 58.3%) were seen in the diagnosis of small hepatocellular carcinoma (HCC) group (less than 2 cm) with the addition of the hepatobiliary phase images.

Conclusion: Gadoxetic acid-enhanced MR imaging has an excellent diagnostic performance for diagnosis of more than 2 cm-HCCs in cirrhotic liver, but still limited in small HCCs. Added hepatobiliary phase images in dynamic Gadoxetic acid-enhanced MR imaging improve detection of small HCCs.

Keywords: Hepatocellular carcinoma, Gadoxetic acid-enhanced MR imaging, Hepatobiliary phase images