Gadoxetic acid-enhanced MR imaging facilitates the differentiation of FNH from HCA.

Purpose: To assess the value of hepatobiliary phase gadoxetic acid (EOB)-enhanced magnetic resonance imaging (MRI) for the diagnosis of early stage hepatocellular carcinoma (HCC) (<3 cm) compared to triple-phase dynamic multidetector computed tomography (MDCT). Materials and Methods:In all, 52 patients with 60 pathologically proven HCCs underwent both EOBenhanced MRI and triple-phase dynamic MDCT. Two radiologists independently and blindly reviewed three image sets: 1) MDCT, 2) dynamic MRI (unenhanced and EOBenhanced dynamic MR images), and 3) combined MRI (dynamic MRI þ hepatobiliary phase images) using a fivepoint rating scale on a lesion-by-lesion basis. Receiver operating characteristics (ROC) analysis was performed, and sensitivity and specificity were calculated. Results:The area under the ROC curve (Az) of dynamic MRI was equivalent to that of MDCT for both readers. For both readers, Az and sensitivity of combined MRI for smaller lesions (<1.5 cm) were significantly higher than that of dynamic MRI and MDCT (P < 0.0166). The majority of false-negative nodules on dynamic MRI or MDCT (75% and 62%, respectively) were due to a lack of identified washout findings.Conclusion: Hepatobiliary phase images can increase the value of EOB-enhanced MRI in the diagnosis of early stage HCC. The sensitivity and accuracy were significantly superior to MDCT for the diagnosis of lesions less than 1.5 cm.

Purpose: To evaluate whether using MR fluoroscopic triggering technique and slow rate injection improves the quality of arterial phase images in gadoxetic acid-DTPAenhanced (Gd-EOB-DTPA) MR imaging because of proper acquisition timing and reduction of artifacts. Materials and Methods:Two hundred sixteen patients undergoing examination for liver diseases were retrospectively reviewed. All MR images were obtained with two Gd-EOB-DTPA injection protocols: (i) a combination protocol, in which the MR fluoroscopic triggering technique and slow rate injection (1 mL/s) were used; and for comparison, (ii) a conventional protocol, in which adjusted fixed scan delay and ordinary rate injection (2 mL/s) were adopted. Signal-to-noise ratio (SNR) of aorta, portal vein, and liver parenchyma on arterial phase images were calculated. Two blinded readers independently evaluated the obtained arterial phase images in terms of acquisition timing and degree of artifacts.Results: The SNRs of aorta and portal vein on arterial phase images were significantly higher in the combination protocol group (aorta/portal: 221.9 6 91.9/197.1 6 89.8) than that in the conventional protocol group (aorta/portal: 169.8 6 97.4/92.7 6 48.5) (P < 0.05). The acquisition timing for arterial phase images with the combination protocol was significantly better than that with the conventional protocol (P < 0.01). The image quality of the combination protocol was significantly higher than that of the conventional protocol (P < 0.01). The occurrence rate of moderate or severe degree of artifacts in the conventional protocol (38.0%) was more prominent than that in the combination protocol (18.5%). Conclusion:The combination of the MR fluoroscopic triggering technique and slow rate injection provides proper arterial phase images and reduces the artifacts in Gd-EOB-DTPA MR imaging.

Dynamic imaging with Gd-BOPTA provides the same information as with conventional gadolinium-based extracellular contrast agents, while late-phase imaging gives additional information for lesion identification and characterization.