How to Differentiate Focal Nodular Hyperplasia From Hepatocellular Adenomas Diagnostic Accuracy of MRI in Differentiating Hepatocellular Adenoma From Focal Nodular Hyperplasia: Prospective Study of the Additional Value of Gadoxetate Disodium

Objective: To determine if the hepatobiliary phase of gadoxetate disodium-enhanced hepatic MRI increases the sensitivity in differentiating focal nodular hyperplasia from hepatocellular adenomas. Design: Prospective analysis. Methods: This study was comprised of 52 patients (2 men, 50 women) with suspicion of hepatocellular adenomas or with focal nodular hyperplasia on a previous CT or conventional gadolinium-enhanced MRI. All lesions measured at least 2 cm to avoid potential sampling error on subsequent biopsy. Exclusion criteria included contraindication to MRI, pregnancy, hemochromatosis, chronic hepatitis, cirrhosis, previous malignancy or metastatic disease, and elevated alpha fetoprotein or CEA levels. All patients underwent gadoxetate disodium-enhanced hepatic MRI. MRI examinations were performed using the 1.5T system. Sequences included 2D T1-weighted gradient echocardiography inand out-of-phase, T2-weighted images, and diffusion-weighted sequence using b values of 50, 400, and 800 s/mm2; 3D T1-weighted gradient echocardiographic dynamic images were acquired before and after intravenous contrast administration at 30, 60, 90, and 180 seconds. Hepatobiliary phase images were acquired 20 minutes after injection. The images acquired before the hepatobiliary phase were designated as the "standard" images, and images were reviewed by 2 radiologists. The standard images were reviewed separately from the hepatobiliary phase images. The largest or most suitable lesion for biopsy had histopathologic confirmation. The lesion shape, signal intensity, and enhancement pattern were recorded, as was the presence of fat, hemorrhage, and central scar. The standard of reference was the histopathologic diagnosis. Results: There were 24 hepatocellular adenomas and 28 focal nodular hyperplasias. Using the standard sequences, MRI had a sensitivity of 50% in characterizing hepatocellular adenomas and 64% sensitivity in characterizing focal nodular hyperplasia. The average lesion diameter was 7.1 cm. Using the hepatobiliary phase of gadoxetate disodium-enhanced MRI, there was 96% sensitivity in characterizing hepatocellular adenomas and focal nodular hyperplasia. Bleeding, fat, and glycogen were more predictive of hepatocellular adenomas, while a central scar was more predictive of focal nodular hyperplasia. Conclusions: Hepatobiliary phase of gadoxetate disodium-enhanced MRI increases the sensitivity in differentiating focal nodular hyperplasia from hepatocellular adenomas measuring >2 cm. Reviewer's Comments: The results of this study are useful in demonstrating that the increased sensitivity of the hepatobiliary phase of gadoxetate disodium-enhanced hepatic MRI in characterizing hepatocellular adenomas and focal nodular hyperplasia is important, as the management options differ. One of the limitations in this study was that there was a higher prevalence of focal nodular hyperplasia than that found in the general population due to selection bias. (Reviewer-John C. Sabatino, MD).