f i f A man with heavy alcohol history (daily alcohol ingestion of 160 g for 25 years) presented for evaluation of cteric sclera and epigastric pain. Initial laboratory data were as ollows: white blood cell count, 10,500; hemoglobin, 14.3; plateet count, 121,000; aspartate aminotransferase/alanine aminoransferase, 325/66 IU/L; bilirubin level (total/direct), 26.4/12.8 g/dL; alkaline phosphatase, 177 IU/L; -guanosine triphosphate, 670 IU/L. Tumor markers (alpha-fetoprotein, carcinoembryonic antigen, protein induced by vitamin K absence or antagonist II) were within normal limits except carcinoembryonic antigen 199, 418.3 IU/mL. The computed tomography scan (Figure A) revealed hepatosplenomegaly and surface undulation of the liver, which represent alcoholic liver cirrhosis, and large bulging mass-like lesions (arrowheads) at segments I, IV, and V of the liver. Ascites was also noted. Sequential magnetic resonance imaging revealed mass-like enlargement of hepatic segments I, IV, and V, which were hyperintense on the in-phase (Figure B) and hypointense on opposed-phase (Figure C) chemical shift gradient echo images. These mass-like lesions showed no enhancement on contrast-enhanced dynamic studies. Ultrasonography-guided liver biopsy was performed with an 18-gauge gun needle. Obtained hepatic specimen included the area of hyperechoic mass-like lesion and adjacent hepatic parenchyma in segment V. Following pathologic examinations proved scattered neutrophils, severe fatty liver with hepatocyte swelling, and massive bilirubin pigment, which are consistent with alcoholic steatohepatitis (Figure D). Hepatic steatosis is frequently manifested in alcoholic liver disease with or without alcoholic hepatitis or cirrhosis. Even though the biopsy is considered the reference standard for assessment of hepatic steatosis, most patients do not undergo liver biopsy but are usually diagnosed as hepatic steatosis with ultrasonography.1 Ultrasonography is known as a routine noninvasive technique for the initial study of the liver in various clinical circumstances, but it still has severe diagnostic limitations, both in diffuse liver disease and in focal hepatic steatosis, that might lead to a misdiagnosis of hepatic mass.2 Especially in alcoholic liver disease, focal well-marginated fat deposit can be frequently diagnosed as hepatocellular carcinoma or its premalignant lesions such as regenerative or dysplastic nodules.1 Focal fatty deposition of the liver is usually a benign lesion that is developed by aberrant venous supply or focal metabolic change caused by perfusion decrease. It is usually represented as illdefined low attenuating lesions on computed tomography scan. For assessment of fatty accumulation of the liver, magnetic resonance imaging is widely used including chemical shift gradient echo imaging with in-phase and opposed-phase acquisitions.1,2 Hepatic steatosis might be present if there is a signal intensity loss on opposed-phase images in comparison with in-phase images, and the amount of hepatic fat present can also be quantified by assessing the degree of signal intensity loss.1 We can usually find the normal vascular structure that passes through the focal fatty deposition without distortion in contrast to true tumorous lesion (arrows in Figure A). Unlike at-containing lesions in other organs, the application of chemcal fat saturation sequences can be neither sensitive nor specific or the detection of fatty liver.1,2