Purpose:To elucidate whether apparent diffusion coefficient (ADC) values calculated from echo-planar diffusionweighted MR imaging (EPDWI) are useful in the differential diagnosis of ovarian cystic masses. Materials and Methods: EPDWI was performed in 131patients with ovarian cystic masses (54 mature cystic teratomas, 35 endometrial cysts, four other benign cysts, 14 benign neoplasms, and 24 malignant neoplasms). The areas of the highest signal intensity on EPDWI (b ϭ 1000 seconds/mm 2 ) and the lowest ADC values within the cystic component were evaluated. Results:On qualitative and quantitative analyses, mature cystic teratomas tended to show higher signal intensity and had areas of lower ADC values than endometrial cysts and other benign and malignant neoplasms (P Ͻ .005). In vitro scanning of the cystic contents of mature cystic teratomas confirmed that high signal on DWI or low ADC value was attributable to the keratinoid substance within the tumors. The difference in ADC between malignant and benign lesions were significant when mature cystic teratomas and endometrial cysts were included, but was not significant when they were excluded. Conclusion:The ADC value may add useful information to the differential diagnosis of ovarian cystic masses in limited populations, such as those with mature cystic teratomas with a small amount of fat. DIFFUSION-WEIGHTED IMAGING (DWI) and the calculated apparent diffusion coefficient (ADC) were originally used to demonstrate early ischemic change in brain tissue by depicting cytoplasmic edema (1). They have also been applied to differentiate brain tumors, particularly in the diagnosis of epidermoid and lymphoma (2-6). A high signal on DWI or a low ADC value of these tumors have been attributed to intratumoral keratinoid substance in the former, and high cellularity and a high nuclear-to-cytoplasm (NC) ratio without interstitial edematous change in the latter (2-6). Both of these conditions restrict Brownian movement of the free water molecules within the tumor. The ability of high-speed echo-planar imaging (EPI) to minimize artifacts due to respiratory motion has extended the application of DWI and ADC to the diagnosis of breast lesions (7-10) and prostate carcinomas (11,12), and promising preliminary results have been reported. However, there is controversy regarding the usefulness of this technique in cystic ovarian tumors (13-16), particularly as applied to differentiating benign from malignant lesions.In this study we applied echo-planar DWI (EPDWI) and ADC values to 131 cystic ovarian masses and assessed their potential usefulness in the differential diagnosis. MATERIALS AND METHODS SubjectsBetween September 2000 and December 2003, 322 consecutive women underwent MRI for evaluation of known ovarian lesions. Of these patients, 131 (11-75 years old, mean ϭ 33.9 years) underwent surgical resection within 2 weeks after the MRI was performed. Of the 131 patients, 102 had a lateral lesion and 29 had bilateral lesions. In the bilateral cases, the largest lesion was selected for...

Recognition of pseudolesions of the liver at computed tomography (CT) is important because of their close resemblance to primary liver cancers or metastases. Two types of pseudolesion in the noncirrhotic liver include that due to transient extrinsic compression, typically caused by ribs or the diaphragm, and that due to a "third inflow" of blood from other than the usual hepatic arterial and portal venous sources: the cholecystic, parabiliary, or epigastric-paraumbilical venous system. Although the location of both types of pseudolesion are characteristic, their appearances at CT during arterial portography and CT during selective angiography vary from nonenhanced low-attenuation areas to well-enhanced high-attenuation areas, depending on the amount and timing of the inflow and presence or absence of focal metabolic alteration of the hepatocytes. Radiologists need to understand the underlying mechanism of these pseudolesions to better recognize the wide range of their appearances at CT.

During hepatocarcinogenesis, most hepatocellular nodules show deterioration of arterial blood flow before loss of portal blood flow. Vascular imaging of hepatic nodules may predict malignant abnormality via the early loss of hepatic arterial flow seen before portal flow changes.

In patients with intraductal mucin-producing tumors of the pancreas, filling defects are indicative of malignancy. Diffuse main pancreatic duct dilatation greater than 15 mm (main duct type), or any main pancreatic duct dilatation (branch duct type), is strongly associated with malignancy.

Magnetic resonance (MR) cholangiopancreatography (MRCP) is widely used in the evaluation of pancreatobiliary disorders. However, numerous related pitfalls may simulate or mask pancreatobiliary disease. Maximum-intensity-projection (MIP) reconstructed images completely obscure small filling defects and may demonstrate respiratory motion artifacts. T2 weighting may vary with different MR imaging sequences and influence MRCP findings. Incomplete imaging may create confusion regarding ductal anatomy or disease. Furthermore, MRCP yields only static images and thus may fail to depict various anomalies. Limited spatial resolution makes differentiation between benign and malignant strictures with MRCP alone extremely difficult. Susceptibility artifacts may be caused by metallic foreign bodies or gastric-duodenal gas. Fluid accumulation may produce a pseudolesion or pseudostricture, although changing the imaging angle or section thickness may be helpful. Pneumobilia may be misinterpreted as bile duct stones, and true stones may be overlooked. Pulsatile vascular compression can cause pseudo-obstruction of the bile duct. Use of both source and MIP reconstructed images obtained from different angles can help avoid cystic duct-related pitfalls. Repeat MRCP or conventional MR imaging can help avoid pitfalls related to the periampullary region. Segmental collapse of the normal main pancreatic duct may be misinterpreted as stenosis, but administration of secretin is helpful. An awareness of these pitfalls and possible solutions is crucial for avoiding misinterpretation of MRCP images.