Magnetic Resonance Imaging Findings of Portal Biliopathy

PURPOSE: We wanted to evaluate the MR imaging findings of portal biliopathy. MATERIALS AND METHODS: We retrospectively reviewed the MR imaging of twelve patients having biliary abnormalities and portal cavernoma. Two radiologists working in consensus analyzed the presence of extrahepatic portal vein obstruction, cavernous transformation, portosystemic collaterals and bile ducts abnormalities that included irregular contour, stricture, ductal dilatation, wall thickening and enhancement of the ductal wall. RESULTS: Extrahepatic portal vein obstruction was replaced by cavernous transformations in all patients. The cavernous transformations depicted on the MR scans were paracholedochal veins in 12 patients and epicholedochal veins in 6 patients. Three different types of biliary abnormalities were present: 1) the pseudocholangiocarcinoma type (3 patients) that resembled extrahepatic cholangiocarcinoma, 2) the varicoid type (6 patients) that was mainly distinguished by the multiple smooth extrinsic impressions along the common bile duct and 3) the mixed type (3 patients). The highly specific findings for the pseudocholangiocarcinoma type disease were thickening and delayed enhancement of the biliary ductal wall. CONCLUSION: The MR findings of portal biliopathy include cavernous transformation of the portal vein and morphological alterations of the biliary tracts such as irregular contour, biliary stricture with proximal ductal dilatation and thickening of the ductal wall with delayed enhancement. Recognition of these MR findings should facilitate the accurate diagnosis of portal biliopathy and prevent the misinterpretation or inappropriate management of this disease as malignancy or sclerosing chloangitis.

Measurement of Lipid Content in Gallbladder Bile Using in- and opposed-phase MR Images and in vivo Proton MR Spectroscopy

PURPOSE: To evaluate the utility of signal intensity differences between in- and opposed-phase MRI and the lipid peak ratio in in-vivo proton MR spectroscopy of the gallbladder as diagnostic tools for measuring the lipid content of gallbladder bile. MATERIALS AND METHODS: Twenty-six normal volunteers underwent MR imaging (FMPSPGR) and in-vivo proton MR spectroscopy of the gallbladder. In all cases the results of liver function tests were normal, as were cholesterol levels, and ultrasonography of the gaubladder revealed nothing unusual. For MRI and MRS a 1.5T unit (Signa Horizon; GE Medical Systems, Milwaukee, U.S.A.) was used. In-phase and opposed-phase coronal-section MR images(FMPSPGR; TR=125 msec, TE=1.8, 4.2 msec) of the gallbladder were obtained, and differences in signal intensity thus determined. For proton MR spectroscopy of the gallbladder, a localized proton STEAM sequence was employed. A single voxel of 1-8 cm3 was placed at the center of the gallbladder cavity, peak areas at 0.8-1.6 ppm (lipid), 2.0-2.4 ppm, 3.2-3.4 ppm, 3.9-4.1 ppm, and 5.2-5.4 ppm were measured by proton MRS and the relative peak area ratios of peak 0.8-1.6 ppm/other peaks were calculated. The degree of correlation between signal intensity differences at MRI and the relative peak area ratio of lipid in proton MRS was estimated using the p-value and Pearson's correlation coefficient. RESULTS: Signal intensity differences ranged from 11.3 to 43.4% (mean, 26+/-8.9%), and the range of lipid peak area ratio at MRS was 0.10-0.97 (mean, 0.66+/-0.21). There was significant correlation between the two measured values (p=0.014, Pearson's correlation coefficient=0.478). CONCLUSION: In normal cystic bile, signal intensity differences at in- and opposed-phase MRI and relative lipid peak area ratios at MRS varied, though both methods could be used diagnostically for measuring the lipid contents of body tissue.