RESUMO
OBJECTIVE: To compare the usefulness of magnetic resonance cholangiopancreatography (MRCP) and MR angiography (MRA) in differentiating malignant from benign intraductal papillary mucinous tumors of the pancreas (IPMTs), and to determine the findings which suggest malignancy. MATERIALS AND METHODS: During a 6-year period, 46 patients with IPMT underwent MRCP. Morphologically, tumor type was classified as main duct, branch duct, or combined. The diameter of the main pancreatic duct (MPD), the extent of the dilated MPD, and the location and size of the cystic lesion, septum, and communicating channel were assessed. For all types of IPMTs, enhanced mural nodules and portal vein narrowing were evaluated at MRA. RESULTS: Combined-type IPMTs were more frequently malignant (78%) than benign (42%) (p < 0.05). Compared with benign lesions, malignant lesions were larger, and the caliber of the communicating channel was also larger (p < 0.05). Their dilated MPD was more extensive and of greater diameter (p < 0.05), and the presence of mural nodules was more frequent (p < 0.001). CONCLUSION: Combined MRCP and MRA might be useful for the differential diagnosis of malignant and benign IPMTs of the pancreas.
RESUMO
PURPOSE: To evaluate the accuracy of T2-weighted imaging and MR cholangiopancreatography using the single shot fast spin-echo technique for evaluating pancreaticobiliary disease. MATERIALS AND METHODS: B e t ween March and July 1997, axial and coronal T2-weighted images ( T E : 8 0 -2 00 msec) and MR cholangiopancreatograms(TE:800 -1 2 00 msec) were obtained in two ways [single slab (thickness:30 -50 mm) and multislice acquisition under chemical fat saturation] using SSFSE pulse sequencing in 131 cases of suspected pancreati-cobiliary disease. The accuracy of SSFSE MR imaging was assessed in 89 lesions of 74 patients [male,48; female,26; age range, 3 0 -86 (mean,59)years] confirmed surgicopathologically(50 lesions in 39 patients) and clinically (39 lesions in 35 patients). Two radiologists reviewed the MR images and diagnosis was determined by consensus. RESULTS: Correct diagnosis was confirmed in 84 of 89 lesions (94 %). Seven lesions were falsely interpreted, false positive and false negative results accounting for two and five cases, respective l y. Two pancreatic cancers were misdiagnosed as pancreatitis and a cancer of the proximal common bile duct(CBD) was interpreted as a distal CBD cancer. The sensitivity of SSFSE MR imaging for malignancy was 93 %. One CBD stone revealed by endoscopic retrograde cholangiopancreatography (ERCP) was not detected on MR images. In contrast, a stone in the CBD seen on MR images was not apparent on subsequent ERC P. Sensitivity and specificity for calculous disease were 96 % and 99.7 %, respective l y. A benign stricture of the ampulla of Vater was falsely interpreted as normal, and correct diagnosis was possible in two falsely diagnosed cases when MR images were rev i ewed retrospectively. CONCLUSION: The combination of T2-weighted and cholangiographic images using SSFSE is an accurate method for diagnosing pancreaticobiliary diseases.