PANCREATIC ADENOCARCINOMA - ASSESSMENT OF VASCULAR INVASION WITH HIGH-FIELD MR-IMAGING AND A PHASED-ARRAY COIL

OBJECTIVE. We performed this study to determine the value of MR imagin g for assessing vascular invasion in preoperative staging of pancreati c carcinoma. SUBJECTS AND METHODS. In 73 consecutive patients with an established diagnosis of pancreatic duct adenocarcinoma, spin-echo T1- weighted and breath-hold multiplanar gradient-recalled (MPGR) images w ere obtained at 1.5 T with a phased-array coil before and after bolus injection of gadopentetate dimeglumine. Major peripancreatic vessels w ere evaluated for contiguity with tumor and tumor encasement. All pati ents subsequently underwent surgery. Results of unenhanced and enhance d NIR imaging studies were compared with the histologic findings in th e resected specimens of 49 tumors and with the surgical findings of 24 nonresected tumors. RESULTS. In 39 patients, tumor contiguity with ad jacent vessels was found at pathologic examination. Of the remaining 3 4 patients, 10 had vascular encasement by tumor seen on pathologic exa mination of the resected specimen; the other 24 had unresectable tumor s encasing adjacent vessels at surgical evaluation. In the assessment of vascular tumor contiguity, the accuracy was 85% for unenhanced spin -echo T1-weighted imaging, 87% for enhanced spin-echo T1-weighted imag ing, 67% for unenhanced MPGR imaging, and 69% for enhanced MPGR imagin g. In the assessment of vascular tumor encasement, the accuracy was 91 % for unenhanced spin-echo T1-weighted imaging, 94% for enhanced spin- echo T1-weighted imaging, 74% for unenhanced MPGR imaging, and 76% for enhanced MPGR imaging. CONCLUSION. MR imaging is helpful for preopera tive assessment of vascular involvement caused by pancreatic carcinoma . For this purpose, unenhanced and enhanced spin-echo T1-weighted imag es seem to be more accurate than images obtained with other pulse sequ ences.