Abstract

The purpose of this study was to assess the role of spiral computed tomography (SCT) and magnetic resonance imaging (MRI) in the preoperative work-up of patients with pancreatic carcinoma, regarding local resectability and vascular involvement.A total of 28 patients (19 men and 9 women; mean age 58 years) with known or highly suspected carcinoma of the pancreas were included in this study. All patients prospectively underwent MRI ( +/- gadolinium-DTPA ) and SCT (3-phase examination) as preoperative diagnostic imaging studies, and laparotomy was carried out within 7 days, irrespective of the MRI or SCT findings. SCT and MR studies were reviewed independently by two radiologists, without knowing the results of the surgical exploration. Standardized image analysis was correlated with findings at laparatomy.Laparotomy identified 10 patients to be suitable for surgical resection and 18 pancreatic carcinomas to be unresectable. In 17 of 18 non-resectable carcinomas MRI and SCT were able to obtain correct information about unresectability (sensitivity 94%), in 7 (MRI), resp. 8 (SCT) carcinomas were correctly considered to be resectable (sensitivity 70% of MRI and 80% for SCT). The presence of vascular involvement was depicted by SCT with a sensitivity of 82-100% and 62-100% by MRI. The specificity varied between 85-100% for SCT and 77-100% for MRI.Both MRI and SCT are good techniques for the preoperative work-up of pancreatic carcinomas in order to obtain a correct assessment of local resectability.

View details for PubMedID 8778925