PURPOSE: The aim of our study was to compare the accuracy of computed tomography and endoscopic ultrasound (EUS) in pre-operative staging of pancreatic cancer. METHODS: Comparative retrospective study of 86 patients with pancreatic cancer. CT was done in 55 patients, 41 patients were examined by EUS. Each patient underwent surgery and had proven pancreatic cancer by histology. CT and EUS results were correlated to per-operative and histological findings. The main attention was paid to the description of peri-pancreatic lymphadenopathy, para-aortic lymphadenopathy, peri-coeliac lymphadenopathy and tumor relationship to superior mesenteric vein, superior mesenteric artery, portal vein, inferior caval vein and common hepatic artery. A description rate was defined as number of pre-operative findings where the structures and relationships mentioned above were described. RESULTS: The description rates of peri-pancreatic lymph nodes were 11 (20%) at CT and 36 (88.0%) at EUS. Para-aortic lymphadenopathy was described in 9 (16.0%) cases at CT and none at EUS. Peri-coeliac lymphadenopathy was mentioned only one time (2.0%) at CT contrary to 12 (29.0%) at EUS. Relationship of the tumor to the mesenteric vessels was well depicted in nine (16.0%) at CT versus nine (22.0%) at EUS. Portal vein relationship was well described in two (4.0%) at CT and seven (17%) cases at EUS. This description rate in vena cava inferior was one (2%) at CT and three (7.0%) at EUS, in hepatic artery it was one (2%) at CT and six (15%) at EUS. In the group of CT, resectability or non-resectability were well predicted in 33 (60%) patients and wrong predicted in 22 (40%) patients. In the group of EUS, resectability or non-resectability were well predicted in 34 (83%) patients and wrong predicted in 7 (17%) patients. CONCLUSION: According to our study, EUS is more accurate in prediction of local PC resectability than CT.
PURPOSE: The aim of our study was to compare the accuracy of computed tomography and endoscopic ultrasound (EUS) in pre-operative staging of pancreatic cancer. METHODS: Comparative retrospective study of 86 patients with pancreatic cancer. CT was done in 55 patients, 41 patients were examined by EUS. Each patient underwent surgery and had proven pancreatic cancer by histology. CT and EUS results were correlated to per-operative and histological findings. The main attention was paid to the description of peri-pancreatic lymphadenopathy, para-aortic lymphadenopathy, peri-coeliac lymphadenopathy and tumor relationship to superior mesenteric vein, superior mesenteric artery, portal vein, inferior caval vein and common hepatic artery. A description rate was defined as number of pre-operative findings where the structures and relationships mentioned above were described. RESULTS: The description rates of peri-pancreatic lymph nodes were 11 (20%) at CT and 36 (88.0%) at EUS. Para-aortic lymphadenopathy was described in 9 (16.0%) cases at CT and none at EUS. Peri-coeliac lymphadenopathy was mentioned only one time (2.0%) at CT contrary to 12 (29.0%) at EUS. Relationship of the tumor to the mesenteric vessels was well depicted in nine (16.0%) at CT versus nine (22.0%) at EUS. Portal vein relationship was well described in two (4.0%) at CT and seven (17%) cases at EUS. This description rate in vena cava inferior was one (2%) at CT and three (7.0%) at EUS, in hepatic artery it was one (2%) at CT and six (15%) at EUS. In the group of CT, resectability or non-resectability were well predicted in 33 (60%) patients and wrong predicted in 22 (40%) patients. In the group of EUS, resectability or non-resectability were well predicted in 34 (83%) patients and wrong predicted in 7 (17%) patients. CONCLUSION: According to our study, EUS is more accurate in prediction of local PC resectability than CT.