Pancreatic adenocarcinoma is a lethal malignan- cy that is rarely detected in an early stage. CT has been the primary modality used to identify mass lesions of the pancreas when there is a suspicion of adenocarcinoma but has limited sensitivity and specificity. 1-5 CT detection of local vascular involve- ment or distant metastases generally precludes efforts at surgical resection, but sensitivity is, again, limited. Contrasted helical CT has higher resolution than traditional CT and may offer improved accura- cy for local and distant tumor detection and stag- ing. 6,7 New technologies may improve diagnostic accuracy for detection of pancreatic cancer, local vascular invasion, and distant metastases. EUS is useful for evaluation of pancreatic can- cers. 8-12 Placement of the ultrasound transducer within a centimeter of the pancreas provides excel- lent sonographic images. Fine-needle aspiration of pancreatic cancers has diagnostic sensitivity rates as high as 80% in the hands of experienced examin- ers. 13-17 The sensitivity of EUS is superior to CT for the diagnosis of vascular involvement. 5,8,10,18 This evaluation is crucial for determining T stage and feasibility of operative resection. Positron emission tomography (PET) with 18 fluo- ro-deoxyglucose (FDG) exploits increased glucose metabolism by malignant tumors. Increased glucose utilization can be demonstrated for most solid tumors, including pancreatic adenocarcinoma. 19 Cell membrane glucose transport proteins and intracel- lular hexokinase levels are high in tumors, leading to increased uptake and phosphorylation of glucose and FDG. 20-22 The 18 FDG emits positrons that collide with electrons, creating two gamma rays at 180- degree angles. A ring detector reconstructs the point of origin to generate images. 23 The 18 FDG has proven useful in imaging a variety of solid tumors, for both primary tumors and metastatic disease. 24-27 Primary pancreatic tumors also can be diagnosed with FDG-PET scanning with high sensitivity. 28 The sensitivity of helical CT, EUS, and 18 FDG- PET scanning for identification of pancreatic adeno- carcinoma was prospectively compared. 18 FDG-PET scanning and helical CT were compared for M stag- ing (metastatic disease) accuracy. EUS and helical CT were compared for local T staging accuracy, par- EUS, PET, and CT scanning for evaluation of pancreatic adenocarcinoma Howard R. Mertz, MD, Panos Sechopoulos, MD, Dominique Delbeke, MD, Steven D. Leach, MD Nashville, Tennessee Background: Preoperative diagnosis of pancreatic adenocarcinoma can be difficult. Computed tomography (CT) is the standard, noninvasive imaging method for evaluation of suspected pan- creatic adenocarcinoma, but it has limited sensitivity for diagnosis, local staging, and metastases. Endoscopic ultrasound (EUS) and fluoro-deoxyglucose/positron emission tomography (FDG-PET) are imaging methods that may improve diagnostic accuracy. Methods: Thirty-five patients with presumed resectable pancreatic adenocarcinoma were prospectively evaluated with helical CT, EUS, and FDG-PET. Results: Sensitivity for the detection of pancreatic cancer was higher for EUS (93%) and FDG-PET (87%) than for CT (53%). EUS was more sensitive than CT for local vascular invasion of the portal and superior mesenteric veins. EUS diagnosis of vascular invasion was associated with poor out- come after surgery. EUS-guided, fine-needle aspiration allowed tissue diagnosis in 14 of 21 attempts (67%). FDG-PET diagnosed 7 of 9 cases of proven metastatic disease, 4 of which were missed by CT.Two of three metastatic liver lesions suspected by CT were indeterminate for metas- tases. FDG-PET confirmed metastases. Conclusions: EUS and PET improve diagnostic capability in pancreatic adenocarcinoma. EUS is useful in determining local vascular invasion and obtaining tissue diagnosis. FDG-PET is useful in identifying metastatic disease. Both techniques are more sensitive than helical CT for identifica- tion of the primary tumor. (Gastrointest Endosc 2000;52:367-71). VOLUME 52, NO. 3, 2000 GASTROINTESTINAL ENDOSCOPY 367 Received February 2, 2000. For revision March 16, 2000. Accepted April 7, 2000. From the Department of Medicine, Division of Gastroenterology, Departments of Radiology and Radiologic Science and Surgery, Vanderbilt University, Nashville, Tennessee. Presented at the annual meeting of the American Society for Gastrointestinal Endoscopy, Orlando, Florida, May 18, 1999. Reprint requests: Howard R. Mertz, MD, 1501 TVC, Vanderbilt University Medical Center, Nashville, TN 37232-5340. Copyright © 2000 by the American Society for Gastrointestinal Endoscopy 0016-5107/2000/$12.00 + 0 37/1/107727 doi:10.1067/mge.2000.107727