Evaluation of pancreatic tumors with positron emission tomography and F-18 fluorodeoxyglucose: comparison with CT and US.

Abstract

PURPOSE: To assess the clinical value of positron emission tomography (PET) with fluorine-18-labeled fluorodeoxyglucose (FDG) for identification of pancreatic carcinoma. MATERIALS AND METHODS: Forty-six patients suspected of having a pancreatic neoplasm and who were to undergo surgery prospectively underwent FDG PET, computed tomography (CT), and transabdominal ultrasound (US). Endoscopic US was performed in 40 patients. Images were independently interpreted and compared with the histopathologic findings at surgery (41 patients) or with clinical follow-up findings (five patients). RESULTS: In 33 of 35 patients, foci of pancreatic carcinomas (10-100 mm in diameter) were identified as an increase in FDG uptake, whereas CT, transabdominal US, and endoscopic US depicted the foci in 31, 31, and 28, cases, respectively. Among 11 benign lesions, nine showed no increased FDG uptake (specificity = 82%). Specificities of the other modalities were lower. False-positive findings were obtained in a case of chronic active pancreatitis and in a serous cystadenoma. CONCLUSION: FDG PET, which provides "biochemical" information, is accurate in identifying pancreatic carcinoma and may be a method of choice when imaging equivocal masses detected with other "anatomic" imaging studies.