Malignant tumors with high glucose metabolic rates accumulate [18F]-fluorodeoxyglucose (FDG), a positron emitting tracer. The aim of this study was to evaluate FDG positron emission tomography (PET) for detection and staging of human cholangiocarcinoma (CC). Patients with adenocarcinoma of the biliary tree (n = 26), with benign lesions of the bile ducts (n = 8), and 20 control patients underwent FDG-PET (370 MBq [18F]-FDG, Siemens ECAT EXACT HR(+)). In a blinded fashion, 4 independent experts evaluated the PET scans visually and semiquantitatively using the standardized uptake value and a tumor/non-tumor ratio. All adenocarcinomas and benign lesions (sclerosing cholangitis, bile duct adenoma, Caroli's disease) were histologically proven and imaged by magnetic resonance imaging and endoscopic retrograde cholangioscopy. True-positive PET scans were obtained in 24 of 26 CC and false-negative scans in the other 2 (sensitivity 92.3%). The PET scan was true-negative in 18 of 20 controls and in all 8 benign biliary lesions (specificity 92.9%). Visual and semiquantitative evaluation using tumor/non-tumor ratios were equally accurate (accuracy 92.6%) whereas evaluation by standardized uptake value revealed lower accuracy (P <.05). Regional or hepatoduodenal lymph node metastases were detected with PET in only 2 of 15 cases whereas distant metastases (peritoneal carcinomatosis, pulmonary metastases) were diagnosed in 7 of 10 cases. In conclusion, PET is highly sensitive and specific for the detection and localization of CC. It can be helpful for diagnosis of distant metastases but is not suitable for detection of regional lymph node metastases.