Objective: Fluorine-18 fluorodeoxyglucose (FDG) uptake is frequently observed in lung hilus. This finding causes difficulties during the interpretation. Our objective was to evaluate the features of FDG uptake in lung hilus associated with benign or malignant etiology in patients with thoracic and non-thoracic tumors. Methods: We retrospectively evaluated the files of 1172 patients who had undergone FDG positron emission tomography (PET)/computed tomography (CT) examination between January 2008 and June 2009. Forty-eight patients (21 males, 27 females, age range 12-80 years, mean 60.9 ± 15.82 years) with either unilateral or bilateral hilar FDG uptake and who had thorax contrast-enhanced computed tomography (CECT) performed within 1 month of the FDG PET/CT scan were enrolled in the study. Characteristics of FDG uptake were classified according to the pathology and CECT or PET/CT follow-up over 12 months. Results: The characteristics of 71 hilar regions with FDG uptake could be classified. In 30 of 71 (42.3%) hilar regions, FDG uptake was considered to be physiological because no lymph node was observed on CECT. In 19 of 71 (26.8%), FDG uptake was secondary to benign lymph nodes and in 22 (30.9%) to malignant lymph nodes. Significant differences were observed between benign and malignant lymph nodes for SUVhilus and SUVhilus/SUVliver ratio. Using 4.49 as the cut-off value for SUVhilus, a sensitivity of 85.7% and a specificity of 86.4% were achieved (area under curve, AUC: 0.956). For SUVhilus/SUVliver ratio, sensitivity and specificity to detect malignant lymph nodes were 77.6 and 77.3% (AUC: 0.885), respectively, at a cut-off value of 1.75. Conclusion: SUVhilus and SUVhilus/SUVliver ratio were found to be significant parameters for determining malignancy in lung hilus. Combined interpretation with CECT is warranted during the evaluation of lung hilus with FDG PET/CT. © 2010 The Japanese Society of Nuclear Medicine.