Abstract: Diagnostic Value of Qualitative and Quantitative Analysis by F-FDG PET/CT forMediastinal Lymph Nodes of Lung CancerWenchao MA,Wengui XU, Xiang ZHU, JianWANG, Dong DAI, Xiuyu SONG, Lei ZHU, Yanjia ZHUCorrespondence to:Wengui XU, e-mail wenguixy@tom.comKey Laboratory of Breast Cancer Prevention and Therapy, Tianjin Medical University, Ministry of EducationDepartment of Molecular Imaging and Nuclear Medicine, Tianjin Medical University Cancer Institute and Hospital, The Joint Laborato-ry of Molecular Imaging in Oncology, Tianjin Cancer Hospital of Tianjin Medical University and Institute of High Energy Physics ofChinese Academy of Sciences, Tianjin 300060, ChinaThis study was supported by the Key Project of Tianjin Natural Science Foundation (08JCZDJC23700) and Research Project of TianjinEducation Commission (20080133)Abstract Objective: To discuss the diagnostic value of qualitative and quantitative analysis by 18F-FDG PET/CT for mediastinallymph nodes of lung cancer. Methods: The results of CT, PET and PET/CT imaging scan for 142 patients with space occupying lesionsof the lung, who underwent18F-FDG PET/CT before surgery between April 2005 and October 2009 in Tianjin Medical University Can-cer Institute and Hospital were retrospectively analyzed. No anti-tumor treatment was conducted before PET/CT examination in 142 pa-tients. The surgical resection or puncture biopsy was performed a month later. The number of mediastinal node stations was evaluatedin nine groups based on the standard of lymph node station of lung cancer by Mountain and Dresler and the AJCC cancer staging manu-al. The surgical and histological findings served as the reference standard. Chi square test and exact probability were used to comparethe diagnostic value of simple CT, the 18F-FDG uptake of the mediastinal lymph node by visual observation, visual observation afterconsideration of the symmetric distribution and negative calcification, and simple semiquantitative method and the combination of thelatter two for the whole mediastinal lymph nodes and every station. Results: The stations #2, #4 and #7 were the predilection site of ma-lignant mediastinal lymph nodes. There was a high probability of finding out malignant nodes at #1, #2, #3 and #8 stations by 18F-FDGPET/CT. Combination of visual observation and semiquantitative method was the best method, especially after considering that calcifi-cation and symmetric distribution was benign. The sensitivity, specificity, accuracy and the positive and negative predictive value were79.2%, 83.8%, 81.9%, 78.5% and 84.3% for the detection of the mediastinal lymph node metastases. The accuracy was significantlyhigher than that of simple CT, visual observation of the 18F-FDG uptake of the lymph nodes , visual observation after considering calcifi-cation and symmetric distribution was benign, and simple semiquantitative method in determining mediastinal lymph node metastasis (χ2= 44.678, P = 0.000 ). Conclusion: Visual observation is an ideal way of diagnosing the metastasis of the mediastinal lymph nodes oflung cancer, and the SUVmax (the maximum of the standard uptake value) is a useful semiquantitative indicator. The effect of visual ob-servation combining with semiquantitative method is more satisfactory.Keywords18F-FDG; PET/CT; Lung cancer; Mediastinal lymph node; Qualitative and quantitative analysis