目的:比较n 13N－NHn 3、n 11C－蛋氨酸(MET)及n 18F－脱氧葡萄糖(FDG)PET/CT显像在疑似脑胶质瘤患者诊断与评估中的应用价值。n 方法:纳入2010年9月至2017年12月间中山大学附属第一医院的90例临床怀疑脑胶质瘤患者[男54例，女36例；年龄(40.0±14.0)岁]，均行n 13N－NHn 3、n 11C－MET及n 18F－FDG PET/CT显像。所有患者经组织病理或随访确诊。对获得的图像进行视觉分析[病灶放射性摄取高于对侧正常脑实质为阳性(＋)，等于/低于对侧正常脑实质为阴性(－)]和半定量分析[以病灶(L)最大标准摄取值(SUV)除以对侧白质(WM)平均SUV，得到L/WM比值为指标]。采用受试者工作特征(ROC)曲线分析，计算曲线下面积(AUC)，比较3种显像方法及三者联合应用对脑肿瘤的诊断效能和区分高低级别胶质瘤的能力。n 结果:90例患者中，高级别胶质瘤30例、低级别胶质瘤27例、非胶质瘤脑肿瘤10例、非肿瘤性病变(NNL)23例。视觉分析中，n 13N－NHn 3、n 11C－MET及n 18F－FDG PET/CT显像诊断脑肿瘤的灵敏度分别为62.7% (42/67)、94.0% (63/67)和35.8% (24/67)，特异性分别为95.7% (22/23)、56.5% (13/23)和65.2% (15/23)，准确性分别为71.1%(64/90)、84.4%(76/90)和43.3% (39/90)；三者联合使用时，以n 11C－MET/n 13N－NHn 3/n 18F－FDG的＋/＋/＋、＋/＋/－和＋/－/－代谢模式为肿瘤病变诊断标准，特异性和准确性分别提高至73.9%(17/23)和88.9%(80/90)，灵敏度仍为94.0% (63/67)。半定量分析中，L/WM比值ROC曲线分析示，n 13N－NHn 3和n 11C－MET PET/CT显像诊断脑肿瘤的灵敏度分别为64.2%(43/67)和89.6%(60/67)，特异性分别为100%(23/23)和69.6%(16/23)，AUC分别为0.819和0.840(n z＝－0.316，n P>0.05)。n 13N－NHn 3、n 11C－MET及n 18F－FDG PET/CT显像鉴别高低级别胶质瘤的准确性分别为86.0%(49/57)、87.7%(50/57)和93.0%(53/57)，AUC分别为0.896、0.928和0.964(n z值：－0.554～1.334,均n P>0.05)。n 结论:13N－NHn 3 PET/CT显像鉴别脑肿瘤与NNL特异性高，但灵敏度较低；n 11C－MET检测脑胶质瘤的灵敏度最高，但与n 18F－FDG一样，在NNL中有摄取。n 13N－NHn 3、n 11C－MET及n 18F－FDG PET/CT显像均可用于脑胶质瘤的分级评估，联合使用有助于脑胶质瘤的准确诊断。n “,”Objective:To compare the application of n 13N-NHn 3, n 11C-methionine (MET) and n 18F-fluorodeoxyglucose (FDG) PET/CT imaging in the diagnosis and evaluation of suspected cerebral glioma.n Methods:From September 2010 to December 2017, ninety patients (54 males, 36 females; age: (40.0±14.0) years) in the First Affiliated Hospital of Sun Yat-sen University with suspected glioma based on clinical diagnosis, who underwent n 13N-NHn 3, n 11C-MET and n 18F-FDG PET/CT imaging, were prospectively enrolled in the study. All patients were confirmed by histology or clinical and radiological follow-up. Images were interpreted by visual evaluation (higher radioactive uptake in lesions than that in the contralateral normal brain parenchyma was considered as positive (+ ), equal/lower were considered as negative (-)) and semi-quantitative analysis (the maximum standardized uptake value (SUV) of lesion (L) to the mean SUV of normal white matter (WM) (L/WM ratio)). Receiver operating characteristic (ROC) curve analysis was used and the area under curves (AUCs) were calculated and compared. The diagnostic efficacies of 3 imaging methods and the combination for diagnosing gliomas and the abilities to differentiating high-grade gliomas (HGG) and low-grade gliomas (LGG) were compared.n Results:In 90 patients, 30 HGG, 27 LGG, 10 non-glioma brain tumors and 23 non-neoplastic lesions (NNL) were diagnosed. On visual evaluation, the sensitivities for differentiating tumors from NNL were 62.7%(42/67), 94.0%(63/67) and 35.8% (24/67) for n 13N-NHn 3, n 11C-MET and n 18F-FDG PET/CT respectively, while the specificities were 95.7%(22/23), 56.5% (13/23) and 65.2% (15/23), and the accuracies were 71.1%(64/90), 84.4%(76/90) and 43.3% (39/90). Taking the metabolic patterns of + /+ /+ , + /+ /- and + /-/- (n 11C-MET/n 13N-NHn 3/n 18F-FDG) as the diagnosis standard of tumor lesions, the specificity and accuracy of the combined method increased to 73.9%(17/23) and 88.9%(80/90) with the sensitivity remaining the same (94.0%, 63/67). ROC curve analysis (L/WM) showed that the sensitivity, specificity and AUC were 64.2%(43/67), 100%(23/23) and 0.819 for n 13N-NHn 3 PET/CT, and 89.6%(60/67), 69.6%(16/23) and 0.840 for n 11C-MET PET/CT (n z=-0.316, n P>0.05). The accuracy for differentiating high and low grade glioma were 86.0% (49/57), 87.7%(50/57) and 93.0%(53/57) forn 13N-NHn 3, n 11C-MET and n 18F-FDG PET/CT, with the AUC of 0.896, 0.928 and 0.964, respectively (n z values: -0.554 to 1.334, all n P>0.05).n Conclusions:13N-NHn 3 PET/CT imaging has remarkably high specificity but low sensitivity for the differentiation of brain tumors from NNL. n 11C-MET PET/CT imaging was found to be highly useful for detection of brain tumors. However, like n 18F-FDG, high MET uptake is frequently observed in some NNL. n 13N-NHn 3, n 11C-MET and n 18F-FDG PET/CT imaging all appear to be valuable for evaluating the histological grade of gliomas, and the combination of them is more useful for the accurate diagnosis of glioma.n