目的: 比较1990年和2022年美国风湿病学会(American College of Rheumatology，ACR)大动脉炎分类标准在中国人群中的诊断效能。方法: 回顾性分析2012年5月至2022年5月就诊于北京大学第三医院且病例资料完善的大动脉炎患者及动脉粥样硬化所致动脉狭窄或闭塞患者的临床资料和影像学检查。以2位风湿免疫科专科医师的诊断为金标准，比较以上两种分类标准的敏感性、特异性、阳性预测值、阴性预测值、准确度及受试者工作特征曲线(receiver operating characteristics，ROC)的曲线下面积(area under ROC curve，AUC)。此外，还尝试将新的影像学方法，如彩色多普勒超声(color Doppler ultrasound，CDUS)、计算机断层血管造影(computed tomography angiography，CTA)、磁共振血管造影(magnetic resonance angiography，MRA)和18F-氟脱氧葡萄糖(18F-fluorodeoxyglucose，18F-FDG)正电子发射断层扫描(positron emission tomography，PET)/计算机断层扫描显像(computed tomography，CT)加入1990年分类标准的影像学检查中，以了解影像学方法的改进对该分类标准诊断效能的影响，并比较大动脉炎和动脉粥样硬化患者影像学表现的差异。结果: 2022年ACR大动脉炎分类标准在灵敏度(91.75%)、阳性预测值(94.68%)、阴性预测值(92.79%)、准确度(93.66%)和AUC(0.979)方面均优于1990年ACR大动脉炎分类标准(45.36%、91.67%、66.24%、72.20%、0.855)。将CDUS、CTA、MRA和PET/CT纳入1990年ACR大动脉炎分类标准的影像学检查后，其灵敏度、阳性预测值、阴性预测值、准确度和AUC均有大幅提高，分别为63.92%、92.54%、74.64%、80.49%和0.959，但仍低于2022年ACR大动脉炎分类标准(P < 0.001)。大动脉炎组影像学表现为动脉狭窄者更多(P=0.030)，动脉粥样硬化组动脉闭塞更多(P=0.021)，动脉瘤形成或动脉夹层在两组间差异无统计学意义(P=0.171)。大动脉炎组受累血管数量≥3者更多(P=0.013)，而动脉粥样硬化组单一血管受累者更多(P=0.011)。结论: 与1990年ACR大动脉炎分类标准相比，2022年的ACR分类标准诊断效能更高，可能更适合中国人群；纳入更多成像方式可以提高1990年ACR分类标准的诊断效能。

Objective: To compare the diagnostic efficiency of the 1990 American College of Rheumatology (ACR) classification criteria for Takayasu arteritis (TA) and the 2022 ACR classification criteria for TA in Chinese populations. Methods: The clinical and imaging data of TA patients and patients with arterial stenosis or occlusion caused by atherosclerosis who were admitted to Peking University Third Hospital from May 2012 to May 2022 were retrospectively analyzed. Clinical diagnosis of TA by two rheumatologists were defined as the gold standard. The sensitivity, specificity, positive predictive value, negative predictive value, accuracy and the area under the receiver operating characteristics (ROC) curve (AUC) of the above two classification criteria were compared. In addition, this study also attempted to apply new imaging modalities, such as color Doppler ultrasound (CDUS), computed tomography angiography (CTA), magnetic resonance angiography (MRA) and 18F-fluorodeoxyglucose (18F-FDG) positron emission tomography (PET)/computed tomography (CT) in the 1990 ACR classification criteria to find whether this approach would improve the diagnostic efficiency. At the same time, the imaging features of the two groups were compared. Results: The sensitivity (91.75%), positive predictive value (94.68%), negative predictive value (92.79%), accuracy (93.66%) and AUC (0.979) of the 2022 ACR TA classification criteria were better than those of the 1990 ACR TA classification criteria (45.36%, 91.67%, 66.24%, 72.20% and 0.855, respectively). In addition, we included new imaging modalities, such as CDUS, CTA, MRA and PET/CT in the 1990 ACR TA classification criteria, and the sensitivity, positive predictive value, negative predictive value, accuracy and AUC were significantly improved, which were 63.92%, 92.54%, 74.64%, 80.49% and 0.959, respectively, but still lower than those of the 2022 ACR classification criteria of TA (P < 0.001). The TA patients had more arterial stenosis (P=0.030), while the atherosclerosis patients had more arterial occlusion (P=0.021). There was no significant difference in arterial aneurysm or dissection (P=0.171). The TA patients had more involvement of ≥3 arteries (P=0.013), while the atherosclerosis patients had more unique artery involvement (P=0.011). Conclusion: Compared with the 1990 ACR classification criteria for TA, the 2022 ACR classification criteria had higher diagnostic efficiency and might be more sui-table for the Chinese populations. Using more imaging modalities would improve the diagnostic perfor-mance of 1990 ACR classification criteria.