目的:比较限制性抗原亲和力酶联免疫法（LAg-Avidity EIA）和集合核酸法评估MSM的HIV-1新发感染率的可行性n 。方法:对2016-2017年云南省13个州（市）MSM哨点监测采用滚雪球方法采集样本进行HIV-1抗体检测，确证阳性样本进行LAg-Avidity EIA检测，HIV-1抗体阴性样本进行集合核酸法检测，分别采用LAg-Avidity EIA新发感染算法和集合核酸法检测新发感染算法计算HIV-1新发感染率，并进行结果的比较。结果:研究样本共计5 363份，其中HIV-1抗体阳性样本407份（含既往阳性177份），HIV-1抗体阴性样本4 956份，完成LAg-Avidity EIA检测211份（91.7%），判为新近感染69份；完成集合核酸法检测4 469份，判定为HIV-1感染窗口期8份。LAg-Avidity EIA估算的2016年和2017年HIV-1新发感染率分别为3.36%和4.84%，集合核酸法估算的HIV-1新发感染率分别为3.27%和3.02%，2种方法估算的HIV-1新发感染率差异无统计学意义。结论:LAg-Avidity EIA和集合核酸法估算得到2016-2017年云南省哨点MSM HIV-1新发感染率一致性较好，2种方法用于HIV-1新发感染率的连续监测会有较好的稳定性。“,”Objective:To assess and compare the performance of limiting-antigen avidity enzyme immunoassay (LAg-Avidity EIA) and pooling PCR in the surveillance for recent infection rates of HIV-1 in men who have sex with men (MSM).Methods:Blood samples were collected from MSM selected through snowball sampling method in sentinel surveillance in 13 prefectures of Yunnan province from 2016 to 2017. The samples were tested for HIV-1 antibody. The confirmed positive samples were tested by LAg-Avidity EIA. The negative samples were tested by pooling PCR. The recent infection rates of HIV-1 were estimated by the algorithm based on LAg-Avidity EIA and pooling PCR respectively. The two results were compared.Results:During 2016-2017, a total of 5 363 blood samples were collected from MSM, in which 407 samples were HIV-1 positive (including 177 positive tested previously) and 4 956 samples were HIV-1 negative. A total of 211 samples(91.7%) were tested by LAg-Avidity EIA, 69 were confirmed to be recent infections. A total of 4 469 samples were tested by pooling PCR, 8 were confirmed to be acute infections. The recent infection rates of HIV-1 from 2016 to 2017 estimated by LAg-Avidity EIA were 3.36% and 4.84%, and the recent infection rates estimated by pooling PCR were 3.27% and 3.02% respectively. The differences in recent infection rates of HIV-1 estimated by the two algorithms were not significant.Conclusions:The recent infection rates of HIV-1 estimated by LAg-Avidity EIA and pooling PCR in sentinel surveillance in MSM in Yunnan had good consistency from 2016 to 2017. Using the two methods might have a better stability in continuous surveillance for recent infection rates of HIV-1.