目的探讨梅毒逆向筛查程序方案,分析其在实验室诊断及梅毒筛查防治中的应用价值。方法应用梅毒螺旋体明胶颗粒凝集试验(TPPA)、梅毒螺旋体抗体酶联免疫吸附试验(TP-ELISA)、甲苯胺红不加热血清学试验(TRUST),同时对2015年1至6月杭州师范大学附属医院各门诊及病房梅毒筛查者23 389份标本进行检测。以TPPA为标准,方法间一致性比较采用Kappa检验,通过各方法的灵敏度、特异性、总符合率的比较,分析TP-ELISA作为初筛方法的梅毒逆向筛查程序的可行性,并进一步统计TP-ELISA的S/CO值与TPPA阳性结果的相关性,简化逆向筛查程序。结果 23 389份标本中,TPPA筛查阳性率为3.38%(790份),TP-ELISA、TRUST筛查阳性率分别为3.63%(849份)、1.89%(443份)。TP-ELISA与TPPA在梅毒筛查时具有非常好的一致性(K=0.954),TRUST与TPPA一致性一般(K=0.693),TP-ELISA的灵敏度为99.1%,大于TRUST的灵敏度54.7%,差异有统计学意义(P<0.01)。统计发现,当TP-ELISA的S/CO值≥5.0时,其结果与TPPA结果完全一致;当TP-ELISA的S/CO值分布在1.0~4.99之间时,有66例假阳性;当TP-ELISA的S/CO值≤1.0时,所有样本中存在7例假阴性(6例样本TRUST阳性)。结论 TRUST敏感度较低,单独作为梅毒筛查不利于大量隐性梅毒患者的发现;TP-ELISA灵敏度和特异性较高,与TPPA一致,其操作简单,可以自动化批量检测,成本低,适用于临床大样本筛查。 Objective To explore the reverse syphilis program of syphilis and analyze its value in laboratory diagnosis and prevention and treatment of syphilis screening. Methods TPPA, TP-ELISA and TRUST were used in this study.Meanwhile, a retrospective study was carried out to evaluate the clinical efficacy of TPPA, TRTP and TRUST from January to June in 2015, 23 389 specimens from hospitals and ward syphilis screening were tested. Using TPPA as the standard, Kappa test was used to compare the consistency among methods. Through the comparison of the sensitivity, specificity and total coincidence rate of each method, the feasibility of TP-ELISA as the screening method for syphilis reverse screening was analyzed and further statistics TP-ELISA S / CO values ​​and the relevance of TPPA-positive results, simplifying the reverse screening program. Results The positive rate of TPPA screening was 3.38% (790) in 23 389 specimens. The positive rates of TP-ELISA and TRUST were 3.63% (849) and 1.89% (443), respectively. TP-ELISA and TPPA had very good agreement (K = 0.954) for syphilis screening, TRUST and TPPA were generally consistent (K = 0.693), TP-ELISA had a sensitivity of 99.1% and TRUST had a sensitivity of 54.7% The difference was statistically significant (P <0.01). Statistics showed that the results of TP-ELISA were exactly the same as those of TPPA when the S / CO value of TP-ELISA was ≥5.0; 66 cases were false positive when the S / CO value of TP-ELISA was between 1.0 and 4.99; Seven false-negatives were present in all samples (TRUST positive in 6 samples) at ELISA S / CO values ​​<1.0. Conclusion TRUST sensitivity is low, as a syphilis screening alone is not conducive to the discovery of a large number of patients with latent syphilis; TP-ELISA high sensitivity and specificity, consistent with TPPA, its simple operation, automated batch testing, low cost, suitable for Clinical large sample screening.