未知诊断定性试验分析性能评价方法探讨

Study on evaluating the performance of unknown diagnosis qualitative test

目的评价酶联免疫吸附试验（ELISA）检测乙型肝炎e抗原（HBeAg）的分析性能,探讨诊断不明确定性试验分析性能评价方法。方法结合美国临床实验室标准化协会（CLSI）发布的EP12-A2文件《定性试验评价方法用户协议;提议指南》和中国合格评定国家认可委员会（CNAS）颁发的《医学实验室质量和能力认可准则在临床免疫学检验领域的指南》对实验室ALISIE和BEPⅢ2套酶联免疫分析仪HBeAg检测系统进行性能评价,探讨临床实验室诊断不明确定性试验的性能评价方案。结果 ＋20%浓度临界值分析物检测结果阳性率≥95%,ALISIE检测系统变异系数（CV）为14.7%,BEPⅢ检测系统CV为19.2%。-20%浓度临界值分析物检测结果阴性率≥95%,ALISIE检测系统CV为11.7%,BEPⅢ检测系统CV为19.7%。2套检测系统的方法学比较,一致性百分率为97%,阳性一致率为98%,阴性一致率为96%。Kappa检验系数K值为0.94,95%的一致程度评分可信区间为91.5%~98.9%。结论 ALISIE和BEPⅢ检测系统检测HBeAg标本浓度在临界值为±20%的浓度范围之内都能得到稳定的检测结果。2套检测系统同时检测HBeAg的结果一致性强。该诊断不明确定性试验分析性能评价方法可行性强。

Objective To evaluate the performance of hepatitis B e antigen（HBeAg） assay by enzyme-linked immunosorbent assay（ELISA）,and evaluate the qualitative test with unknown diagnosis.Methods According to the Clinical and Laboratory Standards Institute（CLSI） document EP12-A2 User Protocol for Evaluation of Qualitative Test Performance;Approved Guideline and China National Accreditation Service for Conformity Assessment（CNAS） document Guidance on the Application of Medical Laboratory Quality and Competence Accreditation Criteria in the Fields of Clinical Immunology,the performances of two HBeAg ELISA analysis systems named ALISIE and BEPⅢ were evaluated.A protocol for evaluating the performance of unknown diagnosis qualitative test was analyzed.Results The samples with ＋20% concentration from the cut-off point yielded positive rate ≥95%.The coefficient of variation（CV） of ALISIE was 14.7%,and the CV of BEPⅢ was 19.2%.The samples with-20% concentration from the cut-off point yielded negative rate ≥95%.The CV of ALISIE was 11.7%,and the CV of BEPⅢ was 19.7%.The comparison of the two analysis systems showed that the concordance was 97%,the positive concordance was 98% and the negative concordance was 96%.The Kappa K value was 0.94.The 95% confidence interval was 91.5%-98.9%.Conclusions Both the ALISIE and BEPⅢ analysis systems can provide the stable results of HBeAg detection within the range of ±20% concentration at the cut-off point.The concordance of the two systems determined simultaneously is good.The evaluation for qualitative test with unknown diagnosis has a feasibility.