目的

探讨两家实验室采用酶联免疫吸附试验（ELISA）检测丙型肝炎病毒（hepatitis C virus, HCV）抗体的结果可比性，为不同实验室之间检验结果的互认提供实验依据。

方法

参照美国临床实验室标准化协会（Clinical and Laboratory Standards Institute, CLSI) EP12-A2文件，对拟研究的两台仪器高、中、低三个浓度分别进行批内和批间精密度评价，收集2014年9月至2014年11月新疆医科大学第一附属医院患者新鲜血清标本共180例，两家医院实验室均采用国产addcare ELISA 1100全自动酶免分析系统（简称addcare ELISA 1100）随机盲法检测HCV抗体，同时用聚合酶链反应（polymerase chain reaction, PCR）联合重组免疫印迹试验（recombinant immunoblot assay, RIBA）进行确证，以受试者工作特征曲线（receiver operating characteristic curve, ROC曲线）分别确定的两种检测试剂最佳阳性判断阈值（sample/cut-off value, S/CO值）为判断标准，比较两家医院实验室检测HCV抗体的符合率，判断测定结果的可比性。

结果

万泰试剂和科华试剂的高、中、低浓度的批内精密度分别为5.0%、8.1%、9.8%和5.2%、8.3%、9.7%；批间精密度分别为7.9%、12.3%、14.7%和8.0%、12.7%、14.5%。180例待检血清中，金标准共检出HCV抗体阳性结果78例，阴性结果86例，不确定结果16例。通过绘制ROC曲线确定万泰试剂和科华试剂的最佳S/CO阈值分别为5.04和2.44，此时对确证试验的阳性预测率均≥95%；二者阳性符合率为96%，阴性符合率是95%，总符合率是96%，差异无统计学意义（P>0.05）。

结论

两家医院实验室以ROC曲线分别确定的两种试剂最佳阳性判断阈值为判断标准，检测结果显示可比性较好，达到检验结果的互认。对检测结果处于最佳阳性判断阈值附近的高值阴性标本和弱反应性标本应进一步做确证试验，以确定患者是否真正感染HCV，为患者早期诊治提供依据。

Objective

To explore the result comparison using enzyme-linked immunosorbent assay (ELISA) to detect anti-hepatitis C virus antibody (anti-HCV) between two different laboratories and to provide basis for results mutual accreditation.

Methods

A total of 180 serum samples were recruited from the First Affiliated Hospital of Xinjiang Medical University from September 2014 to November 2014 in this study to estimate the precision between two different instruments through testing high, medium, and low concentrations in accordance with EP12-A2 files from the American association of Clinical Laboratory Standards Institute (CLSI). The domestic addcare ELISA 1100 full autokinetic enzyme analysis system(addcare ELISA 1100) was used to test HCV antibody under randomized blind method in different labs;recombinant immunoblot assay (RIBA) and polymerase chain reaction (PCR) were carried out to provide a evidence of true positive for HCV antibody tests. Further, we used receiver operating characteristic (ROC)curves to identify the optimal cut-off values, and compared the coincidence rate of HCV antibody tests,estimated the comparability of test results.

Results

The intra variation for Wantai and Kehua reagents were 5.0%, 8.1%, 9.8%, and 5.2%, 8.3%, 9.7%; the inter variation were 7.9%, 12.3%, 14.7%, and 8.0%, 12.7%,14.5%, respectively. A total of 78 samples were positive, 86 were negative, and uncertain results of 16 samples were found in 180 by the gold method. The optimal S/CO threshold values for Wantai and Kehua were 5.04 and 2.44 by drawing ROC with the positive predictive rate of 95% or higher; positive coincidence rate was 96%, negative coincidence rate was 95%, and total coincidence rate was 96%, without significant difference between the two labs (P>0.05).

Conclusions

The test results have a good comparability and recognition between the two laboratories. The negative samples with higher values and weakly positive specimens near the S/CO threshold values should be determined by the confirmed test to diagnose patients with HCV, which could provide the basis for the early diagnosis and treatment.