不同检测系统抗HCV抗体灰区样本检测一致性分析

Consistency of the results of anti-HCV antibody gray zone samples among different determination systems

目的探讨4种国产检测系统与由ARCHITECT i2000SR全自动免疫分析仪及配套试剂组成的检测系统(简称i2000SR检测系统)抗丙型肝炎病毒(HCV)抗体灰区样本检测结果的一致性和相关性。方法分别采用4种国产检测系统[2种使用酶联免疫吸附试验(ELISA)、1种使用时间分辨荧光免疫分析法(CTRFIA)、1种使用化学发光免疫分析法(CLIA)]测定不同浓度的抗HCV抗体标准品,评价各检测系统的线性关系及不精密度[变异系数(CV)]。分别采用4种国产检测系统和i2000SR检测系统测定120例经重组免疫印迹法(RIBA)确认为阳性的抗HCV抗体灰区样本,评价各系统检测结果的相关性和一致性。结果4种国产检测系统检测不同浓度标准品的批内CV分别为17.6%~25.1%、0.8%~26.7%、7.8%~36.2%和0.7%~20.2%,总CV分别为15.9%~25.3%、1.0%~23.5%、7.7%~34.8%和3.4%~18.9%,线性相关方程分别为Y=0.634X+0.973、Y=2.495X+0.010、Y=1.348X+0.896、Y=2.510X-0.214。对于S/CO值为1.00~3.00的样本,4种国产检测系统与i2000SR检测系统符合率为34.2%~55.2%;对于S/CO值为3.01~6.0的样本,检测符合率为57.5%~80.0%;对于S/CO值为6.01~9.00的样本,4种国产检测系统检测符合率均>90%。4种国产检测系统均阴性的样本数为23份(19.2%),S/CO值的x±3s为2.80±4.35;4种国产检测系统均阳性的样本数为71份(59.1%),S/CO值的x±3s为5.13±7.84。4种国产检测系统两两之间的r值分别为0.7125、0.7600、0.8370、0.7022、0.7694、0.7364,Kappa值分别为0.6663、0.5985、0.7426、0.7593、0.8790、0.6648。结论对于灰区(S/CO值为1.00~9.41)样本,建议应至少使用2种检测系统或使用高特异性的检测系统进行复检。对于低浓度样本或临界值样本,应充分考虑不同检测系统之间可能出现的结果不一致现象,制定合理的灰区范围。

Objective To investigate the consistency and correlation between 4 domestic determination systems and ARCHITECT i2000SR automatic immunoassay analyzer with supporting reagents(i2000SR system)for determining anti-hepatitis C virus(HCV)antibody gray zone samples.Methods Totally,4 domestic determination systems[2 systems using enzyme-linked immunosorbent assay(ELISA),1 system using time-resolved fluorescence immunoassay(TRFIA)and 1 system using chemiluminescence immunoassay(CLIA)]were used to determine different concentrations of anti-HCV antibody standard products and to evaluate the linearity of each determination system and imprecision[coefficient of variation(CV)].The 4 domestic determination systems and the i2000SR system were used to determine 120 samples of anti-HCV antibody gray zone,which were confirmed by recombinant immunoblotting assay(RIBA)to evaluate the correlation and consistency of each determination system.Results For the 4 domestic determination systems,the within-run CV were 17.6%-25.1%,0.8%-26.7%,7.8%-36.2%and 0.7%-20.2%for the standard products with different concentrations,respectively.The total CV were 15.9%-25.3%,1.0%-23.5%,7.7%-34.8%and 3.4%-18.9%,respectively.The linear correlation equations were Y=0.634X+0.973,Y=2.495X+0.010,Y=1.348X+0.896 and Y=2.510X-0.214,respectively.The S/CO value of the i2000SR system was 1.00-3.00 samples,and the determination consistency rate of the 4 domestic determination systems was 34.2%-55.2%.The determination consistency rates of the 4 domestic determination systems with S/CO values of 3.01-6.00 samples were 57.5%-80.0%.The determination consistency rates of the 4 domestic determination systems with S/CO values of 6.01-9.00 samples were>90%.The number of samples with negative results from the 4 domestic determination systems was 23(19.2%),and the S/CO value x±3s was 2.80±4.35.The number of samples with positive results from the 4 domestic determination systems was 71(59.1%),and the S/CO value x±3s was 5.13±7.84.The correlation coefficients(r)between 2 domestic reagents were 0.7125,0.7600,0.8370,0.7022,0.7694 and 0.7364,respectively.The Kappa values among the domestic reagents were 0.6663,0.5985,0.7426,0.7593,0.8790 and 0.6648,respectively.Conclusions For the gray zone samples(S/CO value of 1.00-9.41),it is recommended that at least 2 reagents or highly specific reagents should be used for retesting.For low-concentration samples or cut-off samples,the possible inconsistencies between different determination systems should be fully considered,and a reasonable gray zone range should be formulated.