7家血站实验室抗-HCV ELISA试验临界值评价
|
摘要
目的:评价7家血站实验室抗-HCV ELISA实验灰区的必要性及适宜性。方法:7家血站实验室编码为1,2,3,4,5,6和7,而8种ELISA试剂分别编码为A,B,C,D,E,F,G和H。7家血站实验室使用8种不同厂商ELISA试剂中的1种或2种对同一组(n=1259)标本进行抗-HCV检测。对检测结果有差异的标本进行Western blot补充试验以明确标本血清学状态。统计各实验室真阳性检出率、灰区标本确证阳性率以分析7家血站实验室抗-HCV试验"灰区"设置的必要性。通过绘制ROC曲线确定7家血站实验室抗-HCV ELISA试验的最佳临界值,比较3种不同临界值(实验室工作临界值、厂商推荐临界值、最佳临界值)下灵敏度及特异性的变化以分析7家血站实验室抗-HCV ELISA试验"灰区"的合理性。结果:7家血站实验室(其中编码1号实验室采用了A、B两种试剂)真阳性检出率分别是95.40%(1A)、99.23%(1B)、94.25%(2C)、96.17%(3D)、98.08%(4E)、96.93%(5F)、97.32%(6G因为有1家血站使用了两种ELISA试剂)、93.10%(7H)。除2C(94.25%)、7H(93.10%)外其余均大于95%。设立灰区的6家血站实验室(1A、1B、3D、4E、5F、6G和7H)灰区确证阳性率分别为0.00%、0.00%、21.43%、0.00%、0.00%、0.00%、38.89%。绘制ROC曲线比较3种不同临界值关系显示,5家实验室(1B、2C、4E、5F、6G)抗-HCV最佳临界值>厂商推荐临界值>实验室工作临界值,使用厂商推荐的临界值已可以达到筛查实验室高灵敏的要求; 1A、3D、7H抗-HCV最佳临界值<厂商推荐临界值,考虑使用适宜的灰区。设立灰区的6家实验室,使用实验室,工作临界值与使用厂商推荐临界值相比,仅3D、7H灵敏度略有提高(分别为1.60%、2.70%),其余实验室灵敏度无变化且特异性下降(0.20%-0.50%)。结论:除1A、3D、7H可适当设置灰区外,研究结果支持其他实验室抗-HCV ELISA试验取消灰区;即使是同一实验室也要对灰区设置进行科学评估,不宜盲目使用同一尺度,应建立适宜的判定策略。
Objective: To evaluate the necessity and suitability of the anti-HCV ELISA teot gray zone setted up by 7 blood station laboratories. Methods: 7 blood station laboratories were coded as 1, 2, 3, 4, 5, 6 and 7 respectively; 8 kinds of ELISA reagents were coded as A, B, C, D, E, F, G and H respectively. 1 or 2 of 8 ELISA reagents produced by different manufactories were used to detect the anti-HCV in specimens of same group by 7 blood station laboratories; the Westen blot was used to detect the specimens with difference of detected results so as to difine the serological status of specimens. The true positive rate of specimens detected by laboratories and gray zone-comfirined positive rate of specimens were accounted so as to analyze the necessity of setting up the gray zone for anti-HCV ELISA test of 7 blood station laboratories; the optimal cut-off value for anti-HCV ELISA test was determined in 7 blood station laborafories by ROC curve and the changes of sensitivity and specificity of 3 different cut-off value(laboratory work cut-off value, manifactory-recommended cun-off value and optimal cut-off value) were compared so as to analyze the suitability of gray zone for anti-HCV ELISA test in 7 blood station laboratories. Results: The true positive rate detected by 7 blood station laboratories, out of which coded 1 laboratory used 2 kinds of coded A, B reagents was 95.40%(1A), 99.23%(1B), 94.25%(2C), 96.17%(3D), 98.08%(4E), 96.93%(5F), 97.32%(6G) and 93.10%(7H). Except for 2C(94.25%) and 7H(93.10%), the true positive rate detected by laboratoies which not sutted up gray zone, the gray zone-con-firmed positive rate in 6 blood station laboratories setted up gray zone: was 0.00%, 0.00%, 21.43%, 0.00%, 0.00%, 0.00% and 38.89%. The comparison of 3 different cut-off valuces by ROC curve showed that the anti-HCV cut-off values in 5 laboratories(1B, 2C, 4E, 5F and 6G) were as follows: optimal cut-off value>manufactory recommeded cut-off value>laboratory work cut-off value, thus use of manufactory-recommeded cut-off value abreadly has reached the high sensitivity requinements for laboratory screening; however, the optimal cut-off value in laboratories 1A, 3B and 7H, thas the appropriate gray zone should be used. In 6 laboratories setting up gray zone, the gensitivity in 3D, 7H laboratories only a little improved(1.60% and 2.70% raspectively) in Eamparison between laboratory work cut-off value and manufactorg-recommeded cutoff value; moreover, the sensitivity in other laboratories not is changed, but the specificity decreased(0.20%-0.50%). Conclusion: In addition to setting up the appropriate gray zone in laboratories 1A, 3D and 5H, the gray zone in other laboratories may be cancelled. Even in the same laboratory, the setting up the gray zone also should be scientifically assessed, the same scale cannot be blindly used, thus appropniate strategies should be established.
Objective: To evaluate the necessity and suitability of the anti-HCV ELISA teot gray zone setted up by 7 blood station laboratories. Methods: 7 blood station laboratories were coded as 1, 2, 3, 4, 5, 6 and 7 respectively; 8 kinds of ELISA reagents were coded as A, B, C, D, E, F, G and H respectively. 1 or 2 of 8 ELISA reagents produced by different manufactories were used to detect the anti-HCV in specimens of same group by 7 blood station laboratories; the Westen blot was used to detect the specimens with difference of detected results so as to difine the serological status of specimens. The true positive rate of specimens detected by laboratories and gray zone-comfirined positive rate of specimens were accounted so as to analyze the necessity of setting up the gray zone for anti-HCV ELISA test of 7 blood station laboratories; the optimal cut-off value for anti-HCV ELISA test was determined in 7 blood station laborafories by ROC curve and the changes of sensitivity and specificity of 3 different cut-off value(laboratory work cut-off value, manifactory-recommended cun-off value and optimal cut-off value) were compared so as to analyze the suitability of gray zone for anti-HCV ELISA test in 7 blood station laboratories. Results: The true positive rate detected by 7 blood station laboratories, out of which coded 1 laboratory used 2 kinds of coded A, B reagents was 95.40%(1A), 99.23%(1B), 94.25%(2C), 96.17%(3D), 98.08%(4E), 96.93%(5F), 97.32%(6G) and 93.10%(7H). Except for 2C(94.25%) and 7H(93.10%), the true positive rate detected by laboratoies which not sutted up gray zone, the gray zone-con-firmed positive rate in 6 blood station laboratories setted up gray zone: was 0.00%, 0.00%, 21.43%, 0.00%, 0.00%, 0.00% and 38.89%. The comparison of 3 different cut-off valuces by ROC curve showed that the anti-HCV cut-off values in 5 laboratories(1B, 2C, 4E, 5F and 6G) were as follows: optimal cut-off value>manufactory recommeded cut-off value>laboratory work cut-off value, thus use of manufactory-recommeded cut-off value abreadly has reached the high sensitivity requinements for laboratory screening; however, the optimal cut-off value in laboratories 1A, 3B and 7H, thas the appropriate gray zone should be used. In 6 laboratories setting up gray zone, the gensitivity in 3D, 7H laboratories only a little improved(1.60% and 2.70% raspectively) in Eamparison between laboratory work cut-off value and manufactorg-recommeded cutoff value; moreover, the sensitivity in other laboratories not is changed, but the specificity decreased(0.20%-0.50%). Conclusion: In addition to setting up the appropriate gray zone in laboratories 1A, 3D and 5H, the gray zone in other laboratories may be cancelled. Even in the same laboratory, the setting up the gray zone also should be scientifically assessed, the same scale cannot be blindly used, thus appropniate strategies should be established.
引文
1 Lavanchy D.The global burden of hepatitis C.Liver Int,2009;29(Suppl 1):S74-81.
2王晓艳,伊瑶,陈斯勇,等.丙型肝炎病毒的病毒学检测方法.中华实验和临床病毒学杂志,2011,25(2):158-161.
3中华人民共和国卫生部,中国国家标准化管理委员会.GB18469-2012《全血及成分血质量要求》.2012.
4王瑞,陈瑜,等.抗-HCV酶联免疫吸附试验灰区临界值设置的探讨.中国输血杂志,2016,29(7):696-699
5 Centers for Discase Control and Prevention(CDC).Testing for HCVinfection;an update of guidance for clinicians and laboratorians.MMWR Morb Mortal Wkly Rep,2013;62(18):362-365.
6 Kapucu A,Macmillan NA,Rotello CM.Positive and negative remember and judgments ROCs in the plurals paradigm:evidence for alternative decision strategies.Mem Cognit,2010;38(5):541-554.
7 Zhou XH,Obuchowski NA,McClish DK.Statistical methods in diagnostic medicine.New York:John Wiley&Sons,Inc.2002:274-306.
8张莉,刘志远,张薇,等.利用ROC曲线分析HBsAg试验效能.中国输血杂志,2014,27(7):718-721.
9黄悦勤,李立明,等.临床流行病学[M].北京:人民卫生出版社,2010:118-123.
10李金明.临床酶免疫测定技术.北京:认为军医出版社,2005:95.
11 Boyd JC.Statistical analysis and presentation of data.In:Price CP,ed.Evidence-based laboratory medicine;principles,practice and outcomes,second edition.Washington DC,USA:AACC Press,2007:113-140.
2王晓艳,伊瑶,陈斯勇,等.丙型肝炎病毒的病毒学检测方法.中华实验和临床病毒学杂志,2011,25(2):158-161.
3中华人民共和国卫生部,中国国家标准化管理委员会.GB18469-2012《全血及成分血质量要求》.2012.
4王瑞,陈瑜,等.抗-HCV酶联免疫吸附试验灰区临界值设置的探讨.中国输血杂志,2016,29(7):696-699
5 Centers for Discase Control and Prevention(CDC).Testing for HCVinfection;an update of guidance for clinicians and laboratorians.MMWR Morb Mortal Wkly Rep,2013;62(18):362-365.
6 Kapucu A,Macmillan NA,Rotello CM.Positive and negative remember and judgments ROCs in the plurals paradigm:evidence for alternative decision strategies.Mem Cognit,2010;38(5):541-554.
7 Zhou XH,Obuchowski NA,McClish DK.Statistical methods in diagnostic medicine.New York:John Wiley&Sons,Inc.2002:274-306.
8张莉,刘志远,张薇,等.利用ROC曲线分析HBsAg试验效能.中国输血杂志,2014,27(7):718-721.
9黄悦勤,李立明,等.临床流行病学[M].北京:人民卫生出版社,2010:118-123.
10李金明.临床酶免疫测定技术.北京:认为军医出版社,2005:95.
11 Boyd JC.Statistical analysis and presentation of data.In:Price CP,ed.Evidence-based laboratory medicine;principles,practice and outcomes,second edition.Washington DC,USA:AACC Press,2007:113-140.