摘要
采用IFCC参考方法和A、B两厂家测量系统(以下简称方法),同时测量20份不同浓度的新鲜单人份血清、ERM-AD452(有证酶参考物质)、RELA(Ring-trial,国际参考实验室室间质量评价活动)样本,有效数据采用MVS1.380软件按EP9-A2文件评价两厂家方法测量结果的正确性,并用Bland-Altman图形分析法进行验证。采用Matlab软件根据EP14-A2文件评价A、B两厂家校准品的互通性。通过采用不同方法对此两个GGT厂家测量系统进行的正确性验证和比对,可见A、B两个厂家不同方法所得的测量结果也明显不同:A方法与IFCC参考方法的回归方程为Y=0.995X+0.443,预测偏倚值为0.359,按EP9-A2文件方法评价,A方法与IFCC参考方法正确度性能一致;B方法与IFCC参考方法的回归方程为Y=0.827X-0.566,预测偏倚值为17.258,按EP9-A2文件方法评价,B方法与IFCC参考方法正确度性能不一致。对两种不同来源的样本ERM(114.1±2.4U/L)和RELA-A(111.0±2.2U/L)、RELA-B(203.6±3.3U/L)样本使用IFCC参考方法、A方法、B方法同时测量,测量结果各样本均值分别为114.8U/L、115.5U/L、88.1U/L;110.9U/L、109.5U/L、93.4U/L;203.8U/L、201.0U/L、173.4U/L,即A方法与IFCC参考方法正确度性能一致,而B方法与IFCC参考方法正确度性能不一致。鉴于各厂家测量结果的明显差异,实验室应建立常规测量方法,以评价测量结果的正确性。
20 fresh,single human serum samples with different concentrations,ERM-AD452 and RELA samples were simultaneously measured by using A,B two factories measurement systems and IFCC reference method.The results measured with two methods were evaluated by software MVS and Matlab.The results showed that the GGT concentration in serum samples measured with A and B methods were obviously different.The measured results with method A was consistent with IFCC reference method,and the regression equation between method A and IFCC reference method was Y=0.995X- 0.417,and the prediction bias value was 0.359.The measured results with method B were inconsistent with IFCC reference method,the regression equation between method B and IFCC reference method was Y=0.827X- 0.566,and the prediction bias value was 17.258.In view of obvious difference of results with different measurement systems,the laboratories should establish a method to evaluate routine measurement results with different methods.
出处
《标记免疫分析与临床》
CAS
2011年第5期336-341,共6页
Labeled Immunoassays and Clinical Medicine
