摘要
目的：通过对罗氏配套系统和国产某试剂系统LDL-C测定进行方法比对和结果偏倚评估,探讨不同检测系统LDL-C测定结果是否具有可比性和一致性,同时探讨用新鲜混合血清传递标准结果,实现量值传递的可行性.方法：参考CLSI EP9-A2文件,以Roche Modular PPI分析仪、罗氏配套试剂为目标检测方法,其他不同检测系统为实验方法,检测血清样本LDL-C水平,用线性回归方法分析实验方法（y）和目标检测方法（x）测定结果的可比性并进行评估.结果：检测系统y1与目标检测系统的LDL-C测定结果相关系数（r2）为0.8555,相关性差,以2.07mmol/L、2.59mmol/L、3.37mmol/L、4.14mmol/L为Xc浓度下的偏倚均超过35%,检测系统y2、 用新鲜混合血清校准后的y3系统与目标检测系统测定LDL-C结果的相关系数（r2）分别为0.9786,0.9806,相关性良好,系统y2、y3在四个Xc浓度下的偏倚均小于15%,偏倚可接受.应用罗氏校准品标示值对国产某试剂直接进行校准后组成的y1系统LDL-C检测结果与罗氏配套系统结果不具有可比性,而使用国产某试剂及其相应校准品组成的y2系统和使用国产某试剂及用新鲜混合血清校准后的y3系统LDL-C结果与罗氏配套系统结果具有可比性.结论：不同检测系统存在基质效应,将商业化的校准品标示值直接应用于国产试剂的定标,检测系统会产生偏倚;新鲜混合血清能传递目标系统结果,具有实现量值传递的可行性.
Objective： To explore the comparability and consistency of＇results oFthe LDL-C between Roche biochemical analytical system and certain domestic reagent system for method comparison and bias evaluation analysis, and to explore the Feasibility of quantity transmission by pooled flesh semrrL Methods： Referring to EP9-A2 file of CLSI, Roche biochemical analytical reagent system （the detection system made upwith Roche Modular PPI analyzer, reagent of Roche, cfas calibrator and control of Roche） was target detection method, other different detection systems were experimental methods. The serum levels of＇LDL-C were detected. The linear regression analysis was used to analyze the comparability between experimental method （y） and target detection method （x） and the bias was evaluated. Results： The correlation coefficient （r2） of LDL-C result between detection system （yl） and target detection system （x was 0.8555, which had poor correlation. The bias of Xc concentrations （ 2.07mmol/L, 2.59mmol/L, 3.37mmol/L, 4.14mmol/L ） were more than 35%. The correlation coefficient （r2） of＇the detection system y2 （domestic unified brand reagent and calibrator） , the system y3 calibrated by pooled flesh serum and target detection system For the determination oFLDL-C were 0.9786, 0.9806 respectively, which had well correlation. The bias of System y2 and y3 on Four Xc concentrations are less than 15%, acceptable. The LDL-C results were uncomparable between the system yl and Roche target system （x） by using c^as calibrator marked value to calibrate domestic reagent. However, the LDL-C results were comparable between system y2 （domestic unified brand reagent and calibrator） , system y3 calibrated by pooled flesh serum and Roche target system （x） respectively. COnClUSiON： Different detection systems had matrix effect; The commercial calibrator with marked value directly applied to domestic reagent will engender system bias in calibration analysis. The result can be delivered exactly by using pooled flesh serum.
出处
《医学检验与临床》
2017年第5期21-24,共4页
Medical Laboratory Science and Clinics
关键词
量值传递
结果比对
偏倚评估
检测系统
Quantity transmission
Result comparison
Bias evaluation
Detection system
