应用国产血清胆红素钒酸盐氧化法试剂盒自建检测系统及其评价

Evaluation of the self-established assay system of serum total bilirubin and direct bilirubin concentration measured by vanadate oxidation reagent kits

目的比较分析上海科华钒酸盐氧化法测定血清总胆红素(TBil)、直接胆红素(DBil)试剂盒及配套校准品在Roche Modular及CL-7300全自动生化分析仪上建立的检测系统与Roche Modular可溯源的参考检测系统检测患者血清TBil、DBil结果间的可比性及临床可接受性。方法依据美国临床实验室标准化委员会(NCCLS)EP9-A文件对3个自建检测系统与Roche Modular参考检测系统检测患者新鲜血清TBil、DBil浓度,并进行线性回归与相关及偏差分析。结果3个自建检测系统与Roche Modular参考检测系统的血清TBil、DBil检测结果具有较好的可比性,其相关系数(r)>0.99,平均相对偏差<15%;而各自建检测系统间血清TBil、DBil检测结果可比性明显优于与参考系统间的比较,平均相对偏差<6%,r=1.00。结论3个自建检测系统血清TBil、DBil检测结果与Roche Modular可溯源的参考检测系统相比可被临床接受,但具有一定系统负偏差,而此系统偏差主要与TBil、DBil测定方法有关,可能与校准品及仪器的选择无明显关系。在仪器上输入回归方程,结果可比性明显提高;Roche cfas可作为钒酸盐氧化法测定血清TBil、DBil的校准品。

Objective To evaluate the comparability and clinical acceptability of the results of serum total bilirubin（TBil） and direct bilirubin（DBil） measured by different self-established assay systems with Shanghai Kehua TBil and DBil kits（Vanadate oxidation method） in Roche Modular and Shimadzu CL-7300 clinical chemistry analyzers compared with the traceable Roche Modular TBil and DBil reference assay system. Methods The levels of TBil and DBil in fresh patient sera were analyzed by three selfestablished systems and Roche Modular reference system.The linear regression and correlation were calculated,and the bias were estimated according to National Committee for Clinical Laboratory Standards（NCCLS） document EP9-A files. Results The comparability of serum TBil and DBil measurment results by self-established systems compared with Roche Modular reference system was acceptable,the correlation coefficient（r） was 〉0.99,mean relative bias was 〈15%.The comparability between the self-established systems（bias〈6%,r=1.00） was better than one with Roche Modular reference system. Conclusions The results of serum TBil and DBil measured by our self-established systems could be used in clinical laboratory compared with Roche Modular reference system.However there are some system positive bias,which could be corrected by the regression equation and the comparability could be modificated.The bias mainly comes from the used method and has not significant correlation with the measured instrument and calibrator.Roche cfas could be used to calibrate the Kehua serum TBil and DBil kits.