同室不同生化检测系统测定结果的比对和偏倚评估

摘要

Objective The comparability of serum results of different biochemical detecting system were studied through the method of comparison and bias estimation between Hitachi 7170 and Olympus AU1000.Methods According to the profile NCCLS EP9-A2, two different biochemical testing system,Hitachi 7170(comparison method) and Olympus AU1000(laboratory method),were used respectively to test 16 serological items to obtain the correlation coefficient and the linear equation, and the system bias % at the level of medical decision were calculated.The comparability of different biochemical testing system was judged according to half of the clinical laboratory improvement amendment 88(CLIA`88) 1/2 of the total errors is allowed.Results The testing results of the different biochemical testing system had a great correlation, All correlation coefficients of 16 items were greater than 0.975. The system bias of ALB and UA at a level of medical decision were not acceptable , and the rest could be accepted.Conclusion If the same item is tested in more than two different systems,it is necessary to judge clinical acceptability by the method of comparison and bias evaluation so as to insure the comparability.%目的 通过对本室Hitachi 7170和Olympus AU1000两种生化检测系统的测定结果进行方法比对和偏倚评估,探讨不同生化检测系统间检测结果是否具有可比性.方法 参照美国临床实验室标准化委员会(NCCLS)的EP9-A2文件要求,以Hitachi 7170为参比系统,Olympus AU1000为试验系统,检测40例患者血清的16个生化项目,计算相关系数、直线回归方程和不同医学决定水平处的相对偏差(SE%),并以美国临床实验室改进修正案能力验证(CLIA′88)允许总误差的1/2为标准,评价检测结果是否具有可比性.结果 两种不同生化检测系统的检测结果相关性良好,各项目的 相关系数大于0.975,差异有统计学意义(P<0.01);除了清蛋白及尿酸各有一个医学决定水平处的SE%>1/2 CLIA′88,超出临床可接受范围外,其余项目的 SE%均在临床可接受范围.结论 当同一实验室内同一检测项目存在两种以上分析系统时,应进行方法比对和偏倚评估,以实现检测结果的可比性.