摘要
目的对Bio—RadD-10和TOSOHHLC-73G8糖化血红蛋白分析仪的检测结果进行比对,了解两个不同检测系统的可比性和偏倚评估。方法参考临床实验室标准化协会(CLSI)的标准化文件EP9-A2的要求,在5d内两台仪器对40个样本进行HbA1c的双份检测,记录结果,排除离群点,计算直线回归方程和相关系数,进行偏倚评估。结果两台仪器检测结果的直线回归方程为Y=1.0065X+0.0356,r2=0.9972(〉0.95),两系统在医学决定水平的预期绝对偏倚小于0.5(单位:%)、相对偏倚小于基于生物学变异制定的实验室内总误差(2.7%)。结论两个不同系统在检测HbA1c时具有很好的相关性和一致性,检测结果在医学决定水平处的偏倚符合临床要求。
Objective To know about the bias estimation, comparability of two different testing systems by comparing the detection results of Bio-Rad D-10 and TOSOH HLC-73G8 glycohemoglobin analyzers. Methods Referencing to the standardization document EP9-A2 from CLSI, carried out double detecting 40 samples for HbAlc by two glycohemoglobin analyzers in 5 days, recording the results, excluding outliers, calculating the linear regression equation and correlation coefficient, calculating bias estimation. Results The linear regression equation of two analyzers was: Y=1.0065X+0.0356, r2=0.9972 (〉0.95). The expected absolute bias was less than 0.5 (unit:%), and the expected relative bias was less than TE (2.7%) which was based on the biological variation at the medicine decide level. Conclusion There are good correlation and consistency between two systems in detecting HbA1c, the bias of the test results at the medicine decide level meets the clinical requirement.
出处
《国际医药卫生导报》
2016年第9期1303-1305,共3页
International Medicine and Health Guidance News