免疫透射比浊法与干化学法测定血清白蛋白的方法学比较及偏倚评估  

Turbidimetric Immunoassay Method and Dry-slide Bromocresol Green Method in(ALB)Serum Albumin Determination:A Methodological Comparison and Bias Assessment

作　　者：冯品宁[1] 高玲[1] 喻雄文[1] 林少宾[1] 

机构地区：[1]中山大学附属第一医院检验医学部,510080

出　　处：《中国实用医药》2010年第4期18-20,共3页China Practical Medicine

摘　　要：目的通过评估免疫透射比浊法与干化学法测定ALB的偏倚,探讨两种方法检测ALB结果间的偏差是否在允许的范围,保证检测结果的可比性。方法依据NCCLS标准化文件EP9-A2[1],每天取临床样本8份,分别用免疫透射比浊法与干化学法进行双份测定,共测定5d,去除离群点,计算相关系数及预期偏差,以CLIA’88规定的室间质评(ALB+10%)允许误差的1/2作为方法比较结果系统误差的允许限值,进行偏倚的评估。结果干化学法与免疫透射比浊法比较,r2=0.934,在医学决定水平20g/L,35g/L,52g/L时,干化学法的预期相对偏倚分别10.4%,6.8%,3.2%,在Xc=20g/L及Xc=35g/L时,两种方法存在统计学差异。结论以免疫透射比浊法作为比较方法,干化学法测定ALB的结果存在正偏差,测定结果的偏差随着ALB浓度的减低而增大,在低浓度时差异更显著,两者相关性差。因此,实验室内使用不同分析方法检测同一分析物时,建议对各方法进行方法学比较及偏倚评估,明确方法间的差异,并建立各方法的参考范围,特别是建立不同方法在不同医学决定水平浓度的参考值,以保证检测结果的可比性。Objective The article compares the bias assessments of turbidimetric immunoassay method and dry-slide bromocresol green method in ALB detection, and discusses whether the bias of the two detection results are within the range permitted, so as to ensure the comparability of the results. Method According to NCCLS standardization document EP9-A2 [1], firstly 8 clinical samples are daily taken and tested duplicates under each method, altogether five days. Next the outlier points are excluded and correlation coefficients and ex- pected deviations are calculated. The value is restricted by 1/2 of the error that external quality assessment al- lows（ALB ＋ 10% ）by CLIA＇ 88 as the method comparison result system error, and then it is bias assessed. Re- salts The results of comparison between dry-slide bromocresol green method and turbidimetric immunoassay method are as follows： r2 =0. 934,when medical decisions are 20 g/L,35 g/L and 52 g/L,the expected relative deviations of dry-slide bromocresol green method are 10. 4% ,6. 8% ,3.2% respectively. When Xc = 20 g/L and Xc =35 g/L, there are significant differences between the results of two methods. Conclusion Turbidimetric immunoassay is regarded as the methodology in comparison. Dry-slide bromocresol green method, however, proves positive deviations in the results of ALB determination. Its value of bias increases with the decline of ALB concentration, and it is more apparent when the concentration turns low. These two methods have poor correlation. Therefore, when different analysis methods are utilized in the detection of the same object, it is suggested different methodological comparisons and bias assessments be used in each method to make clear the dis- parities among them and then establish each reference range, especially the concentration reference values of different methods in different medical decisions, to insure the comparability of the detection results.

关 键 词：血清白蛋白 免疫透射比浊法 干化学 偏倚 

分 类 号：R446.1[医药卫生—诊断学] R446.6[医药卫生—临床医学]