Accuracy evaluation of the routine method detected serum total bilirubin

XU Jian-hua,ZHUANG Jun-hua,ZHENG Song-bai,XU Ning,CHEN Dan-na,WANG Jian-bing,KE Pei-feng,CHEN Peng-fei,HUANG Xian-zhang
DOI: https://doi.org/10.13602/j.cnki.jcls.2012.12.014
2012-01-01
Abstract:Objective To explore the evaluation method of accuracy of the routine method for detection of serum total bilirubin(T-Bil).Methods According to the CLSI EP15-A2 documents,the T-Bil levels of samples for reference materials,IFCC Ring Trials for Reference Laboratories(RELA),and external quality assessment from National Center of Clinical Laboratories were detected by the routine method.The biases and standard deviation index(SDI) between the routine method and the same method group from Bio Rad Unity laboratory quality assessment platform were analyzed,and the accuracy of the routine method was evaluated and compared with the reference method.Results The T-Bil levels of RELA samples(2010~2011),NIST SRM 909b reference materials,and samples for external quality assessment(the first time in 2012) from National Center of Clinical Laboratories detected by the routine method were all acceptable.The biases and SDI were also acceptable when compared with the data(January to June,2012) from Bio-Rad Unity laboratory quality assessment platform.There was well correlation between the routine method and the reference method(Y=1.129 9X-5.734 3,r=0.999 8).When the concentration of T-Bil was less than 300 μmol/L,the bias could be accepted.However,when the T-Bil concentration was more than 300 μmol/L,the biases were from 10% to 13%,and could not be accepted.After the samples with high concentration of T-Bil were diluted 5 times with deionized water and redetected,the biases could be accepted and met the clinical requirement.Conclusion The accuracy evaluation of "single point concentration" had its limitations.In clinical laboratories,the accuracy of the whole measurement range should be analyzed,and the comparison with the reference method was one of valid methods to perform the accuracy evaluation of the routine method.
What problem does this paper attempt to address?