两种血清总胆红素参考测量方法检测结果的一致性评价
孟祥兆 , 李敏 , 郭盈盈 , 李曾心 , 赵光轮 , 于洪远
山西医科大学学报 ›› 2025, Vol. 56 ›› Issue (9) : 1096 -1100.
两种血清总胆红素参考测量方法检测结果的一致性评价
Consistency evaluation of results of two reference methods for total bilirubin in serum
目的 评价两种血清总胆红素参考测量方法检测临床血清样本结果的可比性,为实验室选择参考测量方法提供数据支持。 方法 依据CLSI EP9-A3文件要求,选择Doumas总胆红素参考测量方法和德国汉诺威临床化学研究所改良总胆红素候选参考测量方法进行正确度试验,通过后采用两种方法同时检测40份患者已检测完成的剩余血清。统计实验数据,以目测法检测离群值,以偏差图和百分比偏差图分析数据差异变化趋势,以Passing-Baklok回归模型和Deming回归模型评估两种方法检测结果在医学决定水平处的偏倚,以国家卫健委临床检验中心室间质评1/2TEa(7.5%)为可接受标准进行结果评价。 结果 两种参考测量方法对有证参考物质及RELA比对样本的检测结果均值均在合格范围内,偏倚小于±1.0%。两组方法检测结果之间具有良好的相关性(R2 =0.999),目测未有离群值。检测结果差值具有恒定、恒比混合变化趋势。Deming回归方程为:y=0.976 3x-0.476 6,斜率95%CI为0.963 7~0.989 0,截距95%CI为-1.370 5~0.417 3,相关系数R2 =0.999。Passing-Baklok回归方程为:y=-0.096 8x+0.970 7,斜率95%CI为0.964 7~0.979 6,截距95%CI为-0.506 6~0.252 4,相关系数R2 =0.999。依据两个回归方程计算得到医学决定水平处相对偏倚在-3.87%至-2.54%之间,小于国家卫健委临床检验中心室间质评1/2TEa(7.5%)。 结论 总胆红素候选改良参考测量方法与经典参考测量方法检测结果一致性符合临床检测的要求。在标准品SRM916a能够稳定获得的情况下,优先选择Doumas总胆红素参考测量方法。
Objective To evaluate the comparability of the results of two reference measurement methods for total bilirubin in clinical serum samples, and provide data support for choosing reference measurement methods in laboratory test. Methods In accordance with the requirements of the CLSI EP9-A3 document, the Doumas total bilirubin reference method and the modified total bilirubin candidate reference method of German Hannover Institute of Clinical Chemistry were selected for experiment of trueness, and then the remaining serum of 40 patients were simultaneously detected by both the two methods.Outliers were detected by visual inspection. The trend of data variation was analyzed using deviation chart and percentage deviation chart. The bias of the test results of the two methods at the medical decision level was evaluated by Passing-Baklok regression model and Deming regression model. The results were evaluated based on the 1/2 TEa(7.5%) of the external quality assessment of the National Health Commission Clinical Laboratory Center as the acceptable standard. Results The mean values of the test results of the certified reference materials and RELA comparison samples by the two reference measurement methods were all within the qualified range, and the bias was less than ±1.0%.There was a good correlation between the detection results by the two methods(R2=0.999), and no outliers were observed visually. The difference in the test results showed a mixed variation trend of constant and constant ratio. The Deming regression equation was y=0.976 3x-0.476 6, the 95%CI of the slope was 0.963 7-0.989 0, the 95%CI of the intercept was -1.370 5 to 0.417 3, and the correlation coefficient R2=0.999. The Passing-Baklok regression equation was y=-0.096 8x+0.970 7, the 95%CI of the slope was 0.964 7-0.979 6, the 95%CI of the intercept was -0.506 6 to 0.252 4, and the correlation coefficient was R2=0.999. Based on two regression equations, the relative bias at the medical decision level was from -3.87% to -2.54%, which was less than 1/2 TEA(7.5%) of the inter-laboratory quality assessment of the National Health Commission Clinical Laboratory Center. Conclusion The consistency of the test results between the candidate modified reference measurement method for total bilirubin and the classical reference measurement method meets the requirements of clinical testing. When the standard SRM916a can be stably obtained, the Doumas total bilirubin reference measurement method should be given priority.
| [1] |
于婷,张娟丽,孙晶, |
| [2] |
邵燕,刘淑明,孙慧颖, |
| [3] |
孟祥兆,李敏,孙江漫, |
| [4] |
居漪. 检验医学参考测量系统的研究与应用[J]. 检验医学,2021,36(3):237-239. |
| [5] |
唐利萍,刘明锦,刘冠财, |
| [6] |
|
| [7] |
李敏,孟祥兆,孙江漫, |
| [8] |
|
| [9] |
周伟燕,刘庆香,刘珍妮, |
| [10] |
吴立山,沈敏,徐峰, |
| [11] |
李敏,孟祥兆,孙江漫, |
“国家质量基础的共性技术研究与应用”重点专项基金项目(2017YFF0205401)
/
| 〈 |
|
〉 |