摘要
目的评估3种液相色谱串联质谱法检测脂溶性维生素样本结果的相关性及一致性。方法选择山东英盛公司的试剂盒(方法A)、江苏豪斯公司的试剂盒(方法B)、实验室自建方法(方法C)作为待评价方法,对方法的正确度、精密度、线性范围、检出限、定量限等进行验证。采用3种方法同时检测42例临床样本中维生素A(vitamin A,VA)、25-羟基维生素D_(3)[25-hydroxyvitamin D_(3),25(OH)D_(3)]、维生素E(vitamin E,VE)的含量,3种方法两两组合,以Passing-Bablok回归分析方法间的相关性,Bland-Altman计算方法间平均相对偏差,以一致性相关系数(concordance correlation coefficient,CCC)评价方法间的一致性。采用实验室配制的混合标准品作为校准品,对方法A、B进行重新校准,比较校准前后不同方法间检测结果的相关性。结果方法A和B的性能符合试剂说明书中声称的要求,方法C的性能符合指南的要求。Passing-Bablok回归分析显示VA检测结果在各方法间的相关性较差(r=0.753~0.810),25(OH)D_(3)和VE的相关性较好(r=0.970~0.991)。Bland-Altman分析显示,VA、25(OH)D_(3)和VE在不同方法间的平均相对偏差均大于生物学变异的最低要求(8.19%、11.57%、7.01%)。一致性分析显示VA在不同方法间一致性较差(CCC:0.825~0.841),25(OH)D_(3)和VE一致性一般(CCC:0.912~0.935)。重新校准后,VA、25(OH)D_(3)和VE在不同方法间的平均相对偏差分别为1.5%~3.2%、-3.9%~3.6%、-5.2%~3.4%,均低于生物学变异的最低要求。化合物在不同方法间的一致性相关系数为0.915~0.987,高于重新校准前。结论3种LC-MS/MS法检测脂溶性维生素的结果相关性和一致性有待提高,校准品不同引起的校准偏差是一致性不佳的主要来源,提高商品化试剂盒的质量,完善脂溶性维生素的溯源体系,才能实现检验结果准确可比。
Objective To evaluate the correlation and consistency of the results of three liquid chromatography-tandem mass spectrometry for the determination of fat-soluble vitamins.Methods The kits of Shandong Yingsheng Company(method A),Jiangsu Housi Company(method B)and laboratory self-built method(method C)were selected as the subjects to be evaluated.The accuracy,precision,linear range,detection limit and quantitation limit of these methods were verified.The contents of vitamin A,25 hydroxyvitamin D_(3)and vitamin E in 42 clinical samples were simultaneously detected by all three methods pair-wise.The correlation between the methods was analyzed by Passing-Bablok regression.The mean relative deviation between the methods was calculated by Bland-Altman and the consistency of the methods was evaluated by concordance correlation coefficient(CCC).Methods A and B were recalibrated using the mixed standard prepared in the laboratory and the correlation between the test results of different methods before and after calibration was compared.Results The properties of methods A and B met the requirements stated in the reagent specification and the properties of method C met the requirements of the guide.Passing-Bablok regression analysis showed that the correlation between VA detection results was poor(r=0.753-0.810)and the correlation between 25(OH)D_(3)and VE was good(r=0.970-0.991).Blain-Altman analysis showed that the mean relative deviations of VA,25(OH)D_(3)and VE among different pairs were greater than the minimum requirements for biological variation(8.19%,11.57%,7.01%).Consistency analysis showed that VA had poor consistency among different methods(CCC:0.825-0.841)and 25(OH)D_(3)and VE had mediocre consistency(CCC:0.912-0.935).After recalibration step,the mean relative deviations of VA、25(OH)D_(3)and VE were 1.5%-3.2%,-3.9-3.6%and-5.2%-3.4%respectively,which were all lower than the minimum requirements of biological variation.The consistency correlation coefficient between different methods was 0.915-0.987,which was higher than before recalibration.Conclusion The correlation and consistency of the three LC-MS/MS methods for the detection of fat-soluble vitamins need to be further improved.Calibration deviation caused by different calibrators could be the main source of the poor consistency.By improving the quality of commercial kits and the traceability system for fat-soluble vitamins,an accurate comparison of test results can then be achieved.
作者
李敏
孙江漫
孟祥兆
郭盈盈
于洪远
李会强
LI Min;SUN Jiangman;MENG Xiangzhao;GUO Yingying;YU Hongyuan;LI Huiqiang(Beijing Aerospace General Hospital,Beijing 100076,China;College of Laboratory Medicine,Tianjin Medical University,Tianjin 300203,China)
出处
《标记免疫分析与临床》
2025年第2期401-406,共6页
Labeled Immunoassays and Clinical Medicine
基金
国家自然科学基金面上项目(编号:81772259)。
关键词
脂溶性维生素
商品化试剂盒
自建方法
一致性
Fat-soluble vitamin
Commercial kit
Laboratory developed tests
Consistency
作者简介
李敏(1991-),女,本科,主管技师,主要从事临床化学标准化研究。E-mail:1140368364@qq.com;通讯作者:李会强,男,博士研究生导师。
