不同量值溯源5’-核苷酸酶试剂性能验证及临床结果一致性研究  

Performance verification and the consistency of clinical results of 5’-nucleoticlase reagent of different valuation traceability

作　　者：陈蕾[1] 王凤平[1] 杨玉婷[1] 封莉[1] 欧萌萌 方斌斌 吕珏[3] CHEN Lei;WANG Feng-ping;YANG Yu-ting;FENG Li;OU Meng-meng;FANG Bin-bin;LV Jue(2.The Affiliated Wuxi Mental Health Center of Nanjing Medical University,Wuxi 214002,Jiangsu,CHINA;Wuxi No.2 People’s Hospital,Wuxi 214002,Jiangsu,CHINA)

机构地区：[1]苏州市第五人民医院检验中心,江苏苏州215007 [2]南京医科大学附属无锡精神卫生中心,江苏无锡214002 [3]无锡市第二人民医院,江苏无锡214002

出　　处：《海南医学》2019年第15期1984-1989,共6页Hainan Medical Journal

基　　金：江苏省自然科学基金面上项目(编号:0161230);国家科技支撑子科题项目(编号:5BAI32H00);江苏省无锡市卫生计生委资助项目(编号:Q201755)

摘　　要：目的通过性能验证5’-核苷酸酶(5’-NT)试剂相关指标,采用理论k值或赋值血清能否实现多检测系统结果一致性。方法 5’-NT属制造商选定测量程序,比对实验按照ISO15189和GP29-A替代评估方案,利用厂家试剂、配套校准品、Sigma纯品酶校准品、理论K值赋值血清分别与日立7600-P组成6个不同检测系统,实验结果评价可能对选择的商品试剂产生不确定因素,故厂家用英文字母A^F代替,评价5’-NT在不同检测系统结果比对一致性,通过实验对试剂厂家产品主要性能指标进行验证。结果 5’-NT厂家配套与理论K值赋值血清两种校准模式检测30份高、中、低临床样本,厂家配套模式样本均数离散较明显,经One-way analysis方差分析组间差异具有统计学意义(P<0.05);Dunnett’s test以A厂家为对照组与C、D、E组作两两比较差异均具有统计学意义(P<0.05);而理论K值赋值血清校准各组间样本均数趋于一致,经单因素方差分析组间差异无统计学意义(P>0.05);各厂家试剂性能指标精密度、线性、稀释度、相关性比对均达到产品要求,其中D厂家不符合生物参考区间验证程序;抗坏血酸浓度达4.0 mg/dL时,B、C、D出现严重负干扰;血红蛋白浓度达2.0 g/L时除A厂家外其余厂家均有不同程度负干扰;6个厂家甘油三酯达25.3 mmol/L均无干扰现象。结论 5’-NT试剂厂家量值溯源方式不同,终端用户采用理论k值或赋值血清为校准品实验室间进行比对,才能实现多检测系统结果一致性;检测原理过氧化物酶法必须对抗坏血酸和血红蛋白抗干扰验证,以评估临床样本结果可接受。Objective To discuss whether it is possible to achieve the consistency of results in multi-detection system by using theoretical k value or assignment serum, through the performance verification of 5'-nucleoticlase (5'-NT) reagent correlation index. Methods According to the manufacturer's protocol of 5'-NT, the alternative evaluation scheme of ISO15189 and GP29-A, six different detection systems are developed using manufacturer reagent, matching calibrator, Sigma pure enzyme calibrator and theoretical K value assigned serum and Hitachi 7600-P. The manufacturers were designated with English letter A to F. The consistency of 5'-NT results in the different detection systems was evaluated, and the main performance indicators of reagents were verified by experiments. Results Thirty high, medium and low clinical samples were detected by the two calibration modes of 5'-NT manufacturer matching and theoretical K value assigned serum, and the results showed that the mean number of samples of manufacturer matching mode was dispersed obviously and there was significant difference by One-way analysis (P<0.05). Dunnett's test showed significant difference between manufacturer A and group C, D and E (P<0.05). By theoretical K value assigned serum calibration mode, the mean number of samples among the groups tended to be consistent, and there was no significant difference between the groups by One-way ANOVA (P>0.05). The precision, linearity, dilution and correlation of reagent performance indexes of each manufacturer all reached the product requirements, and the D manufacturer did not meet the verification procedure of biological reference interval. When ascorbic acid concentration reached 4.0 mg/dL, B, C and D had serious negative interference. When hemoglobin concentration reached 2.0 g/L, there were different degrees of negative interference except for manufacturer A. When triglyceride reached 25.3 mmol/L, there was no interference in all groups. Conclusion When detecting 5'-NT, different manufacturers have different valuat

关 键 词：5’-核苷酸酶 性能验证 量值溯源 纯酶校准品 理论K值 

分 类 号：R446[医药卫生—诊断学]