均相酶免疫法测定尿激素类项目的性能验证

Methodical Evaluation and Clinical Application of Homogeneous Enzyme Immunoassay for Determining Urinary Hormones

目的验证均相酶免疫法尿游离皮质醇(COR)、17-羟类固醇(17-OHCS)、17-酮类固醇(17-KS)和香草扁桃酸(VMA)试剂盒在全自动生化分析仪Beckman DXC800上的性能,并初步评价其临床应用价值。方法依据WS T420-2013文件(临床实验室对商品定量试剂盒分析性能的验证)进行实验设计。精密度验证:采用各项目试剂盒配套的2水平质控品重复测定3次,连测5天,计算尿游离COR、17-OHCS、17-KS和VMA的重复精密度、期间精密度和实验室内变异系数(coefficient of variation,CV),并与厂家提供的数值进行比较,如果实验数值小于厂家数值,则验证通过。正确度验证:采用各项目试剂盒配套的定值校准品进行批内重复测定2次,验证4个项目各个水平定值校准品的均值与厂家提供的定值是否一致(相对偏移<1/2允许总误差),如果一致则表示正确度验证通过。同时,尿游离COR还采用20例随机住院患者的新鲜尿液标本,通过本试剂盒与Beckman化学发光法试剂盒检测进行比对。另外,尿游离COR和VMA还采用回收试验进行正确度验证。线性范围验证:选择配套定值校准品中的最高值,与最低值按比例进行混合稀释后,每个水平测2次,计算均值和偏移,并计算相关系数平方r2。生物参考区间验证:选取20例表观健康体检人员的新鲜尿液标本,验证尿游离COR、17-OHCS、17-KS和VMA的参考区间。结果测定各项目试剂盒配套的2水平质控品,尿游离COR、17-OHCS、17-KS和VMA低值质控品实验室内CV分别为:5. 01%、3. 75%、7. 52%和6. 11%,高值质控品实验室内CV分别为:5. 24%、6. 32%、6. 26%和4. 38%,4个项目实验室内CV均小于厂家提供的数值(均为10%),精密度验证通过。正确度验证时,4个项目各水平定值校准品均值的相对偏移均小于1/2允许总误差(25%)。尿游离COR本试剂盒与Beckman化学发光法试剂盒检测结果一致。尿游离COR回收率为109. 75%,尿VMA回收率为93. 67%。线性范围验证时,尿游离COR的线性范围为20. 0~1600. 0 ng/ml,尿17-OHCS的线性范围为0. 0~60. 0 mg/L,尿17-KS的线性范围为2. 0~130. 0mg/L,尿VMA的线性范围为2. 0~100. 0 mg/L,各项目相关系数平方r2均大于0. 995。生物参考区间验证,20例表观健康体检人员的新鲜尿液标本,4个项目检测结果均在厂家提供的参考区间范围之内。结论此均相酶免疫法尿游离COR、17-OHCS、17-KS和VMA试剂盒在全自动生化分析仪Beckman DXC800上具有良好的精密度、正确度和线性范围,性能指标符合质量标准,可满足临床测试要求。

Objective To validate the performance of homogeneous enzyme immunoassay protocols for determining urinary hormones on an automatic biochemistry analysis system,Beckman DXC800,based on the approach provided by WS T420- 2013. Methods Five aliquots of two control materials were used as precision verification samples.Each aliquot was measured in triplicate in a run every day during 5 days.The repeatability precision,intermediate precision and coefficient of variation （CV）of urinary free cortisol （COR）,17- hydroxy corticosteroids （17- OHCS）,17- ketosteroid （17- KS）and vanillylmandelic acid （VMA）were estimated and compared with manufacturer′s claim of CV.If the estimated CV was less than the claimed CV,the claim would be verifiable.Trueness verification was carried out whether it was of equivalence statistically between the mean value of routine biochemistry analysis system and the manufacturer′s fixed value.Matching fixed- value calibrators of 4 items were applied as trueness verification samples.Each aliquot was measured in duplicate simultaneously with the precision tests.If the test result was less than the 1/2 allowed total error （25%）,the claim would be verifiable.Furthermore,the fresh urines of 20 random inpatients were used to detect the free COR between this kit and the Beckman chemiluminescent immunoassay kit.In addition,trueness verification of urinary free COR and VMA were carried out through the recovery test.Linearity range verification was performed by the highest of matching fixed- value calibrators.After the highest fixed- value calibrator was proportionately diluted with the lowest,each aliquot was measured in duplicate in a run.Then,the mean,bias and correlation coefficient were calculated.To validate the biological reference interval,20 apparent healthy cases were selected by the physical examination.Urinary free COR,17- OHCS,17- KS and VMA of them were detected. Results The within- laboratory CVs of low level control of urinary free COR,17- OHCS,17- KS and VMA were 5.01%,3.75%,7.52% and 6.11% respectively.The within- laboratory CVs of high level control of 4 items were 5.24%,6.32%,6.26% and 4.38% respectively.All of the estimated CVs were less than the claimed CVs,so the manufacturer′s claims were verifiable.All of the relative biases of fixed- value calibrators were less than the 1/2 allowed total error （25%）.The results of urine free COR from the 20 random inpatients were equivalent between this kit and the Beckman chemiluminescent immunoassay kit.Additionally,the results of recovery test showed that the recovery percentages of urinary free COR and VMA were 109.75% and 93.75% respectively.So,all of the claimed trueness were verifiable.The linearity intervals of urinary free COR,17- OHCS,17- KS and VMA were 20.0- 1600.0 ng/ml,0.0- 60.0 mg/L,2.0- 130.0 mg/L and 2.0- 100.0 mg/L respectively.Squares of all the correlation coefficients of 4 items were more than 0.995.The detection results of 4 items from the 20 apparent healthy cases were in the reference intervals of the manufacturer′s claim. Conclusion The verification of performance for precision,trueness and linear range of 4 items are verifiable.All the performance indexes are in accordance with the quality standards and meet the requirements of clinical testing.