Discussion on quality assurance of ALT primary screening in blood collection and supply institutions
-
摘要: 目的通过受试者工作特征曲线(ROC)分析血清丙氨酸氨基转移酶(ALT)初筛设备的性能，确定最佳临界值，进一步提升ALT初筛质量。方法选取湖州市中心血站进行ALT初筛的3种品牌设备(设备A、B、C)，分别收集每台设备初筛检测数据及对应的实验室检测数据各100份，以实验室检查结果为标准，以40 U/L为临界值，建立3台设备ALT检测的ROC曲线，比较曲线下面积(AUC)，计算最佳临界值(cutoff)，并比较修正cutoff值或参数前、后，3台设备ROC曲线变化情况和实验室ALT检测报废率的差异。结果修正前3台设备的AUC分别为0.751、0.863和0.899，试验准确性一般，3台设备与实验室全自动生化分析仪检测值差异有统计学意义(P＜0.05)。对设备的初筛临界值进行修正后，3台设备ROC曲线均有明显改善，设备A、B初筛ALT报废率显著上升，设备C初筛报废率显著下降，对应实验室ALT检测，设备A、B报废率显著下降(P＜0.05)，设备C报废率变化不明显(P＞0.05)。结论运用ROC曲线可对ALT初筛设备检测准确性进行常规监控，可及时发现初筛设备检测准确性问题，对保证ALT初筛质量，减少血液报废有较好的实用价值。Abstract: ObjectiveTo analyze the performance of ALT primary screening equipment by ROC curve, determine the optimal critical value, and further improve the quality of ALT primary screening.MethodsEach of the three brands of ALT primary screening equipment used in our station was selected(defined as equipment A, B and C), and 100 copys of primary screening test data and corresponding laboratory test data of each device were collected respectively. With laboratory test results as the standard and 40 U/L as the critical value, ROC curves of ALT detection of the three devices were established. The area under the curve(AUC) was compareed and the optimal critical value(cutoff) was calculated. And the ROC curve changes of the three equipments and the difference of ALT scrap rate in the laboratory before and after modifying the cutoff value or parameter were compared.ResultsThe AUC of the three devices before modifying were 0.751, 0.863 and 0.899, respectively. The test accuracy was general, and there were significant differences between the three devices and the automatic biochemical analyzer(P < 0.05). After the initial screening critical value of the equipment was corrected, ROC curves of the three equipments were significantly improved, ALT scrap rate of equipment A and B increased significantly in the initial screening, and ALT scrap rate of equipment C decreased significantly. Corresponding to ALT test in the laboratory, ALT scrap rate of equipment A and B decreased significantly(P < 0.05) and the scrap rate of equipment C does not change significantly(P > 0.05).ConclusionThe ROC curve can be used to routinely monitor the detection accuracy of ALT primary screening equipment and timely find the detection accuracy problems of ALT primary screening equipment. It might have good practical value to ensure the quality of ALT primary screening and reduce the scrap of blood.
-
Key words:
- alanine aminotransferase /
- primary screening /
- quality assurance
-
-
表 1 3台初筛设备ALT初筛和对应实验室检测值数据特征
例 设备 初筛检测结果 实验室检测结果 合格 不合格 合格 不合格 A 63 37 68 32 B 52 48 56 44 C 44 56 38 62 
下载: 导出CSV
表 2 3台初筛设备ROC曲线分析和最佳cutoff结果
设备 SE AUC 95%CI P YI cutoff 上限 下限 A 0.049 0.751 0.655 0.847 ＜0.001 0.409 43.125 B 0.039 0.863 0.786 0.940 ＜0.001 0.655 48.125 C 0.031 0.899 0.838 0.961 ＜0.001 0.677 51.875
下载: 导出CSV
-
[1] Kneeman JM, Misdraji J, Corey KE. Secondary causes of nonalcoholic fatty liver disease[J]. Therap Adv Gastroenterol, 2012, 5(3): 199-207. doi: 10.1177/1756283X11430859
[2] 钟利. 影响无偿献血者ALT不合格相关因素分析[J]. 医药前沿, 2019, 9(9): 230-231.
[3] 党进, 王英. 血站新规程实施前后ALT检测模式的回顾与探讨[J]. 临床血液学杂志, 2015, 28(10): 894-895. http://lcxz.cbpt.cnki.net/WKC/WebPublication/paperDigest.aspx?paperID=859fa996-addc-4b51-8227-39860088e22a
[4] 徐远东, 耿帜, 包一熙, 等. 基于PANA 9600E全自动核酸工作站的超敏HBV DNA定量检测性能评价[J]. 临床血液学杂志, 2022, 35(4): 239-243. http://lcxz.cbpt.cnki.net/WKC/WebPublication/paperDigest.aspx?paperID=8c748545-8f83-495e-b194-5eaee7250988
[5] 耿帜, 肖圣达, 徐远东, 等. 不同商品化试剂对临床血清样本HBV DNA定量检测结果的影响[J]. 临床血液学杂志, 2021, 34(6): 379-384, 389. http://lcxz.cbpt.cnki.net/WKC/WebPublication/paperDigest.aspx?paperID=121f407e-aeef-4d2b-8f4e-462c77cf1efc
[6] 陈燕君. 无偿献血者ALT升高原因分析及应对策略探究[J]. 中国医药科学, 2018, 8(3): 243-245. https://www.cnki.com.cn/Article/CJFDTOTAL-GYKX201803076.htm
[7] Kamarudin AN, Cox T, Kolamunnage-Dona R. Time-dependent ROC curve analysis in medical research: current methods and applications[J]. BMC Med Res Methodol, 2017, 17(1): 53. doi: 10.1186/s12874-017-0332-6
[8] Obuchowski NA, Bullen JA. Receiver operating characteristic(ROC)curves: review of methods with applications in diagnostic medicine[J]. Phys Med Biol, 2018, 63(7): 07TR01. doi: 10.1088/1361-6560/aab4b1
[9] Cao R, López-de-Ullibarri I. ROC Curves for the Statistical Analysis of Microarray Data[J]. Methods Mol Biol, 2019, 1986: 245-253.
-
图(3)
表(2)
计量
- 文章访问数: 646
- PDF下载数: 370
- 施引文献: 0