摘要
目的探讨化学发光免疫分析法检测乙型肝炎病毒核心抗体临界值及其可疑区间,保证不同检测系统检测乙型肝炎病毒核心抗体结果间的一致性。方法选取住院、体检及门诊患者180例标本,采用3种不同化学发光检测系统（Abbott,SH和AA系统）检测HBc Ab,以2种或以上化学发光检测系统的一致性结果作为真阴性和真阳性的判定标准,ROC曲线评价检测系统的诊断性能,并用配对χ2检验和kappa检验各检测系统结果与判定结果间的差异性和一致性;有显著差异的系统用ROC曲线重新设定其临界值和可疑区间,并验证其适用性。结果 180例标本中有阳性标本60例,阴性标本120例。Abbott诊断准确性为99.4%,与判定标准结果间差异无显著性（P〉0.05）且一致性好（kappa〉0.75）;SH和AA诊断准确性分别为94.4%、92.2%,与判定标准结果的一致性好（kappa〉0.75）,但差异有显著性（P〈0.05）。ROC曲线分析表明,SH临界值设定为0.594,可疑区间为0.554—0.634时,其诊断敏感性和特异性分别为98.3%和98.3%;AA临界值设定为1.505,可疑区间为0.765—1.850时,其诊断敏感性和特异性分别为95.0%和98.3%。经验证,SH、AA系统结果与Abbott结果符合率为100%（53例/53例）和98.1%（52例/53例,1例假阳性）,结果间差异无显著性（P〉0.05）且一致性好（kappa〉0.75）。结论不同化学发光检测系统其临界值和可疑区间会不同,须验证后使用。ROC曲线可用来验证和设定临界值及可疑区间,有利于提高不同检测系统间检测结果的一致性。
Objective To explore the cut-off value and its dubious interval in the detection of the hepatitis B core antibody（HBc Ab） by chemiluminescence immunoassay（CLIA）,and to ensure consistent results in different detection systems.Methods A total of 180 inpatient,outpatient and physical examination center patient sera samples were collected. All the samples were detected using Abbott,SH,and AA CLIA systems respectively. Only by the consistent results of 2 CLIA systems could one sample be identified true-positive or true-negative. The diagnostic performance of the CLIA systems were then evaluated by the receiver operating characteristic curve（ROC）. The differences and consistency of the results from these CLIA systems and the ＂ gold standard＂ method were compared by Mc Nemanr and kappa statistics tools. For the significantly different results of the CLIA systems,the cut-off value and its dubious interval would be reset by the ROC tool with its applicability verified. Results There were 60 positive samples and 120 negative samples in the 180 samples. The diagnostic accuracy of Abbott CLIA system was 99. 4%,and the difference of the results between the Abbott system and the gold standard was not significant（P〉0. 05）. Its interrater reliability of the results was very high（kappa〉0. 75）. The diagnostic accuracy of SH and AA CLIA system were 94. 4%,and 92. 2% respectively,and the difference of the results between the CLIA systems and the gold standard were both significant（P〈0. 05） while their interrater reliability of the results were high（kappa0. 75）. The ROC showed that the diagnostic sensitivity and specificity of the SH system were 98. 3%,and 98. 3% while the cut-off value was set as 0. 594 and its dubious interval was0. 554—0. 634. Furthermore, the diagnostic sensitivity and specificity of AA system were 95. 0%,and 98. 3% while the cutoff value was set as 1. 505 and its dubious interval was 0. 765 -1. 850. The validated results show that the coincidence rates between the SH,AA sy
作者
郭永灿
姚娟
赵容
王基策
张丹
王友强
黄晓玲
GUO Yongcan;YAO Juan;ZHAO Rong;WANG Jice;ZHANG Dan;WANG Youqiang;HUANG Xiaoling.(Clinical Laboratory of Traditional Chinese Medicine Hospital Affiliated to Southwest Medical University. Luzhou 646000, China)
出处
《中国输血杂志》
CAS
2018年第4期380-383,共4页
Chinese Journal of Blood Transfusion
基金
四川省卫生计生委重点项目(16ZD036)
四川省教育厅重点项目(16ZA0181)
