化学发光免疫分析法与酶联免疫吸附试验检测外周血γ-干扰素诊断结核分枝杆菌感染的结果分析

Chemiluminescence immunoassay and enzyme-linked immunosorbent assay for the detection of peripheral bloodγ-interferon for the diagnosis of Mycobacterium tuberculosis infection

目的对化学发光免疫分析法(CLIA)检测外周血γ-干扰素(IFN-γ)结果为阴性、灰区、弱阳性、不确定和阳性标本进行分析,采用酶联免疫吸附试验(ELISA)进行复测,同时对γ-干扰素释放试验(IGRA)辅助诊断结核分枝杆菌(MTB)感染的临界值设定及应用价值进行探讨。方法将行CLIA-IGRA检查的患者标本共299例纳入研究,根据检测结果分为3组,A组(T-N<14 pg/mL)118例,B组(T-N:14~50 pg/mL)138例,C组(T-N>50 pg/mL)43例,T为测试培养管,N为本底对照培养管。利用ELISA进行复测,比较两种检测方法的结果一致性。计算不同临界值下的灵敏度和特异度,Kappa检验评价两种检测方法的一致性、灵敏度、特异度;ROC曲线分析两种检测方法的诊断性能。结果两种检测方法对于A组和C组样本的检测结果一致性较高,符合率分别为96.61%和93.02%,与B组样本检测得到的IGRA阳性率比较,差异有统计学意义(χ2=142.04,P<0.001)。CLIA和ELISA检测确诊结核患者、潜伏性MTB感染患者、非结核患者的阳性率分别为68.97%、72.60%、49.23%和44.83%、32.88%、14.21%,两者对潜伏性MTB感染和非结核患者检测的阳性率比较差异具有统计学意义(P<0.05)。当以40 pg/mL为临界值时,两种检测方法的特异度均较高,分别为88.80%和91.37%,一致性较好(总符合率94.60%,Kappa=0.779)。结论两种检测方法在检测MTB感染的样本具有较好的一致性,但对于易与结核混淆的非MTB感染者标本需要提高临界值来判定IFN-γ的结果,CLIA检测具有检测时间短、自动化优势,在MTB感染筛查方面具有较高的临床应用价值。

Objective To analyze the results of peripheral bloodγ-interferon(IFN-γ)in CLIA as negative,gray area,weakly positive,uncertain and positive,these samples were retested by ELISA,and to investigate the threshold setting and application value of interferon-γ-interferon release test(IGRA)for the diagnosis of Mycobacterium tuberculosis(MTB)infection.Methods According to CLIA test results,a total of 299 plasma samples were collected after tuberculosis-specific antigen stimulation in chemiluminescence kits and divided into three groups,including group A(n=118),group B(n=138),group C(n=43),and the results of the two methods were compared.Calculate the sensitivity and specificity at different cut-off values,and evaluate the consistency,sensitivity,and specificity.The ROC curve was used to analyze the diagnostic performance of both detection methods.Results The test results of the two methods for the samples was highly consistent in group A and C,with a compliance rate of 96.61%and 93.02%,respectively,and the IGRA positive rate in group B was statistically significant(χ^(2)=142.04,P<0.001).CLIA and ELISA positive rates of 68.97%,72.60%,49.23%and 44.83%,32.88%,14.21%of the confirmed tuberculosis patients,latent MTB infection patients and non-tuberculous patients,respectively,and the positive rates of CLIA and ELISA in latent MTB infection patients and non-tuberculosis patients were statistically significant.When the threshold value was 40 pg/mL,the specificity of both detection systems was 88.80%and 91.37%,respectively.The consistency is good(the total compliance rate is 94.60%,Kappa=0.779).Conclusion CLIA and ELISA have good consistency in detecting samples of Mycobacterium tuberculosis infection,but for non-tuberculous infection samples that are easily confused with tuberculosis,the threshold value needs to be raised to determine the positive results ofγ-interferon,and CLIA detection has the advantages of short detection time and automation,and has high clinical application value in tuberculosis infection screening.