目的评估实时荧光RNA恒温扩增检测(SAT)、分枝杆菌核酸检测(TB–NTM–PCR)及多色巢式荧光定量PCR(Xpert MTB/RIF)法对结核病诊断的价值。
方法收集2014年4–7月西安市结核病胸部肿瘤医院收治的确诊为肺结核患者的痰标本共378份，同时采用痰直接抗酸染色镜检、SAT、TB–NTM–PCR及MTB 960快速液体培养和Xpert MTB/RIF法对标本进行检测。以MTB 960快速液体培养及痰直接抗酸染色镜检结果为金标准，分析SAT法、TB–NTM–PCR法及Xpert MTB/RIF法对MTB的检出率。采用χ 2检验比较3种分子生物学检测方法的阳性率。
结果SAT、TB–NTM–PCR及Xpert MTB/RIF法检测的总体阳性率分别为37.6%(142/378)、37.8%(143/378)和53.4 %(202/378)；涂阳培阳时，SAT、TB–NTM–PCR及Xpert MTB/RIF法检测的阳性率分别为84.6%(77/91)、91.2%(83/91)和96.7%(88/91)，3种检测方法的阳性率比较差异有统计学意义( P＝0.018 2)；涂阴培阳时，SAT、TB–NTM–PCR及Xpert MTB/RIF法检测的阳性率分别为61.9%(60/97)、44.3%(43/97)和80.4%(78/97)，3种检测方法的阳性率比较差异有统计学意义( P<0.000 1)；涂阴培阴时，SAT、TB–NTM–PCR及Xpert MTB/RIF法检测的阳性率分别为1.6%(3/185)、6.5%(12/185)和16.8%(31/185)，3种检测方法的阳性率比较差异有统计学意义( P＝0.018)；涂阳培阴时，SAT、TB–NTM–PCR及Xpert MTB/RIF法检测的阳性例数分别为3/5、5/5及5/5例。以MTB 960快速液体培养和痰直接抗酸染色涂片镜检为金标准，3种检测方法中Xpert MTB/RI法的敏感度(87.6%，163/186)及阴性预测值最高(88.5%，185/209)且漏诊率最低(12.4%，24/193)；SAT法特异度(98.2%，214/218)及阳性预测值最高(97.2%，138/142)且误诊率最低(1.8%，4/218)。以960药敏试验结果为金标准，Xpert MTB/RIF法耐药检测的正确率为95.48%(148/155)，敏感度为95.5%(128/134)，特异度为95.2%(20/21)。
结论SAT、TB–NTM–PCR和Xpert MTB/RIF法对结核病的诊断均具有良好的辅助作用；Xpert MTB/RIF法无论对涂阳或涂阴肺结核均具有良好的检出率。
ObjectiveTo evaluate the diagnostic value of real-time fluorescent RNA isothermal amplification detection technology (simultaneous amplification and testing, SAT), Mycobacterium nucleic acid detection (PCR- fluorescence probe)method (TB-NTM-PCR) and Xpert MTB/RIF detection in the diagnosis of tuberculosis.
MethodsA total of 378 sputum specimens from pulmonary tuberculosis patients were collected between April to July 2014 in Xi’an Thoracic Tumor and Tuberculosis Hospital. The specimens were detected by 5 methods at the same time including acid-fast stain, SAT method, TB-NTM-PCR method, TB 960 rapid liquid culture and Xpert MTB/RIF. The sensitivity and specificity of SAT method, TB-NTM-PCR method and Xpert MTB/RIF were calculated according to the results of TB 960 rapid liquid culture and staining. The difference among all the 3 methods was analyzed by Chi-squared test.
ResultsThe positive rate of SAT-TB, TB-NTM-PCR and Xpert MTB/RIF were 37.6% (142/378), 37.8%(143/378)and 53.4% (202/378), respectively. In specimens both positive for acid-fast stain and culture, the positive rate of SAT method was 84.6% (77/91), that of TB-NTM-PCR was 91.2% (83/91), and that of Xpert MTB/RIF was 96.7% (88/91), the difference being significant ( P=0.018 2). In specimens negative for acid-fast stain but positive for culture, the positive rate of SAT method was 61.9% (60 /97), that of TB-NTM-PCR was 44.3% (43/97), and that of Xpert MTB/RIF was 80.4% (78/97), the difference being significant ( P<0.000 1). In specimens both negative for acid-fast stain and culture, the positive rate of SAT method was 1.6% (3/185), that of TB-NTM-PCR was 6.5% (12/185), and that of Xpert MTB/RIF was 16.8% (31/185), the difference being significant ( P=0.018). In specimens positive for acid-fast stain but negative for culture, the number of positive samples of SAT, TB-NTM-PCR and Xpert MTB/RIF were 3 (3/5), 5 (5/5), and 5 (5/5), respectively. With the result of TB 960 rapid liquid culture and staining as the reference, Xpert MTB/RIF showed the highest sensitivity of 87.6%(163/186), the minimum rate of missed diagnosis of 12.4%(24/193), and the highest negative predictive value of 88.5%(185/209); SAT-TB showed the highest specificity of 98.2%(214/218), the minimum rate of misdiagnosis of 1.8%(4/218), the highest positive predictive value of 97.2%(138/142). With the result of TB 960 rapid liquid culture as the reference, the sensitivity and the specificity of Xpert MTB/RIF were 95.52%(128/134) and 95.24%(20/21). The accordance rate of Xpert MTB/RIF and TB 960 rapid liquid culture was 95.48%(148/155).
ConclusionThe 3 molecular detection methods showed good results for the auxiliary diagnosis of tuberculosis. Xpert MTB/RIF had the best performance both in smear positive and negative specimens and it can detect rifampicin related rpoB gene mutations at the same time.
黄芳,党丽云,孙惠平,等. 三种分子生物学诊断技术对结核病诊断价值的比较[J]. 中华结核和呼吸杂志,2015,38(9)：680-685.
DOI:10.3760/cma.j.issn.1001-0939.2015.09.013版权归中华医学会所有。
未经授权，不得转载、摘编本刊文章，不得使用本刊的版式设计。
除非特别声明，本刊刊出的所有文章不代表中华医学会和本刊编委会的观点。
文档即答
智问智答
机器翻译
你好，我可以帮助您更好的了解本文，请向我提问您关注的问题。
