Pandey Sushil, Congdon Jacob, McInnes Bradley, Pop Alina, Coulter Christopher
Queensland Mycobacterium Reference Laboratory, Pathology Queensland, Queensland Health, Qld, Australia.
Queensland Mycobacterium Reference Laboratory, Pathology Queensland, Queensland Health, Qld, Australia.
Pathology. 2017 Jan;49(1):70-74. doi: 10.1016/j.pathol.2016.10.004. Epub 2016 Nov 29.
The aim of this study was to assess the performance of the GeneXpert MTB/RIF assay on extrapulmonary (EP) and respiratory (non-sputum) clinical samples of patients suspected of having tuberculosis (TB) from Queensland, Australia. A total of 269 EP and respiratory (non-sputum) clinical samples collected from Qld patients who were suspected of having TB were subjected to the GeneXpert MTB/RIF analysis, Ziehl-Neelsen (ZN) staining, Mycobacterium tuberculosis (MTB) culture and drug susceptibility testing. Phenotypic and genotypic data were compared. The overall performance analysis of the GeneXpert MTB/RIF assay for detection of MTB complex demonstrated sensitivity of 89%, specificity of 95%, PPV of 89% and NPV of 95% using culture as a reference standard. The GeneXpert MTB/RIF analysis of acid-fast bacilli (AFB) smear positive samples and AFB smear negative samples showed sensitivities of 100% and 77%, respectively. Looking at individual EP and respiratory (non-sputum) sample types, the sensitivity ranged from 60% to 100% although the specificity ranged from 33% to 100% with the specificity of lymph node tissue biopsy being the lowest. The GeneXpert MTB/RIF assay detected 11% more TB cases than culture and 27% more cases than ZN microscopy. Due to insufficient numbers of presenting rifampicin resistance cases, performance analysis of the GeneXpert MTB/RIF assay on rifampicin resistance could not be carried out. The GeneXpert MTB/RIF assay is potentially valuable for TB diagnosis in the majority of the EP and respiratory (other than sputum) samples in our setting. Although the GeneXpert MTB/RIF assay provides rapid diagnostic results, the overall sensitivity to rule out the disease is suboptimal for some specimen types. Performance varied according to specimen type and AFB smear status. The sensitivity and specificity of lymph node tissue was 63% and 33%. Care must be taken when using the GeneXpert MTB/RIF assay for detection of MTB in lymph node tissue samples. All samples should be cultured regardless of the GeneXpert MTB/RIF assay result.
本研究的目的是评估GeneXpert MTB/RIF检测法对来自澳大利亚昆士兰州疑似结核病（TB）患者的肺外（EP）和呼吸道（非痰液）临床样本的检测性能。对从昆士兰州疑似患有结核病的患者中收集的269份EP和呼吸道（非痰液）临床样本进行了GeneXpert MTB/RIF分析、萋-尼（ZN）染色、结核分枝杆菌（MTB）培养及药敏试验。对表型和基因型数据进行了比较。以培养作为参考标准，GeneXpert MTB/RIF检测法检测MTB复合群的总体性能分析显示，灵敏度为89%，特异性为95%，阳性预测值为89%，阴性预测值为95%。GeneXpert MTB/RIF对抗酸杆菌（AFB）涂片阳性样本和AFB涂片阴性样本的分析显示，灵敏度分别为100%和77%。查看各个EP和呼吸道（非痰液）样本类型，灵敏度范围为60%至100%，而特异性范围为33%至100%，其中淋巴结组织活检的特异性最低。GeneXpert MTB/RIF检测法检测出的结核病病例比培养法多11%，比ZN显微镜检查法多27%。由于利福平耐药病例数量不足，无法对GeneXpert MTB/RIF检测法的利福平耐药性能进行分析。在我们的研究环境中，GeneXpert MTB/RIF检测法对大多数EP和呼吸道（非痰液）样本的结核病诊断具有潜在价值。尽管GeneXpert MTB/RIF检测法能提供快速诊断结果，但对于某些样本类型，排除疾病的总体灵敏度并不理想。性能因样本类型和AFB涂片状态而异。淋巴结组织的灵敏度和特异性分别为63%和33%。使用GeneXpert MTB/RIF检测法检测淋巴结组织样本中的MTB时必须谨慎。无论GeneXpert MTB/RIF检测法结果如何，所有样本均应进行培养。
