传统组织病理学检查与Xpert MTB/RIF检测在脊柱结核诊断中的作用

Evaluation of the role of traditional histopathology and Xpert MTB/RIF assay in the diagnosis of spinal tuberculosis

中文关键词:  脊柱结核  诊断  病理检查  Xpert MTB/RIF检测

英文关键词:Spinal tuberculosis  Diagnosis  Histopathology  Xpert MTB/RIF

中文摘要:

  【摘要】 目的：评价传统组织病理学（简称“病理”）检查与Xpert MTB/RIF（简称“Xpert”）检测在脊柱结核诊断中的作用。方法：回顾性分析2015年1月~2018年4月在我院诊断为脊柱结核的442例患者的资料，男227例（51.4%），女215例（48.6%）；年龄18~86岁（46.3±18.0岁）；胸椎结核165例，腰椎结核162例，腰骶椎结核53例，胸腰段结核46例，颈椎结核11例，颈胸段结核5例。所有患者通过穿刺或手术获得病灶中的脓液、死骨、肉芽组织及干酪样坏死组织，同时进行传统组织病理学检查与Xpert检测，计算两者的阳性率与阴性率。病理诊断分四类，一类、二类为确诊结核，三类、四类为未诊断结核。以病理诊断为标准，获得Xpert检测的敏感度、特异度，计算kappa值评估两者的一致性。以病理诊断为金标准，绘制Xpert检测的受试者工作特征（reciever operating characteristic，ROC）曲线并计算曲线下面积（area under curve，AUC），评估Xpert检测的价值。结果：442例患者中，病理确诊为结核292例（66.1%，95%CI为61.4%~70.5%），病理未诊断结核150例（33.9%，95%CI为29.5%~38.6%）；Xpert检测阳性363例（82.1%，95%CI为78.2%~85.6%），阴性79例（17.9%，95%CI为14.4%~21.8%），发现rpoB突变43例（9.7%）。Xpert检测阳性率高于病理检查（χ2=46.2477，P<0.001）。以病理诊断为金标准，病理确诊为结核的292例中，Xpert阳性273例，阴性19例；病理为三、四类诊断未确诊结核的150例中，Xpert阳性90例，阴性60例；Xpert的敏感度为93.5%（273/292），特异度为40%（60/150）。病理与Xpert两种检测确诊脊柱结核的kappa值=0.378（95%CI为0.290~0.466），一致性欠佳。Xpert检测诊断脊柱结核的AUC为0.667。病理检查与Xpert检测联合诊断脊柱结核的阳性率为86.4%（382/442）。结论：Xpert检测可以提高脊柱结核的确诊率并发现利福平耐药突变，标本量足够的前提下推荐同时行病理检查和Xpert检测。

英文摘要:

  【Abstract】 Objectives: To evaluate the role of traditional histopathology(pathology) and Xpert MTB/RIF (Xpert) in the diagnosis of spinal tuberculosis(TB). Methods: We retrospectively analyzed the data of 442 patients clinically diagnosed with spinal TB in Beijing Chest Hospital affiliated to Capital Medical University from January 2015 to April 2018. Among the 442 patients, 227(51.4%) were male and 215(48.6%) were female; aged 18-86 years(46.3±18.0 years old). There were 165 cases of thoracic TB, 162 cases of lumbar TB, 53 cases of lumbosacral TB, 46 cases of thoracolumbar TB, 11 cases of cervical TB, and 5 cases of cervical-thoracic TB. For all patients, pus, sequestrum, granulation tissue and caseous necrotic tissue in the focus were obtained through puncture or surgery, and traditional histopathological examination and Xpert examination were performed to calculate the positive and negative rates of both methods. Histopathological diagnosis was classified into four categories, the first and second categories were confirmed TB, and the third and fourth categories were undiagnosed TB. With histopathological diagnosis as the gold standard, the sensitivity and specificity of Xpert were obtained, and the kappa value was calculated to evaluate the consistency rate of the two methods. Taking histopathological diagnosis as the gold standard, the receiver operating characteristic (ROC) curve of Xpert was drawn and the area under curve(AUC) was calculated to assess the value of Xpert assay. Results: Out of the 442 patients, 292 were diagnosed with confirmed TB by histopathology (66.1%, 95%CI: 61.4%-70.5%), and 150 were undiagnosed TB by histopathology(33.9%, 95%CI: 29.5%-38.6%). The Xpert was positive in 363 cases(82.1%, 95%CI: 78.2%-85.6%), and negative in 79 cases(17.9%, 95%CI: 14.4%-21.8%), and rpoB mutations were found in 43 cases(9.7%). The positive rate of Xpert was higher than that of histopathology(χ2=46.2477, P<0.001). Keeping histopathological diagnosis as the gold standard, of the 292 cases diagnosed to be confirmed TB by histopathology, Xpert was positive in 273 cases and negative in 19 cases; while among the 150 cases of undiagnosed TB by pathology, Xpert was positive in 90 cases and negative in 60 cases. The sensitivity and specificity of Xpert were 93.5%(273/292) and 40%(60/150) respectively. The Kappa value of spinal TB diagnosed by histopathology and Xpert was 0.378(95%CI: 0.290-0.466), indicating poor consistency. The AUC of Xpert was 0.667. The positive rate of joint diagnosis of spinal tuberculosis by histopathological examination and Xpert assay was 86.4%(382/442). Conclusions: Xpert can improve the diagnosis rate of spinal tuberculosis and detect rifampicin-resistant mutations. It is recommended to perform both histopathology examination and Xpert assay when the amount of sample is sufficient.