论文部分内容阅读
目的 研究不同技术联合检测痰液和胸腔积液对结核性胸膜炎的诊断价值.方法 收集2014年1月至2017年3月首都医科大学附属北京胸科医院收治的临床诊断为结核性胸膜炎的1955例患者,其中1232例(63.02％)有肺部实质病变,723例(36.无肺部实质病变,通过抗酸杆菌涂片镜检(简称“涂片法”)、改良固体培养法和BACTEC MGIT 960液体培养(简称“培养法”)和Xpert MTB/RIF技术(简称“Xpert法”)单独及联合检测患者痰液和胸腔积液中结核分枝杆菌的阳性检出率,并分析肺部实质病变对结核分枝杆菌检测阳性率的影响.结果 1955例结核性胸膜炎患者同时采用涂片法进行痰液和胸腔积液检测者中有535例(27.37％)患者、353例(18.06％)采用培养法、294例(15.04％)采用Xpert法,188例(9.62％)患者同时进行了涂片法+培养法+ XPert法检查.涂片法、培养法和Xpert法单独检测胸腔积液结核分枝杆菌的阳性率[3.18％ (17/535)、15.30％(54/353)、19.05％(56/294)]明显低于3种方法联合检测痰液+胸腔积液的阳性率[21.87％(117/535)、43.63％(154/353)、47.62％(140/294)](χ2值分别为85.31、68.16、54. 00,P值均<0.01).在有肺部实质病变的患者中,涂片法、培养法、Xpert法单独检测胸腔积液中结核分枝杆菌的阳性率[3.82％ (13/340), 16.30％ (37/227 23.91％ (44/184)]明显低于3种方法单独检测痰液的阳性率[30.29％(103/340)、51.10％ (116/227)、60.33％(111/184)](χ2值分别为84.19、61.53、50.04,P值均<0.01)和联合检测痰液+胸腔积液的阳性率[分别为 32.06％ (109/340)、55.95％ (127/227)、67.93％ (125/184)](χ2值分别为92.06、77.32、71.79,P值均<0.01).以培养为金标准,涂片法、培养法和Xpert法检测痰液+胸腔积液的敏感度[48.36％(59/122)、79.38％(77/97)、80.49％(66/82)]明显高于单独检测胸腔积液的敏感度[8.20％ (10/122)、26.80％ (26/97)、 26.83％ (22/82)](χ2值分别为48.52、53.84、47.47,P值均<0.01);检测痰液+胸腔积液的特异度较单独检测胸腔积液有所下降,其中涂片法检测的差异无统计学意义(χ2=3.65,P=0.056);而Xpert法和涂片法+ Xpert法检测的差异均有统计学意义(χ2=4.71,P=0.030;χ2=5.67,P=0.017).结论 联合检测痰液和胸腔积液中结核分枝杆菌的阳性检出率高于单独检测,对于疑似结核性胸膜炎的患者,应尽量进行痰液标本的检测,以提高对结核性胸膜炎诊断的参考价值.“,”Objective To evaluate the diagnostic value of several tests in Mycobacterium tuberculosis (MTB) detection by the combination of pleural fluid and sputum in tuberculous pleurisy. Methods Of 1955 cases with tuberculous pleurisy admitted in Beijing Chest Hospital from January 2014 to March 2017 were retrospectively analyzed. There were 1232 cases (63.02％) with pulmonary parenchyma and 723 cases (36.98％) without pulmonary parenchyma. According to the positive detection rate of MTB in single and combined detection of sputum and pleural effusion by AFB,modified roche solid culture, BACTEC 960 and Xpert, the effect of pulmonary parenchyma lesion on the positive rate of MTB need to be analyzed. Results Of 1955 cases with tuberculous pleurisy,535 cases (27.37％) of sputum and pleural effusion were examined by AFB and 353 cases (18.06％) by culture and 294 cases(15.04％) by Xpert. Among them, 188 cases (9.62％) were examined simultaneously by AFB, culture and Xpert. The positive rate of AFB, culture and Xpert (3.18％ (17/535), 15.30％ (54/353),19.05％ (56/294)) in detecting MTB in pleural effusion was significantly lower than that in sputum combination with pleural effusion (21.87％(117/535), 43.63％ (154/353), 47.62％ (140/294); χ2 were 85.31, 68. 16 and 54.00,respectively. All P values were less than 0.01). In patients with pulmonary parenchyma disease,the positive rate of detection for MTB in pleural effusion by AFB, culture and Xpert alone (3.82％ (13/340), 16.30％ (37/227) and 23.91％ (44/184)) was significantly lower than that in sputum (30.29％ (103/340), 51.10％ (116/227) and 60.33％ (111/184); χ2 were 84.19,61.53 and 50.04, respectively. All P values were less than 0.01),and also lower than that in sputum combination with pleural effusion (32.06％ (109/340),55.95％ (127/227) and 67.93％ (125/184); χ2 were 92.06,77.32 and 71.79 respectively. All P values were less than 0.01). When using culture as gold standard, the sensitivity of AFB, culture and Xpert (48.36％ (59/122),79.38％ (77/97), 80.49％ (66/82)) in detecting sputum combination with pleural effusion was significantly higher than that in single pleural effusion (8.20％(10/122),26.80％ (26/97), 26.83％ (22/82); χ2 were 48.52, 53.84 and 47.47 respectively. All P values were less than 0.01). The specificity of three tests in detecting sputum combination with pleural effusion was lower than that in single pleural effusion. There was no significant difference in AFB (χ2 =3. 65,P=0.056), but significant difference in Xpert (χ2 =4.71,P=0.030) and AFB combination with Xpert (χ2 =5.67, P=0.017). Conclusion Due to the positive rate of MTB in sputum and pleural effusion was higher than that in single test, the combined detection of sputum and pleural effusion should be carried out in patients suspected of tuberculous pleurisy in order to improve the diagnostic value.