The diagnostic value of pleural biopsy by medical thoracoscopy in tuberculous pleural effusions

Objective To evaluate the diagnostic value of pleural biopsy in tuberculous pleural effusions by medical thoracoscopy. Methods A total of 229 patients with undiagnosed pleural effusions in the four Beijing clinical centers including Beijing Chest Hospital, Beijing Chaoyang Hospital, Beijing Jishuitan Hospital and Beijing Hospital from Jun 2015 to Oct 2018 were enrolled as the subjects in this large sample and multi-central prospective clinical trial. All patients were given medical thoracoscopy and the pleural biopsy specimen were collected. A routine pathological examination of the pleural biopsy specimen was firstly conducted for each patient. Then, both GeneXpert MTB/RIF assay and BACTEC MGIT 960 liquid culture detentions were performed using the suspension of the other part of pleural specimen after grinding. Results Of the 229 patients, 129 patients (56.3%) were confirmed with tuberculous pleurisy, 77 patients (33.6%) were non-tuberculosis pleural effusion and 23 (10.0%) patients were still undiagnosed by the medical thoracoscopy. In comparison with the routine pathological examination (23.1%,53/229), there was no significant difference in the positive rate of Mycobacterim tuberculosis infection detected by GeneXpert MTB/RIF (27.9% (64/229)) or BACTEC MGIT 960 liquid culture (17.0% (39/229)) (χ ²=1.32,P=0.251;χ²=2.67,P=0.103). However, combined with the two detection methods, the positive rate of Mycobacterim tuberculosis infection (32.8% (75/229)) was significantly superior to the routine pathological examination (χ²=5.25,P=0.022). The appearances of fibrous adhesions (70.5%, 91/129), diffuse miliary nodules (41.1%, 53/129) and fibrinous exudate (40.3%, 52/129) under the medical thoracoscopy were showed significantly higher in tuberculous pleurisy than that (32.5% (25/77), 6.5% (5/77), 15.6% (12/77)) in non-tuberculosis pleural effusion. Moreover, the appearance of scattered multiple nodular lesions (26.4%, 34/129) were markedly lower than that in non-tuberculosis pleural effusion (53.2%(41/77)) (χ²=28.41,28.52,24.42,15.06;Ps=0.000). Of the 229 patients, no serious adverse events were found. The major complications after medical thoracoscopy were mild chest pain (97.8%,224/229), a small amount of bleeding (58.5%,134/229) and local subcutaneous emphysema (46.7%,107/229). Chest pain could be relieved 2-3 days after taking painkillers. Bleeding and local emphysema did not need special treatment and could be alleviated 2-3 days after drainage tube indwelling. Among them, only 2 patients had empyema and given the drainage tube indwelling for more than 14 days. Finally, the 2 patients were transferred to surgery for thoracoscopic pleural dissection and drainage tube have been safely extubated after follow-up 6 months. Conclusion Medical thoracoscopy and the pathological and Mycobacterium tuberculosis examination of pleural specimen taken under the medical thoracoscopy have important clinical values in the diagnosis of tuberculous pleurisy. Because of its clinical value and high safety, medical thoracoscopy is worthy of clinical application in the diagnosis of tuberculous pleurisy.