This study examined lesions of suspected pulmonary tuberculoma based on the characteristics of isolated nodules. Our major result was that the diagnostic sensitivity of EBUS-GS was significantly higher than that of BALF when used with Xpert (57.81% vs. 25%), MGIT 960 culture (35.94% vs. 23.44%), and acid-fast staining (17.19% vs. 6.25%). Our results indicated that EBUS-GS with Xpert had the highest sensitivity (57.81%) and specificity (100%) for the diagnosis of pulmonary tuberculoma among all tested methods. This method therefore has significant value for use in clinical practice.
Previous research reported that the etiological diagnosis of smear-negative pulmonary TB was difficult [19]. Khan et al. found that the positive rate of acid-fast staining in BALF was only 25.9%, although the positive rate of BALF with culture reached 63% [20]. Zheng et al. performed a retrospective analysis and reported the sensitivity of BALF with Xpert was 68.8% [3]. Compared with smear-negative pulmonary TB, the diagnosis of pulmonary tuberculoma with isolated nodules is more difficult, resulting in a lower detection rate. Thus, previous studies found that 57.1–92% of TB patients in areas with endemic TB were misdiagnosed as having primary lung cancer [21]. It is common for some patients with pulmonary tuberculoma who were misdiagnosed as lung cancer and then received surgery in clinical practice. Our results showed that the sensitivity of BALF with acid-fast staining was 6.25%, BALF with MGIT 960 culture was 23.44%, and BALF with Xpert was 25.00%. These low values indicated that use of BALF might result in missed diagnoses and misdiagnoses. Therefore, a more accurate and minimally invasive examination is needed to obtain specimens and improve the diagnosis of pulmonary tuberculoma from isolated nodules.
Bronchoendoscopy techniques have developed rapidly during recent years. Ordinary bronchoendoscopy has limited value in the diagnosis of isolated pulmonary nodules [22]. The advantages of EBUS-GS are that it uses an ultra-fine bronchoscope and an annular ultrasound probe. As a minimally invasive diagnostic technology, EBUS-GS has greater flexibility, it leads to fewer complications, it can reach lesions around the central airway, and it can be used to guide entry into peripheral pulmonary lesions. The ultrasonic probe in EBUS-GS can help clinicians to avoid blood vessels in the surrounding regions, and thus improve safety and accuracy [23]. EBUS-GS has been used in clinical practice and in the differential diagnosis of peripheral pulmonary lesions [24], especially in the diagnosis of lung cancer. Zhu et al. found that the diagnostic rate of EBUS-GS for peripheral lung cancer was 64% in 150 patients, and that the postoperative complications from EBUS-GS were significantly less than those from CT-guided transthoracic needle aspiration [25]. Similarly, Bo et al. studied solitary pulmonary nodules (SPNs) and reported that EBUS-GS significantly increased the diagnostic rate of to 72.3% [13]. However, little is known about the use of EBUS-GS for the diagnosis of isolated pulmonary nodules caused by MTB. The present single-center prospective study demonstrated that the sensitivity when EBUS-GS was used with acid-fast staining, MGIT 960 culture, and Xpert were significantly higher when BALF was used with these assays (17.19% vs. 6.25%, 35.94% vs. 23.44%, 57.81% vs. 25%, respectively). In other words, compared with bronchoalveolar lavage using a bronchoscope, EBUS-GS is better in locating the lesion and obtains more valuable specimens based on puncture and biopsy. Thus, the EBUS-GS greatly improved the diagnostic accuracy of pulmonary TB.
Xpert is an automated real-time nucleic acid amplification technology that provides rapid and simultaneous detection of tuberculosis and rifampicin resistance. Because Xpert MTB/RIF is more efficient and rapid than traditional methods, it is now widely used for the diagnosis of TB [26, 27], and the WHO recommends this method as a first-line rapid test for the diagnosis of pulmonary TB [28]. A previous study reported the sensitivity of BALF with Xpert among smear-negative or sputum-scarce patients with pulmonary TB was 58.9% [29], higher than in our study (25%). However, our sensitivity was 57.81% and specificity was 100% when EBUS-GS was used with Xpert. In addition, our AUC for EBUS-GS with Xpert was 0.789, significantly higher than the other methods. Although the sensitivity of EBUS-GS with Xpert in our study was still not ideal, it greatly improved the etiological diagnosis of isolated pulmonary nodules caused by MTB. The combined use of QFT, a PPD skin test, and other examinations makes it much easier to make a correct clinical diagnosis. We also found that EBUS-GS with Xpert and with MIGT 960 culture or pathological examination increased the sensitivity to 62.5%; the specificity of EBUS-GS with Xpert and MGIT 960 culture was 100%, and the specificity of EBUS-GS with Xpert and pathological examination was 95.77%.
Pathological examination is crucial for the diagnosis of isolated pulmonary tuberculoma, and EBUS-GS can obtain pathological specimens from isolated pulmonary nodules. However, other diseases, such as nontuberculous mycobacterial pulmonary disease (NTM-PD) and granulomatous diseases of the lung, have clinical and pathological characteristics that are similar to pulmonary TB [30, 31]. In this study, the sensitivity of EBUS-GS with pathological examination (46.88%) was unacceptable, although the specificity was 95.77%; there were also 3 patients with false positive results, 1 with sarcoidosis and 2 with NTM-PD. In addition, drug resistance cannot be determined using a pathological examination. However, use of EBUS-GS with Xpert can quickly diagnose TB and confirm the presence of rifampicin resistance, suggesting that this method is more useful in clinical practice.
In general, compared with conventional invasive examinations, EBUS-GS is characterized by smaller wounds and fewer complications, and Xpert has a high diagnostic rate and is relatively rapid. The combined use of these two methods led to good diagnostic efficacy in isolated pulmonary nodules caused by MTB, and was also useful in the differential diagnosis of non-tuberculous pulmonary diseases.
There were some limitations in this study. Firstly, it was a single-center prospective study with a small sample size. Thus, the diagnostic value of EBUS-GS with Xpert for isolated pulmonary nodules caused by MTB needs verification by large multi-center studies. Secondly, the operators of EBUS-GS must have a relatively high level of technical expertise, and considering the angle of the bronchoscope, some lesions may not be reached or puncture failure might occur.