Lung cancers in patients with emphysema often present as PPLs inside or near the emphysema area. A report demonstrated that pulmonary infiltrates, such as those resulting from inflammatory cells and emphysematous changes, were increased on the side with lung cancer development compared to the other side [3]. De Torres et al. suggested that the presence of emphysema is an independent risk factor for lung cancer; the incidence of lung cancer in patients with emphysema is higher than that in patients without emphysema (relative risk: 3.33; 95% CI, 1.41–7.85), even in patients without airflow restriction. It has also been shown that the presence of emphysema increases the risk of lung cancer (relative risk, 2.51; 95% CI, 1.01–6.23)) [14]. Similarly, we discovered that malignant lesions were more frequent in patients with emphysema (44% [37/84]), but the diagnostic yield for PPLs located in non-emphysema-area lesions was significantly higher than for PPLs located inside or near emphysema-area lesions (92.3% vs 50%, p = 0.002). The reason for this may be that PPLs located in non-emphysema areas could be more easily reached by the probe than PPLs inside or near emphysema areas, which are characterized by the destruction of the lung parenchyma and alveolar attachment [15]. A systematic review and meta-analysis reported that, assisted by ENB, VB, and X-ray fluoroscopy, the overall weighted diagnostic yield of EBUS-GS-TBB was 70.6% (95% CI, 68–73%) [16]. Similarly, the present study found a mean diagnostic yield of 69% (58/84), higher than that reported by Georgiou et al. who showed a diagnostic yield of 63% in patients with advanced chronic obstructive pulmonary disease (COPD) (mean forced expiratory volume in 1 s (FEV1)/ forced vital capacity ratio of 1.33 and mean predicted FEV1 of 54.2%) utilizing electromagnetic navigation [17]. Lee et al. reported that the diagnostic yield of EBUS-GS under X-ray fluoroscopic guidance in patients with no or mild emphysema was significantly higher than that in those with moderate or severe emphysema (78% vs. 61%, p = 0.007) [18]. In the present study, according to the proximity of PPLs to emphysema lesions, the diagnostic yield for PPLs in non-emphysema areas was higher than that for PPLs in or near emphysema areas (82.6% vs. 52.6%, p = 0.010), although no X-ray fluoroscopic guidance was used for assistance. The EBUS-GS technique can improve the biopsy accuracy rate [19]. In addition, a previous study reported a fairly good diagnostic yield using EBUS-GS-TBB, where X-fluoroscopy alone was incapable of detecting solitary nodules [20].
The results of the present study may be supported by the following. First, EBUS with a GS can ensure that TBBs are performed at the correct location as identified using EBUS. Second, we selected two groups of patients with light emphysema (mean FEV1 of 89–93%). We categorized the patients according to the proximity of PPLs to emphysema lesions and not according to the severity of emphysema. This finding may also support the idea that PPLs in non-emphysema areas were more frequently diagnosed than PPLs located inside or near emphysema areas. Third, we discovered that the diagnostic yield of EBUS images was higher within the lesion than outside the lesion (96.2% vs 63.3%, p = 0.11); this also demonstrates the fact that the probe reached the PPLs appropriately in both groups with nearly the same frequency (54.3% [25/46] vs. 50% [19/30]). If target lesions could be confirmed on EBUS, even with the probe located within the lesion, the diagnostic yield for PPLs located inside or near emphysema areas would be lower than that for PPLs in non-emphysema area lesions. All of the above may reflect the fact that only the surfaces of PPLs inside or near emphysema areas were sampled owing to the destruction of the lung parenchyma, which obliterates the small airways [21]. To improve the diagnostic yield for PPLs inside or near emphysema areas, additional conventional transbronchial lung biopsy (TBLB) after EBUS-GS-TBB may be effective if a larger tissue sample is obtained [22]. Furthermore, transbronchial needle aspiration through a GS using the PeriView FLEX needle may improve the diagnostic yield by penetrating the lesions and collecting samples [23, 24]; the diagnostic yield may also be improved using EBUS-GS transbronchial lung cryobiopsy to obtain adequate samples [25].
Previous studies have reported that malignancy status, lesion size, bronchus signs on CT, and probe position on EBUS images were significant factors affecting diagnostic yield for PPLs using EBUS-GS-TBB [18, 19, 26, 27, 28]. Previous reports have indicated the probe position to be the only significant diagnostic factor [26, 29], and the present multivariate analysis found that probe location within the PPLs was significantly associated with a successful diagnosis using radio EBUS (OR, 2.542; p = 0.007) [29]. Similarly, the present study demonstrated that, in patients with emphysema, the probe position on EBUS images was a significant factor affecting the diagnostic yield of PPLs using EBUS-GS-TBB (OR, 3.394; p = 0.020). Moreover, the positional relationship of PPLs with emphysema lesions or the lesion status was a significant factor in the successful diagnosis of PPLs in patients with emphysema (OR, 5.614; p = 0.008) (Table 3). This factor affecting the diagnosis of PPLs in patients with emphysema has not been previously reported; in this situation, if the probe is located inside or near emphysema areas, CT bronchus signs may not be clearly evident, and the probe may not touch the center of the lesions because of the emphysematous destruction of the lung parenchyma. This may be important to physicians who use EBUS-GS-TBB in patients with emphysema. We also found that the positional relationship of PPLs to emphysema areas, rather than the bronchus sign, was a significant factor affecting the successful diagnostic yield of EBUS-GS-TBB.
Regarding the feasibility of molecular and genetic testing, a previous report showed that EBUS-TBB without GS and fluoroscopy could obtain sufficient tissue samples in almost all cases (94–100%) for testing of EGFR, ALK IHC, and PD-L1 IHC [30]. Similarly, the present study showed that the EBUS-GS-TBB-obtained samples from patients with emphysema and malignancies were sufficient for molecular and genetic tests, including for PD-L1, EGFR, ALK, and ROS1. However, molecular testing of samples from EBUS-guided transbronchial needle aspiration has also shown high efficacy [31]. Another report using 1.5-mm biopsy forceps to obtain samples for genotype analysis showed a 67–89% success rate in NSCLC [32]. Using a coaxial GS, EBUS-TBB can be repeatedly performed in a minimally invasive manner, with local anesthesia, reducing the chance of inserting the biopsy forceps into the wrong bronchioles, and multiple tissue specimens can be obtained. Thus, it is expected that this method will enable physicians to avoid unnecessary surgery and more easily select appropriate therapy.
In the present study, the rate of pneumothorax in the two groups was not significantly different; the total rate was 1/84 (< 1%) in patients with emphysema who underwent EBUS-GS-TBB without intercostal catheter insertion. Similarly, previous studies have reported that the rate of pneumothorax was less than 1% [21, 33]. In a study of 965 PPLs using EBUS-GS, there were eight patients with pneumothorax, three of whom required intercostal catheter insertion [33]. However, in the present study, we performed EBUS-GS and did not use assistance from navigation modalities, such as ENB, X-ray fluoroscopy, or VB. The incidence of pneumothorax was 2.1% for EBUS using electromagnetic navigation without GS [17] in patients with COPD undergoing TBLB [34], even lower than that in patients with advanced COPD.
The present study has certain limitations. First, the age of patients with PPLs in non-emphysema areas was significantly higher than that of patients with PPLs within or near emphysema areas. Second, this was a retrospective and small cohort study conducted at a single facility, and selection bias could have influenced our results; therefore, it is difficult to generalize the findings. In particular, the EBUS-GS-TBB complication rate may have been underestimated in this study. Third, all EBUS-GS-TBB procedures were performed without the assistance of a navigation system, such as ENB, X-ray fluoroscopy, or VB [35, 36]. The method utilized may be more suitable for medical centers without navigation devices. Prospective trials that consider these limitations are recommended.