Usefulness of virtual bronchoscopic navigation combined with radial endobronchial ultrasound for peripheral pulmonary lesions

Background: This study aimed to evaluate the diagnostic value of virtual bronchoscopic navigation combined with radial endobronchial ultrasound for peripheral pulmonary lesions (PPLs). Methods: The 105 patients with PPLs identied by computed tomography in Nanjing Brain Hospital underwent radial endobronchial ultrasound (R-EBUS) with or without virtual bronchoscopic navigation (VBN) randomly from January 2015 to December 2017. The diagnostic yield, operation time and complications were evaluated in the two groups. Results: There was no signicant difference in the diagnostic yield between the VBN+R-EBUS group and the R-EBUS group (76.0% vs. 65.5%, P =0.287). The operation time in VBN+R-EBUS group was less than that in R-EBUS group (20.6±12.8 min vs. 28.6±14.3 min, P =0.016). No severe procedure related complications such as pneumothorax and hemoptysis were observed. Conclusions: VBN cannot improve the diagnostic yield, but it can shorten the operation time. The VBN combined with R-EBUS is a safe and effective technique for PPLs.


Background
How to diagnose the peripheral pulmonary lesions (PPLs) quickly and accurately has always been a clinical di culty [1]. CT-guided percutaneous lung biopsy has a high diagnostic yield and is recommended for the diagnosis of peripheral lung diseases, but it is often accompanied by pneumothorax, hemoptysis and other complications [2,3]. Transbronchial lung biopsy (TBLB) has fewer complications, but the diagnostic yield of peripheral lung lesions with diameter ≤ 20 mm is lower [4]. In recent years, virtual bronchoscopic navigation (VBN) has been gradually applied in clinical practice.
Several researches showed that [5][6][7][8][9], VBN combined with radial endobronchial ultrasound (R-EBUS) can improve the diagnosis yield of peripheral lung diseases, but VBN combined with R-EBUS is rarely used in PPLs. The purpose of this study was to investigate the diagnostic yield of pulmonary biopsy guided by VBN and R-EBUS in the diagnosis of PPLs, and to explore the clinical value of VBN and R-EBUS in the diagnosis of peripheral pulmonary nodules.

Patients
Total of 105 patients with PPLs were recruited, who underwent TBLB guided by R-EBUS alone or VBN combined with R-EBUS in the Nanjing Brain Hospital from January 2015 to December 2017. Inclusion criteria: the diameter of PPLs found by chest CT was less than or equal to 30 mm, and there was lesion were detected under bronchoscope. Exclusion criteria: Patients with cerebral hemorrhage and myocardial infarction in the past 3 months; patients with active hemorrhage and severe cardiopulmonary insu ciency can not tolerate bronchoscopy and do not cooperate with them. The patients were randomly divided into VBN + R-EBUS group (50 cases) and R-EBUS group (55 cases). This study was approved by the ethics committee of Nanjing Brain Hospital. All subjects were fully informed of the examination content, risk and signed the informed consent.

Procedure
All patients underwent multi-slice spiral CT scan. The DICOM data of CT scan in VBN and R-EBUS group were imported into the computer, and the virtual bronchoscope image of the target bronchus was automatically created by VBN software (DirectPath v1.02, Cybernet systems), and the focus guidance path was established ( Figure A, B). Two groups of patients were carried out under local anesthesia, fasting and water prohibition for 6 hours before operation, 2% lidocaine was inhaled by atomization, and 2% lidocaine was dripped into nose and trachea at the same time. In the VBN + R-EBUS group, bronchoscopy (Olympus BF-P260F, outer diameter 4.0 mm, working aperture 2.0 mm) was guided to the target's sub segment bronchi through the VBN system, and then the ultrasound probe (UM-S20-20R, Olympus) was extended to the corresponding sub segment. After detecting the low echo area (Figure C), the ultrasound probe was slowly withdrawn and the sub segment bronchus opening was measured to indicate the focus distance of the area. Then according to the measured distance, use the ultrasonic probe twice repeatedly to observe whether the operation path is correct. Withdraw the ultrasound probe and send it into the biopsy forceps along the positioning bronchial subsegment, and take the biopsy materials at the same location distance from the ultrasound focus. EBUS group: according to the location of the focus determined by preoperative chest CT, push the ultrasonic probe to the corresponding segment, the same as the operation steps of EBUS group. All the above groups were biopsied three times.

Statistical analysis
The data were processed by SPSS 20.0 software, the measurement data were described by mean ± standard deviation, the comparison of measurement data was analyzed by single factor ANOVA, and the comparison between the count data groups was tested by χ2, with P < 0.05 as the difference.

Clinical characteristics
Among the 105 patients with pulmonary nodules, there were 50 cases in VBN + R-EBUS group, 30 males and 20 females, with an average age of (55.8 ± 10.6) years and a diameter of (27 ± 3) mm. Among the 55 patients in R-EBUS group, 30 were male and 25 were female, with an average age of (56.5 ± 10.2) years and a diameter of (28 ± 2) mm. In VBN + R-EBUS group, there were 12 lesions in the right upper lobe  (Table 1).

Diagnostic yield
The diagnosis yield of VBN + R-EBUS group and R-EBUS group was 76.6% and 65.5% respectively, there was no signi cant difference between the two groups (P = 0.287). In the PPLs with diameter < 20 mm, the diagnostic yield of the VBN + R-EBUS group was 70.0%, higher than the diagnostic yield of the R-EBUS group of 40.0%. The difference was statistically signi cant (P = 0.045). Although the diagnostic yield of VBN + R-EBUS for benign lesions was higher than that of R-EBUS (80.0% vs. 55.0%), the difference was not statistically signi cant (P = 0.123). The diagnosis yield of malignant lesions in R-EBUS group and VBN + R-EBUS group was 71.4% and 74.3% respectively, and the difference was not statistically signi cant (P = 0.788). However, there was no signi cant difference in diagnosis yield between R-EBUS group and VBN + R-EBUS group in different location lesions (Table 2).

Operation time
The operation time of VBN + R-EBUS group was (20.6 ± 12.8) min, that of R-EBUS group was (28.6 ± 14.3) min, and that of VBN + R-EBUS group was signi cantly shorter than that of R-EBUS group (P = 0.016).

Complications
In the two groups, 12 patients had bleeding in the lumen, 4 in VBN + R-EBUS group and 8 in R-EBUS group, respectively. During the operation, the bleeding stopped after the treatment of 1:1000 ice salt water, adrenaline, thrombin, etc. through the bronchoscope biopsy channel, no moderate or severe bleeding and pneumothorax complications occurred.

Discussion
Histopathology is the "gold standard" for the diagnosis of the nature of peripheral pulmonary nodules. For the lesions near the chest wall, percutaneous lung biopsy can be guided by CT. However, for the lesions far away from the chest wall or with large blood vessels and other important organs around, the risks of pneumothorax and bleeding are di cult to implement. Conventional bronchoscopy can reach 4-5 grade bronchus. With the help of ultra-ne bronchoscopy, intratracheal ultrasound and virtual navigation, the operation eld can be extended to 6, 7 grade or even more distal bronchus, which makes our positioning of pulmonary nodules more accurate. Lung tissue biopsy guided by multiple technologies may improve the diagnosis rate of pulmonary nodules.
The diagnostic yield of traditional lung biopsy for pulmonary nodules is not ideal, which may be lower than 20% [10]. According to ACCP lung cancer guidelines [9], radial ultrasound guided lung biopsy should be preferred in the diagnosis of pulmonary nodules, which can be used as an important means of diagnosis. A number of research results showed that [10], compared with the traditional lung biopsy technology, EBUS guided lung biopsy can signi cantly improve the diagnosis yield of peripheral pulmonary nodules. Some researcher considered that [11,12], endobronchial ultrasound failed to achieve self navigation positioning, so 8% − 20.8% of the lesions could not be detected. Virtual navigation technology is one of the new technologies developed in recent years. The image data obtained by preoperative high-resolution thin-layer chest CT without septum scanning is guided into the virtual navigation software system. The three-dimensional reconstruction has the same pixel value range to the inner surface of the bronchus, endows arti cial pseudo color and simulates the condition in the lumen, and obtains the dynamic reconstruction image similar to that in the lumen of the bronchus. Before operation, according to the prompt of chest CT, the operation path of bronchoscope can be determined by calibrating the focus of lung. At present, virtual navigation technology can observe the 0-6 grade bronchi. In this study, we found that the diagnostic yield of VBN combined with R-EBUS group was higher than that of R-EBUS group, which was basically consistent with the results of Ishida et al.
The diagnosis yield of VBN combined with R-EBUS group and R-EBUS group in the lesions with diameter < 20.0 mm was lower than that of the lesions with diameter ≥ 20.0 mm, the difference was statistically signi cant. It can be seen that the diameter of the lesions was positively correlated with the diagnosis yield. In the lesions with diameter < 2.0 cm, the diagnosis yield of VBN combined with R-EBUS group was signi cantly higher than that of R-EBUS group, re ecting the accuracy of virtual navigation. In addition, the operation time of VBN combined with R-EBUS group was signi cantly shorter than that of R-EBUS group, suggesting that VBN can shorten the operation time. Finally, this study suggests that there is no difference between the two groups of complications, and no complications directly related to VBN are found. It can be seen that VBN is a safe and effective auxiliary technology.

Conclusion
In conclusion, VBN combined with R-EBUS has a high diagnostic value for peripheral pulmonary nodules, which can reduce the operation time of tracheoscopy and provide a safe and effective method for the diagnosis of peripheral pulmonary nodules.