Endobronchial ultrasonography–guided transbronchial needle aspiration (EBUS-TBNA) is a technique whereby a needle is inserted into a lesion adjacent to an airway and aspiration biopsy is performed from the lumen of an extrapulmonary airway such as the trachea, principal bronchus, or intermediate bronchus, while the process is monitored on ultrasound images obtained from an convex ultrasound probe mounted on the tip of the bronchoscope. This procedure has rapidly come into widespread use worldwide since the first description in 2004.1–3 When EBUS-TBNA is performed according to the standard procedure, harvesting cells and tissue by means of brush cytology or biopsy with biopsy forceps from lesions adjacent to the external walls of the lobar, segmental, and subsegmental bronchi that are more peripheral to the extrapulmonary airways is difficult, given their nature as extrinsic lesions. Because an ultrasound guide cannot be used in these narrow airways, puncturing by means of conventional TBNA may be performed.4–6 However, the course of pulmonary arteries and veins frequently runs in close proximity to lesions outside the walls of the lobar, segmental, and subsegmental bronchi, and if the location of such lesions cannot be accurately determined, then cytologic and histologic diagnosis is difficult. We have used wedge insertion of a convex endobronchial ultrasound bronchoscope (convex scope) into lobar, segmental, and subsegmental bronchi narrower than the 6.9 mm external diameter of the bronchoscope itself, and performed EBUS-TBNA of lesions adjacent to lobar, segmental, and subsegmental bronchi. We therefore investigated the airway branch in which EBUS-TBNA was possible, the narrowest airway diameter adjacent to the lesion for which insertion and diagnosis could be performed, the feasibility of puncture, and techniques for ensuring procedural success, on the basis of our clinical experience.
PATIENTS AND METHODS
The study cohort comprised patients with 15 lesions examined between February 2007 and October 2012. The lesions were identified on thoracic computed tomography (CT) as being adjacent to the airway branch (lobar, segmental, or subsegmental bronchi) having a mean diameter narrower than the 6.9 mm external diameter of the convex scope. This study was approved by the institutional review board (IRB number: 2058), and informed consent was obtained from all patients.
Multislice CT Scanning
All patients were scanned using a 64-detector CT scanner (Aquilion-64; Toshiba Medical, Tokyo, Japan). CT was performed during breath-hold in deep inspiration with the patient in the supine position. Every patient was carefully instructed how to breathe before the scan. Multislice CT scan parameters were as follows: collimation, 0.5 mm; 120 kV; 200 mA; gantry rotation time, 0.5 seconds; and beam pitch, 0.83 (table feed per gantry, 53 mm; collimation beam width, 64 mm). All images were reconstructed using a standard reconstruction algorithm with a slice thickness of 1 mm and a reconstruction interval of 0.5 mm.
Airway Lumen Segmentation and Measurements
Reconstruction images were transferred to the commercially available workstation (Ziostation; Ziosoft, Tokyo, Japan), and airway segmentation and measurements of airway lumen area were performed using the software included on the workstation. On the workstation monitor, using axial images with a window width of 1250 HU and a window level of −650 HU, we were able to obtain short-axis images that were exactly perpendicular to the long axis at any site. From the centroid point of the lumen, rays fanning out over 360 degrees were examined to determine inner airway walls along the rays using the full width at half-maximum principle. After this process, we were able to obtain values for the airway luminal area. Any outline of automatically obtained airway walls that was obviously out of contour was excluded. The cross-sectional area and maximum, minimum, and mean internal diameters of the airway lumen adjacent to the lesion were calculated using the measurement software (Ziostation)7 (Fig. 1).
We investigated the bronchus into which it was possible to insert the tip of a convex endobronchial ultrasound bronchoscope (external diameter, 6.9 mm; UC-160F; Olympus, Tokyo, Japan), the narrowest internal airway diameter adjacent to the lesion into which insertion could be performed, whether visualization and puncture of the lesion were possible, and the diagnostic yield in clinical cases.
Method of Guiding an Oblique Endoscope Into Segmental Bronchi and Other Peripheral Airways
While the convex scope with oblique forward viewing is being advanced into a peripheral airway, the airway into which it should be inserted cannot be viewed directly. The images from the oblique forward viewing can be seen from the bronchus next to the appropriate bronchus into which the scope should be inserted at the airway bifurcation; the tip of the bronchoscope can then be inserted into the appropriate airway. For example, when oblique forward images of the left main bronchus can be seen, the tip of the bronchoscope can be inserted into the right main bronchus and rotated 180 degrees clockwise to view the right upper lobe bronchus. It can then be advanced into the intermediate bronchus while being rotated 90 degrees anticlockwise to view the right middle lobe bronchus. After a further 180-degree clockwise rotation to view B6, it can be inserted into the basal bronchus. When oblique forward images of B8 and B9 are seen, the tip of the bronchoscope can be inserted into B10, and moved forward in the direction of least resistance to manual pressure (Fig. 2).
Method of Performing EBUS-TBNA From Segmental to Subsegmental Bronchi
Immediately before performing EBUS-TBNA as normal from the extrapulmonary airways such as the trachea, main bronchi, and intermediate bronchi, the tip of the puncture needle (22 G, NA-2012SX-4022; Olympus) must be positioned inside the sheath as close as possible to the tip of the sheath. This is because the sheath is pushed out together with the needle when the latter is moved out to perform the puncture, and this movement of the sheath may make puncturing impossible, particularly in narrow airways. First, the bronchoscope is anchored in the airway lumen of a main bronchus or an adjacent airway so that the tip of the puncture needle sheath is just visible on the bronchoscopic image. The puncture needle is then pushed until the sheath has extended and its downward movement is visible. If the needle is stopped at this point, the needle tip will be immediately behind the tip of the sheath, which is then retracted and anchored in the position in which the sheath is just visible on the bronchoscope image. When performing EBUS-TBNA from a lobar to a subsegmental bronchus, this needle adjustment should therefore also first be carried out in the airway lumen of the main or intermediate bronchus before insertion into the segmental bronchus, and the puncture needle withdrawn from the bronchoscope. When watching oblique forward images of the next bronchus, the tip of the bronchoscope can be inserted into the bronchus beside the target, and moved forward in the direction of least resistance to manual pressure (Fig. 3). After guiding the convex scope into the lobar, segmental, or subsegmental bronchus and confirming the orientation with the largest cross-sectional area of the lesion adjacent to the airway, the puncture direction is determined so as to avoid the pulmonary arteries and veins running alongside the airway. As the opening of the working channel is the proximal site from the convex transducer, the bronchoscopist notices that the approach of the needle is sometimes from the proximal bronchus. The puncture needle that has been advanced as close as possible to the tip of the sheath is then reinserted through the forceps channel of the bronchoscope. Puncturing is performed after reconfirming the position of the lesion. Suction pressure is then applied to the puncture needle within the subject lesion to be aspirated, and the operation of inserting and withdrawing the needle is repeated around 7 times. After the negative pressure applied to the puncture needle has been released, the needle is removed from the lesion, and the tissue cells and tissue remaining in the needle are sent for cytologic and histologic diagnosis.
Details of the 15 cases investigated are given in Table 1. Two lesions were located in the right upper lobe, 4 in the right middle lobe, 6 in the right lower lobe, 1 in the left lingular segment, 2 in the left lower segment, 4 in the lobar bronchi, 7 in the segmental bronchi, 3 in the subsegmental bronchi, and 1 in the subsubsegmental bronchi.
Mean cross-sectional area of the bronchial lumen for the 13 lesions that could be punctured was ≥15.9 mm2 and the mean internal diameter of the bronchus was ≥4.5 mm. For all 13 of these lesions, puncturing could easily be performed from the segmental to the subsegmental bronchi without interference from bronchial cartilage.
Cytologic or histologic diagnosis by EBUS-TBNA was possible in 11 (73.3%) of the 15 cases. In 2 of the 4 undiagnosed lesions, the mean internal diameter was <4.5 mm and the convex ultrasound probe at the tip of the bronchoscope was unable to reach the lesion, meaning that puncture itself was impossible. The remaining 2 of these 4 lesions were a necrotic pulmonary metastasis of renal carcinoma that was touching right B3 and a hemangioma that was touching right B9, and little tissue was able to be harvested, therefore cytologic and histologic diagnosis could not be performed.
There was no occurrence of complications such as hemorrhage, pneumomediastinum, or pneumothorax in any case.
Figure 4 shows a representative case of adenocarcinoma (long diameter, 13 mm; short diameter, 12 mm) touching the right B10. The internal diameter of B10, the airway touching the lesion, showed maximum and minimum diameters of 6.1 and 3.4 mm, respectively (mean diameter, 4.6 mm) (Fig. 1). The convex scope was inserted into the lower lobe bronchus while viewing the middle lobe bronchus, and B10 while viewing B8 and B9 (Fig. 2). Scanning with the bronchoscope fitted with the ultrasound probe was performed in B10 while rotating the scope left and right, to determine the axis from which the largest cross-section of the lesion could be visualized. The course of the pulmonary artery around this subject lesion was confirmed, and puncture was performed so as to avoid this vessel (Fig. 5). Adenocarcinoma was detected from cytologic and histologic diagnoses of the cells and tissue was harvested by the puncture needle (Fig. 6).
We were able to perform EBUS-TBNA of lesions adjacent to lobar, segmental, and subsegmental bronchi with s mean internal diameter of ≥4.5 mm using the convex scope that is in current use. The convex scope we used has a large external diameter of 6.9 mm, and the convex ultrasound probe at the tip is itself a hard component that does not change shape, meaning that use is limited by the external diameter and branching angle of the airway concerned. Among airways more peripheral to the lobar bronchus, although the scope could be inserted into the right B3, the right middle lobar bronchus, B7, and bilateral B8, B9, and B10, it could not be inserted into right B1, B2, or B6, or left B1+2, B3, or B6, as these would require flexion of the bronchoscope. Development of a bronchoscope with a finer tip of shorter hard length would reduce the number of locations into which insertion is difficult.
Matsuoka et al7 reported a validation study of the luminal area (range, 1.5 to 45.6 mm2) using the same software with the same conditions of CT. The mean coefficient of variation for measurements of luminal area was 1.9%, and there was a strong correlation between measurements of luminal areas obtained with the software and the size measured with an optical micrometer caliber. Other reports of CT measurements of internal airway diameter include a study by Maki et al,8 who investigated the effect of helical pitch and tube current on the resolution of the diameter of pulmonary arteries and airways in multiplanar reconstruction images of the lungs. Maki and colleagues stated that in high-quality mode (helical pitch, 1:7) and high–tube current mode (250 mAs), it was possible to visualize pulmonary arteries with a diameter of approximately 100 µm and airways with a diameter of approximately 1000 µm. Edward et al9 used spiral CT to measure the internal airway diameter of segmental to subsegmental bronchi, and reported that a slice thickness ≤1.5 mm and pitch 1.5 should be used. All images in this study were reconstructed using a standard reconstruction algorithm with a slice thickness of 1 mm and a reconstruction interval of 0.5 mm.
Which airways have a mean internal diameter of 4.5 mm anatomically? According to Weibel,10 the diameter of lobar and segmental bronchi is 5 to 8 mm, whereas that of subsegmental bronchi and bronchioles is 1.5 to 3.0 mm, meaning that an internal diameter of 4.5 mm corresponds to lobar and segmental to subsegmental bronchi. The 13 cases in our present study with a mean internal diameter of ≥4.5 mm comprised 4 in the middle lobar bronchus, 7 touching a segmental bronchus, and 2 touching a subsegmental bronchus. Despite the presence of bronchial cartilage and other obstructions in the airway wall, we were able to insert the bronchoscope into bronchi narrower than its external diameter by pushing in the hard tip. In our experience of using EBUS to visualize peripheral pulmonary lesions with EBUS, inserting a 2.5-mm-diameter radial probe (UM-3 R; Olympus) as far as the subpleural bronchioles was easy. The mean diameter of the subpleural bronchioles has been reported as around 1 mm,10 and the fact that these 1-mm-diameter bronchioles (cross-sectional area, 0.79 mm2) can be expanded to a diameter of 2.5 mm (cross-sectional area, 4.91 mm2) suggests that airways could be enlarged to 2.5 times their diameter and approximately 6.2 times their cross-sectional area. The cross-sectional area of the tip of the 6.9-mm-diameter bronchoscope used in the present study was 37.37 mm2, and this scope could be inserted into airways with a mean diameter of 4.5 mm (15.90 mm2), enlarging these airways to 1.5 times their diameter and approximately 2.4 times their cross-sectional area. The reasons for the smaller degree of dilation of the lobar, segmental, and subsegmental bronchi compared with the subpleural peripheral bronchioles may include the presence of cartilage in the walls of the lobar, segmental, and subsegmental bronchi and the fact that their walls are comparatively thicker.
Few previous reports have described biopsy of pulmonary peripheral lesions under real-time image monitoring. Shinagawa et al11 reported ultrasound-guided biopsy using the 2 working channels of a double-channel bronchoscope to insert a radial narrow-diameter ultrasound probe and biopsy forceps at the same time. Because the cross-sectional plane produced by the radial probe generates a short-axis image perpendicular to the airway, this method has the disadvantage that the cross-sectional slice image of the biopsy forceps cannot be monitored.
The current limitations of this procedure include the fact that puncture from airways narrower than a mean diameter of 4.5 mm is difficult because of the size of the convex ultrasound probe at the tip, and that the oblique endoscope does not allow forward viewing, making it difficult to guide the bronchoscope into peripheral bronchioles. The development of ultrasound bronchoscopes with a smaller external diameter that are capable of forward viewing would enable this procedure to be used even for peripheral lesions.
EBUS-TBNA can be performed by inserting a 6.9 mm EBUS bronchoscope into airways with a mean diameter ≥4.5 mm as measured on CT before bronchoscopy.
1. Silvestri GA, Tanoue LT, Margolis ML, et al..The noninvasive staging of non-small cell lung cancer: the guidelines.Chest.2003;123suppl147S–156S.
2. Wallace BM, Pascual MJ, Raimondo M, et al..Minimally invasive endoscopic staging of suspected lung cancer.JAMA.2008;299:540–546.
3. De Leyn P, Lardinois D, Van Schil PE, et al..ESTS guidelines for preoperative lymph node staging for non-small cell lung cancer.Eur J Cardiothorac Surg.2007;32:1–8.
4. Katis K, Inglesos E, Zachariadis E, et al..The role of transbronchial needle aspiration in the diagnosis of peripheral lung mass or nodules.Eur Respir J.1995;8:936–966.
5. Reichenberger F, Weber J, Tamm M, et al..The value of transbronchial needle aspiration in the diagnosis of peripheral pulmonary lesions.Chest.1999;116:704–708.
6. Trisolini R, Cancellieri A, Tinelli C, et al..Performance characteristics and predictors of yield from transbronchial needle aspiration in the diagnosis of peripheral pulmonary lesions.Respirology.2011;16:1144–1149.
7. Matsuoka S, Kurihara y, Yagihashi K, et al..Airway dimensions at inspiratory and expiratory multisection CT in chronic obstructive pulmonary disease: correlation with airflow limitation.Radiology.2008;28:1042–1049.
8. Maki D, Takahashi M, Ushio N, et al..Resolution of pulmonary multiplanner reconstruction images from 0.5-mm theoretical isotropic data: a fundamental study using an inflated and fixed lung specimen.Acta Med Okayama.2007;61:63–69.
9. Edward PD, Bull RK, Brown VS, et al..Spiral CT optimization for measurement of bronchial lumen diameter using an experimental model.Br J Radiol.2000;73:715–719.
10. Weibel ER.High Resolution Computed Tomography of the Pulmonary Parenchyma: Anatomical Background.1990.Scottdale, AZ:Fleischner Society Symposium on Chest Diseases.
11. Shinagawa N, Yamade N, Asahina H, et al..Transbronchial biopsy for peripheral pulmonary lesions under real-time endobronchial ultrasonographic guidance.J Bronchol Intervent Pulmonol.2009;16:261–265.