Detection of peripheral lung cancer is on the rise because of the advancements in diagnostic imaging. However, the diagnostic yield remains low at 30% to 73% using conventional bronchoscopy, and is lower for lesions less than 20 mm.1–5
We performed bronchoscopy using endobronchial ultrasound (EBUS) with guide sheath (EBUS-GS) in patients for over 3 years to study the improvement to the diagnostic yield. EBUS-GS is known to be an excellent tool for the confirmation of the biopsy point; however, navigating to the target lesion depends on the analytic ability and the proficiency of each bronchoscopist to interpret computed tomography (CT) images. We hypothesize that combining a navigation system with EBUS-GS might improve the diagnostic yield and shorten procedure times. The purpose of this study is to evaluate whether a navigation system can contribute to the diagnostic yield, and can shorten the time required to determine the biopsy position and fluoroscopy.
PATIENTS AND METHODS
Between September 2006 and May 2009, a prospective study was conducted using EBUS-GS with a flexible bronchoscope and a virtual bronchoscopy navigation (VBN) system on patients with solitary peripheral nodules of the lung. We defined a “peripheral lesion” as a lesion considered to be bronchoscopically invisible. We evaluated CT images before the bronchoscopic procedure to select the lesion located more distal than subsegmental bronchus. The Institutional Review Board approved this study and all patients provided written informed consent. Enrollment criteria for this study stated that peripheral lung lesions need to be confirmed by bronchoscopic procedures with a pathological diagnosis, and that visible lesions using bronchoscope were excluded. Cases deemed benign on CT image and clinical inference were excluded from bronchoscopic procedures.
We enrolled 112 consecutive patients with peripheral lung lesions at St Marianna University School of Medicine Hospital. Patients were allocated into the VBN/EBUS-GS group and into the EBUS-GS group. This study was not randomized. During the design procedure, we decided on a randomized scheme using a randomized number table. However, during the patient inclusion process, the randomized trial was interrupted because of mechanical trouble. Owing to this trouble, this study was changed to a prospective single-center study. Eleven physicians, all with 3 years experience or more using EBUS-GS participated in the bronchoscopic procedures. We compared the groups for the time needed to determine the biopsy position, for the time using fluoroscopy, and for the bronchoscopic diagnostic yield. In 57 cases, we examined diagnostic bronchoscopy using EBUS-GS with VBN (VBN/EBUS-GS group), and in 55 cases we used EBUS-GS alone (EBUS-GS group). In both groups, the required time needed to determine the biopsy position was defined as the start of bronchoscopic procedure to decision of the biopsy point, which was confirmed using EBUS. The time spent in sampling lesions during bronchoscopy was excluded. We also evaluated the time using fluoroscopy. We defined the bronchoscopic diagnostic yield as the ratio of cases diagnosed by bronchoscopy to all enrolled cases in this study.
Virtual Bronchoscopic Navigation
CT examination of all 112 cases were performed first using a helical CT scanner (Aquilion64; Toshiba Electronics; Tokyo, Japan) with the following parameters: 120 kV, 200 mA, 0.5-mm collimation, beampitch 0.83 (helicalpitch 53). and rotation time, 0.5 seconds. We evaluated the bronchus leading to the lesion using axial CT images, which were reconstructed at 0.5-mm intervals. Digital images and communications in medicine data were taken from helical CT results. On the basis of these data, VBN images to the target lesion were produced using specific software (Bf-NAVI, Olympus, Tokyo, Japan). When the bronchus connecting the lesion was unclear, the closest bronchus to the lesion was produced as a VBN image. The route to the lesion was selected by VBN, and the bronchi for bronchoscope advancement were marked. VBN has the following characteristics in terms of image display: (1) VBN images are displayed as animated images and advancement to the branches in the peripheral lung is similar to operating a bronchoscope; (2) at each branch site, VBN can be rotated as freely as rotating a bronchoscope to match the images to the bronchoscopic view; and (3) thumbnails of VBN images at each bronchial branch are displayed as a catalog. The navigation system was created by Fumihiro Asano et al6 and was developed in cooperation with Olympus and is commercially available in Japan.
In all cases, patients were premedicated with 25 mg of hydroxyzine and 0.5 mg of atropine sulfates. Continuous pulse oximetry and electrocardiogram monitoring were performed during the flexible bronchoscopic procedure with blood pressure measured every 5 minutes. Local anesthesia for the upper respiratory tract was administered using lidocaine (2% solution), and oxygen was applied by a nasal cannula with flows adjusted upward from 2 L/min to maintain SpO2 above 95%. All patients were administered 2.5 to 5 mg midazolam for conscious sedation. The fiberoptic bronchoscope was advanced beyond the vocal cords visualizing all segments of the bronchial tree. The bronchoscope was then advanced to the segment suspected to be the location of the lesion. Under local anesthesia, a 20 MHz mechanical-radial-type probe (UM-S20-20R and UM-S20-17S; Olympus) with guide sheath (SG-201C and SG-200C; Olympus) was inserted through the working channel of the bronchoscope (BF1T-260R and BF-P260F; Olympus) and into the bronchus, which seemed most closely related to the lesion by CT or VBN assessment. Once the probe was maneuvered to the targeted position, a clear image of the lesion was detected by EBUS. If the lesion was not detected by EBUS on the first try, we attempted different pathways until the lesion was confirmed. After the lesion was localized by EBUS, the probe was withdrawn leaving the GS in place. Biopsy forceps (BF-19C-1; Olympus) or a bronchial brush (BC-202D-5010; Olympus) was then introduced through the sheath to obtain cytology samples and 2 to 3 biopsy specimens. Lesions could be approached repeatedly through the GS. However, if the target lesion could not be seen by EBUS, biopsy and brush forceps were directly inserted into the bronchus under fluoroscopic guidance and samples were collected. Complete details of this method are reported previously.7
All analyses were performed using SAS software (Release 8.02; SAS Institute, Cary, NC). The time needed to navigate to the target lesion, time for fluoroscopy, and bronchoscopic diagnostic yield were evaluated by paired t tests and the Wilcoxon signed-rank test. Characteristics of patients and the diagnostic yield of each tumor size were compared by paired t tests and the χ2 test. All results are presented as mean±SD. Tests of significance were 2 sided, and P values of less than 0.05 were considered statistically significant.
One hundred twelve peripheral lung lesions were biopsied in 112 participants without any bronchoscopic complications, including pneumothorax and major bleeding. Patients included 79 men with a median age of 73 years (range: 32 to 89 y). The mean tumor size was 29.4 mm (range: 8 to 84 mm).
Of 112 cases with peripheral lesions, 101 (90.2%) were diagnosed as malignant tumors, 54 (94.7%) cases for VBN/EBUS-GS and 47 (85.5%) for the EBUS-GS group. Of these 112 cases, characteristics of sex, age, location of tumor, and tumor size are shown in Table 1. There was no significant difference in patient's sex, age, location of the lesion, and lesion size between groups.
Table 2 shows the final diagnosis of all cases. Adenocarcinoma was the most prevalent at 53.6%, followed by squamous cell carcinoma 20.5%, nonsmall cell lung carcinoma 9.8%, small cell lung carcinoma 4.5%, metastatic lung carcinoma 2.7%, and others 9.8%. Of the 57 cases in the VBN/EBUS-GS group, 35 (61.4%) cases were adenocarcinoma, 11 (19.3%) squamous cell carcinoma, 6 (10.5%) nonsmall cell carcinoma, 3 (5.3%) small cell carcinoma, 2 (3.5%) metastatic lung cancer, and 3 (5.3%) other benign lesions. For the EBUS-GS group, 28 (50.9%) cases were adenocarcinoma, 12 (21.8%) squamous cell carcinoma, 5 (9.1%) nonsmall cell carcinoma, 2 (3.6%) small cell carcinoma, and 8 (14.5%) other benign lesions. Undiagnosed cases were defined as cases that could not be determined by the bronchoscopic procedure; however, final diagnosis was provided by other methods including CT-guided biopsy, surgical procedure, transbronchial needle aspiration, and a 6-month follow-up using CT image.
A representative case (case 1) is shown in Figure 1. A 73-year-old woman was admitted to our hospital with an abnormal opacity on chest radiograph. Chest radiography and CT showed a peripheral pulmonary nodule measuring 8 mm in diameter in the left S1+2b. The bronchoscope was advanced to the B1+2bi bronchus after VBN. Adenocarcinoma was diagnosed from brushing cytology.
Another representative case (case 2) is shown in Figure 2. A 68-year-old man was admitted to our hospital with similar findings as the aforementioned case. EBUS probe was advanced to the B2b bronchus guided by VBN, and the target lesion with a diameter of 14 mm was clearly detected. Adenocarcinoma was diagnosed using EBUS-GS-guided TBB.
A significant difference was observed in the time needed to determine the target bronchus (5.54±0.57 min in VBN/EBUS-GS group vs. 9.27±0.86 min in EBUS-GS group, P<0.01). There was no significant difference in the time using fluoroscopy, total bronchoscopic procedure time, and bronchoscopic diagnostic yield for each group (Table 3).
Lesions were subsequently evaluated according to size in each group (Table 4).
In VBN/EBUS-GS group, of the 22 lesions less than 20 mm in diameter, 21 lesions (95.5%) were visible on EBUS and 16 lesions (72.7%) were diagnosed using this procedure. In 18 lesions measuring 20 to 30 mm, 16 (88.9%) were visible and 18 (100%) were diagnosed. In 17 lesions greater than 30 mm, 16 (94.1%) were visible and 14 (82.4%) were diagnosed.
In EBUS-GS group, of the 16 lesions less than 20 mm, 13 (81.3%) were visible on EBUS and 9 (56.3%) were diagnosed. In 21 lesions measuring 20 to 30 mm, 20 (95.2%) were visible and 18 (85.7%) were diagnosed. In 18 lesions greater than 30 mm, 16 (88.9%) were visible and 17 (94.4%) were diagnosed.
Diagnostic sensitivity, specificity, positive predictive value, negative predictive value, and diagnostic accuracy of VBN/EBUS-GS for malignant lesion were 85.2%, 66.7%, 100%, 20%, and 84.2%. The diagnostic sensitivity was 71.4% for lesions less than 20 mm in diameter, 100% for lesions 20 to 30 mm, and 87.5% for lesions greater than 30 mm. Diagnostic sensitivity, specificity, positive predictive value, negative predictive value, and diagnostic accuracy of the EBUS-GS group for malignant lesion were 83.7%, 50%, 100%, 27.3%, and 80%, respectively. Diagnostic sensitivity was 58.3% for lesions less than 20 mm in diameter, 89.5% for lesions 20 to 30 mm, and 94.4% for lesions greater than 30 mm.
In this study, we implemented the VBN system, as it is simple to use and it produces only 1 route to the lesion.6,8–10 Although we presented that VBN was useful in shortening the time needed to determine on the biopsy position, we could not produce a significant difference in the diagnostic yield. We believe this result is due to the fact that EBUS was used in confirming the location for biopsy in all patients, whereas VBN was used to assist in directing EBUS to the target location.
Compared with VBN, electromagnetic navigation works in real time and the operator is always aware of the probe's exact location.11–14 Moreover with the real-time mapping, there is the option of various approaches. However, as the electromagnetic navigation system is not commercially available in Japan, it was not an option for this study.
The EBUS-GS method has been proven to be valuable in confirming the accuracy of biopsy points and in improving the diagnostic yield of peripheral lung lesions.7,15–18 However, it is difficult to select the bronchus leading to the target lesion and to navigate the bronchoscope and biopsy forceps accurately using EBUS-GS.
In this study, VBN with EBUS-GS was used to navigate the bronchoscope in patients with pulmonary peripheral small lesions. VBN with EBUS-GS was found to be a novel technique for the noninvasive evaluation of the tracheobronchial tree, and for navigating the bronchoscope to the accurate target lesion.
With regard to the effectiveness of VBN/EBUS-GS for small lesions less than 20 mm, the diagnostic sensitivity was higher, although not statistically significant (72.7% in VBN/EBUS-GS group vs. 56.3% in EBUS-GS group). In contrast, lesions greater than 30 mm showed a trend toward lower diagnostic sensitivity in VBN/EBUS-GS (82.4% in VBN/EBUS-GS group vs. 94.4% in EBUS-GS group). This might have been due to the bronchial traction and occlusion of the tumor, which may have led to technical difficulties in VBN software to detect the air layer of the bronchi on CT.
There was no significant difference in time using fluoroscopy. We assume that this was because of the time for biopsy procedures rather than for the selection of an appropriate bronchus to confirm the lesion using EBUS.
Although the planning time for VBN was not an endpoint in this study, it ranged between 15 and 50 minutes. The large time discrepancies between cases were likely caused by the software's inability to detect the target area from CT images. When the software was successful in detecting the peripheral branch, the planning time decreased. Adversely, if the software could not detect a peripheral branch, the planning process required manual input to recognize the peripheral branch, thus lengthening the planning time. In contrast, between 5 and 10 minutes were spent for planning for patients in the EBUS-GS group to identify the bronchus leading to the lesion on CT image. The VBN/EBUS-GS group needed more time for planning. However, we could not always clarify the form of segmental and/or subsegmental bronchi by CT images. Furthermore, “human error” might have occurred in the planning of the EBUS-GS group.
In this study, 11 physicians all with over 3 years experience using EBUS-GS participated in the bronchoscopic procedures. We did not analyze the diagnostic yield of each physician, the time needed to determine biopsy position, or the amount of time using fluoroscopy, as each physician did not have a sufficient number of cases for proper analysis.
We could not produce a significant difference in the bronchoscopic diagnostic yield for each group. As EBUS-GS was used in both groups, the diagnostic yield was high, confirming that EBUS-GS alone is an important method in locating peripheral tumors. However, when VBN was combined with EBUS-GS, procedure times were reduced, and although not significant the diagnostic yield was higher suggesting that VBN was able to direct EBUS-GS to a more optimal position for biopsy collection.
Even though a significant difference was observed in the time needed to determine a biopsy position, there was no significant difference in the total bronchoscopic procedure time. We consider this may be due to not prescribing the number biopsy specimens in this study design, and to the possibility that the number of biopsy specimens increased in the VBN/EBUS-GS group opposed to the EBUS-GS group. With VBN, we were able to confidently insert the probe into the bronchus that was considered to lead to the lesion. The number of biopsies was not prescribed in this study; however, the total procedure time might have been influenced by the number of biopsies performed. If the number of biopsies is equal between 2 groups, there is a possibility that the total procedure time for the VBN/EBUS-GS group might have been shorter.
We believe that VBN combined with EBUS-GS can contribute to shortening the time used to determine the biopsy position; however, further studies are needed.
The investigators thank Mr Jason Tonge of St Marianna University School of Medicine for his careful language review of this manuscript.
1. Stringfield JT, Markowitz DJ, Bentz RR, et al. The effect of tumor size and location on diagnosis by fiberoptic bronchoscopy Chest.. 1977;72:474–476
2. Cortese DA, McDougall JC. Biopsy and brushing of peripheral lung cancer with fluoroscopic guidance Chest.. 1979;75:141–145
3. Torrington KG, Kern JD. The utility of fiberoptic bronchoscopy in the evaluation of the solitary pulmonary nodule Chest.. 1993;104:1021–1024
4. Chechani V. Bronchoscopic diagnosis of solitary pulmonary nodules and lung masses in the absence of endobronchial abnormality Chest.. 1996;109:620–625
5. Wassermann K, Gassanov N, Atay Z, et al. The impact of cytology on the bronchoscopic diagnosis of lung cancer J Bronchol.. 2004;11:154–159
6. Asano F, Matsuno Y, Shinagawa N, et al. A virtual bronchoscopic navigation system for pulmonary peripheral lesions Chest.. 2006;130:559–566
7. Kurimoto N, Miyazawa T, Okimasa S, et al. Endobronchial ultrasonography using a guide sheath increases the ability to diagnose peripheral pulmonary lesions endoscopically Chest.. 2004;126:959–965
8. Asano F, Matsuno Y, Matsushita T, et al. Transbronchial diagnosis of a pulmonary peripheral small lesion using an ultrathin bronchoscope with virtual bronchoscopic navigation J Bronchol.. 2002;9:108–111
9. Asahina H, Yamazaki K, Onodera Y, et al. Transbronchial biopsy using endobronchial ultrasonography with a guide sheath and virtual bronchoscopic navigation Chest.. 2005;128:1761–1765
10. Omiya H, Kikuyama A, Kubo A, et al. A feasibility and efficacy study on bronchoscopy with a virtual navigation system J Bronchol Intervent Pulmonol.. 2010;17:11–18
11. Becker HD, Herth F, Ernst A, et al. Bronchoscopic biopsy of peripheral lung lesions under electromagnetic guidance. A pilot study J Bronchol.. 2005;12:9–13
12. Schwarz Y, Greif J, Becker HD, et al. Real-time electromagnetic navigation bronchoscopy to peripheral lung lesions using overlaid CT images: the first human study Chest.. 2006;129:988–994
13. Eberhardt R, Anantham D, Herth F, et al. Electromagnetic navigation diagnostic bronchoscopy in peripheral lung lesions Chest.. 2007;131:1800–1805
14. Eberhardt R, Gompelmann D, Herth FJ. Electromagnetic navigation in lung cancer: research update Expert Rev Respir Med.. 2009;3:469–473
15. Becker HD, Herth F. Is endobronchial ultrasound indispensable in clinical practice? Pro: endobronchial ultrasound J Bronchol.. 2002;9:145–151
16. Kikuchi E, Yamazaki K, Sukoh N, et al. Endobronchial ultrasonography with guide-sheath for peripheral pulmonary lesions Eur Respir J.. 2004;24:533–537
17. Shirakawa T, Imamura F, Hamamoto J, et al. Usefulness of endobronchial ultrasonography for transbronchial lung biopsies of peripheral lung lesions Respiration.. 2004;71:260–268
18. Yoshikawa M, Sukoh N, Yamazaki K, et al. Diagnostic value of endobronchial ultrasonography with a guide sheath for peripheral pulmonary lesions without x-ray fluoroscopy Chest.. 2007;131:1788–1793