INTERVENTIONAL PULMONOLOGY IN OTHER JOURNALS: Commentary on Selected Publications
Transbronchial biopsy with virtual CT bronchoscopy and nodal highlighting.
Radiology 2001;221:531–6. Hopper KD, Lucas TA, Gleeson K, Stauffer JL, Bascom R, Mauger DT, Mahraj R. Departments of Radiology, Medicine, and Health Evaluation Sciences, Penn State University, Hershey, Pennsylvania, U.S.A.
This study was performed to evaluate whether the addition of information from virtual CT–bronchoscopy (VCB) improved the success rate of bronchoscopic needle aspiration/biopsy (BNA) of subcarinal and aortopulmonary lymph nodes. For this study, a senior radiology investigator experienced in VCB selected 35 sites of adenopathy depicted on CT scans obtained from 14 patients (8 men and 6 women; mean age, 58.6 years). The sites were chosen to provide a spectrum of anatomic sites (7 aortopulmonary window, 9 pretracheal, 6 subcarinal, 10 hilar, and 3 high pretracheal). The transition from pretracheal to high pretracheal adenopathy was 3.0 cm above the carina. The lymph nodes were selected to provide a representation of sizes (<1.5 cm, n = 7; 1.5–3.0 cm, n = 13; >3 cm, n = 15). The study population was limited to patients with normal bronchi and mucosa depicted at actual bronchoscopy and to patients with extrabronchial lymphadenopathy. Each patient selected for the study underwent repeat CT (nonenhanced helical CT) of the tracheobronchial tree from 3 mm above the aortic arch to the top of the right hemidiaphragm. By using the editing software, the selected lymph nodes were highlighted on all transverse images that displayed them. This highlighting process required 3 to 4 minutes by the investigator. Creation of the VCB images required an average 8 to 10 minutes of the investigator's time. The same radiology investigator who selected the sites, marked the targeted lymph nodes and created the cine images. The distance from the marked biopsy site to the center 75% of the lymph node was also measured, as was the distance to the nodal center. The distances from the site selected by the pulmonologists for potential BNA to the edge and center of the actual nodal site were determined both with and without lymph node highlighting for the overall data, specific pathologic sites, and between and within individual pulmonologists. The lymph nodes or tumors selected for this study were all adjacent to the bronchial wall and would have been within easy reach of a biopsy needle with the use of a biopsy depth of at least 5 mm. The results showed that VCB with lymph node highlighting increased biopsy success rates significantly for pretracheal, hilar, and high pretracheal adenopathy (p < 0.5). VCB with lymph node highlighting added little to the choice of a biopsy site of subcarinal adenopathy. The three pulmonologists would have chosen the biopsy sites successfully in 94% of these lymph nodes regardless of whether VCB images were highlighted. The addition of lymph node highlighting by VCB allowed the pulmonologists to place successfully the needle for aortopulmonary window lymph nodes in 71% of patients (15 of 21), as opposed to 62% of patients (13 of 21) without highlighting. However, this increase was not significant (p = 0.73). For pretracheal, hilar, and high pretracheal adenopathy, the addition of VCB with lymph node highlighting increased significantly the successful choice of a biopsy site. The successful choice of a biopsy site for hilar adenopathy increased from 37% (11 of 30) to 83% (25 of 30; p = 0.001) and for pretracheal nodes increased from 59% (16 of 27) to 85% (23 of 27; p = 0.07). For the three high pretracheal lymph nodes, the percentage increased from 11% (1 of 9) to 78% (7 of 9; p = 0.07). Overall, for all 105 measurements (35 nodal sites times three pulmonologists), the addition of VCB with lymph node highlighting increased the successful choice of a biopsy site from 55% (58 of 105) to 83% (87 of 105; p < 0.001). For choice of a biopsy site in the central 75% of the lymph node, the percentage increase increased from 41% (43 of 105) to 70% (74 of 105; p < 0.001). Review of publications reveals that BNA of enlarged thoracic lymph nodes caused by malignancy carries a diagnostic rate of less than 50 to 80%. The wide variation in the success rate is determined by anatomic location of the lymph nodes, size of the enlarged lymph nodes, technique used by the bronchoscopist, and the experience of the bronchoscopist. The availability of helical CT scanners now permits development of VCB pictures so that the bronchoscopist can use this information to improve the diagnostic rate from BNAs. Furthermore, the recent introduction of fluoroscopic CT scanners also permits real-time visualization of thoracic lymph nodes and the BNA procedure. Even though VCB technique has been available for some years, the information derived has not been used consistently by bronchoscopists. The study by Hopper et al. shows that combining data from VCB and lymph node highlighting to obtain specimens from BNA will improve the diagnostic rate significantly.