The pneumologist has a number of procedures available (bronchoscopy, computed tomography [CT], magnetic resonance [MR] imaging, spirometry) that permits the study and diagnosis of diseases of the airways. However, until recently, the mediastinum was inaccessible for bronchoscopy and could only be explored with surgical techniques such as mediastinoscopy, anterior mediastinotomy, lateral sternotomy, hiliostomy, or thoracotomy. Although these procedures have good sensitivity and specificity, they require the inconvenience of locating a hospital with thoracic surgery and general anesthesia departments, which generates a high economic cost and is not without morbidity (4%) and mortality (0.5%). 1
It was therefore necessary to develop a technique without these inconveniences, which could be used by all pneumologists with proper training regardless of the size of the hospital in which they work, without general anesthesia, in either inpatients or outpatients, with low morbidity and no mortality, and with low economic cost.
The first description of bronchial needle aspiration (BNA) was provided by Schieppati 2 in 1949. Thirty years later, flexible needles that could be used with flexible fiberoptic bronchoscopy 3 were developed, and Wang et al. demonstrated the multiple applications of this technique 4–7 using different types of needles.
Despite being a new technique that filled a gap between 2 respiratory pathology specialties, one being pneumology (medical) and the other thoracic surgery (surgical), BNA continues to be underused among pneumologists in Spain, 8 Europe, and the United States. 9 The lack of training available for many pneumologists before the appearance of this technique, the absence of training programs for fellows in many hospitals, the failure to achieve immediate results, and the fear of complications are some of the reasons that can be cited to explain the underuse of BNA. 10
We prospectively studied 58 patients, 17 using a cytologic needle (MW-222 Mill-Rose Laboratory Inc., 7310 Corporate Blvd. Mentoi, Ohio 44060 USA) and 41 using a histologic one (MW-319 Mill-Rose Labs Inc.) according to a previously described technique 10 All patients were included in a protocol that also served to collect information. Patients gave written informed consent before the procedures were outlined and possible risks were explained.
Patients taking oral anticoagulants (warfarin) suspended this therapy 48 hours before puncture, and heparin of low molecular weight was given. As premedication, 1 hour before bronchoscopy atropine (0.05–1 mg intramuscularly) and midazolam (0.07 mg/kg intravenously) were administered until achieving conscious sedation. Flumazenil (Anexate, Roche Farma, SA 42, 28027 Madrid Spain) was available if required.
Bronchoscopy was performed in both inpatients and outpatients after a detailed CT study. Bronchoscopy was used to locate the lesion, which was the objective of the study (tumor, adenopathy), and to avoid damaging vascular structures adjacent to the puncture sites. The map of the mediastinum and lymph nodes described by Wang was borne in mind as the endoscopic reference. 11 Sometimes, an extrinsic compression of the tracheal or bronchial wall oriented us toward one of the sites where puncture was made, above all, if a correlation existed with that observed in the CT.
Two or 3 punctures were made in each suspected area until achieving an adequate sample, which was evaluated by direct observation of the pneumologist who made the puncture and the pathology technician who processed the samples. A chest radiograph was obtained in all patients 2 hours after the bronchoscopic needle aspiration and biopsy (BNAB), and a follow up was made to verify whether complications existed.
Samples were considered adequate when sufficient material was obtained for direct extension and fixing when we used a cytologic needle. Samples were processed immediately according to the “dry smear technique” followed by “wet method–saline flush.”10 We also obtained one or more cores of tissue if we were using the histologic needle, which was processed as biopsy material by wet method–saline flush. Samples in which abundant lymphocytes, abundant neoplastic cells, or any specific histology (eg, granulomas) existed were considered valid samples for diagnosis. 12
The patients in whom we obtained a definitive diagnosis were considered true-positives. Those in whom we did not obtain a diagnosis with a valid sample, who underwent other procedures (mediastinoscopy, evolution, cure with medical treatment, and so on) or who were free of disease were considered true-negatives.
Data from the Economic Evaluation Service at our hospital revealed that the mean cost per patient for mediastinoscopy and other surgical techniques for the diagnosis of mediastinal pathology is 3,398.00 dollars, whereas the cost of BNAB, including the needles, is estimated to be 339.00 dollars.
We prospectively studied 58 patients (45 men, 13 women). Mean age was 62.15 (range, 29–85 y). Patients were referred to a pneumologist from different departments in the hospital: nephrology (1), internal medicine (11), medical oncology (1), pneumology (21), thoracic surgery (4), neurosurgery (2), general surgery (2), vascular surgery (1), ENT (1), and from other hospitals in our autonomous community (14). Bronchoscopies had been previously performed in 29 patients (50%) who were undiagnosed.
Bronchoscopy with BNAB was made in 35 hospitalized patients (60%) and in 23 outpatients (40%).
The findings of CT before bronchoscopy consisted of paratracheal tumors, lymph nodes more than 1cm in the short axis (lower right paratracheal, subcarinal, aortic–pulmonary window, and bilateral hilar lymph nodes), extrinsic compressions of the trachea and main bronchus, and submucous tumors. Indications for BNAB are shown in Table 1.
The BNAB site and the number of patients studied and diagnosed in each area are shown in Figure 1. A good correlation existed between lesions observed with CT (eg, masses, adenopathies, infiltrated submucous) and the findings of bronchoscopy (eg, enlarged carinas, extrinsic compression) in 33 patients (57%); among CT, bronchoscopy and the site chosen for puncture in 29 (50%); between CT and the site chosen for puncture in 52 (90%); and between bronchoscopy and the site chosen for puncture in 32 (55%).
We obtained a valid sample for diagnosis in 58 patients (100%) and a definitive cytohistologic diagnosis in 50 patients (86%) (Table 2). Three false-negatives were confirmed, 2 by thoracotomy and 1 by mediastinoscopy. Five patients were true-negatives: 3 as a result of evolution, 1 as a result of mediastinoscopy, and 1 as a result of thoracotomy. BNAB was the only diagnostic technique in 36 patients (62%), and a visible endobronchial lesion did not exist in any of these cases. The size of the biopsy core obtained using the histologic needle (MW-319) was similar to those obtained with transbronchial forceps biopsy or by transthoracic aspiration biopsy of peripheral nodes/masses with CT, which also permitted us to diagnose nontumoral pathology (Table 2;Fig. 2). Among malignant tumors, we emphasize the series of extrapulmonary metastasis (Table 2). Using this technique, we achieved a global sensitivity of 94%, specificity of 100%, predictive positive value of 100%, predictive negative value of 62.5%, and global accuracy of 95%.
Thus, in 39 patients (67%), we were able to avoid surgical procedures such as mediastinoscopy with diagnostic aims (Table 3), 36 as a result of being the only diagnostic techniques and 3 true-negatives resulting from evolution, This means a savings of 105.300 € for the hospital. No complications were seen.
BNAB is a technique that allows physicians obtain cytohistologic samples of lesions adjacent to the wall of the tracheobronchial tree and the mediastinum. These sites are not accessible using the traditional methods of biopsy and brushing. To reach such sites, which were previously inaccessible to bronchoscopists, BNAB competed with surgical techniques such as mediastinoscopy that were considered the gold standard. BNAB can be performed without general anesthesia, in any hospital, and by any pneumologist trained in the technique. It permits access to all lymph nodes except prevascular ones, aortic or pulmonary, and those located in the anterior mediastinum. These nodes are accessible by transthoracic aspiration (TTA), which can provide upper and lower paratracheal and retrotracheal, subcarinal, and bilateral hilar samples (Fig. 1). Some of those nodes are unreachable by mediastinoscopy, namely, the posterior subcarinals, bilateral hilars, and on occasions, inferior nodes of the aortic arch. 13,14
BNAB also permits a rapid diagnosis in patients with inoperable malignant pathologies who will undergo chemotherapy and/or radiotherapy and in patients with benign pathology who will undergo medical treatment (eg, steroids), thus avoiding unjustified waiting for diagnostic surgical techniques. In our series, this occurred in 29 patients (50%) with malignant pathology and in 7 cases with benign pathology.
No mortality and very low morbidity has been previously reported. However, isolated cases of pneumothorax, 6 pneumomediastinum, 15 hemomediastinum, 16 bacteremia, 17 and pericarditis 18 have been reported. In our study, no cases of mortality or morbidity were seen. Nevertheless, an aspect we need to improve is the low sensitivity in the diagnosis of paratracheal tumors and hilar and mediastinal adenopathies, which has improved with training, adequate processing of the samples, the presence of a cytopathologist during the procedure, 20–25 the histologic needles, 26–29 and use of new technologies. 30–36
De Castro et al. 19 improved sensitivity from 32% to 78% after a period of training, and these authors recommend performing 50 procedures to achieve good results. Haponik et al. 20 increased their performance from 21.4% to 47.6% (P <0.001) after a training program for bronchoscopists and technicians. We began learning the technique by reading books, journal articles, and watching videos, in addition to attending courses and practical workshops and undergoing training with tracheobronchial animal models (pigs and sheep). After performing more than 50 punctures, we have reached 94% sensitivity, 100% specificity, 100% predictive positive value, 62.5% predictive negative value, and 95% global accuracy, which is similar to mediastinoscopy 37,38 but without its complications. 1,39
Some authors advise having a cytopathologist present during the performance of BNA to increase its yield. 20,24 We have observed that an adequate and valid sample can be obtained for diagnosis if, when performing aspiration with the needle completely inside the bronchial wall, we can see small bubbles of air with material not contaminated with blood inside the interior of the translucent catheter that lengthens the needle. If we are using the histologic needle, we can also obtain a core of tissue. To achieve both, we have to select the puncture site, relying first on the CT findings (90%) and also on those of bronchoscopy (57%). Thus, although it would be desirable to have a cytopathologist at hand during selected cases, for example, in patients undergoing repeat puncture procedures, we believe that, in general, it is not necessary in daily clinical practice.
Schenk et al. 26 compared the histologic needle (MV-319) with a cytologic one (MW-222) and demonstrated an 85.5% increase in sensitivity compared with 52.7% in identifying malignant adenopathies (P = 0.0001). We currently are using both types of needle and believe that the histologic needle (MW-319) improves the sensitivity and has certain advantages over the cytologic one (MW-222), particularly when benign pathology is suspected (eg. sarcoidosis, tuberculosis), because, apart from extraction of cells, it also permits the sampling of core tissue for histologic analysis. However, its principal inconvenience is achieving penetration through the bronchial wall to reach the desired areas of the mediastinum. This is easily achieved with the cytologic needle (MW-222), which has a good yield in malignant pathology (eg. primary tumors and metastasis). We therefore recommend to those beginning training in the technique to use a cytologic needle when malignant pathology is suspected and a histologic one when benign pathology is suspected or when there are doubts about the pathology.
Currently, controversy is centered on whether images such as virtual bronchoscopy, ultrasound, or computed tomographic fluoroscopy (CTF) should be used with the aim of improving the sensitivity of BNA. McAdams et al. 35 studied the usefulness of virtual bronchoscopy as a guide for BNA in 17 patients and obtained a global sensitivity of 88%. These investigators suggested that with this method, the bronchoscopist can better correlate the location of the adenopathy and the angle of approximation, which together with the axial images of standard CT, results in the aspiration of smaller and less accessible nodes. The inconvenience of this technique is similar to that of conventional CT. The bronchoscopist should bear in mind the images when performing bronchoscopy, and therefore follow with the procedure even without being certain, in real time, where the final position of the needle is.
Shannon et al. 31 examined the usefulness of endobronchial ultrasound for guiding BNA and found no differences with respect to standard BNA. The sensitivity was good in both procedures, 82.6% and 90.5%, respectively; in addition, these authors suggested that endobronchial ultrasound could have a role in small nodes less than 2 cm in diameter on their smaller axis when a cytopathologist was not present and when standard BNA was negative. The disadvantage of this method is that the catheter with the endobronchial ultrasound should pass through the same channel of the bronchoscope as the needle, and, as a consequence, real-time images are impossible.
Rong and Cui 32 used a CT scan as a guide for BNA in 49 patients with hilar and mediastinal adenopathies and improved their yield from 20% to 60%. The limitation of this method is that 3 to 5 seconds are necessary to process each image of the CT and, therefore, the possibility exists of obtaining artefacts instead of real-time images.
White et al. 33,34 used CTF as a guide in 15 patients with hilar and mediastinal adenopathies. They diagnosed 6 patients, and obtained 2 false-negatives in the surgery, 2 true-negatives resulting from mediastinoscopy, and 2 without pathology in the follow up. Three patients were lost to follow up. The mean duration of the procedure, including moving the patient to the room for CT, was 65 minutes, and the mean duration of fluoroscopy was 3.8 minutes (228 sec). The size of adenopathy was greater than 2 cm in only 1 of the 15 patients. Garpestad et al. 36 studied 32 patients with hilar and mediastinal adenopathies using CTF as guidance for BNA. They obtained an adequate tissue sample in 28 (87.5%) and a specific diagnosis in 22 (68.8%). One patient received a false-negative diagnosis, which was confirmed by mediastinoscopy, 2 were true-negatives confirmed by mediastinoscopy, and malignancy was ruled out in 3 patients during follow up. These authors conclude that CTF should not substitute a “blind BNA” made with a good technique, but that it is a useful instrument to improve the yield in small and less accessible adenopathies. This technique has the advantage of being the only one of those described that uses real-time imaging.
We studied 54 patients with hilar and mediastinal adenopathies using the “standard blind technique” and obtained a valid sample in 52 patients (96%) and a definitive cytohistologic diagnosis in 44 (81%). Three patients were false-negatives, 2 confirmed by thoracotomy and 1 by mediastinoscopy. Two patients were true-negatives confirmed by mediastinoscopy and thoracotomy, respectively, and 3 patients showed no pathology during follow up. We believe that “blind BNA” performed with a good standard technique is as sensitive as that done using real-time imaging guidance such as CTF and has the advantage of being within reach of all pneumologists and all hospitals. Furthermore, BNA can be done in the bronchoscopy room without occupying the CT room. Although in selected cases the new techniques can be useful, it is too premature to recommend them as “new standards” until more longitudinal studies have been performed. 30
One aspect to bear in mind, particularly in public hospitals, is the economic aspect of a new technique, apart from its yield and complications. In our study, the cost of using BNAB to diagnose paratracheal tumors and mediastinal adenopathies, including the cost of disposable needles, was 10 times less than surgical techniques (Table 3). This finding corresponds to data published by other authors. 26,40,41
Our study has certain noteworthy features:
- We studied and diagnosed patients with paratracheal tumors and hilar and mediastinal adenopathies in all lymph node chains accessible with the “standard blind technique” of BNA and CT static imaging.
- Samples obtained with a histologic needle were similar in quality and size to those obtained by transbronchial forceps biopsy and/or by transthoracic aspiration biopsy with CT (Fig.2A, B).
- BNAB was the only diagnostic technique used in 36 of 58 patients (62%), and in none of those was a visible endobronchial lesion present.
- No cases of morbidity or mortality were recorded.
- In 39 patients (67%), surgical techniques with diagnostic aims were avoided (mediastinoscopy, thoracotomy), saving 105.300 € for the hospital. Therefore, if BNAB has so many advantages and so few inconveniences, why does it continue to be “the great unknown” among the majority of the world's pneumologists, as surveys demonstrate. 8,9 It is therefore necessary to identify reasons for the frustration of many pneumologists, who were trained before the availability of this technique, and who at one time attempted to use such but gave up on not being able to achieve the expected results and thus formed a negative opinion of such a technique. The absence of training programs for fellows and fear of complications could lead some to think that BNAB is not safe, when in reality it can be claimed that it is one of the safest biopsy procedures used in the diagnosis of lung diseases. Solutions to these problems include implementing practical training programs in hospitals for both fellows and staff or having at least one specialist trained in BNAB available in each pneumology department.
In summary, BNAB performed without general anesthesia, without real-time imaging, and with a “standard blind technique” is minimally invasive, safe, and economical in both inpatients and outpatients and avoids more invasive and more costly techniques. We therefore believe that it should be included in protocols as the first cytohistologic diagnostic procedure for paratracheal tumors and hilar and mediastinal lymph nodes.
1. Varela G, Jiménez MF, López S, et al. Estudio descriptivo de las complicaciones de la mediastinoscopia. Arch Bronconeumol. 1998; 34:119–122.
2. Schieppati E. La punción mediastinal a través de la carina traqueal. Rev Asoc Med Argent. 1949; 663:497–499.
3. Oho K, Kato H, Ogawa I, et al. A new needle for transfiberoptic bronchoscope use [Letter]. Chest. 1979; 76:492.
4. Wang KP, Terry PB, Marsh B. Bronchoscopic needle aspiration biopsy of paratracheal tumors
. Am Rev Respir Dis. 1978; 118:17–21.
5. Wang KP, Terry PB. Transbronchial needle aspiration in the diagnosis and staging of bronchogenic carcinoma. Am Rev Respir Dis. 1983; 127:344–347.
6. Wang KP. Flexible bronchoscopy with transbronchial needle aspiration: biopsy for cytology specimen. In: Wang KP (ed). Biopsy Techniques in Pulmonary Disorders. New York: Raven Press, 1989:63–71.
7. Wang KP, Brower R, Haponik EF, et al. Flexible transbronchial needle aspiration for staging of bronchogenic carcinoma. Chest. 1983; 84:571–576.
8. Puente-Maestu L, Ruiz de Oña JM. Práctica de la broncoscopia en los hospitales de la Comunidad de Madrid. Rev Patol Respir. 2000; 3:59–65.
9. Colt GH, Prakash UB, Offord KP. Bronchoscopy in North America: survey by the American Association for Bronchology, 1999. Journal of Bronchology. 2000; 7:8–25.
10. Wang KP. Transbronchial needle aspiration. Journal of Bronchology. 1994; 1:63–68.
11. Wang KP. Staging of bronchogenic carcinoma by bronchoscopy. Chest. 1994; 106:588–593.
12. Baker JJ, Solanki PH, Schenk DA, et al. Transbronchial fine needle aspiration of the mediastinum: importance of lymphocytes as an indicator of specimen adequacy. Acta Cytol. 1990; 34:517–523.
13. Lemer J, Malberger E, Köning-Nativ R. Transbronchial fine needle aspiration. Thorax. 1982; 37:270–274.
14. Brynitz S, Strure-Christensen E, Borgeskov S, et al. Transcarinal mediastinal needle biopsy compared with mediastinoscopy. J Thorac Cardiovasc Surg. 1985; 90:21–24.
15. Wang KP, Marsh BR, Summer WR, et al. Transbronchial needle aspiration for diagnosis of lung cancer. Chest. 1981; 80:48–50.
16. Kucera RF, Wolfe GK, Perry ME. Hemomediastinum after transbronchial needle aspiration [Letter]. Chest. 1986; 90:466.
17. Watts WJ, Green RA. Bacteriemia following transbronchial fine needle aspiration [Letter]. Chest. 1984; 85:295.
18. Epstein SK, Winslow CJ, Brecher SM, et al. Polymicrobial bacterial pericarditis after transbronchial needle aspiration. Am Rev Respir Dis. 1992; 146:523–525.
19. De Castro FR, Diaz Lopez F, Serdá GJ. et al. Relevance of training in transbronchial fine-needle aspiration technique. Chest. 1997; 111:103–105.
20. Haponik EF, Cappellari JO, Chin R, et al. Education and experience improve transbronchial needle aspiration performance. Am J Respir Crit Care Med. 1995; 151:1998–2002.
21. Wang KP. Transbronchial needle aspiration for cytology specimens. In: Wang KP, Mehta AC (eds). Flexible Bronchoscopy. Cambridge MA: Blackwell Scientific, 1995:195–205.
22. Rosenthal DL, Wallace JM. Fine needle aspiration of pulmonary lesions via fiberoptic bronchoscopy. Acta Cytol. 1984; 28:204–210.
23. Chan JKC, Kung ITM. Rehydration of air-dried smears with normal saline. Application in fine needle aspiration cytologic examination. Am J Clin Pathol. 1988; 89:30–34.
24. Davenport RD. Rapid on-site evaluation of transbronchial aspirates. Chest. 1990; 98:59–61.
25. Harkin TJ, Wang KP. Bronchoscopic needle aspiration of mediastinal and hilar
lymph nodes. Journal of Broncology. 1997; 4:238–249.
26. Schenk DA, Chambers SL, Derdak S, et al. Comparison of the Wang 19-gauge and 22-gauge needles in the mediastinal staging of lung cancer. Am Rev Respir Dis. 1993; 147:1251–1258.
27. Schenk DA, Strollo PJ, Pickard JS, et al. Utility of the Wang 18-gauge transbronchial histology needle in the staging of bronchogenic carcinoma. Chest. 1989; 96:272–274.
28. Wang KP. Transbronchial needle aspiration to obtain histology specimen. Journal of Bronchology. 1994; 1:116–122.
29. Mehta AC, Kavuru MS, Meeker DP, et al. Transbronchial needle aspiration for histology specimens. Chest. 1989; 96:1228–1232.
30. Wang KP. Continued efforts to improve the sensitivity of transbronchial needle aspiration. Chest. 1998; 114:4–5.
31. Shannon JJ, Bude RO, Orens JB, et al. Endobronchial ultrasound-guided needle aspiration of mediastinal adenopathy. Am J Respir Crit Care Med. 1996; 153:1424–1430.
32. Rong F, Cui B. CT scan directed transbronchial needle aspiration biopsy for mediastinal nodes. Chest. 1998; 114:36–39.
33. White CS, Templeton PA, Hasday JD. CT-assisted transbronchial needle aspiration: usefulness of CT fluoroscopy. AJR Am J Roentgenol. 1997; 169:393–394.
34. White CS, Weiner EA, Patel P, et al. Transbronchial needle aspiration: guidance with CT fluoroscopy. Chest. 2000; 118:1630–1638.
35. McAdams HP, Goodman PC, Kussin P. Virtual bronchoscopy for directing transbronchial needle aspiration of hilar
and mediastinal lymph nodes: a pilot study. AJR Am J Roentgenol. 1998; 170:1361–1364.
36. Garpestad E, Goldberg SN, Herth F, et al. CT fluoroscopy guidance for transbronchial needle aspiration: an experience in 35 patients. Chest. 2001; 119:329–332.
37. Disdier C, Varela J, Sánchez de Cos O, et al. Usefulness of transbronchial punction and mediastinoscopy in mediastinal nodal staging of non-microcytic bronchogenic carcinoma: preliminary study [in Spanish]. Arch Bronconeumol. 1998; 34:237–244.
38. Rami-Porta R, Mateu-Navarro M. Videomediastinoscopy. Journal of Bronchology. 2002; 9:138–144.
39. Kirschener PA. Cervical mediastinoscopy. Chest Surg Clin N Am. 1996; 6:1–20.
40. Fernández Villar JA, Iglesias Río F, Barreiro Barreiro JM, et al. Clinical usefulness and cost-effectiveness of transbronchial needle aspiration for the diagnosis of mediastinal adenopathy [in Spanish]. Rev Clin Esp. 2001; 201:169–173.
41. Jain P, Arroliga A, Metha AC. Cost-effectiveness of transbronchial needle aspiration in the staging of lung cancer [Abstract]. Chest. 1996; 110:24S.
Keywords:© 2003 Lippincott Williams & Wilkins, Inc.
bronchoscopic needle aspiration and biopsy; paratracheal tumors; hilar; mediastinal adenopathies