In recent years, several endoscopic techniques have been introduced to obtain a tissue diagnosis in mediastinal lymph node enlargement. Transbronchial needle aspiration (TBNA) was introduced by Wang and Terry in 1983.1 Esophageal ultrasound-guided fine needle aspiration and real-time endobronchial ultrasound fine needle aspiration have gained popularity, with the first publications dating back to the late 1980s of the previous century. Although these ultrasound techniques have shown high yields of up to 96%,2 these results mainly rely on cytology specimens.2,3 The question arises as to whether relying on cytology alone is safe. In transbronchial or transesophageal puncture techniques, it is unknown to what extent histology has an additional diagnostic value when compared with cytology alone. We conducted a retrospective, double-blinded, randomized study to answer this question.
MATERIALS AND METHODS
All consecutive TBNA procedures in our institution performed between August 2004 and November 2006 were reviewed. All adequate TBNA specimens for which both cytology and histology material was available were included for analysis. TBNA was exclusively performed with histology needles: the dedicated histology needle MW 319 (Mill-Rose Laboratories, Mentor, OH) and, since 2005, the new eXcelon needle (19 gauge; Boston Scientific, Natick, MA). TBNA is routinely executed in all patients with mediastinal lymph nodes with a smallest diameter of ≥1 cm as evaluated by chest computed tomography. The aspiration technique was performed as described before.4 TBNA was considered adequate when the obtained specimen showed malignancy or another specific diagnosis, or when normal lymphoid cells were present. Two pathologists reviewed all specimens in a random and independent way. First all cytology specimens were reviewed and afterward all histology was examined to prevent bias by judging cytology and histology from the same patient successively. The first author (F.H.) combined cytologic and histologic data afterward and related the results to the final diagnosis made in the actual workup of the patient. As this is a retrospective analysis of data obtained during routine medical practice without any additional interventions for research purposes, no specific approval of our Institutional Review Board was required.
A total of 236 TBNA procedures were performed in our hospital from August 2004 to November 2006. In 69 procedures (29%) no adequate TBNA specimens were obtained. In 167 procedures (71%) adequate TBNA material was obtained in cytology alone (110 procedures), in histology specimens only (7 procedures), or in both cytology and histology (50 procedures). In 86% (43 of 50) of these 50 TBNA procedures, both pathologists made the same diagnosis on both specimens. Of these 43 TBNAs, 86% (37 of 43) consisted of small cell lung carcinoma (SCLC) (7 of 43) or non-small cell lung carcinoma (NSCLC; 30 of 43). In the case of NSCLC, cytology defined the correct histotype (squamous cell carcinoma, adenocarcinoma, or NSCLC “not otherwise specified”) in 67% of cases, whereas histology gave the correct diagnosis in 93%. In 14% (6 of 43) normal lymphoid tissue or a granulomatous reaction was found and histology diagnosis was not superior to cytology. However, in 1 of these 6 cases, tuberculosis culture of a histology specimen alone resulted in the bacteriologic diagnosis of a Mycobacterium gordonae infection, whereas the culture of the cytologic specimen remained sterile. Of the remaining 5 patients, 3 had normal lymphoid tissue in their TBNA samples and were diagnosed with lymphadenopathy in the setting of a (nontuberculous) pulmonary infection. The 2 patients with granulomatous reaction were diagnosed with sarcoidosis. In 14% (7 of 50) of all TBNAs, by the judgment of at least 1 pathologist, histology findings altered patient management. Detailed results are given in Table 1. In 4 cases a different type of pulmonary carcinoma was found in histology material compared with cytology alone. In 1 case non-Hodgkin lymphoma was diagnosed and in 2 cases Mycobacterium tuberculosis was found by polymerase chain reaction (PCR) in histology specimens only, whereas PCR on cytologic material was negative.
In the literature on endoscopic fine needle aspiration of mediastinal lymph nodes, the diagnostic yield obtained mainly relies on cytology. In most articles, reported yields are high, but rarely are the histologic and cytologic yields reported separately.5 Needles used in these techniques are mainly developed for cytologic purposes and are usually 22-gauge.3,6
Schenk et al7 conducted a study directly comparing 19-gauge and 22-gauge TBNA needles and showed that the 19-gauge needle has a significantly higher sensitivity compared with the 22-gauge needle. The latter showed false-negative TBNAs in up to 47% of all TBNAs. In addition, sensitivity of the 19-gauge histology results was higher compared with the 19-gauge cytology results (78.2% vs. 63.6%). More recently, Stratakos et al8 showed a 35.3% increase in the diagnostic yield for malignancy for histologic examination of TBNA samples, compared with sole cytologic examination.
In view of these data, obtaining a histologic TBNA specimen increases the diagnostic yield. The question that arises, however, is whether relying exclusively on the cytologic TBNA material for an accurate diagnosis, in the case in which no histology is available, is wise. No specific literature is available to assess this question in TBNA material, although the value of obtaining histology over cytology for diagnostic purposes has been subject to debate, especially when a lymphoma or sarcoma is present.9,10
This study shows that in a significant number of samples, histologic material can be of additional diagnostic value over cytology alone, and not only in patients with lymphoma, as was the case in patient 3. In patients 6 and 7, PCR detection of M. tuberculosis was only positive on the histologic material. In patient 2, both pathologists recognized the presence of another tumor type in histology compared with cytology alone. Discrepant results in the case of a pulmonary malignancy were obtained in patients 1, 4, and 5. In all 3 cases, the final diagnosis was SCLC. Histologic material was incorrectly interpreted by 1 or both pathologists as having a NSCLC component.
There are several limitations to this study that have to be addressed.
First, in clinical practice, TBNA results can be compared with other tissue samples obtained during the bronchoscopy procedure, such as bronchial biopsies, brushes, and washings. In this study, the pathologist was only informed of the origin of the material, being fine needle aspiration from enlarged mediastinal lymph nodes mainly from patients suspected of having lung cancer, without additional information obtained from other diagnostic bronchoscopic procedures in the same patient. Furthermore, immunohistochemical staining of slides was not a part of the study procedure.
Second, TBNA cytology material rarely produces a different diagnosis compared with histology in the same patient, unless a mixed-type tumor is present. One therefore might assume that the cytologic diagnosis could be changed after viewing histology (or the other way around), before a final judgment is given to the clinician. As both pathologists were not informed of the histologic diagnosis when evaluating cytology and vice versa, a biased judgment was prevented.
Third, the inherent heterogeneity of the lung cancer tissue must be taken into account. Stang et al11 performed a large histopathological evaluation study in which lung cancer cases diagnosed earlier were reviewed by several pathologists. Even in adequate histologic samples some interobserver variability is inevitable. This interobserver variability was larger in NSCLC compared with SCLC.
In conclusion, this study shows that the histologic material acquired with conventional TBNA can reveal additional diagnostic information compared with sole cytologic examination in 14% of representative TBNA samples. Relying solely on either cytologic or histologic samples results in cases of discrepant TBNA results seems unjustified and should prompt further investigation to secure an accurate diagnosis.
1. Wang KP, Terry PB. Transbronchial needle aspiration
in the diagnosis and staging of bronchogenic carcinoma. Am Rev Respir Dis. 1983;127:344–347.
2. Herth FJ, Lunn W, Eberhardt R, et al. Transbronchial versus transesophageal ultrasound-guided aspiration of enlarged mediastinal lymph nodes. Am J Respir Crit Care Med. 2005;171:1164–1167.
3. Herth FJ, Ernst A, Eberhardt R, et al. Endobronchial ultrasound-guided transbronchial needle aspiration
of lymph nodes in the radiologically normal mediastinum. Eur Respir J. 2006;28:910–914.
4. Hermens FH, Thunnissen FB, Janssen JP. Diagnostic yield of eXcelon, a new transbronchial histology
aspiration needle in patients with mediastinal lymph node
enlargement. J Bronchol. 2007;14:86–89.
5. Bilaçeroğlu S, Cağiotariotaciota U, Günel O, et al. Comparison of rigid and flexible transbronchial needle aspiration
in the staging of bronchogenic carcinoma. Respiration. 1998;65:441–449.
6. Annema JT, Veselic M, Rabe KF. Analysis of subcarinal lymph nodes in (suspected) non-small-cell lung cancer
after a negative transbronchial needle aspiration
—what's next? Respiration. 2004;71:630–634.
7. 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.
8. Stratakos G, Porfyridis I, Papas V, et al. Exclusive diagnostic contribution of the histology
specimens obtained by 19-gauge transbronchial aspiration needle in suspected malignant intrathoracic lymphadenopathy. Chest. 2008;133:131–136.
9. Chhieng DC, Cangiarella JF, Symmans WF, et al. Fine-needle aspiration cytology
of Hodgkin disease: a study of 89 cases with emphasis on false-negative cases. Cancer. 2001;93:52–59.
10. Oki M, Saka H, Kitagawa C, et al. The size and quality of the histology
specimens obtained with transbronchial needle aspiration
. J Bronchol. 2005;12:71–75.
11. Stang A, Pohlabeln H, Müller KM, et al. Diagnostic agreement in the histopathological evaluation of lung cancer
tissue in a population-based case-control study. Lung Cancer