Asian Cardiovasc Thorac Ann 2000;8:253-255
© 2000 Asia Publishing EXchange Pte Ltd
Role of Mediastinoscopy in Intrathoracic Tuberculous Lymphadenitis
Adnan Sayar, MD,
Aysun Ölçmen, MD,
Muzaffer Metin, MD,
Hakan Güleç, MD,
Adalet Demir, MD,
Müfid Ölçmen, MD
Second Department of Surgery
Yedikule Chest Surgery Centre
Istanbul, Turkey
|
For reprint information contact: Adnan Sayar, MD Tel: 90 542 694 1562 Fax: 90 216 369 4708 email: sayaradnan{at}hotmail.com Ömerpa a Caddesi, Kemal Salih Sel Sokak 3/12, Erenköy, Istanbul 81060, Turkey.
|
 |
Abstract
|
|---|
From 1993 to 1998, 19 patients with mediastinal tuberculosis underwent mediastinoscopy without any demonstrable parenchymal lesion and with negative diagnostic bronchoscopy. The mean age was 34.4 years (range, 15 to 67 years) and 10 were male. The most common symptom was cough in 12 patients, 4 were asymptomatic. Computed tomography showed involvement of the right paratracheal nodal station in 8 cases. Multiple biopsies of 3 to 5 mediastinal nodal stations diagnosed tuberculous lymphadenitis in 16 patients; in 5 of these, nonspecific inflammatory nodes were also sampled. In 3 patients who had biopsies of 1 mediastinal nodal station, the diagnosis could not be established. It was concluded that when used effectively, mediastinoscopy was acceptable as a final diagnostic step in patients with mediastinal tuberculous lymphadenitis.
|
Introduction
|
|---|
Tuberculous mediastinal and hilar lymphadenitis is a manifestation of primary pulmonary tuberculosis in child-hood and it is infrequently seen in adults. In the absence of characteristic parenchymal and intrabronchial lesions, and with the low diagnostic yield of routine laboratory methods such as sputum examination and the tuberculin skin test, invasive diagnostic procedures gain importance. Previous studies examined the role of computed tomo-graphy, bronchoscopy, transthoracic needle aspiration, and mediastinoscopy in the diagnosis of mediastinal tuberculosis.
1
–
4
This study reviewed cases of intrathoracic tuberculous lymphadenitis without a parenchymal lesion where mediastinoscopy was undertaken.
|
Patients and Methods
|
|---|
Nineteen patients who underwent mediastinoscopy and were diagnosed with mediastinal tuberculous lymphade-nitis in a period of 5 years between 1993 and 1998, were reviewed retrospectively. These represented 0.18% of 10,627 tuberculous patients admitted to our hospital in the same period. Excluded from the study were patients with previous intrathoracic tuberculosis (14), peripheral tuberculous adenitis (8), or mediastinal tuberculosis detected coincidentally during mediastinoscopy for preoperative staging of lung cancer or during thoracotomy or median sternotomy for surgical treatment of benign or malign diseases (7). The mean age was 34.4 years (range, 15 to 67 years) and 10 were male. All patients underwent chest radiography, computed tomography, routine laboratory investigations, smear and culture of sputum for acid-fast bacilli, and fiberoptic bronchoscopy. Tu-berculin skin tests were performed in 13 patients and computed tomography-guided fine-needle aspiration in 6. All underwent lavage, transbronchial needle aspiration (2 cases were also biopsied with a Wang needle), mucosal biopsy, and histopathologic and microbiologic examina-tion of postbronchoscopic sputum. Positive culture for acid-fast bacilli or caseating granuloma in histopathologic examination was accepted as a definite diagnosis. The 19 patients who remained undiagnosed by these methods underwent anterior cervical mediastinoscopy under general anesthesia.
|
Results
|
|---|
Fifteen patients were symptomatic on admission, the most common symptom was cough (12/19; 63%). The 4 asymp-tomatic patients were referred because of mediastinal widening found on chest radiographs during routine screening. In all 19 cases, there was mediastinal widening on chest radiography and computed tomography. Pre-dominantly involved nodal stations were the right para-tracheal and hilar in 8 patients, the left in 3, anterior or subcarinal in 2, and more than 1 mediastinal nodal station in 6. There was positivity between 10 mm and 26 mm in patients who underwent a tuberculin skin test. At least 2 sputum smears for acid-fast bacilli in all cases and culture in 11 were negative. Fiberoptic bronchoscopy revealed tracheal or bronchial narrowing by external compression or carinal widening in 14 cases, mucosal hyperemia and edema in 9, and mucosal swelling in 1. Bronchoscopic findings were totally normal in 3 cases. Samples taken during bronchoscopy were not helpful in establishing a diagnosis. In 1 of 2 patients who underwent Wang needle aspiration, histopathologic examination revealed a group of lymphocytes with suspicion of lymphoma.
Due to the negative results of these less invasive methods, all patients underwent cervical mediastinoscopy. In 16 patients (group 1), between 3 and 5 mediastinal nodal stations were explored and all abnormal lymph nodes were sampled several times (
Table
1
). The other 3 patients (group 2) underwent biopsies of 1 lymph node in 1 nodal station (1 punch biopsy in 2 patients and 2 in the third). The stations sampled in this group were the left inferior paratracheal (2) and subcarinal (1). A definite diagnosis was made in all cases in group 1 by demonstration of caseating granuloma in specimens of intrathoracic lymph nodes. In group 2, the results of mediastinoscopy were negative and 2 of the 3 patients underwent a right anterior mediastinotomy to establish the diagnosis; the 3rd patient who had a family history of tuberculosis, refused another surgical procedure and responded to antituberculous treatment after 6 months. The most frequently sampled lymph nodes were those in the right paratracheal area (
Table
2
). In 5 of the 16 patients in group 1, besides lymph nodes with caseating granuloma, those with nonspecific hyperplasia were also sampled. Patients were referred to their chest physicians after 3 to 6 days. One patient had a minor pneumothorax that did not require treatment. There was no mortality.
|
Discussion
|
|---|
Tuberculous lymphadenitis is one of the most common manifestation of extrapulmonary tuberculosis. The cervical lymph nodes are most frequently involved.
5
Intrathoracic lymph nodes, the main drainage nodes after primary infection, were involved in only 5% of all reported cases of tuberculous lymphadenitis.
6
Since Silver and Steel
7
described mediastinal tuberculosis in adults in 1961, there have been reports of similar cases from many parts of the world. The incidence of mediastinal tuberculosis among patients with intrathoracic tuberculosis varies from 0.25% to 26.1%.
1
,
2
,
8
–
10
Due to the selection of patients, the incidence in this study (0.18%) was lower than previously reported. We included only patients without a parenchymal lesion and in whom bronchoscopy and transthoracic needle aspiration did not establish a diagnosis.
In parenchymal and intrabronchial disease, noninvasive methods have significant diagnostic value. Recent reports stated that mediastinal tuberculous lymphadenitis in the absence of a demonstrable parenchymal lesion is a diagnostic challenge for chest physicians.
2
,
8
Chang and colleagues
8
investigated the clinical role of bronchoscopy in 25 patients with intrathoracic tuberculous lymphadenitis, 16 (64%) were diagnosed by examination of sputum (smear or culture) or by peripheral lymph node biopsy. While the diagnostic value of bronchoscopy was 75%, the need for bronchoscopy to establish the diagnosis was 24% (6/25) and there were still undiagnosed patients (12%) who required invasive surgical procedures. Baran and colleagues
2
did not include patients with parenchymal lesions and reported that bronchoscopy has a significant role in the diagnosis of mediastinal tuberculosis with positive results in 9/17 patients (53%); due to selection criteria, the need for invasive surgical procedures was high (35%).
Previous studies have examined the roles of different methods including mediastinoscopy, in the diagnosis of tuberculous lymphadenitis.
1
–
4
It has been stated that mediastinoscopy is a safe and effective procedure but it should be withheld until noninvasive methods have failed to establish the diagnosis.
1
,
8
,
11
,
12
In such cases, mediastinoscopy yielded a definitive diagnosis in 43% to 100%.
1
–
3
,
8
,
13
However, factors leading to negative results of mediastinoscopy were not investigated.
In contrast to reports showing a substantial number of undiagnosed cases after mediastinoscopy, this study demonstrated that mediastinoscopy was acceptable as the final diagnostic tool when adequately applied in patients without parenchymal lesions and with negative bronchos-copy results.
1
,
3
In such patients, the diagnostic yield of noninvasive laboratory methods and bronchoscopy is low as seen in this series and others.
2
,
14
A diagnosis could be established in 100% of patients in group 1 where exploration of 3 or more mediastinal nodal stations with multiple biopsies were performed. In group 2, mediastinal tuberculosis could not be detected by performing 1 or 2 punch biopsy in 1 nodal station. In group 1, besides the diseased nodes, we also sampled nonspecific inflammatory lymph nodes without tuberculous infection in 31% of patients (5/16). It was apparent that insufficient exploration of the mediastinal nodal stations was responsible for the failure to detect diseased lymph nodes, accounting for the false negative results in group 2, and leading to a more invasive surgical approach in 2 patients. This may have been the cause of negative mediastinoscopy necessitating thoracotomy in cases of mediastinal tuberculous lymphadenitis reported previously.
1
,
3
,
13
It was concluded that in the absence of a demonstrable parenchymal lesion, and when bronchoscopy is negative, mediastinoscopic multiple punch biopsies from 3 or more nodal stations may establish the diagnosis in cases of mediastinal tuberculous lymphadenitis, and prevent unnecessary thoracotomies.
|
References
|
|---|
-
Khan J, Akhtar M, Sinner WN, Bouchama A, Bazarbashi M. CT-guided fine needle aspiration biopsy in the diagnosis of mediastinal tuberculosis. Chest
1994; 106
:1329
–32.[Abstract/Free Full Text]
-
Baran R, Tor M, Tahaog lu K, Özvaran K, Kir A, Kizkin Ö, et al. Intrathoracic tuberculous lymphadenopathy: clinical and bronchoscopic features in 17 adults without parenchymal lesions. Thorax
1996; 51
:87
–9.[Abstract/Free Full Text]
-
Farrow PR, Jones DA, Stanley PJ, Bailey JS, Wales JM, Cookson JB. Thoracic lymphadenopathy in Asians resident in the United Kingdom: role of mediastinoscopy in initial diagnosis. Thorax
1985; 40
:121
–4.[Abstract/Free Full Text]
-
Moon WK, Im JG, Yeon MK, Han CM. Mediastinal tuberculous lymphadenitis: CT findings in active and inactive disease. AJR
1998; 170
:715
–8.[Abstract/Free Full Text]
-
Dandapat MC, Mishra BM, Dash SP, Kar PK. Peripheral lymph node tuberculosis: a review of 80 cases. Br J Surg
1990; 77
:911
–2.[Medline]
-
Powell DA. Tuberculous lymphadenitis. In: Schlossberg D, editor. Tuberculosis. New York: Springer-Verlag, 1994:113–20.
-
Silver CP, Steel SJ. Mediastinal lymphatic gland tu-berculosis in Asian and coloured immigrants. Lancet
1961; 1
:1254
–6.
-
Chang CS, Lee PY, Perng PR. Clinical role of bronchoscopy in adults with intrathoracic tuberculous lymphadenopathy. Chest
1988; 93
:314
–7.[Abstract/Free Full Text]
-
Bloomberg TJ, Dow JC. Contemporary mediastinal tuberculosis. Thorax
1980; 35
:392
–6.[Abstract/Free Full Text]
-
Hadlock FP, Park SK, Awe RJ, Rivera M. Unusual radiographic findings in adult pulmonary tuberculosis. AJR
1980; 134
:1015
–8.[Abstract]
-
Jahangiri M, Taggart DP, Goldstraw P. Role of medi-astinoscopy in superior vena cava obstruction. Cancer
1993; 71
:3006
–8.[Medline]
-
Vallieres E, Pagé A, Verdant A. Ambulatory mediasti-noscopy and anterior mediastinotomy. Ann Thorac Surg
1991; 52
:1122
–6.[Abstract]
-
Trinkle KJ, Bryant LR, Hiller J, Playforth HR. Mediastinoscopy experience with 300 consecutive cases. J Thorac Cardiovasc Surg
1970; 60
:297
–300.[Medline]
-
Cameron EWJ. Tuberculosis and mediastinoscopy. Thorax
1978; 33
:117
–20.[Abstract/Free Full Text]
TOP
Abstract
Introduction
