Egyptian Journal of Chest Diseases and Tuberculosis (2015) 64, 277–285

HO ST E D BY

The Egyptian Society of Chest Diseases and Tuberculosis

Egyptian Journal of Chest Diseases and Tuberculosis

www.elsevier.com/locate/ejcdtwww.sciencedirect.com

ORIGINAL ARTICLE

Anatomical and histopathological airway

abnormalities detected during fiberoptic

bronchoscopy in patients with mediastinal

lymphadenopathy

Abbreviations: CT, computed tomography; FOB, fiberoptic broncho-

scopy; TBNA, transbronchial needle aspiration; BAL, bronchoalveo-

lar lavage; TB, tuberculosis; AFB, autofluorescence bronchoscopy;

WLB, white light bronchoscopy; H&E, haematoxylin and eosin; PAP,

Papanicolaou; SPSS, Statistical Package for the Social Science; EBB,

endobronchial biopsy; FOB, fiberoptic bronchoscopy; NSCLC, non

small cell lung cancer* Corresponding author. Mobile: +20 01116444422.

E-mail address: samohamedoctober@yahoo.com (S. Ahmed).

Peer review under responsibility of The Egyptian Society of Chest

Diseases and Tuberculosis.

http://dx.doi.org/10.1016/j.ejcdt.2014.11.0290422-7638 ª 2014 Production and hosting by Elsevier B.V. on behalf of The Egyptian Society of Chest Diseases and Tuberculosis.

K. Eid a, H. Khatab b, S. Ahmed a,*, G. Saad a

a Chest Department, Faculty of Medicine, Cairo University, Egyptb Pathology Department, Faculty of Medicine, Cairo University, Egypt

Received 16 July 2014; accepted 24 November 2014Available online 12 December 2014

KEYWORDS

Mediastinal

lymphadenopathy;

Fiberoptic bronchoscopy;

Airway abnormalities

Abstract Background: The diagnosis in cases of mediastinal and hilar lymphadenitis without

parenchymal involvement of the lung is often difficult. Mediastinal lymphadenopathy may be

due to a variety of benign or malignant reasons. Hence, it is important to establish a diagnosis

and differentiate benign from malignant lymph nodes.

Aim of the work: To study the usefulness of fiberoptic bronchoscopy in diagnosis of patients with

mediastinal lymphadenopathy.

Patients and methods: The present study included 30 patients with mediastinal lymphadenopa-

thy. All were subjected to written informed consent, full history taking, full clinical examination,

tuberculin skin test, chest X-ray and CT chest. Fiberoptic bronchoscopy including autofluorescence

bronchoscopy with mucosal biopsies, TBNA and bronchial lavage were also obtained.

Results: 46.7% of the study patients were diagnosed as malignancy, 20% diagnosed as sarcoid-

osis, 10% diagnosed as TB, 3.33% diagnosed as reactive lymphadenitis and 20% were undiagnosed.

Observed anatomical airway abnormalities included vocal cord paralysis (16.7%), tracheal com-

pression (3.3%), widening of main or second carina (80%) and mucosal abnormalities (46.7%)

in the form of nodules, infiltration with tumour tissue and unhealthy mucosa. Bronchial mucosal

biopsy was the most useful method of diagnosis (56.7%) followed by TBNA (30%) and finally

BAL (13.3%). AFB has no cost effective value over WLB in detection of malignant lesions.

278 K. Eid et al.

Conclusion: The best diagnostic yield was obtained by combination of bronchial mucosal biopsy

and TBNA techniques.

ª 2014 Production and hosting by Elsevier B.V. on behalf of The Egyptian Society of Chest Diseases and

Tuberculosis.

Introduction

Mediastinal lymphadenopathy is a common clinical conditionencountered by chest physicians. Lymph nodes may beenlarged due to a variety of benign or malignant reasons.

Hence, it is important to establish a diagnosis and differentiatebenign from malignant lymph nodes [1].

The diagnosis in cases of mediastinal and hilar lymphadeni-

tis without parenchymal involvement of the lung is oftendifficult. There are many causes of mediastinal and hilar lym-phadenopathy, including infection, neoplasm, granulomatousdisease and reactive hyperplasia. Chest radiographs and com-

puted tomography (CT) have become the standard techniquefor demonstration of mediastinal and hilar lymphadenopathyand in many cases biopsy is required to establish the specific

diagnosis [2].The mechanisms responsible for the airway involvement are

diverse. The entire spectrum of the airways from the nasal and

oral passages to terminal bronchioles can be affected. Theanatomic abnormalities may include extrinsic compression ofairways by the enlarged lymph nodes, various types of mucosal

involvement, luminal stenosis and airway distortion caused byparenchymal changes [3].

Bronchoscopy is an invaluable diagnostic tool for manylung disorders and a safe procedure with low (0.1–2.5%) mor-

bidity and very low (0.05%) mortality. Conventional trans-bronchial needle aspiration (TBNA) has probably been themost important innovation to the ancillary diagnostic devices

(forceps biopsy, bronchial brushing, bronchoalveolar lavageand bronchial washings), as it created the possibility of sam-pling beyond the wall of the central airways. The quality of

the diagnostic bronchoscopic examination is certainly operatordependent, but the pathological analysis and interpretation ofthe acquired samples is also of eminent importance in the diag-nostic process [4].

The aim of the present study was to assess the value offiberoptic bronchoscopy in diagnosis of mediastinal lymphad-enopathy, with special emphasis on:

� Anatomical airway abnormalities (extra luminal compres-sion, widening of the carina, mucosal nodules, unhealthy

mucosa, etc.).� The best sampling method for pathological diagnosis(mucosal biopsy, TBNA and BAL).

� The role of AFB in detection of mucosal abnormalities inpatients with mediastinal lymphadenopathy.

Subjects

The present study was conducted in the Chest Department in

collaboration with the Pathology Department, Kasr AlainyHospital, Cairo University. The study included 30 patients

with mediastinal lymphadenopathy. Patients who were unfit

for bronchoscopy, e.g., patients with refractory hypoxia notresponding to oxygen therapy, acute hypercapnia or bleedingdisorders were excluded from the study.

Methods

All included individuals were subjected to written informed

consent, detailed history taking, full clinical examination, rou-tine chemical and haematological blood analysis includingcomplete blood count, and coagulation profile, arterial bloodgases, tuberculin skin test, plain chest X-ray (P-A view) and

CT chest with IV contrast. Fiberoptic bronchoscopy (FOB)(PENTAX FB18RX, Ashi Optical, Japan) including autofluo-rescence (SAFE 1000, Ashi Optical, Japan) bronchoscopy with

mucosal biopsies, TBNA and bronchial lavage were also doneto all included patients.

Fiberoptic bronchoscopy

Patients were fasting for at least 6 h and intravenous accesswas insured. 0.5–1 mg atropine sulphate was given intramuscu-

lar, half an hour before the procedure to reduce the bronchialsecretions and suppress vagal overactivity. FOB was per-formed via the nasal route. Lidocaine 2% was given as localanaesthetic agent. The patient’s pulse oximetry, heart rate

and blood pressure were monitored throughout the procedureto ensure adequate oxygenation and hemodynamic stability.

Steps of the procedure

After insertion of FOB and visualization of vocal cords, sys-tematic inspection of the tracheobronchial tree was done, with

identification of sites of lesions, such as: widening of the mainor secondary carina, extra luminal compression, mucosalabnormalities (e.g., nodules, infiltration with tumour tissue

or unhealthy mucosa).Autofluorescence bronchoscopy was done routinely after

white light bronchoscopy (WLB) to identify sites of abnormalmucosa which was apparently normal by WLB. Abnormal

mucosa shows a cold image due to lack of autofluorescence.Neoplastic tissues can be distinguished from surrounding nor-mal tissues by their properties of fluorescence when exposed to

blue light [5].TBNA was performed using a 21-gauge cytology needle

consisting of a 140 cm long flexible catheter and a system of

21-G ‘‘pull-out’’ needles. It was performed before any othersampling technique in order to eliminate the possibility of con-tamination or changing the stage of the tumour in cases of sus-pected malignancy. The needle system was advanced through

the working channel of the bronchoscope with the needlepulled out in the catheter. When the metallic tip of the catheter

Role of fiberoptic bronchoscopy in diagnosis of mediastinal lymphadenopathy 279

appeared on the edge of the bronchoscope, the needle wasadvanced and drilled the tracheal or bronchial wall at the tar-get site. 20 ml syringe was used for applying suction to the

proximal edge thus ensuring the safety of the drill site. Undercontinuous suction, the needle was successively pushed in andpulled out (few millimetres) 4–5 times so that an adequate sam-

ple was obtained. Suction had been discontinued before theneedle was pulled out from the target site. The edge of thebronchoscope was aligned and the system was removed from

the bronchoscope with a gentle manoeuvre. The sampleobtained was spread on clean, dry glass slides and immediatelyfixed by 70% alcohol solution and sent for cytological exami-nation. The procedure was repeated 4–6 times (Fig. 1) [6].

BAL: During standard flexible bronchoscopy, the broncho-scope is placed in a wedge position within the selectedbronchopulmonary segment (basal segment of right lower

lobe, if no abnormal segment detected). Normal saline (atroom temperature) is instilled through the bronchoscope, witha total volume of 100–300 ml and divided into three to five ali-

quots. After the instillation of each aliquot, instilled saline isgenerally retrieved using a negative suction pressure less than100 mmHg. For optimal sampling of distal airspaces, the total

volume retrieved should be P30% of the total instilled volumeto avoid increased risk of tissue disruption and/or inflamma-tory mediator release due to over distension of the lavaged seg-ment. An amount of 50 ml of BAL sample was sent for

cytological examination [7].Mucosal biopsies: Multiple mucosal biopsies were obtained

using crocodile 18 mm biopsy forceps, introduced through the

working channel of the bronchoscope. Site of biopsies

Figure 1 Autofluorescence bronchoscopy.

included, the mucosa overlying the widened ridges, theunhealthy and infiltrated mucosa as seen by WLB or AFBand from the healthy mucosa if no detected abnormal mucosa.

Biopsies were immediately immersed in 10% formalin solutionand sent for histopathological examination.

Histopathological technique: Tissue samples were fixed in

10% neutral buffered formalin for 24 h and then underwentroutine processing and wax embedding. Sections (5 lm thick)were stained with haematoxylin and eosin (H&E). The smears

taken from TBNA and BAL were stained with the Papanico-laou (PAP) stain as well as the H&E stains [8].

Statistical analysis

Data were statistically described in terms of mean ± standarddeviation (±SD), median and range, or frequencies (number

of cases) and percentages when appropriate. Comparison ofnumerical variables between the study groups was done usingKruskal Wallis test. For comparing categorical data, Chisquare (v2) test was performed. Exact test was used instead

when the expected frequency is less than 5. P values less than0.05 were considered statistically significant. All statistical cal-culations were done using computer programs SPSS (Statisti-

cal Package for the Social Science; SPSS Inc., Chicago, IL,USA) version 15 for Microsoft Windows.

Results

The present study included 30 patients who fulfilled theselection criteria and formed the study population. According

to final histopathological results, 14 (46.7%) patients werediagnosed as malignancy, 6 (20%) patients diagnosed assarcoidosis, 3 (10%) patients diagnosed as tuberculous

lymphadenopathy, 3.33% cases diagnosed as reactive lym-phadenopathy and 6 (20%) patients were undiagnosed (Fig. 2).

According to the demographics of the study population,there was a statistically significant difference in the mean age

between malignancy (55 ± 12.66), sarcoidosis (35.33 ± 6.25)and tuberculosis (58 ± 8.19) patients. Regarding sex distribu-tion, all malignant patients were males and all sarcoidosis and

tuberculosis patients were females. The difference in sexdistribution was found to be statistically significant. Regardingspecial habits 13 patients (43.3%) were smokers, 12 of them

were diagnosed as malignancy and 1 patient was diagnosedas reactive lymphadenitis. Also 8 patients (26.7%) were raisingbirds, 4 of them were diagnosed as sarcoidosis, 1 patient was

46.7%

3%

20%

10%

20%

Malignancy

Reac�ve lymphadeni�s"

"

"

"

"

Sarcoidosis

TB

Undiagnosed

Figure 2 Distribution of final diagnosis.

280 K. Eid et al.

diagnosed as tuberculosis and 3 patients undiagnosed. The restof patients did not have any special habits (Table 1).

Anatomical airway abnormalities encountered during FOB

included (Table 2): vocal cord paralysis in 5 patients (16.7%), 4(80%) of them were diagnosed as malignancy and 1 patientwas undiagnosed; tracheal compression in 1 patient (3.3%)

and he was diagnosed as malignancy; abnormal mucosa in14 patients (46.7%), 9 of them were diagnosed as malignancy,2 were diagnosed as sarcoidosis and 2 were diagnosed as TB

(Fig. 3) and 1 was undiagnosed.Regarding the effect of lymph node enlargement on the air-

ways, it was unilateral in 15 patients, 9 of them were malignant(64.3% of malignant cases) and bilateral airway affection was

found in 9 patients, 4 of them were diagnosed as sarcoidosis(66.6% of sarcoidosis cases). The difference between unilateraland bilateral effect was statistically non significant. Also the

effect of lymph node enlargement on the airways, was central(involving trachea and main carina) in 24 patients, 12 of themwere diagnosed as malignant (85.7% of malignant cases) and 6

of them were diagnosed as sarcoidosis (100% of sarcoidosis

Table 2 Anatomical airway abnormalities and effect of L.N enlarg

Final diagnosis

Malignancy

(n= 14)

Sarcoidosis

(n= 6)

Vocal cord paralysis (count, % within

final diagnosis)

4 (28.6%) 0 (0.0%)

Tracheal compression (count, % within

final diagnosis)

1 (7.1%) 0 (0.0%)

Mucosal abnormality (nodules, infiltration and

unhealthy mucosa) (count, % within

final diagnosis)

9 (64.3%) 2 (33.3%)

Effect of enlarged L.N on the airways

Unilateral 9 (64.3%) 1 (16.7%)

Bilateral 2 (14.3%) 4 (66.7%)

Normal airways 3 (21.4%) 1 (16.7%)

Effect of enlarged L.N on the airways

Central (trachea and main carina) 12 (85.7%) 6 (100.0%)

Peripheral (2nd carina) 11 (78.6%) 5 (83.3%)

Table 1 Patient characteristics in relation to final diagnosis.

Final diagnosis

Malignancy

(n = 14)

Sarcoidosis

(n= 6)

Tuberculosi

(n= 3)

Sex distribution (count, % within final diagnosis)

Female 0 (0.0%) 6 (100.0%) 3 (100.0%)

Male 14 (100.0%) 0 (0.0%) 0 (0.0%)

Age (mean ± SD) (years) 55 ± 12.66 35.33 ± 6.25 58 ± 8.19

Special habits (count, % within subgroup)

Smoking 12 (85.7%) 0 (0.0%) 0 (0.0%)

Raising birds 0 (0.0%) 4 (66.9%) 1 (33.3%)

No special habits 2 (14.3%) 2 (33.3%) 2 (66.7%)

* P value is significant at the level of <0.0.

patients), while it was peripheral (involving the second carina,lobar and segmental bronchi) in 24 patients, 11 of them werediagnosed as malignant (78.6% of malignant cases), 5 of them

were diagnosed as sarcoidosis (83.3% of sarcoidosis patients)and 2 patients were diagnosed as TB (66.7% of TB cases)(Table 2).

Regarding the comparison between WLB and AFB, thebronchial mucosa appeared abnormal by WLB in 14 patients(46.7%), while it appeared abnormal by AFB in 17 patients

(56.7%). The difference between WLB and AFB findingswas found to be statistically insignificant (Table 3).

Histopathologically, mucosal biopsy results were positive in17 patients (56.7%) out of the 30 patients forming the study

population, 10 of them were diagnosed as malignancy(71.4% of the 14 malignant cases), 4 were diagnosed as sar-coidosis (66.7% of the 6 sarcoidosis patients) and 3 were diag-

nosed as tuberculosis (100% of the 3 TB patients). Mucosalbiopsy results were found to be statistically significant. TBNAcytology results were positive in 9 patients (30%), 7 of them

were diagnosed as malignancy (50% of malignancy patients),

ement on the airways during FOB in relation to final diagnosis.

Total

(n= 30)

P-Value

Tuberculosis

(n = 3)

Reactive

lymphadenopathy

(n = 1)

Undiagnosed

(n= 6)

0 (0.0%) 0 (0.0%) 1 (16.7%) 5 (16.7%) 0.489

0 (0.0%) 0 (0.0%) 0 (0.0%) 1 (3.3%) 0.881

2 (66.7%) 0 (0.0%) 1 (16.7%) 14 (46.7%) 0.200

1 (33.3%) 0 (0.0%) 4 (66.4%) 15 (50.0%) 0.161

1 (33.3%) 0 (0.0%) 2 (33.3%) 9 (30.0%)

1 (33.3%) 1 (100.0%) 0 (0.0%) 6 (20.0%)

0 (0.0%) 0 (0.0%) 6 (100.0%) 24 (80.0%) –

2 (66.7%) 0 (0.0%) 6 (100.0%) 24 (80.0%)

Total

(n= 30)

P-Value

s Reactive

Lymphadenopathy

(n= 1)

Undiagnosed

(n= 6)

0 (0.0%) 6 (100.0%) 15 (50.0%) 0.00*

1 (100.0%) 0 (0.0%) 15 (50.0%)

50.00 52.7 ± 10.5 50.77 ± 12.92 0.015*

1 (100.0%) 0 (0.0%) 13 (43.3%) 0.001*

0 (0.0%) 3 (50.0%) 8 (26.7%)

0 (0.0%) 3 (50.0%) 9 (30.0%)

Figure 3 Bronchoscopic view of mucosal nodules and biopsy

revealed caseating tuberculous granuloma.

Role of fiberoptic bronchoscopy in diagnosis of mediastinal lymphadenopathy 281

2 were diagnosed as tuberculosis (66.7% of TB cases). BALcytology results were diagnostic in 4 patients (13.3%) only;all of them were diagnosed as malignancy (28.6% of malignant

cases). TBNA cytology results were found to be statisticallysignificant while BAL cytology results were found to be statis-tically insignificant (Table 4).

The total number of finally diagnosed cases was 23 patients(14 malignancy, 6 sarcoidosis and 3 tuberculosis). Mucosalbiopsy was diagnostic in 17 (73.9%) patients. TBNA was diag-nostic in 9 cases (39.1%) and BAL was diagnostic only in 4

Table 4 Relationship between mucosal biopsy, TBNA cytology

population.

Final diagnosis

Malignancy

(n= 14)

Sarcoidosis

(n= 6)

Tuberculosis

(n= 3)

Reactiv

(n = 1)

Mucosal biopsy results (count, % within final diagnosis)

Negative 4 (28.6%) 2 (33.3%) 0 (0.0%) 1 (100.

Positive 10 (71.4%) 4 (66.7%) 3 (100.0%) 0 (0.0%

TBNA cytology results (count, % within final diagnosis)

Negative 7 (50.0%) 6 (100.0%) 1 (33.3%) 1 (100.

Positive 7 (50.0%) 0 (0.0%) 2 (66.7%) 0 (0.0%

BAL cytology results (count, % within final diagnosis)

Negative 10 (71.4%) 6 (100.0%) 3 (100.0%) 1 (100.

Positive 4 (28.6%) 0 (0.0%) 0 (0.0%) 0 (0.0%

* P value is significant at the level of <0.05.

Table 3 Statistical comparison between WLB and AFB findings in

Final diagnosis

Malignancy

(n= 14)

Sarcoidosis

(n = 6)

T

(n

Abnormal mucosa by WLB

(count, % within final diagnosis)

9 (64.3%) 2 (33.3%) 2

Abnormal mucosa by AFB

(count, % within final diagnosis)

11 (78.6%) 2 (33.3%) 2

P-Value = 0.438 (non significant).

(17.4%) cases. The combination of mucosal biopsy and TBNAincreased the percentage of malignancy from 71.4% to 78.6%of all diagnosed malignant cases (although statistical difference

between mucosal biopsy alone and the combination betweenmucosal biopsy and TBNA was insignificant). The combina-tion of mucosal biopsy and BAL does not increase the diag-

nostic value (no statistically significant difference betweenmucosal biopsy alone and this combination). The combinationof TBNA and BAL does not add any value to diagnosis by

TBNA alone. Also the diagnostic value of the combinationof mucosal biopsy, TBNA and BAL was equal to the diagnos-tic value of the combination of mucosal biopsy and TBNA(Table 5 and Fig 4).

The difference between mucosal biopsy results and thecytology results of either TBNA or BAL alone or in combina-tion was found to be statistically significant (Table 6).

Discussion

There are many causes of mediastinal and hilar lymphadenop-

athy, including infection, neoplasm, granulomatous diseaseand reactive hyperplasia. Infection consists predominantly oftuberculosis and fungal infection. The neoplastic causes

include lymphoma, leukaemia and metastatic carcinoma. Sar-coidosis is also a common cause of intra thoracic lymphade-nopathy [2].

Bilateral symmetrical hilar lymphadenopathy is typical ofsarcoidosis, whereas it is bilateral but asymmetrical in

, BAL cytology results and final diagnosis within the study

Total

(n= 30)

P-Value

e lymphadenopathy Undiagnosed

(n= 6)

0%) 6 (100.0%) 13 (43.3%) 0.012*

) 0 (0.0%) 17 (56.7%)

0%) 6 (100.0%) 21 (70.0%) 0.038*

) 0 (0.0%) 9 (30.0%)

0%) 6 (100.0%) 26 (86.7%) 0.260

) 0 (0.0%) 4 (13.3%)

relation to final diagnosis.

Total

(n= 30)uberculosis

= 3)

Reactive

lymphadenopathy

(n= 1)

Undiagnosed

(n= 6)

(66.7%) 0 (0.0%) 1 (16.7%) 14 (46.7%)

(66.7%) 0 (0.0%) 2 (33.3%) 17 (56.7%)

Table 5 Diagnostic value of mucosal biopsy, TBNA, BAL and their combinations in relation to final diagnosis.

Malignancy

(n = 14)

Sarcoidosis

(n= 6)

Tuberculosis

(n= 6)

Total

(n= 23)

Mucosal biopsy 10 (71.4%) 4 (66.7%) 3 (100%) 17 (73.9%)

TBNA 7 (50%) 0 (0%) 2 (66.7%) 9 (39.1%)

BAL 4 (28.6%) 0 (0%) 0 (0%) 4 (17.4%)

Mucosal biopsy + TBNA 11 (78.6%) 4 (66.7%) 3 (100%) 18 (78.3%)

Mucosal biopsy + BAL 10 (71.4%) 4 (66.7%) 3 (100%) 17 (73.9%)

TBNA+ BAL 7 (50%) 0 (0%) 2 (66.7%) 9 (39.1%)

Mucosal biopsy + TBNA+ BAL 11 (78.6%) 4 (66.7%) 3 (100%) 18 (78.3%)

Figure 4 Diagnostic value of the combinations of mucosal

biopsy, TBNA and BAL in relation to final diagnosis.

282 K. Eid et al.

lymphoma. Bilateral hilar lymphadenopathy in elderly couldbe due to small cell carcinoma. Unilateral hilar lymphadenop-

athy could be due to tuberculosis in the young and metastasesfrom bronchogenic carcinoma in older people [9].

In the present study, the mean age of malignant patients was

55 years and this agreed with Bahader and Jazieh [10] whofound that mean age at diagnosis of lung cancer was 64 yearswhile the mean age for sarcoidosis cases was 35 years and this

agreed with Iannuzzi et al. [11] who mentioned that sarcoidosis

Table 6 Statistical comparison between the diagnostic value of m

study.

Diagnostic

Mucosal biopsy 17

TBNA 9

Mucosal biopsy 17

BAL 4

TBNA 9

BAL 4

Mucosal biopsy 17

Mucosal biopsy + TBNA 18

Mucosal biopsy 17

Mucosal biopsy + BAL 17

Mucosal biopsy 17

TBNA+ BAL 9

Mucosal biopsy 17

TBNA 9

BAL 4

* P value is significant at the level of <0.05.

usually develops before the age of 50 years, with the incidencepeaking at 20–39 years. The mean age of TB lymphadenopathypatients was found to be 58 years and this agreed with Khanet al. [12] who found that ages of patients with TB mediastinal

lymphadenopathy ranged from 17 to 60 years.In the present study, malignancy was almost exclusively in

males and this agreed with Bahader and Jazieh [10] who found

that lung cancer affected 75.6% of males and 24.4% of femaleswith male to female ratio of 3:1. Also Rabahi et al. [13] men-tioned that there was a higher prevalence of lung cancer among

males, with a male to female ratio of 1.46:1.00. On the otherhand this study disagreed with Bain et al. [14] who found highincidence of lung cancer among females. In the present study,sarcoidosis was exclusive in females, and this agreed with Ian-

nuzzi et al. [11] who stated that a preponderance of cases ofsarcoidosis in females is consistent across racial and ethnicgroups. Also De Moraes et al. [15] reported that sarcoidosis

is prevalent in women, however Byg et al. [16] found that sar-coidosis overall male to female ratio was 1.06. In the presentstudy, all patients with TB mediastinal lymphadenopathy were

females and this agreed with Chang et al. [17] who reportedthat incidence of intra thoracic TB lymphadenopathy was0.4% in males and 1% in females. But this study disagreed

with Khan et al. [12] who mentioned that the ratio of maleto female in patients with TB mediastinal lymphadenopathywas 1.2.

Regarding special habits, 85% of malignancy cases were

smokers, this agreed with Silvestri et al. [18] who stated that

ucosal biopsy, TBNA, BAL and their combinations within the

Non-diagnostic Total P-Value

13 30 0.0371*

21 30

13 30 0.0004*

26 30

21 30 0.1171

26 30

13 30 0.7934

12 30

13 30 No

13 30

13 30 0.0371*

21 30

13 30 0.0015*

21 30

26 30

Role of fiberoptic bronchoscopy in diagnosis of mediastinal lymphadenopathy 283

cigarette smoking is the main cause of lung cancers and itaccounts for 85–90% of all cases. Also Sahin and Yildiz [19]stated that all types of lung cancer were higher in the group

of smokers. 66.7% of sarcoidosis patients were raising birdsand this agreed with Rossman and Kreider [20] who stated thatexposures associated with sarcoidosis included bird exposure

and De Moraes et al. [15] who mentioned that sarcoidosis usu-ally occurs in individuals exposed to environments in whichother granulomatoses are proven to occur, particularly, bird

raising and occupational exposure.During FOB, 80% of cases with vocal cord paralysis were

found to be malignant. In contrast to this study, Myssiorek[21] stated that mediastinal lymph node metastases represent

4% of the causes of vocal cord paralysis. Also Rabahi et al.[13] found vocal fold paralysis in 3% only of the malignantcases.

In the present study, 14 patients (46.7%) showed mucosalabnormalities, 15 patients (50%) showed unilateral airwayaffection, 9 patients (30%) showed bilateral airway affection,

24 patients (80%) showed central airway affection and 24patients showed peripheral airway affection. Also Rai et al.[2] mentioned that bronchoscopy showed abnormality in

39.6% of patients with mediastinal lymphadenopathy andthese abnormalities included widening of main carina in12.5%, widening of secondary carina in 8.3%, bulge into air-ways because of extrinsic compression in 14.6% and endo-

bronchial nodule in 4.2% of patients.In this study, endobronchial abnormalities in patients with

malignant mediastinal lymph nodes included mucosal abnor-

malities in 9 cases (64.3%), unilateral airway affection in 9 cases(64.3%), central airway affection in 12 patients (85.7%) andperipheral airway affection in 11 patients (78.6%). This study

agreed with Bayram et al. [22] who reported that the most com-mon endobronchial pathology in malignant cases was extrinsiccompression (25.5%) and mucosal abnormalities (43.5%). Also

Rabahi et al. [13] mentioned that cases of malignancy showedmucosal infiltration in 35%, mucosal nodules in 2%, wideningof the main carina in 19%, external compression in 18% andunhealthy mucosa in 87% of cases. Again Sahin and Yildiz

[19] reported that out of 485 patients with malignant mediasti-nal lymph nodes undergoing bronchoscopy, 72.5% had muco-sal abnormalities and 30% had the findings of external

pressure. Verma et al. [23] mentioned that the endoscopicappearance of the visible lesions in patients with malignancywas found in 100% of patients.

In the present study, the observed endobronchial abnor-malities in sarcoidosis patients included; mucosal affection(mainly nodules) in 33.3% of patients, widening of main carinain 100% of patients, widening of second carina in 83.3% of

patients and this agreed with Armstrong et al. [24] who foundmucosal nodules in 64% and bronchostenosis in 26% of sar-coidosis cases. Also Lannes et al. [25] observed enlargement

of main carina in 59%, enlargement of secondary carina in16%, right or left paratracheal bulging in 7%, bulging onthe posterior trachea wall in 4%, bulging of the right or left

main bronchus in 7% and 7% of cases showed no abnormal-ities. Also Matsuoka et al. [26] stated that endobronchial nod-ules were found in 0–33% of sarcoidosis patients. In contrast

to this study Ishii et al. [27] study found no mucosal nodularityor bronchial compression in sarcoidosis patients.

In this study, patients with TB lymphadenopathy showedmucosal affection in 2 patients (66.7%) of TB cases and

peripheral airway affection in 66.7% of cases. This agreed withChang et al. [17] who stated that endobronchial involvementwas found in 75% of patients with TB lymphadenopathy

who underwent bronchoscopic examination. Also Baranet al. [28] found that 82.4% of patients with TB lymphadenop-athy had an endobronchial abnormality, widened carina in

47.4%, bronchial or tracheal displacement by lymph nodeenlargement in 41.2%. Again Ayed and Behbehani [29] foundthat 20% of patients with intra thoracic TB lymphadenopathy

had an endobronchial abnormality.In the present study, the difference between AFB (56.7%)

and WLB (46.7%) findings was insignificant. To the best ofour knowledge no studies were done to investigate the role

of AFB in cases of mediastinal lymphadenopathy, but therewere some studies which were performed on cases of lung can-cer. Lam et al. [30] and Kusunoki et al. [31] found that speci-

ficity of AFB to detect premalignant and malignant lesions wasrelatively low and no significant difference than WLB alone.Hanibuchi et al. [5] stated that the sensitivity of AFB plus

WLB to detect premalignant lesions (95.7%) was significantlyimproved than using WLB alone (59.6%) whereas there wasno significant difference between AFB plus WLB and WLB

alone in detecting non malignant lesions. Lam et al. [30] statedthat the combination of AFB/WLB can increase the diagnosticyield of this invasive procedure.

In the present study, the histopathology of bronchial muco-

sal biopsy was diagnostic in 17 patients; 10 of them were diag-nosed as malignancy (71.4% of the 14 malignant patients), 4were diagnosed as sarcoidosis (66.7% of the 6 sarcoidosis

patients) and 3 were diagnosed as tuberculosis (100% of the3 TB patients). Also Armstrong et al. [24] stated that endo-bronchial biopsy of normal appearing airway was diagnostic

to sarcoidosis in 30–37% and Puar et al. [32] mentioned thatendobronchial biopsy technique may demonstrate non necro-tizing granulomas in cases of sarcoidosis even if the airways

are normal on visual inspection and endobronchial biopsy(EBB) resulted in diagnostic tissue in 50–70% of cases. AgainShorr et al. [33] reported that EBB findings were positive in61.8% of the patients with sarcoidosis and without the use

of EBB many cases would have been missed. In other words,regular employment of EBB during FOB resulted in a 20.6%increase in the value of the procedure. On the other hand, Ishii

et al. [27] mentioned that endobronchial biopsy of normalappearing mucosa does not improve diagnostic capacity incombination with other methods in sarcoidosis.

Chang et al. [17] stated that mucosal biopsy was used tomake a diagnosis of TB in 62.5% of patients with endobron-chial abnormalities and bronchoscopy was used to make adiagnosis of tuberculosis in 50% of cases in which broncho-

scopic findings were negative. Bayram et al. [22] stated thatbronchial forceps biopsy is the most frequently preferred pro-cedure that has diagnostic rate around 76–82% in lung cancer.

Verma et al. [23] reported that histopathological examinationof bronchial biopsies yielded positive diagnosis in 81.6% ofcases having endoscopically visible tumors. Schenk et al. [34]

found that endobronchial biopsy afforded the highest yield(56%) in patients with malignancy.

The present study results disagreed with Ayed and Behbeh-

ani [29] who found that mucosal samples taken from endo-bronchial abnormality at bronchoscopy gave a definitediagnosis in 9% of TB endobronchial abnormality. So theyconcluded that bronchoscopy has a low diagnostic yield in

284 K. Eid et al.

mediastinal tuberculous lymphadenopathy in the absence of aparenchymal lesion.

In this study, the TBNA cytology results were diagnostic in

9 (30%) patients; 7 of them were diagnosed as malignancy(50% of malignancy patients), 2 were diagnosed as tuberculo-sis (66.7% of TB cases) and none of sarcoidosis patients were

diagnosed by TBNA. This agreed with Bilaceroglu et al. [35]who stated that TBNA provided positive results in 83% ofpatients with intra thoracic TB lymphadenopathy, but they

also reported positive results in 80% of patients with nontuberculous adenopathy. Also Cetinkaya et al. [36] found thatdiagnosis was made by TBNA in 65% of tuberculosis patientsand 100% of patients with carcinoma. Again, the present study

agreed with Lannes et al. [25] who stated that sarcoidosis canonly be definitively diagnosed through the detection of noncaseous granuloma and TBNA with a fine needle plays a

limited role in the diagnosis of the disease. Using TBNA con-comitantly with transbronchial biopsy or even bronchialbiopsy improves diagnostic efficiency. Also Bayram et al.

[22] reported that TBNA procedure was positive in 76% ofcases with mediastinal and hilar lymph nodes in lung cancer.Szlubowski et al. [37] found that the TBNA technique was

diagnostic in 67.1% of lung cancer patients with mediastinallymph nodes and Punamiya et al. [38] stated that TBNA ishighly specific for detecting mediastinal lymph node metastasisin patients with NSCLC.

On the other hand, this study disagreed with Bilacerogluet al. [35] who reported a diagnostic yield between 42% and90% for TBNA in patients with stage I and II sarcoidosis

(in hilar and/or mediastinal lymph nodes) and in 14–20% ofpatients, it was the only technique to provide a histologic diag-nosis. Also Cetinkaya et al. [36] found that diagnosis was made

by TBNA in 76% of cases with sarcoidosis. Similarly, Fernan-dez-Villar et al. [39] reported that TBNA alone detected sar-coid granulomas in 61.5% of patients, and it was the only

technique to do so in 30.8% of patients.As regards BAL cytology results, it was observed that BAL

was diagnostic in only a small portion (28.6%) of malignantcases. This study agreed with Gaur et al. [40] who found that

BAL cytology showed 14.3% true positive cases and 57.1%true negative cases, as confirmed by biopsy. Also Leonardet al. [41] found that 36.4% of patients with positive biopsies

had negative BAL, underlining the poor sensitivity of BALin sarcoidosis. Also Tang et al. [42] found that BAL alonerevealed positive malignant cells in 18 of 37 cases (sensitivity

48.6%) and the diagnostic value significantly increased to73.0% with combined BAL and TBLB techniques. In contrastto this study, Zaman et al. [43] found that BAL was positive in88.5% of cases of pulmonary sarcoidosis.

In this study, mucosal biopsy was diagnostic in 71.4% ofmalignant patients, 66.7% of sarcoidosis patients and 100%of TB patients. TBNA was diagnostic in 50% of malignant

patients and 66.7% of TB patients while BAL was diagnosticin 28.6% of malignant patients. The combination of mucosalbiopsy and TBNA increased the percentage of diagnosed

malignancy patients to 78.6% while the combination of muco-sal biopsy and BAL does not increase the diagnostic value.Also the combination of TBNA and BAL does not add any

value to diagnoses by TBNA alone. The combination of muco-sal biopsy, TBNA and BAL was equal to diagnostic value ofthe combination of mucosal biopsy and TBNA.

The present study concluded that bronchial mucosal biopsywas the most important in diagnoses (56.7%) followed byTBNA (30%) and finally BAL (13.3%). The best results

(60%) were obtained by combination of bronchial mucosalbiopsy and TBNA, while BAL had no added value. AFBhas little cost effective value over WLB in cases of mediastinal

lymphadenopathy. It is recommended to perform mucosalbiopsy in all patients with mediastinal lymphadenopathy, evenif the mucosa appears grossly normal. If TBNA is not avail-

able mucosal biopsy alone can yield a diagnosis. TBNA is bestcombined with mucosal biopsy to improve the diagnostic yield.BAL is not recommended to be performed routinely in combi-nation with TBNA and mucosal biopsy in cases with medias-

tinal lymphadenopathy. It is better to perform EBUS guidedTBNA for more accurate lymph node sampling and toimprove the diagnostic yield of TBNA. Finally to validate

these recommendations it is better to perform this study on alarger number of patients and in multi-centres.

Conflict of interest

None declared.

References

[1] A. Kumar, R. Dutta, U. Kannan, R. Kumar, G. Khilnani, S.

Gupta, Evaluation of mediastinal lymph nodes using FDG

PET-CT scan and its histopathologic scan, Ann. Thorac. Med. 6

(2011) 11–16.

[2] S. Rai, D. Bhattacharyya, R. Choudhary, Flexible

bronchoscopy in the evaluation of mediastinal and hilar

lymphadenopathy, MAJAFI 63 (2007) 26–28.

[3] V. Polychronopoulos, U. Prakash, Airway involvement in

sarcoidosis, Chest 136 (5) (2009) 1371–1380.

[4] C. Dooms, L. Seijo, S. Gasparini, R. Trisolini, V. Ninane, K.

Tournoy, Diagnostic bronchoscopy: state of the art, Eur.

Respir. Rev. 19 (117) (2010) 229–236.

[5] M. Hanibuchi, S. Yano, Y. Nishioka, T. Miyoshi, K. Kondo, H.

Uehara, S. Sone, Autofluorescence bronchoscopy, a novel

modality for the early detection of bronchial premalignant and

malignant lesions, J. Med. Invest. 54 (2007) 261–266.

[6] M. Charalambatou, B. Katsiva, G. Liapi, D. Karaindros, M.

Komis, S. Tragaras, A. Resvani, F. Nanou, D. Mihailides, K.

Katis, A comparison of transbronchial needle aspiration of

mediastinal lymph nodes with transbronchial lung biopsy in the

diagnosis of sarcoidosis, PNEUMON 19 (2) (2006) 118–123.

[7] American Thoracic Society, An official American thoracic

society clinical practice guideline: the clinical utility of

bronchoalveolar lavage cellular analysis in interstitial lung

disease, Am. J. Respir. Crit. Care Med. 185 (9) (2012) 1004–

1014.

[8] J. Bancroft, M. Gamble, Theory and Practice of Histological

Techniques, sixth ed., Churchill Livingstone, 2008.

[9] R. Chokhani, S. Gesparini, Aspiration (TBNA) in the early

diagnosis and staging of bronchogenic carcinoma, Indian J.

Chest Dis. Allied Sci. 45 (2003) 111–115.

[10] Y. Bahader, A. Jazieh, Epidemiology of lung cancer, Ann.

Thorac. Med. (2008) 65–67.

[11] M. Iannuzzi, B. Rybicki, A. Teirstein, Sarcoidosis, N. Engl. J.

Med. 357 (2007) 2153–2165.

[12] J. Khan, M. Akhtar, W. Von Sinner, A. Bouchama, M.

Bazarbashi, CT-guided fine needle aspiration biopsy in the

diagnosis of mediastinal tuberculosis, Chest 106 (1994) 1329–

1332.

Role of fiberoptic bronchoscopy in diagnosis of mediastinal lymphadenopathy 285

[13] M. Rabahi, A. Ferreira, B. Reciputti, T. Matos, O. De, S. Pinto,

Fiberoptic bronchoscopy findings in patients diagnosed with

lung cancer, J. Braz. Pneumol. 38 (4) (2012) 445–451.

[14] C. Bain, D. Feskanich, F. Speizer, M. Thun, E. Hertzmark, B.

Rosner, G. Colditz, Lung cancer rates in men and women with

comparable histories of smoking, JNCI 96 (11) (2004) 826–834.

[15] B. De Moraes, L. Neves, O. Do Brasil, J. De Sa Pedroso, J. De

Melo, Laryngeal sarcoidosis: literature review, Int. Arch.

Otorhinolaryngol. 15 (3) (2011) 359–364.

[16] K. Byg, N. Milman, S. Hansen, Sarcoidosis in Denmark 1980–

1994: a registry-based incidence study comprising 5536 patients,

Sarcoidosis Vasc. Diffuse Lung Dis. 20 (2003) 46–52.

[17] S. Chang, P. Lee, R. Perng, Clinical role of bronchoscopy in

adults with intra thoracic tuberculous lymphadenitis, Chest 93

(1988) 314–317.

[18] G. Silvestri, A. Alberg, J. Ravenel, The changing epidemiology

of lung cancer with a focus on screening, BMJ (2009) 339.

[19] F. Sahin, P. Yıldız, Radiological, bronchoscopic and

histopathologic characteristics of patients with primary lung

cancer in Turkey (2006–2009), Asian Pac. J. Cancer Prev. 12

(2011) 1947–1952.

[20] M. Rossman, M. Kreider, State of the art: lesson learned from

ACCESS (A Case Controlled Etiologic Study of Sarcoidosis),

Proc. Am. Thorac. Soc. 4 (5) (2007) 453–456.

[21] D. Myssiorek, Recurrent laryngeal nerve paralysis: anatomy and

etiology, Otolaryngol. Clin. North Am. 37 (1) (2004) 25–44.

[22] N. Bayram, S. Borekci, M. Uyar, K. Bakir, O. Elbek,

Transbronchial needle aspiration in the diagnosis and staging

of lung cancer, Indian J. Chest Dis. Allied Sci. 50 (2008) 273–

276.

[23] S. Verma, A. Srivastava, R. Prasad, Study of fiberoptic

bronchoscopy in endoscopically visible bronchial carcinoma,

Int. Infect. Dis. 1 (2011) 7.

[24] J. Armstrong, J. Radke, P. Kvale, M. Eichenhorn, J. Popovich,

Endoscopic findings in sarcoidosis. Characteristics and

correlations with radiographic staging and bronchial mucosal

biopsy yield, Ann. Otol. Rhinol. Laryngol. 90 (1981) 339–343.

[25] D. Lannes, A. Monteiro, E. Toscano, A. Cavalcanti, M.

Nascimento, P. De Biasi, M. Zamboni, Transbronchial needle

aspiration of hilar and mediastinal lymph nodes, Rev. Port.

Pneumol. 13 (5) (2007) 651–658.

[26] R. Matsuoka, H. Kobayashi, S. Kitamura, Bronchoscopic

findings in sarcoidosis and a study of longitudinal changes in

the bronchial mucosa, J. Jpn. Soc. Bronchol. 9 (1988) 340–345.

[27] H. Ishii, S. Otani, A. Iwata, H. Oka, K. Komiya, D. Yoshioka,

K. Hashinaga, J. Kadota, Limited role of auxiliary

endobronchial biopsy in the diagnosis of Japanese patients

with Sarcoidosis, Tohoku J. Exp. Med. 223 (2011) 119–123.

[28] R. Baran, M. Tor, K. Tahaoglu, K. Ozvaran, A. Kir, O. Kizkin,

H. Tuirker, Intrathoracic tuberculous lymphadenopathy:

clinical and bronchoscopic features in 17 adults without

parenchymal lesions, Thorax 51 (1996) 87–89.

[29] A. Ayed, N. Behbehani, Diagnosis and treatment of isolated

tuberculous mediastinal lymphadenopathy in adults, Eur. J.

Surg. 167 (5) (2001) 334–338.

[30] S. Lam, C. MacAulay, J. Hung, J. Le Riche, A. Profio, B. Palcic,

Detection of dysplasia and carcinoma in situ with a lung

imaging fluorescence endoscope device, J. Thorac. Cardiovasc.

Surg. 105 (1993) 1035–1040.

[31] Y. Kusunoki, F. Imamura, H. Uda, M. Mano, T. Horai, Early

detection of lung cancer with laser induced fluorescence

endoscopy and spectrofluorometry, Chest 118 (2000) 1776–1782.

[32] H. Puar, R. Young Jr., E. Armstrong, Bronchial and

transbronchial lung biopsy without fluoroscopy in sarcoidosis,

Chest 87 (1985) 303–306.

[33] A. Shorr, K. Torrington, O. Hnatiuk, Endobronchial biopsy for

sarcoidosis: a prospective study, Chest 120 (2001) 109–114.

[34] D. Schenk, C. Bryan, J. Bower, D. Myers, Transbronchial

needle aspiration in the diagnosis of bronchogenic carcinoma,

Chest 92 (1) (1987) 83–85.

[35] S. Bilaceroglu, O. Gunel, Nur Eris, U. Cagırıcı, A. Mehta,

Transbronchial needle aspiration in diagnosing intra thoracic

tuberculous lymphadenitis, Chest 126 (2004) 259–267.

[36] E. Cetinkaya, P. Yildiz, S. Altin, V. Yilmaz, Diagnostic value of

transbronchial needle aspiration by Wang 22-gauge cytology

needle in intra thoracic lymphadenopathy, Chest 125 (2004)

527–531.

[37] A. Szlubowski, J. Kuzdzal, J. Soja, Transbronchial needle

aspiration as a diagnostic method in lung cancer and non-

malignant mediastinal adenopathy, Pneumonol. Alergol. Pol. 75

(2007) 5–12.

[38] V. Punamiya, A. Mehta, P. Chhajed, Bronchoscopic needle

aspiration in the diagnosis of mediastinal lymphadenopathy and

staging of lung cancer, J. Cancer Res. Ther. 6 (2) (2010) 134–

141.

[39] A. Fernandez-Villar, M. Botana, V. Leiro, C. Represas, A.

Gonzalez, M. Mosteiro, L. Pineiroa, Clinical utility of

transbronchial needle aspiration of mediastinal lymph nodes in

the diagnosis of Sarcoidosis in stages I and II, Arch.

Bronconeumol. 43 (9) (2007) 495–500.

[40] D. Gaur, N. Thapliyal, S. Kishore, V. Pathak, Efficacy of

broncho-alveolar lavage and bronchial brush cytology in

diagnosing lung cancers, J. Cytol. 24 (2) (2007) 73–77.

[41] C. Leonard, V. Tormey, C. O’Keane, C. Burke, Bronchoscopic

diagnosis of sarcoidosis, Eur. Respir. J. 10 (1997) 2722–2724.

[42] C. Tang, C. Hsiao, H. Chen, C. Wang, H. Lin, C. Yu, H. Kuo,

Value of broncho-alveolar lavage combined with transbronchial

lung biopsy in the diagnosis of peripheral lung cancer, Chang

Gung Med. J. 23 (11) (2000) 695–700.

[43] S. Zaman, A. Elshami, P. Gupta, Bronchoalveolar lavage

cytology in pulmonary Sarcoidosis, Acta Cytol. 39 (6) (1995)

1117–1123.