|Ahead of print publication
Utility of bronchoalveolar lavage, bronchial brushing, and transbronchial needle aspiration in nonneoplastic lung diseases – A single-center experience
Gaurav Tyagi1, Ajay Malik1, Sarvinder Singh2, Sridhar Mangalesh1, Sharmila Dudani1
1 Department of Pathology, Army College of Medical Sciences and Base Hospital, New Delhi, India
2 Department of Respiratory Medicine, Army College of Medical Sciences and Base Hospital, New Delhi, India
|Date of Submission||22-Jun-2020|
|Date of Decision||17-Aug-2020|
|Date of Acceptance||14-Dec-2020|
7/41, Vikram Vihar, 3rd Floor, Lajpat Nagar-4, New Delhi - 110 024
Source of Support: None, Conflict of Interest: None
Context: Accurate diagnosis may be challenging in nonneoplastic lung diseases. Newer techniques such as bronchoalveolar lavage (BAL), bronchial brushing (BB), and transbronchial needle aspiration (TBNA) have been studied mainly in lung cancers. Aim: This study assesses the efficacy of BAL, BB, and TBNA in non-neoplastic lung diseases, against a gold standard of transbronchial lung biopsy (TBLB). Settings and Design: Three-year prospective study in a tertiary care center in North India. Methodology: Adult patients (n = 137) with nonimproving respiratory complaints and clinical and radiological suspicion of having a pulmonary pathology were recruited. Patients of malignancy were excluded. Patients for whom all four samples of BAL, BB, TBNA, and TBLB were unavailable were excluded. Statistical Analysis Used: Sensitivity (SN), specificity (SP), positive predictive value, and negative predictive value for all tests were computed and represented with 95% confidence intervals. TBLB was taken to be the gold standard. Two or more tests taken in combination were also analyzed. Results: We studied the patients of tuberculosis, interstitial pneumonia, acute and chronic nonspecific inflammation, sarcoidosis, fungal infection, interstitial lung disease, eosinophilic pneumonitis, and pulmonary alveolar proteinosis in this study. TBNA had higher SN and SP (59.8% and 96.7%) than BB (57.0% and 93.3%) or BAL (54.2% and 86.7%). When combined, SN increased to 68.2% and SP decreased to 76.7%. Diagnostic yield for interstitial pneumonia and chronic inflammation was relatively high but poor for tuberculosis and sarcoidosis. Conclusions: TBNA had the highest SN and SP in diagnosing the various nonneoplastic lesions. However, BAL and BB are useful and complementary tools in the definitive diagnosis of localized and diffuse pulmonary infections.
Keywords: Biopsy, bronchoalveolar lavage, bronchoscopy, cytology, pulmonary disease
|How to cite this URL:|
Tyagi G, Malik A, Singh S, Mangalesh S, Dudani S. Utility of bronchoalveolar lavage, bronchial brushing, and transbronchial needle aspiration in nonneoplastic lung diseases – A single-center experience. Med J DY Patil Vidyapeeth [Epub ahead of print] [cited 2021 Jun 13]. Available from: https://www.mjdrdypv.org/preprintarticle.asp?id=316419
| Introduction|| |
The pulmonary air passages are continuously exposed to numerous environmental influences ranging from various air-borne micro-organisms, natural allergens, and a wide variety of pollutants to automobile exhaust fumes and tobacco smoke. This makes them vulnerable to a plethora of primary allergic, inflammatory, and immunologic respiratory diseases in addition to a variable degree of secondary involvement in virtually all terminally ill patients. These constitute a major cause of morbidity and mortality in developing countries such as India.
The introduction of the flexible fiber-optic bronchoscope in the 1970s has revolutionized the cytological evaluation of lung lesions as the material could be obtained from the peripheral bronchopulmonary lesions which had been virtually inaccessible for sampling earlier. Techniques such as bronchoalveolar lavage (BAL), bronchial brushings (BB), and transbronchial needle aspiration (TBNA) are being increasingly used to diagnose neoplastic diseases of the lung.
While numerous studies have assessed the efficacy of cytological techniques such as BAL, BB, and TBNA in the diagnosis of lung cancer, there is a significant paucity of literature assessing the same in the diagnosis of nonneoplastic diseases of the lung.,,, Moreover, no study till date has evaluated all three cytological techniques and compared it with histopathology in infectious and nonneoplastic diseases of the lung.
This study aimed to assess the efficacy (sensitivity [SN] and specificity [SP]) of BAL, BB, and TBNA alone and in combination in diagnosing nonneoplastic diseases of the lung and comparing it with a histopathological diagnosis made by transbronchial lung biopsy (TBLB), which was taken as a gold standard.
| Methodology|| |
This was a prospective study conducted jointly by the departments of pathology and respiratory medicine at a 1000-bedded tertiary care teaching hospital in Northern India from January 2013 to December 2015.
All adults above the age of 18 years were included in the study. Patients selected had both clinical and radiological evidence of lung pathology. A total of 226 patients with nonimproving respiratory complaints and suspected of having a pulmonary pathology were recruited.
Patients of the pediatric age group were not included. Patients wherein all four samples, namely BAL, BB, TBNA, and TBLB were not available (n = 13), or with inadequacy of sample (n = 13) were excluded from the study. Patients of malignancy (n = 63) were also excluded. Hence, a total of 137 participants were included in the analysis.
The present study was conducted after obtaining Institutional Ethics Committee approval, No. 132-20102-131-107540, dated June 30, 2015. All samples were taken by the pulmonologist after prior informed consent from all patients. A detailed clinical history, baseline laboratory investigations, and radiological investigations were done for each case.
Bronchoscopy was performed through the transoral approach with Olympus BF-IT 30 bronchoscopes. BAL samples were received in 20–25 ml aliquots in normal saline in sterile vials. Samples were centrifuged at 1500 rpm for 5 min. A minimum of two to three air-dried and alcohol-fixed smears each were prepared from the sediment, and stained with May Grunwald Giemsa/Leishman Giemsa, Hematoxylin and Eosin, or Papanicolaou. One unstained slide was preserved for special staining if required, at a later date.
Bronchial brush (BB) samples were obtained by a 1.3 mm nylon brush from suspected lesion after complete visualization of the trachea and bronchial tree. Both air-dried and wet fixed smears as described earlier were received in the laboratory. Transbronchial needle aspirate (TBNA) samples were obtained by inserting a needle EB 2519–A15 (Endoflex), 15 mm in length and 1.8 mm in diameter, under bronchoscopic guidance transbronchially, to the underlying mass and aspirated slowly. Slides prepared from the aspirate were both air-dried and alcohol fixed. TBLB was obtained by a tooth fenestrated forceps with a working length of 1050 mm. Three to four bits of tissue fixed in 10% formalin were processed and sections 4 um in thickness and varying in length from 0.5 to 0.8 cm were cut and stained with H and E. Cytological diagnosis was made and correlated with histology. Chamberlain criteria were used to assess the adequacy of the samples.
Statistical analysis was performed using the SPSS software version 21.0 for Windows (Armonk, NY, USA: IBM Corp). Categorical variables are expressed as a percentage and continuous variables as mean ± standard deviation or median (inter-quartile range). Normality of continuous variables was assessed using the Shapiro–Wilk test. SN, SP, positive predictive value (PPV), and negative predictive value (NPV) were computed and expressed with 95% confidence interval (CI).
| Results|| |
This study was conducted on a total of 137 patients in the age group of 18–80 years with suspected primary lung pathology who underwent bronchoscopic evaluation. All these patients had BAL, BB, TBNA, and TBLB samples collected for further evaluation. The median age of the participants was 53 ± 29 (interquartile range) years. There was a male preponderance, with 95 males (69.3%) and 42 females (30.6%). Tuberculosis and interstitial pneumonia were the most frequently encountered cases, together comprising 46% of the study population. Diagnostic findings for tuberculosis are shown in [Figure 1]. Less common pathologies included sarcoidosis, eosinophilic pneumonitis, pulmonary alveolar proteinosis, and fungal infection [Figure 2], [Figure 3], [Figure 4] and [Table 1].
|Figure 1: Tuberculosis. (a) Contrast-enhanced computed tomography of the chest coronal section with cavitary lesion on left apical segment diagnosed with pulmonary tuberculosis; (b) Pictomicrograph of BAL with LG stain – Granulomatous lesion (×200); (c) Pictomicrograph of transbronchial needle aspiration with MGG stain – Granulomatous lesion (×400); (d) Pictomicrograph of transbronchial lung biopsy with H and E stain – Granulomatous lesion (×100)|
Click here to view
|Figure 2: Fungal infection. (a) Chest radiograph showing nonhomogeneous opacities in right upper and left lower lobes, with a cavitary lesion in the right upper lobe, suggestive of fungal infection; (b) Photomicrograph of transbronchial lung biopsy with PAS stain - Fungal Hyphae (×100); (c) Photomicrograph of transbronchial lung biopsy with Grocott stain - Fungal hyphae (×100)|
Click here to view
|Figure 3: Interstitial pneumonia. (a) Contrast-enhanced computed tomography of the chest axial section – mediastinal window shows consolidation of the left upper lobe. (b) Photomicrograph of bronchoalveolar lavage with LG stain – Mixed Inflammation; (c) Photomicrograph of transbronchial lung biopsy – Interstitial inflammation H and E (×40)|
Click here to view
|Figure 4: Sarcoidosis. (a) Photomicrograph of transbronchial lung biopsy with H and E, ×40; (b) H and E, ×100|
Click here to view
The results of TBLB were compared with those of BAL, BB, and TBNA to estimate the SN and SP of these tests. The overall SN of BAL was 54.2% (44.3%–63.7%) (95% CI) and SP was 86.7% (68.3%–95.6%). PPV for BAL was 93.5% (83.5%–97.9%) and NPV was 34.6% (24.2%–46.6%). The overall SN for BB was 57.0% (47.0%–66.4%) and SP was 93.3% (76.4%–98.9%). The PPV and NPV were 96.8% (88.0%–99.4%) and 37.8% (27.0%–49.9%), respectively. Finally, for TBNA, the overall SN was 59.8% (49.8-69.0%) and SP was 96.7% (80.9%–99.8%). The PPV and NPV were 98.4% (90.5%–99.9%) and 40.2% (29.0%–52.5%), respectively. The three diagnostic tests are compared graphically in [Figure 5].
|Figure 5: Sensitivity, specificity, positive predictive value, and negative predictive value for bronchoalveolar lavage, bronchial brushing, and transbronchial needle aspiration|
Click here to view
Combined SN, SP, PPV, and NPV were also computed, for BAL and BB, BAL and TBNA, BB and TBNA, and for all three tests taken together. Positivity in any one of the parallel tests was taken to be a test positive, and TBLB remained the gold standard. The combined SN for BAL and BB was 63.6% (53.6%–72.4%) and SP was 80.0% (60.8%–91.5%). PPV and NPV were 91.8% (82.5%–96.7%) and 38.0% (26.4%–51.2%), respectively. For BAL and TBNA, combined SN was 65.4% (55.5%–74.1%) and SP was 83.3% (64.5%–93.6%). PPV and NPV were 93.3% (84.4%–97.5%) and 40.3% (28.3 = 53.5%), respectively. For BB and TBNA, combined SN was 65.4% (55.5%–74.1%) and SP was 90.0% (72.3%–97.3%). PPV and NPV were 95.8% (87.6%–98.9%) and 42.1% (30.1%–55.1%), respectively. Finally, for all three tests BAL, BB, and TBNA taken together, the combined SN was highest, at 68.2% (58.4%–76.7%). Combined SP was 76.7% (57.2%–89.3%). PPV was 91.2% (82.2%–96.1%), and NPV was 40.3% (27.8%–54.1%). The results are summarized in [Figure 6].
|Figure 6: Sensitivity, specificity, positive predictive value, and negative predictive value for combined tests|
Click here to view
For each case confirmed by TBLB, a score of + 1 was assigned when the disease was also detected by one of the tests under study; BAL, BB, or TBNA. The total “test-score” therefore ranged from 0 (zero) to + 3. A score of 0 indicates that a particular disease could not be diagnosed by BAL, BB, or TBNA. A score of 3 indicates that the particular disease was successfully diagnosed by all the three tests. The average test score is reported for patients grouped by disease and indicates overall diagnostic utility of the three tests for each particular disease. The results are summarized in [Table 2].
|Table 2: Mean and median test scores grouped by the disease. Test score represents the number of tests among bronchoalveolar lavage, bronchial brushing, and transbronchial needle aspiration that were positive for a particular disease|
Click here to view
Disease-wise SN was also calculated for tuberculosis (n = 29), interstitial pneumonia (n = 34), and others (n = 44) for BAL, BB, and TBNA. Combined SN values were also computed, for BAL and BB, BAL and TBNA, BB and TBNA, and for all three tests taken together. The results are summarized in [Table 3].
|Table 3: Sensitivity of bronchoalveolar lavage, bronchial brushing, transbronchial needle aspiration, and combined tests for tuberculosis, interstitial pneumonia and others|
Click here to view
| Discussion|| |
With the advent of the fiber-optic bronchoscope, the accessibility and reach of distal airways have increased, providing a means to assist in the diagnosis of hitherto inaccessible peripherally situated lung lesions. This study is among the first to evaluate the utility of BAL, BB, and TBNA in nonneoplastic diseases and compare it with the results on histopathology. Most cases in this study comprised of interstitial pneumonia, tuberculosis, and chronic nonspecific inflammation [Table 1].
The SN of BAL, BB, and TBNA was 54.2%, 57%, and 59.8% and SP was 86.7%, 93.3%, and 96.7%, respectively. PPV and NPV values of BAL, BB, and TBNA were 93.5%, 96.8%, and 98.4%, and 34.6%, 37.8%, and 40.2%, respectively, further proving the superiority of TBNA >BB >BAL. Though data for the utility of BAL, BB, TBNA in exclusively non-neoplastic diseases are not available, these figures mirror the findings of authors who have studied both benign and malignant lung pathology.,
BAL alone was efficacious in picking up both acute and chronic inflammatory lesions, including interstitial pneumonia, as these cells being discohesive, are easily shed. However, it had poor SN for other disorders, such as in the diagnosis of eosinophilic pneumonitis, where the criteria of 30% eosinophils in the differential cell count of BAL could not be met in all cases with this technique. Furthermore, BAL had a SN of just 13.8% in diagnosing tuberculosis. Some authors have however reported a higher SN of BAL in diagnosing tuberculosis., As the visualization of granulomas or a Langhans giant cell is required against an inflammatory background, stringent criteria employed for diagnosis may have possibly led to a low yield. In addition, all our patients of tuberculosis were immunocompetent, which could be another reason for cells being shed inconsistently.
Diagnostic yield of BB has been reported to range from 50% to 87.30%., The SN of BB in our study was lower than that described by Tomar et al., but SP was seen to be higher. BB was a superior technique, especially in the yield of endoscopically visible lesions, which has been corroborated by previous studies. We found BB to have the maximum SN for the diagnosis of interstitial pneumonia [Table 3]. However, it was very ineffective in the diagnosis of tuberculosis. TBNA is a relatively safe and sensitive technique in the diagnosis of lung pathology. In our study, TBNA scored best in terms of accuracy and had a SN and SP of 59.8% and 96.7%, which is in agreement with earlier reports [Figure 5]. However, others have reported a lower SN for TBNA, mainly due to the study being undertaken in mediastinal lesions. It has also been documented that SN of TBNA improved when it was done under ultrasound guidance for sarcoid lesions from 48.4% to 74.5%.
Expectedly, the SN and SP increased when any of the two tests were combined. SN was the highest when all three tests were used together; however, a fall in SP was observed on combining tests [Figure 6]. In our study, SN for interstitial pneumonia demonstrated an impressive increase from 70.6% to 88.2% when all three tests were combined. However, in the case of tuberculosis, a modest rise from 13.8% to 27.6% was noted [Table 3] and [Figure 6].
We also assigned a mean and median “test-score” to each pathology to examine the positivity rates for a certain disease with all three diagnostic tests studied [Table 2]. For sarcoidosis, none of the five cases were detected by BAL, BB, or TBNA and were diagnosed only by the gold standard TBLB, giving a mean test-score of zero. Similarly, the cases of fungal infection and pulmonary alveolar proteinosis could not be diagnosed by any of the three tests. Tuberculosis, eosinophilic pneumonitis, and interstitial lung disease had relatively low test-scores, indicating a poorer diagnostic utility of BAL, BB, and TBNA in these disorders.
Acute and chronic inflammation and interstitial pneumonia had higher mean and median test-scores, indicating the relative ease of diagnosis with any of the above techniques used. Inflammatory cells are usually abundant, singly scattered and are retrieved easily with any of the modalities used. However, tuberculosis scored very poorly on these counts and patients were mainly diagnosed by biopsy alone. The proportion of patients afflicted with tuberculosis was 21.1% of the study population, which is higher than that reported by other authors., Although sarcoidosis, fungal infections, and pulmonary alveolar proteinosis had very poor scores as they were only diagnosed on TBLB, their numbers were too sparse to arrive at a definitive conclusion.
| Conclusions|| |
In nonneoplastic/infective lesions, successful treatment depends on the accurate identification of the pathology as well as the offending organisms. Other investigations in routine use such as sputum culture yields are poor and run the risk of contamination with oropharyngeal secretions. In our study, TBNA had the highest SN and SP in diagnosing the various nonneoplastic lesions. However, BAL and BB are useful and complementary tools that emerge as a good option in the definitive diagnosis of localized and diffuse pulmonary infections.
This study was a preliminary analysis of nonneoplastic lung diseases and the utility of BAL, BB and TBNA in their diagnosis. Studies giving special consideration to each disease, with large samples are necessitated.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
| References|| |
Kim D, Chen Z, Zhou LF, Huang SX. Air pollutants and early origins of respiratory diseases. Chronic Dis Transl Med 2018;4:75-94.
Acharya K V, B U, Shenoy A, Holla R. Utility of various bronchoscopic modalities in lung cancer diagnosis Asian Pac J Cancer Prev 2017;18:1931-6.
Garg S, Handa U, Mohan H, Janmeja AK. Comparative analysis of various cytohistological techniques in diagnosis of lung diseases. Diagn Cytopathol 2007;35:26-31.
Shukla S, Malhotra KP, Husain N, Gupta A, Anand N. The utility of cytology in the diagnosis of adenocarcinoma lung: A tertiary care center study. J Cytol 2015;32:159-64.
] [Full text]
Liu C, Wen Z, Li Y, Peng L. Application of thin prep bronchial brushing cytology in the early diagnosis of lung cancer: A retrospective study. PLoS One 2014;9:e90163.
Dong Z, Li H, Jiang H, Wu C. Evaluation of cytology in lung cancer diagnosis based on EBUS-TBNA. J Cytol 2017;34:73-7.
] [Full text]
Bezel P, Tischler V, Robinson C, Baumueller S, Bode-Lesniewska B, Kohler M, et al.
Diagnostic value of bronchoalveolar lavage for diagnosis of suspected peripheral lung cancer. Clin Lung Cancer 2016;17:e151-6.
Chamberlain DW, Braude AC, Rebuck AS. A critical evaluation of bronchoalveolar lavage. Criteria for identifying unsatisfactory specimens. Acta Cytol 1987;31:599-605.
Tomar V, Vijay N, Nuwal P, Dixit R. Comparative study of bronchoalveolar lavage, bronchial brushings and FNAC in diagnosing malignant neoplasms of lungs. J Cytol 2016;33:210-3.
] [Full text]
Akuthota P, Weller PF. Eosinophilic pneumonias. Clin Microbiol Rev 2012;25:649-60.
Ojha PR, Madan R, Bharadwaj R. Correlation between sputum and bronchoscopy guided cytology (bronchoalveolar lavage fluid, transbronchial needle aspiration and bronchial brush) with bronchial biopsy in the diagnosis of pulmonary pathology. Arch Med Health Sci 2019;7:25-32. [Full text]
Sing A, Freudenberg N, Kortsik C, Wertzel H, Klosa B, Hasse J, et al
. Comparison of the sensitivity of sputum and brush cytology in the diagnosis of lung carcinomas. Acta Cytol 1997;41:399-408.
Gaur DS, Thapliyal NC, Kishore S, Pathak VP. Efficacy of bronchoalveolar lavage and bronchial brush cytology in diagnosing lung cancers. J Cytol 2007;24:73-7. [Full text]
Vijay N, Vijeta T, Nuwal P, Dixit R. Comparative study of bronchoalveolar lavage, bronchial brushing and FNAC in diagnosing malignant neoplasms of lungs. J Cytol 2016;33:210-3.
] [Full text]
Sharafkhaneh A, Baaklini W, Gorin AB, Green L. Yield of transbronchial needle aspiration in diagnosis of mediastinal lesions. Chest 2003;124:2131-5.
Gupta D, Dadhwal DS, Agarwal R, Gupta N, Bal A, Aggarwal AN. Endobronchial ultrasound guided transbronchial needle aspiration vs. convential transbronchial needle aspiration in the diagnosis of Sarcoidosis. Chest 2014;146:547-56.
Ghildiyal S, Acharya S, Thakur B, Rawat J, Kumar R. Cytopathology of pulmonary lesions: A tertiary care centre experience. J Cytol 2018;35:212-6.
] [Full text]
Biswas P, Datta A, De A, Sinha LK. Pulmonary mass lesions: CT scan diagnostic impression and FNAC diagnosis – A correlative study. Int J Med Res Rev 2016;4:1052-62.
[Figure 1], [Figure 2], [Figure 3], [Figure 4], [Figure 5], [Figure 6]
[Table 1], [Table 2], [Table 3]