Is There a Role for Bronchial Biopsies in Patients Undergoing Bronchoscopy for Non-Cystic Fibrosis Bronchiectasis?

Background: Diagnosis of bronchiectasis mandates workup for etiology, and microbiological characterization often being sought via Bronchoscopy. However, whether to perform bronchial biopsies, is unknown. We aimed to assess the current practice and yield of different bronchoscopic procedures in this setting. Methods: Data from an adult cohort undergoing bronchoscopy for bronchiectasis was reviewed, including demographics, etiology, imaging and results of different bronchoscopic procedures performed. Results: 127 subjects were analyzed (mean age 61, 56% female). Inspection of the airways was abnormal in 31%; however, management changed in only one with a foreign body. BAL culture was positive in 44%. Frequent pathogens were Hemophilus Inuenza (20%), pseudomonas aeruginosa (8%) and Staphylococcus aureus (7%). NTM and tuberculosis were found in 6% and 1.5% respectively. BAL cytology was performed in 125 procedures, EBB in 51 (40%) and TBLB in 38 (30%). EBB did not contribute to diagnosis in any of our subjects, while TBB showed positive ndings in only a single case. Pathology and tissue culture had no benet over BAL with respect to microbiological diagnosis. Conclusions: In adult subjects with Non-CF bronchiectasis requiring bronchoscopy, BAL cytology and bronchial biopsies are frequently performed but are of minimal additional benet and should probably be avoided.


Introduction
Bronchiectasis is a relatively common and increasingly recognized disease characterized radiologically by permanent dilatation of the airways and clinically by chronic sputum expectoration, and recurrent chest infections [1]. Permanent airway dilatation causes chronic colonization by pathogenic organisms, resulting in the characteristic clinical syndrome of chronic in ammation, chronic chough and sputum production, with episodes of exacerbations thus resulting in further destruction of the airways and further in ammation (the so-called "viscous cycle" mechanism [2]. Although not infrequently idiopathic, various etiologies could cause bronchiectasis, including congenital or acquired immune de ciencies, genetic disorders such as cystic brosis or primary ciliary dyskinesia, systemic in ammatory diseases, previous or chronic infections (including infection with non-tuberculous mycobacteria, NTM) and local airway obstruction [3]. Current guidelines recommend that patients diagnosed with bronchiectasis should undergo a thorough investigation for a possible underlying disease that should include a full history and physical examination, immune status, autoimmune serology, testing for genetic alterations and sometimes bronchoscopy [5].
The role of bronchoscopy in patients with non-CF bronchiectasis is generally limited. It is usually recommended in patients, in whom bronchiectasis is limited to one lobe, thereby raising the possibility of focal obstruction [5]. In addition, an important issue in these patients is identifying the type of pathogen infecting or colonizing the diseased airways, as this effects the severity of disease, the prognosis of the patient, and has implications on the correct management plan for the patient [4]. For example, isolation of Pseudomonas Aeruginosa frequently portends a poorer prognosis, with more frequent exacerbations and worse quality of life [6]. This pathogen also warrants consideration of a guideline based eradication strategy, and may affect the choice of long-term treatment given (e.g. inhaled versus chronic oral antimicrobials) [4]. Another example is diagnosis of Non-tuberculous Mycobacteria infection in the airways, frequently requiring speci c and prolonged antimicrobial treatment while chronic macrolide monotherapy is contraindicated [7]. Sputum samples should be sent for culture (including speci c mycobacterial cultures) in all patients with bronchiectasis, and bronchoscopy should be performed for patients who cannot expectorate sputum or when sputum cultures are sterile [2,4]. The diagnosis of NTM infection is by a combination of clinical, radiological and microbiological criteria [8]. The microbiological criteria include positive sputum, bronchoalveolar lavage (BAL) or tissue cultures, or identi cation of acidfast bacilli or granulomas in tissue biopsies, although negative histology may not exclude this diagnosis [9,10]. However, whether tissue should be obtained routinely in patients suspected of having NTM infection is not clear and not stated in the guidelines for NTM or bronchiectasis [1,11,12].
In cases of bronchiectasis in whom bronchoscopy is indicated , following inspection of the main airways, BAL is almost uniformly performed, with uid obtained sent for routine bacterial, mycobacterial and fungal cultures, acid fast and fungal stains and for cytological examination. In our experience, many pulmonologists also perform additional routine endobronchial and/or transbronchial biopsies (EBB and TBLB), which may prolong the procedure and increase the risk of bleeding and pneumothorax. Performing biopsies may potentially increase the diagnostic yield of bronchoscopy, speci cally for diagnosing an underlying cause or increasing the yield of identifying NTM or fungal infection as compared to performing inspection and BAL alone.
In this study, we assess the diagnostic yield and added bene t of the different bronchoscopic procedures in the work up and management of subjects with bronchiectasis, speci cally hoping to nd whether obtaining tissue samples in these subjects should indeed be considered routine practice.

Methods
This study was conducted in accordance with the amended Declaration of Helsinki, and was approved by the Hadassah medical center local institutional review board (approval number HMO-0424-18). Given the retrospective and non-interventional design of the study, obtaining subject's informed consent was waivered.
We conducted a retrospective analysis of all subjects who underwent bronchoscopy due to bronchiectasis in our center during the years 2009-2017. Following approval of our local ethical research board, charts were obtained for all subjects, and data was extracted using our electronic chart system. We excluded subjects with a known history of cystic brosis, and subjects under 16 years of age.
The following data was obtained for all subjects: demographics, etiology of bronchiectasis (where known), imaging ndings, spirometry and results of sputum cultures prior to bronchoscopy. We also reviewed the bronchoscopy reports for data regarding the exact modality used and complications during or after the procedures. Finally, we obtained microbiological and pathology reports for the specimens sent during the procedure. Clinically signi cant results were considered in cases in which abnormal ndings resulted in change of management or in understanding the etiology of bronchiectasis, when non-invasive measures failed to do so.
Routine practice in our institution includes CT scan prior to performing bronchoscopy and sputum cultures including speci c cultures for mycobacterial and fungal infections. Bronchoscopy is usually performed in cases of localized (single lobe) disease or in cases where sputum cultures are nondiagnostic. When performing bronchoscopy, thorough inspection of all accessible airways is done along with aspiration of visible secretions in the airways. In addition, BAL is almost uniformly performed. Both bronchial secretions and BAL uid are routinely sent for cytological analysis as well as bacterial, fungal and mycobacterial cultures. Endobronchial biopsies are carried out at the discretion of the operator, usually in cases in which gross airway pathology is observed. Transbronchial biopsies are attempted when there is a high suspicion of NTM pulmonary disease an underlying parenchymal lung disease or the desire to complement BAL with direct tissue cultures.

Results
One hundred and twenty seven subjects underwent bronchoscopy for bronchiectasis during the study period. Fifty-six (44%) were male, mean age 60.8 (range 17-89). Eighteen subjects (14%) underwent the procedure due to a combination of bronchiectasis and hemoptysis.
Sixteen subjects (12%) had a known etiology for bronchiectasis prior to the procedure including primary ciliary dyskinesia, in ammatory bowel disease, collagen vascular disease, granulomatosis with polyangiitis, lymphoma, chronic lymphocytic leukemia, hypogammaglobulinemia following bone marrow transplant, previous tuberculosis and foreign body aspiration.
Twenty-seven subjects (21%) had disease localized to only one lobe, while the rest had multilobar disease.
Spirometry results were available for 25 subjects (20%), with an average FVC of 84% predicted, FEV1 of 76% predicted and FEV1/FVC of 0.73. Sputum cultures were available for 14 subjects (11%) prior to bronchoscopy and were positive in four subjects, with isolation of NTM (two subjects), Aspergillus and Pseudomonas Aeruginosa.
Bronchoscopies were performed in these cases in order to verify infection with these pathogens or to assess for co-infection.
Bronchoscopy data (see Table 1 for full details): Abnormal ndings (other than increased sputum production) were observed upon visual inspection of the bronchial tree in 39 subjects (30%). Of those, the most common ndings were mucosal hyperemia and edema (24 subjects) with bronchial stenosis or narrowing in the affected lobe observed in 12 cases. In one subject, a foreign body was found and removed at the time of bronchoscopy. None of the other ndings observed resulted in change of management.
Clinically signi cant BAL cultures were positive in 56 subjects (44%). PA was isolated in seven subjects (6%) and NTM in six (5%). Mycobacterium Tuberculosis was found in two subjects. Other frequently isolated bacteria were Haemophilus In uenza (20%) and Staphylococcus Aureus (7%). All microbiological isolations were new, with the exception of one subject with PA, which was previously also identi ed in sputum (bronchoscopy in this case was performed in order to rule out NTM co-infection).
Cytological analysis was performed on BAL uid in all but two subjects (125 samples in total), all of which resulted in no clinically signi cant ndings.
Bronchoscopic biopsies were performed in 76 subjects (60%). Endobronchial biopsies were performed in 51 subjects (40%) and transbronchial biopsies in 38 subjects (30%). 13 subjects (10%) underwent both EBB and TBLB. Three subjects also underwent ultrasound guided transbronchial needle aspiration of enlarged mediastinal lymph nodes. EBB yielded no clinically signi cant ndings in any of the subjects in whom it was performed, and TBLB yielded a signi cant nding in only a single subject (an unexpected metastatic breast cancer).
Tissue was sent for culture in 34 subjects (27%), with positive cultures seen in fteen of these (44%).
Tissue cultures added no further information over and above BAL cultures with the exception of one case in which Staph. Aureus was cultured in tissue and not in BAL uid.
Complications of bronchoscopy were reported in nine cases (7%). There were three cases of pneumothorax (7.8% of all TBLB) and two cases of signi cant bleeding, both managed locally. Three subjects developed fever following bronchoscopy and required antimicrobial therapy (only one of whom underwent biopsies). One subject had prolonged hypoxemia.

Discussion
Bronchoscopy is frequently used as a diagnostic tool in patients with bronchiectasis in whom adequate or informative sputum cultures cannot be obtained [4,13]. However, there is little data regarding the role or diagnostic value of bronchial or lung biopsies in these patients.
In this study, we evaluate the common practice regarding bronchoscopy in subjects with bronchiectasis in our institute. We also assess whether the practice of performing lung biopsies or BAL cytology during bronchoscopy in patients with bronchiectasis is indeed indicated, and adds any clinical information for patient management. We found that performance of BAL cytology was almost universal and that EBB and/or TBB were commonly performed (60% of procedures). We however found that there is minimal or no added clinical bene t to BAL cytology or performing histological and/or microbiological evaluation of lung biopsies (either EBB or TBLB) over and above BAL cultures in non-CF bronchiectasis subjects.
These ndings have important implications to clinical practice. Guidelines suggest sending BAL uid microbiological analysis, including speci c mycobacterial and fungal cultures [1,4]. As expected, and in concordance with previous studies [14][15][16], this practice proved to be most bene cial in our cohort with positive cultures obtained in nearly half the subjects, most of whom with negative previous sputum cultures. All the NTM positive cultures were new compared to previous sputum mycobacterial cultures, reinforcing the role of BAL in diagnosis of NTM.
Despite the importance of lung tissue cultures in aiding diagnosis of pulmonary NTM suggested by some [17], sending additional tissue cultures (performed in 27% of procedures in our cohort) did not result in any new clinically signi cant information except in one subject, making this practice seem not worthwhile.
Endo-bronchial biopsies were performed in our cohort in 40% of subjects. This was mostly performed in cases in which inspection of the airways revealed signi cant mucosal edema, bronchial narrowing or obstruction, ndings commonly described in such subjects. Pathology reports is these cases were consistently non-speci c, and showed mostly mucosal edema, basement membrane thickening, chronic and acute in ammation and goblet cell hyperplasia, ndings that should be expected in subjects with chronic airway disease [18,19] but without clinical implications to date. Given these ndings, and although this practice may seem warranted, our data shows that performing EBB in these subjects yields no clinically signi cant results, and does not alter patient management.
TBLB was also performed quite often in our cohort (30% of cases), with the aims of obtaining tissue cultures, pathology specimens for diagnosis of mycobacterial infections (as per the NTM diagnosis guidelines) [8], and perhaps an underlying unexpected diagnosis. In a retrospective study by Ikedo, the practice of performing TBLB in subjects with MAC yielded speci c histology in 37%, speci cally epithelioid cell granuloma and/or acid-fast bacilli, however this did not add to bronchial washing cultures [20]. In our cohort, as with the case of EBB, TBLB did not result in any added bene t concerning diagnosis of NTM or TB, nor unravel any new underlying etiologies for bronchiectasis. One subject in our cohort was diagnosed with metastatic breast cancer following TBLB; however, these were performed regardless of the diagnosis of bronchiectasis.
Finally, the practice of performing lung biopsies comes at a price. We found an 8% risk of pneumothorax in our subjects undergoing TBLB (three cases) and two cases of signi cant bleeding. Although our numbers are small, there may be an increased risk of complications following transbronchial biopsies in patients with bronchiectasis compared to other patient groups as frequency of complications for TBLB is usually reported to be as low as 0-4% of procedures [21].
Limitations of our study include its retrospective design, and the single center data, that may not be applicable to other centers or regions in the world. However, we believe that the ndings reported here are important for the decision making process undertaken with each bronchoscopy performed for patients with bronchiectasis.
Another limitation to our study may be the relatively low prevalence of mycobacterial disease in Israel [22,23]. Hypothetically, biopsies may be of greatest bene t in patients with suspected tuberculosis or NTM. Thus, biopsies may be of value in regions of the world with higher prevalence of mycobacterial disease and this may warrant analysis of similar data from other geographical regions.
In conclusion, we found that performing TBLB, EBB or cytology analysis of BAL uid did not result in any signi cant additional clinical data when performing bronchoscopies in subjects with non-CF bronchiectasis. We recommend these practices be routinely avoided in such procedures, which should include performing bronchial inspection and BAL for microbiology alone.

Declarations
Funding: This research did not receive any speci c grant from funding agencies in the public, commercial, or not-for-pro t sectors.
Con icts of interest/Competing interests: All authors declare no con ict of interest regarding this manuscript.
Ethics approval: This study was conducted in accordance with the amended Declaration of Helsinki, and was approved by the Hadassah medical center local institutional review board (approval number HMO-0424-18).
Consent to participate: Given the retrospective and non-interventional design of the study, obtaining subject's informed consent was waivered.

Consent for publication: N/A
Availability of data and material: The data that support the ndings of this study are available from the corresponding author upon reasonable request.   (20) Pseudomonas Aeroginosa (8) Staph. Aureus (7) NTM (6)