The airway microbiota of stable COPD : Association with exacerbation frequency and the risks associated with bronchoscopic data collection

Background: Acute exacerbations of COPD are an important cause of mortality and morbidity in patients with COPD. It is incompletely understood why some COPD patients experience frequent exacerbations, while others rarely or never exacerbate. Studies have suggested that the microbiome of the lungs is different in patients with different exacerbation frequencies. Most studies use sputum samples prone to contamination from the upper airway. Bronchoscopic sampling could improve the quality of the samples, but is a more invasive approach. Aims: The overall aim of the PhD project is to investigate if the airway microbiota in subjects with stable COPD is associated with exacerbation frequency and to assess the complications and discomforts (including rates and predictors) associated with bronchoscopic data collection in participants with and without COPD. Materials and Methods: For the first paper, we performed a systematic literature search on complications and discomfort of non-therapeutic bronchoscopy in PubMed. Titles and abstracts of retrieved search hits were sorted according to inclusion and exclusion criteria. The second and third paper uses data collected in the Bergen COPD Microbiome Study (MicroCOPD). Individuals with and without COPD underwent bronchoscopy including protected bronchoalveolar lavage (BAL) (in participants with FEV1 >30% of predicted), protected specimen brushes (PSB), small volume lavage, and in 1/3 of bronchoscopies, endobronchial biopsies. In addition to bronchoscopic sampling, participants provided oral wash samples. For each bronchoscopic procedure, there was one negative control sample of the phosphate-buffered-saline used for the microbial samples. Some participants underwent more than one bronchoscopy. Light sedation with alfentanil was offered to participants. Immediate complications, defined as any event requiring an unplanned intervention or early termination of the procedure, were recorded. Participants were interviewed after a week regarding discomfort, respiratory symptoms and fever sensation. Participants with COPD were followed with telephone interviews every three months for one year regarding exacerbations. Microbial samples and negative controls went through laboratory processing including DNA extraction, PCR and sequencing of the 16S rRNA gene. Extensive bioinformatic processing of sequencing data and microbiota analyses were performed using QIIME2 and R. Pre-processing included bioinformatic identification and removal of contaminant sequences. We them compared bacterial taxonomy and alpha and beta diversity in individuals with and without COPD exacerbations in the follow-up. Results: Bronchoscopy is generally a safe procedure with low mortality and few severe complications, but the literature shows a wide range of specific complication rates, and it was not possible to conclude on discomfort or predictors. In MicroCOPD, 239 participants underwent bronchoscopy once, 61 underwent more than one bronchoscopy. Complications occurred in 25.9% of first bronchoscopies. The rate of potentially severe complications was 1.3%. Participants with COPD experienced more dyspnoea than participants without lung disease. Sedation and lower age were associated with less complications. 47.7% reported fever. Discomfort was associated with fever, dread of bronchoscopy, high COPD Assessment Test score, and never-smoking. Complications and fever in a first bronchoscopy were often predictive for complications and fever in a second bronchoscopy. We found no difference in alpha and beta diversity between participants with and without COPD exacerbations, and no ASV or genus was found to be consistently differentially abundant or distributed between the groups. Conclusions: Bronchoscopy is a generally safe procedure, even in research into COPD, but is not free of risk. Bronchoscopy was associated with frequent need for unplanned interventions, discomfort and fever sensation in MicroCOPD. We found no association between the lung microbiota at stable state and exacerbations of COPD