Characteristics of Respiratory Microbiome in COPD—A Literature Review Authors: Iga Ciesielska-Markowska, Katarzyna Mycroft-Rzeszotarska, Piotr Korczyński, Kaja Pulik, Katarzyna Górska Published: 2026 ABSTRACT Highlights What are the main findings? Each region of the respiratory tract has a distinct microbial community, with partial overlap across sites.Patients suffering from COPD display differences in the makeup of the airway microbiome when compared to healthy individuals. What are the implications of the main findings? Reduced microbial diversity has been associated with lower FEV1 and may relate to COPD progression and exacerbation risk.Low microbial biomass, contamination susceptibility, technical variability, and methodological limitations remain major challenges in accurately characterising the lung microbiome.Future respiratory microbiome research is expected to drive precision pulmonary medicine through multi-omics integration, AI-based analytics, and personalised microbiome-targeted therapies. AbstractChronic obstructive pulmonary disease (COPD) is a respiratory disease that progressively impairs airway function. Its aetiology and clinical presentation are very complex, resulting in an unpredictable course of the disease. The most important causes include smoking and environmental pollutants. However, upper airway microbiome dysbiosis has been linked with COPD severity. Through this review, we aim to compare the microbiome of the respiratory tract between its sites, and to see if there are any significant differences in the composition of the microbial flora of patients with COPD when compared to healthy individuals. While preparing this review, the PubMed database was searched using keywords such as bacteriome, COPD, exacerbation, and microbiome. Analysis of the airway microbiome shows that the three most abundant phyla are Firmicutes, Proteobacteria, and Bacteroidetes. The severity of the disease and the selected therapeutic methods influence the ratio of Proteobacteria and Firmicutes. It has been observed that a decrease in microbial diversity resulted in lower values of FEV1 in patients and could be related with COPD’s progress and exacerbation events. While exacerbation cases need quick treatment, COPD’s complex background makes it difficult to find a singular, microbial cause. 1. Introduction Chronic obstructive pulmonary disease (COPD) is a respiratory disease characterised by chronic airway inflammation, causing progressive impairment of the airway function and resulting in airflow limitation [1]. Airway obstruction leads to chronic cough, phlegm production, air trapping, and shortness of breath [1]. The disease is common, affecting approximately 10.6% of the world population [2]. The most common risk factors for COPD development include smoking tobacco products, exposure to environmental pollutants, and previous respiratory infections [1]. These factors also contribute to one of the most dangerous events that patients with COPD suffer through—the exacerbation of the disease. Exacerbations can occur during any stage of airflow limitation and, if not properly handled, can lead to premature death [3]. The disease has complex clinical presentations and courses. Based on clinical presentation, symptoms of this disease, such as chronic cough with phlegm production, occurrence of exacerbations, and presence of emphysema, can be divided into specific phenotypes [4]. Unsurprisingly—just like many pathways in the human organism—factors associated with COPD development and exacerbations may interact with each other, including through changes in the airway microbiome [5]. As it has been documented already, the bacteria inhabiting the respiratory tract differ between the patients, but different stages of the disease can be characterised with changes in the composition of the airway biofilm [6]. These changes can also be propagated by other factors causing COPD, such as smoking, or even by a patient’s individual immune response [5,6]. The aim of this narrative review was to characterise the microbiome composition at different respiratory levels and compare the most common methods of obtaining microbial samples from the airway. 2. Microbiome Sampling The airways form a complex system of different sites of the respiratory tract that harbour spatially distinct microbial communities [7,8,9]. Each region of the respiratory tract can be sampled by different diagnostic means. However, each sampling method is associated with disadvantages and possible side effects. Most methods require specialist materials, properly trained personnel, and are available only in an in-patient setting [5,10]. The non-invasive methods of sample collection, which can be obtained with minimal risk to the patient’s wellbeing, include oral and nasal swabbing, oral washings, and induced or spontaneous sputum collection. The invasive methods include bronchoscopy (during which one can collect BAL, cytology, and protected specimen brushings) or surgery. Below, we present the available methods of sample collection for microbial analysis. 2.1. Non-Invasive Sample Collection Methods 2.1.1. Oropharyngeal Swab Sampling Cotton swabs are usually used to sample the posterior wall of the oropharynx. Sometimes swabbing is also done from other areas in the oral cavity. The oral microbiome varies significantly between different microhabitats and swabs must be taken from the same site to be reliable [11]. To sample oral washings, water or mouthwashes are used; patients are asked to rinse their mouth for 30 s and then expectorate into a sample tube [12]. Saliva samples are obtained by asking patients to accumulate saliva in the mouth and spit it into a sample tube over the course of one to several minutes [12,13]. Nasal swabs are inserted into the nostrils to a depth of 1–1.5 cm and rotated on the nasal lining [14]. When the microbial analysis is postponed, samples should be subsequently frozen at −80 °C, but the time between collection and freezing varies from almost immediately to two hours between the studies [11,12,13,15]. Relative abundances of taxa have been shown to vary significantly between different sampling methods in studies performed on healthy individuals [15]. Oral swabs have lower alpha diversity than saliva samples and oral washings, while the latter two are comparable in terms of alpha and beta diversity [12,13,15]. The nasal microbiome demonstrates lower alpha diversity than the oral microbiome [16]. Oral microbiome shows higher diversity when compared to sputum samples, but is generally comparable in terms of taxa detected [15,17]. 2.1.2. Induced Sputum Induced sputum samples (ISS) are obtained by using nebulised hypertonic saline (NaCl). The procedure of sputum induction consists of hypertonic saline inhalation at room temperature with increasing concentrations (3%, 4%, and 5% solutions) via an ultrasonic nebuliser. Then, the patient is encouraged to cough deeply and expectorate. Obtained sputum samples should be collected in a sterile tube for later analysis. As the procedure of sputum induction is performed mostly in patients with obstructive disorders, it should be preceded by inhalation of 400 μg of salbutamol to prevent further bronchospasm caused by inhalation of saline [18]. It should be noted that sputum induction is an established standard for studying many respiratory diseases, including COPD patients during stable states, as the specimen quality is good in terms of cell viability and reproducibility [19,20]. The main disadvantage of sputum induction is poor availability of specialist utensils, such as ultrasonic nebulisers, and trained staff outside the research or academic centres, thus making it difficult to access by patients in primary care. 2.1.3. Spontaneous Sputum Spontaneous sputum is collected during a productive cough. It is easily accessible and does not require the access to specialistic equipment or trained staff. Spontaneous sputum samples (SSS) have been used in several studies since they are easier to obtain, especially during acute exacerbations of COPD (AECOPD) when the intensity of sputum production increases, and may be preferred by some, as sputum induction might worsen airway obstruction [21]. However, SSS cannot be collected in patients unable to spontaneously expectorate sputum. 2.1.4. Comparison Between ISS and SSS Although both spontaneous and induced sputum undergo a similar path during expectoration, there are some possible explanations for dissimilarities in their microbial composition. It is a known fact that different species cultivate in different parts of the respiratory tract [7,22], possibly due to distinction of ventilation–circulation ratios in the upper and lower parts of the respiratory tract or its structure. SSS may be better at reflecting proximal airways while ISS may resemble distal parts. Therefore, SSS and ISS should not be used interchangeably [8]. Table 1 presents the key differences between ISS and SSS. 2.2. Invasive Sample Collection Methods Bronchoscopy is an endoscopic procedure performed transnasally or transorally in both diagnostic and therapeutic purposes [23]. It requires the use of topical anaesthesia, preferably lidocaine, to reduce the occurrence of cough reflex or stridor due to bronchoscope insertion [24]. Despite its invasive character, it is considered an emerging gold standard for pulmonary microbiome sampling [25]. Samples for microbiome analysis can be obtained during bronchoalveolar lavage (BAL) and cytology or protected specimen brushing. It should be noted that some COPD patients might have a higher risk of complications resulting from bronchoscopy, such as desaturation, bronchoconstriction, or even pneumonia [24]; thus, COPD treatment should be optimised prior to the examination. 2.2.1. Bronchoalveolar Lavage Bronchoalveolar lavage (BAL) is a commonly used method of sampling the lower airways [26]. Although BAL collection is fairly invasive, its major advantage is that it is less prone to contamination compared to induced sputum [27]. Although the lung is colonised by the oropharyngeal microbiome through microaspiration and mucosal dispersion [28,29], the diversity of lung microbiome in BAL was found to be lower than in the upper respiratory tract (URT) [30]. Moreover, it has been found that BAL is enriched in URT and oral microbes to a greater extent in COPD patients compared to controls [29]. The enrichment of URT and oral microbes is probably caused by the impaired mucociliary clearance system in the bronchi of COPD patients [31]. In one study, the alpha diversity decreased with decreasing sample exposure to potential oral and bronchoscope contamination [32], though this was contradicted by another study where BAL samples showed higher alpha diversity compared to sputum samples. However, this study had a small sample size of only 6 patients, which may have affected the results [33]. 2.2.2. Cytology Brushing Cytology brushing samples for microbiome analysis are typically collected from the sixth- to eighth-generation airways using a cytology brush [34]. 2.2.3. Protected Specimen Brushing Protected specimen brushing (PSB) is another method of obtaining samples representing microbiome from the lower respiratory tract [35]. This method requires using double-sheathed, wax-plug-protected specimen brushes, which are inserted through a bronchoscope [36]. It is necessary to remember that this sample collection method is still burdened by possible contamination with foreign bacteria. However, the contamination limits itself to the outside area of the bronchoscope because the apparatus must pass through the mouth and pharynx. If the sampling procedure is performed with a mouthguard and without suction, then the contamination risk is minimal [25,35]. 2.2.4. Lung Explant Tissue Excision Tissue excision, while invasive, remains an important method to consider when evaluating approaches to airway sample collection. This method is fraught with risks of complications and damage to patients’ overall health—for instance, pneumothorax or bleeding [25]. Moreover, this procedure involves specialised equipment and trained staff. Interestingly, depending on the samples’ localisation, the presence of different bacterial communities was observed, which leads to the assumption that local microarchitecture of airways has an impact on microbiome composition [7,37,38]. 2.3. Microbiome Methodology Limitations The lung microbiome is characteristically low in biomass, often containing very few bacterial cells, which renders it highly susceptible to contamination and technical artefacts [39]. In such samples, even trace amounts of contaminant DNA from reagents (e.g., extraction kits, polymerases, and sequencing reagents) can dominate the signal, frequently introducing spurious taxa, such as Ralstonia, Pseudomonas, or Bradyrhizobium [40]. Negative controls (reagent-only and sampling blanks) are essential yet inconsistently employed or reported, leading to inflated diversity or spurious taxa [41]. Batch effects arising from different DNA extraction protocols (bead-beating intensity, lysis buffers, column vs. magnetic bead purification) introduce substantial technical variability in lysis efficiency for Gram-positive vs. Gram-negative bacteria and fungi, confounding biological signals [42]. In addition, 16S rRNA gene amplicon sequencing remains a commonly used and cost-effective approach for profiling pulmonary microbial communities. However, 16S offers limited taxonomic resolution (often genus level at best, rarely species or strain), cannot distinguish viable from non-viable organisms (amplifying extracellular or dead DNA), and suffers from primer bias and copy-number variation. Shotgun metagenomics provide higher resolution, functional potential, and better viability proxies when combined with PMA (propidium monoazide) or RNA-seq, but it is more expensive, requires deeper sequencing for low-biomass samples, and remains sensitive to host DNA contamination [39]. Sequencing depth directly influences observed diversity; insufficient depth under-samples rare taxa, while over-sequencing amplifies noise in contaminated samples. Finally, microbiome data are compositional (relative abundances summing to 1), introducing spurious correlations and making absolute quantification or cross-sample comparisons difficult without spike-ins or absolute abundance methods. These issues collectively demand rigorous bioinformatics (e.g., decontamination tools like decontam or SCRuB), standardised protocols, and cautious interpretation, especially when linking pulmonary dysbiosis to chronic diseases [39,41]. 3. Airway Microbiome 3.1. Microbiome of the Oral Cavity and Upper Airways The oral cavity hosts hundreds of species of bacteria, as well as microeukaryotes, archaea, and viruses, making it the second area in the human body with the richest and most diverse microbiome, preceded only by the gut [43]. The oral cavity contains several niches, such as teeth, mucosa, tongue, and hard palates, each enriched by different bacteria [44,45]. The upper respiratory tract comprises the nasal cavity, nasopharynx, and the part of the larynx located above the vocal cord. The nasal cavity and its lining epithelium are the first-line barrier against pollutants, microbes, and allergens, which are inhaled during inspiration [46]. The bacterial composition of the oral and nasal cavities is subject to external factors. While the oral microbiota is mainly influenced by lifestyle habits, including diet and cigarette smoking [47], the nasal microbiota is mainly exposed to environmental factors, such as pollutants and airborne microorganisms [46]. The nasal microbiome is distinct from the oral microbiome. The nasal cavity bacterial communities are dominated by phyla Actinobacteria, Firmicutes, and Proteobacteria [48]. Since oral bacterial populations act as sources for the lung microbiome, oral dysbiosis has been linked to diseases of the lower respiratory tract, including COPD [44]. The upper respiratory tract microbiome of smokers with COPD differs from that of smokers without COPD. It was found that the abundance of Veillonella in the throat was increased in patients with COPD, whereas the abundance of Streptococcus was decreased compared to smokers without chronic obstructive pulmonary disease [49]. The presence of specific bacteria can be associated with better or worse lung function in COPD. Veillonella abundance in the throat was found to be negatively correlated with FEV1%pred, but the abundance of Streptococcus was found to be positively correlated with FEV1%pred [49]. In a different study, the abundance of Clostridiales was higher among those with low FEV1 (