Skip to main content
Browse Subject Areas

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

The airways microbiome of individuals with asthma treated with high and low doses of inhaled corticosteroids

  • Matthew J. Martin ,

    Contributed equally to this work with: Matthew J. Martin, Nur Masirah M. Zain

    Roles Conceptualization, Formal analysis, Project administration, Writing – original draft, Writing – review & editing

    Affiliation The Asthma Centre, Nottingham Respiratory Research Unit, University of Nottingham, Nottingham City Hospital, Nottingham, United Kingdom

  • Nur Masirah M. Zain ,

    Contributed equally to this work with: Matthew J. Martin, Nur Masirah M. Zain

    Roles Data curation, Formal analysis, Writing – original draft, Writing – review & editing

    Affiliation Faculty of Life Sciences & Medicine, School of Cancer & Pharmaceutical Sciences, Institute of Pharmaceutical Science, King’s College London, London, United Kingdom

  • Glenn Hearson,

    Roles Conceptualization, Project administration, Software

    Affiliation The Asthma Centre, Nottingham Respiratory Research Unit, University of Nottingham, Nottingham City Hospital, Nottingham, United Kingdom

  • Damian W. Rivett,

    Roles Formal analysis, Methodology, Writing – review & editing

    Affiliation Faculty of Life Sciences & Medicine, School of Cancer & Pharmaceutical Sciences, Institute of Pharmaceutical Science, King’s College London, London, United Kingdom

  • Garrit Koller,

    Roles Formal analysis

    Affiliation Faculty of Life Sciences & Medicine, School of Cancer & Pharmaceutical Sciences, Institute of Pharmaceutical Science, King’s College London, London, United Kingdom

  • David J. Wooldridge,

    Roles Data curation

    Affiliation National Infection Service, Public Health England, London, United Kingdom

  • Graham Rose,

    Roles Data curation

    Affiliation National Infection Service, Public Health England, London, United Kingdom

  • Saheer E. Gharbia,

    Roles Data curation

    Affiliation National Infection Service, Public Health England, London, United Kingdom

  • Ben Forbes,

    Roles Formal analysis, Writing – review & editing

    Affiliation Faculty of Life Sciences & Medicine, School of Cancer & Pharmaceutical Sciences, Institute of Pharmaceutical Science, King’s College London, London, United Kingdom

  • Kenneth D. Bruce ,

    Roles Conceptualization, Formal analysis, Supervision, Writing – original draft, Writing – review & editing

    Affiliation Faculty of Life Sciences & Medicine, School of Cancer & Pharmaceutical Sciences, Institute of Pharmaceutical Science, King’s College London, London, United Kingdom

  • Tim W. Harrison

    Roles Conceptualization, Formal analysis, Funding acquisition, Writing – review & editing

    Affiliation The Asthma Centre, Nottingham Respiratory Research Unit, University of Nottingham, Nottingham City Hospital, Nottingham, United Kingdom



Inhaled corticosteroids (ICS) are the mainstay of asthma treatment, but evidence suggests a link between ICS usage and increased rates of respiratory infections. We assessed the composition of the asthmatic airways microbiome in asthma patients taking low and high dose ICS and the stability of the microbiome over a 2 week period.


We prospectively recruited 55 individuals with asthma. Of these, 22 were on low-dose ICS and 33 on high-dose ICS (16 on budesonide, 17 on fluticasone propionate). Sputum from each subject underwent DNA extraction, amplification and 16S rRNA gene sequencing of the bacterial component of the microbiome. 19 subjects returned for further sputum induction after 24 h and 2 weeks.


A total of 5,615,037 sequencing reads revealed 167 bacterial taxa in the asthmatic airway samples, with the most abundant being Streptococcus spp. No significant differences in sputum bacterial load or overall community composition were seen between the low- and high-dose ICS groups. However, Streptococcus spp. showed significantly higher relative abundance in subjects taking low-dose ICS (p = 0.002). Haemophilus parainfluenzae was significantly more abundant in subjects on high-dose fluticasone propionate than those on high-dose budesonide (p = 0.047). There were no statistically significant changes in microbiota composition over a 2-week period.


Whilst no significant differences were observed between the low- and high-dose ICS groups, increased abundance of the potential pathogen H. parainfluenzae was observed in patients taking high-dose fluticasone propionate compared to those taking high-dose budesonide. The microbiota were stable over fourteen days, providing novel evidence of the established community of bacteria in the asthmatic airways.

Clinical trial registration NCT02671773


Asthma is a common cause of morbidity and mortality affecting an estimated 334 million individuals worldwide [1, 2]. In the UK, a stepwise approach to patient management recognises five different stages reflecting severity and corresponding treatment strategies. All stages now include inhaled corticosteroids (ICS) [35] which are also commonly prescribed in chronic obstructive pulmonary disease (COPD) [6]. Although generally regarded as safe, considerable evidence suggests their use is associated with an increased risk of respiratory infection in COPD [6], and there are suggestions this may also be the case in asthma [4, 5] although not all studies are in agreement [7].

The role of bacteria in the pathophysiology of asthma is not fully established or well-defined [8, 9]. Fundamentally however, the key initial step towards a better understanding is characterising the abundance and diversity of bacteria in the asthmatic airways. This is now possible due to techniques that characterise microbiome composition. Previous studies have described bacterial species in the lower respiratory tract of individuals with asthma as well as making tentative associations with disease severity [10, 11]. In compositional terms, the most common genera across many microbiome studies of the asthmatic airways are Prevotella, Veillonella, Haemophilus, Streptococcus and Moraxella [1215]. Some of these genera contain potential pathogens which are common causes of pneumonia, such as Haemophilus and Streptococcus spp. and both of these species have been detected at higher frequencies in poorly controlled asthmatics and subjects with treatment-resistant severe asthma [12, 1416].

Here, we report the culture-independent analysis of bacteria present in sputum samples of 55 adults with asthma. Data were collected on the composition of the bacteria forming the microbiome and the total bacterial and taxa-specific loads in the samples were quantified. Our aims were to assess whether long-term use of different doses and types of ICS were associated with differences in microbiome composition in asthma and the short-term stability of the microbiome.



Fifty-five adults (age ≥18) were recruited from Nottingham City Hospital outpatient respiratory clinics or from an existing research subject database. All subjects provided written, informed consent to the study which was approved by the National Research Ethics Committee East Midlands–Derby 1 (Ref 14/EM/0091).

The study included adult patients who had not smoked for ≥10 years and had a <10 pack year smoking history. All participants had a previous physician diagnosis of asthma (see S1 File) with no other respiratory diagnosis and no respiratory infection or antibiotic administration in the month prior to the study. Subjects with a post-bronchodilator FEV1 of <60% were excluded for safety reasons due to the use of nebulised hypertonic saline. Recruitment targeted 20 patients on low-dose ICS (beclomethasone dipropionate (BDP) ≤400 μg/day, fluticasone propionate (FP) ≤200 μg/day or budesonide (BUD) ≤400 μg/day for at least a year) and 30 taking high-dose ICS (FP ≥500 μg/day or BUD ≥800 μg/day for at least a year) with approximately equal numbers taking FP and BUD with or without long-acting β2-agonist.

Clinical measurements

At visit 1 demographic details and medical history were recorded. Spirometry was performed and the Asthma Control [17] and Leicester Cough Questionnaires [18] administered. Exhaled nitric oxide (FENO) measurement was performed prior to sputum induction. A subgroup of 20 patients agreeable to further visits were asked to return for repeat sputum induction after 24 h and 2 weeks (see S1 File). This subgroup of patients were asked about changes in symptoms or signs of active infection prior to each sputum induction and FEV1 prior to sputum induction was assessed on each occasion.

Sample collection

Induced sputum samples were obtained from patients using 3, 4 & 5% hypertonic saline as previously described [19, 20]. Samples were frozen at -80°C pending microbiological analysis.

Culture-independent analyses

DNA extraction.

Sputum samples were disrupted as previously described [21]. Processing followed recommendations for Gram-positive bacterial DNA extraction (GenElute™ Bacterial Genomic DNA Kit) (see S1 File).

16S rRNA gene sequencing.

16S rRNA gene sequencing (V3-V4 region) was performed on the Illumina MiSeq platform. Library preparation was carried out according to Illumina 16S Metagenomic sequencing library preparation manual (see S1 File). The paired end reads were rarefied to 9311 reads and transformed into Operational Taxonomic Units (OTU). This was determined using the QIIME version 1.9.1 pipeline to cluster the 16S rRNA gene sequences.


Total bacterial load qPCR was applied as previously described [22]. Two species-specific qPCR assays were used to quantify the load of Haemophilus influenzae [23] and Streptococcus pneumoniae [24] cells (see S1 File).

Statistical analyses

Microbiota analysis.

To identify multiple interactive effects, normalised OTU relative abundance (%) were multiplied by their corresponding qPCR-derived total bacterial load obtained from the qPCR data to give a CFU/ml value. Diversity indices were calculated from bootstrapped OTU tables using the vegan package of R ( such as richness, Shannon’s and Simpson’s. Prior to analyses, data were assessed for homogeneity of variance and the normal distribution of the residuals. Non-metric multidimensional scaling (NMDS) using Bray-Curtis distances assessed overall compositional dissimilarity. Analysis of Similarity (ANOSIM) was used to identify differences between the groups. The effect of treatment on individual OTU abundance was analysed using non-parametric Wilcoxon rank sum tests with Bonferroni corrections for multiple comparisons.

Quantitative bacterial load analysis.

Quantitative PCR data (CFU/ml) were logarithm (base = 10) transformed prior to analysis (SPSS Statistics v23.0; IBM Inc., USA). Same-day data were analysed based on a non-pairwise comparison (independent T-test or Mann-Whitney test). Data from different timepoints were compared using repeated measures ANOVA.



Induced sputum samples were collected from 55 subjects with asthma (Table 1). Twenty-two and 33 subjects were treated with low- and high-dose ICS, respectively (Tables 1 and 2). In the low-dose group, 10 subjects were taking BDP, 7 BUD and 5 FP; whereas in the high-dose group, 16 subjects were taking BUD and 17 subjects FP (Table 2). None of the subjects were taking LAMAs and 9 in the low-dose group and 30 in the high-dose group took concomitant LABAs. Repeat sputum samples (after 24 h and 2 weeks) were collected from 19 subjects, 8 of whom were taking low-dose ICS and 11 high-dose ICS (Table 2). None of the 19 subjects reported any change in symptoms prior to repeat sputum induction at visits 2 and 3 and all FEV1 measurements at visits 2 and 3 were within 150ml of the baseline values.

Table 2. Frequency distribution of patient groups attending visit 1 and visits 2 and 3 based on treatment group.

16S rRNA gene sequencing.

A total of 5,615,037 sequencing reads were generated from 54 of 55 asthma patients. The paired end reads were rarefied to 9311 reads resulting in one patient sample being excluded due to low reads (4693 reads). Sequences were clustered into 167 OTUs. The OTUs Streptococcus spp. and Prevotella spp. were detected in all patients with high mean abundance (%) (Table 3 and Fig 1). Despite the large number of OTUs identified from this cohort, only 11 OTUs were found in ≥90% of the patients. These included H. parainfluenzae, which was found in 90.7% patients. Based on a previous classification [25], OTUs were divided into aerobes/anaerobes (Table 3). Among the most prevalent OTUs, 27% were identified as aerobes, with Streptococcus spp. the most common. Among other OTUs found from this cohort were the genera Moraxella, Pseudomonas and Staphylococcus. Though Moraxella spp. accounted for only 33% of the cohort, 4 patients (aged 68–71 years) had a very high relative abundance of the genus (16.17–52.59%). S. aureus was identified in 61% of patients; only one patient had this taxon as a high relative abundance (17.65%) with others below 4%. Pseudomonas spp. was found in 81% of the cohort, with all relative abundances below 4%.

Fig 1. Top 20 OTUs identified from 54 patients, with the highest mean abundance (%) being the topmost.

The colours indicate the abundance of OTUs in each patient, with blue regarded as low abundance, red as moderate abundance and green as high abundance.

Table 3. The most prevalent OTUs across samples (90–100% patients) in descending order of mean abundance (%).

Correlation of lung microbiota with ICS dose.

ICS dose (high or low) did not affect species richness (p = 0.23), diversity (Simpson’s Diversity Index, p = 0.134), or community structure (Shannon’s index, p = 0.256). The microbiota composition in patients from the low- and high-dose ICS groups is shown in Fig 2. No significant differences were observed in the microbiota composition between the groups (Bray-Curtis; p = 0.676).

Fig 2. The Non-metric Multidimensional Scaling (NMDS) plot represented the bacterial composition of each patient, grouped by ICS dose; low-dose ICS (red) and high-dose ICS (blue).

The shorter the distance between the dots, the more similar the bacterial compositions were between the patients.

Load changes of individual OTUs between ICS dose groups are presented in Table 4. In total, 32 OTUs were significantly more prevalent in low-dose ICS patients. Among these were 5 of the most prevalent OTUs from the cohort as a whole (Streptococcus spp., V. dispar, Atopobium spp., Selenomonas spp. and P. pallens; Table 3). Streptococcus spp. was the most abundant OTU detected in the low-dose ICS group (2.6 x 107 CFU/ml), at significantly higher levels than were found in the high-dose ICS group (1.72 x 106 CFU/ml, p = 0.002). Three other members of the Prevotella genus (besides P. pallens) were also found to be more abundant in the low-dose ICS group, namely P. tannerae, P. melaninogenica and P. nigrescens (Table 4). Most of the more abundant organisms in low-dose ICS patients were anaerobes. Only 7 OTUs were significantly higher in high-dose ICS patients but with lower mean abundances than those more frequent in the low-dose group (Dysgonomonas spp., Parvimonas spp.).

Table 4. Significant OTU changes between ICS dose groups.

Correlating lung microbiome with type of high-dose ICS.

Species richness was not significantly different between high-dose FP and BUD treated subjects (p = 0.9). Similarly, non-significant values were evident for Simpson’s Diversity Index (p = 0.787) and community structure (p = 0.887). The microbiota composition in patients from different types of high-dose ICS groups is shown in Fig 3. The Bray-Curtis distances between groups showed no significant differences in bacterial composition (p = 0.345).

Fig 3. The Non-metric Multidimensional Scaling (NMDS) plot based on bacterial composition of patients on high-dose ICS taking budesonide (red) and fluticasone propionate (blue).

The shorter the distance between the dots, the more similar the bacterial compositions were between the patients.

Haemophilus and high-dose ICS.

A total of 9 OTUs had significantly higher mean abundance in the high-dose FP group (Table 5). Among these, H. parainfluenzae had a significantly higher abundance in individuals taking high-dose FP (mean abundance: 9.64 x 105 CFU/ml) than those taking high-dose BUD (1.45 x 105 CFU/ml, p = 0.047). This led to the qPCR investigation to determine H. influenzae and S. pneumoniae loads in high-dose FP and BUD groups (described in S1 File and S1 Fig), which determined H. influenzae loads were significantly higher than S. pneumoniae in both high-dose FP and BUD groups but found no significant differences in H. influenzae or S. pneumoniae loads between these groups.

Table 5. OTU changes between type of ICS in the high-dose ICS group.

Changes in microbiota over time.

Both analyses on microbiota composition and qPCR total bacteria data between different time points (Day 1, 2 and 14) of 19 patients demonstrated no significant changes in diversity measures and bacterial load (Fig 4). The overall composition was measured by species richness, diversity (Simpson’s) and community structure (Shannon’s) with no significant changes between different time points.

Fig 4.

The NMDS (non-metric multidimensional scaling) of the microbiota composition of asthma patients (top) and total bacterial load (log CFU/ml +1) (bottom) grouped into ICS dose group of low-dose ICS (red) and high-dose ICS (green) at (a) Day 1, (b) Day 2 (after 24 h) and (c) Day 14 (after 2 weeks).


This study presents a culture-independent analysis of bacteria present in respiratory samples of adults with clinically stable asthma and provides several important and novel findings to the field of asthma microbiome studies. All individuals had been taking the same type and dose of ICS for at least a year allowing the impact of long-term usage to be evaluated. A number of patients were sampled more than once allowing insights into short-term asthma microbiota stability.

From the 167 OTUs identified, a restricted number (eleven) were found in over 90% of individuals studied. This is in keeping with findings for the cystic fibrosis airways microbiome in which core and transient populations have been established [25]. In this study, the taxa identified often matched those previously found in asthma and in cystic fibrosis and bronchiectasis such as Streptococcus spp., Haemophilus spp. and anaerobes (Veillonella spp., Prevotella spp.). These have been detected in poorly-controlled asthma in adults [16, 26], treatment-resistant severe asthma in adults [14] and mild [27] and severe asthma patients [10]. Of these genera, Haemophilus and Streptococcus include species considered in other contexts to be pathogens. The consequence of identifying putative pathogens in the asthmatic airways raises important clinical questions as to their potential contribution to the pathophysiology of asthma and also whether or not to treat potential pathogens when detected by such technologies. In relation to this, it is important to consider the bacterial load carried in the asthmatic airways. The average sputum bacterial load of this cohort (~1x107 CFU/mL) is between that observed in healthy controls (<1x106 CFU/mL) [28] and in cystic fibrosis (~1x108 CFU/mL) [29]. As such, the potential for pathophysiological airway damage is a distinct possibility in asthmatic individuals.

The relationship between ICS dose and lung microbiota composition was explored. The clinical characteristics of the low- and high-dose ICS groups were similar with no significant differences in FEV1, FENO or peak blood eosinophil count/number of severe asthma exacerbations in the last year. In general, ICS dose did not affect ecological community characteristics. In contrast, previously, microbiota from severe asthmatics (taking a similar dose of ICS to our high-dose group) was enriched in 53 genera compared to mild/moderate subjects, the majority being Actinobacteria, with the remaining 5 classifying to Gammaproteobacteria [30]. The differences between previous results and our findings may be due in part by fact that Huang et al. sampled the microbiota via bronchial brushings. It is also possible that the severe asthma group from this study had sub-optimal asthma control whilst our high-dose ICS group were generally well-controlled.

A number of genera had significantly higher relative abundances in low-dose ICS patient samples with a smaller number of OTUs detected that were more abundant in the high-dose group. The most prevalent OTU identified in low-dose samples was Streptococcus spp. This genus contains species that differ in their potential to cause pathophysiological damage and without extensive subsequent analysis it is not possible to assess whether the Streptococcal species being detected are pathogens or not. Cox et al. [31] previously reported that 16S rRNA gene sequencing was unable to discriminate between S. pneumoniae and Streptococcus mitis, the latter being a normal commensal of the oropharynx. A smaller set of genera were more abundant in the high-dose ICS group. Of these, the most prevalent was Parvimonas spp. a genus that contains species regarded as human pathogens [32]. The significance of the differing abundances of these genera between high- and low-dose ICS groups is however unclear. These differences may be part of a process which is either a cause or consequence of disease progression. It is also possible that ICS act as “selecting agents” for certain species. In many studies high doses of specific compounds are associated with selection of a few distinct taxa [33].

No significant differences in overall microbiota composition or bacterial load were identified between patients taking high-dose BUD and high-dose FP. However a higher abundance of H. parainfluenzae was identified in the high-dose FP group than in those using high-dose BUD. A selective effect of long-term FP usage on the abundance of this potential pathogen is therefore a possibility. Goleva et al. [34] found an increased abundance of H. parainfluenzae in asthmatics “resistant” to a treatment trial of oral prednisolone in comparison to those who were steroid “sensitive” and an inhibitory effect of H. parainfluenzae on asthmatic airway macrophages in vitro was also observed. The differences in abundance of other organisms between high-dose BUD and FP are of uncertain significance and include some organisms usually associated with oral (Veillonellaceae, Porphyromonas endodontalis) and GI (Lactobacillus reuteri, Peptococcus spp.) microbiota.

Evidence is accumulating that ICS treatment is associated with an increased risk of pneumonia in COPD [6, 35] and respiratory infection in asthma [4, 5] and that this increased risk is greater with FP [36] than BUD. Our findings, combined with the observation that short-term treatment with FP can cause distinct changes in the microbiota [37], provide a possible mechanism but a prospective study of the microbiota of patients starting different ICS is really required to strengthen these observations.

This study also examined the short-term stability of the asthma microbiota. No statistically significant differences were seen in community composition between the three time points over two weeks, suggesting the asthma microbiome is typically stable over short periods of time in clinical stability.

This study has limitations. Microbial contamination, possible from the oral cavity, has been covered extensively elsewhere [38]. No evidence for gross contamination was observed here with the commonality of species detected across over 90% of individuals with asthma supportive of non-contaminated sample collection. A further limitation could be considered the lack of subject phenotyping. Distinct asthma phenotypes may have distinct airways microbiota [39]. This will need to be considered when designing future studies. Also, the mean age of our patients was higher than previous studies [10, 27, 30]. Whilst the effect of age on the microbiota has not previously been investigated, data from CF patients suggest a decrease in microbiota diversity with age [40]. Finally, a “molecular” limitation was that the differences in taxa were described primarily at genus level making the biological effects less clear.


This study did not demonstrate a significant difference in microbiota composition between asthma patients taking low- and high-dose ICS. However, an association was identified between high-dose FP and increased abundance of the pathogen H. parainfluenzae. The clinical implications for patients are not known but this does provide a possible explanation for the increased risk of pulmonary infection seen in asthma and COPD, particularly with FP.

Supporting information

S1 Fig. Total bacteria, H. influenzae and S. pneumoniae levels (log CFU/ml +1) in subjects receiving high-dose inhaled corticosteroids: Budesonide (BUD) and fluticasone propionate (FP).



We would like to acknowledge the efforts of Wendy Gerrard-Tarpey for her assistance with some of the patient visits and Helen Lee and Carly Clayton for their assistance with processing of sputum samples.


  1. 1. Global Asthma Network. The Global Asthma Report 2018. Available From 2018.
  2. 2. Mukherjee M, Stoddart A, Gupta RP, Nwaru BI, Farr A, Heaven M, et al. The epidemiology, healthcare and societal burden and costs of asthma in the UK and its member nations: analyses of standalone and linked national databases. BMC Med. 2016;14(1):113. pmid:27568881
  3. 3. British Thoracic Society, Scottish Intercollegiate Guidelines Network. British guideline for the management of asthma; a national clinical guideline. July 2019.
  4. 4. McKeever T, Harrison TW, Hubbard R, Shaw D. Inhaled Corticosteroids and the Risk of Pneumonia in people with Asthma: A case control study. Chest. 2013. pmid:23990003
  5. 5. Qian CJ, Coulombe J, Suissa S, Ernst P. Pneumonia risk in asthma patients using inhaled corticosteroids: a quasi-cohort study. British Journal of Clinical Pharmacology. 2017;83(9):2077–86. pmid:28425216
  6. 6. Suissa S, Patenaude V, Lapi F, Ernst P. Inhaled corticosteroids in COPD and the risk of serious pneumonia. Thorax. 2013;68(11):1029–36. pmid:24130228
  7. 7. Bansal V, Mangi MA, Johnson MM, Festic E. Inhaled corticosteroids and incident pneumonia in patients with asthma: Systematic review and meta-analysis. Acta Med Acad. 2015;44(2):135–58. pmid:26702909
  8. 8. Wood LG, Simpson JL, Hansbro PM, Gibson PG. Potentially pathogenic bacteria cultured from the sputum of stable asthmatics are associated with increased 8-isoprostane and airway neutrophilia. Free radical research. 2010;44(2):146–54. pmid:19922242
  9. 9. Simpson JL, Grissell TV, Douwes J, Scott RJ, Boyle MJ, Gibson PG. Innate immune activation in neutrophilic asthma and bronchiectasis. Thorax. 2007;62(3):211–8. pmid:16844729
  10. 10. Zhang Q, Cox M, Liang Z, Brinkmann F, Cardenas PA, Duff R, et al. Airway microbiota in severe asthma and relationship to asthma severity and phenotypes. PloS one. 2016;11(4):e0152724. pmid:27078029
  11. 11. Huang YJ, Nariya S, Harris JM, Lynch SV, Choy DF, Arron JR, et al. The airway microbiome in patients with severe asthma: Associations with disease features and severity. Journal of Allergy and Clinical Immunology. 2015;136(4):874–84. pmid:26220531
  12. 12. Simpson JL, Daly J, Baines KJ, Yang IA, Upham JW, Reynolds PN, et al. Airway dysbiosis: Haemophilus influenzae and Tropheryma in poorly controlled asthma. European Respiratory Journal. 2016;47(3):792–800. pmid:26647445
  13. 13. Slater M, Rivett DW, Williams L, Martin M, Harrison T, Sayers I, et al. The impact of azithromycin therapy on the airway microbiota in asthma. Thorax. 2014;69(7):673–4. pmid:24287164
  14. 14. Green BJ, Wiriyachaiporn S, Grainge C, Rogers GB, Kehagia V, Lau L, et al. Potentially Pathogenic Airway Bacteria and Neutrophilic Inflammation in Treatment Resistant Severe Asthma. PLoS ONE. 2014;9(6):e100645. pmid:24955983
  15. 15. Hilty M, Burke C, Pedro H, Cardenas P, Bush A, Bossley C, et al. Disordered microbial communities in asthmatic airways. PloS one. 2010;5(1):e8578. pmid:20052417
  16. 16. Taylor SL, Leong LEX, Mobegi FM, Choo JM, Wesselingh S, Yang IA, et al. Long-Term Azithromycin Reduces Haemophilus influenzae and Increases Antibiotic Resistance in Severe Asthma. Am J Respir Crit Care Med. 2019;200(3):309–17. pmid:30875247
  17. 17. Juniper EF, O'Byrne PM, Guyatt GH, Ferrie PJ, King DR. Development and validation of a questionnaire to measure asthma control. Eur Respir J. 1999;14(4):902–7. pmid:10573240
  18. 18. Birring SS, Prudon B, Carr AJ, Singh SJ, Morgan MDL, Pavord ID. Development of a symptom specific health status measure for patients with chronic cough: Leicester Cough Questionnaire (LCQ). Thorax. 2003;58(4):339–43. pmid:12668799
  19. 19. Paggiaro PL, Chanez P, Holz O, Ind PW, Djukanović R, Maestrelli P, et al. Sputum induction. European Respiratory Journal. 2002;20(37 suppl):3s–8s. pmid:12361361
  20. 20. Pavord ID, Pizzichini MM, Pizzichini E, Hargreave FE. The use of induced sputum to investigate airway inflammation. Thorax. 1997;52(6):498–501. pmid:9227713
  21. 21. Rogers GB, Zain NM, Bruce KD, Burr LD, Chen AC, Rivett DW, et al. A novel microbiota stratification system predicts future exacerbations in bronchiectasis. Annals of the American Thoracic Society. 2014;11(4):496–503. pmid:24592925
  22. 22. Rogers GB, Carroll MP, Zain NMM, Bruce KD, Lock K, Walker W, et al. Complexity, temporal stability, and clinical correlates of airway bacterial community composition in primary ciliary dyskinesia. Journal of Clinical Microbiology. 2013;51(12):4029–35. pmid:24068019
  23. 23. HPA. Developed in-house at the Health Protection Agency London by Dr. Clare Ling. 2011.
  24. 24. Carvalho MDGS, Tondella ML, McCaustland K, Weidlich L, McGee L, Mayer LW, et al. Evaluation and improvement of real-time PCR assays targeting lytA, ply, and psaA genes for detection of pneumococcal DNA. Journal of Clinical Microbiology. 2007;45(8):2460–6. pmid:17537936
  25. 25. van der Gast CJ, Walker AW, Stressmann FA, Rogers GB, Scott P, Daniels TW, et al. Partitioning core and satellite taxa from within cystic fibrosis lung bacterial communities. ISME J. 2011;5(5):780–91. pmid:21151003
  26. 26. Simpson JL, Daly J, Baines KJ, Yang IA, Upham JW, Reynolds PN, et al. Airway dysbiosis: Haemophilus influenzae and Tropheryma in poorly controlled asthma. Eur Respir J. 2015;47(3):792–800. pmid:26647445
  27. 27. Marri PR, Stern DA, Wright AL, Billheimer D, Martinez FD. Asthma-associated differences in microbial composition of induced sputum. J Allergy Clin Immunol. 2013;131(2):346-52.e1-3. pmid:23265859
  28. 28. Marsh RL, Nelson MT, Pope CE, Leach AJ, Hoffman LR, Chang AB, et al. How low can we go? The implications of low bacterial load in respiratory microbiota studies. Pneumonia. 2018;10(1):7.
  29. 29. Zemanick ET, Harris JK, Wagner BD, Robertson CE, Sagel SD, Stevens MJ, et al. Inflammation and Airway Microbiota during Cystic Fibrosis Pulmonary Exacerbations. Plos One. 2013;8(4). pmid:23646159
  30. 30. Huang YJ, Nariya S, Harris JM, Lynch SV, Choy DF, Arron JR, et al. The airway microbiome in patients with severe asthma: Associations with disease features and severity. J Allergy Clin Immunol. 2015;136(4):874–84. pmid:26220531
  31. 31. Cox MJ, Turek EM, Hennessy C, Mirza GK, James PL, Coleman M, et al. Longitudinal assessment of sputum microbiome by sequencing of the 16S rRNA gene in non-cystic fibrosis bronchiectasis patients. PloS one. 2017;12(2):e0170622. pmid:28170403
  32. 32. Murphy EC, Frick IM. Gram-positive anaerobic cocci—commensals and opportunistic pathogens. FEMS Microbiol Rev. 2013;37(4):520–53. pmid:23030831
  33. 33. Kochling T, Sanz JL, Galdino L, Florencio L, Kato MT. Impact of pollution on the microbial diversity of a tropical river in an urbanized region of northeastern Brazil. International microbiology: the official journal of the Spanish Society for Microbiology. 2017;20(1):11–24.
  34. 34. Goleva E, Jackson LP, Harris JK, Robertson CE, Sutherland ER, Hall CF, et al. The Effects of Airway Microbiome on Corticosteroid Responsiveness in Asthma. Am J Respir Crit Care Med. 2013;188(10):1193–201. pmid:24024497
  35. 35. Kew KM, Seniukovich A. Inhaled steroids and risk of pneumonia for chronic obstructive pulmonary disease. Cochrane Database Syst Rev. 2014(3):Cd010115. pmid:24615270
  36. 36. Janson C, Stratelis G, Miller-Larsson A, Harrison TW, Larsson K. Scientific rationale for the possible inhaled corticosteroid intraclass difference in the risk of pneumonia in COPD. International journal of chronic obstructive pulmonary disease. 2017;12:3055–64. pmid:29089754
  37. 37. Durack J, Lynch SV, Nariya S, Bhakta NR, Beigelman A, Castro M, et al. Features of the bronchial bacterial microbiome associated with atopy, asthma, and responsiveness to inhaled corticosteroid treatment. J Allergy Clin Immunol. 2017;140(1):63–75. pmid:27838347
  38. 38. Dickson RP, Erb-Downward JR, Martinez FJ, Huffnagle GB. The Microbiome and the Respiratory Tract. Annual review of physiology. 2016;78:481–504. pmid:26527186
  39. 39. Taylor SL, Leong LEX, Choo JM, Wesselingh S, Yang IA, Upham JW, et al. Inflammatory phenotypes in patients with severe asthma are associated with distinct airway microbiology. J Allergy Clin Immunol. 2018;141(1):94-103.e15. pmid:28479329
  40. 40. Coburn B, Wang PW, Diaz Caballero J, Clark ST, Brahma V, Donaldson S, et al. Lung microbiota across age and disease stage in cystic fibrosis. Sci Rep. 2015;5:10241. pmid:25974282