SEARCH

SEARCH BY CITATION

Keywords:

  • alveolar;
  • asthma;
  • FcεRI;
  • IgE;
  • mast cell

Abstract

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Conflict of interest
  8. Sources of support
  9. Funding
  10. Authors’ contributions
  11. References

To cite this article: Andersson CK, Tufvesson E, Aronsson D, Bergqvist A, Mori M, Bjermer L, Erjefält JS. Alveolar mast cells shift to an FcεRI-expressing phenotype in mild atopic asthma: a novel feature in allergic asthma pathology. Allergy 2011; 66: 1590–1597.

Abstract

Background:  A unique feature of alveolar mast cells is their low high-affinity IgE receptor (FcεRI) expression. Recent discoveries in uncontrolled asthma suggest that the appearance of FcεRI-expressing alveolar mast cells may be a novel disease-specific feature of allergic asthma. This study investigates whether increased FcεRI-expressing alveolar mast cells are present in patients with mild allergic asthma or even in non-asthmatic allergic rhinitis patients (AR) who have developed bronchial hyperactivity (BHR).

Methods:  Bronchial and alveolar tissues were obtained from healthy controls, AR patients with or without BHR, and AR patients with concurrent asthma. Samples were processed for immunohistochemical identification of MCT and MCTC and expression of FcεRI and surface-bound IgE.

Results:  Bronchial mast cell expression of FcεRI was high in all groups. In contrast, in the alveolar tissue, the expression of FcεRI on mast cells was low in healthy controls and in the AR patient groups, whereas a high expression was present in AR patients with concurrent asthma (P = 0.006 compared to controls). The asthmatics had a 29-fold increase in numbers (P = 0.006) and a 19-fold increase in proportion (P = 0.007) of alveolar mast cells that expressed surface-bound IgE.

Conclusions:  The present data show that alveolar mast cells in patients with mild atopic asthma, but not atopic patients with AR, have turned into a highly FcεRI- and IgE-expressing phenotype. These data support the hypothesis that increased FcεRI expression on alveolar mast cells is a novel disease-specific feature of allergic asthma that is important for understanding asthma phenotypes and designing new therapeutic strategies.

Immunoglobulin E (IgE)- and FcεRI-mediated degranulation of mast cells is a key event in the induction of the allergic inflammation that characterizes allergic rhinitis (AR) and asthma (1). Mast cells are abundant at all airway levels with a gradual increase in density towards the alveolar parenchyma (2, 3). It has been demonstrated that human alveolar mast cells have an exceedingly low expression of FcεRI (2). We recently found a marked increase in mast cell expression of FcεRI (40-fold increase) and IgE (500-fold increase) in alveolar tissues from patients with atopic uncontrolled asthma compared to healthy control subjects (4). It can thus be speculated that the appearance of highly FcεRI-expressing alveolar mast cells is a novel and disease-specific feature of allergic asthma. However, so far, no data exist on this phenomenon in other asthma phenotypes or in patients with AR who have developed a bronchial hyper-responsiveness (BHR) and thus are in the risk zone of developing asthma (5–7).

Using a unique study design that allows alveolar sampling not only from asthmatics but also from healthy control subjects, this study tests the hypothesis that also patients with concurrent AR and mild asthma have increased mast cell expression of FcεRI in the alveolar parenchyma and that this phenomenon is less developed in AR patients who have BHR but not yet developed asthma. This is investigated through analyses of the density of mast cell populations and their expression of FcεRI and mast cell–bound IgE in central airways and alveolar parenchyma in healthy controls, AR patients with and without BHR and AR patients with concurrent mild asthma.

Patients with atopic asthmatic frequently have concurrent AR (8). Reports show that AR patients often present unspecific BHR and bronchial inflammation (9). The similarities in induction, development and type of inflammation in the nasal and bronchial mucosa have led to the view that allergic asthma and AR may represent organ-specific variants of the same underlying disease with a gradual development of an upper airway allergic inflammation (rhinitis) towards involvement of the lower airways (asthma) (10).

Our group have previously reported that patients with mild atopic asthma but not patients with rhinitis have evidence of peripheral airway involvement, measured as increased alveolar nitric oxide (11) and peripheral airflow resistance (12). Our findings significantly advance the novel hypothesis that the appearance of FcεRI- and IgE-expressing alveolar mast cells is a novel and disease-specific feature of allergic asthma (4).

Methods

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Conflict of interest
  8. Sources of support
  9. Funding
  10. Authors’ contributions
  11. References

Allergy screening

Atopy was defined as being positive for skin prick test (SPT) to one or more airborne allergens included in the test panel (birch, timothy, mugwort, cat, dog, horse, Dermatophagoides pteronyssinus, Dermatophagoides farinae, Aspergillus fumigatus and Cladosporium herbarum) (ALK Abello, Copenhagen, Denmark). For subjects with positive SPT to pollen, bronchoscopy procedure was performed outside pollen season.

Subjects

This study included 24 nonsmoking patients with confirmed AR (13). Of these, eight had a concomitant diagnosis of mild atopic asthma according to GINA guidelines (14). Among the 16 AR patients with no concomitant asthma, eight were hyper-responsive to methacholine (PD20 < 2000 μg). Eight healthy, never-smoking nonatopic subjects (no respiratory symptoms, normal lung function, negative SPT and not hyper-responsive to methacholine) were used as controls (Table 1). From each of the 32 subjects, central airway biopsies and transbronchial biopsies were collected during a study period from September 2005 to March 2008 at the Department of Respiratory Medicine, Lund University Hospital. All subjects gave their written informed consent to participate in the study, which was approved by the ethics committee in Lund (LU412-03). The healthy control group and patients with AR without BHR have been used in a previously published study (4).

Table 1.   Subject characteristics
 Controls n = 8AR n = 8AR + BHR n = 8AR + Asthma n = 8
  1. M, male; F, female; ICS, inhaled corticosteroid; p.r.n., pro re nata (as needed); FEV1, forced expiratory volume in 1 s; AR, allergic rhinitis; BHR, bronchial hyper-responsiveness; FeNO50, fractional exhaled NO at flow 50 ml/s.

  2. *Data are given as mean (range).

Males/Females5/35/35/35/3
Age*, years23 (21–39)26 (22–58)27 (22–28)28 (22–37)
ICS, n0000
Nasal GCS p.r.n, n0210
β2 agonist p.r.n, n0015
Antihistamines p.r.n, n0441
Lung function
 FEV1*3.70 (2.4–5.4)3.32 (3.0–4.7)3.59 (3.1–4.8)4.12 (3.1–7.3)
 FEV1% of predicted*98.1 (72–116)106.9 (96–138)99.5 (91–107)96.4 (88–124)
 PD20* (μg)>2000>2000614 (193–965)79.4 (5–2000)
NO measurements
 FeNO50* (ppb)13.9 (8.4–21.1)21.3 (9.8–59.5)25.6 (5.1–57.5)85.9 (15.8–240)
 Bronchial NO flux* (nl/s)0.73 (0.38–1.10)1.01 (0.34–3.05)1.24 (0.08–2.95)4.49 (0.68–12.4)
 Alveolar NO* (ppb)2.30 (1.68–3.70)3.04 (1.74–4.59)3.42 (2.53–5.28)3.42 (2.09–5.62)

Bronchoscopy with collection of bronchial and transbronchial biopsies

From each patient, central airway biopsy specimens (n = 5 per patient) were taken from the segmental or subsegmental carina in the right lower and upper lobes, followed by sampling of transbronchial biopsy specimens (n = 5 per patient) in the right lower lobe. Bronchoscopy was performed as previously described (4).

Spirometry and methacholine inhalation challenge test

For measurements of lung function, a MasterScope spirometer (version 4.5, Erich Jaeger GmbH, Wurzburg, Germany) was used with reference values from the study of Crapo et al. (15). Only patients with an FEV1 baseline value of ≥70% of predicted were included in the study. Presence of BHR was measured with a methacholine challenge test (Aerosol Provocation System, APS; Erich Jaeger GmbH), as described previously (13).

Impulse oscillometry

A Jaeger MasterScreen Impulse Oscillometry System, Erich Jaeger GmbH, was used 90 s after each step of the challenge, prior to FEV1 [as previously described (12)]. Resistance was reported as R5 (total respiratory resistance of the airways) and R20 (resistance of the proximal airways). The ΔR5-R20 parameter is an indicator of peripheral resistance of the respiratory tract. Resonant frequency (Fres) is a good index of changes in the degree of peripheral involvement (12). The IOS parameters were plotted against the methacholine dose at each challenge step, and linear regression analysis was used to calculate a slope value (for example Slope-FresMCh). Thus, an increased slope value (i.e. a steeper slope) represents a higher degree of reactivity to methacholine, and in the case of Slope-FresMCh, this would indicate more involvement of the peripheral airways.

Measurements of exhaled nitric oxide

Measurements were taken as previously described (11). Briefly, FeNO measurements were taken prior to bronchial challenge testing at flow rates of 50, 100, 200 and 400 ml/s using a NIOX nitric oxide analyser (Aerocrine AB, Stockholm, Sweden). FeNO50 is defined as the amount of NO at 50 ml/s. Alveolar NO concentration and bronchial flux of NO were calculated with a two-compartment linear model using a flow rate of 100–400 ml/s.

Processing of tissue

Immediately after collection, the bronchial and transbronchial biopsies were fixed in periodate–lysine containing 1% paraformaldehyde (1% PLP) for 4 h at 4°C (16). Specimens were embedded in OCT (Tissue-Tek, Sakura Finetek Eu, Alphen aan den Rijn, the Netherlands) and stored at −80°C. Cryosections were cut serially at 8 μm, enwrapped in aluminium foil and stored at −80°C until used. Sections stained with Mayer’s haematoxylin were used to select two bronchial and two transbronchial biopsies from each patient for further analysis.

Immunohistochemistry

All antibodies used have been extensively validated for staining of human tissue in research and routine clinical diagnosis (Table 2). Staining was absent in sections using isotype-matched control antibodies (Dako, Glostrup, Denmark).

Table 2.   Primary antibodies used for immunohistochemistry
AntibodySpeciesDilutionCloneOriginSecondary antibody
  1. Heat-induced antigen retrieval was performed in DIVA decloaker (Biocare Medical, Concord, CA, USA), pH 6, in a pressure cooker (2100 Retriever, Prestige Medical Ltd., Blackburn, UK).

Anti-tryptase (27)Mouse1 : 12 000G3Chemicon, Temecula, CAEnVision™ G|2 Double stain System, Dako  or Direct labelling, Zenon, Invitrogen
Anti-chymase (28)Mouse1 : 100CC1Novocastra, Newcastle upon Tyne, UK
Anti-FcεR1α (2)Rabbit1 : 50(polyclonal)Proteintech Group, Chicago, ILBiotinylated Goat anti-Rabbit IgG
Anti-IgE (29)Rabbit1 : 8000(polyclonal)Dako, Glostrup, Denmark 
Double immunohistochemical staining of MCTC and MCT

A double-staining protocol was used for simultaneous visualization of MCTC and MCT cells (2, 4, 17). The immunostaining was performed as previously described (4) using an immunohistochemistry robot (Autostainer; Dako) with EnVision™ G|2 Doublestain System (K5361; Dako). For details, see Table 2.

Immunohistochemical identification of FcεRI+ and IgE+ mast cells

Immunofluorescence triple staining was used to simultaneously visualize both MCTC and MCT populations together with the expression of the IgE receptor (FcεRI) or IgE bound to the surface of the mast cells [as previously described (4)].

Tissue analysis

Quantification of density of mast cell subtypes

High-resolution images of sections double-stained for MCTC and MCT were generated through an automated digital slide-scanning robot (Scanscope CS™; Aperio, Vista, CA, USA). All mast cells of each subpopulation per biopsy were quantified manually and related to the tissue area in randomized, blinded images using an image analysis program (ImageScope, v10.0.36.1805, Aperio) (4).

Quantification of FcεRI+ and IgE+ mast cells

After triple immunofluorescence staining, tryptase-positive mast cells (488 nm), chymase (647 nm) and expression of FcεRIα or surface-bound IgE (555 nm) (2, 17) were analysed, and the proportion (%) of total number of mast cells and MCTC and MCT positive for FcεRI and IgE was calculated. The density of mast cells expressing FcεRI and IgE was calculated by multiplying the percentage of MCFcεRI+ or MCIgE+ with the total mast cell density in the same tissue region (4).

Statistical analysis

Data were analysed using Kruskal–Wallis test with Bonferroni’s multiple comparisons test for comparison between three groups or more, and Mann–Whitney rank sum test was used for comparison between two groups using GraphPad Prism version 5 (GraphPad Software, Inc., La Jolla, CA, USA). The Spearman test (two-tailed) was used to study the correlations. Correlation analysis was performed within pooled atopic patients. Results were considered significant at P ≤ 0.05 (*P ≤ 0.05, **P ≤ 0.01 and ***P ≤ 0.001).

Results

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Conflict of interest
  8. Sources of support
  9. Funding
  10. Authors’ contributions
  11. References

Patient characteristics

The 24 patients (11 men and 13 women) with AR were all atopic (i.e. positive SPT). Two patients had only allergy to pollen. None of the subjects were treated with inhaled GCS (ICS). Three patients were treated with nasal GCS, six with β2 agonists and nine with antihistamines p.r.n. For details, see Table 1. No difference in FEV1 or FEV1% predicted was found between the groups (P = 0.4 and 0.3, respectively). FeNO50 was significantly increased in AR patients with asthma compared to nonatopic controls (P = 0.01). For bronchial flux of NO, levels were increased in AR patients with concurrent asthma compared to those of both nonatopic controls (P = 0.03) and AR (P = 0.03). No significant difference in alveolar NO was observed between the groups (Table 1).

Mast cell densities in central airways and alveolar parenchyma

No significant difference in total mast cell numbers or in the density of the two subtypes, MCT and MCTC, was found between the examined groups: healthy controls, AR with and without BHR, and AR with concomitant asthma in central airways or alveolar parenchyma (Fig. 1).

image

Figure 1.  Scattergrams displaying total (MCtot), MCT and MCTC mast cell densities in central airways (A–C) and alveolar parenchyma (D–F) in healthy nonatopic controls and allergic rhinitis patients with and without bronchial hyper-responsiveness and asthma. Horizontal bars represent median values. Overall significance, using Kruskal–Wallis test, is denoted in each picture.

Download figure to PowerPoint

Mast cell expression of FcεRIα and mast cell–bound IgE in central airways and alveolar parenchyma

As previously shown (2, 18, 19), in healthy controls, mast cells positive for FcεRIα were present in high numbers in central airways. No significant difference was observed between AR patients with or without BHR and AR patients with concurrent asthma compared to nonatopic controls (Fig. 2A). In alveolar parenchyma, the mast cell expression of FcεRIα was low in nonatopic controls and in AR patients with or without BHR. A significant increase was found in AR patients with concomitant asthma compared to healthy nonatopic controls (P = 0.01) and patients with AR without BHR (P = 0.01, Fig. 2B). In central airways and alveolar parenchyma, no difference in the expression of FcεRIα on the MCT and the MCTC subpopulations was observed.

image

Figure 2.  Mast cells (%) staining for FcεRI (A–B) and mast cell–bound IgE (C–D) in central airways and alveolar parenchyma in healthy nonatopic controls and allergic rhinitis (AR) patients with and without bronchial hyper-responsiveness (BHR) and asthma. Horizontal bars indicate median values. Results were considered significant at P ≤ 0.05, and *denotes P ≤ 0.05, **P ≤ 0.01 in comparison with nonatopic controls and #denotes P ≤ 0.05 in comparison with AR without BHR. Overall significance, using Kruskal–Wallis test, is denoted in each picture.

Download figure to PowerPoint

In central airways, the proportion of mast cells with bound IgE was low in healthy controls and gradually increased in AR patients without and with BHR. A significant increase in the proportion of mast cells with positive staining for surface-bound IgE was observed in AR patients with concurrent asthma compared to nonatopic controls (P = 0.002) (Fig. 2C). In the alveolar parenchyma, the proportion of mast cells with bound IgE was low in healthy controls and AR patients without and with BHR. In the alveolar parenchyma, a significant increase in the proportion of mast cell with bound IgE was observed in AR patients with concurrent asthma compared to nonatopic controls (P = 0.002) (Fig. 2D). In central airways and alveolar parenchyma, no difference was observed in the proportion of mast cells with bound IgE between the MCT and the MCTC subpopulations.

Density of mast cells expressing FcεRIα and IgE in alveolar parenchyma

The total tissue density (per mm2) of mast cells positive for FcεRIα and IgE was calculated. In alveolar parenchyma, the number of FcεRIα+ mast cells was increased in AR patients with concomitant asthma compared to that of nonatopic controls (Fig. 3A). An increase in the number of mast cells positive for surface-bound IgE was found in AR patients with concomitant asthma and AR patients with BHR compared to nonatopic controls (Fig. 3B).

image

Figure 3.  Density of mast cells expressing FcεRIα (A) and mast cell–bound IgE (B) in alveolar parenchyma in healthy nonatopic controls and allergic rhinitis patients with and without bronchial hyper-responsiveness and asthma. Results were considered significant at P ≤ 0.05, and *denotes P ≤ 0.05, **P ≤ 0.01 in comparison with nonatopic controls. Overall significance, using Kruskal–Wallis test, is denoted in each picture. Picture C shows a representative micrograph of FcεRIα (red)-positive mast cells (tryptase: green) in a central airway of an asthmatic patient. Panels D–E show IgE positive mast cells in alveolar parenchyma from control (D) and asthmatic patient (E). Insets represents image (600×) of mast cells double positive for tryptase and IgE. Scale bar: C, E = 50 μm and D = 100 μm.

Download figure to PowerPoint

Correlation analyses

Significant negative correlations were found between PD20 and bronchial mast cell expression of FcεRIα (P = 0.02) and IgE (P = 0.02). The bronchial NO flux (P < 0.0001) and FeNO50 (P < 0.0001) correlated positively with mast cell–bound IgE in the bronchi. The alveolar mast cell expression of FcεRIα correlated positively with Slope-FresMCh (P = 0.009). No correlations between mast cell parameters and R5, R20 and R5–R20 were found. Results are presented in Table 3.

Table 3.   Correlations of atopic patient characteristics with mast cell expression of FcεRIα or mast cell–bound IgE
ParameterMast cell expression of FcεRIα or IgErsP-value
  1. *Correlation analysis performed on patients with allergic rhinitis + Asthma only.

  2. Correlation analysis was performed using Spearman rank correlation test. rs, Spearman rank correlation coefficient, P ≤ 0.05 is considered significant.

PD20 (μg)FcεRIα expression in bronchial biopsies (mm2)*−0.860.02
PD20 (μg)IgE expression in bronchial biopsies (%)−0.500.02
Br NO flux (ml/s)IgE expression in bronchial biopsies (%)0.76<0.0001
FeNO50 (ppb)IgE expression in bronchial biopsies (%)0.77<0.0001
Slope-FresMCh (Hz/μg)FcεRIα expression in alveolar biopsies (%)0.540.009

Discussion

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Conflict of interest
  8. Sources of support
  9. Funding
  10. Authors’ contributions
  11. References

Through the exploration of alveolar tissue in atopic asthma, in rhinitis, as well as in age-matched healthy controls, the present study shows that alveolar mast cells in patients with mild asthma, but not in patients with AR and BHR, have acquired a highly FcεRI- and IgE-expressing phenotype.

Allergen-induced and IgE-mediated mast cell degranulation is a central feature of all allergic diseases. It has been shown that the inflammatory response at bronchial, segmental allergen provocation is similar in patients with rhinitis without asthma symptoms and rhinitis patients with concurrent asthma. With the similar nature of many bronchial inflammatory components in asthma and rhinitis, it has been suggested that not only the type of inflammation but also the anatomic distribution is important in discriminating between asthma and other atopic or inflammatory diseases of the airways (18, 20–22). The present study extends that view and forwards the emergence of FcεRI expression on alveolar mast cells as a distinct and disease-specific feature of allergic asthma. The present and our recently published data (4) reveal that a high alveolar FcεRI and IgE expression is present in mild atopic asthma but not in patients with AR (with or without BHR), COPD, CF, IPF or respiratory infections [(4) and data not shown]. We recently showed that a high alveolar mast cell expression of FcεRI is also present in patients with uncontrolled atopic asthma (who virtually all also had AR) (4). Although the tissue handling differed from the present study, which precludes a direct comparison, the relative increase was higher in the uncontrolled asthmatics than for the mild asthmatics, indicating that alveolar mast cell expression of FcεRI relates to asthma severity.

A negative correlation of PD20 with the mast cell positive for FcεRI and mast cell–bound IgE in central airways was found in AR patients with or without BHR and asthma. This might suggest a mast cell involvement in the IgE-induced bronchial inflammation and remodelling, irrespective of airway hyper-responsiveness. Our study shows correlation between mast cell–bound IgE in the central airways and exhaled FeNO50 and bronchial NO flux, thus indicating that mast cells directly or indirectly are associated with inflammation in the central airways.

In the present study, we used IOS to monitor peripheral involvement, by measurements of peripheral resistance and reactance. We found no correlation between mast cell expression of FcεRI or mast cell–bound IgE and peripheral resistance. However, increased expression of FcεRI on alveolar mast cells correlated with increased reactivity (Slope-FresMCh) in the peripheral airways within the atopic patients. This could explain our previous observation indicating a difference in peripheral airway involvement between asthmatics and rhinitis patients, also in those with BHR (12).

Mast cells in most types of tissues, especially those facing the external environment, have high basal expression of FcεRI (1). Mast cells located in the alveolar parenchyma, however, have – under healthy conditions – a very low FcεRI expression (2). The discovery of an altered and highly FcεRI-expressing phenotype of the alveolar mast cells also in mild atopic asthma, but not in AR patients who have developed BHR but not clinical asthma, represents a major finding in the present study. This finding suggests that already in mild asthma, the alveolar parenchyma is more susceptible to an allergen-induced IgE-mediated peripheral inflammation. Interestingly, this phenomenon was not observed in AR patients and is thereby not a phenotype of atopic disorders per se. Whether increased FcεRI expression represents a gradual development of allergic disease severity as stated by the ‘link’ theory or represents some specific conditions in patients developing asthma could only be speculated on. Longitudinal studies are now warranted to explore to what extent the progression of rhinitis to asthma is related to the present type of changes in alveolar mast cell phenotype. The number of patients in our study was relatively low, and more extensive studies on larger patient groups are needed to fully understand the differences in peripheral inflammation between AR and atopic asthma.

It should be noted that alveolar mast cells might also contribute to a local immune orchestration by their capacity to release chemokines and cytokines through IgE-independent mechanisms (23). One such mediator class is cysteinyl leukotrienes (cys-LTs), and the mast cells are believed to be a major source for cys-LT production. This might partly explain our previous observation that asthmatics, compared to rhinitis patients with or without BHR, have increased levels of cysteinyl leukotrienes (cys-LTs) in sputum (24). Our data show that patients with mild asthma with an IgE-driven inflammation in the alveolar compartment may benefit from treatment strategies with improved targeting of the distal lung (25, 26).

In summary, the present study shows that in patients with atopic mild asthma, mast cells in the alveolar parenchyma have markedly increased expression of FcεRI compared to healthy nonatopic controls or patients with AR. Taken together, these observations significantly advances the concept of FcεRI-expressing alveolar mast cells as a novel disease-specific feature of allergic asthma and argue for an improved therapeutic targeting of the peripheral airways at early or mild stages of asthma.

Acknowledgments

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Conflict of interest
  8. Sources of support
  9. Funding
  10. Authors’ contributions
  11. References

We thank Karin Jansner and Britt-Marie Nilsson for skilful technical assistance with tissue processing and immunohistochemical staining.

Sources of support

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Conflict of interest
  8. Sources of support
  9. Funding
  10. Authors’ contributions
  11. References

The Heart & Lung Foundation, Sweden, The Swedish Medical Research Council, The Swedish Asthma and Allergy Associations Research Foundation, and The Crafoord Foundation.

Funding

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Conflict of interest
  8. Sources of support
  9. Funding
  10. Authors’ contributions
  11. References

This study was funded by the Swedish Research Council, the Heart and Lung Foundation and Evy and Gunnar Sandberg’s Foundation.

Authors’ contributions

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Conflict of interest
  8. Sources of support
  9. Funding
  10. Authors’ contributions
  11. References

CA contributed to the study design, data analysis, data collection, data interpretation, figures and writing of the manuscript. ET contributed to the study design, data analysis, data collection, data interpretation, figures and writing of the manuscript. DA contributed to data analysis, data collection and data interpretation. AB contributed to data interpretation and literature search. MM contributed to literature search and provided comments on the manuscript. LB supervised the study, collected the clinical data including all bronchoscopies performed and involved in interpreting data and writing the manuscript. JE supervised the study and contributed to study design, literature search, data interpretation and writing of the manuscript.

References

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Conflict of interest
  8. Sources of support
  9. Funding
  10. Authors’ contributions
  11. References