You are here
Persistent airflow obstruction in patients with asthma: Characteristics of a distinct clinical phenotype
Some patients with asthma present persistent airflow limitation but their clinical and inflammatory characteristics have not been extensively described. In this study we aimed to identify differences in the clinical, functional and inflammatory characteristics between patients with asthma with and without persistent airflow obstruction.
Patients (n = 170) were consecutively recruited from two tertiary Asthma Clinics. Patients' demographics, pulmonary function tests, inflammatory cells in induced sputum, bronchial hyperresponsiveness (BHR, PD15 to methacholine) and treatment regimens were recorded.
Sixty patients (35.3%) presented persistent airflow obstruction. Besides differences in lung function, patients with persistent obstruction presented, lower methacholine PD20, higher exhaled NO, and higher eosinophil and neutrophil counts in induced sputum. The majority (71.7%) of the patients with persistent obstruction fulfilled the ATS criteria for severe refractory asthma (SRA), in contrast to 4.5% in the group without persistent obstruction. A cluster analysis identified three clinically relevant clusters: Cluster 1 (n = 56, not related to persistent airflow obstruction) included non-atopic patients, who did not receive high-dose ICS without SRA; Cluster 2 (n = 53, related to persistent airflow obstruction) included atopic patients, receiving high-dose ICS and/or oral CS, fulfilling SRA criteria; Cluster 3 (n = 61, not related to persistent airflow obstruction) included atopic patients not receiving high-dose ICS, without SRA.
Asthma patients with persistent airflow obstruction present a distinct asthma phenotype, with significant differences in clinical, functional and inflammatory characteristics compared to patients without fixed airway obstruction. These patients present more often severe refractory asthma and require more intense treatment.
Keywords: Asthma, Phenotype, Persistent airflow obstruction, Airways inflammation, Cluster analysis.
Abbreviation List: ANOVA - analysis of variance, ATS - American Thoracic Society, BHR - bronchial hyperresponsiveness, BMI - body mass index, COPD - chronic obstructive pulmonary disease, FEV1 - forced expiratory volume in 1 s, FeNO - fraction of exhaled nitric oxide, FVC - forced vital capacity, DLCO - diffusing capacity for carbon monoxide, FRC - functional residual capacity, GINA - global initiative for asthma, ICS - inhaled corticosteroids, LABA - long-acting β2-agonists, LTRA - leukotriene receptor antagonists, SRA - severe refractory asthma.
Asthma is a clinical syndrome that is characterized by intermittent respiratory symptoms triggered by several stimuli, including viral infections and environmental allergens, and is associated with chronic airway inflammation and bronchial hyperresponsiveness  . Asthma, is a heterogeneous disorder which includes several different phenotypes characterized by differences in age of onset, aetiology, inflammatory profile and response to treatment  , as well as different endotypes characterized by distinct functional or pathophysiological mechanisms  . In the previous years, distinct asthma phenotypes and endotypes have been described, yet their characterization is based on a limited number of features and no specific methods or biomarkers have been identified as useful in this process  and .
Most asthmatic patients have mild or moderate asthma and can be controlled with low doses of anti-inflammatory drugs. However, a small subgroup of patients present with uncontrolled asthma, frequent exacerbations and rapid decline of lung function, despite avoidance of trigger factors, proper management of comorbidities and use of high-dose anti-inflammatory treatment. For such patients the term severe refractory asthma has been used  and . Moreover, some patients with asthma present persistent airflow limitation  . In a recent study, patients with severe asthma and persistent airflow obstruction were characterized by increased airway smooth muscle with ongoing TH1 and TH2 inflammatory responses, but these patients did not present specific characteristics on high-resolution computed tomographic scans or sputum analysis  . On the other hand, in severe asthma, more than 50% of the patients develop irreversible airway obstruction  . In terms of taxonomy, it is sometimes quite difficult to classify these patients properly, as the presence of persistent airflow obstruction may lead to a diagnosis of COPD, especially in smokers with characteristics of asthma and/or atopy  and . From the clinicians' point of view it is important to identify these patients in order to optimize their treatment, and especially to avoid overtreatment based solely on pulmonary function tests.
The aim of the present study was to evaluate the differences in clinical, functional and inflammatory characteristics between asthmatic patients with and without persistent airflow obstruction. We further used cluster analysis in order to identify and characterize different asthma subtypes based on the presence of persistent airflow obstruction.
2.1. Study population
In this study we have recruited 170 consecutive adult patients from an on-going cohort of asthmatic patients who are followed up in the asthma clinics of the1st and 2nd Respiratory Medicine University Departments in Athens. The health system in Greece is based on the assessment by a specialist. The patients were followed up in a tertiary clinic as part of their regular follow-up. Demographic data, smoking status, body mass index (BMI) and type of treatment were recorded for all patients. Diagnosis of asthma was based on the GINA guidelines(1), and early onset asthma was defined as the presence of asthma symptoms before the age of twelve years  . All asthmatic subjects were optimally treated for at least 6 months according to GINA guidelines(1) and were adherent to therapy. The latest was checked through the national insurance system. Exclusion criteria were a diagnosis of other respiratory disease, concomitant malignancy and severe heart, liver, renal or collagen disease. Patients with a respiratory tract infection or an asthma exacerbation in the past 8 weeks prior to admission were also excluded. The study was approved by the ethics committees of the Attikon & Sotiria hospitals (approval number 1915, September 2010) and all subjects provided written informed consent.
2.2. Definition of persistent airflow obstruction and severe asthma
Patients were assigned to the group with persistent airflow obstruction if they presented post-bronchodilation FEV1 values < 70% predicted at all visits, or a single value of post-bronchodilation FEV1 between 70% and 75% predicted during a 1-year follow up  . A FEV1/FVC ratio less than 75% was also a presupposition in order to define persistent airflow obstruction. All other subjects were considered to be able to achieve a normal or near normal FEV1 and were classified to the group without persistent airflow obstruction  . Patients were characterized as having severe refractory asthma (SRA) according to the ATS workshop consensus  .
2.3. Study measurements
2.3.1. Sputum induction and processing
Sputum induction was performed as previously described  and , using all the modifications for safe measurements according to the underlying asthma severity  and . The process had a total duration of 15 min, where subjects inhaled 3% saline at room temperature by an ultrasonic nebulizer (DeVilbiss Co., Heston, UK). Sputum supernatant was removed with centrifugation and total cell count was evaluated with a haemocytometer using Trypan Blue stain. Slides were prepared by cytospin (Shandon, Runcorn, UK) and were stained with May-Grunwald and Giemsa for differential cell counts. A blind observer performed counting of a minimum 500 inflammatory cells in each sample. Sputum samples were considered acceptable if they had a volume of at least 2 ml after final expectoration and the percentage of squamous cells on the prepared slides was <10%. Total cell count was expressed as the number of cells × 106/ml and sputum inflammatory cells as percentage (%) of non-squamous cells.
2.3.2. Lung function
Forced expiratory volume in 1 s FEV1were measured using Master Screen Body (Viasys Healthcare, Jaeger, Hoechberg, Germany) according to the American Thoracic Society guidelines  . Bronchial hyperresponsiveness (BHR) was measured as PD20 to methacholine using a commercially available system (APS; Viasys Healthcare, Jaeger, Hoechberg, Germany) according to the ATS guidelines  .
2.3.3. Measurement of exhaled nitric oxide
The fraction of exhaled NO (FeNO) was measured using a portable NO analyser (NIOX MINO, Aerocrine, Solna, Sweden)  .
2.3.4. Characterization of atopic status
Atopic status was confirmed by a positive skin prick test to any of twenty common aeroallergens.
2.4. Statistical analysis
Categorical variables are presented as n (%), whereas numerical variables are presented as mean ± standard deviation (SD) or median (interquartile ranges) for normally distributed and skewed data, respectively. Normality of distributions was checked with Kolmogorov–Smirnov test. Comparisons between patients with fixed and non-fixed airway obstruction were performed using chi-square tests for categorical data, as well as unpaired t-tests or Mann–Whitney U-tests for normally distributed or skewed numerical data, respectively.
A two-step cluster analysis was performed in order to classify patients in asthma subtypes. Uniform cluster analysis methodology was applied to each population using a two-step approach as previously described  . In the ﬁrst step, hierarchical cluster analysis using Ward's method generated a dendrogram for estimation of the number of likely clusters within the studied population. This estimate was prespeciﬁed in a k-means cluster analysis that was used as the principal clustering technique. Variables chosen for cluster modelling were selected on the basis of their considered contribution to characterizing the asthma phenotype. All variables were dichotomized and a number of four clusters was pre-selected, based on previous publications. Parameters that were used to characterize asthma clusters were use of high-dose ICS, use of oral corticosteroids, atopic status and fulfilment of criteria for SRA. Persistent airflow obstruction was used as an evaluation field to the formed clusters. Between clusters comparisons of baseline parameters were performed using one-way analysis of variance (ANOVA) for parametric variables, the chi-square test for proportions, and Kruskal–Wallis for nonparametric variables. Data were analysed using SPSS 18.0 for Windows (SPSS Inc, Chicago, IL, USA) and p-values <0.05 were considered statistically significant.
One hundred seventy asthmatic patients were included in the study and 60 of them (35.3%) presented persistent airflow obstruction. Demographic and functional characteristics of the subjects according to the presence of airflow obstruction are summarized in Table 1 . It should be noted that PD20 was measured only in 18/60 asthmatic patients with persistent airflow obstruction and in 92/110 without persistent obstruction.
|Variables||Persistent airflow obstruction n = 60||Non-persistent airflow obstruction n = 110||p-value|
|Age (ys)||53.9 ± 10.9||50.2 ± 14.1||0.140|
|Gender (F)||42 (70.0%)||75 (68.2%)||0.864|
|Disease duration (y)||31.1 ± 13.7||23.8 ± 13.4|
|Current smokers||9 (15.0%)||33 (30.0%)||0.071|
|Pack-Years||11.9 ± 24.6||12.6 ± 23.6||0.270|
|PB FEV1 (% pred)||58.7 ± 10||90.1 ± 14.7|
|PB FVC (%pred)||77.1 ± 12.0||99.0 ± 14.7|
|PB FEV1/FVC||60.3 ± 6.4||81.6 ± 6.9|
|Obesity||19 (31.9%)||28 (25.5%)||0.473|
|BMI (kg/m2)||28.2 ± 4.9||27.4 ± 4.6||0.245|
|Late onset of symptoms||8 (13.3%)||25 (22.7%)||0.160|
|Atopy||32 (53.3%)||58 (52.7)||1.000|
|SRA||43 (71.7%)||5 (4.5%)|
|PD20 (mg)||0.08 (0.05, 0.12)||0.20 (0.09, 0.33)|
FeNO ≥20 ppb
|38.3 ± 32.3
|28.3 ± 29.1
|Cells (x106/ml)||2.88 (1.4, 4.78)||2.20 (2.20, 3.73)||0.082|
|Eosinophils (%)||4 (1, 14)||2 (0, 6)|
|Neutrophils (%)||39 (23, 47)||31 (19, 40)|
|Eosinophils ≥3%||37 (61.7%)||49 (44.5%)|
|Neutrophils ≥61%||5 (8.3%)||9 (8.2%)||0.377|
|High ICS||43 (71.7%)||7 (6.4%)|
|Oral CS||31 (51.7%)||3 (2.7%)|
|Omalizumab||10 (16.7%)||1 (0.9%)|
|LABAs||59 (98.3%)||89 (80.7%)|
|LTRAs||19 (31.7%)||7 (6.4%)|
Abbreviations: PB FEV1: Post-bronchodilation forced expiratory volume in 1 s, PB FVC: Post-bronchodilation Forced exhaled vital capacity, BMI: Body mass index, SRA: Severe Refractory asthma, FeNO: Fraction of exhaled NO, ICS: inhaled corticosteroids, CS: corticosteroids, LABAs: Long acting Beta agonists, LTRAs: Leukotriene receptors antagonists.
Bold letters indicate statistical significant differences.
Patients with persistent airflow obstruction had a significantly longer disease duration, lower FEV1, FVC and FEV1/FVC ratio compared to patients without persistent airflow obstruction. Patients with persistent airflow obstruction also presented lower PD20 values and higher FeNO levels, as well as increased eosinophil and neutrophil counts in induced sputum. Interestingly, 61.7% of the patients with persistent airflow obstruction had significant sputum eosinophilia, as expressed by the presence of ≥3% eosinophils in induced sputum. In addition, more subjects with persistent airflow obstruction were treated with high-dose ICS, LABAs, oral corticosteroids, omalizumab, and LTRAs. Finally, 71.7% of patients with persistent airflow obstruction fulfilled the criteria for SRA in contrast to only 4.5% of patients without persistent airflow obstruction.
3.1. Cluster analysis
In the cluster analysis, parameters that were found significant to be included in the different clusters of our population were treatment with high doses of ICS, treatment with oral corticosteroids, atopic status and fulfilment of criteria for SRA. Three clusters were identified. The characteristics of the three clusters and their relation to the presence of persistent airflow obstruction are shown in Table 2 . Briefly:
- Cluster 1 (n = 56 patients, 32.9%): was not related to persistent airflow obstruction, and included non-atopic patients, who did not receive high-dose ICS or oral corticosteroids (CS), and did not fulfil the criteria for SRA;
- Cluster 2 (n = 53 patients, 31.2%): was related to persistent airflow obstruction, included atopic patients, receiving high-dose ICS and oral CS, who fulfilled criteria for SRA; and
- Cluster 3 (n = 61 patients, 35.9%): was not related to persistent airflow obstruction, and included atopic patients not receiving high-dose ICS or oral CS, and did not fulfil criteria for SRA.
|Cluster||Cluster 1||Cluster 2||Cluster 3|
|Size (%)||56 (32.9%)||53 (31.2%)||61 (35.9%)|
|Features||Low dose ICS
|High dose ICS
|Low dose ICS
|No criteria for SRA
|Criteria for SRA
|No criteria for SRA (100%)|
|No oral CS
|No oral CS
|No persistent airflow obstruction
|Persistent airflow obstruction
|No persistent airflow obstruction
Abbreviations: ICS: Inhaled corticosteroids, SRA: Severe refractory asthma, CS: corticosteroids.
3.2. Comparisons between the patients of the 3 clusters
Between clusters comparisons of baseline parameters are shown in Table 3 . No differences were observed between the clusters regarding the patients' age, gender smoking habit and BMI. However, patients in cluster 2 which was related to persistent airflow obstruction had longer disease duration, more impaired lung function, higher values of FeNO, higher bronchial hyperresponsiveness, and higher eosinophil and neutrophil counts in induced sputum.
N = 56
N = 53
N = 61
|Age (ys)||51 ± 12.4||52.5 ± 12.6||50.8 ± 14.4||0.833|
|Gender (F)||36 (64.3%)||39 (73.6%)||42 (68.9%)||0.578|
|Disease duration (years)||24.2 ± 14||30.4 ± 13.6||24.9 ± 13.7|
|Pack-Years||11.8 ± 24.3||8.5 ± 19.0||16.2 ± 25.2||0.113|
|PB FEV1 (%pred)||84.3 ± 19.0||61.4 ± 16.2||84.9 ± 16.0|
|PB FVC (%pred)||97.3 ± 18.5||80.8 ± 14.6||94.0 ± 21.5|
|FEV1/FVC||71.9 ± 9.9||62.1 ± 9.1||72.6 ± 7.9|
|FeNO (ppb)||28.4 ± 23.4||43.5 ± 39.4||24.9 ± 24.3|
|BMI (kg/m2)||28.2 ± 4.5||27.1 ± 4.6||27.7 ± 5.1||0.515|
|PD20 (mg)||0.24 ± 0.16||0.13 ± 0.12||0.20 ± 0.14|
|Total sputum cells (x10*6/ml)||2.7 ± 0.90||4.0 ± 4.5||2.6 ± 2.3||0.221|
|Eosinophils (%)||4.9 ± 10.8||9.0 ± 11.0||4.3 ± 5.7|
|Neutrophils (%)||32.8 ± 19.4||38.6 ± 17.6||31.4 ± 17.1|
|Persistent airflow obstruction||8 (14.3%)||44 (83%)||8 (13.1%)|
|Severe refractory asthma||0 (0%)||48 (90.6%)||0 (0%)|
|High ICS||0 (0%)||50 (94.3%)||0 (0%)|
|Oral CS||0 (0%)||34 (64.2%)||0 (0%)|
|Omalizumab||0 (0%)||11 (20.8%)||0 (0%)|
|LABAs||45 (80.4%)||53 (100%)||50 (82%)|
|LTRAs||4 (7.1%)||17 (32.1%)||5 (8.2%)|
Abbreviations: PB FEV1: Post-bronchodilation forced expiratory volume in 1 s, PB FVC: Post-bronchodilation forced exhaled vital capacity, BMI: Body mass index, SRA: Severe Refractory asthma, FeNO: Fraction of exhaled Nitric oxide, ICS: inhaled corticosteroids, CS: corticosteroids, LABAs: Long acting Beta agonists, LTRAs: Leukotriene receptors antagonists.
Bold letters indicate statistical significant differences.
In the present study we evaluated clinical, functional and inflammatory differences between asthmatic patients with and without persistent airway obstruction. Patients with persistent airflow obstruction had longer disease duration, more impaired lung function (lower FEV1, FVC, FEV1/FVC ratio), increased bronchial hyperresponsiveness, higher FeNO levels, as well as increased eosinophil and neutrophil counts in induced sputum, and received more intense treatment. An important observation is that the vast majority of patients with persistent airway obstruction (71.1%) fulfilled the criteria for SRA in contrast to a small minority of 4.5% in the non-persistent airflow obstruction group. Cluster analysis of our population identified three clusters, one of whom included atopic patients, receiving high-dose ICS and oral CS, who fulfilled criteria for SRA and was associated to persistent airway obstruction. Despite the fact that these patients presented greater bronchial hyperresponsiveness, higher values of FeNO and sputum eosinophil and neutrophil counts, none of these inflammatory markers was important in cluster formation. The fact that BHR was measured in a limited number of asthmatic subjects, particularly those with persistent airflow obstruction, requires caution in the interpretation of BHR as a determinant factor of cluster analysis.
Severe refractory asthma is a hot topic in current literature due to the great heterogeneity and challenging treatment, whereas it remains a major burden for the health care system. The pathogenetic underlying mechanisms in a number of severe asthma phenotypes are not well defined yet. Recent studies have tried to elucidate the pathways leading to irreversible airway obstruction, which seems to characterize a distinct subtype of severe refractory asthma. Kaminska et al. examined a population of severe asthma patients and showed that patients with severe asthma and persistent airway obstruction had earlier disease onset, longer disease duration and increased inflammation with combined sputum eosinophilia and neutrophilia, and greater smooth muscle area, despite treatment with high-dose inhaled or even oral corticosteroids  . In the present study we used the same definition of persistent airflow obstruction but we included asthmatic patients of different severity. The second cluster of our analysis presented significant similarities with the population of the Kaminska study, as this population presented persistent airway obstruction, included more severe asthmatic patients, with greater functional impairment and airways inflammation, despite more intense treatment. We defined persistent airflow obstruction as a post-bronchodilation FEV1<70% and one can argue that this FEV1 is not the maximal attainable FEV1 since we did not include an oral steroid trial in our study. However, the patients were optimally treated (as defined in the methods section) and lung function was measured during a clinically stable period. Therefore, we believe that the post-bronchodilation FEV1 for our patients closely approached the maximal attainable FEV1. Still, we cannot exclude the possibility that a longer course of steroids might have improved further our patients FEV1  .
Several previous studies have defined clinical phenotypes by use of cluster analysis. In the first cluster analysis of severe asthma undertaken in a single-centre study in Leicester, UK, Haldar et al. described four clusters defined as ‘early-onset atopic’, ‘benign’ and ‘obese, non-eosinophilic’ asthma, as well as two severe refractory asthma clusters described as ‘early symptom predominant’ and ‘inflammation predominant’  . Subsequently, the US Severe Asthma Research Program (SARP) identified five clusters of different severity, based on analysis of 726 subjects: ‘early onset atopic asthma with normal lung function on few controller medications’, ‘early-onset atopic asthma and preserved lung function but increased medication requirements’, ‘older obese women with late-onset non-atopic asthma, moderate airflow obstruction, and frequent oral corticosteroid use for exacerbations’, ‘severe airflow obstruction with significant bronchodilator reversibility’ and ‘severe with fixed airflow obstruction’  . A multi-centre Korean study described four asthma clusters that were characterized as ‘smoking’, ‘severe obstructive’, ‘early-onset atopic’ and ‘late-onset mild’ asthma group  . Finally, an observational study of the British Thoracic Society (BTS) Severe Refractory Asthma Registry identified five clusters of an asthmatic population in the UK (n = 349, four centres): ‘early-onset atopic’, ‘late-onset obese’, ‘least severe disease’, ‘late-onset eosinophilic’, and ‘significant severe airflow obstruction’  . The second cluster of our population that included patients with persistent airway obstruction and mostly severe refractory asthma, resemble to the severe asthma and fixed airflow cluster of the SARP cohort  . Additionally, in both UK studies  and  and the Korean study  a ‘severe-obstructive’ subtype is also identified. All these data combined further support that asthmatics with persistent airflow obstruction represent a specific group with more severe disease requiring more intense treatment. Whether persistent airway obstruction reflects the additive effect of a time-dependent inflammatory process or it is an expression of a distinct asthma subtype remains to be elucidated. However we cannot exclude that some other factors may contribute to the presence of persistent airflow obstruction. These factors may be attributed to the presence of persistent air-trapping  or/and to domestic exposure to molds, hospitalization during the last year, and very frequent exacerbations  .
The first cluster of the present study included non-atopic patients with mild-to moderate disease that do not require high doses of inhaled or oral corticosteroids. This group shares similarities with the ‘benign’ type described by Haldar  , as well as with the ‘late onset non atopic’ in the SARP  , the ‘late onset mild’ of the Korean population  and the ‘least severe disease’ of the recent study of the BTS Severe Refractory Asthma Registry  . The third cluster of our study, including atopic subjects with mild-to moderate disease, seems to be similar to the ‘mild-to moderate atopic’ type of the SARP cohort and the ‘atopic’ and ‘early onset atopic’ of the UK studies. An ‘early onset atopic’ cluster was also identified in the Korean study. However, it is possible that due to somewhat low number of patients compared to some of the previous studies  , we were able to identify only 3 clusters in our study population.
BHR was not a major determinant for the identification of persistent airflow obstruction in our study. However a significant number of data is missing for the patients with the lowest FEV1 values. This was in accordance with the international regulations for the assessment of BHR  . However, it may lead to an underestimation of the association between BHR and persistent airway obstruction. In a previous study where BHR was assessed as a determinant of persistent airflow limitation an Odds ratio 3.9 was found. However, the above study recruited only severe asthmatics  . Age of onset was not an important determinant in our cluster analysis in contrast to the other studies, even though patients of the first and third cluster had shorter disease duration compared to those with persistent airflow obstruction of cluster 2. BMI was another characteristic that was not found to be important in cluster formation in contrast to the other studies, with the exception of the Korean study  . In addition, smoking status did not present significant differences among the clusters, even though current smokers were not excluded. Only in the Korean study a subtype annotated as ‘smoking asthma’ has been identified. Differences in the parameters defining the different clusters may be related to differences in populations and this demands further analysis to evaluate its possible clinical significance. Furthermore, if we consider that the mean age of our patients was 50 years old we might speculate that the cluster analysis may be different in a younger population. Finally we should take into consideration the possible presence of asthma –COPD overlap syndrome (ACOS). A syndrome where persistent airflow obstruction may represents a major hallmark with a possible underlying mechanism that of a pro-inflammatory and protelolytic mechanism  . We tried to overcome the above limitation by the fact that a detail assessment of patients was performed in order to exclude COPD.
In conclusion, the results of this study demonstrate that patients with persistent airway obstruction present significant differences in clinical, functional and inflammatory characteristics compared to patients without fixed airway obstruction. A cluster of patients with persistent airflow obstruction was identified and these patients presented more often severe refractory asthma and required more intense treatment. Further longitudinal studies are necessary in order to determine possible pathogenetic mechanisms and the natural history of persistent airflow obstruction in asthma that may lead to more effective interventional strategies in these patients.
KK is the guarantor of the content of the manuscript, including the data and analysis. SL, KK and PB were involved in the study conception and design. EK, GP, AP and GH were involved in the recruitment of patients, samples handling and data collection. AIP, EK and KK drafted the manuscript. SP, NK, PB, SL, and KK reviewed the manuscript and provided important scientific input.
Conflict of interest
No conflicts exist for the specified authors.
-  Global Initiative for Asthma. GINA Workshop Report, Global Strategy for Asthma Management and Prevention. (, 2013) http://www.ginasthma.org Revised 2013, available on
-  S.E. Wenzel. Asthma: defining of the persistent adult phenotypes. Lancet. 2006;368(9537):804-813
-  G.P. Anderson. Endotyping asthma: new insights into key pathogenic mechanisms in a complex, heterogeneous disease. Lancet. 2008;372(9643):1107-1119
-  T.B. Kim, A.S. Jang, H.S. Kwon, J.S. Park, Y.S. Chang, S.H. Cho, et al. Identification of asthma clusters in two independent Korean adult asthma cohorts. Eur. Respir. J.. 2013;41(6):1308-1314
-  V. Siroux, X. Basagana, A. Boudier, I. Pin, J. Garcia-Aymerich, A. Vesin, et al. Identifying adult asthma phenotypes using a clustering approach. Eur. Respir. J.. 2011;38(2):310-317
-  Proceedings of the ATS workshop on refractory asthma: current understanding, recommendations, and unanswered questions. Am. Thoracic Soc. Am. J. Respir. Crit. Care Med.. 2000;162(6):2341-2351
-  E.H. Bel, A. Sousa, L. Fleming, A. Bush, K.F. Chung, J. Versnel, et al. Diagnosis and definition of severe refractory asthma: an international consensus statement from the Innovative Medicine Initiative (IMI). Thorax. 2011;66(10):910-917
-  M. Kaminska, S. Foley, K. Maghni, C. Storness-Bliss, H. Coxson, H. Ghezzo, et al. Airway remodeling in subjects with severe asthma with or without chronic persistent airflow obstruction. J. Allergy Clin. Immunol.. 2009;124(1):45-51 e1–4
-  J.H. Lee, T. Haselkorn, L. Borish, L. Rasouliyan, B.E. Chipps, S.E. Wenzel. Risk factors associated with persistent airflow limitation in severe or difficult-to-treat asthma: insights from the TENOR study. Chest. 2007;132(6):1882-1889
-  P. Lange. Persistent airway obstruction in asthma. Am. J. Respir. Crit. Care Med.. 2013;187(1):1-2
-  J.L. Perret, S.C. Dharmage, M.C. Matheson, D.P. Johns, L.C. Gurrin, J.A. Burgess, et al. The interplay between the effects of lifetime asthma, smoking, and atopy on fixed airflow obstruction in middle age. Am. J. Respir. Crit. Care Med.. 2013;187(1):42-48
-  E. Tseliou, P. Bakakos, K. Kostikas, G. Hillas, K. Mantzouranis, P. Emmanouil, et al. Increased levels of angiopoietins 1 and 2 in sputum supernatant in severe refractory asthma. Allergy. 2012;67(3):396-402
-  A. Efthimiadis, A. Spanevello, Q. Hamid, M.M. Kelly, M. Linden, R. Louis, et al. Methods of sputum processing for cell counts, immunocytochemistry and in situ hybridisation. Eur. Respir. J. Suppl.. 2002;37:19s-23s
-  J.C. Kips, J.V. Fahy, F.E. Hargreave, P.W. Ind, J.C. in't Veen. Methods for sputum induction and analysis of induced sputum: a method for assessing airway inflammation in asthma. Eur. Respir. J. Suppl.. 1998;26:9S-12S
-  D.C. Grootendorst, J.W. van den Bos, J.J. Romeijn, M. Veselic-Charvat, E.J. Duiverman, E.J. Vrijlandt, et al. Induced sputum in adolescents with severe stable asthma. Safety and the relationship of cell counts and eosinophil cationic protein to clinical severity. Eur. Respir. J.. 1999;13(3):647-653
-  Standardization of Spirometry, 1994 update. Am. Thoracic Soc. Am. J. Respir. Crit. Care Med.. 1995;152(3):1107-1136
-  R.O. Crapo, R. Casaburi, A.L. Coates, P.L. Enright, J.L. Hankinson, C.G. Irvin, et al. Guidelines for methacholine and exercise challenge testing-1999. This official statement of the American Thoracic Society was adopted by the ATS Board of Directors, July 1999. Am. J. Respir. Crit. Care Med.. 2000;161(1):309-329
-  ATS/ERS recommendations for standardized procedures for the online and offline measurement of exhaled lower respiratory nitric oxide and nasal nitric oxide, 2005. Am. J. Respir. Crit. Care Med.. 2005;171(8):912-930
-  P. Haldar, I.D. Pavord, D.E. Shaw, M.A. Berry, M. Thomas, C.E. Brightling, et al. Cluster analysis and clinical asthma phenotypes. Am. J. Respir. Crit. Care Med.. 2008;178(3):218-224
-  A. Ten Brinke, A.H. Zwinderman, P.J. Sterk, K.F. Rabe, E.H. Bel. Factors associated with persistent airflow limitation in severe asthma. Am. J. Respir. Crit. Care Med.. 2001;164(5):744-748
-  W.C. Moore, D.A. Meyers, S.E. Wenzel, W.G. Teague, H. Li, X. Li, et al. Identification of asthma phenotypes using cluster analysis in the severe asthma research program. Am. J. Respir. Crit. Care Med.. 2010;181(4):315-323
-  C. Newby, L.G. Heaney, A. Menzies-Gow, R.M. Niven, A. Mansur, C. Bucknall, et al. Statistical cluster analysis of the British thoracic society severe refractory asthma registry: clinical outcomes and phenotype stability. PLoS One. 2014;9(7):e102987
-  S.W. Park, J.S. Park, S.H. Jeong, Y.N. Lee, Y. Hwangbo, J.H. Lee, et al. Air trapping is a major determinant of persistent airway obstruction in asthmatics. Respir. Med.. 2012;106(6):786-793
-  M. Ciebiada, M. Domagala, M. Gorska-Ciebiada, P. Gorski. Risk factors associated with irreversible airway obstruction in nonsmoking adult patients with severe asthma. Allergy Asthma Proc.. 2014;35(5):72-79
-  A.F. Gelb, A. Yamamoto, E.K. Verbeken, J.A. Nadel. Unraveling the pathophysiology of the asthma-COPD overlap syndrome: unsuspected mild centrilobular emphysema is responsible for loss of lung elastic recoil in never smokers with asthma with persistent expiratory airflow limitation. Chest. 2015;148(2):313-320
a 1st Respiratory Medicine Department, University of Athens Medical School, Athens, Greece
b 2nd Respiratory Medicine Department, University of Athens Medical School, Athens, Greece
∗ Corresponding author. 2nd Respiratory Medicine Department, University of Athens Medical School, Attikon Hospital, Rimini 1, 12562 Athens, Greece.
☆ An abstract of this study has been presented in the European Respiratory Society Congress in Barcelona 2013 (Oral Presentation 3040).
© 2015 Elsevier Ltd, All rights reserved.