Jae-Sung Choi and An Soo Jang contributed equally as first authors.
Role of neutrophils in persistent airway obstruction due to refractory asthma
Article first published online: 24 JAN 2012
© 2011 The Authors. Respirology © 2011 Asian Pacific Society of Respirology
Volume 17, Issue 2, pages 322–329, February 2012
How to Cite
CHOI, J.-S., JANG, A. S., PARK, J. S., PARK, S. W., PAIK, S. H., PARK, J. S., UH, S.-T., KIM, Y.-H. and PARK, C. S. (2012), Role of neutrophils in persistent airway obstruction due to refractory asthma. Respirology, 17: 322–329. doi: 10.1111/j.1440-1843.2011.02097.x
- Issue published online: 24 JAN 2012
- Article first published online: 24 JAN 2012
- Accepted manuscript online: 31 OCT 2011 11:52PM EST
- Received 19 April 2011; invited to revise 30 May 2011, 14 August 2011; revised 27 July 2011, 24 August 2011; accepted 5 September 2011 (Associate Editor: Shu Hashimoto).
- chronic airflow obstruction;
- refractory asthma;
Background and objective: One of the clinical manifestations of refractory asthma (RA) in a certain group of patients is persistent airway obstruction (PAO), despite treatment with high doses of inhaled and/or systemic corticosteroids. Airway neutrophilic inflammation is frequently observed in RA; however, the relationship between neutrophilic inflammation and PAO has not been evaluated in this group of patients. The aim of this study was to compare the clinical parameters and patterns of inflammatory cells between patients with or without PAO due to RA, and to identify the factors associated with PAO.
Methods: Seventy-seven patients with RA were recruited from a cohort of 2298 asthmatic patients. Sputum differential cell counts were performed at initial presentation. Clinical and physiological parameters were compared between patients with (n = 19) or without PAO (n = 58).
Results: The group with PAO had a longer duration of asthma and a higher frequency of near-fatal asthma than the non-PAO group, although higher doses of inhaled corticosteroids were used in the PAO group (P = 0.037). Neutrophilic inflammation was predominant in the group with PAO, whereas eosinophilic inflammation was predominant in the non-PAO group (P = 0.003). When both groups were stratified according to smoking status, the non-smoking PAO group had the longest duration of asthma, with early onset of asthma (P < 0.05). The non-smoking PAO group tended to have the highest percentage of sputum neutrophils. Irrespective of smoking status, the percentage of sputum eosinophils was significantly higher in the non-PAO group than in the PAO group.
Conclusions: Patients with PAO due to RA show different clinical manifestations when compared with those without PAO and have neutrophil-dominant airway inflammation.
Asthma is a heterogeneous disease with subgroups defined by aetiology, pathology, severity, physiological parameters and response to treatment.1 Refractory asthma (RA) is a specific phenotype of asthma,2 which comprises less than 5% of all asthma (unpubl. obs.).
This subgroup is characterized by recurrent exacerbations and often persistent airway obstruction, despite use of high doses of medication.2 The clinical manifestations of RA are diverse.3,4 Prolonged treatment with high doses of inhaled and/or systemic corticosteroid relieve airflow obstruction in most asthmatic patients. However, some patients with RA continue to show persistent reduction in airway calibre despite long-term treatment with higher-than-recommended doses of inhaled corticosteroids (ICS) or oral corticosteroids.2 In contrast, some patients with RA, who have a relatively normal FEV1 at baseline but make one or more urgent asthma-related visits per year and require three or more ‘bursts’ of oral corticosteroid per year, are referred to as having brittle asthma.5 The major difference between these two subphenotyes is normal or abnormal baseline airflow, including FEV1 and PEF, when their asthma is stable.
Previous studies that used bronchoscopic biopsies to examine the immunopathology of patients with mild asthma suggested that eosinophilic airway inflammation was the characteristic abnormality in both atopic and non-atopic asthma.6–8 However, other pathology studies suggest that neutrophils are present in higher numbers in the airways of patients with severe asthma compared with those with mild asthma or normal control subjects.9–11 In addition, airway neutrophil numbers are increased in patients who die in status asthmaticus.12
Based on recent studies of subjects with well-characterized asthma, eosinophilic asthma has been defined as a distinct phenotype of asthma that is associated with good pharmacological responsiveness to corticosteroids. In contrast, patients with non-eosinophilic asthma, who represent a sizeable subgroup that includes patients with severe disease, appear to be relatively resistant to corticosteroid therapy.13 Thus, it might be expected that neutrophilic airway inflammation would be associated with more severely remodelled airways, as reflected by persistent airway obstruction (PAO) on lung function testing, whereas eosinophilic airway inflammation might be expected to be associated with reversible airway obstruction. However, there have been few reports on the patterns of inflammation in patients with RA in terms of PAO. Bumbacea and collaborators reported that a group of asthmatic patients with PAO showed increased exhaled nitric oxide concentrations and peripheral blood eosinophil numbers compared with those without PAO.14 In addition, Kaminska and co-workers compared clinical and laboratory markers in 16 patients with PAO and 18 without PAO, and showed that PAO was associated with earlier age of onset of asthma, longer duration of disease and more inflammatory cells in sputum but not with the percentage of neutrophils in sputum.15
In the present study, the inflammatory pattern in the airways of patients with RA was analysed in terms of response to high doses of ICS. Clinical and physiological parameters were compared between patients with the two different inflammatory patterns to identify the factors, including percentage of neutrophils in sputum, that were associated with PAO in patients with RA.
Clinical data on 2298 asthma patients, who were registered in an asthma cohort at the Genome Research Center for Allergy and Respiratory Diseases in Korea, was analysed retrospectively. All patients were recruited from three tertiary hospitals (Soonchunhyang University Bucheon Hospital, Seoul Hospital and Cheonan Hospital). The diagnosis of asthma was based on the Global Initiative for Asthma (GINA) guidelines.16 All subjects had a clinical diagnosis of asthma that was supported by one or more of the following criteria: (i) variability in maximum diurnal PEF of more than 20% over the course of 14 days; (ii) an increase in FEV1 of more than 15% after inhalation of 200–400 µg of albuterol; or (iii) a methacholine PC20 of <10 mg/mL. All patients underwent standardized assessment, which included analysis of induced sputum, full blood count with differential cell counts, measurement of serum IgE, posteroanterior CXR, skin prick tests with allergens, and spirometry. Twenty-four common inhalant allergens (Bencard Co., Brentford, UK), including dust mite (Dermatophagoides farinae and D. pteronyssinus), cat fur, dog fur, cockroach, grass and tree pollens, ragweed, and Aspergillus species, were used for skin-prick testing. Atopy was defined as a wheal reaction to allergen equal to or greater than that for histamine (1 mg/mL) or ≥3 mm in diameter.
Among the patients who had received regular follow-up over 2 years or longer, 98 were diagnosed as having RA according to the American Thoracic Society criteria.2 Patients were included in the RA group if they met one or both major criteria and one minor criterion as follows: requirement for daily controller medication in addition to ICS, for example, long-acting β2-agonist (LABA), theophylline or leukotriene receptor antagonist. Patients with PAO met at least the minor criterion of chronic PAO (post-bronchodilator FEV1 <80% of predicted) even when their asthma was stable, despite use of high doses of ICS and additional asthma medications. Patients with brittle asthma were defined as those who made one or more urgent asthma-related visits per year or required three or more ‘bursts’ of oral corticosteroid per year but with a FEV1 >80% of predicted and diurnal PEF variability of <20% when their asthma was stable. In the present study, they are referred to as ‘the patients without PAO’. Patients with near-fatal asthma (NFA) were defined as those who were admitted to the emergency unit or intensive care unit with acute respiratory failure and met the following criteria: hypercapnia (PaCO2 ≥ 45 mm Hg), signs of hypoxia (cyanosis or PaO2 ≤ 60 mm Hg) and a duration of exacerbation of <7 days.17 The exclusion criteria were respiratory infections at the time of sputum induction, COPD, vocal cord dysfunction, OSA, Churg–Strauss syndrome, cardiac dysfunction, allergic bronchopulmonary aspergillosis and poor adherence to treatment. The study protocol was approved by the local research ethics committee of the Soonchunhyang University Hospital research board.
At the baseline visit, demographic information was collected from all subjects, and spirometry was performed before and after inhalation of a bronchodilator. Baseline FVC and FEV1 were measured when the patient had not used a bronchodilator within the previous 8 h. Baseline and post-bronchodilator FEV1, FVC, FEF25-75% and DLCO were measured. FVC and FEV1 were measured at regular intervals of 1 or 2 months. A Vmax Series 2130 Autobox Spirometer (Sensor Medics, Yorba Linda, CA, USA) was used after calibration every morning at 8 am.
All patients with RA received high-dose ICS plus a LABA, as recommended by the GINA guidelines.16,18 To achieve control or partial control of symptoms, montelukast, theophylline (or doxophylline) and increasing doses of ICS or systemic corticosteroid were prescribed, as required. Exacerbations during treatment were managed in accordance with standard clinical guidelines.16,18 During the follow-up period, patients were asked to record symptom scores and medication use, and were switched to a systemic corticosteroid when they experienced exacerbations, that is, reductions in FEV1 of >20%, or aggravation of symptoms. Systemic corticosteroids, including oral prednisolone, were prescribed at a dose of 1–2 mg/kg per day. The best FEV1 was taken as the highest value when asthma was stable.
Sputum samples for differential cell counts were obtained when the patient's asthma was stable. Sputum was induced using isotonic saline that contained a short-acting bronchodilator, and the samples were processed within 2 h of collection, as previously described.19 Briefly, all portions with visibly greater solidity were carefully selected and placed in a preweighed Eppendorf tube. The samples were treated by adding eight volumes of 0.05% dithiothreitol (Sputolysin, Calbiochem Corp., San Diego, CA, USA) in Dulbecco's PBS. One volume of protease inhibitors (0.1 M ethylenediaminetetraacetic acid and 2 mg/mL phenylmethylsulfonyl fluoride) was added to 100 volumes of homogenized sputum, and the total cell count was determined using a haemocytometer. The sputum cells were collected by cytocentrifugation, and for each slide, 500 cells were examined after staining with Diff-Quick (American Scientific Products, Chicago, IL, USA). Based on the percentages of eosinophils and neutrophils in sputum, subjects were categorized into four inflammatory subtypes, with some modifications from previous studies.20 The eosinophilic type was defined as sputum eosinophils >3.0%, the neutrophilic type as sputum neutrophils >70%, the mixed granulocytic type as eosinophils >3.0% and neutrophils >70%, and the paucigranulocytic type as eosinophils <3.0% and neutrophils <70%.
The data were entered into a statistical software package (SPSS version 10.0, SPSS Inc., Chicago, IL, USA). Data are expressed as means ± SD or SEM. Differences between the RA patients with or without PAO were compared using the two-sample t-test, Mann–Whitney test or Pearson's chi-square test for normally distributed, skewed or categorical data, respectively. Differences among RA patients with or without PAO according to smoking status were compared using anova. Differences in the proportions of patients were analysed using chi-square tests, or Fisher's exact test when expected cell counts were low. A P value <0.05 was considered statistically significant.
Demographic parameters for patients with RA, with or without chronic PAO
A total of 2298 asthmatic patients were screened, and 98 were selected as having RA. Among these, 77 patients provided an adequate sputum sample at the initial visit. The baseline characteristics of these 77 patients are shown in Table 1. All asthma patients were adults older than 20 years at the initial visit. The duration of asthma was much longer in patients with PAO than in those without PAO group (P = 0.036). There were no differences in gender distribution or BMI between the two groups. The frequency of current smokers plus ex-smokers was similar, and there were no differences between the two groups in terms of pack years of smoking. All patients were taking high doses of regular ICS plus LABA. The dose of ICS was higher in the group with PAO than in the non-PAO group (P = 0.037). Seventy-five per cent (15/19) of the patients with PAO and 80% (46/58) of those without PAO received systemic intravenous and oral corticosteroids on a regular basis or as three or more bursts of corticosteroid. Montelukast and/or theophylline or doxophylline was given as add-on therapy to similar proportions of patients in the two groups. During follow-up, the frequency of NFA was slightly higher in the patients with PAO than in those without PAO (5/17 vs 5/55, P = 0.049).
|Number of subjects||19||58|
|Age at first visit, years||57.5 ± 2.18||52.0 ± 1.96||0.2|
|Age at onset of asthma, years||37.7 ± 4.68||41.8 ± 2.17||0.486|
|Duration of asthma, years||19.7 ± 4.55||10.2 ± 1.53||0.036|
|Duration of follow-up, years||5.00 ± 0.96||5.64 ± 0.39||0.224|
|Smoking status, NS/ES/CS||10/4/5||32/14/12||0.868|
|Cigarettes smoked, pack years||29.9 ± 9.86||19.9 ± 3.7||0.374|
|BMI, kg/m2||23.9 ± 0.90||23.8 ± 0.47||0.932|
|Inhaled corticosteroid dose, µg/day of budesonide equivalent†||1995 ± 129.2||1621 ± 65.4||0.037|
|Use of oral corticosteroids, n‡||15||46||>0.99|
|Use of montelukast, n||14||39||0.777|
|Use of theophylline/doxophylline, n||14||47||0.523|
|NFA, % (n)||29.4 (17)||9.1 (55)||0.049|
|WCC, ×109/L||8.68 ± 0.47||7.72 ± 0.25||0.064|
|CRP, mg/L||6.3 ± 0.08||5.0 ± 0.03||0.175|
Physiological parameters in patients with RA, with or without chronic PAO
Initial FVC was similar, whereas initial FEV1 and FEV1/FVC were significantly lower in patients with PAO than in those without PAO (P = 0.004 and P = 0.004, respectively, Table 2). Responses to inhaled short-acting bronchodilators were similar in the two groups. Maximum FEV1 during treatment was significantly lower in the patients with PAO (P < 0.001), although minimum FEV1 was similar for the two groups. DLCO was significantly lower in patients with PAO (P = 0.017). The percentage of eosinophils in peripheral blood was significantly higher in patients without PAO than in those with PAO (P = 0.041).
|Initial FEV1, % predicted||48.1 ± 3.01||64.6 ± 3.09||0.004|
|Initial FVC, % predicted||65.8 ± 3.38||74.2 ± 2.8||0.129|
|Initial FEV1/FVC||53.9 ± 2.13||64.5 ± 1.78||0.004|
|Bronchodilator response, FEV1 %||7.1 ± 3.1||8.3 ± 1.2||0.57|
|Best FEV1, % predicted||68.3 ± 1.92||98.5 ± 1.70||0.001|
|DLCO, % (range)||87.5 ± 3.47 (73–116)||106.7 ± 6.84 (71–121)||0.017|
|Total number of acute exacerbations (range)||16.9 ± 3.9 (1–57)||12.2 ± 1.1 (2–38)||0.692|
|Allergic rhinitis, %||16.7||31.0||0.366|
|IgE, IU/mL (range)||745.2 ± 260.1 (4–4284)||382.2 ± 71.1 (12–3182)||0.79|
|Peripheral blood eosinophils, % (range)||3.85 ± 0.92 (0.0–13.0)||6.86 ± 0.82 (0.0–22.9)||0.041|
Inflammatory cell profiles in sputum of patients with RA, with or without PAO
Analysis of sputum from all patients with RA showed that there were similar numbers with eosinophilic- and neutrophilic-type inflammation (35.1% vs 32.5%, Fig. 1). Mixed granulocytic-type inflammation was observed in 10.4% of the patients, and 22.0% showed paucigranulocytic-type inflammation. When the patients were divided into two groups (PAO and non-PAO), the inflammatory subtypes appeared to be different in these two groups. In the PAO group, neutrophilic inflammation predominated, whereas none of these patients showed eosinophilic inflammation. Next most frequently observed, in descending order, were pauci-granulocytic-type and mixed-type inflammation. In contrast, eosinophilic inflammation predominated in the non-PAO group. Neutrophilic inflammation was the next most frequent, followed by paucigranulocytic- and mixed-type inflammation. The difference in the distribution of neutrophilic and eosinophilic inflammation between the two groups was statistically significant (Pearson's chi-square test, P = 0.003). The percentage of sputum eosinophils was significantly higher in patients without PAO than in those with PAO (15.4 ± 2.80% vs 1.88 ± 0.97%, P = 0.003, Fig. 2). In contrast, the percentage of sputum neutrophils was higher in patients with PAO than in those without PAO (68.5 ± 6.55% vs 53.7 ± 3.3%, P = 0.027, Fig. 2).
Differences in clinical and physiological variables in patients with RA, with or without PAO, according to smoking status
To exclude the effect of smoking on chronic PAO in asthma, each group was stratified into smoker/ex-smoker and non-smoker subgroups (Table 3). BMI and age did not differ significantly according to smoking status. As expected, in both groups, smokers/ex-smokers were predominantly male. The duration of asthma was significantly longer in the non-smoking group with PAO (mean of 30.6 years) as compared with the other three groups (mean of 6–13 years, P < 0.05). The onset of asthma was also significantly earlier in the non-smoking group with PAO (mean age of 26.0 years) as compared with the other three groups (mean age of 42–51 years, P < 0.05). Interestingly, this group showed a low rate of skin test reactivity (10%), whereas the prevalence of atopy was very high (62.5%) in the smoking group with PAO. Analysis of the initial lung function data showed that the non-smoking group with PAO had the worst values for initial FEV1, FEV1/FVC and maximum FEV1 among the four groups. This data suggested that this group manifests with the most severe asthma in terms of lung function. Sputum analysis showed that the non-smoking group with PAO had a higher percentage of neutrophils (60%) compared with the non-smoking group without PAO (43.7%, P < 0.05, Table 3). The number of sputum neutrophils was also significantly higher in non-smoking patients with PAO compared with the other groups (P = 0.02, Table 3).
|Number of subjects||9||10||26||32|
|Age at initial visit, years||58.4 ± 12.2||56.6 ± 6.8||49.2 ± 14.9||54.3 ± 14.8|
|Duration of asthma, years||7.7 ± 5.5||30.6 ± 22.0‡||6.9 ± 9.5||12.9 ± 12.6|
|Age at onset of asthma, years||50.8 ± 12.9||26.0 ± 19.0‡||42.3 ± 16.3||41.4 ± 16.9|
|BMI, kg/m2||24.6 ± 4.07||23.4 ± 3.78||24.4 ± 4.21||23.4 ± 2.98|
|Atopy, % positive||62.5¶||10.0||34.6||25.8|
|Initial FEV1, % predicted||51.2 ± 12.8||45.3 ± 13.5§||66.1 ± 25.1||63.3 ± 22.4|
|Initial FVC, % predicted||65.9 ± 14.0||65.8 ± 16.1||77.1 ± 22.5||71.9 ± 20.5|
|Initial FEV1/FVC, % predicted||57.4 ± 11.5||50.7 ± 5.5**||64.0 ± 14.8||64.8 ± 12.7|
|Best FEV1, % predicted||68.2 ± 5.7††||68.3 ± 10.5**||96.7 ± 13.6||99.9 ± 12.4|
|Sputum total cell count, ×105 cells||6.1 ± 7.7||13.4 ± 15.9||10.2 ± 25.9||9.2 ± 17.1|
|Macrophage numbers||68.3 ± 21.8||58.8 ± 29.7||58.4 ± 15.1||57.6 ± 10.4|
|Neutrophil numbers§§||247.6 ± 42.3||298.7 ± 35.7||174.3 ± 19.8||229.2 ± 19.5|
|Eosinophil numbers§§||2.11 ± 2.1||12.7 ± 7.02||72.6 ± 20.2||52.2 ± 12.6|
|Lymphocyte numbers||3.8 ± 3.5||8.9 ± 4.1||4.5 ± 1.81||6.3 ± 1.8|
|Columnar epithelial cell numbers||23.8 ± 15.0||15.4 ± 13.2||13.6 ± 6.9||8.3 ± 2.7|
|Sputum type, neutrophilic/eosinophilic/mixed/pauci–granulocytic, n||33.3/0.0/11.1/55.6‡‡||60/0.0/20/20§||19.2/46.2/15.4/19.2||43.7/46.9/3.1/6.3|
In the present study, sputum analysis showed that neutrophilic inflammation was predominant in patients with PAO, whereas eosinophilic inflammation predominated in those without PAO. This data indicates that sputum inflammatory cells can be used to differentiate the two subphenotypes of RA. These findings are in good agreement with previous reports that airway eosinophilia is consistently associated with favourable responses to corticosteroid therapy, whereas non-eosinophilic inflammation was associated with a significantly poorer response to ICS in patients with mild asthma, as well as severe asthma.3,11,21
In contrast to the present study, recent studies have suggested that eosinophilic airway inflammation contributes to persistent airflow limitation in severe asthma.22,23 Several factors, including age, gender, age at onset of asthma, sputum inflammatory profile and dosage of ICS, may explain the different results in the present patient group as compared with other studies.11,22
In the present study, the patients with RA had used higher doses of ICS because their FEV1 was below normal and their symptoms were uncontrolled when using the dose of ICS recommended in the guidelines.16 Even the use of higher doses of ICS over a period of 2 years or more did not reverse airway obstruction in the subjects with RA in the present study. The proportion of patients with PAO was about 25% (19/77) of all the patients with RA. In addition, sputum analysis showed that almost all these patients showed neutrophilic inflammation. Thus, the identification of specific subphenotypes, such as patients with PAO and sputum neutrophilia, may extend our understanding of pathophysiology and treatment responses, and lead to better targeting of both existing and novel anti-asthma therapies, especially in patients with RA.20
Some studies on patients dying in status asthmaticus or patients with steroid-resistant asthma have shown an inflammatory pattern of increased eosinophil numbers and Th2-type cytokines, which does not respond to corticosteroid therapy.24,25 The blunted response to corticosteroids may be overcome by increasing the corticosteroid dose sufficiently, which presumably results in a decrease in local inflammation.24 For patients in the present study, the dose of ICS used was higher than recommended and was higher in those with PAO than in those without PAO. However, maximum FEV1 during treatment was much lower in the former group than in the latter group. This data indicated that the patients with PAO had severe airway obstruction despite using higher doses of asthma medications than those without PAO, suggesting that the airways of the former group were steroid-resistant.
The mechanisms underlying PAO, despite corticosteroid therapy in patients with RA, are not obvious. The remodelling of airway structural elements has been regarded as contributing to this irreversibility of airway obstruction.26,27 Irreversibility of airway obstruction is often attributed to long-standing, uncontrolled inflammation that eventually leads to increasingly fibrotic airways.28,29 In the present study, the very long duration of asthma in non-smoking patients with PAO (mean of 30.6 years) may support the idea that airway remodelling is dependent on the duration of asthma. Smoking has also been regarded as a major factor contributing to non-responsiveness to asthma medications and irreversibility of airway obstruction.30
Smoking status should be considered when the morphology and physiology of changes in the small airways are evaluated. Smoking asthmatic patients have a higher risk of developing COPD compared with non-smokers.30 Asthmatic patients who are smokers show more severe asthma symptoms31 and have a greater requirement for rescue medication.32 Cigarette smoking combined with asthma results in acceleration of the decline in lung function to a greater degree than either factor alone.33 Because 45% of patients with RA and PAO in the present study were current and previous smokers, each group was stratified into smoker/ex-smoker and non-smoker groups to exclude a smoking effect. The non-smoking PAO group had an earlier age of onset of asthma (mean age of 26.0 years) and a very long duration of asthma (mean of 30.6 years) compared with the other three groups. This data indicated that even in the absence of smoking, a long duration of asthma alone may be associated with chronic airway obstruction. Recently, Mauad and co-workers observed increased numbers of damaged alveolar attachments and a decreased content of elastic fibres in the adventitial layer of the small airways and peribronchial alveoli in non-smoking fatal asthma, and proposed that structural alterations may contribute to some of the functional abnormalities observed in patients with severe asthma34 Intriguingly, the non-smoking patients with PAO had very high levels of airway neutrophilia compared with the other groups, despite being non-smokers.
Other interesting findings were the very low rate of skin test reactivity (10%) and the female predominance among the non-smoking patients with PAO. Of note, this group had the worst values for initial FEV1/FVC and minimum FEV1 during the follow-up period as compared with the other groups, indicating that this group manifested with the most severe form of long-standing asthma. It is not known which factors, other than atopy and gender, may have detrimental effects on the airways over a long period of time in this type of RA. Sputum analysis showed that the proportion of patients with neutrophilic inflammation was higher in the non-smoking group with PAO than in the non-smoking group without PAO, suggesting that the patients with non-atopic severe asthma of long duration had extensive neutrophilic inflammation and PAO that were unresponsive to high-dose corticosteroid treatment even if they had never smoked.
Although the effect of neutrophils on airway wall remodelling remains to be determined, neutrophils also produce matrix metalloproteinases and oxygen-free radicals, which may profoundly alter the structure and function of the airways.35 In addition, neutrophilic inflammation shows a poor response to corticosteroids. In fact, corticosteroids may prolong the survival of neutrophils by decreasing the apoptosis of these cells.36–38 Neutrophils in the airways produce elastase and matrix metalloprotease-9, which contribute to the destruction of elastin fibres in the distal airways,39 and emphysema-like changes in morphology, with increased air trapping. In a previous study of NFA, we observed extensive abnormalities of the small airways and that these abnormalities were partially reversible with successful control of asthma symptoms.17
In the present study, we did not evaluate the effects of passive smoking that may contribute to corticosteroid unresponsiveness in non-smoking patients with RA and PAO. Besides the effect of smoking, respiratory infections, especially viral infections, and agents encountered in the work place (occupational asthma) are well known to cause and aggravate asthma.40 However, this group of patients was comprised of females, most of whom had no history of any occupation-related asthma. At present, there is no evidence that any infectious agents were associated with their PAO. Further insights into the relationship between genetic variation and neutrophilic inflammation as compared with the role of environmental triggering factors,41 may provide clues to the aetiology of the subtypes of RA.
In summary, to identify the clinical parameters and airway inflammatory pattern associated with PAO in patients with RA, demographic, laboratory and sputum findings were analysed in patients with or without PAO. The patients with PAO had a longer duration of asthma and a more severe asthma than the patients without PAO. Neutrophilic inflammation was predominant in the PAO group, whereas eosinophilic inflammation was predominant in the non-PAO group. Non-smoking patients with PAO had the longest duration of asthma, with early onset of asthma and the highest percentage of neutrophils in sputum. In conclusion, patients with RA and PAO have different clinical manifestations compared with those without PAO and may have neutrophil predominant airway inflammation, regardless of a smoking effect. These findings provide a rationale for individualized asthma therapy with additional new medications for this subtype of RA.
This study was supported by grants to CSP and HBM from the Korea Health 21 R&D Project, Ministry of Health, Welfare and Family Affairs (A090548), Republic of Korea, and by a grant of the Korea Healthcare Technology R&D Project, Ministry for Health and Welfare, Republic of Korea (A102065).
- 2American Thoracic Society. Proceedings of the ATS workshop on refractory asthma: current understanding, recommendations, and unanswered questions. Am. J. Respir. Crit. Care Med. 2000; 162: 2341–51.
- 16Initiative for Asthma (GINA). Global Strategy for Asthma Management and Prevention. NHLBI/WHO Workshop Report. National Institutes of Health, National Heart, Lung, and Blood Institute, NIH publication no. 95-3659. Bethesda, MD, 1995.
- 18National Asthma Education and Prevention Program. Expert panel report: guidelines for the diagnosis and management of asthma update on selected topics—2002. J. Allergy Clin. Immunol. 2002; 110(5 Suppl.): 141S–219S.
- 37Neutrophils and asthma. J. Investig. Allergol. Clin. Immunol. 2009; 19: 340–54..