INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Bronchial asthma is characterized by reversible airflow obstruction, airway mucosal inflammation, and airway hyperresponsiveness (AHR) (1). The role of inflammation in the development of symptomatic asthma and AHR has been emphasized in the past decade, and asthma treatment has moved from bronchodilator to antiinflammatory drugs (2). Asthmatic airways, even in mild or recently diagnosed asthma, show mucosal edema and cellular infiltrate constituted primarily of activated lymphocytes and eosinophils. There is also evidence of mucosal remodeling such as subepithelial collagen deposition, bronchial epithelium disruption, mucous cell hyperplasia, and smooth muscle changes (5). It is plausible that changes in the structure of airway wall components following epithelial damage and airway inflammation lead to the development of chronic asthma and AHR (6). Indeed, the persistence of AHR despite prolonged antiinflammatory treatment with corticosteroids and the analyses of bronchial biopsies obtained in asthma and related conditions suggest that it is not only the inflammation that leads to the development and chronicity of AHR, but also the structural modifications of the airways (5, 9).

Early treatment of airway inflammation could modify the outcome of asthma by preventing a permanent loss of pulmonary function or even induce remission in some patients (13, 14). It seems, however, that by the time symptoms appear, some irreversible structural changes have already occurred (15). Moreover airway inflammation tends to recur with the interruption of antiinflammatory therapy and airway structural changes do not seem to respond well to inhaled corticosteroids (19, 20). Consequently we hypothesized that airway inflammation predominates over structural changes in recently diagnosed asthma and that the opposite occurs in long-standing asthma. In these circumstances, high doses of inhaled corticosteroids (ICS) could be more effective in patients with recently diagnosed mild asthma and could possibly improve or even normalize AHR compared to patients with long-standing asthma.

This study was set to determine the extent of physiological, inflammatory and morphological airway changes in recently diagnosed asthma versus long-standing asthma and evaluate the influence of a high dose of inhaled fluticasone given for 8 wk on those parameters.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Patients' Selection

Thirty-two patients with asthma  18 yr old using only bronchodilators to treat their asthma were included in this study (Table 1). None used leukotriene receptor antagonists nor required any other antiinflammatory medication that could have interfered with the study. Asthma was stable and the medication unchanged for at least 1 mo before the study. Patients with recently diagnosed asthma (RDA) included patients (10 F, 6 M; aged 18 to 36 yr, mean ± SEM: 22.4 ± 1.3) with a diagnosis of asthma  2 yr. No patient had experienced symptoms compatible with asthma for more than 2 yr. They had not received any inhaled or oral corticosteroids in the past. Patients with long-standing asthma (LSA) included patients (9 F, 7 M; aged 21 to 38 yr, mean ± SEM: 27.8 ± 1.6) with symptomatic asthma having been diagnosed at least 13 yr prior to the study. No patients had previously been treated with oral prednisone and three used ICS. Of these three, one was on budesonide turbuhaler 200 µg daily during 2 yr and had stopped 3 mo before the study, the second had received beclomethasone metered dose (MDI) 250 µg daily for 3 wk 6 mo before the study, and the third occasionally took beclomethasone MDI 250 µg daily during asthma exacerbation and had not required it in the last year. We excluded patients with a lower respiratory tract infection within 4 wk of visit 1, smokers of more than 6 pack-years of cigarettes, and patients with other respiratory conditions that could affect the evaluation (example: chronic obstructive pulmonary disease [COPD], bronchiectasis, or chronic pulmonary infection).

Study Design

After their inclusion in the study, patients had a respiratory questionnaire, allergy skin-prick tests, measurements of expiratory flows before and after 200 µg of inhaled salbutamol, and a methacholine challenge for measurement of the provocative concentration resulting in a 20% fall in FEV₁ (PC₂₀). They then underwent a 2-wk run-in period with recording of asthma symptoms, use of ₂-agonists (or any other medication), and measurements of peak expiratory flows (PEF) twice daily (morning and evening). At the end of this period, a bronchoscopy with bronchial biopsies was done.

After the run-in period, the patients of the two groups were given a metered-dose inhaler of fluticasone dipropionate (FP) 250 µg per inhalation, and asked to take two inhalations twice a day for 8 wk. Inhalation technique was taught and supervised and a spacer (Aerochamber; UAD Laboratories, Jackson, MS) was given and its use explained to ensure that the medication was taken adequately. Twice daily measurements of PEF, asthma symptom scores, and the use of study and rescue medication or any other medication were recorded on a diary card during these 8 wk. Compliance to the treatment was verified by checking the diary cards. Pre- and postsalbutamol expiratory flows as well as a methacholine PC₂₀ were measured at the end of the treatment period. A second bronchoscopy with bronchial biopsies was also done.

The study was approved by our Institutional Ethics Committee and all subjects had agreed with the study in signing an informed consent form.

Respiratory Questionnaire, Diary Symptom, and PEF Recording

The respiratory questionnaire included questions about the patients' characteristics, smoking, date of diagnosis of asthma, asthma symptoms and their treatment, the nature and intensity of the respiratory symptoms, asthma triggering factors (particularly allergens and respiratory irritants), medication needs, and associated health problems. Asthma symptoms were evaluated in the morning for nighttime symptoms and in the evening for daytime symptoms. All RDA patients were questionned in regard to childhood symptoms, prolonged chest colds, and coughing episodes, and denied having experienced any of those in the past.

Symptoms of breathlessness, wheezing, cough, sputum, chest tightness, and night awakening were evaluated on a scale of 0 (no symptom) to 5 (very severe). The patients measured PEF with a peak flow meter in the morning and evening before medication was taken. Use of salbutamol (rescue medication), study medication, or any other medications was noted on the diary card.

Skin-prick Tests

Allergy skin-prick tests were performed with a battery of common allergens including animal dander, dust, house-dust mite, pollens, and molds. Atopic status was defined as the presence of at least one positive reaction (wheal diameter  3 mm) 10 min after testing.

Spirometry and Methacholine Challenges

Pulmonary function tests included three reproducible measurements of FEV₁ and forced vital capacity (FVC) before and after inhalation of salbutamol (200 µg) according to a standardized procedure with a spirometer approved by the American Thoracic Society (21). Methacholine inhalation tests were done in the morning at the same time of day according to the method described by Juniper and coworkers (22). The percentage reduction in FEV₁ in response to methacholine was determined from the postsaline value and PC₂₀ was interpolated at 20% fall in FEV₁ on the dose-response curve.

Bronchial Biopsy Sampling and Processing

Bronchoscopies were performed according to standard guidelines (23). Nine to ten bronchial biopsies were taken from segmental or lobar bronchi as previously described (24). Bronchial biopsies were immediately placed in acetone antiprotease solution (20° C) and embedded in GMA resin the same day (25). Samples were stored at 20°C in an airtight container containing a desiccant. Two-micrometer sections were cut on an ultramicrotome (Leica Reichert Ultracut S; Vienna, Austria), then floated on water containing 0.02% ammonia until fixed on poly-L-lysine-coated slides. We used a Dako Autostainer Universal Staining System (Dako Diagnostics, Canada Inc., Mississauga, ON, Canada) to perform all immunostainings with the following antibodies: mouse anti-human CD3, CD4, CD8, CD45RO, CD25, and tryptase (mast cells) from Dako Diagnostics, EG1 and EG2 from Kabi Pharmacia Diagnostics (Baie D'Urfé, PQ, Canada), and types 1 and 3 collagens from Biodesign (Kennebunk, ME). Bronchial biopsy immunostaining was performed as previously described (18). Briefly, sample peroxidase background activity was blocked by incubations with 0.3% H₂O₂-0.1% sodium azide in Tris buffer solution (TBS) for 30 min and normal serum for 30 min. The samples were then incubated consecutively overnight with primary antibody (4° C), 1 h with a biotinylated secondary antibody (23° C), and 40 min with ABC-HRP complex (23° C). The slides were washed three times with TBS between steps. To detect bound antibodies, they were incubated with aminoethylcarbazole (AEC) substrate. Washing with water stopped the reaction and Mayer hematoxylin was used for counterstaining. Slides were mounted with Crystal mount (Biomeda, Foster City, CA). Irrelevant isotypic antibodies served as negative controls.

Quantification of Immunohistochemical Staining

All slides were assigned a number and measurements were made blindly. The same blinded observer counted immunostained cells present in the lamina propria of coded sections. Counts were expressed as number of positive cells/mm² of bronchial lamina propria excluding mucus glands and smooth muscles. These lamina proprial areas were measured with a calibrated image analysis system (Mocha image analysis software; Jandel Scientific, San Raphael, CA). The collagen content of bronchial mucosa was determined by immunoperoxidase staining of serial sections. Each type of collagen was identified using specific antibodies. The extent of collagen deposition underneath the basement membrane (more specifically the collagen deposit localized in the reticular layer of the basement membrane) was measured at regular intervals with a calibrated image analyzer (Mocha image analysis software; Jandel Scientific). Eleven measurements were performed on an unfolded nontangential area from each patient's biopsy specimens.

Data Analysis

For symptom score analysis we added the maximum score of the 24 h for each symptom and the mean of the six symptom scores was considered as the symptom score of the day. For the run-in period, the mean symptom score was the mean of the 2 wk of daily symptom scores; for the treatment period, the mean of the two last weeks on fluticasone was used for comparison. Comparisons on the nature and duration of respiratory symptoms and other characteristics of the disease from the respiratory questionnaire were done using t tests or Wilcoxon rank sum tests for continuous variables and contingency tables for nominal variables. Measurements of expiratory flows, PC₂₀, and data from bronchial biopsies were analyzed using an ANOVA model with values before treatment as covariables. Biopsy cell counts were compared, according to the data, using parameter analysis if normality and homogeneity of variance assumptions were met or Kruskal-Wallis test. Results were considered statistically significant at a p value < 0.05. The Bonferoni inequality formula was applied where needed (26).

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Patients' Clinical Characteristics

Patients' characteristics are summarized in Table 1. Six patients with RDA were diagnosed between a few days to a few weeks (< 1 mo) before entering the study. Two patients from RDA and four from LSA had an animal at home for which they had a positive skin-prick test and 13 RDA and 15 LSA patients had a positive prick test for house dust or house dust mite.

Asthma Symptoms, Expiratory Flows, and Medication Use

Mean score for asthma symptoms, mean baseline A.M. and P.M. PEF, and mean inhaled ₂-agonist use during the run-in period in RDA and LSA, are summarized in Table 2. Six patients of the RDA group had never used a ₂-agonist before. After 2 mo on inhaled fluticasone, asthma symptoms were reduced in RDA (p = 0.055) and in LSA (p = 0.004) compared to baseline and were similar in both groups. Morning (A.M.) PEF were significantly increased in RDA (p = 0.009) and LSA (p = 0.02), respectively, while evening (P.M.) PEF were significantly increased in LSA (p = 0.018) but not in RDA (p = 0.063). Fluticasone treatment reduced the use of ₂-agonist in both RDA (p > 0.05) and LSA (p = 0.02), but the reduction was significant only in LSA.

FEV₁ and Airway Responsiveness

Baseline FEV₁ was 3.55 ± 0.18 L (98.1 ± 2.7%) in RDA and 3.30 ± 0.20 L (94.5 ± 4.6%) in LSA (p > 0.05), and it improved similarly by 7.0 ± 0.9% and 9.7 ± 2.4% 15 min after salbutamol (Table 1). After 2 mo of inhaled fluticasone, FEV₁ improved slightly (p = 0.03) and similarly in RDA (2 ± 1.2%) and LSA (3.9 ± 2.1%) up to a mean of 3.62 ± 0.18 L (100.3 ± 2.5%) and 3.43 ± 0.18 L (98.4 ± 3.4%), respectively, and was not significantly different between the two groups (p > 0.05). Salbutamol induced further increase in FEV₁ of 6.1 ± 1.2% and 6.1 ± 1.3%, respectively. Mean PC₂₀ (mg/ml) in the run-in period was 3.4 in RDA and 0.44 in LSA (p = 0.0009). Improvement in PC₂₀ was significant (p = 0.0001) and of similar amplitude in RDA and LSA with respective mean increases of 1.86 and 1.85 double concentrations of methacholine (Figure 1). PC₂₀ normalized after ICS ( 16 mg/ml) in five patients with RDA (PC₂₀ pre/post: 9.6/ 31.2, 12.7/19.6, 5.8/16.1, 0.39/> 64, 8.0/> 64 mg/ml) and two with LSA (PC₂₀ pre/post: 5.6/23.4, 4.0/17.6 mg/ml). Mean PC₂₀ at the end of the treatment period was 12.2 mg/ml in RDA and 1.6 mg/ ml in LSA groups (p = 0.0002). Baseline PC₂₀ correlated positively with the age at onset of asthma (r_s = 0.59, p = 0.001, Figure 2A) and negatively with the number of years of duration of asthma (r_s = 0.55, p = 0.002, Figure 2B).

View larger version (27K): [in this window] [in a new window]   Figure 1.   Mean baseline PC20 were, respectively, 3.4 and 0.44 mg/ml in patients RDA (open circles) and LSA (closed circles) and improved similarly in both groups after treatment with fluticasone propionate for 8 wk. (p = 0.0002). Bars represent mean.

View larger version (17K): [in this window] [in a new window]   Figure 2.   Correlations between baseline PC20 and age at asthma onset (A) or duration of asthma (B). Baseline PC20 was positively correlated with the age at onset of asthma and negatively correlated with the number of years of duration of asthma. RDA, open circles; LSA, closed circles.

Airway Inflammatory Cell Counts

Adequate material was available for immunohistochemical techniques in all RDA and in 15 LSA. Mean baseline levels of inflammatory cells positive for CD3, CD4, CD8, CD25, CD45ro, EG1, EG2, AA1, and elastase (per mm² connective tissue surface) were not significantly different between RDA and LSA (Figure 3). Inhaled fluticasone treatment decreased EG1⁺ cells (p < 0.001), EG2⁺ cells (p = 0.017), and AA1⁺ cells (p = 0.009) similarly in both RDA and LSA compared with the run-in period, while CD3⁺ (p = 0.07) and CD45RO⁺ cells (p = 0.02) decreased only in LSA. After 2 mo on inhaled fluticasone, the numbers of lymphocytes T-cell subtypes were significantly different between RDA and LSA for CD3⁺ cells: respectively, 39.7 ± 9.8 and 10.7 ± 4.3 (p = 0.004), CD4⁺ cells: 14.2 ± 5.7 and 2.0 ± 0.9 (p = 0.01), and CD8⁺ cells: 7.8 ± 3.1 and 1.2 ± 0.8 (p = 0.026). Mean counts for CD45ro⁺ cells also tended to be higher in RDA (71.2 ± 17.7 compared with 20.7 ± 5.5) (p = 0.058). The increase in PC₂₀ was significantly correlated with the reduction in EG1⁺ cells in LSA only (r = 0.803, p = 0.005) and did not correlate with changes of other evaluated cell types.

View larger version (36K): [in this window] [in a new window]   Figure 3.   Number of inflammatory cells in bronchial connective tissue before and after 2 mo of inhaled FP. Baseline cell counts were similar in both groups for each cell type. RDA, open circles; LSA, closed circles. Bars represent mean.

Subepithelial Deposition of Collagen

Mean baseline type 1 and type 3 subepithelial collagen deposits were not significantly different in RDA (type 1: 12.5 ± 1.1 µm, type 3: 15.1 ± 1.6 µm) and LSA (respectively, 11.9 ± 1.9, p > 0.05 and 12.3 ± 1.3 µm, p > 0.05) in the run-in period and after 2 mo of inhaled fluticasone (RDA, type 1: 12.4 ± 1.4 µm; type 3: 14.5 ± 6.8 µm and LSA, respectively, 11.9 ± 1.3, p > 0.05 and 12.4 ± 1.2 µm, p > 0.05). Inhaled corticosteroids (ICS) had no effect on collagen deposition whatever the group.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

This study showed that bronchial mucosa inflammatory cell counts and subepithelial fibrosis parameters were similar in recently diagnosed patients with asthma compared with a long-standing asthma. Furthermore, high doses of ICS for a period of 2 mo improved airway hyperresponsiveness by the same magnitude, by slightly less than two double concentrations of methacholine, respectively, in patients with either recently diagnosed or long-standing mild asthma. Airway responsiveness normalized in only a few patients, mostly patients with mild baseline AHR and not necessarily those with recently diagnosed asthma. ICS decreased eosinophils and mast cell counts in both groups but only the EG1⁺ cell count decrease correlated with PC₂₀ improvement in LSA. These data show the partial improvement in AHR in most patients with symptomatic asthma, whatever the duration of the disease, and suggest that in mild asthma airway inflammatory process and subepithelial fibrosis seems to remain quite similar over the years. It also suggests that AHR increases with time in these patients and that a similar improvement in AHR could be obtained with ICS independently of the asthma duration.

Our hypothesis that RDA would have more inflammatory and less fibrogenic bronchial mucosa changes than LSA is not supported by our results in mild asthma since we found no difference in measured baseline inflammatory and remodeling parameters in the two groups. Neither is the consequent hypothesis that a greater improvement in AHR would be obtained with ICS in RDA than in LSA. These assumptions were, however, suggested by previous reports. Indeed, children in whom ICS were started early at the time of diagnosis had better pulmonary functions than those in whom ICS were started 2 yr after diagnosis and the loss of pulmonary function observed in this last group was not recovered by initiating a late ICS treatment (27). Selroos and coworkers observed a reduction in bronchodilator response to ICS with the duration asthma suggesting that there was an irreversible component of airflow obstruction that was developing over time (28). More recently, Lange and coworkers confirmed that there was a progressive decline in airway caliber over time in the asthmatic population (29). In these reports no data on airway inflammation and remodeling were, however, available to evaluate the intensity of airway inflammation and remodeling over time. Furthermore, many of these previous studies included patients with moderate to severe asthma compared with the patients included in our study who had mostly very mild asthma. The difference in asthma severity could explain part of the disparities observed between our data and the above mentioned studies.

In our study, patients with mild asthma, even those who normalized their airway responsiveness, presented similar airway inflammation and remodeling parameters and normal FEV₁ compared with normal predicted values despite largely different asthma duration in the two selected groups. These data would suggest that mild asthma can be associated with a normal airway caliber despite persistent bronchial inflammation over many years in the absence of antiinflammatory treatment. On the other hand, AHR was higher in LSA than in RDA, suggesting that, even in mild asthma, AHR can increase with time in the absence of regular antiinflammatory treatment. Obviously these issues would require a follow-up study over many years in a cohort of patients with mild asthma to be specifically addressed. It would also be interesting to determine if any long-term increase in AHR could be prevented by antiinflammatory treatment. Our results may also suggest that parameters other than inflammation and airway collagen deposition such as changes in smooth muscle or other components of the airway wall could be involved in determining the degree of AHR.

The responses of airway inflammatory and remodeling features (this last being represented by the change in subepithelial collagen deposition) to ICS treatment were similar in the two groups but, interestingly, RDA patients maintained higher CD3⁺, CD4⁺, CD8⁺, and CD45ro⁺ lymphocyte counts. The reason for the RDA persistent lymphocytic infiltration compared to LSA is unknown. We recruited patients who used no antiinflammatory treatment or some inhaled ICS for only a short period of time; consequently, no significant effect of previous antiinflammatory therapy should be present. The selection of high doses of inhaled corticosteroids as a treatment was not based on the clinical severity of the disease. This dosage was chosen to offer a powerful, although safe, antiinflammatory treatment in an attempt to try to get rid of as much airway inflammation as possible in order to observe the remaining physiological abnormalities and see, particularly in recently diagnosed asthma according to our hypothesis, if it may be sufficient to reverse some aspects of airway remodeling.

The duration of ICS therapy (8 wk) should have been long enough to obtain a significant antiinflammatory effect but perhaps not long enough to reach the maximal effect, particularly in regard to airway remodeling. In this regard, Sont and coworkers showed that a 2-yr treatment strategy aimed at reducing airway hyperresponsiveness (AHR strategy) in addition to the recommendations in the existing guidelines could improve AHR and lead to a more effective control of asthma while alleviating chronic airways inflammation and reducing airway remodeling (30). In this last study, a greater reduction in thickness of the subepithelial reticular layer was found on bronchial biopsies in the AHR strategy group compared with the reference strategy group (mean difference 1.7 µm). The changes in AHR in both strategy groups were correlated with eosinophil counts in the biopsies.

In regard to the influence of ICS on remodeling, in another study, Hoshino and coworkers found that inhaled BDP 800 µg daily produced a significant decrease in collagen type III deposition and the expression of submucosal MMP-9, and a significant increase in the expression of submucosal TIMP-1 (31). On the other hand, other authors found no effect of corticosteroid on subepithelial collagen deposition (19). This subject therefore remains controversial.

Our method of evaluation of apparent basement thickness could be less sensitive than others to detect small changes. Nevertheless, we were able in the past to identify significant changes in this parameter among subgroups of subjects and the significance of small changes is uncertain, in regard to the technical difficulties associated with such measurement (17, 32).

In this study, there was a lack of correlation between the degree of airway responsiveness and of subepithelial collagen deposition. This is in keeping with our previous observations suggesting that although when you look at different groups including subjects with asthma and nonasthmatic subjects, there is a correlation between those two parameters, but when looking at only patients with asthma, this correlation is lost or, as reported by others, relatively weaker (7, 15). We should also mention that although they had different baseline PC₂₀ methacholine, our two groups of subjects were clinically mild and were using only bronchodilators on demand.

In this regard, Chu and coworkers found no significant differences in the subepithelial basement membrane thickness, submucosal collagen deposition, eosinophil numbers, or transforming growth factor (TGF)-beta positive cells in subjects with different asthma severities and in normal control subjects (33). However, when examining all patients with asthma together, a modest increase in collagen type III was observed as compared with normal controls. Here again, the data available up to now are often contradictory (17, 19).

Although a lack of effect of ICS was found in some studies, as in ours, other authors reported a reduction in subepithelial collagen deposition in subjects with asthma either after ICS or withdrawal from sensitizing exposure (6, 10, 19, 34). This stresses the need to further evaluate the long-term effects of ICS in different subgroups of patients with asthma in regard to airway responsiveness, inflammation, and remodeling.

There are evidences of airway inflammation and remodeling in atopic patients, sometimes before asthma develops (16). In patients with asymptomatic airway hyperresponsiveness, we found a mild airway inflammation and a degree of airway remodeling intermediary between normal control subjects and patients with asthma (17). Some of these patients, particularly those exposed to domestic allergens to which they were sensitized and with a family history of asthma, developed symptomatic asthma during a 3-yr follow-up. This was associated with an increase of airway inflammation, particularly an increase in the ratio of CD4⁺/CD8⁺, and also evidence of airway increased remodeling such as development of an increased subepithelial collagen deposition. These data support the role of increasing airway inflammation and remodeling in the pathogenesis of AHR and asthma and the concept that ICS could be effective to prevent AHR and chronic asthma if given very early in asthma history (35). The present study suggests that this aim could be reached in patients with mild asthma and should be evaluated earlier in asthma development at a presymptomatic stage in targeted groups.

In conclusion, this study showed no significant difference in airway inflammation and a marker of airway remodeling, subepithelial collagen deposition, in steroid-naive patients with mild recently diagnosed versus long-standing asthma. The improvement in AHR after high dose ICS was similar in the two groups. This suggests that in mild asthma of recent onset, there are irreversible airway changes that may explain why AHR could not be normalized in most patients. These data support the need to determine if interventions administered earlier in the natural history of asthma may prevent its development.