Lower leukotriene C4 levels in bronchoalveolar lavage fluid of asthmatic subjects after 2.5 years of inhaled corticosteroid therapy

Long-term treatment with inhaled corticosteroids has been shown to result in improvement of symptoms and lung function in subjects with asthma. Arachidonic acid (AA) metabolites are thought to play a role in the pathophysiology of asthma. It was assessed whether differences could be found in bronchoalveolar lavage (BAL) AA metabolite levels between subjects with asthma who were treated for 2.5 years with inhaled bronchodilators alone or in combination with inhaled corticosteroids. Prostaglandin (PG)D2, PGF2α, 6-keto-PGF1α, thromboxane B2, leukotriene (LT)C4 and LTB4 levels and cell numbers were assessed in BAL fluid from 22 non-smoking asthmatic subjects. They were participating in a randomized, double-blind multicentre drug trial over a period of 2.5 years. Results of the group treated with inhaled corticosteroids (CS+: beclomethasone 200 μg four times daily) were compared with the other group (CS−) which was treated with either ipratropium bromide (40 μg four times daily) or placebo. BAL LTC4 levels of asthmatic subjects were significantly lower after 2.5 years inhaled corticosteroid therapy (CS+, 9(1–17) pg/ml vs. CS−, 16(6-53) pg/ml; p = 0.01). The same trend was observed for the PGD2 levels. The results suggest that inhaled corticosteroids may exert their beneficial effect on lung function via a mechanism in which inhibition of LTC4 synthesis in the airways is involved.


Introduction
Recent long-term studies on the prognosis and morbidity of obstructive airways disease have shown that addition of inhaled corticosteroids to maintenance treatment with a 2-agonist in hyper-responsive patients with obstructive airways disease leads to a significant reduction of respiratory symptoms, exacerbation rates, airway 12 obstruction and airway hyper-responsiveness. These findings are thought to be a consequence of suppression of inflammatory processes in the airways.
No single cell type or mediator in the inflammatory processes in the airways is responsible for the clinical events in asthma. Nevertheless, there is now substantial evidence that arachidonic acid (AA) metabolites play an important role in the pathophysiology of the disease. They are potent airway constrictors, increase mucus secretion, play a role in chemotaxis and may 426 Mediators of Inflammation Vol 4 1995 enhance airway responsiveness, one of the characteristics of the disease. A role for AA metabolites in the modulation of severity of asthma is suggested by the presence of higher levels in BAL fluid as compared to those in healthy subjects, increasing even more after allergen provocation. 4,5 One potential mode of action of inhaled corticosteroids involves the modulation of arachidonic acid (AA) metabolite levels in the airways. They may decrease AA metabolite synthesis by their inhibitory effect on phospholipase A2 or by inhibition of synthesis of cytokines that stimulate AA metabolite release. In order to assess the role of AA metabolites in the effects of long-term treatment of asthmatic subjects with corticosteroids, AA metabolite levels in bronchoalveolar lavage (BAL) fluid were assessed in a subgroup of non-smoking asthmatic patients who participated in a randomized, double-blind multicentre drug trial for 2.5 years.

Materials and Methods
Correlations between BAL data and lung function parameters obtained 1 week preceding the Patient selection: Before entering the multicentre bronchoscopy procedure were investigated. The drug intervention trial, patients were selected study protocol was approved by the medical based on: (1) Baseline FEV larger than 1.21 ethics committees of the participating hospitals.
(ranging between 4.5 and 1.6 residual standard All patients gave written informed consent. deviations below the predicted value), or an FEV1/inspiratory vital capacity (IVC) ratio lower Pulmonary function and inhalation provocation than 1.64 RSD below the predicted value, pro-tests.. FEV was performed on water-sealed spi-7 vided that total lung capacity was higher than rometers according to standardized guidelines, 1.64 RSD below the predicted level; (2) airway before and 20 min after four single inhalations of hyper-responsiveness to histamine, defined as the 250 l-tg of terbutaline administered through a 750provocative concentration of histamine that ml spacer device (Nebuhaler). Histamine provocaused 20% decrease in FEV (PC20) 8 mg/ml; cation tests were performed using a 2 min tidal and (3) exclusion of pregnant women, patients breathing method, as described previously. Meawith a history of occupational asthma or other surements were made only during clinical stable serious diseases, patients who used oral corticos-periods, and not within 4 weeks after the termiteroids, 13-blockers, nitrates, or anticoagulants, nation of a course of prednisolone. All pulmonand patients who continuously used antibiotics, ary medications were discontinued 8 h before each test.
At enrolment, the subjects were assigned to one of three double-blind regimens using iden-Bronchoalveolar lavage: Fibre optic bronchotical metered dose inhalers: all patients received scopy (Olympus B1 IT10, Tokyo, Japan) was an inhaled 132-agonist (terbutaline 500 l.tg) comundertaken according to guidelines of the Amerbined with either an inhaled corticosteroid ican Thoracic Society. 8 Premedication consisted (beclomethasone dipropionate 200 btg), an of intramuscular injection of 0.5 mg atropine and inhaled anticholinergic (ipratropium bromide inhalation of 5001.tg terbutaline 30 min before 401.tg) or an inhaled placebo. All medication was the procedure. Lidocaine 4% was administered taken four times daily, into the upper airways and bronchial tree. Bronchoalveolar lavage was performed with Study design.. AA metabolites and cell numbers 1 x 30ml (pool 1) and 4 x 50ml (pool 2) were analysed in BAL fluid obtained from outsterile phosphate-buffered saline (PBS) at 37C patients from two university pulmonary departwith the bronchoscope wedged in the lateral ments (Groningen and Rotterdam), 2.5 years segment of the fight middle lobe. After recovery after participation in the above-indicated randoby gentle suction (-40cm H20), the BAL fluid mized, double-blind multicentre drug intervenwas collected in a siliconized specimen trap tion trial. At the time of selection of patients for placed on melting ice. this study and bronchoscopy the patients were Immediately after collection of the BAL fluid still receiving their trial medication. Therefore, the laboratory procedures were carried out. The the investigators were blind to the treatment that BAL fluid was centrifuged at 400 x g at 4C for 5 had been received in the preceding 30 months, min. BAL supernatant was separated from the cell Analysis was performed only in BAL fluid pellet. The cell pellets were washed in PBS supobtained from 22 non-smoking patients with plemented with 0.5% heat-inactivated bovine asthma, diagnosed by criteria from a standarserum albumin (BSA). Total leukocyte numbers dized history of respiratory symptoms and with in BAL cell suspensions were counted in a a reversibility of airways obstruction > 9% of Coulter Counter and viability was assessed by FEV1% predicted at entry, cellular exclusion of trypan blue. Cell differentials In accordance with findings in all participants were done on May-GrOnwald-Giemsa stained of the trial, no significant differences in FEV cytocentrifuge preparations. At least 500 cells values, reversibility of airway obstruction and were counted. airway responsiveness to histamine obtained at the last visit preceding the bronchoscopy proce-AA metabolite determination: Immediately after dure were found between asthmatic patients the BAL procedure, 20ml of BAL supernatant assigned to the placebo and anticholinergic from pool 2 was processed on C18 SepPak cargroups. Therefore, differences in BAL paratridges (Millipore, Bedford, USA)as described meters in this study were investigated between previously, 9 eluated with 2.5ml methanol and the groups without (i.e. placebo + anti-stored at -80C until analysis. Samples of 200 l.tl cholinergic therapy) and with inhaled corticos-BAL eluted fluid were pipetted into polypropylene teroid therapy, tubes and dried with a Savant sample con- CSand CS asthma groups treated without and with inhaled corticosteroids, respectively. *Atopy as determined by positive results of intracutaneous tests against house dust mite or two other tested common aeroallergens.

Results
Subjects: Patient characteristics and lung function parameters at entry in the study in the groups with (CS +, n 9) and without (CS-, n 13) corticosteroid treatment are listed in Table 1.
Comparing group data at entry in the trial retrospectively, no significant differences in patient characteristics and lung function parameters were found between the groups. In contrast, after 2.5 years of double-blind, randomized treatment, a significant improvement in FEV1, reduction in reversibility of airway obstruction and reduction in airway responsiveness to PC20 histamine were found in the CS + as compared to the CSgroup (Table 2).

BAL cell numbers and levels of AA metabolites:
The percentage recovery of BAL fluid was significantl higher in the CS + than in the CSgroup (p 0.01). There were no significant differences in median total or differential cell numbers/ml BAL fluid between the groups ( Table 3). The median LTC4 level in the CS + group was significantly lower than the level in the CSgroup (p= 0.01), while the median PGD2 level showed the same trend ( Table 4).
The levels of the other investigated AA metabolites were not significantly different between the CS + and CSgroups.  Correlation with lung function: EWe 4 levels correlated significantly with FEV1 % pred.

Discussion
This study started from the hypothesis that suppression of inflammatory processes in the airways underlies the improvement in lung function observed after long-term treatment with inhaled corticosteroids. AA metabolite levels are considered as biochemical markers of on-going chronic airway inflammation in the airways of asthmatic subjects. -5 Therefore, we investigated whether differences in BAL AA metabolite levels could be detected between subjects treated with [2-agonists and inhaled corticosteroids for 2.5 years and those treated with inhaled 2-agonists alone. Results from this study show that the BAL LTC4 levels are significantly lower in asthmatic patients treated with inhaled corticosteroids. The same trend is observed for PGD2 levels. The median BAL AA metabolite values of the corticosteroid treated group were within the same level as those from a control group of eight non-smoking, non-atopic healthy subjects who were concurrently analysed during the same procedure, m This is the first study in which AA metabolite levels were measured after long-term treatment with inhaled corticosteroids. In contrast to our results, short-term treatment with prednisolone 60mg/day has been reported to have no significan effect on BAL fluid AA metabolite levels in asthmatic subjects, although the in vitro synthesis of AA metabolites by BAL cells was decreased. A lower production of LTC4 by BAL cells was also found in asthmatic subjects who had been treated for more than 2 years with 5-15 mg prednisone, 2 which is in line with our results. A role for cysteinyl leukotrienes in the pathophysiology of asthma is suggested by findings that leukotrienes induce airway obstruction, increase airway hyper-responsiveness and increase mucous secretion. After oral treatment with a leukotriene D4 receptor antagonist a significan reduction in asthma symptoms and improvement of lung function has been reported in asthmatic subjects. In the light of current knowledge, several mechanisms can be postulated to explain the reduced LTC4 levels in the CS-treated group. First, corticosteroids exert a decreased cellular AA metabolite synthesis by inhibiting phospholipase A2 activation through the generation of lipocortin. 4 Results from this study suggest that cells that predominantly release LTC4 (eosinophils, alveolar macrophages, mast cells) are more sensitive to corticosteroid treatment. Secondly, corticosteroids may selectively inhibit the transcription of cytokines from airway cells 6'5'6 that may regulate LTC4 release.
IL-3, ILo5 and GM-CSF have been shown to prime human basophils, eosinophils and neutrophils for augmented release of LTC4 after stimulation by a second agonist. [17][18][19] Results of this study suggest an effect of longterm corticosteroid treatment on BAL PGD2 levels as well. PGD2 is a potent airway constrictor and is implicated in the increase in airway responsiveness. PGD2 is synthesized by a variety of airway cells. It has been observed that the in vitro PGD2 synthesis by human lun2 mast cells was not affected by glucocorticoids. The PGD2 synthesis by human alveolar macrophages upon stimulation by calcium ionophore A23187 was, however, significantly inhibited by methyl prednisolone. 21 If in vitro results can be extrapolated to the in vivo situation, the results favour a role for PGD2 release by alveolar macrophages in the chronic inflammatory process in asthma.
We did not find a difference in total or differential cell numbers between the groups. Previous findings in studies on the effect of short-term corticosteroid treatment (6 weeks to 4 months) on BAL cell numbers are not consistent, although in the majority a reduction of BAL eosinophil numbers was found. 15'16'22'23 It cannot be excluded that we did not find differences as a consequence of group sizes that were too small. Inflammatory processes in the airways are, however, probably better reflected by cell activation than increased cell numbers in the BAL fluid, as has been found in this study.
In conclusion, this study shows that BAL LTC4 levels of asthmatic subjects were significantly lower after 2.5 years inhaled corticosteroid therapy. The results suggest that corticosteroids may exert their beneficial effect on lung function via a mechanism in which inhibition of LTC4 synthesis in the airways is involved.