Asthma guidelines recommend inhaled corticosteroids as the most effective anti-inflammatory treatment for chronic asthma.^1,2 Inhaled corticosteroids in asthma have been shown to reduce symptoms and exacerbation rates, to improve lung function,^3,4 and to decrease the risk of death from asthma.^4,5 Most clinical trials have been conducted on non-smokers with asthma and have excluded active smokers. The influence of smoking status on the therapeutic response to corticosteroids has not been investigated until recently. Two randomised controlled trials have found that the efficacy of short term inhaled or oral corticosteroid treatment was impaired in smokers with chronic asthma.^6,7 This finding is potentially of considerable importance for the management of adults with asthma since approximately 25% of them are current smokers.^8,9,10 Smoking in asthma is associated with an increase in asthma symptoms,¹¹ a faster decline in forced expiratory volume in 1 second (FEV₁) over time,¹² and a higher mortality rate after admission with a near fatal asthma attack.¹³

A study was undertaken to assess the efficacy of inhaled corticosteroid treatment when given for a longer duration than previous studies and at different doses. Our hypothesis was that the therapeutic response to inhaled corticosteroids would be reduced in smokers with asthma compared with non-smokers, despite 12 weeks duration of inhaled corticosteroid treatment. If smokers with asthma are found to be insensitive to inhaled corticosteroid treatment, this will have significant implications for the management of these patients who may require alternative or additional treatments to control airway inflammation.

METHODS

Subjects

Subjects with chronic asthma (current smokers and non-smokers) aged 18–60 years were recruited from general practices and a hospital clinic. Asthma was diagnosed by American Thoracic Society criteria.¹⁴ Subjects were receiving a maximum of 500 μg daily of beclomethasone or equivalent and had an FEV₁ over 50% predicted. All were symptomatic, defined as having one or more of the following: a validated asthma control questionnaire score¹⁵ of six or more (maximum 36); reversibility of FEV₁ of 12% or more¹ after 400 μg inhaled salbutamol; diurnal peak expiratory flow (PEF) variability of 15% on four or more days during the run-in period.¹⁶ Subjects with FEV₁ of 80% predicted or less post bronchodilator and FEV₁/FVC 70% or less had to demonstrate either reversibility of FEV₁ or diurnal PEF variability as described above. Smoking participants were cigarette smokers with a minimum smoking history of 5 pack-years (1 pack-year equates to smoking 20 cigarettes daily for 1 year). Non-smokers had either never smoked or had stopped smoking cigarettes over 5 years ago and had smoked 5 pack-years or less. Subjects were excluded if they had a respiratory tract infection or needed oral corticosteroids within the preceding 6 weeks.

All participants gave written informed consent. The multicentre research ethics committee for Scotland and appropriate local research ethics committees granted ethical approval.

Study design and assignment

This was a randomised, double blind, parallel group study comparing the efficacy of 400 μg or 2000 μg daily inhaled beclomethasone in smokers and non-smokers. Study participants took four puffs twice daily of inhaled beclomethasone from a breath activated pressurised device (Easi-Breathe, IVAX Pharmaceuticals, UK). An independent pharmacist overlabelled identical inhalers. The treatment group was assigned from a computer generated list of randomisation numbers. Researchers and participants were blinded to allocation and had no access to the randomisation code held by the Data Centre until completion of the study. Smokers and non-smokers were recruited at approximately the same rate to avoid seasonal confounding. Subjects entered a 2 week run-in where they continued their usual asthma medication and recorded PEF twice daily. Study visits took place 2, 6, and 12 weeks after randomisation. Spirometric parameters were measured before and after administration of salbutamol 400 μg from a large spacer device and a validated asthma control questionnaire was completed.^15,17 At week 12, returned inhaler canisters were weighed as an assessment of compliance.⁶

Measurements

PEF was measured twice daily (mini-Wright meter, Clement Clark, UK or MicroPeak, Micro Medical Ltd, UK) and recorded in a diary card along with reliever inhaler use. Spirometric parameters were recorded using a dry wedge spirometer (Vitalograph, Buckingham, UK) following American Thoracic Society guidelines.¹⁸ Serum cotinine levels were assayed using a commercially available assay (Cozart Biosciences, Abingdon, UK) to confirm smoking status. Total IgE and specific IgE to house dust mite, grass pollen, and cat dander were assayed by the Unicap 100 system (Pharmacia, Uppsala, Sweden). Total IgE levels of over 120 IU/ml and specific IgE levels of more than 0.35 IU/ml were considered raised. When specific IgE was raised, an individual was considered atopic.¹⁹ Severe exacerbations of asthma were defined as a deterioration requiring a course of oral corticosteroid and a mild exacerbation as a fall in morning PEF of ⩾20% from their baseline morning PEF on two consecutive days.²⁰ Participants used their reliever inhaler during exacerbations but were advised not to double the inhaled corticosteroid dose. Attending doctors prescribed courses of oral corticosteroids where necessary.

Statistical analysis

The primary end point was the mean of the morning PEF measurements as taken from 14 days of diary recording before each study visit. The mean was only calculated if there were three days of completed data within that period. Secondary end points included: evening PEF, exacerbations of asthma, daily reliever inhaler use, spirometric data, and the asthma control questionnaire score. Baseline characteristics of smokers and non-smokers were compared by χ² tests, t tests, or Wilcoxon tests. For the continuous end points, changes were calculated at 2, 6, and 12 weeks for the primary end point and at 12 weeks for the secondary end points.

Since groups of smoking and non-smoking participants would have different risk profiles, we attempted to reduce confounding by statistical adjustment. We modeled the effect of all candidate variables at baseline upon the mean morning PEF at 12 weeks. Variables were included in the models as “possible confounders” that were significant at the 20% level. Thus, weak confounders were not erroneously excluded from the analyses. The adjusted analyses of smoking were carried by normal linear models adjusting for baseline, treatment (as a design factor), and the chosen confounders. These confounders were used in all adjusted analyses. An interaction test was used to assess whether the effect of smoking was different for the two doses of inhaled beclomethasone. The occurrence of asthma exacerbations was compared by χ² tests. Further analyses were carried out within the subgroups defined by treatment group. The intended power of the study was 80% (at the 5% level) to detect a mean treatment difference of 20 l/min in PEF measurements and standard deviation of the difference in PEF measurements of 40 l/min. We attempted to recruit a sample size of 64 patients per group (128 in total).

RESULTS

5421 asthmatic patients from 54 practices were invited to participate between May 2002 and December 2003 (fig 1). Screening visits were arranged for 303 volunteers: 152 did not fulfil the inclusion criteria and 56 decided not to take part or did not attend; 95 individuals were randomised, three withdrew because of non-asthma related illnesses, two were lost to follow up, and one decided not to take part.

There were no significant differences in age, baseline FEV₁ % predicted before and after salbutamol, FEV₁/FVC % after salbutamol, and total IgE between smokers with asthma and non-smokers with asthma (table 1). The smoking group had a higher symptom score, a shorter history of asthma, and fewer were atopic. The improvement in FEV₁ after salbutamol was less in smokers than in non-smokers. However, reversibility to salbutamol did not predict the response to corticosteroid (p = 0.38).

There was an improvement in the morning PEF in the non-smoking group (table 2) but little change in the morning PEF measurements in the smoking group. The difference between the combined 400 and 2000 μg doses in the smoking and non-smoking groups was apparent at each visit after PEF was adjusted for duration of asthma, inhaled corticosteroid at baseline, sex, baseline morning PEF, baseline evening PEF, atopy, FEV₁/FVC ratio pre-salbutamol, and treatment group. After 12 weeks the smoking and non-smoking groups were dissimilar (mean difference −18 (95% CI −35 to −1), adjusted p = 0.035). The non-smoking asthmatics showed greater improvement in symptom scores by 12 weeks. There were no other differences in secondary outcome measures (table 3).

There was an improvement in the morning PEF values of non-smoking asthmatics receiving 400 μg daily beclomethasone but not in smokers (fig 2, table 2). By 12 weeks there was a large difference between the groups receiving 400 μg (mean difference −25 (95% CI −45 to −4), adjusted p = 0.019). Smokers receiving lower dose beclomethasone had more exacerbations of asthma than non-smokers (6 v 1, p = 0.0067). One smoker required oral corticosteroid. There was some evidence of greater improvement in the evening PEF in non-smokers after 12 weeks (10.5 l/min) compared with smokers (−12.1 l/min; mean difference −18 (95% CI −38 to 3), adjusted p = 0.088), and in reduction of reliever inhaler use (mean difference 2 (95% CI 0 to 4), adjusted p = 0.057).

There was an improvement in the morning PEF values of non-smoking asthmatics receiving 2000 μg daily beclomethasone and there was a smaller improvement in the morning PEF values of the smoking group (fig 2, table 2). We did not find a difference in morning PEF values or exacerbation rates of those receiving higher dose beclomethasone between smokers and non-smokers (1 v 2, p = 0.661); one smoker required oral corticosteroid treatment. There was a difference between the symptom scores at 12 weeks (mean difference 1 (95% CI 0 to 2), adjusted p = 0.025). We did not see any difference between the smoking asthma patients receiving 400 μg or 2000 μg daily. We carried out a test of interaction to assess whether the effect of smoking was different for the two doses; this test gave a p value of 0.43.

There was no difference in compliance, assessed by weight of returned canisters and by self-assessment at study visits, between the smoking and non-smoking groups. There was no difference between the number of completed days of PEF measurements present in smokers and non-smokers diary cards. Although the study was not powered to detect differences in exacerbation rates as this was a secondary end point, inclusion of the data was considered important as there was a trend to a difference in exacerbation rates in smokers and non-smokers on low dose beclomethasone but not on the high dose. All participants who experienced exacerbations continued in the study and recovered within 7 days. No serious adverse events occurred during the study.

DISCUSSION

This study has shown that smokers with asthma had a reduced therapeutic response to inhaled corticosteroids over a 3 month period compared with non-smokers. Furthermore, we established that this reduced response is particularly marked in those receiving a low dose of inhaled corticosteroid. Smokers with asthma receiving 400 μg daily inhaled beclomethasone did not show any improvement in morning PEF, in contrast to non-smokers. The greater number of mild exacerbations of asthma experienced by smokers on lower dose inhaled beclomethasone supports our conclusion of corticosteroid insensitivity in smokers with asthma.

Participants receiving 2000 μg daily inhaled beclomethasone had improved morning PEF measurements, although the change was greater for non-smokers. We did not find any difference between the outcome measures of the smoking and non-smoking asthmatic groups at this higher dose of medication. The improvement in the morning PEF level of smokers may indicate that their insensitivity to corticosteroid medication is being overcome by the administration of higher dose medication. The similar numbers of exacerbations of asthma between smokers and non-smokers receiving this higher dose supports this hypothesis. However, the lack of any demonstrable difference between the smoking and non-smoking asthmatics receiving 2000 μg daily inhaled beclomethasone must be interpreted with caution as the subgroups included fewer patients than originally planned and the interaction test to assess whether the effect of smoking was different for the two doses of treatment was not significant.

These findings extend the results from two short term randomised controlled trials that found the efficacy of corticosteroid treatment was impaired in smokers with chronic asthma.^6,7 In the study by Chaudhuri et al,⁷ after 2 weeks of oral prednisolone active cigarette smokers with more severe asthma than those included in the current study had a significantly reduced therapeutic response compared with non-smokers with asthma. Similarly, Chalmers et al⁶ reported a reduced benefit from inhaled fluticasone administered for 3 weeks in smokers with asthma compared with non-smokers. A post hoc analysis of a randomised clinical trial comparing inhaled budesonide or theophylline noted reduced efficacy of inhaled budesonide among smokers with asthma compared with non-smokers with asthma.²¹ In this study we confirmed that smokers with mild asthma have an impaired response to inhaled corticosteroids at a low dose. In addition, the findings suggest that this insensitivity may be overcome by inhaled corticosteroids given at a higher dose after 12 weeks.

This study included all subjects with a physician’s diagnosis of asthma because these are treated for mild asthma in the community. It can be difficult to distinguish smokers with asthma from those with early chronic obstructive pulmonary disease (COPD) who might be less responsive to inhaled corticosteroids. However, smokers with asthma fulfilled the clinical criteria for the diagnosis of asthma and had mean baseline post-bronchodilator FEV₁ of 91% predicted and mean baseline post-bronchodilator FEV₁/FVC of 74%. These spirometric values are well above the upper range for mild COPD.²² Taken together, we believe that smokers with asthma are clearly distinct from subjects with COPD.

It is interesting that smokers with asthma were less likely to be atopic than non-smokers, although the level of atopy was still higher than that generally seen in healthy volunteers. In a study of risk factors associated with asthma and the onset of cigarette smoking, asthma that developed before starting smoking was associated with atopy whereas asthma that developed after starting smoking was associated with a lower FEV₁.²³ The influence on asthma phenotypes in relation to the timing of smoking onset needs to be studied further.

Compliance with measuring PEF can be a problem in all clinical trials. We attempted to minimise this by emphasising the importance of diary card completion. In future studies electronic recording of PEF may provide more reliable evidence of measurements.

The main implication for asthma management from this study is that smokers with mild persistent asthma do not benefit in the medium term from low dose inhaled corticosteroid treatment compared with non-smokers. The smokers with asthma showed some benefit from high dose inhaled corticosteroid therapy, suggesting relative corticosteroid insensitivity. This study has not compared any intermediary dose of inhaled corticosteroid or additional therapy. However, if smokers with asthma require high doses of inhaled corticosteroids to produce a therapeutic effect, then this puts them at risk of developing long term adverse effects from inhaled corticosteroid treatment.⁴ Smokers should be encouraged to stop smoking, although it is unknown whether smoking cessation will restore corticosteroid responsiveness in asthma. This group of patients may require alternative or additional anti-inflammatory drug treatment.