Effects of smoking cessation on lung function and airway inflammation
in smokers with asthma
Rekha Chaudhuri l, Eric Livingston l, Alex D McMahon 2, Jane Lafferty l,
Iona Fraser 3, Mark Spears l, Charles P McSharry 3, Neil C Thomson l
Departments of Respiratory Medicine l and Immunology 3, University of Glasgow
and Robertson Centre for Biostatistics2, Glasgow, UK
Professor Neil C Thomson
Department of Respiratory Medicine
Division of Immunology, Infection and Inflammation
University of Glasgow and Western Infirmary
Glasgow, G11 6NT Scotland, UK
Telephone: 44-141-211-3241 Fax: 44-141-211-3464
E- mail: email@example.com
Funded by: Asthma UK, Scottish Council for Postgraduate Medical and Dental
Education and Chief Scientist Office, Chest Heart and Stroke, Scotland.
Running head: Smoking cessation in smokers with asthma
Descriptor number: 64
Word count: 3035
"This article has an online data supplement for the methods section, which is accessible
from this issue's table of content online at www.atsjournals.org"
AJRCCM Articles in Press. Published on April 27, 2006 as doi:10.1164/rccm.200510-1589OC
Copyright (C) 2006 by the American Thoracic Society.
Rationale: Active smoking in asthma is associated with worsening of symptoms, accelerated
decline in lung function and impaired response to corticosteroids.
Objectives: To examine the short-term effects of smoking cessation on lung function, airway
inflammation and corticosteroid responsiveness in smokers with asthma.
Methods and measurements: Smokers with asthma were given the option to quit or continue
smoking. Both groups underwent spirometry and induced sputum at baseline and at one, three
and six weeks. Cutaneous vasoconstrictor response to topical beclometasone, airway response to
oral prednisolone and sensitivity of peripheral blood lymphocytes to corticosteroids were
measured before smoking cessation and at six weeks.
Main results: Of 32 subjects recruited, 11 opted to continue smoking (smoking controls).
Twenty-one subjects opted for smoking cessation of whom ten quit smoking for six weeks (quit
group). Comparing quitters with controls at six weeks, the mean (Cl) difference in FEV1 was
407 ml (21,793), p=0.040; and the proportion of sputum neutrophils reduced by -29 (-51, -8),
p=0.039. Total cutaneous vasoconstrictor response score to topical beclometasone improved
following smoking cessation with a mean (CI) difference of 3.56 (0.84, 6.28), p=0.042
comparing quitters with control smokers. There was no change in airway corticosteroid
responses after smoking cessation.
Conclusions: Six weeks after smoking cessation, smokers with asthma achieved considerable
improvement in lung function and a fall in sputum neutrophil count compared to smokers who
continued to smoke. These findings highlight the importance of smoking cessation in asthma.
Word count: 236
Key words: Smoking cessation; smoking; asthma; lung function; airway inflammation
Active cigarette smoking has detrimental effects on asthma morbidity. Compared to non-
smokers with asthma, smokers have more severe symptoms (1, 2), increased rates of
hospitalization (3), accelerated decline in lung function (4, 5) and impaired therapeutic responses
to inhaled (6, 7) and oral corticosteroids (8). Furthermore, the prevalence rates for cigarette
smoking in individuals with asthma are similar to that of the general population (9), and in many
developed countries over 20% of adults with asthma are active smokers (1, 2, 9-12). Particularly
high rates have been noted in adults presenting to hospital emergency departments with acute
asthma (3). There is limited information on the effects of cigarette smoking on airway
inflammation in asthma (13). Sputum neutrophil counts are reported to be increased in heavy
smokers with mild asthma compared to non-smokers with asthma (14), and sputum
concentrations of cytokines such as IL-8 are raised (14) and others such as IL-18 are suppressed
(15) in smokers with asthma.
Despite its importance, there is limited published data on the effects of smoking cessation on
symptoms, lung function and corticosteroid responsiveness in smokers with asthma. In an
uncontrolled study, seven subjects who quit smoking for a week showed improvements in
symptoms and lung function (16). In a longer term study, a prospective cohort of smokers with
asthma reported improvements in symptoms and bronchial hyperreactivity after four months of
smoking cessation (17). In ten subjects with asthma who were ex-smokers for at least a year, the
therapeutic response to oral corticosteroids was midway between that of current smokers and
never smokers; suggesting that smoking cessation may partially restore corticosteroid
The effect of smoking cessation on airway inflammation in healthy smokers, shows a dose-
dependent relationship between smoking and airway inflammation (18). In these subjects, two
months smoking reduction lead to a significant fall in bronchoalveolar lavage neutrophil and
macrophage counts (19). In contrast to the improvement in airway inflammation found in
healthy smokers after smoking cessation, there is minimal change in airway inflammation in
patients with COPD after quitting smoking (20-22), and the effect of smoking cessation on
airway inflammation in smokers with asthma is not known.
Our hypothesis is that smoking cessation improves lung function, reduces airway inflammation
and restores corticosteroid responsiveness in smokers with asthma. The aim of the study was to
prospectively study smokers with asthma who successfully quit smoking and compare outcome
measures with asthmatic smokers who continue to smoke; measurements included lung
function, asthma control score, induced sputum cell counts and mediator levels. A secondary
aim was to assess the airway, cutaneous and lymphocyte sensitivity to corticosteroids after
smoking cessation. Some of the results have been reported in abstract form (23).
Smokers with asthma [ATS criteria, (24)] aged 18-60 years, baseline FEV1 ≤85% predicted, ≥
15% reversibility of FEV1 after nebulised albuterol, a smoking history of ≥ 10 pack years and
currently smoking ≥ 10 cigarettes/day were recruited. The study was approved by the West
Glasgow Ethics Committee.
This was a prospective, controlled study. Airway corticosteroid sensitivity to oral prednisolone,
40 mg daily for 14 days, was assessed in all subjects by change in pre-albuterol FEV1, asthma
control score, morning and evening PEF, daily morning and night symptoms and use of β-
agonist inhalers. Subjects were recalled to discuss smoking cessation, 2-12 months after the
corticosteroid trial. Spirometry, induced sputum, exhaled nitric oxide (FENO), exhaled carbon
monoxide (CO), asthma control score, skin vasoconstrictor response to beclometasone, serum
cotinine and peripheral blood lymphocyte proliferation assay (LP) were performed. All subjects
were given the option to quit or continue smoking. Both groups were issued diary cards. Visits
were arranged one, three and six weeks later. Following the six-week visit, both groups
received a second course of oral prednisolone 40mg daily for 14 days, with similar end-points
used to compare results. Spirometry, exhaled gases and asthma control score were recorded at
all visits; induced sputum at three and six weeks; cutaneous vasoconstrictor test and LP at six
weeks. Patient adherence to smoking cessation was accepted by patient history.
ATS asthma impairment score (25) and a validated asthma control questionnaire (26) were
recorded. Patients maintained a validated diary card (27), recording morning and night PEF
(Mini-Wright, Clement Clarke, Harlow, UK), daytime symptoms, night awakenings, use of
inhaled rescue medication and study tablet consumption. Spirometry was measured with a dry
spirometer (Vitalograph Ltd, Buckingham, UK). Sputum was induced as previously described
(14, 28, 29). FENO and CO were measured using a chemiluminescence analyzer (Logan
Research Ltd, Rochester, UK) (30) and serum cotinine by enzyme immunoassay (Cozart
Bioscience Ltd, Abingdon, UK). Treatment compliance was assessed by tablet count.
Cutaneous vasoconstrictor response to topical beclometasone was measured as described
previously (31, 32). Concentrations of 0, 1, 3, 10, 30, 100, 300 and 1000 µg/ml were applied to
the skin in a random double-blind manner and the degree of blanching assessed visually after 18
hours by a single trained observer. Blanching at each concentration was graded according to a 4-
point scale [0-3] and a total score calculated [0-24]. Sensitivity of peripheral blood T-
lymphocytes to glucocorticoids was assessed in a functional assay as described previously (33).
The percentage suppression at the final concentration of corticosteroid used was defined as the
maximum inhibitory dose [Imax %].
Baseline characteristics were compared by chi-squared, Wilcoxon and t-tests. Response to
smoking cessation on lung function, diary data, induced sputum, mediator levels, skin
vasoconstrictor response and oral prednisolone trials for smokers versus never smokers with
asthma was assessed by Analysis of Covariance models that adjusted each factor by its baseline
measure. Correlations were assessed by Spearman’s rank correlations. Only those subjects who
completed the study up to the six week visit were included in the analysis. Significance at a level
of 5% was considered for the primary end-point; the change in FEV1 at six weeks. Tests were
performed using SAS version 9.0 (Cary, NC, USA).
The flow of participants through the study is depicted in Figure 1. 11/32 subjects opted to
continue smoking (smoking controls). Of the 21 who opted to attempt quitting smoking, 14
achieved 1 week off and 10 completed six weeks of smoking cessation.
There were no differences in age, duration of asthma, equivalent dose of inhaled beclometasone,
asthma control score, ATS impairment score, IgE levels, spirometry, skin vasoconstrictor test
score, inhibitory dose of dexamethasone maximally suppressing lymphocyte proliferation in
vitro, induced sputum cell counts and inflammatory mediators between the smoking control and
quit group. Mean (SD) baseline FEV1 percentage predicted was 68 (11) in the quit group and 64
(94) in the controls [Tables 1 and 2]. Baseline reversibility of FEV1 to inhaled albuterol [mean%
(SD)] was 25 (11) in the quit group and 19 (6) in the control group. Age, gender distribution,
atopic status, smoking history, duration of asthma and serum cotinine levels were similar in the
successful quitters and those subjects who opted for the quit group but were unable to achieve a
week off smoking (n=7). In the 10 subjects who successfully quit smoking and completed the
study, five used nicotine patches, one subject used acupuncture and four quit without any aid.
Change in lung function, asthma control and airway inflammatory markers with smoking
Lung function: The mean (SD) change in FEV1, ml in the quit group was 356 (278) at 1 week,
390 (311) at 3 weeks, and 450 (471) at 6 weeks of smoking cessation, p values 0.015, 0.009 and
0.031 respectively. There was no difference in the equivalent FEV1 measures in the smoking
control group. Comparing quitters with control smokers at six weeks’ cessation, there was a
mean improvement of 407 ml in FEV1, 15.2% in FEV1% predicted and 93L/min in PEF (Figure
2, Table 3). There was no correlation with the improvement in FEV1 and age, gender, atopy,
duration of smoking, pack years smoked, dose of inhaled corticosteroids and sputum cell counts.
Asthma control: The mean (CI) asthma control score showed an improvement at six weeks after
smoking cessation of -1.2 (-2.2,-0.2), p=0.021 (Table 3).
Induced sputum: Quitters showed a decrease in proportion of sputum neutrophils compared to
controls at six weeks [mean % difference (CI) -29 (-51, -8), p=0.013; with no change in
inflammatory mediator levels (Table 4). The mean (CI) difference for absolute neutrophil
count in sputum at three weeks was -52 (-141, 36), p=0.224 and at six weeks cessation it was
-121 (-215, -26), p=0.017. The degree of improvement in FEV1 was not related to the
baseline neutrophil level [r= 0.24, p=0.248] or the change in neutrophil proportion [r= -0.11,
p=0.819]. There was no difference in mediator response in those quitters who used nicotine
replacement therapy compared to those who did not.
Exhaled gases: Compared to the control smokers the quit group showed a reduction in exhaled
CO at one, three and six weeks. (Table 3). There was no change in exhaled NO levels.
Response to corticosteroids
Airway response: There was no difference in the FEV1 response to oral prednisolone while
smoking or at eight weeks after smoking cessation in the control smokers with asthma compared
to quitters. The mean (CI) change in FEV1, ml with the first prednisolone course when subjects
were smoking was 105 (-231,441), p=0.518 for the quit group versus smoking control group and
after eight weeks of smoking cessation, the difference was 190 (-121, 501), p=0.214.
Compliance with medication was 89% in controls and 100% in the quit group for the first
corticosteroid course and 89% in controls and 100% in quitters with the second course.
Cutaneous vasoconstrictor response to topical beclometasone: The mean (CI) total cutaneous
vasoconstrictor response score to topical beclometasone improved following smoking cessation
with a mean difference of 3.56 (0.84, 6.28), p=0.014; comparing quitters with control smokers.
Lymphocyte proliferation response: There was no difference in the lymphocyte proliferation
response after six weeks of smoking cessation. The mean difference (CI) Imax % for
dexamethasone suppression of lymphocyte proliferation was -9.3 (-28.4, 9.9), p=0.322 for
quitters compared to smokers and for prednisolone suppression was -4.0 (-17.1, 9.0); p=0.522.
Active cigarette smoking is known to worsen the severity of asthma (13). In this study we have
demonstrated that lung function of smokers with asthma improves by a considerable degree
within a week of stopping smoking. This improvement in FEV1 [mean (CI)] at one week was of
the order of 396 (129,664) ml compared to those that continued to smoke and this improvement
increased up to six weeks after smoking cessation. In addition, there was a fall in the sputum
neutrophil counts, evident at six weeks after smoking cessation.
Subjects in the quit group progressed through the study unless they admitted to smoking. This
method of assessment of smoking status is known to overestimate the number of subjects who
effectively quit smoking. Due to the short half-life, only daily measurements of carbon
monoxide or cotinine could have provided objective evidence for smoking cessation in between
clinic visits. Cotinine can be affected by nicotine replacement therapy used for cessation and
daily measurements for eight weeks were not feasible. However the reduction in exhaled carbon
monoxide in the quit group compared to controls at all visits provided objective evidence for
effective smoking cessation, at least for the previous 24-48 hours prior to each clinic
assessment. If some of the subjects in the quit group had not stopped smoking completely, the
results obtained may underestimate the extent of the improvement in lung function and
neutrophil count possible with smoking cessation.
The improvement in lung function seen following smoking cessation was clinically significant.
This demonstrates that there is a reversible component to the harmful effects of smoking on the
airways in asthma. The degree of improvement noted by smoking cessation far exceeds that of
high dose anti-inflammatory treatment, such as oral prednisolone 40 mg daily for 2 weeks,
which had no effect on lung function in smokers in our current study and in our previous work
(8). The improvement in lung function could be due to removal of the acute bronchoconstrictor
effects of cigarette smoke (34) or a reduction in the pro-inflammatory effects of cigarette smoke
on the airways (14). It is also possible that stopping smoking leads to a reduction in
corticosteroid insensitivity, as ex-smokers who had quit smoking for at least one year have a
better response to oral prednisolone compared to current smokers (8). Reduced histone
deacetylase activity is one of the possible mechanisms of corticosteroid insensitivity in smokers
(34), but there are no reports of measurement of HDAC levels following smoking cessation.
In asthma, there have been two previous studies looking at the effect of smoking cessation on
lung function. The first was a small uncontrolled study where 7 subjects quit smoking for a week
and showed improvements in PEF, symptoms and specific airways conductance (16). There was
a reduction in histamine airway responsiveness after 24 hours of smoking cessation (16). A
prospective cohort study compared the effects of smoking cessation, smoking reduction and
continuing smoking on asthma control and biomarkers of exposure to cigarette smoke (17). This
study showed a significant improvement in symptoms and bronchial hyperreactivity after four
months of smoking cessation, but no change in FEV1. The baseline FEV1 % predicted was much
higher than in our subjects and hence it is possible that scope for further improvement was
In our study, although the asthma control score improved, we did not obtain improvement in
symptoms in the diary card up to two months after cessation. In a study by Hillerdahl and
Rylander, two thirds of subjects with asthma who quit smoking reported no improvement or
worsening of symptoms after smoking cessation (35). A cross-sectional study of 3197
smokers, ex-smokers (stopped smoking for at least one year) and non-smokers with asthma,
reported that ex-smokers had significantly less cough, wheeze, night symptoms and sputum
production compared to current smokers, but a similar degree of shortness of breath (36). This
implies that many, but not all asthma symptoms can return to the level of never-smokers with
The raised sputum neutrophil count found in high intensity smokers with asthma (14, 37)
may be partly responsible for their reduced responsiveness to corticosteroids. Unlike
eosinophils, which are exquisitely sensitive to corticosteroids, neutrophils are poorly
responsive to corticosteroid therapy (37) and their survival and proliferation is promoted by
glucocorticoids (38). We demonstrated a reduction in induced sputum neutrophils with
smoking cessation, but no change in mediator levels. We are unaware of any previous
published studies that have measured airway inflammation in smokers with asthma after
smoking cessation. In healthy heavy smokers bronchoalveolar lavage neutrophil counts were
reduced two months after smoking reduction (19). In contrast, the effect of smoking cessation
on airway inflammation in COPD has shown that most inflammatory cells, including
neutrophils, persist in ex-smokers (20, 22, 39) and can even increase (40). The mechanisms
by which the neutrophil count is reduced following smoking cessation are unclear at present.
It seems unrelated to the continued presence of neutrophil chemo-attractants LTB4 and IL-8
in the airways, and may therefore be related to reduced diapedesis following the down-
regulatory effects of smoking cessation on adhesion molecules (41), perhaps in addition to
more efficient apoptosis of neutrophils by alveolar macrophages after smoking cessation (42,
The short-term improvement in FEV1 was not related to the change in neutrophil count, which
might suggest that the removal of bronchoconstrictor effects of cigarette smoke is more
important than neutrophils themselves; but the small number of subjects studied may explain our
findings. The levels of the inflammatory mediators in induced sputum did not change at six
weeks following smoking cessation. This result is similar to the finding in COPD, where IL-8
levels in sputum were similar in smokers and ex-smokers (44) and chronic bronchitis, where
BAL levels of TNF-alpha were unaltered by smoking cessation (22). It is possible that clinical
improvement may precede changes in inflammation, as in the third national health and nutrition
examination survey (NHANES III), inflammatory markers (CRP, white cell count, fibrinogen
and albumin) resolved more slowly than traditional cardiovascular risk factors (45).
Smoking cessation did not improve corticosteroid responsiveness measured by changes in FEV1
and PEF. However, there was a large improvement in baseline lung function with smoking
cessation and it is possible that this reduced the scope for further improvement in lung function,
after corticosteroids. In a previous study, ex-smokers with asthma who had quit smoking for at
least a year showed an improvement in morning PEF, but not FEV1 after high dose oral
prednisolone compared to placebo (8).
The skin vasoconstrictor response to topical beclometasone (32, 46) is based on the ability of
corticosteroids to cause transient vasoconstriction and skin blanching, has been used as a
screening test to determine the relative anti-inflammatory potency of inhaled corticosteroids (47)
and as an index of systemic sensitivity to corticosteroids (31). Objective methods of detecting
glucocorticoid-induced skin blanching have been compared to the visual scoring system, but the
human eye has been found to be the most sensitive tool to measure dermal blanching (32). The
cutaneous vasoconstrictor test for corticosteroid sensitivity showed an improvement after six
weeks of cessation implying some restoration in peripheral corticosteroid sensitivity within this
period. The mechanism for skin vasoconstriction with corticosteroids is not known, but a
possible explanation is that glucocorticoids increase the sensitivity of vascular smooth muscle to
the vasoconstrictor effects of noradrenaline (48). The blanching is possibly mediated through
glucocorticoid receptors, since oral administration of RU 486, a glucocorticoid receptor
antagonist, can abolish this response (49). Smoking causes vascular dysfunction but the additive
effect of nicotine on corticosteroids has not been studied in skin vasculature. The lymphocyte
proliferative response did not alter with smoking cessation in our study. A lack of correlation
between different tests of tissue sensitivity to corticosteroids has been reported previously in
health human volunteers (50).
In conclusion, in smokers with asthma, improvement in lung function occurs as early as one
week after smoking cessation with a further improvement up to six weeks. There is a reduction
of sputum neutrophil percentage after six weeks of smoking cessation but no change in common
inflammatory mediator levels. These findings highlight the importance of smoking cessation in
We are grateful to Kathy McFall, medical illustration department, for assistance with the
1. Siroux V, Pin I, Oryszcyn MP, Le Moual N, Kauffmann F. Relationships of active
smoking to asthma and asthma severity in the EGEA study. Eur Respir J 2000;15:470-477.
2. Althuis M, Sexton M, Prybylski D. Cigarette smoking and asthma symptom severity
among adult asthmatics. J Asthma 1999;36(3):257-264.
3. Silverman RA, Boudreaux ED, Woodruff PG, Clark S, Camargo CA, Jr. Cigarette
smoking among asthmatic adults presenting to 64 emergency departments. Chest
4. Apostol G, Jacobs D, Tsai A, et al. Early life factors contribute to the decrease in lung
function between ages 18 and 40. Am J Respir Crit Care Med 2002;166:166-172.
5. Lange P, Parner J, Vestbo J, Schnohr P, Jensen G. A 15 year follow-up study of
ventilatory function in adults with asthma. N Engl J Med 1998;339:1194-1200.
6. Chalmers GW, Macleod KJ, Little SA, Thomson LJ, McSharry CP, Thomson NC.
Influence of cigarette smoking on inhaled corticosteroid treatment in mild asthma. Thorax
7. Tomlinson JEM, McMahon AD, Chaudhuri R, Thompson JM, Wood SF, Thomson
NC. Efficacy of low and high dose inhaled corticosteroid in smokers versus non-smokers
with mild asthma. Thorax 2005;60(4):282-287.
8. Chaudhuri R, Livingston E, McMahon AD, Thomson L, Borland W, Thomson NC.
Cigarette smoking impairs the therapeutic response to oral corticosteroids in chronic
asthma. Am J Respir Crit Care Med 2003;168(11):1308-1311.
9. Lemiere C, Boulet L-P. Cigarette smoking and asthma: A dangerous mix. Can Respir J
10. Sippel JM, Pedula KL, Vollmer WM, Buist AS, Osborne ML. Associations of smoking
with hospital-based care and quality of life in patients with obstructive airway disease.