Introduction Smoking increases the risk of asthma and reduces lung function among subjects with and without asthma. We assessed the effects of smoking on lung function reflecting both central and small airways among adults with newly onset asthma.
Methods In a population-based study, 521 (response rate 86%) working-aged adults with clinically defined newly diagnosed asthma answered a questionnaire on personal smoking and other factors potentially influencing lung function, and performed spirometry. We applied multiple linear regression analysis to estimate the relations between smoking and lung function adjusting for confounding.
Results Among asthmatics, FEV1 level was reduced significantly, on average 208 mL, related to regular smoking (adjusted effect estimate −0.208, 95% CI −0.355 to −0.061) and 245 mL in relation to former smoking, that is, among those who quit less than a year ago (−0.245, 95% CI −0.485 to −0.004). In contrast, FEV1 was not significantly related to occasional smoking or former smoking among those who quit over a year ago. Forced expiratory flow (FEF) levels (L/s) were also significantly reduced among regular smokers (FEF25–75%: −0.372, 95% CI −0.607 to −0.137; FEF50%: −0.476, 95% CI −0.750 to −0.202). An exposure–response pattern related to both daily smoking rate and lifetime cumulative smoking was seen both among men and women.
Conclusions This study provides new evidence that among working-aged adults with new asthma, regular smoking and former smoking reduce lung function levels with a dose–response pattern. The lung function parameters applied as outcomes reflect both larger and smaller airways.
- lung function
- small airways
- bronchodilation test
This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/.
Statistics from Altmetric.com
This study aims to assess the effects of smoking on pre-bronchodilator and post-bronchodilator lung function measures that reflect both larger and smaller airways in adults with newly onset asthma.
Among adults with newly onset asthma, lung function measures (FEV1, FEF25–75%) were significantly reduced in current regular smokers and in recent former smokers (ie, quit less than a year ago) when compared with never smokers; furthermore, dose–response patterns were found between daily smoking rate or life-time cumulative smoking and lung function measures.
This study is the first to clarify the effects of smoking on smaller airways lung function, measured as forced expiratory flows, in a larger sample of patients with adult-onset asthma, and to provide effect estimates for lung function parameters measured both prior to and after the bronchodilation test.
Smoking is a major determinant of morbidity and mortality worldwide. It causes many chronic diseases, including coronary heart disease and chronic obstructive pulmonary disease (COPD). Tobacco smoke is a mixture of 4000 different compounds, many of which are carcinogenic or irritative.1 Cigarette smoking is still increasing in many low-income, middle-income and even high-income countries, especially among women.2
Jayes and colleagues summarised recently the evidence of the effect of smoking on the development of asthma among adults.3 The summary-effect estimate, based on eight studies published from 1985 to 2013, was 1.61 (95% CI 1.07 to 2.42). In our population-based incident case–control study of adult-onset asthma, the risk of developing asthma was significantly higher among current smokers, with an adjusted OR of 1.33 (95% CI 1.00 to 1.77), and among ex-smokers, with an adjusted OR of 1.49 (1.12 to 1.97) compared with never-smokers.4 Among current smokers, the risk increased up to 14 cigarettes per day, and a similar trend was observed in relation to cumulative smoking. Asthma is among the most common chronic diseases in working-aged adults. Thus, it would be important to address whether smoking has adverse effects on lung function in adults with asthma. In our systematic search, one previous study had addressed the relation between smoking and lung function in asthmatic adults. Little is known about the effects of tobacco smoke on small airways, especially among smokers with adult-onset asthma.
We assessed the relations between current, former and life-time cumulative smoking and lung function among adults with newly onset asthma. In addition to basic forced expiratory volumes (FEV1 and FVC), our lung function outcomes of interest included also mean forced expiratory flow between 25% and 75% of FVC (FEF25–75%) and forced expiratory flow at 50% of FVC (FEF50%), as estimates of small airways function.
This is a cross-sectional study based on a population-based case–control study of incident asthma, the Finnish Environment and Asthma Study (FEAS).4–13 The study population for this substudy included 521 working-aged adults with newly onset asthma. The ethics committees of the Finnish Institute of Occupational Health and Tampere University Hospital approved the study.
Definition and recruitment of asthma cases
The new cases of asthma were recruited at all healthcare facilities diagnosing asthma in the study area, that is, Pirkanmaa District, including the university hospital clinic, offices of the private-practising pulmonary physicians and public healthcare centres. In addition, the National Social Insurance Institution of Finland invited to participate all patients who had received reimbursement rights for asthma medication in this area during the study period and who had not yet participated. The following diagnostic criteria for asthma were applied for all participants: (1) presence of at least one asthmatic symptom (cough, phlegm production, wheezing and/or breathlessness) and (2) demonstration of reversibility in airways obstruction in lung function investigations.5 These criteria were compatible with the Finnish criteria required for the diagnosis of asthma at the time of the study.14 We confirmed the date and criteria of the asthma diagnosis for all subjects from their medical records. In addition, we also checked that they did not have any previous asthma diagnosis or long-term use of asthma medications, as our interest was in adult-onset asthma. Eligible subjects were asked to sign an informed consent. Altogether, 521 subjects (response rate 86%) had verified asthma, and 486 of these had both spirometry and complete questionnaire information on exposures, and they formed the present study population.
Information on smoking was collected with a self-administered questionnaire inquiring about current and former smoking, smoking rate, and duration and quitting of smoking. The following questions were inquired: Do you smoke currently? (yes, regularly; yes, occasionally; no, I quit less than 12 months ago; no, I quit over 12 months ago; no, I never smoked regularly); How many years have you smoked?; How much do you/did you smoke on average? (cigarettes, cigars and pipefuls per day, or per week for occasional smokers).
The self-administered questionnaire has been described in detail elsewhere.4–13
Lung function measurements
We applied the same lung function protocol to all patients with suspected asthma, including baseline spirometry with a bronchodilation test and 2 weeks of peak expiratory flow follow-up with morning and evening measurements.4–13 Spirometry was recorded before and after bronchodilating medication with a pneumotachograph-type disposable flow transducer connected to a computer (Medikro 905; Medikro, Kuopio, Finland). Measurements were conducted according to the standards of the American Thoracic Society15 at the time of the study. We judged potential presence of obstruction using the reference values derived from the Finnish source population, as described elsewhere.16
Our outcomes of interest were lung functions of (1) larger airways, measured as FEV1 and FVC, and (2) smaller airways, measured as mean forced expiratory flow between 25% and 75% of FVC (FEF25–75%) and forced expiratory flow at 50% of FVC (FEF50%). These were measured before and after the bronchodilation test. We applied multiple linear regression to estimate the relations between the current and previous smoking and cumulative smoking and the average lung function levels. First, we adjusted for the three core covariates: age, sex and height. We built the full model by adding into the model also exposure to dampness and moulds, exposure to secondhand smoking and education as an indicator of socioeconomic status, in addition to the three core covariates mentioned before. We have previously described variables dampness and moulds5 17 and secondhand smoking in detail.8 We fitted a model with occasional and regular smoking and previous smoking quit recently and earlier. We also fitted a model with both current smoking rate and cumulative lifetime smoking. In both analyses, the reference category for the current analyses consisted of never-smokers with adult-onset asthma. We estimated the relations of interest separately for men and women.
Characteristics of the study population and exposures
Characteristics of the study population including smoking behaviour are presented in table 1. A total of 27.6% of the study subjects were current smokers, 5.8% occasional smokers and 25.5% ex-smokers. Among ex-smokers, 6.2% had quit smoking less than 12 months ago. We used post-bronchodilator spirometry values (FEV1:FVC<0.7) to define airflow obstruction. Among the 390 subjects who underwent post-bronchodilator spirometry, the prevalence of airflow obstruction was 19.2%, 27.0% in men and 15.0% in women.
Smoking and lung function in larger airways
In the analyses adjusting for all covariates mentioned and including current regular, occasional and previous smoking, pre-bronchodilator FEV1 level was reduced significantly, on average 208 mL, related to current regular smoking (effect estimate −0.208, 95% CI −0.355 to −0.061) and 245 mL in relation to former smoking quit less than a year ago (−0.245, 95% CI −0.485 to −0.004) (table 2). However, FEV1 was not related to occasional smoking only or to former smoking when quit over a year ago. FEV1:FVC ratio showed a similar pattern to that observed in FEV1, with reduced levels related to regular and former smoking (table 2). The effect estimates related to current regular smoking were greater among men than among women. Among men, FEV1 levels were on average 235 mL lower in current regular smokers compared with never smokers. Among women, such effect on was 151 mL. FEV1:FVC was significantly reduced among both men and women smokers, while the differences in FVC levels were small. Online supplementary table 1 shows that the effect estimates for smoking based on post-bronchodilator lung function levels were slightly smaller for FEV1, but larger for FEV1:FVC compared with corresponding pre-bronchodilator effect estimates.
Table 3 elaborates potential dose–response patterns for the effects of smoking on larger airways lung function parameters. Both daily smoking rate and cumulative life-time smoking show dose–response effect estimates on FEV1 and FEV1:FVC. In the total population, a significant reduction of 168 mL in FEV1 and 3.9% in FEV1:FVC was seen per smoking rate of 10 cigarettes per day. The effect estimates were somewhat larger among women than among men (FEV1 133 mL vs 84 mL; FEV1:FVC 4.0% vs 2.6%). The overall effects of cumulative smoking were 69 mL per 100 cigarette-years on FEV1 and 1.5% on FEV1:FVC. Again, larger effect estimates were detected among women (table 3). Online supplementary table 2 presents the dose-related effect estimates for post-bronchodilator lung function parameters, and these show a similar pattern of effects as described for pre-bronchodilator lung function values.
Smoking and lung function in smaller airways
Table 4 shows the effects of smoking on lung functions reflecting smaller airways, measured before and after the bronchodilation test. Pre-bronchodilator FEF25–75% was reduced significantly, on average 372 mL/s (–0.372, –0.607 to −0.137), as was pre-bronchodilator FEF50%, on average 476 mL/s (–0.476, –0.750 to −0.202), in relation to regular smoking. The effects estimates were similar among both men and women, although they were statistically significant only among women, probably because they constituted a larger group. The effect estimates calculated for post-bronchodilator values were slightly greater.
There was a significant dose–response pattern between both daily smoking rate and cumulative smoking and FEF25–75% and FEF50%, as shown in table 5. There was evidence of effect among both men and in women, but statistically significant effect estimates mainly in women only.
Our large population-based study included 521 adult-onset asthma cases who were recruited over a 2.5-year study period. A total of 486 of them had acceptable spirometry. This is the first study to address potential effects of active smoking on lung function among subjects with new adult-onset asthma. We estimated the effects of smoking on lung functions measured both before and after the bronchodilation test. A total of 27.6% of asthmatics were current smokers, while 25.5% were former smokers. Among these ex-smokers, 6.2% had quit less than 12 months ago.
Our main finding was that among asthmatics, both current regular smoking and recent former smoking were related to a significantly reduced FEV1 level. The effect estimates were on average 208 mL lower in smokers and 245 mL among former smokers who quit less than a year ago compared with never smokers. Interestingly, there was no indication of adverse effects among subjects who quit over a year ago, suggesting a recovery from the adverse effects of smoking. For FVC and FEV1:FVC, we detected similar associations. A dose–response pattern was detected in relation to both daily smoking rate and life-time cumulative smoking.
We also estimated the effects of smoking on lung function parameters reflecting small airways. Small airways lung function parameters were related to both daily smoking rate and cumulative smoking, showing statistically significant dose–response patterns. The effect estimates for women were consistently greater than estimates for men, indicating that a given amount of smoking harms the lungs of women more than the lungs of men.
Validity of results
We were able to recruit a high proportion of new cases of asthma (response rate 86%) in the study area by a thorough recruitment system through the healthcare system and with the help of the National Social Insurance Institution (NSII). The national social insurance covers the whole Finnish population and the medication files of NSII have practically full coverage of asthmatics requiring regular treatment. Thus, any major selection bias is unlikely in our study.
Questionnaire-based exposure information may include some misclassification. To reduce any information bias, we had introduced the study to the participants as a study on environmental factors and asthma in general, with no special focus on smoking.
Some of the smoking categories, especially ex-smokers who quit less than 12 months ago, were small resulting wide CIs often including the null value. These results should be interpreted with caution.
We were able to adjust for a number of potential confounders including sex, age, height, education and exposure to secondhand smoke, in regression analyses, so we were able to eliminate these factors as potential explanations for our results.
Synthesis with previous knowledge
We have previously shown in FEAS that both personal smoking4 and exposure to environmental tobacco smoke8 increase the risk of developing asthma. A Danish study has reported evidence that the lung function levels among subjects with asthma are in general lower and the average annual decline is greater than among subjects without asthma.18 They discussed that the difference detected could reflect a difference detectable at the time of the diagnosis or an increased decline of lung function over time. They also provided evidence that the effect of smoking on lung function is stronger among subjects with asthma.18 In a cohort study of 5057 subjects initially 18–30 years of age, the average FEV1 decline from the peak levels through the age 40 was 8.5% among never-smokers with no asthma, 10%–11% among individuals without asthma and currently smoking ≥15 cigarettes/day as well as among never-smokers with asthma, and 18% among those with both asthma and currently smoking ≥15 cigarettes/day.19 Based on our systematic search of literature, our study is the first one that assesses potential effects of smoking on lung function among subjects with newly diagnosed asthma. Besides studying the effects on forced expiratory volumes (FEV1 and FVC), we also addressed potential effects on smaller airways, using FEF25–75% and FEF50% as the outcome variables.
Our results suggest that the effects of smoking on lung function start operating already before the asthma is diagnosed. This points towards a phenomenon that the effects of smoking on airways inflammation lead to the onset of asthma and reduction in lung function concurrently. Thus, asthmatics seem to constitute a group especially vulnerable to the adverse effects of smoking on lung function.
Reduced forced expiratory flows have been identified as early effects of smoking on smaller airways lung function among previously healthy subjects. Already in the 1970s, McFadden and Linden20 and Walter et al 21 postulated that small airways obstruction contributes to airflow limitation at mid and late phases of FEVs. In asthma, the clinical implication of reduced mid forced expiratory flows is that they increase the risk for long-term persistence of asthma as well as poor asthma outcomes.22 23 These adverse effects seem to be independent of the magnitude of the reductions in FEV1 or FEV1:FVC. Persistently reduced forced expiratory flows (eg, FEF25–75%) have been postulated to reflect the small airways phenotype of asthma.24 On the other hand, the disadvantages related to these lung function measures includes poorer reproducibility compared with FEV1 and their contribution to clinical decision-making has been questioned.25 In a retrospective analysis of community-managed asthma, FEF25–75% was associated with severity of asthma defined by treatment steps.26
Smoking among asthmatics is associated with poor control and increased severity of asthma, as well as with more rapid decline in FEV1 and poor therapeutic response to corticosteroids.27 This was suggested to be explained by an alteration in airway inflammatory cell phenotypes. We were able to identify only two previous studies on the effects of smoking on small airways lung function among patients with asthma. In the Melbourne Atopy Cohort Study,28 early-life tobacco smoke exposure perinatally was associated with reduced pre-bronchodilator FEV1, FEV1:FVC and FEF25–75%. Boulet et al 29 compared small samples of smoking and non-smoking patients with asthma visiting asthma clinic, and found that smokers had more respiratory symptoms, a lower FEF25–75% and FEV1:FVC ratio and lung diffusing capacity. Induced sputum showed higher neutrophil counts. They concluded that smoking patients with asthma showed features that were similar to those detectable in the early stages of COPD. The present study appears to be the first one to clarify the effects of smoking on small airways lung function, measured as forced expiratory flows, in a larger sample of patients with adult-onset asthma, and to provide effect estimates for lung function parameters measured both prior to and after the bronchodilation test.
The trend that we detected showing a reduced FVC in men who were regular smokers may also indicate the presence of small airways obstruction, as this finding in asthma may be due to airway closure and air trapping.30 In a longitudinal adult asthma study, the accelerated annual decline in lung function among smokers included FEV1, FEV1:FVC and also FVC,31 which is consistent with our findings. In a review of pulmonary function tests of patients with newly diagnosed asthma, 8% had a true restrictive impairment.32
This study provides new evidence that regular smoking is related to reduced lung function levels among working-aged adults with newly onset asthma. The study also shows that smoking has harmful effects on lung function parameters reflecting both central and small airways measured both before as well as after the bronchodilation test. Observations of consistent dose–response patterns in the detected harmful effects strengthen the causal inference based on the findings and provide a strong message for both clinicians and public health professionals that those who have asthma should be advised and supported not to take up smoking or to quit if they already have this habit.
Contributors JJKJ participated in conception and design of the study, acquisition of data, planning analyses and data interpretation, and wrote the article and serves as the corresponding author and the guarantor of the paper, taking responsibility for the integrity of the work as a whole. PS participated in data analysis and data interpretation and drafted the article. TKL analysed the data and participated in data interpretation and critical revision of the article. SH participated in data analysis and interpretation of data and critical revision of the manuscript. LPM provided expertise in clinical respiratory physiology. MSJ participated as the PI of the project in conception and design of the study, acquisition of data, planning analyses and data interpretation, and critical revision of the manuscript. All authors approved the final version of the manuscript to be submitted.
Funding This study was funded by the Academy of Finland (grants no. 138691 and no. 266314, and grant no. 129419 of SALVE research programme), the Ministry of Social Affairs and Health of Finland (grant no. STM/1523/2012), the Finnish Anti-Tuberculosis Association, Väinö and Laina Kivi Foundation, and Oulu University Hospital Research Fund.
Disclaimer The sponsors had no influence on the analysis or reporting of the results.
Competing interests None declared.
Patient consent for publication Not required.
Ethics approval The ethics committees of the Finnish Institute of Occupational Health and the Tampere University Hospital approved the study.
Provenance and peer review Not commissioned; externally peer reviewed.
Data sharing statement No additional data are available.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.