Introduction

Tobacco smoking is the leading cause of preventable premature mortality in the world [1]. Indeed, tobacco smoke contains multiple toxic substances with inflammatory, cytotoxic, mutagenic, and carcinogenic properties. These substances produce direct cellular injuries and chemical reactions, with release of oxygen radicals. Oxidative stress activates epithelial cells which produce inflammatory mediators, and in turn activate neutrophils, macrophages, and cytotoxic T cells. In addition, cigarette smoke impairs innate immune responses mediated by epithelial cells, alveolar macrophages, and dendritic cells [2, 3].

On the other hand, asthma is a chronic airway disease, affecting around 7.5% of the adult population, characterized by variable airway obstruction, airway hyperresponsiveness, and airway inflammation [4]. It is important to mention that asthmatic patients start and continue smoking with prevalence rates relatively close to those found in the general population [5]. Smoking prevalence in asthmatic patients is as high as 25% in developed countries [5–7] and one in four asthmatic patients continues to smoke despite their medical condition [8, 9].

In this context, the physiopathological and clinical interactions between smoking and asthma have been analyzed in different studies, suggesting an ‘‘asthma smoking phenotype” [2, 3, 10–14]. Smoking asthmatics have different kinds of bronchial inflammation compared with non-smoking asthmatics [14]. Cigarette smoke induces important changes in the asthma endotype, with a predominance of activated macrophages and neutrophils in the sputum, airways, and lung parenchyma as in early chronic obstructive pulmonary disease [3, 12, 13]. The persistent exposure to cigarette smoke has been shown to drive additive or synergistic inflammatory and remodeling responses (e.g., chronic mucus hypersecretion) in the asthmatic airways, explaining the reported accelerated lung function decline and increased airflow obstruction severity [3, 12, 13].

Cigarette smoking and asthma negatively impact patient’s clinical, prognostic, and therapeutic outcomes, regardless of disease severity [8]. Morbimortality in asthma has been depicted as higher in smokers [6, 10]. Asthmatic smokers present more severe symptoms, a worse quality of life, and a higher number of medical visits and hospital admissions compared with non-smokers or ex-smokers [10, 15]. Corticosteroids’ response is also impaired among smokers [16]. Smoking cessation improves lung function in patients with asthma, demonstrating that the detrimental effect of smoking on the airway is reversible in asthmatic patients [17].

However, all of the results exposed above have not always been consistent, mainly due to substantial methodological differences [18]. Therefore, it is vital to identify differential smoking patterns between asthmatic and non-asthmatic smokers in order to provide the best care strategies.

We designed this study with the following objectives: 1) estimate asthma frequency in smokers attending smoking cessation units, 2) describe asthma features in smokers, and 3) characterize smoking in asthmatic vs. non-asthmatic smokers.

Material and methods

Results

The final sample was composed of 329 smokers. Three of the participating units reached the sample size goal of 68 smokers. Nevertheless, recruitment was stopped because the sample size was sufficient to meet the study objectives.

Regarding sociodemographic characteristics, most smokers were caucasian (98%), 51% were women, with a mean age of 56±10 years old. Thirty-six percent of the participants had a very low educational level (no schooling or elementary school), and 23% were university graduates. We found that 55% of participants were active workers, 75% lived in urban areas, and 58% were physically active (Table 1).

Download:

• PNG
  larger image
• TIFF
  original image

Table 1. Sociodemographic characteristics of the study sample (n = 329).

https://doi.org/10.1371/journal.pone.0294587.t001

Table 2 shows asthma related variables. Among smokers, 18.6% (95% CI [14.3, 24.0]) presented an asthma diagnosis. Asthmatic patients displayed a mean ACT score of 17±5, and 15% reported exacerbations during the last month. The most frequent symptoms were dyspnea (34%), coughing (31%), and wheezing (13%). LABAs were the most frequently prescribed inhaler medications during the previous year (45%), followed by anticholinergics (36%), inhaled steroids (31%), and SABAs (30%). Only three patients (1%) had received biologic therapy (Table 2).

Download:

• PNG
  larger image
• TIFF
  original image

Table 2. Asthma related variables.

https://doi.org/10.1371/journal.pone.0294587.t002

Compared to non-asthmatics (Table 3), asthmatic smokers were mostly women (73.8% vs. 46.0%, p<0.001), significantly younger (50.4 vs. 57.7, p<0.0001), displayed a lower COPD frequency (6.6% vs. 31.7%, p<0.001), and a higher prevalence of anxious-depressive syndrome (54.1% vs. 26.3%, p<0.0001). Up to 13% of the smokers studied had chronic airflow limitation, defined by an FEV1/FVC ratio in spirometry <0.70. Passive exposure to tobacco smoke at home was also more common in asthmatic smokers, although the difference did not reach statistical significance (49.2% vs. 37.9%, p = 0.104). Similarly, no significant differences were observed regarding the pulmonary function tests, nor in the presence of chronic airflow limitation between asthmatic and no asthmatic smokers.

Download:

• PNG
  larger image
• TIFF
  original image

Table 3. Demographic and clinical characteristics in smokers with and without asthma.

https://doi.org/10.1371/journal.pone.0294587.t003

The comparison of smokers with and without asthma showed no significant differences in terms of habit type, age of first cigarette or of smoking onset, and number of cigarettes per day (Table 4). However, non-asthmatic individuals had been smoking for longer than asthmatics (38±10 vs. 31±12, p<0.0001) and had consumed a greater number of packs/year (45±25 vs. 32±20, p = 0.0001). Although 306 patients reported having received smoking cessation treatment, only 263 (80.4%) stated they had tried to go for more than 24 hours without smoking. Even though no differences in smoking cessation attempts were observed between groups, non-asthmatics used smoking cessation treatment more commonly than asthmatics (49% vs. 30%, p = 0.010), mainly varenicline (30% vs. 17%, p = 0.044), had a higher relapse frequency (85% vs 74%, p = 0.029), better support to quit smoking (77% vs. 60%, p = 0.008), and a higher smoking self-efficacy (6±2 vs. 5±2, p = 0.035) (Table 4).

Download:

• PNG
  larger image
• TIFF
  original image

Table 4. Characterization of smoking by asthma status.

https://doi.org/10.1371/journal.pone.0294587.t004

No differences in tobacco dependence, measured by the total FTND score or questions measuring psycho-social dependence, were observed between asthmatics and non-asthmatics. However, there were significant differences in FTND categories (p = 0.046), with higher frequency of the "low dependence" category in asthmatics (19% vs. 8%). The comparison of the individual items did not highlight significant differences between the two groups. Finally, CO and carboxyhemoglobin levels, as well as health resource use (emergency room visits and admissions) were similar in asthmatics and non-asthmatics patients (Table 4).

Discussion

In this project we have found a high asthma frequency in Spanish attending smoking cessation units. We have also described both smoking and asthma, in order to characterize the currently known “asthma smoking phenotype”.

The global frequency of asthma in smokers attending smoking cessation units in Spain is almost 20%. The prevalence of asthma varies widely around the world, ranging from 0.2% to 21.0% in adults, probably due to phenotypic variability, different study populations, and diagnostic criteria [23, 24]. Most of the previous studies were population-based with self-reported or questionnaire-based diagnosis, whereas we have used a highly selected population (smokers attending smoking cessation units) with a diagnosis based on a documented asthma diagnosis evidence. It is worth mentioning that the observed higher frequency of asthmatics among women smokers is consistent with the literature [18, 25–28]. However, despite these limitations and comparison difficulties, these results are critical as both smoking and asthma present pathophysiological mechanisms involved in the development of inflammatory airway diseases and might present synergic or additive effects [29, 30]. In fact, several observational studies suggested that cigarette smoking significantly increases asthma risk and prevalence, especially in females, as women were found to be more vulnerable to the effect of tobacco smoking compared with men [25–28, 31]. Therefore, in daily practice smokers should regularly be evaluated in order to rule out asthma.

The smoking habit in smokers with asthma was characterized by a lower use of cessation treatments and a lower self-efficacy to quit smoking compared with no-asthmatic smokers. Our results are consistent with those of a U.S. study involving smokers using quit lines, which found a significantly lower 30-day quit rate in smokers with asthma compared to those without chronic conditions [32]. A greater cessation failure rate addition in smokers with asthma has been observed by other authors [33]. On the other hand, the presence of asthma in smokers was accompanied by a higher anxiety level and other psychopathological traits, which may make it more difficult to quit smoking despite having a good motivation for smoking cessation [34, 35].

Our data show a high level of passive exposure at home (40%), although without significant differences between asthmatic and non-asthmatic smokers. The relationship between passive exposure and asthma is well known [27, 36, 37] and is important to consider since passive exposure contributes to the adverse health outcomes presented by smokers with asthma [13, 15, 24, 38]. Therefore, is important to involve family members to prevent smoking at home.

We would like to highlight the significantly lower smoking cessation support in asthmatic versus no asthmatic smokers (60% vs 77%) found in our study. The fact that smokers with asthma are not more likely to receive smoking cessation counseling than general smokers has already been observed in the literature [12]. These data could be explained, at least in part, by the fact that these patients smoked less, had been smoking for a shorter time, and had a lower level of dependence. On the other hand, our results are in line with those reported in the literature [39, 40], and no differences were found in the quit rates between both groups. Considering the impact of smoking on this lung disease, perhaps we should consider using other types of intervention, even more intensive, in the group of smokers with asthma. Thus, more efforts are needed to implement efficient strategies to achieve smoking cessation in the whole smoking population, but particularly in those with concomitant asthma wishing to stop smoking, regardless of the dependence level or other smoking features [41, 42].

Finally, the main characteristics of asthma in smokers were related to poor asthma control, as measured by the ACT score, as well as the high exacerbation frequency during the last month, the need for rescue medication, and frequent use of LABAs, inhaled steroids, and SABAs. These results are consistent with the literature. Indeed, evidence suggests that smoking is associated with decreased asthma control, increased exacerbation risk, worsened prognosis, and decreased inhaled corticosteroids effectiveness [6, 10, 12, 15, 18]. Asthmatic smokers constitute a subpopulation with higher risk of poor evolution requiring special attention and control.

Our work presents some limitations. The most important one is probably the representativeness of the sample. The smokers studied attended smoking cessation units and may therefore not be representative of the smoker population. This selection bias may be responsible for an overestimation of the outcomes studied. However, studying this type of selected smokers can be a great opportunity for the implementation of smoking cessation strategies given the reversible effect of pulmonary involvement.

In conclusion, asthma frequency is very high among smokers. Asthma screening in smokers and integral management of smokers with asthma are fundamental to prevent poor outcomes in this subgroup of smokers and to achieve smoking cessation.

Supporting information

Acknowledgments

The authors thank to InMusc for their methodology and statistical support.