Identification of asthma phenotypes based on extrapulmonary treatable traits

Patricia Duarte Freitas, Rafaella França Xavier, Vanessa Marie McDonald, Peter Gerard Gibson, Laura Cordova-Rivera, Karina Couto Furlanetto, Joice Mara de Oliveira, Regina Maria Carvalho-Pinto, Alberto Cukier, Rafael Stelmach, Celso Ricardo Fernandes Carvalho
European Respiratory Journal 2021 57: 2000240; DOI: 10.1183/13993003.00240-2020

Abstract

Asthma is a heterogeneous and complex disease, and a description of asthma phenotypes based on extrapulmonary treatable traits has not been previously reported.

The objective of this study was to identify and characterise clusters based on clinical, functional, anthropometrical and psychological characteristics in participants with moderate-to-severe asthma.

This was a cross-sectional multicentre study involving centres from Brazil and Australia. Participants (n=296) with moderate-to-severe asthma were consecutively recruited. Physical activity and sedentary time, clinical asthma control, anthropometric data, pulmonary function and psychological and health status were evaluated. Participants were classified by hierarchical cluster analysis and the clusters compared using ANOVA, Kruskal­–Wallis and Chi-squared tests. Multiple logistic and linear regression models were performed to evaluate the association between variables.

We identified four clusters: 1) participants with controlled asthma who were physically active; 2) participants with uncontrolled asthma who were physically inactive and more sedentary; 3) participants with uncontrolled asthma and low physical activity, who were also obese and experienced anxiety and/or depression symptoms; and 4) participants with very uncontrolled asthma who were physically inactive, more sedentary, obese and experienced anxiety and/or depression symptoms. Higher levels of sedentary time, female sex and anxiety symptoms were associated with increased odds of exacerbation risk, while being more active showed a protective factor for hospitalisation. Asthma control was associated with sex, the occurrence of exacerbation, physical activity and health status.

Physical inactivity, obesity and symptoms of anxiety and/or depression were associated with worse asthma outcomes, and closely and inextricably associated with asthma control. This cluster analysis highlights the importance of assessing extrapulmonary traits to improve personalised management and outcomes for people with moderate and severe asthma.

Abstract

Cluster analysis based on extrapulmonary treatable traits in people with moderate-to-severe asthma indicates that physical inactivity, higher levels of sedentary time, symptoms of anxiety and depression, and obesity are associated with worse outcomes. https://bit.ly/2ATv1Ce

Introduction

Asthma is a major health concern, causing a high illness burden for individuals and health systems. Asthma is a heterogeneous and complex disease characterised by variability in disease expression and severity [1]. People with asthma have different clinical presentations, but, despite being partially explained by disease severity, these are insufficient to account for the recognised heterogeneity [2]. The current concept of airway diseases neither recognises the complexity of the disease nor promotes individualised management [3]. There is increasing appreciation of the impact of extrapulmonary features of asthma and associated comorbidities [4], highlighting the need to deconstruct and personalise asthma management by identifying measurable and modifiable traits [5]. “Treatable traits” is a taxonomy that has been proposed to recognise the complexity of chronic respiratory diseases. The approach seeks to characterise individuals by the presence of “potentially modifiable elements” that impact symptoms and prognosis [6, 7]. Traits are recognised within three domains: pulmonary, extrapulmonary and behavioural/risk factors [7]. To be considered a trait, the characteristic must be 1) identifiable, 2) clinically relevant and 3) modifiable [8].

Physical inactivity, high levels of sedentary time, psychological disturbances and obesity are extrapulmonary morbidities/traits frequently reported in people with moderate-to-severe asthma [9–12]. People with severe asthma are known to be less active and highly sedentary, and this is associated with worse exercise capacity, poor asthma control and increased systemic inflammation [13]. The lower aerobic capacity in people with asthma is associated with reduced health status and increased symptoms of depression, regardless of lung function and age [14]. Obesity in asthma is also associated with worse prognosis and severity, including increased asthma symptoms and exacerbations, worse lung function and higher use of oral corticosteroids (OCS) [1, 15]. Therefore, identifying asthma phenotypes that consider these treatable traits should be integral to asthma assessment to deliver the most appropriate treatments to patients [6].

Previous cluster analyses have identified asthma phenotypes focusing on characteristics related to disease severity, expiratory airflow limitation, inflammatory biomarkers and age of asthma onset [1, 16]. However, despite these well-conducted cluster analyses, a description of phenotypes based on extrapulmonary traits has not been previously reported. Thus, this study aimed to identify and characterise phenotypes based on clinical, functional, anthropometrical and psychological characteristics in people with moderate-to-severe asthma. We hypothesised that the severity of these clinical characteristics and modifiable behavioural risk factors can guide the classification of clinical asthma phenotypes (clusters).

Methods

A cross-sectional multicentre study involving centres from Brazil and Australia was conducted. Participants were recruited prospectively between July 2012 and March 2019 from tertiary care hospitals during routine medical consultations or using the clinics' research databases.

Adults (≥18 years old) with a diagnosis of moderate-to-severe asthma according to the Global Initiative For Asthma criteria [17] who were clinically stable (free from exacerbation in the past 30 days) and receiving optimal treatment according to the international guidelines [17] were eligible for inclusion. Exclusion criteria included chronic obstructive pulmonary disease or other significant respiratory or cardiovascular diseases, active cancer, uncontrolled hypertension, diabetes or a musculoskeletal condition that could compromise participation in physical activity. The inability to understand the questionnaires, pregnancy and current or past history of smoking (≥10 pack-years) were also listed as exclusions.

Written informed consent was obtained from all participants. Ethical approvals were granted by each centre's respective ethics committees.

Procedures

Data on demographics, anthropometrics (weight, height, body mass index (BMI)) [18], smoking history, comorbidities, lung function and asthma medication use were extracted from participants' medical records or through interview. Further assessments are listed below.

Asthma control

The Asthma Control Questionnaire (ACQ) [19, 20] consists of five questions relating to asthma symptoms, bronchodilator use and lung function (% predicted of forced expiratory volume in 1 s (FEV1) before bronchodilation). Scores range from 0 to 6, where <0.75 and >1.5 are indicative of good and poorly controlled asthma, respectively [21].

Asthma-related health status

The Asthma Quality of Life Questionnaire (AQLQ) [22] comprises four asthma-related domains: activity limitations, symptoms, emotional function and environmental stimuli. Scores range from 0 to 7, with higher scores indicating better health status.

Asthma exacerbations

Exacerbations were defined as worsening of symptoms that led to ≥3 days of OCS treatment or a temporary increase in their OCS maintenance dosage, an asthma-specific hospitalisation or an emergency department (ED) visit requiring systemic corticosteroids [23]. Exacerbations were elicited during standardised structured interviewing regarding the last 12 months.

Anxiety and depression symptoms

The Hospital Anxiety and Depression Scale (HADS) [24] consists of 14 items divided into two domains, seven for anxiety (HADS-A) and seven for depression (HADS-D). Each item is scored from 0 to 3, with a maximum score of 21 for each domain. A score of ≥8 in either domain indicates possible anxiety or depression [25].

All questionnaires were validated and applied in the appropriate language (English or Portuguese).

Physical activity and sedentary time

Movement behaviours were objectively measured by the accelerometer Actigraph GT3X (Actigraph, Pensacola, FL, USA) [26]. The device was initialised via a computer interface to collect data in 60-s epochs on the three axes using specific software (ActiLife 6.13.3 Firmware version). Participants wore the device on their waist (using an elastic belt) during wake time for seven consecutive days. Data from valid days (≥4 days and ≥10 h of recording) were presented as the average number of steps per day, the time spent (min·day−1) in moderate-vigorous physical activity (MVPA; defined as ≥1951 counts·min−1) and the time spent (min·day−1) in sedentary time (defined as <100 counts·min−1) [27]. Participants performing ≥10 000, ≥7500 and ≥5000 steps·day−1 were classified as “physically active”, “somewhat active” and “low-level active”, respectively [28].

Statistical analysis

Data were analysed using the Statistical Package for Social Sciences version 18.0 for Windows (SPSS Inc., Chicago, IL, USA).

A hierarchical cluster analysis was performed to identify the number of clusters and cluster centroids using principal component analysis (PCA) [29, 30]. The Shapiro–Wilk test evaluated data normality. The between-cluster differences were analysed by the Chi-squared test, ANOVA (plus Holm–Sidak test) or Kruskal–Wallis test (plus Dunn's multiple comparison test). Significance was set at p<0.05.

The association analyses were performed using multiple logistic regression between the dependent outcomes (exacerbation, hospitalisation, ED visit and temporary systemic corticosteroid) and the clinical and behavioural characteristics included in the cluster analysis. The stepwise forward process was used to add variables into the model. Multiple linear regression analysis was performed to identify the independent factors associated with ACQ-7 (dependent variable), sex, BMI, the occurrence of exacerbation, inhaled corticosteroid (ICS) use, daily physical activity, sedentary time, HADS and AQLQ score. Detailed statistical analyses are presented in the supplementary material.

Results

Patient characteristics

A total of 414 participants were assessed for eligibility, and 118 excluded. The final study population consisted of 296 participants with moderate-to-severe asthma (243 from Brazil and 53 from Australia). Participants were mostly female, overweight, had low physical activity and high sedentary time levels, and had mild airway obstruction (table 1). The majority (68%) of the participants had uncontrolled asthma, and 64% had experienced at least one exacerbation in the last 12 months (10% hospitalisations, 49% visits to the ED, 60% used systemic corticosteroids). Participants from Australia were older, more physically inactive and spent more time engaged in sedentary time. However, they had better health status and fewer symptoms of anxiety and depression (table 1).

View this table:
TABLE 1

Participant characteristics

There were 15 comorbidities identified: osteoporosis, vocal cord dysfunction, dyslipidaemia, bowel disease, hypothyroidism, diabetes, dermatitis, obstructive sleep apnoea syndrome, sinusitis, musculoskeletal impairment, psychological disturbance, hypertension, obesity, gastro-oesophageal reflux disease (GORD) and rhinitis. The frequency of comorbidities ranged from 3% to 78%. Rhinitis, GORD, obesity, hypertension and psychological disturbance were the most prevalent comorbidities (figure 1a). Almost all participants (98%) had at least one comorbidity and >50% had more than three comorbidities (figure 1b).

FIGURE 1
FIGURE 1

Comorbidities in 296 participants with moderate-to-severe asthma. a) Distribution of 15 comorbidities identified; b) distribution of the number of comorbidities per patient. VCD: vocal cord dysfunction; OSAS: obstructive sleep apnoea syndrome; GORD: gastro-oesophageal reflux disease.

Cluster analysis and description

The Kaiser–Meyer–Olkin (0.6) and the Bartlett's Test of Sphericity (p<0.001) confirmed that the cluster analysis was appropriate. The PCA identified two components: component 1 encompassed the variables BMI, ACQ-7, AQLQ and HADS; component 2 encompassed physical activity levels and sedentary time (supplementary table S1).

Ward's cluster analysis was based on the significant components identified by the PCA. Using the hierarchical cluster analysis described in the Methods, a dendrogram was generated and four clusters were identified. The four clusters differed significantly by sex, BMI, asthma medication, asthma control, physical activity levels, sedentary time, health status and symptoms of anxiety and depression (table 2). Although all clusters had a high frequency of rhinitis (>60%), they differed in the frequency for most other comorbidities (figure 2).

View this table:
TABLE 2

Cluster characteristics of participants with moderate-to-severe asthma

FIGURE 2
FIGURE 2

Frequency of comorbidities per cluster. Data are presented as the percentage of each comorbidity in the four clusters. VCD: vocal cord dysfunction; OSAS: obstructive sleep apnoea syndrome; MSK: musculoskeletal; GORD: gastro-oesophageal reflux disease. #: p<0.05 versus cluster 1; : p<0.05 versus cluster 2; +: p<0.05 versus cluster 3.

Cluster 1, “High movers”

This cluster comprised 76 participants (25%), of whom 88% were from the Brazilian population. The cluster had the largest percentage of physically active participants (41% classified as “physically active”, 79% as “somewhat physically active”) and engaged in less sedentary time than other cluster participants. Additionally, most of the participants (62%) had controlled asthma symptoms, were female, were overweight and used lower doses of ICS compared with the other clusters.

Cluster 2, “Poorly active”

This cluster comprised 80 participants (27.3%, with 60% from the Brazilian cohort), and 99% were classified as low-level active and more sedentary. This cluster had fewer female patients, who were overweight, but with a smaller number of obese patients than clusters 3 and 4 (respectively, 37% versus 69% and 64%). Most participants (65%) presented with uncontrolled asthma symptoms.

Cluster 3, “Moderately active, obese and distressed”

This cluster comprised 69 patients (23%) who were mostly female, and 75% presented with uncontrolled asthma. Obesity was present in 71% of participants; they were more physically active, and engaged in less sedentary time than clusters 2 and 4, and had higher anxiety and depression scores than clusters 1 and 2.

Cluster 4, “Inactive, obese and distressed”

This cluster comprised 71 participants (24%, with 83% from the Brazilian cohort). Obesity was present in 64% of participants, and all were classified as physically inactive and accrued a high volume of sedentary time. Participants in cluster 4 presented with increased anxiety and depression symptom scores compared with those in clusters 1 and 2. They used higher doses of ICS and long-acting β2-agonists. Most participants (91%) presented with uncontrolled asthma.

Our results suggest that only those patients with controlled asthma were physically active, suggesting that the two outcomes are firmly linked. There were no between-cluster differences in age, smoking history, asthma onset, lung function or the frequency of participants treated with long-acting β2-agonists in combination with ICS (table 2).

Comorbidities, asthma control, health status, psychological symptoms and exacerbations within clusters

The most prevalent comorbidities in all clusters were GORD, obesity and rhinitis (figure 2). Diabetes, obesity and psychological disturbances were most prevalent in clusters 3 and 4. Cluster 3 had a lower prevalence of musculoskeletal disorders than all the other clusters.

Clusters 3 and 4 reported worse asthma control than clusters 1 and 2 (figure 3). The use of short-acting bronchodilators was higher in cluster 4. No between-cluster differences were observed for lung function. The odds of uncontrolled asthma were higher in clusters 2 and 3 compared to cluster 1 (respectively, OR 3.3, 95% CI 1.7–6.4 and OR 4.6, 95% CI 2.4–10.1). Cluster 4 presented even higher odds of uncontrolled asthma (OR 17.6, 95% CI 6.7–45.6) (figure 4a).

FIGURE 3
FIGURE 3

Comparison between clusters of each Asthma Control Questionnaire-7 (ACQ-7) question. Data are presented as mean±sd. FEV1: forced expiratory volume in 1 s. #: p<0.05 versus cluster 1; : p<0.05 versus cluster 2; +: p<0.05 versus cluster 3.

FIGURE 4
FIGURE 4

Adjusted odds ratios (OR) between clusters. a) Uncontrolled asthma; b) exacerbation; c) hospitalisation; d) emergency department (ED) visit. Data are presented as the median (95% CI). #: p<0.05 versus cluster 1; : p<0.05 versus cluster 2; +: p<0.05 versus cluster 3.

Cluster 4 reported the highest exacerbation rate. No difference in the odds of exacerbations and hospitalisations was observed between clusters 1, 2 and 3 (figure 4b, c). Cluster 4 presented 2.3 (95% CI 1.1–4.6) higher odds of exacerbation compared with cluster 1 and 5.5 (95% CI 1.5–20.0) higher odds of hospitalisation compared with clusters 1 and 3. Clusters 3 and 4 presented higher odds of ED visits (respectively, OR 1.8, 95% CI 1.0–3.6 and OR 1.9, 95% CI 1.0–3.7) compared with cluster 2. No between-cluster differences were observed for the odds of requiring systemic corticosteroid use for asthma exacerbation.

Clusters 3 and 4 reported lower health status, more symptoms, greater activity limitation and impaired mental health compared with clusters 1 and 2 (table 2).

Clusters 3 and 4 presented with increased anxiety and depression symptom levels compared with clusters 1 and 2. However, patients in cluster 3 presented even higher anxiety and depression symptoms than those in cluster 4 (table 2).

Tree diagram

The tree diagram was performed using discriminatory variables for cluster assignment (physical activity, obesity and anxiety symptoms) using subsets of these variables to assess the classification of participants (figure 5). The results of this analysis demonstrate that a greater number of comorbidities/risk factors (e.g. inactivity, obesity and anxiety) identifies clusters with worse asthma control. The proportion of participants in each cluster is presented in figure 6. These figures suggest that a simple method for phenotyping of asthma subclasses can be based on these clinical variables.

FIGURE 5
FIGURE 5

Tree analysis reporting the distribution of each comorbidity in the clusters considering clinical cut-points (darker colour) for physical activity (in step counting) [28], body mass index [43] and anxiety symptoms (Hospital and Anxiety and Depression Scale subscale) [24]. Participants were assigned to one of the four clusters that range from controlled asthma (cluster 1) to very uncontrolled asthma disease (cluster 4).

FIGURE 6
FIGURE 6

Tree performance. Using the algorithm generated by the tree analysis, 80% of participants were assigned to the correct cluster according to the number of comorbidities. Colours are in line with the tree diagram (blue: high movers; green: poorly active; yellow: moderately active, obese and distressed; red: inactive, obese and distressed). The number of comorbidities (inactive, obese and distressed/higher anxiety scores) ranges from zero to three. The percentage of participants from that cluster that are correctly assigned is indicated numerically within the shape.

Clinical associations

The associations between the dependent outcomes (exacerbation, hospitalisation, ED visit and bursts of systemic corticosteroids) with the characteristics sex, BMI, ACQ-5, sedentary time, daily physical activity, HADS-A, HADS-D, AQLQ and ICS dose are described in supplementary table S2. Higher levels of sedentary time were significantly associated with increased odds of exacerbation (OR 1.83, 95% CI 1.02–3.30, p=0.04) and hospitalisation (OR 1.23, 95% CI 1.01–1.50, p=0.04) and with greater systemic corticosteroids bursts (OR 1.16, 95% CI 1.02–1.32, p=0.02) (supplementary table S2). Being more active was a protective factor for hospitalisation (OR 0.81, 95% CI 0.67–0.97, p=0.03) (supplementary table S2). Female sex was also a risk factor for exacerbation (OR 1.14, 95% CI 1.01–1.30, p=0.04), ED visits (OR 1.90, 95% CI 1.02–3.53, p=0.04) and for greater systemic corticosteroid bursts (OR 1.95, 95% CI 1.06–3.60, p=0.03) (supplementary table S2). Higher levels of anxiety symptoms were also significantly associated with increased ED visits (OR 1.06, 95% CI 1.02–1.14, p=0.01).

Multiple linear regression

The ACQ-7 total score was significantly associated with sex, the occurrence of exacerbation, daily physical activity and AQLQ total score (supplementary table S3):

ACQ-7=4.13+(0.33×exacerbation)−(0.35×sex)−(0.08×steps·day−1)−(0.34×AQLQ)

Discussion

In this study, we performed a hierarchical cluster analysis to identify clinical asthma phenotypes based on extrapulmonary traits and behavioural/risk factors in patients with moderate-to-severe asthma, and to describe the clinical characteristics associated with these phenotypes. We included two populations, from Brazil and Australia. Our analysis identified four distinct phenotypes with relatively even representation of patients within each cluster. These were 1) participants with controlled asthma who were physically active; 2) participants with uncontrolled asthma who were physically inactive and more sedentary; 3) participants with uncontrolled asthma with low physical activity, who were also obese and experienced anxiety and/or depression symptoms; and 4) participants with very uncontrolled asthma, who were physically inactive, more sedentary, obese and experienced anxiety and/or depression symptoms. We examined the clinical associations of each of these clusters and determined cluster 4 to be associated with worse outcomes in terms of exacerbation; they also had the poorest asthma control.

Previous cluster analyses have been performed in asthma, aiming to identify clinical phenotypes in patients with severe disease. Despite the importance of these clinical phenotypes, alternative approaches have been recommended for the classification of asthma. Moore et al. [16] identified five clusters of patients with different clinical, physiological and inflammatory characteristics. Of the 11 most important variables that determined assignment to individual clusters, six were based on pulmonary function tests, two were age-related, two reflected medication use and one was sex. In another analysis, Haldar et al. [1] defined four clusters in patients from secondary care; the variables included airway inflammation, lung function, symptoms, atopy and obesity. Patients were classified as having “early-onset symptoms” and “late-onset inflammation”, and there was observed discordance between asthma symptoms and eosinophilic airway inflammation [1]. To the best of our knowledge, no prior asthma cluster analyses have attempted to phenotype patients based on extrapulmonary treatable traits and risk factors. This analysis is necessary for several reasons. Traits such as physical inactivity and high sedentary time are common in patients with asthma, especially in severe disease [13], and they are significantly associated with poor clinical outcomes and poor health status [4]. Evidence from the general population and in other chronic diseases has confirmed that these traits are importantly modifiable [31, 32]. Therefore, understanding the impacts of these traits and how they cluster is important for the development of treatment interventions beyond the current asthma management paradigm [33, 34].

There have been advances in asthma management for patients with severe asthma over the last decade, including monoclonal antibody therapies [35] and macrolide antibiotics for exacerbation reduction [36]. Nevertheless, patients with severe disease continue to experience poor health status [37], and have a higher comorbidity and risk-factor burden than those with controlled disease [6]. Therefore, a current priority is to develop interventions that target clinically important extrapulmonary traits [6]. Our cluster analysis has identified four clusters based on these traits. We have also determined the clinical relevance of these clusters by assessing their associations with important clinical outcomes such as exacerbations, health status and asthma control. Targeting these traits may improve these outcomes. The treatable traits paradigm proposes the application of multidimensional assessment to identify traits that are clinically important, measurable and modifiable, followed by targeted interventions for each trait identified. A randomised controlled trial of this approach in severe asthma demonstrated its efficacy in terms of improving health status and airway and systemic inflammation and in reducing primary care visits [8]. In this study, all of the traits identified were treated. The presence of the traits of anxiety, depression and obesity are common and tend to be associated with poorer asthma outcomes independent of the clinical features of the disease. In clusters 2 and 4, the age of asthma onset and lung function were relatively similar; however, the outcomes in cluster 4 were worse. The difference was the presence of these additional traits. This finding highlights the importance of these traits on disease outcomes.

The impact of obesity and symptoms of anxiety and depression warrants consideration. Cluster 4 (very uncontrolled asthma, highly physically inactive and sedentary with obesity) and cluster 3 (more active with better asthma control than cluster 4, but still obese with anxiety and depression) were associated with the poorest clinical outcomes. This further highlights the importance of these extrapulmonary traits in terms of their additive deleterious effects on people with moderate and severe asthma. Obesity is common in more severe asthma and is a recognised risk factor for increased asthma exacerbations and worse asthma control [1, 10]. The synergistic relationship between obesity and reduced physical activity is also well characterised [38]. Previous studies in obese [39, 40] and non-obese [41] people with asthma have shown that dietary restriction plus exercise programmes have promising effects on asthma control and health status, highlighting their potential as treatable traits.

This is the first study to identify asthma phenotypes considering clinical, functional, anthropometrical and psychological characteristics collectively, in people with moderate-to-severe asthma. Our findings reinforce the need for an individualised multidimensional assessment of asthma to facilitate the implementation of personalised management [8]. For instance, a behaviour-change intervention aimed at increasing physical activity could lead to improved asthma control for patients of cluster 2 (poorly active) [33], while weight-loss and psychological interventions may be proposed for patients within cluster 3 (moderately active, obese and distressed). Regarding cluster 4 (physically inactive, obese and distressed), a more comprehensive approach, including physical training and programmes that address techniques, may be more effective in improving asthma control [8, 39, 42]. In addition, we have proposed cut-points that allow the identification of these clusters (figure 5), which may be applied in clinical practice. However, further studies are required to validate these cut-points to endorse or refute their applicability.

Asthma phenotypes have been applied in clinical practice to allow more precise endpoints according to the main underlying pathology, with the aim of reducing exacerbations and corticosteroid treatment and improving clinical control, pulmonary function and quality of life [1, 16]. Despite this, people with moderate-to-severe asthma continue to experience frequent symptoms and attacks, suggesting that current pharma- and non-pharmacotherapies are insufficient given the complexity of more severe disease. If the phenotypes identified in this current study are considered in the overall management of people with moderate-to-severe asthma, there may be greater gains in asthma control and outcomes for this population.

Our study has strengths and some limitations. In terms of strengths, we quantified physical activity and sedentary outcomes using objective measures, which are scarce in these populations. Second, the inclusion of participants from two continents increases the generalisability of our findings, even though these populations were not matched in terms of age, lung function or the impact of the disease. Another limitation may be the higher proportion of women; however, this is also reflective of a more severe asthma population. Despite the multicentre nature of this study, the sample size is relatively small, indicating the need for further validation of these clusters. We also acknowledge the imbalance in the number of patients between Brazilian and Australian cohorts; however, this imbalance takes into account the populations of each country (210 and 25 million inhabitants, respectively). Finally, due to the cross-sectional nature of the study design, we cannot establish causality. Further studies are needed to understand the bidirectional nature of these traits in this population.

In conclusion, we have identified four asthma phenotypes based on extrapulmonary characteristics through hierarchical cluster analysis. These distinct clusters based on physical activity levels, obesity and depressive and anxiety symptoms were associated with important clinical asthma outcomes. Our data reinforce the importance of evaluating extrapulmonary traits in clinical practice to individualise treatments with the goal of improving clinical outcomes in people with moderate-to-severe asthma.

Supplementary material

Supplementary Material

Please note: supplementary material is not edited by the Editorial Office, and is uploaded as it has been supplied by the author.

Supplementary material ERJ-00240-2020.SUPPLEMENT

Shareable PDF

Supplementary Material

This one-page PDF can be shared freely online.

Shareable PDF ERJ-00240-2020.Shareable

Acknowledgments

The authors gratefully acknowledge all the patients and health professionals who participated in this study. This research was supported by the Sao Paulo Research Foundation (FAPESP, grant 2016/17093-0), Conselho Nacional de Pesquisa (CNPq, grants 311443/2014-1), the John Hunter Hospital Charitable Trust Grant Scheme, The Hunter Medical Research Institute and the University of Newcastle.

Footnotes

  • This article has supplementary material available from erj.ersjournals.com

  • Author contributions: Study concept and design: P.D. Freitas, R.M. Carvalho-Pinto, A. Cukier, R. Stelmach and C.R.F. Carvalho; data acquisition: P.D. Freitas, V.M. McDonald, L. Cordova-Rivera, K.C. Furlanetto and J.M. de Oliveira; data analysis and/or interpretation: P.D. Freitas, R.F. Xavier, V.M. McDonald, P.G. Gibson, L. Cordova-Rivera, K.C. Furlanetto, J.M. de Oliveira, R.M. Carvalho-Pinto, A. Cukier, R. Stelmach and C.R.F. Carvalho; manuscript writing and/or critical revisions for important intellectual content: P.D. Freitas, R.F. Xavier, V.M. McDonald, P.G. Gibson, L. Cordova-Rivera, K.C. Furlanetto, J.M. de Oliveira, R.M. Carvalho-Pinto, A. Cukier, R. Stelmach and C.R.F. Carvalho. All the authors have read and approved the final version of the manuscript.

  • Conflict of interest: P.D. Freitas has nothing to disclose.

  • Conflict of interest: R.F. Xavier has nothing to disclose.

  • Conflict of interest: V.M. McDonald has nothing to disclose.

  • Conflict of interest: P.G. Gibson has nothing to disclose.

  • Conflict of interest: L. Cordova-Rivera has nothing to disclose.

  • Conflict of interest: K.C. Furlanetto has nothing to disclose.

  • Conflict of interest: J.M. de Oliveira has nothing to disclose.

  • Conflict of interest: R.M. Carvalho-Pinto has nothing to disclose.

  • Conflict of interest: A. Cukier has nothing to disclose.

  • Conflict of interest: R. Stelmach has nothing to disclose.

  • Conflict of interest: C.R.F. Carvalho has nothing to disclose.

  • Support statement: The study was supported by the grant 2016/17093-0 from the Sao Paulo Research Foundation (FAPESP), by the grant 311443/2014-1 from the Ministério da Ciência, Tecnologia e Inovação - Conselho Nacional de Pesquisa (CNPq) and by the John Hunter Hospital Charitable Trust Grant Scheme, The Hunter Medical Research Institute and the University of Newcastle. Funding information for this article has been deposited with the Crossref Funder Registry.

  • Received February 4, 2020.
  • Accepted July 6, 2020.

References

  1. ↵
    1. Haldar P,
    2. Pavord ID,
    3. Shaw DE, et al.
    Cluster analysis and clinical asthma phenotypes. Am J Respir Crit Care Med 2008; 178: 218–224. doi:10.1164/rccm.200711-1754OC
    OpenUrlCrossRefPubMedWeb of Science
  2. ↵
    1. Global Initiative for Asthma (GINA)
    . Global Strategy for Asthma Management and Prevention. Bethseda, MD, National Institutes of Health/National Heart, Lung, and Blood Institute, 2020. www.ginasthma.org. Date last accessed: June 15, 2020.
  3. ↵
    1. Pavord ID,
    2. Beasley R,
    3. Agusti A, et al.
    After asthma: redefining airways diseases. Lancet 2018; 391: 350–400. doi:10.1016/S0140-6736(17)30879-6
    OpenUrlCrossRefPubMed
  4. ↵
    1. Cordova-Rivera L,
    2. Gibson PG,
    3. Gardiner PA, et al.
    Extrapulmonary associations of health status in severe asthma and bronchiectasis: comorbidities and functional outcomes. Respir Med 2019; 154: 93–101. doi:10.1016/j.rmed.2019.06.010
    OpenUrl
  5. ↵
    1. Agusti A,
    2. Bafadhel M,
    3. Beasley R, et al.
    Precision medicine in airway diseases: moving to clinical practice. Eur Respir J 2017; 50: 1701655. doi:10.1183/13993003.01655-2017
    OpenUrlAbstract/FREE Full Text
  6. ↵
    1. McDonald VM,
    2. Hiles SA,
    3. Godbout K, et al.
    Treatable traits can be identified in a severe asthma registry and predict future exacerbations. Respirology 2019; 24: 37–47. doi:10.1111/resp.13389
    OpenUrl
  7. ↵
    1. Agusti A,
    2. Bel E,
    3. Thomas M, et al.
    Treatable traits: toward precision medicine of chronic airway diseases. Eur Respir J 2016; 47: 410–419. doi:10.1183/13993003.01359-2015
    OpenUrlAbstract/FREE Full Text
  8. ↵
    1. McDonald VM,
    2. Clark VL,
    3. Cordova-Rivera L, et al.
    Targeting treatable traits in severe asthma: a randomised controlled trial. Eur Respir J 2020; 55: 1901509. doi:10.1183/13993003.01509-2019
    OpenUrlAbstract/FREE Full Text
  9. ↵
    1. Cordova-Rivera L,
    2. Gibson PG,
    3. Gardiner PA, et al.
    A systematic review of associations of physical activity and sedentary time with asthma outcomes. J Allergy Clin Immunol Pract 2018; 6: 1968–1981. doi:10.1016/j.jaip.2018.02.027
    OpenUrl
  10. ↵
    1. Freitas PD,
    2. Silva AG,
    3. Ferreira PG ADAS, et al.
    Exercise improves physical activity and comorbidities in obese adults with asthma. Med Sci Sports Exerc 2018; 50: 1367–1376. doi:10.1249/MSS.0000000000001574
    OpenUrl
    1. Patel PO,
    2. Patel MR,
    3. Baptist AP
    . Depression and asthma outcomes in older adults: results from the National Health and Nutrition Examination Survey. J Allergy Clin Immunol Pract 2017; 5: 1691–1697. doi:10.1016/j.jaip.2017.03.034
    OpenUrl
  12. ↵
    1. Freitas PD,
    2. Ferreira PG,
    3. da Silva A, et al.
    The effects of exercise training in a weight loss lifestyle intervention on asthma control, quality of life and psychosocial symptoms in adult obese asthmatics: protocol of a randomized controlled trial. BMC Pulm Med 2015; 15: 124. doi:10.1186/s12890-015-0111-2
    OpenUrl
  13. ↵
    1. Cordova-Rivera L,
    2. Gibson PG,
    3. Gardiner PA, et al.
    Physical activity and exercise capacity in severe asthma: key clinical associations. J Allergy Clin Immunol Pract 2018; 6: 814–822. doi:10.1016/j.jaip.2017.09.022
    OpenUrl
  14. ↵
    1. Mendes FA,
    2. Lunardi AC,
    3. Silva RA, et al.
    Association between maximal aerobic capacity and psychosocial factors in adults with moderate-to-severe asthma. J Asthma 2013; 50: 595–599. doi:10.3109/02770903.2013.786724
    OpenUrl
  15. ↵
    1. Scott HA,
    2. Wood LG,
    3. Gibson PG
    . Role of obesity in asthma: mechanisms and management strategies. Curr Allergy Asthma Rep 2017; 17: 53. doi:10.1007/s11882-017-0719-9
    OpenUrl
  16. ↵
    1. Moore WC,
    2. Meyers DA,
    3. Wenzel SE, et al.
    Identification of asthma phenotypes using cluster analysis in the Severe Asthma Research Program. Am J Respir Crit Care Med 2010; 181: 315–323. doi:10.1164/rccm.200906-0896OC
    OpenUrlCrossRefPubMedWeb of Science
  17. ↵
    1. Global Initiative for Asthma (GINA)
    . Global Strategy for Asthma Management and Prevention. Bathesda, MD, National Institutes of Health/National Heart, Lung and Blood Institute of Health, 2012. www.ginasthma.org. Date last accessed: January 21, 2012.
  18. ↵
    1. Gordon CC,
    2. Chumlea WC,
    3. Roche AF
    . Stature, recumbent length and weight. In: Lohman TG, Roche AF, Martorell R, eds. Anthropometric Standardization Reference Manual. Champaign, III, Human Kinetics, 1988; pp. 3–8.
  19. ↵
    1. Juniper EF,
    2. O'Byrne PM,
    3. Guyatt GH, et al.
    Development and validation of a questionnaire to measure asthma control. Eur Respir J 1999; 14: 902–907. doi:10.1034/j.1399-3003.1999.14d29.x
    OpenUrlAbstract/FREE Full Text
  20. ↵
    1. Leite M,
    2. Ponte EV,
    3. Petroni J, et al.
    Evaluation of the asthma control questionnaire validated for use in Brazil. J Bras Pneumol 2008; 34: 756–763. doi:10.1590/S1806-37132008001000002
    OpenUrlCrossRefPubMed
  21. ↵
    1. Juniper EF,
    2. Bousquet J,
    3. Abetz L, et al.
    Identifying ‘well-controlled’ and ‘not well-controlled’ asthma using the Asthma Control Questionnaire. Respir Med 2006; 100: 616–621. doi:10.1016/j.rmed.2005.08.012
    OpenUrlCrossRefPubMedWeb of Science
  22. ↵
    1. Juniper EF,
    2. Guyatt GH,
    3. Epstein RS, et al.
    Evaluation of impairment of health related quality of life in asthma: development of a questionnaire for use in clinical trials. Thorax 1992; 47: 76–83. doi:10.1136/thx.47.2.76
    OpenUrlAbstract/FREE Full Text
  23. ↵
    1. Reddel HK,
    2. Taylor DR,
    3. Bateman ED, et al.
    An official American Thoracic Society/European Respiratory Society statement: asthma control and exacerbations: standardizing endpoints for clinical asthma trials and clinical practice. Am J Respir Crit Care Med 2009; 180: 59–99. doi:10.1164/rccm.200801-060ST
    OpenUrlCrossRefPubMedWeb of Science
  24. ↵
    1. Zigmond AS,
    2. Snaith RP
    . The hospital anxiety and depression scale. Acta Psychiatr Scand 1983; 67: 361–370. doi:10.1111/j.1600-0447.1983.tb09716.x
    OpenUrlCrossRefPubMedWeb of Science
  25. ↵
    1. Botega NJ,
    2. Bio MR,
    3. Zomignani MA, et al.
    [Mood disorders among inpatients in ambulatory and validation of the anxiety and depression scale HAD]. Rev Saude Publica 1995; 29: 355–363. doi:10.1590/S0034-89101995000500004
    OpenUrlCrossRefPubMedWeb of Science
  26. ↵
    1. Trost SG,
    2. McIver KL,
    3. Pate RR
    . Conducting accelerometer-based activity assessments in field-based research. Med Sci Sports Exerc 2005; 37: Suppl. 11, S531–S543. doi:10.1249/01.mss.0000185657.86065.98
    OpenUrlCrossRefPubMedWeb of Science
  27. ↵
    1. Freedson PS,
    2. Melanson E,
    3. Sirard J
    . Calibration of the Computer Science and Applications, Inc. accelerometer. Med Sci Sports Exerc 1998; 30: 777–781. doi:10.1097/00005768-199805000-00021
    OpenUrlCrossRefPubMedWeb of Science
  28. ↵
    1. Tudor-Locke C,
    2. Bassett DR, Jr.
    . How many steps/day are enough? Preliminary pedometer indices for public health. Sports Med 2004; 34: 1–8. doi:10.2165/00007256-200434010-00001
    OpenUrlCrossRefPubMedWeb of Science
  29. ↵
    1. Clatworthy J,
    2. Hankins M,
    3. Buick D, et al.
    Cluster analysis in illness perceptions research: a Monte Carlo study to identify the most appropriate method. Psychol Health 2007; 22: 123–142. doi:10.1080/14768320600774496
    OpenUrl
  30. ↵
    1. Harrison SL,
    2. Robertson N,
    3. Graham CD, et al.
    Can we identify patients with different illness schema following an acute exacerbation of COPD: a cluster analysis. Respir Med 2014; 108: 319–328. doi:10.1016/j.rmed.2013.10.016
    OpenUrlCrossRefPubMed
  31. ↵
    1. Demeyer H,
    2. Louvaris Z,
    3. Frei A, et al.
    Physical activity is increased by a 12-week semiautomated telecoaching programme in patients with COPD: a multicentre randomised controlled trial. Thorax 2017; 72: 415–423. doi:10.1136/thoraxjnl-2016-209026
    OpenUrlAbstract/FREE Full Text
  32. ↵
    1. Martin A,
    2. Fitzsimons C,
    3. Jepson R, et al.
    Interventions with potential to reduce sedentary time in adults: systematic review and meta-analysis. Br J Sports Med 2015; 49: 1056–1063. doi:10.1136/bjsports-2014-094524
    OpenUrlAbstract/FREE Full Text
  33. ↵
    1. Freitas PD,
    2. Xavier RF,
    3. Passos NFP, et al.
    Effects of a behaviour change intervention aimed at increasing physical activity on clinical control of adults with asthma: study protocol for a randomised controlled trial. BMC Sports Sci Med Rehabil 2019; 11: 16. doi:10.1186/s13102-019-0128-6
    OpenUrl
  34. ↵
    1. Franca-Pinto A,
    2. Mendes FA,
    3. de Carvalho-Pinto RM, et al.
    Aerobic training decreases bronchial hyperresponsiveness and systemic inflammation in patients with moderate or severe asthma: a randomised controlled trial. Thorax 2015; 70: 732–739. doi:10.1136/thoraxjnl-2014-206070
    OpenUrlAbstract/FREE Full Text
  35. ↵
    1. Holguin F,
    2. Cardet JC,
    3. Chung KF, et al.
    Management of severe asthma: a European Respiratory Society/American Thoracic Society Guideline. Eur Respir J 2019; 55: 1900588.
    OpenUrl
  36. ↵
    1. Gibson PG,
    2. Yang IA,
    3. Upham JW, et al.
    Effect of azithromycin on asthma exacerbations and quality of life in adults with persistent uncontrolled asthma (AMAZES): a randomised, double-blind, placebo-controlled trial. Lancet 2017; 390: 659–668. doi:10.1016/S0140-6736(17)31281-3
    OpenUrlCrossRefPubMed
  37. ↵
    1. Katsaounou P,
    2. Odemyr M,
    3. Spranger O, et al.
    Still fighting for breath: a patient survey of the challenges and impact of severe asthma. ERJ Open Res 2018; 4: 00076-2018. doi:10.1183/23120541.00076-2018
    OpenUrlAbstract/FREE Full Text
  38. ↵
    1. Trost SG,
    2. Owen N,
    3. Bauman AE, et al.
    Correlates of adults’ participation in physical activity: review and update. Med Sci Sports Exerc 2002; 34: 1996–2001. doi:10.1097/00005768-200212000-00020
    OpenUrlCrossRefPubMedWeb of Science
  39. ↵
    1. Freitas PD,
    2. Ferreira PG,
    3. Silva AG, et al.
    The role of exercise in a weight-loss program on clinical control in obese adults with asthma. Am J Respir Crit Care Med 2017; 195: 32–42. doi:10.1164/rccm.201603-0446OC
    OpenUrlCrossRefPubMed
  40. ↵
    1. Scott HA,
    2. Gibson PG,
    3. Garg ML, et al.
    Dietary restriction and exercise improve airway inflammation and clinical outcomes in overweight and obese asthma: a randomized trial. Clin Exp Allergy 2013; 43: 36–49. doi:10.1111/cea.12004
    OpenUrlCrossRefPubMed
  41. ↵
    1. Toennesen LL,
    2. Meteran H,
    3. Hostrup M, et al.
    Effects of exercise and diet in nonobese asthma patients-a randomized controlled trial. J Allergy Clin Immunol Pract 2018; 6: 803–811. doi:10.1016/j.jaip.2017.09.028
    OpenUrl
  42. ↵
    1. Samdal GB,
    2. Eide GE,
    3. Barth T, et al.
    Effective behaviour change techniques for physical activity and healthy eating in overweight and obese adults; systematic review and meta-regression analyses. Int J Behav Nutr Phys Act 2017; 14: 42. doi:10.1186/s12966-017-0494-y
    OpenUrlPubMed
  43. ↵
    1. Bray GA
    . Classification and evaluation of the obesities. Med Clin North Am 1989; 73: 161–184. doi:10.1016/S0025-7125(16)30697-6
    OpenUrlCrossRefPubMedWeb of Science
Back to top
View this article with LENS
Email
Print
Citation Tools

Identification of asthma phenotypes based on extrapulmonary treatable traits
Patricia Duarte Freitas, Rafaella França Xavier, Vanessa Marie McDonald, Peter Gerard Gibson, Laura Cordova-Rivera, Karina Couto Furlanetto, Joice Mara de Oliveira, Regina Maria Carvalho-Pinto, Alberto Cukier, Rafael Stelmach, Celso Ricardo Fernandes Carvalho
European Respiratory Journal Jan 2021, 57 (1) 2000240; DOI: 10.1183/13993003.00240-2020
Reddit logo Technorati logo Twitter logo Connotea logo Facebook logo Mendeley logo
Full Text (PDF)

Jump To

Subjects