A Systematic Review of the Effect of Physical Activity on Asthma Outcomes

Abstract

Background:

Asthma is a chronic respiratory disease that affects millions worldwide. Medication management is the current mainstay of treatment; however, there is evidence to suggest additional benefit with lifestyle changes, particularly with increased physical activity.

Objective:

Our systematic review aims to discover and evaluate the effects of physical activity on asthma outcomes.

Methods:

Systematic search of PubMed, EMBASE, CINAHL, Cochrane Library, Rehabilitation and Sports Medicine Source, Scopus, and Web of Science identified 11,155 results. Thirty-five articles met our inclusion criteria spanning 20 studies. Data extraction was conducted by six independent reviewers, and final results were evaluated by a seventh reviewer and the senior author.

Results:

Wide variation among selected studies, including the heterogeneity of interventions and outcome variables, did not support a meta-analysis. Mixed results of the effects of physical activity on asthma outcomes were found. Most studies suggest that physical activity improves asthma control, quality of life, lung function parameters, and inflammatory serologies, while three found no improvements in any of these outcomes. No studies reported worsening asthma outcomes.

Conclusion:

This review highlights the emerging and promising role of physical activity as a non-pharmacologic treatment for asthma. Additional high-quality randomized controlled trials are needed to overcome the problems of measurement heterogeneity and the dilution of outcome effect size measurement related to physical activity interventions for asthma.

Background

Asthma is a chronic respiratory disease that effects over 300 million individuals worldwide.(1) Current treatment guidelines for asthma include primarily medication management with anti-inflammatory medications and long-acting bronchodilators.(2) These therapies help manage symptoms, but despite appropriate prescribing practices, some patient’s asthma symptoms remain uncontrolled. (3) Obtaining optimal control of the disease is complicated by the cost and side-effects of the medications, as well as poor adherence.(3–6) To supplement medication management in many chronic diseases, studies describe the benefits of modifying lifestyle factors, such as physical activity.(7) In patients with asthma, this is more complex, as asthma symptoms can worsen or be triggered by physical activity and can lead to avoidance of physical activity.(8, 9) Low levels of physical activity have been associated with increased asthma symptoms, increase in healthcare utilization, and decreased quality of life.(10–14) Research over decades has demonstrated that lung expansion, as prompted by aerobic exercise, is a bronchodilator.(15) As such, current guidelines recommend that all patients with chronic diseases, including asthma, participate in regular physical activity for overall health benefits.(2, 16) Despite this recommendation, patients with asthma are less physically active than gender matched controls.(12, 13, 17, 18) Describing the risks and benefits of physical activity in asthma patients is necessary to improve and implement recommendations regarding this lifestyle intervention.

Previous studies have evaluated the effects of physical activity on asthma to further define its potential harms and benefits. Systematic reviews of this evidence have shown mixed results and do not include recent studies on physical activity in people living with asthma.(3, 19–21) Studies associate physical activity with improvement in days without asthma symptoms, quality of life, forced expiratory volume in 1 second (FEV1), maximal oxygen consumption, and exercise capacity.(22, 23) A prior systematic review found no statistically significant effects on lung function parameters, but did find a trend toward improved health related quality of life.(19) Other systematic reviews found varying results in mixed adult and pediatric populations, including increased aerobic power, and no change in fractional excretion of nitric oxide or bronchial hyperresponsiveness.(3, 20, 21) New evidence has emerged on the impact of physical activity in adult populations with asthma. Of the two more recent systematic reviews, one focused on adults only, but limited outcome data to asthma control, lung function and airway inflammation.(21, 24) Our study aims to comprehensively review and summarize the evidence of the disease-modifying effects of physical activity on asthma outcomes in adults.

Methods

This is a systematic review conducted and reported in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-analysis (PRISMA) and Cochrane guidelines.(25, 26) The protocol is registered in Prospero (registration number: CRD42017058804).

Data Sources

We conducted a comprehensive search on multiple databases for articles in English entered into the datasets from their respective inception dates to November 19, 2019 when the search strategy was implemented: PubMed (1946- ), EMBASE (1947- ), CINAHL (Cumulative Index to Nursing and Allied Health Literature –EBSCOhost, 1937- ), Cochrane Library (1996-), Rehabilitation and Sports Medicine Source (1963- ), Scopus (1788- ), and Web of Science (1970- ). The search strategy was developed by investigators (MK, SN, CC) in collaboration with a clinical librarian (MC) with expertise in systematic search and review (Table E1 in Online Repository).

The search strategy comprised 3 intersecting concept sets: (1) asthmatic and bronchospastic condition concepts, (2) exercise and physical activity interventional concepts, and (3) randomized controlled trials concepts (See Table E1 in the Online Repository for the Pubmed Search Strategy). The search of the electronic databases recalled a total of 11,155 non-unique records. These data were supplemented by manual examination of the reference lists of reviews (identified by a systematic search for reviews) and examination of the reference lists of all the articles that underwent full text review in the selection phase. The results of the electronic and supplemental searches are reported in the PRISMA Flow Diagram.

Study Selection

The study selection occurred over three phases: a screening phase, selection phase and a third data abstraction phase. The screening phase of the review was conducted by four trained reviewers (KW, AS, MK, AD) who screened titles/abstracts. Of the titles/ abstracts reviewed, 91 citations were selected for full-text review. Two independent reviewers (KW, AS, MK, AD) completed a checklist-style form and included articles which met inclusion criteria. The senior author (SN) and independent reviewer (CC) reviewed any discrepancies in the full-text review.

Inclusion criteria for the selection phase included: randomized controlled trials of physical activity interventions evaluating adult subjects (≥18 years old) with a known diagnosis of asthma. Further description of inclusion criteria is available in the online repository.

Data abstraction

Six (KW, SN, VP, MK, AS, IR) independent investigators extracted data from included full-text articles and published abstracts. See details of extracted characteristics in the online repository. An independent investigator (CC) reviewed these characteristics for discrepancies and combined all extracted data into one dataset. Any discrepancies were reviewed with the senior author (SN) and independent investigator (CC). To address the limited data available in published abstracts, we attempted to contact investigators to obtain key study characteristics.

Risk of bias assessment

One independent investigator (SK), with expertise in systematic review methodology evaluated each included study for risk of bias. The principal investigator guided a discussion with the investigator for any discrepancies in assessment of study bias. The independent investigator (SK) used the Cochrane Risk for Bias tool as a guide for his assessment. This tool guided the reviewer in examining random sequence allocation, concealed allocation, blinding of participants and personal, blinding of outcome assessment, incomplete outcome data, selective reporting, and other biases. The principal investigator reviewed any inconsistencies in data extraction.

Data analysis

Due to heterogeneity of the interventions, outcome measurements, and reporting of outcomes, no meta-analysis was conducted. We exported data by outcome of interest and analyzed results by direction of effect and statistical significance. We employed vote counting to qualitatively evaluate these results of the included studies. Vote counting compares specific outcomes by adding the overall effect reported in the results of various studies. For example, it describes the number of studies that demonstrate positive effect for a particular outcome. Additionally, it counts the number of studies that report a statistically significant result for a given outcome. We report the direction of effect and the statistical significance for each outcome of interest in one of five categories. The direction of effects was defined as: 1) Positive effect with statistical significance; 2) Positive effect without statistical significance; 3) No difference between intervention and control groups 4) Negative effect with statistical significance; 5) Negative effect without statistical significance. In cases where authors did not report an effect size or direction of effect in a qualitative manner and only reported no statistical significance between groups, we coded those as having no difference between the intervention and control groups. Although vote counting is not a universally accepted means of data synthesis, we apply this method to our data, as there were no adequate alternative methods to summarize the diverse data set.(26)

Results

Study Selection

A total of 11,155 citations were retrieved from the searches of the electronic databases (Figure 1). After deduplication, a total of 6051 unique citations remained. After the screening review of titles and abstract, three additional articles were added from the manual review of references and one article published outside of our search range through contact with an author of a trial report. In total, we included 95 full text articles and abstracts for selection phase. Sixty screened-in articles were excluded for the following reasons: not randomized control(14, 27–42), protocol only(43–48), inappropriate study population(49–52), inappropriate control(53–62), inappropriate intervention(63–76), duplicate article(14, 57, 70, 72, 77–79), unable to attain article(80) and recruitment data only(81) (Table E2 in Online Repository). Thirty-five publications from twenty studies met all inclusion criteria and underwent data extraction and synthesis (Figure 1; Table 1).(77–79, 82–113)

Figure 1.

PRISMA Flow Diagram

Table 1.

Characteristics of included studies

Study	Clinical Setting	Participants	Intervention	Comparison Group	Main Outcomes	Time Points Measured
Abd El-Kader et al. (2016)83	University hospital in Saudi Arabia	80 obese (BMI 31–37), asthmatic patients aged 41–53 years old whom had been using ICS at least 6 months prior to the start of the study	Treadmill exercise of progressive intensity and frequency 3 times per week for six months	Maintain ordinary lifestyle and medication management for six months	Bone mineral density and serologic inflammatory markers (HDL-c, LDL-c, TNF-alpha, IL-2, IL-4, IL-6, leptin)	All outcomes assessed prior to the intervention and at six months
Arandelovic et al. (2007)82	University of Nis, Departments of the Health Center	65 adults with mild, persistent asthma (per GINA guidelines) with documented treatment with low dose ICS and short-acting beta-2 agonists	Swimming twice weekly for six months at a local indoor pool, as well as asthma education	Continued treatment with low-dose ICS and short-acting beta-2 agonists	Lung function (spirometry), Lung inflammation (bronchoprovocation test)	All outcomes measured prior to the intervention and after six months
Bacon et al. (2015)85*	Not described 2 Hospitals in Montreal, Canada and the Montreal Chest Institute, Montreal, Canada	66 adults with objectively confirmed asthma and poor asthma control	Cycle, treadmill, or elliptical exercise, three sessions per week for 12 weeks	Usual care	Asthma control (ACQ, and short-acting bronchodilator usage), waist circumference, BMI, depressive symptoms, and FEV1	All outcomes assessed prior to the intervention and after 12 weeks
Boyd et al. (2012)100*, 101	UAB Lung Health Center in Birmingham, AL	19 adults ages 19–70 years old with mild-moderate persistent asthma and evidence of reversibility on spirometry	Walking program for 30 minutes, three times weekly, as well as a 30 min coordinator-led educational intervention	30 min coordinator-led educational intervention	Asthma control (ACQ), lung function (spirometry), serologic and nasal-lavage inflammatory markers (ECP, IL-1beta, IL-4, IL-5, IL-6, IL-13, TNF-alpha and total IgE)	All outcomes assessed prior to intervention and after 12 weeks
Evaristo et al (2020)106*,107*, 112*, 113	University setting in Brazil, not further described	54 adults with moderate/severe asthma	Intervention consists of 40 minute twice weekly session of treadmill exercises beginning at 60% of the maximum predicted heart rate for 3 months and asthma education (two 2-hour classes/week for 2 weeks)	Breathing exercises (Pranayama Yoga breathing technique) twice a week for three months occurring in 40-minute intervals and asthma education (two 2-hour classes/week for 2 weeks)	Asthma control (ACQ-6), daily asthma symptoms, asthma exacerbations, psychosocial (HAD), health related quality of life (HRQL, AQLQ) exercise capacity (ISWT), airway inflammation (sputum), spirometr and daily physical activity levels (accelerometry)	Outcomes measured at baseline and 3 months (post-intervention)
Franca-Pinto et al. (2015)94, 95*, 96*, 97*	Outpatients at a University Hospital in Brazil	58 adults with severe persistent asthma (GINA guidelines), under the stable care of a pulmonologist, between 20–59 years old	Aerobic training program consisting of 12 weeks of twice weekly indoor treadmill exercise sessions	Yoga breathing group consisting of 12 weeks of twice weekly supervised breathing instruction	Asthma control (ACQ; diary of daily asthma symptoms and exacerbations), health related quality of life (AQLQ), lung function (pulmonary function testing), serologic inflammatory mediators (IL-4, IL-5, IL-6, IL-10, tumor necrosis factor (TNF)-alpha, IL-12p70, IL-8/CXCL-8, MCP-1/CCL2, TANTES/CCL5 and serum IgE)	All outcomes measured over four weeks before and after the 12 weeks of intervention.
Freitas et al. (2017)77, 99*, 104*, 110*, 111	Outpatients at university setting in Brazil, not further described	55 adults ages 30–60 years old with moderate/severe asthma and a BMI between 35–40	Aerobic and resistance exercise program twice weekly for three months, as well as a weight-loss program with 12 individual hypocaloric diet counseling sessions	Sham exercises, described as stretching and breathing, two times weekly for three months, as well as a weight-loss program with 12 individual hypocaloric diet counseling sessions.	Asthma control (ACQ), health related quality of life (AQLQ), lung function (spirometry; lung volumes), exercise capacity (CPET), depression/anxiety (HAD), steps/day (accelerometer), body weight/fat (bioelectrical impedance), lung inflammation (FeNO), and serologic inflammatory markers (IL-1, IL-2, IL-4, IL-5, IL-6, IL-10, IL-12, IL-13, TNF-alpha, VEGF, and chemokines)	All outcomes were measured prior to intervention, and after completion of intervention at 3 months
Haas et al. (1987)105	New York University laboratory setting in New York, New York	22 adults with a diagnosis of asthma and demonstratable EIB (>15% reduction in FEV1 and reversed with bronchodilator).	Aerobic exercise training with early training sessions designed to provide a minimum of 20 minutes of aerobic exercise, and both session length and training heart rate were systematically increased up to 40 minutes and 80% or maximal predicted heart rate.	Not described	Lung function (maximal voluntary ventilation12) and exercise capacity (minute ventilation, oxygen consumption, heart rate)	All outcomes measured before intervention and at 12 weeks.
Ma et al. (2015)91	Kaiser Permanente Medical Centers in Northern California	330 adults aged 18–70 years old with uncontrolled persistent asthma (confirmed through a multistage screening process), and a BMI greater than or equal to 30	A lifestyle intervention (targeting modest weight loss and increased physical activity) curriculum spanning 12-months, consisting of 13 in-person weekly meetings, then monthly, then bi-monthly meetings, encouraged throughout for at least 150 minutes per week of moderate-intensity physical activity	Usual care from their physician, as well as receiving a pedometer, a body weight scale, a list of routinely offered weight management services, and an asthma education DVD	Asthma control (ACQ), health related quality of life (mini AQLQ), lung function (spirometry), health care utilization (medical record abstraction of encounters, diagnoses, and pharmacy dispensing), BMI.	All outcomes were measured at baseline, and at 6- and 12-months post-randomization
Mendes et al. (2010)98, 78, 93	University Hospital in Brazil, not further described	101 adults between ages 20–50 years with asthma diagnosis (based on GINA guidelines), and under medical treatment for greater than or equal to six months	Aerobic training program for 30 minutes per session twice a week, as well as an educational program consisting of two classes once weekly for 2 hours and a breathing exercise program involving 30 min session twice weekly for 3 months	Educational program consisting of two classes once weekly for 2 hours and a breathing exercise program involving 30 min session twice weekly for 3 months	Asthma control (daily diary of symptoms), lung function (spirometry), health related quality of life (HRQoL), depression (Beck Depression Inventory), anxiety (State-Trait Anxiety Inventory), exercise capacity (CPET)	Not specifically defined, but assesses all outcomes prior to and after the intervention
Meyer et al. (2015)92	Hamberg metropolitan area, not further described	21 adults with asthma diagnosis according to “standard criteria”	Group training session weekly for 12 months	Usual care	Exercise capacity (CPET), lung function (pulmonary function testing), health related quality of life (SF-36 and AQLQ), bronchodilator use	All outcomes were measured at baseline and after 12 months
Paul et al. (2013) 87*	University Setting, Kerala, India	80 adults aged 18–40 years old with “partially controlled” asthma	Brisk walking 30 minutes per day for six months	Usual care	Lung function (spirometry), health related quality of life (AQLQ)	At baseline, one month after the onset of the study, and after 6 months
Razavi et al. (2011)84	Allergy clinic in Ilam, Iran	32 women aged 20–30 years old with asthma confirmed by clinical evaluation, pulmonary function test, skin testing for aeroallergens, and a six-minute walk test	30 minutes weekly of aerobic exercise for 8 weeks, sessions included stretching, walking, and running.	Either vitamin D supplementation or usual care	Lung function (spirometry), exercise capacity via 6MWT	All outcomes were measured prior to the intervention and at 8 weeks
Refaat et al. (2015)86	Respirology Department at the Farwanyia Hospital and Kuwait Physiotherapy center in the State of Kuwait	68 adults aged 25–65 years old with moderate to severe asthma (based on GINA guidelines)	Supervised physical training three times weekly for six weeks, training involved a 10-minute warm up, 20–30 minutes of an exercise circuit (cycling, step ups, wall squats etc), and a 5 minute cool down	Usual care	Health related quality of life (AQLQ), lung function (pulmonary function testing)	AQLQ was completed prior to study, at the sixth week of the study, and after 3 months, PFTs were completed at initiation of study and after 3 months
Scichilone et al. (2012)88	Pulmonary and Allergy outpatient clinics at the Institute of Respiratory Diseases at the University of Palermo in Palermo, Italy	Nine adults with mild asthma diagnosed by a pulmonologist (based on GINA guidelines)	Two to three weekly sessions of indoor rowing	Usual care	Airway responsiveness (single dose methacholine bronchoprovocation)	For the training group, evaluation was conducted at week 0, week 5 (mid-training), week 10 (post-training), and 4–6 week safter completion, controls underwent testing at week 0 and week 10
Scott et al. (2013)89	John Hunter Hospital in Australia	46 overweight and obese (BMI 28–40) adults with asthma, diagnosed by a doctor and documented history of airway hyperresponsiveness	12-week membership to a gymnasium and a one-hour personal training session once weekly, additionally encouraged to independently attend the gym three times or more weekly	A dietary intervention involved a calorie-restricted diet with two meal replacements daily, as well as weekly counselling with a dietician	Asthma control (ACQ), health related quality of life (AQLQ), lung function (pulmonary function testing), airway inflammation (induced sputum cytology/cell differential) and serologic inflammatory markers (CRP, IL-6, leptin, and adiponectin)	All outcomes measured prior to the intervention and post-intervention (specific timeline not provided)
Shaw et al. (2010)102, 103, 79	South Africa not further described.	88 adults aged 18–34 years old with moderate-persistent asthma based on NIH guidelines, with daily asthmatic symptoms	Aerobic Exercise Group: Three times weekly exercise program involving 5 minutes warm up, walking or jogging for 30 minutes, and a 5-minute cool down Aerobic Exercise AND Diaphragmatic breathing group: 15 minutes of walking or jogging and a reduced number of breaths during diaphragmatic breathing training	Diaphragmatic breathing group: Three times weekly instructed on breathing technique with practice breathing. Non-exercise control group: Usual care, phoned three times weekly to ensure compliance	Lung function (spirometry), BMI, exercise capacity (VO2 max)	All outcomes measured at baseline and after 8 weeks
Toennesen et al. (2018)109	Copenhagen, Denmark. Academic medical center	149 non-obese adult asthma patients with at least 1 positive diagnostic test demonstrating variable airflow obstruction (mannitol, methacholine, or reversability test).	Exercise only: High intensity interval training using the “10–20–30” concept on indoor spinning bikes. Diet only: 8-week of 5 group counseling sessions (2–6 patients/group) and 1 individual counseling session with a trained study dietician Exercise + Diet: Received both the exercise and diet interventions	Usual Care	Asthma control (ACQ), health related quality of life (AQLQ), airway responsiveness (mannitol), serologic inflammatory markers (blood eosinophils, IL-6, hs-CRP), airway inflammatory markers (sputum eosinophils, FeNO), body composition, exercise capacity (CPET), urine urea excretion	Outcomes assessed at baseline and 8 weeks.
Turk et al. (2017)108*	The Netherlands, not further described	31 adults with suboptimal controlled asthma (ACQ>0.75).	Lifestyle intervention program including high intensity interval training occurring three times a week for 3 months.	Control group participants were advised to lose weight and to exercise.	Asthma control (ACQ), asthma exacerbations, weight loss, waist circumference, exercise capacity (6MWT)	Outcomes assessed at baseline and 3 months post intervention.
Turner et al. (2011)90	Australia, not further described	35 adults older than age 40 with diagnosis of moderate or severe persistent asthma, under the care of a respiratory physician	Three exercise classes weekly for six weeks, that involved walking at different speeds	Usual care	Asthma control (ACQ), health related quality of life (SF-36), exercise capacity (6MWT), anxiety and depression (HADS).	All outcomes measured at baseline, immediately after intervention (6 weeks) and at 3 months following completion of the intervention

Asthma Control Questionnaire (ACQ), Asthma Quality of Life Questionnaire (AQLQ), chemokine ligand (CCL), C-related protein (CRP), chemokine ligand (CXCL), eosinophil cationic protein (ECP), Forced expiratory volume in one second (FEV1), Forced vital capacity (FVC), high density lipoprotein-c (HDL-c), incremental shuttle walk test (ISWT), interleukin (IL), low density lipoprotein-c (LDL-c), monocyte chemoattractant protein (MCP-1), 36-item short form survey (SF-36), tumor necrosis factor-alpha (TNF-alpha), Global Initiative for Asthma (GINA), Body Mass Index (BMI), Hospital Anxiety and Depression scale (HAD), Health related Quality of life (HRQoL), Six Minute Walk Test (6MWT), High Density Lipoproteins (HDL), Low Density Lipoproteins (LDL)

^*Indicates abstract

Risk of bias

Many studies did not adequately describe key features required to assess risk of bias. Randomization, allocation concealment, and blinding were difficult to assess in many studies (Figure 2). Most studies described missing data well, and were found to have low risk of selective outcome reporting (Figure 3). Further information regarding bias in this literature is available in the online repository.

Figure 2:

Risk of bias summary- review author’s judgements about each methodological domain for each included study

Figure 3:

Risk of bias summary graph- review author’s judgments about each methodological domain presented as percentages across all included studies

Population

Table 1 shows the main characteristics of the included studies. The criteria for asthma diagnosis varied considerably, including the regular use of an inhaled corticosteroid or other asthma medication, physician diagnosis, lung function parameters, or meeting the Global Initiative for Asthma (GINA) guidelines.(82, 86, 94, 98) Study size ranged from 19 participants to 330 participants and took place in fourteen different countries. Asthma outcomes varied widely, including lung function parameters, serologic inflammatory markers, health-related quality of life and asthma control measures. Other outcomes assessed included exercise capacity (such as six minute walking distance or incremental shuttle walk test)(84, 90, 108, 113), cardiopulmonary exercise testing including maximum work rates and maximum oxygen uptake(77, 92, 98, 103, 105, 109), weight loss parameters (such as body mass index)(77, 85, 91, 102, 108, 109) and depression/anxiety levels.(77, 85, 90, 98, 113)

Study Interventions

Physical activity interventions varied considerably across studies (Table 1 and Table E3 in the Online Repository). The interventions ranged from walking/running to circuit training or other aerobic and resistance exercises. Nine studies specified that the exercise intervention was supervised.(85, 86, 88, 90, 92, 102, 105, 109, 113) Frequency of physical activity varied from daily to once weekly, and duration varied from six weeks to one year.

Nine studies included other interventions apart from physical activity, either in the control group, intervention group, or both.(77, 84, 89, 91, 94, 98, 103, 109, 113) This ranged from breathing exercises/training (77, 94, 98, 102, 113), a weight loss or dietary component (77, 89, 91, 109) or vitamin supplementation.(84)

One study reported a negative outcome associated with exercise. A participant fell while walking and suffered a fractured wrist.(91) No study attributed any worsening asthma symptoms or exacerbations to the physical activity intervention.

Outcomes

Results of included studies are summarized in Table 2. We recorded all outcomes of included studies for the fullness of the review record, but synthesized results of asthma outcomes. Other outcomes not included in our analysis were weight change, mental health, bone mineral density, aerobic capacity and physical activity levels. Comprehensive study outcomes are described in Table 1. All studies included at least one of the asthma outcomes we evaluated.

Table 2:

Results of Included Studies

STUDIES
	LUNG FUNCTION	ASTHMA CONTROL	HEALTH-RELATED QUALITY OF LIFE
Abd El-Kader et al. (2016)
Arandelovic et al. (2007)	+*
Bacon et al. (2015)	N	+*
Boyd et al. (2012)	N	N
Evaristo et al (2020)	N	+*	+*
Franca-Pinto et al. (2015)	N	+*	+*
Freitas et al. (2017)	N	+*	+
Hass et al. (1987)	N
Ma et al. (2015)	N	N	+
Mendes et al. (2010)	N	+*	+*
Meyer et al. (2015)	N	N	N
Paul et al. (2013)	+
Razavi et al. (2011)	+*
Refaat et al. (2015)	+*		+*
Schichilone et al. (2012)	N
Scott et al. (2013)	+*	N	N
Shaw et al. (2010)	+*
Toennesen et al. (2018)	N	+*	+*
Turk et al. (2017)		+*
Turner et al. (2011)	N	N	+*

+*: Statistically significant positive effect in variable between intervention and control group

+: Positive effect in variable between intervention and control group

N: No difference between intervention and control groups reported

−: Negative effect in variable between intervention and control group

−*: Statistically significant negative effect in variable between intervention and control group

Blank cell indicates the outcome was not measured

Lung function

All but two studies reported lung function information, primarily with spirometry measurements.(83, 108) Physical activity effects on lung function values varied across studies. Six of twenty-two studies reported improvement in at least one spirometry measurement (forced vital capacity-FVC, forced expiratory volume in 1 second-FEV1, FEV₁/FVC or total lung capacity) between the intervention and control groups.(82, 84, 86, 87, 89, 102) Twelve studies showed no between-group differences in any lung function parameter with physical activity. (77, 85, 88, 90–92, 94, 98, 101, 105, 109, 113)

Asthma Control

Twelve studies reported data on asthma control. (77, 85, 89–92, 94, 98, 101, 108, 109, 113) Outcome measures ranged from the Asthma Control Questionnaire (ACQ) scores, to rescue inhaler use, to emergency room visits, to daily diary symptom reports, to number of reported asthma exacerbations. The results of effect of physical activity on asthma control varied across studies. Ten studies assessed participants with the ACQ (77, 85, 89–91, 94, 101, 108, 109, 113) with five studies showed a significant improvement in the ACQ score between groups.(77, 85, 108, 109) One study showed no difference in bronchodilator inhaler use between groups post-intervention,(92) while two showed a significant improvement.(85, 113) Another study showed a significant increase in asthma symptom-free days at 30, 60, and 90 days post-intervention between intervention and control groups.(98) Overall, seven studies showed a statistically significant improvement in at least one outcome measure of asthma control (Table 2).

Health Care Related Quality of Life

Eleven studies reported data regarding asthma-related quality of life. Nine out of eleven studies reported data using the asthma quality of life questionnaire (AQLQ or mini- AQLQ). (77, 86, 87, 89, 91, 92, 94, 109, 113) This is a four-domain questionnaire to evaluate the physical and emotional impact of asthma. The four domains include symptoms, activity limitation, emotional function, and environmental exposure.(114)

Two studies used the asthma-related health care quality of life questionnaire (ARHQoL).(98, 106) This tool assesses four domains of asthma-related quality of life, but rather than the emotional functions and environmental exposure domains, it assesses socioeconomic and psychosocial domains.(115) Two studies additionally reported the 36-item Short Form Survey (SF-36) data.(90, 92)

Of the nine studies that used the AQLQ to assess asthma-related quality of life, five showed a statistically significant improvement with the intervention in the mean AQLQ or at least one domain.(86, 87, 94, 109, 113) Improvement in the activity limitation (77, 86, 91, 94, 113), symptoms (86, 91, 94), and emotional function domains were found.(86)

One study employed the ARHQoL questionnaire and found a statistically significant improvement in the physical limitation and symptom frequency domains.(98) Another study found improvement in the physical component of the SF-36.(90) This between group difference was no longer significant at 3-month follow up. The other study evaluating SF-36 did not publish between group analyses.(92) Nearly half of the studies found some improvement in quality of life, while the remainder showed no statistically significant difference (Table 2).

Serologic Inflammatory Markers (Table 3)

Table 3:

Serologic Inflammatory Markers Results

STUDIES	Serologic Inflammatory Markers
	Th2 Inflammation	Th1 Inflammation	TH17 Inflammation	Pro Inflammatory Markers (hormones and IL8)	Other
Abd El-Kader et al. (2016)	↓IL-4*	↓IL-2*		↓Leptin*	↓LDL-c*
	↓IL-6*	↓TNF-alpha*			↓Calcium*
Boyd et al. (2012)	↔ECP		↓IL-1beta
	↓Eosinophils
Franca-Pinto et al. (2015)	↓FeNO			↓IL-8
	↓MCP-1*
	↓IL-6*
Freitas et al. (2017)	↓IL-4*	↓TNF-alpha*		↓Leptin*	↓CCL*
	↓IL-6*
Scott et al. (2013)				↓Leptin*
Toennesen et al. (2018)	↔Eosinophils

^*Significant

↓reduction

↔ no change

Six studies examined the effect of physical activity on serologic inflammatory markers.(77, 83, 89, 94, 101, 109) The specific serologic markers varied across studies. Three studies showed a significant reduction in at least one inflammatory marker between the intervention and control group.(77, 83, 94) IL-6 was the only inflammatory marker that was consistently reduced in the intervention groups (reported as statistically significantly reduced between groups in three studies).(77, 83, 94) Three studies found no significant between-group differences in any inflammatory marker.(89, 101, 109)

Discussion

This systematic review aimed to discover and evaluate the available evidence to better understand the effects of physical activity in adults with asthma. Twenty-three randomized controlled trials of varying sizes were included in our analysis. Of our included studies, thirteen articles from seven studies were published abstracts. These were included for the fullness of the record as there is limited research in this area. Similar to other systematic reviews, our findings are limited by the presence and quality of data provided in the included studies. We found heterogeneity in the type, frequency and duration of physical activity interventions as well as the outcomes measures which precluded a meta-analysis. Nonetheless, our qualitative findings suggest that physical activity can improve asthma control, lung function and health care-related quality of life. None of the studies found worsening of asthma control during the physical activity interventions.

Asthma symptom control is an important patient-reported outcome and a key component of assessing asthma control.(116, 117) Of the thirteen studies that reported changes in asthma symptom control, over half found improvements in self-reported outcomes. Asthma control was assessed using different self-reported measures of asthma control with the most commonly used measure being Asthma Control Questionnaire (ACQ). In over half of the included studies, blinding of the outcome assessors was unclear, resulting in a potential threat to the validity of the assessments. Future physical activity interventions in asthma should include asthma control as a core outcome measure, utilize validated instruments and improve reporting of their trial methods, especially blinding of assessors and allocation concealment.(118) This would allow not only for more reliable results, but also ease in future collation of data for meta-analysis.

Our study suggests that physical activity can improve patients’ quality of life. Most of the studies included used the same standardized quality of life measure, the Asthma Quality of Life Questionnaire. Quality of life is a key factor when considering treatment interventions, as it is meant to provide an assessment of the patient’s perceived wellbeing.(119) It is generally viewed as a more holistic approach for patient assessment.(119) In asthma, the improvement in quality of life is important to note, as prior studies have demonstrated a lack of direct correlation between clinical measures and patients’ measured QoL. (120–122) Our study suggests that physical activity provides an improvement in patients’ perceived wellbeing, as most studies showed an improvement in overall quality of life. This aligns with the findings of a previous systematic review in physical activity training in asthma which found positive effects on the quality of life of asthma patients.(109) Our findings on improvement in quality of life also add to the literature in adults with asthma as the recently published systematic review on PA and asthma in adults did not assess quality of life outcomes.(21)

In our analysis, almost half of the studies showed evidence of improved resting lung function post-intervention. This is consistent with the findings from a review by Avallone and colleagues in 2012 who reported improvements in lung function (e.g., PEF, FEV₁, FEF₂₅.(21) Additionally, in the meta-analysis conducted by Hansen et al, lung function slightly increased, in favor of exercise training (Standard Mean Difference: −0.36, −0.72 to 0.00).(21) This suggests physical activity improves lung function parameters, but further research is needed to better define this change/effect.

Our systematic review was unique from prior studies in that it also evaluated systemic markers of inflammation as an outcome of interest. Physical activity’s effect on inflammatory markers could serve as a pathway to better understand the physiological benefit of exercise. Six of the studies included in our analysis assessed for changes in serum inflammatory markers with physical activity. Interleukin-6 (IL-6) was the most frequently measured biomarker and was found to decrease with the physical activity intervention in three studies. IL-6 has been shown to be elevated in obese patients with asthma.(123) Recent studies have also shown that IL-6 levels are associated with severe asthma, and together with the results reviewed here (77, 83, 94), suggest that this marker should be further considered as a target for intervention studies.(124)

This adds to prior data evaluating changes in systemic inflammatory markers with physical activity in asthmatic individuals. When comparing physical activity versus sedentary behavior in asthmatic individuals, one study found no difference in airway inflammation, but found a significant decrease in serologic C-reactive protein levels with increased physical activity.(125) This reinforces the importance of evaluating not only direct lung parameters, but also systemic markers of inflammation. Further studies evaluating changes in IL-6 and other systemic inflammatory markers is needed to better understand this relationship.

Limitations of the studies found in our review include short follow-up intervals, lack of assessment of general physical activity levels and generalizability of results. None of the studies included in our analysis assessed for maintenance of physical activity or outcomes longer than one year. Our study and others suggest these short-term interventions have positive effects on multiple asthma outcomes, but the potential for increased long-term benefits has not been evaluated. Studies in the future should assess the long-term benefits of these physical activity interventions on asthma outcomes.

Only three studies reported physical activity outcomes such as change in steps/day, leisure-time physical activity or total physical activity levels.(77, 89, 91) In order better understand the mechanisms of physical activity on asthma outcomes, overall physical activity measures need to be assessed and reported in studies. Additionally, many studies employed a supervised, gym-based physical activity intervention that may not be feasible or accessible for most asthma patients. These clearly defined interventions suggest physical activity improves asthma outcomes, and future research will need to investigate if this extends to more accessible modes of physical activity, such as lifestyle physical activity.(126)

A limitation related to our review was the inability to conduct a meta-analysis due to the heterogeneity of results. Hansen et al. reported a meta-analysis assessing lung function, asthma control, and airway inflammation.(21) However, deriving a summary effect size for this highly heterogeneous body of literature could be misleading because of the different nature and intensity of the interventions. Instead, we used vote counting synthesis to analyze our findings as we were able to summarize the diverse results across the studies. This method has its known limitations including not being able to determine a small or large effect.(127) Using a vote counting method also allowed us to comprehensively evaluate the role of physical activity in asthma outcomes, so we did not need to limit our outcome measures.

Despite the heterogeneity in design and outcomes, our summative report of literature in physical activity’s effect on asthma found improvements in asthma control, quality of life, lung function and serum inflammatory markers. This aligns with recent observations, which note the growing body of evidence supporting aerobic exercise as a treatment for asthma.(15) Currently, the Global Initiative in Asthma nor the National Heart, Lung, Blood Institute (NHLBI) Asthma Guidelines do not provide specific recommendations for physical training or exercise as a means of asthma control.(2, 128) Even the December 2020 NHLBI update to the asthma management guidelines does not mention the potential benefit of physical activity.(129) Physical training was well tolerated among people with asthma in the included studies. Based on this data, individuals with asthma should be encouraged to participate in regular exercise training to improve asthma outcomes, without fear of symptom exacerbation. More research is needed to optimize these exercise interventions and determine the best methods to incite behavior change.

Highlights Box:

1. What is already known about this topic? Low levels of physical activity are associated with poorer asthma outcomes. Engaging in regular physical activity might have a beneficial impact in adults with asthma.

2. What does this article add to our knowledge? Physical activity is safe among adults with asthma with varying severity levels. Physical activity favors improvements in lung function, asthma control, quality of life and serologic inflammatory markers.

3. How does this study impact current management guidelines? Our findings suggest, in addition to the general health benefits of physical activity, regular physical activity may be a potential non-pharmacologic treatment in adults with asthma.

Abbreviations

ACQ

Asthma Control Questionnaire

AQLQ

Asthma Quality of Life Questionnaire

CCL

chemokine ligand

CINAHL

Cumulative Index to Nursing and Allied Health Literature

CRP

C-related protein

CXCL

chemokine ligand

ECP

eosinophil cationic protein

EMBASE

Excerpta Medica database

FEV1

Forced expiratory volume in one second

FVC

Forced vital capacity

GINA

Global Initiative for Asthma

HDL-c

high density lipoprotein-c

IL

interleukin

LDL-c

low density lipoprotein-c

MCP-1

monocyte chemoattractant protein

PRISMA

Preferred Reporting Items for Systematic Reviews and Meta-analysis

PTH

parathyroid hormone

SF-36

36-item short form survey

TNF-alpha

tumor necrosis factor-alpha