Charles O.N.Winn ,Kelly A.Makintosh ,William T.B.Eddolls,Gareth Stratton ,Andrew M.Wilson ,Melitta A.MNarry ,*,Gwyneth A.Davies

a Swansea University Medical School,Singleton Campus,Swansea University,Swansea SA2 8PP,UK

b Applied Sports Technology,Exercise and Medicine(A-STEM)Research Centre,College of Engineering,Bay Campus,Swansea University,Swansea SA1 8EN,UK

c Norwich Medical School,University of East Anglia,Norwich,England NR4 7TJ,UK

Abstract Background:Higher levels of cardiorespiratory fitness are associated with reduced asthma severity and increased quality of life in those with asthma.Therefore,the purpose of this study was to evaluate the effectiveness of a 6-month high-intensity interval training(HIIT)intervention in adolescents with and without asthma.Methods:A total of 616 adolescents (334 boys;13.0 ± 1.1 years,1.57 ± 0.10 m,52.6 ± 12.9 kg,mean ± SD),including 155 with asthma(78 boys),were recruited as part of a randomized controlled trial from 5 schools(4 control and 1 intervention).The 221 intervention participants(116 boys;47 asthma)completed 6 months of school-based HIIT(30 min,3 times per week,10－30 s bouts at ＞90%age-predicted maximum heart rate with equal rest).At baseline,mid-intervention,post-intervention,and 3-month follow-up,measurements for 20-m shuttle run,body mass index (BMI),lung function,Pediatric Quality of Life Inventory,Paediatric Asthma Quality of Life Questionnaire,and Asthma Control Questionnaire were collected.Additionally,69 adolescents(39 boys(of the 36 with asthma there were 21 boys))also completed an incremental ramp test.For analysis,each group’s data(intervention and control)were divided into those with and without asthma.Results:Participants with asthma did not differ from their peers in any parameter of aerobic fitness,at any time-point,but were characterized by a greater BMI.The intervention elicited a signif icant improvement in maximal aerobic fitness but no change in sub-maximal parameters of aerobic fitness,lung function,or quality of life irrespective of asthma status.Those in the intervention group maintained their BMI,whereas BMI signif icantly increased in the control group throughout the 6-month period.Conclusion:HIIT represents an effective tool for improving aerobic fitness and maintaining BMI in adolescents,irrespective of asthma status.HIIT was well-tolerated by those with asthma,who evidenced a similar aerobic fitness to their healthy peers and responded equally well to a HIIT program.

Keywords:Body mass index;Cardiorespiratory fitness;Intermittent exercise;Intervention;Quality of life

1.Introduction

The prevalence of asthma and obesity have both increased dramatically over the past few decades,making them two of the most common chronic conditions in the UK.1,2This concomitant increase has led to suggestions that the 2 conditions may be causatively linked,3,4with overweight and obesity more prevalent in those who suffer from asthma.5Cardiorespiratory fitness has been suggested to be a key inf luential factor in the relationship between asthma and obesity,6although the nature and extent of this inf luence remains to be elucidated.Indeed,the inf luence of asthma on cardiorespiratory fitness requires clarif ication,with little consensus currently available in the literature.5,7－10These equivocal findings may be attributable,at least in part,to the exercise testing methodologies used to determine cardiorespiratory fitness.Specif ically,some studies reporting a lower aerobic fitness in those with asthma have used indirect estimates obtained from tests such as the 20-m shuttle run test.11,12Recent reports have highlighted the limitations associated with this measure,13issues that may be exacerbated in those with asthma given the commonly cited fear of exercise-induced bronchoconstriction,14leading to erroneous conclusions with regard to the pathophysiological inf luence of asthma.It is also pertinent to note the exclusive focus on peak oxygen uptake(VO2)in earlier studies concerning the inf luence of asthma on aerobic fitness.Whilst VO2is accepted as a strong prognostic tool in many clinical conditions,15it lacks direct applicability to many everyday functional abilities.

In addition to providing improvements in fitness,16－18exercise may elicit additional health benef its in those with asthma,such as reduced symptoms and severity and an improved quality of life.19－21Specif ically,a higher level of aerobic fitness in children is associated with a better quality of life,22while a greater body mass index(BMI)is related to a poorer quality of life.23,24Therefore,these measures should be targeted in future exercise interventions aimed at improving a population’s quality of life.However,whilst adolescents with asthma have identif ied exercise as one of their favorite activities,14few adolescents actively engage in exercise on a regular basis.5This finding may be attributable to the use of conventional,moderate-intensity,continuous exercise in previous exercise interventions in children with asthma.20,25Winn et al.14recently reported that adolescents with asthma prefer varied exercises,such as circuits or team games,with apprehension expressed towards long-distance running.Indeed,such variation would avoid monotony during sessions,which is associated with increased dropout rates.26

High-intensity interval training(HIIT)has received considerable attention in recent years because it has been identif ied as a time-eff icient method of exercise that can elicit signif icant improvements in both cardiorespiratory fitness and body composition in youth.27,28Given the potential relationship between asthma,obesity,and fitness and the decreased likelihood of exercise-induced bronchoconstriction owing to its intermittent nature,29HIIT represents a promising management strategy for those with asthma.However,it is important to acknowledge that some studies have raised concerns regarding the safety of HIIT,suggesting that it may be an inappropriate exercise modality for non-athlete populations.30In contrast with these concerns,children with asthma have previously been reported to tolerate HIIT similarly to their healthy peers.16,31Furthermore,whilst comparable data is not available in youth with asthma,healthy children and adolescents perceive HIIT as being more enjoyable to participate in compared with constant-intensity exercise,32with enjoyment a key component in eliciting the effort required for reaching high intensities.33Indeed,in adults with asthma,interval exercise is associated with lower ratings of perceived exertion and dyspnea,which is likely due to the rest periods.34Whether HIIT is similarly perceived to be enjoyable among adolescents with asthma remains to be elucidated,and the debate will continue regarding whether HIIT is associated with feelings of considerable discomfort that would prevent long-term adherence.35

Therefore,the aim of the present study was to ascertain the effectiveness of 6-month,f ield-based HIIT intervention in adolescents with asthma compared with their healthy peers.Furthermore,a secondary aim of this study was to use a 3-month follow-up to determine the sustainability of any adaptations elicited by the intervention.It was hypothesized that HIIT would lead to improvements in cardiorespiratory fitness and quality of life and a decrease in BMI in adolescents,irrespective of asthma,but that these benef icial adaptations would be lost within 3 months after the intervention cessation.This study is the first to implement a HIIT intervention in adolescents with asthma,which may be less monotonous than traditional continuous intensity exercise.If no differences are evident between participants with and without asthma,this finding will aid in informing future interventions for those with asthma and to decrease stigmatization and exclusion of those with asthma from everyday activities.Moreover,identifying a nonpharmacological intervention to reduce asthma symptoms and improve control and quality of life would be valuable.

2.Methods

2.1.Experimental design

The eXercise for Asthma with Commando Joe’s®(X4ACJ)program used in this study was a randomized controlled trial.Cluster randomization was used to select 1 intervention and 4 control schools in South Wales,matched for free school meal status.The exercise intervention began at the start of the school year in September and ended in March,with data collection continuing to July.Ethical approval was granted by Swansea University Medical School and the College of Engineering research ethics committees (Ref:140515 and PG/2014/29).Parent/guardian and head teacher written consent,in addition to child written assent,were obtained before participation.

2.2.Participants

To calculate the number of participants required to power the intervention study,the Paediatric Asthma Quality of Life Questionnaire was used as the primary outcome variable.It was calculated that,to achieve 80% power with an effect size of 0.5,which is deemed the minimal change considered clinically signif icant (conf idence level 0.05),a sample of 132 participants would be required.Owing to the prevalence of asthma and the pragmatic nature of only being able to conduct the intervention in 1 school,44 participants with asthma were required for the intervention,with the remaining 88 with asthma required as controls.To increase the statistical power of the study,2 healthy participants were sought to be recruited for every adolescent with asthma from both the intervention and the control schools.For the more sensitive subsample measures,to achieve an 80% power and 0.05 conf idence level,8 participants were required in each group.In total,616 adolescents(334 boys;Table 1),of which 155 had asthma(78 boys),agreed to participate in the study.A total of 221 participants (116 boys) were recruited from the intervention school,of which 47 suffered from asthma (24 boys).Asthma severity was assessed using the Global Initiative for Asthma guidelines36and was classif ied as mild,moderate,or severe according to the medication step required to achieve asthma control.For the purpose of analysis,moderate and severe asthma were grouped to power the statistics.Participants were excluded if they did not have stable asthma(n=3),if the participant had been admitted to hospital owing to their asthma in the last 6 weeks,visited their doctor because of their asthma becoming worse in the last 3 weeks,had a severe attack of asthma owing to exercise,or if they had ever been admitted to intensive care because of their asthma.

Table 1Anthropometric measures for participants within intervention and control for asthma and without asthma.

2.3.Intervention

The intervention design was devised based on formative work.14The intervention consisted of a 6-month HIIT program,delivered by a Commando Joe’s®personal trainer,involving 30-min sessions 3 times per week (Monday,Wednesday,and Friday).Participants were able to attend sessions before or after school but were asked to attend only 1 session per day.The sessions consisted of a combination of circuits and game-based activities (Table 2) lasting between 10 s and 30 s,followed by an equal period of rest (1:1 work-to-rest ratio).Throughout each exercise bout,participants were asked to exercise maximally,with exercise activities designed to elicit a heart rate (HR) of ＞90% of HR maximum (HRmax).37HRmaxwas predicted according to Tanaka et al.,38whose predictions have been validated for use in children and adolescents.39During each session the participants’HR was continuously monitored(Activio Sport;Activio AB,Stockholm,Sweden),and those who were not achieving the target HRs were individually encouraged to do so.Attendance and effort were further incentivized by a reward-based system whereby those who regularly engaged were entered into a prize drawing at the mid-intervention and end-of-intervention points.Those in the control group engaged in their usual day-to-day activities.

Table 2Detailed examples of exercises.

2.4.Procedures

Measurements were taken from both intervention and control groups at 4 time-points (baseline,mid-intervention,post-intervention,and 3-month follow-up)irrespective of condition.

2.4.1.Anthropometrics

Stature and body mass were measured according to the techniques outlined by the International Society for the Advancement of Kinanthropometry.40Stature,sitting stature and waist circumference were measured to the nearest 0.1 cm(Seca 213;Seca GmbH,Hamburg,Germany) and body mass to the nearest 0.1 kg (Seca 876;Seca GmbH).BMI wassubsequently calculated and grouped using age-and sex-specif ic child percentiles.41Maturity offset was calculated according to Mirwald et al.,42and lower limb length was calculated as the difference between stature and sitting stature

2.4.2.Lung function

Forced expiratory volume in 1 s(FEV1),forced vital capacity (FVC),FEV1/FVC ratio,peak expiratory flow,and forced expiratory flow between 25% and 75% of vital capacity(FEF25－75) was measured using a portable dry spirometer(Alpha Spirometer;Vitalograph Ltd.,Buckingham,UK).Participants were asked to sit up straight,breathe in as deeply as possible,place their lips around the mouthpiece tube and,when they were instructed,“blow out” into the mouthpiece as hard and as fast as possible until no further air could be exhaled;this was explained and demonstrated before the test.Each participant was asked to complete 3 acceptable tests,def ined as each exhalation being within 5% of the other two.The best of the 3 acceptable measurements was used,as recommended by the American Thoracic Society guidelines43and by the standardized protocol.44The best value was then expressed as a percentage of the age－sex－stature predicted value.45

2.4.3.Fractional exhaled nitric oxide(FeNO)

FeNO was measured before spirometric testing.The FeNO test was performed in accordance with the American Thoracic Society guidelines.46Participants were asked to completely exhale and then inhale to total lung capacity through the device (NIOX MINO;Aerocrine AB,Solna,Sweden) before immediately exhaling for 10 s at 50 ± 5 mL/s (mean ± SD).Visual and audio cues were provided by the computer software throughout.One test was completed at all time-points except the 3-month follow-up.The final 3 s of exhalation were evaluated.

2.4.4.Asthma control

Asthma control was assessed using the Asthma Control Questionnaire(ACQ),47which consists of 7 items focusing on reliever inhaler use and symptoms over the previous week and the participants’ FEV1score.Items on the ACQ are scored from 0 to 6,with ACQ scores of ≤0.75 or ≥1.5 indicating well-controlled or poorly controlled asthma,respectively.The ACQ has been validated in children between the ages of 6 and 16 years47and was found to be responsive to change in asthma control with a minimal important difference (MID) of 0.52 ±0.45.Internal consistency for the ACQ,measured using Cronbach’s α coeff icients,48was deemed acceptable(α:0.73－0.82).

2.4.5.Asthma-related quality of life

The Paediatric Asthma Quality of Life Questionnaire(PAQLQ) was used to compare the asthma-specif ic quality of life between those in the intervention and the control groups,as well as to assess the changes over the course of the intervention.Specif ically,the participants were asked to recall the previous week in response to 23 questions (scored on a Likert scale from 1 to 7),with a higher score indicative of a better asthma status.The PAQLQ questions are divided into 3 domains,including activity limitations (5 questions),symptoms (10 questions),and emotional function (8 questions),with a mean score for each domain and a total overall score.The PAQLQ has been validated in children between the ages of 6 and 16 years49and was found to be responsive to change in quality of life with a MID of 0.5.Internal reliability for the PAQLQ was deemed excellent(α:0.96－0.97).

2.4.6.Quality of life

The Pediatric Quality of Life Inventory(PedsQL)Teenager Report Version 4.050was used to compare the perceived quality of life between those participants with and without asthma and to assess any changes throughout the intervention.The participants were asked to recall their previous week and answer questions accordingly.A widely validated measure in adolescents aged 12－18 years,51－53the 23-item PedsQL consists of domains on the participants’ physical,emotional,social,and school functioning quality,with higher scores indicating a better quality of life.Internal reliability for the PedsQL was deemed excellent(α:0.89－0.90).

2.4.7.Cardiorespiratory fitness

2.4.7.1.Twenty-meter shuttle run

Cardiorespiratory fitness was estimated using the 20-m progressive shuttle run test,a previously validated field measure in children.11The test required participants to walk or run between 2 lines 20-m apart in time with pre-recorded beeps that progressively increased in speed throughout the test.The number of shuttles completed before voluntary exhaustion was recorded.

2.4.7.2.Peak VO2

A total of 69 adolescents (39 boys) inclusive of 36 with asthma (21 boys) were selected using stratif ied randomization to complete incremental ramp tests.The groups were stratif ied for age,sex,and condition to provide a representative sample of the wider population.Participants performed an incremental ramp exercise test to volitional exhaustion on an electromagnetically braked cycle ergometer (Ergoselect 200;Ergoline GmbH,Lindenstrasse,Germany),with seat and handlebar height individually adjusted for each participant.The ramp protocol consisted of 3 min of“unloaded”pedaling(0 watt(W))followed by an increase in work rate of 12－24 W/min depending on the age and height of the participant.Participants were asked to maintain a constant cadence (75 ± 5 revolutions per min) until voluntary exhaustion.Breath-by-breath pulmonary ventilation and gas exchange (VO2and VCO2) were recorded throughout (Oxycon Mobile;Jaeger GmbH,Hoechberg,Germany).

2.5.Data analysis

The peak VO2was taken as the highest 10-s mean attained before the end of the test.The gas exchange threshold (GET)was determined using the V-slope method.54The GET was also expressed relative to peak VO2(GET%VO2).To account for the inf luence of body mass on peak VO2,data were log transformed and population-specif ic power function ratios calculated using analysis of covariance (ANCOVA).Breath-by-breath data were then averaged into 10-s time bins,and the mean response time (MRT) and gain (ΔVO2/ΔW)were calculated according to the methods reported by Barstow et al.55Specif ically,the gain was determined by linear regression over 3 segments:S1,from 1 min into the ramp to GET;S2,from GET to peak VO2;and ST,over the total range of S1+S2.The baseline VO2was taken as the mean of the first 45 s of the last minute before the increase in work rate.The MRT was calculated as the point of intersection between the baseline VO2and a backwards linear extrapolation of the VO2by time slope from the onset of the ramp protocol.The MRT was also determined using 2 segments,S1(MRT1)and ST(MRTT).

2.6.Statistical analysis

Shapiro－Wilk tests were used to assess normality.After identif ication of normal distribution,the inf luence of asthma and the intervention,and their interaction,was assessed using a mixed-model analysis of variance(ANOVA) (groups were asthma intervention,non-asthma intervention,asthma control,and non-asthma control;time was collected as baseline,mid-intervention,post-intervention,and follow-up).Tukey’s post hoc analyses were conducted to ascertain where differences in time were found.If signif icant differences were found,mixed-design ANCOVA tests were run to adjust for baseline maturity because this factor may in part explain any changes in parameters.Baseline maturity was used in the ANCOVA because the test did not allow for time-varying covariates.Asthma-specif ic measures were analyzed using repeated measures ANOVAs.Data presented within the tables include the number of participants providing data at every time-point;therefore,the number of participants differ between measurements.All analyses were conducted using an intention-to-treat approach,thus including all participants with measures at any time-point.Data were subsequently analyzed using sensitivity analyses on participants who participated in the majority of the intervention sessions (＞70%).Eta-squared (ηp2) effect sizes were determined from baseline to 3-month follow-up.All statistical analyses were conducted using SPSS Version 22.0 (IBM Corp.,Armonk,NY,USA).All data are presented as mean± SD,with statistical signif icance accepted as p ＜0.05.

3.Results

The participants with asthma in the intervention group consisted of 87.2% with mild persistent asthma and 12.8% with moderate or severe asthma.The percentages of participants with mild persistent asthma and moderate or severe asthma in the control group asthma population were 77.1% and 22.9%,respectively.These percentages were similar in both the intention-to-treat and sensitivity analyses.Where no differences between the intention-to-treat and sensitivity analyses were found,results refer to the results obtained from the intentionto-treat analysis.Furthermore,no differences were observed when covarying for maturity offset or Tanner stages and are therefore not reported in this article.

3.1.Lung function

Lower FEV1% and FEF25－75values were found in participants with asthma at baseline,indicating more airway obstruction and more marked small airways obstruction,respectively.Those with asthma did not have an obstructed FEV1/FVC ratio.There were no changes in lung function over time in any group as highlighted by the mixed methods ANOVAs,which revealed no differences between intervention and control,asthma and nonasthma for lung function(p ＞0.05),according to group or time or time-by-group interaction.There was,however,a trend for FeNO to reduce in the intervention asthma group(Table 3).

3.2.Asthma control and quality of life

The intervention had no effect on asthma control or asthma-related quality of life.The MID for both the ACQ and PAQLQ was a change in score of 0.5.Both intervention and control asthma participants demonstrated similar results,with 33.4% and 35.1% scoring above the MID for the ACQ and 18.6%and 16.4%scoring above the MID for the PAQLQ.The results of the PedsQL revealed no signif icant differences between those with and without asthma in either the intervention or control group.The intervention was not associated with any signif icant change at any time-point in any of the groups(Table 4).

Table 3Lung function measures for participants within the intervention and control groups with asthma and without asthma.

3.3.BMI

BMI was found to be signif icantly greater in participants with asthma at baseline in comparison with their peers(21.7±4.4 kg/m2vs.20.0±3.4 kg/m2).There was a signif icant effect of time on BMI (F(2.23,782)=15.4,p ＜0.05,ηp2=0.04)and a signif icant difference between groups (F(3,351)=5.29,p ＜0.05,ηp2=0.04),but no interaction between time and group (F(6.68,782)=1.16,p=0.33,ηp2=0.01).Specif ically,although the intervention participants maintained their baseline BMI to post-intervention,BMI in the control participants,both with and without asthma,increased throughout the intervention (asthma:21.4 ± 4.4 kg/m2to 21.8 ± 4.4 kg/m2vs.non-asthma:19.8±3.3 kg/m2to 20.3±3.4 kg/m2,p ＜0.05).At the 3-month follow-up,all groups(intervention and control,asthma and non-asthma) had signif icantly greater BMI than they had at baseline(Table 1).

3.4.Twenty-meter shuttle run

No signif icant effects were found for group or time and no interaction was observed between group and time for the 20-m shuttle run.However,when applying sensitivity analysis,there was a signif icant effect of time (F(3,386)=5.44,p ＜0.05,ηp2=0.04) and a signif icant interaction of group by time(F(9,386)=3.23,p ＜0.05,ηp2=0.06).Post hoc analyses revealed a signif icant increase in the number of shuttles completed in both asthma and non-asthma intervention participants with time,which returned to baseline at the 3-month follow-up(Table 4).

3.5.Incremental ramp test

A signif icant effect of time and interaction between time and the group was observed,with no signif icant effect of group on peak VO2.When scaled for body size,these differences were maintained with time (F(3,138)=8.47,p ＜ 0.05,ηp2=0.16),group by time (F(9,138)=2.70,p ＜ 0.05,ηp2=0.15),and group (F(3,46)=1.55,p=0.22,ηp2=0.09).Post hoc analyses revealed signif icant increases in scaled peak VO2in both asthma and non-asthma intervention groups,with 3-month follow-up results showing a return to baseline levels.No differences were observed in either the asthma or nonasthma control groups across the intervention for peak or scaled peak VO2(Table 5).

There were no differences in GET between groups;however,there was a signif icant increase over time in all groups(F(2.23,138)=41.56,p ＜0.05,ηp2=0.48).There was no signif icant between-group difference for GET as a percentage of peak VO2.Post hoc analyses showed signif icant increases at post-intervention for the non-asthma intervention group and for both the asthma and non-asthma control groups.However,inclusive of the asthma intervention group,all groups signif icantly increased GET from baseline to 3-month follow-up.A sensitivity analysis also showed that there were no signif icant increases throughout the intervention in GET%VO2for participants in the non-asthma intervention group.There were no signif icant differences to either section of the MRT according to time,group,or time-by-group interaction across all timepoints.The gain,however,was found to signif icantly increase in the intervention asthma group for both S2and ST,with no signif icant differences observed in any of the other groups(Table 5).

3.6.Intervention intensity

Throughout the intervention sessions,exclusive of warm-up and cool-down,participants’ mean HR (155 ± 18 beats/min;78%HRmax±9%HRmax)and mean HRmax(188±18 beats/min;95%HRmax±6%HRmax)were calculated for each session.During the main body of the sessions,inclusive of both the exercise and rest intervals,HR exceeded the threshold(＞90%HRmax)for 24%of the total time.

3.7.Correlations

All measures were positively correlated with themselves between baseline and post-intervention,with the exception of the MRT and gain.A weak negative correlation was observed between BMI and fitness (r=－0.34,p ＜0.05),quality of life (r=－0.11,p ＜0.05),and lung function (r=－0.21,p ＜0.05) at baseline,but only between BMI and fitness(r=－0.33,p ＜0.05) at post-intervention.Fitness was also weakly correlated with quality of life (r=0.26,p ＜0.05) and lung function(r=0.34,p ＜0.05)at all time-points.However,scaled peak VO2was not associated with quality of life or lung function(p ＞0.05).

4.Discussion

This study was the first to evaluate the effectiveness of a 6-month field-based HIIT intervention in adolescents with asthma compared with their healthy peers.The main findings of this study were that(1)adolescents with asthma did not differ from their healthy counterparts in cardiorespiratory fitness at baseline,despite having a greater BMI,and (2) adolescents with asthma and their healthy peers demonstrated a similar response to the HIIT intervention.Specif ically,HIIT elicited signif icant improvements in cardiorespiratory fitness and maintained BMI in adolescents,irrespective of whether participants had asthma.However,HIIT did not elicit signif icant improvements in lung function,asthma control,or quality of life.These findings have important implications for the design of future interventions for those with asthma,highlighting that those with asthma are able to tolerate,and benef it from,similar exercise stimuli recommended for their healthy counterparts.This study demonstrates the fallacy of the perception that adolescents with asthma should be excluded from exercise,including HIIT because they are unable to participate and keep up with their peers when involved in similar activities.56,57

In accord with previous research and recent systematic reviews,27,58the present study found that HIIT elicited increases in cardiorespiratory fitness in adolescents.Specif ically,in the overall population,20-m shuttle run scores signif icantly improved,irrespective of condition,with no signif icant changes noted for the controls.Furthermore,both absolute and body size-scaled peak VO2increased throughout the intervention,providing evidence of true physiological improvements in cardiorespiratory fitness.Interestingly,participants in the asthma intervention group increased their scaled VO2to a greater extent than their non-asthma peers (19.2% vs.9.4%)and increased it considerably more than previously reported through conventional training programs for healthy adolescents.59This greater increase may be related to the (non-signif icantly) lower baseline fitness in those with asthma since baseline fitness has been reported to inf luence the magnitude of change elicited by an intervention in youth.60－62Although improvements in peak VO2after moderate-intensity exercise over a shorter time-frame16,17,63,64have been noted in those with asthma,the suitability of continuous exercise for those with asthma is questionable.Indeed,research has suggested that prolonged continuous exercise is not enjoyable26and may trigger the onset of asthma symptoms,29both of which are key barriers to exercise for those with asthma.14Furthermore,traditional endurance training,which typically involves a greater time commitment than HIIT,may also be less appealing than the suggested HIIT format to time-poor adolescents.27Importantly,the benef icial adaptations in the peak VO2for those with asthma were sustained in the 3 months after intervention cessation.Although it is beyond the scope of the present study to ascertain whether this was because these participants maintained a higher level of exercise post-intervention,this finding is encouraging for the long-term eff icacy of HIIT in adolescents with asthma.

In contrast with suggestions that submaximal parameters of aerobic fitness may demonstrate greater sensitivity to exercise stimuli than peak VO2,but in agreement with previous studies,65the absolute GET was unaffected by the intervention in the present study,irrespective of asthma status.This finding may indicate that training above the GET for short intermittent periods is not an effective strategy for increasing the GET in youth.These apparent age-and/or maturation-related changes in the relative GET are in contrast to previous reports26;thus,further research that ascertains the inf luence of growth and maturation on the GET is required.

Like the GET,the MRT did not show signif icant improvement following HIIT,irrespective of condition.These findings are perhaps surprising in that HIIT involves repeated transitions from rest to vigorous-intensity exercise.The MRT in the present study was longer than previously reported in healthy children,66,67but did not differ between those with and without asthma.The longer MRT may ref lect a lower level of aerobic fitness,although,given that aerobic fitness increased throughout the intervention with no concomitant speeding of the MRT,this explanation seems unlikely.The lack of effect of asthma in the present study is in contrast with the slower MRT reported in those with cystic fibrosis.68This finding may be attributable to the different etiologies of the 2 diseases and their inf luences on exercise tolerance.However,it may also be related to the relatively mild asthma of the majority of the participants in the present study.Additional inter-study comparisons are precluded because the ramp rate of the incremental test,which differs signif icantly between studies,profoundly affects the MRT.69Interestingly,there were no differences in the MRT between participants with and without asthma,suggesting that asthma does not impede the response to exercise.

In participants with asthma,the increase in gain observed over the intervention is suggestive of a positive adaptation in the delivery and utilization of oxygen by the muscles during exercise.70Although no differences in gain were observed at baseline,it is of note that S2and STgain increased post-intervention for participants with asthma to levels similar to those reported elsewhere in healthy adolescents.68This increase in gain may indicate that HIIT elicits different adaptations in those with and without asthma,although it may also be a function of the lower baseline level of gain in those with asthma,allowing greater capacity for improvement.The lower levels of aerobic eff iciency in participants with asthma may be related to a decreased lung function and may be a contributory mechanism for the onset of early fatigue and the perception that people with asthma are not as fit as their peers,although it is worth noting that gain and measures of lung function were not correlated in the current study.Indeed,Fielding et al.68found a reduced gain in patients with cystic fibrosis patients and suggested that this explained,at least in part,the reduced exercise intolerance in individuals with cystic fibrosis compared with their healthy peers.Importantly,the current study demonstrates that the gain for participants with asthma,but not for adolescents without asthma,can be improved with a HIIT program.

The findings in the present study are in accord with previous findings in non-asthma populations71in that cardiorespiratory fitness at baseline for those with asthma was found to have a weak but signif icant correlation with quality of life,highlighting the importance of exercise as a management strategy for those with asthma.However,despite this correlation and the increase in cardiorespiratory fitness observed in the current study,quality of life did not change over time,irrespective of treatment group or asthma status,which was contrary to the findings from previous studies involving exercise interventions.20,21Furthermore,in the present study there was no change over time for perceived asthma-related quality of life,symptoms,or asthma control.It could be postulated that the lack of improvement in asthma-related quality of life may be due to the mildness of the participants’asthma or to participants having high baseline values for quality of life,1,21,72which decreased the likelihood of an effect,or indeed the need for an effect.Finally,the lack of improvement in quality of life in the present study may have been due to the fact that HIIT decreased,or rather may not have increased,the participants’total time in physical activity as a result of the compensation effect.73This suggests that increased global physical activity may be more benef icial for quality of life rather than a specif ic HIIT intervention or increases in cardiorespiratory fitness.

Although our study is consistent with the majority of the literature in that we found that exercise did not affect lung function,1it is pertinent to note that 2 studies reported a signif icant increase in FEV1% (8%－20%) and that both studies implemented intermittent training.29,31This discrepancy may be related to the mildness of asthma among participants in our study or to the duration of our HIIT intervention.Although the duration in our study was longer than in many previous studies,72,746 months may not have been suff icient to elicit signif icant adaptations in lung function.It is interesting to note that both studies that previously reported benef icial adaptations in lung function involved younger,largely pre-pubertal,children.29,31Furthermore,the actual exercise time,despite being based on intermittent bouts,was signif icantly longer in Latorre-Romˊan et al.,31although the participants in Sidiropoulou et al.29had exercise-induced bronchoconstriction rather than asthma per se.These factors therefore limit additional conclusions that can be drawn as to the discrepancy in these findings with regards to lung function.

In accordance with previous studies,the current findings suggest that BMI increases with age in youth.75It is important to note that the intervention was able to maintain the baseline BMI and prevent an increase in BMI in those with and without asthma.Given that childhood obesity is known to track strongly into adolescence and adulthood,with evidence suggesting that 80% of obese adolescents will become obese adults,76the current findings may have important implications in terms of effective exercise interventions that may help to ameliorate this increase.Furthermore,exercise and physical activity have previously been suggested to be inf luential in the self-management of asthma.77The present study is the first to address whether HIIT may aid in the non-pharmacological management of asthma.Although the maintenance of BMI and increased fitness are promising findings,HIIT did not improve lung function,asthma control,and quality of life.Therefore,taking all findings together,6 months of HIIT may not be effective at improving mild asthma in adolescents.Nevertheless,maintenance of BMI and increased fitness as a result of HIIT may make it an important non-pharmacological strategy in the management of asthma.

There were no differences in fitness or trainability between adolescents with and without asthma;these findings have important practical implications for the design of future exercise interventions.Indeed,this study should decrease stigmatization of adolescents with asthma.Moreover,there were no exacerbations throughout the study,demonstrating that HIIT is safe to use in adolescents with mild asthma.A key strength of the present study was the use of more sensitive measures of aerobic fitness (GET,MRT,and gain),which have not previously been assessed across multiple time-points in adolescents with asthma.Nonetheless,several limitations to the study should be acknowledged.As with any exercise intervention,there may have been a self-selection bias since recruited participants volunteered to participate.Furthermore,although the HIIT intervention used formative research in its design,14participants who signed up for the intervention either committed fully to attending a large proportion of the sessions throughout the 6 months or attended only minimally over the intervention.This finding may indicate that the HIIT intervention is effective for those who fully engage in it,but that it is not acceptable to everyone.Although the utility of the intervention may be questioned,it may be that the timing of the exercise sessions (both before and after school) reduced participation and that if more optimal timing was possible,a stronger adherence could be achieved.

5.Conclusion

HIIT,a previously underused method of managing asthma in adolescents,may be an effective tool for increasing peak aerobic fitness and preventing an increase in BMI in adolescents,irrespective of asthma.This study adds to the literature by demonstrating that adolescents with asthma elicited similar physiological adaptations in comparison to their healthy peers,thereby demonstrating that asthma does not inf luence aerobic fitness or trainability in adolescents.Furthermore,the lack of exercise-induced asthma attacks suggests that HIIT is safe for,and well-tolerated by,adolescents with asthma.

Acknowledgments

The authors thank the students and staff of the schools involved with the planning and execution of the study.Also,the authors would like to thank Nicholas Wade and all others that assisted with data collection.This work was funded by the Asthma UK Centre for Applied Research(AUK-AC-2012-01)and Swansea University Medical School.Commando Joe’s®implemented the intervention and also assisted in funding for co-author WTBE.

Authors’contributions

CONW conceptualized and designed the study,collected data,conducted the statistical analysis,interpreted the data,and drafted the manuscript;MAM,KAM,and GAD conceptualized and designed the study,supervised the work,interpreted the data,and critically revised the manuscript;GS and AMW interpreted the data and critically revised the manuscript;WTBE collected data,interpreted the data,and critically revised the manuscript.All authors have read and approved the final version of the manuscript,and agree with the order of presentation of the authors.

Competing interests

The authors declare that they have no competing interests.