Intervention Review

You have free access to this content

Inspiratory muscle training for cystic fibrosis

  1. Brian W Houston1,*,
  2. Nicola Mills2,
  3. Arturo Solis-Moya3

Editorial Group: Cochrane Cystic Fibrosis and Genetic Disorders Group

Published Online: 21 NOV 2013

Assessed as up-to-date: 10 OCT 2013

DOI: 10.1002/14651858.CD006112.pub3


How to Cite

Houston BW, Mills N, Solis-Moya A. Inspiratory muscle training for cystic fibrosis. Cochrane Database of Systematic Reviews 2013, Issue 11. Art. No.: CD006112. DOI: 10.1002/14651858.CD006112.pub3.

Author Information

  1. 1

    Teesside University, School of Health & Social Care, Middlesbrough, Cleveland, UK

  2. 2

    Glenfield Hospital, Physiotherapy Department, Leicester, UK

  3. 3

    Hospital Nacional de Niños, Servicio de Neumología, San José, Costa Rica

*Brian W Houston, School of Health & Social Care, Teesside University, Victoria Road, Middlesbrough, Cleveland, TS1 3BA, UK. brian.houston@tees.ac.uk.

Publication History

  1. Publication Status: New search for studies and content updated (no change to conclusions)
  2. Published Online: 21 NOV 2013

SEARCH

 
Characteristics of included studies [ordered by study ID]
Albinni 2004

MethodsParallel design over 12 weeks.
Single centre in Austria.


ParticipantsTotal cohort: n = 27.

Age range: 6 - 18 years.

Gender mix: no information.


InterventionsIMT: no details; plus, cycle ergometer training 3 times per week.

Control: cycle ergometer training 3 times per week.


OutcomesFEV1, FVC, IMS, IME, MEC, perceived breathlessness, antibiotic use and ease or degree of expectoration.


NotesIME protocol: abstract only, no details given.


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Unclear riskDescribed as randomised, no details given.

Allocation concealment (selection bias)Unclear riskNot described.

Blinding (performance bias and detection bias)
All outcomes
High riskPerformance bias: clear difference between the interventions received.

Dectection bias: No reference to any blinding.

Incomplete outcome data (attrition bias)
All outcomes
Unclear riskNo information provided. Intention-to-treat: unclear.

Selective reporting (reporting bias)Unclear riskInsufficient information available to arrive at a conclusion.

Other biasUnclear riskInsufficient information available to arrive at a conclusion.

Amelina 2006

MethodsParallel design over 6 weeks.

Single centre in Russia.


ParticipantsTotal cohort: n = 20. Treatment group: n = 10; control group: n = 10.

Age range was not stated, but all were adults.

Gender mix: no information.

States no significant differences between groups in terms of gender, age, weight, height, pulmonary function.


InterventionsThreshold loading device:

Intervention group: 30% of PImax

Control group: 7cm H2O

Training regimen: 10 to 15 minutes twice daily for 6 weeks.


OutcomesFEV1, FVC, PImax, IC, RMS, RME and exercise capacity.


NotesAbstract only.


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Unclear riskThe authors only state that the allocation was random without explaining the process involved.

Allocation concealment (selection bias)Unclear riskNo details are provided.

Blinding (performance bias and detection bias)
All outcomes
High riskPerformance bias: The comparison group are referred to only as the "control group" with no mention of the intensity of the training used; i.e. if it was at "sham" or sub-maximal levels.

Dectection bias: No reference to any blinding.

Incomplete outcome data (attrition bias)
All outcomes
High riskNo statistical data is presented for the control group.

1 participant from the intervention group did not complete the trial; it was not stated whether they were included or excluded from the final analysis.

Selective reporting (reporting bias)High risk2 outcomes (respiratory muscle strength and dyspnoea) are mentioned as having been analysed, but no data are provided for them.

Other biasUnclear riskInsufficient information available to arrive at a conclusion.

Asher 1983

MethodsConsecutive, self-control design over 8 weeks.

Trial run in Canada, unclear if single or 2 centre.


ParticipantsTotal cohort: n = 11.

Age range: 9 - 24 years.

Gender split: no information.


InterventionsIMT: Inspiratory resistance, 15 minutes twice daily, no dosage.

Control: no details provided.


OutcomesIMS, Wmax, VO2max, VE and heart rate.


Notes


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Unclear riskDescribed as randomised, no details given.

Allocation concealment (selection bias)Unclear riskNot described.

Blinding (performance bias and detection bias)
All outcomes
High riskPerformance bias: no details of the control training regimen are provided = high risk.

Dectection bias: observer blind = low risk.

Incomplete outcome data (attrition bias)
All outcomes
High riskTwo participants were unable to satisfactorily perform the outcome measure PIMax, due to expiration up to residual volume resulting in coughing. The authors do not stipulate whether this occurred during the intervention or control phase of the trial.

Intention-to-treat: unclear.

Selective reporting (reporting bias)Unclear riskInsufficient information available to arrive at a conclusion.

Other biasUnclear riskInsufficient information available to arrive at a conclusion.

Chatham 1997

MethodsParallel design over 8 weeks.

Trial run in UK, not clear if single centre or 2 centres.


ParticipantsTotal cohort: n = 18. Treatment group: n = 9; control group: n = 9.

All participants were adults, but no specific age details or information on gender split given.


InterventionsIntervention: computer-generated through range IMT (TIRE) at 80% of individual capacity.

Control: threshold loading device at 30% of peak; the measure used is not named.


OutcomesChronic Respiratory Disease Questionnaire ('mastery' and 'emotion' elements), RMS and RME.


Notes


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Unclear riskDescribed as randomised, no details given.

Allocation concealment (selection bias)Unclear riskInsufficient information available to arrive at a conclusion.

Blinding (performance bias and detection bias)
All outcomes
High riskPerfomance bias: the training intensities employed (80% and "threshold" 30% training) could, potentially, have led the participants to know which group they were in.

Dectection bias: no reference to any blinding.

Incomplete outcome data (attrition bias)
All outcomes
Unclear riskInsufficient information available to arrive at a conclusion; no statistical data is presented for the control group.

Intention-to-treat: 3 from 18 (17%).

Selective reporting (reporting bias)Unclear riskAs this study (to date) is only published in abstract form it is unclear whether the reported outcomes are all that were analysed.

Other biasUnclear riskInsufficient information available to arrive at a conclusion.

de Jong 2001

MethodsParallel design over 6 weeks.

Single centre trial in the Netherlands.


ParticipantsTotal cohort: n = 16. Treatment group: n = 8; control group: n = 8.

Age range for total cohort 10 - 25 years. Treatment group: mean (SD) age = 17 (5.2) years; control group: mean (SD) age = 19 (5.5) years.

Gender split for total cohort: 8 male, 8 female. Treatment group: 4 male, 4 female; control: 4 male, 4 female.


InterventionsIMT: threshold loading: 20 minutes a day, 5 days per week, at 40% of PImax.

Control: threshold loading: 20 minutes a day, 5 days per week. at 10% of PImax.


OutcomesFEV1, FVC, Wmax, VO2max, VEmax, IME, perceived breathlessness, general fatigue, physical fatigue, reduced activity score, reduced motivation score, mental fatigue and dyspnoea.


NotesIME protocol: a commercially-available threshold-loading device (Threshold, Healthscan Products, Inc. USA) was used during an incremental loading procedure. In order to obtain pressures over 41 cm H2O an additional spring was inserted with a double-spring constant. Participants started inspiring from a threshold-loading device set at 30% of PImax for 2 min. The threshold load was then increased every 2 min in increments of 10% of Pimax. The maximal load was defined as the highest load which could be reached and maintained for at least 1 min as a percentage of PImax. The breathing pattern was not regulated.


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Low riskPatients were assigned to 1 of 2 groups by 5 patient factors: gender; age; FEV1; FVC; and BMI using the minimization method.

Allocation concealment (selection bias)Unclear riskNot described.

Blinding (performance bias and detection bias)
All outcomes
High riskPerformance bias: both training intensities were low; however, no attempt was made to ascertain whether the participants knew if the received the training intensity.

Detection bias: no reference to any blinding.

Incomplete outcome data (attrition bias)
All outcomes
High risk1 participant in the intervention group was withdrawn due to earache experienced whilst training at 40% of PImax.

Intention-to-treat: 1 from 15 (6%).

Selective reporting (reporting bias)Unclear riskInsufficient information available to arrive at a conclusion.

Other biasUnclear riskInsufficient information available to arrive at a conclusion.

Enright 2004

MethodsParallel design over 8 weeks.

Single centre in UK.


ParticipantsTotal cohort: n = 29. Treatment group 1: n = 9; treatment group 2: n = 10; control group: n = 10.

Age of total cohort (all adults): mean (SD) age = 22 (4.2) years. Treatment group 1: mean (SD) age = 24.8 (5.5) years; treatment group 2: mean (SD) age = 20 (4.7) years; control group: mean (SD) age = 21.3 (2.7) years.

Gender split of total cohort: 16 male, 14 female. Treatment group 1: 4 male, 6 female (according to table in paper); treatment group 2: 6 male, 4 female; control group: 6 male, 4 female.

All had similar age, height, weight and lung function at baseline.


InterventionsIntervention 1: IMT at 80% of "maximal inspiratory effort".
Intrevention 2: IMT at 20% of "maximal inspiratory effort".

Control: "No Training"

IMT is incremental maximal effort with progressively shorter rest periods, 3 times a week.


OutcomesFEV1(% predicted), FVC (% predicted), PImax, SPImax, heart rate, perceived exertion, dyspnoea and Chronic Respiratory Disease Questionnaire.


NotesSample size calculation undertaken such that study needed at least 9 patients in each group.


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Unclear riskNo information provided.

Allocation concealment (selection bias)Unclear riskNo information provided.

Blinding (performance bias and detection bias)
All outcomes
High riskPerformance bias: the comparison was "no training" making it clear to the participants which arm they were in.

Dectection bias: outcome assessors at the final data collection session, although they did not state whether this was the case at the initial assessment or even if the same assessors carried out all the assessments.

Incomplete outcome data (attrition bias)
All outcomes
Unclear riskNo mention is made of whether all participants completed the trial or not. Nor are there any statistical indications.

Intention-to-treat: unclear.

Selective reporting (reporting bias)Unclear riskInsufficient information available to arrive at a conclusion.

Other biasUnclear riskInsufficient information available to arrive at a conclusion.

Heward 2000

MethodsParallel design over 8 weeks.

Not clear if single or multicentre, authors from UK and USA.


ParticipantsTotal cohort: n = 39 (19 with CF and 20 matched healthy controls). Treatment group CF: n = 9; control group CF: n = 10.

Age of CF adults: mean (SD) age = 22.5 (3.5) years; age of healthy adults: mean (SD) age = 21.5 (3.5) years.

Gender matched groups


InterventionsTreatment: IMT at 80% of "maximal effort", no dosage stated.

Control: no training.


OutcomesVC, TLC.


NotesOnly CF patients eligible in this review, 10 healthy adults in another IMT group and another control group.

Abstracts only available.


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Unclear riskDescribed as randomised, no details given.

Allocation concealment (selection bias)Unclear riskNot described.

Blinding (performance bias and detection bias)
All outcomes
High riskPerformance bias: the comparison was "no training" making it clear to the participants which arm they were in.

Detection bias: no reference to any blinding.

Incomplete outcome data (attrition bias)
All outcomes
Unclear riskNo information provided.

Intention-to-treat: unclear.

Selective reporting (reporting bias)High riskThe post-training pulmonary function results were not presented.

Other biasUnclear riskInsufficient information available to arrive at a conclusion.

Sawyer 1993

MethodsParallel design over 10 weeks.

Single centre trial in USA.


ParticipantsTotal cohort: n = 20. Treatment group: n = 10; control group: n = 10.

Treatment group: mean (SD) age = 11.46 (2.45) years; control group: mean (SD) age = 9.76 (2.57) years.

No information on gender split.


InterventionsIMT: IMT at 60% PImax.

Control: Sham IMT at 10% PImax.


OutcomesFEV1, VC, FRC, IC, RV, TLC, RV/TLC, FEV1/FVC, MVV, exercise time.


Notes


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Unclear riskDescribed as randomised, no details given.

Allocation concealment (selection bias)Unclear riskNot described.

Blinding (performance bias and detection bias)
All outcomes
High riskPerformance bias: there was a clear difference in the intensity of training although no attempt was made to ascertain whether the participants in the training groups knew if they received the training intensity.

Dectection bias: outcome assessors at the final data collection session, although they did not state whether this was the case at the initial assessment or even if the same assessors carried out all the assessments.

Incomplete outcome data (attrition bias)
All outcomes
High risk2 participants removed from analysis and the reasons for this were explained; however, it is unclear which group(s) they were in.
Intention-to-treat: unclear.

Selective reporting (reporting bias)Unclear riskInsufficient information available to arrive at a conclusion.

Other biasUnclear riskInsufficient information available to arrive at a conclusion.

 
Characteristics of excluded studies [ordered by study ID]

StudyReason for exclusion

Howard 2000Inappropriate intervention - ACBT versus TIRE not IMT.

Irons 2012Inappropriate intervention - singing training and not IMT.

Keens 1977Study excluded as allocation not randomised.

Santana-Sosa 2013Although a form of IMT was used as an intervention it was used in combination with another exercise based intervention; therefore, it was impossible to attribute any observed changes to IMT alone.

Sartori 2008Observational study, no randomisation.

Vivodtzev 2013Inappropriate intervention - neuromuscular electrical stimulation prior to endurance training not IMT.

 
Comparison 1. IMT (80% of maximal effort) versus control

Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size

 1 Forced expiratory volume at one second (litres)1Mean Difference (IV, Fixed, 95% CI)Totals not selected

    1.1 Two to six months
1Mean Difference (IV, Fixed, 95% CI)0.0 [0.0, 0.0]

 2 Forced vital capacity (litres)1Mean Difference (IV, Fixed, 95% CI)Totals not selected

    2.1 Two to six months
1Mean Difference (IV, Fixed, 95% CI)0.0 [0.0, 0.0]

 3 Chronic Respiratory Disease Questionnaire (mastery)1Mean Difference (IV, Fixed, 95% CI)Totals not selected

 4 Chronic Respiratory Disease Questionnaire (emotion)1Mean Difference (IV, Fixed, 95% CI)Totals not selected

 
Comparison 2. IMT (60% of maximal effort) versus control

Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size

 1 Forced expiratory volume at one second (litres)1Mean Difference (IV, Fixed, 95% CI)Totals not selected

    1.1 Two to six months
1Mean Difference (IV, Fixed, 95% CI)0.0 [0.0, 0.0]

 2 PImax (cmH2O)1Mean Difference (IV, Fixed, 95% CI)Totals not selected

    2.1 Two to six months
1Mean Difference (IV, Fixed, 95% CI)0.0 [0.0, 0.0]

 
Comparison 3. IMT (40% of maximal effort) versus control

Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size

 1 Forced expiratory volume at one second (litres)1Mean Difference (IV, Fixed, 95% CI)Totals not selected

    1.1 Less than two months
1Mean Difference (IV, Fixed, 95% CI)0.0 [0.0, 0.0]

 2 Forced expiratory volume at one second (% predicted)1Mean Difference (IV, Fixed, 95% CI)Totals not selected

    2.1 Less than two months
1Mean Difference (IV, Fixed, 95% CI)0.0 [0.0, 0.0]

 3 Forced vital capacity (litres)1Mean Difference (IV, Fixed, 95% CI)Totals not selected

    3.1 Less than two months
1Mean Difference (IV, Fixed, 95% CI)0.0 [0.0, 0.0]

 4 Forced vital capacity (% predicted)1Mean Difference (IV, Fixed, 95% CI)Totals not selected

    4.1 Less than two months
1Mean Difference (IV, Fixed, 95% CI)0.0 [0.0, 0.0]

 5 Inspiratory muscle endurance (%PImax)1Mean Difference (IV, Fixed, 95% CI)Totals not selected

    5.1 Less than two months
1Mean Difference (IV, Fixed, 95% CI)0.0 [0.0, 0.0]

 
Comparison 4. IMT (20% of maximal effort) versus control

Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size

 1 Forced expiratory volume at one second (litres)1Mean Difference (IV, Fixed, 95% CI)Totals not selected

    1.1 Two to six months
1Mean Difference (IV, Fixed, 95% CI)0.0 [0.0, 0.0]

 2 Forced expiratory volume at one second (% predicted)1Mean Difference (IV, Fixed, 95% CI)Totals not selected

    2.1 Two to six months
1Mean Difference (IV, Fixed, 95% CI)0.0 [0.0, 0.0]

 3 Forced vital capacity (litres)1Mean Difference (IV, Fixed, 95% CI)Totals not selected

    3.1 Two to six months
1Mean Difference (IV, Fixed, 95% CI)0.0 [0.0, 0.0]

 
Table 1. Explanation of terms

TermExplanation

Continuous trainingTraining at 70% to 80% of maximum effort for 30 to 45 minutes. The percentage of maximal effort or the duration of the training (or both) may be adjusted depending on the goal of the training.

Elastic loadRefers to the load imposed by the stiffness of the lung and chest wall that must be overcome by the inspiratory muscles in order to generate inspiratory flow. Elastic loads are greater when breathing from a higher lung volume as a consequence of the associated decrease in lung and chest wall compliance. Imposing elastic loads has not been used to train the inspiratory muscles most likely due to the need for complicated equipment and poor clinical utility.

Forced expiratory volume at 1 second (FEV1)The volume of air expelled during the 1st second of forced exhalation from total lung capacity.

Forced vital capacity (FVC)The total volume of air expelled during a forced exhalation from total lung capacity.

Inspiratory capacity (IC)The maximum volume of air taken into the lungs during a maximal inhalation from functional residual capacity.

Forced expiratory flow 25-75% (FEF25-75%)The speed of the air leaving the lungs during the middle section of a forced exhalation.

Interval trainingPeriods of intense training interspersed with periods of recuperation; as with continuous training, the level of effort required during the training period may be adjusted to suit the individual and the intended goal. The period of recuperation will be adjusted accordingly.

Maximal inspiratory pressure (PImax)The maximum pressure generated by the inspiratory muscles against an occluded airway.

Resistive loadingRequires person to breathe through a narrow inspiratory pathway or aperture. The load imposed is dependent on inspiratory flow, i.e. when using resistive training devices, participants can reduce the load imposed by manipulating their breathing pattern. Breathing pattern, specifically inspiratory flow, should be controlled when using resistive inspiratory muscle training devices.

Threshold loadingRequires the person to inspire through a device which imposes a threshold load via either a weighted plunger system or a spring-loaded valve. The person needs to generate a critical inspiratory pressure, prior to the threshold valve opening and allowing inspiratory flow. Once the threshold valve is open, pressure and flow are largely independent and therefore the person is unable to reduce the load imposed by the device by manipulations in breathing pattern.

Total lung capacity (TLC)The maximum amount of air the lungs can hold when they are fully inflated.

Voluntary isocapnic (normocapnic) hyperpnoeaRequires the person to maintain a high level of minute ventilation for a specified period. Imposes a high flow, low pressure load on the inspiratory muscles which is analogous to the loads borne by the inspiratory muscles during periods of increased minute ventilation (i.e. during exercise). Requires the use of complex equipment to ensure stable levels of carbon dioxide in the arterial blood (PaCO2), so is rarely used in the clinical setting.