Children with neuromuscular diseases (NMD) are at risk of morbidity and mortality because of progressive respiratory muscle weakness and ineffective cough. Inspiratory muscle training (IMT) aims to preserve or improve respiratory muscle strength, thereby reducing morbidity and improving health-related quality of life (HRQoL).
To describe the safety and feasibility of a 6-week IMT programme using an electronic threshold device (Powerbreathe®). Any adverse events and changes in functional ability, spirometry, peak expiratory cough flow (PECF), inspiratory muscle strength and HRQoL (Pediatric Quality of Life [PedsQL]) were recorded.
A convenience sample of eight participants (
There were significant pre- to post-intervention improvements in upper limb function and coordination (
Inspiratory muscle training (at an intensity of 30% Pimax) appears safe, feasible and acceptable, in a small sample of children and adolescents with NMD and was associated with improved inspiratory muscle strength, PIF and upper limb function and coordination.
Larger, longer-term randomised controlled trials are warranted to confirm the safety and efficacy of IMT as an adjunct respiratory management strategy in children with NMD.
Children with neuromuscular diseases (NMD) often present with respiratory morbidity because of underlying progressive respiratory muscle weakness (Chatwin et al.
Inspiratory muscle weakness leads to limited sigh capacity, shortening and atrophy of respiratory muscles and chest wall, with associated poor ventilation, ineffective cough and decreased cardiorespiratory function (Benditt
Variable reports on the effectiveness of IMT have been published, with outcomes potentially influenced by learning effect; the type, nature and severity of NMD; as well as the specificity, dosage and intensity of training (Gozal & Thiriet
Despite the possible advantages, IMT in NMD remains controversial. The reason for this is that the use of IMT is mostly supported by low level evidence and there is a concern about its safety, particularly in patients with dystrophinopathies because of an associated increased risk of muscle damage, overexertion and fatigue during exercise (Eagle
Our pre-experimental, observational study aimed to determine the safety, adherence to and acceptability of a 6-week IMT programme using an electronic threshold device (Powerbreathe®). Changes in functional ability, spirometry, peak expiratory cough flow (PECF), inspiratory muscle strength, adverse events and HRQoL (Pediatric Quality of Life [PedsQL]) were recorded.
This was a pre-experimental, observational pre-test post-test study conducted in South Africa (SA) to determine the safety and feasibility of implementing IMT in children with NMD. In SA, specialised NMD centres are limited and therefore a non-probability convenience, purposive sampling frame was used. The two sites for recruitment were schools in Pretoria (Gauteng) that cater for a variety of children with special needs (physical and cognitive) and had full-time physiotherapists available to monitor IMT on a daily basis.
A convenience sample of nine children (5–18 years), with a confirmed diagnosis of NMD, was initially identified (from February to June 2017) from two schools catering for children with special needs. Children were excluded from participating if they were terminally ill, had a vital capacity (VC) < 25% predicted, or were enrolled in another clinical trial. One child with VC < 25% predicted was excluded after enrolment and eight children completed our study.
The first author performed a baseline bio-demographic assessment using routinely collected data extracted from patients’ files and measured weight and height. As a result of the majority of the participants being non-ambulant and four presenting with scoliosis, alternative measures for height measurement were implemented (left ulna length) in order to calculate height for spirometry test interpretation.
The level of ambulation was recorded and the functional ability of the lower and upper limbs was assessed. Lower limb function was measured subjectively using the 10-point Vignos scale, where 10 is the lowest possible score (confined to a bed) and one the highest (walks and climbs stairs without assistance). All participants were cognitively able to identify an appropriate Vignos score. Upper limb function was assessed objectively with the Brooke Scale, a six-point scale, where six is the lowest value (cannot raise hands to the mouth and have no functional hand movement) and one the highest (child can abduct their arms above their head in a full circle) (Lue et al.
Ten selected items of the Motor Function Measure (MFM) (items 14–23), with participants in a seated position (either in a chair or their wheelchair), were used to determine upper limb function and coordination. These items were selected so that all participants could be tested in a similar manner, despite their level of ambulation and could be performed in approximately 10 min (Bushby & Connor
Pulmonary function, safety, adherence, acceptability and HRQoL were assessed as follows:
Relaxed and forced spirometry:
Slow vital capacity (VC), FVC, forced expiratory volume in one second (FEV1) and PEF were measured pre- and post-intervention, with participants in a sitting position and using a portable spirometer (MicroLoop TM Spirometer, Carefusion, Germany).
At least three successful spirometry, cough ability (PECF) and inspiratory muscle strength (Pimax; sniff nasal inspiratory pressure [SNIP]) attempts were recorded, but testing at times continued up to five repetitions (Caruso et al.
In addition, the FVC and FEV1 z-scores were calculated (Global Lung Function Initiative [GLI]):
Cough ability:
Unassisted PECF was assessed pre- and post-intervention using a Mini-Wright Peak Flow Meter (Clement Clarke International Ltd, UK) (Chiang, Mehta & Amin
Inspiratory muscle strength:
Inspiratory muscle strength (Pimax) was measured every 2 weeks in order to inform titration of IMT intensity.
Inspiratory mouth pressure (Pimax) (from residual volume [RV]) and SNIP (from functional residual capacity [FRC]) were measured, in a sitting position, using an electronic handheld mouth pressure meter (MicroRPMTM; Carefusion, United Kingdom) (Chiang et al.
Peak inspiratory flow (PIF) (L/s) and strength (S)-index (cmH2O), with performance of one maximum inhalation from RV, were measured with the Powerbreathe K3® inspiratory threshold training device (HaB International Ltd, United Kingdom).
Any adverse events related to IMT were documented daily. Adverse events were classified as mild or moderate, which could include nausea or vomiting; pain or discomfort and gastric distension; whilst severe adverse events could include bradycardia, desaturation and/or hypoxia or barotrauma such as a pneumothorax. To monitor the participants’ perceived exertion during assessment and IMT (before and after training), a visual adapted 10-point Borg scale (OMNI scale) was used (
Adherence:
The number of completed IMT training episodes over the study period was downloaded from the threshold IMT devices (Powerbreathe K3®) and noted in the IMT adherence diaries.
Acceptability:
Satisfaction with the IMT intervention was rated on a 10-point visual analog scale (VAS) and the experience of IMT as verbally reported by participants and physiotherapists at the school was noted as open-text responses, after the 6-week intervention period.
Health related quality of life:
Self-reported, age-appropriate PedsQL InventoryTM (PedsQL Generic Score ScaleTM) questionnaires, for participants were completed pre- and post-intervention, with assistance from the first author as needed. The questionnaire consists of four subdomains: physical, social, emotional and school domain. The total PedsQL score and the Physical and Psychosocial (combined score of emotional, social and school functioning) domains were documented (Iannaccone et al.
The OMNI scale used for participants to describe their level of perceived exertion.
Inspiratory muscle training was performed for 6 weeks with an electronic handheld tapered-flow threshold device (Powerbreathe K3 ®, HaB International Ltd, United Kingdom). This device provides a visual stimulus for training, as every breath is counted down from 30, provided the breath quality is adequate to reach the preset threshold value in order to open the valve.
Participants performed IMT twice a day, 5 days a week, unless the child was absent from school or could not train because of other reasons. Participants started with three sets of 10 breaths (3 × 10 breaths), with a rest-interval of less than 60 s in-between sets, twice a day and progressed until they were able to complete 30 breaths consecutively without resting. The physiotherapists at the respective schools monitored the IMT sessions daily. Intensity for training was set at 30% of the participant’s best Pimax value, as measured at baseline and every 2 weeks by one of the authors and a research assistant. The intensity level was based on both evidence and the manufacturer guidelines suggesting that this is the minimum intensity required for improved inspiratory muscle strength and/or endurance (Hill et al.
Data were tested for normality using the Shapiro–Wilks
Cohen’s
This study was designed as a pilot study of a registered clinical trial (PACTR201506001171421). Approval was obtained from the institutional Human Research Ethics Committee (513/2015), school boards and principals, as well as Department of Education (Gauteng). Informed consent was obtained from parents or legal guardians of all participants and assent was obtained from the child participants.
Eight IMT-naïve, non-ventilated children (four boys and four girls; median [interquartile range {IQR}] age 12.21 [9.63–16.05] years), with a variety of NMD (Duchenne muscular dystrophy [DMD] [
The mean (± standard deviation [SD]) weight for age (WFA) z-score was -2.64 (± 2.66) and Body mass index (BMI) of -7.92 (± 7.53), suggesting severe wasting in most cases. For level of mobility, most participants (
Baseline bio-demographic data and functional ability of participants (
Participant | Gender | Age (years) | NMD type | Weight for age (z-scores) | BMI (z-scores) | Mobility | Baseline (Brooke scale) | Baseline (Vignos scale) |
---|---|---|---|---|---|---|---|---|
1 | Female | 17.58 | Myopathy | −3.31 | −4.61 | PA | 2 | 7 |
2 | Male | 11.25 | DMD | 0.51 | −0.42 | A | 1 | 2 |
3 | Female | 9.83 | SMA | −5.62 | −22.86 | NA | 3 | 9 |
4 | Female | 9.42 | Neuropathy | 1.06 | 0.04 | NA | 3 | 9 |
5 | Male | 15.42 | DMD | −5.95 | −14.28 | NA | 3 | 9 |
6 | Male | 16.67 | DMD | −3.64 | −7.51 | NA | 3 | 9 |
7 | Female | 13.17 | SMA | −3.49 | −7.29 | NA | 3 | 9 |
8 | Male | 8.33 | SMA | −0.66 | −6.42 | NA | 3 | 9 |
Regarding functional ability, the median for upper limb function was three on the Brooke scale (ability to raise a glass of 180 mL water to the mouth, but unable to raise hands above their heads) and nine on the Vignos scale for the lower limbs (using a wheelchair). Both the Brooke (
Pre-post inspiratory muscle training variables.
Variable | Baseline | 6 weeks (post-intervention) | |||
---|---|---|---|---|---|
Central tendency | Dispersion measure | Central tendency | Dispersion measure | ||
VC (L) | 1.44 | 0.66 | 1.56 | 0.83 | 0.18 |
FEV1 (L) | 1.37 | 0.56 | 1.37 | 0.65 | 0.99 |
FEV1 z-scores | −3.84 | 0.87 | −3.94 | 0.96 | 0.62 |
FVC (L) | 1.54 | 0.62 | 1.55 | 0.62 | 0.98 |
FVC z-scores | −4.31 | 1.26 | −4.33 | 1.07 | 0.94 |
PEF (L/min) |
150.5 | 98.5–168 | 161 | 121–196 | 0.05 |
FEV1/FVC (%) | 93.38 | 5.50 | 92.00 | 5.55 | 0.65 |
Pimax (cmH2O) | 37.88 | 14.60 | 48.13 | 16.45 | 0.01 |
Pimax % predicted | 60.66 | 23.02 | 76.38 | 24.70 | 0.01 |
SNIP (cmH2O) | 37.13 | 20.68 | 40.38 | 20.07 | 0.60 |
PIF (L/s) | 1.20 | 0.60 | 2.24 | 0.98 | 0.02 |
S-index (cmH2O) | 23.88 | 9.88 | 41.88 | 16.15 | 0.02 |
PECF (L/min) | 198.13 | 100.43 | 214.38 | 102.42 | 0.63 |
Brooke scale (/6) |
3.00 | 3.50–3.00 | 2.5 | 1.50–3.00 | 0.11 |
Vignos scale (/10) |
9.00 | 8.00–9.00 | 9.00 | 8.00–9.00 | 1.00 |
MFM (/30) | 25.75 | 1.83 | 26.75 | 1.39 | 0.03 |
Child PedsQL total (%) | 64.99 | 16.04 | 73.56 | 16.97 | 0.15 |
PedsQL (physical domain) (%) | 65.23 | 30.45 | 65.50 | 23.0 | 0.97 |
PedsQL (psychosocial) (%) | 64.80 | 12.83 | 77.08 | 17.10 | 0.10 |
All continuous variables are mean (± SD), unless otherwise stated;
, Median (IQR);
, Wilcoxon.
Participants were also asked if they followed any specific home programme. Two indicated that they were performing range of motion (ROM) exercises, which included passive movements and stretches daily or at least twice a week. Half of the participants performed some form of breathing exercise at least once a day. Most used manually assisted cough (MAC) techniques only during acute infections. None of the participants used mechanical insufflation-exsufflation (MI-E), lung volume recruitment (LVR) techniques, or other cough augmentation techniques such as glossopharyngeal breathing (GPB) or breath-stacking with a resuscitation bag or ventilator, which is similar to a previous SA survey (Human, Corten & Morrow
All participants had a restrictive pulmonary pattern of varying degrees, which remained unchanged pre- and post-intervention. There was also no change in spirometry values or cough ability (PECF) from baseline to 6 weeks post intervention (
Measures of inspiratory muscle strength of Pimax (
Mean sniff nasal inspiratory pressure change over training period (ANOVA,
Maximum inspiratory mouth pressure (Pimax) and the ability to create airflow when taking a deep breath (PIF) increased significantly throughout the intervention period and from pre- to post-test (
Mean maximum inspiratory pressure change over training period (ANOVA,
On
Mean peak inspiratory flow change over training period (ANOVA,
We did not find any association between any of the bio-demographic data (age, sex, height and BMI) or any pulmonary function measure (spirometry, PECF and inspiratory muscle strength) and no significant association between sex and change in Pimax (ΔPimax).
The mean (± SD) baseline IMT intensity was 11.4 (± 4.5) cmH2O, increasing to 13.6 (± 3.9) cmH2O and 13.4 (± 4.4) cmH2O at 2 and 4 weeks, respectively. Similar to Pimax, S-index and PIF; a significant change was observed over the period of IMT (
No adverse events directly attributable to IMT were reported during the 6-week intervention period and on average, over the first 2 weeks of IMT, the majority of participants reported a low OMNI score, which indicates low levels of perceived exertion.
Although there was a slight improvement in HRQoL (total PedsQL [%]) as reported by the children from baseline to post-intervention, this was not statistically significant (
Based on subjective feedback post-intervention, overall satisfaction with the IMT programme, on a 10-point visual analog scale, was extremely high, with a median (IQR) of 9 (9–10). All participants (
Open-text responses from participants and physiotherapists (
What did the participants have to say about IMT? | What did the physiotherapists have to say about IMT? |
---|---|
It is nice and relaxing! | Improvement noticed in his posture, breathing pattern and endurance. |
It helps me with breathing, I am less out of breath and my lungs work better. | Improved posture, confidence he even corrected others who participated in the study. He also presented with improved endurance and academic performance. |
My lungs are better, I breathe better (more normal) and I can sing louder in the choir! | Improved posture, confidence and increase in voice volume/projection, as she speaks and sings louder. Decreased use of asthma medication (cortisone) since she started with IMT. Regarding morbidity, she presented with a decreased frequency of respiratory illness during the winter period as compared with previous years: only one upper respiratory tract infection and one case of gastro-enteritis during the 6-week training period. |
Liked the training, it was nice. | When he started training, he was ill and had difficulty with IMT, after he received medication, he was much better and could start with training again. He did not fall ill again after he started training and his respiratory morbidity during the wintertime decreased (frequency of respiratory illness). |
Helps me to improve and keeps me healthier. I can go shopping with my mother and come to the physiotherapy department without having to stop every now and again because I am short of breath. | Inspiratory muscle training will be more beneficial in some patients as compared with others. For this patient specifically, it was clinically relevant, and she was very dedicated and will continue with training (patient adherence has a positive influence on the outcome of the intervention). |
It was fun, and now I can run faster and longer! | Similar to the previous participant: certain patients might benefit more from IMT than others, depending on their diagnosis, clinical presentation and motivation. This patient was very dedicated, motivated to exercise and seems to benefit from the intervention (clinical presentation). |
It makes my lungs better. | Any effort of maintenance is valuable for these patients and making them aware of effective breathing is to their advantage/provides clinical benefit for them. |
Liked the training, it helps me with my breathing. | Any effort of maintenance is valuable for these patients and making them aware of effective breathing is to their advantage/provides clinical benefit for them. |
IMT, inspiratory muscle training.
This is the first SA study to investigate the potential utility of IMT in children with NMD. This was a short-duration, pre-experimental, feasibility study with a small, heterogeneous convenience sample from one region, limiting internal and external validity. Despite these limitations, the results support the use of IMT in children with NMD with no adverse events, high participant adherence and satisfaction. Evidence of positive outcomes for pulmonary function (in particular Pimax, S-index and PIF) and suggested improvements in functional ability and HRQoL were found, similar to other studies (Human et al.
The clinical and bio-demographic profile of this cohort was similar to previous reports with the majority already non-ambulant, as expected for patients with DMD and SMA in this age range (median age of 12.2) (Gozal & Thiriet
Participants presented with moderate upper limb function (median Brooke scale = 3/6), generally good upper limb function and coordination (mean MFM score = 26/30) and poor lower limb function (median Vignos 9/10) at baseline. These findings were similar to a study amongst children and adults (
The baseline Brooke upper limb score of 3/6 suggests that participants were in a time of transition in their upper limb function, with progressive inability to reach overhead. This transition period coincides with pulmonary functional loss (Birnkrant et al.
Reduced upper limb function has been associated with progressive loss of pulmonary function particularly, FEV1 and FVC (Birnkrant et al.
As a result of their advanced disease progression and decreased diaphragmatic function, low spirometry measurements were expected in this population. The mean FVC values were however notably lower than a cohort study of children with DMD (7–15.5 years) (Aslan et al.
Cough ability (PECF) for participants at baseline was within normal range for children of 4–18 years (Bianchi & Baiardi
The inspiratory muscle strength values (Pimax and SNIP) at baseline for participants were similar. A slight, non-significant, improvement in mean SNIP post-intervention was observed, whilst Pimax showed a significant improvement after IMT, similar to other studies (Gozal & Thiriet
When the oropharyngeal muscles are affected in people with NMD, alternative inspiratory muscle strength measures, such as SNIP, might need to be considered because of poor oral control, which can render Pimax testing inaccurate (Birnkrant et al.
Conversely, it has also been reported that the diaphragm activation pattern in patients with NMD during a sniff manoeuvre is unknown (Stefanutti et al.
We also found a significant improvement in the ability to create airflow (PIF) when taking a deep breath (
No adverse events directly attributable to IMT were reported, similar to other IMT studies in children and adults with NMD (Wanke et al.
To the best of our knowledge, this is the first published report of HRQoL (based on PedsQL) of children with NMD in SA. Although we showed an improvement in the child-reported PedsQL total scores, this was not statistically significant (
Non-adherence might pose a challenge with IMT programmes, especially in children and adolescents (Eagle
Although our study suggests possible advantages of IMT in patients with NMD, this was a feasibility study with a small sample size, which included children of both sexes, with a variety of NMD and large age range (8.33–17.58) sampled from only one province in SA, and without a control group. The heterogeneity of the conditions included, the varying levels of physical and cognitive maturity, and the lack of a control group precludes conclusions being made as to the direct association between IMT and the reported outcomes. Factors such as sex, age, ethnicity and growth can influence absolute spirometry values (Birnkrant et al.
Furthermore, the variation in home programme adherence including the performance of stretches, passive movements and breathing exercises on a regular basis, could be confounding factors, which might influence the outcomes of upper limb and pulmonary function. The possible cognitive involvement commonly observed in children with dystrophinopathies such as DMD (Guglieri & Bushby
The convenience sample selection may also have introduced bias.
As a result of these limitations, external validity is compromised and although the results indicated possible clinical benefit of IMT, these findings cannot be extrapolated to larger NMD samples. To determine if IMT can bring a true change in inspiratory muscle strength and be sufficient to counterbalance the natural decline of pulmonary function in patients with NMD, larger and longer-term randomised controlled trials are needed before recommendations for clinical practice can be made.
This was a safety and feasibility study that showed short-term IMT intervention with a threshold tapered flow handheld device, set at an intensity of 30% Pimax, was associated with improved inspiratory muscle strength, PIF and upper limb function and coordination amongst participants. Despite the fact that child-reported HRQoL did not improve significantly following the intervention, the qualitative responses from participants and physiotherapists suggested that the effects of IMT may translate into both improved function and quality of life. This pre-experimental study provides preliminary data supporting the need for adequately powered clinical trials to confirm the safety, feasibility and efficacy of the use of IMT in children with NMD.
The authors would like to thank all the participants, their caregivers, the physiotherapists at the schools that assisted with the IMT and data collection, the principals of the schools involved and Department of Education (Gauteng) for providing permission to conduct this study. The authors would also like to acknowledge and thank Prof. Jennifer Jelsma for supervisory input as co-supervisor with the development of the protocol; Ms Sjaan Flanagan for her support, advice and assistance with the use of the Powerbreathe® devices; Dr Engela Honey who assisted with confirmation of diagnoses of participants and medical input where needed and Dr Lieselotte Corten for clinical input and assistance with development of the data collection sheets.
The authors declare that they have no financial or personal relationships that may have inappropriately influenced them in writing this article.
A.H. was responsible for conceptualisation, design and compilation of the protocol, designing and collation of data collection tools, data collection, analysing and interpreting data (with assistance of the supervisor), drafting and correcting the manuscript. B.M.M. was the primary supervisor, contributed with conceptualisation and design of the protocol, advisory capacity (clinical and academic), analysing and interpreting the data, revising and editing the manuscript.
This work forms part of a PhD study that was supported by the URC Equipment Grant (Western Cape); Sefako Makgatho Health Sciences University Research Development Grant; and the South African Society of Physiotherapy (PhD grant).
The data that support the findings of this study are available from the corresponding author, A.H., upon reasonable request.
The views and opinions expressed in this article are those of the authors and do not necessarily reflect the official policy or position of any affiliated agency of the authors.