Вы находитесь на странице: 1из 8

Archives of Physical Medicine and Rehabilitation

journal homepage: www.archives-pmr.org

Archives of Physical Medicine and Rehabilitation 2018;-:-------


High-Intensity Respiratory Muscle Training Improves

Strength and Dyspnea Poststroke: A Double-Blind
Randomized Trial
Kênia Kiefer Parreiras de Menezes, PT, PhD,a Lucas Rodrigues Nascimento, PT, PhD,a,b
Louise Ada, PT, PhD,c Patrick Roberto Avelino, PT, MSc,a Janaine Cunha Polese, PT, PhD,d
Maria Tereza Mota Alvarenga, PT,a Mariana Hoffman Barbosa, PT, MSc,a
Luci Fuscaldi Teixeira-Salmela, PT, PhDa
From the aDepartment of Physiotherapy, Universidade Federal de Minas Gerais, Belo Horizonte, Minas Gerais, Brazil; bDiscipline of
Physiotherapy, Universidade Federal do Espı´rito Santo, Vitória, Espı´rito Santo, Brazil; cDiscipline of Physiotherapy, University of Sydney, New
South Wales, Australia; and dDepartment of Physiotherapy, Faculdade Ciências Médicas de Minas Gerais, Belo Horizonte, Minas Gerais, Brazil.

Objective: To examine whether high-intensity home-based respiratory muscle training, that is, with higher loads, delivered more frequently and
for longer duration, than previously applied, would increase the strength and endurance of the respiratory muscles, reduce dyspnea and respiratory
complications, and improve walking capacity post-stroke.
Design: Randomized trial with concealed allocation, blinded participants and assessors, and intention-to-treat analysis.
Setting: Community-dwelling patients.
Participants: Patients with stroke, who had respiratory muscle weakness (NZ38).
Interventions: The experimental group received 40-minute high-intensity home-based respiratory muscle training, 7 days per week, for 8 weeks,
progressed weekly. The control group received a sham intervention of similar dose.
Main Outcome Measures: Primary outcome was inspiratory muscle strength (via maximal inspiratory pressure), whereas secondary outcomes
were expiratory muscle strength (maximal expiratory pressure), inspiratory muscle endurance, dyspnea (Medical Research Council score),
respiratory complications (hospitalizations), and walking capacity (6-minute walk test). Outcomes were measured at baseline, after intervention,
and 1 month beyond intervention.
Results: Compared to the control, the experimental group increased inspiratory (27cmH2O; 95% confidence interval [95% CI], 15 to 40) and
expiratory (42cmH2O; 95% CI, 25 to 59) strength, inspiratory endurance (33 breaths; 95% CI, 20 to 47), and reduced dyspnea (-1.3 out of
5.0; 95% CI, -2.1 to -0.6), and the benefits were maintained at 1 month beyond training. There was no significant between-group difference
for walking capacity or respiratory complications.
Conclusion: High-intensity home-based respiratory muscle training was effective in increasing strength and endurance of the respiratory muscles
and reducing dyspnea for people with respiratory muscle weakness post-stroke, and the magnitude of the effect was higher, than that previously
reported in studies, which applied standard protocols.
Archives of Physical Medicine and Rehabilitation 2018;-:-------
ª 2018 by the American Congress of Rehabilitation Medicine

Stroke is the second leading global cause of death and the leading
Supported by Coordenação de Aperfeiçoamento de Pessoal de Nı́vel Superior (CAPES, grant
cause of disability.1,2 Post-stroke, weakness affects not only the
code 001), Conselho Nacional de Desenvolvimento Cientı́fico e Tecnológico (CNPq, grant no. upper and lower limb muscles, but also the respiratory ones.3
304434/2014-0), and Fundação de Amparo à Pesquisa de Minas Gerais (FAPEMIG, grant no. PPM People with stroke can exhibit considerable weakness of the res-
Clinical Trial Registration No.: NCT02400138.
piratory muscles, which may result in atypical breathing patterns
Disclosures: none. and decreased respiratory function.3-5 Furthermore, individuals

0003-9993/18/$36 - see front matter ª 2018 by the American Congress of Rehabilitation Medicine
2 K.K. Parreiras de Menezes et al

with stroke typically report dyspnea, not only during high-effort post-stroke? (2) Are any effects maintained beyond the training
tasks, but even under low effort demands, which in turn, may and/or carried over to walking capacity?
interfere with the performance of daily activities and limit com-
munity participation.5-8
Respiratory muscle training has been used to increase the Methods
strength of the inspiratory and expiratory muscles in people with
stroke.3,7-10 This training consists of repetitive breathing exercises
against external loads, using flow-dependent resistance or pressure Design
thresholds, to control the training intensity.4,7 Respiratory muscle A double-blinded, randomized controlled trial with concealed
training is based on the premise that respiratory muscles respond to allocation was undertaken. Patients with stroke were recruited
training stimuli by undergoing adaptations to their structure in the from the general community of a metropolitan city in Brazil, by
same manner as any other skeletal muscles, that is, when their fibers means of advertisements and by screening out-patient clinics in
are overloaded, both the proportion of Type I and the size of Type II public hospitals. A research assistant, not involved in recruitment,
fibers are increased.11 Therefore, to obtain training responses, res- compiled a computer-generated, random allocation sequence, in
piratory muscles need to work for longer, at higher intensities, and/ blocks of 4 participants. To ensure similar values of inspiratory
or more frequently, than usual. Furthermore, resistance training has pressure between groups at baseline, the randomization was
shown to increase the capacity of the muscle tissue to store energy stratified by the strength of the inspiratory muscles into weak
and enhance the secretion of neurotransmitters. In addition, the (45cmH2O) and very weak (<45cmH2O), which represents
smoothing of neural facilitation results in increases in muscle approximately half of the inspiratory pressure value to charac-
endurance.12 Previous studies also confirmed these findings, that is, terize weakness.19,20 The allocation sequence was placed in opa-
that respiratory muscle training resulted in improvements in que, sequentially numbered, and sealed envelopes, which were
strength and endurance of the respiratory muscles,8,13 mainly when held offsite by an independent researcher. After baseline mea-
protocols with higher training loads (30%) were applied.13 surement, the envelope was opened and group allocation was
Four systematic reviews with meta-analyses of randomized trials revealed. Participants were blinded to group allocation, by
examined the effects of respiratory muscle training post- informing them that they would receive respiratory training with
stroke.3,7,10,14 However, only 2 of the included trials of inspiratory or without progression, but not whether 1 method was deemed
training with substantial statistical heterogeneity (I2Z95%) led to superior to the other. All training devices were covered with
inconclusive findings.3,14 The 2 other reviews7,10 included 5-7 opaque material, so that the participants were blinded to the
randomized trials of reasonable quality and reported increases in training loads. Outcomes were collected within a laboratory
inspiratory strength of 7 cmH2O and in expiratory strength of 6 setting at baseline (wk 0), after training (wk 8), and after 1-month
cmH2O10 and 13 cmH2O.7 Although statistically significant, the follow-up (wk 12) by trained and blinded assessors. The study
clinical magnitude of the effect is relatively small.15,16 This may be reporting followed the CONSORT statement guidelines.20
due to the low dose of training (about 35% of baseline strength), its
short duration (about 4wk), and often not systematically progressed.
In addition, only 1 trial combined inspiratory and expiratory Participants and therapists
training.9 Because inspiratory and expiratory weakness is associated
with symptoms, such as dyspnea and ineffective cough with Participants were eligible if they were >3 months and <5 years
increased risk of aspiration pneumonia,4 training both inspiratory poststroke, were >20 years of age, had maximal inspiratory
and expiratory muscles could have the potential to optimize the pressure (MIP) <80 cmH2O or maximal expiratory pressure
benefits. Furthermore, findings were inconclusive regarding inspi- (MEP) <90 cmH2O,19 and were not undertaking any respiratory
ratory endurance, respiratory complications, and whether there was training. They were excluded if they had cognitive deficits, facial
any carry over to everyday activities. palsy, and/or any conditions, which prevented measurement or
A randomized controlled trial was undertaken to examine training. The training was delivered by 2 physiotherapists, who
whether high-intensity respiratory muscle training targeting both had >4 years of clinical and research experience in neurologic
inspiratory and expiratory muscles was effective. The benefit of rehabilitation. The trial was approved by the Institutional
high-intensity respiratory training on strength, endurance, and Research Ethical Committee (#40290114.8.0000.5149), and all
dyspnea has already been demonstrated in people with heart participants provided written consent.
failure.17 However, no studies which investigated the effect of
high-intensity training in patients with stroke were found. Interventions
Furthermore, if benefits carried over to activity, community
participation may be enhanced, given that walking capacity is a Participants undertook home-based training. They all received an
strong predictor of community walking in people with stroke.18 Orygen-dual valve,a a threshold-loading device, which allows
Therefore, the specific research questions were, as follows: (1) simultaneous training of both inspiratory and expiratory muscles
Is high-intensity home-based respiratory training effective with loads independently adjusted.17 The dose was 40 min/day, split
in increasing the strength and endurance of the respiratory into two 20-minute sessions (morning and afternoon), 7 times per
muscles and decreasing dyspnea and respiratory complications week, over 8 weeks. Each 20-minute session comprised 4-minute
sets of training, followed by 1-minute rest between the sets. Phys-
iotherapists taught the participants to perform the training in the
List of abbreviations: laboratory, but the training was home based. When required, a proxy
95% CI 95% confidence interval was instructed to help them. The physiotherapists visited the par-
MEP maximal expiratory pressure ticipants once a week in their homes. To record compliance, par-
MIP maximal inspiratory pressure
ticipants from both groups received a diary to register the time and

Respiratory muscle training improves strength and dyspnea 3

days of all training sessions and daily training volume. To were given 3 possible answers: “group with progression of training
encourage compliance, all participants signed a symbolic contract loads,” “without progression of training loads,” or “unsure.”28
of commitment to the training. Success of blinding was also determined by asking the assessors
if participants had revealed their group allocation or if they had been
Experimental group unblinded in any other way.
The training load was individually tailored and progressed, as
follows: the initial load was set at 50% of the participants’ Sample size
maximal baseline strength, for both inspiratory and expiratory
training. Once a week, during the home visit, the treating phys- The number of participants was calculated to reliably detect a
iotherapist measured the current values of inspiratory and expi- between-group difference of 15 cmH2O in inspiratory strength
ratory strength and readjusted the training load. (80% power, 2-tailed significance level of 0.05, expected dropout
rate of 15%). In a similar randomized trial,8 the strength of the
Control group inspiratory muscles was 5715 cmH2O. Based on the assumption
The control group received sham respiratory training using the that about 15% of participants could dropout during the course of
same device, without any resistance (0 cmH20) or progression. the study, the least number of participants needed to detect a 15
The participants received the same training schedule, including cmH2O difference between 2 independent groups was 18 per
the home visits, to avoid bias related to amount of attention. group. Thus, a target of at least 36 participants, in total, was set.29

Outcome measures Data analyses

All analyses were conducted on an intention-to-treat basis by an
Primary outcome independent researcher. Missing data were interpolated, by car-
Inspiratory muscle strength, that is, MIP, in cmH2O, was measured rying forward the last known value. Data collection returned 6
with a digital manovacuometer,b which was developed and vali- outcomes: inspiratory and expiratory strength (cmH2O), inspira-
dated at our university.21-23 For the MIP, participants were tory endurance (number of breaths), dyspnea (Medical Research
instructed to expire at near-residual volume. The participants were Council scale, 1-5), respiratory complications (number of hospital
allowed to practice twice and, immediately after, they were asked admissions), and walking capacity (m). Between-group differ-
to repeat the trials until 3 acceptable measurements were obtained ences over time (wk 8 0 and wk 12 0) for outcomes of
without an air leak for the duration of at least 1 second each. Two continuous or ordinal data were analyzed using 2*2 repeated
reproducible attempts were retained with a maximum difference
of 10%, and the highest MIP values were recorded and stored for Table 1 Baseline characteristics of the participants of both
analyses.21-23 experimental and control groups

Secondary outcomes Experimental Control

Expiratory muscle strength (MEP), in cmH2O, was also measured Group Group
by the same manovacuometer.21-23 The procedures were similar to Characteristic (nZ19) (nZ19)
those used for the evaluation of inspiratory pressure; however, the Age (y), mean  SD 6014 6711
participants were instructed to inspire closer to total lung capacity Time since stroke (mo), 2420 1612
and then, expire.21-23 mean  SD
Inspiratory endurance (number of breaths) was measured using Sex (men), n (%) 8 (42) 8 (42)
the flow-resistive loading POWERbreathe KH1.c Participants were >1 stroke episode, n (%) 6 (32) 5 (26)
asked to breathe against a submaximal load, that is, 50% of their Type of stroke, n (%)
baseline MIP, until task failure or up to maximum 7 minutes, Ischemic 12 (63) 15 (79)
following recommended procedures.24 Hemorrhagic 3 (16) 3 (16)
Dyspnea was measured using the Medical Research Council Unknown 4 (21) 1 (5)
scale, which is a 5-point scale. Scores range from 1 to 5, where Paretic side, n (%)
one indicates breathless only with strenuous exercise and 5 too Right 12 (63) 6 (32)
breathless to leave the house or when dressing/undressing.25,26 Left 6 (32) 11 (58)
The occurrence of respiratory complications (number of hos- Unknown 1 (5) 2 (10)
pital admissions) was measured by asking the participants whether Walking speed (m/s), 0.910.40 0.930.35
and how often they were admitted to a hospital due to respira- mean  SD
tory reasons. MIP (cmH2O), mean  SD 5817 5214
Walking capacity was measured by the distance covered, in - Weak participants 13 (68) 13 (68)
meters, during the 6-minute walk test. Participants were instructed (45-80cmH2O), n (%)
to walk along a 30-minute hallway and cover the maximum - Very weak participants 6 (32) 6 (32)
possible distance over 6 minutes following recommended (<45cmH2O), n (%)
procedures.27 MIP (% of the predicted), 6421 6018
Success of blinding was determined by asking the participants mean  SD
after the completion of the trial: “You participated in a trial with two MEP (cmH2O), mean  SD 7420 7824
groups receiving respiratory training. In one group, the training MEP (% of the predicted), 6024 6425
loads were weekly adjusted, whereas in the other, the loads did not mean  SD
change. Do you know to which group you were allocated?” They

4 K.K. Parreiras de Menezes et al

measures analysis of variance and reported as mean differences and/or MEP>90 cmH2O). Thirty-eight (16 men) were eligible and
(95% confidence interval [95% CI]). Between-group differences willing to participate. Table 1 gives the baseline characteristics of
over time for dichotomous data were reported as risk differences the participants of both groups. Significant between-group dif-
(95% CI), where the CI was calculated using the Newcombe- ferences in baseline measures were found only for variables
Wilson method.30 This type of reporting assists in determining related to the time since the onset of the stroke and inspiratory
the clinical interpretation and importance of the observed differ- endurance. Week 8 data are missing from 1 participant from the
ences, as well as the statistical significance of the findings.31 All control group, who withdrew from the trial before starting
analyses were performed with SPSS software (version 17.0).32,d training. Week 12 data are missing from 5 other participants (3
from the experimental group and 2 from the control group), who
did not return for the measurements. Figure 1 shows the flow of
Results participants through the trial.

Flow of participants Compliance

Fifty-six patients with stroke were screened between March 2016 Of the 37 participants, who returned to post-intervention assess-
and May 2017. Of these, 18 were ineligible (MIP>80 cmH2O ments, one of the experimental group ceased training after 2

Fig 1 Design and flow of participants through the trial.

Respiratory muscle training improves strength and dyspnea 5

weeks due to chest pain, the only adverse event. However, this


participant returned for all assessments (post-intervention and
follow-up measures). The 36 participants, who completed all the

1.2 ( 2.2 to
Wk 0

44 ( 13 to
training sessions, because 1 withdrew from the trial and another

Minus Control
24 (11-36)
30 (15-45)
26 (11-42)
Between-Group Differences

one did not complete the full training protocol, reported that they
performed the training, as recommended. However, 15 (32%) di-
Wk 12
aries were either lost or incomplete. The analyses of the diaries
which had complete information revealed that participants of both

groups performed at least 80% of the training. The main reasons

for skipping a training session were lack of time, forgetfulness,

1.3 ( 2.1 to

38 ( 22 to
and sickness. The mean number of home visits was 6.11.9 out of
Minus Control
27 (15-40)
42 (25-59)
33 (20-47)
Wk 0

the 7 planned visits. Week 8 assessments were conducted 1 week

later than intended in 6 cases.
Wk 8

Experimental Control
Wk 0

About 58% and 47% of the experimental and control groups,

respectively, thought that they were in the experimental group,
with most of the remaining participants being unsure. The asses-
Within-Group Differences
Wk 12



sors were unblinded in 11 (29%) of 38 cases (6 from the experi-

mental and 5 from the control group).


Effect of high-intensity home-based respiratory

Mean  SD of groups, mean  SD differences within groups, and mean (95% CI) differences between groups

(nZ19) Experimental Control

muscle training
Wk 0

The group mean  SD, within-group differences  SD, and

between-group differences (95% CI) for all outcomes are provided
Wk 8



in table 2. Analysis of variance revealed significant training effects

on MIP, MEP, and inspiratory endurance. Moreover, significant
interactions were found between time and group factors for the MIP

(FZ9.59, P<.001), MEP (FZ12.10, P<.001), and inspiratory

Experimental Control Experimental Control Experimental Control

endurance (FZ12.77, P<.001), indicating that the groups showed

different behaviors over time. Significant between-group differ-
Wk 12

ences were also found for dyspnea at post-intervention (P<.01) and

328162 369121
1.31.4 1.11.4

follow-up (P<.05) measures. No significant differences were found

(nZ19) (nZ19)
6121 9523
8629 11726
1817 3833

for the other evaluated measures: walking capacity (FZ1.28,

P<.29) and respiratory complications (PZ.99).
Figure 2 illustrates the changes in MIP for both control and
experimental groups, over time. As shown in fig 2, improvements
in MIP at post-training were only observed for the experimental
Wk 8

group and these gains were maintained at follow-up. By week 8,

the experimental group was 27 cmH2O (95% CI, 15-40) stronger
329143 375141
1.41.5 0.61.1
(nZ19) (nZ19)
5214 9424
7824 12533
1918 4328

than the control group. By week 12, they were still 24 cmH2O
(95% CI, 11-36) stronger. Graphs demonstrating the individual
behavior of each participant of both control and experimental
groups can be obtained from the authors on request.
In MEP, by week 8, the experimental group was 42 cmH2O
(95% CI, 25-59) stronger than the control group. By week 12, they
Wk 0

were still 30 cmH2O (95% CI, 15-45) stronger.


By week 8, in inspiratory muscle endurance, the experimental


group took 33 breaths (95% CI, 20-47) more than the control
group. By week 12, they took 26 breaths (95% CI, 11-42) more.
Dyspnea (Medical Research
muscles (no. of breaths)

By week 8, in dyspnea, the experimental group was 1.3 out

Walking capacity 6-minute

of 4.0 (95% CI, 2.1 to 0.6) less breathless than the control
Endurance inspiratory

Council scale: 0-4)

group. By week 12, they were 1.2 (0-4) (95% CI, 2.2 to 0.1)
less breathless.
walk test (m)

In walking capacity, there were no significant between-group

MEP (cmH2O)
MIP (cmH2O)

differences either at week 8 (MD 38 m; 95% CI, 22 to 98) or

Table 2

week 12 (MD 44 m, 95% CI, 13 to 100). There was 1 case of

respiratory complication per group over the 8 weeks, giving a risk
difference of 0% (95% CI, e20 to 20).

6 K.K. Parreiras de Menezes et al

Discussion the respiratory muscles for a longer time, which might result in
positive benefits regarding exercise tolerance and everyday
This trial provides evidence that high-intensity home-based res- activities.38
piratory muscle training increases strength and endurance of the The improvements in strength and endurance are important,
respiratory muscles and reduces dyspnea. The benefits were because their impairments have been associated with dyspnea and/
largely maintained 1 month after the cessation of the training. or nocturnal alveolar hypoventilation.39,40 The training resulted in
However, there was no significant effect on respiratory compli- a decrease in dyspnea, with a mean between-group difference of
cations or walking capacity. 1.3 points on the Medical Research Council scale. This scale is
Respiratory training is gaining interest as a target for rehabil- widely used and changes 1 point indicate perceived clinical
itation.33 The present results demonstrated that respiratory training improvements.41 These findings are in line with previous studies,
increased the strength of the inspiratory muscles in the experi- which also found significant improvements in dyspnea, as deter-
mental group by 62% (up to 94 cmH2O) and that of the expiratory mined by the Borg scale, after respiratory training.37,42
muscles by 68% (up to 125 cmH2O). These large increases Although weakness of the respiratory muscles and dyspnea are
brought the final respiratory muscle strength within the realm of commonly associated with limitations in performing daily activ-
normal, thereby making them definitely clinically worthwhile, ities, the benefits were not carried over to walking capacity. There
because the smallest detectable differences for inspiratory and may be several explanations for this finding. The 6-minute walk
expiratory pressures range from 18% to 22%.16 test requires that participants walk as far as possible in a specified
Importantly, the between-group differences of 27 cmH2O for time. After a stroke, the performance on this test may reflect the
inspiratory strength and 42 cmH2O for expiratory strength are severity of motor impairments, such as muscle weakness and loss
higher than previously found as a result of respiratory of balance and motor coordination, more than breathlessness. On
training.8,9,34-37 The most recent systematic reviews reported the other hand, it may be also due to lack of power. In the present
changes in inspiratory strength of 7 cmH2O and in expiratory study, sample size calculation was based on the primary outcome,
strength of 6 and 13 cmH2O.7,10 The higher effects found in this that is, MIP, and was not sufficient to detect a clinically worth-
trial may be due to several factors. First, the dose of delivered while effect in walking capacity. However, the magnitude of the
training was about twice as higher, than that applied in previous difference observed after training (38m) and follow-up (44m)
trials. Second, training resistance began at 50% of the baseline suggests that there might be a worthwhile effect on walking ca-
MIP, opposite to 35% in previous trials. This was possible because pacity, which should be examined in a larger trial.
the device allowed both inspiratory and expiratory training at In this trial, there were very few respiratory complications that
higher loads (up to 70 cmH2O).17 Last, training targeted both resulted in hospital admissions. This low occurrence may be
inspiratory and expiratory muscles. explained by the characteristics of the participants, who were
This trial also found large improvements in endurance of the community-dwelling and at the sub-acute and chronic post-stroke
inspiratory muscles, with the experimental group increasing their stages. It is expected that higher occurrence of respiratory com-
endurance by 300% (33 breaths). This is in line with 1 other trial plications would happen at the acute stages.43
that investigated the effects of inspiratory training on inspiratory It is important to highlight that although significant improve-
endurance post-stroke8 and also reported positive results (MD 15; ments in respiratory strength, endurance, and dyspnea were
95% CI, 2-27). This means that the individuals were able to recruit observed, the carryover effects were not observed for walking

Fig 2 Mean  SD MIP values (cmH2O) for the experimental and control groups at baseline, post-training, and 1-month follow-up.

Respiratory muscle training improves strength and dyspnea 7

capacity and respiratory complications, which are also important c. POWERbreathe KH1; POWERbreathe International Ltd.
clinical outcomes. Thus, more studies evaluating the effects of d. SPSS software, version 17.0; SPSS, Inc.
high-intensity respiratory muscle training on these outcomes are
recommended, before drawing definitive conclusions regarding
clinically worthwhile effects.
The main strength of the current study is that it is a randomized Keywords
trial that was prospectively registered and followed the Consort
guidelines. It included concealed allocation, an intention-to-treat Breathing exercises; Cerebrovascular disease; Clinical trial;
analysis, blinded participants and assessors, and was powered to Dyspnea; Exercise; Muscle strength; Rehabilitation; Stroke
detect between-group differences in the primary outcome. In addi-
tion, it included a successful sham intervention. Not only did the
control participants not realize they were in the control group, but Corresponding author
there was little or no change in their outcomes. These small changes
may be attributed to the intervention, because, even without load, the Luci Fuscaldi Teixeira-Salmela, PT, PhD, Department of
participants spent 40 minutes a day on controlled inspiratory and Physiotherapy, Universidade Federal de Minas Gerais, Avenida
expiratory exercises, similar to breathing exercises, which may have Antônio Carlos, 6627 Campus Pampulha, 31270-910 Belo
generated some gains for these patients. Finally, both groups un- Horizonte, Minas Gerais, Brazil. E-mail address: lfts@ufmg.br.
derwent the same procedures, which reduced bias related to famil-
iarization with training effect of repeated assessments and had the
same amount of attention from the study personnel, which reduced References
social desirability and performance biases.
1. Mozaffarian D, Benjamin EJ, Go AS, et al. Heart disease and stroke
statisticse2015 update: a report from the American Heart Association.
Study limitations Circulation 2015;131:e29-322.
It is difficult or impossible to blind therapists during the delivery 2. Norrving B, Kissela B. The global burden of stroke and need for a
continuum of care. Neurology 2013;80:S5-12.
of complex interventions.44 Furthermore, due the non-compliance
3. Pollock RD, Rafferty GF, Moxham J, Kalra L. Respiratory muscle
with diary completion of many participants, we cannot be sure of
strength and training in stroke and neurology: a systematic review. Int
the actual dose of the undertaken training, although it was high J Stroke 2013;8:124-30.
enough to be effective, even though the training was home based, 4. McConnell A. Respiratory muscle training: theory and practice.
without direct supervision. In addition, some variance in de- Philadelphia: Churchill Livingstone; 2013.
mographic data, such as time since stroke, and baseline measures 5. Teixeira-Salmela LF, Parreira VF, Britto RR, et al. Respiratory pres-
of endurance and dyspnea, may have influenced the results, sures and thoracoabdominal motion in community-dwelling chronic
although they were not determinant. The loss of 6 participants for stroke survivors. Arch Phys Med Rehabil 2005;86:1974-8.
follow-up assessments also prevents an assertive conclusion 6. Pinheiro MB, Polese JC, Faria CD, et al. Inspiratory muscular
regarding the maintenance of the effects. Also, no adjustments for weakness is most evident in chronic stroke survivors with lower
walking speeds. Eur J Phys Rehabil Med 2014;50:301-7.
multiple comparisons were carried out and this should be taken
7. Menezes KK, Nascimento LR, Ada L, Polese JC, Avelino PR, Teixeira-
into consideration, especially when interpreting the results for the
Salmela LF. Respiratory muscle training increases strength of respira-
secondary outcomes. Finally, measures of spirometry, such as tory muscles and reduces the occurrence of respiratory complications
forced vital capacity, were not assessed to describe the sample, after stroke: a systematic review. J Physiother 2016;62:138-44.
which may limit the comparisons of the present results with other 8. Britto RR, Rezende NR, Marinho KC, Torres JL, Parreira VF, Teixeira-
populations. Salmela LF. Inspiratory muscular training in chronic stroke survivors: a
randomized controlled trial. Arch Phys Med Rehabil 2011;92:184-90.
9. Messaggi-Sartor M, Guillen-Sola A, Depolo M, et al. Inspiratory and
Conclusions expiratory muscle training in subacute stroke: a randomized clinical
trial. Neurology 2015;85:564-72.
The present findings have important implications for neurological 10. Gomes-Neto M, Saquetto MB, Silva CM, Carvalho VO, Ribeiro N,
rehabilitation. High-intensity home-based respiratory muscle Conceição CS. Effects of respiratory muscle training on respiratory
training, that is, with a load of 50% of the maximal respiratory function, respiratory muscle strength and exercise tolerance in post-
stroke patients: a systematic review with meta-analysis. Arch Phys
pressures, 40 min/day, over 8 weeks, is effective in increasing the
Med Rehabil 2016;97:1994-2001.
strength and endurance of the respiratory muscles and reducing
11. Ramı́rez-Sarmiento A, Orozco-Levi M, Guell R, et al. Inspiratory
dyspnea for people with respiratory muscle weakness post-stroke, muscle training in patients with chronic obstructive pulmonary disease
and the magnitude of the effects was higher than that previously - structural adaptation and physiologic outcomes. Am J Respir Crit
reported. Further trials aiming at comparing the effects of low- Care Med 2002;166:1491-7.
versus high-intensity training, powered to detect carryover effects 12. Hong A-R, Hong S-M, Shin Y-A. Effects of resistance training on
to walking capacity, and better comprehend the maintenance of muscle strength, endurance, and motor unit according to ciliary neu-
the benefits are warranted. rotrophic factor polymorphism in male college students. J Sports Sci
Med 2014;13:680-8.
13. Larson JL, Kim MJ, Sharp JT, Larson DA. Inspiratory muscle training
with a pressure threshold breathing device in patients with chronic
Suppliers obstructive pulmonary disease. Am Rev Respir Dis 1988;138:689-96.
14. Xiao Y, Luo M, Wang J, Luo H. Inspiratory muscle training for the
a. Orygen-dual valve; Forumed S.L. recovery of function after stroke. Cochrane Database Syst Rev 2012;
b. UFMG digital manovacuometer; UFMG. 16:CD009360.

8 K.K. Parreiras de Menezes et al

15. Maillard JO, Burdet L, van Melle G, Fitting JW. Reproducibility of 30. Newcombe RG. Interval estimation for the difference between inde-
twitch mouth pressure, sniff nasal inspiratory pressure, and maximal pendent proportions: comparison of eleven methods. Stat Med 1998;
inspiratory pressure. Eur Respir J 1998;11:901-5. 17:873-90.
16. Dimitriadis Z, Kapreli E, Konstantinidou I, Oldham J, Strimpakos N. 31. Nakagawa S, Cuthill IC. Effect size, confidence interval, and statistical
Test/retest reliability of maximum mouth pressure measurements with significance: a practical guide for biologists. Biol Rev Camb Philos
the MicroRPM in healthy volunteers. Respir Care 2011;56:776-82. Soc 2007;82:591-605.
17. Marco E, Ramı́rez-Sarmiento AL, Coloma A, et al. High-intensity vs. 32. Herbert R, Elkins M. Publishing code: an initiative to enhance trans-
sham inspiratory muscle training in patients with chronic heart failure: parency of data analyses reported in Journal of Physiotherapy. J
a prospective randomized trial. Eur J Heart Fail 2013;15:892-901. Physiother 2017;63:129-30.
18. Fulk GD, Reynolds C, Mondal S, Deutsch JE. Predicting home and 33. Schoser B, Fong E, Geberhiwot T, et al. Maximum inspiratory pres-
community walking activity in people with stroke. Arch Phys Med sure as a clinically meaningful trial endpoint for neuromuscular dis-
Rehabil 2010;91:1582-6. eases: a comprehensive review of the literature. Orphanet J Rare Dis
19. Schulz KF, Altman DG, Moher D; CONSORT Group. CONSORT 2017;12:52.
2010 statement: updated guidelines for reporting parallel group ran- 34. Chen PC, Liaw MY, Wang LY, et al. Inspiratory muscle training in
domized trials. J Clin Epidemiol 2010;63:834-40. stroke patients with congestive heart failure: a CONSORT-compliant
20. Farrero E, Antón A, Egea CJ, et al. Guidelines for the management of prospective randomized single-blinded controlled trial. Medicine
respiratory complications in patients with neuromuscular disease. (Baltimore) 2016;95:e4856.
Arch Bronconeumol 2013;49:306-13. 35. Guillén-Solà A, Messagi Sartor M, Bofill Soler N, Duarte E,
21. Pessoa IM, Pereira HL, Aguiar LT, Tagliaferri TL, Silva LA, Barrera MC, Marco E. Respiratory muscle strength training and
Parreira VF. Test-retest reliability and concurrent validity of a digital neuromuscular electrical stimulation in subacute dysphagic stroke
manovacuometer. Fisioter Pesqui 2014;21:236-42. patients: a randomized controlled trial. Clin Rehabil 2017;31:761-71.
22. Pessoa IM, Coelho CM, Mendes LP, Montemezzo D, Pereira DA, 36. Kulnik ST, Birring SS, Moxham J, Rafferty GF, Kalra L. Does res-
Parreira VF. Comparison of three protocols for measuring the maximal piratory muscle training improve cough flow in acute stroke? Pilot
respiratory pressures. Fisioter Mov 2015;28:31-9. randomized controlled trial. Stroke 2015;46:447-53.
23. American Thoracic Society/European Respiratory Society. ATS/ERS 37. Sutbeyaz ST, Koseoglu F, Inan L, Coskun O. Respiratory muscle
statement on respiratory muscle testing. Am J Respir Care Med 2002; training improves cardiopulmonary function and exercise tolerance in
166:518-624. subjects with subacute stroke: a randomized controlled trial. Clin
24. Charususin N, Gosselink R, Decramer M, et al. Inspiratory muscle Rehabil 2010;24:240-50.
training protocol for patients with chronic obstructive pulmonary 38. Syabbalo. Respiratory muscle function in patients with neuromuscular
disease (IMTCO study): a multicentre randomised controlled trial. disorders and cardiopulmonary diseases. Int J Clin Pract 1998;52:319-29.
BMJ Open 2013;3:e003101. 39. Sansone VA, Gagnon C. 207th ENMC Workshop on chronic respira-
25. Kovelis D, Segretti NO, Probst VS, Lareau SC, Brunetto AF, Pitta F. tory insufficiency in myotonic dystrophies: management and impli-
Validation of the Modified Pulmonary Functional Status and Dyspnea cations for research, 27-29 June 2014, Naarden, The Netherlands.
Questionnaire and the Medical Research Council scale for use in Neuromuscul Disord 2015;25:432-42.
Brazilian patients with chronic obstructive pulmonary disease. J Bras 40. Ambrosino N, Carpene N, Gherardi M. Chronic respiratory care for
Pneumol 2008;34:1008-18. neuromuscular diseases in adults. Eur Respir J 2009;34:444-51.
26. Stenton C. The MRC breathlessness scale. Occup Med 2008;58:226-7. 41. de Torres JP, Pinto-Plata V, Ingenito E, et al. Power of outcome
27. Holland AE, Spruit MA, Troosters T, et al. An official European measurements to detect clinically significant changes in pulmonary
Respiratory Society/American Thoracic Society technical standard: rehabilitation of patients with COPD. Chest 2002;121:1092-8.
field walking tests in chronic respiratory disease. Eur Respir J 2014; 42. Kim J, Park JH, Yim J. Effects of respiratory muscle and endurance training
44:1428-46. using an individualized training device on the pulmonary function and
28. Lambert TE, Harvey LA, Avdalis C, et al. An app with remote support exercise capacity in stroke patients. Med Sci Monit 2014;20:2543-9.
achieves better adherence to home exercise programs than paper 43. Martino R, Foley N, Bhogal S, Diamant N, Speechley M, Teasell R.
handouts in people with musculoskeletal conditions: a randomised Dysphagia after stroke: incidence, diagnosis, and pulmonary compli-
trial. J Physiother 2017;63:161-7. cations. Stroke 2005;36:2756-63.
29. Menezes KK, Nascimento LR, Polese JC, Ada L, Teixeira-Salmela LF. 44. Rodrigues-Baroni JM, Nascimento LR, Ada L, Teixeira-Salmela LF.
Effect of high-intensity home-based respiratory muscle training on Walking training associated with virtual reality-based training in-
strength of respiratory muscles after stroke: a protocol for a rando- creases walking speed of individuals with chronic stroke: systematic
mised controlled trial. Braz J Phys Ther 2017;21:372-7. review with meta-analysis. Braz J Phys Ther 2014;18:502-12.