Review

Respiratory muscle strength and training in stroke and

neurology: a systematic review
Ross D Pollock1*, Ged F Rafferty2, John Moxham2, and Lalit Kalra1

We undertook two systematic reviews to determine the

levels of respiratory muscle weakness and effects of respiratory muscle training in stroke patients. Two systematic
reviews were conducted in June 2011 using a number
of electronic databases. Review 1 compared respiratory
muscle strength in stroke and healthy controls. Review 2 was
expanded to include randomized controlled trials assessing
the effects of respiratory muscle training on stroke and other
neurological conditions. The primary outcomes of interest
were maximum inspiratory and expiratory mouth pressure
(maximum inspiratory pressure and maximum expiratory pressure, respectively). Meta-analysis of four studies
revealed that the maximum inspiratory pressure and
maximum expiratory pressure were significantly lower
(P < 000001) in stroke patients compared with healthy
individuals (weighted mean difference -4139 and
-5462 cmH2O, respectively). Nine randomized controlled
trials indicate a significantly (P = 00009) greater effect of
respiratory muscle training on maximum inspiratory pressure in neurological patients compared with control subjects
(weighted mean difference 694 cmH2O) while no effect on
maximum expiratory pressure. Respiratory muscle strength
appears to be impaired after stroke, possibly contributing to
increased incidence of chest infection. Respiratory muscle
training can improve inspiratory but not expiratory muscle
strength in neurological conditions, although the paucity of
studies in the area and considerable variability between
them is a limiting factor. Respiratory muscle training may
improve respiratory muscle function in neurological conditions, but its clinical benefit remains unknown.

Correspondence: Ross D. Pollock*, Kings College London, Denmark

Hill Campus, Academic Neuroscience Centre, PO41 Institute of
Psychiatry, London SE5 8AF, UK.
E-mail: ross.pollock@kcl.ac.uk
1
Department of Clinical Neurosciences, Kings College London,
London, UK
2
Division of Asthma, Allergy and Lung Biology, Kings College London,
London, UK
Conflict of interest: None declared.
DOI: 10.1111/j.1747-4949.2012.00811.x

2012 The Authors.

International Journal of Stroke 2012 World Stroke Organization

Key words: neurology, rehabilitation, respiratory, stroke,

weakness

Introduction
Stroke is one of the leading causes of morbidity and mortality
in the Western world (1) with up to 78 % of stroke patients
presenting with dysphagia (2,3). Dysphagia is associated with
a threefold increase in the risk of developing a chest infection,
which increases to 11-fold in those with definite aspiration
(2,4). Cough is an important mechanism to guard against
aspiration, which is often impaired in stroke patients (5) and
results in greater incidences of aspiration and chest infection
(6,7).
A strong cough is dependent on the ability to draw air into
the lungs, generate high pressures and air flow velocities, while
maintaining the patency of the airways (8), each of which are
influenced by respiratory muscle function. In stroke patients,
respiratory muscle weakness and altered chest wall kinematics
(9,10) may be responsible for impaired cough. Although
muscle weakness is often present in the acute stages of stroke,
this is likely due to impaired central drive to the muscles (11)
rather than reductions in the intrinsic strength of the muscle
(12). Whatever the mechanism, a means of improving respiratory muscle strength and/or central drive to the muscle may
be beneficial for stroke patients.
Inspiratory muscle training (IMT) or expiratory muscle
training (EMT) have been found to improve respiratory
muscle strength and function in multiple sclerosis (MS)
(13,14) and Parkinsons disease (PD) (15). These results
suggest that respiratory muscle training (RMT) can have a
beneficial effect on respiratory muscle function in neurological conditions. If similar results were to occur in stroke
patients, this could provide a potential treatment to improve
muscle function and cough and reduce the incidence of chest
infections.
Relatively little is known about the effects of stroke on respiratory muscle function or effective rehabilitation strategies
to improve muscle function. The aim of this review is twofold:
To perform a systematic review of studies in which respiratory muscle strength has been assessed in stroke patients

Vol , March 2012,

Review

To conduct a meta-analysis of studies investigating the

effects of RMT on individuals who have had a stroke.

Methods
Two literature reviews without any time restrictions were
performed in June 2011 using a number of electronic databases (PubMed, EMBASE, ISI Web of Science; review 1 also
included Scopus, while review 2 included the Cochrane
Central Register for Controlled Trials). The key words in
review 1 were stroke or cerebrovascular accident in combination with inspiratory, expiratory, or respiratory and strength
or weakness. Review 2 included randomized controlled
trials (RCTs) investigating the effect RMT on stroke patients;
however, initial searches revealed only two articles of this
nature. In light of this, review 2 was expanded to include other
neurological conditions. The key words used in review 2 were
stroke, cerebrovascular accident, multiple sclerosis, Parkinsons
disease, motor neurone disease, and neurology in combination
with inspiratory, expiratory, respiratory, or ventilatory and
training, loading, and muscle. The reference list of each article
identified from these searches were checked for any further
publications, while a forward search using the Science Citation
Index was also conducted.

Study selection criteria

R. D. Pollock et al.

The primary outcome of interest in both reviews was

maximal inspiratory and expiratory mouth pressure (PImax
and PEmax, respectively). For each outcome in review 2, data
were extracted from before the intervention and immediately
after the intervention for treatment and control groups. When
required, authors were contacted and original data were
sought. When these were not available, values were inputted
from the results in these papers using previously described
methods (18,19).

Results
Review 1
Four articles were included in the review (Fig. S1) (5,9,12,20).
PImax and PEmax were recorded in 57 stroke patients and
64 control subjects. Study characteristics and outcomes are
reported in Table 1. PImax was significantly lower in the
stroke patients than in the control subjects [weighted mean
difference (WMD) -4139 cmH2O, 95% confidence interval
(CI) -5374 to -2903, P < 000001] as was PEmax (WMD
-5462 cmH2O, 95% CI -6124 to -4781, P < 000001; Fig. 1).
The one study conducted in patients greater than threemonths after stroke onset (20) reported higher values of
PImax and PEmax suggesting improvement with time.

Review 2

Review 1
Only articles that assessed respiratory muscle strength in
stroke patients compared with healthy controls were included
in the review. No age limit for stroke patients was defined.

Nine articles were included in the review (Fig. S2 to be found

online); two assessed stroke patients (21,22), four MS
(13,14,23,24), one PD (15), one amyotrophic lateral sclerosis
(ALS) (25), and one myasthenia gravis (MG) (26).

Review 2
This review was restricted to RCTs conducted in subjects 18
years or older. Only studies investigating RMT (IMT or EMT)
were included. Articles in which subjects were not randomly
assigned to an intervention or control group were excluded.
Abstracts, letters to the editor, and commentaries were also
excluded due to insufficient reported details.
For both reviews, only original articles written in English
and published in peer-reviewed journals were included. After
the initial search, duplicates were removed with titles and
abstracts of the remaining articles assessed for eligibility. Any
uncertainty was discussed between authors until a consensus
was reached. The risk of bias of each study included in review
2 was determined using the Cochrane Collaborations tool
for assessing risk of bias (16). The quality of randomization,
blinding, and description of dropouts was assessed using the
scale reported by Jadad et al. (17). One point each is awarded
for randomization, double-blinding, and adequate description of withdrawals; one further point can be added for
randomization and blinding if the method used to do this is
described, while points are deducted if it is done inappropriately. A maximum score of 5 can be obtained.

Characteristics of studies
All of the studies were RCTs, which investigated RMT in neurological subjects. The duration of the interventions varied
between studies and ranged from six-weeks to 12 weeks
(Table 2). Six studies performed IMT, two EMT, and one
studied combined IMT with diaphragmatic breathing and
pursed lip breathing; the intensity and duration of training
sessions is listed in Table 2.
The median methodological score of all studies was 3
(range 1 to 5). Three trials (13,14,23) scored less than 3, indicating low methodological quality with the remainder being
good methodological quality. All trials were described as randomized; however, only two were classified as low risk of bias
(22,25) with the rest being unclear, except Fry et al. (13),
which was classified as high risk as randomization was based
on the date of recruitment. Three studies were double-blinded
(15,21,25). Where dropouts occurred, explanation was given
in all studies with outcome data always being reported.

Vol , March 2012,

Effect of intervention
All nine studies reported PImax in a total of 103 neurological
subjects and 100 healthy controls. Overall postintervention,

2012 The Authors.

International Journal of Stroke 2012 World Stroke Organization

Review

15:5
56 (16)
20
control

*Significantly greater than stroke group (P < 005).

Values obtained from author.

PEmax, maximum expiratory pressure; Pgas, gastric pressure; PImax, maximum inspiratory pressure; TwT10 Pgas, gastric pressure recorded during magnetic stimulation of the 10th thoracic nerve roots.

1087 (1659)*
951 (33)*

13416 (5676)*

525 (874)
389 (251)

9921 (2905)*

6 (3)
9:9
9:10

11:7
62 (15)

6021 (447)
19

18

control

stroke
Ward et al. (2010) (5)

8944 (4127)
>273
10:6
8:8
5837 (1547)
16
stroke

Teixeira-Salmela et al.
(2005) (20)

9
control

467 (121)

9:0

4:4
8:0
519 (102)
8
stroke
Lanini et al. (2003) (9)

8:8
16
control

758 (7)

7:8
689 (98)
15
stroke
Harraf et al. (2008) (12)

7362 (206)

1218 (181)*
994 (84)*

30 (12)

5343 (214)

616 (16)

Pgas recorded after TMS of injured side was significantly lower

than the uninjured side. No difference between stroke and
control in TwT10 Pgas
No asymmetry in chest wall kinematics of controls. During quiet
breathing, no difference between paretic and healthy side in
stroke patients, although paretic side movement decreased
during voluntary hyperventilation
Tidal volume, minute ventilation, and respiratory rate are similar
between the groups. Greater involvement in rib cage but lower
abdomen movement in the stroke patients during breathing
TwT10 Pgas similar between the groups. Voluntary and reflex
cough peak flow rates impaired in stroke. Pgas was decreased
during voluntary but not reflex cough in stroke subjects
626 (28)

1027 (30)*
758 (195)*

367 (282)
8 (5)

Days after
onset
Stroke
side (R : L)
Gender
(m : f)
Age (years)
n
Group
Study

Table 1 Characteristics and outcomes of studies assessing respiratory muscle strength

PImax
(cmH2O)

PEmax
(cmH2O)

Other

R. D. Pollock et al.

2012 The Authors.

International Journal of Stroke 2012 World Stroke Organization

PImax was significantly greater in the neurological patients

than in the control subjects (WMD 694 cmH2O, 95 % CI 284
to 1104, P = 00009; Fig. 2). The stroke subjects showed a
significant improvement compared with controls with RMT
(WMD 693 cmH20, 95% CI 176 to 1209, P = 0009), whereas
no difference was found between MS and the control subjects.
As only one study investigated PD, ALS, and MG patients each,
they could not be assessed individually. Two studies performed EMT (14,24), which showed no difference in PImax
between groups (P = 093), while the remaining studies, all
using IMT, showed a significantly greater increase in neurological subjects (WMD 839 cmH2O, 95% CI 390 to 1287,
P = 00002).
In total, seven trials reported PEmax in 84 neurological
and 81 control subjects. Overall, there was no difference in
PEmax values recorded in neurological and control subjects
(P = 045). In the MS patients, there was no effect of RMT on
PEmax (P = 091), while lack of studies did not allow this to be
tested in other neurological conditions (Fig. 3). Analysis of the
two studies where EMT was performed showed no difference
between groups in PEmax (P = 040). The same results for
studies performing IMT (P = 022) were recorded.

Adverse events
In one study, it was reported that one subject suffered
from light headedness at the start of the intervention (13).
Two further studies explicitly stated that no adverse events
occurred (23,25). It is unknown if this repeated in the remaining studies.

Discussion
Pooled data from existing studies suggest that respiratory
muscle strength is decreased after stroke by 4139 and
5462 cmH2O for PImax and PEmax, respectively. There is
some evidence that RMT may improve inspiratory but not
expiratory muscle strength in stroke survivors, with PImax, on
average, increasing by 693 cmH2O. Limitations of current
studies include the paucity of studies in the area, small samples
sizes and heterogeneity in patient selection, study design,
interventions, and outcome measurement. Although evidence
suggests that respiratory muscle weakness may contribute to
increased chest infections (4,7), no studies have assessed
whether RMT is clinically meaningful or makes a difference to
clinical outcomes.
Inspiratory and expiratory muscle weakness has a serious
impact on cough function. Inspiratory muscle weakness leads
to a reduced lung volume at the beginning of a cough and
expiratory muscle weakness leads to reduced intrathoracic
pressure needed to produce adequate airflow (27). Inspiratory
and expiratory muscle strength after acute stroke is approximately half of that recorded in healthy age-matched controls,
and studies show PImax values of considerably less than
80 cmH2O, the threshold for clinically meaningful weakness

Vol , March 2012,

Vol , March 2012,

MG

De Freitas Fregonezi
et al. (2005) (26)

IMT + DB + PLB (14)

Control (13)

IMT (9)
Placebo control (10)

IMT (10)
Placebo control (10)

IMT (20)
Control (21)

IMT (7)
Control (8)

EMT (9)
Control (9)

EMT (10)
Placebo control (5)

IMT (9)
Placebo control (9)

IMT (15)
Control (15)

Groups (number
of subjects)

30 mins per day, six-days a week for 12

weeks. Intensity increased from 15 to
60 % PImax in first month (reassessed
PImax each month and altered
accordingly)
10 mins, three times daily for 12 weeks.
Increased from 15 to 60 % maxSNIP
(assessed each month and adjusted
intensity)
Thrice weekly for eight-weeks. 10 mins
DB, 10 mins PLB, 10 mins IMT. IMT
intensity was increased from 20% to
60% baseline PImax each week

Three sets of 15 breaths daily for 10

weeks. Started at 30% PImax (increased
weekly depending on RPE)

Three sets of 10 breaths twice every other

day for 10 weeks. Intensity increased
from 40 to 60 % PImax (assessed every
two-weeks and increased based on RPE)

Three sets of 15 breaths twice daily for 12

weeks. Trained at highest percentage of
PEmax that could be maintained
(reassessed each week)
Three sets of 15 breaths twice daily for 12
weeks; 60% PEmax (unknown if altered
throughout study)

Six sets of five-minutes five times per

week for eight-weeks; 30% PImax
(reassessed every two-weeks)

15 mins twice daily for six-weeks. Intensity

increased weekly from 40% to 60 % of
baseline PImax

Training Characteristics

PImax, FVC, VC, TLC all increased after the

intervention, but this was no statistically greater than
the control group. PEmax, SF36 was unchanged
throughout the study in either group
PImax and PImax increased by 27% and 12%,
respectively, in the intervention group but not
control. No change in spirometry variables was
noted. Respiratory pattern and chest wall movement
improved with training but not control.

PImax improved 9 cmH2O after the intervention but

was not statistically different from the control.
PEmax was unchanged after the intervention. PI
improved significantly after the intervention and
compared to the control
PImax and PEmax improved from 42 to 67 and 46 to
63 cmH2O in the intervention group while remaining
unchanged in the control group, the improvement in
PEmax failed to reach significance compared to
control. FVC, FEV1, VC, PEFR were unchanged after
the intervention
PImax increased 80% in the intervention group, which
was significantly greater than the control group. No
change in PEmax or MVV occurred after the
intervention. FVC, FEV1, VC, and FEF 25 to 75%
increased after the intervention but not control.
PImax increased from 62 to 78 cmH2O, IME increased
20 to 29 cmH2O in IMT but not control group. No
change in FVC, FEV1. Dyspnea rating decreased in
IMT but not control group

PImax increased by 79 cmH2Oafter IMT which was

greater than the control group. PEmax increased
after the intervention but was no different to the
control group. FVC, FEV1, FEF 25% to 75%, MVV,
VO2 peak all increased after the intervention
compared to the control. No change in PEFR after
the intervention
PImax and IME significantly increased from 68 to 102
and 32 to 49 cmH2O, respectively, after the
intervention with no change in the control group.
Functional performance and quality of life were
unchanged after the intervention
PEmax increased by 19 cmH2O in the intervention but
not control group. No change in PImax occurred in
either group

Results

Control group was given one session

on DB and PLB at start of study and
encouraged to use techniques
without supervision, whereas the
intervention group attended the
hospital each week

Placebo control

Placebo control

Control group was sedentary

Control group was sedentary

Control group was instructed to

perform breathing exercises

Control group used an IMT device

set at very low threshold

Placebo control

Both groups underwent normal

stroke rehabilitation

Cointervention

Score

FEF 2575%, forced expiratory flow rate 2575%; FEV1, Force expired volume in 1 s; FVC, Force Vital Capacity; IME, inspiratory muscular endurance; MVV, maximum voluntary ventilation; PEFR, peak expiratory flow rate; PEmax, maximum expiratory
pressure; PI, pulmonary index; PImax, maximum inspiratory pressure; RPE, rate of perceived exertion; VO2 peak, peak oxygen consumption; IMT, Inspiratory muscle training; EMT, expiratory muscle training; DB, diaphragmatic breathing; PLB, pursed lip
breathing; SF, short form.

ALS

MS

Klefbeck & Nedjad

(2003) (23)

Cheah et al. (2009)

(25)

MS

Gosselink et al.
(2000) (24)

PD

MS

Smeltzer et al. (1996)

(14)

Inzelberg et al.
(2005) (15)

Stroke

Britto et al. (2011)

(21)

MS

Stroke

Sutbeyaz et al.
(2010) (22)

Fry et al. (2007) (13)

Condition

Study

Table 2 Characteristics and outcomes of the studies included in review 2

Review
R. D. Pollock et al.

2012 The Authors.

International Journal of Stroke 2012 World Stroke Organization

Review

R. D. Pollock et al.

Fig. 1 Comparison of PImax and PEmax in stroke patients and control subjects.

Fig. 2 Comparison or PImax values for neurological and control subjects after RMT.

(28). This weakness is not due to intrinsic loss of peripheral

muscle function but due to reduced cortical motor output
to these muscles (5,12) and may underlie significant
impairments of cough seen in acute stroke patients (5,6).
Furthermore, the only longitudinal study reported that a
PImax > 80 cmH2O is still not achieved at nine-months
poststroke (20).

2012 The Authors.

International Journal of Stroke 2012 World Stroke Organization

The findings of this review need to be interpreted in the

context of the limitations of studies included. A major limitation, common to all studies, is the method used to assess
respiratory muscle strength. PImax and PEmax are volitional
tasks, which require maximum effort from the subject. Hence,
it is likely to be influenced by subjects motivation and alertness as well as their ability to make an airtight seal around the

Vol , March 2012,

Review

R. D. Pollock et al.

Fig. 3 Comparison of PEmax values for neurological and control subjects after RMT.

mouthpiece of the measuring device. Many studies did not

control for practice effects in using measurement devices or
difference in contact time between intervention and control
groups.
RMT improves respiratory muscle strength, but training
appears to be task-specific and improvements in PImax
(13,15,2123,25,26) or PEmax (14,24) were seen depending
whether training was aimed toward inspiratory or expiratory
muscle function, respectively. The duration of training ranged
from six-weeks to 12 weeks, the number of sessions from three
daily to one session performed three-days per week, and the
duration of training sessions from five-minutes to 30 mins
(Table 2). Only four studies monitored compliance with training schedules (13,21,23,25). Training tended to be performed
between 30% and 60% of maximum inspiratory and expiratory pressure; however, how this was altered or progressed in
each study varied markedly (Table 2). A number of studies
increased training loads based on the baseline measurements
of PImax/PEmax, and two studies altered training intensity
based on the rate of perceived exertion during training
(13,23). Only four studies had control groups that followed
the same training protocol but at lower pressure thresholds
(14,15,21,25). In the remaining studies, the control group was
sedentary (13,22,23) or instructed to perform nonspecific
breathing exercises (24,26).
Although various measures of functional status, fatigue
severity, and quality of life were used, measurement of these

Vol , March 2012,

was variable and did not allow meta-analysis to be performed.

Aspiration risk or cough function was not assessed in any
intervention study. None of the studies reported adverse
events associated with RMT despite the theoretical risks of
elevated thoracic pressures or repeated Valsalvas maneuvers
in patients with stroke and cardiovascular comorbidities.

Conclusion
Good evidence exists that respiratory muscle strength is
significantly impaired after stroke because of decreased
corticorespiratory outflow from the damaged cortex. This has
also been shown to result in a weak cough, with decreased
ability to clear airways and increased risk of chest infections.
Extrapolation of findings from studies in other neurological
diseases suggests that RMT may improve respiratory function
in stroke patients, but this remains to be proven and its clinical
benefits remain unknown. The review identified several methodological challenges, which need to be met when designing
intervention studies to assess the effectiveness of RMT in
stroke patients.

Acknowledgements
This paper presents independent research funded by the
National Institute for Health Research (NIHR) under its
Research for Patient Benefit (RfPB) Programme (Grant

2012 The Authors.

International Journal of Stroke 2012 World Stroke Organization

Review

R. D. Pollock et al.

Reference Number PB-PG-0408-16096). The views expressed

are those of the author(s) and not necessarily those of the
NHS, the NIHR or the Department of Health.

References
1 Wolfe CD, Rudd T. The Burden of Stroke White Paper. London, Stroke
Alliance for Europe (SAFE), 2007.
2 Martino R, Foley N, Bhogal S, Diamant N, Speechley M, Teasell R.
Dysphagia after stroke: incidence, diagnosis, and pulmonary complications. Stroke 2005; 36:275663.
3 Mann G, Hankey GJ, Cameron D. Swallowing disorders following
acute stroke: prevalence and diagnostic accuracy. Cerebrovasc Dis
2000; 10:3806.
4 Smith Hammond CA, Goldstein LB, Horner RD et al. Predicting
aspiration in patients with ischemic stroke: comparison of clinical
signs and aerodynamic measures of voluntary cough. Chest 2009;
135:76977.
5 Ward K, Seymour J, Steier J et al. Acute ischaemic hemispheric stroke
is associated with impairment of reflex in addition to voluntary cough.
Eur Respir J 2010; 36:138390.
6 Smith Hammond CA, Goldstein LB, Zajac DJ, Gray L, Davenport
PW, Bolser DC. Assessment of aspiration risk in stroke patients with
quantification of voluntary cough. Neurology 2001; 56:5026.
7 Addington WR, Stephens RE, Gilliland KA. Assessing the laryngeal
cough reflex and the risk of developing pneumonia after stroke: an
interhospital comparison. Stroke 1999; 30:12037.
8 Lasserson D, Mills K, Arunachalam R, Polkey M, Moxham J, Kalra L.
Differences in motor activation of voluntary and reflex cough in
humans. Thorax 2006; 61:699705.
9 Lanini B, Bianchi R, Romagnoli I et al. Chest wall kinematics
in patients with hemiplegia. Am J Respir Crit Care Med 2003; 168:
10913.
10 Cohen E, Mier A, Heywood P, Murphy K, Boultbee J, Guz A.
Diaphragmatic movement in hemiplegic patients measured by
ultrasonography. Thorax 1994; 49:8905.
11 Newham DJ, Hsiao S-F. Knee muscle isometric strength, voluntary
activation and antagonist co-contraction in the first six months after
stroke. Disabil Rehabil 2001; 23:37986.
12 Harraf F, Ward K, Man W et al. Transcranial magnetic stimulation
study of expiratory muscle weakness in acute ischemic stroke. Neurology 2008; 71:20007.
13 Fry DK, Pfalzer LA, Chokshi AR, Wagner MT, Jackson ES. Randomized control trial of effects of a 10-week inspiratory muscle training
program on measures of pulmonary function in persons with multiple
sclerosis. J Neurol Phys Ther 2007; 31:16272.
14 Smeltzer SC, Levietes MH, Cook SD. Expiratory training in multiple
sclerosis. Arch Phys Med Rehabil 1996; 77:90912.
15 Inzelberg R, Peleg N, Nisipeanu P, Magadle R, Carasso RL, Weiner P.
Inspiratory muscle training and the perception of dyspnea in Parkinsons disease. Can J Neurol Sci 2005; 32:2137.
16 Higgins J, Altman D, Sterne J. Chapter 8: assessing risk of bias in
included studies; in Higgins J, Green S (eds): Cochrane Handbook for
Systematic Reviews of Interventions Version 510 (Updated March
2011). The Cochrane Collaboration, 2011. Available at http://www.
cochrane-handbook.org

2012 The Authors.

International Journal of Stroke 2012 World Stroke Organization

17 Jadad AR, Moore RA, Carroll D et al. Assessing the quality of reports
of randomized clinical trials: is blinding necessary? Control Clin Trials
1996; 17:112.
18 Hozo S, Djulbegovic B, Hozo I. Estimating the mean and variance
from the median, range, and the size of a sample. BMC Med Res
Methodol 2005; 5:13.
19 Higgins J, Deeks J, Altman D. Chapter 16: special topics in statistics;
in Higgins J, Green S (eds): Cochrane Handbook for Systematic
Reviews of Interventions Version 510 (updated March 2011). The
Cochrane Collaboration, 2011. Available at http://www.cochranehandbook.org
20 Teixeira-Salmela LF, Parreira VF, Britto RR et al. Respiratory pressures
and thoracoabdominal motion in community-dwelling chronic stroke
survivors. Arch Phys Med Rehabil 2005; 86:19748.
21 Britto RR, Rezende NR, Marinho KC, Torres JL, Parreira VF, TeixeiraSalmela LF. Inspiratory muscular training in chronic stroke survivors:
a randomized controlled trial. Arch Phys Med Rehabil 2011; 92:18490.
22 Sutbeyaz ST, Koseoglu F, Inan L, Coskun O. Respiratory muscle training improves cardiopulmonary function and exercise tolerance in subjects with subacute stroke: a randomized controlled trial. Clin Rehabil
2010; 24:24050.
23 Klefbeck B, Hamrah Nedjad J. Effect of inspiratory muscle training
in patients with multiple sclerosis. Arch Phys Med Rehabil 2003;
84:9949.
24 Gosselink R, Kovacs L, Ketelaer P, Carton H, Decramer M. Respiratory
muscle weakness and respiratory muscle training in severely disabled
multiple sclerosis patients. Arch Phys Med Rehabil 2000; 81:74751.
25 Cheah BC, Boland RA, Brodaty NE et al. Inspirational inspiratory
muscle training in amyotrophic lateral sclerosis. Amyotroph Lateral
Scler 2009; 10:38492.
26 De Freitas Fregonezi GA, Resqueti VR, Guell R, Pradas J, Casan P.
Effects of 8-week, interval-based inspiratory muscle training and
breathing retraining in patients with generalized myasthenia gravis.
Chest 2005; 128:152430.
27 McCool FD. Global physiology and pathophysiology of cough. Chest
2006; 129(1 Suppl.):48S53S.
28 American Thoracic Society/European Respiratory Society. ATS/ERS
statement on respiratory muscle testing. Am J Respir Crit Care Med
2002; 166:518624.

Supporting Information
Additional Supporting Information may be found in the
online version of this article:
Fig. S1 Flow diagram of the review process with reasons for
article exclusion for review 1.
Fig. S2 Flow diagram of the selection process for articles in
review 2.
Please note: Wiley-Blackwell are not responsible for the
content or functionality of any supporting materials supplied
by the authors. Any queries (other than missing material)
should be directed to the corresponding author for the article.

Vol , March 2012,