652446

research-article2016
CRE0010.1177/0269215516652446Clinical RehabilitationGuillén-Solà et al.

CLINICAL
Original Article REHABILITATION

Clinical Rehabilitation

Respiratory muscle strength 1­–11

© The Author(s) 2016
Reprints and permissions:
training and neuromuscular sagepub.co.uk/journalsPermissions.nav
DOI: 10.1177/0269215516652446

electrical stimulation in subacute cre.sagepub.com

dysphagic stroke patients:

A randomized controlled trial

Anna Guillén-Solà1,2,3, Monique Messagi Sartor1,

Neus Bofill Soler1, Esther Duarte1, Mª Camelia Barrera1
and Ester Marco1,2,3

Abstract
Objective: To evaluate the effectiveness of inspiratory/expiratory muscle training (IEMT) and
neuromuscular electrical stimulation (NMES) to improve dysphagia in stroke.
Design: Prospective, single-blind, randomized-controlled trial.
Setting: Tertiary public hospital.
Subjects: Sixty-two patients with dysphagia were randomly assigned to standard swallow therapy (SST)
(Group I, controls, n=21), SST+ IEMT (Group II, n=21) or SST+ sham IEMT+ NMES (Group III, n=20).
Interventions: All patients followed a 3-week standard multidisciplinary rehabilitation program of SST and
speech therapy. The SST+IEMT group’s muscle training consisted of 5 sets/10 repetitions, twice-daily, 5 days/
week. Group III’s sham IEMT required no effort; NMES consisted of 40-minute sessions, 5 days/week, at 80Hz.
Main outcomes: Dysphagia severity, assessed by Penetration-Aspiration Scale, and respiratory muscle
strength (maximal inspiratory and expiratory pressures) at the end of intervention and 3-month follow-up.
Results: Maximal respiratory pressures were most improved in Group II: treatment effect was 12.9
(95% confidence interval 4.5-21.2) and 19.3 (95% confidence interval 8.5-30.3) for maximal inspiratory
and expiratory pressures, respectively. Swallowing security signs were improved in Groups II and III at
the end of intervention. No differences in Penetration-Aspiration Scale or respiratory complications were
detected between the 3 groups at 3-month follow-up.
Conclusion: Adding IEMT to SST was an effective, feasible, and safe approach that improved respiratory
muscle strength. Both IEMT and NMES were associated with improvement in pharyngeal swallowing
security signs at the end of the intervention, but the effect did not persist at 3-month follow-up and no
differences in respiratory complications were detected between treatment groups and controls.

1Physical Medicine and Rehabilitation Department, Parc Corresponding author:

de Salut Mar (Hospital del Mar/Hospital de l’Esperança), Anna Guillén-Solà, Physical Medicine and Rehabilitation
Barcelona, Catalonia, Spain Department, Hospital de l’Esperança, C/Sant Josep de la
2Rehabilitation Research Group, Institut Hospital del Mar Muntanya 12, 08024 Barcelona, Catalonia, Spain.
d’Investigacions Mèdiques (IMIM), Barcelona, Catalonia, Spain Email: [email protected]
3Department of Medicine, Universitat Autònoma de

Barcelona, Catalonia, Spain

Downloaded from cre.sagepub.com at DALHOUSIE UNIV on June 19, 2016

2 Clinical Rehabilitation 

Keywords
Neuromuscular electrical stimulation, respiratory muscle training, oropharyngeal dysphagia, stroke,
rehabilitation

Received: 6 October 2015; accepted: 6 May 2016

Introduction
New therapeutic modalities to improve the man- Based on these considerations, a randomized
agement of swallowing in neurologic disorders controlled trial was designed to assess the thera-
have appeared in recent years. Neuromuscular peutic effectiveness of neuromuscular electrical
electrical stimulation aims to improve the strength stimulation and of inspiratory and expiratory mus-
of muscle groups that were disabled by stroke but cle training in dysphagic subacute stroke patients,
preserved motor innervation. The available studies compared to standard swallow therapy. A second
observed contradictory results, with some authors objective was to evaluate their potential influence
reporting that sensory and motor stimulation of on the occurrence of respiratory complications at
peripheral nerves can accelerate swallowing 3-month follow-up.
recovery1-5 while others concluded that, depend-
ing on electrodes positioning, surface electrical
Methods
stimulation could increase the risk of bronchoaspi-
ration by reducing hyolaryngeal elevation during The study had a randomized, single-blind design
swallowing therapy.6,7 with 3 parallel groups, following the Consolidated
Respiratory muscle training is another therapeu- Standards of Reporting Trials guidelines.15 The set-
tic strategy to be considered in patients with dys- ting was a tertiary subacute care hospital in
phagia. As impaired cough function in stroke is Barcelona (Catalonia, Spain) between December
related to respiratory muscle weakness,8 an inter- 2013 and June 2015.
vention aimed to strengthen respiratory muscles Inclusion criteria were subacute ischemic stroke
might improve cough effectiveness and reduce within 1 to 3 weeks of inclusion and dysphagia
aspiration risk. Two randomized clinical trials dem- confirmed by videofluoroscopic study with a score
onstrated significant improvement in inspiratory ⩾3 in the 8-point Penetration Aspiration Scale.16
muscle strength and other physiologic parameters Patients with cognitive impairment (Short Portable
after inspiratory muscle training.9,10 Some studies Mental Status Questionnaire <3)17 and/or history
have suggested that expiratory muscle strength of previous neurological diseases that might be
training can improve respiratory function in patients associated with the presence of dysphagia were
with Parkinson disease, and also improve swallow- excluded.
ing function and avoid chest infections.11 Two Sample size was overestimated to allow for a
recent studies showed contradictory results on both potential loss of 25% and the calculation was based
respiratory muscle and cough functions in stroke on data from a previous inspiratory muscle training
patients after respiratory muscle training.12,13 trial in patients with chronic heart failure.18 Thus, a
The relationship between swallowing and minimum of 13 patients were necessary in each
breathing is based on the physiology of these com- group, accepting a mean difference between treat-
plementary functions, which must be closely coor- ments of 14.7 cmH2O and SD of 16 on predicted
dinated because they cannot be performed inspiratory pressures with an α-risk of 0.05 and
simultaneously. After stroke, muscular dysfunction β-risk of 20% in a 2-tailed test.
and lack of the needed coordination secondary to Eligible patients were randomly assigned to
central nervous lesion leads to dysphagia in the one of three treatment arms: Group I, standard
acute phase of stroke onset.14 swallow therapy; Group II, inspiratory and

Downloaded from cre.sagepub.com at DALHOUSIE UNIV on June 19, 2016

Guillén-Solà et al. 3

Figure 1. CONSORT 2010 Flow diagram.

SST, Standard Swallow Therapy; IEMT, Inspiratory and Expiratory Muscle Training; NMES, Neuromuscular Electrical Stimulation.

expiratory muscle training with the use of a res- Randomization was performed independently
piratory muscles trainer in addition to standard by a collaborator blinded to patient identity, using a
swallow therapy; and Group III, neuromuscular computer-generated randomization list. The
electrical stimulation of suprahyoid muscles, researcher responsible for assessment was not
sham inspiratory and expiratory muscle training, aware of the study group allocation; all clinical
and standard swallow therapy (Figure 1). assessments were carried out by the same

Downloaded from cre.sagepub.com at DALHOUSIE UNIV on June 19, 2016

4 Clinical Rehabilitation 

rehabilitation specialist, who also was blinded to Care Fusion, Kent, UK). To determine respiratory
study group assignments. pressures, patients were urged to perform a maxi-
All patients followed a multidisciplinary inpa- mum inspiratory effort from residual volume
tient rehabilitation program consisting of physi- against an occluded airway and a maximum expira-
cal, occupational and speech therapy targeting tion from total lung capacity. The highest value of
specific impairments in mobility, activities of three reproducible manoeuvres (<10% variability
daily living, swallowing and communication between values) was used for analysis and com-
skills (3 hours per day, 5 days a week, during 3 pared with reference values previously published
weeks). All three groups received standard swal- for a Mediterranean population;19 respiratory pres-
low therapy, which consisted of an educational sures >80% were considered normal. Dysphagia
intervention aimed to improve self-management severity was assessed with the Penetration-
of dysphagia and protect the airway, oral exer- Aspiration Scale: scores of 1-2 indicate normal
cises to improve lingual praxis, and compensatory swallowing; 3-5, penetration; >6, aspiration. For
techniques based on videofluoroscopic findings. easier interpretation of the results, Penetration-
These swallowing manoeuvres, oral exercises, Aspiration Scale scores <5 were considered as
and compensatory techniques were individualized non-aspiration and ⩾5 as aspiration. Respiratory
according to intrinsic patient characteristics. complications were assessed by the presence of
Additionally, Group II received respiratory lung infections shown on chest x-ray or by fever
training sessions, which consisted of 5 sets of 10 with abnormal respiratory signs, according to
respirations followed by 1 minute of unloaded information obtained from medical reports and/or
recovery breathing off the device (Orygen Dual telephone interview at 3-month follow-up.
Valve®, Forumed SL, Barcelona, Catalonia, Other swallowing parameters collected included
Spain),18 twice a day, 5 days per week for 3 weeks, signs of security (changes in tone of voice, cough-
with the assistance of a therapist. Training loads ing and/or desaturation greater than 3% compared
were set at a pressure equivalent to 30% of maxi- to baseline pulse oximetry during or after eating)
mal inspiratory and expiratory pressures and and efficacy (piecemeal deglutition and oropharyn-
increased weekly at intervals of 10 cmH2O. geal residue) assessed with the Volume Viscosity
In addition to standard swallow therapy, Group Swallow Test;20 the Functional Oral Intake Scale, a
III received sham respiratory muscle training, with 7-point functional scale which documents the level
the workloads fixed at 10 cmH2O throughout the of oral feeding and liquids, ranging from 1 (npo =
3-week-intervention period, and neuromuscular nil per mouth) to 7 (normal diet);21 and the
electrical stimulation using the Intelect VitalStim Dysphagia Outcome Severity Scale, a 7-point scale
device (VitalStim®, Chattanooga Group, Hixson, based on independence level and type of nutrition,
TN, USA). Under supervision by a speech-lan- determined by videofluoroscopic examination that
guage therapist, two electrodes were placed on ranges from 1 (tube-dependent diet due to massive
suprahyoid muscles in 40-minute daily sessions (5 aspirations at any viscosity tested) to 7 (normal diet
days per week for 3 weeks) and 80 Hz of transcuta- with no aspirations detected at any viscosity).22
neous electrical stimulus was applied, according to Respiratory muscle function and swallowing
VitalStim® instructions; patients were instructed assessments were performed at baseline and at 3
to swallow when they felt muscle contraction. weeks and 3 months post-intervention, except the
Main outcome measures were related to respira- videofluoroscopy, which was performed at baseline
tory muscle function, severity of dysphagia, and at 3-month follow-up. Data on age, sex, smok-
and occurrence of respiratory complications. ing history, and comorbidities were collected, as
Respiratory muscle strength, defined as the abil- well as stroke characterization: aetiology according
ity to achieve a brief maximal effort, was assessed to the ischemic stroke classification, severity accord-
with maximal inspiratory and expiratory pressures ing to the National Institutes of Health Stroke Scale,
at the mouth using a MicroRPM (Micro Medical/ location according to clinically identifiable subtypes

Downloaded from cre.sagepub.com at DALHOUSIE UNIV on June 19, 2016

Guillén-Solà et al. 5

of cerebral infarction, and acute disability assessed (Group II) or neuromuscular electrical stimulation
with the modified Rankin scale and dependence in (Group III), compared to standard swallow therapy
daily living activities with the Barthel Index. (Group I). All groups had improved respiratory
The clinical trial was approved by the local muscle pressures at both follow-up assessments. In
Clinical Research Ethics Committee and performed a comparison of the intervention groups with
in accordance with the Declaration of Helsinki. standard therapy, a positive treatment effect was
Signed informed consent was obtained from all detected for inspiratory and expiratory muscle
patients. The trial was registered in ClinicalTrials. training at 3-week follow-up.
gov (NCT 02473432). Patient distribution according to disorders in
Quantitative variables were descriptively the security and efficacy signs of deglutition is
expressed as mean and SD, unless otherwise described in Table 3. After the 3-week interven-
stated. Univariate analysis was performed using tion, patients in both the neuromuscular electrical
Chi-square, Fisher exact, Student t, or Mann- stimulation and the inspiratory and expiratory
Whitney U tests, depending on the variables ana- muscle training groups significantly improved
lysed. Treatment effect was analysed by changes security signs, but this beneficial effect was lost at
in maximal respiratory pressures pre- and post- 3-month follow-up. At 3 months, efficacy signs
intervention. Changes at 3-month follow-up were improved for inspiratory and expiratory muscle
assessed by analysis of variance using mixed training (p=0.037), and a marginally significant
repeated measures and a one-factor design for the improvement was observed for neuromuscular
analysis of values over time. Effect size was stimulation (p= 0.079).
reported using the Cohen d index. The level of sig- Table 4 shows contingency tables for video-
nificance was set at p ⩽0.05. Data analysis was fluoroscopic results dichotomized as aspirators
performed using IBM SPSS Statistics v.21. (Penetration-Aspiration Scale ⩾5) and non-aspira-
tors (Penetration-Aspiration Scale <5) at admission
to the rehabilitation unit and at 3-month follow-up.
Results Although the number of non-aspirators improved
Figure 1 describes the flow of participants through- at 3 months, no statistically significant differences
out the study period; the main outcome measures were observed between groups.
were analysed by intention-to-treat at 3-week and The Functional Oral Outcome Scale improved
3-month follow-up. Twenty-one patients were not 0.76 (SD 1.1) points after completing the interven-
able to perform the respiratory and/or swallowing tion, and 1.76 (SD 1.1) points at 3-month follow-up.
assessment after the 3-week intervention. Eleven Mean improvement on the Dysphagia Outcome
of these patients were lost to 3-month follow-up Severity Scale at 3 months was 0.96 (SD 1.4). No
and no clinical information was available from significant differences in the Functional Oral
their medical records for analysis. Outcome Scale and the Dysphagia Outcome Severity
Table 1 describes the demographic characteris- Scale were observed between the study groups.
tics of participants. At recruitment, patients had a During the intervention period, only 2 (3.2%)
mean age of 69 (SD 8.7) years; gender distribution patients, both in the standard swallow therapy (con-
was 38 men and 24 women. The majority of strokes trol) group, presented with lung infections. At
were in the anterior circulation (61.3%) and of 3-month follow-up, there were 13 patients with
moderate severity according to the National medical complications: 2 new strokes (1 fatality), 1
Institutes of Health Stroke Scale. Time elapsed hip fracture, 1 seizure, and 9 respiratory events.
from stroke onset to admission to the rehabilitation Therefore, lung infections constituted 15.5% of the
unit was 10.4 (SD 6.5) days. dysphagic sample, which was distributed across the
Table 2 describes changes in maximal respira- 3 study groups as follows: 4 in standard swallow
tory muscle pressures at 3 weeks and 3 months therapy, 3 in neuromuscular electrical stimulation
after inspiratory and expiratory muscle training and 2 in inspiratory and expiratory muscle training.

Downloaded from cre.sagepub.com at DALHOUSIE UNIV on June 19, 2016

6 Clinical Rehabilitation 

Table 1. Baseline demographic and clinical characteristics of the sample.

Total Sample SST IEMT NMES p

(n= 62) (n= 21) (n= 20) (n= 21)
Age (years) 69.0 (8.7) 68.9 (7.0) 67.9 (10.6) 70.3 (8.4) >0.05
Sex, n (%):
- Male 38 (61.3%) 12 (57.1%) 16 (76.2%) 10 (47.6%) >0.05
- Female 24 (38.7%) 9 (42.9%) 4 (23.8%) 11 (52.4%)
Body Mass Index (Kg/m2) 26.5 (4.3) 26.1 (4.1) 25.7 (2.9) 27.6 (5.4) >0.05
Stroke hemisphere, n (%):
- Right 26 (41.9%) 8 (38.1%) 8 (40%) 10 (47.6%) >0.05
- Left 32 (51.6%) 12 (57.1%) 11 (55%) 9 (42.8%) >0.05
- Bilateral/ataxia 4 (6.5%) 1 (4.8%) 1 (5%) 2 (9.6%) >0.05
Stroke aetiology, n (%)
- Atherosclerosis 17 (27.4%) 6 (28.6%) 6 (30%) 5 (23.8%) >0.05
- Cardioembolism 19 (30.7%) 7 (33.3%) 7 (35%) 5 (23.8%) >0.05
- Lacunar 3 (4.8%) 2 (9.5%) 0 (0%) 1 (4.8%) >0.05
- Other determined aetiology 1 (1.6) 0 (0%) 0 (0%) 1 (4.8%) >0.05
- Undetermined aetiology 21 (33.9%) 5 (23.8%) 7 (35%) 9 (42.9%) >0.05
- Missing data 1 (1.6%) 1 (4.8%) 0 (0%) 0 (0%) >0.05
Oxford Topographic Classification, n (%)
- TACI 16 (25.8%) 6 (28.6%) 5 (25%) 5 (23.8%) >0.05
- PACI 22 (35.5%) 4 (19.05%) 10 (50%) 8 (38.1%) >0.05
- LACI 10 (16.1%) 4 (19.05%) 2 (10%) 4 (19%) >0.05
- POCI 11 (17.8%) 5 (23.8%) 3 (15%) 3 (14.3%) >0.05
- Missing data 3 (4.8%) 2 (9.5%) 0 (0%) 1 (4.8%) >0.05
NIHSS score on admission 6.8 (3.8) 6.5 (3.8) 6.4 (3.4) 7.6 (4.3) >0.05
Modified Rankin Scale 3.7 (0.7) 3.7 (SD 0.8) 3.9 (0.5) 3.6 (0.8) >0.05
Barthel Index on admission at 42.8 (15.0) 44.0 (18.5) 42.7 (14.6) 41.8 (12.2) >0.05
Rehabilitation
Stroke onset (days) 10.4 (6.5) 9.3 (5.1) 10.8 (8.7) 11.0 (5.5) >0.05
Stay in rehabilitation unit (days) 18.0 (6.4) 16.9 (5.9) 16.8 (6.9) 18.6 (2.3) >0.05
Respiratory muscle strength
- PImax (cmH20) 36.1 (21.6) 32.5 (20.7) 40.6 (18.9) 35.4 (25.1) >0.05
- PImax (%) 37.0 (20.3) 32.4 (19.5) 40.5 (14.5) 38.4 (25.3) >0.05
- PEmax (cmH20) 53.7 (28.9) 52.4 (26.0) 60 (25.5) 49.1 (34.5) >0.05
- PEmax (%) 34.5 (16.2) 33.2 (14.0) 37.5 (14.7) 32.8 (19.5) >0.05
Swallowing assessment
- FOIS 4.4 (0.7) 4.3 (0.6) 4.5 (0.5) 4.4 (1.0) >0.05
- PAS 5.3 (2.2) 5.4 (2.3) 5 (2.7) 5.5 (2.2) >0.05
- DOSS 4.0 (1.1) 4.0 (1.46) 4.2 (0.9) 4.9 (0.2) >0.05
- V-VST efficacy signs altered, n (%) 50 (90.6%) 15 (71.4%) 16 (80%) 18 (85.7%) >0.05
- V-VST security signs altered, n (%) 49 (90.7%) 17 (81.0%) 16 (80%) 17 (81%) >0.05

Data are shown as mean (standard deviation) unless otherwise stated (%).
SST, Standard Swallow Therapy; IEMT, Inspiratory and Expiratory Muscle Training; NMES, Neuromuscular electrical stimulation;
TACI, Total Anterior Circulation Infarction; PACI, Partial Anterior Circulation Infarction; POCI, Posterior Circulation Infarction;
LACI, Lacunar Infarction; NIHSS, National Institutes of Health Stroke Scale; PImax, Maximal Inspiratory Pressure; PEmax, Maximal
Expiratory Pressure; FOIS, Functional Outcome Intake Scale; PAS, Penetration Aspiration Scale; DOSS, Dysphagia Outcome and
Swallow Scale; V-VST, Volume Viscosity Swallow Test.

Downloaded from cre.sagepub.com at DALHOUSIE UNIV on June 19, 2016

Guillén-Solà et al. 7

Table 2. Changes in maximal respiratory muscle pressures at 3 weeks and 3 months after inspiratory/expiratory
muscle training and neuromuscular electrical stimulation, compared with standard swallow therapy.
Standard Inspiratory and expiratory muscle training Neuromuscular electrical stimulation
swallow therapy

Mean change Mean change Mean difference p* Mean change Mean difference p*
(95% CI) (95% CI)

At 3 weeks:
- Δ PImax 8.2 (SD 7.2) 21.1 (SD 13.1) 12.9 (4.5 to 21.2) 0.015 9.6 (SD 6.4) 11.5 (2.7 to 20.2) >0.05
(cmH2O)
- Δ %PImax 8.0 (SD 7.5) 21.1 (SD 11.8) 13.1 (5.2 to 20.9) 0.014 10.4 (SD 6.8) 10.7 (2.5 to 18.8) >0.05
- Δ PEmax 7.1 (SD 8.6) 26.4 (SD 16.9) 19.3 (8.5 to 30.3) 0.044 13.5 (SD 12.9) 12.9 (0.4 to 25.4) >0.05
(cmH2O)
- Δ %PEmax 6.4 (SD 3.7) 16.8 (SD 11.0) 10.3 (3.7 to 17.0) 0.076 9.8 (SD 8.8) 7.5 (–0.8 to 15.8) >0.05
At 3 months:
- Δ PImax 9.4 (SD 12.5) 18.3 (SD 14.5) 12.4 (3.07 to 21.6) >0.05 13.8 (SD 15.5) 4.48 (–6.2 to 15.1) >0.05
(cmH2O)
- Δ %PImax 6.9 (SD 14.1) 17.0 (SD 15.8) 10.0 (0.25 to 20.3) >0.05 14.5 (SD 17.4) 2.55 (–9.3 to 14.4) >0.05
- Δ PEmax 18.1 (SD 19.2) 32.4 (SD 21.2) 14.3 (0.33 to 28.2) >0.05 24.2 (SD 22.5) 8.26 (–7.2 to 23.8) >0.05
(cmH2O)
- Δ %PEmax 12 (SD 12.8) 19.5 (SD 16.5) 7.11 (2.96 to 17.2) >0.05 15.3 (SD 14.0) 4.17 (–6.7 to 15.0) >0.05

*Significant difference, compared to ‘Standard swallow therapy’ reference group.

Quantitative variables are presented as mean (standard deviation).
CI, Confidence Interval; PImax, Maximal Inspiratory Pressure; PEmax, Maximal Expiratory Pressure.

Table 3. Disorders in swallowing security and efficacy signs at admission to rehabilitation, upon study completion,
and at three-month follow-up: comparison of neuromuscular electrical stimulation and respiratory muscle training
with standard swallow therapy.

Volume-Viscosity Standard Neuromuscular p* Inspiratory p*

Swallow Test swallow therapy electrical and expiratory
(reference) stimulation muscle training
Baseline:
-  security signs 17 17 0.757 16 0.5
-  efficacy signs 15 18 0.114 16 0.5
3 weeks:
-  security signs 16 10 0.049 9 0.011
-  efficacy signs 14 13 0.620 15 0.5
3 months:
-  security signs 6 2 0.112 3 0.219
-  efficacy signs 13 8 0.078 7 0.037

*Significant difference compared to standard swallow therapy reference group.

No adverse effects were reported during the reported hyperventilation-induced dizziness, with
trial. Patient acceptance was quite good. Only a no clinical impact.
few patients receiving neuromuscular electrical
stimulation reported any discomfort related to the
electrode position and motor stimulation threshold;
Discussion
none withdrew from the study. Three patients in the After the 3-week intervention period, both inter-
inspiratory and expiratory muscle training group vention groups showed improvement in clinical

Downloaded from cre.sagepub.com at DALHOUSIE UNIV on June 19, 2016

8 Clinical Rehabilitation 

Table 4. Contingency tables showing patients’ distribution according to Penetration-Aspiration Scale at admission
to rehabilitation and at three-month follow-up.

Penetration- Standard swallow Neuromuscular Inspiratory and expiratory p

Aspiration Scale* therapy (reference) electrical stimulation muscle training

<5 ⩾5 <5 ⩾5 <5 ⩾5

Baseline 9 12 8 13 10 10 0.74
3-month follow-up 15 5 11 6 13 4 0.7

*Penetration-Aspiration Scale <5, Nonaspirators; ⩾5, Aspirators.

signs of swallowing security, but only inspiratory Over the past decade, many authors have investi-
and expiratory muscle training was associated with gated the effects of respiratory muscle training on
improvements in respiratory muscle strength. More the management of dysphagia. The finding of
research will be required to determine the effects of increased motor unit recruitment of the submental
this training on the incidence of respiratory muscle complex in healthy volunteers after expira-
complications. tory muscle training generated discussion of a
A systematic review published in 2009 reported potential benefit of expiratory muscle training in
promising results on the use of surface neuromus- patients with dysphagia.27,28 Thereafter, other stud-
cular electrical stimulation as a motor and sensory ies in patients with Parkinson disease,11 multiple
facilitation modality.3 A meta-analysis aimed to sclerosis, 29 and myasthenia gravis30 also showed
evaluate its effects on swallowing rehabilitation improvement in swallowing function after expira-
revealed a small but significant effect size in the tory muscle training. To the authors’ knowledge,
treatment of swallowing disorders.23 Most of the only two randomized clinical trials have evaluated
studies included in these reviews had been carried the effect of inspiratory and expiratory muscle train-
out in chronic stroke samples; evidence regarding ing in subacute stroke. The first concluded that
subacute stroke is much more limited. The combi- improvements observed in respiratory muscle func-
nation of neuromuscular electrical stimulation and tion and cough flow are part of the natural history of
conventional therapy has been shown to improve stroke and are not influenced by respiratory muscle
swallowing function, as assessed with the training;12 the second found that inspiratory and
Functional Dysphagia Scale, but not on the expiratory muscle training induces a significant
Functional Oral Intake Scale and Penetration- improvement in respiratory muscle strength, which
Aspiration Scale. Another24,25 randomized trial could have contributed to the decrease in respiratory
compared the effect of adding thermotactile stimu- complications observed at 6-month follow-up.13
lation to neuromuscular electrical stimulation, and These studies were conducted in subacute stroke,
showed greater improvement in the Penetration- but none focused specifically on dysphagic patients.
Aspiration Scale of patients who underwent the The present study showed improvements in the
combined therapy. A clinical trial conducted in security signs assessed with Volume Viscosity
acute stroke patients randomly assigned to neuro- Swallow Test at the end of both interventions.
muscular electrical stimulation or usual dysphagia Nevertheless, the lack of differences between
therapy showed a mean gain of 1.4 points on the groups at 3-month follow-up might be explained
Functional Oral Intake Scale at 3 weeks and of 2.4 by the reversibility of the training effect and/or the
points at 6 weeks, but did not provide information natural evolution of dysphagia. The distribution of
on videofluoroscopic parameters.25 Finally, a non- patients with impaired efficacy signs was
randomized study reported improvements in diet unchanged at the end of the 3-week interventions,
and patient satisfaction, as well as a reduction in although an improvement was observed in inspira-
respiratory complications.26 tory and expiratory muscle training patients at 3

Downloaded from cre.sagepub.com at DALHOUSIE UNIV on June 19, 2016

Guillén-Solà et al. 9

months. We cannot offer a definitive explanation thyrohyoid position used in this study has been
for this finding, although the limited sample size evaluated in other studies with good results,
might have been a factor. although a depressive effect on the hyolaryngeal
The prevalence of lung infections in our study complex has been described and authors have re-
was surprisingly low, especially considering its commended a submental position.7,34
homogeneous sample of patients with dysphagia. In our opinion, it might be useful to search for
The continuous supervision of patients participating indirect indicators associated with the presence of
in clinical trials might have contributed to this low lung infections, such as delayed triggering pharyn-
prevalence. Dysphagia and aspiration are probably geal reflex and number of bronchoaspirations in
the most important factors predisposing to lung the videofluoroscopic assessment, or peak cough
infections, but are not the only ones.31 In our study, flow, among others. Further research should
only 9 patients had lung infections at 3-month fol- include larger samples of both subacute and chronic
low-up, and distribution was similar across the study patients in order to avoid confounding factors due
groups, but our sample estimation was not powered to natural stroke evolution during the first weeks
to predict lung infections. The sample size required after event onset.
to show a 5% decrease in post-stroke pneumonia has In conclusion, a 3-week intervention of inspira-
been estimated to exceed 22,000 patients,12 which tory and expiratory muscle training added to stand-
provides a good rationale for a larger study ade- ard swallow therapy improved respiratory muscle
quately powered to assess respiratory complications. strength and security in swallowing during the
There are some limitations to be considered. subacute stroke phase. Swallowing security was
This trial included stroke patients with dysphagia also improved by neuromuscular electrical stimu-
diagnosed with the videofluoroscopic approach lation after the 3-week intervention. The use of
and without previous history of disorders in swal- combined therapies in the management of dyspha-
lowing or respiratory muscle function. Apart from gia appeared to accelerate swallowing recovery.
the initial bias inherent to samples from rehabilita- The low prevalence of bronchoaspirative events
tion units, in which patients tend to be pre-selected prevented the evaluation of potential impact on the
for their potential to follow a rehabilitation pro- presence of respiratory complications.
gram, some potential limitations concerning the
assessment methods should be considered. The Clinical messages
determination of maximum inspiratory and expira- •• Inspiratory and expiratory muscle trai-
tory pressures requires volitional manoeuvres that ning in subacute stroke patients improved
are influenced by individuals’ cooperation as well respiratory muscle strength and swallow-
as their ability to make an airtight seal around the ing security, compared to standard swal-
manometer mouthpiece, which often requires the low therapy, after 3-week intervention;
assistance of a therapist. In order to minimize no difference in swallowing security was
unnecessary patient exposure to radiation accord- observed at 3-month follow-up.
ing to the ‘as-low-as-reasonably achievable’ prin- •• Neuromuscular electrical stimulation
ciple,32 the videofluoroscopic swallow study was improved swallowing security in sub-
not repeated at the end of the 3-week intervention. acute stroke patients, compared to stan-
Although previous studies have shown the reliabil- dard swallow therapy, after 3-week
ity of the volume-viscosity swallow test compared intervention; no difference was observed
with a videofluoroscopic swallow study,20,33 the at 3-month follow-up.
test can be useful in clinical practice but cannot •• Neuromuscular electrical stimulation and
replace videofluoroscopic swallow study. Another inspiratory and expiratory muscle train-
potential limitation is that electrodes positioning ing showed no effect on the incidence of
could affect the potential of neuromuscular electri- respiratory complications.
cal stimulation to achieve better results:6 the

Downloaded from cre.sagepub.com at DALHOUSIE UNIV on June 19, 2016

10 Clinical Rehabilitation 

Acknowledgements 7. Ludlow CL, Humbert I, Saxon K, Poletto C, Sonies B and

Crujido L. Effects of surface electrical stimulation both at
This work has been conducted within the framework of rest and during swallowing in chronic pharyngeal dyspha-
the Doctorate in Medicine from Universitat Autònoma gia. Dysphagia. 2007;22:1–10.
de Barcelona. The authors gratefully acknowledge 8. Hannawi Y, Hannawi B, Rao CPV, Suarez JI and Bershad
Elaine Lilly, PhD, for English language revision and EM. Stroke-associated pneumonia: Major advances and
suggestions, and Pere Ortiz (pharmacist) for randomiza- obstacles. Cerebrovasc Dis. 2013;35:430–443.
tion. The authors appreciate the contributions to the 9. Britto RR, Rezende NR, Marinho KC, Torres JL, Parreira
study made by Sandra Chiarella, Martha Alvarado, Núria VF and Teixeira-Salmela LF. Inspiratory muscular train-
Bas and Marina Depolo, who helped to test participants. ing in chronic stroke survivors: a randomized controlled
trial. Arch Phys Med Rehabil. 2011;92:184–190.
Clinical Trial Registration: ClinicalTrials.gov (NCT
10. Sutbeyaz ST, Koseoglu F, Inan L and Coskun O.
02473432).
Respiratory muscle training improves cardiopulmonary
function and exercise tolerance in subjects with suba-
Declaration of Conflicting Interests cute stroke: a randomized controlled trial. Clin Rehabil.
2010;24:240–250.
The author(s) declared no potential conflicts of interest 11. Troche MS, Okun MS, Pitts T, Rosenbek JC, Musson
with respect to the research, authorship, and/or publica- N, Fernandez HH, et al. Aspiration and swallowing in
tion of this article. Parkinson disease and rehabilitation with EMST: a rand-
omized trial. Neurology. 2010;75:1912–1919.
Funding 12. Kulnik ST, Birring SS, Moxham J, Rafferty GF and Kalra
L. Does respiratory muscle training improve cough flow
The author(s) disclosed receipt of the following financial in acute stroke? Pilot randomized controlled trial. Stroke.
support for the research, authorship, and/or publication 2015;46:447–453.
of this article: Partially supported by grants from: ISCIII 13. Messaggi-Sartor M, Guillen-Solà A, Depolo M, Duarte E,
(PI10/01560)/FEDER. Rodríguez DA, Barrera M-C, et al. Inspiratory and expira-
tory muscle training in subacute stroke: A randomized
clinical trial. Neurology. 2015;85:564–572.
References 14. Martín-Valero R, De La Casa Almeida M, Casuso-
1. Terré R and Mearin F. A randomized controlled study Holgado MJ and Heredia-Madrazo A. Systematic Review
of neuromuscular electrical stimulation in oropharyn- of Inspiratory Muscle Training After Cerebrovascular
geal dysphagia secondary to acquired brain injury. Eur J Accident. Respir Care. 2015;60:1652–1659.
Neurol. 2015;22:687–e44. 15. Schulz KF, Altman DG and Moher D. CONSORT 2010
2. Terré R, Martinell M, González B, Ejarque J and statement: Updated guidelines for reporting parallel group
Mearin F. Treatment of oropharyngeal dysphagia with randomized trials. Ann Intern Med. 2010;152:726–732.
neuromuscular electrostimulation. Med Clin (Barc). 16. Rosenbek JC, Robbins JA, Roecker EB, Coyle JL and
2013;140:157–160. Wood JL. A penetration-aspiration scale. Dysphagia.
3. Clark H, Lazarus C, Arvedson J, Schooling T and 1996;11:93–98.
Frymark T. Evidence-based systematic review: effects 17. Pfeiffer E. A short portable mental status questionnaire for
of neuromuscular electrical stimulation on swallowing the assessment of organic brain deficit in elderly patients.
and neural activation. Am J Speech-Language Pathol. J Am Geriatr Soc. 1975;23:433–441.
2009;18:361–375. 18. Marco E, Ramírez-Sarmiento AL, Coloma A, Sartor
4. Bülow M, Speyer R, Baijens L, Woisard V and Ekberg O. M, Comin-Colet J, Vila J, et al. High-intensity vs. sham
Neuromuscular electrical stimulation (NMES) in stroke inspiratory muscle training in patients with chronic heart
patients with oral and pharyngeal dysfunction. Dysphagia. failure: a prospective randomized trial. Eur J Heart Fail.
2008;23:302–309. 2013;15:892–901.
5. Rofes L, Arreola V, López I, Martin A, Sebastián M, 19. Morales P, Sanchis J, Cordero PJ and Díez JL. Maximum
Ciurana A, et al. Effect of surface sensory and motor static respiratory pressures in adults. The reference val-
electrical stimulation on chronic poststroke oro- ues for a Mediterranean Caucasian population. Arch
pharyngeal dysfunction. Neurogastroenterol Motil. Bronconeumol. 1997;33:213–219.
2013;25:888–e701. 20. Clavé P, Arreola V, Romea M, Medina L, Palomera E and
6. Humbert IA, Poletto CJ, Saxon KG, Kearney PR, Crujido Serra-Prat M. Accuracy of the volume-viscosity swallow
L, Wright-Harp W, et al. The effect of surface electrical test for clinical screening of oropharyngeal dysphagia and
stimulation on hyolaryngeal movement in normal indi- aspiration. Clin Nutr. 2008;27:806–815.
viduals at rest and during swallowing. J Appl Physiol. 21. Hansen TS, Engberg AW and Larsen K. Functional oral
2006;101:1657–1663. intake and time to reach unrestricted dieting for patients

Downloaded from cre.sagepub.com at DALHOUSIE UNIV on June 19, 2016

Guillén-Solà et al. 11

with traumatic brain injury. Arch Phys Med Rehabil. 28. Wheeler Hegland KM, Huber JE, Pitts T, Sapienza CM,
2008;89:1556–1562. Hegland KMW, Huber JE, et al. Lung volume during
22. O’Neil KH, Purdy M, Falk J, Gallo L, Neil KHO, Purdy swallowing: single bolus swallows in healthy young
M, et al. The dysphagia outcome and severity scale. adults. J Speech Lang Hear Res. 2009;52:178–187.
Dysphagia. 1999;14:139–145. 29. Chiara T, Martin D and Sapienza C. Expiratory muscle
23. Huang K-L, Liu T-Y, Huang Y-C, Leong C-P, Lin W-C and strength training: Speech production outcomes in patients
Pong Y-P. Functional outcome in acute stroke patients with with multiple sclerosis. Neurorehabil Neural Repair.
oropharyngeal Dysphagia after swallowing therapy. J Stroke 2007;21:239–249.
Cerebrovasc Dis. 2014;23:2547–2553. 30. Martín-Valero R, Zamora-Pascual N and Armenta-
24. Lim KB, Lee H-JJ, Lim S-SS and Choi Y-II. Peinado JA. Training of Respiratory Muscles in Patients
Neuromuscular electrical and thermal-tactile stimulation With Multiple Sclerosis: A Systematic Review. Respir
for dysphagia caused by stroke: A randomized controlled Care. 2014;59:1764–1772.
trial. J Rehabil Med. 2009;41:174–178. 31. Langmore SE, Terpenning MS, Schork A, Chen Y,
25. Lee KW, Kim SB, Lee JH, Lee SJ, Ri JW and Park JG. Murray JT, Lopatin D, et al. Predictors of aspiration
The effect of early neuromuscular electrical stimulation pneumonia: how important is dysphagia? Dysphagia.
therapy in acute/subacute ischemic stroke patients with 1998;13:69–81.
dysphagia. Ann Rehabil Med. 2014;38: 153–159. 32. Norris TG. Radiation safety in fluoroscopy. Radiol
26. Crary MA, Carnaby-Mann GD and Faunce A. Electrical Technol. 2002;74:132.
stimulation therapy for dysphagia: Descriptive results of 33. Rofes L, Arreola V and Clavé P. The volume-viscosity
two surveys. Dysphagia. 2007 Jul;22:165–173. swallow test for clinical screening of dysphagia and aspi-
27. Wheeler KM, Chiara T and Sapienza CM. Surface elec- ration. Nestle Nutr Inst Workshop Ser. 2012;72:33–42.
tromyographic activity of the submental muscles during 34. Kim SJ and Han TR. Effect of surface electrical stimula-
swallow and expiratory pressure threshold training tasks. tion of suprahyoid muscles on hyolaryngeal movement.
Dysphagia. 2007;22:108–116. Neuromodulation. 2009;12:134–140.

Downloaded from cre.sagepub.com at DALHOUSIE UNIV on June 19, 2016