See editorial on pages 6–7 in this issue.

IMPACT OF EXPIRATORY STRENGTH TRAINING IN AMYOTROPHIC

LATERAL SCLEROSIS: RESULTS OF A RANDOMIZED,
SHAM-CONTROLLED TRIAL
EMILY K. PLOWMAN, PHD ,1,2,3 LAUREN TABOR-GRAY, PHD,1 K. MICHELLE ROSADO, MS,1 TERRIE VASILOPOULOS, PHD,4
RAELE ROBISON, MS, JENNIFER L. CHAPIN, MS,1 JOY GAZIANO, MA,5 TUAN VU, MD,6 and CLIFTON GOOCH, MD6
1

1
Swallowing Systems Core, University of Florida, Gainesville, Florida, USA
2
Department of Speech, Language and Hearing Sciences, College of Public Health and Health Professions, University of Florida,
P.O. Box 117420, Gainesville, Florida, 32610, USA
3
Department of Neurology, University of Florida, Gainesville, Florida, USA
4
Department of Anesthesiology, University of Florida, Gainesville, Florida, USA
5
Joy McCann Culverhouse Center for Swallowing Disorders, University of South Florida, Tampa, Florida, USA
6
Department of Neurology, University of South Florida, Tampa, Florida, USA
Accepted 1 July 2018

ABSTRACT: Introduction: The purpose of this study was to deter- both dysphagia and tracheal aspiration in ALS,23,24
mine the impact of an in-home expiratory muscle strength training management strategies that improve or maintain
(EMST) program on pulmonary, swallow, and cough function in
individuals with amyotrophic lateral sclerosis (ALS). Methods: subglottic air-pressure generation, airway clearance
EMST was tested in a prospective, single-center, double-blind, ability, and peak cough flow may be beneficial.25
randomized, controlled trial of 48 ALS individuals who completed We recently reported improvements in maximum
8 weeks of either active EMST (n = 24) or sham EMST (n = 24).
The primary outcome to assess treatment efficacy was change in expiratory pressure (MEP) and swallowing kinemat-
maximum expiratory pressure (MEP). Secondary outcomes ics after a well-tolerated 5-week exercise program tar-
included: cough spirometry; swallowing; forced vital capacity; and geting expiratory force generation, cough, and
scoring on the ALS Functional Rating Scale—Revised. Results:
Treatment was well tolerated with 96% of patients completing the airway protection (expiratory muscle strength train-
protocol. Significant differences in group change scores were ing, EMST) in 15 individuals with ALS.20 This pre-
noted for MEP and Dynamic Imaging Grade of Swallowing Toxic- sent double-blind, sham-controlled, randomized
ity scores (P < 0.02). No differences were noted for other second-
ary measures. Discussion: This respiratory training program was clinical trial extends these preliminary findings, test-
well-tolerated and led to improvements in respiratory and bulbar ing our hypothesis that 8 weeks of daily, moderate-
function in ALS. intensity resistance EMST at home, applied early in
Muscle Nerve 59:40–46, 2019
the course of ALS, will actively engage the expiratory
and submental musculature, deter disuse atrophy,
Impaired swallowing, cough, and respiratory func-
and prolong more effective cough function, airway
tion are common in amyotrophic lateral sclerosis
protection, swallowing, and functional oral intake.
(ALS), accounting for 91.4% of mortality via malnu-
trition, aspiration, pneumonia, and neuromuscular
METHODS
respiratory failure.1–5 Although the use of global Participants. Fifty-two individuals with a diagnosis of proba-
exercise programs in ALS is still being explored, ble or definite ALS according to the revised El-Escorial cri-
recent preliminary evidence suggests that mild-to- teria26 were screened for potential enrollment in this study at
moderate intensity exercise, applied early in the dis- the University of South Florida’s ALS Center. The diagnosis
ease, may have beneficial effects in both animal was confirmed in all patients by academic neuromuscular neu-
models6–10 and humans.11–22 As impaired airway rologists specializing in ALS before screening for enrollment.
Inclusion criteria were: (1) confirmed ALS diagnosis;
defense mechanisms (dysfunctional expiratory phase
(2) forced vital capacity > 65% predicted; (3) Amyotrophic
of voluntary cough production) are associated with Lateral Sclerosis Functional Rating Scale—Revised (ALSFRS-
R) score > 30; (4) adequate cognition to follow simple com-
mands as evidenced by a score of >24 points on the Mini-
Abbreviations: ALS, amyotrophic lateral sclerosis; ALSFRS-R, Amyo- Mental Status Examination27; (5) no allergies to barium;
trophic Lateral Sclerosis Functional Rating Scale—Revised; DIGEST, (6) no tracheostomy or mechanical ventilation; and (7) no
Dynamic Imaging Grade of Swallowing Toxicity; EAT-10, Eating Assess- diaphragmatic pacer. This study was approved by the univer-
ment Tool-10; EMST, expiratory muscle strength training; FOIS, Functional
Oral Intake Scale; FVC, forced vital capacity; MEP, maximum expiratory sity institutional review board and conducted in accordance
pressure; PAS, Penetration–Aspiration Scale; VFSS, videofluoroscopic with the Declaration of Helsinki. All included patients pro-
swallowing study vided informed written consent.
Key words: amyotrophic lateral sclerosis, cough, EMST, exercise, expira-
tory muscle strength training, rehabilitation, swallowing
Funding: This study was funded by a grant from the National Institutes of Design. As noted, this study was a prospective, double-blind,
Child Health and Development (1R21 HDO75327).
randomized, sham-controlled clinical trial. Once enrolled, par-
Conflicts of Interest: None of the authors have any conflict of interest to ticipants were assigned a study number and randomized to
disclose. either the experimental group (EMST) or control (sham
Correspondence to: E.K. Plowman; e-mail: [email protected]fl.edu EMST) group. Baseline assessments of respiratory, swallow,
and cough physiologic domains were performed, and partici-
© 2018 Wiley Periodicals, Inc.
Published online 7 July 2018 in Wiley Online Library (wileyonlinelibrary. pants immediately commenced the assigned 8-week treatment
com). DOI 10.1002/mus.26292 regimen. After treatment, participants were again assessed

40 Expiratory Training in ALS MUSCLE & NERVE January 2019

using the entire assessment battery. Figure 1 illustrates this the first training session of each week, a research therapist vis-
timeline. ited the patient in their home to reassess MEP using a hand-
held digital manometer (MP01 Micro Respiratory Pressure
Randomization and Blinding. Participants were assigned Meter; Micro Direct, Inc., Lewiston, Maine) and recalibrate
to the EMST or control (sham EMST) group using a per- the EMST trainer to represent 50% of the current MEP value.
muted block randomization schedule. Participants and the After this adjustment, each patient performed an EMST ther-
clinical researchers performing and interpreting evaluations apy session with the therapist. The remaining 4 daily therapy
were blinded to group assignment. Cough spirometry and sessions were performed during the week at home with the
videofluoroscopic data were de-identified and coded to blind assistance of the participant’s caregiver. Patients therefore
participant identification and evaluation testing time-point. completed 125 targeted exhalations each week and a total of
The home therapist was responsible for introducing and per- 1,000 exercise repetitions during the 8-week program. To
forming treatments with the appropriate respiratory trainers encourage compliance and accountability, participants and
and was, therefore, knowledgeable on group assignment their caregivers were provided with a home therapy log and
throughout the trial. asked to track daily therapy sessions by marking off each exer-
cise set performed.
EMST Training Protocol. EMST was completed using a
handheld, one-way, spring-loaded valve trainer. EMST was per- Sham Training Protocol. Patients assigned to the sham
formed with the trainer set at 50% of the patient’s individual EMST group completed an 8-week training protocol with a
maximum expiratory pressure (MEP), providing a moderate trainer that looked identical to the high-physiological-load
training load on the expiratory muscles. To accommodate trainer (EMST 150), but had the internal spring removed.
individualized training loads with thresholds that ranged Therefore, these patients performed exercises against no phys-
between 14 and 107 cm H2O, 2 devices were utilized. Specifi- iological load. The training protocol was in every other respect
cally, a lower resistance threshold trainer with a resistance identical to the EMST group.
load between 5 and 20 cm H2O was used by participants
whose MEPs were <40 cm H2O (Philips Threshold PEP
Evaluation Procedures and Outcomes. Participants com-
Trainer; Philips Respironics, Cedar Grove, New Jersey). For
pleted 2 comprehensive evaluations of pulmonary, cough, and
participants whose MEPs were >40 cm H2O, a higher thresh-
swallowing function at baseline and immediately after the
old trainer with a range of 20–150 cm H2O was utilized
assigned 8-week training program. Both patients and evalua-
(EMST 150; Aspire Products, Gainesville, Florida). Given that
tors were blinded regarding sham vs. active treatment.
labial weakness is typical in ALS patients, a flanged rubber
mouthpiece was attached to the opening of the expiratory
trainer to facilitate labial seal during exercises. EMST therapy Primary Outcome Measure: MEP. MEP was the primary
sessions were completed at home on 5 days of the patients outcome measure. It was assessed using the MP01 handheld
choosing per week. During training, the participants sat in a digital manometer. To minimize labial leakage in patients who
comfortable position and wore a nose-clip to prevent air were unable to create a tight lip seal due to facial weakness, a
escape due to nasopharyngeal weakness. They took a deep flanged rubber mouthpiece was attached to the manometer.
breath, held their cheeks lightly (or had a caregiver do so), During MEP testing, the participant was seated with a nose-
and blew forcefully into the training device. A single daily clip in place to occlude the nasal cavity. After inhaling to total
training session consisted of 25 targeted forced exhalations lung capacity, participants were instructed to place their lips
through the trainer, performed in 5 sets of 5 repetitions. This around the mouthpiece and blow out as forcefully as possible.
was described to participants as the “Rule of Fives” and has Three trials were collected, and the highest obtained MEP was
been used previously in other patient populations.28 Partici- used in the analysis.
pants were instructed to rest between each 5-breath set, and a
typical training session lasted approximately 20 minutes. On Swallowing Function. Three measures of swallowing func-
tion were obtained to assess: (1) global swallowing function;
(2) oral intake; and (3) patient report.

Global Swallow Function. To examine physiological

swallowing function, the “gold standard” videofluoroscopic
swallowing study (VFSS) was performed. Participants were
seated upright in a lateral viewing plane using a properly colli-
mated Philips BV Endura fluoroscopic C-arm unit (GE OEC
8800 Digital Mobile C-Arm System, Type 718074). A Swallow-
ing Signals Lab Unit (Pentax, Lincoln Park, New Jersey) digi-
tally recorded the fluoroscopic images at 29.97 frames per
second using a scan converter. A standardized bolus presenta-
tion protocol was used and consisted of two 1-ml boluses of liq-
uid contrast agent (barium), one 3-ml bolus of thin liquid
contrast, one 3-ml bolus of paste, one 20-ml bolus of liquid
contrast, one 90-ml bolus of thin liquid contrast in the lateral
view, and a 20-ml bolus of liquid contrast in the anterior/pos-
terior view. To ensure patient safety, the VFSS was terminated
immediately after a second aspiration event on any of the
FIGURE 1. Study design flowchart. EMST, expiratory muscle strength 7 swallow trials. Images were recorded digitally onto the swal-
training. [Color figure can be viewed at wileyonlinelibrary.com] low workstation (Pentax) for subsequent analysis.

Expiratory Training in ALS MUSCLE & NERVE January 2019 41

 The validated Dynamic Imaging Grade of Swallowing Toxic- after tidal volume breathing to the end of inspiration before
ity (DIGEST) was utilized to index global swallowing function.29 the compression phase of cough); (2) inspiratory peak flow
The DIGEST includes a swallowing efficiency subscore (0–4), (peak inspiratory flow during the inspiratory phase preceding
based on amount and frequency of pharyngeal residue, and a the cough); (3) compression phase duration (time from the
swallowing safety subscore (0–4), based on amount, depth, and end of the inspiratory phase to the beginning of the expiratory
frequency of penetration/aspiration. The total DIGEST score phase); (4) expiratory rise time (time from the beginning of
reflects the interaction between efficiency and safety subscores expiratory phase to the peak expiratory flow); (5) expiratory
to yield a global grade of pharyngeal dysphagia (0 = normal peak airflow (peak airflow during the expiratory phase of the
swallowing function, 1 = mild dysphagia, 2 = moderate dyspha- cough); and (6) cough volume acceleration (expiratory peak
gia, 3 = severe dysphagia, 4 = life-threatening dysphagia).29 For flow divided by expiratory rise time). Cough spirometry mea-
statistical analyses, DIGEST scores ≥1 were classified as dyspha- sures were assessed on the first cough epoch across the 3 trials
gic and those <1 as non-dysphagic. (i.e., the first cough attempt for each of the 3 trials) by a
Airway safety was further evaluated using the validated Pen- blinded rater.
etration–Aspiration Scale (PAS), an 8-point ordinal scale of
airway safety that describes the degree of airway invasion, the Forced Vital Capacity. Forced vital capacity (FVC) was
participant’s response, and whether the invasive material is assessed using a handheld spirometer (Micro I; Carefusion,
successfully ejected from the airway.30 Using previously Yorba Linda, California) in accordance with the American
defined thresholds for safe vs. unsafe swallowing, individuals Thoracic Society guidelines. Percent predicted values for FVC
whose PAS score was ≥3 were considered “unsafe” and those were recorded and used for statistical analysis.
whose PAS was ≤2 were considered “safe” swallowers for data
analysis purposes. All objective ratings were performed in a ALSFRS-R. The ALSFRS-R34 was administered before and
blinded fashion on all swallows and the worst PAS score was after treatment to index global disease progression. A total
utilized for statistical analysis. A primary rater analyzed all score was calculated as well as a respiratory subscale (out of
VFSE data with a secondary rater scoring 25% of collected 12) and bulbar subscale (out of 12) for comparative analysis.
data for the purposes of reliability. If discrepancies were noted
for either the PAS or DIGEST scores, a consensus meeting was Statistical Analysis. Baseline differences between groups
held between raters with the option of a third expert rater were assessed by t-test (continuous outcomes) and chi-square
available if consensus in scoring could not be met. test (categorical outcomes). To assess differences in pre to post
change over time between groups in the primary outcome of
Daily Oral Intake. The Functional Oral Intake Scale MEP, linear regression models were used. These models
(FOIS)31 was used as an index of daily oral intake. This vali- included posttest scores as the outcome, and group status and
dated 7-point ordinal scale measures what foods an individual pretest scores as the predictors. Using pretest scores as a predic-
consumes to meet their daily nutritional and hydration tor creates a “residual change score” relative to the outcome. If
requirements and ranges from a 1 (nothing by mouth) to the effect of group status is statistically significant (P < 0.05),
7 (full oral diet with no restrictions). then it can be interpreted that the groups differed in their
change from pre to post. The F-statistic (F), with numerator
Patient Report. Finally, the Eating Assessment Tool-10 degrees and denominator degrees of freedom = 1 and 45, was
(EAT-10)32 was administered as a patient-reported outcome used to test the statistical significance of between-group differ-
(PRO) of swallowing function. The EAT-10 is a validated ences. In addition, the pretest score as a predictor controls for
10-item patient-rated questionnaire with each question rated potential baseline performance differences was included. These
on a 5-point ordinal scale. A total EAT-10 score ranges from analyses also controlled for other baseline covariates, including
0 (indicative of no self-perceived swallowing impairments) to age, gender, disease-onset type, disease duration, and ALSFRS-
40 (indicative of severe swallowing impairments). We have R. This approach was also used for secondary outcomes that
previously established that a cut-point of 8 on the EAT-10 dis- were continuously measured. For ordinal outcomes (DIGEST,
tinguishes between ALS patients with dysphagia vs. those with- PAS, FOIS, and EAT-10), previously determined clinical rele-
out dysphagia.33 Therefore, in the current study, this vant cutoffs were used to dichotomize these variables:
threshold was utilized for statistical analysis. DIGEST > 1; PAS ≥3; FOIS ≤ 6; and EAT-10 > 8.29,31,33 For
these outcomes, logistic regression analyses were conducted in
a similar manner to that described previously. Between-group
Voluntary Cough Spirometry. Cough was assessed using an differences were tested with likelihood-ratio chi-square test, with
oral pneumotachograph (MLT 1000; ADInstruments, Inc., Col- 1 degree of freedom; P < 0.05 considered statistically signifi-
orado Springs, Colorado), connected to a spirometry mouth- cant. Both raw and corrected P-values are reported for second-
piece filter (MQ 304 Spirometer Filter; Vacumed; Ventura, ary outcomes. P-values were corrected for multiple comparisons
California) during voluntary cough production. Each partici- using the false discovery rate approach.35 All analyses were con-
pant was seated with a mouthpiece filter held in place by the ducted by the study biostatistician using JMP Pro version 13.0
examiner, nose-clips in place, and instructed to “cough hard (SAS Institute, Inc., Cary, North Carolina).
like something is stuck in your throat.” The patient was pro-
vided with an example by the examiner who ensured the RESULTS
patient understood the requirements of this task, and 3 trials Forty-eight individuals met the inclusion criteria
were obtained. Airflow signal was measured, low-pass filtered at
and were enrolled in this study. Demographic details
60 HZ, digitized at 1,000 HZ, and displayed on a portable laptop
computer using LabChart version 7 (Microsoft Corp., Red-
for this cohort are summarized in Table 1. Over the
mond, Washington). Physiological voluntary cough airflow mea- 2-month study period, 2 patients withdrew from the
sures deduced from the cough flow waveforms included: study (1 from each group). Reasons for withdrawal
(1) inspiratory phase duration (time from onset of inspiration were: (1) diagnosis of cancer requiring immediate
42 Expiratory Training in ALS MUSCLE & NERVE January 2019
 Table 1. Patients’ demographics for the entire cohort and the active and sham groups
Entire cohort (N = 48) Active group (N = 24) Sham group (N = 24) P-value*

Age (years) 61.6 (10.2) 63.1 (10.0) 60.1(10.3) 0.31

Gender (M/F) 29/19 17/7 12/12 0.14
Disease duration† 18.9 (11.4) 20.9 (14.5) 16.9 (6.8) 0.22
Onset (spinal/bulbar/mixed) 35/11/2 17/5/2 18/6/0 0.24

Data expressed as mean (standard deviation) or as number.

*Values are from t-tests for continuous measures and chi-square tests for categorical measures.
†
Mean number of months from symptom onset.

radiation and chemotherapy; and (2) patient no lon- DISCUSSION

ger wished to participate. Adherence for completing EMST was well tolerated and led to significant
prescribed exercises was excellent and ranged improvements in MEP (the primary outcome mea-
between 95% and 100% (between 950 and 1,000 sure), as well as total DIGEST scores and DIGEST
completed repetitions). Patients’ characteristics were efficiency subscale scores. Clinically significant
well balanced across groups with no statistically sig- trends for maintenance or improvements in peak
nificant baseline differences across treatment groups cough flow and oral intake, respectively, were also
for any of the demographic variables (see Table 1). noted. No significant changes were noted in other
A summary of results is provided in Table 2. measures of voluntary cough production, FVC, swal-
lowing safety, patient report of swallow (EAT-10),
and disease progression (ALSFRS-R).
Primary Outcome. For the primary outcome of MEP,
This study extends the findings of our earlier pilot
there were no statistically significant baseline differ- study that moderate intensity respiratory training
ences in MEP across treatment groups (see Table 2). using EMST is safe, feasible, and well-tolerated for
Group status had a significant influence on change early-stage ALS patients.20 Our patient retention rate
in MEP from pre to post time-points. Specifically, of 96% is not only higher than that of our earlier
those in the active EMST treatment group demon- pilot trial, but also substantially better than retention
strated a significantly higher increase from pre- to rates in most ALS clinical therapeutic trials, possibly
posttreatment compared with those in the sham due to the convenience of home-based therapy, facil-
group (Fig. 2). itated by weekly visits from a home therapist. On
average, MEP increased by 25% for patients assigned
Swallowing. After a consensus meeting, rater agree- to the active EMST group compared with a 6%
ment for DIGEST ratings was 100% without the need increase in the sham group over the 8-week treat-
for a third expert rater. After adjusting for multiple ment period. The small increase in MEPs observed
comparisons, significant group differences between for the sham group may be due to a potential learn-
pre to post time-points were observed for global swal- ing or familiarity effect of MEP testing procedures or
lowing function (total DIGEST score) and the due to variability of performing maximum pulmo-
DIGEST efficiency subscale score. Little to no change nary function tests and highlights the methodologic
occurred for ALS patients in the active group, whereas need for a comparison group in intervention-based
those in the sham group worsened from the pre to clinical trials in ALS. Increased MEP for the EMST
post time-points (Table 2). Although non-significant group supports our previous findings of a 29%
after adjusting for multiple comparisons, functional increase in MEPs in 15 ALS patients after 5 weeks of
oral intake (FOIS scores) improved by 14.4% in the EMST (compared with a 9% MEP reduction in these
active group and worsened by 11.8% over the 2-month patients during a 5-week lead-in period). Further-
period in the sham group (Table 2). No group differ- more, in the current study, 70% of patients (n = 16)
ences were observed in swallowing safety metrics in the active EMST group demonstrated increases in
(DIGEST safety subscore or PAS score) or in the MEPs of >15%. These increases are encouraging and
patient-reported swallowing outcome (EAT-10). suggest that this simple, non-invasive, at-home train-
ing program may not only maintain, but also
improve, function over the short term. Further work
Cough. After adjusting for multiple comparisons, no
is needed to explore the long-term impact of this
group differences were observed from pre to post
intervention.
time-points for cough spirometry measures (Table 2).
Although becoming statistically non-significant
after adjusting for multiple comparisons, of clinical
ALSFRS-R and FVC. No group differences were significance was the finding that peak expiratory
noted for ALSFRS-R scores or FVC (Table 2). cough flow remained stable (0% change) for those
Expiratory Training in ALS MUSCLE & NERVE January 2019 43
 Table 2. Mean group values across pre- and post- testing time points with raw and adjusted P Values.
ACTIVE GROUP: SHAM GROUP:
Raw Corrected
Mean Pre Mean Post Change* Mean Pre Mean Post Change F P P
(SD) (SD) (95% CI) (SD) (SD) (95% CI) or χ Value
2
Value Value

Maximum Expiratory Pressure 98.5 125.5 25.5 89.8 96.6 6.6 7.5 0.009* -
MEP
(cmH20) (45.7) (61.3) (14.3,36.7) (37.4) (45.9) (-3.4,16.5)
Inspiratory Peak Flow Rate -2.0 -2.3 -0.4 -1.7 -2.0 -0.2 0.8 0.38 0.61
(L/s) (1.1) (1.5) (-1.0,0.2) (0.8) (1.0) (-0.8,0.3)
Peak Expiratory Flow Rate 6.2 6.2 0 5.2 4.8 -0.6 5.2 0.03* 0.09
(L/s) (2.8) (2.7) (-1.3,1.3) (2.4) (2.4) (-1.5,0.4)
Cough Volume Acceleration (L/s/s) 147.2 147.2 -0.2 121.0 114.9 -11.9 2.6 0.12 0.32
COUGH: (92.2) (78.5) (-35.6,35.2) (86.9) (74.5) (-36.1,12.2)
Inspiratory Phase Duration 1.7 1.1 -0.1 1.4 1.4 0.1 1.1 0.30 0.59
(sec) (0.8) (0.7) (-0.6,0.4) (0.8) (0.7) (-0.5,0.6)
Expiratory Phase Duration 0.05 0.04 -0.01 0.06 0.06 0 1.7 0.20 0.46
(sec) (0.03) (0.02) (-0.02,0.01) (0.03) (0.04) (-0.01,0.01)
Compression Phase Duration (sec) 0.21 0.21 0 0.23 0.23 0 0.2 0.69 0.85
(0.11) (0.10) (-0.05,0.04) (0.12) (0.14) (-0.05,0.05)
Global: 2 3 (13.0%) 4.3% 2 (10.5%) 6 18.1 10.9 0.001* 0.02*
n=dysphagic (DIGEST>1) (8.7%) (-15.7,24.4) (28.6%) (-7.5,40.7)
(% dysphagic)
Safety: 5 4 (16.7%) -4.1% 9 (37.5%) 5 (22.7%) -14.8% 0.05 0.83 0.86
n= unsafe (PAS>3) (20.8%) (-28.9,20.0) (-38.2,11.6)
(% unsafe)
SWALLOWING: Efficiency: 2 3 (13.0%) 4.3% 2 (10.5%) 5 13.3% 9.3 0.002* 0.02*
n= inefficient (DIGEST-E>1) (8.7%) (-15.7,24.4) (23.8%) (-11.4,35.9)
(% inefficient)
Oral Intake, n= FOIS ≤ 6 10 (41.7%) 6 (27.3%) -14.4% 7 (29.2%) 9 11.8% 4.6 0.03* 0.10
(% FOIS ≤ 6) (38.5,12.7) (41.0%) (-14.9,35.6)
EAT-10, n=EAT-10 > 8 7 (29.2%) 6 -1.9% 7 8 6.0% 0.03 0.86 0.86
(%, EAT-10 > 8) (27.3%) (-26.4,23.4) (30,4%) (36.4%) (-20.4,31.4)
Total Score 36.6 (6.3) 35.7 -1.3 37.5 (6.1) 35.9 -2.1 0.24 0.63 0.84
(6.5) (-3.1, 0.4) (8.2) (-3.7,-0.5)
Respiratory Subscale 10.4 (1.8) 9.9 -0.4 10.0 (1.0) 10.5 -0.7 0.31 0.58 0.84
ALSFRS-R:
(1.7) (-1.1,0.3) (1.9) (-1.7,0)
Bulbar Subscale 10.3 10.1 -0.1 10.1 9.7 -0.6 0.33 0.89
(1.4) (1.5) (-0.5,0.3) (1.6) (2.3) (-1.1,-0.30)
Forced Vital Capacity: 80.7 73.7 -7.6 81.4 73.7 -8.3 0.04 0.84 0.86
FVC
(% Predicted) (10.0) (19.7) (-14.9,-0.3) (16.2) (18.4) (-14.7,-1.9)

Key: *Significant difference; F: F statistic with df = 1 (numerator), 45 (denominator); χ2: chi-square statistic with df = 1; ALSFRS-R: ALS Functional Rating Scale-Revised;
DIGEST: Dynamic Imagine Grade of Swallowing Toxicity; FOIS: Functional Oral Intake Scale; EAT-10: Eating Assessment Tool-10.

ALS patients performing active EMST exercises, yet improvements in voluntary cough after EMST in Par-
decreased and became weaker for those who did kinson’s disease,36 multiple sclerosis,37 and the sed-
not. This is particularly significant for individuals entary elderly,38 as well as in reflexive cough and
with ALS who commonly have difficulties with secre- urge-to-cough metrics after EMST in individuals with
tion management and excessive mucus build-up in stroke.39 Given that peak cough flow depends on
the lower airways. Other studies have reported expiratory pressure–generating forces, it is plausible
that the increased MEPs in the active group afforded
the maintenance of peak expiratory flow (i.e., cough
strength) in the present study.
No group differences were noted for swallowing
safety in this study. However, given that our ALSFRS-
R and FVC inclusion criteria were designed to enroll
early to mid-stage subjects, and that specific impair-
ments in swallowing safety were not part of our inclu-
sion/exclusion criteria, our cohort had a relatively
low incidence of unsafe swallowers (29%). In fact, less
than one third of the participants were unsafe swal-
lowers, with only 5 individuals in the active group
(20%) being penetrator/aspirators, substantially limit-
ing our ability to assess the impact of this intervention
on swallowing safety. Group differences, however,
FIGURE 2. Mean (95% confidence interval) maximum expiratory were observed across time-points for global swallowing
pressure (MEP, cm H2O) values for active (gray triangle) and
sham (black circle) groups across time-points. Regression analy-
function (total DIGEST score) and swallowing effi-
sis for change scores denotes a significant group difference in ciency (DIGEST efficiency subscale score) such that
change scores (P < 0.05). there was only a 4% increase in number of individuals
44 Expiratory Training in ALS MUSCLE & NERVE January 2019
demonstrating impaired DIGEST scores (>1) in the and death. Similarly, no findings were noted for
active group, whereas there was an 18% increase in FVC. Given that the FVC maneuver is dependent on
the sham group (i.e., this metric worsened in the both inspiratory and expiratory muscle strength and
sham group but remained relatively preserved in the capacity, it is not necessarily surprising given that the
EMST group). Previous studies have demonstrated current intervention only targeted expiratory force
that EMST facilitates repeated, submaximal, and load- generation, and not inspiratory muscle strength or
dependent demands on the submental, lingual, pala- volume recruitment observed with other therapies
tal, velopharyngeal, and pharyngeal musculature.40–43 (e.g., lung volume recruitment). Future efforts may
Given that synchronous firing and coordination of aim to determine the impact of a combined inspira-
these muscles is imperative for adequate pressure gen- tory and expiratory muscle strength training pro-
eration to facilitate bolus transport during swallowing, gram or the differential or perhaps synergistic
functional improvements or maintenance of swallow- impact of lung volume recruitment therapies com-
ing safety and efficiency after EMST are physiologi- pared with respiratory strength training paradigms.
cally supported. For example, coordinated activation In conclusion, a mild-to-moderate intensity,
of the extrinsic lingual muscles represent the main 8-week, at-home EMST program was well tolerated in
“pressure drivers” to transport the consumed bolus individuals with early ALS and resulted in significant
posteriorly, whereas the velopharyngeal and palatal improvements in MEP and oral intake, as well as
muscles elevate and retract to seal the nasopharynx, maintenance of peak cough flow and swallowing
to aid in pharyngeal pressure generation and subse- function. Further studies are warranted to validate
quent bolus flow. Finally, the submental musculature these findings and to investigate the long-term
facilitates superior and anterior laryngeal movement impact of this program on ALS disease progression.
during deglutition, which is imperative for epiglottic
The authors thank Dr. Christine Sapienza and Dr. Paul Davenport
inversion and adequate upper esophageal sphincter for methodological advice and expertise related to EMST.
relaxation. Given the documented impact EMST is We (the authors) confirm that we have read the Journal’s posi-
noted to have on the aforementioned bulbar muscu- tion on issues involved in ethical publication and affirm that this
lature, it is possible that the repetitive application of a report is consistent with those guidelines.
submaximal load on these muscles may have provided
muscular conditioning for ALS patients enrolled in REFERENCES
the active EMST group and contributed to mainte- 1. Desport JC, Preux PM, Truong TC, Vallat JM, Sautereau D, Couratier
P. Nutritional status is a prognostic factor for survival in ALS patients.
nance of swallowing efficiency (i.e., the degree of resi- Neurology. 1999;53(5):1059-1063.
due remaining in the pharynx) and our global 2. Hadjikoutis S, Wiles CM. Respiratory complications related to bulbar
dysfunction in motor neuron disease. Acta Neurol Scand. 2001;103(4):
swallowing measure (total DIGEST score). 207-213.
Nutrition is of the upmost importance in ALS, and 3. Kuhnlein P, Gdynia HJ, Sperfeld AD, et al. Diagnosis and treatment of
bulbar symptoms in amyotrophic lateral sclerosis. Nat Clin Pract Neu-
malnutrition (due primarily to dysphagia) confers a rol. 2008;4(7):366-374.
7.7-fold increased risk of death.44 Although becom- 4. Ruoppolo G, Schettino I, Frasca V, et al. Dysphagia in amyotrophic lat-
eral sclerosis: prevalence and clinical findings. Acta Neurologica Scandi-
ing statistically non-significant after correcting for navica. 2013;128(6):397-401.
multiple comparisons, a clinically meaningful finding 5. Wijesekera LC, Leigh PN. Amyotrophic lateral sclerosis. Orphanet J
Rare Dis. 2009;4:3.
was that 42% of active EMST group patients 6. Carreras I, Yuruker S, Aytan N, et al. Moderate exercise delays the
reported being on an altered or restricted diet at motor performance decline in a transgenic model of ALS. Brain Res.
2010;1313:192-201.
baseline that decreased to 27% posttreatment. In 7. Deforges S, Branchu J, Biondi O, et al. Motoneuron survival is pro-
comparison, oral intake was noted to worsen in the moted by specific exercise in a mouse model of amyotrophic lateral
sclerosis. J Physiol. 2009;587(Pt 14):3561-3572.
sham group, with 29% of individuals reporting con- 8. Gerber YN, Sabourin JC, Hugnot JP, Perrin FE. Unlike physical exer-
sumption of an altered or restricted diet at baseline cise, modified environment increases the lifespan of SOD1G93A mice
however both conditions induce cellular changes. PLoS One. 2012;7
that increased to 41% posttreatment. Future longer (9):e45503.
term studies should examine this more closely with 9. Kaspar BK, Frost LM, Christian L, Umapathi P, Gage FH. Synergy of
insulin-like growth factor-1 and exercise in amyotrophic lateral sclerosis.
measures such as body mass index, a known predic- Ann Neurol. 2005;57(5):649-655.
tor of survival in ALS.45–48 10. Veldink JH, Bar PR, Joosten EA, Otten M, Wokke JH, van den Berg LH.
Sexual differences in onset of disease and response to exercise in a
No significant changes were noted for the global transgenic model of ALS. Neuromuscul Disord. 2003;13(9):737-743.
disease progression marker (ALSFRS-R). It is likely 11. Bello-Haas VD, Florence JM, Kloos AD, et al. A randomized controlled
trial of resistance exercise in individuals with ALS. Neurology. 5 2007;
that a 2-month time period was not adequate to doc- 68(23):2003-2007.
ument disease-related progression or the potential 12. Bohannon RW. Results of resistance exercise on a patient with amyo-
trophic lateral sclerosis. A case report. Phys Ther. 1983;63(6):965-968.
impact of EMST on this metric. Future studies are 13. Cheah BC, Boland RA, Brodaty NE, et al. INSPIRATIonA-
needed to evaluate the impact of EMST on disease L--INSPIRAtory muscle training in amyotrophic lateral sclerosis. Amyo-
troph Lateral Scler. 2009;10(5-6):384-392.
progression over longer intervals and additional 14. Clawson LL, Cudkowicz M, Krivickas L, et al. A randomized controlled
important related variables including quality of life, trial of resistance and endurance exercise in amyotrophic lateral sclero-
sis. Amyotroph Lateral Scler Frontotemporal Degener. 2017:1-9.
hospitalization rates, aspiration pneumonia rates, 15. de Almeida JP, Silvestre R, Pinto AC, de Carvalho M. Exercise and
time to ventilator, time to feeding tube dependence, amyotrophic lateral sclerosis. Neurol Sci. 2012;33(1):9-15.

Expiratory Training in ALS MUSCLE & NERVE January 2019 45

16. Drory VE, Goltsman E, Reznik JG, Mosek A, Korczyn AD. The value of 34. Cedarbaum JM, Stambler N, Malta E, et al. The ALSFRS-R: a revised
muscle exercise in patients with amyotrophic lateral sclerosis. J Neurol ALS functional rating scale that incorporates assessments of respiratory
Sci. 2001;191(1-2):133-137. function. BDNF ALS Study Group (Phase III). J Neurol Sci. 1999;169
17. Pinto AC, Alves M, Nogueira A, et al. Can amyotrophic lateral sclerosis (1-2):13-21.
patients with respiratory insufficiency exercise? J Neurol Sci. 1999;169 35. Vasilopoulos T, Morey TE, Dhatariya K, Rice MJ. Limitations of Signifi-
(1-2):69-75. cance Testing in Clinical Research: A Review of Multiple Comparison
18. Pinto S, de Carvalho M. Can inspiratory muscle training increase sur- Corrections and Effect Size Calculations with Correlated Measures.
vival in early-affected amyotrophic lateral sclerosis patients? Amyotroph Anesth Analg. 2016;122(3):825-830.
Lateral Scler Frontotemporal Degener. 2013;14(2):124-126. 36. Pitts T, Bolser D, Rosenbek J, Troche M, Okun MS, Sapienza C. Impact
19. Pinto S, Swash M, de Carvalho M. Respiratory exercise in amyotrophic of expiratory muscle strength training on voluntary cough and swallow
lateral sclerosis. Amyotroph Lateral Scler. 2012;13(1):33-43. function in Parkinson disease. Chest. 2009;135(5):1301-1308.
20. Plowman EK, Watts SA, Tabor L, et al. Impact of expiratory strength 37. Chiara T, Martin AD, Davenport PW, Bolser DC. Expiratory muscle
training in amyotrophic lateral sclerosis. Muscle Nerve. 2016;54(1): strength training in persons with multiple sclerosis having mild to mod-
48-53. erate disability: effect on maximal expiratory pressure, pulmonary func-
21. Sanjak M, Bravver E, Bockenek WL, Norton HJ, Brooks BR. Supported tion, and maximal voluntary cough. Arch Phys Med Rehabil. 2006;87
treadmill ambulation for amyotrophic lateral sclerosis: a pilot study. (4):468-473.
Arch Phys Med Rehabil. 2010;91(12):1920-1929. 38. Kim J, Davenport P, Sapienza C. Effect of expiratory muscle strength
22. Tabor LC, Rosado KM, Robison R, Hegland K, Humbert IA, Plowman training on elderly cough function. Arch Gerontol Geriatr. 2009;48(3):
EK. Respiratory training in an individual with amyotrophic lateral 361-366.
sclerosis. Annals of Clinical and Translational Neurology. 2016;3(10): 39. Hegland KW, Davenport PW, Brandimore AE, Singletary FF, Troche
819-823. MS. Rehabilitation of Swallowing and Cough Functions Following
23. Plowman EK, Watts SA, Robison R, et al. Voluntary Cough Airflow Dif- Stroke: An Expiratory Muscle Strength Training Trial. Arch Phys Med
ferentiates Safe Versus Unsafe Swallowing in Amyotrophic Lateral Scle- Rehabil. 2016;97(8):1345-1351.
rosis. Dysphagia. 2016;31(3):383-390. 40. Wheeler KM, Chiara T, Sapienza CM. Surface electromyographic activ-
24. Gaziano J, Tabor L, Watts SA, Plowman EK. Prevalence, Timing and ity of the submental muscles during swallow and expiratory pressure
Source of Aspiration in Individuals with ALS. Dysphagia Research Soci- threshold training tasks. Dysphagia. 2007;22(2):108-116.
ety. Chicago, IL, 2015. 41. Wheeler-Hegland KM, Rosenbek JC, Sapienza CM. Submental sEMG
25. Plowman EK. Is There a Role for Exercise in the Management of Bul- and hyoid movement during Mendelsohn maneuver, effortful swallow,
bar Dysfunction in Amyotrophic Lateral Sclerosis? J Speech Lang Hear and expiratory muscle strength training. J Speech Lang Hear Res. 2008;
Res. 2015;58(4):1151-1166. 51(5):1072-1087.
26. Brooks BR, Miller RG, Swash M, Munsat TL. El Escorial revisited: 42. Hutcheson KA, Hammer MJ, Rosen SP, Jones CA, McCulloch TM. Expi-
revised criteria for the diagnosis of amyotrophic lateral sclerosis. Amyo- ratory muscle strength training evaluated with simultaneous high-reso-
troph Lateral Scler Other Motor Neuron Disord. 2000;1(5):293-299. lution manometry and electromyography. Laryngoscope. 2017;127(4):
27. Folstein MF, Folstein SE, McHugh PR. "Mini-mental state". A practical 797-804.
method for grading the cognitive state of patients for the clinician. J 43. Yanagisawa Y, Matsuo Y, Shuntoh H, Mitamura M, Horiuchi N. Change
Psychiatr Res. 1975;12(3):189-198. in tongue morphology in response to expiratory resistance loading
28. Troche MS, Okun MS, Rosenbek JC, et al. Aspiration and swallowing in investigated by magnetic resonance imaging. J Phys Ther Sci. 2013;25
Parkinson disease and rehabilitation with EMST: a randomized trial. (6):667-669.
Neurology. 2010;75(21):1912-1919. 44. Chio A, Logroscino G, Hardiman O, et al. Prognostic factors in ALS:
29. Hutcheson KA, Barrow MP, Barringer DA, et al. Dynamic Imaging A critical review. Amyotroph Lateral Scler. 2009;10(5-6):310-323.
Grade of Swallowing Toxicity (DIGEST): Scale development and valida- 45. Paganoni S, Deng J, Jaffa M, Cudkowicz ME, Wills AM. Body mass
tion. Cancer. 2017;123(1):62-70. index, not dyslipidemia, is an independent predictor of survival in
30. Rosenbek JC, Robbins JA, Roecker EB, Coyle JL, Wood JL. A penetra- amyotrophic lateral sclerosis. Muscle Nerve. 2011;44(1):20-24.
tion-aspiration scale. Dysphagia. 1996;11(2):93-98. 46. Peter RS, Rosenbohm A, Dupuis L, et al. Life course body mass index
31. Crary MA, Mann GD, Groher ME. Initial psychometric assessment of a and risk and prognosis of amyotrophic lateral sclerosis: results from the
functional oral intake scale for dysphagia in stroke patients. Arch Phys ALS registry Swabia. Eur J Epidemiol. 2017;32(10):901-908.
Med Rehabil. 2005;86(8):1516-1520. 47. Shimizu T, Nagaoka U, Nakayama Y, et al. Reduction rate of body mass
32. Belafsky PC, Mouadeb DA, Rees CJ, et al. Validity and reliability of the index predicts prognosis for survival in amyotrophic lateral sclerosis: a
Eating Assessment Tool (EAT-10). Ann Otol Rhinol Laryngol. 2008;117 multicenter study in Japan. Amyotroph Lateral Scler. 2012;13(4):
(12):919-924. 363-366.
33. Plowman EK, Tabor LC, Robison R, et al. Discriminant ability of the 48. Yang R, Huang R, Chen D, et al. Causes and places of death of patients
Eating Assessment Tool-10 to detect aspiration in individuals with amyo- with amyotrophic lateral sclerosis in south-west China. Amyotroph Lat-
trophic lateral sclerosis. Neurogastroenterol Motil. 2016;28(1):85-90. eral Scler. 2011;12(3):206-209.

46 Expiratory Training in ALS MUSCLE & NERVE January 2019