□ ORIGINAL ARTICLE □

Lung Inflation Training Using a Positive End-expiratory

Pressure Valve in Neuromuscular Disorders

Tsuyoshi Matsumura, Toshio Saito, Harutoshi Fujimura, Susumu Shinno and Saburo Sakoda

Abstract
Objective Respiratory muscle weakness causes alveolar hypoventilation and reduced lung compliance in
neuromuscular disorders. Lung inflation is important to prevent secondary pulmonary complications however
respiratory and laryngeal dysfunction often hamper lung inflation. There is a need for a convenient and low-
cost device that enables effective lung inflation. We tested a lung inflation training method using a positive
end-expiratory pressure (PEEP) valve.
Methods Vital capacity (VC), maximum insufflation capacity (MIC) and peak cough flow (PCF) as well as
PEEP lung inflation capacity (PIC) were assessed in 93 neuromuscular patients. Consecutive PIC training
was done for 4 months in six tracheostomized Duchenne muscular dystrophy (DMD) patients and PIC was
assessed before and after training.
Results PIC training was practicable in all participants and no serious adverse events were detected. PIC
was significantly higher than VC or MIC in all disorders, although MIC was higher than VC in DMD only.
Patients with dysphagia showed lower MIC and PCF compared with non dysphagic patients. PIC was as low
as 345±77 mL in tracheostomized DMD patients; however consecutive training increased it up to 619±205
mL.
Conclusion The PEEP valve enabled effective lung inflation regardless of laryngeal function. Consecutive
training can improve lung condition even in advanced cases. Early introduction of PIC training could be ef-
fective at preventing respiratory complications in patients with neuromuscular disorders.

Key words: neuromuscular disorders, respiratory physical therapy, lung inflation training, positive end-
expiratory pressure, resuscitation bag

(Intern Med 51: 711-716, 2012)

(DOI: 10.2169/internalmedicine.51.6258)

maintenance of airway clearance is essential to prevent res-

Introduction
Respiratory muscle weakness causes alveolar hypoventila-
tion and static lung compliance reduction in neuromuscular
disorders (NMD) such as amyotrophic lateral sclerosis
(ALS) and muscular dystrophy. Dysphagia and ineffective
cough frequently co-occur in these patients, and these pa-
tients were vulnerable to recurrent respiratory infections.
Respiratory management has markedly improved the life ex-
pectation of patients with NMD (1-3) and currently, these
patients commonly receive mechanical ventilation over a
decade. Most of these patients usually live at home. There-
fore not only appropriate ventilatory management but also
piratory infection and improve the prognosis. For this goal,
respiratory physical therapy in patients with NMD places
priority on lung inflation training and coughing train-
ing (1, 4, 5). Lung inflation training, which prevents alveo-
lar collapse, helps maintain lung compliance and increase
cough flow. Because adequate peak cough flow is required
to expel airway mucus (6), manual and/or mechanical assis-
tance is introduced to produce more cough flow when pa-
tients cannot cough effectively (1, 4, 6, 7).
Maximum insufflation capacity (MIC) (4) has been
widely used for lung inflation training. However MIC train-
ing is not completely effective in patients with bulbar palsy
or tracheotomy due to laryngeal malfunction (4). Alternative

Department of Neurology, National Hospital Organization Toneyama National Hospital, Japan

Received for publication July 26, 2011; Accepted for publication December 4, 2011
Correspondence to Dr. Tsuyoshi Matsumura, matumura@toneyama.hosp.go.jp

711
 Intern Med 51: 711-716, 2012 DOI: 10.2169/internalmedicine.51.6258

Table 1. Participants Profiles
n (M/F) Age (y.o.) DD (year) VC (mL) PCF (L/sec) Alb (g/dL) Diet Ventilation
(mean ± SD) (mean ± SD) (mean ± SD) (mean ± SD) (mean ± SD) (Ord/Arr/Tube) (none/NIV/TIV)
ALS 40 (19/21) 63.0 ± 12.4 3.5 ± 2.3 1200 ± 794 122 ± 92 3.65 ± 0.37 12/11/17 23/13/4
DM 21 (14/7) 51.7 ± 9.2 20.8 ± 10.3 1277 ± 566 177 ± 84 3.60 ± 0.37 16/5/0 10/11/0
DMD 32 (32/0) 18.0 ± 5.6 15.0 ± 4.6 1157 ± 636 172 ± 84 4.35 ± 0.49 30/2/0 16/16/0
TIV patients Age (y.o.) DD (year) VC (mL) PCF (L/sec) Alb (g/dL) Diet MVD
ALS 4 (4/0) 61.8 ± 9.7 5.6 ± 3.1 425 ± 530 43 ± 75 3.53 ± 0.34 0/0/4 2.3 ± 2.8
DMD 6 (6/0) 41.7 ± 7.1 38.0 ± 6.5 <50 unmeasurable 3.78 ± 0.56 2/2/3 13.7 ± 2.9
Abbreviations: M, male; F, female; y.o., years old; DD, disease duration; VC, vital capacity; PCF, peak cough flow; Alb, serum albumin;
Ord, oral intake of ordinary diet; Arr, oral intake of arranged food (thickening diet, blender diet, jelly, etc.); Tube, tube feeding; NIV,
non-invasive ventilation; TIV, tracheal intermittent positive pressure ventilation; MVD, duration receiving mechanical ventilation therapy.

lung inflation techniques for these patients include a resusci-
tation bag equipped with a one-way valve (8); a volume
controlled respirator with a deep inspiration mode; a rela-
tively high-pressure support/controlled ventilator (5, 9); and
a mechanical in and ex-sufflator (MIE) (7). However, these
methods are associated with some difficulties. Although one-
way valves may be substituted for poor laryngeal function,
they increase the risk of barotrauma. A respirator with deep
inspiration mode can be used only in patients with a
volume-controlled respirator. Similarly, the use of higher
pressure can be also done in only patients using pressure a
support/controlled ventilator. The Japanese medical insur-
ance approves the use of MIE only in NMD patients receiv-
ing home mechanical ventilation (HMV). Considering the
fact that asymptomatic pneumothorax is not rare in muscular
dystrophy (10) and MIE can be associated with pneumotho-
rax (11), we must focus our attention on lung inflation train-
ing using positive pressure.
We hypothesized that a positive end-expiratory pressure
(PEEP) valve can resolve these problems. A PEEP valve can
stack air until the airway pressure reaches the set value,
which enables lung inflation independent from laryngeal
function. In addition, when the airway pressure exceeds the
set value, air leak occurs automatically. Thus, we can avoid
unexpected high airway pressure, which causes barotrauma.
The low cost (approximately US $25) is also favorable for
home-use. To certify its efficacy and safety, we tested lung
inflation training using a PEEP valve.
Materials and Methods
The institutional ethical review board of Toneyama Na-
tional Hospital approved the protocol of this study, and all
participants provided informed written consent.
Subjects were selected from patients with ALS, myotonic
dystrophy (DM) or Duchenne muscular dystrophy (DMD),
because respiratory dysfunction is inevitable in these disor-
ders and bulbar dysfunction is common in ALS and DM.
Those with obvious bullae, bronchial asthma, other active
pulmonary illness, a history of pneumothorax, or severe
congestive heart failure were excluded.
Single training effects were analyzed in 93 patients (ALS,
40; DM, 21; DMD, 32; Table 1) whose vital capacity (VC)
and MIC (only non-tracheostomized patients) exceeded 50
mL. In this study, we measured VC, MIC (4), and PEEP
lung insufflation capacity (PIC) in the supine position using
a Wright/Haloscale respirometerⓇ. Peak cough flow (PCF)
was also measured using a peak flow meter. Tra-
cheostomized patients were excluded from MIC assessment,
as air stacking is not possible in these patients. All measure-
ments were made by respiratory therapists. We measured
PIC by attaching the PEEP valve (20 cm H2O) to the resus-
citation bag, tightly fitting the mask on the patient’s face
(for tracheostomized patients, the resuscitation bag was con-
nected to the tracheal cannula), delivering air until a leak
started from the PEEP valve, releasing the PEEP valve, and
inducing maximum expiration. We performed each proce-
dure three times and recorded the maximum value. Because
intraindividual variability in VC of these patients was less
than 15%, we regarded MIC and/or PIC as effective when
MIC or PIC exceeded 120% of VC, no change when MIC
or PIC was between 80 and 120% of VC, and deteriorated
when MIC or PIC was less than 80% of VC. To assess the
effect of dysphagia (bulbar dysfunction) on lung inflation,
patients were divided according to their diet; those without
obvious dysphagia and eating an ordinary diet (ordinary
group) and those with dysphagia and eating an arranged diet
(thickening diet, blender diet, jelly etc.) or receiving tube
feeding (dysphagic group).
Long-term effects were assessed in six tracheostomized
DMD patients over 4 months; all six were inpatients with a
mean age of 41.7±7.1 years and a mean duration of respira-
tory management of 13.7±2.9 years. They could hardly per-
form deep breathing and VC was all under 50 mL (Table 1).
In terms of cardiac function, brain natriuretic peptide was
23.0±8.8 pg/mL, left ventricular diastolic diameter was
45.3±6.0 mm, and left ventricular ejection fraction was
35.3±8.0%. None showed obvious cardiac symptoms. Two
sets of five lung inflations at 20 cm H2O of PEEP were
done by nursing staff 5 days per week for 4 months, and the
maximum PIC before and after the training were compared.
The number of febrile events due to respiratory infection in
the 4 months before and during the study was also counted.
Statistical analysis
We used the paired t-test to compare VC, MIC, and PIC
and to evaluate PIC before and after training; the Wilcoxon
signed-rank test to compare the effect of MIC and PIC;
analysis of covariance (ANCOVA) using VC as covariate to
compare MIC, PIC, and PCF between the ordinary and

712
 Intern Med 51: 711-716, 2012 DOI: 10.2169/internalmedicine.51.6258

(mL) p<0.001
p<0.001
p<0.001
p<0.001
2500 p<0.001
p<0.001
p<0.001

2000

1500

1000

500

VC MIC PIC VC MIC PIC VC MIC PIC mean±SD

0
ALS n:40 DM n:21 DMD n:32

Figure 1. Comparison of vital capacity (VC), maximum insufflation (MIC), and positive end-expi-
ratory pressure (PEEP) insufflation capacity (PIC) in each disorder. Paired t-test was used to assess
differences. In all disorders, PIC was higher than VC or MIC. In Duchenne muscular dystrophy
(DMD), MIC was significantly higher than VC.

dysphagic group; and the student’s t test to compare PIC be-
tween tracheostomized ALS and DMD patients. All analyses
were performed using SPSS version 18.0J (SPSS, Chicago,
IL) statistical software. Results are presented as means ±
SDs. For all statistical tests, the level of statistical signifi-
cance was set at p<0.05.
Results
Single training
The PIC training was practicable for all participants. PIC
exceeded 400 mL in all subjects. VC, MIC, and PIC were
1,200±794, 1,226±868 and 1,644±727 mL, respectively for
ALS patients; 1,277±566, 1,400±572 and 1,752±536 mL for
DM patients; and 1,157±636, 1,441±762 and 1,689±815 mL
for DMD patients. For DMD patients, MIC was significantly
greater than VC, but no significant differences between MIC
and VC were detected for ALS (p=0.693) and DM (p=
0.131) patients. In all disorders, PIC was significantly
greater than VC or MIC (Fig. 1).
Among non-tracheostomized patients, PIC training
showed a higher rate of effectiveness than MIC training in
all disorders, although significant differences were detected
in only ALS (p=0.002) and DM (p=0.021) patients (Ta-
ble 2). PIC was higher than 120% of MIC in 20 of 34 pa-
tients with ALS, 8 of 18 with DM and 13 of 30 with DMD.
No subjects presented PIC under 80% of VC, although the
MIC of nine patients was below 80% of VC. PIC was lower
than 80% of MIC in one patient who complained strongly
of discomfort and was reluctant to do PIC training.
VC was lower in the dysphagic group compared with the
ordinary group, although differences were not significant.
MIC, PIC, and PCF were compared with ANCOVA with
VC as a covariate. As a result, MIC (p=0.001) and PCF (p=
0.001) were significantly lower in the dysphagic group. Al-
though PIC was also slightly lower in the dysphagic group,
the difference was not significant (p=0.105) (Fig. 2).
No serious complications were detected during the study,
although seven patients who did not receive mechanical ven-
tilation and were unfamiliar with positive pressure com-
plained of the difficulty in expiration that was caused by the
mask and/or the PEEP valve.
Long-term training
The mean PIC of tracheostomized DMD patients was
345±77 mL, which was significantly lower than that of tra-
cheostomized ALS patients (1,128±724 mL, p=0.027). In
these patients, sustained PIC training for 4 months signifi-
cantly increased PIC up to 619±205 mL (p=0.009). PIC was
increased in all patients, although one patient with severe
deformity (Kobb angle : 105°) showed minimum gain
(Fig. 3A). No patient complained, and no obvious changes
in vital signs (mean changes between before and after train-
ing was 0.3±5.1 mmHg in systolic blood pressure, -0.6±5.0
mmHg in diastolic blood pressure and 2.1±4.4/min in pulse
rate) were observed during the training. The number of feb-
rile events was decreased in three patients, although one pa-
tient presented with fever only after the start of training
(Fig. 3B).

713
 Intern Med 51: 711-716, 2012 DOI: 10.2169/internalmedicine.51.6258

Table 2. Cross-tabulation Table Showing Effectiveness of MIC and

PIC

MIC
Effective No change Deteriorated
PIC
ALS (n=34)
Effective 9 10 3
No change 2 9 1
Deteriorated 0 0 0
DM (n=18)
Effective 8 5 1
No change 1 2 1
Deteriorated 0 0 0
DMD (n=30)
Effective 16 4 1
No change 1 6 2
Deteriorated 0 0 0
MIC, maximum insufflations capacity; PIC, positive end-expiratory pressure
inflation capacity; ALS, amyotrophic lateral sclerosis; DM, myotonic dystrophy;
DMD, Duchenne muscular dystrophy.
Each subgroup was determined as follows: effective, value exceeded 120% of
VC; no change, the value was within 80-120% of VC; deteriorated, value was
below 80% of VC.

p=0.001
(mL) (L/min)
2500 p=0.001
250 safety, portability and low cost are important factors in lung
inflation. MIC is a common method, although many patients
cannot achieve adequate inflation because of laryngeal dys-
2000 200
function or technical problems. In this study, the effect of
MIC was poor in patients with ALS and DM. Although MIE
1500 150 is an effective tool, Japanese medical insurance allows its
Ordinary group use only for NMD patients receiving HMV. Thus it is not
n=58
suitable for patients who are not undergoing respiratory
1000 100
Dysphagic group management. It is noteworthy that PIC training meets most
n=35
of the requirements specified above. All equipment used in
500 50 this study is commercially available at a low cost, also it is
very compact and light, making it convenient and easily
mean±SD transportable.
VC MIC PIC PCF
In this study, all subjects, including tracheostomized pa-
Figure 2. Comparison of the effects of maximum insuffla- tients could perform PIC training. At first, we were con-
tion capacity (MIC) and positive end-expiration pressure cerned that 20 cm H2O would be insufficient for lung infla-
(PEEP) insufflation capacity (PIC) between patients with and tion because previous studies (5, 9, 12, 13) employed higher
without dysphagia. Non-tracheostomized patients (n=93) were pressure. Although there is another tool that allows for
subdivided into two groups: the ordinary group (patients higher pressure (14), this tool has not been approved as a
without obvious dysphagia and eating an ordinary diet) and medical device in Japan. Despite our concern, PIC was sig-
the dysphagic group (patients with dysphagia and eating an nificantly higher than VC or MIC in all disorders. The rate
arranged diet or undergoing tube feeding). Because there was of effectiveness was also higher in PIC than MIC in ALS
no significant difference in VC between groups, analyses of and DM. No patients were assessed as deteriorated in PIC
covariance using VC as covariate was used to assess the differ- compared to VC. These facts confirm that PIC is an effec-
ences. MIC and peak cough flow (PCF) were lower in the dys- tive technique for lung inflation and independent from la-
phagic group (p=0.001). PIC was also lower in the dysphagic ryngeal function or techniques.
group, although differences were not significant. No serious adverse events were detected during the single
or long-term study. Vital signs were not influenced even in
advanced cases. We believe the safety of PIC training is
high, although it is important to exercise caution in patients
Discussion with contraindications such as active pneumothorax, or se-
vere congestive heart failure.
Lung inflation training is essential for NMD patients to PIC training was less effective when patients resisted it.
prevent secondary lung diseases. Practicability, availability, PIC was less than 80% of MIC in a patient who was reluc-

714
 Intern Med 51: 711-716, 2012 DOI: 10.2169/internalmedicine.51.6258

A. B.
PIC (mL) Febrile event (time)
p=0.009 6
1000

800
4

600 3

2
400

1
200
mean±SD 0
Before training After trainig (n=6) Before training During training
4 months 4 months (n=6)

Figure 3. Effects of long-term positive end-expiratory pressure (PEEP) lung inflation training in

tracheostomized Duchenne muscular dystrophy (DMD) patients. Six tracheostomized DMD patients
underwent continuous PEEP lung inflation training for 4 months. (A) We compared the maximum
PIC before and after training by the paired t-test. The mean PIC was as low as 345±77 before train-
ing; sustained training for 4 months significantly increased the mean PIC to 619±205 mL (p=0.009).
Although PIC increased in all patients, one patient (open circle) who had a severe deformity (Kobb
angle : 105°) showed the smallest increase of PIC (pre 320 mL, post 428 mL). (B) The number of fe-
brile events in 4 months before and after training was also counted. Febrile events were decreased in
three patients, although one patient presented fever only after starting training.

tant to train. Some patients who were not on mechanical
ventilation complained about the discomfort associated with
PIC training. In such cases, an adequate explanation and re-
laxation are important to maximize the effects.
PIC was low in six tracheostomized DMD patients. Al-
though not significant, dysphagic patients showed lower PIC
than non-dysphagic patients. These findings suggest that
lung compliance may have been damaged in patients with
advanced disease. Higher pressure might enable more infla-
tion, although it is critical to avoid barotrauma (10, 11). To
date, both positive (5, 9) and negative (12, 13) findings have
been reported for the effect of positive pressure training, and
it is suggested that a brief period of lung inflation may not
be enough to reverse long-standing alterations in the
lung (12, 15). Thus we assessed the long-term effect of the
PIC training for tracheostomized DMD patients. Similar to a
recent report (8), PIC was improved in all patients with sus-
tained training. In addition, febrile events were also reduced.
It seems that consecutive PEEP lung inflation is effective at
improving lung conditions even in advanced cases.
Conclusion
Lung inflation training with a PEEP valve is effective,
convenient, and safe in NMD patients. Although consecutive
training was effective even in advanced cases, we recom-
mend early preventive introduction of this training. We hope
that wider use of this lung inflation method will improve the
health and quality of life of NMD patients.
The authors state that they have no Conflict of Interest (COI).
Acknowledgement
We thank all participants and rehabilitation and nursing staff
who cooperated with this study. This study was supported by a
Research Grant for Nervous and Mental Disorders (20A-12)
from the Ministry of Health, Labour, and Welfare of Japan. We
received no other financial support and there are no conflicts of
interest to declare.
References
1. Gomez-Merino E, Bach JR. Duchenne muscular dystrophy: pro-
longation of life by noninvasive ventilation and mechanically as-
sisted coughing. Am J Phys Med Rehabil 81: 411-415, 2002.
2. Tatara K, Shinno S. Management of mechanical ventilation and
prognosis in Duchenne muscular dystrophy. Iryo 62: 566-571,
2008 (in Japanese).
3. Ishikawa Y, Miura T, Aoyagi T, Ogata H, Hamada S, Minami R.
Duchenne muscular dystrophy: survival by cardio-respiratory in-
terventions. Neuromuscul Disord 21: 47-51, 2011.
4. Kang SW, Bach JR. Maximum insufflation capacity: vital capacity
and cough flows in neuromuscular disease. Am J Phys Med Reha-
bil 79: 222-227, 2000.
5. Lechtzin N, Shade D, Clawson L, Wiener CM. Supramaximal in-
flation improves lung compliance in subjects with amyotrophic
lateral sclerosis. Chest 129: 1322-1329, 2006.
6. Bach JR. Amyotrophic lateral sclerosis: predictors for prolongation
of life by noninvasive respiratory aids. Arch Phys Med Rehabil
76: 828-832, 1995.
7. Bach JR. Mechanical insufflation-exsufflation. Comparison of
peak expiratory flows with manually assisted and unassisted
coughing techniques. Chest 104: 1553-1562, 1993.
8. Bach JR, Mahajan K, Lipa B, Saporito L, Goncalves M, Komaroff

715
 Intern Med 51: 711-716, 2012 DOI: 10.2169/internalmedicine.51.6258

E. Lung insufflation capacity in neuromuscular disease. Am J
Phys Med Rehabil 87: 720-725, 2008.
9. Sinha R, Bergofsky EH. Prolonged alteration of lung mechanics in
kyphoscoliosis by positive pressure hyperinflation. Am Rev Respir
Dis 106: 47-57, 1972.
10. Yamamoto T, Kawai M. Spontaneous pneumothorax in Duchenne
muscular dystrophy. Rinsho Shinkeigaku 34: 552-556, 1994 (in
Japanese).
11. Suri P, Burns SP, Bach JR. Pneumothorax associated with me-
chanical insufflation-exsufflation and related factors. Am J Phys
Med Rehabil 87: 951-955, 2008.
12. De Troyer A, Deisser P. The effects of intermittent positive pres-
sure breathing on patients with respiratory muscle weakness. Am
Rev Respir Dis 124: 132-137, 1981.
13. McCool FD, Mayewski RF, Shayne DS, Gibson CJ, Griggs RC,
Hyde RW. Intermittent positive pressure breathing in patients with
respiratory muscle weakness. Alterations in total respiratory sys-
tem compliance. Chest 90: 546-552, 1986.
14. Pitts T, Bolser D, Rosenbek J, Troche M, Okun MS, Sapienza C.
Impact of expiratory muscle strength training on voluntary cough
and swallow function in Parkinson disease. Chest 135: 1301-1308,
2009.
15. Estenne M, Gevenois PA, Kinnear W, Soudon P, Heilporn A, De
Troyer A. Lung volume restriction in patients with chronic respi-
ratory muscle weakness: the role of microatelectasis. Thorax 48:
698-701, 1993.

Ⓒ 2012 The Japanese Society of Internal Medicine

http://www.naika.or.jp/imindex.html

716