Effects of Inspiratory and Expiratory Muscle Training in Normal Subjects

Abstract

The present study aimed to clarify the effects of inspiratory muscle training (IMT) and expiratory muscle training (EMT) on ventilatory muscle strength, pulmonary function and responses during exercise testing. Young healthy women were randomly assigned to 3 groups: IMT (n=16); EMT (n=16); or untrained normal controls (NC, n=8). Subjects in the IMT and EMT groups trained for 15 minutes twice daily over 2 weeks at loads of 30% maximal inspiratory and expiratory muscle strength, respectively. Ventilatory muscle strength (maximal inspiratory and expiratory muscle strength; PI_max and PE_max, respectively), pulmonary function and progressive exercise testing was performed. Both PI_max and PE_max increased in the IMT group, and PE_max increased in the EMT group. Neither trained group demonstrated any change in pulmonary function or peak values during exercise testing. In the IMT group, exercise-induced increases in heart rate, oxygen uptake (VO₂/kg) and rating of perceived exertion (RPE) decreased with training, as did increases in VO₂/kg and RPE in the EMT group. The increased ventilatory muscle strength in both IMT and EMT groups might improve ventilatory efficacy during exercise, and increased inspiratory muscle strength might facilitate oxygen delivery through improved circulatory responses.

Ventilatory muscle training (VMT) represents one component of respiratory rehabilitation¹⁾²⁾. The rationale behind VMT is that enhancing respiratory muscle function can potentially reduce the severity of breathlessness and improve exercise tolerance¹⁾. From evidence-based guidelines produced by a joint panel of the American College of Chest Physicians (ACCP) and the American Assoc ia tion of Cardiovascular and Pulmonary Rehabilitation (AACVPR)¹⁾, VMT may be considered for individual patients with chronic obstructive pulmonary disease (COPD) who remain symptomatic despite optimal therapy, although scientific evidence does not support routine use of VMT in pulmonary rehabilitation. After a meta-analysis of VMT by Smith and coworkers³⁾, whose analysis was referred to in the evidence-based guidelines, Lötters and colleagues⁴⁾ concluded that VMT significantly increases respiratory muscle strength and respiratory muscle endurance, decreases sensations of dyspnea at rest and during exercise, and tends to improve functional exercise capacity. Conversely, Salman and colleagues⁵⁾ undertook meta-analysis, and concluded that VMT did not improve walking distance or sensation of dyspnea.

All the above studies of VMT referred to inspiratory muscle training (IMT). However, respiratory muscle fatigue is generated in both expiratory and inspiratory muscles⁶⁾, and experiments have attempted to verify the effects of expiratory muscle training (EMT)⁷⁾⁸⁾. Suzuki et al.⁷⁾ reported that EMT increased expiratory muscle strength and decreased sensation of respiratory effort during exercise, and Akiyoshi et al.⁸⁾ indicated that EMT increased both expiratory and inspiratory muscle strength.

To clarify the effects of EMT and IMT on ventilatory muscle strength measured by mouth pressure, pulmonary function and responses during exercise testing, both methods were examined in young healthy subjects.

Methods

Subjects

A total of 40 young healthy women participated in this study. Mean (± SD) age, height, weight and body mass index (BMI) were 19.3 ± 1.4 years, 158.7 ± 4.6 cm, 54.5 ± 7.6 kg and 21.6 ± 2.6 kg/m² respectively. Subjects were randomly assigned to the IMT group (n=16), EMT group (n=16) or untrained normal control (NC) group (n=8). Informed consent was obtained from each subject. Table 1 presents characteristics and peak values during exercise testing for each group before the trial. No significant difference in age, physique, pulmonary function or peak values during exercise testing were observed among the three groups.

Table 1. Characteristics and peak values during exercise testing for each group at baseline.

	IMT group (n=16)	EMT group (n=16)	NC group (n=8)	ANOVA (significance)
Age (years)	18.8 ± 0.7	19.3 ± 1.0	20.4 ± 2.3	N.S.
BMI (kg/m2)	20.5 ± 1.4	22.7 ± 3.0	21.5 ± 2.8	N.S.
PImax (cmH2O)	68.6 ± 19.8	75.4 ± 18.6	76.1 ± 16.0	N.S.
PEmax (cmH2O)	68.6 ± 21.2	76.6 ± 18.6	72.8 ± 15.6	N.S.
VC (l)	3.2 ± 0.4	3.4 ± 0.4	3.3 ± 0.4	N.S.
FEV1.0 (l)	2.8 ± 0.5	3.0 ± 0.3	2.9 ± 0.3	N.S.
Peak VE (l/min)	37.3 ± 6.7	43.1 ± 8.9	38.9 ± 2.4	N.S.
Peak TV (l)	1.3 ± 0.2	1.4 ± 0.3	1.3 ± 0.3	N.S.
Peak RR (breaths/min)	29.3 ± 4.6	31.2 ± 4.5	31.0 ± 5.1	N.S.
Peak VO2/kg (ml/kg/min)	25.5 ± 3.7	26.4 ± 4.1	25.2 ± 3.4	N.S.
Peak RPE	6.8 ± 1.9	7.6 ± 2.4	6.3 ± 2.1	N.S.

mean ± SD, IMT; inspiratory muscle training, EMT; expiratory muscle training, NC; normal control, BMI; body mass index, PI_max; maximal inspiratory pressure, PE_max; maximal expiratory pressure, VC; vital capacity, FEV_1.0; forced expiratory volume in one second, VE; minute ventilation, TV; tidal volume, RR; respiratory rate, VO₂/kg; oxygen uptake/body weight, RPE; rating of perceived exertion, N.S.; not significant.

Device for VMT

Subjects in the IMT group used Threshold-IMT (HealthScan, New Jersey, USA) (Fig. 1A). This device comprises a plastic tube incorporating a spring-loaded valve occluding the inspiratory orifice at one end and a mouthpiece at the other end. When the subject breathes through the mouthpiece, they must generate sufficient vacuum pressure to open the spring-loaded valve and initiate airflow; exhalation is unloaded. The amount of resistance can be adjusted by varying the compression of the spring-loaded valve. Adjustment from 7 cmH₂O to 41 cmH₂O is possible.

Fig 1.

Training devices.

Subjects in the EMT group used Souffle (Kayaku, Tokyo, Japan) (Fig. 1B). This device comprises a central container with a dead space of 400 or 800 ml, a mouthpiece and a metal positive end-expiratory pressure (PEEP) plate elastic-loaded valve occluding the expiratory orifice. When subjects exhale through the mouthpiece they must generate positive pressure equivalent to 5, 10 or 15 cmH₂O to open the PEEP plate elastic-loaded valve. As subjects re-inhale a portion of the expired gas in the container during inhalation, ventilation is promoted. The present study utilized the 800-ml container, adjusted to an appropriate pressure by modifying load pressure and measuring inner pressure of the Souffle during expiration using a manometer (Bird; 3M, Tokyo, Japan).

Experimental protocol

IMT subjects trained using a Threshold-IMT at 30% of maximal inspiratory muscle strength for 15 minutes twice daily. EMT subjects trained using Souffle at 30% of maximal expiratory muscle strength for 15 minutes twice daily. Training was continued for two weeks, and no training was undertaken by the NC group. Subjects in the two training groups predominantly performed under supervision by an examiner, and trained at home on holidays. After the trial period, a record of performance was made to indicate state of training.

Measurement of ventilatory muscle strength, pulmonary function testing (using spirometry) and progressive exercise testing were undertaken before and after training. Ventilatory muscle strength was measured using Vitalopower KH-101 equipment (Chest, Tokyo, Japan), and mouth pressure was considered as ventilatory muscle strength. According to the methods of Black and Hyatt⁹⁾, subjects utilized a nose-clip and mouthpiece, and maintained inspiratory and expiratory effort against a closed valve at least 1 second from residual volume (RV) and total lung capacity (TLC), respectively. In addition, both efforts were measured at functional residual capacity (FRC). All efforts were repeated three times, with peak values recorded as maximal inspiratory and expiratory muscle strength (PI_max and PE_max, respectively). Pulmonary function testing was performed using an Autospiro AS-300 electrical spirometer (Minato Medical Science, Osaka, Japan), using vital capacity (VC) and forced expiratory volume in one second (FEV_1.0). Exercise tolerance testing involved a progressive walking exercise test on a treadmill (Rehabilitation Tread Mill; BIODEX, New York, USA). A modification of the 3MPH protocol by Balk¹⁰⁾ was selected for the exercise testing protocol. This protocol was up to stage 7, with slope increasing every 2 minutes from 0% to 2%, 5%, 7%, 10%, 12% and 15% at a steady speed of 3 miles/hour after a 3-minute warm-up at 2 miles/hour. Subjects lay quietly on a platform for at least 20 minutes prior to exercise testing. The exercise test was performed after fitting a transmitter for electrocardiography (Life Scope 8; Nihon Kohden, Tokyo, Japan) and mask for an expiratory gas analyzer (Metabolic Measurement System 2900c; Sensor Medics, California, USA). Baseline data were measured in the standing position on the treadmill. Target heart rate was considered (220 − age) × 85%, and testing was ceased when target heart rate was reached. The air conditioner was set at 23°C, so the environmental temperature in the experiment room could be kept relatively constant.

Statistical analysis

All values were reported as mean ± SD. For ventilatory muscle strength (PI_max and PE_max) and pulmonary function (VC and FEV_1.0), pre- and post-training comparisons were analyzed using paired t-tests. Differences in changes with training between groups were analyzed using two-way repeated measures analysis of variance (ANOVA). Heart rate (HR), minute ventilation (VE), tidal volume (TV), respiratory rate (RR), oxygen uptake/body weight (VO₂/kg) and rating of perceived exertion (RPE) using the Borg score¹¹⁾ were selected as parameters of exercise testing. For these parameters, comparisons of peak values between preand post-training were analyzed using paired t-tests, and differences in exercise-induced increases between pre- and post-training within each group were analyzed using two-way repeated measures ANOVA.

Results

The trained state in the training groups and the lifestyle in all subjects

Based on training state records, while of 28 times in the trial period, subjects were performing training 26.6 times on the average (range ; 24–28 times) in IMT group, and 27.1 times on the average (range ; 25–28 times) in EMT group. Those who have a movement custom from the investigation in question paper were one competition level and two recreation levels. The number of those with a smoking custom was eight. All subjects were reported to have been changeless to which lifestyle during the trial.

Changes in ventilatory muscle strength

Changes in ventilatory muscle strength with training are shown Table 2. In the NC group, neither PI_max nor PE_max changed over the two weeks. In the IMT group, PI_max differed significantly compared to that in the NC group (df 1, F=10.844, p<0.01), and had increased significantly from 68.6 ± 19.8 cmH₂O to 79.7 ± 20.4 cmH₂O (p<0.01). PEmax in the IMT group increased significantly from 68.6 ± 21.2 cmH₂O to 73.6 ± 22.2 cmH₂O (p<0.05). In the EMT group, PE_max increased significantly from 76.6 ± 18.6 cmH₂O to 84.5 ± 22.1 cmH₂O (p<0.05), whereas no changes in PE_max were observed compared to the NC group. PI_max was unchanged in the EMT group.

Table 2. Change of ventilatiory muscle strength for ventilation muscle training.

	IMT group (n=16)	EMT group (n=16)	NC group (n=8)	ANOVA (interaction)
PImax (cmH2O)
Pre	68.6 ± 19.8	75.4 ±18.6	76.1 ± 16.0
Post	79.7 ± 20.4	81.4 ±17.9	78.0 ± 19.6	p<0.01
t test	p<0.01	N.S.	N.S.
PEmax (cmH2O)
Pre	68.6 ± 21.2	76.6 ± 18.6	72.8 ± 15.6
Post	73.6 ± 22.2	84.5 ± 22.1	76.9 ± 11.4	N.S.
t test	p<0.05	p<0.05	N.S.

mean ± SD, IMT; inspiratory muscle training, EMT; expiratory muscle training, NC; normal control, PI_max; maximal inspiratory muscle strength, PE_max; maximal expiratory muscle strength, N.S.; not significant.

Changes in pulmonary function

Changes in pulmonary function with training are shown in Table 3. In the NC group, changes were observed in neither VC (pre; 3.3 ± 0.4 l, post; 3.2 ± 0.4 l) nor FEV_1.0 (pre; 2.9 ± 0.3 l, post; 2.9 ± 0.3 l) over the training period. In both trained groups, no changes in pulmonary function were identified compared to the NC group and pre- and post-training.

Table 3. Change of pulmonary function for ventilation muscle training.

	IMT group (n=16)	EMT group (n=16)	NC group (n=8)	ANOVA (interaction)
VC (l)
Pre	3.2 ± 0.4	3.4 ± 0.4	3.3 ± 0.4
Post	3.2 ± 0.4	3.4 ± 0.3	3.2 ± 0.4	N.S.
t test	N.S.	N.S.	N.S.
FEV1.0 (l)
Pre	2.8 ± 0.5	3.0 ± 0.3	2.9 ± 0.3
Post	2.9 ± 0.5	3.1 ± 0.3	2.9 ± 0.3	N.S.
t test	N.S.	N.S.	N.S.

mean ± SD, IMT; inspiratory muscle training, EMT; expiratory muscle training, NC; normal control, VC; vital capacity, FEV_1.0; forced expiratory volume in one second, N.S.; not significant.

Changes in peak values during exercise testing

In the NC group, peak values during progressive exercise testing were unchanged for all parameters: VE (pre; 38.9 ± 2.4 l, post; 39.5 ± 6.8 l); RR (pre; 31.0 ± 5.1 breaths/min, post; 29.8 ± 3.2 breaths/min); VO₂/kg (pre; 25.2 ± 3.4 ml/kg/min, post; 24.5 ± 5.6 ml/kg/min); and RPE (pre; 6.3 ± 2.1, post; 6.4 ± 2.7). In both trained groups, no changes in peak values during exercise testing were observed compared to the NC group and pre- and post-training (Table 4).

Table 4. Change of peak values during exercise testing for ventilation muscle training.

	IMT group (n=16)	EMT group (n=16)	NC group (n=8)	ANOVA (interaction)
Peak VE (l)
Pre	37.3 ± 6.7	43.1 ± 8.9	38.9 ± 2.4
Post	39.2 ± 5.8	44.8 ± 7.0	39.5 ± 6.8	N.S.
t test	N.S.	N.S.	N.S.
Peak TV (l)
Pre	1.3 ± 0.2	1.4 ± 0.3	1.3 ± 0.3
Post	1.4 ± 0.1	1.5 ± 0.3	1.3 ± 0.3	N.S.
t test	N.S.	N.S.	N.S.
Peak RR (breaths/min)
Pre	29.3 ± 4.6	31.2 ± 4.5	31.0 ± 5.1
Post	28.9 ± 4.3	30.8 ± 6.0	29.8 ± 3.2	N.S.
t test	N.S.	N.S.	N.S.
Peak VO2/kg (ml/kg/min)
Pre	25.5 ± 3.7	26.4 ± 4.1	25.2 ± 3.4
Post	25.9 ± 4.2	27.1 ± 3.2	24.5 ± 5.6	N.S.
t test	N.S.	N.S.	N.S.
Peak RPE
Pre	6.8 ± 1.9	7.6 ± 2.4	6.3 ± 2.1
Post	6.1 ± 2.3	6.7 ± 2.0	6.4 ± 2.7	N.S.
t test	N.S.	N.S.	N.S.

mean ± SD, IMT; inspiratory muscle training, EMT; expiratory muscle training, NC; normal control, VE; minute ventilation, TV; tidal volume, RR; respiratory rate, VO₂/kg; oxygen uptake/body weight, RPE; rating of perceived exertion, N.S.; not significant.

Changes in exercise-induced increases for parameters during exercise testing

Changes in exercise-induced increases for each parameter during progressive exercise testing are shown in Figs. 2–4. Since all subjects could reach stage 4, data up to this stage was analyzed. In the IMT group, magnitude of increases in HR decreased from pre- to post-training (df 5, F=2.503, p<0.05). In the other groups, no change in HR increases was identified (Fig. 2). Increases in ventilatory parameters (VE, RR and TV) during exercise were unchanged by training. In the IMT group, the magnitude of exercise-induced increases in VO₂/kg decreased from preto post-training (df 5, F=2.693, p<0.05). In the EMT group, VO₂/kg was lower at all stages in post-training compared with pre-training (df 1, F=5.584, p<0.05). In the NC group, no changes were identified (Fig. 3). Transitions of RPE decreased in both the IMT and EMT groups (df 1, F=4.201, p<0.05 and df 1, F=7.527, p<0.05, respectively) (Fig. 4).

Fig. 2.

Change of heart rate increment during exercise testing for ventilation muscle training.

Fig. 4.

Change of rating of perceptive exertion (RPE) increment during exercise testing for ventilation muscle training. * p<0.05; comparison with increment during exercise testing at pre training.

Fig. 3.

Change of oxygen uptake/body weight increment during exercise testing for ventilation muscle training. * p<0.05; comparison with increment during exercise testing at pre training.

Discussion

Both training methods were found to strengthen ventilatory muscles. In the IMT group, in which both inspiratory and expiratory muscles were strengthened, HR, VO₂/kg and RPE at the same exercise intensity were decreased after training. In the EMT group, in which only expiratory muscles were strengthened, VO₂/kg and RPE at the same exercise intensity were decreased after training, despite no change in HR.

Effect of ventilatory muscle strengthening

Inspiratory muscle training increased both inspiratory and expiratory muscle strength. Studies of patients with respiratory disease (including many with COPD) have variously reported that IMT does not¹²⁾¹³⁾ and does^14–20) strengthen inspiratory muscles. In a meta-analysis of 11 studies, Smith and coworkers³⁾ concluded that inspiratory muscle training does not offer significant treatment effects. However, meta-analysis of only five of the 11 studies controlled for training flow rate found positive training effects. Joint guidelines in the USA¹⁾ have indicated that adequate training loads (i.e., intensity ≥30% of PI_max) are needed acquire treatment effects. In the studies which found the effect of inspiratory muscle strengthening by IMT, patients trained for 30 minutes once¹⁵⁾¹⁶⁾ or 15 minutes twice^14)18)20) daily. Since the meta-analysis by Smith et al., most studies^21–26) have revealed effects of increased inspiratory muscle strength, and a recent metaanalysis⁴⁾ has established this effect. Also in the relative recent studies^24–26), the training setting was prescribed at 30% or more of the maximal inspiratory muscle strength for 30 minutes once or 15 minutes twice daily.

In patients with COPD, no increases in expiratory muscle strength have been found with IMT^12)18)26). A study of healthy subjects by Akiyoshi et al.⁸⁾ reported that no significant change was observed in expiratory muscle strength following IMT for two weeks at 30% of PI_max, although expiratory muscle strength increased by about 18 cmH₂O. Sato and colleagues²⁷⁾ found expiratory muscle strength significantly increased by about 39 cmH₂O after IMT for three weeks at 40% of PI_max. Intercostal interni muscles, which are generally considered muscles of expiration, also function in inspiration during forced ventilation. In the present study, depth expiration reached to approximately RV may have been promoted at introduction of inspiration during training. This effect, which would not be expected in patients with COPD due to increased compensatory participation of muscles assisting ventilatory and pulmonary hyperinflation, could probably appear in healthy individuals.

In our study, only expiratory muscle strength increased after EMT, whereas inspiratory muscle strength was unchanged. Akiyoshi and coworkers⁸⁾ revealed that EMT significantly increases not only expiratory but also inspiratory muscle strength. This may be attributable to the function of abdominal muscles in inspiration, as well as expiration⁸⁾²⁷⁾. EMT did not increase inspiratory muscle strength in studies by Suzuki et al.⁷⁾ and Sato et al.²⁷⁾, in which training was performed at 30% or 40% of PE_max for four weeks, respectively.

Influence of VMT on pulmonary function

Most studies^15)18)19)28) have reported that IMT does not influence pulmonary function in patients with respiratory disease. Meta-analysis³⁾ confirmed this, and subsequent studies^21)23)25) found that pulmonary function was unaffected if inspiratory muscle strength was also increased. These reports correspond to our results.

The effect of EMT on pulmonary function was examined in healthy subjects. In all three studies investigating this effect^7)8)27), EMT leading to increased ventilatory muscle strength exerted no influence on pulmonary function. Again, these results correspond to our data.

Influence of VMT on exercise tolerance and dyspnea

Many studies^{16)18)20–24)} have stated that exercise tolerance and dyspnea during exercise can be improved by increasing ventilatory muscle strength. A recent metaanalysis⁴⁾ concluded that VMT improves dyspnea during exercise and may tend to increase exercise capacity. Many subsequent studies²⁵⁾²⁶⁾ have also demonstrated that increased ventilatory muscle strength improves exercise tolerance. Wanke et al.²³⁾ analyzed expiratory gases during exercise testing, and found that VO_2max during bicycle ergometry increased with increased inspiratory muscle strength. Conversely, most studies^14)15)19)25) have found that peak values of expiratory gas data during exercise testing remain unchanged. Suzuki and coworkers⁷⁾ examined the effects of EMT on exercise capacity. In their report, VO₂ at each stage tended to decrease and Borg scale at each stage was decreased during progressive exercise testing. Findings such as these are basically in accordance with the results of the present study.

In relation to mechanisms behind the effects of increased ventilatory muscle strength on exercise capacity and dyspnea in patients with respiratory disease, Wanke et al.²³⁾ suggested that conditioning of the respiratory muscles might enable patients to tolerate higher PaCO₂ levels, and improved inspiratory muscle performance might increase ventilatory efficacy for physiological dead space/tidal volume (V_D/V_T). Furthermore, Larson et al.¹⁶⁾ suggested that IMT might desensitize patients to fear of dyspnea, while Preusser et al.²²⁾ considered that IMT might improve ventilatory limitation during endurance exercises in patients. However, since these mechanisms do not apply to the subjects of the present study, who were healthy individuals, other factors were considered responsible, such as improved neuromuscular coordination¹⁶⁾, changes to ventilatory patterns²⁴⁾, reduced inspiration time and longer expiration time⁷⁾¹⁶⁾, and delayed ventilatory muscle fatigue²⁰⁾.

In our study, the magnitude of HR increases with increasing exercise intensity reduced following IMT. As this result was likely associated with reduced VO₂/kg increases, exercise load might be assumed to decrease through reduced oxygen demand with improvements in ventilatory efficacy. However, EMT did not influence HR, despite reduced oxygen demand during exercise. As a contrast between the two types of VMT, we consider that strengthening inspiratory muscles might increase intra-thoracic negative pressure during exercise, while stroke volume increases with improved venous return, thus contributing to reduced HR following IMT.

Limitations of the present study

Subjects in this study were selected young healthy women, because we thought that it was easy to recruit them as participants in our study. They were students of a physical and occupational therapy school whose male-to-female ratio of students was about 1:4. In patients with respiratory disease, however, there are more males than females and most of them are elder. Healthy subjects could take effects by training term only for two weeks. But patients with respiratory disease will need the longer training period in order to acquire the training effects. In addition, the effects of VMT on respiratory muscle strengthening may differ by healthy subjects and patients, since ventiratory phase in total lung capacity is not same in both subjects during VMT.

Another limitation is the difference between both subjects of the mechanism to which the effects of ventiratory muscle strengthening affects the responses during exercise testing. Generally, it is unthinkable that a healthy individual's exercise limitation factors are abnormality of artery blood gas, alveolar hypoventilation, fear of dyspnea and so on. We mentioned above, when the factors of change of the responses during exercise testing by VMT might be an improvement of respiratory efficacy. In this study, however, since we could not present data of respiratory pattern and ventiration cycle during exercise, thus was still a matter of conjecture.

We are confident that the further study for respiratory disease patients will prove the result of this study.

Conclusion

We examined the effects of IMT and EMT on ventilatory muscle strength, pulmonary function and responses during exercise testing in young healthy subjects. Both types of training increased ventilatory muscle strength, but did not affect pulmonary function or peak values during exercise testing. However, significant reductions were observed in HR, VO₂/kg and RPE at same load during exercise testing after IMT, and in VO₂/kg and RPE after EMT. Increased ventilation muscle strength might improve ventilatory efficacy during exercise in both IMT and EMT, and increased inspiratory muscle strength might facilitate oxygen delivery through advantageous circulation responses.