This study compared the effects of 12 weeks of BTE and IMT interventions on measures of clinical efficacy in middle-aged and elderly individuals with asthma. Our results suggested that IMT improves respiratory muscle strength as reflected by increased PImax. In addition, BTE and IMT resulted in similar positive effects on functional capacity and physical activity. Interestingly, the PImax recorded in middle-aged and elderly patients with stable asthma in this study, was only 72 to 82% predicted. This suggests that there was a need to improve the strength of respiratory muscles in this patient population.
4.1. Primary Outcome
Although the FVC statistically increased in the IMT group, it failed to reach the minimal clinically important difference (MCID) [32
]. Accordingly, the pulmonary function, FVC and FEV1, did not change in both BTE and IMT groups. Previous studies measuring static and dynamic lung volumes in asthmatics pre- and post-IMT have demonstrated controversial results, as some studies have shown increases in FVC and FEV1 [33
] while others observed no changes in FVC or FEV1 [27
]. These differences may depend on duration and type of IMT, as well as participant population. Our results in FVC and FEV1 were consistent with the later studies [27
]. Compared to our study, the former studies used higher training intensity (≤80% PImax vs. our 50–60% PImax) [33
], longer training period (30 min vs. our 10 min) [33
], and duration (6 months vs. our 3 months) [33
], as well as younger participants (40.5 and 21.9 vs. our 55.1 years) [33
In contrast, the impact of IMT on asthma almost uniformly demonstrated an improvement in inspiratory muscle strength (PImax) and endurance [33
] that were in line with our findings. These improvements were observed across a wide spectrum of populations with varying degrees of asthma severity [33
]. Our findings suggest that increases in inspiratory muscle strength may mediate the improvement of functional capacity and physical activity in asthma patients. Moreover, the interventions themselves were also varied. Therefore, despite the fact that inspiratory muscles could adapt differently depending on the techniques and type of demands placed upon them, the variability in methods may not be important as long as the stress on the respiratory muscle is sufficient. This supports the efficacy of IMT as a management strategy for asthma that could reduce the severity of the disease and be a complementary strategy to traditional treatments.
Our results showed that both BTE and IMT increased respiratory muscle strength, PEmax (PImax only in IMT), which is comparable to the results from previous studies investigating different age populations [8
], including BTE programs in elderly asthma patients (>65 years) for 16 weeks [16
], asthmatic adults for 12 weeks [43
], and moderately asthmatic children (6–17 years) for 6 weeks [44
]. Similarly, IMT was reported to cause significant increases in respiratory muscle strength for 6 weeks [36
] and 12 weeks [40
]. A meta-analysis was carried out in favor of increasing PImax during IMT, but PEmax results were controversial [4
]. There are three possible explanations. (1) IMT may elicit subtle improvements in airway diameter and airflow limitation during exercise, thereby reducing dynamic hyperinflation and increasing exercise capacity [45
]. The rising mechanical action on the inspiratory muscles including external intercostals, which affirmatively have an accessory participation in expiration, caused greater thoraco-abdominal mobility, as a result of consequent mechanical reorganization of all of the muscles involved in respiration. Moreover, during IMT and BTE, inspiration and expiration are active throughout the respiratory cycle, fostering muscle function optimization, as evidenced by increased muscle strength [42
]. (2) The inspiratory pressures imposed on the airways may lead to reduced lung hyperinflation, and thus facilitate the action of the inspiratory muscles. Lung hyperinflation, which causes an increase in final expiratory volume, is a characteristic of asthma. It happens because of the premature closing of small-caliber airways that increases the activity of the respiratory muscles at the end of exhalation [8
]. In addition, IMT may offset the functional weakening of the inspiratory muscles that arises due to dynamic hyperinflation. Otherwise, that would lead to the recruitment of high-force-generating highly fatigable accessory muscle fibers, which worsen the development of respiratory muscle fatigue. The reduction of inspiratory muscle fatigue may also have contributed to the improvement in exercise time to the limit of exercise tolerance by delaying the onset of the inspiratory muscle metaboreflex in asthma patients [45
]. (3) Another reason may be the loading intensity. IMT involves resistance training on respiratory muscles while BTE is simply to educate patients with appropriate breathing patterns without resistance loads. IMT has been shown to increase diaphragm thickness and increase the proportion of type I fibers and the size of the type II fibers in the accessory inspiratory muscles. Increases in muscle fiber cross-sectional area could reverse or delay the deterioration of inspiratory muscle function and improve inspiratory muscle economy. Consequently, IMT could stimulate hypertrophy of the diaphragm and accessory inspiratory muscles, facilitating more force for a given level of respiratory muscle drive [8
]. Therefore, the increase in inspiratory muscle strength in the IMT groups in the present study likely occurred because the respiratory mechanics of the patients was favored and a reduction in respiratory work appeared [46
BTE is performed using an appropriate breathing pattern to reduce hyperventilation and hyperinflation, thereby normalizing CO2
levels, which may reduce bronchospasm and breathlessness [3
]. It is related to a positive effect on the effectiveness of air released during expiration. The releasing of air during expiration can be maximized using respiratory muscles correctly [47
]. Moreover, these changes in lung volume seem to relax the smooth muscle in the airway [48
]. In an asthma animal model, the airway resistance is reduced when the animals are ventilated with a greater tidal volume and lower breathing frequency, compared with those animals ventilated with a smaller tidal volume and higher breathing frequency [49
]. However, this hypothesis remains to be confirmed in humans.
The ACT and ACQ were used to investigate asthma control in the subjects. Our results demonstrated that BTE and IMT both had statistically significant effects on asthma control. However, the change in ACT and ACQ score in this study did not reach the respective MCID. The baseline ACT and ACQ scores of participants showed that their asthmatic condition was well controlled prior to the 12-week programs because the ACT scores were all above 19, and ACQ scores were all around 1. The high baseline characteristics in asthma control may indicate a potential ceiling effect, and consequently limit the training effect in this study. In further study, we suggest that recruiting the lower ACT and ACQ participants may be reasonable to expect further improvement. However, our findings showed that both BTE and IMT statistically significantly increase ACT and ACQ scores, which may shed light on the future direction of such studies.
4.2. Secondary Outcomes
Functional capacity was similar between the BTE and IMT groups over the course of the study. In the 6MWT, walking distance, blood oxygen saturation, and heart rate can provide indicators of functional capacity, pulmonary gas exchange, and cardiovascular stress, respectively [50
]. Our results suggested that both BTE and IMT induced similar increases in the 6MWD, which is consistent with the results of Majewski et al. [26
], in which the subjects with bronchial asthma received an 8-week stretching exercise followed by abdominal breathing training. They state that BTE may relieve the uncomfortable symptoms of asthma patients that increased the tolerance for walking during the test [26
]. Contrary to our results, Rondinel et al. [27
] showed that IMT did not improve functional capacity assessed by the 6MWD, possibly because the subjects had good prior exercise tolerance and no aerobic intervention. In addition, no significant differences in blood oxygen saturation or heart rate were found between the two groups in our study, which is consistent with the results of Pereira et al., who found no significant drop in blood oxygen saturation [50
]. Although our data indicated no significant differences in the 6MWD between the two groups, the IMT group increased by 37 m, which tended to surpass the 25-m increase obtained in the BTE group, and also exceed the MCID for 6MWD.
BTE and IMT showed similar results in physical activity. The 3-D PAL was used to determine physical activity, including daily energy expenditure and time spent in moderate- and high-intensity physical activities. Individuals with asthma had lower levels of physical activity than healthy individuals assessed by a physical activity questionnaire (International Physical Activity Questionnaire-Short Form), suggesting that such patients have reduced physical performance stemming from a sedentary lifestyle [51
The retired and employed groups both showed significant differences in 3-D PAL after 12 weeks of BTE and IMT interventions. Specifically, the IMT group tended to show greater increases in their physical activity time than the BTE group did. The adoption of health education programs and professional guidance to improve a patient’s respiratory muscle strength, therefore improve the ventilatory control may be associated with confidence in performing physical activity [48
]. Our findings also suggested that the retired group tended to make more progress in physical activity than the employed group. The results are partially consistent with the previous literature [51
]. A possible explanation may be that the amount of physical activity in the retired group did not meet the standard recommended by the Health Promotion Administration, Ministry of Health and Welfare in Taiwan. On the other hand, subjects in the employed group were more likely to have maintained regular physical activity, so they had already reached the standard of physical activity before the intervention, i.e., they had reached a ceiling effect.
3-D PAL was used to measure the physical activity in older adults previously [54
], and provide reasonable validity compared with an accelerometry-based monitor [56
]. It is also the most used physical activity scale from 1990 to 2009 [57
], and there were no clear trends in the degree to which physical activity measured by self-report and direct measures differ [58
]. Self-report activity logs require participants to record in real time which provides the most detailed data, and can overcome some limitations of questionnaires, such as less susceptible to recall errors, social desirability bias, and measurement bias [59
]. Therefore, we asked the participants to record their physical activity in real time or the next day within the 12-week intervention to eliminate the confounding factor. Moreover, to ensure participants’ adherence, they reported to the laboratory on a monthly basis, and the training diaries were completed by them or their family throughout the intervention. However, we agree that employment might affect the recording of the 3-D PAL in workers. Work-based and leisure time activities mixed may probably dilute the training effect. Nevertheless, the three-day period includes a weekend day and two weekdays, which might balance some occupational activities.
For practical reasons related to the implementation, this study was unable to incorporate a double-blind design of the evaluators and subjects. The participants knew which exercise they performed. Additionally, the non-blind design of the evaluations leads to the possibility of assessment bias. However, the 6MWT was adopted, and walking distance is an objective, observer-independent measure [60
]. Moreover, the FVC, FEV1, PImax, and PEmax were measured with quantitative, objective, and monitoring devices. ACT, ACQ, and 3-D PAL are self-reported measures by participants. Furthermore, IMT increased PImax in this study was supported by the related works [8
]. Confounding factors have been eliminated to the best of our efforts, but there were still some limitations.
The number of samples in this study was small, so the findings may be difficult to generalize to all asthma patients. Larger samples should be collected in the future. In addition, features of pulmonary function testing can be added into the inclusion criteria to reduce the baseline difference in subject characteristics between groups.
With a lack of a no-treatment control group, this study may demonstrate the natural course of the condition, but a more active intervention, such as BTE and IMT, may have led to better long-term outcomes [8
]. However, we are unaware of any studies that perform a no-treatment control group in an asthma study due to ethics of research involving human subjects: the principles of beneficence.
Last, the long-term effects of the training are still unknown. Whether the training had differential effects based on different training times and training intensities remains to be revealed. In future studies, the effects of the intervention can be followed up for a longer time to be validated as clinical indicators.