Introduction
According to a systematic analysis of the Global Burden of Disease, chronic obstructive pulmonary disease (COPD) is the fourth leading cause of death on the global scale.1 The Global Initiative for Chronic Obstructive Lung Disease guidelines categorize the comprehensive management of COPD into four main objectives: assessment and monitoring of the disease, reduction of risk factors, management of stable COPD, and prevention and management of exacerbations.2 Management plans are not limited to pharmacological interventions, and should be supplemented by proper non-pharmacological treatments, in which exercise interventions are included.3
The respiratory control center, through complex physiological mechanisms, adjusts signal(s) that command the respiratory pump to produce specific alveolar ventilation. The command signal is called, “respiratory drive,” and various outcomes have been used to measure this physiological parameter.4 In particular, since the diaphragm is one of the most significant respiratory muscles, it has been considered that the quantification of its activation can reflect commanding signals to the respiratory pump. Thus, this can be a measure of the respiratory drive. Indeed, diaphragm activation can be accurately measured at rest with the diaphragm electromyogram (EMGdi) using a multipair esophageal electrode positioned at the diaphragm crus.5–7 Through the utilization of this technique, it was found that the amplitude of EMGdi is greater in patients with COPD, when compared to healthy subjects, after this is normalized to each subject’s volitional maximum, that is, when the EMGdi activity is expressed as a percentage of maximum (EMGdi%max).8,9
To the best of the authors’ knowledge, no systematic review has been conducted to assess the effects of exercise on respiratory drive, as evaluated using EMGdi%max, in patients with COPD. The overall accumulation of present scientific evidence has a direct impact on clinical practice. This not only provides conclusive findings regarding the potentially beneficial physiological effect of exercise on respiratory drive, but also further supports the notion that COPD specialists should include exercise in the rehabilitation routine for the management of this disease. Therefore, the present systematic review and meta-analysis aimed to investigate the effects and clinical significance of exercise interventions on respiratory drive (EMGdi%max) in patients with COPD.
Materials and methods
This systematic review was carried out according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) recommendations (Supplemental File 1),10 and was registered on the International Platform of Registered Systematic Review and Meta-analysis Protocols (INPLASY, Registration no.: INPLASY202070021).
Search strategy
The eligibility assessment of the titles, abstracts and full texts of the retrieved articles was performed in a 3-step process, based on the Bettany-Saltikov model.11,12 The literature search was independently performed by two investigators (AD and AP). The Pubmed, PEDro, Science direct, and Cochrane Central Register of Controlled Trials databases were searched from inception up to 25 January 2022. Any inconsistencies in the search procedure were resolved by consensus. The search algorithms used for the selected databases are listed in the Supplementary File 2.
Selection process
The selection process was independently performed by two investigators (AD and PD), and any conflict was resolved by a referee investigator (AP).
Inclusion and exclusion criteria
Studies that met the following criteria were included: (1) human subject studies; (2) the participants were patients with COPD; (3) studies presented in the English language; (4) no limit was set for the time of publication; (5) both acute and chronic interventions were investigated; (6) randomized and non-randomized controlled trials; (7) studies that included exercise interventions, including aerobic and/or resistance exercise, and/or respiratory muscle training. Review articles, and grey literature and trials with pharmacological or surgical, or “mechanical ventilation” intervention were excluded.
Data extraction
Two authors (AD and CC) independently extracted the data, and entered the data in an appropriate table. The extracted data included the first author’s name, year of publication, type of clinical trial, number of participants assessed at the end of the trial, and exercise interventions.
Study quality (risk of bias assessment)
The evaluation of study quality was independently performed by two investigators (AD and CC), and inconsistencies were resolved by consensus. The quality of the included studies was assessed using the PEDro methodological quality scale (Table 1).9,13–26 This 11-item scale evaluates (internal and external) the validity and interpretability of studies, and identifies the potential bias with good reliability.13,27 The score was calculated as the sum of scores for each item, except for the first item. Studies with scores of <4 were considered to be of low methodological quality, studies with scores within 4–6 were considered to be of moderate methodological quality, and studies with scores of ≥7 were considered to be of high methodological quality.28
Table 1Classification of studies using the PEDro scale.
| Study | 1* | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | Total |
|---|
| Langer et al.16 (2018) | – | 1 | 1 | 1 | 1 | 0 | 1 | 1 | 1 | 1 | 1 | 9a |
| Frazão et al.20 (2021) | – | 1 | 0 | 0 | 0 | 0 | 0 | 1 | 1 | 1 | 1 | 5b |
| James et al.19 (2021) | – | 0 | 0 | 1 | 0 | 0 | 0 | 1 | 1 | 1 | 1 | 5b |
| Louvaris et al.21 (2021) | – | 0 | 0 | 1 | 0 | 0 | 0 | 0 | 1 | 1 | 1 | 4b |
| Luo et al.22 (2020) | – | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 1 | 1 | 1 | 4b |
| Faisal et al.9 (2016) | – | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 1 | 1 | 1 | 4b |
| Ciavaglia et al.18 (2014) | – | 1 | 0 | 1 | 0 | 0 | 0 | 0 | 0 | 1 | 1 | 4b |
| Sinderby et al.26 (2001) | – | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 1 | 1 | 1 | 4b |
| Dacha et al.15 (2019) | – | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 1 | 0 | 1 | 3c |
| Wu et al.17 (2017) | – | 0 | 0 | 1 | 0 | 0 | 0 | 0 | 0 | 1 | 1 | 3c |
| Jolley et al.23 (2015) | – | 0 | 0 | 1 | 0 | 0 | 0 | 0 | 0 | 1 | 1 | 3c |
| Luo et al.14 (2011) | – | 0 | 0 | 1 | 0 | 0 | 0 | 0 | 0 | 1 | 1 | 3c |
| Guenette et al.24 (2014) | – | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 1 | 2c |
| Qin et al.25 (2010) | – | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 1 | 2c |
Data analysis
A random-effect model inverse variance continuous meta-analysis was conducted using the RevMan 5.329 software, based on the calculated mean differences. The means and standard deviations (SDs) of EMGdi%max at rest, and during constant intense exercise (≥75% of peak work rate) or incremental exercise to subjective exhaustion were used. In the study conducted by Luo et al.,14 EMGdi%max was calculated, as follows: EMGdi × 100 / EMGdimax. When the data was not available in the tables or text,15,16 this was extracted from the figures using a web digitizer (automeris.io/Web Plot Digitizer), or the corresponding authors were conducted by E-mail to request for the data. The 95% confidence interval (CI) and heterogeneity between studies were calculated using the I2 statistic. A significant result for heterogeneity was considered when p<0.10, and the interpretation of the I2 index was made based on previous guidelines.30 Publication bias was assessed using the asymmetry identification of the funnel plot.30 Trial sequential analysis was performed using the TSA software 0.9.5.10 beta version to determine whether the available sample size for the meta-analysis is optimal to reach a statistical significance.31
Sensitivity analysis
For the acute effect of exercise, the study conducted by Wu et al.17 presented data on the effects of exercise using two different respiratory devices (a respiratory resistance device and a respiratory threshold load device) on EMGdi%max in 12 patients with COPD. Ciavaglia et al.18 examined the effects of incremental cycle and treadmill exercise tests on EMGdi%max in 12 patients with COPD. Similarly, Luo et al.14 investigated the effects of incremental and constant treadmill exercise on EMGdi%max in 16 patients with COPD. Furthermore, Langer et al.16 presented the baseline data on EMGdi activity in two groups of patients, with 10 participants in each group, at rest and during cycling at a constant work rate. These studies had one control group and one intervention group, and the latter underwent inspiratory muscle training (IMT). Each of the above four studies was split into two separate “sub-studies” for the subsequent meta-analysis. Specifically, for the first sub-study conducted by Wu et al.,17 data on the effects of exercise with the respiratory resistance device was included, while for the second sub-study, data on the effects of exercise with the respiratory threshold load device was included. Similarly, for the first sub-study conducted by Ciavaglia et al.,18 data on the effects of incremental cycle exercise was included, while for the second sub-study, data on the effects of incremental treadmill exercise was included. Furthermore, for the first sub-study conducted by Luo et al.,14 data on the effects of incremental treadmill exercise was included, while for the second sub-study, data on the effects of constant treadmill exercise was included. Moreover, for the first sub-group study conducted by Langer et al.,16 the baseline (i.e. before IMT intervention) data for the control group was included, while for the second sub-study, the corresponding data of the intervention group was included. In addition, James et al.19 examined two groups of COPD patients during the symptom-limited incremental cycle exercise test: one group had a normal lower limit resting diffusing capacity for carbon monoxide (DLCO), and the other group had less than the lower limit of normal (DLCO-LLN).
Results
Search results
The search retrieved a total of 196 articles. After removing duplicate records, 169 articles remained, while 148 articles were subsequently excluded based on the titles and abstracts. After screening the full texts, 14 articles satisfied the eligibility criteria, and were included in the systematic review. Furthermore, among these 14 articles, 12 articles were included in the quantitative analysis (meta-analysis). The search process is presented in the PRISMA flow diagram (Fig. 1).
Characteristics of the included studies
The characteristics of studies included in the present review are presented in Table 2.9,14–26 The overall sample size was 238 participants, while the sample size of each study ranged from 7–32 subjects. For the 14 eligible selected studies, one study (approximately 7%) was a randomized control trial (RCT), 12 studies (86%) were cross-sectional studies, and one study (approximately 7%) was a case-control study. All studies included COPD patients, except for one study22 that reported the EMGdi data in µV and presented data on the effects of exercise interventions on EMGdi%max, which was used as an index of respiratory drive.9,14–21,23–26 Furthermore, two of the included studies presented data on both acute and chronic (training) effects,15,16 while the remaining 12 studies examined the acute effects of exercise on EMGdi%max.9,14,18–26 Specifically, one of the 12 studies investigated the acute effects of respiratory training,17 while the other studies investigated the acute effects of cycle or treadmill exercise.
Table 2Characteristics of the included studies
| Study | Type of clinical trial | Number of assessed subjects | Exercise interventions | Main outcomes |
|---|
| Faisal et al.9 (2016) | Cross-sectional study | 16 | Incremental cycle test to subjective exhaustion | EMGdi%max gradually increased without a plateau during incremental exercise |
| Frazão et al.20 (2021) | Cross-sectional study | 12 | Incremental cycle test up to the tolerance limit | Healthy subjects presented higher respiratory neuromuscular efficiency during exercise, when compared to COPD |
| James et al.19 (2021) | Cross-sectional study | 28 | Symptom-limited incremental cycle ergometer test | EMGdi%max was higher during exercise in COPD patients with low resting diffusing capacity for carbon monoxide (DLCO), when compared to COPD patients with normal DLCO |
| Louvaris et al.21 (2021) | Cross-sectional study | 11 | Cycle at 80% WRpeak to symptom limitation | EMGdi%max was similar between exercise and hyperpnoea, with similar ventilatory responses to exercise, but the dyspnoea was higher during exercise due to the impairment of extra diaphragmatic respiratory muscle perfusion |
| Luo et al.22 (2020) | Cross-sectional study | 26 | Incremental cycle test to symptom limitation | EMGdi in inspiratory capacity maneuver gradually increased from rest to the end of exercise |
| Dacha et al.15 (2019) | Cross-sectional/Longitudinal study | 7 | Cycle at 75% WRpeak to symptom limitation; Eight weeks of inspiratory muscle training in four patients | EMGdi%max initially increased and reached a plateau during constant work rate exercise; EMGdi%max decreased after inspiratory muscle training in four patients |
| Langer et al.16 (2018) | Randomized control trial | 20 | Cycling at constant work rate (75% WRpeak) to symptom limitation; Eight weeks of controlled inspiratory muscle training | EMGdi%max significantly decreased before vs. after inspiratory muscle training |
| Wu et al.17 (2017) | Cross-sectional study | 12 | Exercise using respiratory resistance device and respiratory threshold load device | EMGdi%max measured using the respiratory resistance device was significantly lower than the EMGdi%max measured using the respiratory threshold load device in all exercise intensity levels. |
| Jolley et al.23 (2015) | Cross-sectional study | 12 | Incremental cycle and treadmill exercise | Dyspnea intensity during exercise correlated best with EMGdi%max |
| Guenette et al.24 (2014) | Cross-sectional study | 32 | A symptom limited cycle test | Dyspnea intensity during exercise correlated best with EMGdi%max |
| Ciavaglia et al.18 (2014) | Cross-sectional study | 12 | Incremental cycle and treadmill exercise tests | Dyspnea intensity during exercise correlated best with EMGdi%max |
| Luo et al.14 (2011) | Cross-sectional study | 16 | Incremental and constant (80% of maximal oxygen consumption) treadmill exercise. | EMGdi%max initially increased and reached a plateau during constant work rate exercise, while EMGdi%max gradually increased without a plateau during incremental exercise |
| Qin et al.25 (2010) | Case-control study | 24 | Constant work treadmill exercise at 80% of maximal oxygen consumption | EMGdi%max initially increased and reached a plateau during constant work rate exercise |
| Sinderby et al.26 (2001) | Cross-sectional study | 10 | Incremental cycle test to subjective exhaustion | EMGdi%max gradually increased without a plateau during incremental exercise |
Risk of bias of the included studies
The risk of bias of the included studies is presented in Table 1. According to the PEDro methodological quality scale,13 among the 14 studies, six14,15,17,23–25 studies were considered to be of low methodological quality, seven studies9,18–22,26 were considered to be of intermediate methodological quality, and one16 study was considered to be of high methodological quality.
Quantitative data synthesis
For the effects of acute exercise, 12 articles were included in the quantitative analysis.9,14–21,24–26 However, since five of these studies14,16–19 were split into two separate “sub-studies” each (refer to the Sensitivity analysis section), a total of 17 studies were used for the subsequent meta-analysis (Fig. 2). Significant heterogeneity was found (I2 = 89%), and the meta-analysis results revealed a significant difference between EMGdi%max during intense exercise and EMGdi%max at rest, in patients with COPD (p<0.00001, Fig. 2).
Qualitative data synthesis
Chronic effects
Merely two studies examined the training effects of exercise on the main outcome of the systematic review, which was EMGdi%max. Langer et al.16 investigated the effects of eight weeks of IMT on EMGdi%max during constant work rate cycle exercise in COPD patients with activity-related dyspnea. The subjects were randomized into the IMT or sham training (control) group, and it was revealed that EMGdi%max significantly decreased after IMT, when compared to control conditions, and when compared to pre-IMT levels in the exercise training group. Similarly, Dacha et al.15 employed a semi-automated method to analyze the EMGdi data, and reported a reduction in EMGdi%max after eight weeks of IMT in four COPD patients. Interestingly, in both studies, the reduction was accompanied by the improvement in exercise tolerance after the IMT program.
Acute effects of respiratory exercise training
Wu et al.17 compared two different respiratory devices: a respiratory resistance device (PFLEX; Respironics Inc., Pittsburgh, PA, USA), which provides a constant and predetermined inspiratory load preserved during inspiration, and a respiratory threshold load device (Threshold Inspiration Muscle Trainer; Respironics Inc., Pittsburgh, PA, USA), which does not provide a constant inspiratory load to maintain the attainment of inspiratory exercise intensity.32 The present study reported that the EMGdi%max measured using the respiratory resistance device was significantly lower, when compared to the EMGdi%max measured using the respiratory threshold load device, in all exercise intensities (low, moderate and high).
Acute effects of the cycle or treadmill exercise
Thirteen studies9,14–16,18–26 investigated the effects of the cycle or treadmill exercise on EMGdi%max in patients with COPD. It was revealed that EMGdi%max initially increased, and reached a plateau during constant work rate exercise,14,15,25 while EMGdi%max gradually increased without reaching a plateau during incremental exercise.14,23,26 A similar gradual increase in EMGdi data, which was expressed in µV, was recently reported by Luo et al.22 during the cycling incremental exercise to volitional fatigue. Furthermore, Ciavaglia et al.18 reported that EMGdi%max at the peak of the cycle exercise was significantly lower, when compared to EMGdi%max at the peak of the treadmill exercise. Nevertheless, for the given ventilation (VE) during exercise, both exercise modalities had a similar EMGdi%max. In addition, Guenette et al.24 reported that at rest, EMGdi%max was significantly lower in the control group, when compared to the COPD group. Similarly, at the highest equivalent work rate (60 W), EMGdi%max was significantly lower in the control group, when compared to the COPD group. It is noteworthy that the dyspnea intensity during exercise was found to correlate best with EMGdi%max.9,18,23,24 Specifically, Faisal et al.9 examined the effects of incremental cycle exercise on EMGdi%max in patients with COPD and healthy controls, revealing that EMGdi%max was higher in patients with the most severe airflow obstruction and hyperinflation, and that this was attributed to the neuromechanical uncoupling in patients with COPD.
Discussion
The present study conducted a systematic review with meta-analysis to investigate the effects of various exercise interventions on the important functional and clinical outcome in patients with COPD, the neural respiratory drive, which was assessed using EMGdi%max. The analysis revealed that exercise increased the EMGdi%max, when compared to that at rest, in patients with COPD. Furthermore, the increase in EMGdi%max was highly correlated with the intensity of dyspnea during exercise. Specifically, the meta-analysis for the included studies in the quantitative analysis revealed that EMGdi%max was higher during intense exercise, when compared to that at rest (I2 = 89%, Fig. 2), while breathlessness during exercise correlated best with EMGdi%max in patients with COPD.18,23,24 These findings may indicate the value of EMGdi%max as an indirect measure of the key component of respiratory function, which can be utilized as a complementary index of exercise intolerance in COPD patients with different severities. Since the monitoring of EMGdi%max gives a breath-by-breath measure of the load on the respiratory system,33 this can be a useful tool in research, and this may also be used in clinical practice to provide continuous measurements during exercise,14,25,26 given that the use of esophageal catheters needed for this technique has been reported to be acceptable in 95% of patients, and are usually well-tolerated.33
Twelve studies were included in the qualitative analysis of the present systematic review.9,14–21,24–26 Among these, two studies presented data on the chronic effects of exercise intervention on EMGdi%max.15,16 Specifically, these studies revealed that EMGdi%max significantly decreased after IMT, when compared to both the control group and pre-IMT levels,16 or when compared to the pre-IMT values in the small sample of four patients.15 Furthermore, despite the reduction in EMGdi%max, the patients presented with improved exercise tolerance, and were able to exercise longer at a constant load equivalent to 75–80% of the peak work rate. The positive training effect of exercise on the respiratory drive is in accordance with the exercise training-induced changes reported by similar research approaches in patients with chronic heart failure.34 Τherefore, EMGdi%max may be used as a measure to assess the respiratory adaptations of COPD patients to exercise, both at rest and during exercise.
On the other hand, 13 studies9,14–16,18–26 investigated the acute effects of cycle or treadmill exercise on EMGdi%max in patients with COPD. Specifically, it was revealed that EMGdi%max initially increased and reached a plateau during constant work rate exercise,14,15,25 while EMGdi%max gradually increased without reaching a plateau during incremental exercise.14,23,26 The different EMGdi%max responses induced during constant work rate vs. incremental exercise might suggest the use of EMGdi%max as an alternative monitoring index for the ventilatory response of COPD patients to steady-state vs. incremental exercise. Furthermore, it was observed that EMGdi%max was higher in COPD patients with the most severe airflow obstruction and hyperinflation,9 suggesting that EMGdi%max can be an outcome for the evaluation of the severity of COPD.33
In addition, merely one study investigated the acute effects of respiratory exercise on respiratory drive,17 and reported that the EMGdi%max measured when using the respiratory resistance device was significantly lower, when compared to the EMGdi%max measured when using the respiratory threshold load device, in all assessed exercise intensities.
Strengths, limitations and future directions
The present systematic review has limitations. The main limitation is the limited number of existing studies that examined the effects of exercise interventions on EMGdi%max in patients with COPD. Furthermore, six14,15,17,23–25 of the 11 included studies were assessed to be of low methodological quality.13 In addition, the studies analyzed in the present review had a relatively small sample size. Hence, the extrapolation of these findings should be undertaken with caution. Moreover, merely one16 of the included studies was an RCT that investigated the chronic effects of exercise, and another study reported the training exercise effects in only four patients.15 Thus, more RCTs are needed, and future studies should follow the CONSORT guidelines for reporting RCTs,35 in order to improve key methodological features, such as the comprehensive reporting of participant adherence rates, since these are essential for appraising the efficacy of exercise interventions in this clinical population. Another limitation of the present systematic review was that this merely utilized published literature, since the grey literature was excluded. In light of this fact, there was potential publication bias in the present review, and the inclusion of the grey literature could itself introduce bias.30 Furthermore, the investigators were unable to compare the present results with other similar systematic reviews or meta-analyses, since, to the best of our knowledge, this is the first systematic review and meta-analysis that evaluated the effects of exercise on neural respiratory drive (EMGdi%max).
Nevertheless, the present review has some strengths. First, the appropriate algorithm with standardized indexing terms was used for the PubMed database.30 Second, a systematic method was applied to identify articles,10 and a well-established tool13 was used to evaluate the included studies. Overall, more research is warranted on both the acute and chronic effects of exercise interventions on EMGdi%max in patients with COPD. In particular, future high-quality RCTs are necessary to determine the type of exercise and its specific characteristics for optimum effectiveness in EMGdi%max adaptations. This would reduce the dyspnea symptoms, both at rest and during physical activity, and improve the quality of life of patients with COPD. Furthermore, the trial sequential analysis in the present meta-analysis indicated the optimal sample size of the meta-analysis to reach statistical significance (Supplementary File 2, Fig. 2). Finally, publication bias was not detected in the meta-analysis.
Conclusions
The respiratory drive, expressed in EMGi%max, was higher during intense exercise, when compared to at rest, in patients with COPD. This was significantly correlated with the dyspnea intensity during exercise. Furthermore, the eight-week IMT reduced the EMGdi%max and improved the exercise tolerance of patients with activity-related dyspnea. Moreover, different EMGdi%max responses were induced during constant work rate exercise, when compared to incremental exercise, in patients with COPD. However, more studies, especially RCT studies, with better methodological quality should be performed to further investigate the EMGdi activity in response to exercise in COPD patients. The EMGdi%max measured during exercise in this population is a useful clinical tool for health professionals, because this is associated with dyspnea severity, and is sensitive to exercise interventions. Hence, the practical importance of EMGdi%max can be its utilization as a complementary index of exercise intolerance in COPD patients with different severities. This can provide continuous measurements during exercise and be applied as a useful tool in both clinical practice and research.
Supporting information
Supplementary material for this article is available at https://doi.org/10.14218/ERHM.2021.00077 .
Supplemental File 1
PRISMA checklist.
(PDF)
Supplemental File 2
Search algorithms.
(PDF)
Abbreviations
- COPD:
chronic obstructive pulmonary disease
- EMGdi:
diaphragm electromyogram
- EMGdi%max:
percentage of maximum diaphragm electromyogram
- IMT:
inspiratory muscle training
- PEDro:
physiotherapy evidence database
- PRISMA:
preferred reporting items for systematic reviews and meta-analyses
- RCT:
randomized control trial
Declarations
Data sharing statement
The search strategy (algorithms) used in the three databases, the Funnel Plot figure of the meta-analysis, and the trial sequential analysis graph used to support the findings of the study are included in the supplementary information file.
Funding
This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.
Conflict of interest
The authors declare that they have no conflicts of interest.
Authors’ contributions
Conceptualization (AD, AP and SN); Analysis and interpretation of data (AD, PD and CC); Manuscript writing (AD, PD and CC); Critical revision (AP, SN and CC); Statistical analysis (PD, AD and CC); Technical support (PD). All authors have made a significant contribution to the study and have approved the final manuscript.