Effects on Respiratory Pressures, Spirometry Biomarkers, and Sports Performance after Inspiratory Muscle Training in a Physically Active Population by Powerbreath®: A Systematic Review and Meta-Analysis

Simple Summary There is currently a growing interest in respiratory muscle training in athletes, so we set out to systematically assess with meta-analyses the effects of IMT with PowerBretah® (PwB), a threshold work device for IMT, on respiratory parameters and athletic performance in healthy physically active adults. Eleven studies were included in the systematic review and nine in the meta-analysis. IMT by PwB significantly increased maximal inspiratory pressure (MIP) and substantial improvements in forced vital capacity (FVC) in the results of the meta-analysis, and sports performance was significantly increased. In conclusion, the IMT with PwB would improve respiratory, MIP, FVC, and sports performance. Abstract Sports performance in athletes can be limited by respiratory factors, so it is understandable to propose that inspiratory muscle training (IMT) can improve respiratory function and exercise performance. Power-Breathe® (PwB) is a sectorized respiratory muscle training tool that uses a resistive load to train IMT. There is currently a growing interest in respiratory muscle training, so we set out to systematically assess the effects of IMT with PwB on respiratory parameters and athletic performance in physically active, healthy adults. Based on the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guideline, the Cochrane and PEDro scales to assess methodological quality, effect size using the Rosenthal formula, and the Cochrane tool for estimation of risk of bias, studies searchable in Medline, Web of Science, and Cochrane. In addition, for the performance of the meta-analysis, the documentation and quantification of the heterogeneity in each meta-analysis were directed through the Cochran’s Q test and the I2 statistic; in addition, a publication bias analysis was performed using funnel plots. Of the total of 241 studies identified in the search, 11 studies for the systematic review and nine for the meta-analysis met the exclusion and/or inclusion criteria. IMT, with PwB, showed significant improvements in maximal inspiratory pressure (MIP) and substantial improvements in forced vital capacity (FVC) in the meta-analysis results. Also, sports performance was significantly increased by IMT with PwB. In conclusion, the use of PwB is an IMT tool that improves respiratory and sports performance.


Introduction
The inspiratory muscles are histologically and functionally human skeletal muscle fibers and, therefore, are capable of being trained, like any muscle of the musculoskeletal system [1]. Inspiratory muscle training (IMT) is a technique used to increase the strength or endurance of the diaphragm and accessory muscles of inspiration [2]. IMT are nonpharmacological interventions that are low cost, easy to apply, safe, and are considered important adjuncts to the treatment of some lung diseases [3]. IMT has demonstrated improvements in functional capacity, health-related quality of life (HRQoL), pulmonary function, and dyspnea in patients with respiratory conditions [4].
Physical performance in athletes may be limited by respiratory factors such as respiratory muscle dysfunction, exercise-induced hypoxemia, or the initiation of the respiratory metabolic reflex mechanism of the respiratory muscles [5]. It is therefore understandable to propose that IMT can improve exercise performance. The adaptations brought about by a correct training program can influence the energy metabolism of the respiratory muscles, increasing their efficiency and leading to a lower oxygen demand with respect to skeletal muscles [2]. Thus, IMT is a work-breathing procedure that has a possible ergogenic effect on physical exercise in untrained [6] and trained [7] individuals.
IMT devices, which perform sectorized training of the respiratory muscles, can be divided into three categories, such as: resistive charge, voluntary isocapnic hyperpnea, and threshold devices. In this sense, Power-Breathe ® (PwB) [PowerBreathe International Ltd. Southam, Warwickshire; England UK] (IMT Technologies LTD) is a resistive loading IMT instrument. Thus, the IMT effect is generated by the adaptation of the inspiratory muscles to overcome the resistance generated by the PwB during inspiration [8]. A recent metaanalysis reported significant improvements in maximal oxygen consumption (VO 2 max) and non-significant improvements in plasma lactate concentration, but imputed a publication bias with respect to these two parameters [3]. In addition, Fernández-Lázaro et al. [3] demonstrated only an increase in respiratory pressures in athletes after IMT with PwB but not in other parameters of pulmonary function.
Thus, and due to the growing interest in the IMT, we set out to critically assess the effects of the IMT with PWB on respiratory pressures, the mechanics of respiratory system biomarkers, and athletic performance in physically active, healthy adults. The research question was defined using the PICO model according to the standard methods proposed by the Preferred Reporting Items for Systematic Reviews and Meta-Analyses Guidelines (PRISMA) [9] as follows: Population: physically active healthy adults (without any respiratory condition); Intervention: inspiratory muscle training through the PwB device; Comparison: placebo/control group or pre/post comparison data group; Outcomes: respiratory pressures (maximal inspiratory pressure [ . These parameters were included as outcomes as they are commonly investigated in studies of IMT.

Search Strategy
We developed a structured search using Medline (PubMed), Web of Science (WOS), and Scopus for articles published from the database's inception to 31 August 2022, restricted to the English and Spanish languages. The search strategy contained terms related to TT and the different outcome biomarkers as well as a combination of these with the Medical Subject Headings (MeSH) index and Boolean operators:: ("Powerbreath") AND ("muscle" OR "inspiratory muscle training" OR "inspiratory muscle strength") AND ("performance" OR "athletic performance") AND ("exercise" OR "physical activity" OR "aerobic capacity" OR "resistance") ("pulmonary function" OR respiratory parameters") AND random* (inspiratory muscle training AND atere* AND random*). Two reviewers (D.F.-L. and J.S.-C.) independently screened titles and abstracts, and full texts were sourced for relevant articles. Inclusion criteria were independently assessed, and disagreements were resolved by a third reviewer (L.C.). Additional records were obtained through reference lists that included relevant articles.

Selection Criteria
The selection of records was based on the following criteria: (a) physically active, healthy adults with moderate levels of physical activity or sports practice (recreational, amateur, or professional) without any respiratory condition (including individuals with moderate levels of physical activity and excluding animals); (b) studies that assessed the effects of IMT with PwB as the only method of respiratory training; (c) clinical trials, randomized and not randomized trials, and pre-test/post-test design studies (excluding records, editorials, reviews, notes, and any other non-original study); (d) studies that evaluated as outcomes (primary or secondary) any respiratory pressure (MIP), spirometry biomarkers (FVC, MVV, PEF, PIF), and sports performance; (e) studies with clear information on the intervention protocol of IMT with PwB. Records that did not meet the criteria were excluded from this systematic review.

Quality Assessment
We used the critical review form for quantitative studies developed by updated method guidelines for systematic reviews in the Cochrane Collaboration back review group [10] and the PEDro scale for rating the quality of randomized controlled trials [11].

Quantitative Assessment
A statistical analysis was performed to obtain the main descriptive statistics of central tendency (mean) and dispersion (standard deviation) of the methodological variables of the treatment and of the subjects recruited in the articles analyzed. We calculated the effect size of the results obtained with respect to the different study variables using Rosenthal's formula [12].

Risk of Bias Assessment
The Cochrane risk bias assessment tool [13] was used to evaluate the quality of the literature by two reviewers (D.F.-L., J.S.-C.). It is a specific tool to assess the risk of bias in clinical trials. For this, it includes a description and assessment for each item in the risk of bias, which offers us a final evaluation that determines the quality of the article [13]. The eight major sources of bias (sequence generation; sssignment concealment; personal blinding; evaluator blinding; incomplete tracking; data reporting; publication bias; observer bias) were classified into three grades: "Yes" indicates a low risk of bias, "No" indicates a high risk of bias, and "Unclear" indicates a lack of information or uncertainty about the potential for bias, which were assessed using the Cochrane risk of bias assessment tool.

Meta-Analysis Data Analysis
To carry out the meta-analysis, the methodology used in the study by Fernández-Lázaro et al. [3] was followed. In this way, the heterogeneity calculation was quantified with Cochrane's I2 statistic. Without heterogeneity, a fixed effect meta-analysis model was used. However, with heterogeneity, a random effect meta-analysis model was used for evaluating tau squared (τ2) by the DerSimonian-Laird method [14]. Regarding the analysis of publication bias, it was performed using funnel plots, whose asymmetry was quantified through Egger's regression [15]. The process of performing meta-analysis was used with the metaphor package (version 2.1-0) in R (The R Foundation for Statistical Computing, Vienna, Austria).

Data Extraction
Two reviewers (D.F.-L. and J.S.-C.) reviewed and synthesized the data from all selected studies into a comprehensive table using standardized data extraction. Disagreements were resolved by a third reviewer (D.G.G.). Information extracted from the selected studies included: the name of the first author, year of publication, the country in which the study was conducted, sample size, age, height, and sport activity (Table 1). Also, interventions, instruments, outcomes, and results ( Table 2) were included in the review.      Figure 1 details the process of study selection. The literature search resulted in 240 studies, of which 234 studies were included in the WOS, SCOPUS, and PubMed databases and the remaining six studies were from other sources, such as ResearchGate and reference lists of relevant studies. Thirty-nine duplicates were eliminated, and 195 identified articles were evaluated. Thirty-one manuscripts were included as prospective studies after analysis by title/abstract. After this process, the full text was revised to include a total of 11 records [16][17][18][19][20][21][22][23][24][25][26] in the systematic review.

Study Selection
Biology 2023, 12, x FOR PEER REVIEW 9 of 20 Figure 1 details the process of study selection. The literature search resulted in 240 studies, of which 234 studies were included in the WOS, SCOPUS, and PubMed databases and the remaining six studies were from other sources, such as ResearchGate and reference lists of relevant studies. Thirty-nine duplicates were eliminated, and 195 identified articles were evaluated. Thirty-one manuscripts were included as prospective studies after analysis by title/abstract. After this process, the full text was revised to include a total of 11 records [16][17][18][19][20][21][22][23][24][25][26] in the systematic review.  Table 1 shows the main characteristics of the participants included in the selected studies and the athletes' weekly training. The age of the subjects ranged between 18-32 years, and the height range of the participants was 160-182 cm. Five studies included in the review included soccer players [16,18,21,24,23], three studies rugby and basketball players [,21,25,26], one study for each sport: field hockey [21], swimming [20], running [17], and cycling [22]. Table 2 summarizes the contents of the studies contained in this systematic review.

Sports Performance
All studies [16][17][18][19][20][21][22][23][24][25][26] included in this systematic review evaluated sports performance. Significant [16,18,20,21,23,24,26] (p < 0.05) and non-significant [17,19,22,25] (p > 0.05) increases in performance have been reported in IG from baseline to study completion. In addition, significant improvements in sports performance in IG compared to CG have been described in the study conducted by Edwards et al. [17] in athletes. Table 3 shows the analysis of the methodological biases of the studies analyzed in this systematic review by the Cochrane risk bias assessment tool [13]. Two registers presented five biases [17,24], mainly related to allocation concealment, blinding of the personnel, and blinding of the evaluator. Also, three studies presented with four biases [18,19,25] and three studies [20,22,23] presented three biases. Finally, three records showed only two biases [16,21]. Abbreviations = 1: sequence generation; 2: assignment concealment; 3: personal blinding; 4: evaluator blinding; 5: incomplete tracking; 6: data reporting; 7: publication bias; 8: observer bias; the rating for each item includes the response to a question", where "Yes" indicates a low risk "f " bias, "No" indicates a high risk of bias, and "Unclear" indicates a lack of information or uncertainty about the potential for bias; the higher the score, the higher the risk of bias.

Methodological Variables Assessment
The evaluation of the quantitative analysis of the selected studies can be seen in Table 6. Regarding the sample size, experimental deaths in both the experimental and control groups in the different evaluations (pre-treatment and post-treatment) indicate a low number of dropouts during the interventions. As for the experimental mortality data, all the articles provide data after the treatment period, presenting a mean percentage of 2.05% (SD: 4.71) of dropouts due to different causes. However, none of the studies included a follow-up evaluation. Table 7 shows the results of the calculation of the effect size of the selected studies. The effect size offered disparate results depending on the study analyzed. According to this statistical analysis, the treatment had a large effect size (d > 0.8) and thus a greater magnitude of effect. In relation to the improvement of sports performance, Romer et al. [21] (d = 1.81), Tong et al. [24] (d = 0.89), Tong et al. [23] (d = 1.03), and Archiza et al. [16] (d = 1.19) showed a large effect size in their interventions.    For MIP, seven studies showed improvements with a large effect size [16,17,[21][22][23][24][25], with an effect size between 1.56 d and 0.95 d. Regarding lung function (FVC), only the study conducted by Romer et al. [21] showed a high effect size (d = 1.29). Vascocelos et al. [26], Salazar-Martínez et al. [22] and Tranchita et al. [25], who obtained significant improvements in FVC, reported results of moderate magnitude (d > 0.5).

Effect Size Assessment
3.8. Evaluation of the Results of the Studies Included in the Synthesis and Meta-Analysis (n = 9 Included Studies) 3.8.1. Maximal Inspiratory Pressure (n = 8 Included Studies) Figure 2 shows the effect of using the PwB on MIP, a statistically significant increase effect is produced (p < 0.05): ROM 1.24; 95% CI, 1.17 to 1.32; Z = 6.94; p = 4 × 10 −12 for the studies [16][17][18][21][22][23][24][25] analyzed in the meta-analysis. The publication bias analysis (Figure 3) for the MIP presented a relatively symmetric funnel plot, without any imputed studies, which could indicate that there is no publication bias, although this asymmetry was not statistically significant 'value of Egger's p = 0.238). Through the "Trim and fill" method, there was no imputation of studies at levels higher than ES and under standard error.
3.8.2. Forced Vital Capacity (n = 6 Included Studies) Figure 4 shows the effect of using the PwB on FVC, a statistically non-significant increase effect is produced (p > 0.05): ROM 1.02; 95% CI, 1.00 to 1.03; Z = 2.47; p = 0.013 for the studies [16][17][18]21,22,26] analyzed in the meta-analysis. The publication bias analysis ( Figure 5) for the MIP presented a relatively symmetric funnel plot, without 2 imputed studies, which could indicate that there is publication bias, and this asymmetry was not statistically significant 'value of Egger's p = 0.369). Through the "Trim and fill" method, there was imputation of 2 studies at levels higher than ES and under standard error, which could indicate a possible lack of studies at this level.       [16][17][18]21,22,26] analyzed in the meta-analysis. The publication bias analysis ( Figure 5) for the MIP presented a relatively symmetric funnel plot, without 2 imputed studies, which could indicate that there is publication bias, and this asymmetry was not statistically significant 'value of Egger's p = 0.369). Through the "Trim and fill" method, there was imputation of 2 studies at levels higher than ES and under standard error, which could indicate a possible lack of studies at this level.

Discussion
The aim of this systematic review was to critically evaluate the effects of IMT with PWB on respiratory pressures, mechanics of the respiratory system, biomarkers, and sports performance in physically active, healthy adults. Eleven studies met the pre-specified inclusion/exclusion criteria. Overall, subjects who performed IMT with PwB showed significant improvements in MIP and substantial improvements in FVC in the meta-analysis results. In addition, this systematic review found significant improvements in sports performance [16,18,20,21,23,24,26] and other spirometry biomarkers such as peak inspiratory flow [25] and peak expiratory flow [26]. The energetic commitment of the respiratory muscles (RMs) with respect to the skeletal muscles that develop active physical activity and the fatigue of the RMs are the two main factors that restrict the respiratory function, causing a decrease in respiratory performance [27]. RMs demand a high percentage of cardiac output (15-20%) in situations of vigorous physical activity, which requires a reduction in the availability of oxygen (O 2 ) for the skeletal muscle involved in the exercise and therefore produces a notable decrease in performance in these athletes [28,29]. Thus, the high O 2 needs of RMs during intense exercise conditions generate a competitive demand with respect to active skeletal muscles, which see a significant decrease in the supply of oxygen to their cells, and this causes a decrease in sports performance [30].
During intense and/or prolonged exercise, the appearance of RMs fatigue is possible, following the depletion of its energy substrates [31] and the response of the sympathetic nervous system that activates the respiratory metabolic reflex [3]. The respiratory metabolic reflex involves vasoconstriction that triggers a decrease in blood flow and an increase in the severity of skeletal muscle fatigue, induced by flow and redistributed to preserve respiratory function without compromising RM energy demand [32]. Furthermore, significant decreases in MIP and maximum expiratory pressure (PEM) have been described after long-term and/or intense aerobic physical exercise [33]. However, training the respiratory musculature, particularly IMT, which completes the athlete's training, attenuates the competition for blood flow, improves tolerance to respiratory fatigue, and increases the efficiency of RMs, all of which have a positive influence on athletic performance [3,34,35].
Our findings indicating benefits in MIP, FVC, and sport performance could be a consequence of adaptations induced by proper IMT program implementation. Especially important could be the improvements in the MIP determined in the meta-analysis, which are also consistent with those reported in other studies [7,[36][37][38][39] and in a recent metaanalysis of IMT that found that administering a resistive load of 15% of the MIP achieved significant improvements although they did not reach the pre-established threshold of statistical significance [3]. The substantial improvements in FVC reported in our metaanalysis in athletes are relatively surprising because, in athletes prior to IMT, lung function is close to supraphysiological limits, that is, the values of the main pulmonary ventilation parameters are ≥100% of physiological level [40]. IMT usually induces improvements in FVC in adults with limited lung function [41,42], due to the state of weakness of their respiratory muscles and decreased lung volumes [43,44]. In this sense, IMT in patients with heart failure reduces the magnitude of the respiratory metabolic reflex, which could prevent functional deterioration and MR atrophy and therefore significantly improve lung function [45].
These adaptations to IMT try to modulate homeostatic processes [46] that lead to changes in the oxidative energy metabolism of RMs, gaining in efficiency and causing a lower oxygen demand with respect to skeletal muscles [28,29], tissue remodeling of RMs (hypertrophy of the diaphragm, increase of type II fibers) that increases their strength and functionality [34], and the delay of the respiratory metabolic reflex [3]. Also, PIM s improvements could also indicate optimization of the neuro-motor control of the respiratory musculature, maintaining the generation of pressure with a lower motor impulse and greater economy of the respiratory musculature [23]. The achievement of these biological adaptations, which involve totally different mechanisms, would be responsible for improvements in sports performance [47,48]. Significant improvements in IG from baseline and compared to CG athletes [17] have been found in the studies included in the systematic review of different sports modalities as soccer [16,18,21,23,24], swimming [20], rugby [21,23,24], basketball [21], and field hockey [21].
In addition to serving as adjuvant therapy in pathological states [41,42,45,49], IMT could be used in populations that perform jobs with high physical demands, such as the military, emergency services, or high mountain rescuers, who usually carry backpacks and heavy loads [50]. In addition, these populations usually use masks or respiratory devices, thus increasing respiratory work, which would induce early muscle fatigue and a reduction in work tolerance [51].
Evidence presented in this systematic review and meta-analysis suggests that IMT by PwB device is safe and provides significant improvements in MIP and substantial improvements in FVC and sports performance. The mechanisms of improvement of respiratory biomarkers by IMT could have a multifactorial etiology, mainly the attenuation of the respiratory metabolic reflex and the modulation of respiratory muscle fatigue. However, these results on IMT with PwB in physically active, healthy adults without chronic diseases need to be confirmed.

Conflicts of Interest:
The authors declare no conflict of interest.