respiratory muscle endurance after respiratory muscle ... · literature review respiratory muscle...

Literature review

Respiratory muscle endurance after respiratory muscle training inathletes and non-athletes: A systematic review and meta-analysisQ6

Q5 Ana Tereza do N. Sales a, *, Guilherme A. de F. Fregonezi a, Andrew H. Ramsook c,Jordan A. Guenette c, Illia Nadinne D.F. Lima a, W. Darlene Reid b

a Department of Physical Therapy, University Federal of the Rio Grande do Norte, Natal, Rio Grande do Norte, Brazilb Department of Physical Therapy, University of Toronto, Toronto, Ontario, Canadac Centre for Heart Lung Innovation, University of British Columbia and St. Paul's Hospital, Vancouver, British Columbia, Canada

a r t i c l e i n f o

Article history:Received 5 December 2014Received in revised form3 July 2015Accepted 7 August 2015

Keywords:Breathing exercisesMaximal voluntary ventilationRespiratory musclesPulmonary function test

a b s t r a c t

The objectives of this systematic review was to evaluate the effects of respiratory muscle training (RMT)on respiratory muscle endurance (RME) and to determine the RME test that demonstrates the mostconsistent changes after RMT. Electronic searches were conducted in EMBASE, MEDLINE, COCHRANECENTRAL, CINHAL and SPORTDiscus. The PEDro scale was used for quality assessment and meta-analysiswere performed to compare effect sizes of different RME tests. Twenty studies met the inclusion criteria.Isocapnic hyperpnea training was performed in 40% of the studies. Meta-analysis showed that RMTimproves RME in athletes (P ¼ 0.0007) and non-athletes (P ¼ 0.001). Subgroup analysis showed dif-ferences among tests; maximal sustainable ventilatory capacity (MSVC) and maximal sustainablethreshold loading tests demonstrated significant improvement after RMT (P ¼ 0.007; P ¼ 0.003respectively) compared to the maximal voluntary ventilation (MVV) (P ¼ 0.11) in athletes whereassignificant improvement after RMT was only shown by MSVC in non-athletes. The effect size of MSVCwas greater compared to MVV in studies that performed both tests. The meta-analysis results provideevidence that RMT improves RME in athletes and non-athletes and tests that examine endurance overseveral minutes (eg. MSVC) are more sensitive to improvement after RMT than the shorter MVV.

© 2015 Published by Elsevier Ltd.

1. Introduction

Respiratory muscle endurance (RME) has been evaluated overthe years in a wide range of patient populations such as spinal cordinjury (Silva, Neder & Chiurciu, 2000) chronic obstructive pulmo-nary disease (COPD) (Dias et al., 2013; Weiner et al., 2003), myas-thenia gravis (Rassler et al., 2011), as well as in healthy individuals(Bell et al., 2013; Johnson, Cowley, & Kinnear, 1997; Johnson,Sharpe, & Brown, 2007; Spengler et al., 1999). This evaluation hasbeen used as an outcome measurement following different treat-ment interventions and to determine normative values for respi-ratory muscle performance (Fischer et al., 2014; Fiz et al., 1998;Kroff & Terblanche, 2010; Verges, Boutellier, & Spengler, 2008).However, a variety of tests have been used to evaluate RME with

some studies reporting conflicting results. This difference ispartially attributable to very diverse outcomes and their respectiveinterpretations. Hence, it is essential to standardize RME tests inorder to derive more ready comparisons.

Leith and Bradley (1976) evaluated RME by examining the timeto exhaustion during a partial rebreathing method and termed thismaneuver the “sustained ventilatory capacity”. In a differentapproach, Nickerson and Keens (1982) developed a test thatrequired sustaining a threshold inspiratory pressure termed themaximal sustainable threshold loading test. The measure requiredless apparatus and did not require high flow rates, which facilitatedcomparisons between those with airflow limitation versus healthysubjects. This test was later modified to an incremental thresholdtest devised by McElvaney et al. (1989) whereby subjects started ata low load and weights were added at two-minute intervals untiltask failure. Some investigators also used the maximal voluntaryventilation (MVV) as a RME measure; however, it is questionablewhether such a short duration test is reflective of RME (ATS, 2002;Driller & Panton, 2012; Freedman, 1970).

* Corresponding author. Departamento de Fisioterapia, Universidade Federal doRio Grande do Norte, Campus Universit!ario, Lagoa Nova, Caixa Postal 1524, CEP:59072-970, Natal/RN, Brazil. Tel.: þ55 84 3342 2027.

E-mail address: [email protected] (A.T.N. Sales).

Contents lists available at ScienceDirect

Physical Therapy in Sport

journal homepage: www.elsevier .com/ptsp

http://dx.doi.org/10.1016/j.ptsp.2015.08.0011466-853X/© 2015 Published by Elsevier Ltd.

Physical Therapy in Sport xxx (2015) 1e11

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354

555657585960616263646566676869707172737475767778798081828384858687888990919293949596979899

100101102103104105106107108109110111112113114115116117118119

YPTSP676_proof ■ 17 August 2015 ■ 1/11

Please cite this article in press as: Sales, A. T. N., et al., Respiratory muscle endurance after respiratory muscle training in athletes and non-athletes: A systematic review and meta-analysis, Physical Therapy in Sport (2015), http://dx.doi.org/10.1016/j.ptsp.2015.08.001

Closer scrutiny reveals that the assortment of tests used toevaluate RME require different metabolic demands (Bradley &Leith, 1978; Nickerson & Keens, 1982), distinctive recruitmentpatterns of motor units (ATS, 2002), and diverse activation of syn-ergistic muscle groups. The maximal sustained ventilator capacity(MSVC) evaluates the endurance of inspiratory and expiratorymuscles over several minutes and requires relatively low pressuresand high speeds of muscle shortening. Because of these demands, itmimics ventilatory requirements during aerobic whole body exer-cise in healthy people. The MVV, used in several RMT clinical trialsis a ventilatory sprint that requires high velocity, unloaded inspi-ratory and expiratory muscle contractions for a very short duration(i.e., 12 or 15 s). Due to the shorter duration of the MVV, its energydemands are likely met by intramuscular ATP and PCr rather thananaerobic and aerobic energy sources required for longer durationtests that last several minutes (McArdle, Katch & Katch, 2001).Alternatively, some endurance tests apply external loads such asthe maximal sustainable threshold loading and maximal incre-mental threshold loading tests; these require higher pressures andlower velocities of shortening than the MSVC (ATS, 2002). Thedifferences in the tests durations, amount of loading and contrac-tion velocities emphasize the importance of task specificity, suchthat the RME test chosen should consider the population beingstudied and the ventilatory demands of the physical activity tar-geted. Ideally, the RME test and RMT should match the ventilatoryand muscle recruitment demands of the target exercise or physicalactivity for optimal benefits.

Besides the different demands of the RME tests, equipment canalso vary considerably. Non-standardized devices are used tomeasure endurance and ensure isocapnia. Over time, differentdevices were constructed to measure test variables such as gasconcentrations, flow rates, pressures and minute ventilation(Bradley& Leith,1978; Fairbarn et al., 1991; Holm, Sattler,& Fregosi,2004; Leith & Bradley, 1976; Nickerson & Keens, 1982). TheAmerican Thoracic Society (ATS) (2002) highlights the lack ofstandardized equipment for the evaluation of RME; however, theylist several essential issues to ensure their reliability of thesedevices.

Unremitting interest in RMT continues because evidence pointsto a link between respiratory muscle fatigue and performanceduring exercise and rehabilitation (Enright & Unnithan, 2011;Gething, Williams, & Davies, 2004; HajGhanbari et al., 2013;Mancini et al., 1994; McConnell & Romer, 2004; McMahon et al.,2002; Vogiatzis & Zakynthinos, 2012). Thus, major outcomesevaluating the effects of RMTare RME and sports performance (Bellet al., 2013; Enright & Unnithan, 2011; Gething et al., 2004).However, whether RMT affects exercise performance is equivocal(McConnell & Romer, 2004). HajGhanbari et al. (2013) concluded,after analyzing 21 articles in their systematic review, that RMT canimprove sports performance and respiratorymuscle endurance andstrength. The improvement in a particular sport, however, maydepend on how closely RMT matches the specificity of training foreach sport category. Thus, standardizing techniques for the evalu-ation of RME is crucial to determine accuracy and comparability ofresults obtained in clinical trials.

Several types of endurance tests have been performed todetermine the efficacy of RMT in healthy individuals, athletes andclinical populations. The considerable differences in recruitmentpatterns, energy/metabolic systems, and apparatuses associatedwith these tests may contribute to the disparity of study outcomesthat evaluate RME after RMT. Accordingly, the purpose of thissystematic review was to assess: a) the effects of RMT on RME, b)the type of RME test that can show the most consistent changesafter RMT in athletes and non-athletes; c) the agreement of RMEtest methodology to that recommended by ATS (2002) and d) if the

type of RMT influences RME test outcomes in athletes and non-athletes.

2. Methods

2.1. Search strategy

Electronic searches were performed on the following databases:MEDLINE, EMBASE, CINAHL, Cochrane Central Register ofcontrolled trials and SPORTdiscus from their inception to February7, 2014. Two concepts were combined for the search strategy (usingthe Boolean operator “and”): (1) RME test used the terms “endur-ance”, “respiratory muscles”, “training”, “test”, or “maximalvoluntary ventilation”; (2) RMT used the terms “training”, “inspi-ratory”, “expiratory”, “ventilatory”, “hyperpnea”, “endurance”,“strength”, “respiratory muscles”, “Powerlung”, “Spirotiger” or“Powerbreathe”. Different terms within each of these conceptswere combined with the Boolean operator “or”. The terms weremodified according to the requirements of each database.

2.2. Study criteria and data abstraction

Studies were included if: a) participants were athletes (elite orrecreational levels) or non-athletes, healthy; and young adults asdefined by #18 and $46 years old; b) any type of RMT (inspiratory,expiratory or both) was performed; c) it was a randomizedcontrolled trial (RCT) design; d) RME measures were reported preand post training; e) full text was available and f) the articles werepublished in English. The first author (ATNS) reviewed all titles andabstracts to determine whether they met the inclusion criteria. Asecond reviewer (AR, JAG, WDR) independently reviewed the ci-tations to determine consensus on their inclusion. Any disagree-ments were discussed with one reviewer (AR) to determine theirinclusion. Full text articles were retrieved for review if articlesshowed potential for inclusion or if there was insufficient infor-mation in the abstract and title to make a decision about inclusion.Inclusion of full-text articles was determined independently by tworeviewers and any disagreements were discussed until consensuswas reached. The flow chart outlining the search strategy, screeningand included articles is presented in Fig. 1.

Data abstraction was performed by two reviewers using stan-dardized forms that included information about study citation,description of participants, inclusion and exclusion criteria, studydesign, descriptions of interventions, and outcomes of RME mea-sures. When datawas missing, authors were contacted (Holm et al.,2004; Markov et al., 2001). However, data was not subsequentlyprovided.

2.3. Definitions

The classification of the athletes was based on the authors'report and by VO2max value according to Wilmore and Costill(2005) classification, although some studies did not reportVO2max values. “Non-athletes” were defined as able-bodied peo-ple, with no injuries and without chronic disease that were notinvolved in sports at recreational or elite levels.

RME tests were defined as follows: a) MVV is the volume of airexpired during 12 or 15 s (Silva et al., 1998); b) MSVC is themaximal ventilation sustained while isocapnea maintained andthe test duration is usually 8e15 min (Dias et al., 2013); c)Maximal sustainable threshold loading test is the submaximalthreshold load sustained as long as possible or the maximal loadendured for a finite amount of time (Dias et al., 2013;Mickleborough et al., 2010; Vogiatzis & Zakynthinos, 2012) and;d) Maximal incremental threshold loading test is a progressive test

A.T.N. Sales et al. / Physical Therapy in Sport xxx (2015) 1e112

1234567891011121314151617181920212223242526272829303132333435363738394041424344454647484950515253545556575859606162636465

66676869707172737475767778798081828384858687888990919293949596979899

100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130



whereby the subject begins at a low threshold load and thethreshold load is increased every two or three minutes until theparticipant is no longer able to inspire against the load (ATS, 2002;McElvaney et al., 1989).

2.4. Quality assessment

Methodological quality assessment was performed by twoindependent reviewers using the PEDro scale (HajGhanbari et al.,2013; Maher et al., 2003). The PEDro scale consists of 11 itemsrelated to: eligibility criteria, random allocation, concealed allo-cation, presence of follow-up, baseline comparability, blindedsubjects, blinded therapists, blinded assessors, intention to treat,analysis between groups, and both point and variability mea-sures. The maximum score is 10 rather than 11 points becausethe first item (evaluates eligibility) relates to external validityand is excluded from the total score (Maher et al., 2003). Thestudy was considered to be of high quality if the PEDro score washigher than 5; moderate quality if the score was 5 or 4, and oflow quality if the score was 3 or lower (Roig, Shadgan, & Reid,2008)Q1 . In addition to the PEDro scale, RME methodology wasevaluated using a checklist based on ATS recommendations ofprocedural issues and device components for RME tests (ATS,2002).

2.5. Statistical analysis

The characteristics of the participants, RMT intervention, andRME tests were described with narrative and summary statistics.When data was available, meta-analyses were performed withReview Manager Version 5.2 (Copenhagen, The Nordic CochraneCentre, The Cochrane Collaboration) in order to compare effectsizes of RMT on different RME tests in athletes and non-athletes.For these comparisons, subgroup analyses were performed onstudies that utilized (a) MVV; (b) MSVC test; (c) maximal sus-tainable threshold loading test and (d) maximal incrementalthreshold loading test. A second set of meta-analyses were per-formed to compare effect sizes of each RME test with subgroupanalyses according to different types of training: (a) thresholdRMT; (b) isocapnic hyperpnea (IH) RMT; (c) targeted resistive RMT.Due to the variety of testing methods, pre- and post-training RMEtest means and standard deviations were normalized by convert-ing values to percentages using baseline data as 100%. The com-parisons groups used in meta-analyses were RMT groups versuscontrol/sham groups. If a study had two groups that performedRMT that met the inclusion criteria, both groups were comparedto the control/sham group(s) and listed separately in the meta-analyses (Enright & Unnithan, 2011; Verges, Renggli, et al., 2009;Wylegala et al., 2007).

Fig. 1. Flow chart of the search strategy.

A.T.N. Sales et al. / Physical Therapy in Sport xxx (2015) 1e11 3

1234567891011121314151617181920212223242526272829303132333435363738394041424344454647484950515253545556575859606162636465

66676869707172737475767778798081828384858687888990919293949596979899

100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130



The results were analyzed as a continuous outcome using arandom effects model to calculate a weighted standardized meandifference and 95% confidence intervals (CI). A P value $ 0.05 wasconsidered as statistically significant for an overall effect (Z value).Heterogeneity was investigated using the I-squared test and a Pvalue $ 0.10 indicated statistical significance.

3. Results

3.1. Study selection

The search strategy produced 6346 citations (Fig. 1). Sixty-fivecitations were retrieved for full test screening; after review, 20articles met the inclusion criteria. The main reasons for exclusionswere: a) articles did not meet inclusion criteria such as age range,participants' characteristics or language of publication; b) articleswere not RCTs; c) article summarized data from other articles; d)the methods did not include a RME test and; e) the results did notprovide sufficient data from the RME test.

3.2. Quality assessment

The mean score of the PEDro scale for RCTs performed in ath-letes was 6 and ranged from 4 to 9, with the highest score obtainedby Romer, McConnell, and Jones (2002) (Table 1). For studies withnon-athletes, the mean score was 5.6 and ranged from 5 to 7, withthe highest score achieved by Verges et al. (2008) (Table 1). Theitems most often not performed or omitted in reporting were: the

allocation was not concealed (18 studies) and the lacking ofblinding of participants (18 studies), investigators (19 studies), andassessors (18 studies).

3.3. Characteristics of participants

The characteristics of the participants are described in Table 2.The total number of participants in the studies were 256 athletesand was 192 non-athletes. Athletes ranged in age from 18 to 34years old and non-athletes ranged in age from 19 to 43 years. Ninestudies included only men and 1 study included only women. Theathletic level ranged from recreational athletes to elite athletes.Four studies that involved athletes did not report the VO2 maxvalues. Nine of 20 articles reported values of MIP at baseline.

3.4. Characteristics of interventions

The characteristics of the interventions including the trainingdevices are presented in Table 3. IH was the most common type oftraining among the studies (40% of reports). One study performedcombined training (threshold combined with IH) in the RMT groupand two studies performed threshold and IH RMT in separategroups while five studies used threshold training applied onlyduring inspiration. Eight studies that performed IH training usedcustom devices. Except for 3 studies that did not report informationconcerning the training progression, all remaining studies includedin analysis reported that the intensity of training graduallyincreased for the duration of the study. The duration of the trainingperiod ranged from 3 to 15 weeks. Eight articles described sets andrepetitions for training, and 13 reports described a training timethat ranged from 28 to 30 min per session. With the exception of 6studies (4 studies with non-athletes), the others supervised all orsome of the training sessions.

3.5. Characteristics of RME test technique

The characteristics of RME test techniques are described inTable 4. Methodology varied considerably between the studiesanalyzed; however, the most commonly performed test was theMSVC (10 studies). MVV was performed in 5 studies, 1 study used amaximal incremental threshold loading test and 4 performedmaximal sustainable threshold loading tests. Custom devices wereused for all of the tests that evaluated the MSVC, maximal sus-tainable threshold loading and maximal incremental thresholdloading. The studies that performed the MVV used a spirometer fortest execution according to ATS Guidelines (ATS, 1995).

3.6. Meta-analyses

Meta-analyses examining the effects of RMT in athletes revealedimprovements in RME in favor of the training group as shown bythe overall effect in Fig. 2 (Z ¼ 3.39, P ¼ 0.0007) and a significantdifferences among tests (P ¼ 0.04); MSVC and the maximum sus-tainable threshold loading test demonstrated greater changes afterRMT (Z ¼ 2.71, P ¼ 0.007; Z ¼ 2.93, P ¼ 0.003, respectively)compared to the MVV (Z ¼ 1.58, P ¼ 0.11). Of 5 studies that usedboth MVV and MSVC (Leddy et al., 2007; McMahon et al., 2002;Morgan et al., 1987; Sonetti et al., 2001; Wylegala et al., 2007),three showed greater improvements in the longer duration test, theMSVC, compared to the MVV. In the study where the MSTL and theMVVwere performed (Mickleborough et al., 2010), again the longermaximum sustained threshold loading test showed greaterimprovement that the changes in the MVV test.

Meta-analysis demonstrated an overall improvement in RMEafter RMT compared to the control group in non-athletes (Fig. 2;

Table 1Description of quality assessment with PEDro scale.

PEDro scores

First author, year 1 2 3 4 5 6 7 8 9 10 11 Total

AthletesForbes, 2011 Yes 1 1 1 1 1 1 6Holm, 2004 Yes 1 1 1 1 1 5Inbar, 2000 Yes 1 1 1 1 1 1 6Leddy, 2007 Yes 1 1 1 1 1 5McMahon, 2002 Yes 1 1 1 1 1 1 1 7Mickleborough,

2008Yes 1 1 1 1 1 1 6

Mickleborough,2010

Yes 1 1 1 1 1 1 6

Morgan, 1987 Yes 1 1 1 1 4Riganas, 2008 Yes 1 1 1 1 1 1 6Romer, 2002 Yes 1 1 1 1 1 1 1 1 1 9Sonetti, 2001 Yes 1 1 1 1 1 1 6Wylegala, 2007 Yes 1 1 1 1 1 1 6Non-athletesEnright, 2011 Yes 1 1 1 1 1 1 6Gething, 2004 Yes 1 1 1 1 1 5Keramidas, 2010 Yes 1 1 1 1 1 5Markov, 1996 Yes 1 1 1 1 1 1 6Markov, 2001 Yes 1 1 1 1 1 1 6Stuessi, 2001 Yes 1 1 1 1 1 1 6Suzuki, 1993 Yes 1 1 1 1 1 5Verges, 2009 Yes 1 1 1 1 1 1 1 7Total 20 2 20 2 1 2 18 16 17 20 118

1 ¼ eligibility criteria were specified; 2 ¼ subjects were randomly allocated togroups; 3 ¼ allocation was concealed; 4 ¼ the groups were similar at baselineregarding the most important prognostic indicators; 5 ¼ blinding of all subjects;6 ¼ blinding of all therapists who administered the therapy; 7 ¼ blinding of allassessors whomeasured at least 1 key outcome; 8¼measure of 1 key outcomewereobtained from >85% of subjects initially allocated to groups; 9 ¼ all subjects forwhom outcome measures were available received the treatment or control condi-tions as allocated or, where this was not the case, data from at least 1 key outcomewas analyzed by “intention to treat”; 10 ¼ the results of between-group statisticalcomparisons are reported for at least 1 key outcome; 11 ¼ the study provide bothpoint measures and measures of variability for at least 1 key outcome.


1234567891011121314151617181920212223242526272829303132333435363738394041424344454647484950515253545556575859606162636465

66676869707172737475767778798081828384858687888990919293949596979899

100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130



Z ¼ 3.28, P ¼ 0.001) although subgroups were not significantlydifferent (P ¼ 0.21). Among the tests, however, the MSVC andmaximal sustainable threshold loading tests showed significantimprovements in RME after RMT (Z ¼ 2.14, P ¼ 0.03 and Z ¼ 2.20,P ¼ 0.03, respectively) when compared with the MVV (P ¼ 0.29)and MITL (0.06) tests. Two studies (Markov, Orler, & Boutellier,1996; Stuessi et al., 2001) performed both MSVC and MVV testsin their evaluations and the effect sizes of MSVC were greatercompared to the MVV data (Fig. 2).

Forest plots in Fig. 3 showed the meta-analyses that examinedthe RME tests according to the type of RMT subgroups. Examinationof the overall effect sizes, showed that MVV just met significancefor all types of training (Z ¼ 1.94, P ¼ 0.05; mean and 95% CI: 0.31[0.00, 0.62] whereas the overall effect sizes of MSVC and maximalsustainable threshold loading reached greater significance(Z ¼ 3.61, P ¼ 0.0003 and Z ¼ 3.20, P ¼ 0.001, respectively). None ofthe RMT subgroups reached significance for the MVV test (upperpanel). In contrast, the type of training subgroup analysis for theMSVC test showed differences between the types of training(Z ¼ 3.61, P ¼ 0.0003). IH training showed significant effect sizes

(Z ¼ 3.77, P ¼ 0.0002) whereas threshold training combined IH didnot (Z ¼ 0.05, P ¼ 0.96).

4. Discussion

This systematic review, a synthesis of 20 articles and 448 par-ticipants, confirms that RMT improves RME in athletes and non-athletes. This improvement was demonstrated when RME wasevaluated using longer tests such as the MSVC and maximal sus-tainable threshold loading tests in athletes and in non-athletescompared to the MVV test, which exhibited non-significant over-all effects in these two groups of individuals. In addition, IH traininginduced greater improvements in RME compared with thresholdtraining and target resistive training.

The overall quality of the included articles was good as shownby a mean PEDro score of 6, similar to that found by HajGhanbariet al. (2013) in their systematic review of RMT in athletes. Illiet al. (2012) performed quality assessment of studies examiningthe effects of RMT on exercise performance in healthy individuals,and their scores ranged between 2 and 7 out of 7 points. Most

Table 2Characteristics of participants.

First author, year N (M/F) Comparison groups (N) Age mean ± SD *VO2 max (ml kg%1 min%1) Type of sport *MIP baseline (cmH2O)

AthletesForbes, 2011 21 (9/12) RMT (12); sham (9) M: 23 ± 11

F: 22 ± 9NR Rowing Reported in graph

Holm, 2004 20 (16/4) RMT (10); sham (4) Con (6) 28 ± 7 RMT: 54.0 ± 4.6Con/Sham: 56.8 ± 3.0

Cyclist and/orTriathlete

NR

Inbar, 2000 20 (20/0) RMT (10); Con (10) 29 ± 9 RMT: 58.0 ± 4.6Sham: 61.2 ± 4.7

Enduranceathletes

NR

Leddy, 2007 22 (22/0) RMT (15); Con (7) T: 29 ± 8Con:34 ± 6

RMT: 56.4 ± 6.7Sham: 52.0 ± 2.7

Runners NR

McMahon, 2002 20 (20/0) RMT (10); Con (10) T: 26 ± 4Con: 28 ± 6

RMT: 73.6 ± 15.0Con: 70.1 ± 12.6

Cyclist NR

Mickleborough,2008

30 (15/15) ST þ RMT (10); ST þSham (10); Con (10)

18 ± 2 NR Swimmers 174.4 ± 23.2

Mickleborough,2010

24 (12/12) RMT (8); sham (8) Con (8) 22 ± 2 NR Road runners 128.9 ± 17.9

Morgan, 1987 9 (9/0) RMT (4); Con (5) T: 24 ± 1Con: 25 ± 0

RMT: 3.89 ± 0.52 L min%1

Con: 3.90 ± 2.79 L min %1Cyclist NR

Riganas, 2008 19 (12/7) RMT (11); Con (8) T: 22 ± 5Con: 20 ± 2

RMT: 51.8 ± 6.5Con: 51.0 ± 6.4

Rowers RMT z139 Con z 125

Romer, 2002 24 (24/0) RMT (12); sham (12) T: 21 ± 1S: 20 ± 1

RMT: 56.3 ± 3.1Sham: 55.8 ± 5.9

Repetitivesprint sports

RMT:130.3 ± 3.7Sham: 133.4 ± 3.6

Sonetti, 2001 17 (17/0) RMT (9); sham (8) 25 ± 5 RMT: 55.0 ± 5.0Sham: 54.2 ± 2.5

Cyclists RMT: 168.5 ± 39.6Sham:154.4 ± 31.5

Wylegala, 2007 30 (30/0) Thres RMT (10);IH RMT (10); sham (10)

23 ± 4 NR Swimmers Thres.: 124.6 ± 43.1IH: 120.7 ± 35.4Sham: 117.4 ± 23.7

Non-athletesEnright, 2011 40 (20/20) RMT80% (10); RMT60% (10);

RMT40% (10);Con (10)

22 ± 4 NR NA RMT80%: 68; RMT60%: 73RMT40%: 76; Con: 68

Gething, 2011 15 (10/5) RMT(5); sham (5); Con (5) 23 ± 2 VO2 peak 3.18 ± .69 L/min NA RMT: 134; S: 136; Con: 127Keramidas, 2010 18 (18/0) RMT (9); Con (9) RMT: 22 ± 4.0

Con: 23 ± 4RMT: 47.0 ± 5.0Con: 45.8 ± 6.2

NA NR

Markov, 1996 16 (8/8) RMT (8); Con (8) RMT: 28 ± 2Con: 31 ± 7

NR NA NR

Markov, 2001 37 (20/17) RMT (13); ET (9)Con (15) RMT: 43 ± 7ET: 40 ± 10Con: 37 ± 9

RMT: 36 ± 11ET: 32 ± 9Con: 35 ± 9

NA NR

Stuessi, 2001 28 (16/12) RMT (13) Con (15) RMT: 43 ± 7Con: 37 ± 9

RMT: 1.79 ± 0.52 L min%1

Con: 1.75 ± 0.39NA NR

Suzuki, 1993 12 (0/12) RMT (6) Con (6) RMT: 20 ± 0Con: 19 ± 0

NR NA NR

Verges, 2009 26 (26/0) RMT-Thres (10)RMT-IH (8)Sham (8)

Thres: 26 ± 6IH: 28 ± 4Sham: 26 ± 6

NR NA Thres.:119 ± 29 IH: 134 ± 32Sham:112 ± 23

Values are expressed as mean ± SD.Con: control group; ET: endurance training; F: female; IH: isocapnic hyperpnea training; M: male; NA: not applicable; NR: not reported; RMT: respiratory muscle training; S:Sham; ST: swimming training; Thres: threshold.


1234567891011121314151617181920212223242526272829303132333435363738394041424344454647484950515253545556575859606162636465

66676869707172737475767778798081828384858687888990919293949596979899

100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130



Table 3Characteristics of interventions.

First author, year Type of training(device)

Starting intensity Progression of trainingintensity

Session/day,week

No. ofweeks

Duration ofsession

Supervision Control and/or sham

AthletesForbes 2011 Thres

(PowerLung™)8e10 RM [ every 2 wk 1/d

(1e4 wk)BID(5e10 wk)

10 8e10 reps(3e4 s each)

Unsupervised Sham: Samefrequency but setat 10e15% of load

Holm 2004 IH (customdevice)

Max VT and fB fromcycle test

VT or fB [ every 1e2 d.Target 19 on 20-pointrespiratory effort scale.

5/wk 4 30 min Supervised Con: No trainingSham 5 min at65% of max

Inbar, 2000 Thres (threshold) 30% MIP [5% per session to 80% ofMIP by 4 wk; Thereafter, [to 80% of MIP weekly

6/wk 10 30 min Supervised Sham: Trainedwith same deviceat no load

Leddy 2007 IH (customdevice)

30 breaths/min~50% MVV

[ by 1e2 breaths/min persession

5/wk 4 30 min Unsupervised Sham: Inhaled toTLC followed by10 s breath hold

McMahon 2002 IH (customdevice)

~60% MVV 15 s [ VE till training-VE couldbe held constant for at least30 but not 35 min

20/wk 4e6 30 min Supervisedevery5th session

Con: No training

Mickleborough2008

Target resistive(RT2 device)

80% SMIP SMIP evaluated eachsession and the trainingwas adjusted to 80% SMIP

3/wk 6 36 reps Supervised Con: No trainingSham: 30% Ofthe SMIP

Mickleborough2010

Target resistive(RT2 device)

80% SMIP SMIP evaluated eachsession and the trainingwas adjusted to 80% SMIP

3/wk 6 36 reps Supervised Con: No trainingSham: 30% Ofthe SMIP

Morgan 1987 IH (customdevice)

85% MVV 15 s [ by 5% MVV the next daywhen 85% MVV performedfor all 4 bouts

5/wk 3 28 min Supervised Con: No specified

Riganas 2008 Thres (threshold) 30% MIP [ 5% each session to 80% ofMIP by wk 2. Then adjustedweekly.

5/wk 6 30 min Supervised Con: No training

Romer 2002 Thres(POWERbreath®)

50% Of MIP Periodically [ load so thatonly 30maneuvers could becompleted.

BID 6 30 reps(2 times/day)

Supervised Sham: Similarprotocol at15% MIP

Sonetti 2001 Thres or IH(POWERbreath®)

Resistive training:50% Of MIP;Endurancetraining: 50e60%Of MVV

Resistive: [ fB every 1e2 dand increase load once awk. Endurance training: [fB once a wk

5/wk 5 Resistive training: ~40 insp.; Endurancetraining:30 min/Session

Supervisedevery5th session

Sham: 30 minWith no load

Wylegala 2007 Thres (threshold)or IH (Customdevice)

Thres.: ±50 cmH2Oon inspiration andexpiration; IH: 55%Of SVC and variablefB

Thres.: NR; IH: At first [ fBby 1e2 permin after 20minof training until fB ¼ 50.Thereafter VT increased by0.1 L and fB adjusted toachieve same VE

5/wk 4 30 min/day Supervise 1sessionper wk

Sham: Inhaled toTLC for 10 s breathhold on amodified apparatus.

Non-athletesEnright 2011 Target resistive

(RT2 device)80% SMIP; 60%SMIP; 40% SMIP

SMIP was evaluated eachsession and the trainingwas adjusted to 80%, 60% or40% of the SMIP

3/wk 9 36 reps Supervised Con: No training

Gething 2004 Target resistive(Flow resistivedevice)

80% SMIP SMIP was evaluated eachsession and the trainingwas adjusted to 80% SMIP

3/wk 10 36 reps Supervised Sham: Mouthpiecewhich had asmall loadCon: No training

Keramidas 2010 IH (SpiroTiger®) 55% SVC andfB ¼ 50% MMV/bagvolume

[ the fB until achieved themaximum. Thereafter VT [by 0.1 L and fB adjusted toachieve same VE

5/wk 4 30 min Supervised Con: No training

Markov 1996 IH (customdevice)

35e40 breaths/min NR 4e5/wk 4e5 30 min Unsupervised Con: No training

Markov 2001 IH (customdevice)

50e60% VC, fB waschosen such that VE

60% MVV

[ fB or tidal volume 40 sessions 15 30 min Unsupervised ET: Running orcycling/40 sessionsin 15 wks Con:No training

Stuessi 2001 IH (customdevice)

VE ¼ 70% MVV NR 40 sessions ~15 30 min Unsupervised Con: No training

Suzuki 1993 Thres (threshold) 30% MIP NR BID, for 15min each

6 30 min Unsupervised Con: No training

Verges 2009 IH (SpiroTiger®)and Thres(DeVilbiss RT-Trainer)

IH: TI/TTOT ¼ 0.5, VE

(first session) wasset at 60% of MVV15 s, with VT ¼ 50e60% VC and fR wasadjustedaccordingly \Thres.:80% MIP

IH:1 to 2 breaths/min eachsession Thres.: NR

IH: 20sessions/RRThres.: 60sessions

4 IH: 30 min/Thres.: 15 min

Supervisedevery5th session

Sham: Voldyne5000 R, twicea day, 70% VC

BID: twice daily; Con: Control group; d: day; ET: endurance training; fB: breathing frequency; IH: Isocapnic hyperpnea training group; NR: not reported; MIP: maximalinspiratory pressure; MVV: maximal voluntary ventilation; RM: repetition maximal; SMIP: sustained maximal inspiratory pressure; SVC: sustained vital capacity; Thres:Threshold training group; VE: minute ventilation; VT: tidal volume; TI/TTOT: cycle time; wk: week.

1234567891011121314151617181920212223242526272829303132333435363738394041424344454647484950515253545556575859606162636465

66676869707172737475767778798081828384858687888990919293949596979899

100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130



studies (70%) were considered as “high quality” (Roig et al., 2008) inthe present systematic review; however, the lack of blinding andconcealed allocation detracted from the strength of their design.

Our results are in agreement with the systematic review byHajGhanbari et al. (2013) who also demonstrated significantimprovement in RME after RMT in athletes. However, the formersystematic review did not differentiate between athletes and non-athletes. Nor did it scrutinize detailed comparisons among fourRME tests. In contrast to systematic reviews, several studies did notshow statistically significant changes of RME after RMT (Forbeset al., 2011; Keramidas et al., 2010; Romer et al., 2002; Sonettiet al., 2001). This discrepancy may be explained, in part, by theintensity of training which was lower when compared with otherstudies (Mickleborough et al., 2010; Mickleborough et al., 2008;Stuessi et al., 2001), and by the ineffective training progression.Another consideration is that some studies had a small sample size(Gething et al., 2004; Suzuki et al., 1993) and thus, may have beenunderpowered to show a treatment effect.

The effects of RMTon RMEwere greater when it was assessed bytests of longer duration (MSVC and maximal sustainable thresholdloading), as indicated by larger standardizedmean differences (1.86and 1.57, respectively) and smaller P values (0.005 and 0.001,respectively) compared to the standardizedmean difference of 0.31

and smaller alpha level shown by the shorter duration MVV (0.05).This is further supported by examination of individual studieswhen only two of fifteen comparisons (Riganas et al., 2008; Suzukiet al., 1993), showed significant improvements in MVV after RMT.These data provide solid evidence that the MSVC test is moresensitive than theMVV test for evaluation of RME in this populationas per the ATS recommendations (ATS, 2002). The shorterMVVmaybe less sensitive to RMT benefits because this 12e15 s ventilatorysprint does not tax similar metabolic and contractile protein de-mands to the cellular elements overloaded and show a positiveadaptation during RMT.

The linkage between training and RME outcomes may beattributable to the specificity of training. IH training showed greaterimprovements in RME as evaluated by the MSVC than other typesof training (threshold and target resistive training), which might bedue to the high velocity, low load contractions required of both theMSVC testing and IH training. However, issues regarding theoptimal RMT protocol remain equivocal. In addition, these gainswere better demonstrated by the MSVC test when compared withthe MVV test. Two studies (Morgan et al., 1987; Wylegala et al.,2007) that performed IH training did not show changes in theMVV test but showed significant improvements when using theMSVC test in the same studies. These data stress the importance of

Table 4Technical Aspects that influence RME tests technique.

First author, Year Type of test Device Starting intensity Isocap Systema Air Humida A/V Cuesa

AthletesForbes, 2011 MVV Spirometer NR NR NR NRHolm, 2004 MSVC Custom device Maximum ventilation for at least

10 minYes Yes Yes

Inbar, 2000 MITL Custom device No load NR NR NRLeddy, 2007 MSVC Custom device 60% MVV Yes Yes YesMcMahon, 2002 MSVC Custom device 70% MVV for a min of 6 min But no

longer than 15 minYes NR NR

Mickleborough, 2008 MSTL RT2 trainer device Sum of SMIP's generated for eachtraining load successfully completed.

NR NR Yes

Mickleborough, 2010 MSTL RT2 trainer device Sum of SMIP's generated for eachtraining load successfully completed.

NR NR Yes

Morgan, 1987 MSVC Custom device 100% MVV as long as possible Yes NR YesRiganas, 2008Q3 MVV12s Spirometer Maximum effort during 12 s NR NR NRRomer, 2002 MVV15s Pneumotachograph

spirometerMaximum effort during 15 s NR NR NR

Sonetti, 2001 MSVC 13.5 L water-sealedspirometer

90% MVV Yes Yes Yes

Wylegala, 2007 MSVC Custom device VT ¼ 50% SVC; fB ¼ 60% MVV/VT;breathe until exhaustion

NR NR Yes

Non-athletesEnright, 2011 MSTL Electronic manometer No defined intensity. NR NR YesGething, 2004 MSTL Flow-resistive device Max pressure that could be sustained

over time (SMIP)NR NR NR

Keramidas, 2010 MVV12s Cardiovit AT-2 plusspirometer

Maximum effort during 12 s NR NR NR

Markov, 1996 MSVC Custom device VT ¼ 60% VC; fB ¼ VE corresponding to70% MVV; breathe until exhaustion

Yes Yes Yes

Markov, 2001 MSVC Custom device 70% MVV until exhaustion (at least15 min)

NR NR NR

Stuessi, 2001 MSVC Custom device 70% MVV until VE drops by more than10% or by the subjects stoppingexercise.

Yes NR NR

Suzuki, 1993 MVV Dry seal spirometer NR NR NR NRVerges, 2009 MSVC SpiroTiger® At 70% of MVV with VT ¼ 50e60% VC

and fB adjusted accordingly, 8 min withintervals of 6 min of rest until taskfailure.

Yes NR Yes

Air humid ¼ presence or absence of the air humidification system; A/V Cues ¼ presence or absence of the Audio/visual feedback system; fB ¼ breathing frequency; IsocapSystem¼ Presence or absence of the isocapnic system; NR¼ not reported; MILT¼maximal incremental threshold loading test; MSTL¼maximal sustainable threshold loadingtest; MSVC¼maximal sustained ventilatory capacity; MVV¼maximum voluntary ventilation; SMIP¼ sustainedmaximal inspiratory pressure; SVC¼ sustained vital capacity;VC ¼ vital capacity; VE ¼ minute ventilation; VT ¼ tidal volume.

a ATS (2002) recommendation for devices which evaluate RME.


1234567891011121314151617181920212223242526272829303132333435363738394041424344454647484950515253545556575859606162636465

66676869707172737475767778798081828384858687888990919293949596979899

100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130



Fig. 2. Forest plots of RME outcomes in athletes and non-athletes with subgroup analyses of different RME tests.


1234567891011121314151617181920212223242526272829303132333435363738394041424344454647484950515253545556575859606162636465

66676869707172737475767778798081828384858687888990919293949596979899

100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130



Fig. 3. Forest plots of the RME outcomes with subgroup analyses of different types of training.


1234567891011121314151617181920212223242526272829303132333435363738394041424344454647484950515253545556575859606162636465

66676869707172737475767778798081828384858687888990919293949596979899

100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130



careful selection of a RME test that matches the imposed recruit-ment demands of RMT and more importantly, corresponds to therequirements of the sport or physical activity that is limited.

In conclusion, this systematic review provides evidence thatRMT improves RME in athletes and non-athletes. Matching theventilatory requirements of the test to the specificity of the trainingis a key consideration to ensure a sensitivemeasure that reflects thegreatest RME gains. In addition to test duration, muscle recruitmentpatterns appear to be key components to ensure a sensitiveoutcome. For example, the MSVC showed greater improvementsafter isocapnic hyperpnea RMT when compared with the gainsafter threshold RMT (Wylegala et al. 2007). Other RME tests oflonger duration, such as the maximal sustainable threshold loadingand maximal incremental threshold loading, appear to be respon-sive measures to RMT but to a lesser extent than the MSVC test. Animportant clinical consideration is that the MSVC, although supe-rior in healthy persons may not be optimal in patients with airflowlimitation in whom the ability for maximal ventilation may belimited more so by airways obstruction rather than respiratorymuscle performance. Of further note, the apparatus required for theMSVC is more elaborate and considerably more expensive than thatrequired for threshold loading tests. In closing, MSVC is the rec-ommended RME test in healthy persons based on the evidencefrom the meta-analyses of this systematic review. The maximalsustainable or incremental threshold loading tests, although morepractical, are superior to the MVV but less responsive than theMSVC.

Conflict of interestNone declared.

FundingNone declared.

Uncited reference

Inbar et al., 2000, Lindholm et al., 2007, Verg"es, Flore, et al.,2009Q4 .

Acknowledgments

We thank Charlotte Beck librarian of the University of BritishColumbia for the support during the search strategy steps.

References

American Thoracic Society. (1995). Standardization of spirometry: 1994 update. TheAmerican Journal of Respiratory and Critical Care Medicine, 152, 1107e1136.

American Thoracic Society/European Respiratory Society. (2002). ATS/ERS state-ment on respiratory muscle testing. The American Journal of Respiratory andCritical Care Medicine, 166, 518e624.

Bell, G. J., Game, A., Jones, R., et al. (2013). Inspiratory and expiratory respiratorymuscle training as an adjunct to concurrent strength and endurance trainingprovides no additional 2000 m performance benefits to rowers. Research inSports Medicine, 21, 264e279.

Bradley, M. E., & Leith, D. E. (1978). Ventilatory muscle training and the oxygen costof sustained hyperpnea. Journal of Applied Physiology: Respiratory, Environmentaland Exercise Physiology, 45, 885e892.

Dias, F. D., Sampaio, L. M. M., Silva, G. A., et al. (2013). Home-based pulmonaryrehabilitation in patients with chronic obstructive pulmonary disease: a ran-domized clinical trial. The International Journal of Chronic Obstructive PulmonaryDisease, 8, 537e544.

Driller, M., & Panton, C. (2012). The effects of respiratory muscle training in highly-trained rowers. The Journal of Exercise Physiology Online, 15, 93e102.

Enright, S. J., & Unnithan, V. B. (2011). Effect of inspiratory muscle training in-tensities on pulmonary function and work capacity in people who are healthy: arandomized controlled trial. Journal of Physical Therapy, 91, 894e905.

Fairbarn, M. S., Coutts, K. C., Pardy, R. L., et al. (1991). Improved respiratory muscleendurance of highly trained cyclists and the effects on maximal exercise per-formance. International Journal of Sports Medicine, 12, 66e70.

Fischer, G., Tarperi, C., George, K., et al. (2014). An exploratory study of respiratorymuscle endurance training in high lesion level paraplegic handbike athletes.The Clinical Journal of Sport Medicine, 24, 69e75.

Fiz, J. A., Romero, P., Gomez, R., et al. (1998). Indices of respiratory muscle endur-ance in healthy subjects. Respiration, 65, 21e27.

Forbes, S., Game, A., Syrotuik, D., et al. (2011). The effect of inspiratory and expi-ratory respiratory muscle training in rowers. Research in Sports Medicine, 19,217e230.

Freedman, S. (1970). Sustained maximum voluntary ventilation. Respiration Physi-ology, 8, 230e244.

Gething, A. D., Williams, M., & Davies, B. (2004). Inspiratory resistive loading im-proves cycling capacity: a placebo controlled trial. The British Journal of SportsMedicine, 38, 730e736.

HajGhanbari, B., Yamabayashi, C., Buna, T. R., et al. (2013). Effects of respiratorymuscle training on performance in athletes: a systematic review with meta-analyses. The Journal of Strength and Conditioning Research, 27, 1643e1663.

Holm, P., Sattler, A., & Fregosi, R. F. (2004). Endurance training of respiratorymuscles improves cycling performance in fit young cyclists. BMC Physiology, 4,1e14.

Illi, S. K., Held, U., Frank, I., et al. (2012). Effect of respiratory muscle training onexercise performance in healthy individuals: a systematic review and meta-analysis. Sports Medicine, 42, 707e724.

Inbar, O., Weiner, P., Azgad, Y., et al. (2000). Specific inspiratory muscle training inwell-trained endurance athletes. Medicine & Science in Sports & Exercise, 32,1233e1237.

Johnson, P. H., Cowley, A. J., & Kinnear, W. J. M. (1997). Incremental thresholdloading: a standard protocol and establishment of a reference range in naivenormal subjects. The European Respiratory Journal, 10, 2868e2871.

Johnson, M. A., Sharpe, G. R., & Brown, P. I. (2007). Inspiratory muscle trainingimproves cycling time trial performance and anaerobic work capacity but notcritical power. The European Journal of Applied Physiology, 101, 761e770.

Keramidas, M. E., Debevec, T., Amon, M., et al. (2010). Respiratory muscle endurancetraining: effect on normoxic and hypoxic exercise performance. The EuropeanJournal of Applied Physiology, 108, 759e769.

Kroff, J., & Terblanche, E. (2010). The kinanthropometric and pulmonary de-terminants of global respiratory muscle strength and endurance indices in anathletic population. The European Journal of Applied Physiology, 110, 49e55.

Leddy, J. L., Limprasertkul, A., Patel, S., et al. (2007). Isocapnic hyperpnea trainingimproves performance in competitive male runners. The European Journal ofApplied Physiology, 99, 665e676.

Leith, D. E., & Bradley, M. (1976). Ventilatory muscle strength and endurancetraining. The Journal of Applied Physiology, 41, 508e516.

Lindholm, P., Wylegala, J., Pendergast, D. R., et al. (2007). Resistive respiratorymuscle training improves and maintains endurance swimming performance indivers. The Undersea and Hyperbaric Medicine Journal, 34, 169e180.

Maher, C. G., Sherrington, C., Herbert, R. D., et al. (2003). Reliability of the PEDroscale for rating quality of randomized controlled trials. Physical Therapy, 83,713e721.

Mancini, D. M., Henson, D., LaManca, J., et al. (1994). Evidence of reduced respira-tory muscle endurance in patients with heart failure. Journal of the AmericanCollege of Cardiology, 24, 972e981.

Markov, G., Orler, R., & Boutellier, U. (1996). Respiratory training, hypoxic ventila-tory response and acute mountain sickness. Respiration Physiology, 105,179e186.

Markov, G., Spengler, C. M., Knӧpfli-Lenzin, C., et al. (2001). Respiratory muscletraining increases cycling endurance without affecting cardiovascular responsesto exercise. The European Journal of Applied Physiology, 85, 233e239.

McArdle, W. D., Katch, F. I., & Katch, V. L. (Eds.). (2001). Exercise physiology: Energy,nutrition and human performance (5th ed.). Philadelphia, PA: Lippincott Wil-liams & Wilkins.

McConnell, A. K., & Romer, L. M. (2004). Respiratory muscle training in healthyhumans: resolving the controversy. International Journal of Sports Medicine, 25,284e293.

McElvaney, G., Fairbarn, M. S., Wilcox, P. G., et al. (1989). Comparison of two-minuteincremental threshold loading and maximal loading as measures of respiratorymuscle endurance. Chest, 96, 557e563.

McMahon, M. E., Boutellier, U., Smith, R. M., et al. (2002). Hyperpnea training at-tenuates peripheral chemosensitivity and improves cycling endurance. TheJournal of Experimental Biology, 205, 3937e3943.

Mickleborough, T. D., Nichols, T., Lindley, M. R., et al. (2010). Inspiratory flowresistive loading improves respiratory muscle function and endurance capacityin recreational runners. Scandinavian Journal of Medicine and Science in Sports,20, 458e468.

Mickleborough, T. D., Stager, J. M., Chatham, K., et al. (2008). Pulmonary adaptationsto swim and inspiratory muscle training. The European Journal of AppliedPhysiology, 103, 635e646.

Morgan, D. W., Kohrt, W. M., Bates, B. J., et al. (1987). Effects of respiratory muscleendurance training on ventilatory and endurance performance of moderatelytrained cyclists. International Journal of Sports Medicine, 8, 888e893.

Nickerson, B. G., & Keens, T. G. (1982). Measuring ventilatory muscle endurance inhumans as sustainable inspiratory pressure. Journal of Applied Physiology: Res-piratory, Environmental and Exercise Physiology, 52, 768e772.

Rassler, B., Marx, G., Hallebach, S., et al. (2011). Long-term respiratory muscleendurance training in patients with myasthenia gravis: first results after fourmonths of training. Autoimmune Diseases, 2011, 1e7.


1234567891011121314151617181920212223242526272829303132333435363738394041424344454647484950515253545556575859606162636465

66676869707172737475767778798081828384858687888990919293949596979899

100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130



Riganas, C. S., Vrabas, I. S., Christoulas, K., et al. (2008). Specific inspiratorymuscle training does not improve performance or VO2max levels in welltrained rowers. The Journal of Sports Medicine and Physical Fitness, 48,285e292.

Roig, M., Shadgan, B., & Reid, W. D. (2008). Eccentric exercise in patients withchronic health conditions: a systematic review. Physiotherapy Canada, 60,146e160.

Romer, L. M., McConnell, A. K., & Jones, D. A. (2002). Effects of inspiratory muscletraining upon recovery time during high intensity, repetitive sprint activity.International Journal of Sports Medicine, 23, 353e360.

Silva, A. C., Neder, J. A., Chiurciu, M. V., et al. (1998). Effect of aerobic training onventilatory muscle endurance of spinal cord injured men. Spinal Cord, 36,240e245.

Sonetti, D. A., Wetter, T. J., Pegelow, D. F., et al. (2001). Effects of respiratory muscletraining versus placebo on endurance exercise performance. Respiration Physi-ology, 127, 185e199.

Spengler, C. M., Roos, M., Laube, S. M., et al. (1999). Decreased exercise blood lactateconcentrations after respiratory endurance training in humans. European Jour-nal of Applied Physiology and Occupational Physiology, 79, 299e305.

Stuessi, C., Spengler, C. M., Knӧpfli-Lenzin, C., et al. (2001). Respiratory muscleendurance training in humans increases cycling endurance without affectingblood gas concentrations. The European Journal of Applied Physiology, 84,582e586.

Suzuki, S., Yoshike, Y., Suzuki, M., et al. (1993). Inspiratory muscle training andrespiratory sensation during treadmill exercise. Chest, 104, 197e202.

Verges, S., Boutellier, U., & Spengler, C. M. (2008). Effect of respiratory muscleendurance training on respiratory sensations, respiratory control and exerciseperformance: a 15 year experience. Respiratory Physiology & Neurobiology, 161,16e22.

Verg"es, S., Flore, P., Nantermoz, G., et al. (2009). Respiratory muscle training inathletes with spinal cord injury. International Journal of Sports Medicine, 30,526e532.

Verges, S., Renggli, A. S., Notter, D. A., et al. (2009). Effects of different respiratorymuscle training regimes on fatigue-related variables during volitional hyper-pnoea. Respiratory Physiology & Neurobiology, 169, 282e290.

Vogiatzis, I., & Zakynthinos, S. (2012). Factors limiting exercise tolerance in chroniclung diseases. Comprehensive Physiology, 2, 1779e1817.

Weiner, P., Magadle, R., Beckerman, M., et al. (2003). Comparison of specific expi-ratory, inspiratory, and combined muscle training programs in COPD. Chest, 124,1357e1364.

Wilmore, J. H., & Costill, D. L. (Eds.). (2005). Physiology of sports and exercise (3rded.). Champaign, IL: Human Kinetics Publishers.

Wylegala, J. A., Pendergast, D. R., Gosselin, L. E., et al. (2007). Respiratory muscletraining improves swimming endurance in divers. The European Journal ofApplied Physiology, 99, 393e404. Q2


12345678910111213141516171819

20212223242526272829303132333435363738



respiratory muscle endurance after respiratory muscle ... · literature review respiratory muscle...

Documents