Article Text

This article has a correction. Please see:

Download PDFPDF
Concurrent inspiratory muscle and cardiovascular training differentially improves both perceptions of effort and 5000 m running performance compared with cardiovascular training alone
  1. A M Edwards1,
  2. C Wells2,
  3. R Butterly3
  1. 1
    UCOL Institute of Technology, Palmerston North, New Zealand
  2. 2
    Sheffield Hallam University, Sheffield, UK
  3. 3
    Leeds Metropolitan University, Leeds, UK
  1. Dr Andrew M Edwards, UCOL Institute of Technology, School of Applied Health Sciences, Cnr of Princess and Queen St, Palmerston North, 4412, New Zealand; a.m.edwards{at}


Objective: To examine whether inspiratory muscle training (IMT) is a useful additional technique with which to augment cardiovascular exercise training adaptations.

Methods: 16 healthy untrained males agreed to participate in the study and were randomly assigned to training (TRA; n = 8) and placebo (PLA; n = 8) groups. Pre- and post-training measurements of spirometry and maximal inspiratory mouth pressure (MIP) were taken in addition to i) maximal aerobic power (VO2max) and ii) 5000 m run time-trial. All subjects completed the same 4 week cardiovascular training programme which consisted of three running sessions (CV1: 5×1000 m, CV2: 3×1600 m, SP1: 20 min run) in each of the 4 weeks. IMT was performed daily by both groups using an inspiratory muscle trainer (POWERbreathe). TRA completed 30 maximal inspirations while PLA inspired 30 times against a negligible resistance.

Results: Mean MIP increased significantly in both groups (TRA: 14.5 (SD 6.8)% change, PLA: 7.8 (7.4)% change) from pre- to post-training (p<0.01) but was not significantly related to changes in running performance. Mean CV1 training-repetition runs improved similarly in both groups, but RPE evaluations were significantly reduced in TRA (15.7 (0.7)) compared with PLA (16.6 (0.8)) at week 4 (p<0.05). Pre- to post-training changes in VO2max were well-matched between both TRA (+2.1 (2.3)%) and PLA (+1.3 (2.4)%) while post-intervention 5000 m performance was significantly augmented in TRA compared with PLA (TRA: 4.3 (1.6)%, PLA: 2.2 (1.9)%, p<0.05).

Conclusions: The addition of IMT to a cardiovascular training programme augments 5000 m running performance but exerts no additional influence over VO2max compared with a cardiovascular-training group. This is probably due to IMT-induced reduction in perceived effort at high ventilatory rates, which is of greater consequence to longer duration time-trial performances than incremental tests of VO2max.

Statistics from

Studies on the effect of respiratory muscle training on whole body exercise training in healthy humans have so far demonstrated conflicting results.13 One common observation among researchers has been that isolated inspiratory muscle training (IMT) modifies the structure and function of the respiratory muscles (eg4 5&rpar.;). However, a well-defined role for IMT has not yet been established as it remains less clear whether this form of training has an ergogenic effect on endurance performance.1 3

Numerous studies have measured both maximal aerobic power (VO2max) and submaximal oxygen dynamics (VO2 kinetics) in response to isolated IMT but have not found cardiovascular fitness to be improved (eg57). This is unsurprising as the cardiovascular load of training the respiratory muscles is minimal. Nevertheless, significant IMT-induced improvements have more often been reported in response to constant-power and time-trial tests over various exercise modalities such as rowing,2 cycling,8 and treadmill running,5 9 although the cause of such improvements remains unclear (eg3).

Several putative mechanisms have been suggested to explain IMT-induced performance adaptations and these include a delay of respiratory muscle fatigue,10 11 a redistribution of blood flow from respiratory to locomotor muscles,3 and a decrease in the perceptions of respiratory and limb discomfort during exercise to fatigue.2 1214 Conversely, it has also been suggested that respiratory muscle training might compromise blood flow (increasing blood lactate) during maximal exercise due to an increased cost of breathing.15

The controversy surrounding the influence of IMT on fatigue is probably further complicated by the prevalence among researchers to infer a reductionism interpretation to research findings. Reductionism implies that performance is controlled by the limiting capacity of a physiological system, where either metabolite accumulation or substrate depletion causes catastrophic system failure and thus fatigue.16 However, it has recently been suggested that this model of physiological control of exercise is flawed, and that, in contrast, metabolic activity in different physiological systems (such as respiratory dynamics) is continually regulated by a brain algorithm throughout exercise in order to ensure that these systems are never maximally utilised.17 18 Therefore, it is possible that reduced effort sensations from respiratory receptors to the brain19 might be one influential feedback system (of many) within a complex control system without necessarily being the cause of fatigue. As muscle recruitment appears to be manipulated by central control,20 it is likely that reduced respiratory effort could lead to greater locomotor muscle recruitment (through a lower perception of physical effort) and consequently enhance endurance performance. These considerations are also likely to have greater consequence for time-trial or constant-power performance tests than relatively brief and standardised incremental tests of VO2max.3 21

The most plausible explanation for IMT-induced performance adaptations probably lies within the reports of reduced effort sensations.2 1214 It is therefore surprising that experimental models examining the effects of IMT have tended to do so in isolation rather than utilising IMT as an additional aid to a cardiovascular training programme. We have previously proposed that the inclusion of IMT within a training programme might positively alter the perception of effort during training bouts and consequently augment the volume and quality of work accomplished.5 This in turn could be expected to enhance endurance performance.

It is the aim of this study to examine whether the concurrent performance of IMT and repetition-based endurance training induces additional endurance performance adaptations in comparison with endurance training alone.



Sixteen healthy male subjects agreed to participate in the study (table 1). All subjects were informed of the procedures in advance of the study and informed consent was provided prior to any data collection in accordance with the Institution’s Research Ethics Committee and the Declaration of Helsinki.

Table 1 Pre- and post-training anthropometric and cardiovascular characteristics of the subjects in the training (TRA) and in the placebo (PLA) groups

Experimental design

All tests were performed at similar times of the day and were conducted in the same routine order prior to and following 4 weeks of training. The subjects were instructed to refrain from additional organised physical activity during the 4 week period. Training took place over the summer period and outdoor trials were conducted over similar environmental conditions at the same routine time of day (evening). There were occasional variations in wind speed (0.2–12 k/h) during training, but 5000 m performance tests were conducted in similar conditions (17–19°C; RH 8–40%; wind 1.8–4.2 k/h).

At the beginning of the study, maximal static inspiratory mouth pressure (MIP) was measured using a portable hand-held mouth pressure meter (Precision Medical, London, UK) and baseline spirometry measurements were recorded (FVC, FEV1, MVV) (Vitalograph, Compact, Buckingham, UK). These measures were followed by an incremental test to volitional exhaustion on a treadmill (Woodway PPS 55, Germany) for the combined assessment of maximal aerobic power and ventilatory threshold (Tvent). On a separate and subsequent occasion, the participants were randomised into training (TRA) (n = 8) and placebo (PLA) (n = 8) groups and randomly assigned places into two races for the assessment of 5000 m run time-trial performances on an all-weather athletics track. Performance times were recorded by the duplicate use of standardised laboratory stopwatches. The same sequence of tests occurred post-training but the participants of the two races were again randomised to avoid preset pacing strategies based on knowledge from the previous race.

Inspiratory muscle training

Both groups performed inspiratory efforts daily using an inspiratory pressure-threshold device (POWERbreathe, Gaiam, UK). Group TRA were required to complete inspiratory manoeuvres at a load that would permit them to only just complete the 30 breaths.5 PLA inspired 30 times daily using a POWERbreathe device at the lowest resistive setting (∼15% of maximal resistance).5 The use of negligible inspiratory resistance has been used as a control measure in other studies,2 5 and is thought to elicit only a minor training effect.22

Cardiovascular training intervention

All participants performed the cardiovascular conditioning programme; the addition of IMT was therefore the differential factor between TRA and PLA. This training regime consisted of two supervised track running sessions in each of the 4 weeks (5×1000 m (CV1), and 3×1600 m (CV2)) in which CV1 was targeted for rigorous evaluation of performance times and RPE responses.23 The participants were asked to rate their perceived level of exertion as a representative measure of the overall session and not in response to each repetition. All participants were also instructed to perform one self-paced (self-directed) 20 min run (SP1) in each week and training diaries were completed to confirm adherence.

The track-running repetition training sessions CV1 and CV2 were performed in mixed-condition groups which were randomised from the pre-training 5000 m time-trial. No participants were informed of either their repetition or race performance times during the experimental process. The post-training 5000 m time-trials were performed as two races (n = 8) and the composition of the two race trials was again randomised to minimise the influence of peer-pacing in response to over familiarity within training groups.

Measurement of gas exchange, blood lactate and heart rate responses

Gas exchange and minute ventilation were continuously recorded breath-by-breath at the mouth for the measurement of VO2max and Tvent. Gases were continuously drawn through a capillary line and analysed for O2 and CO2 concentrations by fast-response analysers utilising principles of electrochemical reactions for the detection of O2 and absorption by CO2 of appropriate wavelengths of infrared light (Cortex MetaMax 3B, Cortex Biophysik, Germany). The system was calibrated before and verified after each test with standard calibration gases. Expired volume was measured by a volume-measuring turbine which was calibrated with a 3 l syringe (Hans Rudolph, Kansas City, Missouri, USA).

Capillary whole blood samples were drawn from the finger tip at the conclusion of time-trial performances for the immediate analysis of blood lactate concentration (Lactate Pro, USA). Heart rates were recorded at 15 s intervals (S610i, Polar, Kempele, Finland) throughout all exercise tests.

Statistical analyses

The statistical software package SPSS (version 11.0, SPSS, Chicago, Illinois) was used for all statistical analysis. Parametric pre- and post-training results and group interactions were statistically compared using one-way or two-way repeated measures analyses of variance (ANOVA) and post hoc Tukey tests of Honestly Significant Difference as appropriate. Other group comparisons were made using paired Student t tests. Non-parametric data were assessed using Friedman’s analysis of variance and Mann–Whitney U tests. The relationships between data sets were examined using Pearson Product Moment Correlations. Probability values of less than 0.05 were considered significant. All results are expressed as means (SD) unless otherwise stated.


There were no significant differences in standard spirometry measurements either between the two groups or as a consequence of the training intervention (table 1). MIP increased significantly in both TRA (14.5 (6.8)% change; p<0.01) and PLA (7.8 (7.4)% change; p<0.01) while the extent of performance change (%) in MIP was significantly greater in TRA than PLA (p<0.05).

Training-based repetition performance times in CV1 did not improve significantly for either group over the 4 week period, although there was a trend towards improvement in TRA in session CV1 at week 4 (p = 0.09) (fig 1). RPE evaluations in response to CV1 were reduced in TRA from week 1 to week 4 (p<0.01) and were significantly lower than PLA at week 4 (p<0.05) (fig 2). Training-based repetition performance times for the second weekly repetition session (CV2) were not recorded in duplicate and, although participant adherence to the session was good, this session was considered purely as a training activity and was not targeted for rigorous analysis. Consequently, the results in response to that session are not reported in this study.

Figure 1 Mean 1000 m repetition times for CV1 (5×1000 m) weekly training sessions. Mean 1000 m repetition efforts at week 4 demonstrate a trend towards improved performances in TRA (p = 0.09).
Figure 2 Mean (SD) rating of perceived exertion (RPE) as an overall rating of perception of effort for the CV1 (5×1000 m) weekly training session.

Maximal aerobic power (TRA: 2.2 (2.3)%; PLA: 1.5 (2.4)%, p<0.05) and Tvent (TRA: 3.1 (3.2)%; PLA: 2.5 (3.6)%, p<0.05) improved similarly in both groups over the 4 week period (table 1) and there was no between-group difference in performance change (%) in either measurement. There were also no significant differences between groups in maximal ventilatory rates and breathing frequency in response to laboratory tests.

Performance times in the post-training 5000 m races improved in both TRA (p<0.01) and PLA (fig 3) (p<0.01). Performance change was significantly enhanced in TRA compared with PLA (p<0.05).

Figure 3 Mean pre- to post-training 5000 m performance times for experimental (TRA; 4.3 (SD 1.6)%) and placebo (PLA; 2.2 (1.9)%) groups. *significant post-training change in performance, p<0.01. †significantly improved post-training performance change in TRA compared with PLA, p<0.05.

Maximal heart rates did not change over the 4 week intervention and were not different between groups (TRA: 189.7 (2.7) beats/min; PLA: 191.2 (3.1) beats/min). There was also no difference in maximal blood lactate concentrations in response to the 5000 m race either between groups or from pre-training (TRA: 8.4 (3.7) mmol/l; PLA 9.1 (3.2) mmol/l) to post-training (TRA: 8.7 (3.2) mmol/l; PLA: 9.0 (2.9) mmol/l).


The main finding from this study was that the concurrent use of IMT and cardiovascular training resulted in differentially improved mean outdoor 5000 m time-trial performances (TRA: 4.3 (1.6)%) compared with a placebo condition (2.2 (1.9)%) (p<0.05), but does not appear to enhance VO2max to a greater extent than cardiovascular training alone (TRA: +2.1 (2.3)%, PLA: +1.3 (2.4)%).

The similarity of maximal heart rates responses and pre- to post-training change in VO2max across both groups in this study is consistent with previous observations where VO2max has been shown to be insensitive to IMT.5 6 8 This is unsurprising, as central circulatory adaptations such as increases in maximal stroke volume or cardiac output have also not been shown to occur with IMT6 24 and probably require sustained and direct cardiovascular stimulation.5 In addition, typical incremental tests of VO2max are relatively brief, use standardised criteria to identify maximal performance, and only require high ventilatory rates as subjects approach the final stages of the test. In contrast, endurance time-trial performances constantly require individuals to upregulate and downregulate running velocity in order to effectively pace the exercise bout.25 26 Consequently, time-trial performances require performers to sustain high velocities for much greater durations and as such probably place greater demand on respiratory dynamics than VO2max.

Post-training measurements of maximal blood lactate concentrations were not different between groups in response to either laboratory tests (p = 0.63) or track-running time-trials (p = 0.47). Neither group experienced any pre- to post-training performance-related adaptation in maximal blood lactate concentration, and although the augmented post-training 5000 m performances by TRA presumably required greater glycolytic metabolism (assuming no net improvement in cardiovascular fitness) this was probably offset by the shorter (faster) performance duration. Several studies have reported attenuated blood lactate responses at the same relative work intensity following IMT, but this has not been a consistent observation (eg). As there was no difference between lactate responses, it is not possible to additionally infer meaning from this finding, particularly as metabolic acidosis is known to exhibit considerable interindividual and intraindividual variations at fatigue.27

Our results demonstrate that 4 weeks of repetition-based cardiovascular training either with, or without, IMT is an adequate stimulus to elicit a significant change in inspiratory muscle function (MIP). Interestingly, changes in MIP were not related to running performance, which is surprising, as MIP probably minimises the effects of exercise-induced diaphragmatic fatigue.11 However, our observations are in agreement with others2 8 14 15 and it seems likely that changes in respiratory muscle function are more closely related to psychological factors than physiological adaptations.

The reduced effort sensations reported in response to CV1 by TRA at week 4 of the concurrent IM and cardiovascular training intervention are consistent with previous observations following isolated IMT.2 12 From this finding, it is possible that the additional repeated generation of large respiratory pressures during IMT may have invoked a desensitising effect upon the sensory feedback mechanisms between the respiratory muscles and the brain.20 As changes within the respiratory muscles underlie interactions with the brain and working locomotor muscles,8 19 it is possible that this feedback mechanism may have positively influenced the subconscious recruitment patterns of skeletal muscle during the time-trial,20 leading to the observed 5000 m performance improvement. However, as muscle recruitment was not measured in this experiment it is not possible to speculate further on this issue.

Mean 1000 m training-based repetition performances did not improve significantly, although a trend towards faster performances was observed in TRA at week 4 (p = 0.09; fig 2). The trajectory of performance change suggests that a longer training intervention might have identified significant change by TRA. However, the weekly RPE evaluations in response to CV1 identified a significant difference in perceived effort between groups at the same stage (fig 3), which indicates that psychological factors had altered prior to (or in the absence of) physiological adaptation. It seems likely that the observed reduction in RPE had a positive subsequent impact on TRA’s postintervention 5000 m performance, which clearly required much longer (∼17–21 min) sustained effort than individual 1000 m repetitions (∼3–5 min). Both distances would have required a pacing strategy25 but the lower effort sensations were probably more influential over the longer distance. Therefore, lower levels of perceived effort may have positively contributed to the 5000 m performance. RPE evaluations at the end of the 5000 m time-trial were not different between groups and, in our experience, are of limited value following a single maximal race. Participants inevitably provide a maximal rating on the RPE scale following a maximal race, and the value of the RPE scale appears to lie within training and submaximal repetition efforts.

Previous studies have identified that pacing strategies are implicit in maximising endurance time-trial performances25 28 and these are based on factors such as previous experiences, the health status of the athlete and also dynamic considerations during a race such as respiratory dynamics, blood pH, skin temperature, plasma osmolality, and metabolic fuel stores.28 In the case of this study, the influence of prior experience was positively manipulated by the combination of IMT and repetition training over shorter distances (1000 m to 1600 m). It remains unclear whether this type of regulatory “deception”29 might result in a sustainable effect or whether it might be a short-term response to a temporarily (and favourably) altered perception of physical capabilities. The presence of a regulatory central (brain) governor suggests that no one single variable controls performance17 18 28 and it is therefore possible that our observations represent a short-term performance change rather than a meaningful physiological adaptation.


The positive 5000 m performance adaptation supports the use of IMT as an additional training device when used concurrently with cardiovascular training. However, the lack of supporting performance change in either VO2max or Tvent complicates a reductionism interpretation of positive IMT-induced performance adaptations. It seems more likely that various factors improving perception of effort are crucial to performance change in time-trials and IMT may be one which positively influences conscious sensations of fatigue as perceived through ventilatory factors. Repeated post-training tests are required to confirm whether IMT-induced performance change is a temporary feature of altered perception or a sustained physiological adaptation.

What is already known on this topic

Numerous studies have identified that inspiratory muscle training (IMT) is more influential in time-trial or constant-power performance tests than relatively brief and standardised incremental tests of maximal aerobic power. However, it is currently unclear whether IMT offers any additional training benefit when added to a cardiovascular training programme. In addition, there is no consensus of opinion in the scientific literature on the putative mechanisms which might explain IMT-induced performance adaptations.

What this study adds

This study examined whether IMT-induced training adaptations are best explained by the theory of central fatigue rather than reductionism. Positive postintervention alterations in 5000 m running performances supported the use of IMT as an additional aid to cardiovascular training. Alterations observed in ratings of perceived exertion (RPE) support the view that the primary mechanism for IMT-induced performance adaptations may be to positively influence conscious sensations of fatigue as perceived through ventilatory factors. The findings of this study may prove influential in the interpretation of similar future interventions.


View Abstract


  • Competing interests: None.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.