Effect of Inspiratory Muscle Training Intensities on Pulmonary Function and Work Capacity in People Who Are Healthy: A Randomized Controlled Trial

Stephanie J. Enright, Viswanath B. Unnithan


Background Inspiratory muscle training (IMT) has been shown to improve inspiratory muscle function, lung volumes (vital capacity [VC] and total lung capacity [TLC]), work capacity, and power output in people who are healthy; however, no data exist that demonstrate the effect of varying intensities of IMT to produce these outcomes.

Objectives The purpose of this study was to evaluate the impact of IMT at varying intensities on inspiratory muscle function, VC, TLC, work capacity, and power output in people who are healthy.

Design This was a randomized controlled trial.

Setting The study was conducted in a clinical laboratory.

Participants Forty people who were healthy (mean age=21.7 years) were randomly assigned to 4 groups of 10 individuals.

Interventions Three of the groups completed an 8-week program of IMT set at 80%, 60%, and 40% of sustained maximum inspiratory effort. Training was performed 3 days per week, with 24 hours separating training sessions. A control group did not participate in any form of training.

Measurements Baseline and posttraining measurements of body composition, VC, TLC, inspiratory muscle function (including maximum inspiratory pressure [MIP] and sustained maximum inspiratory pressure [SMIP]), work capacity (minutes of exercise), and power output were obtained.

Results The participants in the 80%, 60%, and 40% training groups demonstrated significant increases in MIP and SMIP, whereas those in the 80% and 60% training groups had increased work capacity and power output. Only the 80% group improved their VC and TLC. The control group demonstrated no change in any outcome measures.

Limitations This study may have been underpowered to demonstrate improved work capacity and power output in individuals who trained at 40% of sustained maximum inspiratory effort.

Conclusion High-intensity IMT set at 80% of maximal effort resulted in increased MIP and SMIP, lung volumes, work capacity, and power output in individuals who were healthy, whereas IMT at 60% of maximal effort increased work capacity and power output only. Inspiratory muscle training intensities lower than 40% of maximal effort do not translate into quantitative functional outcomes.

In people who are healthy, the inability to sustain high levels of ventilation has been established to be a factor in limiting maximal aerobic capacity.13 Although previous studies have shown that the pulmonary system is unaffected by whole-body exercise,4,5 evidence now suggests that a regimen of high-intensity inspiratory muscle training (IMT) without the addition of systemic exercise may result in quantitative outcomes. These outcomes include increased lung volumes, diaphragm thickness, and work capacity in people who are healthy and moderately trained6 and in improved running performance7,8 and recovery time during sprint activity.9 In addition, inspiratory resistive loading has been shown to enhance cycling capacity10 and swimming performance11 and improve respiratory muscle function in wheelchair athletes.12 These functional improvements have been associated with decreased blood lactate concentrations during whole-body exercise in highly trained individuals.13

The inspiratory muscles are morphologically and functionally skeletal muscles and, therefore, should respond to training in the same way as would any locomotor muscle if an appropriate physiological load is applied.14 However, controversy exists regarding the mode and intensity of training required to result in improvements in specific indexes of pulmonary function and work capacity. Generally, training theory suggests that gains in inspiratory muscle strength (force-generating capacity) can be achieved at intensities of 80% to 90% of maximum inspiratory pressure (MIP). Strength-endurance gains (maximal effective force that can be maintained) can be achieved at 60% to 80% of MIP, and gains in endurance (the ability to continue a dynamic task for a prolonged period) can be achieved at approximately 60% of peak pressure, which equates to high-intensity training regimens used in systemic exercise.15 However, earlier studies have suggested that quantitative improvements in work capacity following IMT regimens can occur with intensities as low as 40% of peak pressure.16,17 Recent published data using intensities of 80% of peak pressure have shown an increase in lung volumes (vital capacity [VC] and total lung capacity [TLC]), diaphragm thickness, and work capacity in patients with cystic fibrosis18 and in people who are healthy.6,7 However, the effect specifically of lower intensities of IMT in people who are healthy is yet to be determined. Therefore, the primary objective of this investigation was to determine the optimal training intensity required to improve quantitative outcomes (ie, lung volumes, work capacity, power output, and inspiratory pressures) in people who are healthy.


Design Overview

This was a randomized controlled study in which 40 people were allocated to 4 groups. Three training groups comprising of 10 participants in each group (n=30) were required to complete an 8-week supervised program of IMT in which the training intensity was set at 80%, 60%, or 40% of each participant's sustained maximal inspiratory effort. The participants performed no other forms of exercise training during the study period. A fourth group of individuals did not participate in any form of training and acted as a control group (n=10). At the initial screening visits, body composition, pulmonary function, and physical activity status were determined.19 In addition, measurements of inspiratory muscle function, work capacity, and power output were taken in all participants (Tab. 1). These measurements (excluding body composition) were repeated at the end of the 8-week training period and were obtained by independent laboratory-based data collectors who were blinded to the group allocation.

Table 1.

Baseline Measures of Anthropometric Data, Pulmonary Function, Inspiratory Muscle Function, and Work Capacitya

Setting and Participants

This study was conducted in a university-based human movement laboratory. Forty moderately trained university students of both sexes (20 male, 20 female) who were healthy volunteered to take part in this investigation. Each participant's level of physical activity was assessed by questionnaire,19 and an individual was defined as being recreationally active by participating in at least 4 hours per week of sporting activity that was of sufficient intensity to elevate his or her heart rate to within 80% of the age-predicted maximum. The mean age of the participants was 21.7 years (SD=4.0). All participants were nonsmokers and had no evidence of pulmonary pathology (eg, asthma) or any known metabolic or endocrine disorder. All participants were informed of the nature of the study and gave full written consent prior to the study. The flow diagram in Figure 1 contains details of participant eligibility, randomization, and study design.

Figure 1.

Diagram of flow of participants in the study. MIP=maximum inspiratory pressure, SMIP=sustained maximum inspiratory pressure, IMT=inspiratory muscle training.

Randomization and Interventions

This was a single-center controlled study in which the participants were allocated to 4 groups using random number tables.20 Prior to the study, all participants' stature (in centimeters) and weight (in kilograms) were determined using a stadiometer (600–2,100-mm model,* accurate to 1.5 mm) and an electronic beam scale (Inscale electro scale, model MRP200P, accurate to 0.1 kg), respectively. Participants were measured wearing lightweight clothing and no shoes. Percentage of body fat was estimated using skinfold calipers (0- to 48-mm model*) at 4 sites: biceps, triceps, subscapular region, and supra-iliac crest. Three measurements for each site were taken, with the mean used for body fat determination.21 Body fat measurements were calculated according to the equations of Grant et al22 using different formulas based on sex (Appendix).

Lung Function Measurements

Vital capacity, expiratory reserve volume (ERV), functional residual capacity (FRC), TLC, and residual volume (RV) were calculated as per British Thoracic Society standards23 using the helium dilution technique (Vitalograph). All participants were asked to refrain from vigorous exercise for at least 24 hours prior to the tests. During all measurements, participants were seated and a single experienced technician obtained recordings. All lung function measurements were expressed in liters and as percentage of the predicted value for age, height, and sex.24

Assessment of Physical Activity Status

The level of physical activity was assessed before and after IMT using a recall questionnaire.19 Activity scores were calculated over a 24-hour period and expressed in metabolic equivalents (1 MET=3.5 mLO2/kg/min or the resting energy expenditure in one person). Following the completion of the recall questionnaire, all participants were encouraged not to change their physical activity patterns during the study period.

Inspiratory Muscle Function

Each participant's MIP and sustained maximum inspiratory pressure (SMIP) were determined using an electronic manometer and computer connected by serial interface to a laptop computer. This setup had been programmed with a specifically designed computer software package (Respiratory Trainer, model 2 [RT2 device]§). The manometer had a fixed leak via a 2-mm-diameter aperture to prevent glottal closure during the inspiratory maneuver.25 This feature set a maximum flow of 450 mL/s and allowed continuous measurement of pressure over a full range of lung volumes, from RV to TLC, until no further pressure could be generated. This pressure was recorded over time by a computer.

The MIP was the maximum pressure (cm H2O) developed in the first second of inspiration and represented a measure of inspiratory muscle strength. The SMIP represented the integrated area under the pressure-time curve, measured in pressure-time units (PTUs) (Fig. 2).26 All data were stored on the computer database for later retrieval and analysis.

Figure 2.

Sustained maximal inspiratory pressure profile: measured from residual volume to total lung capacity.

Assessment of Work Capacity

At the time of scheduling, all participants were instructed to avoid caffeine and refrain from eating and participating in vigorous activity for at least 3 hours before the test. A progressive, incremental exercise test was performed on an electronically braked cycle ergometer (Excalibur Sport) to measure work capacity as described by Godfrey and Mearns.27 Participants began pedaling with no added resistance and at 1-minute intervals at a self-selected pace. Resistance was added in increments by the technician administering the test and was adjusted for each participant depending on his or her height, with 15-W increments for those participants shorter than 125 cm, 20 W for those 125 to 149 cm tall, and 25 W for those at or above 150 cm. The participants were instructed to continue until they could no longer pedal due to volitional exhaustion. All participants, therefore, exercised to a self-determined maximum. Work capacity was defined as the duration of exercise achieved (in minutes), and power output was defined as the energy expended (in watts) at the end of the protocol. Accuracy of the incremental load was achieved by using microprocessors, which checked the actual workload 5 times per second. The system also contained a feedback mechanism that eliminated the influence of temperature, thereby guaranteeing accuracy of workload up to 1,000 W. The incremental loads for each participant were calculated, and the workload was programmed manually into the system using the Excalibur WorkLoad Programmer according to the manufacturer's instructions. At each work level, heart rate (measured with a Polar RS100 Heart Rate Monitor, model RS800sd#) and ratings of perceived exertion (measured using a modified 0–10 Borg scale)28 were recorded to assess the perceived exercise intensity.

Sample Size Determination and Reliability of the Main Outcome Measures

The reproducibility of the principal outcome variables was determined in 10 individuals who were healthy on consecutive days using methods and an experimental protocol identical to those used in the present study. An adequate sample size was found to be at least 9 individuals in the experimental group at α=.05 and 1−β=90%. Recently published observations from our laboratory29 have demonstrated reproducibility coefficients of .87 for measurements of MIP and .99 for SMIP.

IMT Protocol

A pressure manometer and specifically designed computer software (RT2 device§) were used in the training program. Training was performed 3 times weekly on nonconsecutive days (with at least 24 hours separating training sessions) over 9 weeks, although inspiratory pressure data were not collected until the second week of training to allow the participants to become familiar with the training equipment and protocol.

For each of the training groups, 3 SMIPs were recorded at the commencement of each training session, and the highest sustainable profile was selected automatically and redrawn by the computer as a training template equal to 80%, 60%, or 40% of the maximum pressure profile. This profile was determined by manipulation of the computer software prior to the study and remained constant throughout the training period. Therefore, subsequently at each training session, each participant was re-tested to determine his or her maximal effort, and the computer software was adjusted according to the required training intensity. Inspiratory training maneuvers were repeated using a regimen of 6 repetitions performed within each of the training groups at 80%, 60%, or 40% of the SMIP. Thus, the participants were required to complete a total of 36 repetitions at each training session. During each set, the rest time between repetitions was progressively reduced from 60 to 45, 30, 15, 10, and 5 seconds, as this method has been shown to recruit a larger proportion of muscle fibers and, thus, a larger pool of fibers are trained for subsequent lower but potentially fatiguing loads.30 During the training sessions, the accumulated area under the pressure-time curve was calculated and used as a measure of inspiratory muscle endurance. All training sessions were conducted in a quiet room with no distractions, and the same instructions were given to all participants, thereby ensuring that they were being motivated in a consistent manner during the training period.

Data Analysis

Between-group baseline characteristics (eg, age), anthropometric data (mass, stature, body fat, and body mass index [BMI]), lung function data (VC, FRC, TLC, and RV), inspiratory pressure data (MIP and SMIP), work capacity, and power output were compared with a 1-way analysis of variance (ANOVA). Prior to all analyses, normality of the data was assessed by the one-sample Kolmogorov-Smirnov test. A 2-way repeated-measures ANOVA was used to identify differences before and after training between and within groups for inspiratory pressure data, lung function, work capacity, and power output. For all significant data, unplanned, pair-wise multiple comparisons were made using the Tukey critical difference test. Differences were considered to be significant at P<.05. All statistical calculations were performed using SPSS, version 16.**

Role of Funding Source

The Physiotherapy Research Foundation provided funding to purchase the IMT equipment used in this study.


Study Group Characteristics: Baseline Analysis

All participants had complete data sets of baseline and posttraining measurements of lung function and inspiratory pressure data, including MIP and SMIP. In addition, Borg scale scores, work capacity, power output, and peak heart rate (bpm) were obtained in both training and control groups. There were no differences in age, mass, stature, body composition (BMI and percentage of fat), and physical activity status among the groups at baseline. In addition, there were no differences in the dependent variables of inspiratory pressure, lung volumes (VC and TLC), power output, and peak heart rate among the groups (Tab. 1).

Effects of IMT on MIP and SMIP

There was 100% adherence to the IMT protocols by all participants in the 3 training groups. Following 8 weeks of IMT, an increase in MIP was observed in the groups who trained at 80%, 60%, and 40% of their maximum sustained inspiratory effort. The 80% training group increased MIP from a mean of 68 to 163 cm H2O, and the 60% training group MIP increased MIP from a mean of 73 to 127 cm H2O, representing an increase of approximately 50% for both these training intensities. In the 40% training group, MIP increased with less magnitude (from 76 to 91 cm H2O) compared with the 80% and 60% training groups. The SMIP values also improved in all of the training groups in a similar magnitude to MIP. The 80% training group increased SMIP from a mean of 528 to 1,176 PTUs. The SMIP values increased from 635 to 884 PTUs in the 60% training group and from 544 to 619 PTUs in the 40% training group. There was no change in the control group over time in either MIP or SMIP, resulting in a different group effect following training (Tab. 2).

Table 2.

Group Comparisons for Lung Volumes and Inspiratory Muscle Function Before and After Training Interventiona

Effects of IMT on Lung Function

There were no changes in lung volumes in the 60% or 40% training groups or the control group. However, VC increased in the 80% training group from a mean of 3.4 to 3.8 L, and TLC increased in this training group from a mean of 5.1 to 5.4 L, representing a 7% increase in these variables from pretraining levels. Furthermore, both VC and TLC were different among the groups over time, resulting in a difference in the groups following training. No significant changes occurred in either VC or TLC in any of the other training groups or in the control group (Tab. 2).

Effects of IMT on Work Capacity and Power Output

There were no significant changes in either Borg scale scores or peak heart rate following IMT in any of the training groups or the control group, although significant improvements in work capacity occurred in the 80% and 60% training groups. There was an improvement in the duration of exercise from a mean of 5.0 to 6.2 minutes in the 80% training group. In addition, power output increased from a mean of 125 to 155 W from pretraining levels in the 80% training group. In the 60% training group, the duration of exercise increased from a mean of 5.4 to 6.8 minutes, and power output increased from a mean of 135 to 170 W. No significant change in either work capacity or power output was observed in the 40% training group or the control group over time. There was, therefore, a difference in work capacity and power output between the 3 training groups following IMT and the control group from baseline values (Tab. 3).

Table 3.

Group Comparisons for Peak Work Capacity Before and After Training Interventiona


This study demonstrated that IMT improved inspiratory muscle strength (measured as the MIP) and endurance (measured as the accumulated SMIP achieved during the training protocol). These changes in inspiratory pressures were achieved in all participants who underwent an 8-week period of training at 80%, 60%, or 40% of each individual's MIP, with no changes in these indexes in the participants who acted as a control group. However, quantitative improvements in lung volumes, work capacity, and power output were evident in the 80% training group, whereas the 60% training group improved work capacity and power output only. No improvements in lung volumes, work capacity, or power output were evident in the 40% training group. These data are in agreement with previous research, where lung volumes and work capacity were shown to increase in people who were healthy6 and in age-matched individuals with cystic fibrosis18 who utilized an 8-week, high-intensity inspiratory training protocol. In addition, these data suggest that improvements in volitional tests of inspiratory muscle function alone, which may be evident following IMT, do not necessarily translate into quantitative improvements in pulmonary function or work capacity.31

There is now considerable evidence that IMT improves pulmonary function,6 exercise performance,712 and recovery times following sprint performance in healthy athletic populations.9 These outcomes have been achieved using a variety of training methods, including 4 weeks of isocapnic hypernea,7 6 weeks of volitional hypernea,13 and 6 weeks of using a respiratory resistance (threshold) device equivalent to 50% MIP.9,11 The functional improvements in work capacity and power output in the participants who trained at 60% and 80% of MIP are in agreement with the findings of these earlier investigations. However, in the present investigation, only participants in the 80% training group achieved improvements in lung volumes.

In accordance with the early work of Belman and Shadmehr,32 and in contrast to the use of a threshold-loading device, the present study used a pressure-flow–based training program. This program is designed to increase both pressure generation and inspiratory flow throughout the training maneuver and as an outcome of training. The training intervention in this study was successful in achieving a sustained training intensity, which is principle consistent with the overload principle.14 However, during IMT regimens, the effect of a learning response cannot be ignored. Analysis of the MIP and SMIP data for all of the training groups indicated a learning response in the first few weeks of training despite a 1-week habituation period. This learned response could be attributable to an improved neuromuscular recruitment pattern, which is a well-described mechanism for early improvements in strength training, and may partially explain the large magnitude of change in MIP and SMIP over the 8-week training period.33 However, the participants in the group that trained at 80% of MIP increased their VC and TLC, which indicates an increased ability of the inspiratory muscles to expand the thorax following training. The increase in these lung volumes also may result from a greater contribution of the upper thorax and neck muscles to the inspired volume after training.6,18

These findings of increases in VC and TLC are in agreement with the findings of an earlier study by Leith and Bradley.34 Their participants trained for a 5-week period for gains in either endurance (4 participants performed voluntary normocarbic hyperpnea to exhaustion) or strength (4 participants performed repeated static maximum inspiratory and expiratory maneuvers against obstructed airways). Although this study34 was designed to demonstrate how ventilatory muscle strength or endurance can be increased by appropriate ventilatory muscle training programs, increases in VC and TLC of 4% were observed only in the participants who trained for strength at an appropriate intensity. The finding of no increase in lung volumes in the participants who trained for endurance only (at an intensity of approximately 20%) is in agreement with the findings of previous studies where a similar training intensity failed to elicit changes in lung volumes in people who were healthy6 or in patients with cystic fibrosis.18

In conjunction with the improvements in lung volumes in the group who trained at 80% of SMIP, there also were increases in work capacity and power output as assessed by cycle ergometry.27 These increases also were evident in the group that trained at 60% of SMIP. The nonsignificant change in Borg scale scores before and after the IMT protocol in both training groups may reflect the participants' ability to sustain a higher workload without an increase in breathlessness. However, it has been established that respiratory muscle fatigue occurs during exercise at an intensity of at least 85% of maximal oxygen consumption3539 and has been shown to impair exercise performance.36 This impairment in exercise capacity has been attributed to possible limb muscle vasoconstriction and reduction in limb blood flow elicited by a metaboreflex originating in the diaphragm, causing systemic vasoconstriction during periods of inspiratory muscle fatigue.2,38,39

The participants included in this study were healthy and recreationally active and, therefore, should have been able to cycle for a relatively long duration until the onset of leg fatigue. However, they achieved only approximately 75% of predicted peak heart rate (200 − age) and achieved only approximately 5 to 6 minutes of exercise. According to the American College of Sports Medicine,40 the protocol used achieved only a measure of work capacity rather than a measure of aerobic exercise capacity, which may have been attributable to lack of motivation by the participants during the exercise test. However, despite this lack of adherence to the protocol as described in this study,27 these findings further add strength to the rationale for IMT, particularly in patients with inspiratory muscle weakness or fatigue. Accepting the limitations of cycle ergometry without the analysis of expired gas analysis to measure functional improvements following IMT, the impact of this regimen of IMT on the improved work capacity and power output suggests that IMT may improve functional capacity in people who are healthy.

Studies of IMT have remained controversial due to the inadequacy of some study designs. For example, some studies have omitted control groups,16,32 thus preventing the efficacy of IMT from being fully identified. The criteria for methodological quality established by Smith et al,41 namely, the use of random sampling, comparable groups, comparable co-intervention, and standardization of testing techniques, were all observed in the present study. Consequently, the true efficacy of IMT could be judged. In addition to the attention to study design, the training program used in this study utilized a technique of incremental loading of the inspiratory muscles where the workload was fixed and reassessed at each training session. This technique was achieved by selecting the best of 3 maximum sustained inspiratory efforts at the commencement of each training session in each participant to maintain overload. The program also required the participants to work through their full inspiratory volume from RV to TLC, thereby maintaining consistency with the volume at which the training was applied. Furthermore, the use of computer software to run the training program maintained consistency of effort and loading, with the additional advantage of accurate recordings of training levels that were independent of observer input, allowing checks on adherence to the training process.

Unlike previous investigations,6,18,42,43 however, this study failed to evaluate the effects on functional capacity of IMT with intensities below 40% of SMIP. It was considered justified to exclude these training groups, as suboptimal training loads of 20%6,18 and 10%42,43 have failed to elicit changes in any quantitative functional outcomes other than changes in indexes of inspiratory muscle function. This finding has been evident both in patients with chronic obstructive pulmonary disease42 and in people who are healthy.43 However, although this study failed to demonstrate changes in work capacity and power output in the group who trained at 40% of SMIP, it may be possible that this study was underpowered to detect changes in functional capacity in this group of participants. Indeed, it may be possible that intensities of 50% of SMIP or even lower may improve exercise capacity, as the sample size calculation for the present investigation was based upon previous studies that used diaphragm thickness and inspiratory pressure data as outcomes for sample size estimation.6,18 This contention is supported by earlier studies16,17 that demonstrated modest improvements in work capacity, albeit in participants with impaired exercise capacity due to chronic obstructive pulmonary disease.

In conclusion, this study showed that improvements in volitional tests of muscle function alone may not be adequate evidence that IMT is effective.31 However, this study showed that if substantial pressures are generated during a regimen of high-intensity IMT, significant improvements in lung volumes, work capacity, and power output may be achieved in people who are healthy.

The Bottom Line

What do we already know about this topic?

Inspiratory muscle training has been shown to improve inspiratory muscle function, lung volumes, and physical work capacity in people with chronic lung disease and in people who are healthy; however, the optimal training intensity to produce these outcomes is unclear.

What new information does this study offer?

Inspiratory muscle training at low, moderate, or high intensities (40%, 60%, and 80% of sustained maximal inspiratory effort) is beneficial in improving inspiratory muscle function, and training at moderate or high intensities improves physical work capacity. Only high-intensity training (80%) also provides gains in lung volume.

If you're a patient, what might these findings mean for you?

Improvements in tests of inspiratory muscle function alone may not be adequate evidence that inspiratory muscle training is effective; however, if substantial pressures are generated during inspiratory muscle training, significant improvements in your lung volume and physical work capacity may be achieved.



Formulas for Calculation of Body Mass Index and Percentage of Body Fat


  • Dr Enright provided the concept and research design, data collection and analysis, project management, fund procurement, participant recruitment, facilities and equipment, consultation and institutional liaisons, and preparation of the manuscript for publication. Dr Unnithan provided original concept/idea/research design, writing, data analysis, and consultation (including review of the manuscript before submission).

  • This study was approved by the School of Healthcare Studies Research and Development Ethics Committee at Cardiff University.

  • A platform presentation of this study was given at the 19th Annual Congress of the European Respiratory Society; September 13, 2009; Vienna, Austria.

  • This study was supported by the Physiotherapy Research Foundation.

  • This study is registered with the ISRCTN: no. 26277638.

  • * Holtain Ltd, Crosswell, Crymych, Swansea, SA41 3UF United Kingdom.

  • Inscale Measurement Technology Ltd, 7 Heron Close, St-Leonards-on Sea, East Sussex TN38 8DX, United Kingdom.

  • PK Morgan Ltd, Rainham, Kent KF62 5MD, United Kingdom.

  • § DeVilbiss UK Ltd, Sunrise Business Park, High Street, Wollaston, Stourbridge, West Midlands DY8 4PS, United Kingdom.

  • Medical Graphics Corp, 350 Oak Grove Pkwy, St Paul, MN 55127.

  • # Polar RS Sports,4802 Glenwood Rd, Brooklyn, NY 11234.

  • ** SPSS Inc, 233 Wacker Dr, Chicago, IL 60606.

  • Received December 15, 2009.
  • Accepted February 18, 2011.


View Abstract