The Effect of Breathing Pattern

Retraining on Performance in
Competitive Cyclists

Rachel Vickery

A thesis submitted to Auckland University of Technology in partial

fulfillment of the requirements for the degree of Master of Health Science
(MHSc)

October 2007

Division of Rehabilitation and Occupational Therapy

Primary Supervisor: Dr Andrew Kilding

Table of Contents

Table of Contents .............................................................................................................................i

List of Tables .................................................................................................................................. iii
List of Figures .................................................................................................................................iv
List of Appendices ...........................................................................................................................v
Attestation of Authorship.................................................................................................................vi
Dedication...................................................................................................................................... vii
Acknowledgements....................................................................................................................... viii
Intellectual Property Rights.............................................................................................................ix
Abstract ...........................................................................................................................................x
Chapter One: Introduction ............................................................................................................. 1
1.1. Aim of The Study.......................................................................................................... 5
Chapter Two: Literature Review. The Development, Consequence and Managementof Abnormal
Respiratory Mechanics in Athletes.................................................................................................. 6
2.1. Cycling Performance, Ventilation and Breathing .......................................................... 6
2.2. Breathing Pattern Disorders and Abnormal Respiratory Mechanics............................. 8
2.2.1. Normal Breathing at Rest ................................................................................. 9
2.2.2. Abnormal Breathing at Rest ........................................................................... 10
2.2.3. Respiratory Response to Exercise ................................................................ 12
2.2.4. Abnormal Breathing Patterns in Athletes........................................................ 20
2.2.4.1 Environmental and Pathological Effects on Breathing................................. 20
2.2.4.2 Psychological Effects on Breathing ............................................................. 21
2.2.4.3 Postural effects on breathing....................................................................... 22
2.3. Consequences of Altered Respiratory Mechanics ...................................................... 24
2.3.1. Respiratory Timing ......................................................................................... 25
2.3.2. Dynamic Hyperinflation .................................................................................. 26
2.3.3. Expiratory Flow Limitation .............................................................................. 27
2.3.4. Work of breathing........................................................................................... 28
2.3.5. Respiratory Muscle Fatigue............................................................................ 29
2.3.6. Blood Stealing ................................................................................................ 33
2.3.7. Dyspnoea ....................................................................................................... 39
2.4. Breathing as a Limiting Factor to Exercise and Strategies to Overcome it ................. 42
2.4.1. Breathing as a Limiting Factor........................................................................ 42
2.4.2. Inspiratory Muscle Training ............................................................................ 44
2.4.3. Breathing Pattern Retraining .......................................................................... 53
2.5. Summary .................................................................................................................... 60
Chapter Three: Methods.............................................................................................................. 62
3.1. Participants................................................................................................................. 62
3.2. Equipment .................................................................................................................. 63
3.2.1. Ergometer ...................................................................................................... 63
3.2.2. Pulmonary Gas Exchange and Spirometry .................................................... 64
3.2.3. Blood Lactate ................................................................................................. 64
3.2.4. Borg Scale...................................................................................................... 65
3.2.5. Nijmegen Questionnaire................................................................................. 65
3.2.6. Breathing Harness ......................................................................................... 65

i
 3.3. Experimental Procedures ........................................................................................... 66
3.3.1. General Protocol ............................................................................................ 66
3.3.2. Spirometry...................................................................................................... 68
3.3.3. Incremental Power Step Test ......................................................................... 70
3.3.4. Nijmegen Questionnaire................................................................................. 71
3.3.5. 20km Time Trial ............................................................................................. 72
3.3.6. Intervention Sessions - Randomisation .......................................................... 73
3.3.7. Intervention Sessions - Intervention Group .................................................... 73
3.3.8. Intervention Sessions - Control Group ........................................................... 74
3.4. Data Analysis ............................................................................................................. 75
3.4.1. Pmax determination.......................................................................................... 75
3.4.2. Lactate ........................................................................................................... 75
3.4.3. Sub-maximal and maximal VO2...................................................................... 75
3.4.4. Nijmegen Questionnaire................................................................................. 76
3.4.5. Statistics......................................................................................................... 77
Chapter Four: Results.................................................................................................................. 78
4.1. Participant Characteristics.......................................................................................... 78
4.2. Effects on cycling performance (Incremental power and 20km-TT performance)....... 79
4.3. Effect on Physiological Measures............................................................................... 79
4.4. Effect on Respiratory Measures ................................................................................. 80
4.4.1. Comparison of measures at absolute workloads pre and post-BPR .............. 81
4.5. Changes in RPE measures ........................................................................................ 84
4.5.1. RPE................................................................................................................ 84
4.5.2. Reason for terminating exercise..................................................................... 85
4.5.3. Correlations between RPEbreathing, RPElegs, fb and performance changes....... 85
Chapter Five: Discussion............................................................................................................. 86
5.1. Effect of BPR on cycling performance ........................................................................ 87
5.1.1. Effects on 20km-TT performance................................................................... 87
5.1.2. Effects on Incremental Peak Power (Pmax) ..................................................... 90
5.2. Effects of BPR on Respiratory Measures ................................................................... 91
5.2.1. Breathing frequency (fb) and Tidal Volume (VT) ............................................. 92
5.2.2. Inspiratory and Expiratory Time ..................................................................... 93
5.2.3. Minute Ventilation........................................................................................... 94
5.2.4. Inspiratory Drive and Timing ........................................................................ 100
5.3. Effects of BPR on Perceived Effort........................................................................... 101
5.3.1. Perceived Breathing Effort ........................................................................... 101
5.3.2. Perceived Leg Effort..................................................................................... 104
5.4. Effects of BPR on Metabolic Measures .................................................................... 105
5.4.1. Lactate, Power and Heart Rate – LT and LTP ............................................. 105
5.4.2. Measures of Oxygen Uptake - sub-maximal (economy) and VO2peak ........... 106
5.5. Limitations ................................................................................................................ 108
5.6. Future Directions ...................................................................................................... 110
5.7. Conclusion............................................................................................................... 111
Appendices................................................................................................................................. 112
References ................................................................................................................................. 123

ii
List of Tables

Table 1. Muscles and Structures of Respiration (From Chaitow et al., 2002) ................................ 9

Table 2. Effect of Breathing Rate and Depth on Alveolar Ventilation in Three Scenarios (from
Marieb, 1992) ................................................................................................................ 15

Table 3. Different Methods of IMT ............................................................................................... 45

Table 4. A summary of Exercise Performance Studies using Inspiratory Muscle Training (IMT). 46

Table 5. Overview of Breathing Pattern Retraining Studies......................................................... 59

Table 6. Participant Anthropometric Characteristics .................................................................... 63

Table 7. Base Line Performance and Physiological Characteristics for Intervention and Control
Groups*......................................................................................................................... 78

Table 8. Mean Changes in Performance Measures post Breathing Pattern Retraining and Control
and Chances That the True Difference in the Changes Is Substantial .......................... 79

Table 9. Mean changes in Physiological Measures post Breathing Pattern Retraining and Control
and Chances That the True Difference in the Changes is Substantial .......................... 80

Table 10. Mean Changes in Respiratory Measures (non-exercise and exercise) post Breathing
Pattern Retraining and Control and Chances That the True Difference in the Changes Is
Substantial .................................................................................................................... 81

Table 11. Mean Changes in Rate of Perceived Exertion Post Breathing Pattern Retraining and
Control and Chances That the True Difference in the Change is Substantial ............... 85

iii
List of Figures

Figure 1. Ventilatory response of the elite (E) and professional (P) cyclists during the tests:
pulmonary ventilation (VE), breathing frequency (fb) and) tidal volume (VT). Max Maximal
power output. * P < 0.05 ** P < 0.01, § significant (P < 0.05) interactive effect (group x
exercise intensity) (From Lucia et al., 1999).................................................................. 17

Figure 2. Inspiratory (tI) and expiratory (tE) times. Definitions as for Figure 2.1. * P < 0.05, ** P <
0.01, § significant (P < 0.05) interactive effect (group x exercise intensity. (From Lucia et
al., 1999). ...................................................................................................................... 17

Figure 3. A typical athlete’s cycling position during a road time-trial (reprinted with the permission
of Stephen Sheldrake, NZ Triathlete) ............................................................................ 22

Figure 4: Breathing Harness......................................................................................................... 66

Figure 5. Time line (weeks) of familiarisation trials (F), formal incremental step test (ST) and
20km time trial (TT) and intervention period (I). ............................................................ 67

Figure 6. Effect of breathing pattern retraining on breathing frequency, tidal volume and minute
ventilation in the intervention (left panel) and control groups (right panel). ................... 82

Figure 7. Effect of Breathing pattern retraining on inspiratory and expiratory time in the
intervention (left panel) and control (right panel) groups. .............................................. 83

Figure 8. Effect of breathing pattern retraining on inspiratory drive (VT/tI) and inspiratory timing
(tI/tTOT ) in the intervention (left panel) and control (right panel) groups. ........................ 82

Figure 9. Effect of breathing pattern retraining on blood lactate in the intervention (left) and
control (right) groups. .................................................................................................... 82

Figure 10. Effect of breathing pattern retraining on VO2 and VCO2 in the intervention (left panel)
and control (right panel) groups. ................................................................................... 84

iv
List of Appendices

Appendix 1 Nijmegen Questionnaire ......................................................................................... 113

Appendix 2 Clinical Breathing Retraining................................................................................... 115

Appendix 3 Breathing Retraining ............................................................................................... 118

Appendix 4 Training Diary ......................................................................................................... 119

Appendix 5 Modified Borg Scale................................................................................................ 120

Appendix 6 Patient Handout – Potential Symptoms associated with BPR................................. 121

Appendix 7 Participant Handout – Rationale for Exercises........................................................ 122

v
Attestation of Authorship

I hereby declare that this submission is my own work and that, to the best of my knowledge and

belief, it contains no material previously published or written by another person nor material which

to a substantial extent has been accepted for the qualification of any degree or diploma of a

university or other institution of higher learning, except where due acknowledgement is made in

the acknowledgements.

Signed: _______________________________

Date: __________/__________/_________

vi
Dedication

This thesis is dedicated in the first instance to Neil Dearberg, one of life’s true gentlemen,

whose love, support, encouragement and patience know no boundaries.

I also dedicate this thesis to my parents, Evan and Heather Vickery, who are not only

incredible parents, but are amazing, inspirational people. I am proud to be your daughter.

vii
Acknowledgements

My primary supervisor, Dr Andrew Kilding. For taking on a physio with no knowledge of exercise
physiology – what were you thinking! For your guidance, teaching, critiquing and encouragement
to always find that next level – thank you.

My secondary supervisor, (soon to be Dr) David Nicholls. For never being afraid to say “I told you
so”. For always being there to help, advise, inspire. For encouraging me to soar, or to reign me
in when I needed it. For your years of friendship – thank you.

My family, especially my sister Andrea – thank you for everything!

To Judith Paterson. Thank you for your friendship and incredible generosity in providing me a
home away from home where I have always been welcome. This past 10 months would not have
been possible without you.

To Quinton and Vicky Swanson, and Jamie Singe at Contact Self Defence. For providing the
physical outlet to balance this mental challenge! For making me part of the family and for your
friendship. You will never truly know what a difference you have made to my year.

To the post-grad students, especially Joe McQuillan, for taking me under their wing and showing
me the physiology ropes.

To all the participants in the study. What an amazing bunch of guys and such commitment! It
wouldn’t have happened without you all.

To the Auckland University of Technology Ethics Committee for granting ethical approval – 21st
December 2006; 06/207.

To Zephyrtech (Auckland, NZ) for providing the breathing harness used with the control group.

viii
Intellectual Property Rights

This thesis may be utilised for the purpose of research or private study provided that due

acknowledgement is made where appropriate and that the author’s written permission is obtained

before any material from the thesis is published.

I agree that the AUT University library may copy this thesis, in electronic or photocopied form, to

supply the collection of other libraries on request from those libraries. I also agree that this thesis

may be photocopied for supply to any person in accordance with the provision of Section 56 of the

Copyright Act 1994.

ix
Abstract
Background: The increased work of breathing associated with intense cycling has been identified
as a factor that may negatively affect cycling performance. The aerodynamic position, abnormal
respiratory mechanics either at rest or during exercise, and the development of a tachypnoeic
breathing pattern are factors known to increase the work of breathing. Breathing pattern
retraining aims to decrease the work of breathing by delaying the onset of dynamic hyperinflation
and the recruitment of accessory breathing muscles. To date no studies have investigated the
performance, physiological and perceptual consequences of manipulating breathing pattern in
trained cyclists. Purpose: The aim of the present study was to investigate the effect of breathing
pattern retraining on 20-km time trial performance and respiratory and metabolic measures in
competitive cyclists. Method: Twenty-four competitive male cyclists (age 37.7 ± 8.6 years, mean
± SD; V&O 2 peak 4.34 ± 0.47 L—min-1) were match paired on 20-km time trial performance and
assigned at random to either an intervention group (breathing pattern retraining; N = 12) or control
group (N = 12). 20-km time trial performance, pulmonary function and the physiological and
perceptual response during a maximal incremental cycle step test were assessed pre- and post-
intervention. The intervention group underwent four weeks of specific breathing pattern retraining
using exercises designed to reduce dynamic hyperinflation and optimise respiratory mechanics.
The control group attended the laboratory once a week during this period and performed a 10
minute sub-maximal ride wearing a biofeedback breathing harness. The control group was led to
believe the purpose for their participation was to investigate the effect that maximal exercise had
on breathing pattern, and to test the reliability of the breathing harness. There was no attempt to
modify the breathing pattern of the control group. Data were analysed using an MS Excel
spreadsheet designed for statistical analysis. The uncertainty in the effect was expressed as 90%
confidence limits and a smallest worthwhile effect of 1.0% was assumed. Results: The
intervention group showed substantial improvements in 20-km time trial performance (-1.5 ±
1.1%) and incremental power (3.2 ± 3%). Additionally, breathing frequency (-13.2 ± 8.9%; -9.5 ±
8.4%), tidal volume (10.6 ± 8.5%; 9.4 ± 7.6%), inspiratory time (10.1 ± 8%; 9.4 ± 7.7%), breathing
RPE (-30 ± 33.9%; -24.7 ± 28.1%) and leg RPE (-27.9 ± 38.5%; -24.7 ± 28.2%) were all
positively affected at lactate threshold and lactate turn point. No positive changes were observed
in the control group for 20-km time trial performance (0.0 ± 1.0%), incremental power (-1.4 ±
3.5%), breathing frequency (-1.6 ± 8.0%; -2.0 ± 7.9%), tidal volume (0.9 ± 7.2%; 2.9 ± 9.4%),
breathing RPE (16.1 ± 50.2%, 24.8 ± 43%) or leg RPE (13.4 ± 39.6%; 19.9 ± 43.2%) .
Conclusion: These results provide evidence of the performance enhancing effect of four weeks
of breathing pattern retraining in cyclists. Furthermore, they suggest breathing pattern can be
retrained to exhibit a controlled pattern, without a tachypnoeic shift, during high intensity cycling.
Additionally, these results indicate breathing pattern retraining attenuates the respiratory and
peripheral perceived effort during incremental exercise.

Key words: Breathing pattern disorders, retraining, blood stealing, cycling, performance, power
output, respiratory mechanics, perceived exertion, 20km-TT

x
Chapter One: Introduction

In the 2004 Olympic men’s road cycling time trial (48km), the time difference between winning the

gold medal and second place was 18.84 seconds (0.5%). Because small improvements in

performance at the top level of cycling can make a significant difference to overall results, cyclists,

coaches and sports scientists are continually looking for ways to facilitate these performance

gains such as employing new physical training methods (Paton & Hopkins, 2005), consuming

nutritional supplements (Wiles, Coleman, Tegerdine, & Swaine, 2006) and manipulating

equipment (Belen, Habrard, Micallef, & Le Gallais, 2007). An area that appears to be overlooked

is the effect that the mechanics of breathing may have on a cyclist’s performance.

Abnormal breathing patterns, or breathing pattern disorders, may be the long term consequence

of altered respiratory mechanics. Dysfunctional respiratory mechanics during exercise may be

due to:

i) pathological factors such as exercise-induced asthma, exercise-induced

bronchospasm, vocal cord dysfunction, acute respiratory illness;

ii) emotional factors, such as anxiety;

iii) biomechanical factors for example the postures adopted by athletes in their

sport and/or;

iv) physiological factors, such as the changes that occur to breathing frequency

and tidal volume during the normal ventilatory response to exercise.

Either alone or in combination, these factors may cause the development of a chronic breathing

pattern disorder. Of particular interest to this study is the development, and potential

1
consequences, of a chronically abnormal breathing pattern in cyclists whose breathing pattern

becomes maladaptive.

The consequences of a breathing pattern disorder may have a detrimental effect on a cyclist’s

performance for several reasons. Altered respiratory mechanics during exercise may lead to a

tachypnoeic shift (increased breathing frequency at the expense of maintaining tidal volume; Lucia,

Carvajal, Calderon, Alfonso, & Chicharro, 1999), increased dynamic hyperinflation (Johnson,

Saupe, & Dempsey, 1992; McConnell, 2005) and a subsequent increase in the work of breathing

(Harms et al., 1998). Consequently, an increased work of breathing may cause blood stealing

(Sheel et al., 2001), a phenomenon where blood is diverted from working peripheral muscles to

the respiratory muscles to meet increased demand. Such decreased blood flow may negatively

affect the ability of the peripheral muscles to generate power (Romer & Dempsey, 2006).

Additionally, an increased work of breathing can result in respiratory muscle fatigue (Johnson,

Aaron, Babcock, & Dempsey, 1996) and increased dyspnoea (Grazzini, Stendardi, Gigliotti, &

Scano, 2005), both of which may have an adverse affect on performance.

Traditionally, the idea that breathing may be a limiting factor in high intensity exercise has been

disregarded due to the apparent breathing reserve observed in athletes even at the highest level

of exertion (Dempsey, 1986). However, with evidence suggesting expiratory flow limitation may

occur in elite athletes (Johnson, Saupe, & Dempsey, 1992) and that the work of breathing during

intense exercise may cause blood stealing (Harms, Wetter, St Croix, Pegelow, & Dempsey, 2000;

Sheel et al., 2001), attention has been brought to the possibility that breathing may be a limitation

to performance in athletes (Dempsey, 2006b). The apparent success of techniques such as

inspiratory muscle training (IMT) (Dellaca et al., 2001; Romer, McConnell, & Jones, 2002a) and

2
voluntary isocapnic hyperpnoea (VIH) (Holm, Sattler, & Fregosi, 2004; Leddy et al., 2007;

McMahon, Boutellier, Smith, & Spengler, 2002) in improving performance in endurance sport

suggests that specific intervention strategies can be employed to overcome this limitation.

Although strong evidence exists to suggest that IMT is effective in increasing performance by

increasing the strength and endurance of the respiratory muscles and reducing inspiratory muscle

fatigue (Romer, McConnell, & Jones, 2002a, , 2002b; Volianitis et al., 2001), IMT in itself does not

address the pattern of breathing. This could be likened to increasing the leg strength and power

of a cyclist, without addressing pedalling technique.

The ability to maintain an even, controlled breathing pattern during high exercise intensities,

without defaulting to a tachypnoeic shift, has been identified as an important characteristic of

professional cyclists (Lucia, Carvajal, Calderon, Alfonso, & Chicharro, 1999). It is not known,

however, whether this breathing pattern is a physiological adaptation (e.g. to years of high

intensity training,) or whether it could be likened to a learned skill and could therefore be

manipulated.

Breathing pattern retraining is a technique used by physiotherapists to correct abnormal

respiratory mechanics and breathing pattern disorders (Blager, 2000; Chaitow, Bradley, & Gilbert,

2002; Jones, Dean, & Chow, 2003). Although used typically in the management of patients with

respiratory pathology, the concept and application is similar for athletes. Breathing pattern

retraining may benefit cyclists in two ways - firstly by correcting any abnormal respiratory

mechanics or breathing pattern disorders that are present at rest, and secondly by attenuating the

consequences of poor respiratory mechanics that often occur during intense exercise (Fallon,

2004; 2007).

3
Whilst breathing pattern retraining techniques are used clinically by physiotherapists to address

breathing pattern disorders in athletes, there are no known published studies documenting the

ergogenic potential of this technique in this population. Further, the effect that optimising

respiratory mechanics may have on exercise performance in cyclists, by attempting to attenuate

the negative consequences of an increased work of breathing, has not been reported. The aim of

Chapter Two (Review of Literature) is to introduce the reader to the concept and consequences of

abnormal breathing patterns in athletes and to discuss how breathing pattern retraining may not

only improve the efficiency of breathing in cyclists, but also facilitate increased cycling

performance. In consideration of how or why breathing pattern retraining may potentially enhance

cycling performance, a brief description of normal and abnormal breathing patterns is provided

and the prevalence of breathing disorders in athletes will be discussed. In addition, the theories

and evidence that underpins the ventilatory response to exercise, subsequent changes to the

work of breathing and consequences of this, will be presented. Finally, strategies that may

decrease the effect of altered respiratory mechanics during exercise, including respiratory muscle

training and breathing pattern retraining, will be explored.

Using breathing pattern retraining to manipulate breathing in athletes is a novel area of research.

Indeed there is a paucity of literature that objectively defines breathing pattern retraining and its

effectiveness in addressing breathing pattern disorders. There is currently no evidence to either

support or refute its effectiveness as a performance enhancing technique in athletic populations.

The aim of this thesis is to address these gaps by investigating the use of breathing pattern

retraining techniques in cyclists.

4
1.1. Aim of The Study

The aim of this study was to investigate the effect of four weeks of breathing pattern retraining on

20-km time-trial performance, and on respiratory and metabolic measures during an incremental

step test in competitive cyclists.

5
Chapter Two: Literature Review. The Development, Consequence
and Management of Abnormal Respiratory Mechanics in Athletes

2.1. Cycling Performance, Ventilation and Breathing

The difference between winning an Olympic gold medal and second place in the men’s 48km time

trial in 2004 was 18.84 seconds (0.5%) (www.cyclingnews.com). Even the smallest improvement

in performance at the top level can make a significant difference to the overall result (Jeukendrup

& Martin, 2001). Cycling performance is dependent on physiological factors which influence

mechanical power production (for example training modification, nutrition and possibly breathing

pattern) (Lucia et al., 1999), in addition to mechanical and environmental factors that affect power

demand (for example body mass, body position, bicycle) (Jeukendrup & Martin, 2001). These

factors are even more important in an individual time trial (TT) where race tactics play a lesser

part (Jeukendrup & Martin, 2001) and physical performance factors come to the fore. For

example, the effects of modified training on a 40km-TT in already well-trained individuals has

been reported to be 2-4% (Jeukendrup & Martin, 2001) which could make a difference of 70-140

seconds over 40km. This time difference applied to the 2004 Olympic results described above

would be the difference between first and fourteenth place (www.cyclingnews.com).

Competitive road cyclists may take part in a one day race, or multistage events which are

characterised by back-to-back days of racing including mass start races, individual and team time

trials. Of all the road race formats, the individual time trial is considered the most physically

demanding (Mujika & Padilla, 2001) and is a component of major tour races (e.g. the Tour de

France), or in some cases is a stand-alone event (e.g. World Championships and Olympic

Games). An individual time trial demands each cyclist cover the required distance in the shortest

6
possible time by sustaining the highest average power output that will allow them to gain a time

advantage over their opponents (McDonald, 2007).

Mass start races allow drafting, which means a cyclist can ride in the slipstream of the rider(s)

ahead to decrease their resistance (Faria, Parker, & Faria, 2005b). Drafting can decrease energy

utilization by up to 40% (Faria, Parker, & Faria, 2005b), however drafting is not allowed in time

trial events, meaning cyclists must be able to sustain high energy outputs for prolonged periods.

Although many factors influence cycling performance, including years of training (Coyle et al.,

1991), cycling economy (Lucia, Hoyos, & Perez, 2002) and V&O 2 max (Lucia et al., 2006), power

output at lactate threshold and peak power output indicating a power-to-weight ratio of greater or

equal to 5.5W/kg have been suggested to be the physiological markers most predictive of cycling

performance (Faria, Parker, & Faria, 2005).

In an attempt to make small improvements to performance, numerous studies have investigated

ways in which power output can be maximised in cyclists. These include hypoxic training (Clark

et al., 2007), recovery strategies (Maxwell, Castle, & Spencer, 2007), aerodynamics (Grappe,

Candau, & Belli, 1997) and inspiratory muscle training (Johnson, Sharpe, & Brown, 2007). Whilst

attention has been paid to ventilatory aspects, including the changes that occur in breathing

pattern during exercise (Dempsey, 2006a; Lucia, Carvajal, Calderon, Alfonso, & Chicharro, 1999;

Lucia, Hoyos, & Perez, 2002; Lucia et al., 2006), no attention has been paid to manipulating

breathing pattern on a long term basis. Given the reported high prevalence of breathing disorders

in athletes (Fallon, 2004; Mendelli, Lounana, Messan, Menuet, & Petitjean, 2006; Rundell &

Spiering, 2003) and the changes to respiratory mechanics that occur during exercise (Johnson,

7
Saupe, & Dempsey, 1992; Verges, Notter, & Spengler, 2006; Ward, 2007), it is surprising that no

study has focused on the potential improvements in cycling work output and performance that can

be made as a result of a breathing pattern intervention, or by improving a maladaptive breathing

pattern which is commonly induced during high intensity exercise.

2.2. Breathing Pattern Disorders and Abnormal Respiratory Mechanics

A breathing pattern disorder (BPD) is a general term used to describe the consequences of long-

standing abnormal respiratory mechanics. Little is published in the medical and scientific

literature describing BPDs; what they are, how they are established and how they affect otherwise

healthy individuals. Even less is known about how BPDs and abnormal respiratory mechanics

may affect athletes and, specifically to this review, cyclists. Breathing in cyclists may be

compromised in a number of ways (e.g. altered mechanics of breathing caused by the cyclists

unique exercising posture, changes in respiratory timing and/or muscle recruitment pattern due to

the normal physiological response to exercise, ventilation-perfusion mismatching, diffusion

limitations and increased intra- and post-pulmonary shunt) (Dempsey & Wagner, 1999; Stickland

& Lovering, 2006), and while most of these are beyond the scope of this review, it is possible that

with regard to respiratory mechanics, a cyclist’s breathing may be compromised by the presence

of a breathing pattern disorder at rest, or the changes to respiratory timing and respiratory muscle

recruitment pattern that occurs with the normal respiratory response to exercise. To gain further

understanding of the aetiology of BPDs, this section will focus on the mechanics of normal and

abnormal breathing at rest and during exercise. The prevalence of abnormal breathing patterns in

athletes and contributing factors will also be discussed.

8
 2.2.1. Normal Breathing at Rest

Breathing at rest, in people without respiratory illness, requires less than five percent of whole-

body oxygen consumption ( V&O 2 ) (Jones, Dean, & Chow, 2003). This proportion of V&O 2 enables

the respiratory muscles to generate appropriate pressure changes within the thoracic cavity to

overcome a combination of airway resistance and lung compliance, resulting in inspiration. As the

primary muscle of inspiration, the diaphragm is normally responsible for 95% of the work of quiet,

tidal breathing, with minimal contribution from the accessory muscles (Faithfull, Jones, & Jordan,

1979) (Table 1).

Table 1. Muscles and Structures of Respiration (From Chaitow et al., 2002)

Muscle of Inspiration Muscles of Expiration
Primary Primary
Diaphragm Elastic recoil of lungs, diaphragm, pleura and
Intercartilaginous Internal intercostals costal cartilages
Upper and more lateral external intercostals
Levatores costarum
Scalenes

Accessory Accessory
Sternocleidomastoid Interosseous internal intercostals
Upper trapezius Abdominal muscles (rectus abdominus, internal
Serratus anterior (arms elevated) and external obliques, transverse abdominus)
Latissimus dorsi (arms elevated) Transversus thoracis
Serratus posterior superior Subcostales
Iliocostalis Iliocostalis lumborum
Subclavius Quadratus lumborum
Omohyoid Serratus posterior inferior
Latissimus dorsi

Assuming optimal respiratory mechanics, the diaphragm descends into the abdominal cavity

during inspiration, pushing the abdominal contents downwards and forwards, increasing the

vertical dimensions of the chest cavity, as the ribs are lifted and move out to increase the

transverse dimensions (Chaitow & Bradley, 2002). Expiration during quiet breathing is normally

passive and as the diaphragm relaxes the abdomen and chest wall return to their starting position.

9
Minute ventilation (VE) is the volume of gas ventilated in one minute and is the product of tidal

volume and respiratory rate. The primary aim of VE is to meet the metabolic demands of the body

by moving oxygen (O2) into and carbon dioxide (CO2) out of the lungs in the most efficient way

(Abernethy, Mackinnon, Neal, Kippers, & Hanrahan, 1996). It is suggested that meeting the

metabolic demands of the body at rest (which requires approximately 6 L•min-1; Webber & Pryor,

1994) in the most efficient way, is achieved using a steady, rhythmical pattern of the diaphragm; a

respiratory rate of 10-14 breaths—min-1; and an inspiratory to expiratory ratio of 1:1.5-2 (Chaitow &

Bradley, 2002). Ideally, a person naturally adopts an optimal breathing pattern that requires the

least amount of mechanical stress and work from the respiratory musculature (Jones, Dean, &

Chow, 2003). However in some instances this optimal pattern can be disrupted such that

abnormal, and potentially inefficient, respiratory mechanics become the default pattern.

2.2.2. Abnormal Breathing at Rest

According to Chaitow et al. (2002) abnormal respiratory mechanics in a healthy human at rest

have been defined as:

i) breathing to the upper chest (apical breathing);

ii) a breathing frequency (fb) of greater than 16 breaths—min-1;

iii) an inspiratory to expiratory ratio of greater than 1.5:1;

iv) chronic mouth breathing, and;

v) absence of the end-expiratory pause.

Abnormal respiratory mechanics may occur as the result of, but not limited to, an acute or chronic

illness, prolonged exposure to stressful situations, occupational factors (for example occupations

10
that require large amounts of talking), or poor posture (Blager, 2000; Chaitow, Bradley, & Gilbert,

2002).

One consequence of abnormal respiratory mechanics, especially an apical pattern with an

increased fb, is an acute decrease in the level of CO2 in the bloodstream (Garssen, de Ruiter, &

van Dyke, 1992). The physiology underpinning how breathing pattern may be altered chronically

is not known. One theory that has been presented however, is related to the body’s sensitivity to

CO2 (Garssen, de Ruiter, & van Dyke, 1992; Gilbert, 2005). Whilst the body has compensatory

strategies to buffer the decreased CO2 in the short term, it is thought that if abnormal respiratory

mechanics persist chronically, the brainstem respiratory centres are ‘reset’ to respond to the new,

decreased level (Garssen, de Ruiter, & van Dyke, 1992). Accordingly, the centrally mediated

automatic breathing pattern is ‘reprogrammed’ such that an apical breathing pattern with an

elevated fb is accepted as ‘normal’ and a decreased level of circulating blood CO2 is maintained.

Consequently, a chronic BPD is established. Symptoms and signs of a BPD can include

shortness of breath inappropriate to the level of activity, difficulty inhaling during exercise,

inspiratory wheeze or stridor, air hunger, chest wall pain and heart palpitations (Chaitow & Bradley,

2002; Fallon, 2004). It is not known why some people develop BPDs whilst others exposed to the

same situation or environment do not.

Diagnosis of a BPD is usually made by clinical observation comparing a patient’s breathing

pattern with ‘normal’ respiratory mechanics (Chaitow, Bradley, & Gilbert, 2002). Attention is paid

to a patient’s breathing frequency, whether they are breathing apically or diaphragmatically, via

their mouth or nose, the degree to which their breathing pattern is rhythmical and the presence of

frequent sighs, yawns and/or abdominal splinting (Chaitow & Bradley, 2002). There is currently

11
no formal, objective, reliable and valid single test to clinically diagnose abnormal respiratory

mechanics and a BPD. Spirometry can give an indication of compromised respiratory mechanics,

for example airflow limitations, as can a bronchoscopy, however both of these tests require a

patient to be symptomatic at the time of testing (Blager, 2000). The Nijmegen questionnaire

(Appendix 1) has been used as a supportive diagnostic tool for identifying breathing pattern

disorders in research studies (Han, Stegen, De Valck, Clement, & Van de Woestijne, 1996;

Thomas et al., 2003), and is used by some clinicians to confirm diagnosis of hyperventilation

syndrome (HVS) (Chaitow, Bradley, & Gilbert, 2002). This noninvasive questionnaire identifies

anxiety related breathing pattern disorders. It has a sensitivity of 91% and a specificity of 95% in

patients with a clinical diagnosis of HVS, and enables patients to see the widespread nature of

symptoms and progress with treatment (Van Dixhoorn & Duivenvoorden, 1984).

The prevalence of abnormal respiratory mechanics and BPDs in athletes, in the absence of

respiratory pathology, has not been reported in the literature. It would be reasonable to speculate

that if abnormal respiratory mechanics and/or BPDs are present in athletes at rest, the effect and

consequences of such might be exaggerated during exercise due to the increased physiological

demands upon the respiratory, cardiovascular and musculoskeletal systems during exercise.

Before addressing these factors in more detail, a brief review of the normal respiratory response

to exercise is provided.

2.2.3. Respiratory Response to Exercise

The increased metabolic demands experienced by the body during exercise must be met by

significantly increasing VE (Gallagher, Brown, & Younes, 1987). Compared to untrained

individuals who elicit a VE of 112 ± 19.9 L—min-1 (Sheel et al., 2006) during maximal intensity

12
exercise (VEmax), studies of top level athletes have demonstrated that VE can increase to 190

L—min-1 in cycling (Harms, Wetter, St Croix, Pegelow, & Dempsey, 2000), and often above 200

L—min-1 in rowers (Sheel, Lama, Potvin, Coutts, & McKenzie, 1996), sometimes peaking as high as

270 L—min-1 (Shephard, 1998). To achieve a breathing pattern that will enable the ventilatory

demand of high intensity exercise to be met usually causes an athlete’s respiratory pattern to shift

from diaphragmatic via the nose, to apical via the mouth in an attempt to shift larger volumes of air

faster (Bradley, 2002). The workload at which this occurs has not been documented in the

literature. It is possible that this shift in pattern corresponds with the respiratory compensation

point – the onset of hyperventilation during incremental exercise that is thought to represent the

body’s attempt to compensate for metabolic acidosis (Meyer, Faude, Urhausen, & Kindermann,

2004). During exercise VE can be met by increasing fb and tidal volume (VT) and it is not known

whether the shift from diaphragmatic to apical breathing is a physiological necessity, or a ‘learned

habit’. Further, an optimal breathing pattern during exercise (as opposed to the ‘normal

physiological response’) has not been identified.

To achieve such large increases in VE, the normal respiratory response to exercise is to increase

the depth of breathing (increased VT) and as the workload further increases, by increasing fb

(Gallagher, Brown, & Younes, 1987). The initial increase in VT may cause the end-expiratory lung

volume (EELV) to decrease below tidal EELV (Johnson, Saupe, & Dempsey, 1992). This

lengthens the diaphragm and other inspiratory muscles at end-expiration, allowing them to

operate near their optimal length for force generation as exercise intensity increases (Ward,

Eidelman, & Stubbing, 1988). In addition, a decreased EELV (below functional residual volume)

may increase the proportion of “passive” inspiration due to the elastic recoil of the chest wall,

13
further enhancing the attainment of a high VE whilst preserving respiratory muscle length (Verges,

Notter, & Spengler, 2006).

As ventilatory demands increase further, fb continues to rise, and VT will either plateau at

approximately 50-60% of vital capacity (Clark, Hagerman, & Gelfand, 1983; Hey, Lloyd,

Cunningham, Jukes, & Bolton, 1966) or start to decrease (Spengler, Knopfli-Lenzin, Birchler,

Trapletti & Boutellier, 2000; Gallagher, Brown & Younes, 1987). Potentially EELV may increase

above resting levels resulting in dynamic hyperinflation of the lungs (Johnson, Saupe, & Dempsey,

1992). Although dynamic hyperinflation permits increases in expiratory flow rates (Miller et al.,

2005), which may assist the attainment of the high VE necessary during maximal exercise, it

comes at the expense of an increased work of breathing (the effect of which will be discussed

later in this review) (Witt, Guenette, Rupert, McKenzie, & Sheel, 2007).

It has been observed that as exercise intensity rises, fb increases at the expense of maintaining VT

(this is commonly referred to as a tachypnoeic shift or a tachypnoeic pattern) and an irregular,

inefficient pattern develops (Gallagher, Brown, & Younes, 1987; Scheuermann & Kowalchuk,

1999). A tachypnoeic shift is considered to be highly inefficient, both mechanically and

metabolically (Lucia, Carvajal, Calderon, Alfonso, & Chicharro, 1999). The likely consequence of

a tachypnoeic shift is dynamic hyperinflation due to the decreased expiratory time and subsequent

breath stacking which ultimately increases the work of breathing caused by breathing at increased

lung volumes. Further, VE may remain unchanged despite further changes to fb and VT, whilst the

amount of O2 actually reaching the alveoli for gas exchange decreases due to increased dead

space volume (Hanson, Claremont, Dempsey, & Reddan, 1982) (Table 2).

14
Table 2. Effect of Breathing Rate and Depth on Alveolar Ventilation in Three Scenarios
(from Marieb, 1992)
Breathing Dead Tidal Respiratory Minute Alveolar % of TV =
pattern space volume rate respiratory ventilation dead space
volume (TV) volume rate volume
I - Normal rate 150ml 500ml 20/min 10,000 ml/min 7000 ml/min 30%
and depth

II - Slow, deep 150ml 1000ml 10/min 10,000 ml/min 8500 ml/min 15%
breathing

III - Rapid, 150ml 250ml 40/min 10,000 ml/min 4000 ml/min 60%
shallow breathing

It could be suggested that the ability to maintain an even, controlled breathing pattern, rather than

allowing a tachypnoeic shift to develop, would be advantageous to athletes during high exercise

intensities. Given the significant increase in VE (often 10-15 times resting; Johnson, Aaron,

Babcock, & Dempsey, 1996) that must be achieved during maximal exercise, it would seem

advantageous to achieve this in the most efficient way. Certainly, it appears that some athletes

can maintain an effective respiratory pattern (as defined by lack of tachypnoeic shift; Lucia,

Carvajal, Calderon, Alfonso, & Chicharro, 1999) whilst others do not. In the few studies that have

compared the respiratory pattern of different levels of athletes, it appears that only the very highly

trained or experienced athletes demonstrate this desired characteristic (Lucia, Carvajal, Calderon,

Alfonso, & Chicharro, 1999; Mahler, Shuhart, Brew, & Stukel, 1991; Vaughan, 1984). More

specifically, a clear difference in respiratory pattern has been observed between professional and

non professional cyclists (Lucia, Carvajal, Calderon, Alfonso, & Chicharro, 1999) at high exercise

intensities. Lucia et al. (1999) compared fb, VT, VE, inspiratory and expiratory time in 14

professional Tour de France level male road cyclists ( V&O 2 peak: 73.2 ± 6.6 ml—kg—min-1, Pmax: 471

± 30.9W) and 11 amateur, elite male road cyclists ( V&O 2 peak: 73.8 ± 5 ml—kg—min-1, Pmax: 437 ±

25.2W) during incremental exercise from sub-maximal to maximal intensities (exhaustion). The

15
VEmax did not differ between the two groups (170 vs. 175 L—min-1, Figure 1), although at intensities

between 300 and 400W professional cyclists exhibited a 15% (P < 0.01) lower VE than elites. At

power outputs less than 200W fb tended to be higher in professionals than elite cyclists (for

example 25 versus 20 breaths—min-1; P < 0.05 at 100W, Figure 1) but from 200W to maximum

intensities the elite group had a significantly higher fb than professionals (58 vs. 51 breaths—min-1

at maximal; P < 0.05, Figure 1). Inspiratory time reached a plateau in professionals at low

intensities (300W; 63.7% Pmax) and then decreased with increasing exercise intensity, whereas it

consistently decreased in elite from the lowest to the highest intensities (50 to 437.5W; Figure 2).

Expiratory time was significantly higher in professionals than elites at most moderate-to-high

power outputs (P < 0.05 for 250 and 300W, and P < 0.01 for 350 and 400W, Figure 2). These

findings indicate that although the maximum volume of air delivered to the lungs was not

statistically different between the two groups, professional cyclists did not adopt a tachypnoeic

shift to maintain their high VE.

16
 Figure 2. Inspiratory (tI) and expiratory (tE) times.
Definitions as for Figure .1. * P < 0.05, ** P < 0.01, §
significant (P < 0.05) interactive effect (group x
exercise intensity. (From Lucia et al., 1999).

Figure 1. Ventilatory response of the elite (E) and

professional (P) cyclists during the tests:
pulmonary ventilation (VE), breathing frequency
(fb) and) tidal volume (VT). Max Maximal power
output. * P < 0.05 ** P < 0.01, § significant (P <
0.05) interactive effect (group x exercise intensity)
(From Lucia et al., 1999).
It should be acknowledged that the results of Lucia et al. (1999) were obtained in a laboratory

environment. The subjects breathing was measured using a mouth piece and nose clip which

have been shown to influence the breathing pattern in normal subjects by causing an increase in

VT while fb is unchanged or decreased (Askanazi et al., 1980). Lucia et al. (1999) believed their

17
subjects, having participated in previous studies using the same respiratory apparatus, were

familiar with the equipment and it would have been unlikely to have affected their breathing.

Additionally, both groups of cyclists were tested using the same equipment meaning both groups

would have been affected in the same way. Lucia et al. (1999) set the cadence for their subjects,

meaning exercise intensity and power could only be generated by increasing the force applied to

each pedal stroke, whereas in field conditions cyclists increase power by increasing the cadence

in addition to the force applied to each stroke (Conconi et al., 1996). Lucia et al. (1999)

acknowledge that in real cycling conditions at high intensities, increasing cadences might alter

breathing patterns through the sensory input from muscle mechanoreceptors to the respiratory

centre, which could result in a higher fb than they observed.

In sports with a rhythmic component such as rowing, respiration is coordinated with the action of

the locomotor system (entrainment) to maximise respiratory efficiency, however entrainment has

only been observed in high level rowers (Mahler, Shuhart, Brew, & Stukel, 1991; Vaughan, 1984).

Mahler et al. (1991) compared the respiratory pattern of 16 novice and 18 experienced rowers and

found entrainment at maximal effort in 78% of elite female rowers, but only in 30% of novices.

Similar to the professional cyclists in Lucia et al.’s (1999) study, the elite rowers maintained a

respiratory pattern that was void of a tachypnoeic shift, whilst the novice rowers increased

ventilation by continuing to increase fb and decrease VT until exhaustion.

It is possible that maintaining optimal levels of effective ventilation is one factor that enables

athletes to achieve levels of performance necessary to excel in their own sport. Lucia et al.

(1999) speculated that a breathing pattern devoid of a tachypnoeic shift, and with a prolonged tE,

is required to perform at the highest level of cycling. Currently, there is insufficient research to

18
identify which physiological characteristics enable professional cyclists to breathe more efficiently,

whether it is a genetic predisposition, or whether their ability to breathe efficiently is an

independent factor that could be likened to a learned technique or skill that optimises performance.

In an attempt to explain the observed differences in breathing patterns between professional and

elite athletes, Lucia (1999) initially suggested that physiological factors related to the more

demanding training carried out by the professional, as compared with the elite, cyclists was

responsible. However, in a subsequent longitudinal study, Lucia et al. (2001) later demonstrated

that a professional cyclist’s breathing pattern did not vary during the three periods (rest, pre-

competition and competition) of a competitive cycling season, despite the fact that a significant

overall training effect on power output and V&O 2 was shown throughout the season. These

findings suggest that training volume was not the causative factor, or if it was, it possibly occurred

over years of training, and was not influenced by the differences occurring throughout a

professional cyclist’s season. Another suggestion is that the high fb and decreased VT often

observed at high intensity exercise (Clark, Hagerman, & Gelfand, 1983; Folinsbee, Wallace, Bedi,

& Horvath, 1983; Gallagher, Brown, & Younes, 1987) could simply be attributed to respiratory

muscle fatigue, and that high level athletes develop a superior level of resistance to such fatigue

due to their greater years of experience and volume of training (Syabbalo, Krishnan, Zintel, &

Gallagher, 1994). In a group of competitive, but not professional, cyclists, respiratory muscle

fatigue (measured by inspiratory pressure at zero flow) has been recorded at 17 ± 4% and13 ±

3% following a 20 and 40km-TT respectively (Romer et al., 2002). The consequences of

respiratory muscle fatigue will be discussed further in section 2.3.5.

19
In contrast, one theory that has been supported across decades (Hey, Lloyd, Cunningham, Jukes,

& Bolton, 1966; Scheuermann & Kowalchuk, 1999) is that the integration of the depth and

frequency of breathing required to achieve a given ventilatory demand appears to be independent

of known respiratory stimuli, and may simply reflect a pattern of breathing that results in a minimal

amount of work by the respiratory muscles. This theory may support the suggestion that

breathing pattern may be a learned skill and could therefore be retrained. Clearly, experimental

work is needed to determine whether an efficient breathing pattern can be taught, or whether it is

an adaptive response secondary to changes in other physiological variables after training.

2.2.4. Abnormal Breathing Patterns in Athletes

Ventilation during exercise may be compromised in athletes for a number of reasons including,

but not exclusively, mechanistic (ventilation-perfusion mismatching, diffusion limitations and

increased intra- and post-pulmonary shunt (Dempsey & Wagner, 1999; Stickland & Lovering,

2006)), environmental and pathological (exercise-induced asthma, bronchospasm and vocal cord

dysfunction), psychological (anxiety-induced hyperventilation) and biomechanical (cycling

postures). Whilst further details of the mechanistic factors are beyond the scope of this review,

the latter three will be explored in more detail below.

2.2.4.1 Environmental and Pathological Effects on Breathing

The prevalence of disordered breathing in athletes is unknown; however the literature suggests

that athletes are just as much at risk as the general population, if not more, of developing

breathing related problems, which will subsequently affect their breathing pattern during exercise

(Rundell & Spiering, 2003). The incidence of asthma in athletes (11-50%; (Larsson,

Hemmingsson, & Boethius, 1994; Rundell & Spiering, 2003)), is known to be greater than in the

20
general population (10-12%; (AIHW, 2005)) which could be due to increased exposure to cold, dry

environments as well as pollen and pollution whilst training (Mendelli, Lounana, Messan, Menuet,

& Petitjean, 2006). For example, Rundell and Spiering (2003) reported a 30% incidence of EIB in

370 developmental and elite athletes. In Olympic athletes, Fallon (2003) suggested that the

incidence of EIB may be as high as 10-15%. Breathlessness (36.8%), exercise–induced phlegm

production (14.9%), cough (7.3%), wheezing (6.2%) and chest tightness (4.9%) were amongst

respiratory symptoms reported in a sample of 698 athletes despite only 15.3% having diagnosed

asthma (Turcotte, Langdeau, Thibault & Boulet, 2003) suggesting that EIA is not the only cause of

respiratory symptoms in athletes. Inspiratory stridor is another respiratory sign seen in athletes

and is often associated with Vocal Cord Dysfunction (VCD) - the paradoxical closure of the vocal

cords that occurs during inspiration (Rundell & Spiering, 2003). Mouth breathing, increased

activation of the scalene muscles (accessory respiratory muscles), anxiety and air pollutants are

all triggers of VCD. Symptoms usually present at a high VE during both intense training and/or

competition (Fallon, 2003), causing stridor, air hunger and chest tightness – all symptoms that

closely mimic EIA (Brugman & Simons, 1998). Not surprisingly, VCD is often misdiagnosed in

athletes as EIA and treated as such (Brugman & Simons, 1998; Fallon, 2004; Rundell & Spiering,

2003). VCD is unresponsive to traditional asthma medications, potentially meaning that athletes

never receive the appropriate treatment and subsequent resolution of their symptoms (Fallon,

2004).

2.2.4.2 Psychological Effects on Breathing

In addition to the environmental and pathological factors presented above, psychological factors

may also affect an athlete’s breathing pattern (Taylor, Rehder, Hyatt, & Parker, 1989). Anxiety is

known to increase respiratory rate through stimulation of the sympathetic nervous system (Ley &

21
Yelich, 1998; Bradley, 2002). Athletes competing in a sporting environment are subjected to

stressful, anxiety-inducing environments on a regular basis (Baker, Cote, & Hawes, 2000), which

may predispose them to BPDs. It is possible that the incidence of BPDs is higher in athletes than

the general population due to the high levels of performance anxiety to which they are exposed.

2.2.4.3 Postural effects on breathing

Breathing may also be compromised in cyclists through biomechanical factors. More specifically,

it is possible that the posture adopted by a cyclist when riding may compromise breathing

efficiency. For example, time trial riders adopt an aerodynamic position (Figure 3) to minimize

frontal area in order to maximize speed (Capelli, Rosa, & Butti, 1993; Faria, Parker, & Faria,

2005b).

Figure 3. A typical athlete’s cycling position during a road time-trial (reprinted with the permission of Stephen
Sheldrake, NZ Triathlete)

22
When cyclists adopt this position, the elbows are brought in close to the body and the trunk close

to the thighs – a position that compromises the ability of the thoracic cavity to expand fully to

enable breathing to maximal capacity (Hogg, 2005). This aerodynamic position may induce an

increase in abdominal impedance, and therefore in diaphragmatic work to overcome it, leading to

premature recruitment of the accessory respiratory muscles accelerating the onset of respiratory

muscle fatigue (Boussana et al., 2007; Johnson, Babcock, Suman, & Dempsey, 1993).

While gains are to be made by adopting an aerodynamic position, some evidence exists to

suggest that the aerodynamic position may not be metabolically efficient. Gnehn, Reichenbach,

Altpeter, Widmer & Hoppeler (1997) observed a significant increase in V&O 2 (3%; P = 0.02), heart

rate (3.3%; P = 0.02), and respiratory exchange ratio (1.3%; P = 0.02) when a cyclist’s upper body

configuration was changed from the upright to the time trial position at 70% V&O 2 max. Although

these are small percentage changes, they may be important if they carry over to the cyclist’s

performance. Conversely, in two sub-maximal conditions (70 and 80% V&O 2 max), Johnson and

Shultz (1990) found no evidence to suggest that riding in an aerodynamic position interfered with

breathing mechanics in any way (70% VO2max), nor did they show a significant difference in V&O 2

when cyclists rode at 80% of V&O 2 max during a 10 minute steady state ride.

Clearly, the apparent increased metabolic cost that may be associated with the time trial position,

is offset by the gains made by decreasing the aerodynamic resistance (Gnehm, Reichenbach,

Altpeter, Widmer, & Hoppeler, 1997). At speeds greater than 50kph (often achieved during time

trials in elite competition) some authors argue that aerodynamic resistance is the most

23
performance-determining variable (Faria, Parker, & Faria, 2005). The effect of the time trial

position on breathing has primarily been investigated at sub-maximal intensities (Gnehm et al.,

1990; Johnson & Shultz, 1997). However, time trial competitions are raced at high intensities and

it is unclear from the literature what effect the aerodynamic time trial position would have on

breathing pattern at higher intensity domains (i.e. closer to V&O 2 max). It is possible that adopting

an efficient breathing pattern whilst riding in the time trial position may reduce the elevated

metabolic cost previously reported in this position and, added to the gains of optimising

aerodynamics, may further enhance performance.

A combination of environmental, pathological, physiological and psychological factors, in addition

to the potential contribution that posture makes to altered respiratory mechanics, may affect an

athlete’s breathing. This may occur acutely, for example increased fb immediately prior to a race

or during training, but more detrimentally may affect it chronically through alteration of the

subconscious breathing pattern and altered respiratory mechanics. The consequences of altered

respiratory mechanics are multifactorial and, in addition to compromising breathing efficiency, may

ultimately have a negative effect on performance. These consequences will be further discussed

in the section below.

2.3. Consequences of Altered Respiratory Mechanics

Respiratory mechanics during exercise may be altered by the presence of BPDs at rest, the

normal respiratory response to exercise, development of a tachypnoeic shift and postures

adopted when riding. The consequences of altered respiratory mechanics (changes to respiratory

timing and muscle recruitment patterns) may be detrimental to breathing efficiency and ultimately

24
exercise performance by initiating dynamic hyperinflation of the lungs and increasing the work of

breathing. Dynamic hyperinflation and increased work of breathing may contribute to respiratory

muscle fatigue, blood stealing and dyspnoea – all factors that have been associated with

decreased exercise performance. This section will explore how altered respiratory mechanics

increase the work of breathing and lead to such consequences, and in turn how exercise

performance may be negatively affected.

2.3.1. Respiratory Timing

The “normal” timing of the inspiratory to expiratory ratio at rest is 1:1.5 - 2 (West, 1995). To

achieve the increased fb and decreased VT associated with exercise, the ratio and length of

inspiratory (tI) and expiratory (tE) time has to change. In normal breathing, the lengthened

expiratory phase ensures the inspiratory muscles are returned to their optimal length prior to the

next inspiration. Previous research with sedentary individuals and elite cyclists (Folinsbee,

Wallace, Bedi, & Horvath, 1983) indicates that at high respiratory rates the reduction of expiratory

time makes a significantly greater contribution to the rise in fb, than the decrease in tI, effectively

resulting in a reverse inspiratory to expiratory ratio of 2:1. This breathing ratio leads to “breath

stacking” – a phenomenon where a subsequent inspiration is taken prior to the full exhalation of

the previous breath, resulting in tidal breathing occurring closer to total lung capacity (the volume

of air in the lungs at the height of a maximal inspiration) (Chaitow & Bradley, 2002; McConnell,

2005). Increased end-inspiratory (Grazzini, Stendardi, Gigliotti, & Scano, 2005; Johnson, Aaron,

Babcock, & Dempsey, 1996) and increased end-expiratory lung volumes have been observed in

athletes (Sutton, 1992), suggesting that breathing at higher lung volumes or “dynamic

hyperinflation” may be a consequence of the respiratory response to exercise (McConnell, 2005).

25
 2.3.2. Dynamic Hyperinflation

Whilst the initial increase in VT at the onset of exercise may cause end-expiratory lung volume

(EELV) to decrease, increases to fb and the subsequent changes to respiratory timing discussed

above may increase EELV, resulting in dynamic hyperinflation (Johnson, Saupe, & Dempsey,

1992). Dynamic hyperinflation affects the mechanics of the respiratory muscles in three ways - by

altering the length-tension, the pressure-volume and the force-velocity relationships. Muscles

have an optimal length at which they can generate their maximal force. This is referred to as the

length-tension relationship of muscles (Rack & Westbury, 1969). When a muscle is required to

contract in a position that is shorter or longer than its optimal length, its force generating potential

is compromised. Hyperinflation induces functional weakening of the inspiratory muscles by

shortening the operating lengths of the diaphragm and accessory muscles. The inspiratory

muscles therefore move to a weaker portion of the length tension relationship and their ability to

generate a forceful contraction is compromised (Grazzini, Stendardi, Gigliotti, & Scano, 2005).

Additionally, because the operating length of the muscles is shortened, the time available for

contraction through range is less. The shorter time through which a muscle must contract, the

lower its force generating potential (Grazzini, Stendardi, Gigliotti, & Scano, 2005) therefore the

inspiratory muscles are also in a weaker portion of their force-velocity relationship.

Lung compliance refers to the capability of the lungs to distend under pressure as measured by

pulmonary volume change per unit pressure change (West, 1995). As inspiration approaches

total lung capacity the lungs are almost at maximal distension and the lungs become ‘stiffer’.

Therefore, for a given inspiratory pressure (effort), there is a smaller increase in lung volume

during dynamic hyperinflation, than at normal lung volumes, resulting in decreased pulmonary

ventilation. Additionally, the inward elastic recoil of the respiratory system is greater at higher

26
volumes, imposing an added load on the inspiratory muscles, which are already contracting

inefficiently (ATS, 1998).

The alteration in the length-tension, force-velocity and pressure-volume relationships means that if

dynamic hyperinflation occurs (through an inefficient breathing pattern) at a time when the

demands on the inspiratory muscles are greatest (i.e. during intense exercise), then the muscle’s

force generation capacity will be significantly compromised (Johnson, Aaron, Babcock, &

Dempsey, 1996). The capacity of the inspiratory muscles to generate dynamic pressure has been

predicted to decrease by 17% for each 10% of the total lung capacity that is accounted for by an

increased tidal volume above functional residual capacity (El-Manshawi, Killian, & Summers,

1986). Thus, even a small increase in end-expiratory lung volume will reduce the contractile

properties of the inspiratory muscles, and add to the load presented to them such that the work

required to sustain ventilation (work of breathing), especially during intense exercise, is

significantly increased.

2.3.3. Expiratory Flow Limitation

Although traditionally the mechanics of inspiration have been the main subject of investigation in

ventilation during exercise (Grazzini, Stendardi, Gigliotti, & Scano, 2005; Johnson, Saupe, &

Dempsey, 1992), the mechanics of expiration may also be negatively affected in some athletes.

At high levels of ventilation, an expiratory flow limitation has been identified in some athletes,

specifically elite, endurance athletes (Johnson, Saupe, & Dempsey, 1992). As expiration

becomes active to meet the increasing ventilatory demands experienced with high intensity

exercise, the abdominal muscles contract to force the diaphragm up and the resulting pleural

pressure can become positive. Positive pleural pressure may temporarily collapse the bronchi

27
and cause limitation of expiratory airflow. Thus, further increases in expiratory pressure limits

expiratory flow, and requires additional metabolic work in an attempt to overcome the flow

limitation (Kayser, Sliwinski, Yan, Tobiasz, & Macklem, 1997). It is suggested that the presence

of expiratory flow limitation promotes dynamic pulmonary hyperinflation with the concomitant

increase of work of breathing and impairment of inspiratory muscle function discussed above,

which may contribute to dyspnoea and limitation to exercise (Koulouris & Kosmas, 2002).

2.3.4. Work of breathing

Breathing at rest in healthy individuals requires less than 5% of the body’s total V&O 2 (Jones,

Dean, & Chow, 2003). However, in highly trained individuals, the work of breathing during intense

exercise (92-100% V&O 2 max) may require up to 15% of the body’s total V&O 2 (Aaron, Seow,

Johnson, & Dempsey, 1992B), to enable the respiratory muscles to meet the metabolic demands

of this level of exercise. It would be plausible to expect that to meet such an increase in metabolic

demand, the respiratory muscles would also require a significant increase in the percentage of

cardiac output (Q). In an attempt to demonstrate this, Harms et al (1998) manipulated the work of

breathing using a proportional assist ventilator (PAV) to either decrease or increase the work of

breathing. Using this technique during maximal intensity exercise, Harms et al. (1998) showed

that a substantial portion of Q (14-16%) was indeed directed to the respiratory muscles to meet

their metabolic requirements. Given the importance of a high VE during maximal intensity

exercise, and the consequent increase in the work of breathing, changes to breathing pattern,

respiratory mechanics and work of breathing at maximal exercise could potentially play a part in

limiting exercise performance.

28
 2.3.5. Respiratory Muscle Fatigue

The diaphragm is the most fatigue resistant of all the skeletal muscles (Dempsey, 2006b). A

healthy diaphragm is composed of approximately 45% type I and 55% type II muscle fibres

(McConnell, 2005) and unsurprisingly, has a very high aerobic enzymatic capacity and blood

supply (Dempsey, Romer, Rodman, Miller, & Smith, 2006). Despite these fatigue-resistant

qualities, the diaphragm has been shown to fatigue during exercise to exhaustion at intensities

greater than 85% of V&O 2 max. Using bilateral phrenic nerve stimulation (BPNS) (1-20Hz) to

measure transdiaphragmatic pressure before and after exercise to exhaustion at 80% V&O 2 max in

healthy sedentary subjects, Mador et al. (1993) demonstrated a 17.2% decrease in

transdiaphragmatic pressure 10 minutes after cycling exercise to exhaustion. After 60 minutes of

rest, transdiaphragmatic pressure had recovered to 93 ± 7% of pre-exercise levels. Using similar

methodology in active individuals exercising at 85 and 95% of V&O 2 max, Johnson et al. (1996)

demonstrated that transdiaphragmatic pressure immediately after exercise was reduced 25-50%

below pre-exercise measures and took two hours to recover. Johnson et al. (1996) concluded

that heavy, whole-body exercise to exhaustion in relatively fit, normal subjects resulted in

significant diaphragmatic fatigue. Comparing these two studies, it appears that the degree to

which the diaphragm fatigues, and the speed at which it recovers, is significantly affected by the

intensity of exercise. Conversely, diaphragm fatigue has not been shown to occur below 80%

V&O 2 max (Gallagher & Younes, 1989) or when the diaphragm is fatiguable with short term (less

than eight minutes) progressive exercise to exhaustion unless a resistive load is added (Levine &

Henson, 1988). This would suggest that diaphragm fatigue is determined by a combination of the

work demanded of it, and the time that it is under load.

29
The actual work performed by the inspiratory muscles, specifically the diaphragm, per se is,

however, not enough to explain the diaphragm fatigue observed by Mador et al. (1993) or

Johnson et al. (1996). When resting subjects were asked to mimic (using auditory and visual

feedback) the duration and magnitude of diaphragm work achieved during maximal intensity,

endurance exercise, fatigue did not occur. Only by increasing diaphragm work voluntarily to twice

that required during maximal exercise did diaphragm fatigue occur (Babcock, Pegelow, McClaran,

Suman, & Dempsey, 1995). This would suggest that other changes that occur, perhaps

elsewhere in the body, during intense exercise contribute to respiratory muscle fatigue, rather

than just an inability of the respiratory muscles to meet the work demand.

Global inspiratory muscle fatigue has also been demonstrated to occur immediately following

exercise both in untrained (Boutellier & Piwko, 1992) and trained individuals (Romer, McConnell,

& Jones, 2002c; Volianitis et al., 2001). For example, Romer et al. (2002) demonstrated

substantial decreases (compared to baseline measures) in Po (pressure generated by the

respiratory muscles at zero flow) of -17 ± 4% and -13 ± 3% following a simulated 20 and 40km-TT

respectively in competitive cyclists. Furthermore, Po had not returned to baseline levels at 30

minutes post-exercise. Similarly, but in competitive swimmers, Lomax and McConnell (2003)

reported a 29% deficit from baseline measures in maximal inspiratory pressure (MIP) following a

200m freestyle swim at 90-95% race pace. In competitive rowers, Volianitis et al. (2001)

demonstrated a 10% decline in MIP after six minutes of rowing ergometry at 100%. One

explanation for the large differences in deficits observed in these three studies (apart from the

obviously important postural differences) could be due to the nature of the sports and the

ventilatory requirements of each. For example, entrainment has been observed in high level

rowers and it is possible that the rowers used in the study subconsciously used entrainment to

30
optimise breathing and subsequently attenuate respiratory muscle fatigue. Conversely, in

freestyle swimming the mouth is out of water for a very short time, during which an inspiratory

effort must be taken. This short, fast inspiratory effort may predispose to dynamic hyperinflation

and exaggerate the pressure generating deficit from respiratory muscle fatigue, due to alterations

of the length – tension relationship (discussed earlier). Another factor that may exacerbate

respiratory muscle fatigue in swimming, and contribute to the different degrees of inspiratory

muscle fatigue observed by Romer et al. (2002) and Volianitis et al. (2001), is the hydrostatic

compression around the chest wall when the inspiratory muscles are relaxed (Withers & Hamdorf,

1989). Hydrostatic pressure counteracts inspiratory muscle force (Frangolias & Rhodes, 1996)

which may increase inspiratory muscle work in an attempt to overcome this.

Exercise duration, exercise intensity and inefficient respiratory mechanics can increase the work

of breathing to the degree that respiratory muscle fatigue may occur, with negative consequences

on exercise performance (Harms, Wetter, St Croix, Pegelow, & Dempsey, 2000). Harms et al.

(2000) demonstrated this concept by unloading the respiratory muscles during high intensity

exercise (90% V&O 2 max) using a PAV. Subjects were able to cycle longer (14.4 ± 4.9%; p<0.05)

than the control duration, whereas increasing respiratory muscle work significantly reduced (-15.1

± 3.3%; p<0.05) exercise time from control values. It is possible that there were other factors

contributing to the effect on exercise time in Harms et al.’s (2000) study in addition to respiratory

muscle fatigue (for example blood stealing, which will be discussed later) however certainly when

load was removed from the respiratory muscles their capability for sustained power generation

was increased.

31
Respiratory muscle fatigue may negatively affect exercise performance by altering the way the

respiratory muscles are recruited (Johnson, Aaron, Babcock, & Dempsey, 1996). Indirect

evidence from pressure recordings suggest that the fatiguing diaphragm actually reduces its

output as a force generator during the later stages of sustained endurance exercise and

accessory inspiratory and expiratory muscles become the dominant effecter of time dependant

hyperventilation (Dempsey, 2006b). Johnson et al. (1996) suggested that although total VE and

respiratory muscle pressure may continue to rise, the diaphragm is deactivated in an attempt to

reserve function. Diaphragm fatigue may indirectly influence performance by a reduction in the

relative contribution of the diaphragm; and a subsequent increase of accessory muscle use may

negatively affect performance. This change in muscle recruitment pattern may lead to dynamic

hyperinflation, inefficient respiratory mechanics, subsequent increased work of breathing and

ultimately further respiratory muscle fatigue (Harms, Wetter, St Croix, Pegelow, & Dempsey,

2000) and blood stealing (Harms et al., 1997) – the latter being detrimental to limb function during

strenuous exercise.

The ability of professional cyclists, unlike elite amateur and lower level cyclists, to maintain the

diaphragmatic breathing pattern from low to high intensity (exhaustive) exercise (Lucia, Carvajal,

Calderon, Alfonso, & Chicharro, 1999) demonstrates that a mechanism may exist that enables

high intensity exercise to be maintained, without an increase in the work of breathing to the point

that respiratory muscle fatigue, and the subsequent negative effect on performance, occur. It has

been suggested that this controlled breathing pattern is one of the major factors that enables

professional cyclists to perform better than the tier of cyclists below them (Faria, Parker, & Faria,

2005b). Lucia et al.’s (1999) findings suggest that either respiratory muscle fatigue did not occur

in professional cyclists, or, if it did, it did not result in deactivation of the diaphragm with a

32
subsequent shift to an apical breathing pattern. This observation would suggest that the

decreased diaphragm recruitment speculated by Johnson et al. (1996) is more an inefficient but

common response during exercise, rather than an efficient physiological necessity. Johnson et al.

(1996) commented that although the diaphragm appeared to be deactivated, total VE and

respiratory muscle output continued to rise. However, it is not known what, if any, greater

increases in these variables would have occurred had the diaphragm continued to be the primary

pressure generator. It is therefore possible that the presence of an altered breathing pattern is the

instigator that then leads to respiratory muscle fatigue through the altered respiratory muscle

mechanics discussed above, rather than the result of respiratory muscle fatigue itself.

Regardless of whether altered respiratory recruitment pattern is the cause or effect of respiratory

muscle fatigue, the implications of it may not just be limited to the mechanical effects. Increased

work of breathing and respiratory muscle fatigue may have a further effect of causing reflex

vasoconstriction to exercising muscles, reducing blood flow to the muscles responsible for power

generation during exercise, and subsequently decrease exercise performance.

2.3.6. Blood Stealing

A further consequence of altered respiratory mechanics, and the subsequent increased work of

breathing, is the potential competition for blood supply between the peripheral locomotor and

respiratory muscles. During exercise, an adequate blood supply is vital to ensure required

nutrients are delivered to, and exercise metabolites are removed from muscle (Barclay, 1986;

Frisbee, Murrant, Wilson, & Barclay, 1999). At maximal intensity exercise, competition exists

between the respiratory and locomotor muscles for a supply of energy that is insufficient to meet

the needs of both (Aliverti & Macklem, 2001). Subsequently blood flow may be diverted from the

33
exercising (locomotor) muscles to the respiratory muscles under certain respiratory conditions

(blood stealing) to maintain the vital supply of respiratory muscle V&O 2 (Harms et al., 1997).

The concept of blood stealing was first investigated by Secher et al. (1977) who demonstrated

that recruiting additional muscle mass during sub-maximal leg cycling ( V&O 2 legs 67% V&O 2 max)

by simultaneously arm cranking ( V&O 2 arms 44%; total body exercise 78% V&O 2 max) decreased

blood flow to the already exercising leg muscle by approximately 15% (12.4 to 10.5 L—min-1).

Since that pioneering work further studies have been undertaken that both refute (Kowalchuk,

Rossitter, Ward, & Whipp, 2002; Richter, Kiens, Hargreaves, & Kjaer, 1992) and support (Harms

et al., 1997; Harms et al., 1998; Sheel et al., 2001) the existence of the blood stealing

phenomenon. In contrast to Secher et al.’s (1977) findings, Richter et al. (1992) was unable to

demonstrate a decrease in Qlegs when arm cranking was added to cycling in untrained cyclists.

Further, Kowalchuk et al. (2002) demonstrated that increasing the work of breathing by adding a

load to both inspiratory and expiratory muscles during heavy intensity exercise (93 ± 3% V&O 2 max),

also in untrained cyclist, did not increase leg de-oxygenation (the variable these authors chose to

indicate decreased Qlegs), implying that Qlegs was not reduced consequent to the increased work of

breathing.

In contrast to the studies in untrained cyclists that refute blood stealing, two well designed studies

that support its existence were completed by Harms et al. (1997; 1998) involving competitive male

cyclists (N = 7; N = 8). These authors demonstrated that using a PAV to unload the respiratory

muscles and decrease the work of breathing (by 63%, compared to control) during maximal

incremental cycling to exhaustion, resulted in a decrease in Q (2.4 ± 0.7 l/min) and limb vascular

34
resistance (LVR) (0.5 ± 0.1mmHg—1-1.min), whilst blood flow to the leg muscles (Qlegs) increased

(0.8 ± 0.3L—min-1). Conversely, when the work of breathing was artificially increased (by 28%) by

adding inspiratory resistance (3-5cmH2O—L-1—s-1), both LVR (1.5 ± 0.3mmHg.1-1—min; p = 0.002)

and Qlegs (1.3 ± 0.2 l/min; p = 0.002) decreased (Harms et al., 1997). These results clearly show

that a relationship exists between increased work of breathing and Qlegs such that increasing the

work of breathing has a detrimental effect on locomotor muscle blood flow. This could suggest

that factors that increase the work of breathing in cyclists, for example posture and altered

respiratory mechanics, may compromise leg blood flow – the consequences of which are

discussed below. Conversely, factors that prevent the work of breathing from increasing may

preserve Qlegs.

Blood stealing is thought to be due to sympathetically mediated vasoconstriction in the locomotor

muscles induced in a reflex manner, possibly originating in fatigueing respiratory musculature

(Harms et al., 1998). It is postulated that during maximal exercise, the work and / or aerobic

status of the respiratory muscles may stimulate type III-IV afferents in the diaphragm and other

respiratory muscles which then causes reflex vasoconstriction of limb vasculature during control

or loaded conditions and local vasodilation with unloading of the respiratory muscles (Wetter,

Harms, Nelson, Pegelow, & Dempsey, 1999). Increased sympathetic excitation of limb

vasculature is measured by norepinephrine (NE) spillover (Harms et al., 1997). Harms et al.

(1997) demonstrated a 78 ± 5% increase in NE with respiratory muscle loading (28% increased

work of breathing) during maximal exercise, which was significantly related to LVR (r = 0.71, P <

0.05) suggesting a significant increase in limb vasoconstriction with increased work of breathing.

To identify the specific role that the respiratory muscles had in the aetiology of blood stealing,

Sheel et al. (2001) investigated the effect that respiratory muscle fatigue had on Qlegs at rest.

35
Using a combination of varying inspiratory resistance and respiratory patterns to fatigue the

respiratory muscles in different ways, (i.e. changes to fb and duty cycle) Sheel et al. (2001)

demonstrated decreased Qlegs (-30%) and increased LVR (50-60%), when compared with control

conditions, only at the point of diaphragm fatigue. Voluntary increases in inspiratory effort without

respiratory muscle fatigue had no effect on either Qlegs or LVR, indicating that the work of

breathing must be sufficient to cause respiratory muscle fatigue in order for blood stealing to occur.

The contrasting results above that support and refute the existence of blood stealing may be

explained, however, by methodological factors. Wetter et al. (1999) observed that the effect of

respiratory muscle unloading on Qlegs did not occur unless exercise intensity was greater than

75% V&O 2 max. During sub-maximal (50% and 75% of V&O 2 max) exercise, V&O 2 total increased with

increasing work of breathing and decreased with decreasing work of breathing but Qlegs and LVR

did not change with changing work of breathing. Even substantial increases in the work of

breathing (50-70%), although raising whole body V&O 2 , did not cause alterations in LVR or Qlegs

during sub maximal exercise (Wetter, Harms, Nelson, Pegelow, & Dempsey, 1999). Therefore, it

would appear that blood stealing only occurs during very high intensity exercise, and under

conditions that fatigue the respiratory muscles. Although Kowalchuk et al. (2002) used a much

greater inspiratory resistance than Harm et al. (1997; 1998) (7cmH2O—L-1—s-1 vs. 3-5cmH2O—L-1—s-1

respectively) at a work load that had previously been shown to induce decreased Qlegs, they

controlled both fb and VT during all experimental protocols. It is possible that by doing this the

respiratory muscles were not fatigued enough to invoke the metaboreflex and subsequent

changes to Qlegs and LVR. Further, Richardson et al. (1995) and Richter et al. (1992) increased

exercising muscle mass by adding another peripheral muscle group, rather than increasing the

36
work of breathing. The respiratory muscles were not specifically fatigued in these studies, and as

the exercise intensity (<75% V&O 2 max) was unlikely to be sufficient to induce respiratory muscle

fatigue indirectly, it is unlikely that metaboreflex was invoked.

When methodological conditions are taken into consideration, the evidence presented above

suggests that blood stealing does occur when the work of breathing is increased. The work of the

respiratory muscles in normal physiological conditions at maximal exercise intensity appears to

have two effects on the cardiovascular response – a) a substantial proportion of cardiac output is

directed to the respiratory muscles to support their metabolic requirements as discussed above,

and b) blood flow is reduced to (or ‘stolen from’) working locomotor muscles (Harms et al., 1997;

Harms et al., 1998). Limb muscle force output and fatigue are highly responsive to changes in

limb muscle blood flow under conditions of high intensities of muscle contraction (Romer &

Dempsey, 2006). The effects of changing limb blood flow have been attributed to changes in O2

transport and/or changes in the washout of local metabolites (Barclay, 1986). Given that blood

flow to the legs during maximal intensity exercise may be in the order of 15.4 ± 0.3 L—min-1 (Harms

et al., 1997), a decrease of 1.3 L—min-1 (8.4%) would likely have a large impact on performance.

To investigate this effect, Harms et al. (2000) again used a PAV to offload, (and inspiratory

resistance to load) the respiratory muscles during incremental tests to exhaustion at V&O 2 max in

competitive cyclists. When inspiratory muscle work was decreased by 63%, subjects were able to

exercise 1.3 ± 2 minutes longer (range -1.2 – 5.3 min; +14.4% ± 4.9%, p<0.05) than during

control conditions, whereas with loading, subjects exercised 1.0 ± 0.8 minutes less (range -0.4-

4.7 min; 15.1 ± 3.3%; p<0.05) (Harms, Wetter, St Croix, Pegelow, & Dempsey, 2000). Clearly this

suggests that the decrease in Qlegs caused by increased work of breathing was sufficient to alter

37
cycling performance. To further investigate the effect of blood stealing on performance, and using

similar methodology, Romer and Dempsey (2006) recently measured quadriceps power, as a

measurement of peripheral fatigue in eight endurance trained male cyclists, before and after

incremental cycling in a randomised series of control, inspiratory muscle loading and unloading.

Exercise at 90% of maximum work rate (292W for 13.2 minutes) under control conditions reduced

force output of the quadriceps by 28%, indicating an expected presence of quadriceps fatigue

after exercise to exhaustion. This fatigue was gradually recovered with 70 minutes of rest

following exercise. After completion of an identical amount and duration of exercise as in the

control, but with unloading sufficient to reduce the work of breathing by 50-60%, limb fatigue was

8% less than the control, whereas respiratory muscle loading increased fatigue by 20% greater

than the control and quadricep strength had not recovered by 70 minutes. This suggests that

preventing the work of breathing from increasing to the point that the Qlegs is compromised, is an

important consideration for cyclists in order to optimise power output and exercise performance.

The findings by Harms et al. (1997, 2000), Sheel et al. (2001) and Romer et al. (2006) clearly give

strength to the suggestion that increased respiratory muscle work, particularly to the point of

respiratory muscle fatigue, will decrease blood flow to the exercising peripheral muscles to the

degree that physical exercise performance is impaired. If the altered respiratory muscle

mechanics experienced by cyclists due to factors such as breathing pattern disorders,

development of a tachypnoeic breathing pattern and the normal physiological response to

exercise was such that respiratory muscle fatigue was invoked, theoretically performance at

maximal intensity exercise would be compromised. Conversely, if strategies or interventions were

employed to optimise respiratory mechanics and prevent unnecessary respiratory muscle fatigue,

performance at maximal intensity exercise could be improved by decreasing blood stealing.

38
Clearly, maintaining optimal respiratory mechanics may be a simple way to provide the small

improvement in performance that is so sought after by cyclists and sports scientists alike.

Maintaining optimal respiratory mechanics to prevent respiratory muscle fatigue may have a

second benefit with regards to optimising cycling performance – quite separate to preventing

blood stealing. Respiratory muscle fatigue is thought to be an important component in the

development of dyspnoea, which is a major factor in limiting exercise in healthy humans.

2.3.7. Dyspnoea

Dyspnoea is one of the primary reasons given by healthy individuals for the termination of

exercise to exhaustion (Hamilton, Killian, Summers, & Jones, 1996; Killian et al., 1992) indicating

that some aspect of ventilation limits exercise. Dyspnoea is the subjective awareness of an

increased work of breathing (Webber & Pryor, 1994, p. 5). It is a general term used to

characterise a range of different descriptors that vary in intensity and are influenced by a wide

variety of factors, such as cultural expectations and previous experience (Grazzini, Stendardi,

Gigliotti, & Scano, 2005). In healthy humans, dyspnoea limits exercise when it reaches a level

that is deemed no longer tolerable (Grazzini, Stendardi, Gigliotti, & Scano, 2005).

The pathophysiology underlying dyspnoea is both complex and uncertain and a detailed

discussion of all potential contributing factors to dyspnoea during exercise (ATS, 1998; McConnell

& Romer, 2004a) is beyond the scope of this review. Given the complexity of changes to

respiratory mechanics during exercise it is difficult to be sure which alterations contribute most

strongly to the sensation of dyspnoea. However, with regards to limiting exercise in athletes, it

appears that the contractile properties of the respiratory muscles, combined with operational high

39
lung volumes and the conscious awareness of the outgoing respiratory motor command are likely

contributing factors (Grazzini, Stendardi, Gigliotti, & Scano, 2005). Factors that impair the

contractile properties of the respiratory muscles (for example functional weakening, pattern of

recruitment and fatigue) have the potential to increase the intensity of dyspnoea, whereas factors

that improve the contractile properties of these respiratory muscles have the potential to reduce

the intensity of dyspnoea (McConnell & Romer, 2004a).

In the 1960’s Moran Campbell coined the term length-tension inappropriateness (Campbell, 1966)

and suggested that humans have a quantitative appreciation of the degree of effort associated

with breathing. Further, dissociation or a mismatch between central respiratory motor activity and

the mechanical response of the respiratory system may produce a sensation of respiratory

discomfort (dyspnoea) (McConnell & Romer, 2004a). As such, the intensity of dyspnoea is

increased when changes in respiratory muscle length (i.e. volume) or tension (i.e. pressure) are

inappropriate for the outgoing motor command. During exercise, the central outgoing drive for

ventilation is increased, however the decrease in respiratory muscle efficiency through altered

respiratory mechanics (i.e. breathing occurring on a less compliant portion of the pressure-volume

curve, increased elastic recoil of the respiratory system at end expiration, and respiratory muscle

shortening) (ATS, 1998) does not enable appropriate volume and pressure changes to occur. In

other words, although the drive to breathe has been increased in response to intensifying exercise,

the capability of the respiratory muscles to respond appropriately is compromised. It is possible

that this inappropriate response is perceived in the brain as ‘not breathing enough’ resulting in the

sensation of dyspnoea. Theoretically, interventions that restore more efficient respiratory

mechanics could enable the respiratory muscles to respond more appropriately to the increased

central output and therefore decrease perceptions of dyspnoea (McConnell & Romer, 2004a).

40
The order in which the respiratory muscles are recruited in response to exercise may also

contribute to the intensity of dyspnoea. Assuming a diaphragmatic breathing pattern is present,

the diaphragm is recruited first, followed by increasing activation of the respiratory accessory

muscles as breathing frequency increases. Animal studies show that the diaphragm is less well

supplied with proprioceptors than the intercostal muscles (accessory muscles) (Corda, von Euler,

& Lennerstrand, 1965). If the same is assumed in humans, more sensory input as to the work of

the muscles would be received centrally from the accessory muscles, than from the diaphragm,

potentially giving the impression that more work was being done, with a subsequent increase in

perceived effort, when the accessory muscles were recruited.

The importance of the inspiratory accessory muscles in the aetiology of dyspnoea is further

supported by evidence that the increase in respiratory effort during fatiguing inspiratory resistive

loading correlated better with the level of activation of muscles of the rib cage and neck muscles

than with the level of activation of the diaphragm (Ward, Eidelman, & Stubbing, 1988). Further,

loaded breathing tasks preferentially fatigue the inspiratory accessory muscles rather than the

diaphragm (Hershenson, Kikuchi, & Tzelepis, 1989). As the force generating capacity of the

diaphragm exceeds that of the accessory muscles (Hershenson, Kikuchi, & Loring, 1988), a

recruitment strategy that brings in the accessory muscles, especially sooner than necessary

through a faulty breathing pattern, may necessitate a greater motor outflow since recruitment of

weaker muscles to achieve a given intra-thoracic pressures change would require a higher level of

motor outflow (McConnell & Romer, 2004a). Minimising, or at least delaying, the recruitment of

the accessory muscles by encouraging maintenance of a diaphragmatic breathing pattern could

41
attenuate or delay the onset of dyspnoea by decreasing sensory input received from the

accessory muscles regarding the work being done.

If dyspnoea was due solely to mechanical or physical changes to the respiratory pump, then one

would expect patients with similar FEV1 (forced expiratory volume in one second; a measure of

lung function) values to report similar levels of dyspnoea. However, Cooper (2006) reported there

was no correlation between these two variables, supporting the theory that dyspnoea may also be

influenced by cerebral factors (ATS, 1998). It is acknowledged that the conscious awareness of

the outgoing respiratory motor command has an important role in the perception of dyspnoea

(McConnell & Romer, 2004a), and that sensory adaptation to intensity of dyspnoea may also

occur (ATS, 1998). Whilst the magnitude of these topics puts them beyond the scope of this

review, it is important to note that dyspnoea, by definition, is a person’s perception of

breathlessness (Grazzini, Stendardi, Gigliotti, & Scano, 2005). It is a personalised experience,

likely affected by a raft of psychological and cerebral components, and any strategies used in an

attempt to attenuate dyspnoea must recognise the subjectivity of the sensation and manage this

appropriately (Grazzini, Stendardi, Gigliotti, & Scano, 2005).

2.4. Breathing as a Limiting Factor to Exercise and Strategies to Overcome it

2.4.1. Breathing as a Limiting Factor

Traditionally, the idea that breathing may be a limiting factor to exercise has been disregarded

due to the apparent breathing reserve observed in athletes even at the highest level of exertion

(Dempsey, 1986). This argument was based on the fact that even at exhaustion both trained and

sedentary subjects breathed below their maximal voluntary ventilation (MVV; maximal voluntary

42
ventilation, usually measured in 15 seconds) and therefore still had a significant breathing reserve

to increase ventilation further if needed (breathing reserve is calculated as MVV – VEmax, (Markov

et al., 2001)). For example, immediately after exercise to exhaustion at 85 and 95% of V&O 2 max,

Johnson et al. (1996) recorded values of 70 and 83% MVV respectively in 19 healthy individuals

( V&O 2 max range 40-80ml.kg-1.min-1) after exercise to exhaustion. Similarly, elite cyclists reached

86.9% MVV and professional cyclists 90% MVV (Lucia, Carvajal, Calderon, Alfonso, & Chicharro,

1999) indicating that even highly trained athletes, performing at maximal levels of exhaustion,

have an ability to increase their ventilation further.

However, in large studies (N= 578; N=417) comparing reasons for volitional termination in

maximal intensity exercise tests in healthy individuals and those with respiratory pathology, both

Hamilton et al. (1996) and Killian et al. (1992) identified that breathlessness (either alone or in

combination with leg fatigue) contributed equally in both groups. Despite the healthy individuals

having no underlying respiratory pathology and, based on the above studies, presumably ample

breathing reserve (Johnson, Saupe, & Dempsey, 1992; Lucia, Carvajal, Calderon, Alfonso, &

Chicharro, 1999), breathlessness still caused some healthy individuals to terminate exercise.

Taking into consideration the apparent breathing reserve, this would suggest that although the

physical capability exists to increase ventilation to meet the metabolic demands of maximal

exercise, factors exist that prevent the person from maximizing their ventilatory effort.

The literature presented in the previous section indicates that breathing may be a factor that

compromises maximal exercise performance if the work of breathing is such that respiratory

muscle fatigue occurs. Increased work of breathing has been shown to increase respiratory

43
muscle fatigue (Johnson, Babcock, Suman, & Dempsey, 1993), decrease blood flow to the

locomotor muscles (Harms et al., 1997) and increase dyspnoea (Grazzini, Stendardi, Gigliotti, &

Scano, 2005) – a combination of which negatively affect exercise performance (Harms, Wetter, St

Croix, Pegelow, & Dempsey, 2000). It would therefore be plausible to suggest that strategies that

decrease the work of breathing would attenuate these negative consequences and have a

positive effect on exercise tolerance. Certainly, Harms et al. (2000) demonstrated that decreased

work of breathing increased exercise performance. More specifically, these authors demonstrated

that decreasing the work of breathing using a PAV significantly decreased dyspnoea (0-10 Borg)

at both isotonic (5 minutes, 8.5 ± 0.3 unloaded; 9.6 ± 0.2 control) and at end-exercise

(approximately 9 minutes, 9.6 ± 0.1; 9.9 ± 0.1). Of particular relevance to athletes and sports

performance was that the significant increase in time to exhaustion was most strongly correlated

with the change in dyspnoea (r=-0.54, P ≤ 0.05) rather than changes to RPEleg (r = -0.42, P ≤

0.05), VE (r = 0.29, P ≤ 0.05) or V&O 2 (r=-0.39, P ≤ 0.05).

2.4.2. Inspiratory Muscle Training

One strategy that appears to decrease the symptoms and exercise limiting consequences

associated with poor respiratory mechanics and increased work of breathing is inspiratory muscle

training (Markov et al., 2001; Romer, McConnell, & Jones, 2002c; Witt, Guenette, Rupert,

McKenzie, & Sheel, 2007). Inspiratory muscle training (IMT) aims to increase the strength and

endurance of the respiratory muscles and requires a person to breathe against a variable

resistance, usually via a hand-held device. A variety of IMT methods exist including voluntary

isocapnic hyperpnoea (Leddy et al., 2007; McMahon, Boutellier, Smith, & Spengler, 2002), flow

resistive loading (Hanel & Secher, 1991) and pressure threshold loading (Romer, McConnell, &

44
Jones, 2002a; Volianitis et al., 2001); each with their own methodological strengths and

weaknesses (Table 3).

Table 3. Different Methods of IMT

Voluntary Isocapnic Flow Resistive Pressure Threshold

Hyperpnoea (VIH) Loading (IFRL) Loading (IPTL)
Procedure Maintain high targets of Inspire via variable Individuals produce a
ventilation for up to 30 diameter orifice whereby negative pressure to
minutes. for a given flow the smaller overcome a threshold load
the orifice, the greater the and therefore initiate
load. Up to 30 minutes inspiration

Intensity 50-90% MVV 80% MIP 50% P0

Benefits ↑Time to exhaustion ↑ Inspiratory muscle ↑ Maximal force
↑ Maximum sustained strength production
ventilatory capacity ↑ Total lung capacity ↑ Maximal velocity of
↑ VE (small) muscle contraction
↑ MVV ↑ Maximal rate of
shortening
↑ Maximal power output
↑ Respiratory muscle
endurance

Weaknesses Time consuming Harder to ensure training Does not address

Doesn’t affect maximum load is reached secondary breathing pattern. May
pressure generating to changes to inspiratory exacerbate poor
capacity. pressure, which varies with respiratory mechanics.
Possibly dries airways flow.
secondary to high
ventilatory flow rates

Effects on respiratory Only affects velocity axis of Only affects force axis of Affects both force and
mechanics force-velocity relationship force-velocity relationship velocity axis of force-
velocity relationship.

An elegant discussion of the strengths, weaknesses and methodological considerations of IMT is

provided by McConnell and Romer (2004b) and therefore will not be covered in depth in this

review. In addition to different modes of IMT, a variety of training durations, intensities and

protocols are typically used in studies investigating the effect of IMT upon exercise performance

and are presented in Table 4.

45
Table 4. A summary of Exercise Performance Studies using Inspiratory Muscle Training (IMT)
Study Age (years) Subject Data Study Design IMT Intervention Measures Collected Main findings of the effect
of RMT intervention
Morgan, DW Trained male RCT 3/52 VIH VO2max No change in Endurance time
Kohrt, W M cyclists IMT N = 4 Cycling end. 95% VO2max No change VO2max
Bates, B J Control N = 5 RME
Skinner, J S MVV
(1987)
Fairbarn, M S Experienced RCT 4/52 VIH VO2max RME 12%
Coutts, K C male cyclists IMT N = 5 (16 sessions) RME No change VO2max
Pardy, R L Control N = 5 Cycling end. @ 90% Pmax No change cycling endurance
McKenzie, D C time
(1991)
Hanel, B Healthy Humans RCT 27.5 days, 12 min run ↑ MIP 18%
Secher, NH IMT N = 10 30mins, 2 x daily IFRL PImax ↑12 min run 8% (control ↑ 6%)
1991 Control N = 10 VEmax No change VEmax
VO2max No change VO2max
fb ↓fb 5%
Spenger, C M 26.3 ± 3.3 Athletic males No control 4/52 VIH Cycling endurance ↑Cycling endurance 27%
Roos, B N = 20 30 mins, 5 x week Lung function No change lung function
Laube, S M VO2peak ↑VE 23%
Boutellier, U Pmax ↑fb 37.5%
(1999) Lactate ↓lactate after endurance 7%
MVV ↓lactate after incremental 15%
fb ↑MVV 19%
VE No change lung function
Chatham, K 26 ± 14 Healthy, active RCT 8/52 IFRL RRPE ↑ VO2max (Pred.) 3.78%
Baldwin, J 10 Males (5 IMT N = 11 80% MIP PRPE ↑ shuttles
Griffiths, H active) Control N = 11 3 x week VO2max Predicted ↑ RMS 31%
Sumers, L 12 females (6 Shuttle runs ↑ RME 42%
Enright, S active) RMS No change to RRPE
1999 RME No change to PRPE
No change in control
Inbar, O 28.9 ± 8.9 Well trained RCT 10/52 IPTL using threshold PmPeak ↑ PmPeak 10%
Weiner, P endurance track SIMT inspiratory muscle trainer, PImax ↑ PImax 24.9%
Azgad, Y athletes N = 10 30min, 6x / week VEmax No change to VEmax or VO2max
Rotstein, A Placebo N= 10 30-80% WImax VO2max
Weinstein, Y
2000

46
Study Age (years) Subject Data Study Design IMT Intervention Measures Collected Main findings of the effect
of RMT intervention
Markov, G RMT 43 ± 7 Healthy, RCT 40 training sessions over 14 Wpeak ↑ Breathing endurance
Spengler, C ET 40 ± 10 sedentary IMT N = 15 weeks of: VO2peak ↑ Cycling endurance
Knopfli-Lenzin, C Placebo 37 ± 9 subjects ET N = 10 1. VIH 30mins (60%MVV, Cycling Endurance SV unchanged
Steussi, C Males = 20 Placebo N = 15 continuous breathing) Stroke Volume
Boutellier, U Females = 18 2. ET 30 mins running or
2001 cycling
3. No training
Sonetti, D A Competitive male RCT 4/52 IPTL AND VIH MIP ↑ MIP 8%
Wetter, T J cyclists IMT N = 9 3-5 mins IPTL and 30 mins Fixed work rate cycling ↑ Fixed work rate 26% (control
Pegelow, D F Control N = 8 VIH day, 5 x week 8km TT 16%)
Dempsey, J A Pmax ↑ 1.8%
(2001) Lactate ↑Pmax 9% (control 6%)
HR No significant difference
VE between control and IMT in
above variables
No change lactate
No change HR
No change VE
Steussi, C 43 ± 7 Sedentary RCT 40 x 30 min sessions Static lung function ↑ Cycling end 23%
Spengler, C M Males = 16 IMT N = 13 VIH Cycling end @ 70% Pmax ↑ RME 632%
Knopfi-Lenzin, C Females = 12 Control N = 15 Pmax No change lung function
Markov, G HR No change HR
Boutellier, U VE No change VE
(2001) No change Pmax
Volianitis, S 23.8 ± 3.8 Competitive RCT 11/52 IMT 6 min all out 4/52:
McConnell, A K female rowers IMT N = 7 30 inspiratory efforts @ 50% Sub max incremental load ↑ MIP 40.9 %
Koutedakis, Y Control N = 7 MIP 5000m TT 6 min 3.4%
McNaughton, L 2 x daily RRPE 5000m 3.1%
Backx, K Lactate RRPE no change
Jones, D A VT 11/52
(2001) fb ↑ MIP 45.3 % (from baseline)
6 min 3.5% (from baseline
5000m N/A
↓Lactate 1.3% (not different
from control)
↑ VT
No change fb

47
Study Age (years) Subject Data Study Design IMT Intervention Measures Collected Main findings of the effect
of RMT intervention
Williams, J S 20.9 ± 1.2 Cross country No placebo 4/52 IMT with inspiratory PImax ↑31% PImax
Wongsathikun, J runners. resistive device. 25 mins breathing endurance time ↑128% breathing endurance
Boon, S M N=7 4-5 x week VO2peak time
Acevedo, E O Males = 5 50% MIP ↑ by 5% each run time @ 80% VO2peak VO2peak No change
2002 Females = 2 week Run time no change
McMahon, M E RMT = 26 ± 4 Experienced RCT 4-6/52 VIH Wpeak Wpeak no change
Boutellier, U Control 28 ± 6 male cyclists RMT N = 10 IMT = 30 min sessions x Cycling endurance time to ↑ Cycling endurance time
Smith, R M Control N = 10 20, VE = 60% MVV exhaustion @ 85% Wpeak 3.26%
Spengler, C Control = unclear Breathing endurance ↑ Breathing endurance
2002 Peripheral chemoreceptor ↓ Peripheral chemoreceptor
sensitivity sensitivity
Romer, L M IMT 29.5 +/- 3.3 Male DBRCT 6/52 IMT using Inspiratory muscle pressure pre ↓Po change after 20 &
McConnell, A K Placebo 30.3 +/- 2.6 competitive IMT N=8 POWERbreathe and post 20km and 40kmTT. 40kmTT. Intervention group
Jones, D A cyclists Placebo N=8 30 dynamic inspiratory returned to baseline at 30
2002C (5 triathletes) efforts, twice a day. mins, placebo >30mins
Pressure threshold 50% of
WImax
Romer, L M IMT 29.5 +/- 3.3 Male DBRCT 6/52 IMT using VT ↑VT= 0.16L (90%Wmax)
McConnell, A K Placebo 30.3 +/- 2.6 competitive IMT N=8 POWERbreathe Respiratory RPE ↑VT= 0.17L(100%Wmax)
Jones, D A cyclists Placebo N=8 30 dynamic inspiratory Peripheral RPE ↑20kmTT = 65 ± 30s
2002A (5 triathletes) efforts, twice a day. 20kmTT & 40kmTT time ↑40kmTT = 114 ± 38s
Pressure threshold 50% of VO2max ↓RRPE 16 ± 4%
WImax Power output ↓PRPE 18 ± 4%
No change to power or
VO2max
Romer, L M IMT 21.3 ± 1.1 Male sprint DBRCT 6/52 IMT using Recovery time ↓recovery time 6.9 ± 1.3%
McConnell, A K Placebo 20.2 ± 0.7 athletes IMT N=12 POWERbreathe RRPE ↓RRPE 7.9 ± 0.6%
Jones, D A Placebo N=12 30 dynamic inspiratory PRPE ↓PRPE 7.2 ± 0.6%
2002B efforts, twice a day.
Pressure threshold 50% of
WImax
Holm, P Experienced IMT N = 10 4/52 VIH RME ↑ RME 12%
Sutter, A cyclists or Placebo N = 4 20 sessions 30 mins VE ↑ VE 20%
Fregosi, R F triathletes Control N = 6 Gradually increasing fb r VT Fb ↑ Fb significantly” (P = 0.013)
(2004) Males = 4 to achieve 18-19 or RPE VT No change VT
Females = 16 scale HR No change HR
Approx 40km-TT ↑ TT 4.7%

48
Study Age (years) Subject Data Study Design IMT Intervention Measures Collected Main findings of the effect
of RMT intervention
Edwards, A M Healthy, IMT = 10 4/52 IMT using MIP ↑ significantly (P < 0.01)
Cooke, C B physically active Control = 8 POWERbreathe (IPTL) Run time to exhaustion (treadmill) ↑ Run time 4.3% (P < 0.05)
(2004) males 30 dynamic inspiratory Blood lactate No change lactate
efforts, twice a day. Lung function No change lung function
Pressure threshold 50% of VO2 No change VO2
MIP
McConnell, A K IMT = 23.2 ± 2.64 Recreationally RCT 6/52 IMT using MIP ↑ Increased 26%
Sharpe, G R Placebo = 23.8 ± active unclear N = 12 POWERbreathe Cycling Power at MLSS Cycling Power at MLSS no
2005 4.75 whether male vs IMT N = 6 30 dynamic inspiratory Lactate at MLSS change
female? Sham N = 6 efforts, twice a day. ↓ Lactate at MLSS
Pressure threshold 50% of
MIP
Leddy, J J RMT 29 ± 8 Experienced RCT 4/52 VIH PImax PImax no change
Limprasertkul, A Control 34 ± 6 male distance RMT N = 15 RMT = 30 mins, 5 x week PEmax PEmax no change
Patel, S runners Control N = 7 Control = similar protocol VO2peak ↑VO2peak
Modlich, F with no rebreathing bag. Run time @ 80% VO2peak ↑Run time @ 80% VO2peak
Buyea, C 4 mile time ↑4 mile time
Pendergast, D R
Lundgren, C E G
2007
DBRCT = double blind, randomised controlled study; RCT = randomised controlled trial; VIH = Voluntary Isocapnic Hyperpnoea; IFRL = Inspiratory Flow Resistive Loading; IPRL
= Inspiratory Pressure Resistive Loading; WImax = Maximum Inspiratory Work; IMT = Inspiratory Muscle Training; RMT = Respiratory Muscle Training; TT = Time Trial; Po =
Pressure generated by inspiratory muscles at zero flow; VO2max = Maximum oxygen uptake; Vmax = maximum flow; RRPE =respiratory rate of perceived exertion; PRPE =
peripheral rate of perceived exertion; Wmax = Maximum power; SIMT = specific inspiratory muscle training; VEmax = maximal minute ventilation; ET = endurance training; MVV =
maximum voluntary ventilation; VT = tidal volume; SV = stroke volume; Wpeak = Peak power; VO2peak = peak oxygen uptake; VIH = Voluntary Isocapnic Hyperpnea; MLSS =
maximum lactate steady state; RME = respiratory muscle endurance

49
Whilst it is generally accepted that IMT is effective in increasing the strength and endurance of the

respiratory muscles (8 - 45.3% and 12-632% respectively) regardless of mode (see Table 4), some

studies have failed to report positive ergogenic (performance enhancing) effects as a result of

these changes (Fairburn, Coutts, Pardy, & McKenzie, 1991; Hanel & Secher, 1991; Morgan, Kohrt,

Bates, & Skinner, 1987). The lack of performance changes reported by these authors may be due

to methodological factors, rather than inadequacy of the intervention itself. For example, Fairburn

et al. (1991) reported no change to time to exhaustion in a fixed rate (90% V&O 2 max) ride following

RMT despite changes to inspiratory muscle strength and endurance in highly trained cyclists

( V&O 2 max = 66.1± 4.7 ml—kg—min-1). However on closer inspection McConnell & Romer (2005)

identified large performance changes were in fact made – for example, a mean 25% increase in

time compared with 4% in the control, but the study had insufficient power due to low subject

numbers (n=10) and the results failed to reach significance.

Conversely, there are a number of studies that demonstrate IMT produces a range of positive

performance benefits; for example, improved cycling endurance ((24%, P < 0.01) (Markov,

Spengler, Knopfli-Lenzin, Stuessi, & Boutellier, 2001), (3.26 ± 4.29%) (McMahon, Boutellier, Smith,

& Spengler, 2002) (23%; P < 0.05) (Steussi, Spengler, Knopfli-Lenzin, Markov, & Boutellier, 2001))

cycling time trial time (3.8 ± 1.7) (Romer, McConnell, & Jones, 2002a), recovery time following high

intensity sprints (6.9 ± 1.3%; Romer, McConnell, & Jones, 2002b), rowing (3.1 ± 0.8%; Volianitis et

al., 2001), running (4.3%; P < 0.05; Edwards & Cooke, 2004) (50%; P < 0.001; Leddy et al., 2007)

and swimming performance (66%; P < 0.001; Wyelegala, Pendergrast, Gosselin, Warkander, &

Lundgren, 2007) (Table 4).

50
A further benefit of IMT appears to be the ability to decrease dyspnoea – another consequence of

altered respiratory mechanics that may compromise exercise performance. Just as factors that

impair the contractile properties of the respiratory muscles are thought to contribute to dyspnoea,

factors that improve the contractile properties are thought to attenuate it (McConnell & Romer,

2004a). Inspiratory muscle training appears to be effective in decreasing dyspnoea by increasing

inspiratory muscle strength and therefore enabling the inspiratory muscles to create greater length

and tension changes for each breath, in response to exercise (McConnell, 2005; McConnell &

Romer, 2004a; Romer, McConnell, & Jones, 2002a).

It is not clearly understood how and/or why IMT improves performance. Inspiratory muscle training

has consistently been shown to have no effect on V&O 2 max or traditional parameters of respiratory

function i.e. PEF, FEV1 (Inbar, Weiner, Azgad, Rotstein, & Weinstein, 2000; Romer, McConnell, &

Jones, 2002c). One theory receiving increasing interest is that by delaying or decreasing

respiratory muscle fatigue by increasing inspiratory muscle strength and endurance, the reflex

activity from the type III/IV chemo-sensitive receptors in the respiratory muscles thought to trigger

sympathetic vasoconstriction in the limbs, may be attenuated (McConnell & Romer, 2004b). Witt et

al. (2007) supported this theory by demonstrating a significant decrease of sympathetically-

mediated heart rate (83 ± 4 vs. 74 ± 2 beats—min-1) and mean arterial pressure (99 ± 3 vs. 89 ± 2

mmHg) during a eucapnic resistive breathing task designed to invoke respiratory muscle fatigue

after a five week intervention of IMT. In providing an explanation for why whole-body exercise

improvements are seen following IMT, these authors suggested that the attenuated cardiovascular

response they observed indicates IMT causes blunted sympatho-excitation to resistive inspiratory

51
work, which may subsequently protect locomotor muscle blood flow. McConnell and Sharpe

(2005) demonstrated a significant decrease in blood lactate concentrations (B[La]) at MLSS power

(power at maximal lactate steady state) following six weeks of IMT and suggested that the

ergogenic effect of IMT may be also be related to reduced B[La].

Whether IMT is effective in eliciting improved exercise performance and decreased dyspnoea by

decreasing the work of breathing (McConnell & Romer, 2004b; Witt, Guenette, Rupert, McKenzie,

& Sheel, 2007) or by some other means, it appears from studies to date that increasing the

strength and endurance of the respiratory muscles is a common and important finding (Inbar,

Weiner, Azgad, Rotstein, & Weinstein, 2000; Markov et al., 2001; Romer, McConnell, & Jones,

2002a, , 2002b, , 2002c). Interestingly, it does not appear from the methodology of any of the

studies that attention is paid to the pattern of breathing used by subjects when completing their IMT.

In some instances, it might also exacerbate (or reinforce) an already poor breathing pattern. A

recent case history presented by Dickinson, Whyte & McConnell (2007) is the only known paper

that mentions the breathing pattern that is used during IMT. These authors specifically instructed

the athlete to breathe to her diaphragm and to minimise cranial (upwards) shoulder movement.

With exception of the case history of Dickinson et al. (2007), IMT does not appear to specifically

address dynamic hyperinflation or the recruitment order of the inspiratory muscles – rather it looks

to increase inspiratory muscle strength and endurance within the positional constraints of altered

respiratory mechanics. However for athletes (who are presumably able to attain optimal

respiratory mechanics in the absence of respiratory pathology) it would seem advantageous to

specifically address factors that are known to contribute to respiratory muscle fatigue and

dyspnoea, for example dynamic hyperinflation and the respiratory muscle recruitment order by

52
addressing the mechanics of breathing, to decrease the work of breathing and optimise exercise

performance.

2.4.3. Breathing Pattern Retraining

A strategy that specifically addresses dynamic hyperinflation and respiratory muscle recruitment

pattern, and pays specific attention to breathing pattern both at rest and under load, is breathing

pattern retraining (Blager, 2000; Jones, Dean, & Chow, 2003). Physiotherapists have been using

breathing exercises as a treatment modality for respiratory conditions for many years, with

breathing pattern retraining being recommended for breathing pattern disorders as early as 1938

(Soley & Shock, 1938). Clinically, physiotherapists used diaphragmatic breathing pattern retraining

and/or pursed lip breathing 1 to relieve the symptoms and exercise limitations associated with

COPD (Jones, Dean, & Chow, 2003), VCD (Blager, 2000) and BPDs (Chaitow, Bradley, & Gilbert,

2002; Lum, 1977). Diaphragmatic breathing retraining involves educating the patient of its

importance, discussion of potential feelings of discomfort that may arise during the retraining

process, and demonstration and instruction of how to diaphragmatically breathe correctly (Chaitow,

Bradley, & Gilbert, 2002; Lum, 1977) (refer to Appendix 2 and 3 for a further description of clinical

breathing pattern retraining and BPR exercises). The beneficial effects of diaphragm breathing

may be due to decreased work of breathing through improved respiratory mechanics, decreased

dynamic hyperinflation and increased tolerance and normalisation of PaCO2 through a slower,

deeper respiratory pattern (Falling, 1995; Thoman, Stoker, & Ross, 1966; Tisp et al., 1986).

Pursed Lip Breathing (PLB) is taught clinically to assist airflow in patients with COPD (Jones, Dean,

& Chow, 2003) and VCD (Blager, 2000) and to help elongate the expiratory phase in athletes to

maintain ventilation at appropriate lung volumes (Fallon, 2004). The emphasis is on expiring

1 Normal inhalation with expiration through slightly parted lips, exhaling air over the bottom lip

53
longer than the inspiratory phase to enhance lung emptying (Jones, Dean, & Chow, 2003), avoid a

tachypnoeic shift as exercise intensity increases and to avoid breath stacking. The benefits of PLB

may be due to decreased respiratory rate, changes in pattern of respiratory muscle recruitment,

longer tE and larger VT (Ito, Kakizaki, Tsuzura, & Yamada, 1999; Sharp, Drutz, & Moisan, 1980).

These changes may also lead to a decrease in dyspnoea and work of breathing (ATS, 1998).

Whilst BPR studies have been conducted with success in individuals with respiratory related

pathologies for example asthma (Grossman, DeSwart, & Defares, 1985; Ram, Holloway, & Jones,

2003; Thomas et al., 2003), HVS (Grossman, DeSwart, & Defares, 1985) and VCD (Banez &

Culbert, 2005) no study has been published detailing the potential efficacy of such techniques in

athletic populations. This is surprising given the importance of optimal breathing in athletes, the

numerous factors that can compromise breathing in athletes and the consequence of these on

respiratory mechanics and, ultimately, exercise performance. In cyclists, breathing may be

compromised by the presence of a BPD at rest, alterations to respiratory timing and pattern caused

by the normal respiratory response to exercise, psychological and postural factors, resulting in an

inefficient breathing pattern. It has been shown that an inefficient breathing pattern may increase

the work of breathing (Jones, Dean, & Chow, 2003). Increased work of breathing may cause blood

stealing (Harms, Wetter, St Croix, Pegelow, & Dempsey, 2000), decreased power output (Romer &

Dempsey, 2006), increased dyspnoea (McConnell & Romer, 2004a) and decreased exercise time

(Harms, Wetter, St Croix, Pegelow, & Dempsey, 2000). Optimising a cyclist’s breathing pattern

could potentially decrease the work of breathing such that blood stealing is delayed or prevented,

power output is maintained for longer, and a decrease or delayed perception of dyspnoea.

Dyspnoea has been identified as one of the limiting factors to exercise performance (Aliverti &

Macklem, 2001). If dyspnoea could be delayed or prevented, this could enable cyclists to both

54
train and compete at a greater intensity. In elite cyclists the margins for improvement through

training alone is thought to be 1-3% (Jeukendrup & Martin, 2001) and in a 40km-TT this small

alteration could decrease race time by up to 1.33 minutes. If dyspnoea could be minimized, higher

quality, more intense and longer training sessions might be possible and the physiological benefits

of training could be maximised.

Although diaphragmatic breathing pattern retraining and PLB are used clinically, there are no

studies reported in the scientific literature that investigate their use in the healthy, athletic

population and the true effect of these breathing exercises in respiratory pathology is unclear. For

example, Gosselink et al (1995) demonstrated that diaphragmatic breathing reduced rather than

enhanced breathing efficiency in patients with severe COPD, concluding that diaphragmatic

breathing contributed to inappropriate chest wall motion and decreased mechanical efficiency while

increasing dyspnoea. Conversely, PLB when combined with diaphragmatic breathing has been

shown to enhance pulmonary mechanics and breathing efficiency, relieve dyspnoea, slow the

respiratory rate, increase tidal volume and help to restore diaphragmatic function (Falling, 1995;

Thoman, Stoker, & Ross, 1966; Tisp et al., 1986). These are all changes that would theoretically

assist athletes in containing the work of breathing, and the consequences associated with this, and

potentially enhance exercise performance.

More recently, Jones, Dean and Chow (2003) investigated the differences in V&O 2 and respiratory

rate during three commonly described breathing exercises (PLB, diaphragmatic breathing and a

combination of the two) and spontaneous breathing at rest in patients with COPD. Subjects (N=30)

were ‘experienced’ in all three breathing exercises and could perform them on command. Over a

55
ten minute recording period, respiratory rate was decreased from 17.3 ± 4.23 breaths/min in

spontaneous breathing to 15 ± 4.32 in diaphragmatic breathing, 12.8 ± 3.53 during PLB and 11.2

± 2.7 during combination. V&O 2 decreased significantly from 174.5 ± 25.2 mL—O2—min-1 in

spontaneous breathing to 165.8 ± 22.3 in diaphragmatic breathing, 164.8 ± 20.9 in PLB and 167.7

± 20.7 in combination breathing indicating that people with COPD adopt a breathing pattern at rest

that is not associated with the least V&O 2 . Whilst significant, the differences observed by Jones et

al. (2003) were recorded at rest, and were only small changes. Theoretically the differences

between spontaneous breathing and the different breathing patterns could be larger at higher

workloads, when more efficient breathing patterns would be desirable. The findings of Jones et al.

(2003) may suggest that the components of respiratory mechanics and respiratory muscle

efficiency may play a greater role in supporting spontaneous breathing than simply minimising

V&O 2 . Interestingly, despite the fact that all patients were “experienced” in the techniques and

were able to adopt the required breathing patterns on command, none of the patterns were

adopted by subjects as their automatic breathing pattern, despite the apparent improved efficiency.

This could indicate that subjects had only learnt the patterns as a motor skill, not achieved the

subconscious level of attainment necessary to reprogram the respiratory centres.

An explanation for the variable results in studies investigating breathing retraining in COPD may be

due to lack of ‘reprogramming’ or ‘resetting’ of diaphragmatic breathing and PLB as the default or

automatic breathing pattern. No indication was given in either study regarding the length of the

intervention period. It is possible that subjects had ‘learnt’ the techniques and could voluntarily

reproduce them at will, but were unable to maintain these patterns indefinitely therefore reverted

back during spontaneous breathing. According to Grossman et al. (1985), unless the respiratory

56
centres are reprogrammed, patients can only tolerate small decreases in respiratory rate and

increased CO2 can be tolerated for approximately 10 minutes before the urge to revert to

spontaneous breathing becomes overwhelming2. It is possible that neither Gosselink et al. (1995)

or Jones, Dean and Chow (2003) incorporated a sufficient intervention period to allow central

adaptation to occur. Maclennan et al. (1994) and Fabre et al. (2007) attempted to investigate the

effect of entraining breathing with exercise during rowing and roller-ski skating respectively.

Neither author provided an intervention period for participants to adjust to the new breathing

pattern, testing time was greater than 10 minutes, and unsurprisingly neither author identified

positive changes to breathing efficiency or perception of respiratory effort as a result of entrainment.

It seems apparent that regardless of the mechanisms underpinning how breathing patterns are

‘retrained’, an unknown period of time is required for the new breathing pattern to become the

automatic pattern and therefore become beneficial. This time frame may be longer in athletes who

not only need to adjust to a new breathing pattern at rest, but also need to assimilate it to their

exercise environment and associated increased ventilatory demand.

Another factor that needs to be considered when analysing studies of breathing pattern retraining

in COPD is that often the altered breathing pattern is the effect of an impairment in respiratory

mechanics due to the disease process, rather than the symptoms of COPD (for example dyspnoea

and hyperinflation) being caused in the first instance by altered respiratory mechanics. Healthy

humans who have no underlying respiratory pathology should theoretically have no need for

altered respiratory mechanics and should therefore have the ability to achieve an optimal breathing

pattern.

2 The concept of respiratory centres being ‘reprogrammed’ is only a theoretical notion. The precise mechanisms

underpinning how breathing pattern is altered in a BPD, and subsequently ‘retrained’ to demonstrate optimal
respiratory mechanics, is not known.

57
At this stage there is insufficient evidence, predominantly due to poor methodology of those studies

that have been undertaken, to strongly support or reject the use of breathing pattern retraining in

athletes or healthy humans. For example, Lum (1983) reported the effectiveness of breathing

retraining offered by physiotherapists to patients with hyperventilation syndrome (a chronic

breathing pattern disorder). In a group of more than 1000 patients, 80% became symptom free.

However, the value of this study is limited as no details were given on the exact nature of the

treatment or the instruments used for evaluating therapeutic outcome and no control was used.

Small subject numbers, lack of control, unclear separation of treatment modalities and poor

outcome measures are weaknesses of the six studies critiqued (Table 5) that have investigated the

effect of breathing pattern retraining in individuals with breathing pattern disorders.

58
Table 5. Overview of Breathing Pattern Retraining Studies

Study Treatment Group Intervention Control Group Intervention Outcome measure Findings
(N) (N)
Lum (1983) N = 1350 Clinical judgement 80% symptom free
Intervention unclear
Grossman et al. (1985) AR BI Self-report questionnaires e.g. AR > BI, on psychological
7 sessions 7 sessions STAI, EPI measures at 1 month post
N=5 N = 22 Respiratory measures treatment
Van Doorn et al. (1982) BF BI Self report HVS symptoms, BF > BI on HVS symptoms at
7 sessions 4 sessions PetCO2 post-treatment
N = 10 N = 10 BF = BI on HVS symptoms at
3 month follow up
BF = BI on PetCO2 at post
treatment and follow up
Kraft & Hoogduin (1984) 1. Voluntary hyperventilation No control Self report HVS symptoms, 1 = 2 = 3, on psychological
N = 6; 6 sessions HVS attacks, depression measures, at post-treatment
2. Relaxation breathing retraining and one month follow up
N = 6; 6 sessions
3. Stress management
N = 7; 6 sessions
Rapee (1985) B No control Self report panic attacks, Reduction in panic attacks,
3 sessions anxiety, physical symptoms symptoms and anxiety
N=1
De Ruiter et al. (1989) B+E 1. E Self report panic attacks, B + C = 1 = 2, on
8 sessions 8 sessions agoraphobia, PCO2, breathing psychological and respiratory
N = 13 N = 13 frequency measures at post treatment
2. B + C + E
8 sessions
N = 13
AR = auditory regulation; BI = breathing instruction; BF = biofeedback; R = relaxation; C = cognitive restructuring; B = breathing retraining; CT = cognitive therapy; E = exposure;

STAI = State-Trait Anxiety Inventory; AR > BI = auditory regulation scored significantly better than breathing instruction, PetCO2 = End-tidal CO2 pressure

59
Clearly there is a paucity of well-designed studies that investigate the effect of breathing pattern

retraining in individuals with breathing pattern disorders, or on altered respiratory mechanics. In

addition to the methodological issues discussed above, nothing can be found in the scientific literature

to indicate how long each individual training session needs to be, the optimal frequency of sessions per

day or per week, or total number of weeks an intervention should be followed in order to retrain a

persons automatic breathing pattern. Further, and more specifically, no studies have been identified

that investigate the effect of breathing pattern retraining on optimising respiratory mechanics in healthy

athletes. Given the high demands placed on the respiratory system during racing and intense training,

and the potential for maladaption in the person’s mechanics of ventilation highlighted above, it has been

previously suggested that optimising an athlete’s breathing pattern would have beneficial effects on

cycling performance (Lucia, Hoyos, Pardo, & Chicharro, 2001).

2.5. Summary

Competitive road cycling is a sport where very small improvements in physiology can have a significant

affect on overall performance. Much attention is given to find ways that may enhance a rider’s mean

power during time trials and peak aerobic power during incremental endurance exercise.

An area that has not been explored to date is the mechanisms of breathing and the effect that

optimising a cyclist’s breathing pattern may have on performance. This is surprising given the

significant increases in ventilatory demands observed during maximal intensity cycling. To meet these

ventilatory demands both breathing frequency and tidal volume are altered. In most athletes this is

achieved by shortening the expiratory time rather than the inspiratory time for each breath – a change

that is associated with a tachypnoeic shift, breath stacking, dynamic hyperinflation and a subsequent

increase in the work of breathing. Increased work of breathing, particularly to the point of respiratory

60
muscle fatigue, has been associated with decreased blood flow to the leg muscles during high intensity

cycling, a consequence of which is a decreased power output and a drop in performance.

If increased work of breathing has a negative effect on cycling performance, it would be reasonable to

argue that strategies that decrease the work of breathing may have a positive effect. Whilst respiratory

muscle training has been shown to decrease the work of breathing and increase exercise performance,

it fails to specifically address the pattern of breathing. Breathing pattern retraining is an intervention

strategy used by physiotherapists to decrease the work of breathing by decreasing dynamic

hyperinflation and delaying the onset of respiratory muscle fatigue. Theoretically, optimising an efficient,

diaphragmatic breathing pattern in cyclists would serve to decrease the work of breathing, preserve

blood flow to the legs, decreases perceptions of effort which collectively would enable a cyclist to

maintain power output.

Despite the apparent benefit to cycling performance, no studies have investigated, to date, the effect

that breathing pattern retraining could have on performance in cycling. Given the strength of theoretical

argument presented in this review and the lack of existing research into the area of breathing pattern

retraining in cyclists, an investigation into the effect of breathing pattern retraining on exercise

performance in cyclists is clearly warranted.

61
Chapter Three: Methods

3.1. Participants

Twenty-four competitive male cyclists and triathletes from various cycling and triathlon clubs within the

greater Auckland area were recruited for this study by flyer distribution and direct contact to clubs

(Intervention Group N = 12, Control Group N = 12; match paired on pre-intervention 20km-TT time ).

Two participants withdrew from the intervention group during the study – one due to an acute respiratory

illness, the other due to the inability to comply with standardising his training schedule. All participants

were informed of the risks associated with the testing and the requirements of participation both verbally

and in written form, and were given the opportunity to have any questions answered. Prior to

participation, all participants gave their written informed consent in accordance with Auckland University

of Technology’s Ethics Committee guidelines and completed a medical questionnaire.

Since it is not possible to teach a person “sham breathing” without affecting their breathing pattern in

some way, some deception as to the true aims of the study was applied. In order to have a true control

group, full disclosure of the real purpose of the study was not provided to participants at its onset.

Further written information about the true study purpose was provided and written consent was gained

from the participants randomised into the intervention group. All participants were aware they could

withdraw from the study at any time with no consequences.

Participants’ characteristics are presented in Table 6. The majority of participants competed regularly in

club level races or higher, in either road cycling or triathlon. Participants were excluded from the study if

they were current smokers, had suffered an acute respiratory illness or injury that had affected training

within the previous four weeks, or had previously received any formal breathing retraining or advice.

62
The study took place in the competitive phase of the athlete’s season. Participants were training and

competing regularly at the time of the testing, with an average riding volume of 246 ± 121.8 km per

week (range 145-580 km). Training volume over the period of the study was not set by the investigator,

however a requirement of inclusion in the study was that participants were in a stable phase of training

and were requested to keep their training volume consistent over the duration of the study. To monitor

the actual training load throughout the study, participants were given a heart rate monitor (PolarS625x;

Polar Electro Oy, Kemplele, Finland) and a training diary (Appendix 4) to record training data.

Table 6. Baseline Participant Anthropometric Characteristics

Intervention* Control*
Age (yr) 38.6 ± 5.9 37 ± 10.6
Body Mass (kg) 78.3 ± 11.2 84 ± 10
Height (cm) 176.1 ± 6 182 ± 4.3
Years Competing 5.3 ± 2.8 3.9 ± 3
Hours per week training 8.8 ± 2.3 7.3 ± 4.1

*All values are mean ± SD

3.2. Equipment

3.2.1. Ergometer

An electro-magnetically braked cycle ergometer (Velotron, Racemate Inc, Seattle, USA) was used for all

physiological assessments and sub-maximal rides during the intervention period. This model of

ergometer utilises an electro-magnetic braking system to increase the amount of force applied to the

rear wheel of the machine, where the assessor can predetermine the distance required to complete (as

with the 20km-TT), or the level of resistance (i.e. power output) on the flywheel, independent of pedaling

cadence (as with the incremental step test). This ergometer allowed for the design of different

‘protocols’ or ‘courses’, which in this study were the incremental step test and sub-maximal intervention

period rides, and the 20km-TT respectively.

63
 3.2.2. Pulmonary Gas Exchange and Spirometry

An automated breath-by-breath system (Metamax 3B, Cortex, Leipzig, Germany) was used to record

pulmonary gas exchange measures during the pre- and post-intervention incremental test. The same

system was used to perform spirometry assessment in the same sessions. This system has a proven

reliability and validity when compared to the Douglas bag as the traditional gold standard measurement

(Larsson, Wadell, Jakobsson, Burlin, & Henriksson-Larsen, 2004).

Calibration of the gas analysis system was performed prior to and immediately after all assessments,

using a two-point calibration procedure. This involved calibrating the apparatus to ambient air, which

was assumed to be 20.93% O2 and 0.03% CO2, and then to a known mixture of high-tolerance

calibration gas (BOC, Auckland, NZ), composed of 14.82% O2 and 4.80% CO2. The two-point

calibration and ambient air checking procedure was repeated until acceptable values (± 0.02%) were

reached. Following the gas calibration, the flow-volume transducer was calibrated using a three litre

syringe (Hans Rudolph, US). This calibration was verified with three different ventilation rates,

according to the manufacturer’s instructions.

3.2.3. Blood Lactate

Blood lactate concentration was measured using the Lactate Pro lactate analyser (Akray, Tokyo, Japan).

Studies by Pyne, Boston, Martin, and Logan (2000) and Buckley, Bourdon and Woolford (2003) support

the accuracy, reliability and versatility of this device when compared to other devices such as the Abl

700 Series Acid-Base Analyser 2, the Accusport Lactate Meter, the YSI 2300 Stat lactate analyser and

the YSI 1500 Sport analysers.

Prior to each use, calibration of the lactate pro device was checked using the magnetic strip provided by

the manufacturer of the lactate testing strips. Blood samples were taken from the right ear lobe after the

64
lobe had been cleaned with an alcohol swab to ensure an adequate blood sample was returned, and

that this sample was not contaminated by sweat. The same Lactate Pro device was used for all

participants.

3.2.4. Borg Scale

Rating of perceived effort for leg and breathing effort was recorded using the modified Borg Scale

(MBS) (Appendix 5). This rating scale (0-10) was used in the incremental power test and the 20km time

trial. Although the Borg scale is typically used for patients with respiratory disease, normal healthy

people with exercise induced dyspnoea are also able to distinguish between different levels of

breathlessness using the MBS (Wilson & Jones, 1989). The same scale, modified to reflect leg effort

rather than breathlessness, was used so as not to confuse participants with different scales.

3.2.5. Nijmegen Questionnaire

The Nijmegen Questionnaire (Appendix 1) was used as a general screening tool to identify the presence

of an underlying breathing pattern disorder (BPD) in the participants. The Nijmegen Questionnaire is a

list of sixteen signs and symptoms that, grouped together, are used as a screening tool for anxiety-

based BPDs. There are no specific objective measures to identify abnormal respiratory mechanics or

BPDs in athletes. The sensitivity of the questionnaire in relation to the clinical diagnosis of anxiety-

based BPDs is reported to be 91% and the specificity 95% (Van Dixhoorn & Duivenvoorden, 1984).

3.2.6. Breathing Harness

A newly developed breathing harness (Zephyrtech, Auckland, New Zealand) was used with the control

group during the intervention period. The harness is worn as two straps, one around the upper chest

and the other around the abdomen (Figure 4). Each strap has a sensor that detects expansion and

65
contraction of the underlying thoracic cavity. The upper chest sensor is positioned on the lateral aspect

of the chest wall, and the lower sensor is positioned midline, on the upper abdomen.

Figure 4: Breathing Harness

The harness was used to differentiate between upper and lower chest breathing. Each of the sensors

capture the amplitude of the expansion and contraction of the ribcage during inspiration and expiration

respectively, and this data is presented in real time on a computer screen. The breathing harness is still

in the developmental stages therefore there is no reliability or validity data to support its use. However,

because the primary reason for using the harness was to create a “research purpose” for the control

group (see Intervention Sessions - Control Group, 3.3.8.), this lack of data was not deemed important.

3.3. Experimental Procedures

3.3.1. General Protocol

All tests were performed at AUT University’s Human Performance Laboratory located at the Millennium

Institute of Sport and Health, Auckland, New Zealand. The laboratory was well ventilated and

environmental conditions were standardised. The laboratory was temperature-controlled within the

range 19-21°C, and humidity was consistently in the range of 60-75%.

66
Participants were required to attend the laboratory ten times over a period of seven weeks (Figure 5).

During session one the medical prescreening questionnaires were analysed, height and body mass (in

their riding clothes, without shoes) were recorded and participants underwent pulmonary function testing

(spirometry), immediately followed by a familiarisation incremental power step test. During session two

participants performed a 20km time trial (TT) familiarisation test. Sessions three and four were the

formal baseline (pre-intervention) sessions for spirometry and the incremental power test, and the

20km-TT test respectively. Following baseline assessments, participants completed the intervention

(sessions 5-8), then post-intervention assessments (sessions 9 and 10) in the order above.

F F ST TT I ST TT

-2 -1 0 1 2 3 4 +1

Figure 5. Time line (weeks) of familiarisation trials (F), formal incremental step test (ST) and 20km
time trial (TT) and intervention period (I).

The incremental step test (including spirometry) and 20km-TT (for each of the familiarisation, pre- and

post-intervention testing sessions) were completed at least 48 hours apart. To minimise sources of

variation, the time between each of the tests, time of day and day of the week were kept consistent for

each participant between their pre- and post-intervention test sessions.

Prior to the first assessment the participant’s own bike dimensions were measured and recorded so that

the cycle ergometer resembled the riders preferred bike setup as closely as possible. This included the

height, reach and angle of the handle bars, as well as the height, fore and aft position, and angle of the

seat. Additionally, participant’s own pedals were attached to the cranks to allow them to ride in their

own shoes. Prior to the first assessment participants were able to request changes to any of the bike’s

dimensions during their warm-up. Once confirmed, this setup was recorded and held constant for all

future testing sessions.

67
To minimise the effect of nutrition on performance, food and fluid intake were kept consistent for pre-

and post-testing. Participants were instructed to submit a food diary of the food and fluid consumed and

at what time during the 24 hours preceding the formal incremental power and 20km-TT pre tests. This

food diary was returned to participants for replication prior to their post-intervention tests. Participants

were instructed not to consume any caffeinated beverages in the 12 hours preceding each test and to

abstain from exhaustive or prolonged exercise for at least 48 hours prior to each test. All participants

claimed to have adhered to these criteria.

3.3.2. Spirometry

Each participant’s vital capacity (VC), forced vital capacity (FVC), forced expiratory volume in one

second (FEV1), peak expiratory flow (PEF), maximum voluntary ventilation (MVV), inspiratory and

expiratory lung volumes (IRV and ERV respectively) and tidal volume (VT) were determined in

accordance with American Thoracic Society guidelines (Miller et al., 2005). For all tests, participants sat

in an upright position and wore a nose-clip to eliminate nose breathing. Participants completed three

different procedures with a small rest (one to two minutes) between each test. A procedure was

stopped and repeated if a participant bent forward at the waist in an attempt to maximise the expiratory

phase, or if there were symptoms suggestive of upper airway occlusion during the maximal flow-volume

loop procedure (such as pronounced inspiratory stridor, cough or expiratory wheeze).

The first procedure required participants to be quietly seated with the mouth piece and nose clip in situ,

breathing normally. When they were ready a maximal inhalation was taken, followed by a controlled but

maximal exhalation. Normal tidal breathing with the mouth piece and nose clip still in situ was

commenced immediately after the maximal exhalation. After three or four tidal breaths the maneuver

68
was repeated. This procedure was repeated three times before the mouth piece was removed. The

highest values for VC, IRV, ERV and VT were recorded from this procedure.

The second procedure (maximal flow-volume loop) required a maximal inhalation to be taken with the

nose clip on but without the mouth piece in situ. The participant then inserted the mouth piece at the

peak of the inhalation, ensuring a firm seal was achieved, and exhaled as hard, fast and completely as

possible. When the participant felt they could not exhale any further (approximately 6s) a forceful, fast

and maximal inspiration was performed. When no further inhalation could be taken the participant

removed the mouth piece and breathed normally until they were ready to repeat the procedure. This

procedure was repeated until three consistently reproducible efforts were achieved (within 0.15L of each

other; Miller et al., 2005)), as determined by computer analysis. The highest values for FVC, FEV1 and

PEF were recorded from this procedure.

The final spirometry measure was the MVV test. In the same testing position used above, participants

were instructed to “breathe as hard, fast and deeply (in and out) as possible” for 12 seconds.

Participants were instructed to sit quietly after completion of the 12 seconds and recover their breathing.

They were advised prior to the test that they may feel dizzy or light headed during the last few seconds

or on completion of this test and that this feeling would soon pass. Lucia et al (1999) in their study with

competitive elite cyclists measured FVC and MVV in a standing position. Although the present study

was also done with cyclists it was decided to use a seated position for two reasons: a) cycling is

performed in a seated (albeit forward leaning) position so it seemed more appropriate to and b) since

the MVV test is a maximal hyperventilatory test it is not uncommon for participants to feel faint or light

headed during this test and it was simply deemed safer to have participants in a seated rather than

standing position.

69
 3.3.3. Incremental Power Step Test

Each participant’s peak oxygen uptake ( V&O 2 peak), lactate threshold (LT), lactate turn point (LTP) and

maximum aerobic power (Pmax) were determined using an incremental step test to volitional exhaustion.

Prior to this test participants had completed spirometry procedures, as described above. The face mask

was then attached, ensuring a firm seal with no air leaks. The ergometer was calibrated in accordance

with the manufacturer’s instructions and when acceptable variations were achieved (<1%) participants

commenced their 6 minute self selected warm-up. Participants did not have to use the whole 6 minutes

and were able to perform any final stretches before commencing the test. The self selected warm up for

the pre-intervention tests was recorded for each participant, and repeated prior to their post-intervention

tests. Final calibration checks of the gas analysis system were performed during this warm-up period

and recalibration was performed if required.

The incremental test for most participants started at 150W and consisted of a continuous step test

protocol with three minute stages. However, to cater for a range of fitness levels, the starting power

was adjusted to ensure that each participant completed at least seven but not more than nine stages.

The starting power for each participant was the same for the pre- and post-test. Power output was

progressively increased by 30W at the end of each three minute stage until exhaustion. The test was

terminated when the participant stopped completely, or when the participant could no longer hold a

cadence above 60 rev.min-1. Participants were instructed to stay seated throughout the duration of the

test, and to try to maintain a consistent cadence. A designated set cadence was not used for either the

incremental test or the 20km-TT since each cyclist develops power at his own pedaling rate (Zavorsky

et al., 2007) and a fixed cadence may not have resulted in an accurate reflection of a participant’s

maximum aerobic power.

70
Prior to commencement of the test participants were shown the modified (0-10) Borg scale (Kendrick,

Baxi, & Smith, 2000), its use was explained to them and they had time to become familiar with it.

Twenty seconds before the end of each stage participants were asked to rate their perceived leg effort,

followed by their breathing effort. This order was maintained during all testing, to ensure participants

were clear whether they were describing their leg or breathing effort. In addition, a blood lactate sample

was taken from the participant’s right ear lobe in the last 15 seconds of each stage. Heart rate and

respiratory rate were also collected during this time frame.

Consistent feedback was given to each participant at the end of each power stage. Maximal verbal

encouragement was given to each participant when it was obvious they were in the final stage of their

test. There were always two assessors (primary investigator and research assistant) present for each

test. To avoid bias in the feedback, and therefore to avoid the influence of externally motivating factors,

the assistant assessor provided the maximal encouragement in the last stage and was blinded as to

whether the participant had been in the control or intervention group.

3.3.4. Nijmegen Questionnaire

The Nijmegen questionnaire was completed prior to the pre-intervention 20km-TT. As no anxiety based

BPDs were identified using this tool in either group, the questionnaire was not repeated post-

intervention. The primary investigator read out each of the 16 signs and symptoms listed in Appendix 1

and participant was asked to rate them in accordance with the frequency chart provided (Appendix 1),

over the previous six month period. Participants were only told that it was a breathing related

questionnaire, not one that was specific to anxiety.

71
 3.3.5. 20km Time Trial

Laboratory based time trials (TT) of varying distances have previously been shown to be both reliable

and valid measures of cycling performance (Laursen, Shing, & Jenkins, 2003). More recently, Zavorsky

et al (2007) has demonstrated that a familiarisation session improves the reliability of the formal pre-

and post-testing measurements. A familiarisation TT followed by a second TT has a larger CV than that

observed between the second and a third TT. Given that the smallest worthwhile meaningful change

can be as little as 0.5-1.5% (Paton & Hopkins, 2001) it was considered important to include a

familiarisation trial so as not to confuse typical error with performance improvement. Therefore, over the

course of the study participants complete three 20km-TT’s – a familiarisation trial, and both a pre- and

post-intervention trial.

Participants presented to the laboratory at least 48 hours after their incremental power test, having

followed the same guidelines as the incremental test with respect to exhaustive exercise, food and

caffeinated beverages. The TT protocol allowed participants to change gears like they would on a road

bike. Prior to the familiarisation trial participants had the opportunity to become familiar with the gear

changing and to establish what gears they would like to use.

After calibration of the ergometer and a six minute self-selected warm-up (consistent for pre- and post-

testing), participants were asked to bring the ergometer to a stop, so a “standing start” (seated, with no

movement of the cranks) could be achieved. The ergometer had been programmed to run a 20km-TT

protocol, such that when the program was started (five second count down for participants) the time

clock ran until automatically stopping once the 20kms had been reached. Participants were able to

select their own gearing and cadence and could alter either of these as needed throughout the test.

Participants remained seated throughout the entire TT, with their hands on either the brake hoods or the

top bar. These riding positions were chosen, as opposed to the aerodynamic position, to enable a

72
better comparison to the breathing patterns observed by Lucia et al. (1999) who also used these testing

positions. Water was available as required.

At five minute intervals HR, power, speed, leg and lung RPE were recorded. Standardised verbal

encouragement was given to participants at the 5, 10, 15, 19 and 19.5 km mark (for example “that’s 5km,

going well”). Standardised maximal verbal encouragement was given in the last 500m of the test.

3.3.6. Intervention Sessions - Randomisation

In an attempt to have two even groups based on performance measures, participants were match

paired based on their pre-intervention 20km-TT time. Randomisation to either the intervention or control

group was done by the toss of a coin, such that the individual in the pair whose christian name was

alphabetically first was a “head”, the other “tails”. Whoever corresponded with the side of the coin that

landed face up was put into the intervention group.

3.3.7. Intervention Sessions - Intervention Group

During session five, participants in the intervention group were told the true purpose of the study (i.e.

that the investigators were interested in whether breathing pattern could be retrained in cyclists) and

given a concise but comprehensive overview of efficient versus inefficient breathing, the consequences

of each, the basic mechanics of breathing and how breathing can be retrained. In an attempt to

minimise the placebo effect on the outcome of the study participants were not told that the effect on

performance of changing respiratory pattern was of primary interest to the investigators. Participants

were taught the two breathing exercises (static exercises) (Appendix 3) to be completed at home. The

ways in which the static breathing exercises differed from the ideal breathing pattern for on the bike

were discussed and demonstrated (dynamic exercise) but the participant did not perform the dynamic

exercises during this session. Participants had the opportunity to have any questions answered and

73
were given the opportunity to withdraw from the study if they did not wish to continue with the new study

protocol. Participants were provided with a CD with spoken instructions and a counted rhythm for the

10 minute static breathing exercise, 10 gold star stickers to be used as visual triggers, and written hand

outs of the exercises (Appendix 3), potential symptoms they may experience during BPR (Appendix 6)

and main points that were discussed during this session (Appendix 7).

During session six, participants demonstrated their 10 minute static breathing exercise technique and

feedback was provided with appropriate changes made. Following this, participants completed a 10

minute sub-maximal ride (60% of Pmax) during which they were taught to apply the diaphragmatic

pattern with emphasis on the expiratory phase (dynamic breathing exercise). Sessions seven and eight

each involved a 10 minute sub maximal (60% of Pmax) ride with two 30 second maximal efforts. During

each maximal effort the emphasis was on breathing control and on recovering one’s respiratory rate

when the effort was over.

Participants were encouraged to continue using their dynamic breathing skills during training in addition

to their static breathing exercises.

3.3.8. Intervention Sessions - Control Group

Participants who were randomised into the control group were led to believe the purpose of the study

was twofold:

1) to investigate the effect that breathing pattern had on performance in cyclists

2) to test the reliability and validity of a newly developed breathing harness.

Over the four weeks of the intervention period participants completed one 10 minute sub maximal (60%

of Pmax) ride each week wearing the breathing harness. Participants were asked to rate their leg and

breathing RPE at five and ten minutes.

74
When participants wore the breathing harness for the first time they were shown how it worked and

were able to view the computer screen to observe the different amplitudes of the apical and

diaphragmatic breaths in real time. However, during the sub-maximal rides the screen was turned away

from them. Participants were told that this was to prevent them from pacing or consciously altering their

breathing. No formal data was collected from the breathing harness, or from the participants, during this

intervention period. No feedback on adopted breathing pattern was given to the control group.

3.4. Data Analysis

3.4.1. Pmax determination

As the stages of the peak power incremental test were three minutes in duration, final maximum power

(Pmax) was measured as the participants’ last fully completed three minute stage, plus 0.16W for each

additional second completed of the next power stage.

3.4.2. Lactate

The LT was defined as the first deflection point of 1 mmol.L-1 or greater above baseline (Yoshida, Chida,

Ichioka, & Suda, 1987) in the Lactate (B[La]) vs. power output curve. LTP was defined as the point in

the B[La] vs. power output curve before the observation of a second sudden and sustained increase in

B[La] that coincided with a B[La] of between approximately 2-5 mmol.L-1 (Davis, Bassett, Hughes, &

Gass, 1983). All LT and LTP analyses were performed by two independent examiners, and where

discrepancies occurred, a third examiner was used.

3.4.3. Sub-maximal and maximal VO2

V&O 2 max represents the maximum rate of oxygen that can be inhaled and delivered to and used by the

working muscles during maximal exercise (Astrand & Rodahl, 1986; Saltin & Astrand, 1967). The

V&O 2 max is often identified by a plateau in V&O 2 despite continued increases in exercise intensity.

75
However, a plateau in V&O 2 is not always observed (Doherty, Nobbs, & Noakes, 2003; Lucia et al.,

2006) and along with the finding that different protocols (e.g. discontinuous step versus continuous

ramp) and different exercise modes (e.g. cycling versus running) may yield different maximal results,

(Lucia et al., 2006) it has been proposed that it may be more accurate to term this the peak, rather than

the maximum, V&O 2 (Day, Rossiter, Coats, Skasick, & Whipp, 2003); i.e. the V&O 2 peak. For the purpose of this study,

V&O 2 peak was defined as the peak value achieved during the incremental assessment, averaged over a

30 second period. To quantify the O2 cost of cycling (cycling economy) the mean V&O 2 in the last 30

seconds of each three minute stage was determined.

All V&O 2 and V& C O2 data was recorded breath-by-breath, and to reduce the effect of breath-by-breath

variability, outliers were removed using in-built functions in the Cortex software, thereby reducing the

‘noise’. For determination of V&O 2 peak data during the incremental assessment was averaged on a 30

second basis. Each participant’s absolute V&O 2 peak was recorded as the highest 30-second V&O 2 value

obtained during the test.

3.4.4. Nijmegen Questionnaire

The Nijmegen Questionnaire is scored based on the frequency that a participant experiences the

described signs and symptoms (Appendix 1). The maximum score is 64, and a score greater than or

equal to 23/64 is considered indicative of an anxiety based breathing pattern disorder.

76
 3.4.5. Statistics

Once individual values had been obtained they were entered into an MS Excel statistics spreadsheet

designed by Hopkins (2003) and values were log-transformed. To make inferences about true

(population) values of the effect of BPR on cycling performance, the uncertainty in the effect was

expressed as 90% confidence limits and as likelihoods that the true value of the effect represents

substantial change (Batterham & Hopkins, 2005). An effect was deemed unclear if its confidence

interval overlapped the thresholds for substantiveness; that is, if the effect could be substantially positive

and negative. An estimate of the smallest substantial change in power output was required to make

these inferences. Paton and Hopkins (2001) estimated smallest effects of 0.5-1.5% in mean power,

based on variability in competitive performance of elite cyclists in various time trials where drafting and

group tactics did not contribute. A smallest worthwhile effect of 1.0% was assumed. Pearson

correlations were performed to identify the strength of relationship between changes in RPE, fb, Pmax,

20km-TT speedAVE and 20km-TT time.

77
Chapter Four: Results

4.1. Participant Characteristics

The baseline performance and physiological characteristics of the 22 participants are presented in Table

7.

Table 7. Base Line Performance and Physiological Characteristics for

Intervention and Control Groups*

Measure Intervention Control

(N=10) (N=12)

20km-TT time (min:sec) 32:31 ± 1:31 32:36 ± 1:07

20km-TT PAVE (W) 263 ± 5 269 ± 8
20km-TT SpeedAVE (km·hr-1) 37.5 ± 0.8 37.8 ± 0.9
Pmax (W) 331 ± 33 338 ± 31
Pmax (W—kg—-1) 4.2 ± 0.7 4.1 ± 0.6
VO2peak (L—min-1) 4.3 ± 0.6 4.4 ± 0.5
VO2 peak (ml—kg—min-1) 55.3 ± 8.5 53 ± 7.5
HRpeak (b—min-1) 181 ± 10 184 ± 11
B[La]peak (mmol—L) 10.2 ± 1.5 13.2 ± 1.8
Power at LT (W) 193 ± 26 180 ± 33
Power at LT (%Wmax) 60.1 ± 7.4 54.2 ± 6.4
VO2 at LT (L—min-1) 2.9 ± 0.5 2.9 ± 0.4
VO2 at LT (% VO2peak) 68.5 ± 6.4 65.5 ± 5.6
HR at LT (b—min-1) 140 ± 12 135 ± 12
HR at LT (% HRpeak) 77 ± 4 74 ± 5
fb at LT (breaths.min-1) 29 ± 6 30 ± 6
Power at LTP (W) 270 ± 40 257 ± 39
Power at LTP (%Wmax) 81.6 ± 6.5 75.5 ± 6.3
VO2 at LTP (L—min-1) 3.7 ± 0.6 3.7 ± 0.4
VO2 at LTP (% VO2peak) 87.2 ± 5.5 83.6 ± 6.5
HR at LTP (b—min-1) 165 ± 13 158 ± 13
HR at LT (% HRpeak) 91 ± 4 87 ± 3
fb at LTP (breaths—min-1) 35 ± 5 38 ± 8

*All values are mean ± SD

The Nijmegan Questionnaire did not identify any anxiety-based breathing pattern disorders in either

group, with the mean scores for both the intervention group (6.4 ± 5.06) and the control group (8.55 ±

3.75) being well below a score of 23 necessary to be indicative of an anxiety-based breathing pattern

disorder.

78
Attendance to each intervention session was high for both the intervention (100%) and control (98.3%)

groups. Compliance with the BPR was 89%, and all participants were able to accurately and

appropriately perform their static and dynamic exercises, indicating they had practiced regularly.

4.2. Effects on cycling performance (Incremental power and 20km-TT

performance)

Table 8 shows the effect of BPR on cycling performance. After 4 weeks of BPR there were substantial

changes to Pmax (3.2 ± 3%) in the incremental step test and to Pave (3.1 ± 2.9%), average speed (1.3 ±

1.2%) and time (-1.5 ± 1.1%) in the 20km-TT.

Table 8. Mean Changes in Performance Measures post Breathing Pattern Retraining and Control and
Chances That the True Difference in the Changes Is Substantial

Change in Measure*
Intervention Control Difference Cohen’s ES Practical Inference
(N=10) (N=12) ± 90% CL ± 90% CL† Descriptive
Incremental
Pmax 1.8 ± 4.5 -1.4 ± 3.5 3.2 ± 3.0 0.32 ± 0.3 Very likely +ve
20kmTT
PAve 3.1 ± 4.5 0.0 ± 2.9 3.1 ± 2.9 0.3 ± 0.27 Very likely +ve
SpeedAve 1.3 ± 2.0 0.0 ± 1.1 1.3 ± 1.2 0.34 ± 0.32 Very likely +ve
Time -1.5 ± 1.8 0.0 ± 1.0 -1.5 ± 1.1 -0.38 ± 0.28 Very likely +ve

*Units of change are percentage means ± SD for all measures.

†± 90% CL: add and subtract this number to the difference to obtain the 90% confidence limits for the true difference.
‡Based on a smallest beneficial or harmful change in performance of 1%.

4.3. Effect on Physiological Measures

A comparison of the physiological measures determined during the incremental test to exhaustion in the

two groups revealed no clear differences (Table 9). Although the effect of BPR on lactate and heart rate

at LT and LTP was unclear, there was a trend for both of these variables to be decreased in the

intervention group at LTP (-4.6 ± 16.6%; -2.3 ± 2.5% respectively). All other measures were unclear.

79
Table 9. Mean changes in Physiological Measures post Breathing Pattern Retraining and
Control and Chances That the True Difference in the Changes is Substantial

Change In Measure*
Intervention Control Difference Cohen’s ES Practical Inference
(N=10) (N=12) mean ± SD ± 90% CL† Descriptive‡
Power at LT -2.7 ± 6.0 -0.5 ± 13.4 -2.2 ± 7.3 -0.13 ± 0.41 unclear
Power at LTP 1.7 ± 6.6 -1.4 ± 10.0 3.1 ± 6.1 0.21 ± 0.40 unclear
HRmax 0.1 ± 2.2 -0.1 ± 2.0 0.0 ± 1.6 -0.01 ± 2.80 unclear
B[La] at LT 13.8 ± 21.7 10.8 ± 22.8 2.7 ± 16.0 0.08 ± 0.46 unclear
B[La] at LTP -0.9 ± 22.0 3.9 ± 24.2 -4.6 ± 16.6 -0.17 ± 0.54 unclear
HR at LT 0.9 ± 3.7 2.5 ± 3.9 -1.5 ± 2.8 -0.19 ± 0.34 unclear
HR at LTP -0.7 ± 3.0 1.7 ± 3.5 -2.3 ± 2.5 -0.3 ± 0.31 unclear
HRave 2.5 ± 2.3 0.2 ± 4.9 2.3 ± 2.9 0.32 ± 0.40 unclear
VO2 at LT (L—min-1) -0.5 ± 7.1 0.0 ± 6.7 -0.4 ± 5.1 -0.03 ± 0.35 unclear
VO2 at LTP (L—min-1) -0.7 ± 6.2 0.3 ± 7.6 1.0 ± 5.1 0.08 ± 0.38 unclear
VCO2 at LT (L—min-1) 0.7 ± 6.9 -1.5 ± 7.2 2.3 ± 5.2 0.14 ± 0.31 unclear
VCO2 at LTP (L—min-1) 0.7 ± 7.0 0.7 ± 6.4 1.4 ± 4.9 0.09 ± 0.31 unclear
VO2 at LT (% VO2peak) -1.0 ± 5.9 4.3 ± 11.1 -5.1 ± 6.3 0.58 ± 0.68 unclear
VO2 at LTP (% VO2peak) -0.2 ± 5.0 0.9 ± 6.9 -1.2 ± 4.3 -0.15 ± 0.57 unclear
VO2max (L—min-1) 0.5 ± 6.9 -0.4 ± 6.4 1.0 ± 4.9 0.09 ± 0.45 unclear
VO2max (ml—kg—min-1) 0.7 ± 6.6 -0.5 ± 6.2 1.2 ± 4.7 0.08 ± 0.31 unclear

*Units of change are percentage means ± SD for all measures.

†± 90% CL: add and subtract this number to the difference to obtain the 90% confidence limits for the true difference.
‡Based on a smallest beneficial or harmful change in performance of 1%.

4.4. Effect on Respiratory Measures

The effect of BPR on static and dynamic lung function tests was unclear for all measures, except for

FVC where the effect was deemed unlikely to be beneficial. Whilst there was a trend of both IRV (18 ±

36.4%) and ERV (9.8 ± 44.3%) to increase, and for FEV1 and MVV to decrease, the 90% CI associated

with these measures was large.

The effect of BPR on tI (10.1 ± 8%; 9.4 ± 7.7%) and VT (11.6 ± 14%; 12.6 ± 11.2%) at both LT and LTP

respectively was substantial. There was a strong trend for tE to be increased at both LT and LTP (5.0 ±

9%; 5.9 ± 7.1%). There was a clear trend for VE at both LT (-5.6 ± 4%) and LTP (-5.2 ± 4%) to be lower

post-BPR. Similarly, breathing frequency (fb) was also substantially lower both at LT (-13.2 ± 8.9%) and

LTP (-9.5 ± 8.4%). The effect on respiratory drive (VT/tI) and respiratory timing (tI/tTOT) at LT and LTP

was unclear. Table 10 shows the mean changes in respiratory measures for the BPR relative to the

control condition and the statistics for the differences in the changes.

80
Table 10. Mean Changes in Respiratory Measures (non-exercise and exercise) post Breathing
Pattern Retraining and Control and Chances That the True Difference in the Changes Is
Substantial

Change in Measure*
Intervention Control Difference Cohen’s ES Practical Inference
(N=10) (N=12) ± 90% CL ± 90% CL Descriptive‡
Lung Function Tests
IRV 6.3 ± 35.3 -10 ± 62.7 18 ± 36.4 0.33 ± 0.61 unclear
ERV 8.7 ± 79.7 -1.1 ± 37.7 9.8 ± 44.3 0.2 ± 0.78 unclear
FVC -2.8 ± 5.9 2.7 ± 8.1 -5.4 ± 5.1 0.44 ± 0.39 probably -ve
FEV1 -1.7 ± 8.1 0.9 ± 6.5 -2.7 ± 5.4 -0.19 ± 0.37 unclear
PEF -0.6 ± 5.9 -2.5 ± 16.3 1.98 ± 8.7 0.09 ± 0.39 unclear
MVV -4.9 ± 16.7 5.5 ± 30.6 -9.9 ± 17.1 -0.71 ± 1.07 unclear

Incremental Test
tE at LT (sec-1) 12.1 ± 13.6 6.8 ± 10.3 5.0 ± 9 0.25 ± 0.43 unclear
tE at LTP (sec-1) 10 ± 10.7 3.9 ± 8.2 5.9 ± 7.1 0.29 ± 0.35 unclear
tI at LT (sec-1) 10.8 ± 13.4 0.7 ± 6.3 10.1 ± 8 0.42 ± 0.34 Very likely +ve
tI at LTP (sec-1) 11.8 ± 12.4 2.1 ± 7.2 9.4 ± 7.7 0.43 ± 0.35 Very likely +ve
VT at LT (L) 11.6 ± 14 0.9 ± 7.2 10.6 ± 8.5 0.53 ± 0.42 Very likely +ve
VT at LTP (L) 12.6 ± 11.2 2.9 ± 9.4 9.4 ± 7.6 0.47 ± 0.39 Very likely +ve
VE at LT (L·min-1) -5-5 ± 5.7 -0.1 ± 5 -5.6 ± 4 -0.33 ± 0.22 Very likely +ve
VE at LTP (L·min-1) -4.1 ± 4 1.2 ± 6.7 -5.2 ± 4 -0.32 ± 0.23 Very likely +ve
VEmax (L·min-1) -2.2 ± 11 -2.2 ± 11.5 0.0 ± 8.2 0.0 ± 0.52 unclear
VT/tI at LT (L·sec-1) 0.7 ± 16.8 0.2 ± 3.5 0.5 ± 9.6 0.03 ± 0.46 unclear
VT/tI at LTP (L·sec-1) -2.9 ± 9.8 0.8 ± 9.4 -3.6 ± 7 -0.2 ± 0.37 unclear
fb at LT (1·min-1) -14.5 ± 14.4 -1.6 ± 8 -13.2 ± 8.9 -0.71 ± 0.43 Very likely +ve
fb at LTP (1·min-1) -11.3 ± 13.5 -2.0 ± 7.9 -9.5 ± 8.4 -0.51 ± 0.41 Very likely +ve
tI/tTOT at LT -1.2 ± 4.1 -2.1 ± 4.1 0.9 ± 3.8 0.14 ± 0.57 unclear
tI/tTOT at LTP -0.5 ± 5 -0.8 ± 3.2 0.3 ± 3.2 0.05 ± 3.2 unclear

*Units of change are percentage means ± SD for all measures.

†± 90% CL: add and subtract this number to the difference to obtain the 90% confidence limits for the true difference.
‡Based on a smallest beneficial or harmful change in performance of 1%.

4.4.1. Comparison of measures at absolute workloads pre and post-BPR

Several measures were compared over a range of absolute workloads pre- and post-intervention

(Figures 6 to 10).

At each workload, fb decreased (Figure 6) and VT increased following BPR (Figure 6). There was a

trend for VE to slightly decrease following BPR (Figure 6).

81
 Breath in g freq u en cy (b reath s•m in 1 )
60.00 70.00

Breath in g frequ ency (breaths•m in 1)

50.00 60.00

50.00
40.00
40.00
30.00
30.00
20.00
20.00 Pre BPR
Pre BPR Post BPR
10.00 10.00
Post BPR
0.00 0.00
90 120 150 180 210 240 270 300 330 360 90 120 150 180 210 240 270 300 330 360

Power (W) Power (W)

3.50 4.00
3.50
3.00 3.00

T id al V o lu m e (L )
T id al V o lu m e (L )

2.50
2.50
2.00
2.00 1.50
Pre BPR Pre BPR
1.00
1.50 Post BPR Post BPR
0.50

1.00 0.00
90 120 150 180 210 240 270 300 330 360 90 120 150 180 210 240 270 300 330 360

Power (W) Power (W)

160 160
M inu te V entilation (L •m in-1)
M in ute V entilation (L•m in -1)

140 140

120 120

100 100

80 80
Pre BPR
60 60
Pre BPR Post BPR
40 Post BPR 40

20 20
90 120 150 180 210 240 270 300 330 360 90 120 150 180 210 240 270 300 330 360

Power (W) Power (W)

Figure 6. Effect of breathing pattern retraining on breathing frequency, tidal volume and minute ventilation in the
intervention (left panel) and control groups (right panel).

As anticipated, the tI and tE decreased with the progressive increase in workload. After BPR, both tI

(13%) and tE (12%) were longer (Figure 7).

82
 1.80 1.80
Pre BPR 1.60
1.60 Pre BPR
Post BPR
1.40 Post BPR
In sp irato ry T im e (s)

In sp irato ry T im e (s)
1.40
1.20
1.20 1.00
1.00 0.80
0.60
0.80
0.40
0.60 0.20
0.40 0.00
90 120 150 180 210 240 270 300 330 360 90 120 150 180 210 240 270 300 330 360
Power (W) Power (W)

2.00 2.00
1.80 1.80
Pre BPR Pre BPR
1.60
E xp irato ry T im e (s)
E xp irato ry T im e (s)

1.60 Post BPR

Post BPR 1.40
1.40 1.20
1.20 1.00
1.00 0.80
0.60
0.80
0.40
0.60 0.20
0.40 0.00
90 120 150 180 210 240 270 300 330 360 90 120 150 180 210 240 270 300 330 360
Power (W) Power (W)

Figure 7. Effect of Breathing pattern retraining on inspiratory and expiratory time in the intervention (left panel)
and control (right panel) groups.

Inspiratory drive (VT/tI) and inspiratory timing (tI/tTOT) did not change substantially throughout the

incremental test as a result of BPR (Figure 8).

83
 6.0 6.0

5.0 5.0
V T /tI (L •m in -1)

V T /tI (L •m in -1)
4.0 4.0

3.0 3.0

2.0 2.0
Pre BPR Pre BPR
1.0 Post BPR 1.0 Post BPR

0.0 0.0
90 120 150 180 210 240 270 300 330 360 90 120 150 180 210 240 270 300 330 360
Power (W) Power (W)

0.55
0.54
Pre BPR
0.52 Post BPR
0.50
0.50
T i/T to t

T i/T to t
0.48
0.45
0.46

0.40 Pre BPR 0.44

Post BPR 0.42

0.35 0.40
90 120 150 180 210 240 270 300 330 360 90 120 150 180 210 240 270 300 330 360
Power (W) Power (W)

Figure 8. Effect of breathing pattern retraining on inspiratory drive (VT/tI) and inspiratory timing (tI/tTOT ) in the
intervention (left panel) and control (right panel) groups.

There was a slight trend for lactate levels to be decreased from 240 to 330W in the intervention group

(Figure 9).

12.00

10.00
Blo o d L actate (m m o l.L )

14.00
8.00
Blo o d L actate (m m o l.l)

Pre BPR 12.00

Pre BPR
6.00 Post BPR 10.00 Post BPR
8.00
4.00
6.00

2.00 4.00
2.00
0.00
0.00
90 120 150 180 210 240 270 300 330 360
90 120 150 180 210 240 270 300 330 360
Power (W)
Power (W)

Figure 9. Effect of breathing pattern retraining on blood lactate in the

intervention (left panel) and control (right panel) groups.

82
End-tidal partial pressures for oxygen (PETO2) and carbon dioxide (PETCO2) (data not presented), V&O 2

and V& C O2 were not affected by BPR at any stage throughout the incremental test (Figure 10).

5.00 5.00
4.50 4.50
4.00 4.00
3.50 3.50
V O 2 (L .m in )

V O 2 (m in .L )
3.00 3.00
Pre BPR
2.50 2.50
Post BPR
2.00 2.00
Pre BPR
1.50 1.50
1.00 Post BPR 1.00
0.50 0.50
0.00 0.00
90 120 150 180 210 240 270 300 330 360 90 120 150 180 210 240 270 300 330 360
Power (W) Power (W)

6.0 6.0

5.0 5.0
V CO 2 (L .m in )

V CO 2 (L .m in )

4.0 4.0

3.0 3.0
Pre BPR
2.0 Pre BPR 2.0
Post BPR
Post BPR
1.0 1.0

0.0 0.0
90 120 150 180 210 240 270 300 330 360 90 120 150 180 210 240 270 300 330 360
Power (W) Power (W)

Figure 10. Effect of breathing pattern retraining on VO2 and VCO2 in the intervention (left panel)
and control (right panel) groups.

4.5. Changes in RPE measures

4.5.1. RPE

Table 11 shows the effect of BPR on mean RPE (legs and the breathing) during both the

incremental test and 20km-TT. The RPE at LT for both the legs (-27.9 ± 38.5%) and breathing (-

30 ± 33.9%) were lower post-intervention. Similarly, leg and breathing RPE were also lowered at

LTP (-24.7 ± 28.2%; -24.7 ± 28.1%). It was unclear whether BPR had a substantial effect on leg

or breathing RPE at any point during the 20km-TT.

84
 4.5.2. Reason for terminating exercise

Prior to BPR, 60% of the intervention group rated their lungs/breathing as the terminating factor in

the incremental test, 30% rated their legs, and 10% both legs and lungs. In the control group,

42% rated their lungs, 42% rated their legs and 10% both legs and lungs as the terminating factor.

Following BPR, only legs (90%) or both legs and lungs (10%) were rated as terminating factors in

the intervention group whilst the control group continued to rate lungs (58%) or legs (42%) as the

terminating factor.

Table 11. Mean Changes in Rate of Perceived Exertion Post Breathing Pattern Retraining
and Control and Chances That the True Difference in the Change is Substantial
Change in Measure*
Intervention Control Difference Cohen’s ES Practical Inference
(N=10) (N=12) ± 90% CL ± 90% CL† Descriptive‡
Incremental Test
LT LegRPE -18.2 ± 64 13.4 ± 39.6 -27.9 ± 38.5 -0.63 ± 0.62 Very likely +ve
LTP LegRPE -9.8 ± 37.1 19.9 ± 43.2 -24.7 ± 28.2 -0.69 ± 0.60 Likely +ve
LT BreathingRPE -18.8 ± 47 16.1 ± 50.2 -30.0 ± 33.90 -0.77 ± 0.63 Likely +ve
LTP BreathingRPE -6.1 ± 37.1 24.8 ± 43 -24.7 ± 28.1 -0.69 ± 0.60 Likely +ve

20km TT
15 min LegRPE 3.8 ± 22.2 1.4 ± 33.5 2.5 ± 19.9 0.08 ± 0.58 unclear
Final LegRPE -0.1 ± 10.8 5.2 ± 31.2 -5.1 ± 16.1 -0.18 ± 0.52 unclear
15 min BreathingRPE -4.0 ± 28.4 6.7 ± 40.0 -10.0 ± 24.1 -0.39 ± 0.80 unclear
Final BreathingRPE -6.0 ± 14.4 1.6 ± 38.0 -7.4 ± 19.6 -0.34 ± 0.78 unclear

*Units of change are percentage means ± SD for all measures.

†± 90% CL: add and subtract this number to the difference to obtain the 90% confidence limits for the true difference.
‡Based on a smallest beneficial or harmful change in performance of 1%.

4.5.3. Correlations between RPEbreathing, RPElegs, fb and performance changes

There were no clear correlations between RPEbreathing, RPElegs, or fb and the performance

changes observed in either the 20km-TT or the incremental step test.

85
Chapter Five: Discussion

The aim of this study was to investigate the effect of breathing pattern retraining (BPR) on cycling

performance, and on respiratory, metabolic and perceptual responses during an incremental test

in competitive cyclists. The main findings were that four weeks of specific BPR clearly enhanced

endurance performance and incremental peak power and positively affected breathing pattern

and perceived exertion.

To the best of the author’s knowledge, this is the first study that has investigated the effect of

BPR on endurance performance in athletes. Consequently, it is not possible to directly discuss

the results from the present study in the context of those that have been previously reported.

However, where appropriate, the findings of the present study have been compared to other

cycling studies, or studies that have attempted to influence or modify ventilation in athletes using

related methods such as IMT (Johnson, Sharpe, & Brown, 2007; Romer, McConnell, & Jones,

2002a) and VIH (Holm, Sattler, & Fregosi, 2004; Sonetti, Wetter, Pegelow, & Dempsey, 2001).

Furthermore, in some instances reference has been made to studies that have acutely imposed a

change to breathing pattern (i.e. entrainment), during a one off test (Fabre, Perrey, Arbez, &

Rouillon, 2007; Maclennan, Silvestri, Ward, & Mahler, 1994) as opposed to specifically retraining

breathing pattern.

The large standard deviations reported in tables 8, 9, 10 and 11 demonstrate a degree of

individual response (intervention and control groups) to the intervention period. Whilst these

responses are noted, the following discussion will focus on group changes and trends.

86
5.1. Effect of BPR on cycling performance

In the present study, cycling performance was measured in two-ways: i) a simulated (laboratory-

based) 20km-TT and ii) an incremental step test to exhaustion. Both tests were performed on the

same electromagnetically-braked cycle ergometer.

5.1.1. Effects on 20km-TT performance

The four week BPR intervention had a clear beneficial effect both on average 20km-TT power

(3.1 ± 2.9%), average speed (1.3 ± 1.2%) and 20km-TT performance time (-1.5 ± 1.1%). These

results meet the criteria for worthwhile effect size (0.5-1.5%) estimated by Paton and Hopkins

(2001) in mean power, based on variability in competitive performance of elite cyclists. In terms

of changes in performance time, the observed performance improvement is somewhat less than

the cycling performance changes reported by Romer et al. (2002) (-3.8 ± 1.7%; 20km-TT) and

Johnson et al. (2007) (-2.66 ± 2.5%; 25km-TT) after six weeks of IMT, or the improvement in

approximate 40km-TT (-4.7 ± 1.6%) after four weeks of VIH (Holm, Sattler, & Fregosi, 2004).

Conversely it is consistent with the findings of Sonnetti et al. (2001) who reported a 1.5%

improvement in 8km-TT after four weeks of combined VIH and IPTL (although this latter study did

not show a significant difference between the control and intervention groups). Similarly, but in

rowing, the magnitude of performance change has been greater for IMT than the present study (-

3.1 ± 0.8%, 5000m-TT after 11 weeks of IMT; Volianitis et al., 2001).

Whilst the substantial enhancement in cycling performance after BPR is encouraging, the smaller

change compared to those reported should be considered. Differences could be explained by

several methodological reasons including: i) the differences in intervention period and number of

intervention sessions; ii) the time of competitive season that the intervention took place; and iii)

physiological reasons including initial (pre-intervention) aerobic fitness. In terms of duration, the

87
present study involved a four week (28 sessions) intervention period. In contrast, the cycling

studies that reported larger performance changes have typically used six week interventions both

involving 84 sessions (Johnson, Sharpe, & Brown, 2007; Romer, McConnell, & Jones, 2002a).

Holm et al. (1994) used a four week intervention which involved 40 sessions and recorded a

larger change than the present study. The cycling study that reported a similar performance

change (Sonetti, Wetter, Pegelow, & Dempsey, 2001) also used a four week intervention period

with 20 sessions. The greater number of sessions could benefit athletes in two ways; first by

enhancing the benefits seen during exercise (acute response to intervention), and second, by

allowing athletes to train at a higher intensity, further enhancing performance (long term benefit).

In support of the latter, Dickinson, Whyte and McConnell (2007) recently commented that

improvements to breathing after IMT may enable athletes to train at higher intensities, with

subsequent performance gains.

Adopting a new automatic breathing pattern is not achieved immediately upon commencement of

BPR (Grossman, DeSwart, & Defares, 1985). Indeed, a period during which respiratory exertion

is perceived as being greater is often experienced during the adaptation phase (Ley & Yelich,

1998). Interestingly, participants in the BPR group of the present study reported it took

approximately three weeks (21 sessions) until they felt they were using their new breathing

pattern comfortably and automatically during training. It is possible therefore that the

performance improvements seen in the present study solely reflected a response to the

participants’ breathing pattern during exercise, but was probably not long enough for participants

to benefit in terms of increasing their training effort. Thus additional weeks and/or sessions of

BPR may have allowed participants to train at a higher intensity with their newly adopted

breathing pattern, potentially resulting in physiological adaptation and further performance

enhancement.

88
The time of the competitive season that the studies were performed may also have contributed to

the variation observed in performance. As highlighted by Paton and Hopkins (2004), most

intervention research involving competitive athletes is conducted in the non-competitive part of

the season and that smaller performance gains are to be made by cyclists during their

competition season due to the high level of intensity that they are training at. During this time of

the season, the magnitude of change of some interventions may be inflated since athletes are

further away from their peak condition. The present study was purposely undertaken during the

competition season so that the true benefit of BPR could be determined. In comparison, Romer

et al. (2002) conducted their study during the maintenance phase. In other studies, it is not

stated what part of the season studies were performed (Holm, Sattler, & Fregosi, 2004; Johnson,

Sharpe, & Brown, 2007).

It is unlikely that the difference (approximately 1-2%) in TT performance between the present

study and those previously reported are due to the experience or ability of the participants. It

would be reasonable to expect that higher level athletes may make smaller performance gains

due to the smaller margins for improvement they possess (Hamlin & Hellemans, 2006). This is

unlikely to be the case in this instance as the athletes recruited in the present study were of

similar fitness level ( V&O 2 peak 4.34 ± 0.47L—min-1) to those recruited by Romer et al. (4.58 ± 0.17

L•min-1, 2002a) and should theoretically have had the same potential to improve their aerobic

performance. Further, despite recruiting cyclists with fitness levels lower than the present study

Johnson et al. (3.65 ± 0.86 L•min-1, 2007) and Holm et al. (3.78 L—min-1, 2004) demonstrated

larger performance gains than the present study, again suggesting that pre-intervention fitness

level was not a contributing factor to the variation in performance change observed between the

present study and those reported above.

89
Finally, it is probable that the difference between the results of the present study and those

previously reported may be due to different aspects of respiratory muscle function that IMT and

BPR attempt to have an effect on. For example, IMT primarily aims to increase both the strength

and endurance of the inspiratory muscles in an attempt to reduce inspiratory muscle fatigue (as

measured by changes to MIP) (Romer, McConnell, & Jones, 2002b; Volianitis et al., 2001).

Conversely, BPR does not specifically train the strength and endurance of the inspiratory muscles

but instead aims to optimise the mechanics of breathing thus decreasing the work of breathing

during exercise. Therefore BPR may affect TT performance by primarily altering the recruitment

pattern of the accessory muscles and enhancing the mechanical advantage of the respiratory

muscles (the benefits of which will be discussed later), with changes in B[La] being secondary. It

is possible that if IMT was combined with BPR such that an optimal breathing pattern (automatic)

was established prior to commencing and throughout the duration of IMT, there could be an

accumulative effect on endurance performance.

5.1.2. Effects on Incremental Peak Power (Pmax)

In addition to enhanced 20km-TT performance, a substantial improvement in Pmax (3.2 ± 3%,

Table 8) was also observed following BPR. In contrast, neither IMT nor VIH have been shown to

induce an ergogenic effect on incremental power in rowing (Volianitis et al., 2001) or cycling

(Holm, Sattler, & Fregosi, 2004; McConnell & Sharpe, 2005; McMahon, Boutellier, Smith, &

Spengler, 2002; Romer, McConnell, & Jones, 2002a) and there are conflicting results that both

refute (Romer, McConnell, & Jones, 2002b) and support (Chatham, Baldwin, Griffiths, Summers,

& Enright, 1999) its effect on incremental running. For example, in a well controlled, double blind

study Romer et al. (2002b) were unable to demonstrate changes to incremental shuttle running

performance after four weeks of IMT. Conversely, Chatham et al. (1999) demonstrated a 3.78%

90
improvement in shuttle running. However this latter study has been criticised because the IMT

group was led to believe that IMT would enhance their performance, which inevitably may have

motivated them to perform better, compared to the control group who had no expectation.

It is possible that, as with Chatham et al. (1999), participants in the BPR group of the present

study had a higher expectation of improved performance and therefore their motivation to do

better was elevated. To address this potential confounding variable from the outset, participants

in the present study were at no stage led to believe that the aim of BPR was to improve

performance. Rather, they were informed that the purpose was solely to investigate whether

breathing pattern could indeed be altered during exercise. The combination of increased

performance, yet decreased perceived exertion (see Effects of BPR on Perceived Effort) supports

the assertion that increased motivation was not the primary influence on performance in the

present study.

5.2. Effects of BPR on Respiratory Measures

An important finding in the present study was that BPR, as anticipated, resulted in a substantial

change in breathing pattern during exercise. Several respiratory measures reflecting breathing

pattern were determined during the incremental step test to exhaustion and for direct comparison

were compared at similar absolute workloads pre- and post-BPR. These included VE, fb, VT, tI, tE,

inspiratory drive (VT / tI) and inspiratory timing (tI/tTOT). Dynamic measures of lung function were

measured at rest.

As anticipated, no changes were observed in most dynamic lung function measures post-BPR

(benefit unlikely, Table 10) – a finding that coheres well with IMT studies (Romer, McConnell, &

91
Jones, 2002a) (McMahon, Boutellier, Smith, & Spengler, 2002). However, respiratory measures

during exercise were altered. BPR resulted in a substantial decrease in fb and VE, combined with

an increase in VT and tI and it is possible that these changes explain the observed performance

improvements.

5.2.1. Breathing frequency (fb) and Tidal Volume (VT)

Most healthy humans increase ventilation at high exercise intensities by initially increasing VT,

then fb. As exercise intensity increases, the continued increase in fb compromises VT such that

VT plateaus or even slightly decreases, resulting in a tachypnoeic shift (Eastwood, Hillman, &

Finucane, 2001; Gallagher, Brown, & Younes, 1987; Scheuermann & Kowalchuk, 1999). This

ventilatory response has been demonstrated in untrained subjects with a wide range of fitness

levels (Folinsbee, Wallace, Bedi, & Horvath, 1983; Gallagher, Brown, & Younes, 1987) as well as

elite (but not professional) cyclists (Lucia, Carvajal, Calderon, Alfonso, & Chicharro, 1999). In the

present study the BPR and control groups demonstrated similar breathing patterns pre-

intervention, with similar fb (51 vs. 56 breaths—min-1 respectively) and VT (2.7 vs. 2.7 L—min-1) being

achieved at termination of exercise (330 vs. 340W) however substantial differences in breathing

pattern were observed post-BPR. Whilst fb increased and VT decreased across workloads in the

incremental test, the BPR group showed a significant decrease in fb and an increase in VT

compared with the breathing pattern they adopted pre-intervention (Figure 6). Conversely, the

breathing pattern of the control group did not change except for a decreased tE post-intervention

(which is not thought to be a desired change). The biggest difference between the two groups

was the change in fb, with the intervention group reaching a maximal fb of 47 ± 7 breaths—min-1 at

exhaustion (335 ± 35W) whilst the control group demonstrated a maximal fb of 60 ± 8

breaths—min-1 (333 ± 30W). These fb values are comparable to those observed by Lucia et al.

(1999), who reported a fb of approximately 51 and 58 breaths—min-1 in professional and elite

92
cyclists respectively, though it should be acknowledged that a greater power output was reached

(>400W) by Lucia et al.’s (1999) subjects (Figure 1). It is likely that the emphasis during BPR of

maintaining a diaphragmatic breathing pattern, with forced exhalation, enabled the BPR group to

keep their fb substantially lower as exercise intensity increased.

5.2.2. Inspiratory and Expiratory Time

In an extensive investigation of the breathing pattern measures in elite and professional cyclists,

Lucia et al. (1999) identified that tI and tE were the two parameters that differentiated elite and

professional cyclists and that such difference may possibly contribute to the superior level of

performance attained by the professionals. Inspiratory time reached a plateau in professionals at

low intensities (300W) and then decreased with increasing exercise intensity, whereas it

continuously decreased in elites from the lowest to the highest intensities (Figure 2). Expiratory

time was significantly higher in professionals than elites at most moderate-to-high power outputs

(Figure 2). Prior to the intervention of the present study, both BPR and control groups displayed

a decrease in tI from low to high intensities (Figure 7) consistent with the elite cyclists of Lucia et

al. (1999). After intervention the control group, as expected, retained the same breathing pattern,

however the BPR group’s breathing pattern changed. Specifically, the tI initially increased but

then plateaued between 150 and 240W (early to mid workloads), before gradually decreasing

(Figure 7) in a pattern similar to professional cyclists. It is likely that the BPR allowed subjects to

maintain control of the length of tI and to ultimately control their breathing pattern during

incremental exercise.

Although the difference in tE pre- versus post-intervention was unclear, there was a trend for tE to

be lengthened in the intervention group post-BPR at both LT and LTP (5.0 ± 9; 5.9 ± 7.1%

respectively; Table 10). An increased tE may also have been an important factor in preventing a

93
tachypnoeic breathing pattern developing, as evident by the increased VT and decreased fb

discussed above. Lucia et al. (1999) suggested that the ability to maintain a breathing pattern

during maximal intensity exercise void of a tachypnoeic shift is an important differential between

professional and amateur elite cyclists, with the prolongation of tE being an important adaptation

made by professional cyclists during high exercise intensities. Most healthy humans increase fb

by decreasing tE rather than tI (Folinsbee, Wallace, Bedi, & Horvath, 1983) – a breathing pattern

that reverses the normal inspiratory to expiratory ratio, and may lead to breath-stacking and

dynamic hyperinflation (Chaitow & Bradley, 2002), the consequence of which may ultimately

decrease exercise performance. Prolonging tE in the BPR group may have aided in preventing a

tachypnoeic breathing pattern from developing and subsequently contributed to enhanced

exercise performance.

The change in tI and tE observed in the present study is exciting as it suggests that it is possible

to change an athletes breathing pattern, and for them to maintain this during high intensity

exercise, in a way that appears to be beneficial to endurance performance. Previously it had

been suggested that this difference was an acquired physiological response that perhaps

developed over years of training (Lucia, Carvajal, Calderon, Alfonso, & Chicharro, 1999), rather

than something that could be learned in a relatively short period of time. It is believed that no

other study has reported changes in tI or tE after a respiratory related intervention in athletes.

5.2.3. Minute Ventilation

The largest physiological change observed in the present study after BPR was the decrease in VE

at both LT and LTP (Table 10). Although VEmax was unchanged in the present study (Benefit

likely, Table 10), there was a trend for VE to be decreased at workloads above LTP (240 – 300W;

Figure 6). This finding is consistent with that of Lucia et al. (1999) who showed that VEmax did not

94
differ between elite and professional cyclists (Figure 1), although it remained lower in professional

than elite cyclists at higher workloads (300-400W). The change in fb and VT discussed above,

coupled with an unchanged VEmax suggest that cyclists in the present study were able to achieve

the same VE in a more ‘controlled’ way, which prevented the occurrence of a tachypnoeic shift.

Lack of tachypnoeic shift is one of the major differences in breathing pattern highlighted by Lucia

et al. (1999) and is thought to be beneficial to exercise performance by decreasing the work of

breathing (Lucia, Carvajal, Calderon, Alfonso, & Chicharro, 1999), reducing respiratory muscle

fatigue (Johnson, Aaron, Babcock, & Dempsey, 1996) and at high workloads reducing blood

stealing (Sheel et al., 2001).

Considering VE responds more closely to demands for CO2 clearance than for O2 uptake (Ward,

2007) it would be reasonable to expect that a change in B[La] could serve to decrease VE.

However, whilst significant changes were observed in VE at LT and LTP, no clear change in B[La]

at these points was observed, although there was a trend for both VE and B[La] to be decreased

in the higher workloads (240-300W; Figures 6 and 9). This would suggest that the changes in VE

were not solely in response to metabolic changes but that they were likely due to the change in fb

and VT as a result of BPR.

The respiratory compensation point (RCP) marks the onset of hyperventilation during incremental

exercise and is thought to be mediated by metabolic acidosis (Meyer, Faude, Urhausen, &

Kindermann, 2004). It could be speculated that one of the mechanisms underlying the ergogenic

effect of BPR is the ability to control ones breathing pattern at the RCP such that the work of

breathing is decreased, rather than due to any real metabolic changes. This suggestion is

supported by the results of the present study demonstrating a decrease in fb and VE, and increase

in VT, post-BPR at LTP (275 ± 26W; Figure 6). In contrast, other breathing interventions such as

95
IMT do not appear to consistently influence VE, (Griffiths & McConnell, 2007; Romer, McConnell,

& Jones, 2002a; Spengler, Roos, Laube, & Boutellier, 1999; Steussi, Spengler, Knopfli-Lenzin,

Markov, & Boutellier, 2001; Volianitis et al., 2001; Williams, Wongsathikun, Boon, & Acevedo,

2002) despite an abatement in B[La] being a consistent finding (Griffiths & McConnell, 2007;

McConnell & Sharpe, 2005; Romer, McConnell, & Jones, 2002b; Volianitis et al., 2001). This is

surprising as theoretically decreased B[La] would result in decreased metabolites and as such a

decreased demand to clear CO2 (Ward, 2007).

Maintaining a low VE may be beneficial for at least two reasons, the first by potentially preventing

expiratory flow limitation. Expiratory flow limitation occurs when the breathing reserve (MVV - VE)

becomes small, or even negative, such that despite further increases in ventilatory drive in

response to increasing exercise, ventilation can not increase further unless lung operating

volumes increase. The ensuing CO2 retention and arterial hypoxaemia predispose to decreased

exercise tolerance (Ward, 2007). Although increasing lung operating volume in the presence of

expiratory flow limitation can enable the necessary increased ventilatory demands to be met, the

resulting increase in end-expiratory lung volume may result in dynamic hyperinflation which

initiates a cascade of events (i.e. respiratory muscle fatigue, increased work of breathing, blood

stealing; see section 2.3.6, pp33) that ultimately reduces exercise tolerance. In addition,

decreased VE may directly decrease the metabolic and therefore perfusion cost of the respiratory

muscles which may also positively benefit exercise performance through attenuating the

peripheral vasoconstriction metaboreflex.

The clear decrease in VE that occurred during incremental exercise may reflect changes to the

dead space-VT ratio (see section 2.2.3, pp13). When a tachypnoeic shift occurs, the dead space-

VT ratio rises, and alveolar ventilation may be reduced, resulting in increased PCO2 (Hanson,

96
Claremont, Dempsey, & Reddan, 1982). To preserve alveolar ventilation VE may be increased by

10-30% (Hanson, Claremont, Dempsey, & Reddan, 1982). It has been demonstrated however,

that such an increase in VE is not required when VT is maintained during high exercise intensities

(Hanson, Claremont, Dempsey, & Reddan, 1982). It is possible that the intervention group did

not need to increase their VE to the extent they did pre-BPR as they were able to maintain a more

effective VT post-BPR.

The changes after BPR described above may have contributed to the improved 20km-TT

performance and incremental power observed in the present study by attenuating the

metaboreflex associated with increased work of breathing (Sheel et al., 2001), although no

correlation was identified between the breathing pattern changes and exercise performance.

Harms et al. (1997) demonstrated that increasing the work of breathing decreased blood flow to

the legs in cyclists. Such decreased leg blood flow has been shown to decrease exercise

tolerance and power output in cyclists (Harms, Wetter, St Croix, Pegelow, & Dempsey, 2000). A

tachypnoeic breathing pattern is associated with increased dynamic hyperinflation which alters

the mechanical contractile properties of the inspiratory muscles, increases the work of breathing

and causes greater respiratory muscle fatigue (Scheuermann & Kowalchuk, 1999). Therefore, it

is possible that BPR in the present study prevented the onset of a tachypnoeic breathing pattern

and dynamic hyperinflation (in comparison to pre-intervention breathing pattern) thus containing

the work of breathing and preserving leg muscle blood flow ultimately allowing subjects to

increase both their 20km-TT and incremental step test performance. Direct measures of

respiratory work and blood flow distribution during intense pre- and post-BPR would support (or

refute) this suggestion.

97
It is thought that the sympathetically mediated reflex vasoconstriction in the limbs may be induced

by reflex activity from chemo-sensitive type III / IV receptors of fatigueing respiratory muscle

(Sheel et al., 2001). It is speculated that respiratory muscle fatigue may be caused by the

respiratory muscles working in a mechanically disadvantaged position (i.e. under conditions of

high elastic loads and velocity of muscle shortening and working at large percentages of their

available capacity for muscle pressure generation (Johnson, Babcock, Suman, & Dempsey,

1993)), as occurs with dynamic hyperinflation, and that preventing or delaying dynamic

hyperinflation may decrease respiratory muscle fatigue. In cyclists this would be an important

change given the possibility of increased inspiratory muscle fatigue caused by the crouched

cycling position (Boussana et al., 2007; Johnson, Babcock, Suman, & Dempsey, 1993).

Unfortunately, respiratory muscle fatigue was not measured in the present study and therefore

the ability of BPR to attenuate respiratory muscle fatigue is not known.

The exciting importance of the breathing pattern changes observed in the present study is that

following BPR no tachypnoeic shift was observed, which reflects a breathing pattern usually

observed only in very high level athletes (professional cyclists), and supposedly only achievable

after years of training (Eastwood, Hillman, & Finucane, 2001; Lucia, Carvajal, Calderon, Alfonso,

& Chicharro, 1999; Shephard, 1998; Steinacker, Both, & Whipp, 1993). Lucia et al. (Lucia, Hoyos,

Pardo, & Chicharro, 2001) suggested that level of endurance training was not the causative factor

for the difference in breathing pattern they observed in their 1999 study, unless it was a change

that occurred over a number of years. Syabbalo et al. (1994) suggested that a tachypnoeic

breathing pattern was caused by respiratory muscle fatigue and that high level athletes simply

developed a superior level of resistance to such fatigue due to their greater years of training and

therefore did not demonstrate such a pattern. This suggestion is unsupported by Romer et al.

(2002c) or Griffiths and McConnell (2007) who did not identify changes to breathing pattern

98
despite identifying changes in maximal inspiratory effort of 28 ± 7% and 26% respectively – a

measure used by these authors to identify respiratory muscle fatigue. In contrast, Hey et al.

(1966) theorised that the integration of the depth and frequency of breathing required to achieve a

given ventilatory demand is independent of known respiratory stimuli, and may simply reflect a

pattern of breathing that results in a minimal amount of work by the respiratory muscles. The

findings of the present study may give strength to this theory, and it is possible that the

participants who underwent BPR were able to learn, or relearn, a breathing pattern that

decreased the work of the respiratory muscles and which led to a positive effect on their exercise

performance.

It is not known how breathing pattern is reprogrammed, and prior to this study whether in fact it is

possible to reprogramme breathing pattern in trained athletes. However the results of this study

may indicate that a change at some level of automatic ventilatory control did occur. Grossman et

al. (1985) suggested that ones automatic breathing pattern can only be overridden for

approximately ten minutes before one reverts back to a pattern that is comfortable and natural,

yet the average incremental test in the present study lasted 21 minutes. The changes in

breathing pattern discussed above were evident throughout the entire duration of each

participant’s test. This could suggest that the brainstem respiratory centres, or whatever factor(s)

control the maintenance of automatic breathing pattern (currently unknown), were in some way

altered. It could be argued that the participants in the BPR group simply altered their breathing

pattern in response to verbal cues provided by the primary investigator. To address this issue, an

assistant, blinded to which research group the participants belonged, was present for all post

testing and provided the verbal encouragement. The assistant had no way of knowing which

participants would benefit from verbal cueing. Further, studies that have altered breathing pattern,

without giving an intervention period during which an athlete could acclimatise to, or automatically

99
adopt the new pattern, have failed to demonstrate changes to exercise performance (Fabre,

Perrey, Arbez, & Rouillon, 2007; Maclennan, Silvestri, Ward, & Mahler, 1994) suggesting that a

period of time is necessary (but currently unknown) for a new breathing pattern to become

automatic. Based on the findings of this study it is suggested that if BPR is to be used in

combination with IMT to further enhance the benefits of IMT, that a period of time is allowed for

BPR and the (assumed) changes to automatic breathing pattern to occur prior to the introduction

of IMT.

5.2.4. Inspiratory Drive and Timing

Despite observing changes in tI and tE, Lucia et al. (1999) found no difference in inspiratory drive

or timing between the two groups. Similarly, no change was observed in the present study post-

intervention (Figure 8) for either variable despite the changes to fb and VT. It should be borne in

mind however that changes in the mechanical properties of the respiratory system can alter VT/tI

for a given neural output of the respiratory centres (Milic-Emili & Grunstein, 1976). The changes

in the present study to both components of inspiratory drive (i.e. VT and tI), combined with the

possibility that BPR altered the mechanical properties of the respiratory system by decreasing

hyperinflation may mask any changes to VT/tI that may have occurred. It is suggested that mouth

occlusion pressure, diaphragm electromyography or phrenic nerve neurography are better

indicators of inspiratory drive than VT/tI (Lucia, Carvajal, Calderon, Alfonso, & Chicharro, 1999).

Mean values for inspiratory timing were consistently smaller than 0.5 in both groups pre and post-

BPR (Figure 8) and similar to those observed by Lucia et al. (1999).

100
5.3. Effects of BPR on Perceived Effort

The results from the present study show that four weeks of BPR is effective in decreasing the

perception of breathlessness and leg effort during maximal intensity cycling.

5.3.1. Perceived Breathing Effort

Following BPR RPEbreathing during the incremental test was decreased (benefit likely, Table 11) at

both LT (-30 ± 33.9%) and LTP (-24.7 ± 28.1%;) when compared with the control group. The

changes to RPEbreathing in the present study are larger than those reported previously. For

example, RPEbreathing was decreased following IMT by -16 ± 4% in cycling (Romer, McConnell, &

Jones, 2002a), -7.9 ± 6% in repeated sprint activity (Romer, McConnell, & Jones, 2002b) and

approximately -22%, (precise data not provided) during an incremental rowing test, however no

changes were observed at the end of this test, or during an all-out six minute test (Volianitis et al.,

2001). The effects of IMT on perceived respiratory effort has been strongly correlated (r = 0.64; P

< 0.05) with improved performance (Romer, McConnell, & Jones, 2002b) however no correlation

was identified in the present study between these two variables. Unlike the incremental test,

RPEbreathing was not substantially different during the time trial at either 15 minutes (isotime) or at

completion of the time trial. This finding is not totally surprising as the incremental test is ridden

at prescribed workloads whereas the TT has no set workload, enabling participants to ride at their

highest tolerated intensity. It is more likely that changes in perceived exertion could be identified

during the incremental test where effort during equal absolute workloads is compared, than in the

TT where workloads are variable. It is interesting to note that RPEbreathing for the control group

was increased at LT and LTP (Table 11). This is likely due to a slight detraining effect as shown

by a 1.5% decrease in power at LTP and incremental Pmax (Table 8).

101
Perceived respiratory exertion, or dyspnoea, is one of the primary reasons given for the

termination of exercise in healthy humans (Hamilton, Killian, Summers, & Jones, 1996; Killian et

al., 1992). Theoretically if dyspnoea could be decreased, exercise tolerance may be increased.

The degree to which RPEbreathing was decreased at LT and LTP in the present study may have

been a contributing factor in increased exercise performance. Just as factors that impair the

contractile properties of the respiratory muscles are thought to increase dyspnoea (for example

respiratory muscle fatigue, dynamic hyperinflation, increased accessory muscle recruitment)

factors that improve the contractile properties have the potential to decrease the intensity of

dyspnoea. BPR aims to decrease dynamic hyperinflation by maintaining an appropriate

inspiratory:expiratory ratio, and delay recruitment of the accessory muscles by encouraging a

diaphragmatic breathing pattern. It is speculated that BPR was effective in optimising respiratory

muscle function to the degree that RPEbreathing was significantly decreased, allowing participants

to exercise longer with less discomfort. Because pacing strategies are thought to be affected by

afferent (incoming central) information (Tucker et al., 2006) it is possible that the BPR group were

able to ride at a higher intensity due to changes in afferent information that was reflecting the

decreased work of breathing i.e. they rode harder because it felt easier. It is interesting to note

that post-BPR only one participant in the intervention group terminated the incremental test due

to his breathing (in combination with his legs) compared with seven pre-BPR, whilst seven in the

control group terminated exercise due to their breathing and this was unchanged pre and post-

intervention. This suggests that BPR was effective in decreasing dyspnoea to the point it was no

longer perceived as the limiting factor during exercise. This could potentially mean cyclists could

train at higher intensities without being restricted by their breathing.

Dyspnoea is a subjective experience that is partially contributed to by psychological and / or

emotional influences (Grazzini, Stendardi, Gigliotti, & Scano, 2005). During the initial stages of

102
BPR participants were exposed to feelings of breathing discomfort as they attempted to alter

(reprogramme) their VT and fb at rest and during exercise. In the early stages of retraining

participants subjectively reported feelings of air-hunger and inability to inhale deeply enough.

During this time they were educated and re-assured that these feelings were normal and just a

perceived experience, not an indication of their body’s requirement for oxygen. It is possible that

through increased exposure to unpleasant respiratory sensations in a controlled environment the

BPR group became more familiar and comfortable with the feeling of dyspnoea, knew they were

in control of their breathing and no longer perceived respiratory exertion as such an intense

experience.

Changes to the automatic breathing pattern do not occur immediately once BPR is undertaken

(Grossman, DeSwart, & Defares, 1985). Indeed it is apparent from studies that have attempted

to acutely manipulate breathing pattern that such short term changes do not affect perceived

respiratory exertion. For example, paced breathing during roller-ski skating did not change

RPEbreathing (Fabre, Perrey, Arbez, & Rouillon, 2007), nor did entraining breathing on command

during a rowing trial in rowers who were unfamiliar with entrainment (Maclennan, Silvestri, Ward,

& Mahler, 1994), which is an interesting observation considering entrainment of breathing during

upper limb exercise is thought to be a more efficient pattern (Maclennan, Silvestri, Ward, &

Mahler, 1994; Steinacker, Both, & Whipp, 1993). As anticipated, the BPR group in the present

study reported an increase in RPEbreathing during the first 2-3 weeks of BPR, both in training and

when doing their BPR exercises. This is possibly due to increased awareness of their breathing

while they were consciously trying to change it (Ley & Yelich, 1998). The findings of Fabre et al.

(2007) and Maclennan et al. (1994) in context with the subjective increased, prior to decreased

RPEbreathing in the present study suggests that an adjustment period is needed in order for

conscious changes to breathing pattern to have a beneficial effect on RPEbreathing during high

103
intensity exercise. It is not known how long this adjustment period needs to be in order to be

effective although it would have been interesting to observe what changes may have occurred

had participants in Fabre et al.’s (2007) and Maclennan et al.’s (1994) studies had time to adjust

to the altered breathing pattern.

5.3.2. Perceived Leg Effort

In addition to decreasing RPEbreathing, four weeks of BPR also decreased (benefit likely, Table 11)

perceptions of leg effort (RPElegs) in the incremental test. The percent change in RPElegs was

similar at LT (-27.9 ± 38.5%) and LTP (-24.7 ± 28.3%). As observed for RPEbreathing, the

decrease in RPElegs is larger than that observed after IMT in cyclists (-18.4%) (Romer, McConnell,

& Jones, 2002a) or repeated sprint activity (7.2 ± 0.6%) (Romer, McConnell, & Jones, 2002b).

Similar to breathing RPE, the control group in the present study rated their RPElegs as greater at

LT (13.4 ± 39.6%) and LTP (19.9 ± 43.2%) during the post-intervention test, demonstrating a

potential detraining effect.

The decrease in RPElegs may have been simply due to the same mechanism that resulted in

decreased RPEbreathing. It is possible that BPR prevented or decreased dynamic hyperinflation

which subsequently decreased the work of breathing as discussed above. Harms et al. (1997)

demonstrated that decreasing the work of breathing significantly increased blood flow to the legs

in cyclists, which may aid the clearance of metabolites usually associated with peripheral fatigue

(Cain, 1973). Blood pH is a mediator of perceptions of peripheral exertion, especially at exercise

intensities that equal or exceed the lactate threshold as in the present study (Kostka & Cafarelli,

1982). By preventing blood stealing, blood flow to the legs may have been preserved allowing

appropriate clearance of metabolites and affecting pH to a degree that peripheral exertion was

perceived as less.

104
5.4. Effects of BPR on Metabolic Measures

Breathing pattern did not clearly influence any of the physiological variables measured in the

present study, although trends suggest that BPR may positively influence B[La], power and heart

rate at LTP (Table 9). This may be significant given the nature of physiological events that occur

when exercising at and above LTP.

5.4.1. Lactate, Power and Heart Rate – LT and LTP

Ventilatory work during heavy endurance exercise may contribute to the accumulation of

metabolites (Johnson, Sharpe, & Brown, 2007) that exacerbate respiratory and locomotor muscle

fatigue. One of the physiological parameters that is most often associated with improved

exercise performance following IMT is an attenuated B[La] response (Boutellier, Buchel, Kundert,

& Spengler, 1992; Boutellier & Piwko, 1992; McConnell & Sharpe, 2005; Romer, McConnell, &

Jones, 2002b; Spengler, Roos, Laube, & Boutellier, 1999) which suggests that metabolic

responses can be modified by a breathing intervention. The present study failed to show a clear

reduction in B[La] at absolute workloads pre- and post-BPR, however there was a trend for B[La]

to be decreased at LTP (-4.8 ± 16.6%) but not LT (2.7 ± 16%) after BPR, though unclear

because of the sizeable 90% confidence intervals. It is difficult to understand why IMT affected

B[La] whilst the present study failed to show such a response. Spengler et al (1999) suggested

that IMT was effective in decreasing B[La] by increasing the ability of the respiratory muscles to

use more lactate as fuel to maintain work output. Because BPR does not specifically train the

strength or endurance of the respiratory muscles it would be unlikely to induce an effect on

lactate turnover, which may explain the difference. Rather, BPR served to decrease the work of

breathing by optimising respiratory mechanics (non-physiological) which decreased the

perception of dyspnoea and work of breathing and may lead to a reduction of blood stealing.

105
The present study measured B[La] during incremental exercise – a performance test that

McConnell and Romer (2005) suggested would overwhelm any alterations to B[La] turnover that

had been induced by IMT, which may explain why this parameter did not significantly change in

the present study. Volianitis et al. (2001) also measured B[La] during incremental exercise and,

like the present study, demonstrated a trend for B[La] to decrease, without reaching statistical

significance. In contrast, previous studies that reported large changes in B[La] were measured

during an endurance or sub maximal test – where it is thought the ability to limit the effect of

gradually increasing blood lactate may be enhanced (McConnell, 2005). It is possible that a

larger change, more consistent with previous findings, may have been observed if B[La] had been

measured during the 20km-TT. Of course, it is possible that because BPR does not specifically

increase the strength and endurance of the respiratory muscles, its ergogenic effect is not related

to minimising the B[La] response regardless of the nature of the test protocol.

Whilst power at LTP showed a trend to increase (Table 9), this was not a clear difference – a

finding consistent with McConnell and Sharpe (2005) who demonstrated no change to MLSS

power after four weeks of IMT. Further, heart rate was unchanged in the present study (despite

the trend to be decreased at LTP; Table 9) which is consistent with the change observed after

IMT (Romer, McConnell, & Jones, 2002a). Collectively, this data suggests that the ergogenic

effect of BPR is unlikely to be related to changes in these physiological (metabolic) variables.

5.4.2. Measures of Oxygen Uptake - sub-maximal (economy) and VO2peak

The present study measured sub-maximal V&O 2 and V&O 2 peak in an attempt to determine whether

BPR could reduce or affect the O2 cost (work) of breathing associated with incremental exercise.

Regardless of expression (relative or absolute), BPR had no effect on the O2 of cycling

(comparison of absolute V&O 2 at same power output pre- and post-intervention), V&O 2 peak (or

106
V& C O2 ) in either the intervention or the control group, a finding that is consistent with the lack of

effect reported by other authors who have demonstrated performance improvements after IMT

(Griffiths & McConnell, 2007; Romer, McConnell, & Jones, 2002a), VIH (Holm, Sattler, & Fregosi,

2004; Leddy et al., 2007; McMahon, Boutellier, Smith, & Spengler, 2002) or acute breathing

pacing strategies (Fabre, Perrey, Arbez, & Rouillon, 2007).

Potentially, changes in sub-maximal V&O 2 during cycling could be achieved by reducing the O2

cost of breathing by optimizing the work of breathing. Wetter et al. (1999) demonstrated that

decreasing the work of breathing by 40 ± 6% using a proportional assist ventilator at 75%

V&O 2 max significantly decreased V&O 2 from 3.2 ± 0.1 to 3.1 ± 0.1 L—min-1. At V&O 2 max intensity,

Harms et al. (1997) demonstrated a significant decrease in whole-body V&O 2 (6.2%) when

respiratory muscles were offloaded by 63%, which resulted in an 8% increase in V&O 2 legs

showing that blood (and hence O2) stealing occurred at maximal intensity. When respiratory

muscle work at V&O 2 max intensity was increased by 28%, V&O 2 legs was significantly decreased

(9.4%) but no change in whole-body V&O 2 was evident. In the present study, however, no

changes in O2 cost of exercise were observed after BPR (Table 9). To expect BPR to decrease

the work of breathing by 63%, or even 40%, would be ambitious. It is likely that the effects of

BPR on the work of breathing were much smaller and this could explain why no changes were

detected in either V&O 2 or V&O 2 peak in the present study. Although no changes to V&O 2 peak were

observed, the distribution of V&O 2 between the legs and respiratory muscles may have changed

such that V&O 2 legs was enhanced even if only by a few percent (compared with the much greater

8% increase with PVA). To be able to detect such changes was beyond the scope of this study.

107
However, the ability to measure and differentiate Qlegs from V&O 2 TOT may enhance our

understanding of the influence of BPR on blood and oxygen distribution and utilisation.

In summary, the minimal physiological changes observed after BPR are generally consistent with

previous findings for other breathing interventions such as IMT and VIH. This could be the result

of actual negligible change or a meaningful change but a lack of sensitivity of measures to detect

such small changes in the physiological measures included as part of this study. The

mechanisms underpinning how BPR contributed to improved time trial and incremental

performance may be more related to changes in respiratory muscle work, than to any

physiological adaptations given the unremarkable changes to physiological variables observed in

the present study.

5.5. Limitations

The findings presented in this thesis are considered unique insofar as this is the first study to

investigate the use of a physiotherapy intervention traditionally used with chronic respiratory

conditions, to enhance performance in sport. Consequently, there were no methodological

blueprints to guide the direction of this study. Perhaps understandably, there are a number of

limitations in this study.

To further understand the effect of BPR on performance it would have been useful to measure

respiratory muscle recruitment pattern, dynamic hyperinflation and locomotor muscle blood flow.

Understanding whether changes to these variables occurred would provide a greater

understanding of whether, and then how, BPR may affect exercise performance. Unfortunately,

the resources were not available to measures these parameters, therefore it is not possible to

objectively claim that BPR altered or delayed recruitment of the accessory muscles. Similarly, as

108
objective measures of dynamic hyperinflation during exercise were not utilised it is unclear

whether dynamic hyperinflation was i) present pre-BPR or ii) altered post-BPR. Advanced

physiological measures such as limb blood flow were not measured, adding to the difficulty

ascertaining how BPR may have influenced performance from a mechanistic perspective.

The extent to which the respiratory muscles fatigued following endurance and incremental cycling

was not established in the present study. In this thesis, it has been speculated that some of the

decrease in MIP observed by others after exhaustive exercise may not be solely due to

inspiratory muscle fatigue, but may be decreased due to the shift in mechanical advantage of the

respiratory muscles caused by a tachypnoeic breathing pattern. To this end, MIP could have

been measured pre- and post-BPR to determine whether BPR did indeed have an effect on this

variable. In addition, it is possible that the four week intervention period did not allow the full

effect of BPR to be realised (and observed/measured) for the reasons discussed in section 5.1.1.

However this time frame was chosen to ensure compliance of the participants with i) their BPR

exercises, and ii) was the maximum compliable time for standardisation of training during the

subjects competitive season. Although training diaries were provided to all participants,

compliance with accurately recorded heart rate and training volume was insufficient to accurately

quantify consistency of training volume during the intervention period. However, analysis of the

available data indicated there were no large variations in weekly training load in either group

throughout the study.

Finally, it could be argued that failing to set the cadence in the present study may have

introduced a confounding variable that affected breathing pattern. However, Prabhu et al. (1992)

found no change in fb, VT or tI/tTOT between pedalling rates from 40-80rpm and participants in the

109
present study were required to replicate their self-selected cadence between pre- and post-

intervention assessments.

5.6. Future Directions

The mechanisms underpinning how BPR resulted in enhanced endurance and incremental

cycling performance remain unclear. It is speculated that BPR results in changes to the

automatic breathing pattern, and that collectively these changes to respiratory mechanics may

benefit an athlete in three ways. Firstly, prevention of a tachypnoeic shift after BPR may have a

direct effect on inspiratory muscle fatigue by enhancing the mechanical advantage of the

respiratory muscle during intense exercise. Secondly, improved breathing mechanics may have

an indirect effect of improving blood flow distribution to the leg muscles during intense exercise

thus optimising power output. Thirdly improved respiratory mechanics may have an effect on the

intensity with which respiratory and peripheral efforts are perceived allowing the athlete to exert

themselves more.

In light of the positive outcome of the present study the list of questions still deserving attention is

long. For example, does BPR actually influence the inspiratory muscle recruitment pattern,

dynamic hyperinflation and respiratory muscle fatigue? Does BPR actually affect locomotor

muscle blood flow? What mechanisms underlie how the automatic breathing pattern is retrained,

how long does it take and once changes are established are they lasting? What is the incidence

of breathing pattern disorders in athletes in the absence of respiratory pathology?

To address some of these questions a number of suggestions for future studies can be made.

Increasing the intervention period would allow further understanding of whether BPR had a

greater effect on exercise performance once the new pattern was assimilated into training.

110
Maximal inspiratory pressure or some measure of respiratory muscle fatigue needs to be

recorded to assess whether BPR influences respiratory muscle fatigue. Similarly, recording leg

blood flow during cycling pre- and post-BPR would help the understanding as to whether BPR

enhances exercise performance by preventing blood stealing. An objective tool to measure the

incidence of breathing pattern disorders in athletes needs to be developed or the current practice

of diagnosing breathing pattern disorders based on pattern observation needs to be validated.

5.7. Conclusion

The results of the present study suggest BPR has a positive influence on 20km-TT performance

and incremental power during cycling in trained athletes. In the absence of any strong

physiological changes, it appears the observed changes in performance were due to altered

respiratory mechanics and a reduced perception of exertion. Specifically, BPR substantially

decreased fb and VE, and increased VT and tI compared with pre-intervention values, resulting in a

more controlled breathing pattern at high exercise intensities. The mechanisms underpinning the

ergogenic effect of BPR remain unclear. Further research to enhance our understanding of the

role that BPR has in sport performance is clearly warranted.

111
Appendices

112
Appendix 1 Nijmegen Questionnaire

Nijmegen Questionnaire
Never Rare Sometimes Often Very Often
1 2 1 2 1 2 1 2 1 2
Chest Pain
Feeling Tense
Blurred Vision
Dizzy spells /
light headed
Confusion
Faster / deeper
breathing
Shortness of
Breath
Tight feelings in
the chest
Bloated feeling in
the stomach
Tingling Fingers
Unable to
breathe deeply
Stiff fingers or
arms
Tight feelings
around the
mouth, jaw or
throat
Cold hands or
feet
Heart racing
(palpitations)
Feelings of
Anxiety

Never = 0Rare = 1 Sometimes = 2 Often = 3 Very Often = 4

Total 1 / 64 date……………………..

Total 2 / 64 date……………………..

113
Nijmegen Descriptors3

Never: Never.

Rare: Once a month.

Sometimes: Once a week.

Often: More than once a week but not every day.

Very often: Daily or more.

3 Participants have a copy of these descriptors in front of them when they complete the Nijmegen Questionnaire

114
Appendix 2 Clinical Breathing Retraining

Despite the fact that breathing pattern retraining has been used by physiotherapists since the

1930’s (Soley & Shock, 1938) there are no “gold standard” treatment methods. Instead the basic

method described by Lum (1983) has been modified and integrated into clinical practice by

clinicians working in the area of breathing pattern disorders and retraining. The description

provided below is a general overview, summarizing current clinical practice. The reader is

reminded that, to date, no studies have investigated time frames, appropriate intervention

strategies or physiological reasons for how and/or why breathing pattern retraining may be an

effective tool for optimising respiratory mechanics. Therefore much of this description is provided

without scientific rational for its existence, instead it is based on the experience of the author and

other physiotherapists working in the field of BPDs (Bradley, 1992; Chaitow, Bradley, & Gilbert,

2002).

There are two components of breathing pattern retraining. The first involves learning a new

motor pattern, to train the inspiratory muscles to be recruited in the optimal way. Just like

learning a new golf swing or swimming technique, the initial phase of learning and retraining

requires intense concentration. With further practice the new muscle firing pattern will become

more efficient and eventually automatic (Wilmore & Costill, 1999) and based on studies

investigating neural plasticity and motor skill acquisition this may occur anytime in the first four to

six weeks of retraining (Enorka, 1997). There are two main techniques involved in breathing

pattern retraining – diaphragmatic breathing and pursed lip breathing. Diaphragmatic breathing

retraining requires a conscious appreciation of inspiring air to the lung bases with a slight forward

abdominal displacement and passive, relaxed expiration (Jones, Dean, & Chow, 2003) and is

used to delay recruitment of the accessory breathing muscles and to encourage breathing at

optimal lung volumes. Pursed lip breathing is taught to help elongate the expiratory phase of

115
breathing to maintain an optimal inspiratory to expiratory ration, prevent breath stacking and to

avoid a tachypnoeic shift in breathing pattern (Jones, Dean, & Chow, 2003).

A patient is often able to demonstrate a diaphragmatic breathing pattern within the first treatment

session, albeit not comfortably, and to continue to demonstrate it correctly when requested. At

this point the diaphragm pattern is a learned motor skill that can be used to voluntarily override

the automatic pattern.

The second part of breathing pattern retraining is more difficult as it involves re-programming the

brainstem respiratory centres and based on Grossman et al.’s theory (1985) involves retraining

the body to respond to a higher level of CO2. The normal reaction of the body to a rising CO2

level is to increase the respiratory rate to bring the CO2 level back to appropriate parameters

(Wilmore & Costill, 1999). When a breathing pattern disorder is well established, the CO2 in the

blood stream may be lower than normal (Garssen, de Ruiter, & van Dyke, 1992). One aim of

breathing pattern retraining is to decrease the respiratory rate to 12-14 breaths per minute. As a

result, PaCO2 starts to rise relative to the brainstems incorrect perception of “normal” and the

appropriate but undesired response of trying to increase respiration is initiated. Clinicians find

that this part of breathing pattern retraining takes patients the most time, and is most difficult to

overcome as it feels uncomfortable and unnatural to the patient. It has been suggested that

decreased respiratory rate can only be tolerated for up to ten minutes at a time, and only small

increases in CO2 can be withstood before the urge to breathe is overwhelming (Grossman,

DeSwart, & Defares, 1985). A session of breathing pattern retraining is usually broken into small

groups of “correct” breathing, interspersed with the patient’s comfortable pattern, to build

tolerance to the new CO2 levels. The length of time required to adjust to the appropriate PaCO2

is unknown. However, it is likely that a breathing pattern disorder will not be overcome, despite

116
that ability to demonstrate a diaphragmatic breathing pattern on demand, until the PaCO2 level

has been “reset”.

It is common, and almost expected, that patients will feel significant discomfort when trying to

correct their breathing pattern (Innocenti, 1998). These symptoms are explained to patients at

the outset of retraining and further discussion can take place if they do arise. If a breathing

pattern disorder is well established it will feel comfortable and even normal to the patient, despite

the symptoms generated. Correction of breathing pattern will be opposed by the decreased

respiratory rate causing increased CO2 which may initiate feelings of ‘air hunger’. Further, a

patient may be used to breathing at high lung volumes secondary to dynamic hyperinflation. A

full, relaxed exhalation will alter operating lung volumes and give the impression that no enough

air is being taken into the lungs. Additionally, if mouth breathing is a habit, the increased (but

appropriate) resistance of nose breathing will feel difficult at first (Garssen, de Ruiter, & van Dyke,

1992). All of these symptoms will decrease with time and practice. If retraining is consistent and

regular the diaphragmatic breathing pattern appears to be relearned and re-programmed. It is

possible that once automatic breathing has been reprogrammed, and causative factors have

been addressed, (for example, anxiety or postural factors) no ongoing breathing intervention

would be necessary. This is unlike strength training, including respiratory muscle training, which

requires an ongoing maintenance program to preserve strength gains (Wilmore & Costill, 1999).

117
Appendix 3 Breathing Retraining

10 minutes Daily

• Lie on your back, both hands firmly on the top of your

chest.
• First breath - breathe in through your nose, out through
your mouth, pressing down on your upper chest as you
breathe out. Feel the top of your chest deflate.
• All the rest of your breaths MUST be in and out of
your nose.
• Keep the firm pressure on the top of your chest so that when you breathe, you cannot
move your upper chest. Your stomach will move up as you breathe in and fall as you
breathe out. This is not a push of your stomach, it just moves as your lungs expand and
deflate.
• Try to breathe to the following 1,2,3,4 rhythm:
BREATHE IN for the count of “1”,
BREATHE OUT for the count of “2 and 3”,
PAUSE for “4”.
• Aim for your out breath to be twice as long as your in breath with a slight pause after you
breathe out.
• Initially start with 5 breaths in this rhythm, then revert to your comfortable breathing.
Spend 10 minutes alternating between the correct pattern and your comfortable
breathing.
• As you get used to it, try to breathe with the correct pattern for more breaths in a row until
you can comfortably breathe for 10 minutes with the correct rhythm.
• Continue with 10 minutes every day. The correct pattern WILL get easier the more you
practice.

Gold Stars
• These are to remind you to correct your breathing through out the day.
• Put them in various places in your environment e.g. on your head stem, on your
computer monitor, dashboard, phone etc.
• When you see one do the following:
- Stop mentally what you are thinking about,
- Drop your shoulders, upper chest and breathe out,
- Take three normal size breaths into your stomach.
• It is very important that you do this to “reprogram” your breathing.

When Exercising
• Try to keep breathing into your stomach for as long as possible. At some point you may
revert to breathing to your upper chest but try to delay this as long as possible.
• When you breathe in, think of “sipping” the air in over your bottom lip. As you breathe out,
blow the air out over your bottom lip as if blowing out a straw. Allow your out breath to
become more forceful as exercise intensity increases.
• Try to keep your mouth opening as small as possible, for as long as possible.

118
Appendix 4 Training Diary
TRAINING DIARY
AUT University
Contact: Rachel Vickery, 0211779309, rachinaus@hotmail.com
• Fill in the date, the duration, the distance in kms, heart rate (average and max) and average speed.
• Then mark the intensity of the session, and how breathless you felt when you did it (using the 0-10 scale at the front of your diary).
• If you do sessions other than on the bike please indicate what type of session you did.
• Record training as soon as possible after each session.
• Bring this diary with you each week you come to the Millennium Institute
HR
(Average & Average Intensity Breathlessness: Any Comment?
Date Duration Distance Max) Speed 0-10 0-10 (Breathing Exercises)*
Monday

Tuesday

Wednesday

Thursday

Friday

Saturday

Sunday

* The comment on breathing exercises was only included in the diaries given to the Intervention Group participants

Total _______ _______

119
Appendix 5 Modified Borg Scale

Modified BORG scale

SCALE SEVERITY
0 No Breathlessness At All
0.5 Very, Very Slight (Just Noticeable)
1 Very Slight
2 Slight Breathlessness
3 Moderate
4 Some What Severe
5 Severe Breathlessness
6
7 Very Severe Breathlessness
8
9 Very, Very Severe (Almost Maximum)
10 Maximum

120
Appendix 6 Patient Handout – Potential Symptoms associated with
BPR

What You May Feel

When you start to change your breathing pattern it is normal for it to feel uncomfortable. Some
people describe feelings like:

• Air hunger
• “not getting enough air in”
• “not breathing in long enough”
• Light headed
• Dizziness
• Needing to take a deep breath in after finishing the “correct” breaths.
• General discomfort

All these feelings are normal. When you breathe slower (correctly) your body is retaining more
carbon dioxide than what it is used to. Usually when our carbon dioxide level gets too high we
feel the need to breathe more (think about how you feel when you have held your breath as long
as possible). It takes time for your body to adjust to the new carbon dioxide level. Over time
these feelings will decrease and the new way of breathing will feel much easier. The only thing
that will speed the change over time is spending your 10 minutes daily breathing with the 1,2,3,4
pattern, using your gold stars and trying to breathe to your stomach when you are riding.

Initially you may feel more out of breath as you exercise, or you may feel that it is harder to
breathe when you ride. Again this is normal, as you are breathing through a smaller hole,
however this will also improve and the more you practice the quicker you will adjust.

121
Appendix 7 Participant Handout – Rationale for Exercises

Why do I have to do my 10 minutes practice every day?

There are two parts to changing your breathing pattern. The first is the easy part and is as simple
as learning to stride out more when you run. It is what we call a motor skill or technique. The
gold stars are used as a reminder for you to be aware of how you are breathing and to correct
your technique if necessary. It is important to break the habit of how you are breathing, and use
the correct technique. This will be achieved through awareness of how you are breathing, and
consciously changing to the new technique of breathing to your belly. It is important to use the
visual triggers (gold stars). From experience you will not correct your breathing technique as
often if you don’t use them, and it will then be harder to breathe properly when you ride.

The second part is the hardest part as you need to reprogram the part of your brain that controls
breathing. When you breathe incorrectly over a period of time your brain gets used to a certain
level of carbon dioxide in your blood stream (usually a lower level) and bases future breathing
rates and depths on this. Although you can override your brain’s program (for example, holding
your breath under water, or breathing really quickly on command), over-riding your brain’s
program can not be tolerated for longer than about 10 minutes. The only way you can achieve a
correct breathing pattern long term is to reset the level of carbon dioxide that your brain responds
to. This is why you do your 10 minutes of correct breathing! By consciously slowing your
breathing rate, your carbon dioxide level will slowly normalise and you will be able to tolerate the
correct breathing pattern longer.

Combining the awareness and correction of how you are breathing (gold stars), with the tolerance
to maintain this correct pattern, will help you breathe more efficiently on the bike.

122
References

Aaron, E. A., Seow, K. C., Johnson, B. D., & Dempsey, J. A. (1992B). Oxygen cost of
exercise hyperpnea: implications for performance. Journal of Applied
Physiology, 72(5), 1818-1825.
Abernethy, A. B., Mackinnon, L. T., Neal, R. J., Kippers, V., & Hanrahan, S. J. (1996).
The biophysical foundations of human movement. Melbourne: Human Kinetics
Publishers.
AIHW. (2005). Australian Centre for Asthma Monitoring. Asthma in Australia 2005.
AIHW Asthma series 2. Canberra.
Aliverti, A., & Macklem, P. T. (2001). How and why exercise is impaired in COPD.
Respiration, 68, 229-239.
Askanazi, J., Silverberg, P. A., Foster, R. J., Hyman, A. I., Milic-Emili, J., & Kinney, J.
M. (1980). Effects of respiratory apparatus on breathing pattern. Journal of
Applied Physiology, 48, 577-580.
Astrand, P. O., & Rodahl, K. (1986). Textbook of work physiology: physiological bases
of exercise (3rd ed.). New York: McGraw-Hill.
ATS. (1998). Dyspnea. Mechanisms, assessment and management: a consensus
statement. American Journal of Respiratory and Critical Care Medicine, 159,
321-340.
Babcock, M. A., Pegelow, D., McClaran, S. R., Suman, O. E., & Dempsey, J. A. (1995).
Contribution of diaphragm power output to exercise-induced diaphragm fatigue.
Journal of Applied Physiology, 78(5), 1710-1719.
Baker, J., Cote, J., & Hawes. (2000). The relationship between coaching behaviours and
sport anxiety in athletes. Journal of Science and Medicine in Sport, 3(2), 110-
119.
Banez, G. A., & Culbert, T. P. (2005). An integrative approach to vocal cord
dysfunction in young athletes. Biofeedback, 33(2), 54-57.
Barclay, J. K. (1986). A delivery-independent blood flow effect on skeletal muscle
fatigue. Journal of Applied Physiology, 61(3), 1084-1090.
Batterham, A. M., & Hopkins, W. G. (2005). Making meaningful inferences about
magnitudes. Sportscience, 9, 6-13.
Belen, L., Habrard, M., Micallef, J. P., & Le Gallais, D. (2007). The performance and
efficiency of cycling with a carbon fiber eccentric chainring during incremental
exercise. Journal of Sports Medicine and Physical Fitness, 47(1), 40-45.
Blager, F. B. (2000). Paradoxical vocal fold movement: diagnosis and management.
Current Opinion in Otolaryngology & Head and Neck Surgery, 8, 180-183.
Boussana, A., Galy, O., Hue, O., Meatecki, S., Varray, A., Ramonatxo, M., et al. (2007).
The effects of prior cycling and a successive run on respiratory muscle
performance in triathletes. International Journal of Sports Medicine, 24(63-70).
Boutellier, U., Buchel, R., Kundert, A., & Spengler, C. (1992). The respiratory system
as an exercising limiting factor in normal trained subjects. European Journal of
Applied Physiology, 65, 347-353.
Boutellier, U., & Piwko, P. (1992). The respiratory system as an exercising limiting
factor in normal sedentary subjects. European Journal of Applied Physiology, 64,
145-152.
Bradley, D. (1992). Hyperventilation Syndrome. Auckland: Random House.
Brugman, S. M., & Simons, S. M. (1998). Vocal Cord Dysfunction. Don't mistake it for
asthma. The physician and sports medicine, 26(5), 63-85.

123
Buckley, J. D., Bourdon, P. C., & Woolford, S. M. (2003). Effect of measuring blood
lactate concentrations using different automated lactate analysers on blood
lactate transition thresholds. Journal of Science and Medicine in Sport, 6(4),
408-421.
Cain, W. S. (1973). Nature of perceived effort and fatigue: roles of strength and blood
flow in muscle contractions. Journal of Motor Behaviour, 5, 33-47.
Campbell, E. J. M. (1966). The relationship of the sensation of breathlessness to the act
of breathing. London: Blackwell Scientific Publications.
Capelli, C., Rosa, G., & Butti, F. (1993). Energy cost and efficiency of riding
aerodynamic bicycles. European Journal of Applied Physiology, 67, 144-149.
Chaitow, L., & Bradley, D. (2002). The structure and function of breathing. In L.
Chaitow, D. Bradley & C. Gilbert (Eds.), Multidisciplinary Approaches to
Breathing Pattern Disorders (pp. 1-41). New York: Churchill Livingston.
Chaitow, L., Bradley, D., & Gilbert, C. (2002). Multidisciplinary approaches to
breathing pattern disorders. London: Harcourt Publishers Limited.
Chatham, K., Baldwin, J., Griffiths, H., Summers, L., & Enright, S. (1999). Inspiratory
muscle training improves shuttle run performance in healthy subjects.
Physiotherapy, 85, 676-683.
Clark, J. M., Hagerman, F., & Gelfand, R. (1983). Breathing patterns during
submaximal and maximal exercise in elite oarsman. Journal of Applied
Physiology, 55(2), 440-446.
Clark, S. A., Bourdon, P. C., Schmidt, W., Singh, B., Cable, G., Onus, K. J., et al.
(2007). The effect of acute simulated moderate altitude on power, performance
and pacing strategies in well-trained cyclists. European Journal of Applied
Physiology, 20(Epub ahead of print).
Conconi, F., Grazzi, G., Casoni, I., Guglielmi, C., Borsetto, C., Ballarin, E., et al. (1996).
The conconi test; methodology after 12 years of application. International
Journal of Sports Medicine, 17, 509-519.
Corda, M., von Euler, C., & Lennerstrand, G. (1965). Proprioceptive innervation of the
diaphragm. Journal of Physiology, 178, 161-178.
Coyle, E. F., Feltner, M. E., Kautz, S. A., Hamilton, M. T., Montain, S. J., Baylor, A.
M., et al. (1991). Physiological and biomechanical factors associated with elite
endurance cycling performance. Medicine & Science in Sports & Exercise, 23(1),
93-107.
Davis, H. A., Bassett, J., Hughes, P., & Gass, G. C. (1983). Anaerobic threshold and
lactate turnpoint. European Journal of Applied Physiology and Occupational
Physiology, 50(3), 383-392.
Day, J. R., Rossiter, H. B., Coats, E. M., Skasick, A., & Whipp, B. J. (2003). The
maximally attainable VO2 during exercise in humans: the peak versus maximal
issue. Journal of Applied Physiology, 95(5), 1901-1907.
Dellaca, R. L., Aliverti, A., Pelosi, P., Carlesso, E., Chiumello, D., Pedotti, A., et al.
(2001). Estimation of end-expiratory lung volume variations by optolelectronic
plethysmography. Critcal Care Medicine, 29(9), 1807-1811.
Dempsey, J. A. (1986). Is the lung built for exercise? Medicine & Science in Sport &
Exercise, 18(143-155).
Dempsey, J. A. (2006a). Challenges for future research in exercise physiology as
applied to the respiratory system. Exercise and Sport Sciences Review, 34, 92-98.
Dempsey, J. A. (2006b). Is the healthy respiratory system (always) built for exercise.
The Journal of Physiology, 576, 339-340.

124
Dempsey, J. A., Romer, L. M., Rodman, J., Miller, J., & Smith, C. (2006).
Consequences of exercise-induced respiratory muscle work. Respiratory
Physiology & Neurobiology, 151, 242-250.
Dempsey, J. A., & Wagner, P. D. (1999). Exercise-enduced arterial hypoxaemia.
Journal of Applied Physiology, 87, 1997-2006.
Dickinson, J., Whyte, G., & McConnell, A. K. (2007). Inspiratory muscle training: a
simple cost -effective treatment for inspiratory stridor. British Journal of Sports
Medicine, 41, 694 - 695.
Doherty, M., Nobbs, L., & Noakes, T. D. (2003). Low frequency of the plateau
phenomenon during maximal exercise in british elite athletes. European Journal
of Applied Physiology 89(6), 619-623.
Eastwood, P. R., Hillman, D. R., & Finucane, K. E. (2001). Inspiratory muscle
performance in endurance athletes and sedentary subjects. Respirology, 6, 95-
194.
Edwards, A. M., & Cooke, C. B. (2004). Oxygen uptake kinetics and maximal aerobic
power are unaffected by inspiratory muscle training in healthy subjects where
time to exhaustion is extended. European Journal of Applied Physiology, 93,
139-144.
El-Manshawi, A., Killian, K. J., & Summers, E. (1986). Breathlessness during exercise
with and without resistive loading. Journal of Applied Physiology, 61(3), 896-
905.
Enorka, R. M. (1997). Neural adaptations with chronic physical activity. Journal of
Biomechanics, 30, 477-455.
Fabre, N., Perrey, S., Arbez, L., & Rouillon, J. D. (2007). Paced breathing in roller-ski
skating: effects on metabolic rate and poling forces. International Journal of
Sports Physiology and Performance, 2(46-57).
Fairburn, M. S., Coutts, K. C., Pardy, R. L., & McKenzie, D. C. (1991). Improved
respiratory muscle endurance of highly trained cyclists and the effect on
maximal exercise performance. International Journal of Sports Medicine, 12,
66-70.
Faithfull, D., Jones, J. G., & Jordan, C. (1979). Measurement of the relative
contributions of rib cage and abdomen / diaphragm to tidal breathing in man.
British Journal of Anaesthesia, 51(5), 391-398.
Falling, L. (1995). Principles and practices of pulmonary therapy. Philadelphia: W B
Saunders.
Fallon, K. E. (2004). Upper airway obstruction masquerading as exercise induced
bronchospasm in an elite road cyclist. British Journal of Sports Medicine, 38(e9).
Fallon, K. E. (2007). Commentary on inspiratory muscle training: a simple cost-
effective treatment for inspiratory stridor. British Journal of Sports Medicine, 41,
695.
Faria, E. W., Parker, D. L., & Faria, I. E. (2005). The science of cycling. Physiology
and training - part 1. Sports Medicine, 35(4), 285-312.
Faria, E. W., Parker, D. L., & Faria, I. E. (2005b). The science of cycling. Factors
affecting performance - part 2. Sports Medicine, 35(4), 313-337.
Folinsbee, L. J., Wallace, E. S., Bedi, J. F., & Horvath, S. M. (1983). Exercise
respiratory pattern in elite cyclists and sedentary subjects. Medicine and Science
in Sports and Exercise, 15(6), 503-509.
Frangolias, D., & Rhodes, E. (1996). Metabolic responses and mechanisms during water
immersion running exercise. Sports Medicine, 22(1), 38-53.

125
Frisbee, J. C., Murrant, C. L., Wilson, B. A., & Barclay, J. K. (1999). Polycythemia
decreases fatigue in tetanic contractions of canine skeletal muscle. Medicine &
Science in Sport & Exercise, 31(9), 1293-1298.
Gallagher, C. G., Brown, E., & Younes, M. (1987). Breathing pattern during maximal
exercise and during submaximal exercise with hypercapnia. Journal of Applied
Physiology, 63(1), 238-244.
Gallagher, C. G., & Younes, M. (1989). Effect of pressure assist on ventilation and
respiratory mechanics in heavy exercise. Journal of Applied Physiology, 66(4),
1824-1837.
Garssen, B., de Ruiter, C., & van Dyke, R. (1992). Breathing retraining: a rational
placebo? Clinical Psychology Review, 12, 141-153.
Gilbert, C. (2005). Better chemistry through breathing: the story of carbon dioxide and
how it can go wrong. Biofeedback, 33(3), 100-104.
Gnehm, P., Reichenbach, S., Altpeter, E., Widmer, H., & Hoppeler, H. (1997).
Influence of different racing positions on metabolic cost in elite athletes.
Medicine & Science in Sports & Exercise, 29(6), 818-823.
Gosselink, R. A. A. M., Wagenaar, R. C., & Rijswijk, H. (1995). Diaphragmatic
breathing reduces efficiency of breathing in patients with chronic obstructive
pulmonary disease. American Journal of Respiratory and Critical Care
Medicine, 151, 1136-1142.
Grappe, F., Candau, R., & Belli, A. (1997). Aerodynamic drag in field cycling with
special reference to the obrees position. Ergonomics, 40, 1299-1322.
Grazzini, M., Stendardi, L., Gigliotti, F., & Scano, G. (2005). Pathophysiology of
exercise dyspnoea in healthy subjects and in patients with chronic obstructive
pulmonary disease. Respiratory Medicine, 99, 1403-1412.
Griffiths, L. A., & McConnell, A. K. (2007). The influence of inspiratory and expiratory
muscle training upon rowing performance. European Journal of Applied
Physiology, 99, 457-466.
Grossman, P., DeSwart, J. C. G., & Defares, P. B. (1985). A controlled study of a
breathing therapy for treatment of hyperventilation syndrome. Journal of
Psychosomatic Research, 29(1), 49-58.
Hamilton, A. L., Killian, K. J., Summers, E., & Jones, N. L. (1996). Symptom intensity
and subjective limitation to exercise in patients with cardiorespiratory disorders.
Chest, 110(5), 1255-1263.
Hamlin, M. J., & Hellemans, J. (2006). Effect of intermittent normobaric hypoxic
exposure at rest on haemotological, physiological, and performance parameters
in multi-sport athletes. Journal of Sports Sciences, 1-11.
Han, J. N., Stegen, K., De Valck, C., Clement, J., & Van de Woestijne, K. P. (1996).
Influence of breathing therapy on complaints, anxiety and breathing pattern in
patients with hyperventilation syndrome and anxiety disorders. Journal of
Psychosomatic Research, 41(5), 481-493.
Hanel, B., & Secher, N. H. (1991). Maximum oxygen uptake and work capacity after
inspiratory muscle training: a controlled study. International Journal of Sports
Medicine, 9, 43-52.
Hanson, P., Claremont, A., Dempsey, J. A., & Reddan, W. (1982). Determinants and
consequences of ventilatory responses to competitive endurance running.
Journal of Applied Physiology (Respiratory and Environmental Exercise
Physiology), 52, 615-623.

126
Harms, C. A., Babcock, M. A., McClaran, S. R., Pegelow, D. F., Nickele, G. A., Nelson,
W. B., et al. (1997). Respiratory muscle work compromises leg blood flow
during maximal exercise. Journal of Applied Physiology, 82(5), 1573-1583.
Harms, C. A., Wetter, T. J., McClaran, S. R., Pegelow, D. F., Nickele, G. A., Nelson, W.
B., et al. (1998). Effects of respiratory muscle work on cardiac output and its
distribution during maximal exercise. Journal of Applied Physiology, 85(2),
609-618.
Harms, C. A., Wetter, T. J., St Croix, C. M., Pegelow, D. F., & Dempsey, J. A. (2000).
Effects of respiratory muscle work on exercise performance. Journal of Applied
Physiology, 89(131-138).
Hershenson, M. B., Kikuchi, Y., & Loring, S. H. (1988). Relative strengths of the chest
wall muscles. Journal of Applied Physiology, 66(2), 750-754.
Hershenson, M. B., Kikuchi, Y., & Tzelepis, G. E. (1989). Preferential fatigue of the rib
cage muscles during inspiratory resistive loaded ventilation. Journal of Applied
Physiology, 66(2), 750-754.
Hey, E. N., Lloyd, B. B., Cunningham, D. J. C., Jukes, M. G. M., & Bolton, D. P. G.
(1966). Effects of various respiratory stimuli on depth and frequency of
breathing in man. Respiration Physiology, 1, 193-205.
Holm, P., Sattler, A., & Fregosi, R. F. (2004). Endurance training of respiratory muscles
improves cycling performance in fit young cyclists. BMC Physiology, 4(9), 1-14.
Inbar, O., Weiner, P., Azgad, Y., Rotstein, A., & Weinstein, Y. (2000). Specific
inspiratory muscle training in well-trained endurance athletes. Medicine &
Science in Sports & Exercise, 32(7), 1233-1237.
Innocenti, D. M. (1998). Hyperventilation. Edinburgh: Churchill Livingstone.
Ito, M., Kakizaki, F., Tsuzura, Y., & Yamada, M. (1999). Immediate effect of
respiratory muscle stretch gymnastics and diaphragmatic breathing on
respiratory pattern. Internal Medicine, 38, 126-132.
Jeukendrup, A. E., & Martin, J. (2001). Improving cycling performance. How we
should spend our money. Sports Medicine, 31(7), 559-569.
Johnson, B. D., Aaron, E. A., Babcock, M. A., & Dempsey, J. A. (1996). Respiratory
muscle fatigue during exercise: implications for performance. Medicine &
Science in Sports & Exercise, 28(9), 1129-1137.
Johnson, B. D., Babcock, M. A., Suman, O. E., & Dempsey, J. A. (1993). Exercise
enduced diaphragm fatigue in healthy humans. Journal of Physiology, 460, 385-
405.
Johnson, B. D., Saupe, K. W., & Dempsey, J. A. (1992). Mechanical constraints on
exercise hyperpnea in endurance athletes. Journal of Applied Physiology, 73(3),
874-886.
Johnson, M. A., Sharpe, G. R., & Brown, P. I. (2007). Inspiratory muscle training
improves cycling time trial performance and anaerobic work capacity but not
critical power. European Journal of Applied Physiology, 15(Epub ahead of
print).
Jones, A. Y. M., Dean, E., & Chow, C. C. S. (2003). Comparison of the oxygen cost of
breathing exercises and spontaneous breathing in patients with stable chronic
obstructive pulmonary disease. Physical Therapy, 83(5), 424-431.
Kayser, B., Sliwinski, P., Yan, S., Tobiasz, M., & Macklem, P. T. (1997). Respiratory
effort sensations during exercise with induced expiratory-flow limitation in
healthy humans. Journal of Applied Physiology, 83(3), 936-947.

127
Kendrick, K., Baxi, S., & Smith, R. (2000). Usefulness of the modified 0-10 Borg scale
in assessing the degree of dyspnoea in patients with COPD and asthma. Journal
of Emergency Nursing, 26(3), 216-222.
Killian, K. J., Leblanc, P., Martin, D. H., Summers, E., Jones, N. L., & Campbell, E. J.
M. (1992). Exercise capacity and ventilatory, circulatory and symptom
limitation in patients with chronic airflow limitation. American Review of
Respiratory Disease, 146, 935-940.
Kostka, C. E., & Cafarelli, E. (1982). Effect of pH on sensation and vastus lateralis
electromyogram during cycling exercise. Journal of Applied Physiology, 52,
1181-1185.
Koulouris, N. G., & Kosmas, E. N. (2002). Negative expiratory pressure: a new
diagnostic in chest medicine. Pneumon, 15(3), 254-262.
Kowalchuk, J. M., Rossitter, H. B., Ward, S. A., & Whipp, B. J. (2002). The effect of
resistive breathing on leg muscle oxygenation using near-infrared spectroscopy
during exercise in man. Experimental Physiology, 87(5), 601-611.
Kraft, A. R., & Hoogduin, C. A. L. (1984). The hyperventilation syndrome. A pilot
study on the effectiveness of treatment. British Journal of Psychiatry, 145, 538-
542.
Larsson, L., Hemmingsson, P., & Boethius, G. (1994). Self-reported obstructive airway
symptoms are common in young cross-country skiers. Scandinavian Journal of
Medicine and Science in Sports, 4, 124-127.
Larsson, L., Wadell, K. M., Jakobsson, E. J., Burlin, L. U., & Henriksson-Larsen, K. B.
(2004). Validation of the metamax II portable metabolic measurement system.
International Journal of Sports Medicine, 25(2), 115-123.
Laursen, P. B., Shing, C. M., & Jenkins, D. G. (2003). Reproducibility of a laboratory-
based 40-km cycle time trial on a stationary wind-trainer in highly trained
cyclists. International Journal of Sports Medicine, 24(7), 481-485.
Leddy, J. J., Limprasertkul, A., Patel, S., Modlich, F., Buyea, C., Pendergrast, D. R., et
al. (2007). Isocapnic hyperpnea training improves performance in competitive
male runners. European Journal of Applied Physiology, Epub ahead of print.
Levine, S., & Henson, D. (1988). Low-frequency diaphragmatic fatigue in
spontaneously breathing humans. . Journal of Applied Physiology, 64(2), 672-
680.
Ley, R., & Yelich, G. (1998). Fractional end-tidal CO2 as an index of the effects of
stress on math performance and verbal memory of test-anxious adolescents.
Biological Psychology, 49(1-2), 83-94.
Lomax, M. E., & McConnell, A. K. (2003). Inspiratory muscle fatigue in swimmers
after a single 200 m swim. Journal of Sports Sciences, 21, 659-664.
Lucia, A., Carvajal, A., Calderon, F. J., Alfonso, A., & Chicharro, J. L. (1999).
Breathing pattern in highly competitive cyclists during incremental exercise.
European Journal of Applied Physiology, 79, 512-521.
Lucia, A., Hoyos, J., Pardo, J., & Chicharro, J. L. (2001). Effects of endurance training
on the breathing pattern of professional cyclists. Japanese Journal of Physiology,
51, 133-141.
Lucia, A., Hoyos, J., & Perez, M. (2002). Inverse relationship between VO2max and
economy / efficiency in world-class cyclists. Medicine and Science in Sports and
Exercise, 34, 2079-2084.
Lucia, A., Rabadan, M., Hoyos, J., Hernandez-Capilla, M., Perez, M., San Juan, A. F.,
et al. (2006). Frequency of the VO2 max plateau phenomenon in world-class
athletes. International Journal of Sports Medicine, 27(12), 984-992.

128
Lum, C. (1977). Breathing exercises in the treatment of hyperventilation and chronic
anxiety states. Chest, Heart and Stroke Journal, 2, 6-11.
Maclennan, S. E., Silvestri, G. A., Ward, J., & Mahler, D. A. (1994). Does entrained
breathing improve the economy of rowing. Medicine and Science in Sports and
Exercise, 26(5), 610-614.
Mador, M. J., Magalang, U. J., Rodis, A., & Kufel, T. J. (1993). Diaphragmatic fatigue
after exercise in healthy human subjects. American Review of Respiratory
Disease, 148(6 part 1), 1571-1575.
Mahler, D. A., Shuhart, C. R., Brew, E., & Stukel, T. A. (1991). Ventilatory responses
and entrainment of breathing during rowing. Medicine and Science in Sports and
Exercise, 23(2), 186-192.
Markov, G., Spengler, C. M., Knopfli-Lenzin, C., Stuessi, C., & Boutellier, U. (2001).
Respiratory muscle training increases cycling endurance without affecting
cardiovascular responses to exercise. European Journal of Applied Physiology,
85, 233-239.
Maxwell, N. S., Castle, P. C., & Spencer, M. (2007). Effect of recovery intensity on
peak power output and the development of heat strain during intermittent sprint
exercise while under stress. Journal of Science and Medicine in Sport, 11(Epub
ahead of print).
McConnell, A. K. (2005). The role of inspiratory muscle function and training in the
genesis of dyspnoea in asthma and COPD. Primary Care Respiratory Journal,
14, 186-194.
McConnell, A. K., & Romer, L. M. (2004a). Dyspnoea in health and obstructive
pulmonary disease. The role of respiratory muscle function and training. Sports
Medicine, 34(2), 117-132.
McConnell, A. K., & Romer, L. M. (2004b). Respiratory muscle training in healthy
humans: resolving the controversy. International Journal of Sports Medicine, 25,
284-293.
McConnell, A. K., & Sharpe, G. R. (2005). The effect of inspiratory muscle training
upon maximum lactate steady-state and blood lactate concentration. European
Journal of Applied Physiology, 94, 277-284.
McDonald, W. (2007). Discussion on time trial criteria. In R. Vickery (Ed.). Sunshine
Coast.
McMahon, M. E., Boutellier, U., Smith, R. M., & Spengler, C. (2002). Hyperpnea
training attenuates peripheral chemosensitivity and improves cycling
performance. The Journal of Experimental Biology, 205, 3937-3943.
Mendelli, J., Lounana, J., Messan, F., Menuet, J. J., & Petitjean, M. (2006). Testing of
pulmonary function in a professional cycling team. Journal of Sports Medicine
and Physical Fitness, 46(2), 298-306.
Meyer, T., Faude, O., Urhausen, A., & Kindermann, W. (2004). Is lactic acidosis a
cause of exercise induced hyperventilation at the respiratory compensation point.
British Journal of Sports Medicine, 38, 622-625.
Milic-Emili, J., & Grunstein, M. M. (1976). Drive and timing components of ventilation.
Chest, 70 (suppl), 131-133.
Miller, M. R., Hankinson, J., Brusasco, V., Burgos, F., Casaburi, R., Coates, A., et al.
(2005). Standardisation of spirometry. European Respiratory Journal, 26, 319-
338.
Morgan, D. W., Kohrt, W. M., Bates, B. J., & Skinner, J. S. (1987). Effects of
respiratory muscle endurance training on ventilatory and endurance performance

129
 of moderately trained cyclists. International Journal of Sports Medicine, 8, 88-
93.
Paton, C. D., & Hopkins, W. G. (2001). Tests of cycling performance. Sports Medicine,
31(7), 489-496.
Paton, C. D., & Hopkins, W. G. (2005). Combining explosive and high-resistance
training improves performance in competitive cyclists. Journal of Strength and
Conditioning Research, 19(4), 826-830.
Prabhu, B., Zintei, T., Marciniuk, D., Ciemens, R., & Gallagher, C. G. (1992). Lack of
effect of pedalling frequency on breathing pattern during bicycle ergometry.
American Review of Respiratory Disease, 145, A582.
Pyne, D. B., Boston, T., Martin, D. T., & Logan, A. (2000). Evaluation of the lactate
pro blood analyser. European Journal of Applied Physiology, 82(1-2), 1120116.
Rack, P. M. H., & Westbury, D. R. (1969). The effects of length and stimulus rate on
tension in the isometric cat soleus muscle. Journal of Physiology, 204, 443-460.
Ram, F. S. F., Holloway, E. A., & Jones, P. W. (2003). Breathing retraining for asthma.
Respiratory Medicine, 97, 501-507.
Rapee, R. M. (1985). A case of panic disorder treated with breathing retraining. Journal
of Behavior Therapy and Experimental Psychiatry, 16, 63-65.
Richter, E. A., Kiens, B., Hargreaves, M., & Kjaer, M. (1992). Effect of arm-cranking
on leg blood flow and noradrenaline spillover during leg exercise in man. Acta
Physiologica Scandinavica, 144, 9-14.
Romer, L. M., & Dempsey, J. A. (2006). Effects of exercise-induced arterial
hypoxaemia on limb muscle fatigue and performance. Clinical and Experimental
Pharmacology and Physiology, 33(4), 391-394.
Romer, L. M., McConnell, A. K., & Jones, D. A. (2002a). Effects of inspiratory muscle
training on time-trial performance in trained cyclists. Journal of Sports Sciences,
20, 547-562.
Romer, L. M., McConnell, A. K., & Jones, D. A. (2002b). Effects of inspiratory muscle
training upon recovery time during high intensity, repetitive sprint activity.
International Journal of Sports Medicine, 23, 353-360.
Romer, L. M., McConnell, A. K., & Jones, D. A. (2002c). Inspiratory muscle fatigue in
trained cyclists: effects of inspiratory muscle training. Medicine & Science in
Sports & Exercise, 34, 785-792.
Ruiter, C. d., Rijken, H., Garssen, B., & Kraaimaat, F. (1989). Breathing retraining,
exposure, and a combination of both, in the treatment of panic disorder with
agoraphobia. Behaviour Research and Therapy, 27, 647-655.
Rundell, K. W., & Spiering, M. S. (2003). Inspiratory stridor in elite athletes. Chest,
123(2), 468-474.
Saltin, B., & Astrand, P. O. (1967). Maximum oxygen uptake in athletes. Journal of
Applied Physiology, 23(3), 353-358.
Scheuermann, B. W., & Kowalchuk, J. M. (1999). Breathing patterns during slow and
fast ramp exercise in man. Experimental Physiology, 84, 109-120.
Secher, N. H., Clausen, J. P., Klausen, K., Nore, I., & Trap-Jensen, J. (1977). Central
and regional circulatory effects of adding arm exercise to les exercise. Acta
Physiologica Scandinavica, 100, 288-297.
Sharp, J., Drutz, W., & Moisan, T. (1980). Postural relief of dyspnoea in severe chronic
obstructive pulmonary disease. American Review of Respiratory Disease, 122,
201-211.

130
Sheel, A. W., Derchak, P. A., Morgan, B. J., Pegelow, D. F., Jacques, J. A., & Dempsey,
J. A. (2001). Fatigueing inspiratory muscle work causes reflex reduction in
resting leg blood flow in humans. Journal of Physiology, 537(1), 277-289.
Sheel, A. W., Koehle, M. S., Guenette, J. A., Foster, G. E., Sporer, B. C., Diep, T. T., et
al. (2006). Human ventilatory responsiveness to hypoxia is unrelated to maximal
aerobic capacity. Journal of Applied Physiology, 100(4), 1204-1209.
Sheel, A. W., Lama, I., Potvin, P., Coutts, K. D., & McKenzie, D. C. (1996).
Comparison of aero-bars versus traditional cycling postures on physiological
parameters during submaximal cycling. Canadian Journal of Applied
Physiology, 21, 16-22.
Shephard, R. J. (1998). Science and medicine of rowing: a review. Journal of Sports
Sciences, 16, 603-620.
Soley, M. H., & Shock, N. W. (1938). The etiology of effort syndrome. American
Journal of Medical Science, 196, 840-851.
Sonetti, D. A., Wetter, T. J., Pegelow, D. F., & Dempsey, J. A. (2001). Effects of
respiratory muscle training versus placebo on endurance exercise performance.
Respiration Physiology, 127, 185-199.
Sonnetti, D. A., Wetter, T. J., Pegelow, D. F., & Dempsey, J. A. (2001). Effects of respiratory
muscle training versus placebo on endurance exercise performance. Respiration
Physiology, 127, 185-199.
Spengler, C. M., Roos, M., Laube, S. M., & Boutellier, U. (1999). Decreased exercise
blood lactate concentrations after respiratory endurance training in humans.
European Journal of Applied Physiology, 79, 299-305.
Steinacker, J. M., Both, M., & Whipp, B. J. (1993). Pulmonary mechanics and
entrainment of respiration and stroke rate during rowing. International Journal
of Sports Medicine, 14, S15-S19.
Steussi, C., Spengler, C., Knopfli-Lenzin, C., Markov, G., & Boutellier, U. (2001).
Respiratory muscle endurance training in humans increases cycling endurance
without affecting blood gas concentrations. European Journal of Applied
Physiology, 84, 582-558.
Stickland, M. K., & Lovering, A. T. (2006). Exercise-induced intrapulmonary
arteriovenous shunting and pulmonary gas exchange. Exercise and Sport
Sciences Review, 34, 99-106.
Sutton, J. R. (1992). Limitations to maximal oxygen uptake. Sports Medicine, 13(2),
127-133.
Syabbalo, N. C., Krishnan, B., Zintel, T., & Gallagher, C. G. (1994). Differential
ventilatory control during constant work rate and incremental exercise.
Respiration Physiology, 97, 175-187.
Taylor, A., Rehder, K., Hyatt, R., & Parker, J. (1989). Clinical Respiratory Physiology.
Philadelphia: W B Saunders.
Thoman, R., Stoker, G., & Ross, J. (1966). The efficacy of pursed lips breathing in
patients with chronic obstructive pulmonary disease. American Review of
Respiratory Disease, 93, 100-106.
Thomas, M., McKinley, R. K., Freeman, E., Foy, C., Prodger, P., & Price, D. (2003).
Breathing retraining for dysfunctional breathing in asthma: a randomised
controlled trial. Thorax, 58, 110-115.
Tisp, B., Burns, M., Kro, D., Kao, D., Madison, R., & Herrera, J. (1986). Pursed lip
breathing using ear oximetry. Chest, 90, 218-221.

131
Tucker, R., Bester, A., Lambert, E. V., Noakes, T. D., Vaughan, C. L., & St Clair
Gibson, A. (2006). Non-random fluctuations in power output during self-paced
exercise. British Journal of Sports Medicine, 40, 912-917.
Van Dixhoorn, J., & Duivenvoorden, H. J. (1984). Efficacy of nijmegan questionnaire
in recognition of the hyperventilation syndrome. Journal of Pscyhosomatic
Research, 29(2), 199-206.
van Doorn, P., Folgering, H., & Colla, P. (1982). Control of the end-tidal PCO2 in the
hyperventilation syndrome. Effects of biofeedback and breathing instructions
compared. Bulletin Europeen de Physiopathologie Respiratoire, 18(829-836).
Vaughan, C. L. (1984). Biomechanics of Sport. Florida: CRC Press, Inc.
Verges, S., Notter, D., & Spengler, C. (2006). Influence of diaphragm and rib cage
muscle fatigue in healthy humans. Respiratory Physiology and Neurobiology,
154(3), 431-442.
Volianitis, S., McConnell, A. K., Koutedakis, Y., McNaughton, L., Backx, K., & Jones,
D. A. (2001). Inspiratory muscle training improves rowing performance.
Medicine & Science in Sports & Exercise, 33(5), 803-809.
Ward, M. E., Eidelman, D., & Stubbing, D. G. (1988). Respiratory sensation and pattern
of respiratory muscle activation during diaphragm fatigue. Journal of Applied
Physiology, 65(5), 2181-2189.
Ward, S. A. (2007). Ventilatory control in humans: constraints and limitations.
Experimental Physiology, 92(2), 357-366.
Webber, A. B., & Pryor, J. A. (1994). Physiotherapy for Respiratory and Cardiac
Problems. Singapore: Churchill-Livingston.
West, J. B. (1995). Respiratory Physiology - the essentials (5th ed.). Baltimore:
Williams & Wilkins.
Wetter, T. J., Harms, C. A., Nelson, W. B., Pegelow, D. F., & Dempsey, J. A. (1999).
Influence of respiratory muscle work on VO2 and leg blood flow during
submaximal exercise. Journal of Applied Physiology, 87(2), 643-651.
Wiles, J. D., Coleman, D., Tegerdine, M., & Swaine, I. L. (2006). The effects of
caffeine ingestion on performance time, speed and power during a laboratory-
based 1 km cycling time-trial. Journal of Sports Sciences, 24(11), 1165-1171.
Williams, J. S., Wongsathikun, J., Boon, S. M., & Acevedo, E. O. (2002). Inspiratory
muscle training fails to improve cadence capacity in athletes. Medicine and
Science in Sports and Exercise, 34(7), 1194-1198.
Wilmore, J. H., & Costill, D. L. (1999). Physiology of Sport and Exercise (2 ed.).
Auckland: Human Kinetics.
Wilson, R. C., & Jones, P. W. (1989). A comparison of the visual analogue scale and
modified borg scale for the measurement of dyspnoea during exercise. Clinical
Science, 76(3), 277-282.
Withers, R., & Hamdorf, P. (1989). Effect of immersion on lung capacities and
volumes: implications for the densitometric estimation of relative body fat.
Journal of Sports Sciences, 7, 21-30.
Witt, J. D., Guenette, J. A., Rupert, J. L., McKenzie, D. C., & Sheel, A. W. (2007).
Inspiratory muscle training attenuates the human respiratory metabolreflex.
Journal of Physiology, 13(Epub).
www.cyclingnews.com.
http://www.cyclingnews.com/road/2004/olympics04/?id=results/road_men_tt.
Wyelegala, J. A., Pendergrast, D. R., Gosselin, L. E., Warkander, D. E., & Lundgren, C.
E. G. (2007). Respiratory muscle training improves swimming endurance in
divers. European Journal of Applied Physiology, 99(4), 393-404.

132
Yoshida, T., Chida, M., Ichioka, M., & Suda, Y. (1987). Blood lactate parameters
related to aerobic capacity and endurance performance. European Journal of
Applied Physiology and Occupational Physiology, 56(1), 7-11.
Zavorsky, G. S., Murias, J. M., Gow, J., Kim, D. J., Poulin-Harnois, C., Kubow, S., et al.
(2007). Laboratory 20-km cycle time trial reproducibility [Electronic Version].
International Journal of Sports Medicine.

133