Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Physiological mechanisms of dyspnoea in patients with fibrotic interstitial lung disease and the role… Schaeffer, Michele Rebecca 2018

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
24-ubc_2018_may_schaeffer_michele.pdf [ 2.85MB ]
Metadata
JSON: 24-1.0364703.json
JSON-LD: 24-1.0364703-ld.json
RDF/XML (Pretty): 24-1.0364703-rdf.xml
RDF/JSON: 24-1.0364703-rdf.json
Turtle: 24-1.0364703-turtle.txt
N-Triples: 24-1.0364703-rdf-ntriples.txt
Original Record: 24-1.0364703-source.json
Full Text
24-1.0364703-fulltext.txt
Citation
24-1.0364703.ris

Full Text

PHYSIOLOGICAL MECHANISMS OF DYSPNOEA IN PATIENTS WITH FIBROTIC INTERSTITIAL LUNG DISEASE AND THE ROLE OF HYPEROXIA AS AN EXERCISE INTERVENTION by  Michele Rebecca Schaeffer  B.A., The University of California Berkeley, 2010 M.Sc., McGill University, 2013  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF  DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE AND POSTDOCTORAL STUDIES (Rehabilitation Sciences)  THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver)  April 2018   © Michele Rebecca Schaeffer, 2018 ii Abstract Problem:  Dyspnoea is a common and debilitating symptom in patients with fibrotic interstitial lung disease (ILD).  Unfortunately, there are no therapies that consistently reduce exertional dyspnoea in this population.  Hyperoxia is a potential intervention to acutely address many of the pathophysiological mechanisms thought to be associated with dyspnoea and exercise intolerance in fibrotic ILD.  However, additional research is needed to clarify the role and specific physiological and perceptual effects of hyperoxia during exercise in these patients. Methods:  Study 1:  Twenty fibrotic ILD patients performed two symptom limited constant work-rate cycle exercise tests at 75% of peak work-rate while breathing room air or hyperoxia, in randomized order.  Ventilatory responses as well as both the intensity and qualitative dimensions of dyspnoea were measured throughout exercise.  Study 2:  Fourteen patients with fibrotic ILD completed an incremental cycle exercise test while breathing room air and two constant work-rate cycle exercise tests while breathing room air or hyperoxia.  Diaphragmatic electromyography (EMGdi), a surrogate of neural respiratory drive (NRD), was measured with an oesophageal catheter.  Neuromechanical uncoupling (NMU) was calculated as the ratio between EMGdi (% max) and tidal volume (% vital capacity).  Dyspnoea intensity was recorded throughout exercise.  Study 3:  Sixteen patients with fibrotic ILD performed incremental and constant work-rate cycle exercise tests while breathing room air until exhaustion, wherein dyspnoea quality was evaluated throughout exercise.   Conclusions:  Study 1 demonstrated that hyperoxia results in clinically significant improvements in exercise tolerance, dyspnoea intensity, and dyspnoea quality.  Study 2 found that dyspnoea intensity was more strongly associated with estimates of NRD than NMU during normoxic incremental cycling.  However, improvements in dyspnoea intensity with hyperoxia were more strongly correlated with NMU than NRD.  Study 3 showed increased work/effort was the dominant descriptor of dyspnoea throughout incremental and constant work-rate cycle exercise, but there was an increase in the selection of unsatisfied inspiration once further tidal volume expansion was constrained.  Collectively, these results may contribute to the development and enhancement of symptom management in patients with ILD.  In the context of rehabilitation, this may translate into improvements in patient outcomes from exercise training programs. iii Lay Summary We do not have a complete understanding of what causes breathing discomfort (dyspnoea) in patients with fibrotic interstitial lung disease (ILD).  The purpose of this thesis was to evaluate the effect(s) of supplemental oxygen, a potential therapeutic intervention, on exertional symptoms and exercise performance.  Collectively, we found that supplemental oxygen improves both the intensity and quality of dyspnoea during exercise as well as exercise endurance time in fibrotic ILD patients.  Our approach to supplemental oxygen delivery may augment current exercise rehabilitation programs by allowing patients to exercise for longer durations and or/at a higher intensity, which could translate in to greater improvements in fitness.  In addition, the proposed mechanisms of improvement may inform the design of future research and ultimately therapies for symptom relief in this population. iv Preface This thesis contains the work of the candidate, Michele R. Schaeffer, under the supervision of Dr. Jordan A. Guenette.  Dr. Guenette and the candidate’s thesis committee member, Dr. Chris J. Ryerson, conceptualized the studies.  The candidate and Drs. Guenette and Ryerson collaboratively developed the study designs, including the testing protocols and specific methods.  Data collection, analysis, interpretation and document preparation are primarily the work of the candidate.   All experiments presented in this thesis received ethical approval from the Providence Health Care Research Ethics Board (UBC-PHC REB #: H13-00059) and were registered on ClinicalTrials.gov (NCT01781793).  Data were collected at the Cardiopulmonary Exercise Physiology Laboratory in the Centre for Heart and Lung Innovation at St. Paul’s Hospital, Vancouver, British Columbia.   A version of the study rationale presented in Chapter 1 was published as part of a correspondence and reproduced with permission of the © European Respiratory Society 2017: Schaeffer MR, Molgat-Seon Y, Ryerson CJ, Guenette JA. (2017). Supplemental oxygen and dyspnoea in interstitial lung disease: absence of evidence is not evidence of absence. Eur Respir Rev, 26: 170033 [dx.doi.org/10.1183/16000617.0033-2017].  All authors played a role in the content and writing of the document. A version of Chapter 2 has been published and reproduced with permission of the © European Respiratory Society 2017:  Schaeffer MR, Ryerson CJ, Ramsook AH, Molgat-Seon Y, Wilkie, SS, Dhillon SS,  Mitchell RA, Sheel AW, Khalil N, Camp PG, Guenette JA. (2017). Effects of hyperoxia on dyspnoea and exercise endurance in fibrotic interstitial lung disease. Eur Respir J, 49: 1602494 [dx.doi.org/10.1183/13993003.02494-2016] (Impact Factor: 10.569; v Ranked 4th out of 136 Pulmonary and Respiratory Medicine journals). All authors played a role in the content and writing of the manuscript. JAG was the principal investigator and contributed the original idea for the study; MRS, CJR, AWS, NK, PGC, and JAG had input into the study design and conduct of the study; MRS, CJR, AHR, YMS, SSW, RAM, and SSD collected the data; and MRS, CJR, AHR, SSD, RAM, and JAG performed data analysis. A version of Chapter 3 has been published and reproduced with the permission of the © European Respiratory Society 2018: Schaeffer MR, Ryerson CJ, Ramsook AH, Molgat-Seon Y, Wilkie, SS, Dhillon, SS,  Mitchell, RA, Sheel, AW, Khalil N, Camp PG, Guenette JA. (2018). Neurophysiological mechanisms of exertional dyspnoea in fibrotic interstitial lung disease. Eur Respir J, 51: 1701726 [10.1183/13993003.01726-2017] (Impact Factor: 10.569; Ranked 4th out of 136 Pulmonary and Respiratory Medicine journals). All authors played a role in the content and writing of the manuscript.  MRS, CJR, AWS, NK, PGC, and JAG had input into the study design and conduct of the study; MRS, CJR, AHR, YMS, SSW, SSD, and RAM collected the data; and MRS, CJR, AHR, SSD, RAM, and JAG performed data analysis.  All 14 patients that participated in this study also participated in the study presented in Chapter 2. A version of Chapter 4 has been written and will be submitted to a journal in April of 2018:  Schaeffer MR, Guenette JA, Ramsook AH, Molgat-Seon Y, Wilkie SS, Dhillon SS, Mitchell RA, Sheel AW, Ryerson CJ. Qualities of dyspnoea during exercise in fibrotic interstitial lung disease.  All authors played a role in the content and writing of the manuscript.  MRS, JAG, and CJR had input into the study design and conduct of the study; MRS, AHR, YMS, RAM, SSW, SSD, and CJR, collected the data; and MRS, JAG, AHR, RAM, and CJR performed data analysis.  All 16 patients that participated in this study also participated in the study presented in Chapter 2, and 12 of these patients also participated in the study presented in Chapter 3. vi Table of Contents Abstract .......................................................................................................................................... ii	Lay Summary ............................................................................................................................... iii	Preface ........................................................................................................................................... iv	Table of Contents ......................................................................................................................... vi	List of Tables ................................................................................................................................ ix	List of Figures ................................................................................................................................. x	List of Abbreviations .................................................................................................................. xii	Acknowledgements .................................................................................................................... xiv	Dedication ..................................................................................................................................... xv	Chapter 1: Introduction ................................................................................................................1	1.1	 Neurophysiology of dyspnoea ........................................................................................... 2	1.2	 Qualities of dyspnoea ......................................................................................................... 3	1.2.1	 Work/effort ................................................................................................................. 3	1.2.2	 Tightness ..................................................................................................................... 4	1.2.3	 Air hunger/unsatisfied inspiration ............................................................................... 5	1.3	 Mechanisms of exercise intolerance in ILD ...................................................................... 7	1.3.1	 Pulmonary gas exchange ............................................................................................. 8	1.3.2	 Respiratory mechanics ................................................................................................ 9	1.3.3	 Haemodynamic factors ............................................................................................. 11	1.3.4	 Peripheral muscle dysfunction .................................................................................. 13	1.4	 Potential treatment options for the management of dyspnoea in fibrotic interstitial lung disease ....................................................................................................................................... 13	vii 1.4.1	 Disease modifying .................................................................................................... 13	1.4.2	 Non-disease modifying ............................................................................................. 15	1.5	 Supplemental oxygen and dyspnoea in ILD .................................................................... 17	1.6	 Summary .......................................................................................................................... 20	1.7	 Purpose ............................................................................................................................. 21	1.8	 Specific aims .................................................................................................................... 22	1.9	 Hypotheses ....................................................................................................................... 22	Chapter 2: Effects of hyperoxia on dyspnoea and exercise endurance in fibrotic interstitial lung disease ...................................................................................................................................24	2.1	 Introduction ...................................................................................................................... 24	2.2	 Methods............................................................................................................................ 25	2.3	 Results .............................................................................................................................. 29	2.4	 Discussion ........................................................................................................................ 30	Chapter 3: Neurophysiological mechanisms of exertional dyspnoea in fibrotic interstitial lung disease ...................................................................................................................................42	3.1	 Introduction ...................................................................................................................... 42	3.2	 Methods............................................................................................................................ 43	3.3	 Results .............................................................................................................................. 47	3.4	 Discussion ........................................................................................................................ 49	Chapter 4: Qualitative dimensions of exertional dyspnoea in fibrotic interstitial lung disease............................................................................................................................................57	4.1	 Introduction ...................................................................................................................... 57	4.2	 Methods............................................................................................................................ 58	viii 4.3	 Results .............................................................................................................................. 61	4.4	 Discussion ........................................................................................................................ 63	Chapter 5: Conclusion .................................................................................................................77	5.1	 Overall summary .............................................................................................................. 77	5.2	 Significance ...................................................................................................................... 78	5.3	 Strengths and limitations .................................................................................................. 80	5.4	 Future directions .............................................................................................................. 81	5.5	 Overall conclusion ........................................................................................................... 83	Bibliography .................................................................................................................................85	              ix List of Tables Table 2-1. Patient characteristics and maximal exercise data ....................................................... 36	Table 2-2.  Selected sensory and physiological parameters at rest, iso-time, and peak exercise for constant work-rate cycle exercise with room air vs. hyperoxia .................................................... 37	Table 2-3. Spearman’s correlations for select parameters with the change in exercise endurance time with hyperoxia vs. room air .................................................................................................. 38	Table 2-4. Spearman’s correlations for select parameters with the change in dyspnoea intensity at iso-time with hyperoxia vs. room air ............................................................................................ 38	Table 3-1. Patient characteristics and maximal exercise data ....................................................... 53	Table 3-2. Selected sensory and physiological parameters at iso-time and peak exercise for constant work-rate cycle exercise tests with room air and hyperoxia .......................................... 54	Table 3-3.  Spearman’s correlations for selected parameters with the change in dyspnoea intensity at iso-time with hyperoxia vs. room air ......................................................................... 55	Table 4-1. Patient characteristics .................................................................................................. 69	Table 4-2. IPF vs. non-IPF patient characteristics ........................................................................ 70	Table 4-3. Selected sensory and physiological parameters at the VT/V̇E inflection and peak exercise for incremental and constant work-rate exercise tests .................................................... 71	 x List of Figures Figure 1-1. A physiological model of dyspnoea in health and disease ........................................... 3	Figure 2-1. Flow diagram of the single-blind, randomized, placebo-controlled, cross-over study design ............................................................................................................................................ 39	Figure 2-2. Box plots for exercise endurance time for constant work-rate cycle exercise tests with room air and hyperoxia ................................................................................................................. 40	Figure 2-3. Individual changes in exercise endurance time for constant work-rate cycle exercise tests with hyperoxia vs. room air .................................................................................................. 40	Figure 2-4. Individual changes in dyspnoea intensity at iso-time during constant work-rate cycle exercise tests with hyperoxia vs. room air .................................................................................... 41	Figure 2-5. Reasons for stopping constant work-rate cycle exercise tests with room air and hyperoxia ....................................................................................................................................... 41	Figure 3-1. Trend lines for individual patients demonstrating the relationship between dyspnoea intensity and (A) electromyography of the diaphragm (EMGdi), (B) neuromechanical uncoupling, and (C) tidal volume (VT) expressed as a percentage of vital capacity (VC) ........... 56	Figure 4-1. A) tidal volume (VT), B) breathing frequency (FB), C) inspiratory reserve volume (IRV), and D) operating lung volumes relative to minute ventilation (V̇E) during incremental (INCR) and constant work-rate (CWR) exercise .......................................................................... 72	Figure 4-2. Dyspnoea intensity relative to A) minute ventilation (V̇E), B) tidal volume (VT), and C) inspiratory reserve volume (IRV) during incremental (INCR) and constant work-rate (CWR) exercise ......................................................................................................................................... 73	Figure 4-3. Selection frequency of dyspnoea descriptors during A) incremental and B) constant work-rate exercise ......................................................................................................................... 74	xi Figure 4-4. Selection frequency of A) work/effort and B) unsatisfied inspiration relative to dyspnoea intensity during incremental (INCR) and constant work-rate (CWR) exercise ............ 75	Figure 4-5. Selection frequency of dyspnoea descriptors measured via a 15-item questionnaire and clustered into 10 experiences of breathing discomfort evoked by different respiratory stimuli at peak A) incremental (INCR) and B) constant work-rate (CWR) exercise ............................... 75	Figure 4-6. Ventilatory responses to exercise in patients that selected unsatisfied inspiration (selectors) and patients that did not select unsatisfied inspiration (non-selectors) during incremental exercise ...................................................................................................................... 76	 xii List of Abbreviations 6MWD  Six-minute walk distance 6MWT Six-minute walk test BMI  Body mass index CO   Cardiac output COPD  Chronic obstructive pulmonary disease DLCO  Diffusing capacity of the lungs for carbon monoxide EELV   End-expiratory lung volume EET  Exercise endurance time EILV  End-inspiratory lung volume EIAH   Exercise induced arterial hypoxaemia EMGdi Crural diaphragm electromyography FB   Breathing frequency FEV1  Forced expiratory volume in one second FiO2   Fraction of inspired oxygen FVC   Forced vital capacity HP  Hypersensitivity pneumonitis HRCT  High resolution computed tomography  HRQOL  Health-related quality-of-life IC  Inspiratory capacity ILD   Interstitial lung disease IPF  Interstitial pulmonary fibrosis IRV  Inspiratory reserve volume xiii LIP  Lymphoid interstitial pneumonia MCID   Minimal clinically important difference MRC  Medical Research Council MVV  Maximal voluntary ventilation NMU   Neuromechanical uncoupling NRD   Neural respiratory drive NSIP  Nonspecific interstitial pneumonia O2   Oxygen PETCO2 Partial pressure of end tidal carbon dioxide PR   Pulmonary rehabilitation PVR   Pulmonary vascular resistance RER  Respiratory exchange ratio RMS  Root mean square  SpO2  Arterial oxygen saturation TLC  Total lung capacity V̇O2   Metabolic rate of oxygen consumption VC   Vital capacity V̇CO2   Metabolic rate of carbon dioxide production VD   Physiological dead space V̇E   Minute ventilation V̇E/V̇CO2  Ventilatory equivalent for metabolic rate of carbon dioxide production V̇E/V̇O2  Ventilatory equivalent for metabolic rate of oxygen consumption VT   Tidal volume xiv Acknowledgements This would not have been possible without the support of my supervisor, Dr. Jordan Guenette.  Thank you for your outstanding mentorship.  You helped me develop a skillset that will allow me to be successful in multiple avenues.  You have also shown me that it is possible to have a high level career in academia while maintaining a positive life-work balance.  I would like to thank Drs. Chris Ryerson and Bill Sheel for serving on my thesis committee and for their valuable contributions to the design and interpretation of my research projects.  I appreciate the immense amount of additional time Dr. Ryerson dedicated to recruiting patients in clinic, supervising exercise tests, and helping with statistical data analysis.  I would like to thank Dr. Nasreen Khalil for her help with patient recruitment as well.  I must acknowledge the financial support I received throughout my doctoral studies from the British Columbia Lung Association, the Institute of Heart + Lung Health at St. Paul’s Hospital, and the University of British Columbia.  I also received travel bursaries to attend and present at several domestic and international scientific conferences from the Graduate Studies in Rehabilitation Sciences and the Faculty of Graduate and Postdoctoral Studies, both at the University of British Columbia.   I would like to thank the past and current members of the Cardiopulmonary Exercise Physiology Laboratory for creating a positive and collaborative learning environment: Sabrina Wilkie, Andrew Ramsook, Yannick Molgat-Seon, Nafeez Syed, Satvir Dhillon, Reid Mitchell, and Kyle Boyle.  Thank you for your guidance and friendship. Finally, I would like to thank my family and friends for always fully supporting my endeavors, even if they did not always understand them.  xv Dedication This thesis is dedicated to all of the patients who graciously contributed to this study.1 Chapter 1: Introduction Interstitial lung disease (ILD) includes a number of entities that result in inflammation and/or fibrosis of the lung parenchyma.  Idiopathic pulmonary fibrosis (IPF) is the most common fibrotic ILD, with a prevalence of approximately 42/100,000 (Hopkins et al., 2016), and most of the literature is derived from this population.  Median survival is only three years (Ley et al., 2012), with similar mortality in other ILDs when significant fibrosis is present (Churg et al., 2009; Ryerson et al., 2013).  While the pathways to fibrosis differ between patients with IPF and each of the abovementioned subtypes, the pathophysiology of these patients converges with similar symptoms and physiology at the time of diagnosis (Ryerson et al., 2014).  Collectively, prognosis for these patients is poor and treatment options are minimal.  Dyspnoea, a “subjective experience of breathing discomfort that consists of qualitatively distinct sensations that vary in intensity” (Parshall et al., 2012c), is a major source of crippling distress and is the hallmark symptom of fibrotic ILD, with over 90% of these patients reporting dyspnoea at the time of diagnosis (Bjoraker et al., 1998).  The symptom of dyspnoea is an independent predictor of morbidity and mortality and is associated with reductions in functional capacity and adverse health-related quality-of-life (King et al., 2001; Swigris et al., 2005).  However, there are no therapeutic interventions available to consistently reduce dyspnoea in this population (Raghu et al., 2011), and the mechanisms of dyspnoea remain poorly understood and understudied.  As such, the primary aim of the proposed thesis will be to further understand dyspnoea in ILD. Dyspnoea is commonly experienced during physical activity in healthy individuals and in patients with cardiopulmonary disease.  Furthermore, there are several well-established and distinct qualitative sensations of dyspnoea that include, but are not limited to, work/effort, 2 tightness, and air hunger/unsatisfied inspiration.  There is evidence that humans can reliably discriminate between these qualitative dimensions and that these sensations can be targeted independently with different physiological interventions (Parshall et al., 2012c).  It has therefore been suggested that these distinct qualities of dyspnoea likely have different neurophysiological underpinnings (O'Donnell et al., 2009; Parshall et al., 2012c).  Accordingly, the neurophysiological mechanisms of exertional dyspnoea are complex, multifactorial, and include alterations in both central and peripheral sensory inputs and their integration within higher brain centres.   1.1 Neurophysiology of dyspnoea A diagram of the proposed neurophysiology of the symptom of dyspnoea based on the available literature is presented in Figure 1-1.  Respiratory motor drive is relayed via both automatic (brainstem) and voluntary (cortical) pathways (Parshall et al., 2012c).  There is evidence that the somatosensory cortex calibrates and interprets the appropriateness of the muscular/mechanical response of the respiratory system, conveyed via multiple afferents in the lungs, the airways, and respiratory musculature, to the level of increased neural respiratory drive (NRD) during exercise (Jensen et al., 2009).  Previous studies in animals and humans have shown that the mid-brain also transmits signals in parallel to the abovementioned motor neurons to higher brain centres such as the thalamus and somatosensory cortex, which may importantly contribute to the intensity and quality of perceived dyspnoea; this is known as corollary discharge (Chen et al., 1991; Gandevia et al., 1993; Flume et al., 1994).  It has been proposed that different sources of corollary discharge may give rise to the different qualitative dimensions of dyspnoea (Parshall et al., 2012c). 3  Figure 1-1. A physiological model of dyspnoea in health and disease Reproduced with permission of the © European Respiratory Society 2016 (Jensen et al., 2016)  1.2 Qualities of dyspnoea  1.2.1 Work/effort There is evidence that the perception of an increased work/effort to breathe is evoked by a combination of respiratory muscle afferents and perceived cortical motor command or corollary discharge (el-Manshawi et al., 1986).  In healthy individuals, there are many acute physiological adaptations that occur during exercise to ensure that the respiratory system fulfills its primary role of matching alveolar ventilation to metabolic demand.  As exercise intensity increases, healthy individuals commonly become aware that their breathing requires more work/effort, just 4 as they are aware that their exercising muscles need to work harder, compared to the normal resting state (Parshall et al., 2012c).  Until the capacity to match ventilation to metabolic demand has been reached, and while the mechanical/muscular response of the respiratory system is matched to the prevailing level of increased NRD, dyspnoea intensity ratings have been shown to increase in direct proportion to indices of NRD.  While these sensations may be intense, they are not inherently unpleasant and may not elicit a negative emotional or behavioral compensatory response, such as exercise cessation (Jensen et al., 2009).  However, the sensation of increased work/effort has been demonstrated to be much greater in patients with cardiopulmonary disease, and often limits exercise in these populations (Jensen et al., 2009).  A sense of increased work/effort to breathe can be experimentally induced in healthy subjects via external resistive or elastic loads (Chapman & Rebuck, 1983; Killian et al., 1984; Lane et al., 1987; Simon et al., 1989), voluntary hyperpnoea (Lansing et al., 2000), or weakening the respiratory musculature through the alteration of the length-tension relationship, induced fatigue, or partial neuromuscular blockade (Remmers et al., 1968; Campbell et al., 1980; Moosavi et al., 2000).  When the respiratory muscles are disadvantaged by the abovementioned methods or as a result of cardiopulmonary disease, the required NRD to achieve a given ventilation is increased, and the perception of increased work/effort is magnified even in the absence of an attendant increase in ventilation (Parshall et al., 2012c).    1.2.2 Tightness The sensation of tightness is most commonly experienced in the context of bronchoconstriction (Simon et al., 1990; Moy et al., 1998; Moy et al., 2000; Lougheed et al., 2002; Lougheed et al., 2006; Harver et al., 2011), for example, in patients with asthma.  While bronchoconstriction has 5 been shown to elicit the sensation of increased work/effort in addition to tightness (Lougheed et al., 1993; Lougheed et al., 2006), blocking pulmonary afferents has been shown to reduce the perception of tightness but not work/effort (Petit & Delhez, 1970; Parshall et al., 2012c), and mechanical ventilation has been shown to relieve the perception of increased work/effort but not tightness (Binks et al., 2002).  Collectively, these findings suggest that tightness is likely related to the simulation of airway receptors and conveyed via pulmonary afferents, whereas work/effort is related to an increased NRD in the context of airflow obstruction.  1.2.3 Air hunger/unsatisfied inspiration Air hunger/unsatisfied inspiration is described as the sensation of not being able to get enough air in to the lungs (Parshall et al., 2012c).  Similar to the abovementioned sensation of increased work/effort to breathe, air hunger/unsatisfied inspiration is also experienced in the setting of an increased NRD.  However, the latter is in the context of an inability to further increase ventilation to meet the demand (Parshall et al., 2012c).  This disparity is known as neuromechanical uncoupling (NMU).  When the ventilatory demand exceeds the physiological capacity to provide further increases in ventilation, an imbalance between NRD, as sensed via corollary discharge, and afferent feedback from the respiratory mechanoreceptors develops.  This has been shown to occur only at very high levels of exercise in healthy individuals, primarily in elite athletes, but routinely in patients with cardiopulmonary or neuromuscular disease (Parshall et al., 2012c).   The sensation of air hunger/unsatisfied inspiration has been experimentally induced via chest wall strapping (Wright & Branscomb, 1954; Harty et al., 1999; O'Donnell et al., 2000), limiting VT and respiratory rate during hypoxia (Wright & Branscomb, 1954), limiting volume 6 available at the airway opening (Wright & Branscomb, 1954; Banzett et al., 1990; Moosavi et al., 2003), or elastic mechanical loading (Simon et al., 1989).  Interestingly, similar sensations of air hunger/unsatisfied inspiration have been reported during symptom limited exercise in patients with restrictive (O'Donnell et al., 1998a) and obstructive (O'Donnell et al., 1997b; Lougheed et al., 2006) lung diseases, particularly when dynamic hyperinflation restricts inspiratory capacity in the latter (Parshall et al., 2012c).  For example, Laveneziana et al. (2011) demonstrated that during both incremental and constant work-rate exercise tests in chronic obstructive pulmonary disease (COPD) patients, there is an inflection point, or plateau, in tidal volume (VT) whereby dyspnoea increases abruptly to intolerable levels, and the quality of dyspnoea transitions from a perceived increase in work/effort to a sensation of air hunger/unsatisfied inspiration.  These findings suggested that the intensity and quality of dyspnoea evolve separately and are strongly related to mechanical constraints on VT expansion in COPD.  This concept was further supported in a study by O'Donnell et al. (2012a) showing a similar inflection point in dyspnoea intensity when VT approached approximately 75% of the concurrent dynamic inspiratory capacity (IC) during exercise in patients with COPD.  When ventilatory reserve is diminished, end-inspiratory lung volume (EILV) is positioned close to total lung capacity.  This results in shortening and functional weakening of the inspiratory muscles whereby perceived respiratory work/effort increases in tandem.  This inability to further increase VT in the context of an increasing NRD results in an unpleasant perception of air hunger/unsatisfied inspiration. Air hunger/unsatisfied inspiration is intensified by stimulating an increase in spontaneous ventilatory drive (Lane et al., 1990; Chen et al., 1992; Lane & Adams, 1993) (e.g., hypoxia, hypercapnia, and acidosis (Mitchell & Babb, 2006)), and even more so if the ventilatory response is constrained (Mitchell & Babb, 2006).  This suggests that spontaneous NRD coming 7 from the brainstem is conveyed to the cerebral cortex via corollary discharge (Adams et al., 1985; Banzett et al., 1990; Chen et al., 1992).  In contrast, voluntary, or cortical (Colebatch et al., 1991; Ramsay et al., 1993), increases in NRD primarily induce a sensation of increased work/effort (Lansing et al., 2000; Moosavi et al., 2000), which in healthy individuals has been shown to be less unpleasant than air hunger/unsatisfied inspiration (Banzett et al., 2008).  Furthermore, air hunger/unsatisfied inspiration is not specific to any particular disease or stimulus.  However, a consistent finding, in both patients with a variety of cardiopulmonary diseases and healthy subjects with chest wall strapping, is that these individuals report greater difficulty and discomfort during inspiration compared to expiration (Lougheed et al., 1993; Mahler et al., 1996; O'Donnell et al., 1997b; O'Donnell et al., 1998a; O'Donnell et al., 2000; Parshall, 2002; O'Donnell et al., 2006; O'Donnell & Webb, 2008; Jensen et al., 2009; Smith et al., 2009).  Per the above, air hunger/unsatisfied inspiration is not limited to the awareness of an increase in ventilation, as that is not necessarily unpleasant (Adams et al., 1985), rather, it is an awareness that a given drive to breathe does not result in the expected level of minute ventilation (Parshall et al., 2012c).  1.3 Mechanisms of exercise intolerance in ILD Compared to healthy age- and sex-matched individuals, exercise tolerance in patients with ILD is severely diminished as characterized by a relatively low maximal oxygen uptake (V̇O2) and a reduced peak work-rate (Spiro et al., 1981; Vogiatzis & Zakynthinos, 2012).  Primary patient-reported reasons for early exercise cessation are intolerable dyspnoea and/or leg discomfort (O'Donnell et al., 1998a; Vogiatzis & Zakynthinos, 2012), with dyspnoea as the dominant limiting symptom (Laveneziana, 2010).  Diminished exercise tolerance is a clinically relevant 8 (i.e., directly relating to patient management and/or outcomes) characteristic of patients with ILD.  Reductions in baseline exercise capacity, as well as longitudinal reductions in exercise capacity, have been shown to be strong independent predictors of mortality in patients with IPF (du Bois et al., 2014). In response to exercise, these patients typically demonstrate: (i) low peak ventilations (VE) with high peak V̇E/MVV ratios (>85%); (ii) increased ventilatory equivalents for metabolic rate of oxygen uptake or carbon dioxide production (V̇E/V̇O2 and V̇E/V̇CO2, respectively); (iii) low arterial oxygen saturation (SaO2); and (iv) a rapid and shallow breathing a pattern (Vogiatzis & Zakynthinos, 2012).  Key mechanisms suggested to limit exercise tolerance include pulmonary gas exchange abnormalities, restrictive lung mechanics, circulatory impairment, and peripheral muscle dysfunction (Marciniuk et al., 1994a; Marciniuk et al., 1994b; Hansen & Wasserman, 1996; Harris-Eze et al., 1996; Vogiatzis & Zakynthinos, 2012; Panagiotou et al., 2016), although their respective relative contributions remain unknown.  1.3.1 Pulmonary gas exchange Impairments in pulmonary gas exchange lead to functionally reduced oxygen transport capacity, which can limit the ability to perform and sustain aerobic exercise (Hansen & Wasserman, 1996; Vogiatzis & Zakynthinos, 2012).  Destruction of pulmonary capillary beds and hypoxic vasoconstriction increase physiological dead space (VD) and result in increased V̇E-perfusion mismatching, while thickening of the alveolar capillary membranes impose a diffusion limitation and promotes ventilatory inefficiency.  Collectively, these structural changes increase ventilatory requirements (Austrian et al., 1951; Lourenco et al., 1965) and reduce the capacity to deliver 9 oxygen to the exercising muscles, as evidenced by a low mixed venous partial pressure of oxygen (Bush & Busst, 1988; Agusti et al., 1991). During exercise, the arterial partial pressure of carbon dioxide is largely unchanged in patients with ILD, but the arterial partial pressure of oxygen decreases, demonstrating a gas exchange impairment (Vogiatzis & Zakynthinos, 2012).  This exercise induced arterial hypoxaemia (EIAH), defined as a minimum reduction in arterial oxygen saturation of 3-4% (Dempsey & Wagner, 1999), is a result of the abovementioned increases in V̇E-perfusion mismatching and diffusion limitation, with potential contributions of intracardiac and intrapulmonary shunts (Hamer, 1964; Agusti et al., 1991; Hughes et al., 1991).  EIAH compromises oxygen delivery to exercising muscle, which could result in premature muscle fatigue, metabolic acidosis, increased perception of leg discomfort, and, ultimately, volitional exercise cessation.  Indeed, there is strong evidence that EIAH significantly impairs exercise performance in patients with ILD.  A study by Harris-Eze et al. (1994) showed that supplemental oxygen with added VD reduced submaximal V̇E and increased peak V̇O2, peak WR, and exercise time in ILD patients.  1.3.2 Respiratory mechanics Because patients with ILD have relatively low peak V̇E, it has been largely assumed that their impaired exercise performance is a direct result of pathological respiratory mechanics (Harris-Eze et al., 1996).  Lung stiffening due to fibrosis decreases compliance and increases static recoil pressure of the lung (Gibson & Pride, 1977).  Additionally, resting IC and inspiratory reserve volume (IRV) in these patients are reduced (O'Donnell et al., 2009).  During exercise, these patients adopt a more rapid and shallow breathing pattern where the observed reductions in VT 10 are proportional to their decreased vital capacity (VC) (Gowda et al., 1990).  Less baseline reserve translates to less potential for VT expansion. In contrast to health, end-expiratory lung volume (EELV) in patients with ILD stays at resting levels, despite increasing V̇E during progressive exercise (Marciniuk et al., 1994a).  Based on previous research in obese individuals (O'Donnell et al., 2012b), a condition that has also been shown to have restrictive mechanical effects on the respiratory system, it is reasonable to predict that ILD patients are already near the lower limit of EELV at rest, where further decreases during exercise would likely induce flow limitation, and is therefore avoided.  With no room to expand VT in either direction, increases in V̇E must be achieved via increases in breathing frequency. Due to the ventilatory constraints on VT expansion in ILD, patients with ILD breathe relatively closer to TLC where there is an increase is elastic loading (O'Donnell et al., 2009).  As a result, greater respiratory muscle force must be generated for inspiration (Leblanc et al., 1986; O'Donnell et al., 1998a), which increases the work of breathing (O'Donnell et al., 1998a).  Previous studies in healthy subjects have illustrated that increased demand of the respiratory system can cause blood flow redistribution away from the locomotor muscles and effectively reduce exercise capacity (Harms et al., 1997).  It is therefore possible, yet unproven, that respiratory muscle oxygen “stealing” may also contribute to exercise intolerance in patients with ILD. Research on whether or not activity limitation in patients with ILD is primarily due to respiratory factors is inconclusive.  Marciniuk et al. (1994b) investigated the role of respiratory factors in limiting exercise performance in patients with ILD by increasing ventilatory demand during exercise via dead space loading.  Exercise time, peak work-rate, and peak V̇O2 were all 11 reduced compared to the control condition.  Peak dyspnoea was higher with dead space loading and peak V̇E and SaO2 were not different between conditions.  It was therefore reasoned that exercise limitation was primarily due to respiratory abnormalities.  However, in a follow-up study, Marciniuk et al. (1994a) examined the tidal inspiratory and expiratory flow-volume loops of patients with ILD during cycle exercise at different intensities.  They observed that, on average, these patients had large ventilatory reserves at maximal exercise.  While this suggests that exercise limitation in patients with ILD is not primarily due to abnormal respiratory mechanics, it is possible that dyspnoea induced as a result of these abnormalities may lead to volitional exercise cessation before a true ventilatory limitation is achieved.  These findings are consistent with the results from their previous study with dead space loading (Marciniuk et al., 1994b), and the proposed neurophysiology of the symptom of dyspnoea (O'Donnell et al., 2009).  However, these studies have not been reproduced and are not definitive given their very small sample sizes of 6-7 patients.  Nonetheless, and as previously described, increased NRD in the context of impaired respiratory mechanics can result in increased intensity of perceived dyspnoea.  More specifically, the widening disparity between NRD and the ventilatory output has been shown to elicit an unpleasant sensation of air hunger and/or unsatisfied inspiration.  Once a critical level, or lower limit, of ventilatory reserve is achieved, this sensation can reach intolerable levels and lead to exercise cessation (O'Donnell et al., 2012a; Laveneziana et al., 2013).  1.3.3 Haemodynamic factors It has been argued that the pathophysiological mechanisms that account for the reduced peak V̇O2 in patients with ILD are more likely circulatory than ventilatory (Hansen & Wasserman, 12 1996).  High correlations of peak V̇O2 with lactate threshold, oxygen pulse, and the change in V̇O2 relative to the change in work-rate suggest exercise limitation is due to impaired oxygen transport (Hansen & Wasserman, 1996). The ability to increase cardiac output (CO) is compromised during exercise in patients with ILD.  Severe pulmonary capillary destruction, progressive parenchymal fibrosis, hypoxic vasoconstriction, and reduced lung volumes increase pulmonary vascular resistance (PVR) (Hansen & Wasserman, 1996; Vogiatzis & Zakynthinos, 2012; Panagiotou et al., 2017).  This is associated with right ventricular hypertrophy and the development of cor pulmonale (Vogiatzis & Zakynthinos, 2012; Panagiotou et al., 2017).  Furthermore, structural pathophysiological abnormalities of the lung tissue interfere with vascular recruitment and distension in the lungs during exercise.  This inhibits PVR from decreasing during exercise (Agusti et al., 1991), and consequently, ILD patients have a reduced cardiac reserve (Vogiatzis & Zakynthinos, 2012).  In accordance with Ohm’s law, PVR is equal to the difference between mean pulmonary artery pressure and left ventricular pressure divided by CO (Chemla et al., 2015).  Thus, in order for CO to increase in the setting of a relatively increased PVR, mean pulmonary artery pressure must increase to support left ventricular filling and stroke volume (Vogiatzis & Zakynthinos, 2012).  As a result, the CO response to exercise may therefore be inadequate to match increased muscle oxygen requirements, despite the aforementioned right ventricular hypertrophy (Vogiatzis & Zakynthinos, 2012; Panagiotou et al., 2017).  This provides explanation for why CO has been observed as normal during low intensity exercise, but reduced at peak exercise in ILD patients relative to age-matched healthy controls (O'Donnell et al., 2009; Vogiatzis & Zakynthinos, 2012).  13 1.3.4 Peripheral muscle dysfunction Quadriceps muscle force has been shown to be relatively reduced in patients with fibrotic ILD compared to normative values (Nishiyama et al., 2005).  This is clinically important as quadriceps muscle force is considered to be an independent predictor of peak V̇O2 (Nishiyama et al., 2005; Vogiatzis & Zakynthinos, 2012).  While it is difficult to draw definitive conclusions from the available literature, there are multiple factors that may promote reduced muscle function in ILD patients.  For example, chronic hypoxaemia, systemic inflammation and oxidative stress, use of steroids for treatment, physical inactivity with subsequent deconditioning, as well as malnutrition (Panagiotou et al., 2016).  These factors, individually or in combination with one another, may have negative implications on skeletal muscle function in this population (Panagiotou et al., 2016).  However, there is a significant lack of research in this area and more studies are needed to expand our knowledge on peripheral muscle dysfunction in ILD patients.  1.4 Potential treatment options for the management of dyspnoea in fibrotic interstitial lung disease Treatments options for dyspnoea in patients with fibrotic ILD are minimal, and include both disease modifying and non-disease modifying therapies.    1.4.1 Disease modifying No studies have shown a benefit of disease modifying drug treatment on dyspnoea.  However, there is evidence that they can slow disease progression and/or the decline in forced vital capacity (FVC).  Since FVC is correlated with dyspnoea, slowing progression may improve this symptom.   14  Pirfenidone has been shown to be a relatively safe drug with an acceptable side-effect profile (i.e., the physiological benefits outweigh the potential adverse reactions) that slows disease progression and is associated with fewer deaths in patients with IPF (Noble et al., 2011; King et al., 2014).  Specifically, pirfenidone down regulates signaling pathways for fibroblast proliferation and collagen synthesis, and in doing so, has been shown to decrease cellular and histological markers of lung fibrosis in animal models (Iyer et al., 1995; Iyer et al., 1998; Iyer et al., 1999; Hirano et al., 2006; Oku et al., 2008).  Rates of decline in FVC and six-minute walk distance (6MWD) were consistently reduced with treatment vs. placebo in several multinational, double-blind, placebo-controlled phase 3 randomized trials (Noble et al., 2011; King et al., 2014), with these changes well above the minimal clinically important difference (MCID) for each measure (du Bois et al., 2011a; du Bois et al., 2011b).  Pooled data between studies showed a reduced risk of death from any cause or pulmonary fibrosis in patients taking the drug (King et al., 2014).  No significant differences were found between treatment and placebo in regards to dyspnoea as measured by the UCSD Shortness of Breath Questionnaire (Noble et al., 2011; King et al., 2014).  Most common side effects were gastro-intestinal (nausea, dyspepsia, vomiting, anorexia), skin related (rash, photosensitivity), and dizziness, but were mild-to-moderate and generally without any clinically significant consequences (Noble et al., 2011; King et al., 2014).  Nintedanib is another comparatively safe drug that has also been shown to slow disease progression in patients with IPF (Richeldi et al., 2014).  It functions as tyrosine kinase inhibitor that disrupts known pathways involved in lung fibrosis, therefore slowing the progression of fibrosis (Richeldi et al., 2011).  Double-blind, placebo-controlled phase 3 randomized trials showed a significantly reduced rate of decline in FVC in treatment vs. placebo (Richeldi et al., 2014).  Treatment effect was inconsistent with respect to risk of exacerbations or health-related 15 quality-of-life (HRQOL) as measured by the St. George’s Respiratory Questionnaire (Richeldi et al., 2014).  Treatments effects on dyspnoea were not evaluated.  The most frequent side effects were gastro-intestinal (diarrhea) and elevated liver enzymes (Richeldi et al., 2014). While the collective results of the abovementioned trials are not striking, the direction of benefit is in favor of the drug in all studies.  Collectively, based on the available evidence, both pirfenidone and nintedanib demonstrate a consistently mild benefit.  However, this benefit is not seen in all patients.  Additionally, pirfenidone does not appear to have an effect on perceived dyspnoea, while the effect of nintedanib on dyspnoea has yet to be assessed.  1.4.2 Non-disease modifying Pulmonary rehabilitation (PR) is an intervention that includes exercise training and patient education that has a weak positive recommendation in the most recent ATS/ERS guidelines for IPF management (Raghu et al., 2011).  There is evidence for short-term improvements in dyspnoea, 6MWD, and HRQOL immediately following PR in patients with IPF.  However, these benefits were only modest and no longer evident 6 months post training (Dowman et al., 2017b; Nakazawa et al., 2017).  This is likely a product of using the same training principles derived from research in patients with COPD, which are not optimized for patients with ILD (Raghu et al., 2011).  Maintenance programs have been proposed as a means of prolonging the benefit, but the long-term effects of PR are still unclear (Dowman et al., 2017b).    While lung transplant is strongly recommended for patients with end-stage IPF (Raghu et al., 2011), the direct effect of transplant on dyspnoea has not been evaluated.  Evidence suggests favorable long-term survival post-transplant in patients with IPF, with 5-year survival rates estimated to be 50-56% (Mason et al., 2007; Keating et al., 2009).  While transplant is likely an 16 appropriate treatment for dyspnoea, it is costly and resources are highly limited (Ryerson et al., 2012). No studies have shown a benefit of anxiolytics and antidepressants on dyspnoea in fibrotic ILD.  However, there is evidence that anxiety and depression intensify the perception of dyspnoea disproportionately to the magnitude of the physiological impairment in COPD patients (Mahler, 2013).  Consequently, modulating these symptoms will likely improve dyspnoea as well (Ryerson et al., 2012).  Anxiolytics and antidepressants, specifically, have the potential to target this symptom more centrally and translate into improvements in HRQOL outcome measures.  Similarly, the role of opioids for palliative care and treatment of refractory dyspnoea in patients with fibrotic ILD in unclear (Kohberg et al., 2016).  However, based on the observed effect of opioids on dyspnoea in other chronic lung diseases (Jennings et al., 2002; Abernethy et al., 2003), it may be an appropriate treatment for dyspnoea in patients with fibrotic ILD.  Opioids decrease neural respiratory drive, which positively effects the central perception of dyspnoea and/or associated respiratory distress (Mahler, 2013).  The most commonly reported side effects from oral opioid treatment include drowsiness, nausea, vomiting, dizziness, and constipation (Jennings et al., 2002).  Opioid withdrawal syndrome is also possible but less common (Jennings et al., 2002).  Because very few studies examining the role of opioids in the treatment of dyspnoea included ILD patients, larger randomized placebo-controlled trials are still needed to determine if opioids therapy effectively reduces dyspnoea in this population (Kohberg et al., 2016).  Supplemental oxygen is routinely prescribed to patients who are hypoxaemic at rest (SpO2<88%).  The effect of long-term oxygen therapy on dyspnoea has not been evaluated in in patients ILD (Bell et al., 2017).  In the context of acute physical activity, supplemental oxygen 17 has the potential to alleviate dyspnoea by attenuating hypoxaemia and reducing the drive to breathe.  Nevertheless, the role of supplemental oxygen in the treatment of dyspnoea in ILD remains controversial as described below.    1.5 Supplemental oxygen and dyspnoea in ILD There are multiple mechanisms that may contribute to improvements in exertional dyspnoea and/or exercise tolerance with hyperoxia.  These include but are not limited to: (1) chemoreceptor stimulation of the carotid bodies; (2) decreased respiratory drive with attendant reductions in minute ventilation and consequent respiratory muscle fatigue; and (3) increased blood flow to the locomotor muscles allowing for a greater proportion of aerobic vs. anaerobic metabolism, which could delay acidosis and fatigue in the peripheral muscles (Dean et al., 1992; O'Donnell et al., 1997a; O'Donnell et al., 2001; Somfay et al., 2001).  There is strong evidence that hyperoxia improves exertional dyspnoea and exercise endurance in COPD patients with and without EIAH (Dean et al., 1992; O'Donnell et al., 1997a; O'Donnell et al., 2001; Somfay et al., 2001).  On the contrary, the results of previous work examining the effects of hyperoxia during exercise in ILD patients are inconclusive (Bye et al., 1982; Harris-Eze et al., 1996; Visca et al., 2011; Marti et al., 2013; Nishiyama et al., 2013); there is a lack of sufficient evidence from which to draw sound conclusions. Past methodological approaches to evaluate the effect of supplemental oxygen on exertional dyspnoea in patients with ILD are problematic.  These studies only report peak or end-exercise dyspnoea ratings (Bye et al., 1982; Harris-Eze et al., 1994; Visca et al., 2011; Frank et al., 2012; Nishiyama et al., 2013), thereby ignoring important changes occurring at submaximal exercise intensities where patients are more likely to be performing their activities of daily living 18 (Puente-Maestu et al., 2016).  The exercise testing modalities previously employed are variable, such as self-paced walk tests (Visca et al., 2011; Frank et al., 2012; Nishiyama et al., 2013; Cao et al., 2015) or incremental cycle exercise tests (Harris-Eze et al., 1994), which are often insensitive to detect changes in dyspnoea.  An additional concern relates to the measures in place to reduce experimental bias.  For example, a study by Visca et al. (2011) showed a beneficial effect of oxygen on dyspnoea, but did not have an appropriate control condition; therefore, it is impossible to rule out the placebo effect. The measurement of dyspnoea is founded on the principle of psychophysics, whereby a stimulus is linked to a sensation (Zechman & Wiley, 2011).  As previously mentioned, it has been established that humans can reliably detect, quantify, and discriminate qualitatively distinct sensations of dyspnoea provoked by different respiratory stimuli either applied experimentally or as a result of disease (Zechman & Wiley, 2011).  Accordingly, a more appropriate way to evaluate the effects of any therapeutic intervention on exertional dyspnoea is to standardize the stimulus intensity.  Specifically, this requires the assessment of dyspnoea at the same absolute exercise intensity and measurement time.  It is extremely challenging to standardize exercise intensity during self-paced walk tests and thus, cycle exercise tests are more appropriate by design.  Additionally, peak or end-exercise dyspnoea ratings are often insensitive to change following an intervention.  This may be attributed to exercise tolerance improving as a result of the intervention and/or the fact that symptom-limited exercise cessation typically occurs at a similar level of dyspnoea for a given individual (Puente-Maestu et al., 2016).  In other words, peak exercise is not the same stimulus if exercise time is increased (i.e., for incremental and constant work-rate cycle exercise tests) or if walking pace and/or distance changes (i.e., during self-paced walking tests).  19 In addition to standardizing exercise intensity and measurement times, the magnitude of FiO2 delivered should be the same between individuals and should be of sufficient magnitude to reverse or, at the very least, increase arterial oxygen saturation.  With the exception of one study (Harris-Eze et al., 1994), all previous studies that examined dyspnoea in ILD used nasal cannulae or venturi masks to deliver the oxygen at various flow rates (Visca et al., 2011; Frank et al., 2012; Nishiyama et al., 2013; Cao et al., 2015; Dowman et al., 2017a).  These gas delivery methods have well-established limitations (Markovitz et al., 2010; Maggiore et al., 2014) and the effective FiO2 is likely to vary between individuals based on breathing pattern.  These systems also have modest effects on improving arterial oxygen saturation compared to control conditions (Visca et al., 2011; Frank et al., 2012; Nishiyama et al., 2013; Cao et al., 2015), which may explain, at least in part, previous negative studies of oxygen on dyspnoea in ILD.  By contrast, administering oxygen through a reservoir bag connected to a two-way non-rebreathing valve, while less practical, allows for the precise control of FiO2 and is more effective at preventing arterial oxygen desaturation.  This design is critical to testing the potential benefit of increased FiO2 on dyspnoea.  If a benefit of supplemental oxygen during exercise is proven, there will be a need for additional research to identify the best means of translating these findings into clinical practice within a standard pulmonary rehabilitation setting.  Thus, in order to make an unbiased evaluation of supplemental oxygen on dyspnoea, dyspnoea ratings should be measured at the same absolute submaximal work-rate and exercise time (i.e., iso-time) while administering identical levels of effective FiO2 compared to an appropriate placebo condition.   Based on the above, it is not surprising that a recent systematic review by Bell et al. (2017) reported that while supplemental oxygen increases exercise capacity, it does not improve dyspnoea in patients with ILD.  This conclusion regarding the lack of benefit of supplemental 20 oxygen on dyspnoea in ILD needs to be interpreted with caution as methodological factors discussed previously, such as the mode of gas delivery, FiO2, the timing of dyspnoea measurements, and the exercise testing protocols, have been adequately considered.  More appropriately designed studies should be conducted to properly evaluate the impact of supplemental oxygen on dyspnoea.  1.6 Summary The available literature suggests that there is a respiratory mechanical effect on the intensity and unpleasantness of perceived dyspnoea in health vs. patients with cardiopulmonary disease.  Dyspnoea can be broken down into multiple distinguishable sensations that vary in intensity, which can be explained, at least in part, by a mismatch between the drive to breathe and the muscular/mechanical response of the respiratory system.  These distinct sensations generally do not occur independently of one another or in an isolated fashion, and while dyspnoea is not limited to the abovementioned categories, there is evidence that the sensations of work/effort, tightness, and air hunger/unsatisfied inspiration are more reliably characterized than others (i.e., rapid or heavy breathing) (Parshall et al., 2012c).  The aforementioned gas exchange, ventilatory mechanical, and circulatory abnormalities associated with ILD all contribute importantly to exercise limitation in patients with ILD.  Peripheral muscle dysfunction is also a potential contributor, though it is an understudied area that requires further investigation.  These impairments manifest as increased sensations of dyspnoea and/or leg discomfort, and these symptoms may reach intolerable levels, resulting in volitional exercise cessation, before true physiological maxima have been reached.   21  Current treatments for dyspnoea in patients with fibrotic ILD include both disease modifying and non-disease modifying therapies.  The results of recently completed phase 3 clinical trials support the use of anti-fibrotic pharmacotherapies in this population.  However, there is no clear evidence of subsequent improvements in dyspnoea or HRQOL.  Lung transplants are appropriate for patients with end-stage disease, yet it is a costly and a limited resource.  The effects of PR are modest and temporary, and the efficacy could likely be improved if the design of the intervention was population specific. While there is strong theoretical support for the use of supplemental oxygen during exercise, previous studies were not adequately designed to evaluate the acute effects of oxygen on dyspnoea and the long-term benefits in this context have not been explored.  Therefore, symptom based therapy remains an important treatment objective.  1.7 Purpose Dyspnoea is the primary symptom of patients with ILD and is a significant clinical indicator of disease severity and progression.  Prognosis for ILD patients is poor, and effective treatment options are minimal.  Supplemental oxygen has the potential to alleviate dyspnoea and improve exercise performance in patients with fibrotic ILD.  Previous studies demonstrate conflicting results on the effects of supplemental oxygen for exertion.  This likely reflects limitations in study design and we believe that the potential magnitude of improvement has been consistently underestimated.  We therefore seek to determine if supplemental oxygen can improve dyspnoea and exercise tolerance and to identify potential physiological mechanisms of improvement in these clinical outcome measures. 22 In order to develop more effective population specific symptom-based strategies for dyspnoea management, detailed pathophysiological mechanisms must first be investigated.  Thus, another objective of the proposed thesis is to comprehensively evaluate the pathophysiological mechanisms of the intensity and qualitative dimensions of exertional dyspnoea in fibrotic ILD.  1.8 Specific aims 1. To determine if hyperoxia (60% oxygen) improves exertional dyspnoea and exercise tolerance in fibrotic ILD and to examine potential mechanisms of improvement. 2. To determine whether dyspnoea intensity is primarily dictated by NRD or NMU in fibrotic ILD and to evaluate the effect of hyperoxia on each of these parameters. 3. To evaluate the evolution of the qualitative dimensions of dyspnoea in fibrotic ILD during incremental and constant work-rate cycle tests.  1.9 Hypotheses 1. Hyperoxia will result in clinically significant improvements in exertional dyspnoea intensity and endurance time. 2. NMU is (i) a significant and independent predictor and (ii) a stronger predictor vs. NRD of dyspnoea intensity in fibrotic ILD throughout exercise.  Improvement in dyspnoea intensity with hyperoxia will be independently associated with reductions in NMU. 3. Perceived “work/effort” of breathing will be the dominant qualitative descriptor of dyspnoea up until the inflection/plateau in VT.  However, “unsatisfied inspiration” will be the dominant 23 qualitative descriptor of dyspnoea after the VT inflection/plateau.  This relationship will be independent of cycle protocol (i.e., incremental vs. constant work-rate). 24 Chapter 2: Effects of hyperoxia on dyspnoea and exercise endurance in fibrotic interstitial lung disease  2.1 Introduction Dyspnoea is a common and distressing symptom in patients with ILD with adverse health implications (King et al., 2001; Swigris et al., 2005).  Currently, there are no therapeutic interventions that consistently reduce exertional dyspnoea in this population. Supplemental oxygen may alleviate dyspnoea by attenuating arterial oxygen desaturation, increasing oxygen delivery, and reducing the drive to breathe.  However, previous studies show conflicting results on the effectiveness of supplemental oxygen on dyspnoea and exercise performance in ILD (Bye et al., 1982; Swinburn et al., 1991; Harris-Eze et al., 1994; Visca et al., 2011; Frank et al., 2012; Nishiyama et al., 2013).  Methodological factors in these studies likely led to underestimation of the potential magnitude of improvement, including an insufficient fraction of inspired oxygen (FiO2) to prevent exercise induced arterial hypoxaemia and/or the use of self-paced walking tests and incremental cycle tests rather than constant work-rate exercise protocols (Harris-Eze et al., 1994; Visca et al., 2011; Frank et al., 2012; Marti et al., 2013; Nishiyama et al., 2013; Puente-Maestu et al., 2016).  Dyspnoea was also either not evaluated or only evaluated at peak exercise (Bye et al., 1982; Harris-Eze et al., 1994; Visca et al., 2011; Frank et al., 2012; Nishiyama et al., 2013), which is insensitive to change compared to more clinically relevant submaximal exercise (Puente-Maestu et al., 2016).  Finally, some studies were retrospective and did not include a blinded room air exercise trial, making it difficult to rule out the placebo effect (Visca et al., 2011; Frank et al., 2012).  Further research is needed to clarify the role of supplemental oxygen on dyspnoea and exercise capacity in ILD as 25 concluded recently in a Cochrane review (Sharp et al., 2016).  Additionally, there is evidence that humans can reliably discriminate between the qualitative dimensions of dyspnoea, and that these distinct sensations can be varied independently with different physiological interventions (Parshall et al., 2012c).  To date, no study has examined the impact of supplemental oxygen on the qualitative dimensions of dyspnoea during exercise in ILD. The purpose of this study was to determine the effects of hyperoxia on exercise endurance as well as the intensity and qualitative dimensions of exertional dyspnoea in patients with fibrotic ILD.  We hypothesized that hyperoxia would reduce dyspnoea intensity, delay the onset of unpleasant qualitative dyspnoea descriptors, and that these benefits would translate into clinically significant improvements in cycle endurance time.  2.2 Methods Participants:  Twenty fibrotic ILD patients with isolated lung involvement participated in this study.  Patients satisfying inclusion and exclusion criteria (summarized below) were recruited from the St. Paul’s Hospital ILD Clinic, Vancouver General Hospital ILD Clinic, and St. Paul’s Hospital Pulmonary Rehabilitation Program. Inclusion criteria: • A diagnosis of IPF (Raghu et al., 2011), idiopathic fibrotic NSIP (Travis et al., 2008), chronic HP (Lacasse et al., 2003), or unclassifiable ILD with a differential diagnosis that consists of the above diagnoses (Ryerson et al., 2013) • Fibrosis on high resolution computed tomography (HRCT): honeycombing, reticulation, or traction bronchiectasis • Oxygen saturation ≥92% by pulse oximetry at rest while breathing room air 26 Exclusion criteria: • Concurrent participation in or recent completion (<6 weeks) of pulmonary rehabilitation  • Significant pleural, chest wall, musculoskeletal, or other extra-pulmonary disease that, based on clinical assessment, could impair exercise capacity and/or oxygenation  • Significant emphysema: >10% volume on HRCT • Pulmonary hypertension: pulmonary artery systolic pressure >40mmHg by echocardiography or mean pulmonary artery pressure >25mmHg by right heart catheterization (patients with DLCO <50% predicted and >20% lower than FVC % predicted had mandatory echocardiographic evaluation prior to enrolment) • Prednisone >10mg/day for at least two weeks within three months of the first study visit Study design:  This prospective, single-blind, randomized, placebo-controlled, crossover study (ClinicalTrials.gov ID#: NCT01781793; Figure 2-1) received ethical approval (H13-00059).  All patients provided informed written consent.  Participation involved four visits separated by a minimum of 48 hours.  Visit 1 included medical history, chronic activity-related dyspnoea questionnaires (modified Medical Research Council scale and Oxygen Cost Diagram (Mahler & Wells, 1988)), pulmonary function testing, and a symptom-limited incremental cycle exercise test for familiarization purposes.  Visit 2 included pulmonary function testing and the same symptom-limited incremental cycle exercise test to determine peak work-rate.  Visits 3 and 4 each included pulmonary function testing followed by a symptom-limited constant work-rate cycle exercise test at 75% of peak work-rate while breathing room air (FiO2=21%) or hyperoxia (FiO2=60%), in randomized order.  Gas mixtures were delivered into a non-diffusing Douglas 27 bag connected via large bore tubing to a two-way non-rebreathing valve (Hans Rudolph, Inc., Shawnee, KS, USA).  The gas delivery system was identical for both conditions, and the gas cylinders and SpO2 monitor were obstructed from patient view.     Pulmonary function.  Spirometry, plethysmography, 12 second maximal voluntary ventilation, single-breath diffusing capacity of the lungs for carbon monoxide (DLCO), and maximal respiratory pressures were performed according to established guidelines using a commercially available testing system (Vmax Encore 229, V62J Autobox; Carefusion, Yorba Linda, CA, USA).  Values were expressed as percent predicted (Knudson et al., 1983; Morris et al., 1988; Gutierrez et al., 2004; Tan et al., 2011). Exercise testing protocol.  All exercise tests were conducted on a calibrated electronically-braked cycle ergometer (Ergoselect 200P; Ergoline, Bitz, Germany) preceded by six minutes of quiet breathing and one minute of unloaded pedaling.  Incremental tests consisted of 15-watt stepwise increases in work-rate every two minutes until symptom limitation.  Peak work-rate was defined as the highest work-rate sustained for at least 30 seconds.  Symptom-limited constant work-rate exercise tests consisted of an immediate increase in work-rate to 75% of peak incremental work-rate after the warm-up period.  Patients were required to maintain a self-selected cadence ≥60 revolutions/min throughout all exercise tests and no verbal encouragement was provided during constant work-rate tests.  Predicted work-rate and V̇O2 were calculated in accordance with Blackie et al. (1989).   Symptom evaluation.  Patients rated the intensity of “breathing discomfort” (dyspnoea) and “leg discomfort” using the Borg 0-10 category-ratio scale (Borg, 1982).  Immediately following intensity ratings, patients were asked to select the best description of their breathing using the following four options (Laveneziana et al., 2011): (1) “My breathing requires more 28 work and effort” (work and effort); (2) “I cannot get enough air in” (unsatisfied inspiration); (3) “I cannot get enough air out” (unsatisfied expiration); (4) None apply.  Patients were permitted to select more than one phrase if the phrases applied equally.  After exercise cessation, patients were asked to report their main reason(s) for stopping exercise (i.e., breathing discomfort, leg discomfort, a combination of breathing and leg discomfort, or other) and to attribute a percentage to each reason totaling 100. Cardiorespiratory responses to exercise.  Metabolic and ventilatory responses were measured on a breath-by-breath basis (Vmax Encore 229; Carefusion, Yorba Linda, CA, USA).  Arterial oxygen saturation and heart rate were measured at rest and throughout exercise by pulse oximetry and 12-lead electrocardiography, respectively.  Clinically significant desaturation was defined as a decrease in SpO2 ≥4% to a nadir of ≤88% during exercise, regardless of the resting SpO2 (Knower et al., 2001).  Operating lung volumes were assessed by inspiratory capacity manoeuvers (Guenette et al., 2013).   Statistical analyses.  The primary outcome for this study was iso-time dyspnoea intensity, defined as the highest equivalent submaximal time achieved during constant work-rate exercise tests performed by a given patient under both conditions.  Using a two-tailed paired subject formula with α=0.05 and β=0.80, we estimated that a minimum of 16 participants were needed to detect a minimal clinically important difference (MCID) of ±1 Borg units (Ries, 2005) at iso-time between treatments, assuming a standard deviation of  ±1 Borg units.  Paired t-tests were used to compare outcomes between the room air and hyperoxic conditions at rest, iso-time, and at peak exercise.  Spearman’s correlation coefficients were used to examine the association between selected physiological variables and change in cycle endurance time and dyspnoea intensity.  Reasons for stopping exercise were analyzed using a McNemar’s Exact Test.  Data 29 were analyzed using STATA® 11.2 (StataCorp LP, College Station, TX, USA).  Data are presented as mean±SD or median (interquartile range) unless otherwise specified.  Statistical significance was set at p<0.05.  2.3 Results Patient characteristics are summarized in Table 2-1.  Idiopathic pulmonary fibrosis was the most common diagnosis (55%).  Patients had markedly reduced DLCO, peak work-rate, and oxygen uptake relative to predicted values.  None of the patients met regional criteria for supplemental oxygen at rest or with ambulation (Sandberg & Fleetham, 2013). Mean values for selected sensory and physiological parameters with room air and hyperoxia are shown in Table 2-2.  The mean work-rate for the constant work-rate exercise tests was 67±27 watts, and the median iso-time was 6.0 (4.0-14.0) minutes.  Exercise endurance time increased significantly with hyperoxia compared to room air (21.9±12.9 vs. 11.6±10.0 minutes, p<0.001; Figure 2-2 and  Figure 2-3).  The duration between constant work-rate exercise tests was 7±2 days.  Eleven patients (55%) were randomized to exercise with room air first.  There were no significant differences in endurance time between randomization groups (p=0.31). Sensory responses.  Dyspnoea intensity was similar between conditions at peak exercise (p=0.69), but significantly reduced at iso-time with hyperoxia vs. room air (4.4±3.1 vs. 2.5±2.1 Borg units, p=0.001; Table 2-2 and Figure 2-4).  Fourteen patients (70%) selected unsatisfied inspiration as a qualitative descriptor of dyspnoea at some point during exercise with room air.  Three of these patients did not select this descriptor at all with hyperoxia, while the onset of this selection was significantly delayed with hyperoxia vs. room air in the remaining 11 patients (13.8±10.8 vs. 5.6±4.3 minutes, p=0.02).  Patients reported significantly lower relative 30 contribution of breathing discomfort to exercise cessation in the hyperoxia vs. room air condition (47% vs. 60%; p=0.01).  Reasons for stopping exercise included: breathing (25% in hyperoxia vs. 55% in room air, p=0.05); legs (25% vs. 20%, p=0.70); combination (35% vs. 20%, p=0.29); other (15% vs. 5%, p=0.29) (Figure 2-5).  Physiological responses to exercise.  Physiological responses to exercise are summarized in Table 2-2.  Thirty percent of patients developed clinically significant exertional arterial oxygen desaturation, per the study’s definition, during constant work-rate exercise with room air, but none with hyperoxia.  Hyperoxia significantly reduced iso-time minute ventilation (p<0.001), tidal volume (p<0.01), breathing frequency (p<0.001), and heart rate (p<0.001), and increased SpO2 (p<0.001).   Correlations.  Change in cycle endurance time was significantly correlated with between-condition changes in peak exercise SpO2 (r=0.54, p=0.01) and iso-time dyspnoea intensity (r=-0.59, p=0.006), end-inspiratory lung volume (% of total lung capacity) (r=-0.47, p=0.04), and minute ventilation (r=-0.49, p=0.03).  There were no other correlates with iso-time dyspnoea ratings.  Additional Spearman’s correlation coefficients are summarized in Table 2-3 and Table 2-4.    2.4 Discussion This randomized controlled crossover study is the most comprehensive evaluation of hyperoxia on ventilatory, sensory, and exercise performance outcomes in fibrotic ILD.  The main results are as follows: (1) Hyperoxia results in clinically significant improvements in dyspnoea intensity with favorable alterations in the qualitative dimensions of dyspnoea during exercise; (2) hyperoxia results in consistent and clinically significant improvements in exercise endurance 31 time; and (3) changes in endurance time were associated with improvements in SpO2, ventilatory responses, and dyspnoea intensity. Dyspnoea intensity.  While dyspnoea intensity was similar between the room air and hyperoxia conditions at rest and peak exercise, ratings were significantly reduced at iso-time with hyperoxia by 1.9 Borg units.  This is nearly twice the proposed COPD-derived MCID of one Borg unit (Ries, 2005).  The few studies that have investigated the acute effects of hyperoxia on exercise performance in patients with ILD did not report dyspnoea intensity (Bye et al., 1982) or only evaluated dyspnoea at peak exercise (Harris-Eze et al., 1994; Visca et al., 2011; Frank et al., 2012; Nishiyama et al., 2013).  Peak dyspnoea intensity is typically not responsive to therapy in patients with chronic respiratory diseases, including the present study, because most acute interventions (e.g., bronchodilators, helium, non-invasive ventilation, etc.) increase exercise performance and may even increase peak ventilation, thus resulting in similar or potentially higher dyspnoea intensity at peak exercise (O'Donnell et al., 1988; O'Donnell et al., 2004; Eves et al., 2006).  Had we only evaluated dyspnoea intensity at peak exercise as done previously (Visca et al., 2011; Frank et al., 2012; Nishiyama et al., 2013), we would have also concluded that hyperoxia does not improve dyspnoea.  Instead, we evaluated dyspnoea during submaximal exercise at standardized measurement times and work-rates in order to more appropriately standardize the intensity of the stimulus (Puente-Maestu et al., 2016).   Dyspnoea quality.  To our knowledge, this is the first study to examine the effects of hyperoxia on the qualitative dimensions of dyspnoea throughout exercise in any population.  A previous study demonstrated that a greater proportion of COPD patients selected unsatisfied inspiration as the dominant qualitative descriptor of dyspnoea after reaching an inflection/plateau in VT (Laveneziana et al., 2011).  After this point, inspiratory effort continues to increase but is 32 met by little or no increase in VT.  This increase in the effort-to-displacement ratio is associated with an increased sensation of unsatisfied inspiration (O'Donnell et al., 2006).  Despite different specific causes of restricted VT expansion, COPD and ILD both adopt a rapid breathing frequency when VT becomes constrained (Faisal et al., 2016).  Patients in the present study exercised with a significantly lower V̇E and were likely able to remain below their VT plateau/inflection longer with hyperoxia compared to room air.  This more favorable respiratory pattern may have contributed to the improvement in the qualitative descriptors of dyspnoea during exercise, and specifically the delayed onset of unsatisfied inspiration during exercise with hyperoxia. The reduction in ventilatory demand during exercise with hyperoxia is also reflected in patient-reported reasons for exercise cessation.  Selection of breathing discomfort as the primary reason for exercise cessation tended to be less frequent with hyperoxia compared to room air.  Patients exercised for significantly longer durations with hyperoxia, thus increasing the probability of leg muscle fatigue contributing alone or in combination with breathing discomfort as the primary locus of symptom limitation. Exercise performance.  Cycle endurance time as a clinical outcome measure has not been fully investigated in patients with ILD.  In patients with COPD, cycle endurance time is both reproducible and more responsive to interventions compared to other forms of exercise testing such as the six-minute walk test (O'Donnell et al., 1998b; Oga et al., 2000).  A recent study suggested that cycle exercise time was the most responsive measure for evaluating the efficacy of pulmonary rehabilitation in patients with idiopathic pulmonary fibrosis (Arizono et al., 2014).  However, it is still unknown if this responsiveness holds true for other interventions, and an ILD-specific MCID has not been established.  33 Previous studies in ILD demonstrate conflicting results regarding improvements in exercise tolerance with hyperoxia (Harris-Eze et al., 1994; Visca et al., 2011; Frank et al., 2012; Nishiyama et al., 2013), due to differences in exercise testing modalities, oxygen delivery systems, and the inclusion of appropriate room air placebo trials.  These studies have used either a fixed FiO2 of 60% or 100% oxygen via nasal cannula at various flow rates.  Endurance tests, such as the constant work-rate cycle exercise test used in the present study, are more responsive to interventions compared to incremental tests and the six-minute walk test (6MWT) in COPD (Puente-Maestu et al., 2016).  The lack of clinically relevant improvements (Nishiyama et al., 2013) and/or only modest improvements (Visca et al., 2011) in exercise tolerance in previous hyperoxia studies in ILD may also reflect the mode of gas delivery and the inability of nasal cannulae to deliver sufficient oxygen to adequately reverse exertional arterial oxygen desaturation.  Patient-specific optimization of flow-rates for oxygen delivery via nasal cannula can increase 6MWD compared to standard practice (Frank et al., 2012).  However, clinically significant arterial oxygen desaturation still occurs and the magnitude of the potential benefit from hyperoxia could therefore be underestimated.  In contrast, delivery of 60% oxygen via a two-way non-rebreathing valve can fully prevent exertional arterial oxygen desaturation in ILD patients, and is accompanied by significant improvements in exercise capacity (Bye et al., 1982; Harris-Eze et al., 1994; Harris-Eze et al., 1996).  Using this more rigorous oxygen delivery system, we showed that 85% of ILD patients improved more than the COPD-derived MCID for cycle exercise time of 105 seconds, and 80% improving more than 33% compared to room air (Puente-Maestu et al., 2009).  There are several potential mechanisms that could account for the observed improvements in exercise endurance time with hyperoxia.  The reduction in V̇E at iso-time and 34 peak exercise may have reflected, at least in part, a reduction in the drive to breathe, with attendant improvements in dyspnoea intensity.  Attenuating dyspnoea intensity to more tolerable levels was associated with improvements in cycle endurance time in the present study.  In addition, hyperoxia improves oxygen delivery to the skeletal muscles via enhanced blood flow and/or arteriovenous oxygen content difference, thereby increasing the anaerobic threshold and delaying the onset of metabolic acidosis and skeletal muscle fatigue (Powers et al., 1989; Harris-Eze et al., 1994). Exercise endurance time improved in 90% of our cohort, including 12 of the 14 patients that did not develop clinically significant desaturation, indicating that clinically hypoxaemic ILD patients are not the only individuals that can derive physiological and symptomatic benefits from hyperoxia.  We speculate that hyperoxia has the potential to augment the beneficial effects of pulmonary rehabilitation by enabling patients to exercise at higher intensities, even in patients with normal oxygen saturation during exercise.  Although 60% oxygen is not used routinely in pulmonary rehabilitation, these data suggest that it should be studied as a potential intervention.  This is now being investigated in a large multi-centre clinical trial (Ryerson et al., 2016). Limitations.  Our study was limited by the absence of a familiarization constant work-rate cycle test.  However, the lack of difference between randomization groups indicates that a learning effect on the constant work-rate test did not have a significant impact on our results.  Furthermore, subjects were familiarized with symptom limited exercise tests by performing two incremental tests prior to the constant work-rate tests.  Due to safety reasons, it was not feasible to blind all study personnel to the study condition.  However, we do not believe this impacted our results as all procedures were rigorously standardized, including an absence of verbal encouragement during exercise.  The metabolic cart used in the present study has not been 35 validated with an FiO2 of 0.60.  Accordingly, we were unable to compare between-condition differences in gas exchange or gas exchange-derived parameters (e.g., oxygen consumption, carbon dioxide production, ventilatory equivalents, and the respiratory exchange ratio).  Finally, this study focused exclusively on 60% oxygen as this amount of oxygen results in the greatest improvement in endurance time in COPD (Somfay et al., 2001) and has been used in similar studies in ILD (Bye et al., 1982; Harris-Eze et al., 1994).  Future studies are needed to examine the impact of a lower FiO2 and to evaluate the cost-effectiveness and practicality of our method of delivering hyperoxia during pulmonary rehabilitation.  Conclusion.  The results of this study strongly support the notion that 60% oxygen improves exercise endurance, as well as both the intensity and qualitative dimensions of exertional dyspnoea.  Mean improvements in dyspnoea intensity and exercise endurance time substantially exceeded the MCID of these outcome variables.  We speculate that our method of supplemental oxygen delivery may augment traditional pulmonary rehabilitation programs by allowing patients to train at higher intensities, even in those patients with relatively preserved pulmonary gas exchange.    36 Table 2-1. Patient characteristics and maximal exercise data Descriptive characteristics    Age 66 ± 9 Male:Female 15 : 5 BMI, kg/m2 29 ± 5 Diagnosis, n    HP 2    IPF 11   Idiopathic NSIP 1   Polymyositis-associated ILD* 1   Scleroderma-associated ILD* 1   Unclassifiable ILD 3   Undifferentiated connective tissue disease* 1   Time since diagnosis, months 34.4 (20.6-51.5) Surgical lung biopsy, n 10   Modified MRC Dyspnoea Scale 1.3 ± 1.0 Oxygen Cost Diagram, mm 63 ± 14 Pulmonary Function    FEV1, % predicted 79 ± 16 FVC, % predicted 72 ± 16 FEV1/FVC, % 79 ± 7 TLC, % predicted 64 ± 11 DLCO, % predicted 46 ± 13 Peak incremental exercise    V̇O2, ml/kg/min 19.8 ± 6.4 V̇O2, % predicted 68 ± 22 RER 1.05 ± 0.08 Work-rate, Watts 89 ± 36 Work-rate, % predicted 57 ± 17 Heart rate, bpm 130 ± 19 Heart rate, % predicted 84 ± 10 V̇E, l/min 71 ± 20 V̇E/MVV, % 66 ± 21 SpO2, % 91 ± 4 Dyspnoea, 0-10 Borg scale 5.6 ± 2.3 Leg discomfort, 0-10 Borg scale 5.7 ± 1.9 Values are means±SD or median (interquartile range). *, these patients had no extra-pulmonary manifestations; BMI, body mass index; DLCO, diffusing capacity of the lungs for carbon monoxide; FEV1, forced expiratory volume in 1 s; FVC, forced vital capacity; HP, hypersensitivity pneumonitis; IPF, idiopathic pulmonary fibrosis; LIP; lymphoid interstitial pneumonia; RER, respiratory exchange ratio; SpO2, oxygen saturation by pulse oximetry; TLC, total lung capacity; V̇E, minute ventilation; V̇O2, oxygen consumption.37 Table 2-2.  Selected sensory and physiological parameters at rest, iso-time, and peak exercise for constant work-rate cycle exercise with room air vs. hyperoxia  Rest Iso-time Peak Parameter Room Air Hyperoxia Room Air Hyperoxia Room Air Hyperoxia Dyspnoea,  0-10 Borg scale 0.1 ± 0.4 0.2 ± 0.7 4.4 ± 3.1 2.5 ± 2.1† 5.6 ± 3.0 5.4 ± 3.6 Leg discomfort,  0-10 Borg scale 0.0 ± 0.1 0.1 ± 0.2 4.5 ± 2.7 2.9 ± 1.8† 5.7 ± 2.5 6.4 ± 2.7 SpO2, % 95 ± 2 99 ± 1‡ 91 ± 3 98 ± 1‡ 90 ± 4 98 ± 1‡ Heart rate, bpm 82 ± 14 77 ± 13* 125 ± 20 116 ± 20‡ 127 ± 18 124 ± 20 Heart rate, % predicted 54 ± 10 52 ± 10* 82 ± 11 77 ± 12‡ 84 ± 11 82 ± 13 V̇E, l/min 18 ± 5 17 ± 5 69 ± 19 55 ± 15‡ 72 ± 22 62 ± 15‡ V̇E/MVV, % 17 ± 7 16 ± 6 65 ± 21 51 ± 14‡ 68 ± 22 58 ± 16‡ VT, l 0.79 ± 0.20 0.75 ± 0.21* 1.51 ± 0.45 1.41 ± 0.45† 1.51 ± 0.51 1.35 ± 0.39‡ FB, breaths/min 24 ± 9 24 ± 9 47 ± 12 40 ± 10‡ 50 ± 15 47 ± 12 EELV, %TLC 61 ± 8 60 ± 8 54 ± 7 55 ± 8 55 ± 8 55 ± 10 EILV, %TLC 80 ± 9 77 ± 8 90 ± 6 88 ± 7 91 ± 9 86 ± 11‡ PETCO2, mmHg 34 ± 5 35 ± 5 32 ± 5 37 ± 6‡ 32 ± 5 35 ± 6‡ Values are means±SD.   EELV, end-expiratory lung volume; EILV, end-inspiratory lung volume; MVV, maximal voluntary ventilation; PETCO2, partial pressure of end tidal carbon dioxide; SpO2, oxygen saturation by pulse oximetry; VC, vital capacity; V̇E, minute ventilation; VT, tidal volume. Significantly different from room air: *, p<0.05; †, p<0.01; ‡, p<0.001.38 Table 2-3. Spearman’s correlations for select parameters with the change in exercise endurance time with hyperoxia vs. room air   r p         FVC, % predicted -0.20 0.41     DLCO, % predicted -0.21 0.38     Peak incremental V̇O2, %pred 0.11 0.66     ∆SpO2 at iso-time % 0.49 0.03     ∆SpO2 at peak exercise, % 0.54 0.01     ∆V̇E at iso-time, l/min -0.49 0.03     ∆VT at iso-time, l -0.25 0.29     ∆FB at iso-time, breaths/min -0.36 0.12     ∆EELV at iso-time, %TLC -0.17 0.48     ∆EILV at iso-time, %TLC -0.47 0.04     ∆Dyspnoea at iso-time,  0-10 Borg scale -0.59 0.006     ∆, difference between selected parameter with hyperoxia vs. room air; DLCO, diffusing capacity of the lungs for carbon monoxide; EELV, end-expiratory lung volume; EILV, end-inspiratory lung volume; FB, breathing frequency; FVC, forced vital capacity; SpO2, oxygen saturation by pulse oximetry; V̇E, minute ventilation; V̇O2, oxygen consumption; VT, tidal volume.   Table 2-4. Spearman’s correlations for select parameters with the change in dyspnoea intensity at iso-time with hyperoxia vs. room air   r p        FVC, % predicted -0.02 0.93     DLCO, % predicted 0.21 0.37     Peak incremental V̇O2, %pred 0.41 0.07     ∆SpO2 at iso-time % -0.24 0.31     ∆SpO2 at peak exercise, % -0.27 0.25     ∆V̇E at iso-time, l/min 0.21 0.38     ∆VT at iso-time, l -0.23 0.33     ∆FB at iso-time, breaths/min 0.29 0.23     ∆EELV at iso-time, %TLC 0.16 0.49     ∆EILV at iso-time, %TLC 0.07 0.77        ∆, difference between selected parameter with hyperoxia vs. room air; DLCO, diffusing capacity of the lungs for carbon monoxide; EELV, end-expiratory lung volume; EET, exercise endurance time; EILV, end-inspiratory lung volume; FB, breathing frequency; FVC, forced vital capacity; SpO2, oxygen saturation by pulse oximetry; V̇E, minute ventilation; V̇O2, oxygen consumption; VT, tidal volume.39   Figure 2-1. Flow diagram of the single-blind, randomized, placebo-controlled, cross-over study design  Visit 3: Constant work-rate cycle exercise test with 60% oxygen (n=10) Visit 3: Constant work-rate cycle exercise test with room air (n=11) Allocation Enrollment Visit 1: Familiarization (n=25) Visit 2: Incremental cycle exercise test (n=21) Randomized (n=21) Visit 4: Constant work-rate cycle exercise test with room air • Discontinued participation (unrelated injury) (n=1)  Visit 4: Constant work-rate cycle exercise test with 60% oxygen Crossover Analyzed (n=9) Analyzed (n=11) Analysis Excluded (n=4) • Did not meet inclusion criteria (n=2) • Other reasons (n=2) 40  Figure 2-2. Box plots for exercise endurance time for constant work-rate cycle exercise tests with room air and hyperoxia     Figure 2-3. Individual changes in exercise endurance time for constant work-rate cycle exercise tests with hyperoxia vs. room air MCID, minimal clinically important difference (in patients with COPD).  0 5 10 15 20 25 30 35 40 45 50 Room air Hyperoxia Exercise Endurance Time (min) p<0.001 -15 -10 -5 0 5 10 15 20 25 30 35 40 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 ∆ Exercise Endurance Time min Patient MCID Mean 41  Figure 2-4. Individual changes in dyspnoea intensity at iso-time during constant work-rate cycle exercise tests with hyperoxia vs. room air MCID, minimal clinically important difference (in patients with COPD).    Figure 2-5. Reasons for stopping constant work-rate cycle exercise tests with room air and hyperoxia 0 10 20 30 40 50 60 Breathing Discomfort Leg Discomfort Combination Other Selection Frequency (%) Reasons for Stopping Room Air Hyperoxia -8 -7 -6 -5 -4 -3 -2 -1 0 1 2 3 4 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 ∆ Dyspnoea (0-10 Borg scale) Patient MCID Mean 42 Chapter 3: Neurophysiological mechanisms of exertional dyspnoea in fibrotic interstitial lung disease  3.1 Introduction Dyspnoea is a common and debilitating symptom in patients with fibrotic ILD.  In this population, dyspnoea is an independent predictor of morbidity and mortality, and is associated with decreased HRQOL and reductions in functional capacity (King et al., 2001; Swigris et al., 2005).  Despite its clinical importance, our understanding of the mechanisms of dyspnoea in fibrotic ILD remains incomplete. The mechanisms of dyspnoea, in both health and disease, are complex and multifactorial (Parshall et al., 2012c).  However, accumulating evidence suggests that NRD and the imbalance between NRD and the mechanical response of the respiratory system (i.e., NMU) are key contributors (Jensen et al., 2016).  A strong relationship has recently been established between NRD, estimated using crural diaphragm electromyography (EMGdi), and dyspnoea intensity in patients with ILD (Faisal et al., 2016).  Furthermore, a study in COPD showed that dyspnoea intensity was strongly associated with both NRD and NMU.  However, the correlation between dyspnoea intensity and NRD was more significant (Jolley et al., 2015).  It is unknown whether these findings are also translatable to patients with ILD. Treatment options to reduce dyspnoea in ILD are limited.  Recent evidence suggests breathing a hyperoxic inspirate (FiO2 = 0.60) significantly reduces dyspnoea intensity and can eliminate or delay the onset of “unsatisfied inspiration” during exercise (Schaeffer et al., 2017).  To our knowledge, no study has investigated the effect of supplemental oxygen on NRD and/or NMU in ILD in order to determine which outcome is more strongly associated with reduced 43 exertional dyspnoea intensity.  Accordingly, the aims of this study were to determine if dyspnoea intensity in fibrotic ILD is better predicted by NRD or NMU (Aim 1), and to examine the effect of 60% oxygen on EMGdi-derived measures of NRD and NMU (Aim 2).  3.2  Methods Participants.  This study included 14 fibrotic ILD patients with isolated lung involvement that simultaneously participated in a previously published study (n=20) examining the effects of hyperoxia on cycle endurance time and dyspnoea (Schaeffer et al., 2017) (see Chapter 2).  All patients had resting arterial oxygen saturation ≥92%.  Patients who recently completed pulmonary rehabilitation (< six weeks prior to enrollment), or with significant emphysema, pulmonary hypertension, extra-pulmonary disease, and/or other contraindications to exercise were excluded from participating. Study design.  This study was a component of a larger randomized, single-blind, placebo-controlled, cross-over study (ClinicalTrials.gov ID#: NCT01781793) with institutional ethical approval (H13-00059).  Written informed consent was obtained from all patients prior to enrollment.  Patients visited the laboratory on four occasions separated by ≥48 hours between visits.  Medical history, chronic activity-related dyspnoea questionnaires, pulmonary function testing, and a symptom-limited incremental cycle exercise test with dyspnoea assessment for familiarization purposes were completed at Visit 1.  During Visit 2, patients performed pulmonary function testing and a symptom-limited incremental cycle exercise test whilst instrumented with an EMGdi catheter.  Visit 2 data were used to address the primary aim. During Visits 3 and 4, patients performed pulmonary function testing and a symptom-limited constant work-rate cycle exercise test whilst breathing room air (FiO2 = 0.21) or hyperoxia (FiO2 = 0.60), 44 in randomized order.  Gases were delivered into a non-diffusing Douglas bag connected to a two-way non-rebreathing valve via large bore tubing (Hans Rudolph, Inc., Shawnee, KS, USA).  The gas delivery system was identical for both conditions and patients were blinded to the inspired gas condition.  Data from Visits 3 and 4 were used to address the secondary aim of the study. Pulmonary function.  Patients performed spirometry, plethysmography, 12 second maximal voluntary ventilation, and single-breath diffusing capacity of the lungs for carbon monoxide (DLCO) according to established guidelines (Macintyre et al., 2005; Miller et al., 2005; Wanger et al., 2005).  Values were measured using a commercially available system (Vmax Encore 229, V62J Autobox; Carefusion, Yorba Linda, CA, USA) and expressed as a percentage of predicted values (Gutierrez et al., 2004; Tan et al., 2011). Exercise testing protocol.  Exercise was conducted using an electronically-braked cycle ergometer (Ergoselect 200P; Ergoline, Bitz, Germany) at a self-selected cadence ≥60 revolutions per minute.  All tests started with 6 minutes of rest followed by a one minute warm-up of unloaded pedaling.  Incremental tests proceeded with 15-watt stepwise increases in work-rate, starting at 15 watts, every two minutes until symptom limitation.  Peak work-rate was defined as the highest work-rate sustained for at least 30 seconds.  Constant work-rate exercise tests proceeded with an immediate increase in work-rate to 75% of peak incremental work-rate, which was sustained until symptom limitation.  Iso-time was defined as the highest equivalent submaximal time achieved on both constant work-rate exercise tests performed by a given patient under both conditions.  No verbal encouragement was provided during constant work-rate exercise tests. Symptom evaluation.  Patients rated the intensity of “breathing discomfort” (dyspnoea) and “leg discomfort” at rest, every two minutes during exercise, and at peak exercise using the 45 modified Borg 0-10 category-ratio scale (Borg, 1982).  The scale’s endpoints were anchored such that 0 represented “no breathing/leg discomfort” and 10 represented “the most severe breathing/leg discomfort ever experienced or imagined.”  Cardiorespiratory responses to exercise.  Standard cardiorespiratory responses were measured on a breath-by-breath basis at rest and during exercise (Vmax Encore 229; Carefusion, Yorba Linda, CA, USA).  This metabolic cart has not been validated with an FiO2 of 0.60; therefore, differences in gas exchange parameters (i.e., oxygen consumption and carbon dioxide production) and gas exchange-derived parameters (i.e., ventilatory equivalents and the respiratory exchange ratio) were not compared between conditions (Visits 3 and 4).  Flow was calibrated using a three litre calibration syringe connected to the breathing apparatus and using the appropriate gas mixture from the Douglas bag at a wide range of flow rates.  The inflection point in tidal volume relative to minute ventilation (VT/V̇E) was determined for each patient by examining 30-second averaged data throughout incremental exercise (Visit 2).  The time at which the corresponding VT was achieved during Visits 3 and 4, if at all, was also identified using the 30-second averaged data from those visits.  Arterial oxygen saturation was estimated via finger pulse oximetry (Radical-7 Pulse CO Oximeter, Masimo Corporation, Irvine, CA, USA).  Heart rate was recorded using 12-lead electrocardiography.   Diaphragm electromyography.  Crural diaphragm electromyography was measured using a multi-pair oesophageal electrode catheter (Guangzhou Yinghui Medical Equipment Ltd, Guangzhou, China) and was used as a surrogate of NRD (Luo et al., 2008).   Diaphragm electromyography was collected, processed, and analyzed as previously described (Schaeffer et al., 2014).  Briefly, the catheter consists of five consecutive recording pairs and was passed through the nose and positioned based on the amplitude of the EMGdi signals obtained 46 simultaneously from these recording pairs during tidal breathing.  To minimize discomfort, a lidocaine hydrochloride non-aerosol spray (Lidodan™; Odan Laboratories, Montréal, QC, Canada) was administered in the nasal and pharyngeal passages.  Raw EMGdi signals were sampled at two kilohertz using a PowerLab 16/35 analog-to-digital converter (ADInstruments, Castle Hill, Australia), processed with a 60 hertz notch filter (Model 08-GL biological Amplifier, Guangzhou Yinghui Medical Equipment Ltd, Guangzhou, China) as well as a band-pass filter between 20 hertz and one kilohertz using LabChart Pro Version 7.3.7 software (ADInstruments, Castle Hill, Australia), and converted to a root mean square (RMS) using a time constant of 0.1 seconds.  The maximum RMS value for each inspired breath was manually selected between QRS complexes to avoid the influence of cardiac artifact on the EMG signal.  The electrode pair with the largest amplitude for a given breath was used for the analysis.  Data were reported as a percentage of maximal EMGdi, which was defined as the highest root mean square value recorded during an inspiratory capacity manoeuvre at rest or during exercise.  We used inspiratory capacity manoeuvres exclusively instead of using multiple maximal inspiratory manoeuvres (e.g., sniffs and maximal inspiratory pressure manoeuvres) to obtain maximal EMGdi.  This was done to standardize the normalization procedure across all subjects and because our previous work shows that inspiratory capacity manoeuvres always results in the largest EMGdi activity (Ramsook et al., 2015).  NMU was defined as the ratio of EMGdi expressed as a percentage of maximum to VT expressed as a percentage of VC (i.e., EMGdi (%max):VT (%VC)). Statistical analyses.  Exercise variables were measured continuously with mean values determined over a 30-second epoch linked to the time of dyspnoea assessment.  We determined the association of dyspnoea intensity with NMU and its individual components (i.e., EMGdi 47 (%max) and VT (%VC)), using an unadjusted repeated measures linear regression.  All of these predictors were then forced into a multivariate model, also using repeated measures analysis, to determine independent predictors.  Variables included in the multivariate model were assessed for multicollinearity.  Paired t-tests were used to compare outcomes between the room air and hyperoxic conditions at iso-time and peak exercise.  Spearman’s correlation coefficients were used to make intra-subject, between-treatment comparisons of dyspnoea intensity with select physiological variables during constant work-rate exercise at iso-time.  Data are presented as mean±SD or median (interquartile range) where appropriate.  Statistical significance was set at p<0.05.  Data were analyzed using STATA® 11.2 (StataCorp LP, College Station, TX, USA).  3.3 Results Participants.  Patient characteristics and peak exercise data are summarized in Table 3-1.  On average, patients had moderate physiological impairment as demonstrated by reduced forced vital capacity and diffusing capacity of the lungs for carbon monoxide, and reduced functional capacity as evidenced by reduced peak aerobic capacity and work-rate relative to predicted values (Blackie et al., 1989; Gutierrez et al., 2004; Tan et al., 2011).  The VT/V̇E inflection point occurred at 76±28 watts during incremental exercise.  None of the patients qualified for supplemental oxygen, at rest or with ambulation, according to regional criteria (Sandberg & Fleetham, 2013).  No cardiac abnormalities were noted during exercise testing. Associations with dyspnoea.  A total of 97 observations were obtained during the incremental cycle exercise test on Visit 2.  One patient was excluded from this analysis (Aim 1) due to lack of EMGdi data on Visit 2; the patient refused the catheter.  Dyspnoea intensity was significantly associated with NMU, EMGdi (%max), and VT (%VC) (all p<0.001, Figure 3-1).  48 With adjusted analysis, EMGdi (%max), NMU, and VT (%VC) were all still significantly associated with dyspnoea intensity (p<0.001, p=0.006, and p=0.02, respectively), with EMGdi having the strongest correlation (Figure 3-1). Effects of hyperoxia.  Sensory and physiological effects of hyperoxia during constant work-rate cycle exercise are summarized in Table 3-2.  Three patients were excluded from this analysis (Aim 2); these patients refused the catheter for at least one of these visits and/or did not complete both visits.  The average time between constant work-rate cycle exercise tests was 6±2 days.  The mean work-rate for the constant work-rate cycle exercise tests was 71±27 watts, and the median iso-time was 6.0 (4.0-10.0) minutes.  Constant work-rate cycle exercise endurance time increased significantly with hyperoxia compared to room air (24.2±11.1 vs. 9.6±8.0 minutes, p<0.001). Dyspnoea intensity was similar between room air and hyperoxia at peak exercise (p=0.38), but significantly reduced at iso-time with hyperoxia vs. room air (1.8±1.9 vs. 3.8±3.0 Borg 0-10 units, respectively, p=0.005; Table 3-2.  EMGdi was significantly reduced during constant work-rate cycle exercise at iso-time (p<0.001) and peak exercise (p=0.005) with hyperoxia vs. room air (Table 3-2).  NMU was significantly improved at iso-time with hyperoxia vs. room air (p=0.003, Table 3-2).  The between-treatment change in iso-time dyspnoea intensity during constant work-rate cycle exercise was only correlated with the change in exercise endurance time (r=-0.67, p=0.02) and NMU (r=0.63, p=0.04).  Additional Spearman’s correlation coefficients are shown in Table 3-3. During constant work-rate cycle exercise with room air, all eleven patients included in the analysis for Aim 2 achieved or surpassed a VT that corresponded to the VT/V̇E inflection point identified on the incremental test.  Two of these patients did not achieve this VT with 49 hyperoxia, and the time at which this VT was achieved was delayed with hyperoxia vs. room air in the remaining nine patients (5.5±5.1 vs. 1.9±0.7 minutes, respectively, p=0.06).  Measures of dyspnoea, EMGdi (%max), and NMU were not different at this measurement time between treatments (p=0.12, p=0.12, and p=0.09; respectively).  3.4 Discussion The main findings of this study were as follows: 1) dyspnoea intensity was significantly correlated with both EMGdi and NMU during incremental cycle exercise, but with a stronger correlation between dyspnoea and EMGdi; and 2) the improvements in dyspnoea intensity at iso-time during constant work-rate cycle exercise with hyperoxia were significantly correlated with the change in NMU, but not with the change in EMGdi. It has been proposed that increased dyspnoea is a reflection of increased levels of NRD in both health and disease (Jensen et al., 2016).  There is also evidence that central and peripheral afferent feedback pathways may neuromodulate the intensity of perceived dyspnoea through direct effects on NRD (Jensen et al., 2016).  Accordingly, increased dyspnoea intensity in patients with cardiopulmonary disease likely corresponds to the awareness of an increased NRD needed to meet an increased ventilatory demand in the context of increased mechanical loading and/or reduced capacity of the respiratory muscles.  For example, patients with ILD demonstrate reduced compliance and increased static recoil pressure of the lung due to fibrosis.  The static pressure-volume curve of the lungs is therefore shifted downward and to the right (Yernault et al., 1975; Gibson & Pride, 1977).  The pressure-volume relationship of the entire respiratory system is therefore contracted along its volume axis (O'Donnell et al., 2009), whereby resting inspiratory capacity and inspiratory reserve volume are generally diminished.  This reduction in 50 baseline reserve translates to less potential for VT expansion during exercise (Jones & Rebuck, 1979; Gowda et al., 1990).  The ILD patients in the present study achieved critical constraints on VT expansion at markedly lower absolute work-rates than what has been previously documented in healthy men and women of similar ages (Guenette et al., 2011).  Thus, ILD patients are likely to have increased elastic loading of the respiratory muscles (O'Donnell et al., 2009), which necessitates greater respiratory muscle force for inspiration (O'Donnell et al., 1998a), and therefore an increased NRD (Faisal et al., 2016), in order to meet ventilatory requirements. Faisal et al. (2016) recently showed that the relationship between EMGdi and dyspnoea intensity is similar in patients with ILD and COPD, as well as healthy controls.  This supports the hypothesis that NRD is an important component of the physiological basis of increased dyspnoea intensity, as disease-specific differences in respiratory function and/or capacity, and therefore afferent feedback pathways, do not influence this association (Jensen et al., 2016).  Indeed, another recent study in COPD patients (Jolley et al., 2015) examined the relationships of dyspnoea intensity with NRD and NMU, concluding that dyspnoea intensity is closely related to EMGdi and NMU, but that the relationship between dyspnoea intensity and EMGdi was comparatively stronger.  This is consistent with our findings, suggesting that increased NRD largely contributes to increased dyspnoea intensity in patients with fibrotic ILD, at least during normoxic incremental exercise. We recently demonstrated significant reductions in iso-time dyspnoea intensity during constant work-rate exercise with hyperoxia compared to room air (Schaeffer et al., 2017) (see Chapter 2).  However, there were no signification correlations between the reduction in dyspnoea intensity and the between-treatment changes in standard ventilatory measures during exercise.  We reasoned that a reduction in dyspnoea intensity with hyperoxia may therefore be associated 51 with improvements in neurophysiological parameters such as NRD and/or NMU.  However, despite EMGdi being more significantly correlated with dyspnoea intensity during normoxic incremental exercise, the improvement in dyspnoea intensity with hyperoxia was only significantly correlated with an improvement in NMU and exercise performance (i.e., endurance time).  This suggests that hyperoxia reduces dyspnoea via multiple mechanisms, with a combined beneficial effect on both NRD and respiratory mechanics. Previous studies have shown that hyperoxia improves oxygen delivery to the skeletal muscles by widening the arteriovenous oxygen content difference, thereby increasing the anaerobic threshold and delaying the onset of metabolic acidosis and skeletal muscle fatigue (Powers et al., 1989; Harris-Eze et al., 1994).  The subsequent reduction in V̇E at iso-time and peak exercise in the present study may have reflected, at least in part, both a reduction in EMGdi as well as VT, with attendant improvements in perceived breathlessness.  This is further supported by the delayed onset of the VT/V̇E inflection with hyperoxia compared to room air.  NMU was defined in this study as the ratio of EMGdi (%max) to VT (%VC), and therefore a direct representation of the interaction between EMGdi and VT.  In this context, had only EMGdi been reduced and VT held constant with hyperoxia, NMU would have decreased (i.e., improved).  Alternatively, had only VT been reduced and EMGdi held constant with hyperoxia, NMU would have increased (i.e., worsened).  We observed reductions in both EMGdi and VT with hyperoxia compared to room air, with a relatively larger reduction in EMGdi (26±11 vs. 12±6 %, respectively).  Accordingly, our study suggests that hyperoxia reduces dyspnoea mainly via an improvement in NMU, which reflects an optimized relationship between EMGdi and VT.  The interaction between the improvement in NRD and the change in VT seems to have a more important effect than the improvement in NRD alone. 52 Limitations.  There are some limitations to this study that must be acknowledged.  First, we used EMGdi as a surrogate of NRD and the limitations of this method have been well described (Faisal et al., 2016).  Second, while the diaphragm is the primary inspiratory muscle, we cannot ignore the potential contribution of extra-diaphragmatic inspiratory muscles, such as the sternocleidomastoids, scalenes, and intercostal muscles.  Changes in the ventilatory response with hyperoxia could elicit changes in respiratory muscle recruitment patterns and their relative contributions, which could influence measures of NMU.  Third, due to safety reasons, it was not feasible to blind all study personnel to the study condition.  However, we do not believe this impacted our results as all test procedures were rigorously standardized, including an absence of verbal encouragement for all constant work-rate exercise tests.  Fourth, we acknowledge that this invasive study had a small sample size, and therefore may not be representative of all patients with fibrotic ILD.  Finally, we included a variety of ILD subtypes.  However, all patients had isolated lung involvement.  Due to the similar physiology of these diseases, we expect the physiological findings to be similar across these subtypes. Conclusion.  Our data suggest that the intensity of dyspnoea more likely reflects increased NRD, as estimated by EMGdi, rather than NMU, in patients with ILD during normoxic incremental exercise.  However, attenuation of dyspnoea intensity with hyperoxia was correlated with improvements in NMU, but not EMGdi.  This suggests that the combined benefits of hyperoxia on the coupling between NRD and the mechanical output of the respiratory system are more important in relieving dyspnoea than the effects on NRD alone.  Future research is needed to better isolate the independent effects of EMGdi and the mechanical output of the respiratory system (i.e., VT) on dyspnoea during exercise in patients with ILD.  This knowledge may aid in the development of symptom-based treatment options in this population. 53 Table 3-1. Patient characteristics and maximal exercise data Descriptive characteristics    Age 66 ± 10 Male:Female 10 : 4 BMI, kg/m2 29 ± 5 Diagnosis, number of patients    IPF 7   HP 2    Idiopathic LIP 1   Polymyositis-associated ILD* 1   Scleroderma-associated ILD* 1   Unclassifiable ILD 1   Undifferentiated connective tissue disease* 1   Time since diagnosis, months 40.1 (31.1-64.3) Surgical lung biopsy, number of patients 7   Modified MRC Dyspnoea Scale 1.5 ± 1.2 Oxygen Cost Diagram, mm 62 ± 17 Pulmonary Function    FEV1, %predicted 77 ± 15 FVC, %predicted 71 ± 16 FEV1/FVC, % 80 ± 8 TLC, %predicted 64 ± 11 DLCO, %predicted 45 ± 15 Peak incremental exercise    Dyspnoea, 0-10 Borg scale 5.7 ± 2.3 Leg discomfort, 0-10 Borg scale 5.6 ± 2.1 V̇O2, ml/kg/min 21.1 ± 6.2 V̇O2, % predicted 70 ± 23 RER 1.08 ± 0.08 Work-rate, watts 93 ± 39 Work-rate, %predicted 59 ± 17 Heart rate, bpm 132 ± 20 Heart rate, %predicted 87 ± 10 V̇E, l/min 76 ± 20 V̇E/MVV, % 72 ± 23 SpO2, % 90 ± 5 EMGdi, %max 69 ± 15 NMU, EMGdi (%max):VT (%VC) 1.4  ± 0.4 Values are mean±SD or median (interquartile range).   *, no extra-pulmonary manifestations; BMI, body mass index; DLCO, diffusing capacity of the lungs for carbon monoxide; EMGdi, electromyogram of the crural diaphragm; FEV1, forced expiratory volume in 1 second; FVC, forced vital capacity; HP, hypersensitivity pneumonitis; IPF, idiopathic pulmonary fibrosis; LIP; lymphoid interstitial pneumonia; MRC, Medical Research Council; MVV, maximal voluntary ventilation; NMU, neuromechanical uncoupling; RER, respiratory exchange ratio; SpO2, arterial oxygen saturation; TLC, total lung capacity; VC, vital capacity; V̇E, minute ventilation; V̇O2, oxygen consumption; VT, tidal volume. 54 Table 3-2. Selected sensory and physiological parameters at iso-time and peak exercise for constant work-rate cycle exercise tests with room air and hyperoxia  Iso-time Peak  Room Air Hyperoxia Room Air Hyperoxia Dyspnoea, 0-10 Borg scale 3.8 ± 3.0 1.8 ± 1.9† 5.0 ± 2.7 4.5 ± 3.7 Leg discomfort, 0-10 Borg scale 3.4 ± 2.1 2.1 ± 1.5 4.7 ± 1.8 5.9 ± 2.2 SpO2, % 89 ± 3 98 ± 2‡ 88 ± 4 98 ± 1‡ Heart rate, beats/min 125 ± 22 114 ± 19‡ 127 ± 21 125 ± 22 Heart rate, %predicted 83 ± 14 76 ± 13‡ 85 ± 13 83 ± 15 V̇E, l/min 74 ± 18 53 ± 14‡ 76 ± 20 62 ± 13† V̇E/MVV, % 71 ± 26 50 ± 18‡ 73 ± 26 59 ± 20† VT, l 1.58 ± 0.48 1.39 ± 0.44‡ 1.50 ± 0.46 1.29 ± 0.36† FB, breaths/min 49 ± 15 40 ± 12† 54 ± 19 50 ± 15 EELV, %TLC 55 ± 5 56 ± 7 54 ± 7 57 ± 7 EILV, %TLC 93 ± 5 89 ± 5* 91 ± 6 88 ± 6* PETCO2, mmHg 33 ± 6 39 ± 6‡ 32 ± 7 37 ± 7‡ EMGdi, %max 67 ± 16 49 ± 12‡ 68 ± 18 53 ± 13† NMU, EMGdi (%max):VT (%VC) 1.3 ± 0.3 1.0 ± 0.2† 1.3 ± 0.4 1.2 ± 0.2 Values are mean±SD.   EELV, end-expiratory lung volume; EILV, end-inspiratory lung volume; EMGdi, electromyography of the crural diaphragm; FB, breathing frequency; MVV, maximal voluntary ventilation; NMU, neuromechanical uncoupling; PETCO2, partial pressure of end tidal carbon dioxide; SpO2, oxygen saturation by pulse oximetry; TLC, total lung capacity; VC, vital capacity; V̇E, minute ventilation; VT, tidal volume.  Significantly different from room air: *, p<0.05; †, p<0.01; ‡, p<0.001.    55 Table 3-3.  Spearman’s correlations for selected parameters with the change in dyspnoea intensity at iso-time with hyperoxia vs. room air  r p        ∆EET, min -0.67 0.02     FVC, %predicted -0.43 0.18     DLCO, %predicted -0.11 0.75     Peak incremental V̇O2, %predicted 0.11 0.75     ∆SpO2 at iso-time, % -0.04 0.91     ∆SpO2 at peak exercise, % -0.06 0.86     ∆V̇E at iso-time, l/min 0.14 0.68     ∆VT at iso-time, l -0.46 0.16     ∆FB at iso-time, breaths/min 0.23 0.50     ∆EELV at iso-time, %TLC 0.37 0.26     ∆EILV at iso-time, %TLC 0.27 0.43        ∆EMGdi, %max 0.41 0.21     ∆NMU, EMGdi (%max):VT (%VC) 0.63 0.04     ∆, difference between selected parameter with hyperoxia vs. room air; DLCO, diffusing capacity of the lungs for carbon monoxide; EELV, end-expiratory lung volume; EET, exercise endurance time; EILV, end-inspiratory lung volume; FB, breathing frequency; FVC, forced vital capacity; SpO2, oxygen saturation by pulse oximetry; V̇E, minute ventilation; V̇O2, oxygen consumption; VT, tidal volume.            56   Figure 3-1. Trend lines for individual patients demonstrating the relationship between dyspnoea intensity and (A) electromyography of the diaphragm (EMGdi), (B) neuromechanical uncoupling, and (C) tidal volume (VT) expressed as a percentage of vital capacity (VC) Reported r values are expressed as median (interquartile range) from Spearman correlations within each patient. 0 1 2 3 4 5 6 7 8 9 10 0 20 40 60 80 Dyspnoea (0-10 Borg scale) VT (%VC) 0 1 2 3 4 5 6 7 8 9 10 0.0 0.5 1.0 1.5 2.0 2.5 Dyspnoea (0-10 Borg scale) EMGdi (%max):VT (%VC) 0 1 2 3 4 5 6 7 8 9 10 0 20 40 60 80 100 Dyspnoea (0-10 Borg scale) EMGdi (%max) r=0.93(0.85-0.94) r=0.81(0.72-0.91) r=0.83(0.73-0.88) B A C 57 Chapter 4: Qualitative dimensions of exertional dyspnoea in fibrotic interstitial lung disease  4.1 Introduction There is increasing evidence that humans can reliably distinguish varying intensities and qualities of dyspnoea (Parshall et al., 2012a; Parshall et al., 2012b; Parshall et al., 2012c; Laviolette et al., 2014).  Healthy individuals predominantly report a steady worsening of dyspnoea intensity that is typically described as increased work/effort of breathing throughout incremental cycle exercise (Cory et al., 2015).  In patients with COPD, there is also a progressive increase in dyspnoea intensity during incremental exercise, but the predominant descriptor of dyspnoea changes from increased work/effort to unsatisfied inspiration, particularly at the VT/V̇E inflection (O'Donnell et al., 2006; Laveneziana et al., 2011).  At this point, IRV is reduced such that any further VT expansion results in substantial increases in respiratory muscle work (O'Donnell et al., 2006).  This ventilatory pattern and associated sensory consequences have been observed during both incremental and constant-load cycle exercise in patients with COPD (Laveneziana et al., 2011).  However, the relationship between constraints on VT expansion and dyspnoea quality has not been evaluated in patients with ILD. Recent studies have shown that patients with COPD and ILD have similar physiological mechanisms of exertional dyspnoea intensity (Faisal et al., 2016).  Accordingly, the physiological mechanisms responsible for the more frequent selection of unsatisfied inspiration in COPD and ILD may also be similar (Scano et al., 2010).  The purpose of this study was to examine the progression of dyspnoea quality during exercise and its relationship to ventilatory constraints in patients with fibrotic ILD. 58 4.2 Methods Participants.  Patients with mild-to-moderate fibrotic ILD were recruited from two specialized ILD clinics, including some patients who concurrently participated in two previously published studies (Schaeffer et al., 2017, 2018).  All patients provided informed written consent prior to enrollment (University of British Columbia Providence Health Care Research Ethics Board H13-00059).  Patients with fibrotic ILD and an arterial oxygen saturation (SpO2) ≥92% by finger pulse oximetry at rest while breathing room air were eligible.  Patients were excluded if they had concurrent participation in or recent (<6 weeks) completion of pulmonary rehabilitation, other diseases that could impair exercise capacity and/or SpO2, significant emphysema, pulmonary hypertension, and/or the use of prednisone >10mg/day for at least two weeks within three months of enrollment.  Four additional patients were excluded from analysis because they did not complete all three visits and/or did not have an identifiable VT/VE inflection on both study tests. Study design.  During Visit 1, patients completed chronic activity-related dyspnoea questionnaires (modified Medical Research Council scale and Oxygen Cost Diagram (Mahler & Wells, 1988)) followed by pulmonary function testing and a symptom-limited incremental cycle exercise test with dyspnoea assessment for familiarization purposes.  During Visit 2, patients performed another symptom-limited incremental cycle exercise test.  During Visit 3, patients performed a symptom-limited constant-load cycle exercise test at 75% of the peak work-rate determined at Visit 2. Pulmonary function.  Spirometry, whole-body plethysmography, maximal voluntary ventilation, single-breath diffusing capacity of the lungs for carbon monoxide (DLCO), as well as maximal inspiratory and expiratory pressures were measured using a commercially available 59 cardiopulmonary testing system (Vmax Encore 229, V62J Autobox; Carefusion, Yorba Linda, CA, USA).  Values were expressed as percentages of predicted (Knudson et al., 1983; Morris et al., 1988; Gutierrez et al., 2004; Tan et al., 2011). Exercise testing protocol.  All exercise tests were conducted on an electronically-braked cycle ergometer (Ergoselect 200P; Ergoline, Bitz, DE) after six minutes of quiet breathing and one minute of unloaded pedaling.  Incremental exercise tests started at 15 watts with 15-watt stepwise increases in work-rate every two minutes until volitional exhaustion.  Peak work-rate was defined as the highest work-rate sustained for ≥30 seconds.  Constant work-rate exercise tests started with an immediate increase to 75% of peak incremental work-rate until volitional exhaustion.  Patients self-selected a cadence ≥60 revolutions per minute for all tests. Symptom evaluation.  Patients rated the intensity of “breathing discomfort” (dyspnoea) and “leg discomfort” at rest, every two minutes throughout exercise, and at peak exercise using the Borg 0-10 category-ratio scale (Borg, 1982).  Patients were also asked to select the phrase that best described their breathing at that moment from a previously-described list of four items: “My breathing requires more work and effort” (work/effort); “I cannot get enough air in” (unsatisfied inspiration); “I cannot get enough air out” (unsatisfied expiration); and “None apply” (Laveneziana et al., 2011).  Patients were allowed to select multiple phrases if equally applicable.  Immediately after exercise, patients selected all descriptors that applied to how their breathing felt at the end of exercise from a standardized list consisting of 15 qualitative descriptors (O'Donnell et al., 2000).  Patient responses were grouped into 10 clusters corresponding to distinct experiences of breathlessness evoked by different respiratory stimuli, as previously described (Simon et al., 1990; Mahler et al., 1996; O'Donnell et al., 2000).  However, if more than one applied to the same cluster, only one rating was attributed to that cluster (i.e., 60 the rating for the same cluster was not counted twice).  Patients were also asked to identify the three best descriptors of their breathing at the end of exercise from this list if they selected more than three in total. Cardiorespiratory responses to exercise.  Metabolic and ventilatory responses were measured breath-by-breath and averaged over 30-second epochs (Vmax Encore 229; Carefusion, Yorba Linda, CA, USA).  Arterial oxygen saturation was estimated using finger pulse oximetry (Radical-7 Pulse CO Oximeter, Masimo Corporation, Irvine, CA, USA) and heart rate was monitored by 12-lead electrocardiography (Cardiosoft Diagnostics System v6.71, GE Healthcare, CA).  Operating lung volumes were derived from serial dynamic IC maneuvers (Guenette et al., 2013).  The VT/VE inflection was determined for each patient by visual examination of individual Hey plots by three different observers from incremental exercise at Visit 2 (Hey et al., 1966); if multiple inflection points were present, the selection made by the majority was chosen. Statistical analyses.  A sample size of 16 was chosen based on previous literature (Laveneziana et al., 2011).  Data are presented as mean ± SD or median (interquartile range) unless otherwise specified.  Paired t-tests were used to compare outcomes between the incremental and constant work-rate exercise tests at rest, VT/VE, and peak exercise.  Repeated measures linear regression was used to evaluate between-protocol differences in dyspnoea intensity throughout exercise, and to determine the association between dyspnoea and each of VT, VE, IRV, and SpO2.  Between-protocol differences in qualitative descriptors of dyspnoea at peak exercise were assessed using a McNemar’s Exact Test.  Unpaired t-tests were used to assess differences between patients that selected unsatisfied inspiration at any point during exercise on either test protocol and those that never selected unsatisfied inspiration. Analyses were 61 performed using IBM® SPSS® Statistics for Macintosh, version 25.0 (IBM Corp., Armonk, NY, USA) and STATA® 11.2 (StataCorp LP, College Station, TX, USA).  4.3 Results Participants.  Characteristics of the 16 included patients are summarized in Table 4-1.  Patients had mild reduction in forced vital capacity (76±15 %predicted) and moderate reduction in DLCO (48±14 %predicted).  None of the patients that participated in this study met the regional criteria for home oxygen supplementation (Sandberg & Fleetham, 2013).  Eight patients had IPF, with no significant difference in baseline pulmonary function or functional capacity compared to patients with non-IPF fibrotic ILD (Table 4-2). Physiological responses to exercise.  Physiological responses to exercise are summarized in Table 4-3.  Patients had reduced peak aerobic capacity compared to predicted values.  Breathing pattern and operating lung volumes relative to V̇E were similar for incremental and constant work-rate exercise (Figure 4-1).  The VT/V̇E inflection occurred sooner during constant work-rate exercise compared to incremental exercise (2.5±1.8 vs. 9.0±3.4 min, p<0.001).  Beyond the VT/V̇E inflection, further increases in V̇E were achieved via increases in breathing frequency (Figure 4-1). Dyspnoea intensity.  On average there were 8±3 dyspnoea intensity ratings during incremental exercise and 8±5 ratings during constant work-rate exercise.  Baseline dyspnoea intensity was above zero in three patients for incremental exercise and one patient for constant work-rate exercise.  There were strong associations between dyspnoea intensity and each of V̇E, VT, and IRV in both exercise protocols (all p <0.001).  The relationship of dyspnoea intensity with both VT and IRV was biphasic (Figure 4-2), with dyspnoea intensity gradually increasing 62 until the VT/V̇E inflection, after which there was a sharp increase in dyspnoea intensity until peak exercise.  Median dyspnoea intensity at peak exercise was 6 Borg units (range 4-7) for incremental and 6 Borg units (3-7) for constant work-rate exercise tests, respectively.  Dyspnoea intensity during incremental exercise was associated with IRV independently of SpO2 (p=0.002). Dyspnoea quality.  Increased work/effort accounted for 50% and 79% of all selections for the description of dyspnoea during incremental and constant work-rate exercise, respectively (Figure 4-3).  Increased work/effort was selected at least once by all 16 patients (100%) during incremental exercise and 15 patients (94%) during constant work-rate exercise.  Selection frequency of work/effort increased proportionally to increasing dyspnoea intensity up until the VT/V̇E inflection, after which the response appeared to plateau (Figure 4-4A).    Unsatisfied inspiration was selected at least once by eight patients (50%) during incremental exercise, ten patients (63%) during constant work-rate exercise, and seven patients (44%) on both exercise tests.  Selection frequency of unsatisfied inspiration increased after the VT/V̇E inflection during both incremental and constant work-rate exercise (Figure 4-3).  Unsatisfied expiration had the lowest selection frequency of any descriptor throughout exercise, which was never selected during incremental exercise and selected by four patients (25%) accounting for 7% of all selections during constant work-rate exercise (Figure 4-3).  None apply accounted for 38% and 4% of all selections for the description of dyspnoea during incremental and constant work-rate exercise, respectively.  However, this option was selected early during exercise and no patient selected this descriptor at symptom limitation.  Selection frequency of unsatisfied inspiration increased proportionally to increasing dyspnoea intensity throughout exercise (Figure 4-4B). 63 Selection frequencies of dyspnoea descriptors derived from the 15-item list administered at the end of exercise were similar between exercise protocols (all p>0.05; Figure 4-5).  The three best descriptors at peak incremental exercise corresponded to “unsatisfied inspiration” (50% of patients made this selection), “work” (44%), and air “hunger” (38%), respectively.  The three best descriptors at peak constant work-rate exercise corresponded to “work” (62%), “unsatisfied inspiration” (56%), and “inspiratory difficulty” (44%). Sub-group analysis.  There were no statistically significant differences in resting pulmonary function between patients that selected unsatisfied inspiration (“selectors”, n=11) at any point during exercise compared to patients that never made this selection (“non-selectors,” n=5).  Breathing pattern was similar between selectors and non-selectors (Figure 4-6A and B).  However, our sample size precluded a rigorous statistical comparison of these sub-groups.  The mean IRV at the VT/V̇E inflection tended to be lower in the selectors compared to the non-selectors (0.3±0.2 vs. 0.6±0.4 liters, respectively), while selectors had a higher VT/IC at baseline and throughout exercise (Figure 4-6C).  4.4 Discussion The main findings of this study are: (1) Ventilatory and perceptual responses to exercise were similar between incremental and constant work-rate protocols in patients with fibrotic ILD, and were likely dictated by mechanical constraints on VT expansion; (2) perceived work/effort of breathing was the dominant qualitative descriptor of dyspnoea throughout both incremental and constant work-rate exercise; and (3) selection frequency of unsatisfied inspiration markedly increased at the VT/V̇E inflection, independent of cycle protocol.  Collectively, these findings have important implications for targeted symptom management of patients with fibrotic ILD. 64 Ventilatory response and constraints.  Breathing pattern and operating lung volumes relative to VE were similar in the two exercise protocols.  However, there were, not surprisingly, greater metabolic and therefore ventilatory requirements at the beginning of constant work-rate compared to incremental exercise.  This resulted in constraints on VT expansion being reached significantly sooner during the constant work-rate protocol.  Despite this difference, the relationships between dyspnoea intensity and each of V̇E, VT, and IRV during exercise were preserved across protocols, with all showing a similar biphasic response.  Accordingly, dyspnoea intensity ratings increased linearly to only moderate levels prior to the VT/V̇E inflection, after which dyspnoea intensity ratings rose steeply to severe levels.  These data suggest that breathing pattern, dynamic operating lung volumes, and perceived dyspnoea intensity were a consequence of an increased ventilatory requirement in the setting of mechanical constraints on VT expansion in patients with fibrotic ILD, with these changes occurring independently of protocol, the rate of increase in V̇E from rest, and oxygenation.  Our findings are similar to previously published literature in COPD (Laveneziana et al., 2011), despite different mechanisms through which mechanical constraints on VT expansion occurs in these diseases. Work/effort and unsatisfied inspiration.  Work/effort and unsatisfied inspiration occurred independently of one another during both incremental and constant work-rate exercise.  Work/effort was the dominant descriptor throughout exercise for both protocols, but selection frequency plateaued despite both increasing ventilation and dyspnoea intensity after the VT/V̇E inflection.  On the contrary, selection frequency of unsatisfied inspiration increased in direct proportion to increasing dyspnoea intensity throughout exercise across protocols.  While there was a marked increase in the selection of unsatisfied inspiration after the VT/V̇E inflection for both protocols, the percentage of subjects that identified unsatisfied inspiration as the primary 65 descriptor of dyspnoea during exercise never surpassed that of work/effort.  This differs from previous observations in COPD where after the VT/V̇E inflection unsatisfied inspiration became the dominant descriptor of dyspnoea during exercise (Laveneziana et al., 2011).  It has been suggested that unsatisfied inspiration is a reflection of increased neuromechanical uncoupling (i.e., a disproportionate increase in neural respiratory drive relative to VT expansion) (O'Donnell et al., 1998a).  Patients with ILD and COPD have greater neuromechanical uncoupling during exercise compared to age-matched healthy controls, which can largely be attributed to a reduced ventilatory reserve (Faisal et al., 2016).  As demonstrated in our study, end-expiratory lung volume slightly decreased during exercise in patients with ILD (-0.19 and -0.21 litres) from rest to peak during incremental and constant work-rate exercise, respectively), while end-inspiratory lung volume increases up to the point of the VT/V̇E inflection.  At this point, patients are likely breathing on the upper less-compliant portion of the respiratory system’s pressure-volume relationship where greater neural respiratory drive is needed to generate a given level of inspiratory pressure.  The awareness of this mismatch is conveyed by afferent feedback from respiratory muscle mechanoreceptors and may form the psychophysical basis of unsatisfied inspiration in ILD (O'Donnell et al., 1998a).   Descriptors of dyspnoea at peak incremental and constant work-rate exercise in the present study were selected via a 15-item list.  The two most common descriptors identified were “work” and “unsatisfied inspiration,” with the former having a higher selection frequency compared to the latter.  The observed selection of dyspnoea descriptors at the end of exercise may be related to increased neural respiratory drive, central corollary discharge, and/or afferent feedback from the respiratory muscles.  Future studies are needed to explore these parameters as potential targets for therapeutic interventions seeking to reduce the unpleasant sensation of 66 unsatisfied inspiration in these patients.  Additionally, it is important to note that patients were asked to select the predominant descriptor of dyspnoea during exercise from a list confined to three descriptors: work/effort, unsatisfied inspiration, and unsatisfied expiration.  Thus, we cannot evaluate whether other sensations were important or absent during exercise. Unsatisfied inspiration and inspiratory reserve volume.  Previous studies in COPD and asthma have demonstrated that the VT/V̇E inflection, in the setting of a pathologically reduced IRV, corresponds to an increased perception of unsatisfied inspiration (Laveneziana et al., 2011; Laveneziana et al., 2013), and our results suggest that IRV is similarly important in patients with ILD.  Patients in the present study who selected unsatisfied inspiration at any point during incremental or constant work-rate exercise reached a mean minimal IRV of 0.3 litres at the VT-/V̇E inflection, whereas those that never made this selection failed to reach this threshold and had a higher mean minimal IRV of 0.6 litres.  In comparison, the critical, minimal IRV appears to be 0.5 litres in both COPD and asthma (O'Donnell et al., 2006; Laveneziana et al., 2013).  A larger sample size is needed to determine if this threshold is both similar in patients with ILD and independent of disease severity.  However, our data suggest that the perception and corresponding sensory consequence of this mechanical event is likely not disease specific, with increased perception of unsatisfied inspiration as a common outcome in patients that reach the VT/V̇E inflection.    We did not observe any statistically significant differences in resting pulmonary function between selectors of unsatisfied inspiration and non-selectors; however, selectors tended to have less pulmonary restriction and worse gas exchange.  Future studies with larger sample sizes are necessary to determine whether resting pulmonary function can predict perceived dyspnoea quality during exercise.  Patients that selected unsatisfied inspiration adopted breathing patterns 67 that further encroached on their ventilatory capacity, as evidenced by a consistently higher VT/IC throughout incremental exercise.  This is supported by previous work in ILD that identified VT/IC as the dominant independent contributor to dyspnoea intensity and inspiratory difficulty (O'Donnell et al., 1998a).  However, we were limited in our ability to make strong comparisons between patients that selected unsatisfied inspiration and those that did not; therefore, we must be cautious in making any generalizations to the larger ILD population. Clinical implications.  Based upon our findings, interventions that reduce ventilatory requirements and therefore delay the time required to reach the VT/V̇E inflection or even prevent reaching it altogether are likely to reduce dyspnoea intensity and the perception of unsatisfied inspiration in patients with ILD, which may have implications in a rehabilitation setting for increasing exercise tolerance.  For example, hyperoxia improves exercise tolerance, and dyspnoea intensity during an acute bout of exercise in patients with fibrotic ILD (Schaeffer et al., 2017).  Hyperoxia also results in a more favourable dyspnoea quality that may translate into better adherence to a rehabilitation program.  This possibility is currently being explored in a large multi-centre clinical trial (Ryerson et al., 2016).  Additionally, the perceived quality of dyspnoea could be used to gauge whether or not a critical IRV has been reached during exercise to help monitor exercise intensity during a training session, making it symptomatically more tolerable.  These implications are particularly important given the absence of effective symptom-modifying pharmacotherapy for most subtypes of fibrotic ILD. Conclusions.  The VT/V̇E inflection is an important ventilatory-mechanical event that marks a substantial increase in dyspnoea intensity and a change in dyspnoea quality during exercise in patients with fibrotic ILD, regardless of exercise protocol.  Furthermore, the VT/V̇E inflection in the setting of a reduced IRV resulted in an increased selection of unsatisfied 68 inspiration as the predominate descriptor of dyspnoea.  These findings may inform the management of exertional symptoms in patients with fibrotic ILD. 69 Table 4-1. Patient characteristics  All patients  Selectors Non-selectors Descriptive characteristics          Age 65 ± 9 66 ± 7 61 ± 13 Male : Female 12 : 4 7 : 4 5 : 0 BMI, kg/m2 29 ± 5 28 ± 5 33 ± 5 Diagnosis, number of patients          IPF 8   6   2   Non-IPF 8    5   3   Time since diagnosis, months 37 (27-46) 40 (27-51) 32 (30-40) CPI 47 ± 11 48 ± 10 45 ± 14 Modified MRC Dyspnoea Scale 1.3 ± 1.1 1.4 ± 1.2 1.0 ± 0.7 Oxygen Cost Diagram, mm 64 ± 15 62 ± 17 68 ± 10 Pulmonary Function          FVC, %predicted 71 (66-86) 81 (66-94) 67 (66-71) FEV1, %predicted 77 (71-87) 77 (75-98) 72 (65-74) FEV1/FVC, % 80 (78-82) 80 (78-82) 80 (79-81) TLC, %predicted 67 (59-75) 67 (61-75) 62 (59-73) DLCO, %predicted 47 (39-57) 46 (42-53) 60 (35-64) MVV, l 114 ± 26 112 ± 26 119 ± 29 Values are mean ± SD or median (interquartile range).   BMI, body mass index; CPI, composite physiologic index; DLCO, diffusing capacity of the lungs for carbon monoxide; FEV1, forced expiratory volume in 1 second; FVC, forced vital capacity; IPF, idiopathic pulmonary fibrosis; MRC, Medical Research Council; Non-selectors, patients that did not select unsatisfied inspiration at any point during exercise; Selectors, patients that selected unsatisfied inspiration at any point during exercise; TLC, total lung capacity.   70 Table 4-2. IPF vs. non-IPF patient characteristics  IPF Non-IPF  Descriptive characteristics          Age 69 ± 8 60 ± 8    Male : Female 8 : 0 4 : 4    BMI, kg/m2 29 ± 5 30 ± 5    Time since diagnosis, months 34.2 (25.1-20.8) 39.7 (32.0-66.5)  CPI 51.1 ± 12.5 43.0 ± 7.9    Modified MRC Dyspnoea Scale 1.8 ± 1.3 0.8 ± 0.5    Oxygen Cost Diagram, mm 58 ± 18 70 ± 10    Pulmonary Function          FVC, %predicted 73 (63-92) 71 (67-82)    FEV1, %predicted 82 (69-107) 75 (72-78)    FEV1/FVC, % 80 (79-82) 79 (71-82)    TLC, %predicted 63 (56-75) 68 (63-75)    DLCO, %predicted 43 (32-52) 52 (44-61)    Functional capacity          Peak incremental work-rate, %predicted 59 ± 15 63 ± 18    Peak incremental V̇O2, %predicted 78 ± 21 63 ± 18    Values are mean ± SD or median (interquartile range).   BMI, body mass index; CPI, composite physiologic index; DLCO, diffusing capacity of the lungs for carbon monoxide; FEV1, forced expiratory volume in 1 second; FVC, forced vital capacity; IPF, idiopathic pulmonary fibrosis; MRC, Medical Research Council; TLC, total lung capacity.   71 Table 4-3. Selected sensory and physiological parameters at the VT/V̇E inflection and peak exercise for incremental and constant work-rate exercise tests   VT/V̇E Inflection   Peak   INCR CWR INCR CWR Exercise time, min 9.0 ± 3.4 2.5 ± 1.8‡ 12.3 ± 4.9 13.1 ± 10.6 Work-rate, watts 74 ± 26 73 ± 26 97 ± 35 73 ± 26‡ Dyspnoea, 0-10 Borg scale 2.7 ± 2.0 1.7 ± 2.2 5.7 ± 2.2 5.2 ± 2.9 Leg discomfort, 0-10 Borg scale 2.6 ± 1.8 1.6 ± 1.5 5.5 ± 1.9 5.6 ± 2.2 SpO2, % 92 ± 4 93 ± 3 90 ± 5 90 ± 4 Heart rate, bpm 116 ± 19 112 ± 15 131 ± 20 128 ± 18 Heart rate, %predicted 76 ± 12 74 ± 11 86 ± 11 84 ± 11 V̇O2, l/min 1.5 ± 0.4 1.6 ± 0.3 1.8 ± 0.6 1.9 ± 0.5 V̇O2, ml/kg/min 17.2 ± 4.8 18.4 ± 5.2 20.9 ± 6.6 22.2 ± 7.0 V̇O2, % predicted 58 ± 16 64 ± 20 70 ± 21 76 ± 23 RER 0.99 ± 0.07 1.00 ± 0.11 1.06 ± 0.08 1.03 ± 0.10 V̇E, l/min 54 ± 12 55 ± 15 73 ± 18 75 ± 22 V̇E/MVV, % 50 ± 19 51 ± 21 66 ± 21 68 ± 23 VT, l/min 1.53 ± 0.45 1.58 ± 0.47† 1.61 ± 0.51 1.60 ± 0.52 FB, breaths/min 37 ± 11 37 ± 13 47 ± 10 49 ± 14 EELV, %TLC 56 ± 4 55 ± 5 56 ± 6 55 ± 7 EILV, %TLC 91 ± 6 91 ± 6 92 ± 7 92 ± 8 IRV, l 0.41 ± 0.29 0.39 ± 0.26 0.37 ± 0.26 0.38 ± 0.28 Values are mean ± SD.   CWR, constant work-rate exercise test; EELV, end-expiratory lung volume; EILV, end-inspiratory lung volume; FB, breathing frequency; INCR, incremental exercise test; MVV, maximal voluntary ventilation; RER, respiratory exchange ratio; SpO2, oxygen saturation by pulse oximetry; TLC, total lung capacity; V̇E, minute ventilation; V̇O2, oxygen consumption; VT, tidal volume. †, p<0.01 INCR vs. CWR; ‡, p<0.001, INCR vs. CWR.   72  	 Figure 4-1. A) tidal volume (VT), B) breathing frequency (FB), C) inspiratory reserve volume (IRV), and D) operating lung volumes relative to minute ventilation (V̇E) during incremental (INCR) and constant work-rate (CWR) exercise Values are mean ± SEM.  Grey markers, VT/V̇E inflection; TLC, total lung capacity.   0 10 20 30 40 50 60 0 20 40 60 80 100 F B (breaths/min) VE (l/min) B 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0 0 20 40 60 80 100 VT (l) VE (l/min) INCR CWR A 0 10 20 30 40 50 60 70 80 90 100 0 20 40 60 80 100 Lung volume (%TLC) VE (l/min) D 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 0 20 40 60 80 100 IRV (l) VE (l/min) C . . . . 73   Figure 4-2. Dyspnoea intensity relative to A) minute ventilation (V̇E), B) tidal volume (VT), and C) inspiratory reserve volume (IRV) during incremental (INCR) and constant work-rate (CWR) exercise Values are mean ± SEM.  Grey markers, VT/V̇E inflection.    0 1 2 3 4 5 6 7 8 9 10 0.0 0.5 1.0 1.5 Dyspnoea (0-10 sclae) IRV (l) 0 1 2 3 4 5 6 7 8 9 10 0.0 0.5 1.0 1.5 2.0 Dyspnoea (0-10 scale) VT (l) 0 1 2 3 4 5 6 7 8 9 10 0 20 40 60 80 100 Dyspnoea (0-10 scale) VE (l) INCR CWR A B C . 74   Figure 4-3. Selection frequency of dyspnoea descriptors during A) incremental and B) constant work-rate exercise Dashed line, tidal volume relative to minute ventilation (VT/V̇E) inflection.   0 10 20 30 40 50 60 70 80 90 100 0 2 4 6 8 10 12 14 Descriptor (% of subjects) Time (min) 0 10 20 30 40 50 60 70 80 90 100 0 2 4 6 8 10 12 14 Descriptor (% of subjects) Time (min) Increased work and effort Unsatisfied inspiration Unsatisfied expiration A B 75   Figure 4-4. Selection frequency of A) work/effort and B) unsatisfied inspiration relative to dyspnoea intensity during incremental (INCR) and constant work-rate (CWR) exercise Values are mean ± SEM.  Grey markers, tidal volume relative to minute ventilation (VT/V̇E) inflection.     Figure 4-5. Selection frequency of dyspnoea descriptors measured via a 15-item questionnaire and clustered into 10 experiences of breathing discomfort evoked by different respiratory stimuli at peak A) incremental (INCR) and B) constant work-rate (CWR) exercise 0 10 20 30 40 50 60 70 80 90 100 Inspiratory difficulty Unsatisfied Inspiration Work Rapid Expiratory difficulty Heavy Hunger Shallow Suffocating Tight Selection frequency (%) Dyspnoea Descriptors INCR CWR 0 10 20 30 40 50 60 70 80 90 100 0 1 2 3 4 5 6 7 8 9 10 Work/Effort (% of subjects) Dyspnoea (0-10 scale) INCR CWR 0 10 20 30 40 50 60 70 80 90 100 0 1 2 3 4 5 6 7 8 9 10 Unsatisfied inspiration (% of subjects) Dyspnoea (0-10 scale) A B 76   Figure 4-6. Ventilatory responses to exercise in patients that selected unsatisfied inspiration (selectors) and patients that did not select unsatisfied inspiration (non-selectors) during incremental exercise Values are mean ± SEM.  Grey markers, tidal volume relative to minute ventilation (VT/V̇E) inflection; IC, inspiratory capacity.       0.0 0.5 1.0 1.5 2.0 2.5 0 20 40 60 80 100 VT (l) VE (l/min) Selectors Non-selectors 0 10 20 30 40 50 60 0 20 40 60 80 100 F B (breaths/min) VE (l/min) 0 10 20 30 40 50 60 70 80 90 100 0 20 40 60 80 100 VT/IC  (%) VE (l/min) A B C . . . 77 Chapter 5: Conclusion  5.1 Overall summary Dyspnoea is an important and common symptom in patients with fibrotic ILD that often results in poor quality of life and an inability to perform physical activity.  Unfortunately, there are few treatment options that improve dyspnoea and therefore exercise tolerance in these patients (Noble et al., 2011; Raghu et al., 2011).  A major limitation of the current approaches to symptom management in ILD is that the guidelines, and corresponding recommendations, are based on research conducted in COPD patients.  The application of COPD-specific strategies to patients with ILD has no sound physiological basis.  Thus, there is a strong need for research to develop and evaluate dyspnoea-relieving strategies specifically in ILD, which was the premise of this thesis.  This work is imperative for the development of more effective population-specific therapies that reduce this debilitating symptom, enable patients to become more physically active, and enhance their quality of life.  First, we conducted a randomized controlled crossover study to evaluate the effect(s) of hyperoxia on ventilatory, sensory, and exercise performance outcomes in patients with fibrotic ILD (see Chapter 2).  Contrary to previously published studies on supplemental oxygen in ILD, our results suggest that breathing 60% oxygen during exercise results in clinically significant improvements dyspnoea intensity ratings.  We also found substantial improvements in exercise endurance time and the qualitative dimensions of dyspnoea with hyperoxia. Second, we performed an invasive and technically demanding study to assess whether dyspnoea intensity in patients fibrotic ILD is primarily determined by NRD or NMU (see Chapter 3).  Similar to previous findings in COPD, our results suggest that exertional dyspnoea 78 is strongly related to the level of NRD, estimated via EMGdi, in patients with fibrotic ILD.  Interestingly, when we experimentally reduced dyspnoea by administering supplemental oxygen during exercise, the reduction in dyspnoea was more closely related to an improvement in the relationship between NRD and the mechanical output of the respiratory system (i.e., NMU) compared to neural respiratory drive alone.     Third, we conducted a study examining dyspnoea quality in patients with fibrotic ILD during incremental and constant work-rate exercise (see Chapter 4).  This study demonstrates the importance of mechanical constraints on VT expansion on dyspnoea intensity and quality during exercise in patients with fibrotic ILD.  Specifically, the VT/V̇E inflection marked the onset of an increase in dyspnoea intensity as well as selection frequency of unsatisfied inspiration as the predominate dyspnoea descriptor.  The results of this second study extend those of the first, as in order for us to more fully understand the cause(s) and distinct manifestations of dyspnoea, it is important to study its multidimensional components.    5.2 Significance The work presented in thesis provides meaningful insight into the effects of hyperoxia on exercise tolerance in patients with fibrotic ILD.  Additionally, this work improves our fundamental understanding of dyspnoea in patients with fibrotic ILD and has important clinical implications for the management of exertional symptoms in this population. The most comprehensive evaluation of hyperoxia on ventilatory, sensory, and exercise performance outcomes in patients with fibrotic ILD, to date, is presented in Chapter 2.  This is also the first study to examine the effects of hyperoxia on the qualitative dimensions of dyspnoea during exercise in patients with any chronic respiratory disease.  Our experimental approach to 79 supplemental oxygen delivery may serve to optimize pulmonary rehabilitation programs by enabling patients to train at higher exercise intensities, regardless of their degree of exertional arterial oxygen desaturation. Our use of advanced and invasive physiological measurement techniques for the study presented in Chapter 3 provide novel insight into the neurophysiological underpinnings of exertional dyspnoea in patients with fibrotic ILD.  This was the first study to demonstrate that dyspnoea intensity in patients with fibrotic ILD more strongly reflects increased neural drive, as opposed to the imbalance between neural drive and the mechanical response of the respiratory system.  This is consistent with Figure 1-1 (Chapter 1), which depicts perceived dyspnoea as the awareness of increased neural drive needed to achieve a given ventilation in the setting of increased mechanical loading (e.g., from reduced lung compliance) and/or reduced capacity of the respiratory muscles to generate pressure (e.g., sub-optimal length-tension relationship due to relatively high operating lung volumes).  Improving our understanding of this debilitating symptom may aid in the development of new and/or optimized, targeted, population-specific therapies to reduce dyspnoea and subsequently, improve functional capacity and/or quality of life in this population.  The study presented in Chapter 4 is the first study to examine the evolution and corresponding physiologic basis of the perception of unsatisfied inspiration during acute bouts of exercise in patients with fibrotic ILD.  Our results suggest that interventions that reduce ventilatory requirements, and therefore delay or prevent the VT/V̇E inflection from occurring, should reduce dyspnoea intensity and perception of unsatisfied inspiration in this population.  Furthermore, there may be a role for dyspnoea quality as a clinical indicator to gauge relative exercise intensity.  In combination with the findings in Chapter 2, these studies have important 80 implications for exercise rehabilitation programs and symptom management.  Collectively, this thesis (1) provides insight on the effects of hyperoxia on exercise tolerance in this population; (2) contributes to what is known about the mechanisms and manifestation of the intensity and quality of dyspnoea in patients with ILD; and (3) proposes potential physiological targets for symptom-relieving interventions.  5.3 Strengths and limitations The strengths of this thesis relate to its methodology, as the current most validated and objective measures were used to address the primary study aims.  Our study design presented in Chapter 2 was more appropriate than all previously published studies in evaluating the effects of hyperoxia on exertional dyspnoea and exercise performance in patients with ILD, as we examined iso-time responses at a standardized stimulus intensity.  Additionally, the multipair oesophageal catheter used to measure EMGdi in Chapter 3 is the best available surrogate measure of neural respiratory drive during exercise in humans. The underlying limitation of the work presented in this thesis relates to the generalizability of our findings.  The studies presented in this thesis have a seemingly small sample size.  However, my studies were of a similar or larger sample size compared to previously published physiological studies in patients with ILD, which have included between seven and 16 patients.  There are significant challenges associated with recruiting patients that meet such strict inclusion criteria, as well as the invasiveness, time intensiveness, and technical nature of our experiments.  While IPF is the most common sub-type, it still qualifies as a rare disease in most countries.  Additionally, there are only two ILD specialty clinics in British Columbia and we actively recruited from both.  Nevertheless, patient recruitment for this thesis 81 still took over two years to complete.  As a consequence, our findings may not be representative of all patients with fibrotic ILD.  Future studies with larger sample sizes would allow for more translatable findings, as well higher-powered sub-group comparisons.  For example: between the many different sub-types of ILD, sexes, and responders vs. non-responders to interventions such as hyperoxia.  5.4 Future directions The main application of this thesis is how we can enhance the exercise component of pulmonary rehabilitation for patients with fibrotic ILD.  Severe dyspnoea makes it challenging for ILD patients to withstand the amount of training they need to achieve a meaningful and enduring benefit.  In this thesis, we have shown that in an acute setting, hyperoxia has profound effects on dyspnoea intensity, dyspnoea quality, and exercise tolerance.  However, we still do not know if this oxygen delivery strategy can be applied to a long-term exercise program.  More specifically, if these patients exercise for longer and/or with a greater intensity with the help of supplemental oxygen, will they attain greater physiological adaptations (e.g., in the peripheral muscles) from an exercise training program compared to if they were to do the same regimen without the additional oxygen?  To answer this question, the next logical step is to conduct a randomized control trial to evaluate changes in exercise tolerance, dyspnoea, and quality of life after a longer-term exercise training program with hyperoxia. Indeed, the preliminary findings from this thesis formed the premise of a large clinical trial entitled: “High Oxygen delivery to Preserve Exercise capacity in IPF patients treated with nintedanib study” (i.e., The HOPE-IPF study) (Ryerson et al., 2016).  This double-blind, randomized, placebo-controlled study will include 88 IPF patients at 8 research centres in 82 collaboration with more than 30 clinicians and scientists across Canada.  After initial assessment, including pulmonary function testing, an incremental cycle test to determine peak work-rate, and a constant-load cycle exercise test corresponding to 75% of peak incremental work-rate, patients will undergo exercise training three times a week for eight weeks.  The intervention group will exercise while breathing a hyperoxic gas mixture (FiO2 = 60%) as used in the present thesis; the control group will breathe room air (FiO2=21%) while exercising unless their oxygen saturation decreases below 88%, in which case they will receive the minimum amount of supplemental oxygen needed to maintain an oxygen saturation ≥88%, up to a maximum FiO2 of 40%, per standard of care.  The exercise training protocol will be identical for both groups.  The mode of oxygen delivery will be identical to that developed for this thesis (see Chapter 2).  Follow up tests will take place 1 week, 2 months, and 4 months post-training, and will including another constant-load cycle exercise test at 75% of peak incremental work-rate.  The primary outcome will be based on the pre-training to post-training change in constant-load cycle exercise endurance time, which is the most responsive parameter for assessing the efficacy of pulmonary rehabilitation in patients with IPF (Arizono et al., 2014).  Exertional dyspnoea intensity will be measured during all cycle exercise tests at rest, throughout exercise, and at peak exercise.  Dyspnoea, physical activity, and quality of life will also be evaluated using validated questionnaires.  Twenty-six patients have been enrolled.   While the HOPE-IPF study will help elucidate whether or not prevention of hypoxaemia will allow IPF patients to achieve greater benefits from exercise training, and if these improvements in the exercise setting translate into improvements in quality of life, participation is limited to IPF patients treated with nintedanib.  Thus, the findings cannot be extrapolated to the broader ILD population or patients treated with other pharmacotherapy.  The HOPE-IPF 83 study will also not evaluate the safety of hyperoxia during exercise, the impact of a lower FiO2, or the feasibility of this specific method of oxygen delivery during pulmonary rehabilitation.  Therefore, additional studies are needed to examine the efficacy of hyperoxia during exercise training in a more representative population as well as to determine its safety.  These data may inform guidelines around the provision of supplemental oxygen during pulmonary rehabilitation in patients with ILD. Additionally, this thesis identified specific physiological events underlying dyspnoea intensity and quality.  Future studies are needed to manipulate these variables, in the context of exercise, to develop new concepts for symptom management.  For example, eccentric cycling or single leg exercise has been not explored in this population, but similar to hyperoxia, may also reduce the ventilatory requirement for a given exercise intensity.  5.5 Overall conclusion The purpose of this thesis was three-fold: (1) to appropriately and comprehensively evaluate the effects of hyperoxia on dyspnoea in patients with fibrotic ILD; (2) to determine if dyspnoea intensity is more strongly reflects NRD or NMU in patients with fibrotic ILD; and (3) to evaluate the qualitative dimensions of exertional dyspnoea in patients with fibrotic ILD.  We found that breathing 60% oxygen results in clinically significant improvements in exertional dyspnoea intensity, dyspnoea quality, and exercise performance in this population.  Our data also suggests that exertional dyspnoea intensity in these patients is better predicted by NRD compared to NMU.  However, hyperoxia as an intervention appears to reduce dyspnoea via improvements in NMU.  Lastly, the VT/V̇E inflection is an important respiratory-mechanical event in patients with fibrotic ILD that has implications for exertional dyspnoea intensity and quality.  Future work is 84 needed to translate these findings during acute bouts of exercise in to the context of a longer-term exercise training program, as well as to extrapolate them to the larger ILD population.85 Bibliography Abernethy AP, Currow DC, Frith P, Fazekas BS, McHugh A & Bui C. (2003). Randomised, double blind, placebo controlled crossover trial of sustained release morphine for the management of refractory dyspnoea. BMJ 327, 523-528.  Adams L, Lane R, Shea SA, Cockcroft A & Guz A. (1985). Breathlessness during different forms of ventilatory stimulation: a study of mechanisms in normal subjects and respiratory patients. Clin Sci 69, 663-672.  Agusti AG, Roca J, Gea J, Wagner PD, Xaubet A & Rodriguez-Roisin R. (1991). Mechanisms of gas-exchange impairment in idiopathic pulmonary fibrosis. Am Rev Respir Dis 143, 219-225.  Arizono S, Taniguchi H, Sakamoto K, Kondoh Y, Kimura T, Kataoka K, Ogawa T, Watanabe F, Nishiyama O, Nishimura K, Kozu R & Tabira K. (2014). Endurance time is the most responsive exercise measurement in idiopathic pulmonary fibrosis. Respir Care 59, 1108-1115.  Austrian R, Mc CJ, Renzetti AD, Jr., Donald KW, Riley RL & Cournand A. (1951). Clinical and physiologic features of some types of pulmonary diseases with impairment of alveolar-capillary diffusion; the syndrome of "alveolar-capillary block". Am J Med 11, 667-685.  Banzett RB, Lansing RW, Brown R, Topulos GP, Yager D, Steele SM, Londono B, Loring SH, Reid MB, Adams L & et al. (1990). 'Air hunger' from increased PCO2 persists after complete neuromuscular block in humans. Respir Physiol 81, 1-17.  Banzett RB, Pedersen SH, Schwartzstein RM & Lansing RW. (2008). The affective dimension of laboratory dyspnea: air hunger is more unpleasant than work/effort. Am J Respir Crit Care Med 177, 1384-1390.  Bell EC, Cox NS, Goh N, Glaspole I, Westall GP, Watson A & Holland AE. (2017). Oxygen therapy for interstitial lung disease: a systematic review. Eur Respir Rev 26, 160080 [dx.doi.org/10.1183/16000617.0080-2016].  Binks AP, Moosavi SH, Banzett RB & Schwartzstein RM. (2002). "Tightness" sensation of asthma does not arise from the work of breathing. Am J Respir Crit Care Med 165, 78-82.  Bjoraker JA, Ryu JH, Edwin MK, Myers JL, Tazelaar HD, Schroeder DR & Offord KP. (1998). Prognostic significance of histopathologic subsets in idiopathic pulmonary fibrosis. Am J Respir Crit Care Med 157, 199-203.  Blackie SP, Fairbarn MS, McElvaney GN, Morrison NJ, Wilcox PG & Pardy RL. (1989). Prediction of maximal oxygen uptake and power during cycle ergometry in subjects older than 55 years of age. Am Rev Respir Dis 139, 1424-1429. 86 Borg GA. (1982). Psychophysical bases of perceived exertion. Med Sci Sports Exerc 14, 377-381.  Bush A & Busst CM. (1988). Cardiovascular function at rest and on exercise in patients with cryptogenic fibrosing alveolitis. Thorax 43, 276-283.  Bye PT, Anderson SD, Woolcock AJ, Young IH & Alison JA. (1982). Bicycle endurance performance of patients with interstitial lung disease breathing air and oxygen. Am Rev Respir Dis 126, 1005-1012.  Campbell EJ, Gandevia SC, Killian KJ, Mahutte CK & Rigg JR. (1980). Changes in the perception of inspiratory resistive loads during partial curarization. J Physiol 309, 93-100.  Cao M, Wamboldt FS, Brown KK, Hickman J, Olson AL, Solomon JJ & Swigris JJ. (2015). Supplemental oxygen users with pulmonary fibrosis perceive greater dyspnea than oxygen non-users. Multidiscip Respir Med 10 [dx.doi.org/10.1186/s40248-015-0035-y].  Chapman KR & Rebuck AS. (1983). Inspiratory and expiratory resistive loading as a model of dyspnea in asthma. Respiration 44, 425-432.  Chemla D, Lau EM, Papelier Y, Attal P & Herve P. (2015). Pulmonary vascular resistance and compliance relationship in pulmonary hypertension. Eur Respir J 46, 1178-1189.  Chen Z, Eldridge FL & Wagner PG. (1991). Respiratory-associated rhythmic firing of midbrain neurones in cats: relation to level of respiratory drive. J Physiol 437, 305-325.  Chen Z, Eldridge FL & Wagner PG. (1992). Respiratory-associated thalamic activity is related to level of respiratory drive. Respir Physiol 90, 99-113.  Churg A, Sin DD, Everett D, Brown K & Cool C. (2009). Pathologic patterns and survival in chronic hypersensitivity pneumonitis. Am J Surg Pathol 33, 1765-1770.  Colebatch JG, Adams L, Murphy K, Martin AJ, Lammertsma AA, Tochon-Danguy HJ, Clark JC, Friston KJ & Guz A. (1991). Regional cerebral blood flow during volitional breathing in man. J Physiol 443, 91-103.  Cory JM, Schaeffer MR, Wilkie SS, Ramsook AH, Puyat JH, Arbour B, Basran R, Lam M, Les C, MacDonald B, Jensen D & Guenette JA. (2015). Sex differences in the intensity and qualitative dimensions of exertional dyspnea in physically active young adults. J Appl Physiol 119, 998-1006.  Dean NC, Brown JK, Himelman RB, Doherty JJ, Gold WM & Stulbarg MS. (1992). Oxygen May Improve Dyspnea and Endurance in Patients with Chronic Obstructive Pulmonary Disease and only Mild Hypoxemia. American Review of Respiratory Disease 146, 941-945. 87 Dempsey JA & Wagner PD. (1999). Exercise-induced arterial hypoxemia. J Appl Physiol 87, 1997-2006.  Dowman LM, McDonald CF, Bozinovski S, Vlahos R, Gillies R, Pouniotis D, Hill CJ, Goh NSL & Holland AE. (2017a). Greater endurance capacity and improved dyspnoea with acute oxygen supplementation in idiopathic pulmonary fibrosis patients without resting hypoxaemia. Respirology 22, 957-964.  Dowman LM, McDonald CF, Hill CJ, Lee AL, Barker K, Boote C, Glaspole I, Goh NSL, Southcott AM, Burge AT, Gillies R, Martin A & Holland AE. (2017b). The evidence of benefits of exercise training in interstitial lung disease: a randomised controlled trial. Thorax 72, 610-619.  du Bois RM, Albera C, Bradford WZ, Costabel U, Leff JA, Noble PW, Sahn SA, Valeyre D, Weycker D & King TE. (2014). 6-minute walk distance is an independent predictor of mortality in patients with idiopathic pulmonary fibrosis. Eur Respir J 43, 1421-1429.  du Bois RM, Weycker D, Albera C, Bradford WZ, Costabel U, Kartashov A, King TE, Jr., Lancaster L, Noble PW, Sahn SA, Thomeer M, Valeyre D & Wells AU. (2011a). Forced vital capacity in patients with idiopathic pulmonary fibrosis: test properties and minimal clinically important difference. Am J Respir Crit Care Med 184, 1382-1389.  du Bois RM, Weycker D, Albera C, Bradford WZ, Costabel U, Kartashov A, Lancaster L, Noble PW, Sahn SA, Szwarcberg J, Thomeer M, Valeyre D & King TE, Jr. (2011b). Six-minute-walk test in idiopathic pulmonary fibrosis: test validation and minimal clinically important difference. Am J Respir Crit Care Med 183, 1231-1237.  el-Manshawi A, Killian KJ, Summers E & Jones NL. (1986). Breathlessness during exercise with and without resistive loading. J Appl Physiol 61, 896-905.  Eves ND, Petersen SR, Haykowsky MJ, Wong EY & Jones RL. (2006). Helium-hyperoxia, exercise, and respiratory mechanics in chronic obstructive pulmonary disease. Am J Respir Crit Care Med 174, 763-771.  Faisal A, Alghamdi BJ, Ciavaglia CE, Elbehairy AF, Webb KA, Ora J, Neder JA & O'Donnell DE. (2016). Common mechanisms of dyspnea in chronic interstitial and obstructive lung disorders. Am J Respir Crit Care Med 193, 299-309.  Flume PA, Eldridge FL, Edwards LJ & Houser LM. (1994). The Fowler breathholding study revisited: continuous rating of respiratory sensation. Respir Physiol 95, 53-66.  Frank RC, Hicks S, Duck AM, Spencer L, Leonard CT & Barnett E. (2012). Ambulatory oxygen in idiopathic pulmonary fibrosis: of what benefit? Eur Respir J 40, 269-270.  88 Gandevia SC, Killian K, McKenzie DK, Crawford M, Allen GM, Gorman RB & Hales JP. (1993). Respiratory sensations, cardiovascular control, kinaesthesia and transcranial stimulation during paralysis in humans. J Physiol 470, 85-107.  Gibson GJ & Pride NB. (1977). Pulmonary mechanics in fibrosing alveolitis: the effects of lung shrinkage. Am Rev Respir Dis 116, 637-647.  Gowda K, Zintel T, McParland C, Orchard R & Gallagher CG. (1990). Diagnostic value of maximal exercise tidal volume. Chest 98, 1351-1354.  Guenette JA, Chin RC, Cory JM, Webb KA & O'Donnell DE. (2013). Inspiratory capacity during exercise: measurement, analysis, and interpretation. Pulm Med 2013, 956081 [dx.doi.org/10.1155/2013/956081].  Guenette JA, Jensen D, Webb KA, Ofir D, Raghavan N & O'Donnell DE. (2011). Sex differences in exertional dyspnea in patients with mild COPD: physiological mechanisms. Respir Physiol Neurobiol 177, 218-227.  Gutierrez C, Ghezzo RH, Abboud RT, Cosio MG, Dill JR, Martin RR, McCarthy DS, Morse JL & Zamel N. (2004). Reference values of pulmonary function tests for Canadian Caucasians. Can Respir J 11, 414-424.  Hamer J. (1964). Cause of Low Arterial Oxygen Saturation in Pulmonary Fibrosis. Thorax 19, 507-514.  Hansen JE & Wasserman K. (1996). Pathophysiology of activity limitation in patients with interstitial lung disease. Chest 109, 1566-1576.  Harms CA, Babcock MA, McClaran SR, Pegelow DF, Nickele GA, Nelson WB & Dempsey JA. (1997). Respiratory muscle work compromises leg blood flow during maximal exercise. J Appl Physiol 82, 1573-1583.  Harris-Eze AO, Sridhar G, Clemens RE, Gallagher CG & Marciniuk DD. (1994). Oxygen improves maximal exercise performance in interstitial lung disease. Am J Respir Crit Care Med 150, 1616-1622.  Harris-Eze AO, Sridhar G, Clemens RE, Zintel TA, Gallagher CG & Marciniuk DD. (1996). Role of hypoxemia and pulmonary mechanics in exercise limitation in interstitial lung disease. Am J Respir Crit Care Med 154, 994-1001.  Harty HR, Corfield DR, Schwartzstein RM & Adams L. (1999). External thoracic restriction, respiratory sensation, and ventilation during exercise in men. J Appl Physiol 86, 1142-1150.  Harver A, Schwartzstein RM, Kotses H, Humphries CT, Schmaling KB & Mullin ML. (2011). Descriptors of breathlessness in children with persistent asthma. Chest 139, 832-838. 89 Hey EN, Lloyd BB, Cunningham DJ, Jukes MG & Bolton DP. (1966). Effects of various respiratory stimuli on the depth and frequency of breathing in man. Respir Physiol 1, 193-205.  Hirano A, Kanehiro A, Ono K, Ito W, Yoshida A, Okada C, Nakashima H, Tanimoto Y, Kataoka M, Gelfand EW & Tanimoto M. (2006). Pirfenidone modulates airway responsiveness, inflammation, and remodeling after repeated challenge. Am J Respir Cell Mol Biol 35, 366-377.  Hopkins RB, Burke N, Fell C, Dion G & Kolb M. (2016). Epidemiology and survival of idiopathic pulmonary fibrosis from national data in Canada. Eur Respir J 48, 187-195.  Hughes JM, Lockwood DN, Jones HA & Clark RJ. (1991). DLCO/Q and diffusion limitation at rest and on exercise in patients with interstitial fibrosis. Respir Physiol 83, 155-166.  Iyer SN, Gurujeyalakshmi G & Giri SN. (1999). Effects of pirfenidone on procollagen gene expression at the transcriptional level in bleomycin hamster model of lung fibrosis. J Pharmacol Exp Ther 289, 211-218.  Iyer SN, Margolin SB, Hyde DM & Giri SN. (1998). Lung fibrosis is ameliorated by pirfenidone fed in diet after the second dose in a three-dose bleomycin-hamster model. Exp Lung Res 24, 119-132.  Iyer SN, Wild JS, Schiedt MJ, Hyde DM, Margolin SB & Giri SN. (1995). Dietary intake of pirfenidone ameliorates bleomycin-induced lung fibrosis in hamsters. J Lab Clin Med 125, 779-785.  Jennings AL, Davies AN, Higgins JPT, Gibbs JSR & Broadley KE. (2002). A systematic review of the use of opioids in the management of dyspnoea. Thorax 57, 939-944.  Jensen D, Ofir D & O'Donnell DE. (2009). Effects of pregnancy, obesity and aging on the intensity of perceived breathlessness during exercise in healthy humans. Respir Physiol Neurobiol 167, 87-100.  Jensen D, Pattinson K & Jolley C. (2016). Mechanisms of breathlessness. In Palliative Care in Respiratory Disease, ed. Bausewein C, Currow DC & Johnson MJ, pp. 111-133. European Respiratory Society.  Jolley CJ, Luo YM, Steier J, Rafferty GF, Polkey MI & Moxham J. (2015). Neural respiratory drive and breathlessness in COPD. Eur Respir J 45, 355-364.  Jones NL & Rebuck AS. (1979). Tidal volume during exercise in patients with diffuse fibrosing alveolitis. Bull Eur Physiopathol Respir 15, 321-328.  90 Keating D, Levvey B, Kotsimbos T, Whitford H, Westall G, Williams T & Snell G. (2009). Lung transplantation in pulmonary fibrosis: challenging early outcomes counterbalanced by surprisingly good outcomes beyond 15 years. Transplant Proc 41, 289-291.  Killian KJ, Gandevia SC, Summers E & Campbell EJ. (1984). Effect of increased lung volume on perception of breathlessness, effort, and tension. J Appl Physiol Respir Environ Exerc Physiol 57, 686-691.  King TE, Jr., Bradford WZ, Castro-Bernardini S, Fagan EA, Glaspole I, Glassberg MK, Gorina E, Hopkins PM, Kardatzke D, Lancaster L, Lederer DJ, Nathan SD, Pereira CA, Sahn SA, Sussman R, Swigris JJ, Noble PW & Group AS. (2014). A phase 3 trial of pirfenidone in patients with idiopathic pulmonary fibrosis. N Engl J Med 370, 2083-2092.  King TE, Schwarz MI, Brown K, Tooze JA, Colby TV, Waldron Jr. JA, Flint A, Thurlbeck W & Cherniack RM. (2001). Idiopathic pulmonary fibrosis: relationship between histopathologic features and mortality. Am J Respir Crit Care Med 164, 1025-1032.  Knower MT, Dunagan DP, Adair NE & Chin Jr. R. (2001). Baseline oxygen saturation predicts exercise desaturation below prescription threshold in patients with chronic obstructive pulmonary disease. Arch Intern Med 161, 732-736.  Knudson RJ, Lebowitz MD, Holberg CJ & Burrows B. (1983). Changes in the normal maximal expiratory flow-volume curve with growth and aging. Am Rev Respir Dis 127, 725-734.  Kohberg C, Andersen CU & Bendstrup E. (2016). Opioids: an unexplored option for treatment of dyspnea in IPF. Eur Clin Respir J 3 [dx.doi.org/10.3402/ecrj.v3.30629].  Lacasse Y, Selman M, Costabel U, Dalphin JC, Ando M, Morell F, Erkinjuntti-Pekkanen R, Muller N, Colby TV, Schuyler M, Cormier Y & HP Study Group. (2003). Clinical diagnosis of hypersensitivity pneumonitis. Am J Respir Crit Care Med 168, 952-958.  Lane R & Adams L. (1993). Metabolic acidosis and breathlessness during exercise and hypercapnia in man. J Physiol 461, 47-61.  Lane R, Adams L & Guz A. (1987). Is low-level respiratory resistive loading during exercise perceived as breathlessness? Clin Sci 73, 627-634.  Lane R, Adams L & Guz A. (1990). The effects of hypoxia and hypercapnia on perceived breathlessness during exercise in humans. J Physiol 428, 579-593.  Lansing RW, Im BS, Thwing JI, Legedza AT & Banzett RB. (2000). The perception of respiratory work and effort can be independent of the perception of air hunger. Am J Respir Crit Care Med 162, 1690-1696.  Laveneziana P. (2010). Qualitative aspects of exertional dyspnea in patients with restrictive lung disease. Multidiscip Respir Med 5, 211-215. 91 Laveneziana P, Bruni GI, Presi I, Stendardi L, Duranti R & Scano G. (2013). Tidal volume inflection and its sensory consequences during exercise in patients with stable asthma. Respir Physiol Neurobiol 185, 374-379.  Laveneziana P, Webb KA, Ora J, Wadell K & O'Donnell DE. (2011). Evolution of dyspnea during exercise in chronic obstructive pulmonary disease: impact of critical volume constraints. Am J Respir Crit Care Med 184, 1367-1373.  Laviolette L, Laveneziana P & ERS Research Seminar Faculty. (2014). Dyspnoea: a multidimensional and multidisciplinary approach. Eur Respir J 43, 1750-1762.  Leblanc P, Bowie DM, Summers E, Jones NL & Killian KJ. (1986). Breathlessness and exercise in patients with cardiorespiratory disease. Am Rev Respir Dis 133, 21-25.  Ley B, Ryerson CJ, Vittinghoff E, Ryu JH, Tomassetti S, Lee JS, Poletti V, Buccioli M, Elicker BM, Jones KD, King TE, Jr. & Collard HR. (2012). A multidimensional index and staging system for idiopathic pulmonary fibrosis. Ann Intern Med 156, 684-691.  Lougheed MD, Fisher T & O'Donnell DE. (2006). Dynamic hyperinflation during bronchoconstriction in asthma: implications for symptom perception. Chest 130, 1072-1081.  Lougheed MD, Flannery J, Webb KA & O'Donnell DE. (2002). Respiratory sensation and ventilatory mechanics during induced bronchoconstriction in spontaneously breathing low cervical quadriplegia. Am J Respir Crit Care Med 166, 370-376.  Lougheed MD, Lam M, Forkert L, Webb KA & O'Donnell DE. (1993). Breathlessness during acute bronchoconstriction in asthma. Pathophysiologic mechanisms. Am Rev Respir Dis 148, 1452-1459.  Lourenco RV, Turino GM, Davidson LA & Fishman AP. (1965). The regulation of ventilation in diffuse pulmonary fibrosis. Am J Med 38, 199-216.  Luo YM, Moxham J & Polkey MI. (2008). Diaphragm electromyography using an oesophageal catheter: current concepts. Clin Sci 115, 233-244.  Macintyre N, Crapo RO, Viegi G, Johnson DC, van der Grinten CP, Brusasco V, Burgos F, Casaburi R, Coates A, Enright P, Gustafsson P, Hankinson J, Jensen R, McKay R, Miller MR, Navajas D, Pedersen OF, Pellegrino R & Wanger J. (2005). Standardisation of the single-breath determination of carbon monoxide uptake in the lung. Eur Respir J 26, 720-735.  Maggiore SM, Idone FA, Vaschetto R, Festa R, Cataldo A, Antonicelli F, Montini L, De Gaetano A, Navalesi P & Antonelli M. (2014). Nasal high-flow versus Venturi mask oxygen therapy after extubation. Effects on oxygenation, comfort, and clinical outcome. Am J Respir Crit Care Med 190, 282-288. 92 Mahler DA. (2013). Opioids for refractory dyspnea. Expert Rev Respir Med 7, 123-134.  Mahler DA, Harver A, Lentine T, Scott JA, Beck K & Schwartzstein RM. (1996). Descriptors of breathlessness in cardiorespiratory diseases. Am J Respir Crit Care Med 154, 1357-1363.  Mahler DA & Wells CK. (1988). Evaluation of clinical methods for rating dyspnea. Chest 93, 580-586.  Marciniuk DD, Sridhar G, Clemens RE, Zintel TA & Gallagher CG. (1994a). Lung volumes and expiratory flow limitation during exercise in interstitial lung disease. J Appl Physiol 77, 963-973.  Marciniuk DD, Watts RE & Gallagher CG. (1994b). Dead space loading and exercise limitation in patients with interstitial lung disease. Chest 105, 183-189.  Markovitz GH, Colthurst J, Storer TW & Cooper CB. (2010). Effective inspired oxygen concentration measured via transtracheal and oral gas analysis. Respir Care 55, 453-459.  Marti S, Pajares V, Morante F, Ramon MA, Lara J, Ferrer J & Guell MR. (2013). Are oxygen-conserving devices effective for correcting exercise hypoxemia? Respir Care 58, 1606-1613.  Mason DP, Brizzio ME, Alster JM, McNeill AM, Murthy SC, Budev MM, Mehta AC, Minai OA, Pettersson GB & Blackstone EH. (2007). Lung transplantation for idiopathic pulmonary fibrosis. Ann Thorac Surg 84, 1121-1128.  Miller MR, Hankinson J, Brusasco V, Burgos F, Casaburi R, Coates A, Crapo R, Enright P, van der Grinten CP, Gustafsson P, Jensen R, Johnson DC, MacIntyre N, McKay R, Navajas D, Pedersen OF, Pellegrino R, Viegi G & Wanger J. (2005). Standardisation of spirometry. Eur Respir J 26, 319-338.  Mitchell GS & Babb TG. (2006). Layers of exercise hyperpnea: modulation and plasticity. Respir Physiol Neurobiol 151, 251-266.  Moosavi SH, Golestanian E, Binks AP, Lansing RW, Brown R & Banzett RB. (2003). Hypoxic and hypercapnic drives to breathe generate equivalent levels of air hunger in humans. J Appl Physiol 94, 141-154.  Moosavi SH, Topulos GP, Hafer A, Lansing RW, Adams L, Brown R & Banzett RB. (2000). Acute partial paralysis alters perceptions of air hunger, work and effort at constant PCO2 and VE. Respir Physiol 122, 45-60.  Morris JF, Koski A, Temple WP, Claremont A & Thomas DR. (1988). Fifteen-year interval spirometric evaluation of the Oregon predictive equations. Chest 93, 123-127.  93 Moy ML, Lantin ML, Harver A & Schwartzstein RM. (1998). Language of dyspnea in assessment of patients with acute asthma treated with nebulized albuterol. Am J Respir Crit Care Med 158, 749-753.  Moy ML, Woodrow Weiss J, Sparrow D, Israel E & Schwartzstein RM. (2000). Quality of dyspnea in bronchoconstriction differs from external resistive loads. Am J Respir Crit Care Med 162, 451-455.  Nakazawa A, Cox NS & Holland AE. (2017). Current best practice in rehabilitation in interstitial lung disease. Ther Adv Respir Dis 11, 115-128.  Nishiyama O, Miyajima H, Fukai Y, Yamazaki R, Satoh R, Yamagata T, Sano H, Iwanaga T, Higashimoto Y, Nakajima H, Kume H & Tohda Y. (2013). Effect of ambulatory oxygen on exertional dyspnea in IPF patients without resting hypoxemia. Respir Med 107, 1241-1246.  Nishiyama O, Taniguchi H, Kondoh Y, Kimura T, Ogawa T, Watanabe F & Arizono S. (2005). Quadriceps weakness is related to exercise capacity in idiopathic pulmonary fibrosis. Chest 127, 2028-2033.  Noble PW, Albera C, Bradford WZ, Costabel U, Glassberg MK, Kardatzke D, King TE, Jr., Lancaster L, Sahn SA, Szwarcberg J, Valeyre D, du Bois RM & Group CS. (2011). Pirfenidone in patients with idiopathic pulmonary fibrosis (CAPACITY): two randomised trials. Lancet 377, 1760-1769.  O'Donnell DE, Bain DJ & Webb KA. (1997a). Factors contributing to relief of exertional breathlessness during hyperoxia in chronic airflow limitation. Am J Respir Crit Care Med 155, 530-535.  O'Donnell DE, Bertley JC, Chau LK & Webb KA. (1997b). Qualitative aspects of exertional breathlessness in chronic airflow limitation: pathophysiologic mechanisms. Am J Respir Crit Care Med 155, 109-115.  O'Donnell DE, Chau LK & Webb KA. (1998a). Qualitative aspects of exertional dyspnea in patients with interstitial lung disease. J Appl Physiol 84, 2000-2009.  O'Donnell DE, D'Arsigny C & Webb KA. (2001). Effects of hyperoxia on ventilatory limitation during exercise in advanced chronic obstructive pulmonary disease. Am J Respir Crit Care Med 163, 892-898.  O'Donnell DE, Fluge T, Gerken F, Hamilton A, Webb K, Aguilaniu B, Make B & Magnussen H. (2004). Effects of tiotropium on lung hyperinflation, dyspnoea and exercise tolerance in COPD. Eur Respir J 23, 832-840.  94 O'Donnell DE, Guenette JA, Maltais F & Webb KA. (2012a). Decline of resting inspiratory capacity in COPD: the impact on breathing pattern, dyspnea, and ventilatory capacity during exercise. Chest 141, 753-762.  O'Donnell DE, Hamilton AL & Webb KA. (2006). Sensory-mechanical relationships during high-intensity, constant-work-rate exercise in COPD. J Appl Physiol 101, 1025-1035.  O'Donnell DE, Hong HH & Webb KA. (2000). Respiratory sensation during chest wall restriction and dead space loading in exercising men. J Appl Physiol 88, 1859-1869.  O'Donnell DE, Lam M & Webb KA. (1998b). Measurement of symptoms, lung hyperinflation, and endurance during exercise in chronic obstructive pulmonary disease. Am J Respir Crit Care Med 158, 1557-1565.  O'Donnell DE, O'Donnell CD, Webb KA & Guenette JA. (2012b). Respiratory consequences of mild-to-moderate obesity: impact on exercise performance in health and in chronic obstructive pulmonary disease. Pulm Med 2012, 818925 [dx.doi.org/10.1155/2012/818925].  O'Donnell DE, Ora J, Webb KA, Laveneziana P & Jensen D. (2009). Mechanisms of activity-related dyspnea in pulmonary diseases. Respir Physiol Neurobiol 167, 116-132.  O'Donnell DE, Sanii R & Younes M. (1988). Improvement in exercise endurance in patients with chronic airflow limitation using continuous positive airway pressure. Am Rev Respir Dis 138, 1510-1514.  O'Donnell DE & Webb KA. (2008). The major limitation to exercise performance in COPD is dynamic hyperinflation. J Appl Physiol 105, 753-755; discussion 755-757.  Oga T, Nishimura K, Tsukino M, Hajiro T, Ikeda A & Izumi T. (2000). The effects of oxitropium bromide on exercise performance in patients with stable chronic obstructive pulmonary disease. A comparison of three different exercise tests. Am J Respir Crit Care Med 161, 1897-1901.  Oku H, Shimizu T, Kawabata T, Nagira M, Hikita I, Ueyama A, Matsushima S, Torii M & Arimura A. (2008). Antifibrotic action of pirfenidone and prednisolone: different effects on pulmonary cytokines and growth factors in bleomycin-induced murine pulmonary fibrosis. Eur J Pharmacol 590, 400-408.  Panagiotou M, Church AC, Johnson MK & Peacock AJ. (2017). Pulmonary vascular and cardiac impairment in interstitial lung disease. Eur Respir Rev 26, 160053 [dx.10.1183/16000617.0053-2016].  Panagiotou M, Polychronopoulos V & Strange C. (2016). Respiratory and lower limb muscle function in interstitial lung disease. Chron Respir Dis 13, 162-172.  95 Parshall MB. (2002). Psychometric characteristics of dyspnea descriptor ratings in emergency department patients with exacerbated chronic obstructive pulmonary disease. Res Nurs Health 25, 331-344.  Parshall MB, Carle AC, Ice U, Taylor R & Powers J. (2012a). Validation of a three-factor measurement model of dyspnea in hospitalized adults with heart failure. Heart Lung 41, 44-56.  Parshall MB, Meek PM, Sklar D, Alcock J & Bittner P. (2012b). Test-retest reliability of multidimensional dyspnea profile recall ratings in the emergency department: a prospective, longitudinal study. BMC Emerg Med 12 [dx.doi.org/10.1186/1471-227X-12-6].  Parshall MB, Schwartzstein RM, Adams L, Banzett RB, Manning HL, Bourbeau J, Calverley PM, Gift AG, Harver A, Lareau SC, Mahler DA, Meek PM, O'Donnell DE & American Thoracic Society Committee on Dyspnea. (2012c). An official ATS statement: update on the mechanisms, assessment, and management of dyspnea. Am J Respir Crit Care Med 185, 435-452.  Petit JM & Delhez L. (1970). [Some recent experimental data on the origin of dyspnea and ventilatory regulation in asthmatics]. Acta Tuberc Pneumol Belg 61, 169-186.  Powers SK, Lawler J, Dempsey JA, Dodd S & Landry G. (1989). Effects of incomplete pulmonary gas exchange on VO2 max. J Appl Physiol 66, 2491-2495.  Puente-Maestu L, Palange P, Casaburi R, Laveneziana P, Maltais F, Neder JA, O'Donnell DE, Onorati P, Porszasz J, Rabinovich R, Rossiter HB, Singh S, Troosters T & Ward S. (2016). Use of exercise testing in the evaluation of interventional efficacy: an official ERS statement. Eur Respir J 47, 429-460.  Puente-Maestu L, Villar F, de Miguel J, Stringer WW, Sanz P, Sanz ML, de Pedro JG & Martinez-Abad Y. (2009). Clinical relevance of constant power exercise duration changes in COPD. Eur Respir J 34, 340-345.  Raghu G, Collard HR, Egan JJ, Martinez FJ, Behr J, Brown KK, Colby TV, Cordier JF, Flaherty KR, Lasky JA, Lynch DA, Ryu JH, Swigris JJ, Wells AU, Ancochea J, Bouros D, Carvalho C, Costabel U, Ebina M, Hansell DM, Johkoh T, Kim DS, King TE, Jr., Kondoh Y, Myers J, Muller NL, Nicholson AG, Richeldi L, Selman M, Dudden RF, Griss BS, Protzko SL, Schunemann HJ & ATS/ERS/JRS/ALAT Committee on Idiopathic Pulmonary Fibrosis. (2011). An official ATS/ERS/JRS/ALAT statement: idiopathic pulmonary fibrosis: evidence-based guidelines for diagnosis and management. Am J Respir Crit Care Med 183, 788-824.  Ramsay SC, Adams L, Murphy K, Corfield DR, Grootoonk S, Bailey DL, Frackowiak RS & Guz A. (1993). Regional cerebral blood flow during volitional expiration in man: a comparison with volitional inspiration. J Physiol 461, 85-101. 96 Ramsook AH, Schaeffer MR, Syed N, Wilkie SS & Guenette JA. (2015). Which respiratory maneuvers generate the greatest electromyography activity of the inspiratory muscles? Am J Respir Crit Care Med 191, A2682.  Remmers JE, Brooks JE, 3rd & Tenney SM. (1968). Effect of controlled ventilation on the tolerable limit of hypercapnia. Respir Physiol 4, 78-90.  Richeldi L, Costabel U, Selman M, Kim DS, Hansell DM, Nicholson AG, Brown KK, Flaherty KR, Noble PW, Raghu G, Brun M, Gupta A, Juhel N, Kluglich M & du Bois RM. (2011). Efficacy of a tyrosine kinase inhibitor in idiopathic pulmonary fibrosis. N Engl J Med 365, 1079-1087.  Richeldi L, du Bois RM, Raghu G, Azuma A, Brown KK, Costabel U, Cottin V, Flaherty KR, Hansell DM, Inoue Y, Kim DS, Kolb M, Nicholson AG, Noble PW, Selman M, Taniguchi H, Brun M, Le Maulf F, Girard M, Stowasser S, Schlenker-Herceg R, Disse B, Collard HR & Investigators IT. (2014). Efficacy and safety of nintedanib in idiopathic pulmonary fibrosis. N Engl J Med 370, 2071-2082.  Ries AL. (2005). Minimally clinically important difference for the UCSD shortness of breath questionnaire, borg scale, and visual analog scale. COPD 2, 105-110.  Ryerson CJ, Camp PG, Eves ND, Schaeffer M, Syed N, Dhillon S, Jensen D, Maltais F, O'Donnell DE, Raghavan N, Roman M, Stickland MK, Assayag D, Bourbeau J, Dion G, Fell CD, Hambly N, Johannson KA, Kalluri M, Khalil N, Kolb M, Manganas H, Moran-Mendoza O, Provencher S, Ramesh W, Rolf JD, Wilcox PG & Guenette JA. (2016). High oxygen delivery to preserve exercise capacity in IPF patients treated with nintedanib: methodology of the HOPE-IPF study. Ann Am Thorac Soc 13, 1640-1647.  Ryerson CJ, Donesky D, Pantilat SZ & Collard HR. (2012). Dyspnea in idiopathic pulmonary fibrosis: a systematic review. J Pain Symptom Manage 43, 771-782.  Ryerson CJ, Urbania TH, Richeldi L, Mooney JJ, Lee JS, Jones KD, Elicker BM, Koth LL, King TE, Jr., Wolters PJ & Collard HR. (2013). Prevalence and prognosis of unclassifiable interstitial lung disease. Eur Respir J 42, 750-757.  Ryerson CJ, Vittinghoff E, Ley B, Lee JS, Mooney JJ, Jones KD, Elicker BM, Wolters PJ, Koth LL, King TE, Jr. & Collard HR. (2014). Predicting survival across chronic interstitial lung disease: the ILD-GAP model. Chest 145, 723-728.  Sandberg D & Fleetham J. (2013). Home oxygen therapy in British Columbia. BC Med J 55, 149-152.  Scano G, Innocenti-Bruni G & Stendardi L. (2010). Do obstructive and restrictive lung diseases share common underlying mechanisms of breathlessness? Respir Med 104, 925-933.  97 Schaeffer MR, Mendonca CT, Levangie MC, Andersen RE, Taivassalo T & Jensen D. (2014). Physiological mechanisms of sex differences in exertional dyspnoea: role of neural respiratory motor drive. Exp Physiol 99, 427-441.  Schaeffer MR, Ryerson CJ, Ramsook AH, Molgat-Seon Y, Wilkie SS, Dhillon SS, Mitchell RA, Sheel AW, Khalil N, Camp PG & Guenette JA. (2017). Effects of hyperoxia on dyspnoea and exercise endurance in fibrotic interstitial lung disease. Eur Respir J 49, 1602494 [dx.doi.org/10.1183/13993003.02494-2016].  Schaeffer MR, Ryerson CJ, Ramsook AH, Molgat-Seon Y, Wilkie SS, Dhillon SS, Mitchell RA, Sheel AW, Khalil N, Camp PG & Guenette JA. (2018). Neurophysiological mechanisms of exertional dyspnea in fibrotic interstitial lung disease. Eur Respir J 51, 1701726 [dx.doi.org/10.1183/13993003.01726-2017].  Sharp C, Adamali H & Millar AB. (2016). Ambulatory and short-burst oxygen for interstitial lung disease. Cochrane Database Syst Rev 7, CD011716.  Simon PM, Schwartzstein RM, Weiss JW, Fencl V, Teghtsoonian M & Weinberger SE. (1990). Distinguishable types of dyspnea in patients with shortness of breath. Am Rev Respir Dis 142, 1009-1014.  Simon PM, Schwartzstein RM, Weiss JW, Lahive K, Fencl V, Teghtsoonian M & Weinberger SE. (1989). Distinguishable sensations of breathlessness induced in normal volunteers. Am Rev Respir Dis 140, 1021-1027.  Smith J, Albert P, Bertella E, Lester J, Jack S & Calverley P. (2009). Qualitative aspects of breathlessness in health and disease. Thorax 64, 713-718.  Somfay A, Porszasz J, Lee SM & Casaburi R. (2001). Dose-response effect of oxygen on hyperinflation and exercise endurance in nonhypoxaemic COPD patients. Eur Respir J 18, 77-84.  Spiro SG, Dowdeswell IR & Clark TJ. (1981). An analysis of submaximal exercise responses in patients with sarcoidosis and fibrosing alveolitis. Br J Dis Chest 75, 169-180.  Swigris JJ, Kuschner WG, Jacobs SS, Wilson SR & Gould MK. (2005). Health-related quality of life in patients with idiopathic pulmonary fibrosis: a systematic review. Thorax 60, 588-594.  Swinburn CR, Mould H, Stone TN, Corris PA & Gibson GJ. (1991). Symptomatic benefit of supplemental oxygen in hypoxemic patients with chronic lung disease. Am Rev Respir Dis 143, 913-915.    98 Tan WC, Bourbeau J, Hernandez P, Chapman K, Cowie R, FitzGerald MJ, Aaron S, Marciniuk DD, Maltais F, O'Donnell DE, Goldstein R, Sin D & LHCE study investigators. (2011). Canadian prediction equations of spirometric lung function for Caucasian adults 20 to 90 years of age: results from the Canadian Obstructive Lung Disease (COLD) study and the Lung Health Canadian Environment (LHCE) study. Can Respir J 18, 321-326.  Travis WD, Hunninghake G, King TE, Jr., Lynch DA, Colby TV, Galvin JR, Brown KK, Chung MP, Cordier JF, du Bois RM, Flaherty KR, Franks TJ, Hansell DM, Hartman TE, Kazerooni EA, Kim DS, Kitaichi M, Koyama T, Martinez FJ, Nagai S, Midthun DE, Muller NL, Nicholson AG, Raghu G, Selman M & Wells A. (2008). Idiopathic nonspecific interstitial pneumonia: report of an ATS project. Am J Respir Crit Care Med 177, 1338-1347.  Visca D, Montgomery A, de Lauretis A, Sestini P, Soteriou H, Maher TM, Wells AU & Renzoni EA. (2011). Ambulatory oxygen in interstitial lung disease. Eur Respir J 38, 987-990.  Vogiatzis I & Zakynthinos S. (2012). Factors limiting exercise tolerance in chronic lung diseases. Compr Physiol 2, 1779-1817.  Wanger J, Clausen JL, Coates A, Pedersen OF, Brusasco V, Burgos F, Casaburi R, Crapo R, Enright P, van der Grinten CP, Gustafsson P, Hankinson J, Jensen R, Johnson D, Macintyre N, McKay R, Miller MR, Navajas D, Pellegrino R & Viegi G. (2005). Standardisation of the measurement of lung volumes. Eur Respir J 26, 511-522.  Wright GW & Branscomb BV. (1954). The origin of the sensations of dyspnea. Trans Am Clin Climatol Assoc 66, 116-125.  Yernault JC, de Jonghe M, de Coster A & Englert M. (1975). Pulmonary mechanics in diffuse fibrosing alveolitis. Bull Physiopathol Respir 11, 231-244.  Zechman FW & Wiley RL. (2011). Afferent inputs to breathing: respiratory sensation. In Compr Physiol, pp. 449–474. John Wiley & Sons, Inc.     

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.24.1-0364703/manifest

Comment

Related Items