UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Pulmonary oedema following exercise in humans Hodges, Alastair Neil Hugh 2006

You don't seem to have a PDF reader installed, try download the pdf

Item Metadata

Download

Media
ubc_2006-130040.pdf [ 6.11MB ]
Metadata
JSON: 1.0077013.json
JSON-LD: 1.0077013+ld.json
RDF/XML (Pretty): 1.0077013.xml
RDF/JSON: 1.0077013+rdf.json
Turtle: 1.0077013+rdf-turtle.txt
N-Triples: 1.0077013+rdf-ntriples.txt
Original Record: 1.0077013 +original-record.json
Full Text
1.0077013.txt
Citation
1.0077013.ris

Full Text

P U L M O N A R Y O E D E M A FOLLOWING E X E R C I S E IN H U M A N S by A L A S T A I R NEIL H U G H HODGES M . A . , M c G i l l University, 2000 B.H.K., The University of British Columbia, 1998  A THESIS SUBMITTED IN P A R T I A L F U L F I L L M E N T OF T H E REQUIREMENTS FOR T H E D E G R E E OF DOCTOR OF PHILOSOPHY in T H E F A C U L T Y OF G R A D U A T E STUDIES (Human Kinetics)  T H E UNIVERSITY OF BRITISH C O L U M B I A February 2006 © Alastair Neil Hugh Hodges, 2006  Abstract In order to determine if transient pulmonary oedema occurs after strenuous exercise, 10 well trained male athletes were challenged in normoxic and hypoxic conditions. To determine the minimal tolerable F1O2 for hypoxia, ten aerobically trained male athletes (VC^max = 57.2 ± 7.95mL-kg" min') performed graded cycling work to maximal effort 1  under four conditions of varying FT0 (21%, 18%, 15%, 12%). Mean VChmax was 2  significantly reduced while breathing 15 and 12% oxygen (VC^max = 48.2 ± 7.9 and 31.5 ± 7.4 mL-kg'-min respectively). In the 12% oxygen condition, the majority of the 1  subjects were not able to complete maximal exercise without SaO"2 falling below 70%. Ten highly trained males (V0 max = 65.0 ± 7.5mL-kg"'-min" ) then underwent 1  2  assessment of lung density by quantified magnetic resonance imaging prior to and 54.0 ± 17.2 and 100.7 ± 15.1 min following 60 min of cycling exercise (61.6 ± 9.5% VO^max). The same subjects underwent an identical measure prior to and 55.6 ± 9.8 and 104.3 ± 9.1 min following 60 min cycling exercise (65.4 ± 7.1% hypoxic VChmax) in hypoxia (F1O2 = 15.0%).  Two subjects demonstrated mild exercise-induced arterial hypoxaemia  (EIAH) (minSa0 = 94.5 & 93.8%), and 7 demonstrated moderate E I A H (minSa0 = 2  2  91.4 ± 1.1%) during a preliminary VC^max test in normoxia. No significant differences (p<0.05) were found in lung density following exercise in either condition. Mean lung densities, measured once pre- and twice post-exercise, were 0.177 ± 0.019, 0.181 ± 0.019 and 0.173 ± O.O^g-mL" in the normoxic condition, and 0.178 ± 0.021, 0.174 ± 0.022 1  and 0.176 ± 0.019g-mL"' in hypoxic condition. These results indicate that transient interstitial pulmonary oedema does not occur following sustained steady-state cycling  exercise in normoxia or hypoxia. This diminishes the likelihood of transient oedema as a mechanism for changes in SaC>2 during sustained exercise.  111  T A B L E OF CONTENTS Abstract  ii  List of tables  v  List of figures  viii  List of symbols, nomenclature and abbreviations  x  Acknowledgement  xi  CHAPTER ONE: General Introduction  1  1.1 Introduction 1.2 Statement of the problem 2.1 Exercise-Induced Arterial Hypoxaemia 2.2 Pulmonary Oedema 2.3 Pulmonary Capillary Leakage 2.4 Pulmonary Capillary Stress Failure 2.5 Assessment of Pulmonary Oedema 2.7 Pulmonary Diffusing Capacity 2.8 Pulmonary Diffusing Capacity & Exercise 2.9 Pulmonary Oedema & Exercise 2.10 Animal Studies 2.11 Human Studies 2.12 Exercise In Acute Hypoxia 2.13 High Altitude Pulmonary Oedema 2.14 Magnetic Resonance Imaging (MRI) 2.15 Conclusion  1 2 3 7 8 9 10 14 15 18 19 22 28 30 30 31  CHAPTER T H R E E : Incremental Exercise Under Varying Hypoxic Conditions... 32 3.1 Introduction 3.2 Methods 3.3 Results 3.4 Discussion  32 33 35 42  CHAPTER FOUR: Effects of Prolonged Exercise on Extravascular Lung Water . 48 4.1 Introduction 4.2 Methods 4.3 Results 4.4 Discussion  ;  48 51 59 71  CHAPTER FIVE: General Summary and Conclusions  80  REFERENCES  82  APPENDIX A. Individual data from Chapter 3  95  APPENDIX B. Individual exercise data from Chapter 4  99  APPENDIX C. Individual data: lung densities, gradients, & areas  104  iv  List of tables Table 2.1. Summary of major extravascular water studies  28  Table 3.1. VC^max, heart rate, peak power, and minimum SaO"2 during maximal exercise in various conditions of hypoxia (means ± SD); N = 10  39  Table 3.2: Ventilatory data during maximal exercise in various conditions of hypoxia (means ± SD); N = 10  39  Table 4.1. Maximal exercise data (means ± SD); N = 10  59  Table 4.2. Exercise intervention data averaged over the last 50 min (means ± SD); N = 10 Table 4.3: Total work performed in 60 minutes (means ± SD); N = 10  62 62  Table 4.4: Lung densities (density threshold = 0.3 g-mL" ) pre and post-exercise intervention (means ± SD); N = 10  66  Table 4.5: Heart rate (HR), mass, and time data at initiation of each M R scan (means ± SD); N = 10  67  Table 4.6. Lung slice areas (density threshold = 0.3 g-mL" ) pre and post-exercise intervention (means ± SD); N = 10  69  1  1  Table 4.7. Lung density gradients (density threshold = 0.3 g-mL" ) pre and post-exercise intervention (means ± 1  S D ) ; N = 10  70  Table A . l . Individual descriptive data  95  Table A.2. Individual maximal exercise data (F1O2 = 21%)  95  Table A.3. Individual maximal exercise data (F1O2 = 18%)  96  Table A.4. Individual maximal exercise data (F1O2 = 15%)  96  Table A.5. Individual maximal exercise data (F1O2 = 12%)  97  Table B . l . Individual descriptive and spirometry data  99  Table B.2. Individual maximal exercise data (F1O2 = 21%)  99  v  Table B.3. Individual maximal exercise data (F1O2 = 15%)  100  Table B.4. Individual exercise intervention data (normoxic condition, averaged over last 50 min; mean ± SD)  100  Table B.5. Individual exercise intervention data (hypoxic condition, averaged over last 50 min; mean ± SD)  101  Table B.6: Individual exercise intervention data (total work performed)  101  Table C . l . Individual lung densities (g-mL" ) with a density threshold = 0.25 g-mL"  104  Table C.2. Individual lung densities (g-mL" ) with a density threshold = 0.30 g-mL' :  105  Table C.3. Individual lung densities (g-mL" ) with a density threshold = 0.35 g-mL"  105  Table C.4. Individual lung densities (g-mL" ) with a density threshold = 0.40 g-mL"  106  1  1  1  1  1  1  1  1  Table C.5. Individual lung densities (g-mL" ) with a density threshold = 0.45 g-mL" 1  1  :  106  Table C.6. Individual lung densities (g-mL" ) with a density threshold = 1.0 g-mL"  107  Table C.l. Individual lung density gradients (mg-mL" -cm" ) with a density threshold = 0.25 g-mL"  107  Table C.8. Individual lung density gradients (mg-mL" -cm" ) with a density threshold = 0.30 g-mL"  108  Table C.9. Individual lung density gradients (mg-mL" -cm" ) with a density threshold = 0.35 g-mL"  108  1  1  1  1  1  1  1  1  1  1  1  Table C I O . Individual lung density gradients (mg-mL" -cm" ) with a density threshold = 0.40 g-mL"  '.. 109  Table C. 11. Individual lung density gradients (mg-mL" -cm" ) with a density threshold = 0.45 g-mL"  109  Table C. 12. Individual lung density gradients (mg-mL" -cm" ) with a density threshold =1.0 g-mL"  110  Table C.13. Individual lung slice areas (mm )with a density threshold = 0.25 g-mL"  Ill  1  1  1  1  1  1  1  1  1  2  1  vi  Table C.14. Individual lung slice areas (mm )with a density threshold = 0.30 g-mL"  Ill  Table C.15. Individual lung slice areas (mm )with a density threshold = 0.35 g-mL"  112  Table C. 16. Individual lung slice areas (mm )with a density threshold = 0,40 g-mL"  112  Table C.17. Individual lung slice areas (mm )with a density threshold = 0.45 g-mL"  ....113  Table C.18. Individual lung slice areas (mm )with a density threshold = 1.0 g-mL"  113  2  1  2  1  2  1  2  1  2  1  vii  List of figures Figure 2.1. Radiograph of pulmonary oedema  11  Figure 2.2. CT of pulmonary oedema  12  Figure 3.1. Peak power / peak normoxic power (mean ± SD); N = 10  36  Figure 3.2. V 0 m a x with varying F r 0 (mean ± SD); N = 10  37  Figure 3.3. Minimal SaC>2 with varying F i 0 (mean ± SD); N = 10  38  Figure 3.4. Mean decline in VO"2max (for all levels of hypoxia) vs. normoxic VChmax, R = - 0.41, p = 0.23  41  2  2  2  Figure 3.5. Mean decline in VO"2max (for all levels of hypoxia) vs. normoxic Sa0 , R = 0.48, p = 0.16  42  2  Figure 3.6. Hypoxic VChmax / normoxic VC^max vs. F1O2 Figure 4.1. Timeline of experimental protocol for each • condition (normoxia and hypoxia) Figure 4.2. Arterial oxyhaemoglobin saturation during normoxic VC^max test  43 52 60  Figure 4.3. Arterial oxyhaemoglobin saturation during hypoxic VC^max test  61  Figure 4.4. Normoxic exercise data. (Means ± SD); N = 10  63  Figure 4.5. Hypoxic exercise data. (Means ± SD); N = 10  64  Figure 4.6. Sa0 during the exercise intervention. (Means ± S D ) ; N = 10 Figure 4.7 Lung densities (density threshold = 0.3 g-mL" ). (Means); N = 10 2  65  1  Figure 4.8. (a) Sample M R image of right sagittal lung slice, (b) - (g): Computerized image of the lung scan showing areas of pixel removal. Dark areas within the lung represent areas of pixel removal using a threshold of (b) 0.25 g-mL" (c) 0.3 g-mL" (d) 0.35 g-mL" (e) 0.4 g-mL" (f) 0.45 g-mL" (g) 1.0 g-mL" 1  1  1  1  66  1  1  68  vm  Figure 4.9. Sample lung density gradient for the resting normoxic condition with a threshold density of 0.3 g-mL-'  71  Figure A . l . Individual minimum SaC>2 with varying F1O2  98  Figure B. 1. Individual lung densities following normoxic exercise (density threshold = 0.3 g-mL"')  102  Figure B.2. Individual lung densities following hypoxic exercise (density threshold = 0.3 g-mL" )  103  Figure C . l . Infra-observer reliability of lung density, r = 0.98, p< 0.05, N = 20  114  Figure C.2. Inter-observer reliability of lung density r = 0.98, p< 0.05, N = 20  115  Figure C.3. Lung density vs. lung density threshold  116  1  ix  List of symbols, nomenclature and abbreviations  ATP  a-v0 diff. 2  BAL CT  DLco D M  ECG EIAH EVLW FEVi  F,0  2  FVC HAPE HRCT HRmax MIGET MRI  PaC0 Pa0  2  2  PAP PEFR  Qc R f  ROI RV  Sa0  2  TE TLC TR Tvent  V V /Qc A  A  V  E  V V0 V0 max c  2  2  V  T  adenosine triphosphate arterial-venous oxygen difference broncho-alveolar lavage computerized tomography diffusing capacity of carbon monoxide diffusing capacity of the alveolar-capillary membrane electrocardiograph exercise induced arterial hypoxaemia extravascular lung water forced expiratory volume in one second fraction of inspired oxygen forced vital capacity high altitude pulmonary oedema high-resolution computerized tomography maximal heart rate multiple inert gas elimination technique magnetic resonance imaging arterial partial pressure of carbon dioxide arterial partial pressure of oxygen pulmonary artery pressure peak expiratory flow rate cardiac output respiratory frequency region of interest residual volume arterial oxyhaemoglobin saturation time to echo total lung capacity repetition time ventilatory threshold alveolar ventilation ventilation / perfusion ratio ventilation pulmonary capillary blood volume oxygen consumption maximal oxygen consumption tidal volume  Acknowledgement  I wish to acknowledge the involvement and assistance of my supervisory committee: Dr. A. William Sheel, Dr. John Mayo, Dr. Jack Taunton, and the late Dr. Peter Hochachka. I also wish to express my deepest gratitude to my supervisor Dr. Donald McKenzie for his guidance and inspiration throughout this work and my entire doctoral degree.  I appreciate the participation of the subjects in these studies, all of whom donated a significant amount of time and effort with a minimum of problems or complaints. Dr. Alex MacKay and Mr. Thorarin Bjarnason of the Department of Physics & Astronomy at the University of British Columbia provided valuable assistance with the M R imaging and analysis setup. Additionally, the help of the following M R I technologists at the University of British Columbia Hospital is greatly appreciated: Lesley Costley, Monique Genton, Trudy Harris, Ashwani Kumar, Jennifer McCord, Sylvia Renneburg, Martin Sherriff, Karen Smith, and Linda Zimmer. Benjamin Sporer, Kirstin Lane, and Dr. Michael Koehle were an integral part of the data collection process of part of this work. As with all research conducted at the Allan McGavin Sports Medicine Centre, Diana Jespersen has been indispensable in helping these studies to run smoothly.  Funding for the work in this document was provided by the British Columbia Lung Association and the British Columbia Sports Medicine Foundation.  xi  CHAPTER ONE: General Introduction  1.1 INTRODUCTION The membrane separating the pulmonary capillaries from the alveoli is a remarkable structure in the human lung. The exceptionally thin yet strong nature of this membrane allows gas exchange while maintaining structural integrity under the mechanical force of the pulmonary circulation. Normally, in healthy humans, this structure functions throughout life to allow maintenance of arterial PO2 and P C O 2 within the ranges appropriate to normal physiological function.  There are, however, circumstances such as disease and exposure to altitude, in which the normal function of this membrane is impaired, in particular due to disruption of the structural integrity of the membrane, or accumulation of fluid within the interstitial space (interstitial pulmonary oedema). The focus of this manuscript is the examination of pulmonary oedema in relation to one particular circumstance known as exercise-induced arterial hypoxaemia (EIAH), which sometimes occurs during exercise in aerobically fit individuals.  Chapter two provides a brief review of the literature to date on the topics relevant to this dissertation including exercise-induced arterial hypoxaemia and transient pulmonary oedema following exercise in animals and humans. Chapter three describes an experiment involving the examination of incremental cycling exercise under several  1  conditions of hypoxia. This was necessary to determine the tolerable level of hypoxia that would allow sustained high-intensity exercise. Chapter four describes an experiment designed to assess the effects of exercise, both in normoxia and hypoxia, on the development of E I A H and extravascular lung water measured as lung density by magnetic resonance imaging. Chapter five provides a summary and conclusion of the dissertation.  1.2 S T A T E M E N T O F T H E  PROBLEM  Transient pulmonary oedema leading to diffusion limitation represents a possible mechanism for EIAH. Examination of transient pulmonary oedema in humans has been difficult due to the indirect nature of the measures used, and has led to inconclusive results following exercise intervention. In recent years magnetic resonance imaging (MRI) technology has allowed indirect assessment of lung density to a reasonably sensitive level. This technology allows detection of changes in lung density likely to be reflective of transient pulmonary oedema if it occurs during and following exercise. Before making the case for pulmonary oedema as a mechanism for EIAH, the first step is to establish whether it occurs in a predictable and quantifiable manner. Therefore, the purpose of the following studies is to examine the effects of exercise on transient pulmonary oedema as measured by the most sensitive tool available to date.  2  'CHAPTER TWO: Review of Literature  During continuous intense exercise the ability of the skeletal locomotor muscles to perform work is limited by the regeneration of ATP through oxidative phosphorylation. The proposed limiting factors in this process have been a focus of research during the past century, and have included pulmonary diffusing capacity, cardiac output, skeletal muscle capillary proliferation, and skeletal muscle mitochondrial capacity. As a result of experimental research over the past 30 years, cardiac output has become the most commonly accepted limit to continuous whole-body skeletal muscle work in healthy athletic i n d i v i d u a l s ' ' ' 8  10  21  102  '  107  and the pulmonary system has been accepted as a non-  limiting system in healthy humans during exercise at sea level.  23  2.1 EXERCISE-INDUCED ARTERIAL HYPOXAEMIA Oxyhaemoglobin saturation is maintained within 2 - 4% of resting values during strenuous aerobic exercise in normal healthy human subjects. However, arterial hypoxaemia has been observed during exercise in aerobically trained athletes. This phenomenon has been observed almost half a century a g o ' 53  101  and well described over  the past 15 years, and is termed exercise-induced arterial hypoxaemia (EIAH). The incidence of E I A H is often accepted as approximately 50% of elite male endurance athletes, 9 9  24,90  but may occur more frequently and at lower aerobic work rates in females. '  There is variation between individuals in the degree of severity of E I A H experienced  during exercise and, although E I A H is often considered to be a phenomenon that occurs ' A version of this chapter has been accepted for publication. Hodges, A.N.H., Mayo, J.R., and McKenzie, D.C. (2006) Pulmonary oedema following exercise in humans. Sports Medicine. 3  near maximal work loads, it has also been demonstrated during moderate work. Dempsey and Wagner have proposed three levels of EIAH as defined by the degree of 25  arterial haemoglobin saturation: mild (93-95%), moderate (88-93%), and severe (< 88%). Despite extensive research the mechanism responsible for E I A H has not been completely established. Those most likely to contribute are right to left shunts, relative alveolar hypoventilation, ventilation/perfusion (V /Qc) mismatch, and diffusion limitation. A 25  A  significant amount of research has been dedicated to the study of E I A H in the past two decades with advances in the understanding of the nature of this phenomenon and the mechanisms involved.  Right to left, or venoarterial shunts allow the passage of venous blood into the arterial system causing a slight drop in the arterial oxygen content. Venoarterial shunts are the result of bronchial and Thebesian venous drainage into the left heart and represent 1 3% of cardiac output at rest. It has been found that breathing 100% 0 during exercise 22  2  generally eradicates E I A H suggesting that venoarterial shunts are not the mechanism responsible. ' ' ' 24  44  117  123  Recently, however, two studies involving the use of agitated  saline contrast echocardiography have provided new evidence of intrapulmonary shunting of blood during exercise, in the form of contrast bubbles appearing in the left heart following injection in a systemic v e i n . ' 31  114  Relative alveolar hypoventilation is defined as ventilation below the rate required to maintain arterial blood gases at normal values  and is generally accompanied by an  increased PaCO"2. Dempsey et a l have noted that little or no hyper-ventilatory response 24  4  occurred in the most hypoxemic subjects in their study. Powers et al  have investigated  the role of inadequate hyperventilation in EIAH by examining arterial blood gases and pulmonary gas exchange in cyclists during incremental and sustained exercise while varying the fraction of inspired oxygen (F1O2) and concluded that relative hypoventilation is not the major cause of EIAH. However, there is evidence that relative hypoventilation may contribute to E I A H in some athletes during heavy and submaximal exercise. '  It is likely that another mechanism contributes to E I A H even if relative  28 51  hypoventilation is a factor.  During exercise there is an increased A-aD02, typically caused by V / Q c mismatch, A  right to left shunting of blood, or diffusion limitation.  25  VA/QC  mismatch maybe  responsible for 50% of the A-aD0 at rest , and increases with exercise increases 94  2  123  with  exercise intensity to explain 60% of the A-aDO"2 during moderate to severe exercise. '  44 56  It is speculated that pulmonary diffusion limitation has an increased contribution to A aD02 near maximal exercise. However, the mechanism of increase in V A / Q C mismatch 94  with exercise remains unclear. The investigation of the contribution of V / Q c to a A  widening A-aD0 during exercise is best assessed with the multiple inert gas elimination 2  technique (MIGET), in which inert gases are infused into the blood and the pattern of removal through the lungs is observed and recorded. In their review of EIAH, Dempsey and Wagner have outlined several possible mechanisms for an increased V / Q c 25  A  mismatch during exercise including minor structural differences in airways and blood vessels, bronchoconstriction, airway secretions, variations in the modulation of airway and vascular tone, and mild interstitial oedema.  5  A limitation in the diffusion of oxygen from the alveoli to the red blood cell could contribute to EIAH, and could occur primarily in one of two ways: inadequate red blood cell transit time in the pulmonary capillaries, or an increase in the thickness of the alveolar pulmonary capillary membrane. Hopkins et a l have given evidence of a 54  relationship between shortened pulmonary transit time (2.91 ± 0.3 seconds during maximal exercise vs. 9.32 ± 1.42 seconds at rest) diffusion limitation in subjects with evidence of diffusion disequilibrium from MIGET. In contrast, Warren et a l  124  examined  pulmonary capillary blood volume (Vc), A-aD02, and mean transit time during varying exercise intensities and concluded that the decrease in Pa02 is not due to a plateau in pulmonary capillary blood volume and reduction in pulmonary transit time. Clearly examination of the relationship between pulmonary transit time and diffusion limitation is not precise and the conclusions are at least partly speculative. Powers et a l have 91  investigated the effects of mildly hyperoxic (26% oxygen) exercise on two groups of athletes defined according to aerobic fitness and demonstration of E I A H with the finding that V02max was increased only in the more highly trained group. This is taken as evidence that pulmonary gas exchange may contribute to the limitation of V02max in subjects with EIAH. In females, an increase in V0 max has been observed in a majority 2  of subjects (22 of 25) when hyperoxic gas (26% oxygen) was administered during exercise, with approximately a 2% increase in V02max for every 1% decrease in the arterial saturation below resting values. Hopkins et al have found that the integrity of 49  the pulmonary blood-gas barrier is impaired during intense but not during sustained 57  CO  submaximal  exercise. This is consistent with pulmonary capillary stress failure theory  discussed below which may lead to diffusion limitation during exercise. Edwards et a l  29  6  concluded that alveolar epithelial integrity is maintained during exercise suggesting that any compromise of the alveolar-pulmonary capillary membrane occurs on the capillary side. Two further studies by Hopkins et a l ' 5 5  5 6  have produced results consistent with  diffusion limitation as a contributor to EIAH.  EIAH has been shown to have an affect on maximal oxygen uptake (VOamax) during 68 77 91  exercise ' '  and therefore, although the pulmonary system does not normally limit  exercise performance, it may do so with the presence of EIAH. As described above, pulmonary diffusion limitation has been proposed as one of the contributing mechanisms to EIAH, but the nature of this limitation has remained elusive due to the difficult and invasive nature of investigation in this area. While it is possible that a combination of the mechanisms discussed above contribute to EIAH in different subjects and under different circumstances, a single mechanism has yet to be isolated as the major contributor to EIAH in highly trained athletes during severe exercise. Given the research conducted over the past several years, diffusion limitation has not been ruled out and remains a likely candidate.  2.2 P U L M O N A R Y O E D E M A Excluding blood, the lung is approximately 80% water by weight. Between 30 - 50% of this water is extra cellular consisting of interstitial fluid and lymph.  109  Interstitial  pulmonary oedema is the development of additional extravascular lung water (EVLW) caused by an increase in the filtration of fluid from the pulmonary capillaries into the interstitial space between the alveolar epithelium and capillary endothelium. Pulmonary 7  oedema may occur through changes in the Starling forces (pulmonary capillary leakage) and/or changes in the permeability of the capillary membrane (pulmonary capillary stress failure).  2  Clinically, pulmonary oedema is associated with raised left atrial pressure, intravascular coagulation, or the release of vasoactive substances associated with shock. Microvascular permeability and lymphatic flow may also contribute to pulmonary oedema, but the proposed mechanism during intense exercise is increased pulmonary pressure (> 20 mmHg). A larger increase (> 30 mmHg) in pulmonary pressure may lead to fluid accumulation in the alveoli space (alveolar oedema). Clinical symptoms of pulmonary 69  oedema include coughing, cyanosis, and dyspnoea.  2.3 P U L M O N A R Y C A P I L L A R Y L E A K A G E The hydrostatic and osmotic pressures in the pulmonary capillaries and the interstitial space control fluid movement across the capillary alveolar membrane according to the Starling equation:  113  Q  f  = K [(P f  m v  - Ppmv) -  a  (n - n )] mv  pmv  Where Q f is net water flow out of the vascular compartment, K f is the membrane filtration coefficient, P  m v  is the hydrostatic pressure in the microvasculature, P p  m v  is the  interstitial hydrostatic pressure, a is the microvascular membrane coefficient for plasma proteins, n  mv  is the osmotic pressure in the microvasculature, and Tl  pmv  is the interstitial  8  osmotic pressure.  109  Thus, net water flow is governed by the differences in hydrostatic  and osmotic pressures.  Estimates of the hydrostatic pressures of the pulmonary capillaries and the interstitial space are 10 mm Hg and -3 mm Hg respectively, while the corresponding osmotic pressures are estimated as 25 mm Hg and 19 mm H g . ' 2  resting circumstances, Q f is positive  109  60  It is likely that under normal  indicating some filtration of fluid out of the  pulmonary capillaries at an estimated rate of 20 ml-hr" in normal conditions. 1  126  An  added complication in the practical use of the Starling equation is the discrepancy in pressures between the apices and bases of the lungs.  126  Clearance of alveolar pulmonary oedema is a more complicated process than either a simple reversal of the Starling forces associated with alveolar epithelial leakage or a reduction in pulmonary capillary pressure associated with pulmonary capillary stress failure. Clearance of interstitial oedema involves drainage through lymphatic ducts but alveolar oedema clearance is achieved through an active pumping process involving a 126  sodium-potassium ATPase pump in alveolar epithelial cells.  2.4 P U L M O N A R Y C A P I L L A R Y STRESS F A I L U R E Pulmonary capillary stress failure is a condition that occurs as a result of structural changes in the capillary alveolar membrane.  125  Stress failure may occur in humans as a  result of increased pulmonary capillary pressure in a number of circumstances: severe exercise, altitude exposure, and disease such as left ventricular dysfunction and mitral  stenosis. The capillary alveolar membrane is an exceptionally thin structure (0.2 - 0.3 um) comprised of three layers: capillary endothelium, alveolar epithelium, and a basement membrane or extra cellular matrix between the endothelial and epithelial layers. There is some evidence that the basement membrane provides a large portion of the strength of the entire membrane structure,  118  indicating that structural changes to the  basement membrane may alter the behaviour of the entire structure. Type TV collagen fibres synthesized by epithelial and endothelial cells in the basement membrane likely provide the strength of this structure.  128  The Type TV collagen fibres are arranged in  lattice-shaped structures that may be temporarily altered upon introduction of stress forces. This temporary change in the collagen structure may account for a change in permeability of the membrane with raised pulmonary capillary pressure. The strength of 32  the capillary alveolar membrane is demonstrated by its ability to maintain integrity during all but the most severe physiological stresses including maximal exercise in highly trained humans and animals.  2.5 A S S E S S M E N T O F P U L M O N A R Y O E D E M A Clinically, pulmonary oedema is observed with radiography (Figure 2.1) or computerized tomography (CT) scan (Figure 2.2). Several methods have been used in an attempt to detect and quantify transient pulmonary oedema including wet/dry lung weight ratio, chest radiography, presence of rapid shallow breathing, and MRI. O f these, M R I provides  10  Figure 2.1. Radiograph of pulmonary oedema. Thickened lymphatic channels in the interlobular septa (arrows) can only be seen in tangent in the periphery of the lung on this chest radiograph of a 61 year old male with mild cardiac failure. The 2 dimensional viewing perspective and the limited soft tissue contrast of the chest radiograph make it impossible to appreciate thickened interlobular septa in other regions of the right lung base.  11  Figure 2.2. CT of pulmonary oedema. Thin section (1mm) CT scan of the same patient as Figure 2.1 within 1 hour of radiography better demonstrates distended lymphatic channels in the interlobular septa (arrows) all around the right lung base. The cross sectional perspective and improved soft tissue contrast provided by CT imaging enhances visualization of the increased interstitial fluid in the lung.  12  the greatest potential for detection of the limited oedema likely to be present following severe exercise. High-resolution CT (HRCT) has been used in the evaluation of exerciseinduced pulmonary oedema in cardiac patients. HRCT was used before and following 12  symptom limited exercise (Bruce treadmill protocol) in 2 groups of 10 patients with 13  chronic congestive heart failure and 10 healthy controls. Significantly more visual abnormalities were seen in the heart failure patients than controls and it was concluded that HRCT was a useful tool in the assessment of interstitial pulmonary oedema.  Comparison of three-dimensional M R I and conventional methods (multiple-indicator dilution) of assessing lung water was undertaken by Caruthers et a l following lung 19  injury induction in dogs. Results suggested that the M R I technique used was sensitive to changes in lung water and that the method may be used to assess the time course of oedema formation. M R I has also been used to examine pulmonary inflammation in rats due to allergen exposure. '  9 116  It was found that the M R I signals correlated significantly  with inflammatory parameters determined by broncho-alveolar lavage (B A L ) and the authors conclude that M R I provides a measure of lung oedema and the pulmonary inflammatory response following allergen challenge.  M R I has been used specifically to measure water content in the lung in a study by Estilaei et al. Nineteen juvenile pigs were scanned and the in vitro water content 35  measured by M R I was strongly correlated (r = 0.98) with gravimetric measurements. 2  13  Using the same M R I scanner and sequence, McKenzie et a l  84  found a significant increase  in E V L W following exercise in 8 human subjects.  There is a moderate amount of research that contributes to the literature on diffusion limitation following exercise. The remainder of this review is divided into two main areas: pulmonary diffusing capacity, and pulmonary oedema and the effects of exercise on each. The former area has been the focus of precisely controlled experiments that have come to a reasonable consensus, while the latter has remained a more difficult area to study with no clear conclusions yet on the existence or nature of oedema following exercise.  2.7 P U L M O N A R Y  DIFFUSING  CAPACITY  The rate of diffusion for a gas (Vgas) across the capillary alveolar membrane is determined by the following Fick equation and is controlled by the concentration gradient (Pi - P2), membrane area (A), membrane thickness (T), and a diffusion constant (D):  126  Vgas = ( A / T ) x D x ( P i - P ) 2  During exercise these factors may be influenced by changes in alveolar or capillary gas concentrations, V / Q c inequalities, and the development of pulmonary oedema A  respectively. In the lung, diffusing capacity (D ) takes area, thickness, and the diffusion L  constant into account  126  and may be partitioned into the diffusing capacity of the  alveolar-capillary membrane ( D ) and capillary blood volume ( V ) . M  c  61  Total resistance  14  (1/DL) is the sum of membrane resistance (1/DM) and red cell resistance (l/0Vc) such that:  i/D =i/D L  M  + i./ev  61  c  The measurement of pulmonary diffusing capacity is a relatively easy and non-invasive procedure in the laboratory. Carbon monoxide (CO) is a gas that has a high affinity for haemoglobin and therefore is diffusion limited,  which makes it useful for the  measurement of pulmonary diffusing capacity. The single breath method involves the inhalation of a trace amount of CO, followed by a 10 s breath hold. Upon exhalation the rate of diffusion is calculated from the difference in CO concentration in the inhaled and exhaled breaths.  2.8 P U L M O N A R Y D I F F U S I N G C A P A C I T Y & E X E R C I S E Normally the diffusion rate for CO (DL o) is 25 ml-min^-mmHg" at rest, and may 1  C  increase by 2 or 3 fold during exercise  126  primarily as a result of changes in the  VA/QC  relationship and pulmonary capillary recruitment and distension. Diffusion limitation has been proposed as a contributing mechanism to E I A H , but the role of pulmonary oedema 25  in this limitation remains unclear. While diffusing capacity is reduced following exercise, ' ' ' ' ~ ' 46  47  73  83  95  97  108  this reduction has been associated with a decreased Vc rather  than a reduction in DM, which would be more indicative of pulmonary oedema ' ' . A 46  48  84  number of studies have examined pulmonary diffusing capacity following high intensity exercise.  15  Rasmussen et al have reported a 6.7% decline in D 2.1 hr following a short maximal L  bout of arm exercise. A reduction in D of 15% has been demonstrated independent of 96  L  exercise mode (maximal arm cranking, treadmill running, or ergometer rowing) 2 - 3 hr post-exercise. The decreased D L has been shown to persist for 2.5 days following 97  intense rowing exercise.  95  A study involving 12 elite handball players examined the effects of maximal exercise (incremental cycling) on membrane diffusing capacity ( D ) . M  73  Pulmonary capillary blood  volume (Vc) and D were determined with the single breath method prior to and for 30 M  minutes following progressive maximal exercise. DLco was found to decrease significantly up to 30 minutes post-exercise, and at 30 minutes returned to the control resting value.  Two studies by Hanel et a l ' 4 6  4 7  suggest that the drop in diffusion capacity post-exercise is  the result of a change in central blood volume rather than pulmonary oedema. In the first study, DLco was measured following sub-maximal rowing and maximal rowing. Diffusion capacity was decreased by 6% and 10% following 6 minutes of exercise at 61%) and 76% VC^max respectively, and by 7%, 8%, and 7% following maximal rowing for one, two, and three minutes respectively. The authors concluded that the decrease in DLco following even a short-term exercise bout is unlikely due to change in the pulmonary capillary membrane integrity. The second study involved the measurement of DLco following exercise in 21 subjects divided into three groups. The first group  16  performed two all-out rows on an ergometer for six min, the second group performed one all-out rowing bout followed by the administration of a diuretic (10 mg furosemide) 150 min post-exercise, and the third group performed one all-out rowing bout and served as controls. It was found that DLco was reduced following exercise from a median of 37 to 34 ml-min^-mrnHg" . Both D and V c were reduced, and a second bout of exercise did 1  M  not change DLco or D M - Administration of furosemide did not affect DLco-  McKenzie et a l  83  examined the effects of repeated exhaustive cycling on E I A H in 13  athletic male subjects. D L o was 36.3 ± 4.6, 32.4 ± 6.0, and 30.4 ± 5.4 ml-min"'•mmHg'  1  C  at rest and following the first and second bouts of exercise respectively. D and V c were M  also progressively reduced with each bout of exercise. No significant difference was found between the minimum SaC>2 in the two tests, and therefore the authors concluded that, if pulmonary oedema developed during the initial exercise bout, it was of no clinical significance. Further, the authors suggest that the changes in DLco may reflect changes in blood flow rather than oedema.  Hanel et a l hypothesized that the decreased DLco following exercise is partly the result 48  of a redistribution of blood away from the central vascular bed. DLco measures were made in nine male oarsmen prior to and 120 minutes following six minutes of maximal rowing exercise, and in six male controls with no exercise intervention. Blood volume and thoracic and thigh blood flow activity was imaged by labelling with  9 9 m  TC  pertechnetate. DLco was decreased by 6% post-exercise and was unchanged in the  17  control group, and a shift in fluid from the thoracic to peripheral area occurred confirming the hypothesis.  A study by Sheel et a l  108  examined the time course of DLco 1,2,4,6, and 24 h post-  exercise in three groups of subjects defined according to aerobic fitness (group one > 65, group two 50 - 60, group three < 50 ml-kg^-min" ). The exercise intervention consisted 1  of high intensity cycling. DLco was decreased one hour post-exercise, with a minimal value of 88% of baseline values at 6 h post-exercise. DLco following 24 h was not different than baseline values. The authors conclude that the decreased DLco following intense exercise is largely due to a decreased Vc.  2.9 P U L M O N A R Y O E D E M A & E X E R C I S E During exercise there is an observed increase in the alveolar-arterial oxygen difference (A-aD02) resulting from V / Q c mismatch and diffusion disequilibrium. 4  56  117  A  It has been  hypothesized that pulmonary oedema may contribute to the V / Q c mismatch during A  75, 104  exercise.  103  Pulmonary oedema has been demonstrated following exercise in animals.  There is  evidence that during severe exercise in Thoroughbred horses, the high pulmonary 129  capillary pressures cause stress failure and interstitial oedema.  Evidence in humans is  less consistent and, while there is evidence of membrane damage and/or pulmonary 57  oedema following both severe  82 84  and prolonged exercise, '  it is not clear that all  18  humans demonstrate this phenomenon nor, when present, is it clear the degree to which it may contribute to impairment of gas exchange.  There is limited data on the study of pulmonary oedema in animals and humans following exercise. The most accurate measures involve invasive techniques that often limit study of this phenomenon in humans. (See Table 2.1 for a summary of the following findings).  2.10 A N I M A L STUDIES Marshall et a l  75  examined canine lungs for changes in lung water by weighing of lung  tissue drained of blood. Four dogs performed maximal treadmill exercise (running at 10 km-h" for 20 min) and four dogs served as resting controls. Immediately following 1  exercise the animals were sacrificed and the lungs were removed and prepared for weighing. No significant difference was found in lung water between exercised and control animals.  In a study on the detection of pulmonary oedema in dogs through the use of radiography, densitometry and lung water, Snashall et a l  1 1 0  found a positive correlation between  radiological grade and lung water. Pulmonary oedema was induced in three groups of dogs with the intention of determining the ability to detect extravascular lung water by chest radiography. Intravenous injection of Alloxan to increase micro vascular permeability was used in one group, extra cellular fluid volume expansion with Hartmarm's solution was used in a second group, and the third group underwent  19  pulmonary angiography with 70 ml of sodium iothalamate. A fourth group of dogs acted as controls. In addition to radiography, the lungs were removed and the extravascular water/dry lung weight ratio was measured. Further, change in the opacity of the radiograph films was assessed. In the Alloxan group, perivascular cuffs were observed in all dogs and peribronchial cuffs in cases of severe oedema, and mean extravascular water / dry lung weight ratio was 8.9 ± 3.8. In the angiography group, one dog developed severe oedema in one lung, three of the dogs were normal, and small perivascular cuffs were visible in the final subject. The mean extravascular water / dry lung weight ratio was 4.8 ± 0.6 in this group. In the Hartmann's solution group perivascular cuffing was seen in all dogs and two dogs had peribronchial cuffs. Mean extravascular water / dry lung weight ratio was 7.3 ± 4.6. In the control group macroscopic examination of the frozen lungs was normal in all animals. Microscopically, perivascular cuffs were seen in two animals and the mean extravascular water / dry lung weight ratio was 4.1 ± 0.6 in the upper lobes. The authors conclude that the radiograph is a sensitive method for the detection of acute pulmonary oedema with an increase in extravascular water of greater than 35% being detectable as oedema by chest radiography. With any measure that relies on the subjective analysis of an observer, the reproducibility of the measures are important. In this study, there were no statistically significant differences between the two radiologists' assessments of the lower zones of the lungs, but their upper zone assessment did differ significantly, which may call into question the reliability of this conclusion. However, the authors concluded that this method may be appropriate for the detection of clinical pulmonary oedema in humans.  20  Examination of swine has provided evidence of pulmonary oedema following heavy exercise. In a study involving high-intensity treadmill exercise,  five pigs ran at the  highest speed maintainable for 6 - 7 minutes while five pigs acted as controls. Control and exercised animals had identical previous treadmill training. Upon completion of the exercise period, the animals were sacrificed and four lung tissue blocks from each animal were prepared for microscopic examination and documentation of perivascular or peribronchial cuffing. The exercised animals showed significantly more periarterial cuffing in general, as well as a higher percentage of periarterial cuffing in the lower lobes and ventral areas of the lung than the control animals. Interstitial oedema is characterized by peribronchial and perivascular engorgement, and the authors suggest this, and the 126  observation of cuffing primarily around the larger vessels, provides evidence of early stage interstitial oedema and conclude that mild pulmonary oedema can occur during heavy exercise in the pig.  Manohar has extensively researched the effects of exercise on Thoroughbred horses. In a study involving repeat treadmill exercise bouts in seven Thoroughbred horses, Manohar et a l reported significantly higher PaC>2 during the second exercise performed at the 74  same workload (galloping at 14 m-s" on a 3.5% uphill grade) following six minutes of 1  rest. These results suggest that a structural change is not the mechanism for E I A H in these animals, or at least that there is no worsening of oedema during subsequent exercise.  21  2.11 H U M A N STUDIES Evidence of increases in lung water following exercise in humans is inconclusive. While there is evidence suggesting that transient pulmonary oedema may occur, direct measurement has proven difficult and studies have drawn conclusions in both directions about the occurrence of oedema following exercise in humans.  One of the earliest studies on pulmonary oedema in humans involved the case studies of two ultra-marathon runners following completion of a 90-km race. Both athletes 82  demonstrated bilateral pulmonary consolidation, upper lobe venous congestion and cardiomegaly with no evidence of hypertrophic cardiomyopathy, mitral stenosis, or aortic valve disease. The authors suggest that the radiological findings, in conjunction with the lack of disease, are likely due to cardiogenic pulmonary oedema following exercise of an extreme duration (9.4 and 9.9 hrs for athletes one and two respectively).  Buono et a l have conducted a series of studies to examine the changes in residual 15  volume and total lung capacity following maximal exercise shown in an earlier study.  14  In the first study, trans-thoracic electrical impedance was decreased following 30 minutes of recovery from exercise suggesting an increase in intra-thoracic fluid. The second study examined the effects of G-suit use on central blood volume and post-exercise R V and • found that the post-exercise increase in R V was not due to increased thoracic blood volume. In the third study, no decrease in DLco / VA was observed despite an increased post-exercise heart rate. The authors conclude that pulmonary oedema is present following exercise and may contribute to the post-exercise increase in R V and TLC.  22  Schaffartzik and colleagues  104  examined the  VA/QC  relationship in 13 male humans at  rest, during exercise, and during recovery in an attempt to explain the changes in V / Q c A  through the occurrence of pulmonary oedema. It was hypothesized that a  VA/QC  mismatch that remained elevated for some time following exercise would be indicative of pulmonary oedema, whereas resolution of a  VA/QC  mismatch that followed the return of  cardiac output and ventilation to resting values would be indicative of changes in ventilation and blood flow per se. V / Q c mismatch was assessed by the measurement of A  Qc through indocyanine green dye dilution and the measurement of V A through the multiple inert gas elimination technique. Two groups of subjects were classified according to the degree of V / Q c mismatch observed from rest to heavy exercise. Group A  one demonstrated an increased V / Q c mismatch during exercise while group two did not. A  During recovery, both groups initially showed a decrease in V A / Q C mismatch consistent with decreases in ventilation and blood flow during recovery. Over the initial 20 min of recovery there was a resolution of the differences in V A / Q C mismatch between the two groups consistent with oedema formation in group one. The authors conclude that, while there is no proof of pulmonary oedema, the findings are consistent with the formation of pulmonary interstitial oedema.  Exercise in hypoxic conditions provides a unique circumstance for the study of pulmonary oedema due to the accentuating role of hypoxia in the development of oedema. Anholm et al have examined cyclists at altitude for radiographic evidence of 3  pulmonary oedema . Chest radiographs were obtained before and following various  23  distances ranging from 70-131.5 km of road cycling at altitude (range 2,097 - 3,369 m), and analyzed by three radiologists for signs of pulmonary oedema (loss of sharp definition of pulmonary vascular markings, hilar blurring, Kerley A , B , or C lines, peribronchial cuffing, thickening of the fissures, diffuse opacification, and pleural effusion). Scores were given according to the degree of oedema observed, and subtle but significant signs of oedema were found (loss of definition of vascular markings). The authors concluded that the findings provide evidence of pulmonary oedema as a result of increased cardiac output and increased filtration beyond clearance ability, rather than as a result of capillary stress failure, which they conclude would appear radiographically more similar to the signs of high altitude pulmonary oedema.  Caillaud and colleagues  1 6  have provided further indirect evidence of pulmonary  interstitial oedema following exercise. Eight male trained subjects (triathletes) and eight untrained subjects performed an incremental cycling VC^max test followed by measures of ventilation and arterial blood gases during recovery. Rapid shallow breathing was defined as a positive A V j , where A V j is the difference in exercise V j and recovery V j . Both groups of subjects developed rapid shallow breathing during recovery and the trained group did so to a significantly greater degree than the untrained group.  A recent study by McKenzie et a l  84  provides evidence of extravascular lung water  following sustained exercise in humans. Eight male cyclists performed a 45-minute cycling test followed by measures of DLco, D M , V C , and M R I of the chest. DLco and Vc were significantly decreased post exercise while D M was unchanged. There was a  24  significant increase in extravascular lung water post exercise (0.223 ± 0.0225 vs. 0.244 ± 0.0506 gml" pre- and post-exercise respectively). A previous study has shown that 1  extravascular water increases should be greater than 35% to be detected by chest radiography,  110  but the detection of an increase of less than 10% in this study provides  promise that MRI technology may be more effective in this area. This use of MRI technology may provide a new technique for quantitative assessment of extra-vascular lung water in humans.  There is indirect evidence of pulmonary oedema as a mechanism for E I A H in the form of pulmonary capillary stress failure as shown by increased concentration of red blood cells in B A L fluid following intense exercise. As indicated earlier in the study of 57  Thoroughbred horses, in theory exercise-induced pulmonary oedema should have an 74  effect on arterial saturation during subsequent exercise if the oedema is of the severity to affect diffusion, and if it persists through the rest period between exercise bouts. Two studies have investigated such effects of repeated exercise bouts on E I A H in humans. ' 74  8 3  '  1 1 2  St. Croix et a l  1 1 2  studied the effects of two high-intensity treadmill exercise bouts  separated by 20 minutes on E I A H (measured by arterial blood samples) in females of various fitness levels. Contrary to the authors' hypothesis, the results showed a slightly less severe E I A H during the second exercise bout. This is taken as evidence of a "functionally based mechanism" which does not persist beyond the exercise, rather than a mechanism resulting from a temporary change in structure of the gas barrier. Further, the authors suggest that the structural change of stress failure in the pulmonary capillaries caused by high pulmonary pressures is not the mechanism responsible for EIAH. In a  25  similar study, McKenzie et a l  83  found no significant difference in minimum arterial  saturation (measured by pulse oximetry at the ear lobe) between two progressive cycling VChmax tests separated by 60 minutes of rest. The authors suggest that the mechanism for EIAH is not aggravated by repeat exercise, and these results suggest that any pulmonary oedema that was present either cleared prior to the second exercise bout, or was not significant enough to interfere with pulmonary gas exchange.  Upon examination with CT, mean lung density has been shown to increase (0.21 ± 0.009 to 0.25 ± 0.01 g-cm" ) following triathlon in male athletes. Manier et a l 3  16  72  studied nine  trained runners in an attempt to verify and quantify the mechanism of this change. Specifically, the previous study by Caillaud et a l examined only a few slices of the 16  lungs, while Manier et a l  72  examined the distribution of density in the whole lung to  provide a measure of lung mass. In this study, no significant changes were found in measured lung density or lung mass following two hours of running at 75% VC^max. No visual observations associated with pulmonary oedema (increase in the observed pulmonary vessels) were present. The authors concede that any change in extravascular lung water could have been below the CT resolution.  In a study involving nine well-trained male subjects, extravascular lung water measured by the double indicator dilution technique, and pulmonary capillary blood volume measured by the carbon monoxide diffusing capacity method were compared between rest and following 10 and 50 min of bicycle ergometer work.  A significant increase in  extravascular lung water was found between rest and following 10 min of exercise (178 ±  26  37 to 219 ± 46 ml), with no significant change between 10 and 50 minutes. Pulmonary capillary blood volume also increased significantly from 140 ± 42 to 220 ± 106 ml from rest to 10 min, with no significant difference after 50 min. It was concluded that the increase from rest to exercise signalled the redistribution of alveolar wall blood flow, and that the lack of difference between the two exercise durations signalled a lack of accumulation of lung fluid.  Marshall et a l  76  found an increase in pulmonary extravascular water (126 ± 15 vs. 155 ± 8  ml), as measured with the double indicator-dilution technique, when male subjects changed posture from sitting to lying supine. A further increase in extravascular water (229 ± 22 ml) was found with supine cycling exercise at a workload of 150 kg-m" . The 1  double indicator-dilution technique is dependant on lung perfusion since the measure of extravascular lung water depends upon the almost instantaneous equilibrium of indicator (tritiated water in this study) with pulmonary extravascular water. Therefore, it cannot be concluded that an increase in extravascular water as measured by this technique necessarily represents pulmonary oedema.  Radiographic examination of five male subjects following maximal exercise (cycling VChmax test) failed to show evidence of pulmonary oedema. Subjects cycled to fatigue 39  on a ramp protocol starting at 50 W and increasing by 50 W every three minutes, immediately followed by a chest radiograph (within two minutes of the end of exercise). Radiographs were examined for signs of pulmonary oedema including redistribution of pulmonary blood flow, loss of sharp definition of pulmonary vascular markings, hilar  27  blurring, and perivascular and peribronchial cuffing. A densitometer was used to measure the radiographic density of six areas of the lungs. No evidence of pulmonary oedema was found, and the authors therefore concluded that any increase in extravascular lung water must have been trivial.  E V L W Change Reference  Measure  Model  Lung weighing  Canine  No  Marshall et a l  75  Radiography & lung  Canine  Yes  Snashall e t a l  110  Microscopic  Swine  Yes  Schaffartzik et a l  Radiography  Human  Yes  McKechnie et a l  Double indicator-  Human  No  Marshall et a l  Double indicator-  Human  No  Vaughan et a l  Trans-thoracic electrical  Human  Yes  Buono et a l  CT  Human  Yes  Caillaud et a l  CT  Human  No  Manier et a l  Radiography  Human  No  Gallagher et a l  Radiography  Human  Yes  Anholm et al  MRI  Human  Yes  McKenzie et a l  103  82  76  120  15  16  72  39  3  84  Table 2.1. Summary of major extravascular water studies.  2.12 EXERCISE IN A C U T E HYPOXIA Any sojourner to altitude has experienced the detrimental effects of hypoxia on exercise, and these have been well documented. With exposure to altitude or decreased F1O2, there is a decrease in arterial PO2, and arterial oxyhaemoglobin saturation falls according to the  28  sigmoidal oxyhaemoglobin dissociation curve. The decrease in exercise capacity is demonstrated as a curvilinear decrease in VC^max with increasing altitude or isobaric 81  hypoxia. Upon exposure to moderate altitudes of approximately 4,000m, this drop in V0 max may appear more linear. ' The exact response to hypoxic exercise is 70  87  2  somewhat variable in the literature. Fulco et a l  38  have reviewed a number of studies  involving exercise testing during exposure to hypoxia and listed potential sources of variation in the responses including: differences in fitness levels, resident altitude prior to study, gender, age, hypoxic ventilatory response, and duration of exposure.  Upon acute exposure to hypoxia during exercise, ventilation is significantly increased compared to sea-level exercise of comparable intensities. Fifty years ago Astrand '  5 6  performed studies on the ventilatory response to hypoxia and demonstrated that ventilation is significantly increased during exercise at altitude, but that this increased ventilation is mitigated by breathing oxygen rather than normal air. Maximal exercise values for V E , and \ V V O 2 have been shown to increase linearly with exposure to decreased F1O2. Nevertheless, the increased ventilation during exercise in hypoxia is 87  not adequate to maintain SaO"2 or VC^max at sea-level values. According to Calbet et al, the decrease in VC^max in moderate hypoxia can be explained by the decrease in 17  arterial oxygen content but in severe hypoxia this only partially explains the decrease in VC^max. Other factors in severe hypoxia (F1O2 < 10.5%) include impaired pulmonary gas exchange and decreased cardiac output (and therefore decreased skeletal muscle blood flow).  29  2.13 HIGH ALTITUDE PULMONARY OEDEMA High altitude pulmonary oedema (HAPE) is a condition that may accompany exposure to altitude (2,500m and greater) or hypoxia, particularly when exposure is acute and without adequate acclimatization, or in combination with exercise. Prevalence of H A P E is <0.2% with a slow ascent and up to 10% with a rapid ascent. There are many intricately 7  connected mechanisms involved in the development of H A P E , some of which maybe a factor in the development of oedema during exercise in hypoxia. According to Bartsch 77  these include pulmonary vasoconstriction, increased endothelin release nitric oxide (NO) synthesis, permeability, ' '  43 64 106  105  structural damage,  122  and decreased  increased pulmonary capillary  and changes in alveolar fluid clearance. ' In general a high 71 89  pulmonary pressure is associated with the development of H A P E . Pulmonary pressure is 7  increased above resting values during exercise in normoxia, and in combination with hypoxia the effect may be accentuated.  2.14 MAGNETIC RESONANCE IMAGING (MRI) MRI or nuclear magnetic resonance (NMR) is a non-invasive medical imaging technique that involves placing the patient within the coil of a strong magnet. Three strengths of magnetic field are commonly used with MRI: low field (0.035-0.3 T), mid field (0.5-1.0 T), and high field (1.0-1.5 T). Field strengths of over 2.0 T are often used during research imaging. The large magnetic field causes the protons of hydrogen atoms in the body to align the poles of their spin in a uniform direction. The protons are then bombarded with electromagnetic radiation of radio frequency, which causes a momentary change in the proton orientation. Upon return to the uniform orientation the protons emit a detectable  30  radio signal reflecting the number of protons in a particular volume of tissue. Typically M R I is used clinically for the visual examination of soft tissues, particularly cerebral areas.  Because M R I technology depends on a proton signal, a water phantom may be used to provide a baseline measure of the signal generated by a known volume of water. In this way MRI may be adapted for research use to examine the quantity of water in a volume of tissue.  2.15 C O N C L U S I O N It may be concluded from these studies that diffusing capacity generally declines following exercise in humans. However, it should not be concluded that the commonly . observed reduction in D following exercise could be definitively attributed to interstitial L  pulmonary oedema. The studies directly examining diffusing capacity suggest that a 45  redistribution of blood flow is the more likely explanation. Those studies that have attempted to measure pulmonary oedema following exercise do not offer such a distinct conclusion. It appears from this review that transient pulmonary oedema following exercise is possible in certain circumstances, but more research is needed before a definitive conclusion may be drawn.  31  CHAPTER T H R E E : Incremental Exercise Under Varying Hypoxic Conditions  3.1 INTRODUCTION Continuous high-intensity exercise such as that undertaken by endurance athletes requires a dramatic increase in the oxygen demand of the aerobic metabolic processes that occur in skeletal muscle mitochondria. In accordance with the Fick equation [VO2 = Qc(av02)diff.], both cardiac output and mean arterial-venous oxygen difference increase during continuous exercise to meet the increased demand for oxygen. For this reason, any interference with these changes stands to influence maximal oxygen consumption. The alteration of F1O2 during exercise has a number of effects on the normal physiological response to exercise, most notably increased ventilation for a given workload in an attempt to maintain Pa0 . During heavy exercise in moderate hypoxia, however, Pa02 2  inevitably falls below values normally seen during exercise of comparable intensity in normoxia.  A number of studies have documented the performance ' ' ' 1  36,59,62,63,66,77,87, 88,  ioo,  127,130 r  e s p o n s e s  t o  m  a  x  i  m  a  26  30  111  and physiological ' ' 17 33  i exercise in hypoxia, but the degree of  hypoxia in which subjects are capable of completing maximal exercise without Sa02 dropping below acceptable levels, remains unclear. In our lab we have set a minimum Sa0 of 70% as the cut-off threshold, dictated mainly by the specifications of the pulse 2  oximeters in use. During exercise, mean pulmonary arterial pressure (PAP) may rise above 30 mmHg from resting values of approximately 15 m m H g . Although the 40  126  mechanisms of increased P A P are different with exercise and exposure to hypoxia, both  32  provide a useful means to examine the effects of raised mean P A P on pulmonary blood flow and the development of interstitial oedema. Of interest is the level of hypoxia that subjects may tolerate without prohibiting them from performing high intensity exercise.  Therefore the purpose of this study was to examine maximal cycling exercise in varying levels of hypoxia and to establish the greatest level of hypoxia while still allowing subjects to perform high intensity cycling work. To elicit a significant physiological response to simulated altitude, the lowest possible F1O2 was desirable for subsequent portions of this thesis (Chapter 4). Therefore in this study, four conditions of varying levels of F1O2 (21, 18, 15, and 12% 0 ) were used while performing cycling exercise of 2  increasing intensity to maximal effort on several visits to the laboratory. It was hypothesized that V02max and Sa02 would be significantly different between each condition, and that not all subjects would manage to achieve volitional fatigue during the F1O2 = 12% condition without Sa0 falling below 70%. 2  3.2 M E T H O D S Subjects Ten healthy, habitually active males (age = 29.6 ± 5.8 y, height = 181.1' ± 8.3 cm, mass = 79.4 ± 5 . 6 kg) with no history of respiratory disease participated in this study. Written informed consent was obtained from all subjects prior to participation as approved by The University of British Columbia Committee on Human Experimentation.  33  Exercise Testing A randomized, blinded design was used with four conditions corresponding to normoxia and three levels of hypoxia during exercise. Under each condition, subjects performed a graded cycle test to exhaustion (VC^max test) on an electronically braked cycle ergometer (Quinton Excalibur, Lode, Groningen, Netherlands). Subjects started at 0 W and increased at a rate of 30 W-min" . Expired gases were collected and analyzed and 1  ventilation was measured (True One, Parvomedics, Sandy, UT) and averaged every 15 seconds. V02max was calculated as the mean of the four highest consecutive readings. Heart rate was measured by telemetry (Polar Vantage X L , Kemple, Finland) and averaged every 15 seconds, and Sa02 was measured every 15 seconds using an ear pulse oximeter (Biox 3740, Ohmeda, Madison, WI). The test stopped when Sa02 dropped below 70%, or when the subject experienced volitional fatigue and could not maintain a constant pedalling rate. There are certain considerations to be taken into account when using pulse oximetry to assess Sa02. These are discussed in the methods section of Chapter 4.  Gas Delivery The four conditions included 21, 18, 15, and 12% F]0 , with the balance nitrogen. 2  Following delivery from a compressed gas tank, the inspired air was humidified and subjects breathed through a two-way valve (Hans Rudolph, Kansas City, MO). Due to limited flow through the tank regulator, a Douglas bag was used to store the humidified air prior to delivery to the subject to allow adequate ventilation during high-intensity exercise. Compressed gas from a tank was used in all four conditions, including  34  normoxia, in order to preserve the blinding effect. Prior to beginning exercise, subjects remained seated on the cycle for five minutes of rest while breathing the selected gas concentration. For safety reasons, medical oxygen was readily available throughout the testing.  Statistics Repeated measures analysis of variance was used to examine the differences between V0 max, peak power, Sa0 , HR, V , V , R , V / V 0 , and V E / V C 0 under each 2  2  E  f  T  E  2  2  condition. Tukey's Honest Significant Difference post-hoc test was performed where a significant difference was found. For all analyses, significance was set at a = 0.05. Mean decline in V 0 m a x for all conditions was calculated for each subject and correlated to 2  normoxic V 0 m a x and to minimum Sa0 in the normoxic condition. 2  2  3 . 3 RESULTS There were significant differences in V0 max, F(3,36) = 24.5, peak power F(3,36) = 2  28.4, and minimum Sa0 , F(3,36) = 142.1 between the four conditions (see figures 3.1, 2  3.2, and 3.3).  35  ON  37  100 -| 95 90 C/3  85  X  — i< a  a cd  o  80 75 70  cd  65 60 0  / A  —I—  21  - 1 —  - 1 —  18  15  12  F , 0 (%) 2  Figure 3.3. Minimal Sa02 with varying F ^ (mean ± SD); N = 10. * Significantly different from the normoxic values (p<0.05).  Post-hoc analyses revealed a significant difference in VChmax from the normoxic condition in the 12% (p = 0.00016) and 15% (p = 0.035) conditions. VChmax was not significantly decreased in the 18% condition. Peak power was significantly reduced in the 12% (p = 0.00016) condition, but not in the 15% or 18% condition. S a 0 was 2  significantly lower in the 12% (p = 0.00016), and the 15% (p = 0.00016) conditions than the normoxic condition. Tables 3.1 and 3.2 summarize the results for each condition in this study.  38  F,0 (%) 21 18 15 12  V0 max (mL-kg'^min" ) 57.2±7.9J» 53.2 ±5.7« 48.2 ±7.9*» 31.5 ± 7.4*tJ  2  2  1  HR (beat'min ) 178.0 ± 18.6* 179.7 ± 11.6176.6 ± 10.5* 153.0 ± 2 0 . 4 * t t 1  Peak Power (W) 401.4 ±47.9* 375.7 ±42.1* 362.2 ±41.4* 201.8 ± 75.l*tt  Minimum S a 0  2  (%)  95.1 ± 1.3 J* 91.5 ± 2.1 J83.3 ± 3 . 7 * f 71.5±3.2*tt  Table 3.1. V02max, heart rate, peak power, and minimum Sa0 during maximal exercise in various conditions of hypoxia (means ± SD); N = 10. * Significantly different from Fi0 = 21% (p < 0.05). t Significantly different from F i 0 = 18% (p < 0.05). X Significantly different from FT0 = 15% (p < 0.05). • Significantly different from F i 0 = 12% (p < 0.05). 2  2  2 2  2  FiO (%) 21 18 15 12  v  z  V  E  (Lmin ) 140.9 ± 19.3132.6 ± 18.7134.2 ± 17.974.8 ± 34.7*tt 1  Rr (br-min" ) 58.1 ± 10.6' 56.8 ± 9.555.8 ±9.232.9 ± 11.2*tJ  T  (L-br ) 3.0 ±-0.62.9 ±0.6« 3.0 ±0.5« 2.5 ± 0.5*t$  1  1  v /vo E  2  31.3 ± 2.1 31.5 ± 1.2 35.5 ± 5 . 8 30.4 ± 6 . 3  v /vco E  2  24.8 ± 1.2 24.9 ± 1.2 26.8 ± 1.6 25.6 ± 3 . 5  Table 3.2: Ventilatory data during maximal exercise in various conditions of hypoxia (means ± S D ) ; N = 10. * Significantly different from FT0 = 21% (p < 0.05). | Significantly different from F i 0 = 18% (p < 0.05). t Significantly different from FT0 = 15% (p < 0.05). • Significantly different from Fi0 = 12% (p < 0.05). 2  2 2  2  With an F i 0 of 12%, only one of ten subjects was able to achieve 70% of peak normoxic 2  power, and mean peak power during the 12% condition was 50.3 ± 17.5% of peak normoxic power (Figure 3.1). With an F i 0 of 15%, however, all ten subjects achieved at 2  least 85%> of peak normoxic power, and mean peak power during the 15% condition was 90.3 ± 3.3% of that in normoxia. During interpretation of these data it is important to note that several subjects did not achieve volitional fatigue during the 12% condition due  39  to termination of the test when S a 0 reached 70%. This had a significant effect on the 2  data recorded in this condition and reported as maximal exercise data. Nonetheless, this data adds important value to the study, in particular with reference to Chapter 4, and therefore all data is reported and included in analyses.  As demonstrated in Figure 3.4, there were no significant correlations between mean decline in VO"2max and normoxic V 0 m a x (R = 0.41, p = 0.23), or between mean decline 2  in VChmax and minimum SaO"2 in the normoxic condition (R = 0.48, p = 0.16) (Figure 3.5).  40  0  -1  -5 -  -30 H 40  1  1  1  1  50  60  70  80  Normoxic V0 max (mL.kg" .min" ) 1  1  2  Figure 3.4. Mean decline in VC^rnax (for all levels of hypoxia) vs. normoxic VC^max, R = -0.41, p =  0.23.  41  0  -i  -5 -  -30 -I 92  1  1  1  94  96  98  Minimum normoxic Sa0 (%) 2  Figure 3.5. Mean decline in VC»2max (for all levels of hypoxia) vs. normoxic Sa02, R = 0.48, p = 0.16.  3.4 DISCUSSION The purpose of this study was to.examine maximal cycling exercise in varying levels of hypoxia and to establish the greatest degree of hypoxia that subjects could tolerate while maintaining sustained high intensity cycling exercise.  42  Exercise in Hypoxia The results of this study show the expected declines in VC^max, peak power, and Sa02 with decreased F1O2 during incremental exercise on a cycle ergometer. This decline in V02max has been well documented in past work, and the results of this study complement the existing data. Figure 3.6 demonstrates the comparison between mean values of VC^max (as a percentage of VC^max in normoxia) of a number of previous studies ' " ' ' ' < ' " ' 18  20  36  59  66  70  77  86  88  11  1 3 0  with similar values in the present study.  100  o .  8  *  90  o  X cd  a  80  CN  o >x  o o  a  a  70  60  o  o >  0  V/10  —I—  12  14  •  Mean previous studies  O  Present Study  — 1 —  — 1 —  16  18  20  FTO Iu2  Figure 3.6. Hypoxic VC^max / normoxic VC^max vs. F1O2.  43  The values in the present study decline in a slightly more curvilinear fashion than do the values in previous studies, and this is likely the result of the failure of subjects in this study to achieve a true V 0 m a x while breathing 12% oxygen. Thus, the values presented 2  for V0 max in the most severe hypoxic condition are lower than expected (as they do not 2  represent true volitional fatigue or maximal exercise exertion). A cut-off value of S a d = 70% may not have been used in the previous studies. Martin & O ' K r o y report a 77  minimum S a 0 = 67.0 ± 7.1% in a group of trained athletes, Ferretti et a l report a 36  2  minimum S a 0 = 66.2 ± 2.9%, and Hughes et a l report multiple individual values of 59  2  Sa0 < 70%). Nonetheless, the general pattern of decline in V 0 m a x is similar and falls 2  2  within the expected range of the previous studies. Of the previous studies referred to, three ' ' 20  87  111  involved treadmill running as the mode of exercise testing, while the  remainder involved cycle ergometer work similar to this study.  Several previous studies have examined the relationships between % A V 0 m a x in 2  70  hypoxia, and normoxic V 0 m a x in normoxia and minimum S a 0 values. Lawler et al 2  2  specifically studied the relationship between V0 max and V 0 m a x decrement during 2  2  exposure to acute hypoxia (Fi0 = 14%) and found that trained subjects demonstrated 2  significantly lower S a 0 during exercise in hypoxia than untrained subjects. Further, a 2  significant linear correlation (r = 0.94) was found between % A V 0 m a x in hypoxia and 2  normoxic V0 max. Martin and O'Kroy also found a significant negative correlation (r = 2  -0.91) between % A V 0 m a x in hypoxia and normoxic V 0 m a x (data presented inversely 2  2  to Lawler et al, but with the same direction of correlation), and a significant negative 70  44  correlation (r = -0.84) between %AV02max in hypoxia and SaC<2 at maximal exercise in hypoxia. In a study of six ice-hockey players and six cross-country skiers a similar 66  significant correlation (r = -0.61) was found between %AV02max in hypoxia and normoxic V G w a x , but a study of seven endurance trained and seven sedentary women,  130  showed significance in this relationship only amongst trained (VC^max = 56.3  ± 4.7 mL-kg^-min" ) females (r = 0.8) but not amongst untrained (VC^max = 34.8 ± 5.6 1  mL-kg^-min" ) females (r = 0.13). This indicates that those subjects with the highest 1  aerobic capacity, suffer the greatest decline in VChmax in hypoxia, and that those subjects with the greatest degree of arterial desaturation in hypoxia also suffer the greatest decline in VC^max in hypoxia. The latter point falls in line with what is known about the relationship between exercise performance and arterial saturation. Similarly, in normoxia exercise performance is related to the degree of exercise-induced arterial hypoxaemia. Contrary to some of the previous findings, the results of this study do not 91  show a significant correlation between % A VC^max in hypoxia and either normoxic V02max or minimum SaO"2. In this study, even the degree of arterial desaturation at an F1O2 = 15% and the % AVO^max at this level of hypoxia are not significantly correlated (R = -0.17). Figure 3.4 demonstrates that the data in this study do follow the pattern of correlation reported in the previous studies, but the correlation did not achieve statistical significance. It is also possible that the range of data (particularly the range of SaCh) reported in this study did not lend itself to statistically powerful correlational analyses, in which case a larger group of subjects may have been helpful for these particular analyses of the data. Statistical calculations demonstrate that powers of 0.38 and 0.44 were achieved for the correlations of decrease in F1O2 with decline in VC^max and SaCh  45  respectively, with a = 0.05. Therefore, a greater sample size would be desirable before making bold claims about the correlations reported. Finally, the group of subjects used in this study were moderately trained individuals (VC^max = 57.2 ± 7.9 mL-kg^-min" ), and 1  as evidenced in the study by Woorons et a l ,  130  these relationships may hold true only for  more highly trained subjects.  Maximal Exercise Tolerance in Hypoxia It was decided that an appropriate level of hypoxia would be indicated by the ability of subjects to reach 70% peak normoxic power. A peak hypoxic power lower than this would likely lead to sustained hypoxic exercise at a level too low to stimulate the desired physiological and pulmonary responses to exercise that were to be studied in Chapter 4. It is clear from the results that an F1O2 of 12% is too severe for subjects to maintain this level of exercise. Indeed, in the 12% condition, only one of the subjects involved managed to achieve 70% of peak normoxic power. In the 15% condition, all ten subjects achieved at least 85% of peak normoxic power. Further, when examining the peak power in each condition as a percentage of peak power in normoxia, the standard deviation increased substantially from 3,3% with F ^ = 15% to 17.5% with  Fi0  2  = 12%. This  would seem to indicate that there is less inter-individual variation and unpredictability in the physiological response to intense exercise with an F1O2 of 15% than 12%, although this increased variation is most likely due to the fact that the subjects were unable to complete exercise to volitional fatigue during the 12% condition without  Sa02  falling  below 70%). This sharp change at 12% oxygen is very evident when the values are displayed graphically as in figure 3.1. It is worth noting that while the F1O2 of 18% was  46  included in the study, it is clear from the results that the physiological response to exercise at 18% is only marginally different from that in normoxia. Indeed, in 18% hypoxia, VOimax, peak power, and SaC>2 were not significantly different than in normoxia. Minimum SaO"2 at 18% oxygen was only 91.5 ± 2.1% compared to 95.1 ± 1.3% in normoxia. One of the purposes of Chapter 4 is to provide the greatest hypoxic perturbation possible while preserving the ability of the subjects to perform sustainable exercise of a moderately high intensity. Therefore it is appropriate that, to elicit the maximal physiological response from hypoxia while preserving the ability of subjects to perform intense sustainable exercise, an F1O2 of 15% be selected as the hypoxic condition in Chapter 4.  47  CHAPTER FOUR: Effects of Prolonged Exercise on Extravascular Lung Water  4.1 INTRODUCTION During intense continuous exercise in humans, oxygen consumption may increase 20fold or more from rest, placing large demands on the pulmonary system to maintain arterial PO2 and PCO2 within normal tolerable values. Typically, in humans, arterial PO2 is effectively maintained near resting values of approximately 100 mmHg during exercise by a pulmonary system that effectively meets this increase in metabolic demand. However, in approximately 50% of trained aerobic athletes (V02max > 150% predicted) exercise-induced arterial hypoxaemia (EIAH) occurs during strenuous exercise. '  24 90  Dempsey and Wagner have proposed three levels of hypoxaemia as defined by the 25  degree of arterial haemoglobin saturation: mild (93-95%), moderate (88-93%), and severe (< 88%). Despite extensive research the mechanisms responsible for E I A H have not been completely established. Those most likely to contribute are relative alveolar hypoventilation, ventilation/perfusion (VA/QC) mismatch diffusion limitation, and right 25  to left shunts. A significant amount of research has been dedicated to the study of EIAH in the past two decades with advances in the understanding of the nature of this phenomenon and the mechanisms involved.  Interstitial pulmonary oedema is the development of additional extravascular lung water caused by an increase in the filtration of fluid from the pulmonary capillaries into the interstitial space between the alveolar epithelium and capillary endothelium, and may interfere with gas exchange across the alveolar-capillary membrane. Pulmonary oedema  48  may occur through changes in the Starling forces (pulmonary capillary leakage) and/or changes in the permeability of the capillary membrane (pulmonary capillary stress failure) . Mean pulmonary artery pressure (PAP) in humans at rest is typically 15 mmHg 2  (systolic = 25 mmHg, diastolic = 8 m m H g ) .  40,126  Mean PAP rises with exercise and near-  maximal exercise values have been measured at 33 mmHg  4 0  Hypoxia also alters mean  PAP with resting and near-maximal exercise values reaching 34 and 54 mmHg respectively.  40  Transient interstitial oedema during exercise could partly explain the A-aC»2 difference observed during exercise in hypoxaemic subjects. Whether transient pulmonary oedema occurs or not has been the subject of a number of studies and remains an unresolved issue. In animals, evidence of pulmonary oedema following heavy exercise has been demonstrated in a study by Schaffartzik et a l  103  using swine. Five pigs ran at the highest  speed maintainable for 6 - 7 minutes while five pigs acted as controls. Lung tissue blocks from each animal were prepared for microscopic examination and documentation of perivascular or peribronchial cuffing, and the exercised animals showed significantly more periarterial cuffing in general, as well as a higher percentage of periarterial cuffing in the lower lobes and ventral areas of the lung than the control animals. In humans, evidence of pulmonary oedema (bilateral pulmonary consolidation, upper lobe venous congestion and cardiomegaly) was observed in two subjects following very long duration exercise. Assessment of pulmonary oedema with modern imaging techniques including 82  radiography, CT, and M R I has led several authors to conclude that transient pulmonary oedema was present in subjects following exercise. ' ' However, a number of studies 3 16 84  49  both in animals and humans have found no evidence of transient pulmonary oedema following exercise including a study involving treadmill exercise in dogs, and several 75  39 72 76 120  exercise studies in humans. ' ' '  The first step to understanding the role of transient pulmonary oedema as a mechanism for diffusion limitation leading to EIAH in highly trained athletes is to establish whether oedema occurs in the lung in exercising humans, and to quantify it. Direct measure of lung density is not possible in humans, and indirect measures are generally only practical or valid following rather than during exercise. Therefore, the purpose of this study was to assess measurement of in vivo lung density by M R I and to describe transient pulmonary oedema, through this measure, following exercise in healthy athletic humans while breathing normoxic and hypoxic air. The hypoxic condition was included in an attempt to increase mean PAP as high as possible during intense exercise. With increasing severity of hypoxia exercise capacity decreases, and a moderate degree of hypoxia (F1O2 = 15%) was selected to combine the effects of hypoxia and heavy exercise on mean PAP. It was hypothesized that following sustained exercise in normoxia, mean lung density would increase over baseline values, and that a similar but greater magnitude of increase would occur in the hypoxic condition. Additionally, it was hypothesized that two hours postexercise, lung density would not be different than the baseline values in each condition.  50  4.2 M E T H O D S Subjects With a minimum difference of 0.03 g-mL-1 being considered a meaningful change in lung density, and an expected standard deviation of 0.02 g-mL" , the required number of 1  subjects were 9.3, given a power value of 0.8 and alpha set at a = 0.05. Therefore, ten male subjects (age = 25.9 ± 4.7 y, height = 184.1 ± 8.2 cm, mass = 79.4 ± 9.5 kg) were used in this study. Subjects reported for testing on four separate testing days. The first two involved maximal cycling (V02inax) tests under normoxic and hypoxic conditions in a randomized order, and a pulmonary function test to assess F V C , F E V i , and PEFR. The second two days involved M R I assessment of lung density prior to and following an exercise intervention in normoxic and hypoxia. In the hypoxic condition, F1O2 was 15% oxygen and balance nitrogen as determined during the study explained in Chapter 3. Figure 4.1 illustrates the order of each of the second two days involving sustained exercise and M R I assessment of lung density. The order of these days was also randomized.  51  Time (min) 0 15  1  MRI  50  110  170 185  Rest  Rest  210  225  Rest  •  Sustained exercise Normoxic & Hypoxic (Randomized Order) - Sa0 , V 0 , V , HR, 2  2  MRI  MRI  E  Figure 4.1. Timeline of experimental protocol for each condition (normoxia and hypoxia).  Exercise Testing Subjects were requested to avoid exercise, alcohol, and caffeine for 24 h prior to each testing session. The V 0 m a x tests were performed on an electronically braked cycle 2  ergometer (Quinton Excalibur, Lode, Groningen, Netherlands) starting at 0 W and increasing at a rate of 30 W-min" until the subject experienced volitional fatigue and 1  could not maintain a constant pedalling rate. Expired gases were collected and analyzed and ventilation was measured (True One, Parvomedics, Sandy, UT) and averaged every 15 seconds. Heart rate was measured by telemetry (Polar Vantage X L , Kemple, Finland) and averaged every 15 seconds, and Sa0 was measured using an earlobe pulse oximeter 2  (Biox 3740, Ohmeda, Madison, WI), and averaged every 15 seconds. During all exercise testing in this study, S a 0 was monitored by a second finger-tip oximeter (Nonin 8500, 2  Nonin Medical Inc., Plymouth, M N ) to ensure that inadequate earlobe perfusion was not an issue in assessment of Sa0 . V 0 m a x was calculated as the average of the four highest 2  2  52  consecutive readings. In the hypoxic condition, following delivery from a compressed gas tank, the inspired air was humidified and subjects breathed through a two-way valve (Hans Rudolph, Kansas City, MO). Due to limited flow through the tank regulator, a Douglas bag was used to store the humidified air prior to delivery to the subject to allow adequate ventilation during high-intensity exercise. In this condition, arterial oxyhaemoglobin saturation was monitored continuously and the exercise was terminated if Sa02 fell below 70%. Following each V02max test, the results were examined for the following criteria to determine whether maximal exercise was achieved: a plateau in VO2 with increasing exercise intensity, achievement of a maximal heart rate within 5% of agepredicted H R  m a x  , and a respiratory exchange ratio of > 1.10.  For measurement of lung density, subjects reported to the Radiology Department of the University of British Columbia Hospital for M R I assessment. Approximately one hour (62.3 ± 9 . 6 min normoxic condition; 62.0 ± 7.5 min hypoxic condition) following the completion of the baseline lung MRI, subjects began the exercise intervention at the Allan McGavin Sports Medicine Centre Exercise Physiology Laboratory at the University of British Columbia. This consisted of 60 min cycling exercise while breathing either normal room air or hypoxic gas of 15% O2 and balance N by volume. 2  The cycling was performed on the same ergometer, and with the same metabolic equipment and gas delivery system as the previous V02max tests. Initial workloads were set at between 55 - 60% of peak power achieved in the V02inax tests, and subjects had control of the workload on the cycling ergometer and were instructed to perform as much work as possible in 60 min. Workload intensities were monitored throughout the exercise  53  and feedback was given to the subjects in an attempt to maximize work accomplished. Approximately 50 min (54.0 ± 17.2 min normoxic condition; 55.6 ± 9.8 min hypoxic condition) and 100 min (100.7 ± 15.1 min normoxic condition; 104.3 ± 9.1 min hypoxic condition) following the exercise intervention subjects underwent identical M R scans as prior to exercise. Therefore, for each condition, each subject underwent three M R scans (one pre-exercise and two post-exercise scans) for a total of six scans per subject. Two of the subjects (subjects 3 and 7) were unable to complete the measure of resting lung density under both conditions. Therefore, for these two subjects data is shown for the hypoxic condition only.  Although subjects were given time to perform a self-selected warm-up prior to the sustained exercise intervention, there was some delay in set-up following warm-up and prior to the start of the exercise. Therefore the initial minutes of the exercise do not represent steady-state exercise at a high workload. For this reason, the exercise data are averaged over the last 50 min of exercise, with the exception of total work which includes all 60 min of exercise.  Subjects were weighed before the first and second M R scans, and immediately prior to and following the exercise interventions in each condition. Following exercise and prior to the first post-exercise M R scan, subjects were required to drink at least the volume of water corresponding to the mass loss during the exercise intervention. Heart rates were recorded at the initiation of each M R scan.  54  Methodological Considerations Pulse oximetry provides a non-invasive measure of Sa02, and is therefore a convenient tool for use during exercise in human subjects. Validation of SaC»2 measured by pulse oximetry and by direct arterial blood analysis has been performed by Martin et al.  78  Specifically, a mean difference of 0.52 ± 1.36%, and a correlation of r = 0.98 was found between examination of SaC>2, as measured by 3 Ohmeda oximeters and direct arterial blood, in eleven aerobically fit cyclists while performing constant load high-intensity exercise and incremental exercise. These findings involved 232 observations of SaO"2 ranging from 72 - 99%. However, pulse oximetry provides only a representation of true arterial oxyhaemoglobin saturation, and several issues should be noted when using pulse oximetry. Firstly, pulse oximetry measures SaC>2 at one peripheral site on the body. During exercise this is normally on the finger tip, ear lobe, or forehead and any change in normal perfusion of these areas may affect the pulse oximeter representation of true arterial saturation. Secondly, normally pulse oximetry does not allow for the correction of values to blood temperature and pH. Nonetheless, in this study the use of direct arterial blood, for analysis of blood gases was not included because precise measurement of changes in arterial PO2 during both incremental and sustained exercise were not critical to the findings. The main issue was the determination of whether transient pulmonary oedema occurred during and persisted following exercise. The primary reason for measuring Sa02 was to describe E I A H and changes in SaC>2 during sustained exercise in normoxia and hypoxia rather than to draw any conclusion from the degree of hypoxaemia. Therefore it was decided that arterial blood gas assessment was an unnecessarily invasive technique on the subjects.  55  Magnetic Resonance Imaging During assessment of lung density, subjects were imaged with the body coil on a 1.5 Tesla General Electric Horizon Echospeed M R scanner (General Electric Medical Systems, Milwaukee, WI, 5.7 software release) while lying supine and during normal tidal volume breathing. Three water phantoms were placed over the right lung immediately below the clavicle. A n eight-echo pulse sequence was used with an echo spacing of 10 msec and a repetition time (TR) of between 3,000 and 4,000 msec, (time to echo (TE) 10, 20, 30, 40, 50, 60, 70, 80). In this single slice 8-echo sequence, the 90° radiofrequency (rf) pulse was slice selective and the 180° rf pulses were non-selective composite pulses. Multiple gradient lobes of alternate sign and incrementally decreasing amplitude were applied in the slice selection direction to eliminate stimulated echo artefacts. The single slice sequence yielded 8 images of a single slice that provided pixelby-pixel data on lung T relaxation times. For each scan, multi-exponential T analysis 2  2  was used with extrapolation to T E = 0. A Ti relaxation time correction was included. Each scan was cardiac-gated after the R wave peak to limit heart and blood flow artefact. The images were obtained using a 320 mm field of view and 256 x 128 mm matrix.  Output was eight images of each sagittal section, which were transferred to Matlab software (MathWorks, Milwaukee, WI). After extrapolation to T E = 0, a water content map was generated on a Linux workstation using multi-exponential T analysis. Each 2  scan was then analysed for density as compared to the known density of one of the water phantoms. A region of interest (ROT) was drawn manually on the water phantom. The  56  phantom pixel intensity was calibrated to the known density of the phantom (1.0 g-mL" ), 1  and an ROI of the sagittal section of the right lung was drawn manually including as much lung tissue as possible, while definitively excluding any non-lung tissues at the margin of the ROI to avoid partial voluming. Pixel thickness, height, and width were 10.0, 1.0, and 1.0 mm respectively providing a slice thickness of 10.0 mm.  To account for vascular components of the lung, all pixels above a set density threshold were removed during the calculation of lung density. Mean lung slice density values were calculated and recorded at density thresholds of 0.25, 0.30, 0.35, 0.4, 0.45, and 1.0 g-mL" . This technique has been used in a previous study, in which pixels above a 1  84  density of 0.30 g-mL" were excluded, however in this study it was decided to record data 1  on a range of threshold densities for comparison purposes (see Appendix C). For each scan, the area removed was compared visually with areas of high density to verify the total area removed (see figure 4.8). Data are presented for each threshold level in Appendix C (Table C . l - C.6); however, it was observed that a threshold of 0.3 g-mL"  1  corresponded to the removal of major blood vessels while avoiding the removal of other tissues. Further, it was noted that for each threshold value, the lung density for each segment of lung on an anterior-posterior scale did not exceed 0.3 g-mL" . For these 1  reasons, a threshold value of 0.30 g-mL" was used for analyses. Ultimately the 1  measurements were performed at the wide variety of threshold densities to ensure that increased extravascular water following exercise was not missed as a result of discarding certain densities of tissues, and as is evident in the data (Appendix C), it can be stated with confidence that this objective was met.  57  In order to assess lung density gradient, lung density was measured at 3 regions anterior posterior: the anterior 5 cm; the middle 5 cm; and the posterior 5 cm of the lung. For each image, an equation of the lung density gradient was calculated by regression using these three density values. The slope of this equation represented the density gradient for the particular image, and was expressed as mg-mL'^cm" . 1  A l l analyses described above were performed in an identical manner by the same observer. For intra-observer reliability purposes, twenty of the images were analysed twice and a correlation was calculated between the mean slice densities of each group of analyses. Twenty images were analyzed by a second observer and inter-observer reliability between two observers was calculated in an identical manner on these twenty different scans.  Statistics Repeated measures analysis of variance was used to examine the differences between the following variables during maximal exercise and during sustained exercise in each condition: VC^max, power,  SaO"2, HR, V E .  A repeated measures 2X3 A N O V A was used  to analyse the density data between the pre- and the two post-measures and between normoxia and hypoxia. Several variables were recorded prior to the M R scans in an attempt to avoid extraneous factors that could have an effect on the measure of lung density. Heart rates were monitored by E C G throughout the scans and were recorded at the beginning of each scan. Subject masses were recorded prior to the first scan of each  58  day, and the first post-exercise scan of each day. The time between the pre-exercise scan and the two post-exercise scans was recorded and controlled. A n A N O V A with Tukey's Honest Significant Difference post-hoc test was performed on these data. No control of heart rates was possible during the study, but the mass of each subject and the scan times were controlled to maintain as much uniformity as possible. Paired one-tailed t-tests were performed on the data of mass loss during exercise for each of the two conditions.  4.3 RESULTS Exercise In the hypoxic condition, V02max, peak power, and minimum Sa02 during maximal exercise were significantly lower than in the normoxic condition (Table 4.1). The individual patterns of oxyhaemoglobin desaturation are shown in Figures 4.2 and 4.3.  Condition Normoxic Hypoxic  HR (bmin ) 186.7 ± 5 . 9 178.4 ± 7 . 6 1  VE (L-min ) 155.2 ±23.9 141.1 ±26.6 1  V02inax (mL-kg'-rnin ) 65.0 ± 7 . 5 54.1 ± 7 . 0 * 1  Peak Power (W) 456.9 ± 6 6 . 6 377.2 ± 2 9 . 3 *  Minimum Sa0 (%) 92.3 ± 1.8 79.8 ± 4 . 5 * 2  Table 4.1. Maximal exercise data (means ± SD); N = 10. * Significantly different from F i 0 = 21% (p < 0.05). 2  59  100 -i  90  •  o T  v  <N  o  •  00  —  •  80  •  —  0 — •  7 0 ^ Vs 'A  Subject 1 Subject 2 Subject 3 Subject 4 Subject 5 Subject 6 Subject 7 Subject 8 Subject 9 Subject 10  —r~  0  100  200  300  400  500  600  Power (W)  Figure 4.2. Arterial oxyhaemoglobin saturation during normoxic VC^rnax test.  60  100  Subject 1 Subject 2 Subject 3 Subject 4 Subject 5 Subject 6 Subject 7 Subject 8 Subject 9 Subject 10  o v •  *  •o— • - A -  70 Jf 0 0  100  200  300  400  500  600  Power (W) Figure 4.3. Arterial oxyhaemoglobin saturation during hypoxic VOamax test.  Two subjects demonstrated mild E I A H (minSaCh = 94.5 & 93.8%), and seven demonstrated moderate E I A H (minSaO"2 = 91.4 ± 1.1%) during the normoxic VC^max test. During the exercise intervention, subjects cycled at 61.6 ± 9.5 and 65.4 ± 7.1% VC^max (56.6 ± 4 . 1 and 57.2 ± 6.9% peak power) in the normoxic and hypoxic conditions respectively (Table 4.2). One subject (subject 8) was unable to complete the hypoxic VOimax test without Sa02 falling below 70%, and therefore the test was terminated prematurely in the 12 minute of exercise. For In the hypoxic condition, th  mean V 0 , Sa02, power, and total work during the sustained exercise were significantly 2  lower than in the normoxic condition. Mean power and total work performed in each  61  condition is demonstrated in Table 4.3. Heart rate, V 0 , and S a 0 data in the normoxic 2  2  and hypoxic exercise intervention are displayed in Figures 4.4, 4.5, and 4.6.  VE (L-min" ) 63.3 ± 4 . 0 64.6 ± 3 . 4  HR  Condition Normoxic Hypoxic  (b-min" ) 156.9 ± 5 . 7 156.3 ± 5 . 3 1  1  vo  2  Sa0  (mL-kg'^min" ) 39.9 ± 1.3 35.4 ± 0 . 9 * 1  2  (%) 95.2 ±0.1. 82.6 ± 1.2*  Table 4.2. Exercise intervention data averaged over the last 50 min (means ± SD); N — 10. * Significantly different from F A - = 21% (p < 0.05).  Condition  Power (W) 260.6 ± 4 . 1 215.2 ± 3 . 3 *  Normoxia Hypoxia  Total Work (kJ) 926.0 ± 199.3 753.9 ± 88.9*  Table 4.3: Total work performed in 60 minutes (means ± SD); N == 10. * Significantly different from F i 0 == 21% (p < 0.05). 2  62  180 160 140 60  120 o > a  Heart Rate (beats.mirf  100  V02(mL.kg" .min" ) 1  1  80  • I—I  a  60 40  u  20 0  — i —  - 1 —  10  20  30  40  50  60  Time (min) Figure 4.4. Normoxic exercise data. (Means ± SD); N = 10.  63  0  10  20  30  40  50  60  Time (min) Figure 4.5. Hypoxic exercise data. (Means ± SD); N = 10.  64  Figure 4.6. Sa0 during the exercise intervention. (Means ± SD); N = 10. 2  Lung Density There were no significant differences in lung density between conditions F(l,54) = 0.13; p = 0.72 or scan times F(2,54) = 0.23; p = 0.80, nor was there an interaction effect between condition and scan time F(2,54) = 0.37; p = 0.70 (see Table 4.4 and Figure 4.7). For clarity, standard deviations are not presented in Figure 4.7, rather individual lung densities are provided in Appendix B (Figures B . l & B.2).  65  Pre Density (g-ml/') 0.177 ±0.019 0.178 ±0.021  Condition Normoxic Hypoxic  Postl Density (g-mL- ) 0.181 ±0.019 0.174 ±0.022 1  Post2 Density (g-mL ) 0.173 ±0.019 0.176 ±0.019 1  Table 4.4: Lung densities (density threshold = 0.3 g-mL" ) pre and post-exercise intervention (means ± SD); N = 10. 1  0.25  0.20  A •o-  0.15  a -59 § 0.10 Q -•— Normoxic Exercise O - - Hypoxic Exercise  0.05  0.00 Baseline  Post 1  Post 2  Figure 4.7 Lung densities (density threshold = 0.3 g-mL" ). (Means); N = 10. 1  During the exercise intervention there were significant (p < 0.05) decreases in mass of 1.4 ± 0.5 and 1.1 ± 0.5 kg in the normoxic and hypoxic conditions respectively. Heart rate was significantly (p < 0.05) increased at the initiation of the first post-exercise scan  compared to the pre-exercise scan in both the normoxic (59.7 ± 8.5 vs. 53.3 ± 5.2 beats-min" ) and hypoxic (58.3 ± 7.0 vs. 51.7 ± 2.1 beats-min" ) conditions. No other 1  1  significant differences in heart rate or mass were found between scans. There were no significant differences between the two conditions in the time between the pre-exercise and the two post-exercise scans. Mean heart rate, mass, and time data are summarized in Table 4.5.  Scan Normoxic Pre Normoxic Postl Normoxic Post2 Hypoxic Pre Hypoxic Postl Hypoxic Post2  H R (b-min f 53.3 ± 5 . 2 *59.7±8.5 62.5 ± 6.4 51.7 ± 2.1 *58.3±7.0 54.0 ± 6 . 6 -1  Mass (kg) 79.4 ± 9 . 5 78.6 ± 10.3 79.0 ± 10.3 78.8 ± 10.3  Time (min) 0 54.0 ± 17.2 100.9 ± 18.3 0 55.6 ± 9 . 8 104.3 ± 9 . 1  Table 4.5: Heart rate (HR), mass, and time data at initiation of each M R scan (means ± SD);N=10. * Significantly different than resting baseline (pre) measure (p<0.05).  As described above, for each scan a range of threshold densities were used, above which density all pixels were discarded before calculation of mean slice density. A sample M R image of a right lung sagittal slice from this study is presented in Figure 4.8, along with images of computer representations of the scan showing the discarded pixels at each threshold used.  67  Figure 4.8. (a) Sample M R image of right sagittal lung slice, (b) - (g): Computerized image of the lung scan showing areas of pixel removal. Dark areas within the lung represent areas of pixel removal using a threshold of (b) 0.25 g-mL" (c) 0.3 g-mL" (d) 0.35 g-mL" (e) 0.4 g-mL" (f) 0.45 g-mL" (g) 1.0 g-mL" . 1  1  1  1  1  1  68  Mean lung densities (0.177 ±0.019 g-mL" ) are consistent with established values in the 1  •  80 84  literature. '  The variation between subjects was relatively small (coefficient of  variation = 8.0%) and, importantly, the repeatability of resting slice density within subjects gave a significant Casewise correlation of r = 0.73 (p=0.007). The Casewise 2  correlations for intra-observer and inter-observer repeatability on 20 randomly selected scans were both r = 0.98 (p<0.05) (see Appendix C Figures C . l and C.2). The mean 2  areas of the ROI used in each M R scan are presented in Table 4.6.  Condition Normoxic Hypoxic  Pre Area (mm ) 18,158 ±3,729 16,670 ±3,603 2  Post Area (mm ) 16,227 ±3,779 16,423 ±4,066 2  Post Area (mm ) 16,365 ±4,010 16,313 ±3,853 2  Table 4.6. Lung slice areas (density threshold = 0.3 g-mL" ) pre and post-exercise intervention (means ± SD); N = 10. 1  Lung Density Gradient Mean lung density gradient (anterior to posterior) in the normoxic condition was 5.7 ± 1.8, 5.6 ± 1.9, and 6.9 ± 1 . 4 mg.mL" .cm" in the resting, first, and second post-exercise 1  1  scans respectively. In the hypoxic condition, the corresponding lung density gradients were 6.5 ± 1.2, 6.5 ± 1.6, and 6.8 ± 1.5 mg.mU'.cm" (Table 4.7). 1  69  Condition Normoxic Hypoxic  Pre Gradient (mg-mL'-cm ) 5.7 ±1.8 6.5 ± 1.2 1  Postl Gradient (mg-mL'-cm" ) 5.6 ± 1.9 6.5 ± 1.6 1  Post2 Gradient (mg-mL' cm' ) 6.9 ± 1.4 6.8 ± 1 . 5 1,  1  Table 4.7. Lung density gradients (density threshold = 0.3 g-mL" ) pre and post-exercise intervention (means ± SD); N = 10. 1  Individual values for lung density gradient at each threshold level are presented in Appendix C Tables C.7 - C. 12. In the normoxic condition, the mean anterior - posterior distances used in the pre-exercise scans were 5.0 ± 0.0, 73.8 ± 8.3, and 144.5 ± 16.3 mm with corresponding tissue densities of 0.137 ± 0.025, 0.179 ± 0.020, and 0.216 ± 0.023 g-mL . These data are illustrated in Figure 4.9, which is included as a sample of lung -1  density gradient for one of six scans for each subject.  70  0.30 -i  0.25 A  0.05 A  0  20  40  60  80  100  120  140  160  Anterior - Posterior Distance (mm) Figure 4.9. Sample lung density gradient for the resting normoxic condition with a threshold density of 0.3 g-mL" . 1  4.4 DISCUSSION The purpose of this study was to assess measurement of in vivo lung density by magnetic resonance imaging and to describe transient pulmonary oedema, through this measure, following exercise in healthy athletic humans while breathing normoxic and hypoxic air.  Lung Density Lung density varies with inspiration and age of subjects. Use of CT allows measurement of lung density at specific lung volumes, and values of 0.0715 ±0.017 and 0.272 ± 0.067  71  g-mL" are reported at T L C and R V respectively in healthy subjects while supine. 1  85  Values obtained in 50 subjects of mean age 50 years in the upper, middle, and lower lung were 0.123 ± 0.46, 0.121 ± 0.033, and 0.154 ± 0.057 g-mU during inspiration and 0.215 1  ± 0.058, 0.228 ± 0.066, and 0.260 ± 0.078 g-mL during expiration. Van Dyk et a l -1  65  1 1 9  have reported lung density in five year-olds of 0.36 and 0.20 g-mL" , and in 80 year-olds 1  of 0.22 and 0.16 g-mL" during inspiration and expiration respectively. Further findings of 1  mean lung density by CT are 0.235 and 0.199 g-mL" for the right and left lungs 1  respectively. Mayo et a l measured lung density by M R I at T L C as 0.21 ± 0.03 and 42  80  0.20 ± 0.03 g-mL" in the prone and supine positions respectively, and McKenzie et a l 1  84  measured resting lung density by M R I as 0.223 ± 0.023 g-mL" . The findings and 1  variation of lung density during normal quiet breathing in this study (resting lung density = 0.177 ± 0.019 g-mL" ) compare with these previous measurements made with both CT 1  and MRI, and the consistency of repeated measurements of lung density in this study lends confidence to the validity and reliability of M R I assessment of lung density.  Previously, M R I technology has been used to measure lung density both in vitro in animals ' ' and in vivo in humans ' ' 34  35  80  52  values for lung density. Estilaei et a l ' 3 4  3 5  80  84  with good results when compared to known  have conducted two validation studies involving  in vitro measurement of porcine lung tissue density by M R I and gravimetric means. In the first study the ratio of wet/dry lung density as measured by N M R and gravimetric means was 1.00 ± 0.08 and 1.00 ± 0.05 respectively. The second study showed a linear correlation between M R I and gravimetric measure of water content of R = 0.98. A study 2  by the same group has found a ratio of lung water measured by M R and gravimetric  72  means was 0.95 ± 0.03, and concluded that MRI allows measurement of lung water content. The methods and equipment used to measure lung density used in this study 80  were very similar to those mentioned above by Mayo et a l , and as such the measure of 80  lung density in this study is further validated.  Lung Density Gradient It was found that the mean resting lung density gradient was 5.7 ± 1 . 8 mg-mL" -cm" . 1  1  Lung density is gravity dependant: while standing upright, the inferior portions of the lung are more dense than the inferior, and while lying supine the posterior portions are more dense that the anterior. ' Figure 4.9 demonstrates this pattern with a very linear 80 85  relationship between anterior - posterior distance and lung density (r = 0.999994). As a 2  result, the mid-lung density is almost identical to the reported overall mean lung density. Most importantly, the consistency of the mean lung density gradient values across all scans indicates that changes in mean lung density were not missed simply through variance in the pattern of density across each scan slice.  EVLW and Exercise Table 2.1 summarizes previous findings of changes in E V L W , as measured in a variety of techniques, following various interventions - generally exercise. There is no clear consensus on change in E V L W following exercise, with f i v e ' ' ' ' 3  1 5  1 6  8 2  8 4  of these studies  in humans indicating that some increase is observed or likely, and t h r e e ' ' 39  76  120  of them  indicating no such change. At first observation, the results of this study could simply be added to the group of studies demonstrating no change in E V L W following exercise.  73  However, this study is different than the majority of these previously cited studies in that the measure of lung density by M R I is presently the most direct measure available for in vivo studies. This study also specifically used exercise as the intervention in an attempt to investigate the link between pulmonary oedema and EIAH. Further, it is imperative to compare these results with those of McKenzie et a l  84  in which an increase in lung density  following exercise was observed using the same technology as this study.  Several possibilities arise when attempting to explain the discrepancy in results between this study and the previous one using M R I assessment of lung density in humans.  84  Firstly, the exercise intervention was different between these two studies. McKenzie et a l used an intervention of 45 min cycling exercise at 77% of VC^max. For three 84  minutes at the end of the exercise, subjects were encouraged to sprint and increased power outputs to only 35 W below peak values obtained in a VO"2max test, and increased heart rate to within 1 beatmin" of maximal values. B y contrast, in this study subjects 1  cycled for 60 min at a workload of 61% of VC^max during the normoxic condition with no sprinting. Theoretically, the greater exercise intensity (for a shorter duration) could produce greater cardiac output and pulmonary capillary pressures leading to increased pulmonary leakage and/or stress failure and increased E V L W , while the lower intensity exercise (for a longer duration) may not have provided these stimuli. However, the inclusion of a hypoxic condition in this study should have adequately provided for this contingency by providing an added effect of raising PAP further than with sustained exercise alone. Further, the results of a recent unpublished study from our lab involving 41  severe short-term exercise (five bouts of approximately four minutes each) in hypoxia  74  also failed to demonstrate increased E V L W as measured by CT scan. Thus, if exercise mode is the key to the different findings, it appears that the very specific mode of 45 min ending with a sprint was the only mode to induce pulmonary oedema.  Secondly, it is possible that demonstration of increased E V L W is an individual response not uniform across the population of aerobically trained human subjects. Interestingly, McKenzie et a l  84  found increased E V L W in only four of eight subjects with the  remaining four showing no change. A significant increase in mean E V L W was found as a result of the changes in these four subjects, but the lack of change in 50% of the subjects is physiologically relevant. Perhaps importantly, the four subjects in whom there was an increase in lung density following exercise had a significantly higher resting lung density than those who showed no change (this significant difference was not reported in McKenzie et al, but was calculated from the data reported in the results). It is possible that a segment of this population demonstrates increased E V L W following exercise while •  84.  the remainder does not, and McKenzie et al happened to include some of these "responder" subjects while the present study did not. This possibility cannot be accounted for in the statistical power calculations made in either study, as there is no estimate of what percentage of this population may fall into this "responder" category. If this is the case, however, there is no way of discriminating these subjects through any physiological measure used in the study: no relationship was found between VC^max or minimum SaC>2 and degree of increased E V L W by McKenzie et a l . That fact alone 84  discredits this theory of "responder" and "non-responder" subjects, at least in as much as it refers to pulmonary oedema as a mechanism for EIAH, and further the fact that no  75  "responders" were found in the present study or the subsequent unpublished study  41  involving severe hypoxic exercise suggests they would be a very small proportion of the population indeed. Given that approximately 50% of aerobically trained male subjects demonstrate EIAH, it is very unlikely indeed that pulmonary oedema is the sole mechanism responsible for EIAH, even if the theory of "responders" and "nonresponders" were true. And if that were the case, it is unlikely that McKenzie et a l  84  would have found 50% "responders", while this study found none. This argument is less convincing for the fact that the present study did not include a subset of "responders" and therefore, in short, the "responder" and "non-responder" scenario is statistically unlikely. Nonetheless, it is possibly worth further examination and comparison of physiological variables of the two groups in further studies, if this pattern is ever reproducible.  This hypothesis of responders and non-responders is worth examining in light of recent evidence of intrapulmonary shunting of blood during exercise. In two recent studies using agitated saline contrast echocardiography, evidence of intrapulmonary shunting in the form of contrast bubbles visible in the left heart following injection in a vein occurred in approximately 90% of subjects (21 of 23 subjects in a study by Eldridge et al,  31  and  seven of eight subjects in a study by Stickland et al ). While merely speculative, it is 114  possible that the development of intrapulmonary shunts provides a protective effect. Therefore the possibility that those subjects who do not demonstrate intrapulmonary shunting of blood during exercise are more likely to develop transient pulmonary oedema due to increased PAP, is worth further investigation.  76  One of the limitations of lung density measures in human is that they are necessarily indirect. Further, the techniques that offer the most sensitive quantitative measure of lung density, including CT and MRI, generally require a period of time between the end of exercise and measurement of lung density to allow pulmonary blood flow to return to normal resting values. As a result there is uncertainty whether transient pulmonary oedema occurred and resolved prior to the M R scan, or whether no transient pulmonary oedema occurred during exercise. Clearance of alveolar oedema is a complicated process involving active transepithelial sodium transport by sodium-potassium A T P a s e ' 11  115  and  is likely to take an extended period of time to occur. But interstitial oedema, which is perhaps the more likely level of oedema to result from the perturbations used in this study given that none of the clinical signs of alveolar oedema were observed, may clear much more quickly. Furthermore, exercise-induced hyperpnoea has been shown to increase lymph clearance of interstitial oedema in sheep  indicating that the exercise  these subjects performed may have paradoxically assisted in the clearance of any oedema that did occur through post-exercise hyperpnoea. A recent study of horses has shown increased transvascular fluid flux in the pulmonary vasculature during exercise when compared to rest,  121  suggesting that, in these animals at least, fluid movement across the  blood gas barrier is increased during exercise even if there is no accumulation of oedema. Unfortunately, at least in human subjects, the limitations involved in direct measurement of pulmonary oedema are unlikely to be resolved in the near future.  The main questions raised in this study are whether sustained exercise can cause increased E V L W , and whether this represents sub-clinical pulmonary oedema that may  77  be a mechanism of EIAH. The first question is answered quite decisively by the lack of difference in lung density between any of the M R scans. Arterial oxyhaemoglobin saturation during exercise in the hypoxic condition was in the range of 80 - 85 (Figure 4.6), indicating that the perturbations of exercise in combination with hypoxia met the conditions associated with raised mean pulmonary arterial pressure. But even following 60 min of sustained intense exercise in hypoxia, no subject in this study demonstrated an increase in lung density of any significance. The greatest increase in lung density by any single subject was 0.011 g-mL" which is much less than the range of changes reported by 1  McKenzie et al  of 0.02 - 0.071 g-mL" for the four subjects that showed increased  E V L W , and the change of 0.04 g-mL" (as measured by CT) reported by Caillaud et a l . 1  16  in subjects following a triathlon. Furthermore, in the present study, the greatest decrease in lung density by a single subject was 0.024 g-mL" which exceeds the greatest single 1  increase. As presented in the results, the variation between subjects was relatively small and the repeatability of resting slice density within subjects gave a significant Casewise correlation of r = 0.73 (p=0.007). Therefore, in this study no change in lung density was found in either condition. The second question is slightly more complex. While this study does not add any strength to the possibility of pulmonary oedema as a mechanism for EIAH, it may not so clearly rule it out. The exercise intervention used was the highest intensity exercise that subjects could sustain for 60 min. E I A H often presents during exercise of short duration and very high intensity, but has also been observed during submaximal exercise of intensities as low as 40% VC^max. While the possibility 98  remains that pulmonary oedema is an integral part of the mechanisms of E I A H during more intense exercise that would elicit higher intrapulmonary capillary pressures, this is  78  not a likely scenario as this study included hypoxic exercise to specifically raise these pressures. The present study provided a strong stimulus for pulmonary capillary leakage to occur over time, during the 60 min of exercise. Increased intrapulmonary pressures during very intense exercise over a short period of time are unlikely to result in more capillary leakage. However, capillary stress failure theory remains a possibility in this situation. If steady-state exercise over a long period of time does not produce an increase in E V L W (as shown in this study), but evidence of E V L W exists following more intense exercise, then stress failure may be the more logical explanation. Signs of stress failure have been observed in horses,  129  dogs, rabbits, ' 79  37 118  and in humans following intense,  57  CO  but not sustained  exercise. The findings of these last two studies of humans by Hopkins  and colleagues are confirmed to some degree by the present finding of no change in lung density following sustained exercise.  79  C H A P T E R F I V E : General Summary and Conclusions  Exercise-induced arterial hypoxaemia (EIAH) is an established and well-studied response to exercise that occurs in approximately 50% of aerobically trained male athletes. The mechanisms involved have been extensively examined to the point that it is commonly accepted that there are four possibilities: right to left intrapulmonary shunts, relative alveolar hypoventilation during exercise, ventilation/perfusion (V /Qc) mismatch, and A  diffusion limitation. The focus of this manuscript was to examine diffusion limitation in 25  the lung as a mechanism for EIAH. Specifically, these studies examined the effects of exercise on lung density. The significance of an exercise-induced change in lung density is that it represents an increase in fluid in the lung and, providing the assumption that intravascular fluid is constant at each measure, this fluid represents increased E V L W indicative of sub-clinical pulmonary oedema.  There is evidence of pulmonary oedema following exercise in animals ' 103  129  and  57 82 84  humans ' '  but, as discussed in Chapter 2, there is no consistency in the development  of oedema following exercise in humans. Initially, the present studies may simply be added to the list of work done on pulmonary oedema following either an exercise or pharmacological intervention. But the results of these studies add to the existing literature in several ways. Firstly, the methods used to assess lung density by M R I in these studies have shown to be consistent and reliable: there was a small coefficient of variation for repeated measures of resting lung density; the process involved some qualitative measurement on the part of the researchers, and this process was shown to be reliable 80  both between and within observers; and finally, the values for lung density achieved in this study match those previously reported. When added to the validation studies of Estilaei et a l , ' 34  35  the use of M R I as a measure of lung density assessment appears sound.  Secondly, these studies clearly demonstrate a lack df change in lung density following sustained exercise both in normoxia and hypoxia, which indicates that pulmonary oedema does not occur as a result of exercise as readily as previously indicated. Diffusion limitation cannot be completely ruled out as a mechanism of E I A H from the results of these studies, as pulmonary oedema and diffusion limitation may occur under different exercise conditions, such as those indicated in the findings by McKenzie et al,  84  and further transient oedema may resolve relatively quickly following exercise making detection post-exercise difficult.  The most comprehensive means of examining the issue of pulmonary oedema following exercise in humans would involve multiple exercise protocols under normoxic and hypoxic conditions followed by MRI assessment of lung density and measures of diffusing capacity. However, MRI is prohibitively expensive when used repeatedly in this way, and there are logistical problems associated with measuring diffusing capacity in a limited time frame between exercise and MRI. Nonetheless, further study of this area is warranted, perhaps combining technologies used in the present study with other indices of pulmonary oedema such as bronchoalveolar lavage, and of right to left shunts such as agitated saline contrast echocardiography in an attempt to expand the knowledge base of pulmonary responses to exercise, and further understand the mechanisms of EIAH.  81  Adams, R. P. and H . G. Welch. Oxygen uptake, acid-base status, and performance with varied inspired oxygen fractions. Journal of Applied Physiology 49:863-868, 1980. Angerio, A. D. and P. A . Kot. Pathophysiology of pulmonary edema. Critical Care Nursing Quarterly 17:21-26, 1994. Anholm, J. D., E. N . Milne, P. Stark, J. C. Bourne, and P. Friedman. Radiographic evidence of interstitial pulmonary edema after exercise at altitude. Journal of Applied Physiology 86:503-509, 1999. Asmussen, E. and M . Nielson. Alveolo-arterial gas exchange at rest and during work at different O2 tensions. Acta Physiologica Scandinavica 50:153-166, 1960. Astrand, P. O. The respiratory activity in man exposed to prolonged hypoxia. Acta Physiologica Scandinavica 30:343-368, 195.4. Astrand, P. O. A study of chemoceptor activity in animals exposed to prolonged hypoxia. Acta Physiologica Scandinavica 30:335-342, 1954. Bartsch, P. High altitude pulmonary edema. Respiration 64:435-443, 1997. Bassett, D. R. and E. T. Howley. Maximal oxygen uptake: "classical" versus "contemporary" viewpoints. Medicine and Sciencein Sports and Exercise 29:591603, 1997. Beckmann, N . , B. Tigani, D. Ekatodramis, R. Borer, L. Mazzoni, and J. R. Fozard. Pulmonary edema induced by allergen challenge in the rat: noninvasive assessment by magnetic resonance imaging. Magnetic Resonance in Medicine 45:88-95,2001. Bergh, U . , B . Ekblom, and P. O. Astrand. Maximal oxygen uptake "classical" versus "contemporary" viewpoints. Medicine & Science in Sports & Exercise 32:85-88,2000. Berthiaume, Y . , H . G. Folkesson, and M . A. Matthay. Lung edema clearance: 20 years of progress: invited review: alveolar edema fluid clearance in the injured lung. Journal of Applied Physiology 93:2207-2213, 2002.  12.  Brasileiro, F. C , F. S. Vargas, J. I. Kavakama, J. J. Leite, A . Cukier, and C. Prefaut. High-resolution CT scan in the evaluation of exercise-induced interstitial pulmonary edema in cardiac patients. Chest 111:1577-1582, 1997.  13.  Bruce, R. A . Exercise testing of patients with coronary artery disease. Annals of Clinical Research 3:323-332, 1971.  14.  Buono, M . J., S. H . Constable, A. R. Morton, T. C. Rotkis, P. R. Stanforth, and J. H . Wilmore. The effect of an acute bout of exercise on selected pulmonary function measurements. Medicine and Science in Sports and Exercise 13:290293,1981.  15.  Buono, M . J., J. H . Wilmore, and F. B. Roby. Evidence of increased thoracic extravascular fluid following intense exercise in man. Physiologist 25:201, 1982.  16.  Caillaud, C , O. Serre-Cousine, F. Anselme, X . Capdevilla, and C. Prefaut. Computerized tomography and pulmonary diffusing capacity in highly trained athletes after performing a triathlon. Journal of Applied Physiology 79:12261232,1995.  17.  Calbet, J. A., R. Boushel, G. Radegran, H . Sondergaard, P. D. Wagner, and B. Saltin. Determinants of maximal oxygen uptake in severe acute hypoxia. American Journal of Physiology Regulatory, Integrative and Comparative Physiology 284:R291-303, 2003.  18.  Cardus, J., R. M . Marrades, J. Roca, J. A . Barbera, O. Diaz, J. R. Masclans, R. Rodriguez-Roisin, and P. D. Wagner. Effects of F(I)02 on leg V 0 2 during cycle ergometry in sedentary subjects. Medicine & Science in Sports & Exercise 30:697-703, 1998.  19.  Caruthers, S. D., C. B . Paschal, N . A. Pou, R. J. Roselli, and T. R. Harris. Regional measurements of pulmonary edema by using magnetic resonance imaging. Journal of Applied Physiology 84:2143-2153, 1998.  20.  Chapman, R. F., M . Emery, and J. M . Stager. Degree of arterial desaturation in normoxia influences V02max decline in mild hypoxia. Medicine & Science in Sports & Exercise 31:658-663, 1999.  21.  Clausen, J. P., K . Klausen, B. Rasmussen, and J. Trapjensen. Central and peripheral circulatory changes after training of the arms and legs. American Journal of Physiology 255:67'5-682, 1973.  83  22.  Cudkowicz, L. Bronchial circulation in man: normal anatomy and response to disease. In: Pulmonary vascular disease. K . M . Moser (Ed.) New York, N Y : Dekker, 1977, pp. 65-71.  23.  Dempsey, J. A . J.B. Wolffe memorial lecture. Is the lung built for exercise? Medicine & Science in Sports & Exercise 18:143-155, 1986.  24.  Dempsey, J. A., P. G. Hanson, and K. S. Henderson. Exercise-induced arterial hypoxaemia in healthy human subjects at sea level. Journal of Physiology 355:161-175, 1984.  25.  Dempsey, J. A . and P. D. Wagner. Exercise-induced arterial hypoxemia. Journal of Applied Physiology 87:1997-2006, 1999.  26.  Dill, D. B . and W. C. Adams. Maximal oxygen uptake at sea level and at 3,090-m altitude in high school champion runners. Journal of Applied Physiology 30:854859, 1971.  27.  Droma, Y . , T. Hayano, Y . Takabayashi, T. Koizumi, K . Kubo, T. Kobayashi, and M . Sekiguchi. Endothelin-1 and interleukin-8 in high altitude pulmonary oedema. European Respiratory Journal 9:1947-1949, 1996.  28.  Durand, F., P. Mucci, and C. Prefaut. Evidence for an inadequate hyperventilation inducing arterial hypoxemia at submaximal exercise in all highly trained endurance athletes. Medicine & Science in Sports & Exercise 32:926-932, 2000.  29.  Edwards, M . R., G. S. Hunte, A . S. Belzberg, A . W. Sheel, D. F. Worsley, and D. C. McKenzie. Alveolar epithelial integrity in athletes with exercise-induced hypoxemia. Journal of Applied Physiology 89:1537-1542, 2000.  30.  Ekblom, B., R. Huot, E. M . Stein, and A . T. Thorstensson. Effect of changes in arterial oxygen content on circulation and physical performance. Journal of Applied Physiology 39:71-75, 1975.  31.  Eldridge, M . W., J. A . Dempsey, H . C. Haverkamp, A . T. Lovering, and J. S. Hokanson. Exercise-induced intrapulmonary arteriovenous shunting in healthy humans. Journal of Applied Physiology 97:797-805, 2004.  32.  Elliott, A . R., Z. Fu, K . Tsukimoto, R. Prediletto, O. Mathieu-Costello, and J. B . West. Short-term reversibility of ultrastructural changes in pulmonary capillaries caused by stress failure. Journal of Applied Physiology 73:1150-1158, 1992.  84  33.  Elliott, P. R. and H . A. Atterbom. Comparison of exercise responses of males and females during acute exposure to hypobaria. Aviation Space and Environmental Medicine 49:415-418, 1978.  34.  Estilaei, M . , A . MacKay, C. Roberts, and J. Mayo. 1H N M R measurements of wet/dry ratio and T l , T2 distributions in lung. Journal of Magnetic Resonance 124:410-419, 1997.  35.  Estilaei, M . , A . MacKay, K . Whittall, and J. Mayo. In vitro measurements of water content and T2 relaxation times in lung using a clinical M R I scanner. Journal of Magnetic Resonance Imaging 9:699-703, 1999.  36.  Ferretti, G., C. Moia, J. M . Thomet, and B. Kayser. The decrease of maximal oxygen consumption during hypoxia in man: a mirror image of the oxygen equilibrium curve. Journal of Physiology 498 ( Pt 1):231-237, 1997.  37.  Fu, Z., M . L. Costello, K . Tsukimoto, R. Prediletto, A. R. Elliott, O. MathieuCostello, and J. B . West. High lung volume increases stress failure in pulmonary capillaries. Journal of Applied Physiology 73:123-133, 1992.  38.  Fulco, C. S., P. B . Rock, and A . Cymerman. Maximal and submaximal exercise performance at altitude. Aviation Space and Environmental Medicine 69:793-801, 1998.  39.  Gallagher, C. G., W. Huda, M . Rigby, D. Greenberg, and M . Younes. Lack of radiographic evidence of interstitial pulmonary edema after maximal exercise in normal subjects. The American Review of Respiratory Disease 137:474-476, 1988.  40.  Groves, B. M . , J. T. Reeves, J. R. Sutton, P. D. Wagner, A . Cymerman, M . K . Malconian, P. B . Rock, P. M . Young, and C. S. Houston. Operation Everest II: elevated high-altitude pulmonary resistance unresponsive to oxygen. Journal of Applied Physiology 63:521-530, 1987.  41.  Guenette, J. A., M . J. MacNutt, J. D. Witt, A . N . H . Hodges, J. R. Mayo, and D. C. McKenzie. Does strenuous exercise induce pulmonary oedema? Unpublished, 2005.  42.  Hachenberg, T., H . Lundquist, L. Tokics, B. Brismar, and G. Hedenstierna. Analysis of lung density by computed tomography before and during general anaesthesia. Acta Anaesthesiologica Scandinavica 37:549-555, 1993.  85  43.  Hackett, P. H., J. Bertman, G. Rodriguez, and J. Tenney. Pulmonary edema fluid protein in high-altitude pulmonary edema. Journal of the American Medical Association 256:36, 1986.  44.  Hammond, M . D., G. E. Gale, K. S. Kapitan, A . Ries, and P. D. Wagner. Pulmonary gas exchange in humans during exercise at sea level. Journal of Applied Physiology 60:1590-1598, 1986.  45.  Hanel, B. Pulmonary function after exercise with special emphasis on diffusion capacity. Danish Medical Bulletin 47:196-217, 2000.  46.  Hanel, B., P. S. Clifford, and N . H . Secher. Restricted postexercise pulmonary diffusion capacity does not impair maximal transport for O2. Journal of Applied Physiology 77:2408-2412, 1994.  47.  Hanel, B., F. Gustafsson, H . H . Larsen, and N . H . Secher. Influence of exercise intensity and duration on post-exercise pulmonary diffusion capacity. International Journal of Sports Medicine 14 Suppl 1:S 11-14, 1993.  48.  Hanel, B., I. Teunissen, A . Rabol, J. Warberg, and N . H . Secher. Restricted postexercise pulmonary diffusion capacity and central blood volume depletion. Journal of Applied Physiology 83:11-17, 1997.  49.  Harms, C. A., S. R. McClaran, G. A. Nickele, D. F. Pegelow, W. B . Nelson, and J. A . Dempsey. Effect of exercise-induced arterial 02 desaturation on V02max in women. Medicine & Science in Sports & Exercise 32:1101-1108, 2000.  50.  Harms, C. A., S. R. McClaran, G. A. Nickele, D. F. Pegelow, W. B . Nelson, and J. A . Dempsey. Exercise-induced arterial hypoxaemia in healthy young women. Journal of Physiology 507:619-628, 1998.  51.  Harms, C. A . and J. M . Stager. Low chemoresponsiveness and inadequate hyperventilation contribute to exercise-induced hypoxemia. Journal of Applied Physiology 79:575-580, 1995.  52.  Hatabu, H., D . C. Alsop, J. Listerud, M . Bonnet, and W. B . Gefter. T2* and proton density measurement of normal human lung parenchyma using submillisecond echo time gradient echo magnetic resonance imaging. European Journal of Radiology 29:245-252, 1999.  53.  Holmgren, A . and H . Linderholm. Oxygen and carbon dioxide tensions of arterial blood during heavy and exhaustive exercise. Acta Physiologica Scandinavica 44:203-215, 1958.  86  Hopkins, S. R., A . S. Belzberg, B . R. Wiggs, and D. G. McKenzie. Pulmonary transit time and diffusion limitation during heavy exercise in athletes. Respiration Physiology 103:67-73, 1996. Hopkins, S. R., T. P. Gavin, N . M . Siafakas, L. J. Haseler, I. M . Olfert, H . Wagner, and P. D. Wagner. Effect of prolonged, heavy exercise on pulmonary gas exchange in athletes. Journal of Applied Physiology 85:1523-1532, 1998. Hopkins, S. R., D. C. McKenzie, R. B. Schoene, R. W. Glenny, and H . T. Robertson. Pulmonary gas exchange during exercise in athletes. I. Ventilationperfusion mismatch and diffusion limitation. Journal of Applied Physiology 77:912-917, 1994. Hopkins, S. R., R. B . Schoene, W. R. Henderson, R. G . Spragg, T. R. Martin, and J. B . West. Intense exercise impairs the integrity of the pulmonary blood-gas barrier in elite athletes. American Journal of Respiratory & Critical Care Medicine 155:1090-1094, 1997. Hopkins, S. R., R. B . Schoene, W. R. Henderson, R. G. Spragg, and J. B . West. Sustained submaximal exercise does not alter the integrity of the lung blood-gas barrier in elite athletes. Journal of Applied Physiology 84:1185-1189, 1998. Hughes, R. L., M . Clode, R. H . Edwards, T. J. Goodwin, and N . L. Jones. Effect of inspired 0 2 on cardiopulmonary and metabolic responses to exercise in man. Journal of Applied Physiology 24:336-347, 1968. Hurley, J. V . Current views on the mechanisms of pulmonary oedema. Journal of Pathology 125:59-79, 1978. Johnson, R. L., G. J. F. Heigenhauser, C. C. W. Hsia, N . L. Jones, and P. D . Wagner. Determinants of gas exchange and acid-base balance during exercise. In: Handbook of physiology Section 12: Exercise: Regulation and integration of multiple systems. L. B . Rowell and J. T. Shepherd (Eds.) New York, N Y : Oxford University Press, 1996, pp. 515-584. Klausen, K., D. B . Dill, and S. M . Horvath. Exercise at ambient and high oxygen pressure at high altitude and at sea level. Journal of Applied Physiology 29:456463, 1970. Knuttgen, H . G. and B . Saltin. Oxygen uptake, muscle high-energy phosphates, and lactate in exercise under acute hypoxic conditions in man. Acta Physiologica Scandinavica 87:368-376, 1973.  87  64.  Kobayashi, T., S. Koyama, K . Kubo, M . Fukushima, and S. Kusama. Clinical features of patients with high-altitude pulmonary edema in Japan. Chest 92:814821, 1987.  65.  Kohda, E. and N . Shigematsu. Measurement of lung density by computed tomography: implication for radiotherapy. The Keio Journal of Medicine 38:454463, 1989. Koistinen, P., T. Takala, V . Martikkala, and J. Leppaluoto. Aerobic fitness influences the response of maximal oxygen uptake and lactate threshold in acute hypobaric hypoxia. International Journal of Sports Medicine 16:78-81, 1995. Koizumi, T., R. J. Roselli, R. E. Parker, C. I. Hermo-Weiler, M . Banerjee, and J. H . Newman. Clearance of filtered fluid from the lung during exercise: role of hyperpnea. American Journal of Respiratory and Critical Care Medicine 163:614-618., 2001. Koskolou, M . D . and D. C. McKenzie. Arterial hypoxemia and performance during intense exercise. European Journal of Applied Physiology & Occupational Physiology 68:80-86, 1994. Kumar, P. J. and M . L . Clark. Clinical medicine. Letchworth, U . K . : Mackays of Chatham, 1987 Lawler, J., S. K . Powers, and D. Thompson. Linear relationship between V02max and V02max decrement during exposure to acute hypoxia. Journal of Applied Physiology 64:1486-1492, 1988. Mairbaurl, H., R. Wodopia, S. Eckes, S. Schulz, and P. Bartsch. Impairment of cation transport in A549 cells and rat alveolar epithelial cells by hypoxia. American Journal of Physiology 273:L797-806, 1997.  72.  Manier, G., M . Duclos, L. Arsac, J. Moinard, and F. Laurent. Distribution of lung density after strenuous, prolonged exercise. Journal of Applied Physiology 87:8389, 1999.  73.  Manier, G., J. Moinard, and H. Stoicheff. Pulmonary diffusing capacity after maximal exercise. Journal of Applied Physiology 75:2580-2585, 1993. Manohar, M . , T. E. Goetz, and A. S. Hassan. Effect of prior high-intensity exercise on exercise-induced arterial hypoxemia in Thoroughbred horses. Journal of Applied Physiology 90:2371-2377, 2001.  88  75.  Marshall, B. E., L. R. Soma, and G. R. Neufeld. Lung water volume at rest and exercise in dogs. Journal of Applied Physiology 39:7-8, 1975.  76.  Marshall, B. E., R. L . Teichner, T. Kallos, H. J. Sugerman, M . Q. Wyche, Jr., and K . R. Tantum. Effects of posture and exercise on the pulmonary extravascular water volume in man. Journal of Applied Physiology 31:375-379, 1971.  77.  Martin, D. and J. O'Kroy. Effects of acute hypoxia on the VC^max of trained and untrained subjects. Journal of Sports Sciences 11:37-42, 1993.  78.  Martin, D., S. Powers, M . Cicale, N . Collop, D. Huang, and D. Criswell. Validity of pulse oximetry during exercise in elite endurance athletes. Journal of Applied Physiology 72:455-458, 1992.  79.  Mathieu-Costello, O., D. C. Willford, Z. Fu, R. M . Garden, and J. B . West. Pulmonary capillaries are more resistant to stress failure in dogs than in rabbits. Journal of Applied Physiology 79:908-917, 1995.  80.  Mayo, J. R., A . L. MacKay, K . P. Whittall, E. M . Baile, and P. D. Pare. Measurement of lung water content and pleural pressure gradient with magnetic resonance imaging. Journal of Thoracic Imaging 10:73-81, 1995.  81.  McArdle, W. D., F. I. Katch, and V . L. Katch. Exercise Physiology. 5th ed. Philadelphia, P A : Lippincott Williams & Williams, 2001,1158.  82.  McKechnie, J. K., W. P. Leary, T. D. Noakes, J. C. Kallmeyer, E. T. MacSearraigh, and L. R. Olivier. Acute pulmonary oedema in two athletes during a 90-km running race. South African Medical Journal 56:261-265, 1979.  83.  McKenzie, D. C., I. L . Lama, J. E. Potts, A. W. Sheel, and K . D. Coutts. The effect of repeat exercise on pulmonary diffusing capacity and EIH in trained athletes. Medicine & Science in Sports & Exercise 31:99-104, 1999.  84.  McKenzie, D. C , T. J. O'Hare, and J. Mayo. The effect of sustained heavy exercise on the development of pulmonary edema in trained male cyclists. Respiratory Physiology & Neurobiology 145:209-218, 2005.  85.  Millar, A. B . and D. M . Denison. Vertical gradients of lung density in healthy supine men. Thorax 44:485-490, 1989.  86.  Ozcelik, O. and H . Kelestimur. Effects of acute hypoxia on the estimation of lactate threshold from ventilatory gas exchange indices during an incremental exercise test. Physiological Research 53:653-659, 2004.  89  87.  Paterson, D. J., H . Pinnington, A . R. Pearce, and A . R. Morton. Maximal exercise cardiorespiratory responses of men and women during acute exposure to hypoxia. Aviation Space and Environmental Medicine 58:243-247, 1987.  88.  Peltonen, J. E., H . O. Tikkanen, and H. K. Rusko. Cardiorespiratory responses to exercise in acute hypoxia, hyperoxia and normoxia. European Journal of Applied Physiology 85:82-88, 2001.  89.  Planes, C , G. Friedlander, A . Loiseau, C. Amiel, and C. Clerici. Inhibition of NaK-ATPase activity after prolonged hypoxia in an alveolar epithelial cell line. American Journal of Physiology 271 :L70-78, 1996.  90.  Powers, S. K., S. Dodd, J. Lawler, G. Landry, M . Kirtley, T. McKnight, and S. Grinton. Incidence of exercise induced hypoxemia in elite athletes at sea level. European Journal of Applied Physiology 58:298-302, 1988.  91.  Powers, S. K., J. Lawler, J. A . Dempsey, S. Dodd, and G. Landry. Effects of incomplete pulmonary gas exchange on V O 2 max- Journal of Applied Physiology 66:2491-2495, 1989.  92.  Powers, S. K . , D . Martin, M . Cicale, N . Callop, D . Huang, and D. Criswell. Exercise-induced hypoxemia in athletes: role of inadequate hyperventilation. European Journal of Applied Physiology 65:37-42, 1992.  93.  Powers, S. K., D. Martin, and S. Dodd. Exercise-induced hypoxaemia in elite endurance athletes. Incidence, causes and impact on V 0 2 m a x - Sports Medicine 16:14-22, 1993.  94.  Prefaut, C , F. Durand, P. Mucci, and C. Caillaud. Exercise-induced arterial hypoxaemia in athletes: a review. Sports Medicine 30:47-61, 2000.  95.  Rasmussen, B. S., P. Elkjaer, and B. Juhl. Impaired pulmonary and cardiac function after maximal exercise. Journal of Sports Sciences 6:219-228, 1988.  96.  Rasmussen, B . S., B. Hanel, K. Jensen, B. Serup, and N . H . Secher. Decrease in pulmonary diffusion capacity after maximal exercise. Journal of Sports Sciences 4:185-188, 1986.  97.  Rasmussen, J., B. Hanel, K . Saunamaki, and N . H . Secher. Recovery of pulmonary diffusing capacity after maximal exercise. Journal of Sports Sciences 10:525-531, 1992.  90  98.  Rice, A. J., G. C. Scroop, C. J. Gore, A . T. Thornton, M . A . Chapman, H . W. Greville, M . D. Holmes, and R. Scicchitano. Exercise-induced hypoxaemia in highly trained cyclists at 40% peak oxygen uptake. European Journal of Applied Physiology & Occupational Physiology 79:353-359, 1999.  99. . Richards, J. C , D. C. McKenzie, D. E. Warburton, J. D. Road, and A . W. Sheel. Prevalence of exercise-induced arterial hypoxemia in healthy women. Med Sci Sports Exerc 36:1514-1521, 2004. 100.  Robertson, W. G., D. G. Quigley, and G. L. McRae. Effect of exercise on oxygen consumption at decreased pressure. Aerospace Medicine 38:617-619, 1967.  101.  Rowell, L. B., L. T. Taylor, Y . Wang, and W. S. Carlson. Saturation of arterial blood with oxygen during maximal exercise. Journal of Applied Physiology 19:284-286, 1964.  102.  Saltin, B . and S. Strange. Maximal oxygen uptake: "old" and "new" arguments for a cardiovascular limitation. Medicine and Science in Sports and Exercise 24:3037, 1992.  103.  Schaffartzik, W., J. Arcos, K . Tsukimoto, O. Mathieu-Costello, and P. D. Wagner. Pulmonary interstitial edema in the pig after heavy exercise. Journal of Applied Physiology 75:2535-2540, 1993.  104.  Schaffartzik, W., D. C. Poole, T. Derion, K. Tsukimoto, M . C. Hogan, J. P. Arcos, D. E. Bebout, and P. D. Wagner. V A / Q distribution during heavy exercise and recovery in humans: implications for pulmonary edema. Journal of Applied Physiology 72:1657-1667, 1992.  105.  Scherrer, U . , L. Vollenweider, A. Delabays, M . Savcic, U . Eichenberger, G. R. . Kleger, A . Fikrle, P. E. Ballmer, P. Nicod, and P. Bartsch. Inhaled nitric oxide for high-altitude pulmonary edema. New England Journal of Medicine 334:624-629, 1996.  106.  Schoene, R. B., E. R. Swenson, C. J. Pizzo, P. H . Hackett, R. C. Roach, W. J. Mills, Jr., W. R. Henderson, Jr., and T. R. Martin. The lung at high altitude: bronchoalveolar lavage in acute mountain sickness and pulmonary edema. Journal of Applied Physiology 64:2605-2613, 1988.  107.  Secher, N . H., J. P. Clausen, K. Klausen, I. Noer, and J. Trap-Jensen. Central and regional circulatory effects of adding arm exercise to leg exercise. Acta Physiologica Scandinavica 100:288-297, 1977.  91  108.  Sheel, A . W., K . D. Coutts, J. E. Potts, and D. C. McKenzie. The time course of pulmonary diffusing capacity for carbon monoxide following short duration high intensity exercise. Respiration Physiology 111 :271-281, 1998.  109.  Snashall, P. D. and J. M . Hughes. Lung water balance. Reviews of Physiology Biochemistry and Pharmacology 89:5-62, 1981.  110.  Snashall, P. D., S. J. Keyes, B. M . Morgan, R. J. McAnulty, P. F. Mitchell-Heggs, J. M . McLvor, and K . A . Howlett. The radiographic detection of acute pulmonary oedema. A comparison of radiographic appearances, densitometry and lung water in dogs. British Journal of Radiology 54:277-288, 1981.  111.  Squires, R. W. and E. R. Buskirk. Aerobic capacity during acute exposure to simulated altitude, 914 to 2286 meters. Medicine and Science in Sports and Exercise 14:36,40, 1982.  112.  St Croix, C. M . , C. A . Harms, S. R. McClaran, G. A. Nickele, D . F. Pegelow, W. B. Nelson, and J. A . Dempsey. Effects of prior exercise on exercise-induced arterial hypoxemia in young women. Journal of Applied Physiology 85:15561563,1998.  113.  Starling, E. H . Absorbtion of fluids from the connective tissue spaces. Journal of Physiology 19:312-326, 1896.  114.  Stickland, M . K., R. C. Welsh, M . J. Haykowsky, S. R. Petersen, W. D. Anderson, D. A . Taylor, M . Bouffard, and R. L. Jones. Intra-pulmonary shunt and pulmonary gas exchange during exercise in humans. Journal of Physiology 561:321-329, 2004.  115.  Sznajder, J. I., P. Factor, and D. H . Ingbar. Invited review: lung edema clearance: role of Na(+)-K(+)-ATPase. Journal of Applied Physiology 93:1860-1866, 2002.  116.  Tigani, B., E . Schaeublin, R. Sugar, A . D . Jackson, J. R. Fozard, and N . Beckmann. Pulmonary inflammation monitored noninvasively by M R I in freely breathing rats. Biochemical & Biophysical Research Communications 292:216221,2002.  117.  Torre-Bueno, J. R., P. D. Wagner, H . A . Saltzman, G. E. Gale, and R. E. Moon. Diffusion limitation in normal humans during exercise at sea level and simulated altitude. Journal of Applied Physiology 58:989-995, 1985.  92  118.  Tsukimoto, K., O. Mathieu-Costello, R. Prediletto, A. R. Elliott, and J. B. West. Ultrastructural appearances of pulmonary capillaries at high transmural pressures. Journal of Applied Physiology 71:573-582, 1991.  119.  Van Dyk, J., T. J. Keane, and W. D. Rider. Lung density as measured by computerized tomography: implications for radiotherapy. International Journal of Radiation Oncology, Biology, Physics 8:1363-1372, 1982.  120.  Vaughan, T. R., Jr., E. DeMarino, and N . C. Staub. Indicator dilution lung water and capillary blood volume in prolonged heavy exercise in normal men. The American Review of Respiratory Disease 113:757-762, 1976.  121.  Vengust, M . , H . Staempfli, L. Viel, and G. Heigenhauser. Transvascular fluid flux from the pulmonary vasculature at rest and during exercise in horses. J Physiol 570:397-405, 2006.  122.  Viswanathan, R., S. K . Jain, S. Subramanian, and B. K . Puri. Pulonary edema of high altitude. I. Production of pulmonary edema in animals under conditions of simulated high altitude. The American Review of Respiratory Disease 100:327333, 1969.  123.  Wagner, P. D., G. E. Gale, R. E. Moon, J. R. Torre-Bueno, B . W. Stolp, and H. A . Saltzman. Pulmonary gas exchange in humans exercising at sea level and simulated altitude. Journal of Applied Physiology 61:260-270, 1986.  124.  Warren, G. L., K . J. Cureton, W. F. Middendorf, C. A . Ray, and J. A . Warren. Red blood cell pulmonary capillary transit time during exercise in athletes. Medicine & Science in Sports & Exercise 23:1353-1361, 1991.  125.  West, J. B . Invited review: pulmonary capillary stress failure. Journal of Applied Physiology 89:2483-2489;discussion 2497, 2000.  126.  West, J. B. Respiratory Physiology The Essentials. 6th ed. Philadelphia, PA: Lippincott Williams and Wilkins, 2000, 171.  127.  West, J. B., S. Lahiri, M . B. Gill, J. S. Milledge, L. G. Pugh, and M . P: Ward. Arterial oxygen saturation during exercise at high altitude. Journal of Applied Physiology 17:617-621, 1962.  128.  West, J. B . and O. Mathieu-Costello. Structure, strength, failure, and remodeling of the pulmonary blood-gas barrier. Annual Review of Physiology 61:543-572, 1999.  93  129.  West, J. B., O. Mathieu-Costello, J. H . Jones, E. K . Birks, R. B . Logemann, J. R. Pascoe, and W. S. Tyler. Stress failure of pulmonary capillaries in racehorses with exercise- induced pulmonary hemorrhage. Journal of Applied Physiology 75:1097-1109, 1993.  130.  Woorons, X . , P. Mollard, C. Lamberto, M . Letournel, and J. P. Richalet. Effect of acute hypoxia on maximal exercise in trained and sedentary women.' Medicine and Science in Sports and Exercise 37:147-154, 2005.  94  APPENDIX A. Individual data from Chapter 3  Age Subject 1 2 3 4 5 6 7 8 9 10  Height (cm) 166.2 183.7 173.1 176.5 182.6 174.5 184.7 186.2 192.9 175.5  (y)  37 26 27 20 33 33 21 33 33 33  Mass (kg) 71.2 83.0 73.2 80.2 81.0 74.7 73.8 85.9 83.3 87.0  Table A. 1. Individual descriptive data.  Subject 1 2 3 4 5 6 7 8 9 10  HR (b-min") 188 183 187 191 180 189 193 184 174 187  VE (L-min ) 119.0 138.8 126.9 174.3 146.3 129.2 181.4 155.3 140.7 151.6 -  VChmax (L-min ) 3.7 3.9 3.7 5.2 4.6 4.5 5.3 5.7 5.8 .4.7 1  V02inax (mL-kg^-min" ) 51.2 46.6 49.4 63.5 56.2 58.8 71.8 65.4 55.8 52.9 1  Peak Minimum Power Sa0 (W) (%) 347 96.9 370 96.9 341 95.4 407 94.9 363 96.1 406 93.8 430 94.6 491 94.5 452 93.0 400 95.0 2  Table A.2. Individual maximal exercise data (F1O2 = 21%).  95  Subject 1 2 3 4 5 6 7 8 9 10  HR (b-min ) 185 181 189 189 181 182 196 178 177 182 1  VE (Lmin ) 107.5 143.0 123.7 123.8 144.7 130.6 165.2 164.1 128.8 147.2 1  VChmax (L-min ) 3.5 4.4 3.6 3.6 4.4 4.3 4.8 5.2 4.3 4.7 1  V02inax (mL-kg^-min" ) 48.2 50.0 48.2 48.2 52.4 55.5 65.7 59.7 50.7 53.3 1  Peak Minimum Power Sa0 (W) (%) 333 91.2 362 94.4 325 92.0 325 92.0 363 95.0 378 88.8 415 90.3 452 92.6 408 90.0 393 89.3 2  Table A.3. Individual maximal exercise data (F1O2 = 18%).  Subject 1 2 3 4 5 6 7 8 9 10  HR (b-min ) 185 184 188 186 187 179 188 178 173 178 1  VE (Lmin ) 118.9 129.3 107.5 158.9 137.6 138.4 163.1 158.1 129.7 138.9 1  V0 max (Lmin ) 3.4 3.9 3.3 4.3 4.1 4.2 4.5 4.5 4.0 2.8 2  1  V0 max (mL-kg^-min ) 48.2 44.6 41.5 53.1 48.3 54.8 59.9 52.8 47.0 31.7 2  1  Peak Power (W) 318 348 291 378 347 355 382 430 413 355  Minimum Sa0 (%) 84.7 87.1 85.0 84.8 86.0 76.5 79.9 87.8 81.0 80.5 2  Table A.4. Individual maximal exercise data (F1O2 = 15%).  96  Subject 1 2 3 4 5 6 7 8 9 10  HR (b-min' ) 177 162 131 153 171 128 174 157 169 141 1  VE (L-min' ) 114.2 65.7 43.0 59.3 137.1 33.7 79.8 111.1 99.4 53.9 1  V0 max (L-min ) 2.7 2.3 2.1 2.4 3.6 1.6 2.8 3.2 3.0 2.1  V0 max (mL-kg^-min ) 37.4 26.5 20.8 28.0 43.2 27.8 37.1 36.6 35.5 22.4  2  2  1  1  Peak Minimum Power SaO (W) (%) 273 74.4 204 72.8 176 75.5 153 68.7 137 73.7 71 70.8 243 69.7 273 74.0 317 64.9 168 70.1 z  Table A. 5. Individual maximal exercise data (Fi0 = 12%). 2  97  Figure A. 1. Individual minimum Sa0 with varying F i 0 . 2  2  APPENDIX B. Individual exercise data from Chapter 4  Subject 1 2 3 4 5 6 7 8 9 10  Age (y) 24 26 23 34 34 23 22 29 21 24  Height (cm) 190.0 177.5 189.0190.0 184.5 180.5 167.5 192.0 175.5 195.0  Mass (kg) 87.5 69.5 89.9 84.5 81.6 67.0 62.6 89.4 80.4 73.5  FVC (L) 7.3 5.3 6.8 6.7 5.6 4.4 5.4 7.2 3.9 6.7  F E V 1 / FVC (%) 80.7 81.9 86.7 81.1 80.8 81.1 86.2 86.4 80.9 81.1  Table B . l . Individual descriptive and spirometry data.  Subject 1 2 3 4 5 6 7 8 9 10  HR (b-min" ) 189 176 184 182 183 192 190 191 184 196 1  VE (L-min ) 188.0 123.4 177.0 153.1 126.1 144.9 134.1 189.3 157.5 158.8 1  V0 max (L-min ) 5.0 4.3 5.5 4.7 4.7 4.7 4.5 5.8 5.1 5.8  V0 max (mL-kg^-min ) 56.8 66.4 61.4 56.2 57.1 69.4 72.2 66.5 64.5 79.5  2  2  1  1  Peak Power (W) 445 355 450 393 420 460 460 604 480 500  Minimum Sa0 (%) 94.5 93.8 90.8 95.3 91.6 92.6 92.5 89.5 91.9 90.6 2  Table B.2. Individual maximal exercise data (Fi0 = 21%). 2  99  Subject 1 2 3 4 5 6 7 8 9 10  HR (b-min ) 180 174 180 180 179 184 178 159 184 186 1  VE (L-min ) 168.0 132.5 158.9 134.3 145.0 132.1 147.0 75.2 165.8 152.1  V0 max (Lmin ) 4.3 3.9 4.4 4.5 4.1 4.0 3.6 3.6 4.7 4.4  VO^max (mL-kq^-min ) 50.6 58.5 49.0 52.3 50.4 60.0 57.3 39.8 58.8 63.9  2  1  1  1  Peak Power (W) 386 326 378 355 369 401 393 341 408 415  Minimum Sa0 (%) . 83.0 83.5 79.0 86.2 77.1 81.6 76.7 72.1 83.4 74.9 2  Table B.3. Individual maximal exercise data ( F i 0 = 15%). 2  Subject 1 2 3 4 5 6 7 8 9 10  HR (b-min ) 161.7 ± 3.9 140.6 ± 6 . 8 146.8 ± 7 . 2 145.7 ± 5 . 3 153.9 ± 4 . 8 160.6 ± 13.0 164.1 ± 9 . 7 163.0 ± 4 . 7 161.5 ± 6 . 9 170.7 ± 5 . 4 1  v  vo  E  (L-min ) 63.9 ± 5 . 2 47.0 ± 3 . 5 74.4 ± 5 . 5 57.9 ± 4 . 6 67.7 ± 4 . 6 70.1 ± 9 . 9 76.3 ± 8.8 40.5 ± 3 . 6 72.9 ± 3 . 5 62.7 ± 10.6 1  2  (mL-kg^-min ) 37.5 ± 1 . 9 37.2 ± 1.5 38.7 ± 2 . 3 33.4 ± 2 . 2 41.0 ± 0 . 9 46.9 ± 2.9 51.7 ± 2 . 9 28.1 ± 1.7 43.0 ± 1.3 40.9 ± 5 . 3 1  Power (W) 246.5 ± 2.2 173.9 ± 1.5 235.8 ± 7 . 5 218.5 ± 5 . 7 237.3 ± 1.7 257.0 ± 2 8 . 1 275.0 ± 10.0 383.0 ± 6 . 8 290.0 ± 0 . 0 289.1 ± 2 7 . 5  SaQ (%) 95.3 ± 0 . 6 96.3 ± 0.2 97.2 ± 0.5 95.8 ± 0 . 2 93.9 ± 0 . 3 95.3 ± 0.4 94.6 ± 0 . 3 93.2 ± 1.1 95.4 ± 0 . 1 95.4 ± 1.0 2  Table B.4. Individual exercise intervention data (normoxic condition, averaged over last 50 min; mean ± SD).  100  Subject 1 2 3 4 5 6 7 8 9 10  HR (b-min" ) 158.4 ± 4 . 4 157.9 ± 3 . 7 152.8 ± 4 . 6 144.6 ± 5 . 7 155.0 ± 4 . 2 155.0 ± 10.3 165.2 ± 6 . 9 149.8 ± 7 . 1 162.6 ± 4 . 7 162.1 ± 4 . 3 1  V (L-min ) 66.9 ± 4 . 0 58.0 ± 2 . 5 77.4 ± 7 . 7 54.0 ± 1.8 71.7±3.6 68.4 ± 3 . 5 70.1 ± 6 . 7 44.7 ± 1.1 64.7 ± 3 . 9 69.9 ± 2 . 6 E  -1  vo  2  (mL'kg'-min ) 32.7 ± 0 . 9 39.1 ± 1.1 33.1 ± 2 . 4 27.4 ± 0 . 6 36.5 ± 0 . 9 38.7 ± 1.2 44.2 ± 2.3 25.5 ± 1.1 32.8 ± 1.9 45.0 ± 2 . 1 1  Power (W) 212.0 ± 0 . 0 181.2 ± 1.4 195.5 ± 3 . 3 185.6 ± 8 . 0 204.1 ± 4 . 1 214.4 ± 14.7 224.6 ± 2.9 257.9 ± 13.0 241.4±3.1 235.0 ± 0 . 0  SaO (%) 87.5 ± 1.5 82.0 ± 0 . 3 81.7 ± 1.5 88.0 ± 0 . 5 80.8 ± 0 . 6 86.1 ± 0 . 6 81.4±0.4 78.8 ± 1.5 81.1 ± 9 . 4 78.2 ± 1.6 z  Table B.5. Individual exercise intervention data (hypoxic condition, averaged over last 50 min; mean ± S D ) .  Subject 1 2 3 4 5 6 7 8 9 10  Total Work in Normoxia (kJ) 889.5 618.0 840.4 785.2 843.7 884.8 936.9 1366.8 1016.6 1078.0  Total Work in Hypoxia (kJ) 763.2 647.4 699.6 636.2 712.3 728.3 759.3 912.6 854.5 753.9  Table B.6: Individual exercise intervention data (total work performed).  101  Subject 1  •  o  Subject 2 Subject 3  T  v •  Post 55min  Subject 5  •  Subject 6  • —  Subject 7  — o  Subject 8  •  Subject 9  A  Pre  Subject 4 —  Subject 10  Post l O l m i n  Figure B . l . Individual lung densities following normoxic exercise (density threshold = 0.3 g-mL" ). 1  102  0.30 0.25  A  •  o •  v • • •  0 • A  Pre  Post 56min  — —  Subject 1 Subject 2 Subject 3 Subject 4 Subject 5 Subject 6 Subject 7 Subject 8 Subject 9 Subject 10  Post 106min  Figure B.2. Individual lung densities following hypoxic exercise (density threshold = 0.3 g-mL- ). 1  103  APPENDIX C. Individual data: lung densities, gradients, & areas  As described in section 4.2, pixels above a specific density were removed from the lung density images to avoid including vascular tissue in the density calculations. Six different thresholds were used for this process. A threshold of 0.30 g-mL" was used for all 1  analyses in the manuscript, but lung densities were calculated for all threshold values used (0.25, 0.30, 0.35, 0.40, 0.45, and 1.0 g-mL" ). Tables C . l - C.6 contain these 1  individual lung density data.  Subject 1 2 3 4 5 6 7 8 9 10  Normoxic Pre 0.141 0.170 —  0.158 0.170 0.181 —  0.165 0.181 0.146  Normoxic Postl 0.146 0.166 . 0.154 0.157 0.166 0.173 0.188 0,164 0.177 .0.148  Normoxic Post2 0.136 0.155 0.142 0.157 0.159 0.176 0.187 0.167 0.160 0.147  Hypoxic Pre 0.132 0.178 0.156 0.147 0.170 0.179 0.184 0.167 0.168 0.150  Hypoxic Postl 0.136 0.158 0.157 0.136 0.171 0.176 0.183 0.165 0.166 0.135  Hypoxic Post2 0.138 0.154  0:159  0.154 0.169 0.176 0.178 0.172 0.174 0.141  Table C . l . Individual lung densities (g-mL" ) with a density threshold = 0.25 g-mL" . 1  1  104  Subject 1 2 3 4 5 6 7 8 9 10  Normoxic Pre 0.148 0.180 —  0.173 0.187 0.194 —  0.179 0.201 0.153  Normoxic Postl 0.156 0.179 0.167 0.172 0.185 0.205 0.209 0.186 0.198 0.159  Normoxic Post2 0.144 0.165 0.156 0.172 0.179 0.189 0.213 0.179 0.177 0.158  Hypoxic Pre 0.138 0.190 0.171 0.161 0.191 0.199 0.207 0.184 0.181 0.160  Hypoxic Postl 0.145 0.174 0.170 0.148 0.196 0.197 0.204 0.179 0.179 0.147  Hypoxic Post2 0.143 0.164 0.174 0.170 0.190 0.197 0.200 0.192 0.179 0.153  Table C.2. Individual lung densities (g-mL" ) with a density threshold = 0.30 g-mL" . 1  Subject 1 2 3 4 5 6 7 8 9 10  Normoxic Pre 0.152 0.185 —  0.185 0.198 0.223 —  0.186 0.214 0.158  Normoxic Postl 0.162 0.188 0.173 0.183 0.197 0.199 0.224 0.200 0.213 0.165  Normoxic Post2 0.148 0.171 0.164 0.183 0.196 0.208 0.229 0.187 0.189 0.164  1  Hypoxic Pre 0.142 0.197 0.180 0.171 0.205 0.212 0.222 0.194 0.188 0.166  Hypoxic Postl 0.150 0.185 0.178 0.157 0.217 0.209 0.220 0.187 0.186 0.154  Hypoxic Post2 0.146 0.169 0.183 0.183 0.205 0.213 0.216 0.206 0.205 0.161  Table C.3. Individual lung densities (g-mL" ) with a density threshold = 0.35 g-mL" . 1  1  105  Subject 1 2 3 4 5 6 7 8 9 10  Normoxic Pre 0.155 0.188  —  0.194 0.203 0.233  —  0.190 0.222 0.160  Normoxic Postl 0.165 0.193 0.178 0.191 0.206 0.206 0.233 0.210 0.225 0.170  Normoxic Post2 0.151 0.174 0.169 0.192 0.208 0.216 0.241 0.191 0.197 0.168  Hypoxic Pre 0.145 0.202 0.185 0.178 0.214 0.220 0.231 0.200 0.192 0.170  Hypoxic Postl 0.153 0.190 0.183 0.165 0.233 0.217  0:230 0.193 0.191 0.160  Hypoxic Post2 0.148 0.172 0.189 0.193 0.215 0.224 0.227 0.216 0.214 0.166  Table C.4. Individual lung densities (g-mL" ) with a density threshold = 0.40 g-mL" . 1  Subject 1 2 3 4 5 6 7 8 9 10  Normoxic Pre 0.156 0.191  —  0.199 0.207 0.214  -0.191 0.227 0.162  Normoxic Postl 0.167 0.196 0.181 0.196 0.213 0.240 0.240 0.215 0.233 0.173  Normoxic Post2 0.152 0.176 0.171 0.197 0.217 0.210 0.248 0.194 0.202 0.170  1  Hypoxic Pre 0.147 0.205 0.188 0.183 0.220 0.225 0.236 0.203 0.195 0.173  Hypoxic Postl 0.156 0.193 0.187 0.171 0.245 0.222 0.236 0.195 0.194 0.164  Hypoxic Post2 0.149 0.174 0.193 0.200 0.222 0.231 0.233 0.222 0.220 0.169  Table C.5. Individual lung densities (g-mL" ) with a density threshold = 0.45 g-mL" . 1  1  106  Normoxic Pre 0.160 0.199  Subject 1 2 3 4 5 6 7 8 9 10  —  0.205 0.213 0.267  —  0.193 0.238 0.167  Normoxic Postl 0.171 0.201 0.186 0.204 0.224 0.233 0.255 0.223 0.251 0.180  Normoxic Post2 0.154 0.180 0.178 0.204 0.233 0.241 0.260 0.196 0.212 0.173  Hypoxic Pre 0.155 0.216 0.194 0.190 0.229 0.237 0.244 0.206 0.199 0.179  Hypoxic Postl 0.162 0.201 0.195 0.187 0.272 0.239 0.247 0.201 0.201 0.171  Hypoxic Post2 0.152 0.177 0.204 0.216 0.236 0.246 0.245 0.231 0.232 0.175  Table C.6. Individual lung densities (g-mL" ) with a density threshold =1.0 g-mL" . 1  1  The gradient of lung density for every image was calculated as the slope of the best fit line between the average density at each of 3 distances from anterior to posterior lung. The mean data for gradient at a threshold density of 0.30 g-mL" are discussed in section 1  4.3. The individual data for each scan at each threshold density are presented in tables C.7-C.12.  Normoxic Pre 5.1 2.4  Subject 1 2 3 4 5 6 7 8 9 10  —  5.3 2.7 4.9 —  4.6 —  6.3  Normoxic Postl 4.1 5.9 4.8 6.2 2.5 6.5 3.5 4.2 2.5 5.7  Normoxic Post2 3.3 6.4 7.1 5.7 6.3 4.9 5.0 4.6 5.0 6.2  Hypoxic Pre 4.9 5.9 7.1 5.1 4.7 5.7 3.4 5.2 5.2 5.3  Hypoxic Postl 4.3 5.9 4.6 6.8 4.1 5.9 4.8 2.7 —  6.6  Hypoxic Post2 4.8 5.8 5.4 —  4.6 6.7 5.9 3.8 4.8 5.7  Table C.7. Individual lung density gradients (mg-mL"'-cm" ) with a density threshold = 0.25 g-mL" . 1  1  107  Normoxic Pre 5.8 3.0  Subject 1 2 3 4 5 6 7 8 9 10  —  7.3 3.0 6.6 —  5.5 6.7 7.3  Normoxic Postl 4.9 7.1 5.4 8.1 2.7 6.8 4.9 5.6 2.6 7.7  Normoxic Post2 3.8 7.2 8.4 7.0 8.4 8.1 6.8 5.8 6.1 7.9  Hypoxic Pre 5.4 6.9 8.8 6.5 6.1 7.6 4.0 6.7 6.4 6.6  Hypoxic Postl 5.2 7.4 5.6 8.7 5.6 7.8 6.7 3.6 5.6 8.4  Hypoxic Post2 5.0 6.5 6.5 8.9 6.0 9.4 7.7 5.3 5.6 7.4  Table C.8. Individual lung density gradients (mg-mL" -cm ) with a density threshold = 0.30 g-mL" . 1  -1  1  Normoxic Pre 6.3 3.4  Subject 1 2 3 4 5 6 7 8 9 10  ~  8.9 3.2 8.5 ~  6.0 7.8 7.7  Normoxic Postl 5.5 7.9 5.7 9.6 2.5 9.1 6.1 6.9 2.8 8.8  Normoxic Post2 4.2 • 7.7 9.1 8.7 10.1 7.8 8.8 6.6 7.1 8.9  Hypoxic Pre 5.8 7.4 9.9 7.5 7.7 9.2 4.3 7.9 6.9 7.5  Hypoxic Postl 5.7 8.5 6.4 10.2 7.3 9.3 8.4 4.5 6.3 9.7  Hypoxic Post2 5.1 6.9 7.2 10.9 7.3 11.8 9.0 6.6 7.3 8.1  Table C.9. Individual lung density gradients (mg-mL" -cm" ) with a density threshold = 0.35 g-mL" . 1  1  1  108  Normoxic Pre 6.5 3.5  Subject 1 2 3 4 5 6 7 8 9 10  ~  10.0 3.5 9.6 ~  6.3 8.5 8.0  Normoxic Postl 5.7 8.4 5.9 10.6 2.5 9.6 7.0 7.6 2.6 9.5  Normoxic Post2 4.4 7.9 9.6 10.0 11.5 8.6 10.2 7.1 7.7 9.5  Hypoxic Pre 6.1 7.7 10.4 8.4 9.0 10.2 4.9 8.6 7.1 8.3  Hypoxic Postl 6.0 9.0 6.7 11.4 8.9 10.5 9.3 5.1 6.8 10.5  Hypoxic Post2 5.1 7.2 7.6 12.4 8.4 13.6 10.3 7.8 8.0 8.7  Table C I O . Individual lung density gradients (mg-mU'-cm" ) with a density threshold = 0.40 g-mL" . 1  1  Normoxic Pre 6.6 3.5  Subject 1 2 3 4 5 6 7 8 9 10  —  10.7 3.8 8.9 ~  6.4 8.8 8.2  Normoxic Postl 5.8 8.7 6.0 11.4 2.4 10.6 7.5 7.9 2.6 9.9  Normoxic Post2 4.4 8.1 9.8 10.6 12.7 9.9 11.6 7.4 8.1 9.7  Hypoxic Pre 6.3 7.9 10.8 9.0 9.8 10.8 5.4 8.9 7.2 8.6  Hypoxic Postl 6.2 9.3 7.0 12.2 10.3 11.0 10.2 5.4 7.0 11.1  Hypoxic Post2 5.2 7.3 7.9 13.8 9.0 14.7 11.0 8.6 8.5 8.9  Table C . l 1. Individual lung density gradients (mg-mL^-cm" ) with a density threshold = 0.45 g-mL" . 1  1  109  Normoxic Pre 6.7 3.6  Subject 1 2 3 4 5 6 7 8 9 10  —  11.2 4.4 13.4 —  6.5 9.3 8.4  Normoxic Postl 5.8 8.9 6.3 12.5 0.6 11.8 8.6 8.3 2.7 10.4  Normoxic Post2 4.3 8.2 10.1 11.2 14.6 10.4 13.2 7.6 9.0 9.9  Hypoxic Pre 6.8 8.3 11.3 9.6 11.6 12.0 6.0 9.1 7.4 9.1  Hypoxic Postl 6.5 9.8 7.6 14.2 13.4 12.2 10.8 5.8 7.3 11.7  Hypoxic Post2 5.2 . 7.4 8.5 16.1 10.4 18.8 12.2 9.5 9.2 9.0  Table C.12. Individual lung density gradients (mg-mU'-cm" ) with a density threshold = •1.0 g-mL" . 1  1  The area of each lung scan slice was measured during the calculation of lung slice density. This area varies depending on subject size, observer slice selection, and the amount of tissue removed above the threshold density. The individual data for each scan at each threshold density are presented in tables C.l3 - C. 18.  110  Subject 1 2 3 4 5 6 7 8 9 10  Normoxic Pre 21,936 15,703 ~  14,569 12,198 9,113  —  19,255 12,064 20,711  Normoxic Postl 20,923 14,844 13,289 15,341 12,202 11,223 7,688 15,239 10,859 18,492  Normoxic Post2 21,020 15,895 14,445 14,711 10,681 11,467 6,238 17,147 13,030 19,313  Hypoxic Pre 21,555 14,402 13,600 15,483 10,738 10,966 8,075 15,833 13,722 19,850  Hypoxic Postl 21,355 13,619 13,928 16,273 8,634 10,820 7,855 16,983 13,756 19,695  Hypoxic Post2 21,745 16,064 13,708 14,008 10,795 9,806 7,653 14,284 10,709 19,550  Table C.13. Individual lung slice areas ( mm )with a density threshold = 0.25 g-mL" . 2  Subject 1 2 3 4 5 6 7 8 9 10  Normoxic Pre 23,169 17,375 —  16,791 14,616 13,833 —  22,102 15,344 22,033  Normoxic Postl 22,706 16,919 14,800 17,583 14,723 12,313 10,200 18,931 13,916 20,181  Normoxic Post2 22,225 17,359 16,125 16,761 12,967 13,406 8,886 19,481 15,356 21,081  1  Hypoxic Pre 22,469 16,567 15,706 17,448 13,497 13,913 10,863 18,811 15,773 21,655  Hypoxic Postl 22,781 15,919 15,698 17,848 11,417 13,750 10,256 19,511 15,614 21,438  Hypoxic Post2 22,603 17,531 15,773 16,181 13,464 12,636 9,992 17,711 15,614 21,628  Table C.14. Individual lung slice areas (mm )with a density threshold = 0.30 g-mL" . 2  1  Ill  Subject 1 2 3 4 5 6 7 8 9 10  Normoxic Pre 23,777 18,056 —  18,273 15,838 14,513 —  23,205 17,231 22,586  Normoxic Postl 23,519 18,039 15,475 19,006 16,222 14,522 11,805 21,227 15,858 21,045  Normoxic Post2 22,842 18,034 16,955 18,108 14,647 15,959 10,486 20,586 16,747 21,911  Hypoxic Pre 22,941 17,470 16,727 18,548 15,119 15,588 12,508 20,338 16,602 22,484  Hypoxic Postl 23,477 17,103 16,547 18,827 13,645 15,305 11,850 20,716 16,473 22,434  Hypoxic Post2 22,975 18,094 16,853 17,606 15,222 14,481 11,461 19,823 14,659 22,672  Table C.15. Individual lung slice areas (mm )with a density threshold = 0.35 g-mL" . 2  Subject 1 2 3 4 5 6 7 8 9 10  Normoxic Pre 24,066 18,364 —  19,153 16,378 15,544 ~  23,670 18,116 22,881  Normoxic Postl 23,913 18,573 15,838 19,802 17,067 15,114 12,573 22,458 17,064 21,516  Normoxic Post2 23,077 18,347 17,356 19,011 15,716 16,705 11,417 21,108 17,445 22,325  1  Hypoxic Pre 23,225 17,938 17,133 19,244 16,022 16,398 13,319 21,031 16,964 22,931  Hypoxic Postl 23,822 17,602 16,984 19,525 15,266 16,070 12,661 21,355 16,906 23,005  Hypoxic Post2 23,163 18,341 17,423 18,602 16,234 15,477 ' 12,297 21,086 15,444 23,252  Table C.16. Individual lung slice areas (mm )with a density threshold = 0.40 g-mL" . 2  1  112  Subject 1 2 3 4 5 6 7 8 9 10  Normoxic Pre 24,202 18,567  —  19,627 16,648 15,997  —  23,845 18,542 23,027  Normoxic Postl 24,103 18,805 16,019 20,291 17,602 16,173 13,009 23,020 17,823 21,806  Normoxic Post2 23,189 18,497 17,536 19,458 16,398 15,420 11,903 21,359 17,864 22,480  Hypoxic Pre 23,400 18,178 17,336 19,613 16,470 16,789 13,677 21,281 17,152 23,189  Hypoxic Postl 24,053 17,878 17,227 20,009 16,320 16,456 13,048 21,620 17,119 23,381  Hypoxic Post2 23,283 18,480 17,697 19,217 16,741 16,047 12,709 21,794 15,881 23,498  Table C.l7. Individual lung slice areas (mm )with a density threshold = 0.45 g-mL' . 2  Subject 1 2 3 4 5 6 7 8 9 10  Normoxic Pre 24,430 19,009 —  19,984 16,975 17,445 ~  23,942 19,191 23,313  Normoxic Postl 24,317 19,086 16,277 20,788 .18,277 16,350 13,669 23,630 18,933 22,188  Normoxic Post2 23,320 18,680 17,813 19,842 17,258 18,094 12,381 21,503 18,375 22,661  1  Hypoxic Pre 23,841 18,719 17,642 19,984 16,978 17,416 14,042 21,483 17,375 23,558  Hypoxic Postl 24,402 18,273 17,634 20,255 17,897 17,263 13,514 21,964 17,481 23,823  Hypoxic Post2 23,447 18,663 18,227 20,145 17,500 16,852 13,208 22,448 16,492 23,823  Table C.l8. Individual lung slice areas (mm )with a density threshold =1.0 g-mL" . 2  1  113  114  0.00  0.14  0.16  0.18  0.20  Observer 2 Lung Density (g.mL  Figure C.2. Inter-observer reliability of lung density r = 0.98, p< 0.05, N = 20.  115  0.20  0.00  1  0.0  0.1  1  0.2  1  0.3  1  0.4  1  1  0.5  0.6  1  0.7  0.8  Threshold density (g.mU ) 1  Figure C.3. Lung density vs. lung density threshold.  116  

Cite

Citation Scheme:

    

Usage Statistics

Country Views Downloads
United States 28 4
China 13 9
Japan 3 0
France 1 0
Sweden 1 0
Ireland 1 0
City Views Downloads
Washington 23 0
Beijing 6 0
Dongguan 5 0
Tokyo 3 0
Ashburn 3 0
Unknown 2 24
Hangzhou 1 0
Redmond 1 0
Shenzhen 1 9
Phoenix 1 0
Stockholm 1 0

{[{ mDataHeader[type] }]} {[{ month[type] }]} {[{ tData[type] }]}
Download Stats

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.831.1-0077013/manifest

Comment

Related Items