LUNG FUNCTION. [7 ed.] 9781118597354, 1118597354

Citation preview

Cotes’ Lung Function

Cotes’ Lung Function Seventh Edition

Edited by Robert L. Maynard, CBE, FRCP, FRCPath Honorary Professor of Environmental Medicine University of Birmingham Birmingham UK

Sarah J. Pearce, FRCP Formerly Consultant Physician County Durham and Darlington NHS Foundation Trust Darlington UK

Benoit Nemery, MD, PhD Emeritus Professor of Toxicology & Occupational Medicine KU Leuven Leuven Belgium

Peter D. Wagner, MD Emeritus Distinguished Professor of Medicine & Bioengineering University of California San Diego La Jolla, CA USA

Brendan G. Cooper, PhD, FERS, FRSB Honorary Professor and Consultant Clinical Scientist in Respiratory and Sleep Physiology University Hospital Birmingham Birmingham UK

This edition first published 2020 © 2020 by John Wiley & Sons Ltd Edition History [6e,2006] All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, except as permitted by law. Advice on how to obtain permission to reuse material from this title is available at http://www.wiley.com/go/permissions. The rights of Robert L. Maynard, Sarah J. Pearce, Benoit Nemery, Peter D. Wagner, and Brendan G. Cooper to be identified as the authors of editorial work has been asserted in accordance with law. Registered Office(s) John Wiley & Sons, Inc., 111 River Street, Hoboken, NJ 07030, USA John Wiley & Sons Ltd, The Atrium, Southern Gate, Chichester, West Sussex, PO19 8SQ, UK Editorial Office 9600 Garsington Road, Oxford, OX4 2DQ, UK For details of our global editorial offices, customer services, and more information about Wiley products visit us at www.wiley.com. Wiley also publishes its books in a variety of electronic formats and by print‐on‐demand. Some content that appears in standard print versions of this book may not be available in other formats. Limit of Liability/Disclaimer of Warranty The contents of this work are intended to further general scientific research, understanding, and discussion only and are not intended and should not be relied upon as recommending or promoting scientific method, diagnosis, or treatment by physicians for any particular patient. In view of ongoing research, equipment modifications, changes in governmental regulations, and the constant flow of information relating to the use of medicines, equipment, and devices, the reader is urged to review and evaluate the information provided in the package insert or instructions for each medicine, equipment, or device for, among other things, any changes in the instructions or indication of usage and for added warnings and precautions. While the publisher and authors have used their best efforts in preparing this work, they make no representations or warranties with respect to the accuracy or completeness of the contents of this work and specifically disclaim all warranties, including without limitation any implied warranties of merchantability or fitness for a particular purpose. No warranty may be created or extended by sales representatives, written sales materials or promotional statements for this work. The fact that an organization, website, or product is referred to in this work as a citation and/or potential source of further information does not mean that the publisher and authors endorse the information or services the organization, website, or product may provide or recommendations it may make. This work is sold with the understanding that the publisher is not engaged in rendering professional services. The advice and strategies contained herein may not be suitable for your situation. You should consult with a specialist where appropriate. Further, readers should be aware that websites listed in this work may have changed or disappeared between when this work was written and when it is read. Neither the publisher nor authors shall be liable for any loss of profit or any other commercial damages, including but not limited to special, incidental, consequential, or other damages. Library of Congress Cataloging‐in‐Publication data Names: Maynard, Robert L., editor. | Pearce, Sarah (Sarah J.), editor. |   Nemery, Benoit, editor. | Wagner, P. D. (Peter D.), editor. | Cooper,   Brendan (Brendan G.), editor. | Cotes, J. E. Lung function. Title: Cotes’ lung function / edited by Robert L. Maynard, Sarah J. Pearce,   Benoit Nemery, Peter D. Wagner, Brendan G. Cooper. Other titles: Lung function Description: Seventh edition. | Hoboken, NJ : Wiley-Blackwell, 2020. |   Preceded by Lung function : physiology, measurement and application in   medicine / J.E. Cotes, D.J. Chinn, M.R. Miller. 2006. | Includes   bibliographical references and index. Identifiers: LCCN 2020000667 (print) | LCCN 2020000668 (ebook) | ISBN   9781118597354 (cloth) | ISBN 9781118597330 (adobe pdf ) | ISBN   9781118597323 (epub) Subjects: MESH: Lung–physiology | Lung Diseases–physiopathology |   Respiratory Function Tests | Respiratory Physiological Phenomena Classification: LCC RC756 (print) | LCC RC756 (ebook) | NLM WF 600 | DDC  616.2/4–dc23 LC record available at https://lccn.loc.gov/2020000667 LC ebook record available at https://lccn.loc.gov/2020000668 Cover Design: Wiley Set in 10/12pt Warnock by SPi Global, Pondicherry, India 10 9 8 7 6 5 4 3 2 1

v

Contents Preface  xxvii Contributors  xxix Part I  1

Introduction  1

How We Came to Have Lungs and How Our Understanding of Lung Function has Developed  3

1.1 The Gaseous Environment  3 1.2 Functional Evolution of the Lung  4 1.3 ­Early Studies of Lung Function  5 1.4 ­The Past 350 Years  5 1.4.1 Lung Volumes  6 1.4.2 Lung Mechanics  6 1.4.3 Ventilatory Capacity  6 1.4.4 Blood Chemistry and Gas Exchange in the Lung  6 1.4.5 Control of Respiration  7 1.4.6 Energy Expenditure during Exercise  8 1.5 ­Practical Assessment of Lung Function  8 1.6 ­The Position Today  11 1.7 ­Future Prospects  11 References  11 Further Reading  15 Part II  2

Foundations  21

Getting Started  23 Michael D.L. Morgan

2.1 ­Brief Description of the Lungs and their Function  23 2.2 ­Deviations from Average Normal Lung Function  24 2.3 ­Uses of Lung Function Tests  24 2.4 ­Assessment of Lung Function  25 2.5 ­Setting up a Laboratory  25 2.6 ­Conduct of Assessments  29 Reference  30 Further Reading  30 3

Development and Functional Anatomy of the Respiratory System  33 Sungmi Jung and Richard Fraser

3.1 ­Introduction  33 3.2 ­Functional Anatomy of the Upper Airways  33 3.3 ­The Lungs  34

vi

Contents

3.3.1 Early Stages in Development  34 3.3.2 Functional Anatomy  35 3.3.3 Bronchopulmonary Anatomy  35 3.3.4 Intrapulmonary Airways  36 3.3.5 Acinus  37 3.3.6 Collateral Channels  38 3.3.7 Alveoli  38 3.3.8 Pulmonary Circulation  39 3.3.9 Bronchial Circulation  39 3.3.10 Pulmonary Lymphatics  39 3.3.11 Lymphoreticular Cells  39 3.3.12 Innervation and Pulmonary Receptors  40 3.4 ­The Pleura  40 References  41 Further Reading  43 Body Size and Anthropometric Measurements  45 4.1 ­Bodily Components are Matched for Size  45 4.2 ­Growth and Ageing  45 4.3 ­Stature (Body Length)  47 4.3.1 Overview  47 4.3.2 Measurement of Stature and Sitting Height  48 4.4 ­Body Width  48 4.5 ­Body Depth and Girth  49 4.6 ­Body Mass and Body Mass Index  50 4.7 ­Body Composition  51 4.7.1 Fat% and Fat‐Free Mass  51 4.7.2 Measurement of Fat% and Fat‐Free Mass  52 4.8 ­Distributions of Fat and Muscle: A Forward Look  54 4.9 ­Conclusion  55 References  55 Further Reading  56

4

5

Numerical Interpretation of Physiological Variables  57 J. Martin Bland

5.1 ­Introduction  57 5.2 ­Simple Arithmetic  57 5.2.1 Manipulating Numbers  57 5.2.2 Averaging Ratios  58 5.2.3 Decimal Age  58 5.2.4 Logarithms  58 5.3 ­Normal and Skewed Distributions  58 5.4 ­Measurement Error  62 5.5 ­Relationship of One Variable to Another  63 5.5.1 Proportional Relationships  64 5.5.2 Linear Relationships  64 5.5.3 Simple Curves Through the Origin  65 5.5.4 Exponential Curves  65 5.6 ­Interpreting a Possible Change in an Index  65 5.6.1 Sample Size Required to Detect a Meaningful Difference  66 5.6.2 Regression to the Mean  66 5.6.3 Choice of Model for Paired Observations  69 5.7 ­Relationship of One Variable to Several Others  69

Contents

5.7.1 Multiple Regression  69 5.7.2 Co‐Linearity  70 5.7.3 Allowing for the Effects of Age  71 5.7.4 Variation about the Regression Equation  71 5.7.5 Other Types of Regression Analysis  72 5.7.6 Principal Component Analysis  72 ­ References  73 6

Basic Terminology and Gas Laws  75 Adrian Kendrick

6.1 ­Glossary of Terms  75 6.2 ­Units  75 6.3 ­Primary Symbols and Suffixes  75 6.4 ­Abbreviations  79 6.5 ­Terminology for Lung Imaging  80 6.6 ­The Gas Laws  80 6.6.1 Boyle’s Law and Charles’ Law (BTPS and STPD Adjustment)  81 6.6.2 Ideal Gas Law  87 6.6.3 Partial Pressure – Dalton’s Law  88 6.6.4 Henry’s Law – Solubility of Gases in Liquids  89 6.6.5 Laws of Diffusion – Graham’s Law and Fick’s First Law  89 6.6.6 Conclusion  89 ­ References  89 7

Basic Equipment and Measurement Techniques  91 Brendan G. Cooper

7.1 ­Introduction  91 7.2 ­Computers  91 7.3 ­Measurement of Gas Volumes and Flows  92 7.3.1 Volume‐Measuring Devices  92 7.3.2 Flow‐Measuring Devices  94 7.4 ­Measurement of Respiratory Pressure  96 7.5 ­Other Electronic Apparatus  96 7.6 ­Connecting the Subject to the Equipment  98 7.7 ­Analysis of Gases  98 7.8 Measurement of Oxygen Consumption and Respiratory Exchange Ratio  99 7.8.1 Oxygen Consumption  99 7.8.2 Respiratory Exchange Ratio  100 7.9 ­Collection and Storage of Blood  100 7.10 ­Analysis of Blood for Oxygen  101 7.10.1 Content of Oxygen and Saturation of Haemoglobin  101 7.10.2 Tension of Oxygen in Blood  102 7.11 ­Analysis of Blood for Carbon Dioxide  103 7.11.1 Direct Methods  103 7.11.2 Indirect Methods  105 7.12 ­Use of Isotopes (Including Radioisotopes) to Study Lung Function  106 7.13 ­Sterilisation and Disinfection of Equipment  109 7.14 ­Care of Gas Cylinders  109 7.15 ­Calibration of Equipment  110 7.15.1 Anthropometry Equipment  111 7.15.2 Linearity of Gas Analysers  111 7.16 ­Quality Control  113

vii

viii

Contents

7.17 ­Manufacturers  113 References  113 Further Reading  116 8

Respiratory Surveys  117 Peter G.J. Burney

8.1 ­The Uses of Epidemiology  117 8.2 ­Study Designs and Sampling  117 8.2.1 Populations and Samples  117 8.2.2 Prevalence Studies  118 8.2.3 Cohort Studies  118 8.2.4 Case–Control Studies  119 8.2.5 Selection Bias  120 8.2.6 The Use and Abuse of Matching  121 8.2.7 Other Stratagems for the Efficient Design of Studies  122 8.3 ­Data Collection  122 8.3.1 The Characteristics of Good Data and the Nature of Error  122 8.3.2 Information Bias  123 8.3.3 Use of Questionnaires  123 8.3.4 Lung Function Measurements  124 8.3.5 Quality Assurance and Quality Control  125 8.4 ­Analysis and Related Issues  125 8.4.1 Analysis Needs to be Appropriate to the Design  125 8.4.2 Confounding  126 8.4.3 Effect Modification  126 8.4.4 Analysis of Lung Function  126 8.5 ­Ethics Considerations  127 ­ References  127 9

The Application of Analytical Technique Applied to Expired Air as a Means of Monitoring Airway and Lung Function  129 Paolo Paredi and Peter Barnes

9.1 ­Exhaled Nitric Oxide  129 9.1.1 Source of Nitric Oxide in Exhaled Air  129 9.1.2 Anatomic Origin of Nitric Oxide  130 9.1.3 Nitric Oxide Measurement  130 9.1.4 Single‐Breath Nitric Oxide Measurement  131 9.1.5 Multiple‐Breath Nitric Oxide Measurement  133 9.1.6 Limitations of the Multiple‐Breath Nitric Oxide Measurement  135 9.1.7 Area Under the Curve Method  136 9.2 ­Conclusions  136 9.2.1 The Role of New Markers of Airway Inflammation  136 9.2.2 Exhaled Breath Temperature  136 9.2.3 Bronchial Blood Flow  139 9.2.4 Clinical Studies  140 9.3 ­Volatile Organic Compounds  141 9.3.1 Ethane and Pentane  141 9.3.2 Methods  142 9.3.3 Clinical Studies  142 9.3.4 Other Volatile Organic Compounds and their Measurement  143 9.3.5 Clinical Studies  143 9.3.6 Electronic Nose  143 9.4 ­Exhaled Carbon Monoxide  144 9.4.1 Measurement  144

Contents

ix

9.4.2 Clinical Studies  144 9.5 ­Conclusions  145 ­ References  145 Part III  10

Physiology and Measurement of Lung Function  149

Chest Wall and Respiratory Muscles  151 André De Troyer and John Moxham

10.1 ­Introduction  151 10.2 ­The Chest Wall  151 10.3 ­The Diaphragm  152 10.4 ­The Intercostal Muscles  156 10.5 ­Interaction Between the Diaphragm and the Inspiratory Intercostals  162 10.6 ­The Neck Muscles  163 10.7 ­The Abdominal Muscles  165 10.8 ­Clinical Assessment of the Respiratory Muscles  166 ­ References  172 11 Lung Volumes  177 11.1 ­Definitions  177 11.1.1 Total Lung Capacity and its Subdivisions  177 11.1.2 Vital Capacity and Variants Thereof  177 11.1.3 Other Volumes  177 11.2 ­Features of Lung Volumes  178 11.2.1 Some Determinants  178 11.3 ­Measurement of Total Lung Capacity and its Subdivisions  180 11.3.1 Closed Circuit Gas Dilution Method  180 11.3.2 Alternative Closed Circuit Methods  182 11.3.3 Open Circuit Gas Dilution Method  182 11.3.4 Radiographic Method  183 11.3.5 Plethysmographic Methods  184 References  184 Further Reading  185 12

Lung and Chest Wall Elasticity  187 G. John Gibson

12.1 ­Introduction and Definitions  187 12.2 ­Lung Elasticity  188 12.2.1 Factors Contributing to Lung Recoil  188 12.2.2 Implications of Lung Elasticity for the Distribution of Ventilation  190 12.2.3 Implications of Lung Elasticity for Airway and Alveolar Patency  190 12.2.4 Inspiratory and Expiratory Pressure–Volume Curves  191 12.2.5 Dynamic Lung Compliance  191 12.2.6 Measurement of Lung Elasticity  192 12.2.7 Physiological Variation in Lung Elasticity  194 12.3 ­Pathological Variation in Lung Elasticity  195 12.4 ­Compliance of the Chest Wall and Respiratory System  196 12.4.1 Clinical Measurements of Respiratory System Elasticity  197 12.4.2 Methods of Measurement in Ventilated Patients [56]  198 12.5 ­Distensibility of Conducting Airways  199 12.5.1 Practical Aspects  199 12.6 ­Concluding Remarks  199 ­References  200

0004504801.INDD 9

2/10/2020 1:06:39 PM

x

Contents

13

Forced Ventilatory Volumes and Flows  203 Riccardo Pellegrino

13.1 ­Introduction  203 13.2 ­Maximal Breathing  203 13.2.1 Definitions 203 13.2.2 Background 204 13.2.3 Measurement 205 13.3 ­Peak Expiratory Flow  205 13.3.1 Background 205 13.3.2 Measurement 205 13.4 ­Indices from Single Breath Volume–Time Curves  206 13.4.1 Indices Based on Volume  206 13.4.2 Indices Expressed as Times  207 13.5 ­Indices from the Relationship of Flow to Volume  208 13.5.1 Expiratory Flow–Volume Curve  209 13.5.2 Inspiratory Flow–Volume Curve  210 13.6 ­Measurement of Single Breath Indices of Ventilatory Capacity  210 13.6.1 General Considerations  210 13.6.2 Measurement of FEV1 and Other Indices from Volume–Time Curves  210 13.6.3 Practical Aspects of Flow–Volume Spirometry  212 13.7 ­Density Dependence  213 13.7.1 Volume of Iso‐Flow  213 13.7.2 Measurement of V‐isov   213 References  214 Further Reading  216 14

Theory and Measurement of Respiratory Resistance  217 Jason H.T. Bates

14.1 ­Introduction  217 14.2 ­Theoretical Basis for Respiratory Resistance  217 14.3 ­Airway Resistance  218 14.3.1 Body Plethysmography  218 14.3.2 Alveolar Capsule  219 14.3.3 Flow Dependence of Airway Resistance  220 14.4 ­Respiratory Resistance and its Components  220 14.4.1 Total Respiratory System Resistance  220 14.4.2 Lung Resistance  221 14.4.3 Tissue Resistance  221 14.5 ­Frequency Dependence of Resistance and Elastance  222 14.5.1 Tissue Viscoelasticity  222 14.5.2 Mechanical Heterogeneities  223 14.6 ­Respiratory Impedance  223 14.6.1 Forced Oscillation Technique  224 14.6.2 Physiological Interpretation of Impedance  224 14.7 ­Summary  227 ­ References  227 15

The Control of Airway Function and the Assessment of Airway Calibre  231 Eric Derom

15.1 ­Introduction  231 15.2 ­Genetics and Airway Calibre  232 15.3 ­Physiological Control of Airway Calibre  232 15.3.1 Parasympathetic Nervous System  232 15.3.2 Sympathetic Nervous System  233

Contents

15.3.3 The NANC System  234 15.3.4 Other Control Mechanisms of Airway Calibre  234 15.4 ­Airflow Limitation in Diseases  235 15.4.1 Asthma  235 15.4.2 Chronic Obstructive Pulmonary Disease  237 15.4.3 Cystic Fibrosis  238 15.4.4 Obesity  238 15.4.5 Allergic Rhinitis  238 15.5 ­Assessment of Airflow Limitation  238 15.5.1 Large Airway Obstruction  238 15.5.2 Small Airway Obstruction  239 15.6 ­Bronchodilator Testing as a Diagnostic Tool  240 15.6.1 Measurements Used in Bronchodilator Testing  241 15.6.2 Bronchodilatation After Inhalation of β2‐Agonists and/or Anticholinergics  241 15.6.3 Expression of the Results  242 15.6.4 Interpretation  243 15.6.5 Bronchodilating Effects of Other Drugs  244 15.7 ­Bronchial Hyper‐Responsiveness as a Diagnostic Tool  245 15.7.1 Methacholine and Histamine Challenge Testing  245 15.7.2 Exercise Challenge Testing  248 15.7.3 Eucapnic Voluntary Hyperpnoea  249 15.7.4 Challenges with Hyperosmolar Aerosols (Hypertonic Saline, Mannitol)  250 15.7.5 Adenosine Challenge Testing  250 15.7.6 Specific Inhalation Challenges (Allergens, Aspirin)  250 Acknowledgements  251 ­ References  251 16

Ventilation, Blood Flow, and Their Inter‐Relationships  259 G. Kim Prisk

16.1 ­Introduction and Basic Concepts  259 16.1.1 Gas Exchange – The Basic Principle  259 16.1.2 Ventilation–Perfusion Ratio  259 16.1.3 Systematic Variation in Ventilation: the Slinky Spring  260 16.1.4 Systematic Variation in Blood Flow: the Zone Model of Perfusion  261 16.2 ­Distribution of Ventilation  261 16.2.1 Anatomical Dead‐Space  261 16.2.2 Gravitationally Induced Heterogeneity and the Effects of Posture  262 16.2.3 Non‐Gravitational Heterogeneity and the Uneven Distribution of Resistance and Compliance  263 16.2.4 Convective Diffusive Interactions  264 16.2.5 Uneven Contraction of the Respiratory Muscles  264 16.2.6 Cardiogenic Motion  265 16.2.7 Airway Obstruction  265 16.3 ­Distribution of Pulmonary Blood Flow  265 16.3.1 The Pulmonary Circulation  265 16.3.2 Gravitational Blood Flow Heterogeneity  265 16.3.3 Non‐Gravitational Blood Flow Heterogeneity  266 16.3.4 Pulmonary Vasomotor Tone  266 16.3.5 Hypoxic Pulmonary Vasoconstriction  267 16.3.6 Local Pharmacological Mechanisms  267 16.4 ­Matching of Ventilation and Perfusion  267 16.4.1 The Three‐Compartment Model  267 16.4.2 Beyond the Three‐Compartment Model  272 16.4.3 Compensations for V ̇A/Q̇ Inequality  272 16.5 ­Assessing the Evenness of Ventilation  274

xi

xii

Contents

16.5.1 Performing the Single‐Breath Wash‐Out  274 16.5.2 Slope of the Alveolar Plateau (Slope of Phase 3)  276 16.5.3 Cardiogenic Oscillations  277 16.5.4 Closing Volume and Closing Capacity  277 16.5.5 Variants on Single‐Breath Methods  278 16.5.6 The Multiple‐Breath Wash‐Out  279 16.5.7 Moment Analysis and ‘Slow’ and ‘Fast’ Space  279 16.5.8 Distribution of Specific Ventilation  280 16.5.9 Lung Clearance Index  281 16.5.10 Scond and Sacin  281 16.6 ­Measuring Pulmonary Blood Flow and its Heterogeneity  282 16.6.1 Total Pulmonary Blood Flow – Cardiac Output  282 16.6.2 Direct Fick  282 16.6.3 Indirect Fick Methods Using CO2  282 16.6.4 Soluble Gas Rebreathing  284 16.6.5 Soluble Gas Open Circuit  284 16.6.6 Indicator Dilution Methods  284 16.6.7 Other Methods of Measuring Cardiac Output  285 16.6.8 Single‐Breath Perfusion Heterogeneity  286 16.7 ­Measuring V̇A/Q̇ Inequality  286 16.7.1 Compartmental Analysis – Dead‐Space and Shunt  287 16.7.2 Intra‐Breath‐R  288 16.7.3 MIGET  289 16.8 ­Imaging V̇A, Q̇, and V̇A/Q̇  291 16.8.1 Ventilation  291 16.8.2 Perfusion  293 16.8.3 V̇A/Q̇  293 ­ References  295 17

Transfer of Gases into the Blood of Alveolar Capillaries  301 Eric Derom and Guy F. Joos

17.1 ­Introduction  301 17.2 ­Diffusion in the Gas Phase  301 17.2.1 Directional Velocity  301 17.2.2 Diffusion Coefficient  303 17.2.3 Behaviour of Gas Mixtures  303 17.2.4 Applications  303 17.3 ­Transfer of Gas Across the Alveolar Capillary Membrane  303 17.3.1 Role of Gas Solubility in Blood  304 17.3.2 Transfer Involving Chemical Reaction with Blood  305 17.3.3 General Gas Equation  306 17.3.4 Concept of Resistance to Transfer of Gas  307 17.3.5 Terminology: Transfer Factor or Diffusing Capacity?  308 17.4 ­Application of the General Gas Equation (Eq. 17.8) to Individual Gases  308 17.4.1 Gases that do not Combine with Haemoglobin  308 17.4.2 Carbon Monoxide  309 17.4.3 Oxygen  309 17.4.4 Nitric Oxide  309 17.5 ­Practical Consequences  310 References  310 Further Reading  311 18

Transfer Factor (Tl) for carbon monoxide (CO) and nitric oxide (NO)  313 Colin D.R. Borland and Mike Hughes

18.1 ­Introduction 

313

Contents

18.1.1 Overview  313 18.1.2 Terminology and Units: Transfer Factor or Diffusing Capacity?  316 18.2 ­Introduction to Diffusion  317 18.2.1 The General Equation for Diffusing Capacity (Dl)/Transfer Factor (Tl)  317 18.3 ­Diffusion in the Gas Phase  318 18.3.1 Molecular Weight Dependence  318 18.3.2 Stratified Inhomogeneity: Does Gas Phase Diffusion Resistance (1/Dg) Affect Tl,CO and Tl,NO?  318 18.4 ­Partitioning Transfer Factor Tl,CO into Membrane (Dm) and Red Cell (θVc) Components  319 18.4.1 The Roughton–Forster Equation  319 18.4.2 Calculating Dm and Capillary Volume (Vc) from the Roughton–Forster Equation  320 18.4.3 Membrane Diffusing Capacity (Dm,CO)  321 18.4.4 Red Cell Resistance (1/θVc)  322 18.4.5 How θNO can be Finite In Vitro, but Infinite In Vivo  322 18.4.6 Dm and Vc: Morphometric–Physiological Comparison  323 18.5 ­Diffusion Limitation for Oxygen  323 18.5.1 Diffusion–Perfusion Interaction: the Tl/βQ Concept  323 18.5.2 Low Diffusion–Perfusion Ratios Cause Hypoxaemia  325 18.6 ­The Transfer Factor (Tl) for Different Gases: Theory  325 18.6.1 Oxygen  325 18.6.2 Carbon Monoxide  326 18.6.3 Nitric Oxide  326 18.6.4 Effects of Heterogeneity  327 18.7 ­Methods for Measuring Tl,CO  328 18.7.1 Principles  328 18.7.2 Single‐Breath with Breathholding (Tl,COsb)  329 18.7.3 Other Methods for Measuring Tl,CO  332 18.8 ­ T l,CO: Extrinsic Variables  333 18.8.1 Alveolar Volume (VA) and Expansion Change  333 18.8.2 Exercise  334 18.8.3 Haemoglobin Concentration and Haematocrit  334 18.8.4 Carbon Monoxide Back Tension and Carboxyhaemoglobin  335 18.8.5 PA,O2 Variation  335 18.8.6 Reference Values for Tl,CO and KCO  335 18.9 ­ T l,COsb: Interpretation  336 18.9.1 Introduction  336 18.9.2 KCO and VA in Lungs with Normal Alveolar Structure  336 18.9.3 KCO and VA in Lungs with Abnormal Alveolar Structure  338 18.9.4 Dm,CO and Vc in Physiology and Pathology  338 18.10 ­Nitric Oxide Transfer Factor: Tl,NO  340 18.10.1 Methodology 340 18.10.2 Normal Values and Physiological Variation  342 18.10.3 Tl,NO/Tl,CO Ratio  343 18.10.4 Tl,NO in Disease  343 18.10.5 Should Tl,NO Become a Routine Lung Function Test?  344 18.11 ­Summary  344 ­ Acknowledgements  345 ­ References  345 18.A Appendices  350 18.A.1 The General Equation for Diffusion  350 18.A.2 Derivation of the Diffusive–Perfusive Conductance Ratio  350 18.A.3 The Single‐Breath Tl,CO Calculation Derived  351 18.A.4 Steady‐State (ss) Tl,CO: Method and Calculation  351 18.A.5 Tl,CO Corrections for Low or High PA,O2  352

xiii

xiv

Contents

19 Oxygen  353 Dan S. Karbing and Stephen E. Rees

19.1 ­Overview  353 19.2 ­Diffusion in the Gas Phase  355 19.2.1 Directional Velocity  355 19.2.2 Fick’s First Law of Diffusion  355 19.3 ­Capacity of Blood for Oxygen  356 19.3.1 Role of Gas Solubility in Blood on Rate of Gas Transfer  356 19.3.2 Reaction of Oxygen with Haemoglobin  356 19.3.3 Oxygen Dissociation Curve  357 19.4 ­Transfer Factor of the Lung  359 19.4.1 General Gas Equation  359 19.4.2 Concept of Resistance to Transfer of Gas  359 19.4.3 Terminology: Transfer Factor or Diffusing Capacity?  360 19.5 ­Oxygen Uptake into Blood  360 19.5.1 Some Features of the Transfer Gradient  360 19.5.2 Oxygen Uptake During Normoxia: a Worked Example  361 Oxygen Uptake During Hypoxia: a Worked Example  362 19.5.3 19.6 ­Measurement of Transfer Factor for Oxygen (Tl,O2)  362 19.6.1 Overview of Methods  362 19.6.2 Derivation of Tl,O2 from Tl,CO  362 19.6.3 V̇A/Q̇ Method for Tl,O2 (Summary)  363 19.6.4 Method of Riley and Lilienthal [40-42]  363 19.6.5 Effects on Tl,O2 of Uneven Lung Function  364 19.7 ­Respiratory Determinants of Arterial Oxygen Tension and Saturation: Some Worked Examples  364 19.7.1 Alveolar Ventilation  365 19.7.2 Two‐Compartment Model of Normal Gas Exchange  365 19.8 ­Investigation of Hypoxaemia, Including Use of Models  367 19.8.1 Introduction  367 19.8.2 Graphical Analysis of Gas Exchange: Oxygen–Carbon Dioxide Diagram  369 19.8.3 Two‐Parameter Models Relating Changes in Inspiratory O2 to Sp,O2  369 Acknowledgement  373 ­References  373 20

Carbon Dioxide  377 Erik R. Swenson

20.1 ­Introduction  377 20.2 ­Gas Exchange for CO2  378 20.2.1 Overview  378 20.2.2 Whole Blood Dissociation Curve for Carbon Dioxide  378 20.2.3 Uptake of Carbon Dioxide by Blood  379 20.2.4 Carbamino‐Haemoglobin  380 20.2.5 Release of CO2 in the Lungs  380 20.2.6 Rate of Tissue and Alveolar Equilibration for CO2  380 20.3 ­Acid–Base Balance  381 20.3.1 Overview  381 20.3.2 Indices: Base Excess, Strong Ion Difference, Anion Gap  382 20.3.3 Respiratory Alkalosis and Acidosis  383 20.3.4 Changes in Cerebrospinal Fluid  384 20.3.5 Renal Mechanisms  385 20.3.6 Metabolic Acidosis and Alkalosis  385 20.3.7 Acid–Base Disturbances of Multiple Aetiologies  386 References  386 Further Reading  388

Contents

21

Control of Respiration  389 Bertien M.‐A. Buyse

21.1 ­Introduction  389 21.2 ­Control of Respiration (Figure 21.1)  389 21.2.1 Brain Stem Neural Respiratory Activity [4]  389 21.2.2 Automatic Breathing  391 21.2.3 Spinal Mechanisms  393 21.2.4 Behavioural Control – Volitional Breathing  394 21.3 ­Clinical Assessment of Respiratory Control  394 21.3.1 Standardisation of the Conditions of Measurement is of Crucial Importance  394 21.3.2 Measurement of Respiratory Output (Figure 21.1)  395 21.3.3 Methods of Evaluating Control of Respiration in Clinical Practice  397 References  403 22

The Sensation of Breathing  407 Mathias Schroijen, Paul W. Davenport, Omer Van den Bergh, and Ilse Van Diest

22.1 ­Introduction  407 22.2 ­Afferent Input of Respiratory Sensory Information  408 22.2.1 Respiratory Sensation  408 22.2.2 Dyspnoea  410 22.3 ­Assessment of Respiratory Sensation and Dyspnoea  411 22.3.1 Neural processing  411 22.3.2 Psychophysical Methods  412 22.3.3 Self‐Reported Dyspnoea  413 22.3.4 Behavioural Measures  413 22.4 ­Factors Modulating the Experience of Respiratory Sensations and Dyspnoea  414 22.4.1 Age  414 22.4.2 Gender  414 22.4.3 Attention  414 22.4.4 Fear and Anxiety  415 22.4.5 Symptom Schemata, Illness Representation, and Illness Behaviour  416 22.4.6 Social Context  417 22.5 ­Conclusion  418 ­ References  419 23

Breathing Function in Newborn Babies  423 Urs P. Frey and Philipp Latzin

23.1 ­Introduction  423 23.2 ­Developmental Respiratory Physiology in Early Life  423 23.3 ­Assessment of Lung Function in Neonates and Infants (Aged 0–2 Years)  424 23.3.1 Standardisation and Measurement Conditions Related to Infant Lung Function Testing  424 23.3.2 Tidal Breathing Parameters  425 23.3.3 Regulation of Breathing, Novel Mathematical Methods  425 23.3.4 Measurement of Forced Expiratory Flow (Rapid Thoraco‐Abdominal Compression Techniques: RTC)  425 23.3.5 Measurement of Lung Volumes  426 23.3.6 Measurement of Ventilation Inhomogeneity  427 23.3.7 Passive Respiratory Mechanics and the Interrupter Technique  427 23.3.8 Forced Oscillation and Interrupter Technique  427 23.3.9 Measurement of Transfer Factor (Diffusing Capacity)  428 23.3.10 Measurement of Exhaled Nitric Oxide  428 23.4 ­Reference Values of Infant Lung Function  428 23.5 ­Potential Role of Lung Function Testing in Infant Respiratory Disease  429 23.6 ­Lung Function in Children Aged 2–6 Years  429

xv

xvi

Contents

23.6.1 Standardisation of Preschool Lung Function Testing  429 23.6.2 Tidal Breathing Parameters  429 23.6.3 Measurement of Forced Expiratory Flows  430 23.6.4 Multiple‐Breath Wash‐Out  430 23.6.5 Plethysmography  430 23.6.6 Forced Oscillation and Interrupter Technique  430 23.6.7 Measurement of Exhaled Nitric Oxide  432 23.7 ­Reference Values in Preschool Age  432 23.8 ­Potential Role of Lung Function Testing in Preschool Respiratory Disease  433 ­ References  433 Part IV  24

Normal Variation in Lung Function  435

Normal Lung Function from Childhood to Old Age  437 Andrew Bush and Michael D.L. Morgan

24.1 ­Introduction  437 24.2 ­Influences on Lung Structure  439 24.2.1 Epigenetic Transgenerational Effects  439 24.2.2 Key Stage 1: Antenatal Lung Development  439 24.2.3 Key Stage 2: Lung Growth in Childhood and Adolescence  439 24.2.4 Key Stage 3: the Phase of Lung Function Decline  443 24.2.5 Summary: How do Developmental Structural Changes Manifest?  444 24.3 ­Physiological Changes in Childhood and Adolescence  444 24.3.1 Early Changes  444 24.3.2 Factors that Influence Development  444 24.4 ­Puberty and Transition to Adult Lung Function  446 24.4.1 Contribution of Gender  446 24.4.2 Girls and Young Women  447 24.4.3 Boys and Young Men  447 24.5 ­Lung Function in Early Adulthood  447 24.6 ­Variation in Lung Function Between Adults  447 24.6.1 Roles of Body Composition  447 24.6.2 Roles of Habitual Activity and Physical Training  448 24.7 ­Cyclical Variation in Lung Function  449 24.8 ­Differences in Function Between Men and Women  449 24.8.1 The Effect of Obesity on Lung Function  450 24.9 ­Menstrual Cycle and Pregnancy  451 24.9.1 Menstrual Cycle  451 24.9.2 Pregnancy  451 24.10 ­Effects of Age on the Lungs  452 24.10.1 Underlying Considerations  452 24.10.2 Lung Function  453 24.11 ­Effects of Age on Responses to Exercise  454 24.11.1 Interaction with Ageing of the Lungs  454 24.11.2 Oxygen Consumption, Ventilation, and Breathlessness  454 24.11.3 Contribution of Cardiovascular System  456 References  456 Further Reading  461 Growth of the lungs  461 Effects of age  461 25

Reference Values for Lung Function in White Children and Adults  463

25.1 ­Basic Considerations  25.1.1 Definitions  463

463

Contents

25.1.2 Reference Subjects  464 25.1.3 Quality Control  464 25.1.4 Models (Mathematical Equations) that Form the Basis for Reference Values  465 25.1.5 Strategies for Selecting Reference Values  465 25.2 ­Preschool Children (Ages 3–6 Years)  466 25.3 ­Children of School Age  466 25.3.1 Models Based on Stature  466 25.3.2 Empirical Models  467 25.3.3 Reference Values Independent of Body Size  470 25.3.4 Longitudinal Growth Charts (Percentiles)  471 25.4 ­Young Persons Aged 16–25 Years  472 25.5 ­Adults Aged 25–65 Years  473 25.6 ­Adults Aged 65 Years Onwards  479 25.7 ­Comprehensive Cross‐Sectional and Longitudinal Models  479 25.7.1 Cross‐Sectional Models  481 25.7.2 Progression of Lung Function Throughout Life  481 25.8 ­Interpreting Reference Values  486 25.8.1 Making Sense of the Results  494 References  494 Further Reading  498 Reference Values for Lung Function in Non‐White Adults and Children  499 26.1 ­Overview  499 26.1.1 Relevance of Race  499 26.1.2 Ethnic Factor in Lung Function  499 26.1.3 Variables Sometimes Linked to Ethnic Group  500 26.1.4 Biases Introduced by Migration  502 26.1.5 Miscegenation: Evidence for Autosomal Inheritance of Lung Function  502 26.1.6 Ethnic Factor in Ventilatory Responses to Exercise and Breathing CO2  502 26.1.7 Inheritance of Lung Function Within Ethnic Groups  503 26.1.8 Reference Variables  503 26.1.9 Technical Factors in Interpretation  503 26.2 ­Lung Function Analysed by Ethnic Group and Geographical Location  504 26.2.1 South Asia, Including the Indian Subcontinent  504 26.2.2 Mexicans and Hispanic Americans  507 26.2.3 Middle East and North Africa  507 26.2.4 East Asian People  507 26.2.5 Sub‐Saharan Africans  508 26.2.6 Oceanians, Including Australian Aboriginals  509 26.3 ­Perspective  512 References  513 Further Reading  515 26

Part V  27

Exercise  517

Physiology of Exercise and Effects of Lung Disease on Performance  519

27.1 ­Some Basic Concepts  519 27.2 ­Oxygen Cost of Exercise  520 27.3 ­Determinants of Exercise Capacity: an Overview  522 27.4 ­Respiratory Response to Exercise  523 27.4.1 Introduction  523 27.4.2 Ventilation During Exercise of Constant Intensity  523 27.4.3 Ventilation During Progressive Exercise  524 27.4.4 Respiratory Exchange Ratio  526

xvii

xviii

Contents

27.4.5 Respiratory Frequency and Tidal Volume  527 27.4.6 Anaerobic Threshold: Useful Index or Dangerous Fallacy?  528 27.4.7 Exercise Ventilation in Medical Conditions  529 27.4.8 Maximal Exercise Ventilation  532 27.5 ­Cardiac Output and Stroke Volume  533 27.5.1 Cardiac Output  533 27.5.2 Stroke Volume  533 27.6 ­Exercise Cardiac Frequency  534 27.6.1 Numerical Values  534 27.6.2 Indices of fcsubmax  536 27.6.3 Clinical Applications  536 27.7 ­Breathlessness on Exertion  537 27.7.1 Sensation of Dyspnoea  537 27.7.2 Mechanisms of Breathlessness  537 27.7.3 Clinical Aspects  537 27.7.4 Speculations  538 27.8 ­Limitation of Exercise  538 27.8.1 Ventilatory Limitation (Including Use of Oxygen)  538 27.8.2 Pulmonary Gas Exchange  540 27.8.3 Cardiac Output and Muscle Blood Flow  540 27.8.4 Tissue Transfer of Oxygen  541 27.9 ­Events in Muscles  542 27.9.1 Muscle Metabolism  542 27.9.2 Lactate and Pyruvate  543 27.9.3 McArdle’s Disease  544 27.10 ­Role of Gender  545 27.11 ­Effects of Age  545 27.12 ­Habitual Activity and Physical Training  545 27.12.1 Overview of Effects  545 27.12.2 Implications for Clinical Exercise Testing  545 References  547 Further Reading  551 Exercise Testing and Interpretation, Including Reference Values  553 28.1 ­Introduction  553 28.2 ­Reasons for an Exercise Test  553 28.2.1 In Apparently Fit Persons  553 28.2.2 In Respiratory and Other Patients  554 28.3 ­Exercise Protocols  554 28.3.1 Submaximal Progressive Protocol  554 28.3.2 Submaximal Steady‐State Protocol  554 28.3.3 Symptom‐Limited Exercise Test (S‐LET)  554 28.3.4 Nearly Maximal Exercise Test (Ergocardiography, Bronchial Lability)  555 28.3.5 Maximal Exercise Test (Aerobic Capacity)  556 28.4 ­Ergometry  557 28.4.1 Choice of Ergometer  557 28.4.2 Treadmill  557 28.4.3 Cycle Ergometry  559 28.4.4 Stepping Exercise  561 28.5 ­Measurements  561 28.5.1 What Should be Measured?  561 28.5.2 Overview of Equipment  561 28.5.3 Ventilation Minute Volume  562 28.5.4 Gas Analysis  562 28

Contents

28.5.5 Other Measurements  562 28.5.6 Respiratory Symptoms  563 28.6 ­Conduct of the Test  563 28.7 ­Data Processing  564 28.8 ­Interpretation of Data  565 28.8.1 Submaximal Exercise  565 28.8.2 Exercise Limitation  566 28.9 ­Non‐Ergometric and Field Tests  569 28.9.1 Observational Tests  569 28.9.2 Walking Tests  569 28.9.3 Shuttle Tests  569 28.9.4 Harvard Pack Test  570 28.10 ­Reference Values for Ergometry in Adults  570 References  573 Further Reading  575 29

Assessment of Exercise Limitation, Disability, and Residual Ability  577

29.1 ­Terminology  577 29.1.1 Respiratory Impairment  577 29.1.2 Respiratory Limitation of Exercise  577 29.1.3 Respiratory Handicap (Participation Restricted)  578 29.2 ­Causes of Respiratory Disablement  578 29.3 ­Preliminaries to the Assessment  579 29.3.1 Medical Considerations  579 29.3.2 Review of the Lung Function  580 29.3.3 When is an Exercise Test Needed?  580 29.4 ­Conduct of the Exercise Test  580 29.4.1 Practical Considerations  580 29.4.2 Avoiding Non‐Cooperation  581 29.4.3 Special Role of the Person Conducting the Test  581 29.5 ­Interpreting the Exercise Test  581 29.5.1 Type of Limitation  581 29.5.2 Scoring Loss of Exercise Capacity (Disability)  583 29.5.3 Underperformance 584 29.6 ­Residual Ability  584 29.7 ­Relevance for Compensation  584 29.8 ­Summary  585 References  585 Further Reading  586 30

Exercise in Children  587 Andrew Bush

30.1 ­Introduction  587 30.2 ­Indications for Exercise Testing in Children  588 30.3 ­Methods  589 30.4 ­Normal Response to Exercise in Children  589 30.5 ­Special Indications for Exercise Testing in  Children  592 30.6 ­When There is a Discrepancy between Symptoms and Baseline Lung Function  592 30.7 ­Assessment of Prognosis in Cases of Respiratory Disease  592 30.8 ­Assessment of EILO  592 30.9 ­Understanding the Physiology of Disease  592 30.10 ­Summary and Conclusions  593 ­ References  593

xix

xx

Contents

Part VI  31

Breathing During Sleep  595

Breathing During Sleep and its Investigation  597 Joerg Steier

31.1 ­Introduction  597 31.2 ­Terminology and Definitions  597 31.3 ­Investigative Techniques  598 31.3.1 Sleep Staging  598 31.3.2 Nasal and Oral Airflow  598 31.3.3 Abdominal/Thoracic Movement  599 31.3.4 Arterial Oxygen Saturation  599 31.3.5 Pa,O2 and Pa,CO2  599 31.3.6 Snoring  599 31.3.7 Heart Rate  599 31.3.8 Movement and Posture  600 31.4 ­Sleep Studies  600 31.4.1 Polysomnography  600 31.4.2 Limited Respiratory Sleep Studies  600 31.4.3 Screening for Sleep‐Disordered Breathing  601 31.5 ­Sleepiness  601 31.6 ­Respiratory Physiology in Sleep  602 31.6.1 Sleep Levels  602 31.6.2 CO2 and O2 Responses  603 31.6.3 Upper Airway  603 31.6.4 Thorax  604 31.7 ­Respiratory Pathophysiology in Sleep  604 31.7.1 Upper Airway Control  604 31.7.2 Central Control  605 31.8 ­Clinical Syndromes of Sleep‐Disordered Breathing  606 31.8.1 Obstructive Sleep Apnoea (OSA)  606 31.8.2 Central Sleep Apnoea (CSA)  606 31.8.3 Mixed Sleep Apnoea  607 31.8.4 Upper Airway Resistance Syndrome (UARS)  607 31.8.5 Obesity Hypoventilation Syndrome  607 31.9 ­Treatment of Sleep‐Disordered Breathing  607 31.9.1 Obstructive Sleep Apnoea  607 31.9.2 Central Sleep Apnoea  609 31.10 ­Respiratory Conditions Affected by Sleep  609 31.10.1 Asthma 609 31.10.2 Chronic Obstructive Pulmonary Disease  609 31.10.3 Neuromuscular and Skeletal Disorders  609 References  610 Further Reading  614 Part VII 

Potentially Adverse Environments  615

32 Hypobaria  617 James Milledge

32.1 ­Introduction  617 32.2 ­The Atmosphere and Physiological Effects of Hypobaria  618 32.2.1 Atmospheric Pressure  618 32.2.2 Atmospheric Temperature  618 32.2.3 Atmospheric Ozone  618

Contents

32.2.4 Cosmic Radiation  618 32.2.5 O2 and CO2 Partial Pressures at Altitude  619 32.2.6 Exercise at Altitude  622 32.3 ­Effects of Altitude on Lung Function in Lowlanders  623 32.3.1 Peak Expiratory Flow  623 32.3.2 Bronchoconstriction, Hypoxia, and Hypocapnia  624 32.3.3 Subclinical Oedema  624 32.3.4 Lung Diffusing Capacity  624 32.4 ­Effects of Lifelong Residence at Altitude on Lung Function  625 32.4.1 Lung Volumes  625 32.4.2 Chronic Mountain Sickness and High‐Altitude Pulmonary Hypertension  625 32.4.3 Genetics of High‐Altitude Residents Compared with Lowlanders  626 32.4.4 Genetics of Chronic Mountain Sickness  626 32.5 ­Coping with Altitude  626 32.6 ­High‐Altitude Illness  627 32.6.1 Acute Mountain Sickness  627 32.6.2 High‐Altitude Cerebral Oedema  627 32.6.3 High‐Altitude Pulmonary Oedema  627 32.7 ­Physiology and Medicine of Flight  628 32.7.1 The Aircraft Cabin  628 32.7.2 Mechanical Effects of Pressure Change  629 32.7.3 Assessment of Aircrew  629 32.8 ­Fitness to Fly as a Passenger  629 32.8.1 On‐Board Oxygen  630 32.8.2 Deep Vein Thrombosis and Venous Thromboembolism  630 32.9 ­Altitude‐Induced Decompression Illness  631 References  632 Further Reading  636 33

Immersion in Water, Hyperbaria, and Hyperoxia Including Oxygen Therapy  639 Einar Thorsen

33.1 Introduction  639 33.2 ­Surviving at the Air–Water Interface (Including Drowning)  640 33.3 ­Effects of Diving on Lung Function  641 33.3.1 Immersion and Dives with Breathholding  641 33.3.2 Deep Dives  642 33.3.3 Pathological and Adaptive Changes in the Lungs  644 33.4 ­Barotrauma in Divers and Submariners  644 33.5 ­Decompression Sickness  645 33.5.1 Features 645 33.5.2 Prophylaxis (Including Saturation Dives)  646 33.5.3 Diving Strategies  646 33.6 ­Screening for Fitness to Dive  646 33.7 ­Hyperoxia  646 33.7.1 Summary of O2 Therapy  646 33.7.2 Pathological Effects of a Raised Oxygen Tension  648 33.7.3 Hyperbaric O2 Therapy  649 References  650 Further Reading  652 34

Effects of Cold and Heat on the Lung  653 Malcolm Sue‐Chu

34.1 34.1.1

Acute Effects of Cooling on the Lungs  653 Lower Airway Cooling  653

xxi

xxii

Contents

34.1.2 Upper Airway Cooling  654 34.1.3 Facial Cooling  654 34.1.4 Whole Body Cooling  654 34.2 ­Chronic Effects of Cooling on the Lungs  655 34.2.1 Residents of Cold Climates  655 34.2.2 Cold Weather Athletes  656 34.2.3 Patients  656 34.3 ­Impact of Hot Air Breathing  656 34.3.1 Breathing Pattern in Hot Environments  656 34.3.2 Airway Calibre in Hot Environments  657 ­References  658

35

Part VIII 

Lung Function in Clinical Practice  661

Strategies for Assessment of Lung Function  663 James Hull

35.1 ­Introduction  663 35.2 ­Techniques Available – Standard Tests  664 35.2.1 Pulse Oximetry and Arterial Blood Gas Measurement  664 35.2.2 Peak Flow, Spirometry, and Flow–Volume Loops  664 35.2.3 Static Lung Volumes  665 35.2.4 Airway Resistance  665 35.2.5 Gas Transfer  666 35.3 ­Techniques Available – Specialist Tests  666 35.3.1 Exercise Testing  666 35.3.2 Airway Responsiveness  666 35.3.3 Airway Inflammation – Exhaled Nitric Oxide  667 35.3.4 Respiratory Muscle Strength  667 35.3.5 Compliance  667 35.3.6 Sleep Studies  667 35.4 ­Imaging Modalities and their Role in Assessment  668 35.5 ­Strategies for Disease Assessment  668 35.6 ­Interpretive Strategy/Algorithm for Patients With Dyspnoea  670 ­References  670 36

Patterns of Abnormal Lung Function in Lung Disease  673 William Kinnear

36.1 ­Classical Patterns  673 36.1.1 Lung Function Tests in the Era of High‐Resolution CT Scanning  673 36.1.2 Test Quality and Normal Ranges  673 36.2 ­Spirometry  673 36.2.1 Restriction  673 36.2.2 Obstruction  674 36.3 ­Lung Volumes  674 36.3.1 Restriction  674 36.3.2 Obstruction  675 36.4 ­Gas Transfer  676 36.4.1 Tl,CO  676 36.4.2 KCO  676 36.4.3 Restrictive Defect, Low Tl,CO, Low KCO  676 ­ References  678

Contents

37

Lung Function in Asthma, Chronic Obstructive Pulmonary Disease, and Lung Fibrosis  681 Wim Janssens and Pascal Van Bleyenbergh

37.1 ­Introduction  681 37.2 ­Asthma  681 37.2.1 Spirometry and Peak Expiratory Flow  681 37.2.2 Reversibility and Airways Resistance  682 37.2.3 Bronchial Hyper‐Responsiveness  683 37.2.4 Lung Volumes and Diffusing Capacity  683 37.2.5 Fractional Exhaled Nitric Oxide (FENO)  684 37.2.6 Illustrative Case  684 37.3 ­Chronic Obstructive Pulmonary Disease and Emphysema  685 37.3.1 Spirometry  685 37.3.2 Reversibility and Airway Resistance  688 37.3.3 Lung Volumes  688 37.3.4 Gas Transfer and Diffusing Capacity  689 37.3.5 Illustrative Case  689 37.4 ­Diffuse Lung Fibrosis (Diffuse Interstitial Lung Disease)  690 37.4.1 Spirometry and Resistance  690 37.4.2 Lung Volumes  691 37.4.3 Gas Transfer  691 37.4.4 Compliance  692 37.4.5 Illustrative Case  692 ­ References  693 38

Lung Function in Specific Respiratory and Systemic Diseases  697 Stephen J. Bourke

38.1 ­Introduction  697 38.2 ­Extrapulmonary Airways  697 38.2.1 Laryngeal Disorders  697 38.2.2 Tracheal Disease  698 38.3 ­Intrapulmonary Airways  698 38.3.1 Bronchiectasis  698 38.3.2 Cystic Fibrosis  699 38.3.3 Bronchiolitis  700 38.4 ­The Alveoli  700 38.4.1 Pneumonia  700 38.4.2 Pneumocystis Pneumonia and HIV Infection  701 38.4.3 Hypersensitivity Pneumonitis  701 38.4.4 Sarcoidosis  701 38.4.5 Drug‐Induced Alveolitis  702 38.4.6 Radiation Pneumonitis  702 38.4.7 Acute Respiratory Distress Syndrome  702 38.5 ­Pulmonary Vascular Disease  702 38.5.1 Pulmonary Embolism  703 38.5.2 Fat Embolism  703 38.5.3 Pulmonary Hypertension  703 38.5.4 Pulmonary Veno‐Occlusive Disease  704 38.5.5 Arteriovenous Malformations  704 38.6 ­The Pleura  704 38.7 ­The Chest Wall  704 38.7.1 Flail Segment  704 38.7.2 Pectus Excavatum and Pectus Carinatum  704

xxiii

xxiv

Contents

38.7.3 Kyphoscoliosis  705 38.7.4 Sternotomy/Thoracotomy  705 38.7.5 Thoracoplasty  705 38.7.6 Ankylosing Spondylitis  705 38.7.7 Obesity  705 38.8 ­Neuromuscular Disease  705 38.8.1 Brain Stem Failure  706 38.8.2 Disease of the Motor Neurone  706 38.8.3 Phrenic Neuropathy, Guillain–Barré Syndrome  706 38.8.4 Myasthenia, Muscular Dystrophies, Myositis  707 38.8.5 Parkinson Disease  707 38.8.6 Multiple Sclerosis  707 38.9 ­Rare Pulmonary Diseases  707 38.9.1 Unilateral Hyperlucent Lung  707 38.9.2 α1‐Antitrypsin Deficiency  707 38.9.3 Pulmonary Alveolar Microlithiasis  708 38.9.4 Pulmonary Alveolar Proteinosis  708 38.9.5 Pulmonary Langerhans Cell Histiocytosis  708 38.9.6 Lymphangioleiomyomatosis and Tuberous Sclerosis  708 38.9.7 Behçet Disease  709 38.9.8 Idiopathic Pulmonary Haemosiderosis  709 38.9.9 Pulmonary Amyloidosis  709 38.10 ­Haematological Diseases  709 38.10.1 Anaemia and Polycythaemia  709 38.10.2 Methaemoglobinaemia 710 38.10.3 Carboxyhaemoglobinaemia 710 38.10.4 Sulfhaemoglobinaemia 710 38.10.5 Sickle Cell Disease  710 38.10.6 Bone Marrow Transplantation  711 38.11 ­Cardiac Disease  711 38.11.1 Left Heart Failure  711 38.11.2 Congenital Heart Disease  712 38.12 ­Connective Tissue Diseases  712 38.12.1 Rheumatoid Disease  712 38.12.2 Systemic Lupus Erythematosus  712 38.12.3 Sjögren Syndrome  713 38.12.4 Systemic Sclerosis  713 38.12.5 Marfan Syndrome  713 38.12.6 Ehlers–Danlos Syndrome and Cutis Laxa  713 38.13 ­Liver Disease  713 38.13.1 Ascites 713 38.13.2 Hepatopulmonary Syndrome  714 38.13.3 Portopulmonary Hypertension  714 38.14 ­Renal Disease  714 38.14.1 Pulmonary Haemorrhagic Syndromes  714 38.14.2 Dialysis and Chronic Renal Failure  714 38.15 ­Gastrointestinal Disease  715 38.15.1 Coeliac Disease and Inflammatory Bowel Disease  715 38.16 ­Endocrine Disease  715 38.16.1 Diabetes 715 38.16.2 Thyroid Disease  715 38.16.3 Pituitary Disease  716 ­References  716

Contents

39

Pulmonary Rehabilitation  729 Sally Singh

39.1 ­Introduction  729 39.2 ­Limitation of Exercise in Lung Disease  729 39.2.1 Ventilatory Limitation  730 39.2.2 Energy Cost of Breathing  730 39.2.3 Impaired Gas Exchange  730 39.2.4 Muscle Dysfunction  731 39.2.5 Cardiovascular Dysfunction  731 39.3 ­Assessment for Pulmonary Rehabilitation Programmes  731 39.3.1 Perspective  731 39.3.2 Assessment of Breathlessness  731 39.3.3 Quality of Life Assessment  732 39.3.4 Functional Assessment  732 39.4 ­Components of Pulmonary Rehabilitation Programmes  733 39.4.1 Exercise Component  733 39.4.2 Respiratory Muscle Training  733 ­ References  733 40

Lung Function in Relation to Surgery, Anaesthesia, and Intensive Care  737 Göran Hedenstierna

40.1 ­Introduction  737 40.2 ­Pre‐Operative Evaluation of Respiratory Function  737 40.2.1 Airway and Thoracic Anatomy  738 40.2.2 Spirometry and Other Lung Function Tests  738 40.2.3 Muscle Function  738 40.2.4 Premedication and General Anaesthesia  739 40.2.5 Special Considerations  739 40.3 ­Respiratory Function During Anaesthesia  740 40.3.1 Upper Airway  740 40.3.2 Respiratory Mechanics  740 40.3.3 Distribution of Ventilation and Lung Perfusion  743 40.3.4 Ventilatory Responses to Anaesthesia  743 40.3.5 Inflammatory Reaction of the Lung  744 40.4 ­Respiratory Function and Mechanical Ventilation in the Critically Ill  744 40.4.1 Acute Lung Injury and Acute Respiratory Distress Syndrome  744 40.4.2 Ventilator‐Induced Lung Injury  744 40.4.3 High Inspired Oxygen Concentration and Infection  745 40.4.4 Ventilatory Management in the Postoperative Period and in Critically Ill Adults  745 ­ References  747

Index  751

xxv

xxvii

Preface The first edition of Lung Function was published in 1965, more than 50 years ago. It was written by John Cotes and reflected his unique knowledge of respiratory physiology and his commitment to the use of lung function tests in clinical medicine. The book was a runaway success. All respiratory physicians and physiologists recognised it as the leading work in the field and there can have been few lung function laboratories in which it was unknown. Further editions followed, the sixth appearing in 2006. By then John was in his ninth decade and was assisted in producing the book by D J Chinn and M R Miller. In about 2012 John, now approaching his tenth decade, began to think about the seventh ­edition. He suggested that the new edition should be produced by a team of editors who should invite leading authorities in the field to contribute individual chapters. It was intended that John would be closely involved with the work. An editorial board was assembled and work began. Sadly John Cotes died in 2018 before the new edition was completed. Taking over a book that has been the leading work in its field and, especially, one that so closely reflected the approach and thinking of its original author has been a challenging task. The new edition is very different from earlier editions in that it is a multi‐author text with the advantages and disadvantages which inevitably attend such an approach. More than 30 experts have contributed to the to the new edition. Some chapters have been allowed to stand unchanged from the sixth edition: need revision in further editions but, for now, they remain both valuable and irreplaceable. Others have been revised or rewritten.

Who is this book for? This simple question is less easy to answer now than it would have been in 1965. Then, the classical approach to lung function taken by John was ‘cutting edge’; now, the applications of molecular biology hold that position, and lung function laboratories are fewer than they were. We recognise the importance of new developments in our field but hold fast to our belief that pulmonology, or respiratory medicine, is and must be based on a clear understanding of how gases are handled by the lung, how gases move across the air–blood barrier, and how these processes can be measured. Thus we think the classical approach to lung function, the approach that John Cotes did so much to encourage, remains indispensable in modern medical practice. It is our hope that this new edition will be consulted, if not read cover to cover, by all involved in respiratory research and clinical care, and, in particular, by trainees in respiratory medicine and clinical respiratory physiology. Producing this new edition has been a long process: we would like to thank our authors as well as Yogalakshmi Mohanakrishnan, Baskar Anandraj, Anglea Cohen, and James Schultz at our publisher, Wiley, and our copy‐editor, Lindsey Williams, for their enthusiasm and their patience. We hope that John would have been pleased with our efforts and we dedicate this new edition, to be entitled Cotes’ Lung Function, to his memory. Robert L. Maynard Sarah J. Pearce Benoit Nemery Peter D. Wagner Brendan G. Cooper

xxix

Contributors Peter Barnes, FRS, FMedSci

Andrew Bush, MD, FRCP, FRCPCH

Margaret Turner‐Warwick Professor of Thoracic Medicine National Heart & Lung Institute Imperial College London London UK

Professor of Paediatrics and Paediatric Respirology Imperial College London and Consultant Paediatric Chest Physician Royal Brompton & Harefield NHS Foundation Trust London UK

Jason H.T. Bates, PhD, DSc

Professor Larner College of Medicine University of Vermont Burlington, VT USA J. Martin Bland, MSc, PhD, FSS

Emeritus Professor of Health Statistics University of York York UK Colin D.R. Borland, MD, FRCP

Honorary Senior Visiting Fellow Department of Medicine University of Cambridge Formerly Consultant Physician Hinchingbrooke Hospital Huntingdon UK Stephen J. Bourke, MD, FRCP, FRCPI

Consultant Respiratory Physician Royal Victoria Infirmary Newcastle upon Tyne UK Peter G.J. Burney, MD, FFPH, FMedSci

Senior Research Fellow National Heart and Lung Institute Imperial College London UK

Bertien M.‐A. Buyse, MD, PhD

Pulmonologist University Hospital Leuven KU Leuven Leuven, Belgium Brendan G. Cooper, PhD, FERS, FRSB

Honorary Professor and Consultant Clinical Scientist in Respiratory and Sleep Physiology University Hospital Birmingham Birmingham UK Paul W. Davenport, PhD

Distinguished Professor Department of Physiological Sciences University of Florida, Gainesville, FL USA Eric Derom, MD, PhD

Professor of Internal Medicine/Pneumology Department of Respiratory Medicine Ghent University Hospital Ghent Belgium André De Troyer, MD, PhD

Formerly Professor of Respiratory Medicine Erasme University Hospital and Respiratory Physiologist Brussels School of Medicine Brussels Belgium

xxx

Contributors

Richard Fraser, MD, CM

Guy G.F. Joos, MD, PhD

Professor of Pathology Department of Pathology McGill University Health Center Montreal Canada

Professor of Internal Medicine/Pneumology Department of Respiratory Medicine Ghent University Hospital Ghent Belgium

Urs P. Frey, MD, PhD, FERS

Sungmi Jung, MD

Chair of Pediatrics University Children’s Hospital Basel University of Basel Basel Switzerland

Assistant professor and pathologist Department of Pathology McGill University Health Center Montreal Canada

G. John Gibson, BSc, MD, FRCP

Dan S. Karbing, PhD

Emeritus Professor of Respiratory Medicine Newcastle University Newcastle UK Philipp Latzin, MD, PhD

Head, Department of Pediatrics University Hospital Bern Inselspital Bern Switzerland Göran Hedenstierna, MD, PhD

Senior Professor in Clinical Physiology Department of Medical Sciences Uppsala university Uppsala Sweden Mike Hughes, DM, FRCP

Emeritus Professor of Thoracic Medicine National Heart and Lung Institute Imperial College London London UK James Hull, FRCP, FHEA, FACSM

Consultant Respiratory Physician Royal Brompton Hospital and Imperial College London UK Wim Janssens, MD, PhD

Associate professor and respiratory physician Clinical Department of Respiratory Diseases, UZ Leuven BREATHE, Department CHROMETA, KU Leuven Leuven Belgium

Associate Professor Aalborg University Aalborg Denmark Adrian Kendrick, PhD

Consultant Clinical Scientist University Hospitals and Physiological Sciences, University of West of England Bristol UK William Kinnear, FRCP

Consultant Physician (Respiratory Medicine) Queens Medical Centre Nottingham, England UK Robert L. Maynard, CBE, FRCP, FRCPath

Honorary Professor of Environmental Medicine University of Birmingham Birmingham UK James Milledge, BSc, MD, FRCP

Emeritus Professor of Respiratory Medicine Faculty of Life Sciences and Medicine King’s College London London UK Michael D.L. Morgan, MB, BChir (Cantab), MD, FRCP

Consultant Respiratory Physician University Hospitals of Leicester NHS Trust and Honorary Professor of Respiratory Medicine University of Leicester Leicester UK

Contributors

John Moxham, MD, FRCP

Joerg Steier, FRCP, MD, PhD

Late Medical Director King’s College Hospital Medical Trust London UK

Professor of Respiratory and Sleep Medicine Guy’s & St Thomas’ NHS Foundation Trust King’s College London London UK

Benoit Nemery, MD, PhD

Emeritus Professor of Toxicology & Occupational Medicine KU Leuven Leuven Belgium Paolo Paredi, MD, PhD

Clinical Research Fellow National Heart & Lung Institute Imperial College London London UK Sarah J. Pearce, FRCP

Formerly Consultant Physician County Durham and Darlington NHS Foundation Trust Darlington UK Riccardo Pellegrino, MD

Allergology and Respiratory Pathophysiology S. Croce e Carle Hospital Cuneo and 2 Respiratory Pathophysiology Dept of Internal Medicine and Medical Specialitie University of Genoa Genoa, Italy G. Kim Prisk, PhD, DSc

Professor Emeritus University of California San Diego La Jolla, CA USA Stephen E. Rees, PhD, DrTech

Professor Aalborg University Aalborg Denmark Mathias Schroijen, MA

Researcher, Health Psychology KU Leuven – University of Leuven Leuven Belgium Sally Singh, PhD

Professor of Pulmonary & Cardiac Rehabilitation Coventry University Coventry UK and University Hospitals of Leicester NHS Trust Leicester UK

Malcolm Sue‐Chu, MB, ChB, PhD

Emeritus Professor Norwegian University of Science and Technology/NTNU and Formerly Consultant Pulmonologist St. Olavs Hospital Trondheim University Hospital Trondheim Norway Erik R. Swenson, MD

Professor of Medicine and Physiology University of Washington Seattle, WA USA Einar Thorsen, MD

Professor, Department of Thoracic Medicine Haukeland University Hospital, University of Bergen Bergen Norway Pascal Van Bleyenbergh, MD

Pulmonologist Clinical Department of Respiratory Diseases UZ Leuven Leuven Belgium Omer Van den Bergh, PhD

Emeritus Professor of Health Psychology KU Leuven – University of Leuven Leuven Belgium Ilse Van Diest, PhD

Professor of Health Psychology KU Leuven – University of Leuven Leuven Belgium Peter D. Wagner, MD

Emeritus Distinguished Professor of Medicine & Bioengineering University of California San Diego La Jolla, CA USA

xxxi

1

Part I Introduction

3

1 How We Came to Have Lungs and How Our Understanding of Lung Function has Developed This chapter describes how the theory and practice of lung function testing have reached their present state of d ­ evelopment and gives pointers to the future.

CHAPTER MENU 1.1 1.2 1.3 1.4 1.5 1.6 1.7

The Gaseous Environment,  3 Functional Evolution of the Lung,  4 Early Studies of Lung Function,  5 The Past 350 Years,  5 Practical Assessment of Lung Function,  8 The Position Today,  11 Future Prospects,  11 References, 11

1.1 ­The Gaseous Environment The basis of respiratory physiology is Claude Bernard’s concept of a ‘milieu interieur’ that remains constant and stable despite changes in the environment. However, the two are not independent since life on Earth has evolved symbiotically with changes in Earth’s atmosphere and this process is continuing. At first, the composition of the atmosphere was determined by physical processes, and then by biological ones. Now changes in the composition of air are being driven by man’s own actions. It remains to be seen how and to what extent the system will adapt. Initially the atmosphere was mainly nitrogen. Then as the Earth cooled, carbon dioxide (CO2) was formed by chemical reactions beneath the Earth’s crust and released by volcanic activity. Some of the gas was taken up by combination with minerals and deposited as sediment at the bottom of the oceans. Oxygen was released, but immediately combined with iron and other elements, and so the atmospheric concentration was very low [1, 2]. Subsequently, the concentration of oxygen increased as a result of biological activity [3]. A hypothesis as to how this happened was proposed by Lovelock [4], whose concept of the living Earth (Gaia) is on a par with evolution as one of the formative influences of our time.

Free oxygen first appeared some 3.5 × 109 years ago, coincidentally with the development of organisms capable of photosynthesis. The organisms multiplied and their growth reduced significantly the atmospheric concentration of CO2. Some organisms (methanogens) developed an ability to form free methane gas. The methane was liberated into the atmosphere, where it shielded the Earth’s surface from ultraviolet light. The shielding allowed ammonia gas to accumulate and this provided a substrate for the growth of photosynthesising organisms; as a result, at the beginning of the Proterozoic era some 2.3 × 109 years ago, the atmospheric concentration of oxygen began to rise. By geological standards the increase was rapid, from 0.1% to 1% over about 1 million years (Figure 1.1). When the ambient oxygen concentration reached 0.2%, aerobic organisms became abundant in the surface layers of lakes and oceans; at 2%, life began to move onto the land. A concentration of 3% may have been attained some 1.99 × 109 years ago. At 10% photosynthesis was at its peak; this further raised the concentration of oxygen and lowered that of CO2. The changes reduced the available substrate (CO2) and increased the formation of hydrogen peroxide, superoxide ions, and atomic oxygen that were potentially lethal to cells. Photosynthesis was reduced in consequence. With other factors, the balance between promotion and

Cotes’ Lung Function, Seventh Edition. Edited by Robert L. Maynard, Sarah J. Pearce, Benoit Nemery, Peter D. Wagner, and Brendan G. Cooper. © 2020 John Wiley & Sons Ltd. Published 2020 by John Wiley & Sons Ltd.

1  How We Came to Have Lungs and How Our Understanding of Lung Function has Developed Earth formed

Land colonised

Life began

Large animals evolved

10% Atmospheric concentration

4

Carbon dioxide

1%

Oxygen

0.1% 100 ppm

Methane

Present

10 ppm 1 ppm

4.5

4.0

3.5

3.0

2.5

2.0

1.5

Time before present (eons = years × 109)

1.0

0.5 10000 2050 BCE CE Date

Figure 1.1  Approximate timescale for the evolution of the gaseous environment. Source: After [4].

inhibition of photosynthesis formed a feedback loop that stabilised the atmospheric concentration of oxygen at its present level (21.93%, FIO2 = 0.2193; Chapter 6). The concentration of oxygen stabilised at the start of the Phanerozoic era some 6 × 108 years ago. It led to the evolution of animals with skeletons. Thereafter, the concentration of CO2, and to some extent that of oxygen, appears to have oscillated in response to secondary factors. These included fluctuations in the balance between the relative dominance of plants and animals. Some 5 × 108 years ago the species that were net consumers of oxygen (e.g. bacteria, fungi, and insects) were in the ascendancy and CO2 levels were relatively high. Then, plants that fix CO2 as lignin appeared and the levels fell. The plants led to the evolution of dinosaurs and other animals that could feed off and digest the cellulose. The species flourished and the cycle was reversed. From time to time the sequence was unsettled by dust clouds from meteors and volcanic eruptions. The dust interfered with photosynthesis by obscuring the sun, but up to the present the equilibrium has always been restored. Currently, the atmosphere is under threat from human activity. Clearance of forests and the replacement of grassland by buildings and roads are reducing the Earth’s capacity for photosynthesis. Hence, the amount of CO2 removed from air is falling. Concurrently, the quantity released is increasing because of massive combustion of fossil fuels. As a result, the Earth’s temperature is rising and this is increasing the formation of methane gas that could raise the temperature further. However it also has other effects, and so the long‐term outcome is unpredictable. In the short term any change in gaseous equilibrium is likely to occur slowly. In summary, living organisms first appeared in an anaerobic environment that they helped to convert to an aerobic one. Hence they were adapting to the new conditions as they were creating them. On this account, the

capacity to tolerate conditions of hypoxia and hypercapnia are part of man’s heritage. How this is achieved is described in subsequent chapters. The evolutionary history also indicates the importance of natural protection against oxygen radicals. However, there is only limited evidence for Berken and Marshall’s suggestion [1] that the relevant mechanisms emerged during periods of what we would now regard as hyperoxia.

1.2 ­Functional Evolution of the Lung Aerobic organisms developed in an aqueous medium where the amount of oxygen is determined by its partial pressure and by the solubility; this is such that the concentration in water is only about one‐fortieth of that in air (Table  1.1). By contrast CO2 is highly soluble, so at physiological partial pressures the concentration in water is nearly as great as in air. The differences in solubility have consequences for gas exchange [5]. For fish the problem of obtaining sufficient oxygen was solved by the evolution of the gill system. This organ is ventilated by a large volume of water from which almost all the oxygen is extracted; the blood leaving the bronchial clefts contains oxygen in a concentration equal to that in blood leaving the lung in man. However, the large volume of water flowing over the gill takes with it much of the CO2 in solution and this lowers the CO2 tension in the blood leaving the gill to less than 0.7 kPa (5 mmHg). Mainly on this account the blood pH is relatively high (approximately 8.0 pH units at a temperature of 20 °C). At higher water temperatures the pH falls to approach that in the blood of man. Concurrently, the solubility of oxygen in water ­delivered to the gill clefts is reduced.

1.4  The Past 350 Years

Table 1.1  Atmospheric concentrations and solubility in water of oxygen and carbon dioxide. Units

Oxygen

Carbon dioxide

Atmospheric concentration Solubility in water at 1 atm:

vol./vol.

0.2093

0.003

Temperature 20 °Ca

vol . of gas STPD

0.031

0.88

0.024

0.55

vol . of water Temperature 37 °Ca

vol . of gas STPD vol . of water

a

 Solubility in blood plasma is approximately 10% less. atm., atmosphere; TPD, standard temperature and pressure dry.

In hot climates a high ambient temperature might cause streams to dry up, leaving any fish stranded. To meet this hazard some fish developed lung‐like pouches in the back of the pharynx; they also developed primitive limbs with which to crawl along streambeds in search of water. For this type of existence a gill for the exchange of CO2 and a primitive lung for exchanging oxygen formed a life‐saving combination. The lung was further developed in reptiles. In birds the pouches were adapted as reservoirs from which air was pumped in a cross‐current manner through parabronchi; these supplied air to the gas exchange zones where the whole of the surface was lined with capillaries. This arrangement resulted in a very compact lung with a high capacity to transfer gas. The amphibians developed in a different way by shedding their scales to leave a soft vascular skin; this replaced the gill as a means of exchanging CO2 with the surrounding water. Somewhere between these diverging species emerged the primitive mammals and eventually man.

1.3 ­Early Studies of Lung Function Erasistratus (c. 280 bce) and Galen (129–201) (see Footnote 1.1) demonstrated the role of the diaphragm as a muscle of respiration, the origin and function of the phrenic nerve, and the function of the intercostal and accessory muscles. The function of the diaphragm was further explored by da Vinci (1452–1519), who observed that during inspiration the lung expanded in all directions following the movement of the thoracic cage. The lung collapse that followed puncture of the pleura was described by Vesalius (1514–1564). The need for fresh air was recognised by Galen, who believed it reacted with the blood in the left heart and arteries to produce the ‘vital spirit’. The absence of a visible communication between the pulmonary artery and the Footnote 1.1. Some historical references are given by the name–date system and can be found in the Further reading section.

pulmonary vein led him to suggest that blood passed through invisible pores between the two sides of the heart; thus, he failed to comprehend the function of the lung. This was surmised by Ibnal‐Naf īs (c. 1210–1288) and by Servetus (1511–1553; see Fulton 1953), who separately recognised the impermeability of the interventricular septum and proposed that blood passed from the pulmonary artery through the lung to the pulmonary vein. Harvey (1578–1657) demonstrated that blood circulated through the lung and Malpighi (1628–1694) showed that the blood capillaries were in close proximity to the smallest air spaces. These observations prepared the way for a correct understanding of lung function. The role of ventilation in maintaining life was demonstrated by Vesalius, who was able to restore the activity of the heart in an apnoeic dog by insufflating air into the trachea through a reed. Hooke (1635–1703) subsequently showed that the essential factor was a supply of fresh air. Boyle (1627–1691) and, to a lesser extent, Mayow (1643–1679) demonstrated that the constituent of air that supported combustion also supported life. Lower (1631–1691) further showed that the uptake of air in the lung caused the blood to change colour. These discoveries laid the foundations for subsequent studies of gas exchange but their importance was not immediately apparent. The confusion was such that on 22 January 1666, after a meeting of the Royal Society on the subject of respiration, Samuel Pepys wrote in his diary: ‘it is not to this day known, or concluded on among physicians how the action is managed by nature, or for what use it is’.

1.4 ­The Past 350 Years The information about the lung that was necessary for the birth of respiratory physiology was available by about the year 1667. Thereafter aspects of the subject developed at different rates, reflecting their immediate ­interest and the techniques that were available for their investigation.

5

6

1  How We Came to Have Lungs and How Our Understanding of Lung Function has Developed

1.4.1  Lung Volumes

1.4.3  Ventilatory Capacity

The volume of air that a man can inhale during a single deep breath was first measured by Borelli (1680 and 1681). Subsequent work established that this quantity in an average adult is about 200–300 in3 (3.3–4.9 l) at ambient temperature. The need for a temperature correction was pointed out by Goodwyn [6]. Thackrah [7] showed the volume of air to be less in women than in men and to be reduced among workers in flax and other occupations owing to the inhalation of dust. The measurement of vital capacity was put on a quantitative basis by Hutchinson [8]. Hutchinson defined it as ‘the greatest voluntary expiration following the deepest inspiration’ and designed a spirometer for its estimation. He showed that the vital capacity is related to the height such that ‘for every inch of height (from 5 to 6 ft.) eight additional cu. inches of air at 60°F are given out by forced expiration’. The equivalent parameter in metric units is 5.8 l m−1, which is similar to values used today (Chapter 25). He further showed that the vital capacity decreased with age and was reduced by excess weight and by disease of the lung. The measurement of residual volume by a gas dilution method was first performed by Davy (1800) (see also [9]). The method using whole body plethysmography was developed by DuBois et al. [10].

The relationship of breathlessness on exertion to vital capacity was considered by Peabody and Wentworth (1917). He also compared the ventilation during exercise with that during breathing CO2. The use of the forced vital capacity was introduced by Strohl (1919) The role of changes in lung distensibility in causing breathlessness was explored by Christie [25]. The maximal breathing capacity was introduced as a dynamic test of lung function by Jansen et  al. [26], who calculated it from the forced vital capacity. The maximal voluntary ventilation was first measured by Hermannsen [27]. The use of the proportion of the vital capacity that could be expired in 1 s as a guide to airway obstruction was introduced by Tiffeneau and Pinelli (1948). The measurement was facilitated through the addition of a timing device to the spirometer by Gaensler [28] and subsequently by McKerrow, McDermott, and Gilson (1960). A convenient and reasonably accurate peak flowmeter was developed by Wright and McKerrow [29] and other instruments followed.

1.4.2  Lung Mechanics The role of the elastic recoil of the lung in causing expiration was demonstrated by Donders [11], who was the first to measure the retractive force. This work was extended by Dixon and Brodie [12] and by Cloetta [13]. Concurrently, Rohrer (1916) was applying the concepts of Newtonian mechanics to explain the relationship between the force exerted by the respiratory muscles and the rate of airflow. This approach was extended by his successors Neergaard and Wirz [14], who used the pneumotachograph of Fleisch [15]. Neergaard and Wirz [14] also demonstrated the role of surface forces in the lung by comparing the relationship of the lung volume with the retractive force when the air in the lung was replaced by water. This work was repeated independently by Radford (1959), who, with Pattle [16], Clements [17], and Avery and Mead [18], established the physiological and chemical significance of lung surfactant. Knowledge of the viscoelastic properties of the lung was extended by Bayliss and Robertson [19], Dean and Visscher [20], Rahn et al. [21], Mead and Whittenberger [22], and their many collaborators; a seminal review was  prepared by Mead [23]. The role of antitrypsin in protecting the lung from proteolytic enzymes was discovered by Eriksson [24].

1.4.4  Blood Chemistry and Gas Exchange in the Lung During the eighteenth century, the lung’s role as an organ of gas exchange was obscured by the belief of Lavoisier (1777) and others that it was the site of combustion. This was disproved by Magnus [30], who used an extraction technique to analyse the gases in arterial and venous bloods. The use of such data for the calculation of cardiac output was proposed by Fick [31], while the true site of oxidation was demonstrated by Pflüger [32]. The techniques for analysing gases were improved by Haldane and described in Methods of Air Analysis (1912); an improved method for determining the concentrations in the blood was described by Barcroft and Haldane (1902). The tonometer methods for measuring the blood gas tensions were developed by Bohr [33] and Krogh [34]; other technical advances were reported by Peters and Van Slyke in Quantitative Clinical Chemistry [35]. The application of these and other techniques to human arterial blood was made possible through the introduction by Hürter (1912) of the procedure of arterial puncture. The relationship of the pressure to the content of oxygen in the blood was explored by Paul Bert and described in La Pression Barometrique (1878); in this he showed that the pressure and not the concentration of gases in the atmosphere is of physiological significance. The oxygen dissociation curve was described by Bohr et  al. (1904). With Hasselbalch and Krogh, Bohr (1904) showed that its shape is greatly influenced by the coexisting tension of CO2. Further advances were made by

1.4  The Past 350 Years

Barcroft and summarised in The Respiratory Function of the Blood (1914). The dissociation curve for CO2 was described by Christiansen et al. (1914) and the chemical reactions were further explored by Hasselbalch, Hastings, Roughton, Sendroy, Stadie, and others. Some of this work is described by L.J. Henderson in Blood: A Study in General Physiology [36]. The exchange of gas across the alveolar capillary membrane was considered by Bohr [37]. He found that the tension of oxygen was sometimes higher in the arterial blood than in the alveolar gas and concluded that oxygen was secreted by the alveolar cells. The measurements were in error, but the hypothesis was supported by Haldane and Smith [38]; these workers inhaled gas containing carbon monoxide (CO), and observed differences between the observed and expected CO tensions in blood. This could best be explained by secretion of oxygen. Their view was opposed by Krogh [34] and by Barcroft (1914), who believed correctly that the transfer of oxygen took place solely by diffusion. The controversy led Bohr [39] to develop his integration method for determining the mean tension of oxygen in the pulmonary capillaries and to calculate the diffusing capacity of the lung for carbon monoxide. It also stimulated physiological expeditions to high altitudes, including to Pikes Peak, Colorado, USA, described by Douglas et  al. [40], and to Cerro de Pasco, Peru, described by Barcroft in the second edition of The Respiratory Function of the Blood. Studies of conditions at high altitude were also undertaken by Dill et al. [41] and by Houston and Riley [42]. Subsequently, interest shifted to the Himalayas, where the physiological adaptations necessary for the ascent of Mount Everest were investigated by Pugh et al. [43] and West et al. [44], among others. Meanwhile, the transfer of oxygen from alveolar gas to pulmonary capillary blood was explored by Lilienthal et al. (1946) and Piiper [45]. Understanding of the transfer of carbon monoxide was advanced by Roughton and Forster [46]. The single‐ breath method for the measurement of transfer factor (diffusing capacity) for carbon monoxide was developed by Marie Krogh [47] and improved under Comroe’s guidance by Forster et al. [48]. The anatomical basis of gas exchange was described in quantitative terms by Weibel [49]. The distribution of gas in the lung was considered by Zuntz [50], who introduced the concept of dead‐space; this was first measured at post‐mortem by Loewy [51]. The dead‐space for CO2 was measured during life by Bohr [37] as well as by Haldane and others who used the method of sampling the alveolar gas devised by Haldane and Priestley [52]. By this method Douglas and Haldane [53] showed that the dead‐space increased with the depth of inspiration, but the magnitude of the increase was disputed by Krogh and Lindhard (1917), who sam-

pled the end‐tidal gas. Part of the increase was believed by Haldane to represent ventilation of the alveolar ducts and atria where the ventilation per unit of perfusion (i.e. the ventilation–perfusion ratio) was higher than in the alveoli. Haldane et al. [54] explored the effects of uneven lung function upon the composition of alveolar gas and arterial blood. The application of these concepts to patients with lung disease was described by Meakins and Davies in Respiratory Function in Disease [55]. The role of the pulmonary circulation was clarified through the application of the newly discovered technique of cardiac catheterisation by Cournand et  al. (1942) and by McMichael and Sharpey‐Schafer [56]. However, there was disagreement as to whether or not it was ethical to apply the technique to healthy people. The mechanisms underlying uneven lung function were further illuminated by the development of bronchospirometry by Jacobaeus et  al. (1932), the concept of regional inhomogeneity by Rauwerda (1946), the respiratory mass spectrometer by Fowler and Hugh‐Jones [57], the oxygen electrode by Clark et  al. [58], and radioisotope assay methods by Knipping et  al. (1955). These techniques were used to good effect by Rahn and Fenn [5], Gilson et al. [59], and West [60], who, with Wagner et al. (1974), developed the multiple inert gas elimination technique for describing ventilation–perfusion inequality. In the current era, ever more powerful imaging techniques have been developed to assess regional ventilation (V A),  their ratio ( VA/Q ) and thus also gas blood flow (Q), exchange. 1.4.5  Control of Respiration Knowledge of the central nervous regulation of respiration stems from the observations of LeGallois (1812) and Flourens (1823) that a lesion in a small area of the medulla oblongata caused breathing to cease. The location of the respiratory region was defined with increasing precision by many workers, including Lumsden [61] and Pitts et al. [62]. At an early stage, Hering and Breuer (1868) separately showed that the region received, via the vagi, sensory information on the distension of the lung. This provided the basis for a mechanism of self‐ regulation whereby the inflation of the lung tended to terminate inspiration and to initiate expiration while deflation of the lung had the opposite effect. Activity in single vagal fibres was recorded by Adrian [63] and others. Their work paved the way for dramatic advances in understanding the role of pulmonary receptors. Subsequent contributors included Whitteridge [64] and his pupils Sears (1964), Paintal and, Widdicombe showed that the muscle spindles in the respiratory muscles played a part in regulation, whilst Campbell and Howell [65] explored the role they might play in the sensation of

7

8

1  How We Came to Have Lungs and How Our Understanding of Lung Function has Developed

dyspnoea. The Hering–Breuer centenary symposium provided a seminal review [66]; it also introduced respiratory physiologists to some psychological techniques for the quantification of breathlessness. The stimulant effects upon respiration of both a relative deficiency of oxygen and a moderate excess of CO2 were known to Pflüger [67], who believed the former to be the more important factor. In this he was in agreement with Rosenthal [68]. Evidence for the role of CO2 was provided by Miescher‐Rusch [69], while Geppert and Zuntz [70] demonstrated the stimulant action of other products of metabolism. The action of CO2 in man was investigated quantitatively by Haldane and Priestley [52], who, over a wide range of barometric pressures, demonstrated that the ventilation was adjusted to maintain the alveolar CO2 tension at a constant level. J.S. Haldane’s great contribution is summarised in Respiration (1922), which was republished jointly with Priestley in 1935 [71]. The role of the blood hydrogen ion concentration in controlling breathing was suggested by Winterstein [72] and elaborated, among others, by Yandell Henderson in Adventures in Respiration [73]. Gesell [74] believed the response of the respiratory region of the brain to be affected by the metabolism of chemosensitive cells. The role of hypoxaemia was advanced through the identification by Heymans and Heymans (1926) and De Castro [75] of the chemoreceptors in the carotid and aortic bodies; their function was further studied by Comroe and Schmidt [76]. The interdependence of the responses of ventilation to hypercapnia and hypoxaemia was demonstrated by Nielsen and Smith [77], while the effects of inhalation of oxygen were studied by Leonard Hill and Flack [78], A.V. Hill et al. [79], Asmussen and Nielsen (1945), Comroe and Dripps [80], Dejours [81], and others. The combined effects on respiration of these and other factors were synthesised into a multiple theory of respiratory regulation by Gray in Pulmonary Ventilation and its Physiological Regulation [82]. 1.4.6  Energy Expenditure during Exercise The rates of exchange of oxygen and CO2 in the lung were measured by Lavoisier (1784),who showed that they varied with the level of activity. The relationship of resting metabolism to body surface area was demonstrated by Robiquet and Thillaye (see Sarrus 1839). The underlying biochemical processes were investigated by von Liebig (1843), Voit [83], Rubner (1839), and others. One important landmark was the demonstration by Fletcher and Hopkins [84] that lactic acid was produced in muscles during anaerobic contractions. The measurement of human metabolism by indirect calorimetry was facilitated by Zuntz [50] when he developed a portable apparatus. The method was validated by Atwater and Rosa (1897) using a human calorimeter. Other

e­quipment was introduced by Tissot [85], Douglas [86], Benedict and Roth (see Roth 1922), Kofranyi and Michaelis [87], Müller and Franz [88], and Wolff [89]. The need to relate the results to the body mass of the subjects was recognised by Frentzel and Reach [90]. The energy expenditure during activity was measured by many workers, including Benedict [91], while the relationship to the speed of locomotion was analysed in detail by Magne [92], A.V. Hill and his colleagues, including Lupton and co‐workers [79], Atzler and Herbst [93], Fenn [94], Margaria [95], and others.

1.5 ­Practical Assessment of Lung Function Most of the physiological concepts described in this chapter were applied to the assessment of patients with respiratory disease, starting with the vital capacity in the early nineteenth century [8]. In the 1930s, Knipping’s laboratory in Hamburg was setting the trend, using a wide range of tests, all of which have their counterparts today (Table 1.2). In 1950, when Comroe and Dripps reviewed the subject, the scope of the tests had broadened to include aspects of lung mechanics. However, the forced expiratory volume was scarcely known outside France and no test had reached its current form. This mainly happened during the next 12 years (Table  1.3); the developments were aided by a transfer of technical expertise from wartime aviation medicine [113]. Photographs of some of those who contributed are given in Figure 1.2. Since the 1960s, the means for calibration have been improved, the convenience of the tests increased and the subtlety of interpretation extended. New tests have mainly emerged in the related fields of medical physics and anthropometry. Some of them are included in the present account. Table 1.2  Tests of respiratory efficiency in Knipping’s laboratory. Aspect

Test

Normal level

Anatomical

Vital capacity

>70% pred.

Physiological

Ventilation equivalent for O2

1 m) is associated with increased risk of cardiovascular disease. In one such study the subjects were drivers and conductors employed by London Transport and the circumferences were obtained from the company records of the sizes of employees’ trousers [19]. Measurement of girth. Girth is measured using a non‐ stretchable steel tape with spring‐loaded handles so that the tension applied to the skin can be constant. The tape should be horizontal, flat against the skin, and not

Figure 4.3  Evolution of physique in a fairly typical sedentary male. The first step change is mainly skeletal. The second can be accentuated by, for example, discontinuing smoking.

MINTS

Age 17

Age 25

Age 40

49

4  Body Size and Anthropometric Measurements

Figure 4.4  Body shapes that affect the lungs. A full abdomen raises the diaphragm, a big bosom encourages shallow breathing, and intrathoracic fat leaves less space for air. Fat on the hips is of no consequence. 120 110 Circumference (cm)

50

120 Men Hip

100

90

80

80

70

70

60

10 20

30

40 50 60 Age (years)

Hip

100

Waist

90

Women

110

70

80

60

Waist

10

20

30

40 50 60 Age (years)

70

80

Figure 4.5  Relationship of hip and waist circumferences (cm) on age in men and women. Source: [15].

twisted. For trunk girths, the subject is partially undressed and stands symmetrically. Except for maximal chest expansion, measurements are made during tidal breathing. Thoracic girth is at the junction of the third and fourth sternebrae. Waist girth is the circumference at the mid‐point between the lowest rib and the iliac crest. Hip girth is the maximal pelvic circumference at the level of the greater trochanters.

4.6 ­Body Mass and Body Mass Index During childhood and adolescence the mass of the body increases in parallel with skeletal growth. In adults through into middle age the mass often continues to increase but at a slower rate. Adults who put on weight usually accumulate

fat. However, in persons who u ­ ndertake physical ­training, a gain in weight is due to an increased quantity of muscle and mineralisation of bone. The quantity of fat may then be relatively small. In later life the body mass often stabilises, and then declines. The reductions affect all bodily compartments, including fat, muscle, fluid, and bone (Figure 4.6). These changes influence the lung function and capacity for exercise and are best described in terms of simple indices of body fat and muscle. In most men, and to a lesser extent in women, a change in fat is the largest single cause for a change in lung function [20]. It is also the commonest cause for a change in body mass. Hence, for interpreting longitudinal measurements of lung function the body mass should be measured as part of the assessment. The component of body mass that is attributable to stature can be allowed for arithmetically. In adults the

4.7  Body Composition 40

40

Fat% Men

Percentage

30

Fat%

Women

30

BMI

BMI 20

20 FFMI

FFMI 10

10

0

10

20

30

40

50

60

70

0

80

10

20

30

Age (years)

40 50 60 Age (years)

70

80

Figure 4.6  Relationship of fat%, body mass index (BMI) (kg m−2), and fat‐free mass index (FFMI) (kg m−2) on age in men and women. Source: [15].

outcome is body mass index (BMI)[21], also called Quetelet’s index after the Belgian mathematician who described it in 1835 [22]: 2 BMI body mass stature (4.1)

In adults, but not in children, BMI is nearly independent of stature. For youngsters, body mass × stature−3 more nearly achieves this goal [23], but BMI is more widely used. In the absence of extra muscle, the range for healthy adults of all ages and both sexes is 18.5–25 kg m−2; a BMI in excess of 30 kg m−2 is evidence for obesity [24]. BMI has the important limitation of not distinguishing between body fat and muscle (Table 4.4). Both compartments can affect the lungs, and so interpretation is difficult when their effects have opposite signs (Table 4.1). Relative weight. Standardisation of mass for frame size can be achieved by using the correlation that exists between bone mass and elbow width. The association is the basis of the classification of the Metropolitan Life Assurance Company of New York [26]. Its relative weights Table 4.4  An example of body mass index (BMI) being a poor guide to its constituents; data from male firefighters before and after physical training. Index

Before

After

Difference

Maximal O2 uptake (mmol min−1)

133

151

18.2a

Fat% Fat‐free mass (kg) −2

Body mass index (kg m ) Source: [25]. a  P  2 [47, 48]. Coffman et al. [47] found that Dm,CO, calculated from combined simultaneous Tl,NO–Tl,CO, increased, as expected, in volunteers on exercise when infinite θNO was used but  no such increase was observed when θNO = 4.5 min−1 mmHg−1; with a rebreathing technique, Dm,CO and Vc significantly increased on exercise, with both finite and infinite θNO. Furthermore they and others [48, 49] found that the ratio of Dm,CO calculated as 0.5 × Tl,NO was 2.06–4 times Dm,CO from their ‘gold standard’ two‐stage Tl,CO at varying PO2 (the classical Roughton–Forster analysis; Figure 18.2). To address the concerns of Coffman et al. [47] regarding θNO (finite or infinite), and Hsia et al. [49] concerning α values, a hypothesis applying basic diffusion and chemical kinetic theory to the red cell was formulated to show how θNO could be finite in the continuous flow rapid mixing apparatus (CFRMA) in vitro but infinite in vivo and how α could be greater than 2.0 [50]. In the CFRMA, dilute but equimolar solutions of NO and red cells in physiological fluid are mixed and travel at high velocity down an observation tube. The reaction is followed until all the reagents have combined. As the reaction proceeds, NO molecules will diffuse past nitrosyl haemoglobin molecules to reach and react with haemoglobin deep in the red cell interior. Therefore, the rate of uptake of NO by red cells reflects a combination of internal diffusion and also chemical reaction. In a single‐breath Tl,NO, an extremely dilute NO gas mixture dissolves in plasma and diffuses into the capillary. The oxyhaemoglobin molecules greatly outnumber those of NO, which only react with the very outermost oxyhaemoglobin layers (illustrated in Figure  18.1b). Penetration of NO into the red cell is almost negligible, the resistance (1/θNO) approaches zero, and the blood conductance θNO approaches infinity. Tl,NO would therefore represent the conductance of the alveolar–­ capillary membrane, plasma, and outermost layers of the red cell. The oxygenator and anaesthetised dog experiments [2] are explained as follows: haemolysis and addition of cell‐free haemoglobin substitutes shorten the

18.5  Diffusion Limitation for Oxygen

path length for diffusion and therefore increase Dm,NO and, by definition, Tl,NO. Conversely, progressive addition of nitrite to an oxygenator model [6] builds up layers of methaemoglobin increasing the length an NO molecule has to travel before reaching oxyhaemoglobin. There are additional implications of the finite θNO in vitro, infinite in vivo hypothesis. First, because of its far slower reaction with oxyhaemoglobin, CO will penetrate deeper into the red cell interior (Figure 18.1b); the distance travelled and hence the volume of distribution of CO in the capillary blood will be greater for CO than NO. In addition, Dm, which is proportional to surface area divided by path length for diffusion, will be greater for NO than CO; α will unknown but will be >2.0. Equations  (18.8) and (18.9) (Section  18.4.2) would give the solution for Dm,CO and Vc (at infinite θNO), once an appropriate value for α has been chosen. This hypothesis, if correct, opposes the extrapolation of in vitro data derived from the CFRMA to the in vivo situation. Nevertheless, whatever the outcome regarding θNO, 1/Tl,CO can be considered as weighted towards 1/θCOVc, and 1/Tl,NO as weighted towards 1/Dm,NO. 1/θVc (a combined diffusive–reactive resistance) contributes 75% of the total resistance in the case of Tl,CO, but at most 33% of the total resistance for Tl,NO. 18.4.6  Dm and Vc: Morphometric– Physiological Comparison The physiological measurements of Tl,O2 are summarised in Table 18.3; the methodologies, which are complex, are described in Section 18.6.1. The physiological measurements of Dm,O2 are mostly derived from Dm,CO using the ‘Krogh’ factor (×1.23; Table  18.1). There is a large ‘gap’ between Tl,O2 morphometric [17] and Tl,O2 physiological [35, 36] for measurements at rest. But, if morphometric values represent a true ‘capacity’, the gap becomes less for morphometric versus exercise Tl,O2 (53 versus 17–25 [39], or 53 versus 30 [41] if particularly high Tl,O2 values are chosen). For Dm,O2 and Vc, we have chosen measurements with the simultaneous Tl,NO–Tl,CO method, using a finite θNO (Table  18.3, footnotes e, g, h) to compute Dm,CO and then Dm,O2. There is a shortfall for physiological Dm,O2 at rest (111 versus 68), but the gap almost closes on nearly maximal exercise (111 versus 77–104) when blood distribution should be more uniform and capillary volume larger owing to recruitment of units and distension of individual capillaries., Measurements of Vc at rest fall short of morphometric, but the difference reduces on severe exercise (197 ml morphometric versus 141 ml physiological) (Table 18.3). To conclude, if morphometric measurements represent the true ‘capacity’, there is reasonable agreement

between morphometry and physiological measurements made on extreme exertion (Table 18.3, footnote h). The morphometric–physiological difference for Dm,O2 at rest may represent uneven distribution of θVc in relation to Dm (analogous to V A/Q mismatch), which is not detected by morphometry. It is necessary to inject a note of caution. Alternative analyses, (i) the simultaneous Tl,NO–Tl,CO with infinite θNO (Eqs. 18.8 and 18.9; Section 18.4.5) or (ii) the classic normoxic–hyperoxic Roughton–Forster analysis with CO (Figure 18.2) with rebreathing [49], would raise exercise Vc to 188 ml (very close to morphometric), but at the expense of reducing Dm,O2 (computed from Dm,CO with finite θNO) from 104 to 30–50 mmol min−1  kPa−1. While absolute values of Dm,CO and Vc should be treated with reservations, changes (from controls to patients, from rest to exercise, from TLC to lower levels of alveolar expansion) are likely to have more validity.

18.5 ­Diffusion Limitation for Oxygen In the pulmonary function laboratory, there is no relationship across the spectrum of pathology between Tl,CO and arterial PO2, both measured at rest. Pa,O2 ­levels are mostly a reflection of maldistribution of V A/Q̇ ratios, particularly poorly ventilated but normally ­perfused units. Venous blood travelling through such units emerges desaturated in the arterial circulation. Even when Tl,CO is significantly reduced, the surface area for gas exchange is sufficient to sustain metabolic (V O2) demands at rest. But, on exercise, in patients with lung fibrosis and a 10, perfusion limits alveolar–capillary transfer; for Tl/ bQ ratios >0.1  expected for metabolic abnormality)

Additional respiratory stimulation (Table 20.1)

­acidaemia the source of the lactic acid is the liver. It is due to this organ changing from being a net consumer of lactic acid, converting it to carbohydrate, to being a producer. The condition can occur in diabetes mellitus, liver disease, acute alcoholic intoxication, and treat­ ment with oral hypoglycaemic biguanides such as phenformin.

20.3.7  Acid–Base Disturbances of Multiple Aetiologies An acid–base disturbance often has more than one cause. Some examples are given in Table 20.3. Diagnosis is often difficult in these circumstances [43–45] and requires a thoughtful integration of the patient’s history, physical examination, and other laboratory and imaging data.

References 1 Siggaard‐Anderson, O. (1963). The acid–base status of the 2

3 4 5

6

7

8

9

blood. Scand. J. Clin. Lab. Invest. 15 (Suppl 70): 1–134. Geers, C. and Gros, G. (2000). Carbon dioxide transport and carbonic anhydrase in blood and muscle. Physiol. Rev. 80: 681–715. Klocke, R.A. (1988). Velocity of CO2 exchanges in blood. Annu. Rev. Physiol. 50: 625–637. Visser, B.F. (1960). Pulmonary diffusion of CO2. Phys. Med. Biol. 5: 155–166. Christiansen, J., Douglas, C.G., and Haldane, J.S. (1914). The absorption and dissociation of carbon dioxide by human blood. J. Physiol. Lond. 48: 244–271. Swenson, E.R. and Maren, T.H. (1978). A quantitative analysis of CO2 transport at rest and during maximal exercise. Respir. Physiol. 35: 129–159. Hlastala, M.P., Swenson, E.R., and Klocke, R.A. (2008). Blood gas transport. In: Pulmonary Diseases and Disorders, 4e (eds. A.P. Fishman, J. Elias, J. Fishman, et al.), 201–206. New York: McGraw Hill. Sowah, D. and Casey, J.R. (2011). An intramolecular transport metabolon: fusion of carbonic anhydrase II to the COOH terminus of the chloride‐bicarbonate exchanger, AE1. Am. J. Physiol. 301: C336–C346. Michel, C.C. (1966). The in vivo carbon dioxide dissociation curve of true plasma. Respir. Physiol. 1: 121–137.

10 Kilmartin, J.V. and Rossi‐Bernardi, L. (1973).

11

12

13

14

15

16 17

Interaction of hemoglobin with hydrogen ions, carbon dioxide, and organic phosphates. Physiol. Rev. 53: 836–890. Lin, K.M. and Cumming, G. (1973). A model of  time‐varying gas exchange in the human lung during a respiratory cycle at rest. Respir. Physiol. 17: 93–112. Schuster, K.D. (1985). Kinetics of pulmonary CO2 transfer studied by using labelled carbon dioxide C16O18O. Respir. Physiol. 60: 21–37. Heller, H., Fuchs, G., and Schuster, K.D. (1988). Single‐breath diffusing capacities for NO, CO and C18O2 in rabbits. Pflugers Arch. 435: 254–258, erratum 581. Endeward, V., Al‐Samir, S., Itel, F., and Gros, G. (2014). How does carbon dioxide permeate cell membranes? A discussion of concepts, results and methods. Front. Physiol. 4: e382. Wagner, P.D. and West, J.B. (972). Effects of diffusion impairment on O2 and CO2 time courses in pulmonary capillaries. J. Appl. Physiol. 33: 62–71. Bidani, A. (1991). Analysis of abnormalities of capillary CO2 exchange in vivo. J. Appl. Physiol. 70: 1686–1699. Carruthers, B., Ponte, J., and Purves, M.J. (1980). Changes in partial pressure of carbon dioxide with

References

18

19

20

21

22

23

24

25

26

27

28

29

30 31

time in carotid arterial blood in cats. J. Physiol. 298: 13–23. Rispens, R., Oeseburg, B., Zock, J.P., and Zijlstra, W.G. (1980). Intra‐aortic decrease in blood plasma pH. Pflugers Arch. 386: 97–99. Johnson, R.L. and Ramanathan, M. (1985). Buffer equilibria in the lung. In: The Kidney: Physiology and Pathophysiology (eds. D.W. Seldon and G. Giebisch), 149–171. New York: Raven Press. Rahn, H., Reeves, R.B., and Howell, B.J. (1975). Hydrogen ion regulation, temperature and evolution. Am. Rev. Respir. Dis. 112: 165–172. Reeves, R.B. (1972). An imidazole alphastat hypothesis for vertebrate acid–base regulation; tissue carbon dioxide content and body temperature in bullfrogs. Respir. Physiol. 14: 219–236. Astrup, P. and Severinghaus, J.W. (1986). History of Acid–Base Physiology. Stockholm: Munksgaard. Bone, J.M., Cowie, J., Lambie, A.T., and Robson, J.S. (1974). The relationship between arterial Pco2 and hydrogen ion concentration in chronic metabolic acidosis and alkalosis. Clin. Sci. 46: 113–123. Goldstein, M.B., Gennari, F.J., and Schwartz, W.B. (1971). The influence of graded degrees of chronic hypercapnia on the acute carbon dioxide titration curve. J. Clin. Invest. 50: 208–216. Arbus, G.A., Herbert, L.A., Levesque, P.R. et al. (1969). Characterization and clinical application of the ‘significance band’ for acute respiratory alkalosis. N. Engl. J. Med. 280: 117–123. Ingram, R.H., Miller, R.B., and Tate, L.A. (1973). Acid–base response to acute carbon dioxide changes in chronic obstructive pulmonary disease. Am. Rev. Respir. Dis. 108: 225–231. Jorgensen, K. and Astrup, P. (1957). Standard bicarbonate, its clinical significance, and a new method for its determination. Scand. J. Clin. Lab. Invest. 9: 122–132. Stewart, P. (1983). Modern quantitative acid–base chemistry. Can. J. Physiol. Pharmacol. 61: 1444–1461. Narins, R.G., Jones, E.R., Stom, M.C. et al. (1982). Diagnostic strategies in disorders of fluid, electrolyte and acid–base homeostasis. Am. J. Med. 72: 496–520. Ishihara, K. and Szerlip, H.M. (1998). Anion gap acidosis. Semin. Nephrol. 18: 83–97. Kowalchuk, J.M. and Scheuermann, B.W. (1994). Acid–base regulation: a comparison of quantitative methods. Can. J. Physiol. Pharmacol. 72: 818–826.

32 Pingree, B.J.W. (1977). Acid–base and respiratory

33

34 35

36

37

38

39

40 41

42

43 44

45

46 47

changes after prolonged exposure to 1% carbon dioxide. Clin. Sci. 52: 67–74. Jones, N.L. (1997; 1657–1672). Acid–base physiology. In: The Lung: Scientific Foundations, vol. 1, 2e (eds. R.G. Crystal, J.B. West, E.R. Weibel and P.J. Barnes). Philadelphia: Lippincott‐Raven. Corey, H.E. (2003). Stewart and beyond: new models of acid–base balance. Kidney Int. 64: 777–787. Story, D.A. (2004). Bench‐to‐bedside review: a brief history of clinical acid–base. Crit. Care 8: 253–258. Swenson, E.R. (1999). Can a strong case be made for the strong ion difference. Respir. Care 44: 26–28. Effros, R.M. and Swenson, E.R. (2010). Acid‐base balance. In: Textbook of Respiratory Medicine, 5e (eds. R. Mason, V.C. Broaddus, T. Martin, et al.). Saunders Elsevier. Siggaard‐Andersen, O. and Fogh‐Andresen, N. (1995). Base excess or buffer base (strong ion difference) as measure of a non‐respiratory acid‐base disturbance. Acta Anaesthesiol. Scand. Suppl. 107: 123–128. van Ypersele de Strihou, C. and Frans, A. (1973). The respiratory response to chronic metabolic alkalosis and acidosis in disease. Clin. Sci. Mol. Med. 45: 439–448. Epstein, S.K. and Singh, N. (2001). Respiratory acidosis. Respir. Care 46: 366–383. Nattie, E.E. (1983). Ionic mechanisms of cerebrospinal fluid acid–base regulation. J. Appl. Physiol. 54: 3–12. Koeppen, B.M. and Steinmetz, P.R. (1983). Basic mechanisms of urinary acidification. Med. Clin. North Am. 67: 753–770. Swenson, E.R. (2001). Metabolic acidosis. Respir. Care 46: 383–402. Galley, H.F. and Webster, N.R. (1999). Acidosis and tissue hypoxia in the critically ill: how to measure it and what does it mean. Crit. Rev. Clin. Lab. Sci. 36: 35–60. Battle, D.C., Hizon, M., Cohen, E. et al. (1988). The use of the urinary anion gap in the diagnosis of hyperchloremic metabolic acidosis. N. Eng. J. Med. 318: 594–599. Flenley, D.C. (1971). Another non‐logarithmic acid– base diagram? Lancet 1: 961–965; 2: 160–161. Dulfano, M.J. and Ishikawa, S. (1966). Quantitative acid–base relationships in chronic pulmonary patients during the stable state. Am. Rev. Respir. Dis. 93: 251–256.

387

388

20  Carbon Dioxide

Further Reading Henderson, L.J. (1928). Blood: A Study in General Physiology. New Haven: Yale University Press. Hill, E.P., Power, G.G., and Longo, L.D. (1973). Mathematical simulation of pulmonary O2 and CO2 exchange. Am. J. Phys. 224: 904–917. Gennari, F.J., Adrogue, H.J., Galla, J.H., and Madias, N.E. (eds.) (2005). Acid–Base Disorders and Their Treatment. Boca Raton: Taylor and Francis. Piiper, J. (1991). Carbon dioxide‑oxygen relationships in gas exchange of animals. In memory of Hermann Rahn. Boll. Soc. Ital. Biol. Sper. 67: 635–658.

Rahn, H. and Prakash, O. (eds.) (1985). Acid–Base Regulation and Body Temperature. Boston: Martinus Nijhoff. Stewart, P.A. (1981). How to Understand Acid–Base: A Quantitative Acid–Base Primer for Biology and Medicine. New York: Elsevier. Stickland, M.K., Lindinger, M.I., Olfert, I.M. et al. (2013). Pulmonary gas exchange and acid–base balance during exercise. Compr. Physiol. 3: 693–739.

389

21 Control of Respiration Bertien M.‐A. Buyse CHAPTER MENU 21.1 21.2 21.3

Introduction, 389 Control of Respiration,  389 Clinical Assessment of Respiratory Control,  394 References, 403

21.1 ­Introduction Breathing is a primal homeostatic process regulating ­levels of O2 and CO2 and H+ in arterial blood and cerebral extracellular fluid. Respiratory movements occur auto­ matically and continuously throughout life and are driven by the rhythmic motor activity generated within neural circuits in the brain stem. These regulatory mechanisms operate over a wide range of physiological ­circumstances such as sleep and pregnancy, but also during speaking, eating, drinking, or other activities. In addition, the regu­ lating mechanisms operate over a wide range of levels of metabolism and altered acid–base b ­ alance and in patho­ logical circumstances inducing hypoxaemia and hyper­ capnia, such as diseases of the lungs, the cardiovascular system, and sleep‐disordered breathing. We currently only possess fragmentary understanding about how the central nervous system generates and controls the act of breathing and there are, in this area, more unresolved questions than clear answers. In the first part of this chapter the current knowledge on the complex matter of respiratory control mecha­ nisms has been summarised in a somewhat simplified way, to allow clinicians, rather than experts, to under­ stand this complex physiology. The second part contains a more detailed review on the available lung function tests to assess the control of respiration in patients. These tests have obtained a place in clinical practice to investigate effects of high altitude (Chapter 32), the patho­ physiology of sleep‐disordered breathing (Chapter 31), and even cardiovascular risk estimation [1–3].

21.2 ­Control of Respiration (Figure 21.1) Breathing is primarily controlled by the activity of net­ works of neurones in the medulla and pons. These net­ works are involved in the rhythmic contractions of upper airway and respiratory muscles, and they respond to affer­ ent information from chemoreceptors, as well as non‐ chemical receptors in lungs and respiratory muscles. Efferent signalling occurs to upper airway muscles and respiratory muscles. Although the muscles of the upper airway do not have a direct action on the chest cage, they are essential for keeping the airways open during inspira­ tion. These muscles are innervated by cranial nerves. The motor pathways from the spinal cord to respiratory mus­ cles are situated in the phrenic nerves that innervate the diaphragm, and the nerves that innervate the intercostal, abdominal, and other respiratory muscles. Their action is ultimately influenced by airway resistance, lung elastic recoil, and the inertance of the lung and chest wall. Additionally, the activity of these networks can be over‐ ridden by spinal cord reflexes in response to local sensory inputs and by cortical/mid‐brain/cerebellar sensorimotor input information coordinating speech, swallowing, etc. 21.2.1  Brain Stem Neural Respiratory Activity [4] Breathing movements are driven by rhythmic neural activ­ ity generated within spatially and functionally organised brain stem neural circuits (Figure 21.2). These networks,

Cotes’ Lung Function, Seventh Edition. Edited by Robert L. Maynard, Sarah J. Pearce, Benoit Nemery, Peter D. Wagner, and Brendan G. Cooper. © 2020 John Wiley & Sons Ltd. Published 2020 by John Wiley & Sons Ltd.

390

21  Control of Respiration Respiratory outcome parameters Spinal cord

Cortical/mid-brain/cerebellar sensorimotor inputs

Peripheral nerve

Neural respiratory network in the brain stem

Peripheral chemoreceptors

Central chemoreceptive areas Brain stem tissue H+

Brain stem blood flow

Respiratory muscle

Diaphragmatic EMG

Respiratory pressure

Pdi, P0.1

Brain stem CSF PCO2 Pa,O2 hypoxia

H+

Pa,CO2 hypercapnia

Metabolism

Airflow

Vt/TI

Ventilation

VE, Vt, f

Alveolar gas exchange

Figure 21.1  Schematic diagram of the control system of breathing. Pa,CO2, Pa,O2, and pH work as input stimuli for this feedback system. Continuous lines with arrowheads are facilitating effects in minute ventilation, and broken lines with arrowheads for suppressive effects. CSF, cerebrospinal fluid; EMG, electromyography; f, frequency; Pdi, transdiaphragmatic pressure; TI, inspiratory time; VE, ventilation; Vt, tidal volume.

Dorsal Cortical/mid-brain/cerebellar sensorimotor inputs Rostral PRG

Peripheral sensory chemoreceptive and mechanical inputs

DRG V

Pons

Rh RG mi ythm com cro og cir en par cu ic t Bötz C it T m Ex Pre- s m ran en icr smis ts Bötz C oc sio Ce In irc n uit re ntal rVRG s ce c In pt he ive mo cVRG Medulla oblongata ar Ex ea s

Spinal cord Ventral

Bu lb m ar ot an on d eu sp ro ina ne l s

Caudal

Figure 21.2  Schematic presentation of the central neural respiratory network in the brain stem. The pontine respiratory group (PRG), dorsal respiratory group (DRG), and ventilator respiratory group (VRG) sites are situated bilaterally in the pons and medulla oblongata. Bötz, Bötzinger; cVRG, caudal ventral respiratory group; In, predominantly inspiratory neurones; Ex, predominantly expiratory neurones; rVRG, rostral ventral respiratory group.

21.2  Control of Respiration

containing excitatory and inhibitory neurones, are the basis for rhythmic motor pattern generation and sensori­ motor integration [4]. These respiratory neurones are organised into three principal collections bilaterally dis­ tributed in the brain stem: the pontine respiratory group (PRG) and, in the medulla, the dorsal respiratory group (DRG) and ventral respiratory group (VRG). 21.2.1.1  The Pontine Respiratory Group

The PRG lies in the dorsolateral pons and includes both expiration‐related, inspiration‐related neurones and neurones active during the transition phase between inspiration and expiration (known as phase spanning). The region also contains laryngeal premotor neurones controlling laryngeal motoneurone activity, and spinal projecting premotor neurones controlling the phrenic motoneurone activity. The region receives afferent inputs from the cortex/ mid‐brain/cerebellum, from the central chemoreception sites, and also, via the medullary DRG, peripheral sen­ sory afferent inputs from peripheral chemoreception cir­ cuits and mechanical feedback. The PRG has afferent and efferent connections with the respiratory groups in the medulla. The PRG appears to be involved in the cortical and hypothalamic control of respiration and modulation of respiration by sensory inputs such as odour and temper­ ature and is thought to play a role in the ‘fine control’ of the inspiratory–expiratory phase transition. 21.2.1.2  The Medullary Respiratory Groups

The DRG lies in the dorsomedial part of the medulla and is rich in inspiration‐related neurones. These neurones can be divided into two main types depending upon their response to inflation [5]. Activity in the Iβ (Rβ) neurones is excited by lung inflation. The Iα (Rα) neurones are actively inhibited by inflation, the so‐called inspiratory switch‐off mechanism (Sections 21.2.2.2 and 21.2.3). The DRG receives afferent signals from peripheral chemoreceptors (Section  21.2.2.1.2) and non‐chemical receptors (Section 21.2.2.2) and projects to the VRG in the medulla. The VRG lies in the medullary ventral column. The current view is that this group contains (interacting) key excitatory and inhibitory inspiratory‐ and expiratory‐ related neurones forming the basis of respiratory rhythm generation. Numerous interconnections with the PRG and DRG control these microcircuits. The group can be divided into several subregions. The most rostral part is formed by the Bötzinger and pre‐Bötzinger complex, mainly containing rhythmogenic microcircuits. The Bötzinger complex contains predominantly expiration‐ related neurones inhibiting the inspiration‐related activity and thus provides the inspiratory–expiratory ­

phase alternation during normal breathing. The pre‐ Bötzinger complex lies below the Bötzinger complex and has been postulated to be the source of respiratory rhythm generation. Neurones in this area appear to pos­ sess pace‐maker‐like properties, at least in animals, with periodic membrane potential depolarisation adequate for rhythmogenesis [6]. Its microcircuits are the primary source of rhythmic inspiratory excitatory drive and pro­ vide inhibition of expiratory neurones during inspira­ tion. Below the pre‐Bötzinger regions, areas are found with predominantly transmission circuits. There is the rostral ventral respiratory group (rVRG), which contains mainly inspiratory bulbospinal premotor neurones relay­ ing inspiratory drive to phrenic motor neurones. These neurones are driven by excitatory pre‐Bötzinger neu­ rones and inhibited by expiratory Bötzinger neurones. The caudal ventral respiratory group (cVRG) contains mainly excitatory bulbospinal expiratory premotor neu­ rones that make synaptic contacts with thoracic and lumbar expiratory motor neurones. 21.2.2  Automatic Breathing 21.2.2.1  Chemical Control of Respiration 21.2.2.1.1  Central Chemoreceptors  Different chemo­­ se­nsitive regions are present in the brain stem; yet, there is still doubt on the specific areas and their extent [7]. These central chemosensitive areas are thought to project to the VRG in the medulla and the PRG in the pons. Both H+ concentration and tension of CO2 in blood are stimuli of these regions. It is generally accepted that H+ is the actual agent and the action of CO2 is mediated through the production of H+. The central chemore­ ceptors themselves are not in direct contact with the arterial blood but are surrounded by brain extracellular fluid that is in direct contact with cerebrospinal fluid [8]. The arterial blood is kept separate by the blood– brain barrier (Figure  21.3). The barrier is readily per­ meable to CO2 but effectively impermeable to HCO3−. In the cerebrospinal fluid, CO2 combines with H2O and via carbonic acid (H2CO3), H+ and HCO3− are pro­ duced. H+ passes the barrier and stimulates the central chemoreception regions. Hypoxia activates the peripheral (carotid) chemore­ ceptors (Section  21.2.2.1.2), but more profound and longstanding hypoxia depresses central inspiratory activity. While hypoxia depresses central inspiratory activity, there is still a normal increase in response to a rise in H+ (and Pa,CO2); so, the sensitivity of the central receptors is not affected by hypoxia. This hypoxic respiratory depression is not the result of a direct excitatory action on specific O2 chemoreceptors, but has been postulated to be due to hyperpolarisation of

391

392

21  Control of Respiration Activation by changes in Pa,CO2 through H+

Medulla oblongata extracellular fluid

H2O CO2

H+ H2CO3

HCO3–

CO 2

respiratory neurones caused by changes in concentration of different neurotransmitter substances. Hypoxia reduces the concentrations of respiratory excitatory neurotrans­ mitters, while the concentrations of inhibitory substances (e.g. adenosine, endogenous opioids) are increased. In addition, the magnitude of the hypoxic respiratory depres­ sion has been ascribed to wash‐out of CO2 at the site of the central chemoreceptors due to an increase in cerebral medullar blood flow. Indeed, it has been established that brain blood flow is sensitive to hypoxia and hypercapnia with both hypoxia and hypercapnia causing vasodilata­ tion; in this context the medulla is even far more sensitive than other brain regions (Figure 21.1) [9]. In adults, central hypoxic depression of respiration occurs only with severe hypoxia and is normally more than compensated for by the drive from the peripheral chemoreceptors. By contrast, in the newborn, respira­ tory depression can occur with lesser degrees of hypox­ aemia. This paradoxical ventilator response to hypoxia may contribute to the aetiology of neonatal apnoea. The hypoxic ventilator depression can be alleviated by nalox­ one, which is evidence for it being due in part to endog­ enous opioids. The production of endogenous opioids is also stimulated by some anaesthetic agents, and by high tensions of CO2; for this reason naloxone can alleviate carbon dioxide narcosis. 21.2.2.1.2  Peripheral Chemoreceptors (Carotid and  Aortic Bodies)  The partial pressures of O2, CO2, and/or H+

concentration in arterial blood are monitored by the carotid bodies and the aortic bodies. These are small nodes closely related to the carotid bulb and to the arch of the aorta from which they obtain a rich blood supply. The function of the aortic bodies is quite similar to that of the carotid bodies, but they do not monitor partial pressures of O2 and contribute very little to respiratory control; the following, therefore, relates only to the carotid body. Information on arterial PO2, PCO2, and H+ is trans­ mitted via the carotid body nerves – most nerve endings

Figure 21.3  Activation of central chemoreceptors across the blood–brain barrier. CSF, cerebral spinal fluid.

CSF

Capillary

arise from the glossopharyngeal (9th cranial) nerve – to the DRG in the medulla and possibly also to other cen­ tres in the brain. It should be emphasised that the peripheral chemore­ ceptors are responsive to the partial gas pressure (ten­ sion) of the gases, not their concentration. Therefore, the chemoreceptors respond to hypoxia and stagnant hypoxia associated with a reduction in systemic blood pressure and histotoxic hypoxia such as that following the administration of cyanide; however, there is little response to a reduction in delivery of O2 caused by anae­ mia or the inhalation of carbon monoxide. A number of transmitter substances appear to contrib­ ute to the response. The effect of released noradrenaline (norepinephrine) is excitatory [10]. Noradrenaline prob­ ably acts by altering the distribution of blood flow within the carotid bodies: this increases the sensitivity to hypoxia and hypercapnia and contributes to the increase in ventilation that occurs during exercise [11]. The carotid bodies are considered to be the O2 chemo­ sensitive receptors [12]. Their cells have a high metabolic rate and require a high O2 tension. Consequently, they become short of O2 when the PO2 in the arterial blood is reduced or the flow of blood to the capillary network is decreased. Either event leads to non‐aerobic metabolism in the node and local production of metabolites that stimulate the adjacent afferent nerve endings. The asso­ ciation between this nerve discharge and PO2 is weak at tensions in excess of approximately 13 kPa (100 mmHg). Below this level the rate of firing varies inversely with O2 tension. The rate of firing of the carotid body nerves is also influenced by the coexisting arterial tension of CO2 and/ or concentration of H+. When these values are increased, the response to hypoxia is augmented in a multiplicative fashion; when they are reduced, the response to hypoxia is diminished. The response to hypoxia is, therefore, markedly affected by the prevailing level of CO2 and pH. However, for the response to hypoxia to be totally sup­ pressed, the CO2 tension must be very low.

21.2  Control of Respiration

21.2.2.2  Non‐Chemical Peripheral Neural Inputs

The principal inspiratory switch‐off mechanism is the Hering–Breuer inflation reflex [13]. The stimulus is lung inflation, primarily an increase in tension of the airway wall, which is detected by the slowly adapting stretch receptors (SARs, originally known as pulmonary stretch receptors). These are located close to smooth muscle cells in both the extra‐ and intrathoracic lower airways. The information is transmitted via the vagi to the brain stem and leads to a reflex inhibition of inspiratory motor output and prolongation of the subsequent expiration with an increase in the calibre of the larynx and trachea. The Hering–Breuer reflex is well developed in many mammals and in newborn babies. In adults the inhibi­ tion has a high threshold and is normally only demon­ strable at tidal volumes in excess of 1 l, despite the receptors themselves responding to all levels of volume change. Yet, the threshold for activation of the reflex may not hold for the calibres of the larynx and trachea, which have been shown to be regulated by the Hering–Breuer mechanism at all levels of tidal volume [14]. The Hering–Breuer inflation reflex is facilitated by hyperthermia and by vibration applied to the chest wall, presumably via activation of intercostal spindles (Section 21.2.3). Both stimuli increase the frequency of breathing by reducing the duration of inspiration. In case of hypoxia, hypercapnia, or exercise the reverse happens; instead, the increase in central inspiratory activity appears to raise the threshold volume at which the infla­ tion reflex operates. In some mammalian species vagotomy results in greater tidal volumes and a slower respiratory rate. In man, denervation of the lungs as occurs during (heart–) lung transplantation is not associated with any apparent deviation from the normal pattern of resting ventilation neither when the subject is awake nor during the several stages of sleep [15]. Consequently, the question arises whether the role attributed to the stretch receptors might be shared with tendon organs and other non‐pulmonary afferents; yet, if so, this has still to be demonstrated. There exist also rapidly adapting receptors stimulated by rapid deflation, pneumothorax, and particularly by lung irritants; hence they became known as lung irritant receptors. However, these rapidly adapting receptors (RARs) are probably mechanoreceptors like the SARs, with the two groups being at opposite ends of one con­ tinuum [16]. The RARs are found in the airway epithelium from the trachea to the respiratory bronchioles. When stimulated, the receptors fire rapidly and then quickly accommodate. Like the SARs, the information is transmitted up the vagi to the brain stem. The reflex responses depend on their location. Stimulation of those in the trachea and upper airways produces bronchoconstriction, increased mucus

production, and possibly coughing. Stimulation of those in the more distal airways reduces the duration of the respiratory cycle but may increase ventilation by an increase in phrenic nerve activity and an increase in dead‐space ventilation resulting from the shallow, rapid breathing (tachypnoea). A third group of receptors are the C‐fibre ending receptors, present in airways and lung parenchyma (the  latter are also called juxta‐pulmonary capillary or J‐receptors) and served by small nerve fibres in the vagi; these also contribute to the control of breathing [17]. Pulmonary C‐fibres are stimulated by pulmonary con­ gestion, embolism, and infection and by a number of chemical substances such as anaesthetic agents (halo­ thane, ether, chloroform, trichlorethylene) but also tobacco smoke, sulfur dioxide, ammonia, etc. The recep­ tors contribute afferent information to the brain stem. The response is an inspiratory apnoea followed by rapid shallow breathing with shortening of the time of expira­ tion. In addition, there is usually bradycardia and hypo­ tension. Bronchoconstriction, somatic motor inhibition, and increased airway secretion can also occur. 21.2.3  Spinal Mechanisms The efferent nerve impulses from the respiratory centres converge upon the respiratory motor neurones in the spinal cord; here they summate with excitatory impulses from other sources including Golgi tendon organs and muscle spindles in the diaphragm and intercostal muscles. Tendon organs are tension receptors. They provide a safety mechanism whereby an undue increase in tension can inhibit the α motor neurone that is causing the mus­ cle to contract. The tendon organs also supply informa­ tion to higher centres in the brain stem. By this route they may contribute to the inspiratory switch‐off mechanism. The muscle spindles, abundant in the intercostals but rare in the diaphragm, are length receptors and they respond to stretching. They consist of sensory nerve endings joined to small muscle fibres enclosed in a fusi­ form sheath (Figure  21.4). The endings respond to stretching of the central portion of the spindle. The intrafusal muscle fibres are innervated from the spinal cord by fibres from γ motor neurones. These sensory endings communicate through afferent nerves with the α motor neurones in the spinal cord that supply the res­ piratory muscles. This arrangement provides the basis for a local control mechanism whereby the intensity of contraction of the respiratory muscles is tightly regu­ lated with respect to local events in the lungs and thorax [18]. At the start of inspiration nerve impulses from the ­ eurones. respiratory centres activate both γ and α motor n

393

394

21  Control of Respiration Dorsal root ganglion

Muscle spindle

Intrafusal fibre

Figure 21.4  Diagram illustrating the role of muscle spindles in regulating the contraction of respiratory muscles. See text for details.

Extrafusal fibre

γ fibre Spinal cord

Respiratory muscle

α fibre

The discharge of the γ motor neurones leads to contrac­ tion of the intrafusal fibres; this causes traction on the sensory endings in the spindles. Concurrently, the dis­ charge of the α motor neurones contracts the extrafusal fibres of the respiratory muscles. When the contraction is accompanied by shortening of the muscle, the length of the spindle, and hence the traction which is exerted on the sensory ending, is reduced. The discharge from the ending reflects the balance between the two forces. When the shortening of the extrafusal fibres lags behind that of the spindles, the discharge from the sensory end­ ings stimulates the α motor neurones to increase the strength of contraction of the respiratory muscles. When the respiratory muscles shorten at a greater rate than the spindles, the stimulation of the α motor neurones is reduced. In this way the strength of the contraction of the respiratory muscles is modified by the extent to which it causes the required chest wall movement. In man these responses have a latency of 33–85 ms [19], which is shorter than the minimum voluntary reaction time; their effect is to stabilise the lung volume or the rate of change of lung volume against the mechanical load which is applied. The muscle spindles can also be stimulated by vibration applied to the chest wall; by this means they can contribute to the inspiratory switch‐off mechanism. 21.2.4  Behavioural Control – Volitional Breathing Breathing is generated and ultimately controlled through activities in the brain stem but breathing can also be modulated volitionally, e.g. to subserve speech, swallow­ ing, during breathholding. These activities entail higher centre control by the motor cortex, thalamus, and cere­ bellum [20]. There is a lot of evidence that the volitional control of breathing acts via corticobulbar pathways that innervate the medullary respiratory centres. However, a ‘short‐cut’ pathway is plausible, e.g. a fast conduction pathway from the motor cortex to the diaphragm has been identified [21]. Moreover, patients with congenital hypoventilation

syndrome apparently lack the medullo‐spinal pathways that facilitate reflexogenic breathing and these patients rely on volitional breathing at all times, suggesting that cortical control is not completely integrated within the brain stem respiratory control centre [22]. Thus, the motor cortex has the neuronal apparatus to act directly on the respiratory apparatus apart from modulating activity in the medullary respiratory control centre. Yet, the available evidence does not delineate the precise form the integration takes or the extent to which it is used.

21.3 ­Clinical Assessment of Respiratory Control 21.3.1  Standardisation of the Conditions of Measurement is of Crucial Importance In everyday life, respiratory drive varies from breath to breath in response to the many factors that influence it. The drive diminishes on closing the eyes and increases momentarily with every external stimulus. Thus, to obtain a representative result the findings for a number of breaths must be averaged and the conditions of meas­ urement should be standardised. This is particularly important at rest when steady‐state measurements tend to be unreliable because some subjects hyperventilate when they become aware that interest is being taken in their breathing. Such hyperventilation is less likely when the subject’s attention is occupied by reading a book or by performing an exercise test. In addition, since the drive to respiration is normally increased during exer­ cise, an assessment while exercising may reveal an abnor­ mality that is not detectable at rest. Precautions need to be taken when attempting to assess respiratory control: the test should not be per­ formed after a meal or after taking alcohol or caffeine, and the subject should not be stimulated by extraneous sights, exciting reading matter or sounds, interesting conversations, or exciting music [23]. Before the test, the subject should empty the bladder, adopt a comfortable posture, and be given time to adjust to the mouthpiece,

21.3  Clinical Assessment of Respiratory Control

nose clip, and other respiratory apparatus, then relax for 30 min. During the test, the subjects should preferably not think about their breathing and not consciously try to influence it. When the test entails switching between gas mixtures, this should be done in such a way that the subject is not aware of its occurrence; the gases should be of the same temperature and humidity and not taste differently. 21.3.2  Measurement of Respiratory Output (Figure 21.1) 21.3.2.1  Arterial Blood Gases

Blood gas tensions are the ‘end‐parameters’ regulated by the chemosensitive homeostatic process. Disturbed val­ ues could be the result of a respiratory control defect, but they are more often due to neural signal transmission alterations, respiratory muscular problems, chest wall/ pulmonary mechanical disturbances, and/or gas exchange disturbances in the lung due to ventilation/perfusion mis­ match or vascular shunt and even metabolic disturbances might interfere (e.g. hypothyroidism). Correct interpretation of arterial blood gases at rest is only possible if a steady state is reached and the stress and any pain related to the arterial puncture have disap­ peared. The blood should, therefore, preferably be sam­ pled via an indwelling arterial cannula and a period of approximately 30 min should be allowed for the subject’s breathing to stabilise before the blood is taken. Normal values at rest exclude gross abnormalities but not changes that occur during sleep or progressive exer­ cise. Hence, measurement in these circumstances can be appropriate. 21.3.2.2  Ventilation and Pattern of Breathing

Minute ventilation (VE) is one of the most frequently used indices of respiratory output. With use of a mouth­ piece and a nose clip the expired volume can be measured directly, or expiratory airflow can be recorded using a res­ piratory flow meter and then the airflow is integrated. Measurements can also be done without the annoying mouthpiece and nose clip, by wearing an expandable coil, i.e. a correctly calibrated respiratory inductive plethys­ mograph [24], around the chest and abdominal walls. In normal subjects, VE can be considered to reflect the res­ piratory output from the respiratory control centre; how­ ever, this is not the case in patients with airflow or volume limitations owing to abnormal resistance or compliance of the airways, the lungs, or thoracic cage and in case of neuromuscular disorders. Ventilation is the product of tidal volume (Vt) and breathing frequency ( f ): V E Vt

f

Tidal volume is determined by the inspiratory flow and time of inspiration (TI, in seconds), so VE can be also expressed as follows, where total respiratory cycle time is Ttot: VE 60 Vt / TI TI / Ttot Vt/TI is said to reflect ‘inspiratory’ drive in normal subjects; yet, this parameter is again not a reliable drive measure in case of neuromuscular or pulmonary/chest wall disease because of interfering mechanical problems. TI/Ttot is a parameter that reflects respiratory timing and is called the inspiratory duty cycle. The respiratory control system also determines the pattern of breathing. An increase in minute volume (VE) is achieved by an increase in inspiratory and expiratory flows and by shortening of the expiratory time (TE); the inspiratory time (TI) is much less affected. Indeed, dur­ ing moderate exercise or hypoxia, TI remains at the rest­ ing level. However, TI diminishes when tidal volume exceeds about half the vital capacity and this reduction is due to the Hering–Breuer inflation reflex; another excep­ tion is the condition of hyperthermia, which also results in reduction of inspiratory time. The pattern of breathing during increased ventilation under steady‐state conditions is constant. However, the constant pattern is not achieved immediately; instead, the initial response varies depending on the stimulus [25]. When the stimulus to increase ventilation acts via the carotid chemoreceptors, the first response is usually a shortening of TE and hence an increase in respiratory frequency. When the stimulus acts via central chemore­ ceptors, the initial response is an increase in tidal volume without much change in respiratory frequency: TE decreases but TI increases and the latter change is responsible for the initial increase in tidal volume. 21.3.2.3  Mouth Occlusion Pressure

A more appropriate method to assess the strength of the inspiratory drive is the rate of change in pressure that develops in the airway when airflow is occluded sud­ denly while the subject starts to breath in through a cir­ cuit [26]. The closure of the valve is abrupt and very short with measurement made 0.1 s from the start of inspiration, hence the term P0.1 (Figure  21.5). Because there are only negligible changes in lung volume during the occlusion period, these pressure measurements can be considered as respiratory output indexes relatively insensitive to the effects of limitations in elastance and resistance. Yet, the pressure that is developed varies with muscle force, the initial length of the muscle fibres, and hence with functional residual capacity. These variables should be taken into account when interpreting the result (e.g. with respiratory muscle weakness, in emphy­ sema, or during exercise).

395

21  Control of Respiration

Figure 21.5  Registration of the mouth occlusion pressure (P0.1).

Expiration

Volume

Inspiration

1 Mouth pressure (cmH2O)

396

0 –1

X

P0.1

–2 –3

Y

X = closure of the valve Y = opening of the valve

–4 0 0.1 0.2 0.3 0.4 0.5 S

This P0.1 measurement has the advantage of requiring only relatively simple equipment, but the disadvantage of interrupting breathing and hence being applicable to only a small proportion of breaths. An alternative is the dP/dtmax, which is the maximal rate of rise of pressure at the start of the inspiration when a small resistance is interposed in the external airway [27]. The dP/dtmax does not disturb ventilation and can be measured at each breath: so, dP/dtmax is more reproducible than P0.1. 21.3.2.4  Work of Breathing

The non‐elastic work done on the lungs during inspira­ tion provides a measure of respiratory drive which, like the P0.1 (Section  21.3.2.3) is almost independent of the functional condition of the lung. The work can be obtained from measurements of oesophageal pressure throughout inspiration; the instantaneous pressures together with the corresponding volumes above func­ tional residual capacity are used to delineate a volume– pressure curve. The inspiratory work is given by the area of the inspiratory part of the curve. The method provides a means for validating other simpler procedures; yet, it is seldom used for clinical assessment because of the need for intubation of the oesophagus. 21.3.2.5  Oxygen Cost of Breathing

The consumption of oxygen by the respiratory muscles reflects the respiratory drive. The consumption can be estimated from measurements of the uptake of oxygen made using a closed circuit spirometer [28]. The spirometer is fitted with an absorber for CO2 and filled with 100% O2. The ventilation is varied by arranging that the subject rebreathes from a length of wide‐bore tubing introduced between the mouthpiece and the

spirometer circuit. In terms of ventilation, the oxygen cost is given by: O2 cost

O2 uptake 1 VE 1

O2 uptake 2 VE 2

mmol l

1



where VE is the ventilation minute volume in l min−1 and (1) and (2) refer, respectively, to measurements made when breathing 100% oxygen and mixtures of CO2 in oxygen obtained by rebreathing. The oxygen cost exhibits a curvilinear relationship to ventilation. The work and the oxygen cost of breathing are increased by diseases of the lung or the chest wall that increase the work of breathing. By contrast, the oxy­ gen cost is within normal limits if the hypoventilation is due to reduced activity of the respiratory centre but the lungs are normal. In such cases the result should be expressed in terms of the hypercapnic drive to respiration: O2 cost O2 uptake 1 Pa, CO2 1

O2 uptake 2 Pa, CO2 2

mmol min 1 kPa

1



This index provides a measure of the activity of the res­ piratory centre that is independent of the structural integrity of the lungs and thoracic cage. 21.3.2.6  Diaphragmatic Electromyography

Diaphragmatic electromyographic activity can serve as a useful index of respiratory neuromotor output and is not directly affected by respiratory muscle weakness or mechanical abnormalities of the thoracopulmonary sys­ tem. Raw electromyographic signals can be ­electronically

21.3  Clinical Assessment of Respiratory Control

21.3.3  Methods of Evaluating Control of Respiration in Clinical Practice

Ventilation or P0.1(l/min–1 or cmH2O)

Ventilation or P0.1(l/min–1 or cmH2O)

The principle of testing is to alter a component of the drive to breathe and measure the response. In clinical practice, most often a change in PA,O2 or PA,CO2 is applied and ventilation or P0.1 are the response parame­ ters used; in more experimental settings, as mentioned above, work or O2 cost of breathing and diaphragmatic EMG are other possible response parameters. Chemoreflex‐mediated responses can be viewed in two different ways [31]. One is the response to changes in PA,O2 or PA,CO2 and the other is the response to actual chemoreceptor stimuli: Pa,CO2 or H+ and Pa,O2 at the peripheral chemoreceptors, and Pa,CO2 or H+ at the central chemoreceptors.

S

VRT

20

B 40

PetCO2 (mmHg)

Isoxic hypoxia

21.3.3.1  Evaluation of Hypercapnic Drive and Central Chemoreceptor Function

CO2 is the most important factor in the control of respi­ ration under normal circumstances. The hypercapnic drive to respiration is, as already mentioned above, mediated via central chemoreception regions and the carotid chemoreceptors. For central chemoreceptors, the usual stimulus is H+ and CO2 (through production of H+) and is independent of O2 tension (except to the extent that severe hypoxaemia depresses central nervous activity). For peripheral chem­ oreceptors, either CO2 tension or H+ concentration are effective stimuli and the response is dependent on and interacts with that of O2. The ventilatory and P0.1 response to PCO2 is linearly proportional to PCO2 over a wide range of values above a lower limit, the so‐called ventilator recruitment thresh­ old (VRT), beyond which the ventilation is apparently independent of the CO2 tension (Figure 21.6) [31]. This dissociation of ventilation from CO2 tension can occur following voluntary overbreathing, which increases the elimination of CO2 and lowers the CO2 alveolar tension. Below the VRT point, while awake, subjects still breathe (due to the cortical drive to breathe during wakefulness); during sleep, this VRT becomes the apnoea threshold. Above the VRT point, the ventilatory or P0.1 response to hypercapnia can be described by a linear equation: Ventilation VE or P0.1

S Pa,CO2

B



In this equation, S is the index of respiratory chemo­ sensitivity to hypercapnia and B represents the position of the regression line. The derivation requires measure­ ments of ventilation or P0.1 at a minimum of two levels of carbon dioxide tension.

Isoxic normoxia Isoxic hyperoxia

PetCO2 (mmHg)

Ventilation or P0.1(l/min–1 or cmH2O)

integrated to provide a quantitative measure of changes in respiratory activity. Either surface electrodes on the chest wall or intra‐oesophageal catheter electrodes posi­ tioned at the level of the diaphragm may be used. The technique can be combined with the technique of magnetic stimulation of bulbospinal pathways in the neck. It has even been extended by using stimulation of the cranium to initiate diaphragmatic contraction via corticospinal pathways as well. The former pathway is direct and virtually automatic. The latter includes a com­ ponent that is at least partly under voluntary control. Comparison of the two responses provides new informa­ tion about the contribution to respiratory control of higher centres in the brain [29, 30]. Yet, the use of diaphragmatic electromyography (EMG) to evaluate respiratory neuromotor output is lim­ ited to experimental settings and is rarely used in clinical practice.

Metabolic acidosis or exercise Control Sleep

PetCO2 (mmHg)

Figure 21.6  Schematic illustration of the response to isoxic changes in CO2; and illustration of the impact of different oxygen tensions, different concentrations of bicarbonate, sleep, and exercise. S, the slope of the regression line or index of respiratory chemosensitivity to hypercapnia; B represents the position of the regression line; VRT, ventilator recruitment threshold.

397

398

21  Control of Respiration

When the range of ventilation–perfusion ratios is within normal limits, the alveolar tension of CO2 may be obtained by rapid analysis of the respired gas using an end‐tidal gas sampler. When pulmonary gas exchange is uneven, the arterial or arterialised venous blood should be measured. The respiratory sensitivity to CO2 (S) varies inversely with the tension of O2 and is also affected by other cir­ cumstances. The sensitivity is increased in response to noradrenaline, progesterone, and almitrine; the sensitiv­ ity is decreased during sleep and anaesthesia. The intercept B is independent of O2 tension and of many of the other factors that influence CO2 sensitivity (S). The B value is related to the concentration of H+/ bicarbonate in blood plasma: the intercept is reduced when bicarbonate is decreased by metabolic acidosis or by persistent overbreathing, such as occurs at high alti­ tude, in pregnancy, and with some diseases of the lung parenchyma; the intercept is reduced by exercise. The B value is increased by depression of respiration sufficient to cause the retention of bicarbonate in the arterial blood such as occurs, for example, in some patients with an obstructive type of ventilatory defect and during sleep as a result of the administration of sedative drugs, including morphine and codeine. To test CO2 sensitivity either a steady‐state or a rebreathing method can be used. In the ‘original’ steady‐state method, the subject inhales two (or more) CO2 concentrations, e.g. 3% and 6%. The different CO2 concentrations are inspired until the ventilation becomes stable; it can take up to 20 min to allow equilibration in all tissues and a stable ventilatory pattern. On each occasion the ventilation is measured over the last 5 min and, if arterial blood is to be collected, it is sampled over 2 min. Breathing mixtures of CO2 in air or in a hypoxic gas mixture stimulates both central and peripheral chemoreceptors; however, the resulting increase in ventilation raises the alveolar O2 tension and (a)

Figure 21.7  (a) The end‐tidal forcing method (DEF); (b) the prospective targeting technique using the sequential gas delivery circuit; (c) the CO2 rebreathing technique; (d) the O2 rebreathing technique.

Flow meter Mouthpiece

hence reduces the peripheral chemoreceptor drive (unless the inspired oxygen tension is adjusted concur­ rently; see next paragraph). When breathing CO2 in hyperoxic mixtures, the peripheral chemoreceptor is suppressed and central chemoreceptor function meas­ ured; yet, although hyperoxia does reduce the peripheral chemoreflex response to CO2 to a minimum, there is evi­ dence that the carotid body still might provide some tonic drive to breathe [32]. Although this technique was the original way to obtain the hypercapnic ventilatory response, there are important drawbacks related to the long exposure time to hypercapnia. Secondary to hyper­ capnia, vasodilatation in the medulla oblongata with concurrent CO2 wash‐out interferes with the CO2 response and has a dampening effect. Moreover, in the case of long‐lasting hypercapnia, renal excretion of bicarbonate is increased; as a consequence, if several gas mixtures are used consecutively, the resulting acid–base changes can modify the ventilatory response. Because of these drawbacks, this original technique is no longer used as such. To address these problems, the end‐tidal forcing method (DEF) has been developed [33]. DEF controls the inspired fractional concentrations of CO2 and O2 so as to maintain constant PetO2 and PetCO2 tensions. The apparatus (Figure 21.7a) operates by measuring PetCO2 and PetO2 and a computer‐based model combined with feedback is used to predict the inspired concentrations necessary to produce the desired PetCO2 and PetO2. A gas mixer/controller provides a flow through a T‐piece with the correct concentrations sufficient to meet or exceed inspiratory flow, which is supplied to the subject to breathe. DEF is capable of producing different step changes in PetCO2 and PetO2, and has, therefore, been used to measure changing CO2 and O2 tensions (Section 21.3.3.2). DEF is not easy to use: the technique requires considerable equipment for controlling the high gas flows needed and a high gas consumption; so these

T-piece

Gas Gas and flow controller/mixer sensors Computer

N2 CO2 O2

21.3  Clinical Assessment of Respiratory Control

(b)

Low-resistance expiratory one-way valve

Expiratory port to atmosphere Expiratory reservoir

Flow meter Cross-over valve

Mouthpiece

Inspiratory reservoir Gas inlet

Low-resistance inspiratory one-way valve

Computerised gas and flow sensors and controllers

(c)

N2 CO2 O2

Two-way valve Gas meter or spirometer or pneumotachograph

7% CO2 93% O2

Mouthpiece

CO2 analyser

(d)

Two-way valve

Mouthpiece

Pump

Gas meter or spirometer or pneumotachograph

Soda lime CO2 analyser

Figure 21.7  (Continued)

O2 analyser

399

400

21  Control of Respiration

experiments are usually only done in an experimental laboratory environment. More recently, the prospective targeting technique using the sequential gas delivery cir­ cuit method has been introduced [34,35]. A flow of mixed gas is provided from a computerised gas sensor– controller. The flow is controlled so as to fill the inspira­ tory bag with a volume slightly less than tidal volume with each breath. During inspiration the subject breathes the blended gas mixture from the inspiratory bag until it is emptied, followed by expired gas from the expiratory bag; the latter via a cross‐over valve (filled arrow in Figure  21.7b) with a slightly higher opening pressure than the other one‐way valves. In this way the flow from the gas blender is equal to alveolar ventilation and the gas mixture controls PCO2 and PO2. The technique con­ sumes less gas flow and is more practical to use. When using ‘modern’ steady‐state DEF or prospective targeting with sequential gas delivery techniques, it is important to ensure that measurements are within the linear range of the response and to choose target PetCO2 levels above the VRT. These ‘modern’ steady‐state techniques are also able to separate the central and peripheral chemoreceptor effects by using the differential temporal responses of the two chemoreceptor systems to step changes in end‐tidal PCO2 [32]. Under conditions of both normoxia and hypoxia, it is generally accepted that the ventilatory response to CO2 has both rapid and slow components. The rapid component of the response (with a time con­ stant of 10–30 s) is associated with the peripheral chem­ oreflex and the slower component with the central chemoreflex. In clinical practice the rebreathing method, originally reported by Read [36], is more frequently used (Figure  21.7c). The subject rebreathes from a bag

c­ omprising 7% CO2 with the remainder (93%) being O2 in order to ‘silence’ the peripheral chemoreceptor response to CO2 and to obtain a response mediated by the central chemoreceptor alone; yet, as already men­ tioned above, even in the hyperoxic situation the carotid body may still provide some tonic drive to breathe [32]. In order to avoid long‐lasting hypercapnia, Read intro­ duced the use of a relatively small bag that initially con­ tains between 3 and 5 l of gas (~vital capacity +1 l) with an initial high CO2 concentration (7%) that would supply enough CO2 to increase Pa,CO2 and PetCO2 rapidly to the mixed venous PCO2, equilibrated with the tissues (Figure  21.8). This initial equilibration is hastened by having the subject take three or four deep breaths. From then on, PetCO2 is driven by tissue production of CO2 (the reverse of what happens when using steady‐state methods, where tissue levels are driven by inspired CO2). The PetCO2 is measured with an analyser having a fast response time of less than 0.2 s. The bag is usually con­ tained in a box with a separate outlet that is connected to a dry gas meter, spirometer, or pneumotachograph for measurement of ventilation. Rebreathing produces pro­ gressively increasing hypercapnia in a setting of hyper­ oxia. Rebreathing is terminated after 4 min or if the PetCO2 has risen to approximately 9.3–10.7  kPa or 70–80  mmHg (or CO2% 9–10%). The ventilatory response to CO2 is the slope of the relationship of venti­ lation to PetCO2 over the range where this is linear; alin­ earity is often apparent at low tensions of CO2 (Figure 21.8) and either these points or the first 45 s of rebreathing should be excluded from the analysis. Reproducibility is influenced by the number and accu­ racy of the measurements of ventilation minute volume; these should be made over finite numbers of whole breaths where it is practicable. Alternatively, timed

PetCO2 (mmHg)

70

r rise

Mixed venous plateau

40

Linea

Resting

4 min rebreathing Start

Figure 21.8  The CO2 rebreathing procedure.

Stop

21.3  Clinical Assessment of Respiratory Control

­intervals of 30 s (or 15 s) can be used. Repeating the test after an interval of 10 min will improve the accuracy, but some subjects develop a headache, so the use of dupli­ cates may be inadvisable. The rebreathing technique can be considered as a test of the central CO2 chemoresponsiveness. However, there is one caveat: the hyperoxic mixture used. As will be discussed below, a very high PO2 can itself stimulate breathing. There is considerable variation in hypercapnic responses among normal subjects. When using ventila­ tion as the outcome parameter, published mean values for S are between 1.11 and 3.80 l min−1 mmHg [37, 38]. The hypercapnic ventilatory response is dependent on vital capacity and consequently often lower in women than in men [38]. The gender difference is less when con­ sidering P0.1 responses [38]. 21.3.3.2  Evaluation of Hypoxic Drive and Peripheral Chemoreceptor Function

Although, as already mentioned above, severe hypoxia can depress the central nervous system, the hypoxic drive to respiration is primarily mediated via the periph­ eral chemoreceptors. Abrupt changes in O2 in the inspired gas are followed within a few seconds by changes in ventilation: the latter initially mirror the changes in peripheral chemoreceptor activity. The immediate changes in ventilation that fol­ low a step change in the PIO2 are subsequently dimin­ ished, or even reversed, by changes in the tension of CO2 that are secondary to the change in ventilation (Figure  21.9). These secondary adjustments are most marked at rest when they lead to the steady‐state ventila­ tion during moderate hypoxaemia not being materially changed compared with breathing air, while during O2 breathing the ventilation is usually increased. Inhalation of gas low in O2 causes an immediate increase in ventila­ tion and consequently reduces the tension of CO2. As a result, the stimulus to inspiration from CO2 decreases, and the ventilation declines towards that obtained when breathing air. Converse changes occur during the inhala­ tion of 100% O2; there is then an immediate decrease in ventilation minute volume, resulting in less CO2 being excreted and a rise in the tension of CO2, stimulating breathing. The rise in the tension of CO2 also dilates cer­ ebral arterioles and increases the flow of blood to the brain. This change causes a rise in cerebral PO2, which is in addition to that due to the initial enrichment of the inspired gas with O2. The two processes combine to increase the concentration of oxyhaemoglobin in the blood in the cerebral capillaries; less reduced haemoglo­ bin is then available to take up and transport CO2 from the brain in the form of carbamino‐haemoglobin. The tension of CO2 in the medulla rises in consequence and this further increases the hypercapnic drive to ­respiration.

Low O2 Rest

High O2 Rest

Exercise

Chemoreceptor drive

Pulmonary ventilation

CO2 drive

Figure 21.9  The sequence of events following a change in oxygen delivery. Inspiration of gas deficient in oxygen increases the drive; this stimulates respiration, washes out carbon dioxide, and decreases the overall stimulus to breathing. The ventilation then declines to a new equilibrium value where the increased chemoreceptor drive is nearly offset by the reduced drive from CO2. When breathing O2 converse changes take place. However, they are partly offset by a concurrent increase in the CO2 drive due to a reduction in the buffering capacity of blood perfusing the brain. At rest this change causes an increase in the ventilation above that when breathing air. On exercise the effect is concealed by the relatively greater importance of the chemoreceptor drive.

At rest, this increase more than compensates for the reduction in peripheral chemoreceptor drive caused by the hyperoxia; as a result, the steady‐state ventilation at rest when breathing O2 exceeds that when breathing air by about 2 l min−1. During exercise, the peripheral chem­ oreceptor drive is increased by a rise in blood noradrena­ line. When O2 is administered during exercise, the reduction in chemoreceptor drive exceeds the gain in hypercapnic drive, so ventilation decreases and the ten­ sion of CO2 rises. The rise can be substantial, of the order of 1.3 kPa or 10 mmHg, and occurs in patients with lung disease as well as healthy subjects. Peripheral chemoreceptor drive is important when the central drive to respiration is reduced, e.g. by narcotic agents or cerebral hypoxia, to the extent that CO2 no longer exerts its central stimulant effect or the ability of the respiratory apparatus to increase ventilation is impaired. In such circumstances, respiration may even depend solely upon the peripheral chemoreceptor drive; if this is removed by administration of O2, the breathing becomes reduced and may cease altogether. A reduced response to hypoxia is a feature of people who were born and grew up at high altitude. It is more frequently observed in subjects with mountain sickness and patients with cyanotic congenital heart disease. An intermediate response is observed in some athletes. The ventilatory response to hypoxia is not linearly proportional to PO2 (Figure  21.10a). In contrast with the controller line for CO2, the O2 controller line is

401

21  Control of Respiration

Ventilation or P0.1 (l min–1 or cmH2O)

(a)

Isocapnic resting PetCO2 + 5 mmHg

VE = VEo + A/PetO2 – C P0.1 = P0.1o + A/PetO2 – C

Isocapnic resting PetCO2 60

PetO2 (mmHg)

(b) Ventilation or P0.1 (l min–1 or cmH2O)

402

Isocapnic resting PetCO2 + 5 mmHg VE = α1 (Sa,O2 + α2) or P0.1 = α1 (Sa,O2 + α2)

Isocapnic resting PetCO2 Sa,O2 (%)

Figure 21.10  The hypoxic ventilatory or P0.1 response: (a) the method of Weil and (b) the method of Rebuck. (a) PetO2, end‐tidal tension of O2; VEO or P0.1O, the horizontal asymptote and C is the vertical asymptote; A is the value that reflects the hypoxic chemosensitivity. (b) α1 (generally expressed as ΔVE or ΔP0.1/ΔSa,O2) is used to express hypoxic chemosensitivity.

virtually flat until the PA,O2 decreases to a level of 6.7–8 kPa or 50–60 mmHg. These differences in shape between the hypoxic and hypercapnic ventilatory responses explain why in healthy subjects the resting ventilation and arterial blood gas values are deter­ mined largely by hypercapnic chemosensitivity rather than by hypoxic chemosensitivity. The hypoxic response plays a major role when there is considerable hypoxia, for example in patients with lung disease. PA,O2 and ventilation are hyperbolically related, and the regression curve, in an isocapnic situation, is as follows: or

V E VE O P0.1

P0.1O

A / PA , O2 C A / PA , O2 C

where VE is ventilation (l min−1); PA,O2 is the tension of O2 in the alveolar gas; VEO is the horizontal asymptote, and C is the vertical asymptote. A is the value that reflects the hypoxic chemosensitivity.

Weil et al. [36] used 32 mmHg for the C value, derived from the mean value in normal subjects. However, some researchers recommend calculating individual C values for each subject, because C values among normal sub­ jects vary considerably. The ventilatory response to hypoxia can also be expressed in logarithmic form or in terms of the extent of desaturation of the arterial blood measured by oximetry. When using Sa,O2 as the outcome parameter [39], the equation for oxygen sensitivity is, in an isocapnic condi­ tion, linear (Figure 21.10b): or

V E P0.1

1 1

Sa , O 2 Sa , O 2

2 2



where α1 (generally expressed as ΔVE or ΔP0.1/ΔSa,O2) is used to express hypoxic chemosensitivity. The rationale for the use of PO2 as the stimulus param­ eter is that the peripheral chemoreceptors do not ­actually

­  References

sense Sa,O2 but Pa,O2. However, the Sa,O2 technique of Rebuck and Campbell [40] is even clinically useful in patients in whom the alveolar–arterial O2 difference (A– aDO2) is large, e.g. in severe lung disease. Moreover, the widespread availability of oxygen saturation meters makes this test more convenient. To test hypoxic response, again, steady‐state and rebreathing methods are possible. Producing a step change in PA,O2 (a steady‐state approach) while holding PA,CO2 at the same level, using the steady‐state DEF or sequential gas delivery circuit method (Section  21.3.3.1), can measure hypoxic drive. Although this test is rarely used in the clinical setting, the technique is really valuable and allows more correct measurements of the peripheral hypoxic chemoreceptor drive than when using the isocapnic progressive hypoxia test or rebreathing test: during the rebreathing test, both the peripheral chemoreceptors and the hypoxic suppres­ sion in the brain stem are measured (Section 21.2.2.1.1) [38]. Using a dynamic measuring protocol (the transient O2 test) the steady‐state method can differentiate between both. Without going into details, the main principle of this test is to deliver 100% O2 abruptly without warning and in one or two breaths, to a subject breathing a hypoxic gas mixture. There is about a 20 s difference in the circu­ lation time between the peripheral chemoreceptors and the central brain stem regions. This procedure removes the hypoxic stimulus at the peripheral chemoreceptor; consequently, measuring the change in ventilation during the initial 10–15 s from the beginning of the 100% O2 inhalation delivers the ventilatory response originating solely from the peripheral chemoreceptors. When using the rebreathing technique (Figure 21.7d) the subject rebreathes from a bag that initially contains between 3 and 5 l of gas comprising 23% O2, 7% CO2, and  70% N2. In order to remain on a constant CO2 level, a CO2 absorber is installed with manual or servo‐­

controlled CO2 addition when needed. The further test­ ing is quite similar to the CO2 rebreathing test described above. During this hypoxic test the PetO2 is gradually decreased to a level of 5.4–6 kPa or 40–45 mmHg or an Sa,O2 minimum of 80%. The response to hypoxic drive can be obtained by an isocapnic technique in which the ventilation is related to the alveolar oxygen tension or oxygen saturation during hypoxia of increasing intensity (isocapnic progressive hypoxia test). Yet, as already men­ tioned above, one should take into account that the development of hypoxic depression during progressive hypoxia causes an underestimation of the true hypoxic response [41]. The PetCO2 is generally maintained iso­ capnic (~at the level of resting room air‐breathing) [39]. It may also be kept at a 5 mmHg higher value (~the mixed venous value) [40]. This hypercapnic hypoxic condition combines the effects of O2 and CO2 and can, therefore, not be considered as a real hypoxic response. Yet, this hypercapnic hypoxic approach has the advantage that, owing to the hypercapnia, the hypoxic curve is shifted to the right, so that a larger part of the hypoxic curve can be measured at less hypoxic levels [42]. Because of the dangers of hypoxia, both monitoring of Sa,O2 by pulse oximetry and an ECG must be done throughout the study. Interindividual variation in the hypoxic response is generally even larger than for the hypercapnic response. When using Sa,O2 and VE as outcome parameters, nor­ mal mean values for S are between −0.15 and −1.02 l min−1% in the case of a PetCO2 level at the resting arterial value [37] and between −1.18 and −3.20 in the case of a PetCO2 level at the mixed venous value [37, 38]. Similar to the hypercapnic ventilatory response, also the hypoxic ventilator response is dependent on vital capacity and, consequently, is often lower in women than in men [38]. The gender difference is less for P0.1 responses [38].

­References 1 Fatemian, M., Gamboa, A., Léon‐Velarde, F. et al.

(2003). Selected contribution: ventilatory response to CO2 in high‐altitude natives and patients with chronic mountain sickness. J. Appl. Physiol. 94: 1279–1287. Ponikowski, P., Chua, T.P., Anker, S.D. et al. (2001). 2 Peripheral chemoreceptor hypersensitivity: an ominous sign in patients with chronic heart failure. Circulation 104: 544–549. Javaheri, S. and Dempsey, J.A. (2013). Central sleep 3 apnea. Compr. Physiol. 3: 141–163. Smith, J.C., Abdala, A.P., Borgmann, A. et al. (2013). 4 Brainstem respiratory networks: building blocks and microcircuits. Trends Neurosci. 36: 152–162.

5 von Baumgarten, R. and Kanzow, E. (1958). The interaction

of two types of inspiratory neurons in the region of the tractus solitarius of the cat. Arch. Ital. Biol. 96: 361–373. Smith, J.C., Ellenberger, H.H., Ballanyi, L. et al. (1991). 6 Pre‐Botzinger complex: a brain stem region that may generate respiratory rhythm in mammals. Science 254: 726–729. Nattie, E. and Li, A. (2009). Central chemoreception is 7 a complex system that involves multiple brain stem sites. J. Appl. Physiol. 106: 1464–1466. Leusen, I.R. (1954). Chemosensitivity of the respiratory 8 centre. influence of CO2 in the cerebral ventricles on respiration. Am. J. Phys. 176: 39–44.

403

404

21  Control of Respiration

9 Berkenbosch, A., Olievier, C.N., and DeGoede, J.

10

11

12

13

14

15

16 17

18

19

20

21

22 23

(1995). Respiratory responses to hypoxia: peripheral and central effects. In: Modelling and Control of Ventilation (eds. S. SJG, L. Adams and B.J. Whipp), 251–256. New York: Plenum. Yamamoto, M., Nishimura, M., Kobayashi, S. et al. (1994). Role of endogenous adenosine in hypoxic ventilatory response in humans: a study with dipyridamole. J. Appl. Physiol. 76: 196–203. Foo, I.T., Warren, P.M., and Drummond, G.B. (1996). Influence of oral clonidine on the ventilatory response to acute and sustained isocapnic hypoxia in human males. Br. J. Anaesth. 76: 214–220. Fidone, S.J. and Gonzalez, C. (1986). Initiation and control of chemoreceptor activity in the carotid body. In: Handbook of Physiology, Section 3: The Respiratory System, Vol. 2: Part 1 (eds. A.P. Fishman, N.S. Cherniack, J.G. Widdicombe and S.R. Geiger), 247– 312. Bethesda, MD: American Physiology Society. Hamilton, R.D., Winning, A.J., Horner, R.L., and Guz, A. (1988). The effect of lung inflation on breathing in man during wakefulness and sleep. Respir. Physiol. 73: 145–154. Cohen, M.I. (1975). Phrenic and recurrent laryngeal discharge patterns and the Hering‐Breuer reflex. Am. J. Phys. 228: 1489–1496. Shea, S.A., Horner, R.L., Banner, N.R. et al. (1988). The effect of human heart‐lung transplantation upon breathing at rest and during sleep. Respir. Physiol. 72: 131–149. Paintal, A.S. (1973). Vagal sensory receptors and their reflex effects. Physiol. Rev. 53: 159–227. Coleridge, J.C.G. and Coleridge, H.M. (1984). Afferent C‐fibre innervation of the lungs and airways and its functional significance. Rev. Physiol. Biochem. Pharmacol. 99: 1–110. Sears, T.E. (1964). Efferent discharges in alpha and fusimotor fibres of intercostal nerves of the cat. J. Physiol. Lond. 174: 295–315. Davis, J.N. and Sears, T.A. (1967). The effects of sudden alterations in load on human intercostal muscles during voluntary activation. J. Physiol. Lond. 190: 36P–38P. Evans, K.C., Shea, S.A., and Saykin, A.J. (1999). Functional MRI localisation of central nervous system regions associated with volitional inspiration in humans. J. Physiol. Lond. 520: 383–392. Gandevia, S.C. and Rothwell, J.C. (1987). Activation of human diaphragm from the motor cortex. J. Physiol. Lond. 384: 109–118. Gozal, D. (1998). Congenital central hypoventilation syndrome: an update. Pediatr. Pulmonol. 26: 273–282. Severinghaus, J.W. (1976). A proposed standard determination of ventilatory responses to hypoxia and hypercapnia. Chest 70: 129–131.

24 Banzett, R.B., Mahan, S.T., Garner, D.M. et al. (1995).

25

26

27 28

29

30

31 32

33

34

35

36

37

38

A simple and reliable method to calibrate respiratory magnetometers and Respitrace. J. Appl. Physiol. 79: 2169–2176. Gardner, W.N. (1980). The pattern of breathing following step changes of alveolar partial pressure of carbon dioxide and oxygen in man. J. Physiol. Lond. 300: 55–73. Whitelaw, W.A., Derenne, J.P., and Milic‐Emili, J. (1975). Occlusion pressure as a measure of respiratory center output in conscious man. Respir. Physiol. 23: 181–189. Milic‐Emili, J. (1982). Recent advances in clinical assessment of control of breathing. Lung 160: 1–17. Campbell, E.J.M., Westlake, E.K., and Cherniack, R.M. (1958). The oxygen consumption and efficiency of the respiratory muscles of young male subjects. Clin. Sci. 18: 55–64. Straus, C., Locher, C., Zelter, M. et al. (2004). Facilitation of the diaphragmatic response to transcranial magnetic stimulation by increases in human respiratory drive. J. Appl. Physiol. 97: 902–912. Demoule, A., Verin, E., Ross, E. et al. (2003). Intracortical inhibition and facilitation of the response of the diaphragm to transcranial magnetic stimulation. J. Clin. Neurophysiol. 20: 59–64. Duffin, J. (2011). Measuring the respiratory chemoreflexes in humans. Respir. Physiol. Neurobiol. 177: 71–79. Pedersen, M.E., Tatemian, M., and Robbins, P.A. (1999). Identification of fast and slow ventilatory responses to carbon dioxide under hypoxic and hyperoxic conditions in humans. J. Physiol. 15: 273–287. Robbins, P.A., Swanson, G.D., Micco, A.J., and Schibert, W.P. (1982). A fast gas‐mixing system for breath‐to‐ breath respiratory control studies. J. Appl. Physiol. Respir. Environ. Exerc. Physiol. 52: 1358–1362. Banzett, R.B., Garcia, R.T., and Mooravi, S.H. (2000). Simple contrivance “clamps” end‐tidal PCO(2) and PO(2) despite rapid changes in ventilation. J. Appl. Physiol. 88: 1597–1600. Somogyi, R.B., Vesely, A.E., Preiss, D. et al. (2005). Precise control of end‐tidal carbon dioxide levels using sequential rebreathing circuits. Anaesth. Intensive Care 33: 726–732. Read, D.J.C. (1967). A clinical method for assessing the ventilatory response to carbon dioxide. Aust. Ann. Med. 16: 20–32. Akiyama, Y. and Kawakami, Y. (1999). Clinical assessment of the respiratory system. In: Control of Breathing in Health and Disease, Lung Biology in Health and Disease (eds. M.D. Altose and Y. Kawakami), 251–287. New York: Marcel Dekker. van Klaveren, R.J. and Demedts, M. (1998). Determinants of the hypercapnic and hypoxic response in normal man. Respir. Physiol. 113: 157–165.

­  References

39 Weil, J.V., Byrnne‐Quinn, E., Sodal, I.D. et al. (1970).

Hypoxic ventilatory drive in normal man. J. Clin. Invest. 49: 1061–1072. 40 Rebuck, A.S. and Campbell, E.J.M. (1974). A clinical method for assessing the ventilatory response to hypoxia. Am. Rev. Respir. Dis. 109: 345–350. 1 Temp, J.A., Henson, L.C., and Ward, D.S. (1994). Effect 4 of a subanesthetic minimum alveolar concentration of

isoflurane on two tests of the hypoxic ventilatory response. Anesthesiology 80: 739–750. 2 van Klaveren, R.J. and Demedts, M. (1998). A 4 mathematical and physiological evaluation of the different hypoxic response models in normal man. Respir. Physiol. 113: 123–133.

405

407

22 The Sensation of Breathing Mathias Schroijen, Paul W. Davenport, Omer Van den Bergh, and Ilse Van Diest CHAPTER MENU 22.1 22.2 22.3 22.4 22.5

Introduction, 407 Afferent Input of Respiratory Sensory Information,  408 Assessment of Respiratory Sensation and Dyspnoea,  411 Factors Modulating the Experience of Respiratory Sensations and Dyspnoea,  414 Conclusion, 418 References, 419

22.1 ­Introduction Breathing is a vital and continuous activity, generating a multitude of somatic, proprioceptive, and interoceptive sensations that are accessible to awareness. Most of the time, however, breathing sensations do not capture attention and do not lead to conscious perception. Typically, they are ‘gated out’ [1], allowing attention to be directed to other motivationally relevant stimuli in the outer or inner environment. This is consistent with two well‐accepted tenets in cognitive psychology and symptom perception research [2]. The first is that attention is a limited resource. This implies that environmental and internal stimuli continuously compete among and with each other to enter awareness. Thus, respiratory sensations are a constant source of information that compete with other types of information in order to enter awareness and to give rise to conscious perception. Second, attention allows for the prioritisation of motivationally relevant stimuli to enter awareness. Therefore, and similar to perception of stimuli within other modalities, motivational relevance and emotions can be expected to play a major role in respiratory perception. Because breathing is a vital function, breathing sensations become motivationally relevant when breathing becomes more difficult. Conscious perception of respiratory dysfunction can be considered the basis for a behavioural layer in a hierarchical defence system to protect respiratory function. When more local, automatic, and reflexive regulatory systems within the body fail, this behavioural system is typically engaged. For example, the

experience of intense and sudden breathlessness is likely to interrupt ongoing behaviour in order to rearrange processing priorities, and to produce a compelling drive to gasp, to open the window, to flee from closed places, and/or to seek medical help  –  all of which can be ­considered behaviours that aim to compensate for a dysfunctional automatic regulation of gas exchange. Humans are capable of verbalising their breathing sensations and of communicating their breathing d ­ iscomfort to others. However, the language used to describe breathing sensations is complex, culturally determined, and endowed with both sensory and affective dimensions [3], making it sometimes difficult to accurately interpret self‐ reported respiratory symptoms. Perceived sensations and symptoms are often not linearly associated with sensory input, pulmonary function, or disease severity [3]. For instance, self‐reported symptoms of patients with asthma or chronic obstructive pulmonary disease (COPD) correlate more strongly with negative affectivity (a personality trait) than with actual respiratory parameters [4, 5]. At the more extreme end, there can be a total lack of correspondence between respiratory symptoms and physiological measures [6]. This is common in patients diagnosed with multiple chemical sensitivity or with the hyperventilation syndrome. For example, symptoms of hypocapnia can occur in a normocapnic state [6,  7]. As self‐reported symptoms are a major source of information for physicians to judge a patient’s condition and treatment effects, it is important to understand how respiratory sensations and verbal

Cotes’ Lung Function, Seventh Edition. Edited by Robert L. Maynard, Sarah J. Pearce, Benoit Nemery, Peter D. Wagner, and Brendan G. Cooper. © 2020 John Wiley & Sons Ltd. Published 2020 by John Wiley & Sons Ltd.

408

22  The Sensation of Breathing

reports thereof are being generated, and which factors modulate respiratory perception. The present chapter aims to describe processes that are related to respiratory symptom perception in accordance with the basic tenets of the biopsychosocial model [8]. This model distinguishes between (i) sensory information, signalling potential physiological dysregulation, and (ii) the transduction and modulation of this information into a subjective emotional experience, eventually leading to behaviour. First, we will describe the afferent input of the respiratory sensory information to the brain, and relate this to a range of various breathing sensations that can be perceived consciously. Second, we will pay attention to how breathing sensations can be assessed. Finally, we will discuss psychosocial factors that have been found to modulate conscious perception and symptom reporting.

r­epresents their cognitive interpretation of the sensory information, respiratory motor behaviour, central neural state, experience, anticipated breathing behaviour, and threat to the individual of the change in breathing. Numerous afferents are modulated by respiratory‐ related stimuli. Table 22.1 lists some of the major afferents that can be stimulated by components of respiration. Respiratory‐related afferents are generally evaluated according to the subcortical, reflexogenic control of breathing effects. Afferents that directly modulate the breathing pattern are chemosensitive, mechanosensitive, or combined Table 22.1  General, non‐exhaustive overview of major respiratory‐modulated afferent populations. The convergence and divergence of these general afferent populations and their subpopulations (not indicated in this table) provide the inter‐ and exteroceptive neural input for respiratory sensation.

22.2 ­Afferent Input of Respiratory Sensory Information

Central nervous system nerves, second‐order neurones in brain stem

22.2.1  Respiratory Sensation

Nasal chemoreceptors

Various sensations are associated with breathing and the respiratory tract. Some of these sensations are specific to the act of breathing (respiratory‐specific sensations), including shortness of breath, the urge to breathe, sense of breathlessness, chest tightness, lung volume, breathing effort, air hunger, heavy (excessive) ventilation, nasal airflow, and suffocation. Other sensations relate to interoceptive states that include the respiratory system (interoceptive sensations), such as the urge to cough, the urge to swallow, the urge to sneeze, sense of exercise effort, sense of nasal congestion, and cardiac dysfunction. Both types of respiratory sensation probably result from the cortical processing and interpretation of the convergent and divergent properties of higher order afferent and neural elements that raise the respiratory sensation to consciousness. For example, the primary respiratory sensations specifically associated with a compromised breathing are experienced as a subjective sense of breathing discomfort [9], also referred to as dyspnoea. In such case, dyspnoea is an interpretation resulting from the processing of sensory input from multiple afferent modalities encompassing different types of infor­ m ation (somatic localisation, discrimination, unpleasantness, escape motor behaviour), and the processing of this information in the context of neural prediction of negative outcomes. A spectrum of dyspnoeic sensations exists that reflects afferent modalities, intensity and duration, and central state modulation of sensory information. The language people use to express their breathing sensations

Nasal mechanoreceptors Nasal thermoreceptors Buccal mechanoreceptors Buccal chemoreceptors Buccal thermoreceptors Genioglossus muscle mechanoreceptors Palatal mechanoreceptors Palatal C‐fibres Pharyngeal mucosal slow adapting mechanoreceptors Pharyngeal mucosal rapidly adapting mechanoreceptors Pharyngeal mucosal C‐fibres Pharyngeal muscle mechanoreceptors Upper oesophageal mechanoreceptors Upper oesophageal C‐fibres Lower oesophageal C‐fibres Lower oesophageal mechanoreceptors Laryngeal muscle mechanoreceptors Laryngeal slow adapting receptors Laryngeal rapidly adapting receptors Laryngeal C‐fibres Tracheal slowly adapting receptors Tracheal rapidly adapting receptors Tracheal C‐fibres Intrathoracic pulmonary stretch receptors Intrathoracic rapidly adapting receptors Lung bronchial C‐fibres (Continued)

22.2  Afferent Input of Respiratory Sensory Information

Table 22.1  (Continued) Lung pulmonary C‐fibres Visceral pleural C‐fibres Mediastinal pleural C‐fibres Aortic arterial chemoreceptors Carotid arterial chemoreceptors Spinal nerves, second‐order neurones in spinal cord and brain stem Intercostal muscle spindles Intercostal muscle tendon organs Intercostal joint receptors Intercostal muscle C‐fibres Costal diaphragm mechanoreceptors Costal diaphragm C‐fibres Crural diaphragm mechanoreceptors Crural diaphragm C‐fibres Abdominal muscle mechanoreceptors Abdominal muscle C‐fibres Cutaneous thoracic mechanoreceptors

mechano‐chemosensitive. These afferents constitute the sensory modalities that transduce the specific exteroceptive or interoceptive modality providing the peripheral afferent input to respiratory neural control centres. Classification of an afferent population as a respiratory‐related afferent requires the afferents to generate action potentials that code respiratory parameters such as lung volume, arterial blood oxygen, respiratory muscle contraction, airway pressure, lung inflammation, and other physical parameters that are part of respiration. Each parameter constitutes a physical modality that, when transduced into a sensory neural code, is a sensory modality. However, a physical modality can also modulate the action potential activity pattern of multiple afferent populations, which results in a specific respiratory sensory modality that is coded by convergent multiple afferent populations. There are also less obvious afferent populations that are activated or inhibited by respiratory physical modalities such as thoracic cutaneous mechanoreceptors that change their action potential pattern with the physical movement of the thoracic skin during respiratory motion. Similarly, for example, nasal or mouth afferents are modulated by the movement of air, respiratory‐ related temperature changes, respiratory‐related pressure changes, chemicals entering the respiratory tract, and external impediments to entry of air into the breathing passages. These afferent populations enter the central nervous system via either cranial or spinal nerves. The primary

afferents converge and/or diverge onto second‐order neurones located throughout the spinal cord, brain stem, and subcortical nuclei. It is the convergent and divergent properties of second‐ and higher order neurones processing primary afferent input in combination with the central neural respiratory status (motor drive) that underlies respiratory sensations. Under the normal physiological state, conscious individuals are mostly unaware of their breathing. This suggests that there is a gating out of respiratory‐related sensory modalities and motor drive from cognitive ­neural pathways involved in conscious perception. When breathing changes sufficiently or in a pattern that raises a respiratory‐related modality to a cognitive threshold, conscious awareness of the change in breathing occurs. This conscious awareness is evaluated by the central nervous system using the sensory afferent neural substrate relayed by second‐order neurones and the brain stem respiratory neural motor state. It is the integration of multiple afferent inputs that provides specificity, somatic localisation, and discrimination of the change in physical state of the respiratory apparatus. Because multiple respiratory and respiratory‐modulated afferent populations have unique action potential patterns/coding of the change in breathing, discrimination of the respiratory disruption occurs, thus resulting in a modality‐specific awareness of respiratory status. For example, an external obstruction of air entry into the respiratory tract will affect multiple afferent populations in the airways, lung, and respiratory muscles, resulting in a characteristic action potential pattern that is directly related to the magnitude, duration, somatic location, and external obstruction effects on ventilatory parameters. The result of such an external obstruction is thus conscious awareness of an increased respiratory effort due to a localised obstruction to the nose or mouth. The cognitive awareness leads to central neural interpretation of the stimulus modalities, strength of the stimuli, duration of the activation of the stimuli, respiratory motor drive, and central neural state modulation of sensory information. The cognitive motor response to the change in breathing is compensatory and escape behaviour is targeted directly towards the physical modality that is affecting breathing. The result can be a reflexive change in breathing pattern and/or cognitive behavioural modulation of breathing. If the physical modality is returned to normal conditions, then respiratory‐related relief occurs and breathing returns to an unconscious state. If, however, the physical modality remains affecting breathing, additional cognitive sensations/feelings and motor behaviours will be recruited in an effort to escape the physical modality and return breathing to ‘normal’. The effects of reflex and behaviour are also monitored by respiratory‐related sensory

409

410

22  The Sensation of Breathing

­ odalities, providing feedback into the central nervous m system about the success or failure of the compensatory effort. Respiratory sensations and dyspnoea can, therefore, be conceptualised as the cognitive interpretation of respiratory motor state and sensory input from the integration of multiple afferent inputs that provide specificity, somatic localisation, and discrimination of a physical change in breathing. As such, a change in breathing can result in an aversive interpretation through higher brain centre processing of the sensory–motor information and neural prediction of negative outcomes. The convergent and divergent properties of higher order neural processing underlie respiratory sensations and dyspnoea. Figure 22.1 provides a heuristic model of respiratory‐ related afferent input underlying conscious respiratory sensations, and how such input can form the basis of both reflexive and non‐reflexive behavioural responding. 22.2.2 Dyspnoea Dyspnoea is defined as ‘a subjective experience of breathing discomfort that consists of qualitatively distinct sensations that vary in intensity’ [9]. Symptoms of dyspnoea may occur in healthy subjects during intense emotional states, heavy labour, or exercise. Dyspnoea is also frequently observed in cardiopulmonary and neuromuscular disease, as well as in anxiety and psychosomatic disorders. Two categories of ventilatory dysfunction have been distinguished as potential causes of dyspnoea: processes that impede efficient breathing, on the one hand, and processes that elevate the ventilatory needs, on the other hand [10–13]. The former include increased airway

Unaware

resistance, reduced elasticity of the lungs and chest wall, dynamic hyperinflation and associated reduced inspiratory capacity, weakness of the respiratory muscles, and increased pulmonary venous congestion and reduced lung compliance. Factors in the last category include reduced diffusion capacity causing hypoxaemia, limited physical fitness, and anaemia with compromised oxygen transport to the muscles. As ‘breathing discomfort’ is a broad and rather vague description of a symptom experience, both clinical and experimental research has tried to describe the sensory qualities in more detail. As a result, a consensus has been reached that dyspnoea encompasses three cardinal sensory qualities [14]. First, among the most consistently reported descriptors of dyspnoea is the sensation of increased work/effort of breathing that results from an increased motor command and activity of the respiratory muscles. This sensation is common to most disorders and intensifies with increasing levels of exercise load on the respiratory system. A sensation of increased work/effort of breathing is very common in COPD and asthma, and in patients with neuromuscular or interstitial disease [15]. Adding extrinsic loads to an external breathing circuit has been used as an experimental model to mimic the sensation of increased work of breathing and has allowed for the application of psychophysical techniques to study the perception of dyspnoea. A second quality of dyspnoea is air hunger, which refers to the conscious perception of the urge to breathe. This sensation results mainly from chemoreceptor stimulation and can thus be experimentally induced by creating hypercapnic and/or hypoxic conditions. Air hunger is particularly prominent

Aware

Behavioural response - Compensatory - Restorative Awareness Gating mechanism Sensory input - Somatic Breathing

- Subjective experience

- Default = gating out - Cognitive - Cortical awareness only interpretation probed when sufficient change in ventilatory state

- Proprioceptive - Interoceptive Reflexive regulation

Attention and expectation - Motivational relevance - Limited resource

Figure 22.1  The left part of the figure illustrates the default mode during which sensory input helps to regulate breathing behaviour in a non‐conscious, reflexive way. Only when breathing changes sufficiently or in a specific manner will sensory input be gated into conscious awareness and motivate behavioural responses.

22.3  Assessment of Respiratory Sensation and Dyspnoea

if at the same time ventilation is mechanically constrained (e.g. by strapping the chest and abdomen or limiting tidal volume) [14]. Clinically, air hunger is encountered in more severe stages of lung disease in which an efficient gas exchange is compromised. The most commonly used techniques to experimentally induce air hunger are rebreathing and inhalation of CO2‐enriched air. Recent evidence indicates that experimentally induced hypercapnia combined with restricted ventilation is a good model for air hunger induced by daily life activities in patients with COPD, and that patients with COPD perceive experimental air hunger in a similar way to healthy persons [16]. Finally, the sensation of chest tightness seems rather specific for asthma and is assumed to arise from bronchoconstriction and the associated stimulation of pulmonary afferents [14]. Interestingly, the above categorical qualities of dyspnoea may differ in their affective connotation, as some evidence suggests that air hunger is experienced as more unpleasant than the work/effort of breathing [17]. Similar to pain, dyspnoea has both a sensory and an unpleasantness aspect, which is consistent with the tenets of the biopsychosocial model. When respiratory sensations are processed, distinct neurobiological pathways are suggested for sensory–discriminative (somatosensory cortex, associated thalamic nuclei, and higher order association areas) and affective–motivational processing (amygdala, insular cortex, and thalamic relay nuclei) [1, 18]. Although both dimensions typically correlate strongly, they can diverge under certain circumstances. For example, the intensity and unpleasantness dimensions of dyspnoea have been differentiated during physical exercise and resistive load breathing [19], rebreathing [20], and bronchoconstriction [21]. Within the unpleasantness dimension of dyspnoea, some authors have further distinguished the immediate unpleasantness of dyspnoea, and the subsequent evaluative/emotional responses to dyspnoea [22]. Respiratory perception consists thus of a sensory–discriminative and an affective–motivational dimension with the latter most influenced by psychological aspects and acting as the main motivator to initiate health behaviour [23]. As such, respiratory sensations are perceived in a highly subjective way, often not linearly related to sensory input or underlying disease and substantially modulated by cognitive and affective factors.

22.3 ­Assessment of Respiratory Sensation and Dyspnoea Respiratory perception can be measured at different levels and in different response channels; each of them characterises a distinct stage and/or aspects of the perceptual process. Traditional psychophysical methods

(e.g. magnitude estimation) can be used to assess perceptual performance, that is, the sensitivity of respiratory perception. In addition, self‐reports are informative to see whether a conscious percept is present, and how it is experienced (e.g. in terms of intensity and unpleasantness). Behavioural measures in turn can indicate to what extent respiratory perception engages motivated behaviour, and is therefore an indicator of the motivational relevance of the respiratory percept. Also, neural activity related to respiratory sensation can be measured using evoked potentials and brain imaging methods. In the remainder, we mainly limit our overview to perceptual measures of dyspnoea, as it is the cardinal respiratory symptom that has been studied from both an experimental and clinical perspective. 22.3.1  Neural processing 22.3.1.1  Respiratory‐Related Evoked Potentials (RREPs)

The cortical processing of a respiratory sensation, created by brief occlusions of the inspiratory airway, can be assessed with event‐related potentials (ERPs) [24]. ERPs are measured at the scalp and reflect electrical changes that occur in the brain in response to a stimulus. A distinction can be made between exogenous (sensory evoked) and endogenous potentials. The sensory‐evoked exogenous potentials vary with stimulus parameters as intensity and frequency, and are thought to reflect the primary sensory pathway. Endogenous potentials are determined by psychological factors including arousal, attention, motivation, and task difficulty. In response to brief inspiratory occlusions, four peaks have been identified consistently (P1, N1, P2, N2) from neurones overlying the somatosensory cortex. Together they have been defined as RREPs. Non‐cephalic references also revealed a short latency frontal peak over the pre‐motor cortex (Nf ) [25, 26]. Logie et al. [27] showed that the generator location of the P1 and Nf components is bilaterally located within primary somatosensory and supplementary motor cortices, respectively, and respiratory muscle afferents have been hypothesised to generate both components P1 and Nf [25]. Early RREP components Nf, P1, and N1 reflect the initial arrival and first‐order sensory processing in sensory motor regions whereas the later components P2 and P3 are related to subsequent higher order cognitive processing in other cortical areas [28]. RREPs are elicited when resistive loads exceed detection thresholds and peak amplitudes are directly related to magnitude estimations [29]. When breathing becomes more difficult and unpleasant for healthy women (12 breaths against a 20 cmH2O l–1  s–1 flow resistor), heightened neural processing is suggested by increased amplitudes of the N1,

411

412

22  The Sensation of Breathing

P2, and P3 components, which may indicate increased attention to the respiratory sensation [30]. Early ERP reports indicated that attention to a stimulus, compared with ignore conditions, is characterised by larger P3 amplitudes, and that earlier P3 latencies are reported in younger than older subjects [31]. Furthermore, P3 amplitudes have been shown to be larger and to occur earlier in response to large resistive loads [32]. Webster and Colrain [26] replicated the observed increased amplitude and decreased latency of P3 for the attend condition and also showed significantly larger N1 and P2 amplitudes for the attend condition. This contrasts somehow with other findings from Harver et al. [31], who found N1 and P2 amplitudes to be unaffected by attention. Because respiratory somatosensation is usually not consciously perceived, it has been hypothesised that respiratory‐related sensory motor activity generated during eupnoeic breathing is gated out most of the time. Only when the ventilatory state changes sufficiently is respiratory information ‘passed through the gate’ to evoke cortical awareness. Consistent with this idea, the RREP seems only elicited when the load intensity exceeds the subject’s detection threshold [29, 33], suggesting an intensity‐ based mechanism governing the postulated ‘gate’. In addition, a frequency‐based gating mechanism reflects the cortical response based on stimulus frequency. For instance, paired inspiratory obstructions, presented with a 500 ms inter‐stimulus interval within a single inspiration, elicit an RREP for both the first (S1) and the second (S2) obstruction, but the S2 occlusion RREP showed decreased P1, N1, and P300 peak amplitudes [34]. The RREP paired occlusion S2/S1 ratio was similar to that in other somatosensory modalities (i.e. mouth and hand) and consistent with the existence of a somatosensory neural gating mechanism. Further research has shown that, similar to gating in other sensory modalities, changes in emotional state affect the S1/S2 ratio [35]. 22.3.1.2  Brain Imaging

Just like pain, hunger, or thirst, dyspnoea is an aversive bodily sensation that motivates behaviour [36]. Given this, it is not surprising to see that the perception of pain and dyspnoea activate similar cortico‐limbic neuronal networks [37]. Similar to pain, an increased work/effort of breathing as induced by inspiratory resistive loading activates an extensive network of sensorimotor and (para) limbic areas, among which activations in the insular cortex, dorsal anterior cingulate cortex, and amygdala [38]. Activation in paralimbic and limbic areas (anterior insular activation, the amygdala, cingulate gyrus), but not in sensorimotor areas has been observed also in response to laboratory‐induced air hunger [38]. Interestingly, insular activation is more pronounced in panic disorder patients than in healthy persons, and greater self‐reported

anxiety to hypercapnic stimulation is associated with greater insular activation [39]. In a similar vein, patients with idiopathic hyperventilation display a greater anterior insular activation than controls, when exposed to transient inspiratory occlusions [40]. Together, these brain imaging findings support the notion that dyspnoea is recruiting the fear network, thereby supporting behavioural responding to deal with a threatened respiratory function. The findings also converge on a crucial role of the anterior insula in the experience of dyspnoea, which is consistent with current neurobiological models that emphasise the relationship between interoception and awareness of the ‘self’ [36]. Future research will hopefully delineate to what extent brain imaging can be informative in the assessment of dyspnoea, beyond what can be derived from self‐reports and behaviour, and to what extent certain brain areas may be useful targets to treat dyspnoea–anxiety interactions [41]. 22.3.2  Psychophysical Methods 22.3.2.1  Classical Psychophysical Techniques

Classical psychophysical techniques can be applied to study detection, discrimination, recognition, and magnitude scaling of a respiratory stimulus. Detection of respiratory stimuli, in particular resistive loads, is possible when they exceed a certain detection threshold. The absolute threshold is the smallest detectable stimulus and corresponds to the just noticeable added resistance (ΔR) to the existing basal resistance (R), indicated by a detection rate of 50% and above. The difference threshold (discrimination threshold) is the smallest detectable change between stimuli. The basal resistance comprises both the respiratory, physiological resistance and the circuit resistance of the breathing tubes. Wiley and Zechman [42] showed that increasing the basal resistance (R) by an acute elevation of inspiratory airflow resistance in healthy participants results in an increased absolute threshold (ΔR). However, when ΔR is plotted against the total basal resistance, R, the 50% detection level for all conditions corresponds to a 25–30% load change during inspiration [42]. This means that, regardless of body position or pulmonary resistance, a constant ratio of added resistance to the basal resistance was observed, suggesting the applicability of Weber’s psychophysical law for the perception of added airflow resistance (ΔI/I = K, cf. other sensory modalities). For basal resistances up to 30 cmH2O l–1 s–1 the Weber fraction is constant and approximately 25–30% [43]. 22.3.2.2  Magnitude Estimation

Judgment of magnitude estimation is a widely used method to assess an individual’s perceptual sensitivity to respiratory stimuli and is based on the assignment of

22.3  Assessment of Respiratory Sensation and Dyspnoea

r­ atings to a series of stimuli. Perceptual performance, in general, can be assessed by using Steven’s power law (ψ = kφn), in which ψ is the subjective/perceived magnitude, φ is the stimulus intensity (e.g. peak mouth pressure, weight of load), k is a constant, and exponent n describes the relation between ψ and φ and, as such, characterises perceptual performance [44]. The n exponent reflects the slope for log (ψ)/log (φ) and provides an index of sensitivity with which a sensation is perceived. Changes in the n exponent are interpreted as reflecting alterations in perceptual performance [43, 45]. Log (ψ)/ log (φ) ratios are interpreted as an expression of the individual perceptual sensitivity. 22.3.2.3  Signal Detection

A limitation of magnitude estimation techniques is that they cannot differentiate between sensory discrimination and response biases. Signal detection theory (SDT) has been proposed to account for this problem, but has been applied only rarely to respiratory perception (but see [46]). In SDT, a response is required to trials in which a stimulus is present or not. Following a number of trials, percentages of correct identifications (hit), incorrect identifications (false alarm), and missed identifications (miss) are calculated, and mathematical treatment of those categories enables the calculation and distinction of sensitivity (d′) and response bias (c). Whereas the former reflects an individual’s capacity to make correct judgements, the latter reflects the tendency to favour one response above another [47]. 22.3.3  Self‐Reported Dyspnoea Several instruments have been developed to measure both clinical and experimentally induced dyspnoea. Instruments differ on several dimensions, so it is recommended to critically define which aspects of dyspnoea are most important to capture when selecting an instrument. Important differences include (i) the use of single versus multiple items; (ii) the extent to which the instruments rely on recall of a past symptom episode rather than on real‐time perception of dyspnoea; (iii) whether they primarily assess the sensory or the affective experience of dyspnoea, or focus more on the impact or burden caused by dyspnoea [14]. Several quick and easy, yet validated, instruments that are often applied in the context of dyspnoea inductions (e.g. cycle ergometer exercise, walking tests, CO2 inhalation, resistive load breathing) are available. For example, the modified Borg scale consists of a vertical scale with 12 points, 10 of which have verbal descriptors, while the visual analogue scale (VAS) consists of a horizontal line with two anchor points (not at all to maximum imaginable breathlessness), one at each extreme end [48]. Both

scales can be used to quantify feelings of breathlessness in terms of intensity, unpleasantness, and severity. Dyspnoea during daily activities can be measured with several validated questionnaires, such as the modified UK Medical Research Council (MRC) Dyspnoea Scale [49], and the Baseline Dyspnea Index (BDI) and Transitional Dyspnea Index (TDI) developed and validated by Mahler and colleagues [50]. The Chronic Respiratory Disease Questionnaire (CRQ) [51] assesses health‐related quality of life (HRQL) and has five items on perceived dyspnoea. Another well‐validated instrument for HRQL is the respiratory‐specific St. George’s Respiratory Questionnaire (SGRQ), containing 50 items in the three domains: symptoms, activity, and social/psychological impact [52]. A problem inherent to most instruments is that they cover only a limited number of facets that characterise dyspnoea. An exception to this may be the Multidimensional Dyspnea Profile (MDP) [53], which is a comprehensive measure of both sensory and affective dyspnoea dimensions. It comprises 12 items covering the following aspects: immediate sensory intensity, immediate unpleasantness, sensory qualities (work/effort, air hunger, mental effort/concentration, tightness, breathing characteristics), and emotions. 22.3.4  Behavioural Measures Tachypnoea has been described as a sign of acute respiratory distress [54], as are flaring of the alae nasi, a marked use of the strap muscles of the neck (resulting in a sternum that tends to move upwards), and both suprasternal and supraclavicular notch retraction. Also, it is well known that dyspnoeic patients avoid effort and physical activity in order not to further increase their dyspnoea [16]. A tendency to hyperventilate is more prominent when breathing is endangered than equally intense stress levels caused by non‐respiratory threats [55]. Furthermore, persons scoring high on the Fear of Suffocation (FoS) scale show a more pronounced decrease in end‐tidal PCO2 during exposure to a claustrophobic situation than persons scoring low [56]. In a similar vein, ventilatory responses to experimentally induced dyspnoea seem to depend largely on the fear response to dyspnoea. Research has demonstrated that persons with FoS defend their ventilation more during sustained breathing through inspiratory resistive loads than persons scoring low [57, 58]. These findings confirm that respiratory responses to consciously perceived dyspnoea are importantly influenced by the unpleasantness dimension of dyspnoea, and suggest that behavioural changes in response to dyspnoea mainly reflect the motivational drive to prevent or overcome a (potentially) threatened breathing.

413

414

22  The Sensation of Breathing

A systematic behavioural assessment of respiratory distress can be used in severely ill patients who are cognitively impaired or unconscious. In a preliminary study on patients who were weaned from the mechanical ventilator, changes in facial expression were observed concomitant with increases in heart and respiration rate, and paradoxical breathing patterns. An observation scale for respiratory distress for use in a nursing context has been developed [59]. It assesses accessory muscle use to breathe, restlessness, nasal flaring, end‐expiratory grunting, and a fearful facial expression.

22.4 ­Factors Modulating the Experience of Respiratory Sensations and Dyspnoea Many findings indicate that the perception of dyspnoea can vary to a large extent, not only between but also within individuals. A blunted or reduced perception may lead to a failure to recognise the severity of an underlying disease, resulting in inadequate disease management and life‐threatening conditions. Consistent with this, a blunted perception of dyspnoea has been observed in patients with near‐fatal asthma [60]. Conversely, exaggerated perception compared with objective disease severity may lead to an overconsumption of medical resources and to excessive suffering and impairment. This phenomenon is often observed in patients with comorbid anxiety. In many cases, a linear association between sensory input or disease severity, on the one hand, and perceived dyspnoea, on the other hand, is absent. Correlations between symptoms and objective signs of pulmonary disease are modest at best [3]. Clearly, symptom perception is based not only on the ability to sensory discriminate but also on the subjective appraisal of a sensory sensation and its expression in a given social context. Therefore, comprehensive models of perceptual performance must account for both the sensory stimulus representation and how that representation is read out and combined with stimulus evaluation, goals, beliefs, and other psychological or person‐related factors to result in a percept that instructs a behavioural response [61]. 22.4.1 Age Both in healthy persons and in patients with asthma, age is related to a higher inspiratory load detection threshold, which may reflect a decrement in proprioceptive acuity and thus a reduction of perceptual sensitivity with increasing age [62]. Older patients with asthma report also less obstruction and chest tightness in response to  methacholine‐induced bronchoconstriction [63].

Furthermore, clinical studies show that patients with asthma [64] and patients with pneumonia [65] report lower symptom levels with older age. Also, older patients with COPD often show hypercapnia at rest without concomitant dyspnoea [66]. In addition, healthy older individuals are less sensitive to stimulation by both hyper‐ and hypocapnia [67, 68]. A few studies, however, have found no or opposite age‐related effects [69, 70]. Only a few longitudinal studies have investigated the effect of age as a within‐subject change over time. In one study an increase in symptom reports was observed, independent of a decline in lung function [71], whereas in another study spanning 5 years more dyspnoea was observed with higher age (above 70 years), but a decrease was observed with very high age (above 89 years) [72]. In sum, whereas most of the evidence shows a reduced sensitivity and a reduction of symptom levels with increasing age, these associations are not always replicated and also opposite results have been reported. Obviously, a multitude of factors can play a role in these effects: peripheral changes in respiratory muscle strength, gas exchange efficiency, and/or receptor sensitivity may be involved, but also age‐related cortical loss, particularly in prefrontal regions in normal ageing. Also habituation processes across repeated distress episodes, implicit shifts in what is considered good health when getting older, and a higher motivation to regulate negative affective reactions to distress in higher age may play a role. 22.4.2 Gender Compared with men, women report more symptoms and seek medical care more frequently, while at the same time they often show a reduced perceptual sensitivity for respiratory signals in psychophysical tasks compared with men [2, 3, 46]. One hypothesis may be that gender differences in fear and anxiety underlie these phenomena, as the perception of dyspnoea in women with respiratory disease seems more influenced by anxiety than in men. Consistent with this idea, women respond with greater negative affect than men to sustained breathing through resistive loads [73], and female patients with COPD show a higher comorbidity for anxiety disorders than male patients [41]. 22.4.3 Attention A general finding is that attention, a process involving the allocation of processing resources to stimuli, amplifies responses to attended stimuli, whereas distraction from these stimuli reduces such responses [74]. Attentional focus on respiratory sensation may be determined by characteristics of the sensation itself (bottom

22.4  Factors Modulating the Experience of Respiratory Sensations and Dyspnoea

up, e.g. novelty) and/or by factors in the person (top‐ down, e.g. expectancy, previous experience, beliefs about the sensation). For example, it has been shown that focusing attention on cough sensations enhances the urge to cough and actual cough frequency compared with directing attention away from it [75]. Several other studies have also demonstrated effects of distraction. When music distracted patients with COPD from their dyspnoea, they persisted exercising for a longer time at a higher workload and they reported less exertion [76]. Listening to music during repeated walking sessions was also instrumental to improve and maintain functional performance [77]. Furthermore, distraction decreased unpleasantness but not intensity of a respiratory sensation during loaded breathing in healthy persons and during exercise in patients with COPD [78]. In addition, also respiratory‐ evoked potentials to brief inspiratory occlusions were smaller when healthy persons were distracted by emotional compared with neutral pictures [79]. 22.4.4  Fear and Anxiety Numerous clinical and experimental findings indicate an intrinsic and reciprocal relationship between anxiety and respiration. On the one hand, dyspnoea is very common in states characterised by fear and anxiety. Dyspnoea is not only a defining symptom of a panic attack [80], also healthy persons with high levels of negative affectivity (that is, tending to experience unpleasant emotions frequently) report experiencing enhanced levels of dyspnoea in daily life [81]. On the other hand, dyspnoeic stimuli trigger fear and anxiety very easily [58, 82, 83], which may explain why patients with lung disease are at greater risk of developing panic disorder [84, 85]. As fear augments respiratory drive and can trigger hyperventilation [55, 86, 87], a process in which fear and dyspnoea maintain each other may be easily installed. Important in the present context, however, are a range of consistent observations that defence system activation – states of fear and anxiety in particular  –  is capable of influencing the perception of dyspnoea. Although fear and anxiety are often used synonymously in daily language, they represent functionally distinct states. Fear is an immediate alarm reaction to a specific, imminent threat, characterised by escape impulses that mobilise for action, and result in a surge of sympathetic activation and attentional narrowing towards the threat [88]. Anxiety, on the other hand, is a future‐oriented, apprehensive state to a range of potential, uncertain threats, characterised by free‐floating attentional vigilance. It is characterised by negative affect, worry, and rumination [80, 88]. Although no research has thus far directly compared the influence of fear versus anxiety

on respiratory perception, the overall evidence in the literature suggests a differential influence which will be outlined in the following two paragraphs. 22.4.4.1 Anxiety

In a series of studies, von Leupoldt and colleagues [19] applied the affective picture viewing paradigm to study whether changing one’s emotional background state influences the perception of dyspnoea caused by resistive loaded breathing. Unpleasantness ratings increased from positive over neutral to negative pictures, whereas intensity ratings for dyspnoea did not change. Respiratory parameters (inspiratory time, respiratory frequency, and oscillatory resistance) were unaffected by the pictures’ valence. These findings suggest that particularly the affective dimension of dyspnoea is vulnerable to background emotional influences, thus matching many other findings that have investigated how affective picture viewing or other negative mood inductions modulate symptom reporting. Interestingly, during anxious states the slope of magnitude estimation tasks decreases while absolute levels of dyspnoea ratings increase, particularly when the rating refers to the affective rather than to the intensity dimension of dyspnoea. This was shown in a recent study on healthy persons, combining the affective picture viewing paradigm with a load magnitude estimation task [89]. Participants rated a range of resistive loads on the experienced ‘difficulty of breathing’ and, compared with pleasant and neutral pictures, unpleasant picture viewing increased the dyspnoea rating of the lowest intensity load (5 cmH2O l–1 s–1). Because of this increased magnitude estimation (ME) for this lowest load, the overall slope of LogME vs LogPmax (mouth pressure) was significantly lower during the unpleasant affective state than during the neutral state. This is consistent with many other findings suggesting that emotional background states modulate symptom perception only in ‘ambiguous’ situations where the experimental interoceptive stimulation is weak or even absent [41, 90, 91]. Not only transient, experimental inductions of anxiety within an individual but also more stable, personality‐ like differences in anxiety between individuals are associated with a reduced perceptual performance (flatter slope), but overall increased dyspnoea ratings in magnitude estimation tasks. During a rebreathing task, interoceptive accuracy, operationalised as the correlation between dyspnoea ratings and ventilation, is reduced in anxious compared with non‐anxious persons [92]. Also studies measuring dyspnoea ratings to resistive loads of different intensities seem to confirm that anxious compared with non‐anxious persons are characterised by a flatter slope and overall increased dyspnoea rating in ME tasks. For example, Tiller and colleagues [93] observed a flatter slope in anxious patients than in healthy controls.

415

416

22  The Sensation of Breathing

Other studies have documented that patients with COPD and comorbid panic disorder gave consistent higher estimations of breathing difficulty when confronted with sustained resistive loaded breathing than matched COPD‐only patients and healthy controls [94, 95]. Together, these findings suggest that, during an anxious state, the subjective experience and reporting of breathing discomfort is based relatively more on one’s emotional state than on sensory aspects of respiratory sensation. 22.4.4.2  Fear of Suffocation/Dyspnoeic Fear

FoS, together with fear of restriction, is a fear component underlying claustrophobia [56]. Such fears follow a continuous distribution in the population, meaning that also in healthy persons a considerable variance exists in the sensitivity to potential threats of a respiratory nature. For example, some persons dislike more than others wearing a face mask, having a stuffy nose, or residing in a warm, non‐ventilated small room. Importantly, persons scoring high on FoS (measured by the FoS scale; [56, 96]) rate breathing against resistive loads as more unpleasant than persons scoring low [57, 58]. Evidence on the role of anxiety and specific dyspnoeic fears suggests that respiratory perception is differentially influenced by both conditions. When people scoring high on FoS breathe against a respiratory load, fear and motivated attention in response to the load may result in an enhanced perceptual sensitivity. In contrast, when creating anxious states by unpleasant picture viewing, the threatening stimulus can be either the pictures or the dyspnoeic sensation, or both. In the latter conditions, it can be expected that motivated attention and perceptual effects may largely depend on what is appraised as the primary threat. Findings from research using RREPs lend some support to this interpretation. In healthy participants, P3 amplitudes to very brief inspiratory occlusions were reduced during pleasant or unpleasant picture series, compared with neutral series [79]. Earlier RREP components (Nf, P1, N1) and P2 showed no modulation by affective picture viewing. According to the authors, these findings indicate that, in low anxious healthy persons, emotion impacts the perception of respiratory sensations by reducing the attention resources available for processing afferent respiratory sensory signals [79]. Interestingly, whereas P3 amplitude was reduced, respiratory symptom reports were increased following unpleasant picture viewing. This is in line with the hypothesis that, during affective picture viewing, persons base their symptom reports less on sensory processing and relatively more on the affective state installed by affective picture viewing. In a follow‐up study, the effects of affective picture viewing on the RREPs were replicated in persons scoring low on state anxiety, but an

opposite pattern was found for anxious persons [97]. Whereas low anxious participants show reduced P2 and P3 magnitudes during the unpleasant, compared with the neutral affective context, higher anxious individuals show increased P2 and P3 magnitudes and amplitudes during the unpleasant context. In summary, RREP findings suggest that unpleasant picture viewing affects later, higher order neural processing of respiratory sensations, and that these effects are quite versatile and depend on what subjects may appraise as the primary threat. Obviously, studies are needed in which the threat value of respiratory sensations is experimentally created. This was done for instance in a fear learning protocol using a ‘benign’ resistive load of low intensity as conditioned stimulus (CS) and an aversive breathing occlusion as unconditioned stimulus (US) [83]. Specific fear for a benign respiratory sensation was learned in a group of participants for whom the low‐intensity sensation (CS) reliably predicted the occurrence of the aversive breathing occlusion (US). As a result of such CS–US pairing, this group of participants showed time‐limited, intense fear responses to the benign respiratory sensation (CS). Such fear learning was also associated with an increased ventilation during the CS load [98]. Another group for whom the aversive occlusion was much harder to predict (no reliable association between the CS and US was presented) displayed continuous, enhanced levels of distress, anxiety, and dyspnoea [83]. Such learning processes that install fear or anxiety are highly relevant for asthma, COPD, and other respiratory diseases characterised by occasional exacerbations of dyspnoea. 22.4.5  Symptom Schemata, Illness Representation, and Illness Behaviour Repeated dyspnoeic experiences are recorded in memory, creating an organised pattern representing the commonalities among several episodes. Such symptom schemata can then act as learned perceptual categories: they facilitate perception and identification of internal sensations, meaning that less information is needed for a symptom to emerge as a conscious experience. However, this facilitation often comes at a cost, as it also increases the probability towards biased perception. Even simple magnitude judgements of respiratory resistance and related affective and behavioural responses are easily biased by placing resistive loads into artificial perceptual categories [99]. The slightest change in respiratory effort may be noticed and interpreted as an asthma symptom by an ‘experienced’ patient with asthma, whereas the same change may not reach awareness in the not‐yet‐ diagnosed or novice patient with asthma. Conversely, a patient with asthma who is concerned about potential attacks may easily misperceive respiratory distress

22.4  Factors Modulating the Experience of Respiratory Sensations and Dyspnoea

caused by stress‐induced hyperventilation as signs of an impending attack. The higher the activation state of symptom schemata, the less evidence from peripheral stimulation is needed to result in a conscious symptom percept. Ultimately, symptom experiences may emerge without peripheral input at all, such as in placebo or nocebo symptoms, which in a way can be conceptualised as ‘somatovisceral illusions’. Several studies showed that, when participants breathed repeatedly an air mixture consisting of a harmless odour and CO2‐enriched air causing bodily symptoms, elevated symptom reports emerged upon subsequently perceiving the harmless odour only. These nocebo symptoms were similar to the symptoms originally induced by CO2 inhalation and resulted from automatically activated symptom schemata biasing the person’s perception of his/her somatic state [100]. Similar mechanisms could be at play in patients with medically unexplained respiratory symptoms, as often encountered in multiple chemical sensitivity, atypical chest pain, vocal chord dysfunction, panic disorder or the hyperventilation syndrome. Symptoms such as dyspnoea are typically experienced as unpleasant and represent a potential threat to the integrity of the body. This prompts additional processing to aggregate different symptoms into an organised set, to identify an appropriate illness label, to contemplate about potential causes and consequences, to solicit advice from significant others, and, possibly, to start emotional worry and rumination or attempts to minimise and neglect the symptoms. In other words, persons develop an illness representation, elaborate on it, and behave accordingly. These processes are captured by the Common‐Sense Model of Self‐Regulation, which states that illness representations include assumptions about the identity, cause, time line (duration), consequences, and controllability [101], which in turn elicit illness behaviours. These are actions intended to cope with a disease and manage the anticipated threats and include a large array of behaviours such as monitoring one’s bodily state, behaving sick, seeking medical care (or delaying it), taking sick leave from work, temporary drop out of social roles, etc. The interpretation of breathing difficulties as a symptom of an acute heart condition will prompt completely different behaviours than believing that it results from stress‐related hyperventilation or asthma. Illness representations are important for clinical outcomes. For example, illness representations held by patients with COPD impact upon functional status and disability, psychological outcomes (i.e. depression, anxiety), and quality of life [102]. The set of beliefs of patients with COPD before rehabilitation impact upon exercise capacity and quality of life after treatment [103]. Patients with COPD who were more convinced after a rehabilitation programme that

they had reached desired outcomes were less concerned about the negative consequences of COPD, were better in perceiving the variability in symptoms (cyclical time line), and were more convinced about personal controllability, after correction for differences in clinical variables [104]. More adequate and positive illness perceptions were associated with better HRQL [105]. Several studies also show that different aspects of the illness representations of pulmonary patients are related to treatment adherence, such as the use of nebulised antibiotics in patients with cystic fibrosis, self‐care during exacerbations in patients with COPD, and reliever and preventive medication use in those with asthma [106]. Illness perceptions can be measured with the Illness Perception Questionnaire‐Revised (IPQ‐R) [107]. It assesses the different aspects proposed in the Common‐ Sense Model of Self‐Regulation as a set of beliefs about one’s illness at one moment in time. A more dynamic measurement approach is described in the Symptom and Illness Attitude Model (SIAM) [108]. It takes into account the relative impact of different illness representation dimensions, how this relative impact may change in different contexts and mood states, and how central the patient aspect is in one’s self‐concept. 22.4.6  Social Context Illness representations do not originate in isolation. Symptoms are shared with significant others and interpreted within a set of beliefs that are part of popular or folk conceptualisations of disease and treatment [109], which may be quite different from scientific evidence. In addition, the internet has not only tremendously increased the availability of evidence‐based knowledge, but also socially shared lay understanding of illness and treatments. Therefore, it is important for experts and caregivers to be aware that their scientifically based approach has to compete with this socially shared belief system. For example, in an attempt to understand why people often reject a seasonal flu vaccine, it was shown that lay people use frames that are quite different from those of experts to differentiate colds from the flu, and that lay people give relatively more weight to concepts such as resistance and immunity than to infection [110]. Humans are social beings and easily engage in social comparison. The experience of breathing difficulty can be influenced by being aware of others who are more or less affected by breathing difficulties. For example, whether a patient with COPD experiences his/her condition as poor or not too bad may be implicitly influenced by standards set by peers in a pulmonary rehabilitation programme. In a set of experimental studies [111], it was found that social comparison can influence reported breathing difficulties, breathing behaviour, and exercise

417

418

22  The Sensation of Breathing

Table 22.2  Glossary of important terms and definitions. Term

Definition

Illness representation

An individual’s cognitive representation of his/her illness in terms of: 1)  Identity; the label for the illness and associated symptoms (e.g. fatigue, dyspnoea, thirst) 2)  Cause; beliefs about the causal factor for disease onset (e.g. feeling dyspnoea due to the fact that I just ran/ because I developed chronic obstructive pulmonary disease [COPD] through smoking) 3)  Timeline; the expected duration in terms of acute, chronic, or cyclical characteristics (e.g. dyspnoea will stop when I calm down/will sustain if I keep smoking) 4)  Consequences; physical, psychological, social, and economic implications for both short term (e.g. COPD affects the way people treat me) and long term (e.g. additional cardiovascular symptoms as a result of COPD) 5)  Treatment effectiveness; beliefs about curability or controllability (in case of chronic disease)

Illness beliefs

Illness beliefs encompass ideas and convictions held by individuals about their disease. These beliefs have an important influence on health behaviour in that they can both facilitate or restrict access to effective treatment

Symptom schemata

Cognitive representations in memory that result from previous symptom experiences and capture the communalities across different experiences. They are automatically activated by associated cues in the environment or in the internal state. Activated memory schemata can both facilitate and bias information processing when fitting new information within existing memory structures

Symptom percept

The cognitive impression of a symptom

Perceptual categories

Discrete subjective categories for perceptual information that are not present at the level of sensory input

Self‐concept

The conceptual idea of one’s self as constructed from (a) beliefs about one’s self (e.g. gender roles, racial identity, academic performance) and (b) the comparison of these cognitions and attitudes against societal and personal norms

Placebo/nocebo symptom

Perceived or actual improvement/worsening of symptoms in response to simulated or physiologically inert treatment (psychogenic effects)

Hierarchical defence system

A hierarchy of structures that constitute our defence system. A defensive response will be determined by the distance of perceived threat and the behavioural orientation (approach vs avoidance). Remote threat typically engages higher corticolimbic regions (e.g. ventromedial prefrontal cortex (vmPFC) and hippocampus) that process contingency and contextual information in order to elicit a survival response via the amygdala and through inhibition of the mid‐brain structures. More imminent threat on the other hand (e.g. circa‐strike) elicits defence responses through mid‐brain regions while forebrain circuits are more inhibited

Neural gating mechanism

A subcortical gating mechanism that controls the neural input to higher cortical regions and awareness. The mechanism is based on both stimulus intensity and frequency; in addition the mechanism is also sensitive to higher cortical modulation (e.g. emotions)

performance. In addition, upward social comparison with peers at the start of a rehabilitation programme predicted exercise capacity more strongly than disease severity in patients with COPD [111]. Table 22.2 provides an overview of the key psychological and psychophysiological terms that were used throughout this chapter.

22.5 ­Conclusion Breathing is a vital function. The sensations generated by the act of breathing typically reach awareness when breathing becomes endangered. From the peripheral afferents to cortical levels, signals become integrated to constitute a multi‐faceted, subjective experience that is often associated with a compelling behavioural

drive to maintain or restore an adequate respiratory function. The experience of and response to respiratory sensations and dyspnoea are importantly modulated by fears, anxiety, attention, illness beliefs, symptom schemata, and social context. Therefore, respiratory perception constitutes an active process that is often influenced by idiosyncratic factors. Problems can arise when patients are not sufficiently or incorrectly aware of precursors of symptoms (e.g. asthma triggers), fail to detect symptoms, base their reported symptoms mainly on their affective state rather than on sensory aspects, misinterpret the meaning of a symptom, and/ or fail to initiate or maintain proper treatment [3]. Since all of these variables are associated with relevant clinical outcomes, the challenge for health care providers is to become aware of these modulating factors and to understand the patient’s illness representations and

­  References

coping behaviours. Patients are not passive recipients of expert information, and nor are they obedient compliers with treatment advice. Improving the match between representations held by the patient and those held by the doctor is an important part of successful

treatment. Only when the match is ­sufficient can a true partnership be built between the patient and the doctor in order to collaboratively counter the biological, psychological, and social adversities associated with the disease.

­References 1 Davenport, P.W. and Vovk, A. (2009). Cortical and

2 3 4

5

6

7

8 9

10

11

12

13

14

subcortical central neural pathways in respiratory sensations. Respir. Physiol. Neurobiol. 167 (1): 72–86. Pennebaker, J.W. (1982). The Psychology of Physical Symptoms. New York: Springer. Harver, A. and Mahler, D.A. (1998). Dyspnea: sensation, symptom, and illness. Lung Biol. Health Dis. 111: 1–33. Put, C., Van den Bergh, O., Van Ongeval, E. et al. (2004). Negative affectivity and the influence of suggestion on asthma symptoms. J. Psychosom. Res. 57 (3): 249–255. Müllerova, H., Lu, C., Li, H., and Tabberer, M. (2014). Prevalence and burden of breathlessness in patients with chronic obstructive pulmonary disease managed in primary care. PLoS One 9 (1): e85540. https://doi. org/10.1371/journal.pone.0085540. Van Diest, I., De Peuter, S., Piedfort, K. et al. (2006). Acquired lightheadedness in response to odors after hyperventilation. Psychosom. Med. 68 (2): 340–347. Hornsveld, H.K., Garssen, B., Dop, M.F. et al. (1996). Double‐blind placebo‐controlled study of the hyperventilation provocation test and the validity of the hyperventilation syndrome. Lancet 348 (9021): 154–158. Engel, G.L. (1977). The need for a new medical model: a challenge for biomedicine. Science 196 (4286): 129–136. American Thoracic Society (1999). Dyspnea. Mechanisms, assessment, and management: a consensus statement. Am. J. Respir. Crit. Care Med. 159: 321–340. Manning, H.L. and Schwartzstein, R.M. (1995). Pathophysiology of dyspnea. N. Engl. J. Med. 333 (23): 1547–1553. O’Donnell, D.E., Revill, S.M., and Webb, K.A. (2001). Dynamic hyperinflation and exercise intolerance in chronic obstructive pulmonary disease. Am. J. Respir. Crit. Care Med. 164 (5): 770–777. Somfay, A., Porszasz, J., Lee, S.M., and Casaburi, R. (2001). Dose‐response effect of oxygen on hyperinflation and exercise endurance in nonhypoxaemic COPD patients. Eur. Respir. J. 18 (1): 77–84. Casaburi, R., Patessio, A., Ioli, F. et al. (1991). Reductions in exercise lactic acidosis and ventilation as a result of exercise training in patients with obstructive lung disease. Am. Rev. Respir. Dis. 143 (1): 9–18. Parshall, M.B., Schwartzstein, R.M., Adams, L. et al. (2012). An official American Thoracic Society

15

16

17

18

19

20

21

22

23

24

25

statement: update on the mechanisms, assessment, and management of dyspnea. Am. J. Respir. Crit. Care Med. 185 (4): 435–452. Mahler, D.A., Harver, A., Lentine, T. et al. (1996). Descriptors of breathlessness in cardiorespiratory diseases. Am. J. Respir. Crit. Care Med. 154 (5): 1357–1363. O’Donnell, C.R., Schwartzstein, R.M., Lansing, R.W. et al. (2013). Dyspnea affective response: comparing COPD patients with healthy volunteers and laboratory model with activities of daily living. BMC Pulm. Med. 13 (1): 27. Banzett, R.B., Pedersen, S.H., Schwartzstein, R.M., and Lansing, R.W. (2008). The affective dimension of laboratory dyspnea: air hunger is more unpleasant than work/effort. Am. J. Respir. Crit. Care Med. 177 (12): 1384–1390. von Leupoldt, A., Sommer, T., Kegat, S. et al. (2008). The unpleasantness of perceived dyspnea is processed in the anterior insula and amygdala. Am. J. Respir. Crit. Care Med. 177 (9): 1026–1032. von Leupoldt, A., Mertz, C., Kegat, S. et al. (2006). The impact of emotions on the sensory and affective dimension of perceived dyspnea. Psychophysiology 43 (4): 382–386. Wan, L., Van Diest, I., De Peuter, S. et al. (2009). Repeated breathlessness experiences induced by hypercapnia: differential effects on intensity and unpleasantness. Chest 135 (2): 455–461. De Peuter, S., Lemaigre, V., Van Diest, I. et al. (2007). Differentiation between the sensory and affective aspects of histamine‐induced bronchoconstriction in asthma. Respir. Med. 101 (5): 925–932. Lansing, R.W., Gracely, R.H., and Banzett, R.B. (2009). The multiple dimensions of dyspnea: review and hypotheses. Respir. Physiol. Neurobiol. 167 (1): 53–60. Banzett, R.B., Dempsey, J.A., O’Donnell, D.E., and Wamboldt, M.Z. (2000). Symptom perception and respiratory sensation in asthma. Am. J. Respir. Crit. Care Med. 162 (3): 1178–1182. Davenport, P.W., Friedman, W.A., Thompson, F.J., and Franzen, O. (1986). Respiratory‐related cortical potentials evoked by inspiratory occlusion in humans. J. Appl. Physiol. 60 (6): 1843–1848. Davenport, P.W., Colrain, I.M., and Hill, P.M. (1996). Scalp topography of the short‐latency components of

419

420

22  The Sensation of Breathing

26

27

28

29

30

31

32

33

34

35

36 37

38

39

the respiratory‐related evoked potential in children. J. Appl. Physiol. 80 (5): 1785–1791. Webster, K.E. and Colrain, I.M. (2000). The relationship between respiratory‐related evoked potentials and the perception of inspiratory resistive loads. Psychophysiology 37 (6): 831–841. Logie, S.T., Colrain, I.M., and Webster, K.E. (1998). Source dipole analysis of the early components of the RREP. Brain Topogr. 11 (2): 153–164. Chan, P.Y. and Davenport, P.W. (2010). Respiratory related evoked potential measures of cerebral cortical respiratory information processing. Biol. Psychol. 84 (1): 4–12. Knafelc, M. and Davenport, P.W. (1997). Relationship between resistive loads and P1 peak of respiratory‐related evoked potential. J. Appl. Physiol. 83 (3): 918–926. von Leupoldt, A., Bradley, M.M., Lang, P.J., and Davenport, P.W. (2010). Neural processing of respiratory sensations when breathing becomes more difficult and unpleasant. Front. Physiol. 1: 144. Harver, A., Squires, N.K., Bloch‐Salisbury, E., and Katkin, E.S. (1995). Event‐related potentials to airway occlusion in young and old subjects. Psychophysiology 32 (2): 121–129. Bloch‐Salisbury, E. and Harver, A. (1994). Effects of detection and classification of resistive and elastic loads on endogenous event‐related potentials. J. Appl. Physiol. 77 (3): 1246–1255. Davenport, P.W., Chan, P.Y., Zhang, W., and Chou, Y.L. (2007). Detection threshold for inspiratory resistive loads and respiratory‐related evoked potentials. J. Appl. Physiol. 102 (1): 276–285. Chan, P.Y. and Davenport, P.W. (2008). Respiratory‐ related evoked potential measures of respiratory sensory gating. J. Appl. Physiol. 105 (4): 1106–1113. Chan, P.Y., von Leupoldt, A., Bradley, M.M. et al. (2012). The effect of anxiety on respiratory sensory gating measured by respiratory‐related evoked potentials. Biol. Psychol. 91 (2): 185–189. Craig, A.D. (2009). A rat is not a monkey is not a human: comment on Mogil. Nat. Rev. Neurosci 10 (6): 466. von Leupoldt, A., Sommer, T., Kegat, S. et al. (2009). Dyspnea and pain share emotion‐related brain network. NeuroImage 48 (1): 200–206. Evans, K.C., Banzett, R.B., Adams, L. et al. (2002). BOLD fMRI identifies limbic, paralimbic, and cerebellar activation during air hunger. J. Neurophysiol. 88 (3): 1500–1511. Evans, K.C., Song, T.Y., Zimmerman, J.P. et al. (2015). Air hunger evokes greater anxiety and insular cortical activity in panic disorder than in healthy individuals: an arterial spin labeled fMRI study. Paper presented at the 20th Annual Meeting of the International Society for the Advancement of Respiratory Psychophysiology in Leuven, Belgium (September 2013). Biol. Psych. 104: 150–151.

40 Jack, S., Kemp, G.J., Bimson, W.E. et al. (2010). Patterns

41

42 43 44

45

46

47 48

49

50

51

52

53

54

55

of brain activity in response to respiratory stimulation in patients with idiopathic hyperventilation (IHV). In: Advances in Experimental Medicine and Biology, Vol. 669: New Frontiers in Respiratory Control: XIth Annual Oxford Conference on Modeling & Control of Breathing (eds. I. Homma, H. Onimaru and Y. Fukuchi), 341–345. New York: Springer. Hayen, A., Herigstad, M., and Pattinson, K.T. (2013). Understanding dyspnea as a complex individual experience. Maturitas 76 (1): 45–50. Wiley, R.L. and Zechman, F.W. (1967). Perception of added airflow resistance in humans. Respir. Physiol. 2 (1): 73–87. Stevens, S.S. (1957). On the psychophysical law. Psychol. Rev. 64 (3): 153–181. Mahutte, C.K., Campbell, E.J., and Killian, K.J. (1983). Theory of resistive load detection. Respir. Physiol. 51 (2): 131–139. Revelette, W.R., Zechman, F.W., Parker, D.E., and Wiley, R.L. (1984). Effect of background loading on perception of inspiratory loads. J. Appl. Physiol. 56 (2): 404–410. Harver, A., Katkin, E.S., and Bloch, E. (1993). Signal‐ detection outcomes on heartbeat and respiratory resistance detection tasks in male and female subjects. Psychophysiology 30 (3): 223–230. Green, D.M. and Swets, J.A. (1966). Signal Detection Theory and Psychophysics, vol. 1. New York: Wiley. Wilson, R.C. and Jones, P.W. (1989). A comparison of the visual analogue scale and modified Borg scale for the measurement of dyspnoea during exercise. Clin. Sci. 76 (3): 277–282. Medical Research Council Committee on the Aetiology of Chronic Bronchitis (1960). Standardized questionnaires on respiratory symptoms. Br. Med. J. 2: 1665. Mahler, D.A., Weinberg, D.H., Wells, C.K., and Feinstein, A.R. (1984). The measurement of dyspnea. Contents, interobserver agreement, and physiologic correlates of two new clinical indexes. Chest 85 (6): 751–758. Guyatt, G.H., Berman, L.B., Townsend, M. et al. (1987). A measure of quality of life for clinical trials in chronic lung disease. Thorax 42 (10): 773–778. Jones, P.W., Quirk, F.H., Baveystock, C.M., and Littlejohns, P. (1992). A self‐complete measure of health status for chronic airflow limitation: the St. George’s Respiratory Questionnaire. Am. Rev. Respir. Dis 145 (6): 1321–1327. Meek, P.M., Banzett, R., Parshall, M.B. et al. (2012). Reliability and validity of the multidimensional dyspnea profile. Chest 141 (6): 1546–1553. Honig, E. (1990; ch. 35). An overview of the pulmonary system. In: Clinical Methods: The History, Physical, and Laboratory Examinations, 3e (eds. H.K. Walker, W.D. Hall and J.W. Hurst). Boston: Butterworth. Van Diest, I., Winters, W., Devriese, S. et al. (2001). Hyperventilation beyond fight/flight: respiratory

­  References

56

57

58

59

60

61

62

63

64

65

66

67

68

69

responses during emotional imagery. Psychophysiology 38 (6): 961–968. Van Diest, I., Smits, D., Decremer, D. et al. (2010). The Dutch Claustrophobia Questionnaire: psychometric properties and predictive validity. J. Anxiety Disord. 24 (7): 715–722. Pappens, M., Smets, E., Van Den Bergh, O., and Van Diest, I. (2012). Fear of suffocation alters respiration during obstructed breathing. Psychophysiology 49 (6): 829–832. Alius, M.G., Pané‐Farré, C.A., von Leupoldt, A., and Hamm, A.O. (2013). Induction of dyspnea evokes increased anxiety and maladaptive breathing in individuals with high anxiety sensitivity and suffocation fear. Psychophysiology 50 (5): 488–497. Campbell, M.L. (2008). Respiratory distress: a model of responses and behaviors to an asphyxial threat for patients who are unable to self‐report. Heart Lung 37 (1): 54–60. Kifle, Y., Seng, V., and Davenport, P.W. (1997). Magnitude estimation of inspiratory resistive loads in children with life‐threatening asthma. Am. J. Respir. Crit. Care Med. 156 (1535): 1530–1535. Gold, J.I. and Ding, L. (2013). How mechanisms of perceptual decision‐making affect the psychometric function. Prog. Neurobiol. 103: 98–114. Allen, S.C. and Khattab, A. (2012). The airflow resistance sensing threshold during tidal breathing rises in old age in patients with asthma. Age Ageing 41 (4): 557–560. Connolly, M.J., Crowley, J.J., Charan, N.B. et al. (1992). Reduced subjective awareness of bronchoconstriction provoked by methacholine in elderly asthmatic and normal subjects as measured on a simple awareness scale. Thorax 47 (6): 410–413. Battaglia, S., Sandrini, M.C., Catalano, F. et al. (2005). Effects of aging on sensation of dyspnea and health‐ related quality of life in elderly asthmatics. Aging Clin. Exp. Res. 17 (4): 287–292. Metlay, J.P., Schulz, R., Li, Y.H. et al. (1997). Influence of age on symptoms at presentation in patients with community‐acquired pneumonia. Arch. Intern. Med. 157 (13): 1453–1459. Mahler, D.A., Ward, J.O., and Mejia‐Alfaro, R.O. (2003). Stability of dyspnea ratings after exercise training in patients with COPD. Med. Sci. Sports Exerc. 35 (1087): 1083–1087. Brischetto, M.J., Millman, R.P., Peterson, D.D. et al. (1984). Effect of aging on ventilatory response to exercise and CO2. J. Appl. Physiol. Respir. Environ. Exerc. Physiol. 56 (5): 1143–1150. Marsden, K.R., Haykowsky, M.J., Smirl, J.D. et al. (2012). Aging blunts hyperventilation‐induced hypocapnia and reduction in cerebral blood flow velocity during maximal exercise. Age 34 (3): 725–735. Wilkie, S.S., Guenette, J.A., Dominelli, P.B., and Sheel, A.W. (2012). Effects of an aging pulmonary system on expiratory flow limitation and dyspnoea during exercise in healthy women. Eur. J. Appl. Physiol. 112 (6): 2195–2204.

70 Boezen, H.M., Schouten, J.P., Postma, D.S., and Rijcken,

71

72

73

74

75

76

77

78

79

80

81

82

83

84

B. (1995). Relation between respiratory symptoms, pulmonary function and peak flow variability in adults. Thorax 50 (2): 121–126. Mahler, D.A., Ward, J., Waterman, L.A., and Baird, J.C. (2012). Longitudinal changes in patient‐reported dyspnea in patients with COPD. COPD. 9 (5): 522–527. Hardie, J.A., Vollmer, W.M., Buist, A.S. et al. (2005). Respiratory symptoms and obstructive pulmonary disease in a population aged over 70 years. Respir. Med. 99 (2): 186–195. Alexander‐Miller, S. and Davenport, P.W. (2010). Perception of multiple‐breath inspiratory resistive loads in males and females. Biol. Psychol. 84 (1): 147–149. Bushnell, M.C., Duncan, G.H., Hofbauer, R.K. et al. (1999). Pain perception: is there a role for primary somatosensory cortex? Proc. Natl. Acad. Sci. USA. 96 (14): 7705–7709. Van den Bergh, O., Van Diest, I., Dupont, L., and Davenport, P.W. (2012). On the psychology of cough. Lung 190 (1): 55–61. Thornby, M.A., Haas, F., and Axen, K. (1995). Effect of distractive auditory stimuli on exercise tolerance in patients with COPD. Chest 107 (5): 1213–1217. Bauldoff, G.S., Hoffman, L.A., Zullo, T.G., and Sciurba, F.C. (2002). Exercise maintenance following pulmonary rehabilitation: effect of distractive stimuli. Chest 122 (3): 948–954. von Leupoldt, A., Seemann, N., Gugleva, T., and Dahme, B. (2007). Attentional distraction reduces the affective but not the sensory dimension of perceived dyspnea. Respir. Med. 101 (4): 839–844. von Leupoldt, A., Vovk, A., Bradley, M.M. et al. (2010). The impact of emotion on respiratory‐ related evoked potentials. Psychophysiology 47 (3): 579–586. American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders, 4ee: Text Revision. Washington, DC: American Psychiatric Publishing, Inc., 1994. Van Diest, I., De Peuter, S., Eertmans, A. et al. (2005). Negative affectivity and enhanced symptom reports: differentiating between symptoms in men and women. Soc. Sci. Med. 61 (8): 1835–1845. Leivseth, L., Nilsen, T.I., Mai, X.M. et al. (2012). Lung function and anxiety in association with dyspnoea: the HUNT study. Respir. Med. 106 (8): 1148–1157. Pappens, M., Smets, E., Vansteenwegen, D. et al. (2012). Learning to fear suffocation: a new paradigm for interoceptive fear conditioning. Psychophysiology 49 (6): 821–828. Hasler, G., Gergen, P.J., Kleinbaum, D.G. et al. (2005). Asthma and panic in young adults: a 20‐year prospective community study. Am. J. Respir. Crit. Care Med. 171 (11): 1224–1230.

421

422

22  The Sensation of Breathing

85 Smoller, J.W., Simon, N.M., Pollack, M.H. et al. (1999).

86

87

88

89

90

91

92

93

94

95

96

97

98

99

Anxiety in patients with pulmonary disease: comorbidity and treatment. Semin. Clin. Neuropsychiatry 4 (2): 84–97. Van Diest, I., Janssens, T., Bogaerts, K. et al. (2009). Affective modulation of inspiratory motor drive. Psychophysiology 46 (1): 12–16. Barlow, D.H., Chorpita, B.F., and Turovsky, J. (1996). Fear, panic, anxiety, and disorders of emotion. Neb. Symp. Motiv. 43: 251–328. Rhudy, J.L. and Meagher, M.W. (2000). Fear and anxiety: divergent effects on human pain thresholds. Pain 84 (1): 65–75. Tsai, H.W., Chan, P.Y., von Leupoldt, A., and Davenport, P.W. (2013). The impact of emotion on the perception of graded magnitudes of respiratory resistive loads. Biol. Psychol. 93 (1): 220–224. Constantinou, E., Bogaerts, K., Van Diest, I., and Van den Bergh, O. (2013). Inducing symptoms in high symptom reporters via emotional pictures: the interactive effects of valence and arousal. J. Psychosom. Res. 74 (3): 191–196. Bogaerts, K., Janssens, T., De Peuter, S. et al. (2010). Negative affective pictures can elicit physical symptoms in high habitual symptom reporters. Psychol. Health 25 (6): 685–698. Bogaerts, K., Millen, A., Li, W. et al. (2008). High symptom reporters are less interoceptively accurate in a symptom‐ related context. J. Psychosom. Res. 65 (5): 417–424. Tiller, J., Pain, M., and Biddle, N. (1987). Anxiety disorder and perception of inspiratory resistive loads. Chest 91 (4): 547–551. Livermore, N., Butler, J.E., Sharpe, L. et al. (2008). Panic attacks and perception of inspiratory resistive loads in chronic obstructive pulmonary disease. Am. J. Respir. Crit. Care Med. 178 (1): 7–12. Giardino, N.D., Curtis, J.L., Abelson, J.L. et al. (2010). The impact of panic disorder on interoception and dyspnea reports in chronic obstructive pulmonary disease. Biol. Psychol. 84 (1): 142–146. Radomsky, A.S., Rachman, S., Thordarson, D.S. et al. (2001). The Claustrophobia Questionnaire. J. Anxiety Disord. 15 (4): 287–297. von Leupoldt, A., Chan, P.‐Y.S., Bradley, M.M. et al. (2011). The impact of anxiety on the neural processing of respiratory sensations. NeuroImage 55 (1): 247–252. Pappens, M., Van den Bergh, O., Vansteenwegen, D. et al. (2013). Learning to fear obstructed breathing: comparing interoceptive and exteroceptive cues. Biol. Psychol. 92 (1): 36–42. Petersen, S., Schroijen, M., Mölders, C. et al. (2014). Categorical interoception: perceptual organization of sensations from inside. Psychol. Sci. 25 (5): 1059–1066.

100 Van den Bergh, O., Stegen, K., and Van de Woestijne,

101

102

103

104

105

106

107

108

109

110

111

K.P. (1997). Learning to have psychosomatic complaints: conditioning of respiratory behavior and somatic complaints in psychosomatic patients. Psychosom. Med. 59 (1): 13–23. Leventhal, H., Breland, J.Y., Mora, P.A., and Leventhal, E.A. (2010). Lay representations of illness and treatment: a framework for action. In: Handbook of Behavioral Medicine (ed. A. Steptoe), 137–154. New York: Springer. Kaptein, A.A., Scharloo, M., Fischer, M.J. et al. (2008). Illness perceptions and COPD: an emerging field for COPD patient management. J. Asthma 45 (8): 625–629. Zoeckler, N., Kenn, K., Kuehl, K. et al. (2014). Illness perceptions predict exercise capacity and psychological well‐being after pulmonary rehabilitation in COPD patients. J. Psychosom. Res. 76 (2): 146–151. Fischer, M., Scharloo, M., Abbink, J. et al. (2010). The dynamics of illness perceptions: testing assumptions of Leventhal’s common‐sense model in a pulmonary rehabilitation setting. Br. J. Health Psychol. 15 (4): 887–903. Weldam, S.W., Lammers, J.W., Decates, R.L., and Schuurmans, M.J. (2013). Daily activities and health‐ related quality of life in patients with chronic obstructive pulmonary disease: psychological determinants: a cross‐sectional study. Health Qual. Life Outcomes 11 (1): 190. Fischer, J., Wimmer, A., and Mahlich, J. (2013). Medication adherence in asthma therapy‐‐a structured review. Pneumologie 67 (7): 406–414. Moss‐Morris, R., Weinman, J., Petrie, K. et al. (2002). The revised illness perception questionnaire (IPQ‐R). Psychol. Health 17 (1): 1–16. Petersen, S., van den Berg, R.A., Janssens, T., and Van den Bergh, O. (2011). Illness and symptom perception: a theoretical approach towards an integrative measurement model. Clin. Psychol. Rev. 31 (3): 428–439. Cameron, L., Leventhal, E.A., and Leventhal, H. (1993). Symptom representations and affect as determinants of care seeking in a community‐dwelling, adult sample population. Health Psychol. 12 (3): 171–179. Prior, L., Evans, M.R., and Prout, H. (2011). Talking about colds and flu: the lay diagnosis of two common illnesses among older British people. Soc. Sci. Med. 73 (6): 922–928. Petersen, S. and Van den Bergh, O. (2015). Relative breathless: comparison with others can affect feelings of breathlessness. Oral presentation at the 20th Annual Meeting of the International Society for the Advancement of Respiratory Psychophysiology in Leuven, Belgium (September 2013). Biol. Psych. 104: 156.

423

23 Breathing Function in Newborn Babies Urs P. Frey and Philipp Latzin Infant and preschool lung function testing is a new and rapidly growing field. This chapter describes lung function methods and the underlying neonatal physiology. Lung function and exercise in older children are reviewed in Chapters 25, 26, and 31.

CHAPTER MENU 23.1 23.2 23.3 23.4 23.5 23.6 23.7 23.8

Introduction, 423 Developmental Respiratory Physiology in Early Life,  423 Assessment of Lung Function in Neonates and Infants (Aged 0–2 Years),  424 Reference Values of Infant Lung Function,  428 Potential Role of Lung Function Testing in Infant Respiratory Disease,  429 Lung Function in Children Aged 2–6 Years,  429 Reference Values in Preschool Age,  432 Potential Role of Lung Function Testing in Preschool Respiratory Disease,  433 References, 433

23.1 ­Introduction There has been huge progress in the development of infant lung function testing in the last two decades. Problems related to measuring small changes in pressures and volumes and gas composition have been overcome by new sensor and computer technology. Large steps have been made in the development of devices enabling lung function measurements even in natural sleep without sedation. International efforts to define minimum technical standards for infant lung function devices with minimal instrumental dead‐space, resistance and impedance, high accuracy, and low signal‐to‐noise ratio have led to further development in the field [1–3]. Various measurement techniques were internationally standardised [3–9] and are regularly updated [10], now allowing multi‐centre trials using comparable protocols. Nevertheless, the implementation of such standards in commercially available equipment is still ongoing. Currently, high‐quality infant lung function testing is still restricted to a few specialised centres. Experience in infant lung function testing cannot be learned from textbooks, but usually requires several months (to years) of training in specialised centres to gain the experience

necessary for calibration, measurement, and quality‐ controlled analysis and interpretation of results. Reliable normative values of various techniques assessed under such standards are emerging, but are still scarce.

23.2 ­Developmental Respiratory Physiology in Early Life There is increasing awareness that many chronic adult respiratory diseases have their origin in early infancy [11]. A variety of genetic, epigenetic, and environmental factors, but also factors relating to lung injury and repair, have been associated with early lung disease and alteration in early lung function. Such factors include lung injury in prematurity, oxidative stress and inflammatory processes, early viral infections, or multi‐trigger wheezing disorders. These factors affect functional growth of the vulnerable infant lung, which can then track throughout life. During pregnancy, nutritional factors including vitamins (e.g. A, D [12]), tobacco exposure, and exposure to air pollution [13] have been associated with altered neonatal or childhood lung function. Similar associations have been described for exposure to postnatal indoor and outdoor air pollutants

Cotes’ Lung Function, Seventh Edition. Edited by Robert L. Maynard, Sarah J. Pearce, Benoit Nemery, Peter D. Wagner, and Brendan G. Cooper. © 2020 John Wiley & Sons Ltd. Published 2020 by John Wiley & Sons Ltd.

424

23  Breathing Function in Newborn Babies

(tobacco smoke, particulate matter, emissions from gas stoves, volatile organic components, and biomass and solid fuels) and lung growth later in childhood [14]; however, more conclusive evidence is still emerging. One particular challenge in paediatric lung function is the rapid change in mechanical properties of the infant lung and the changing ventilatory needs with age and maturation, even in the healthy lung. Many of these changes are gender dependent. While minute ventilation decreases in the first few months of life, lung volume doubles, the chest wall becomes stiffer, airway wall compliance decreases, and airway resistance decreases, leading to a decreasing flow limitation during early childhood. Airway size increases steadily in a fractal‐type manner. With fractal‐type growth, relative proportions of mother and daughter branches are maintained and the relation between the diameter and length of individual branches remains constant. During growth, flow through a given airway increases linearly with body size. This, along with the growth of the alveoli, implies that, theoretically, the deposition of small pollutant matter particles (40 weeks, and for measurements in the supine position during behaviourally defined quiet sleep. In very preterm infants, or during measurements in the intensive care unit, technical issues may become even more critical.

23.3  Assessment of Lung Function in Neonates and Infants (Aged 0–2 Years)

23.3.2  Tidal Breathing Parameters Pneumotachography. Ventilation minute volume and its subdivisions respiratory frequency and tidal volume are best measured using either a pneumotachograph or an ultrasonic flow meter connected to a face mask. A mask is usually simple to apply, well tolerated, and accurate within the limits of the calibration, provided the seal with the face is intact. The seal should be checked frequently. The method has the disadvantage that the presence of the mask or the head positioning may alter the pattern of breathing. The technical standards are described in detail [4]. Important technical issues are sufficient sampling rate, appropriate drift correction, BTPS corrections, and algorithms for breath detection. Respiratory inductance plethysmography (RIP). An inductance plethysmograph senses and interprets changes in the circumferences of the thorax and abdomen throughout the respiratory cycle. The device is arranged to fit snugly, but without restricting respiratory movements. Its main use is for monitoring the pattern of breathing. Where the pattern is stable, the device can be calibrated and then used to measure ventilation minute volume. Calibration entails measuring the redistribution of air between the thorax and abdomen that occurs when the airway is occluded. The procedure has been called qualitative diagnostic calibration (QDC) [17]. Calibration is achieved by measuring tidal volume during the procedure. The calibration is relative to the ventilation at the time and still dependent on the amount of compressible air in the lung, particularly in the presence of a high airway resistance or severe ventilation inhomogeneity. Reliable measurements can then be made for as long as the balance between the respiratory and abdominal components of breathing is maintained. The primary role of the technique is related to monitoring of hypopnoea, apnoea, and related acute life‐threatening events (designated ALTEs). Its diagnostic value can be increased when combined with electromyography, electrocardiography, and nasal flow or pressure measurements. Tidal breathing indices are usually derived from at least 20–30 breaths during quiet tidal breathing [4]. Breathing frequency (  f  ), tidal volume (Vt), and minute ventilation are considered to be markers of the ventilatory need of these infants. Parameters such as TI and TI/Ttot (total inspiratory time/total breath length) are strongly influenced by respiratory control. TPEF/TE (time to peak expiratory flow/total expiratory time) is influenced by obstructive and/or restrictive mechanical properties of the lung as well as by neuro‐respiratory control and the upper airways (e.g. grunting). The TPEF/TE is known to increase during the first year of life up to preschool age.

23.3.3  Regulation of Breathing, Novel Mathematical Methods Variability in breath‐by‐breath tidal breathing parameters is a function of measurement error, neuro‐respiratory control, and lung mechanics. Most measurement variations are random in nature. However, as was recently described, variations resulting from the interaction between lung and upper airway mechanics and the adaptive neuro‐respiratory control system are intrinsically correlated in a fractal‐type manner [17, 18]. These breath‐to‐breath correlations tend to be weak. However, newer mathematical methods, adapted from statistical physics, have been used to describe these so‐called long‐ range correlations with a single correlation exponent: α. In analogy, many other nonlinear parameters have been used to describe the behaviour of the respiratory system over time. Because of the highly adaptive capacity of infants to react to changes in mechanical properties of the lung, such new parameters have been used to characterise the overall abnormality of breathing in various disease states or developmental changes in neuro‐respiratory interactions of breathing. Of particular interest is the description of sighs [19, 20], which help to stretch the tissue and airway smooth muscles and recruit lung volume, in order to improve lung mechanics. 23.3.4  Measurement of Forced Expiratory Flow (Rapid Thoraco‐Abdominal Compression Techniques: RTC) Reliable forced expiratory manoeuvres are mainly performed under sedation, which clearly limits the ease of bedside clinical application of such techniques in severely ill patients. However, on the other hand, the reproducibility of measurements is probably one of the best of all infant lung function tests. Partial expiratory flow–volume curves. Tidal forced expirations [7] can be measured using the RTC in which pressure is applied to the trunk via an inflatable jacket. In an infant who is breathing spontaneously, the pressure is usually applied at the end of inspiration when the lung volume is the FRC plus tidal volume. The procedure then yields a partial expiratory flow–volume curve. It can be used for infants to identify the presence of obstruction to airflow (Figure 23.1). Since the maximal flow (flow limitation) is highly volume dependent, the reference flow is obtained at the end‐ expiratory level FRC, and thus labelled V′maxFRC. Particularly in pre‐term infants, the volume axis can be unstable, in which case the reproducibility of the flow is unsatisfactory. It can be improved by using the elastic equilibrium point as a volume landmark. This is done by preceding the forced expiration with inflations to briefly inhibit spontaneous breathing.

425

23  Breathing Function in Newborn Babies

The drawback, however, is the need for sedation and high technical skills. Although safety monitoring is recommended, unwanted effects such as gastro‐oesophageal reflux, pneumothorax, or respiratory insufficiency have not been observed in large samples [9].

900

600

Flow (ml s–1)

426

300 Vmax,FRC F= 0

0

–300 150

100

50 Volume (ml)

0

Forced inflation followed by tracheal suction. Infants can be assessed under fully controlled conditions if they are relaxed and ventilated via a tracheal catheter. The criteria for the investigation are necessarily strict. This is because the volume history is readily standardised; the attainment of total lung capacity is achieved by applying more negative pressure to the thorax than is practicable in the absence of intubation, and complete exhalation down to residual volume is ensured by suction applied to the endotracheal tube. The volume history is standardised by making three initial inflations, each followed by passive deflation before the full manoeuvre. In these circumstances, the reproducibility of the flow and volume indices is high. The interpretation is as for the previous method.

–50

Figure 23.1  Overlay of the tidal and raised volume forced expiratory flow–volume curves from the same infant. ───: raised volume flow–volume curve; ‐ ‐ ‐: partial flow–volume curve. V′max, functional residual capacity (FRC): maximal flow at FRC; F: flow. Source: European Respiratory Monograph 2006;37.

Lung inflation followed by compression. An even better way to standardise the lung volume at a given pressure from which the forced expiration occurs can be achieved by using the raised volume rapid thoraco‐abdominal compression technique (RV‐RTC) (Figure  23.1) [9]. In sedation, the infant’s lung is inflated several times above FRC. Apnoea occurs as a consequence, owing to a minimal drop in CO2 and the induction of the Hering– Breuer reflex. The latter is stronger at high lung volumes, leading to less muscular interference with the measurements. Standardised inflation pressure (2 or 3 kPa) may not achieve the same proportion of total lung capacity in all infants, but will depend on age and lung compliance. Nevertheless, the technique allows, for the first time, the tracking of flow limitation using measures comparable to those used in school age. Since forced expiration is terminated in 1.0 in highly trained cyclists.

without achieving a plateau. In this circumstance the extrapolation procedure should be repeated using the results for stage 2. The aerobic capacity is then either the observed or extrapolated value [2]. When reporting the result the method of derivation should be stated [28]. Additional information is given in Chapters 27 and 29.

28.4 ­Ergometry 28.4.1  Choice of Ergometer Cycle ergometry is now the preferred choice of both the American Thoracic and European Respiratory Societies [1, 5]. However, the defining criteria are administrative and at odds with the physiological evidence; this favours use of a treadmill because this form of exercise usually reproduces the symptoms experienced by a respiratory patient whereas cycle ergometry often does not (Table 28.6; also [24]). Cycle ergometry maybe more suitable for cardiac patients. It is appropriate for persons who cycle regularly including many children and those who undertake cycle ergometry as a means of keeping fit. 28.4.2 Treadmill Treadmill exercise simulates walking and running. These are familiar activities and can extend over a wide range of energy expenditure. The energy cost varies with the velocity and the angle of incline of the belt. When out of doors the energy expenditure also varies with the terrain. For any combination of these variables the energy cost is proportional to the subject’s mass including that of any load (Figures 28.3 and 28.4b). The energy cost is reduced by practice through the acquisition of a more economi­ cal style of locomotion [34]. Theoretically this could be a problem for serial measurements, but in practice no difficulty appears to have been reported. A treadmill was formerly driven passively by the subject. Now it is usually a continuous belt that is driven across a

Table 28.6  Relative merits of different types of ergometer.

a

Criterion

Treadmill

Cyclea

Step

Cost

Medium

Medium

Low

Space needed

Considerable

Modest

Modest

Safety requirements

Extra care

Normal care

Extra care

Convenience

Moderate

High

Low

Estimate of external work

Approximate [31]b

Precise

Precise

Symptoms at breaking point

Relevant to patient’s experience

Not relevant

Fairly relevant

 Results can be qualitatively different from those for treadmill exercise (Table 28.2).  Divergent views are expressed as to the clinical usefulness of this information!

b

557

28  Exercise Testing and Interpretation, Including Reference Values Walking velocity

5

5

6

7

8

(mph) 9

10

Gr

30

2

4

6

8

1.4

0

1.2

26

1.0

Figure 28.3  Graphical solution to the relationship of Givoni and Goldman describing the oxygen cost of treadmill walking per kilogram body mass:

(kph)

22

0.8

18

0.6

14

n O2



34

ad

1.6 Oxygen consumption per kg body mass (mmol min–1 kg–1)

4

ien t 24 (%) 22 20 18 16 14 12

3

4

3

Oxygen consumption per kg body mass (ml min–1 kg–1)

2

1.65 44.6 M 2.3 0.32 V 2.5 294 G 0.2 0.07 V 2.5 RSD 4.86 mmol min

10

50

100

150

Walking velocity (m min–1)

(b)

Rate of work (kp m min−1) 0 100

500

1000 Stepping

100

Walking (67 m min−1, 10% incline)

2 80

2.0

Cycling 80

60

Stepping (57 watts)

1

40

Cycling (74 watts)

60

1.5

20

0

50 100 150 Rate of work (watts)

40

80 Body mass (kg)

Consumption of oxgen (l min−1)

(a)

1

 2 is oxygen uptake in mmol min−1; η is the where nO terrain factor which for treadmill walking is 1.0, M is body mass (kg), G is gradient as percentage (100 × sine of angle of incline), and V is velocity (km h−1). To convert to oxygen uptake in traditional units (l min−1) divide by 44.6 and to kcal h−1 multiply by 294. Source: [31].

0.4

Consumption of oxgen (mmol min−1)

558

120

Figure 28.4  Oxygen cost of exercise for healthy men.(a) Relationships to rate of work. For cycling, the work is that done in overcoming the restraint applied to the flywheel. For walking and stepping, it is the product of body mass, the vertical movement of the centre of gravity of the body per step, and step frequency. The additional cost of stepping reflects that of restraining the body during stepping down (negative work). (b) Average relationships of the consumption of oxygen to body mass for subjects performing standardised exercise: walking at 67 m min−1 (2.5 mph) up an incline of 1 in 10, stepping at 57 W (350 kp m min−1) and cycling at 73.5 W (450 kp m min−1). Source: [32, 33].

28.4 Ergometry

Table 28.7  Progressive protocols for treadmill and cycle ergometry. Cycle ergometer

Treadmill

Healthy person

Patient

Healthy person

Patient

Test

Submax

Max

Symptom limited

Submax

Max

Symptom limited

Starting point

Rest – 20 W

W at RER = 1.0

Rest

3 kph

Incline ≥4°

2.0 kph

Incline 4°

Increment per minute

15 W

20 W

10 W

1 kph 1°a

1°

0.5 kph

1°

Endpoint

Rest = 1.0

 2 max no

 2 max (SL)a no

RER = 1.0

 2 max no

4 kph

 2 max (SL)b no

Source: [35]; see also Figure 28.2. a  From 4 kph onwards increments of velocity and incline can alternate. b  2 = 45 mmol min−1.  Exercise may also be terminated at no W = watts; kph = kilometres per hour; RER = respiratory exchange ratio; SL = symptom limited.

flat surface by an electric motor fitted with a continu­ ously variable gear. The minimal surface dimensions should be 2 m × 0.5 m and the belt speed 0–14 kph. The incline (i.e. the tangent of the angle to the hori­ zontal expressed as a percentage) should be capable of continuous adjustment from horizontal to at least 15%. The speed and incline indicators should be cali­ brated. The treadmill should be electrically earthed, and equipped with handrails, safety harness, or rear safety chain. It should also have the means for rapid deceleration, and be capable of being operated either automatically or by the subject or observer. There should be an adjacent platform to provided access and a gantry for supporting the mouthpiece or face mask. Agile subjects can be instructed to step onto the belt when it is moving using the handrails for support. Inexperienced or frail subjects should stand on the belt before it is set in motion, which should be done gradually. The subject should be instructed to walk as naturally as possible in an upright posture with head erect, eyes looking ahead, and initially taking relatively long steps. At the start one hand can rest on the rail, but once the predetermined speed has been achieved the arms should be allowed to swing. This encourages the subject to breathe naturally and reduces variability in oxygen uptake because no physical work is then done by the arms. At the end of exercise, either the belt is decelerated or the subject quickly gets off by gripping the handrails then placing the feet astride the belt. Exercise protocols for ergocardiography and assessing the physiological response to exercise are given, respectively, in Tables 28.4 and 28.7. The outcome of the test is usually in terms of oxygen uptake and other physiological indices (Table  28.8). However, for ergocardiography (e.g. Table 28.4) the time into the test or the stage reached can be used instead.

Table 28.8  Measurements for routine ergometry in respiratory patients. The procedures are outlined in this section. Measurements for non‐ergometric and field tests are in Section 28.9. Essential

Selected cases

Ventilation and gas exchange

Oxygen saturation

Cardiac frequency

Transfer factor v –V curve

Electrocardiogram (chest lead) Symptoms at termination

End‐expiratory lung volume

Fat‐free body mass

Systemic blood pressure

FEV1 and FVC

Cardiac output Pulmonary arterial pressure  diaphragmatic Other (e.g. V A/Q, EMG)

EMG, electromyography; FEV1, forced expiratory volume in 1 s; FVC, forced vital capacity.

For this to be representative of the subject’s exercise tol­ erance the treadmill should have been set precisely and the exercise performed in the recommended manner. 28.4.3  Cycle Ergometry A cycle ergometer is a stationary bicycle with the front wheel removed and the rear wheel restrained in a con­ trolled manner; the restraint is usually electromagnetic, formerly it was mechanical. The cycling rate should be displayed for the subject to see but a metronome should not be used because it can influence the respiratory fre­ quency. The loading should be capable of continuous vari­ ation from zero to the equivalent of 1 kW when the subject is cycling at 1 Hz. The work rate will usually be in watts but

559

28  Exercise Testing and Interpretation, Including Reference Values

Figure 28.5  Relationship of oxygen consumption to rate of work and body mass during cycling. A comparable relationship is given in Table 28.15. Source: [38]; see also [39].

Rate of work − cycling (kp m min−1) 400

600 2.0 Body mass (kg) 90 80 70 60 50

80

70

1.8 1.6 1.4

60

1.2 50 1.0 40 0.8 30 0.6 20

Consumption of oxygen (l min−1)

200

Consumption of oxgen (mmol min−1)

560

0.4

10

0.2

0

20

40 60 80 Rate of work − cycling (watts)

may be in kilopond metres per minute (see Footnote 28.1).1 The work rate meter should be calibrated [36, 37]. The restraint on the flywheel is usually controlled by a servomechanism set to maintain either a pre‐set steady work rate or one that increases by linear or loga­ rithmic increments. A setting for zero work can also be incorporated. The increases are usually either continuous (triangular or ramp protocol) or incremental each minute. The physiological responses are similar. Formerly 3 min increments were used on account of combining the features of progressive and nearly steady‐state exercise (staircase exercise). However, for studying responses to steady‐state exercise longer increments (e.g. 5 or 6 min) are necessary. The size of the increment will normally be in the range 10–30 W min–1 and chosen with a view to the subject achieving the chosen endpoint within 10 min. The endpoint will usually be a predetermined rate of work or the symptom‐limited maximum. The Footnote 28.1  One kilopond (kp) is the force acting on 1 kg at the normal acceleration of gravity, hence: 1 kpm min

1

9.81 / 60 0.1635W

where 9.81 is the gravitational acceleration in cgs units, 60 converts from second to minute, and W is rate of work in watts (also see Table 6.6).

100

120

former can constitute either a complete test or be the first stage of a maximal exercise test. The oxygen cost of cycling is influenced by the work done in moving the legs; this is a function of the mass of the legs and hence of body mass (Figure 28.5). The work also depends on the subject’s familiarity with cycling and with the cycling frequency that is adopted. The overall efficiency is the rate of external work relative to the total energy output as measured by direct calorimetry or esti­ mated from oxygen consumption (indirect calorimetry). The efficiency is approximately 25%. Cycle ergometry has the merits that the subject is seated throughout the test, the noise level is low, and the work rate can be specified. However, the last can be only an approximate guide to the energy expenditure for the reason given above. Many ergometers can be adapted for use with the arms. Cycling has the disad­ vantage that for most subjects it is an unfamiliar form of exercise and can cause undue fatigue [24, 40]. Except in trained cyclists the maximal oxygen uptake is less than when a treadmill is used; for most subjects the difference is approximately 10%. Responses to the two forms of exercise also differ in some other respects (Table 28.2).

28.5 Measurements

28.4.4  Stepping Exercise In the traditional step test the subject steps up onto and off a box, using both feet to a count of four, in time with a metronome [41]. The energy expenditure is  then linearly related to the work done; this will ­normally be the product of the height of the step, ­frequency of stepping, and body mass. However, the estimate of work is critically dependent on the subject raising his or her body mass up and down over the full height of the step and not reducing the rate of work by gripping a handrail. The test is normally performed at a single work level but can be adapted for progressive exercise [42]. The energy expenditure of stepping exercise is greater than that for cycle ergometry at the same rate of exter­ nal work (Figure 28.4). The difference is mainly due to the energy expended during stepping down (negative work). Stepping exercise has the disadvantages that the subject may stumble and the movement can dislodge the measuring equipment. The latter disadvantage does not apply to the Harvard pack test of exercise capacity (Section 28.9.4).

28.5 ­Measurements 28.5.1  What Should be Measured? The measurements should be appropriate to the test, the  subject (often a patient), and the local facilities (Table 28.8). Where facilities are limited the investigation Drying agent

CO2 analyser

of patients who are considered to be at moderately high risk might need to be undertaken elsewhere. 28.5.2  Overview of Equipment The equipment (Figure 28.6) measures the responses to exercise either over successive periods of 0.5 or 1 min or breath by breath. In most routine laboratories gas is sampled distal to a gas mixing chamber and analysed using instruments that respond relatively slowly, for example an infrared analyser for carbon dioxide and paramagnetic analyser for oxygen (Chapter  7). Using this approach the results are stable and the equipment is relatively inexpensive. Modern equipment often incorporates gas analysers that respond rapidly, for example a respiratory mass spectrometer. This provides nearly instantaneous analysis of expired gas leav­ ing the lungs, so can be used to provide integrated results for each breath; a gas mixing chamber is not required then. In some equipment the mass spectrometer is also used for other tests. An electrocardiogram is used to monitor the patient during the exercise (Table 28.3). Usually it also provides cardiac frequency. Alternatively, if a pulse oximeter is used to measure arterial oxygen saturation (Chapter  7) the frequency can be obtained concurrently. Other procedures that can be incorporated in the test include flow–volume or simple spirometry [43], arterial blood sampling, cardiac catheterisation, and diaphragmatic electromyography. However, these procedures introduce additional complexity and, depending on circumstances, may not be appropriate for a basic protocol.

O2 analyser

To air FO2 FCO2 fR ECG

Thermistor Mixing chamber Valve box Flow transducer

Vl From electrodes

Subject on ergometer

COMPUTER

PRINTER

Fresh air

Figure 28.6  Essential equipment for assessing the physiological response to exercise. For many assessments a pulse oximeter will be used as well.

561

562

28  Exercise Testing and Interpretation, Including Reference Values

28.5.3  Ventilation Minute Volume This will normally be measured using a pneumotachograph or rotating vane anemometer and an appropriate inte­ grator (Chapter 7). The sensor will also provide respiratory frequency. It should be located proximal to the inspira­ tory port of the valve box. The accuracy and resistance to flow, also the methods for calibration, should conform to current recommendations (Chapter  7). The valve box should have a small dead‐space (preferably 10–15 ml, maximum 50 ml). When measurements are over 30 s intervals the expiratory port is connected to a gas‐mixing chamber (capacity ~5 l) from which gas is sampled con­ tinuously for analysis. The ventilation signals should be in phase with those for gas analysis (see Breath‐by‐breath procedures). In older equipment, expired gas is collected in Douglas bags (Chapter  1). The method is now used mainly to check other equipment. Ventilation is reported in litres BTPS per min (V E) and per breath (Vt). The circumstances should be stated, for example maximal exercise, symptom‐limited exercise, or submaximal exercise. If it is the last, the index used should be defined (Section 27.4.3). 28.5.4  Gas Analysis This will usually be by a paramagnetic or zirconium cell analyser for oxygen and an infrared analyser for carbon dioxide. A respiratory mass spectrometer can also be used. With the former methods, water vapour in the expired gas is absorbed prior to the analyses. An allowance must then be made for the changes in partial pressures of the other gases. The corrected gas con­ centrations are used to calculate O2 uptake and CO2 output (Chapter 7). Breath‐by‐breath procedures. The essential features are that the response times for all the instruments are short and the resulting signals are synchronised to allow for differences in response times and in the time taken for the gas to reach the analysers (sampling delays; Chapter 7). It is helpful if gas analysis is by mass spec­ trometry as the signals for oxygen and carbon dioxide then have the same characteristics. Gas analysis should be accurate to within 1% [1, 5] and to achieve this the channels for oxygen and carbon dioxide should each be calibrated daily using three test gas mixtures of known composition. The use of two test mixtures is not ade­ quate [44] (Chapter 7). The fluctuations in gas flow and composition over the course of each breath should be allowed for by electrical integration of the individual signals. The variability between breaths can be reduced by the calculation of running averages over 5 or 10 breaths. Averaging over

time periods, for example 15 or 30 s, is less satisfactory. The method that is used should be indicated and its accuracy confirmed by comparison with the Douglas bag method [1]; this calibration should be repeated after the equipment has been moved [4] and in other circum­ stances at regular intervals with a log kept of the results. In addition, at weekly intervals a biological check should be carried out by a member of the laboratory staff [45]. The results should be displayed on a wall chart and any divergent results investigated. Results for consecutive whole breaths (obtained with­ out averaging) provide information about the ‘on’ and ‘off’ transients at the beginning and end of exercise. However, interpretation of transients is still not established [46]. 28.5.5  Other Measurements Cardiac frequency. This is best obtained from the elec­ trocardiogram (ECG) that is used to monitor the patient’s condition. The electrodes are placed over the upper ster­ num and cardiac apex (CM5 configuration, [25]), with an earth lead at the base of the sternum and another on the ergometer. The frequency is obtained from the R waves and its accuracy is checked by counting the radial pulse. Anthropometric measurements. The proportion of body mass which is fat and hence the fat‐free mass are obtained by measurements of body mass and subcuta­ neous fat. The latter is usually measured as skinfold thickness using skin calipers but other methods are also appropriate (Chapter 7). The observer should have received appropriate training and the instrument should have been calibrated. The information is necessary for interpretation of submaximal exercise cardiac frequency and maximal oxygen uptake and on this account should form part of the protocol for ergometry. Pulse oximetry. Arterial oxygen saturation is usually measured by a pulse oximeter applied to a finger or lobe of the ear. The part should be maximally vasodilated and the sensor should be secure to avoid movement relative to the skin (Chapter 7). A change in saturation, for example between rest and exercise, is measured with greater accuracy than an absolute level. Accuracy is best if the oximeter is cali­ brated using the subject’s own arterial blood. Blood pressure. Manual measurements are expensive in technician time so an automated method should be used where possible.  curve. The method is described in Flow–volume(v–V) Chapter 13 and the procedure for comparing the curve obtained during exercise with that over the same lung volumes at rest in Section 27.8.1.

28.6  Conduct of the Test Not at all breathless

Very, very breathless

10 cm

Figure 28.7  Visual analogue scale of breathlessness. The subject marks a 10 cm line at the point which reflects the intensity of breathlessness. The distance from the origin can be measured by hand; alternatively the scale can be a chain of lights controlled by a potentiometer. A result is given in Figure 27.17. Table 28.9  Scores and descriptions used in Borg scales. Perceived exertiona

Breathlessnessb

Score

Description

Score

Description

 7

Very, very light

 0

Nothing at all

 9

Very light

 0.5

Very, very slight (just noticeable)

11

Fairly light

1

Very slight

13

Somewhat hard

2

Slight

15

Hard

3

Moderate

17

Very hard

4

Somewhat severe

19

Very, very hard (the limit)

5

Severe

20

Beyond endurance

scales can be automated for use during, as well as at the end of, exercise [52].

28.6 ­Conduct of the Test The quality of the test result is influenced by the technical, interpersonal, and observational skills of the technician. This is particularly so when the procedure includes assessment of disability and is considered under this heading (Section 29.4.3).

28.5.6  Respiratory Symptoms

Safety. The laboratory should have a safety protocol that must be adhered to, a resuscitation trolley, and an alarm bell for summoning the resuscitation team. The trolley should be checked regularly to ensure that the defibrilla­ tor works, the valves of the resuscitator have not stuck, and the drugs are not time‐expired. Depending on the types of patient that are assessed the resuscitation equip­ ment may be used occasionally, rarely, or not at all. Consequently, the safety protocol should specify the training and checking procedures. Conformity with the protocol should be recorded and signed for by a respon­ sible person on a regular basis. More general considera­ tions are discussed in Chapter 29. The technician conducting the test should be compe­ tent to monitor the electrocardiogram, be alert to any symptoms that the subject may develop, and have access both to the controls of the treadmill if one is used and means for summoning assistance. He or she should have received training in electrocardiography and in cardi­ orespiratory resuscitation but need not be a physician. However, if the patient has any evidence of ischaemic heart disease a physician should be in close call. Some cardiovascular indications for terminating the exercise test are given in Table 28.3. Very often the presence of these abnormalities is a result of anxiety not disease, and do not reappear when the exercise in repeated.

The procedure for observing and recording the signs and symptoms that a patient experiences is described in Section  28.6. The principal symptom (usually breathlessness) is rated on a visual analogue scale [49] (Figure  28.7; also see Figure  27.17) or a Borg scale of perceived exertion or breathlessness (Table 28.9). The

Conditions for the test. The ventilatory and cardiovascu­ lar responses to exercise are influenced by the condition of the subject, environmental temperature, time of day, and time since the last meal. The measurements should normally be made at a neutral temperature (usually 17–23 °C), with the subject relaxed, in a post‐absorptive

6 7

Very severe

8 9 10

Very, very severe (almost maximal) Maximal

Source: [50, 51]. a  Score has a linear scale (0.1 × average cardiac frequency for that level of effort). b  Score has a proportional scale. A similar scale can be used for perceived exertion.

Transfer factor(Tl). The measurement on exercise is usu­ ally made during cycle ergometry. A single‐breath (breathholding) method (Chapter  18), with a shortened breathholding time (e.g. 6 s) is recommended [47]. The intra‐breath method can also be used [48]. Several determi­ nations are made, each at the end of a 5 min period of exercise of constant intensity. The result is interpolated to a standarduptakeofoxygen(usually45 mmol min−1,1.0 l min−1) or a standard cardiac frequency (usually 100 min−1[13]).

563

28  Exercise Testing and Interpretation, Including Reference Values

state, and not within 2 h of a heavy meal. To eliminate error due to carboxyhaemoglobin, the subject should be instructed not to have smoked or been otherwise exposed to carbon monoxide on the day of the test. The instruc­ tions should also detail loose clothing and shoes with soft soles and flat heels. Spectacles will normally be worn but false teeth removed. A clinical evaluation, anthropometric measurements, and assessment of lung function will normally precede the exercise test. The assessment should include measurement of ventilatory capacity and transfer factor. In addition, depending on the clinical history, there may be a need for a 12‐lead electrocardiogram. However, cardiovascular dis­ ease, unless it is recent, unstable, or very advanced, should not preclude an exercise test if the patient is ambulant and there is a clear respiratory indication [1]. Where there is doubt a cardiological opinion should be sought. Preliminaries. The technician will ensure that the equip­ ment is in good order, with appropriate parts sterile, toi­ letries to hand, and the instruments calibrated. He or she should study the request form, have a good idea of the clinical features, check that these have not deteriorated recently or contraindicate exercise, and establish the rea­ sons for the test. If an inhaler is used the time of the last dose should be noted and, if appropriate, repeated. Spirometry may need to be repeated and measurements made of body mass and skinfold thickness. The subject should be instructed on how to perform the exercise, how to communicate symptoms to the laboratory staff, and how to discontinue exercise in the event of need­ ing to do so suddenly (Sections 28.4.2 and 28.4.3). If practi­ cable, the instruction should be given on a preliminary visit during which the subject tries out the equipment. During the test. The technician’s prime responsibility is to keep a close eye on the subject, provide instruction and encouragement as appropriate, and, if the subject does not do so, decide when the exercise should be ter­ minated. The decision will be influenced mainly by interaction with the subject and also from the displays of the measuring instruments. In high‐risk cases the latter should be monitored by a second observer. At the end of exercise the technician should immedi­ ately remove the mouthpiece, quickly seat the subject in a chair (if not already seated), ask the reasons for stopping, and observe the extent of breathlessness before and during the reply. Subsequently, the subject should rate the stated symptom using an appropriate scale (Table 28.9). This procedure is necessary because when exercise is limited by symptoms their nature and severity are often the most important outcomes of the test. Accuracy is lost if the information is only obtained after a period of ‘winding down’.

28.7 ­Data Processing A progressive exercise test yields an enormous amount of information. In modern equipment signals are processed on‐line to obtain flows, volumes, gas concentrations, and frequencies, and then ventilation, tidal volume, respira­ tory and cardiac frequencies, oxygen uptake and carbon dioxide output, and respiratory exchange ratio (RER). Each of these steps involves the use of a mathematical model with coefficient and constant terms (parameters) of which some are general (as in Chapter 6) whilst others are specific to the equipment. Before the equipment is brought into use and again after any items have been replaced the models and parameters should be checked carefully as any errors will cause faults in the data that can seldom be corrected retrospectively (Chapter 7). Next, selected indices are related to each other as x/y plots (e.g. Figure 28.1; also Figure 27.5). The plots should be scrutinised immediately so the quality of the informa­ tion can be assessed and a decision made as to whether the result is acceptable or if the test should be repeated. Factors to be taken into account include evidence for anxiety such as initial hyperventilation or tachycardia (Figure  28.8), more than one ventricular ectopic beat, and fluctuations in any of the measured variables. The fluctuations are usually due to a leaking mouthpiece or poor electrode contact with the skin. After the data have been checked the physiological response is reviewed. In one commonly used system of exercise testing the review is based on nine graphs, of which three have two sets of axes, making 12 graphs in all [38]. Some of the graphs are included to permit esti­ mation of the anaerobic threshold (see Table  27.3).

Terminal rise (for VE) fC or VE

564

Too high initial values

Concavity

Aberrant point Intercept fC > 90 min−1 VE > 10 l min−1 nO2

Figure 28.8  Example of a fault that should be detected by scrutiny of an intermediate result. In most instances the exercise should be repeated, but it may be sufficient to exclude one or two points.

28.8  Interpretation of Data

However, the indirect method is not valid for patients with respiratory diseases (Chapter 27) and some graphs can be omitted on this account. Even so, interpretation of such a large quantity of information is not easy and the results cannot readily be incorporated into a lung func­ tion report. In the preceding chapter a case is made for confining the result of a basic exercise test to the symp­ toms and a limited number of physiological indices (e.g. Table 28.10): these are ones that can be defined without ambiguity and for which reference values are available. Ventilation equivalent and oxygen pulse do not meet these criteria.

28.8 ­Interpretation of Data 28.8.1  Submaximal Exercise Where the patient can achieve an oxygen uptake of at least 33 mmol min−1 (0.75 l min−1) the data can be used to explore the reason for an increased exercise ventilation. The exploration is best undertaken one step at a time. Step 1. Ventilation. Establish if the exercise ventilation is within the normal range (e.g. for V E45 in men 19–27 l min−1[53]). If it is then, subject to the symptoms and level of ventilatory capacity, there is unlikely to be a respiratory problem. If the ventilation is abnormal, can it be brought within the normal range by making allowance for the subject’s RER and Vt30? If ‘yes’ any abnormality is unlikely to reside in the lungs. If ‘no’ this may be because the adjustments for RER and Vt30 were inappropriate in the circumstances, alternatively the lungs are responding abnormally. The cause for the increased exercise venti­ lation should now be explored.

Step 2. RER and Sa,O2. Inspection of these variables can point to (but not confirm) the likely mechanism for an increased ventilation. If RER is normal (e.g. RER45 ≤ 0.91) most of the extra ventilation will be to parts of the lungs that are not making a big contribution to gas exchange  ratio). This site will usually be alveoli that (high VA/Qs are poorly perfused (alveolar dead‐space). However, if breathing is shallow and rapid (low Vt30) there will be additional ventilation of the airway dead‐space as well. If RER is increased, much of the excess ventilation is of alveoli that are making a material contribution to gas exchange. The commonest cause is lactacidaemia. The increased RER is then likely to reflect either that the work level is high for that subject or he or she is taking relatively little exercise for social reasons (i.e. is unfit) or there is a respiratory, cardiac, or other medical condi­ tion. In many of these patients the Sa,O2 will not have deviated much from the initial level. If the saturation rises the likely causes include uneven lung function and functional overbreathing, possibly superimposed on an organic disorder. If the saturation falls there may be a defect of gas transfer. In this event the transfer factor is likely to be reduced at rest and may fall further when measured during exercise. The common causes include generalised emphysema and alveolitis or diffuse inter­ stitial fibrosis. If the patient has airway obstruction, the features represent a type 1 and not a type 2 response (patient pink and puffing, not potentially blue and bloated (Chapter  36). If the abnormality is mainly in the lung parenchyma it might fit the somewhat inappropriate designation of ‘alveolar capillary block syndrome’. How­ ever, these syndromes develop progressively and may not be prominent in their early stages. Step 3. Pattern of breathing. In the present scheme, the pattern is interpreted in terms of Vt30 and fR30 (for which purpose the ventilation minute volume needs to

Table 28.10  Data recommended for use in the report on a routine exercise test.

a

Submaximal exercise

At symptom limitation

Basic information

Age, fat‐free mass, Fat% (or height and weight), FVC

Ditto, also grade of habitual activity, smoking status, and FEV1

Ventilationa Respiratory exchange ratio Cardiac frequency

At specified O2 uptake (e.g. 45 mmol min−1, i.e. 1.0 l min−1 in men, possibly less in women) −1

Maximal attained values for all variables

Tidal volume

At ventilation 30 l min

Shape of Hey plot

Saturation by oximetry

Per cent before, during, and at end of exercise

Ditto

Symptoms

Score on visual analogue scale

Comment and scores describing condition

 Measurement of flow–volume curve and end‐expiratory lung volume should be considered for specialist and research laboratories. FEV1, forced expiratory volume in 1 s; FVC, forced vital capacity.

565

566

28  Exercise Testing and Interpretation, Including Reference Values

be at least 30 l min−1). An alternative approach is to consider instead the times of inspiration and expira­ tion and the relationship between them (Section 27.4.5). If Vt30 is normal with respect to the reference value (based on forced vital capacity) the pattern of breathing is unremarkable, but if it is abnormal (either high or low) the cause should be identified (Section 27.4.7). In a person with normal lungs a high Vt30 often reflects a high level of physical fitness now or in the past. In other circumstances it can be associated with asthma in remission or labile airway obstruction. A Vt30 that is reduced relative to vital capacity can be the principal or a subsidiary cause of an increased ventilation leading to breathlessness on exertion. The mechanism is likely to be mechanical, reflex, or functional. Dynamic shal­ low breathing can result from pleural thickening or excess soft tissue on the anterior thorax. Reflex shallow breathing (tachypnoea) arises from receptors in the pulmonary circulation or lung parenchyma (J‐recep­ tors). Functional shallow breathing is a feature of anxi­ ety and malingering. In all these circumstances the subject is likely to be worried by the symptom, which in the case of respiratory patients is often out of propor­ tion to the deterioration in lung function. Some clinical examples are in Tables  28.11 and 28.12; also Figures 27.11 and 27.12. The breathing pattern can be explored further if infor­ mation is available on the relationship of forced expira­ tory flow to lung volume during exercise. Shallow breathing accompanying an increase in end‐expiratory lung volume (EELV) and displacement of the flow–vol­ ume loop in the direction of total lung capacity (see Figure 27.18) are evidence for exercise being limited by respiratory factors; see also the following step. Step 4. Cardiac frequency. The fC at the chosen uptake of oxygen should be compared with that expected for the subject’s muscularity as assessed by fat‐free mass or related variable (Figure  28.11). In a respiratory patient the frequency will usually be relatively normal (see Tables  27.6 and 27.7). A high cardiac frequency may indicate that the patient is unfit, experiencing hypoxae­ mia or toxaemia, or is anxious. Alternatively, it may be a consequence of coexisting cardiac disease. 28.8.2  Exercise Limitation The cause of limitation will be included in the medical report. It is usually either mainly respiratory or mainly circulatory in origin. Other causes are skeleto‐muscular difficulties, anaemia, toxaemia, and portal cirrhosis. In addition, a possible contribution from apprehension, anxiety, or malingering should be borne in mind. Thus, the scrutiny of the data (Table 28.10) should be broadly

based and follow that for submaximal exercise reviewed above. Step 1. Review the symptoms for clues as to the mode of limitation, including: (i) the operator’s impression of the condition of the subject at the point of termination, (ii) the patient’s symptoms, and (iii) the scores for breath­ lessness and effort. If the limitation was likely to have been of cardiopulmonary origin, the scrutiny should indicate if the subject experienced incapacitating breath­ lessness. If so, the physiological response was probably maximal. If the recovery appeared to be slow and the patient was obviously fatigued, the limitation might have been cardiovascular. Step 2. Look for evidence of cardiovascular limitation. If this was the case the maximal observed cardiac frequency should be similar to its reference value (see Eq. 27.2). Information to this effect might already have been obtained from the electrocardiogram. In addition, there might be evidence for a raised blood concentration of lac­ tic acid, including a progressively increasing respiratory exchange ratio towards or above unity, slow recovery of ventilation after exercise (suggesting an oxygen debt), and a high blood concentration of lactic acid (Section 27.4.6). Step 3. If the limitation was not cardiovascular, was there evidence for ventilatory limitation, e.g. did the exercise ventilation approximate to the maximum expected for the level of the forced expiratory volume in 1 s (FEV1) (see Figure 27.13) (dyspnoeic index near to or in excess of 100%)? Did the flow–volume curve show evidence for exercise limitation (see Figure 27.18)? Step 4. Consider what might be learned from the pattern of breathing at near the breaking point of exercise. Here a characteristic pattern is of the terminal part of the Hey plot not being vertical, but inclined to the left (Figure  28.9). This can indicate progressive shallow breathing associated with air trapping and a rise in EELV at near to the breaking point of exercise. The sequence suggests ventilatory insufficiency. Many of the changes that have been indicated in this section can be demonstrated more elegantly in other ways but to do so require additional manoeuvres (e.g. [11, 43]) that the patient may not be able to perform (see Footnote 28.2).2 Practical outcome. Once the nature of the physiological response to submaximal exercise and the cause of Footnote 28.2  A cost–benefit analysis into the resource implications, success rates, and clinical utility of the alternative procedures might be helpful.

Table 28.11  Examples of exercise test results (post bronchodilator, using treadmill) in men. Diagnosis

COPD

Emphysema

Asthma

Functional overbreathing

Patient details

Fitter (smoker) with irreversible airflow limitation. TLC and T l both normal

Furnace operator (smoker) with airway obstruction, large TLC, high compliance and low T l

Baker with asthma, apparently controlled by inhaled salbutamol (albuterol). TLC and T1 both normal

Coalminer with pneumoconiosis and undue breathlessness. Normal lung function

57

39

52

Age (years)

62

FEV1 (l)

1.17

Clinical grade (BMRC)a

4 (750 m)

V E45(l min−1)b

38

(24)

46

(24)

20

(24)

61

(24)

RER

0.77

(0.79)

0.89

(0.79)

0.76

(0.79)

1.0

(0.79)

Vt30 (l)c

0.81

(1.1)

1.75

(1.45)

2.43

(1.4)

0.73

(1.3)

fC45 (min–1)b

104

(108)

103

(111)

103

(105)

98

(115)

Sa,O2

(2.69)

1.48

(2.89)

5 (40 m)

No desaturation

3.75

(3.69)

2

(2.99)

4

No desaturation

No desaturation

No desaturation

Intense breathlessness (dyspnoeic index 104%)

Tightness of the chest associated with fall in FEV1 to 3.0 l

Dizziness

large alveolar dead‐ space. Normal airway dead‐space and alveolar ventilations (Vt30 and RER both normal)

Deep breathing (high Vt30) and mild hyperventilation (RER 0.89) maintained a normal saturation (hence ‘pink puffer’)

Deep breathing (high Vt30) contributed to a normal response to submaximal exercise. The patient had exercise‐induced airflow limitation

Obesity

Pleural thickening

Pulmonary fibrosis

Portal cirrhosis

Patient details

Lagger who was breathless. He had normal lung function but was overweight (BMI 31 kg m−2)

Shipwright (smoker) who was breathless despite normal lung function and quantity of body fat

Fitter (smoker) with reduced TLC, compliance, and T1. No airway obstruction

Lorry driver with low transfer factor and mild airway obstruction. Normal TLC

Age (year)

53

60

70

41

Limitation to exercise Comment

Diagnosis

Breathlessness (V Emax 48 l min−1), dyspnoeic index 113%

V E45 increased by a

FEV1 (l)

4.2

Clinical grade (BMRC)

3

(3.60)

3.21

(3.13)

4 (375 m)

2.0

(2.81)

4(375 m)

The high V E45 was due to increased airway dead‐space ventilation low Vt30 (raised fR) plus hyperventilation (high RER)

2.8

(3.87)

4

V E45 (l min−1)

24.7

(24)

50

(24)

38

(24)

60

(24)

RER

0.61

(0.79)

0.83

(0.79)

0.95

(0.79)

1.13

(0.79)

Vt30 (l)

0.8

(1.52)

1.03

(1.43)

1.02

(1.11)

2.1

(1.4)

101

(90)

88

(102)

100

(111)

117

(94)

–1

fC45 (min ) Sa,O2

No desaturation

No desaturation

96 → 88 → 95

92–79

Limitation to exercise:

Exercise ‘somewhat hard’

Stopped because of breathlessness

Exercise limited by fatigue and moderate breathlessness

Patient extremely breathless

Raised ventilation and hence breathlessness due to both shallow breathing and alveolar hyperventilation

Ventilation increased by large alveolar dead‐ space(VA/Q), increased hypoxic drive and? other factors

Gross hyperventilation (also tachycardia) secondary to hypoxaemia resulting from transfer defect. Abnormalities were corrected by breathing oxygen

Comment:

a

Normal V E45 due to alveolar hypoventilation compensating for increased ventilation of airway dead‐space. Breathlessness possibly due to hypercapnia and shallow breathing

 British Medical Research Council. For breathless grades, see Table 28.16. Stated walking distance is in brackets.  45 mmol min−1 is 1.0 l min−1. c  Tidal volume at a ventilation of 30 l min−1. Reference values for exercise indices are in parentheses. COPD, chronic obstructive pulmonary disease. For definitions of other abbreviations, see Chapter 6. b

2.44

28  Exercise Testing and Interpretation, Including Reference Values

Table 28.12  Examples of exercise test results (post bronchodilator) in women. Functional hyperventilation

COPD

Pleural thickening

Diffuse interstitial fibrosis

Asbestos worker (smoker) with some airway obstruction and reduced transfer factor

Reversible airway obstruction in former asbestos worker. Lung function was otherwise normal

Previous asbestos worker with COPD and rather small total lung capacity. Normal transfer factor

Previous asbestos worker (smoker) with small lungs and low transfer factor

Age (years)

44

FEV1 (l)

1.75

Clinical grade (BMRC)a

3

V E22(l min−1)b

36

(14)

15.5

(14)

18

(14)

24

(14)

RER

1.28

(75% (absolute) for FEV% (FEV1/FVC)

Slight Moderate Severe

Not normal but >60% In range 59–40% (50% for FVC) 0.85)

1.07 (>0.85)

Mild airway obstruction, dynamic restriction to lung expansion

Material airway obstruction, not asbestosis

Transfer defect

38 (24)

32 (24)

30 (24)

Lung function: FEV1 (l)

−1

−1

Comment Response to exercise V E45 (l min−1)a RER45

0.76 (0.85)

0.88 (0.85)

0.78 (0.85)

Sa,O2(%)

95 → 92 → 97

94 → 91 → 96

98 → 93 → 97

Vt30 (l)

0.71 (1.3)

0.90 (1.2)

1.1 (1.4)

fC45a

103 (100)

111 (126)

95 (113)

63 (104) … 1.41 (2.33)

44 (79) … 0.99 (1.77)

72 (101) … 1.61 (2.26)

(76) … (1.70)

(51) … (1.14)

(73) … (1.63)

 2max (mmol min−1) V O2max nO (l min−1) observed (with reference value) Reference less 1.64 RSD (same units) b

Estimated respiration limited max.

62 … 1.39

49 … 1.10

85 … 1.90

Symptom at termination

Struggling to breathe

Slight breathlessness

Legs tired (quite breathless)

Self‐reported effort

Hard

Somewhat hard

Hard

Comment

Breathing very shallow

Not ventilation limited

Not ventilation limited

Respiratory disability Observedc Estimated

d

Did not meet criteria 24% …

7%

Did not meet criteria

41%

29%

Nonee

(confirms observed result)  Ventilation and cardiac frequency at V O2 = 1.0 l min−1.  Calculated as 66.4 + 13.4 FEV1 (l) –0.94 V E45 (l min−1) + 0.45 FFM –0.31 age (year) (see Table 28.15). c  100 (76–63)/(76–22) and 100 (1.7–1.41)/(1.7–1.0) where 22 and 1.0 are 100% disability [2, 4] (see also Table 29.7). d  Calculation as in a but using estimated value for maximal O2 uptake. e  Respiration limited maximal O2 uptake exceeds (reference – 1.64 RSD). For definitions of abbreviations, see Chapter 6. a b

29.5  Interpreting the Exercise Test

pain, or any of a number of other symptoms of which the ­commonest is tiredness. Amongst applicants with presumed respiratory disability who have been referred for assessment with a view to compensation, the proportion who actually meet the criteria can be of the order of 20% [5, 26, 27]. Other subjects may have a respiratory component to their disability. If the limitation is respiratory the appropriateness of the observed maximal oxygen uptake can sometimes be checked using submaximal data obtained during the course of the test (subject A, Table 29.6). Alternatively, the method can be used to estimate the respiratory component of limitation when there is underachievement (subject B, in whom the respiratory component was estimated as 29%) or another factor is dominant (subject C). Additional problems arise if there is a possibility of malingering (Section 29.5.3). 29.5.2  Scoring Loss of Exercise Capacity (Disability) The ‘disability’ can be scored on a linear scale. It is zero when the maximal recorded oxygen uptake is at the lower limit of normal (i.e. reference value minus 1.64 RSD). The disability is 100% when the exercise capacity is at a defined minimal level [28]. For the ERS the level is twice the ­resting level (2 metabolic equivalents or METs, where 1 MET is an oxygen uptake of 3.5 ml kg−1  min−1 or its equivalent in SI units). It reflects that the person is no longer able to live semi‐independently. For the ATS 100% disability is set at 4 METs [2] and the person is no longer considered able to undertake work away from home. The conventions used for defining the limits affect the o ­ utcome (e.g. Table  29.6) and should be stated. The percentage ­disability can be represented on a diagram (Figure 29.4). The outcome is the overall disability; it should be

Rating scale for respiratory disability Rating (%)

Range of values for nO2 max

0

Lower limit of normal (pred. − 1.64 SD)

2 or 4 METs (ERS and ATS respectively)

c­ ompatible with the stated exercise capacity (Table 29.4). Where the limitation is respiratory the symptom‐limited maximal oxygen uptake will be similar to that estimated  from submaximal data (footnote to Table  29.7). Alternatively, if the estimated value materially exceeds the observed value the former can be used to indicate the approximate limitation from respiratory factors. In one study, the method of rating was validated by comparing the objective scores for disability based on treadmill exercise with those obtained independently by an experienced clinician who based his score on clinical grade of breathlessness, forced expiratory volume, and the appearance of the chest radiograph. A significant correlation was observed (R2 = 0.5, [28]). This result not only indicated a 50% overlap in the information obtained by the two methods but also important differences that arose from the clinician using radiographic as well as functional criteria. Ideally the accuracy of the estimated disability should be indicated by the provision of confidence intervals, but these appear not to be available. Alternatively, the ratings can be grouped into grades (Table  29.8). Grades based on  the measured exercise capacity are inevitably more

LLN – subject’s measured max. O2 uptake 100× LLN – O2 uptake at 100% disability

0% (none)

Maximal O2 uptake exceeds LLN

100% (total)

Maximal O2 uptake less than uptake at 100% disability (2 or 4 METs as appropriate)

Reference value for max O2 uptake based on

Age, level of habitual activity, fat‐free mass (FFM), smoker (yes/no) Respiration‐limited V O2max is estimated from: FEV1, V E(submax. ex.), age, and FFM (Table 28.15).

If mixed aetiology:

100

Figure 29.4  Diagram indicating the derivation of the percentage disability. For definitions of abbreviations, see text. Source: [29].

Table 29.7  Calculation of percentage disability from the observed maximal O2 uptake, the lower limit of normal (predicted less 1.64 RSD, designated lower limit of normal [LLN]), and the criterion used for 100% disability (see text)a.

% disability

50

Source: [28]. a  For disability to be respiratory the exercise capacity must be limited by respiratory impairment (Table 29.5).

583

584

29  Assessment of Exercise Limitation, Disability, and Residual Ability

Table 29.8  Grades of respiratory disability: distribution amongst 157 men who met the criteria for respiratory limitation of exercise [5]. % Disability

Grade

Description

Number of men

0

0

None

44

1–39

1

Slight

59

40–59

2

Moderate

40

60–100

3

Severe

14

Table 29.9  Some types of non‐cooperation (deliberate cheating is rare). Spirometry

Inhaler not used Bronchoconstriction induced Previous inspiration not completed Expiration impeded, leaked, or restrained

Transfer factor exercise test

Valsalva manoeuvre performed during test State of agitation induced Ventilation unduly deep or shallow Endpoint contrived Symptoms misrepresented

Detection

Observation, repetition, checks for internal inconsistencies in the physiological indices

reliable than those extrapolated from the lung function (Tables 29.2 and 29.7). 29.5.3 Underperformance In most physicians’ experience the misrepresentation of symptoms is relatively common, especially in the setting of a tribunal, but there may be mitigating factors (Table  29.9). Submaximal effort, hyperventilation, and features of anxiety are also quite common, as is failure to take medication effectively, especially bronchodilators. To avoid the latter, a subject who has been prescribed an inhaler should normally use it in the presence of the technician, prior to the assessment. Subtle manipulation of the performance of physiological tests is rare. Some features that should be watched for are listed in Table 29.9. The principal safeguards are to adhere to protocols, bear in mind that they can be manipulated, and look for inconsistencies in the results.

29.6 ­Residual Ability Residual ability is important for daily living but is not the converse of disability. This is partly because exercise capacity decreases with age. Hence, in healthy persons without ‘disability’, an activity that was easy in youth can become impossible by old age [30] (see Figure 25.3).

In addition, the ability to undertake exercise has different determinants depending on its duration and whether the  mode of limitation is respiratory, from which ­recovery is rapid, or cardiovascular, where recovery is slow (Section 28.8), or a mixture of the two. For whole body exercise that is limited by cardiovascular factors the work level that can be sustained for a period of hours is approximately 40% of maximal oxygen uptake [31]. Hence, this is a useful threshold against which to compare the energy cost of any particular exercise (see Figure 28.1). If the exercise entails raising the centre of gravity of the body, as in walking (Section 28.2), the energy cost is proportional to body mass. This has led the ATS to categorise sustainable work in terms of oxygen uptake per kilogram of body mass [2]. The threshold for a person required to perform a range of physical tasks is set at an oxygen uptake of 15 ml min−1 kg−1, equivalent to an oxygen uptake of 1.051 min−1 for a person weighing 70 kg. For the work to be sustainable this level of V O2 should be less than 40% of maximal oxygen uptake, hence V O2max should exceed 37.5 ml min−1  kg−1 (i.e. 15 × 100/40), equivalent to a maximal oxygen uptake of 2.6  l  min−1 for a man weighing 70 kg. On a superficial view the criterion seems reasonable. However, had it been applied to men in regular employment as welders and other tradesmen in a British shipyard, more than one‐third would have been excluded as having an insufficient exercise c­ apacity (Figure 29.5). The discrepancy was due to V O2max per kilogram body mass not being independent of mass (Section 28.1.1) and to heavy shipyard tasks being performed intermittently. Thus, the method is not satisfactory even when the limitation to exercise is cardiovascular. It is also not applicable to respiratory patients in whom exercise is limited by breathlessness. In both sets of circumstances the ability to undertake a particular task can best be  assessed by the subject trying it to see if he or she can  cope. However, sometimes this information will have  been obtained during the preliminary assessment (Section 30.3.1).

29.7 ­Relevance for Compensation In respiratory patients the extent of exercise limitation (disability) is an important factor in awarding compensation. Another consideration is the likely prognosis. This is related to the loss of lung function both at one point in time and longitudinally. Thus, for purposes of compensation the response to exercise and the lung function make inter‐related but partly separate contributions to the outcome. The two components of the assessment can also contribute to medical treatment and subsequent

Figure 29.5  Irrelevance of rigid criteria of suitability for strenuous work. The data points relate 40% V O2max to body mass for tradesmen in a shipyard [24]. As in other studies [32], the points are negatively correlated with body mass. The horizontal lines are arbitrary boundaries between different categories of task [2]. The boundaries are unrelated to the ability of the men to perform their work. Source: [33].

40% VO2max per kg body mass (ml min–1 kg–1)

References 30 No limitation 25 Can sustain most activities

20 15 10

Uncomfortable getting to work

5 0 50

clinical management, so a holistic approach has practical utility as well as contributing to the quality of the measurements (Section 30.5.3).

29.8 ­Summary The assessment of respiratory limitation of exercise (­respiratory disability) poses a challenge to the respiratory team as it combines a need for high professional and technical standards with the vigilance of a good detective. The procedures include clinical appraisal, assessment of lung function especially dynamic spirometry and transfer

60

70

80 90 100 Body mass (kg)

110

120

130

f­actor, and, where appropriate, the performance of a symptom‐limited maximal exercise test. The exercise should be of a type that the subject performs regularly. For most subjects this entails walking using a treadmill and not cycling using a stationary bicycle. From the results the percentage and grade of disability and the proportion attributable to respiratory insufficiency can be calculated. The information necessary for the assessment can also be used to make allowance for coexisting cardiovascular disease and, in some instances, to overcome the effects of incomplete cooperation by the subject. The assessment of residual ability can make use of the same information, but this topic is best approached empirically.

References 1 World Health Organisation (1980). International

Classification of Impairments, Disabilities and Handicaps. Geneva: WHO. 2 American Thoracic Society (1986). Evaluation of impairment/disability secondary to respiratory disorders. Am. Rev. Respir. Dis. 133: 1205–1209. 3 World Health Organisation (1999). International Classification of Functioning and Disability, Beta‐2 Draft. Geneva: WHO. 4 De Coster, A. (1983). Respiratory impairment and disablement. Bull. Eur. Physiol. Pathol. Respir. 19: 1P–3P. 5 Cotes, J.E., Zejda, J., and King, B. (1988). Lung function impairment as a guide to exercise limitation in work‐ related lung disorders. Am. Rev. Respir. Dis. 137: 1089–1093. 6 O’Donnell, D.E., Lam, M., and Webb, K.A. (1998). Measurement of symptoms, lung hyperinflation, and endurance during exercise in chronic obstructive pulmonary disease. Am. J. Respir. Crit. Care Med. 158: 1557–1565.

7 Ross, R.M. (2003). ATS/ACCP statement on

cardiopulmonary exercise testing. Am. J. Respir. Crit. Care Med. 167: 211–277. 8 Smith, D.D. (1995). Pulmonary impairment/disability evaluation: controversies and criticisms. Clin. Pulm. Med. 2: 334–343. 9 Cotes, J.E. and Steel, J. (1987). Work‐Related Lung Disorders. Oxford: Blackwell Scientific. 10 Morgan, A.D., Peck, D.F., Buchanan, D.R., and McHardy, G.J.R. (1983). Effects of attitudes and beliefs on exercise tolerance in chronic bronchitis. Br. Med. J 286: 171–173. 11 Sprake, C.M., Cotes, J.E., and Reed, J.W. (1984). Correlates of 6 min walking distance and maximal oxygen uptake in chronic lung disease. Clin. Sci. 66: 57. 12 Hesselink, A.E., Penninx, B.W., Schlosser, M.A. et al. (2004). The role of coping resources and coping style in quality of life of patients with asthma or COPD. Qual. Life Res. 13: 509–518.

585

586

29  Assessment of Exercise Limitation, Disability, and Residual Ability

13 Rosser, R. and Guz, A. (1981). Psychological approaches

14

15

16

17 18

19

20 21

22

23

to breathlessness and its treatment. J. Psychosom. Res. 25: 439–447. King, B. and Cotes, J.E. (1989). Relation of lung function and exercise capacity to mood and attitudes to health. Thorax 44: 402–409. Pearce, S.J., Posner, V., Robinson, A.J. et al. (1985). “Invalidity” due to chronic bronchitis and emphysema: how real is it? Thorax 40: 828–831. Musk, A.W., Bevan, C., Campbell, M.J., and Cotes, J.E. (1979). Factors contributing to the clinical grade of breathlessness in coalminers with pneumoconiosis. Bull. Eur. Physiopathol. Respir. 15: 343–353. Wright, G.W. (1949). Disability evaluation in industrial pulmonary disease. J. Am. Med. Assoc 141: 1218–1222. Bestall, J.C., Paul, E.A., Garrod, R. et al. (1999). Usefulness of the Medical Research Council (MRC) dyspnoea scale as a measure of disability in patients with chronic obstructive pulmonary disease. Thorax 54: 581–586. McGavin, C.R., Artvinli, M., Naoe, H., and McHardy, G.J.R. (1978). Dyspnoea, disability and distance walked: comparison of estimates of exercise performance in respiratory disease. Br. Med. J 11: 241–243. Quanjer, P.H. (ed.) (1983). Standardized lung function testing. Bull. Eur. Physiopathol. Respir. 19 (Suppl 5): 1–95. American Thoracic Society (1992). Lung function testing; selection of reference values and interpretative strategies. Am. Rev. Respir. Dis. 145: 1202–1218. Cotes, J.E., Chinn, D.J., and Reed, J.W. (2001). Body mass, fat percentage, and fat free mass as reference variables for lung function: effects on terms for age and sex. Thorax 56: 839–844. Gallagher, C.G. (1994). Exercise limitation and clinical exercise testing in chronic obstructive pulmonary disease. Clin. Chest Med. 15: 305–326.

Further Reading Williams, R.G.A., Johnston, M., Willis, L.A., and Bennett, A.E. (1976). Disability: a model and measurement technique. Br. J. Prev. Soc. Med. 30: 71–78.

24 Weller, J.J., El‐Gamal, F.M., Parker, L. et al. (1988).

25

26

27

28

29

30

31

32 33

Indirect estimation of maximal oxygen uptake for study of working populations. Br. J. Ind. Med. 45: 532–537. Jones, N.L., Makrides, L., Hitchcock, C. et al. (1985). Normal standards for an incremental progressive cycle ergometer test. Am. Rev. Respir. Dis. 131: 700–708. Oren, A., Sue, D.Y., Hansen, J.E. et al. (1987). The role of exercise testing in impairment evaluation. Am. Rev. Respir. Dis. 135: 230–235. Agostoni, P., Smith, D.D., Schoene, R.B. et al. (1987). Evaluation of breathlessness in asbestos workers: results of exercise testing. Am. Rev. Respir. Dis. 135: 812–816. Cotes, J.E., Chinn, D.J., Reed, J.W., and Hutchinson, J.E.M. (1994). Experience of a standard method for assessing respiratory disability. Eur. Respir. J. 7: 875–880. Cotes, J.E. (1990). Rating respiratory disability: a report on behalf of a working group of the European Society for Clinical Respiratory Physiology. Eur. Respir. J. 3: 1074–1077. Cotes, J.E. (1975). Assessment of disablement due to impaired respiratory function. Bull. Eur. Physiopathol. Respir. 11: 210–217. Christensen, E.H. (1953). Physiological valuation of work in Nykroppa steel works. In: Ergonomics Research Society, Symposium on Fatigue (eds. W.F. Floyd and A.T. Welford), 93–108. London: HK Lewis. Åstrand, P.‐O. and Rodahl, K. (1986). Textbook of Work Physiology, 3e. London: McGraw‐Hill. Cotes, J.E. and Reed, J.W. (1999). Relationship of maximal oxygen uptake (VO2max) to body mass (BM), implications for rating exercise ability. Eur. Respir. J. 14 (Suppl 30): 491S.

587

30 Exercise in Children Andrew Bush CHAPTER MENU 30.1 30.2 30.3 30.4 30.5 30.6 30.7 30.8 30.9 30.10

Introduction, 587 Indications for Exercise Testing in Children,  588 Methods, 589 Normal Response to Exercise in Children,  589 Special Indications for Exercise Testing in Children,  592 When There is a Discrepancy between Symptoms and Baseline Lung Function,  592 Assessment of Prognosis in Cases of Respiratory Disease,  592 Assessment of EILO,  592 Understanding the Physiology of Disease,  592 Summary and Conclusions,  593 References, 593

30.1 ­Introduction There is an extensive and increasing literature on this subject; space precludes more than an overview in this chapter. The smaller size of children compared with adults inevitably influences their responses to exercise. However, many of the underlying physiological responses are numerically similar in the two age groups, for example between alveolar ventilation and uptake of oxygen, cardiac output and uptake of oxygen [1–3], and maximal heart rate (HR) and age [1]. Apart from young children (age 14. 5 10–16

160

12–18

140 300 m. b

may expand faster than it can escape via the trachea and other orifices. The pressure within the cavity then rises relative to the surroundings and can cause barotrauma. Failure to equalise pressures can extend to gases dissolved in body fluids, particularly nitrogen. Bubbles of nitrogen form in any of several tissues of the body and can cause decompression sickness. The physiological responses to hypobaria and their implications for life at high altitude and for aviation have excited generations of physiologists [1–3]. Classical studies are reviewed in Chapter  1. Present views on the respiratory effects of hypobaria and the implications for persons who are exposed are summarised in this chapter. More detailed reviews into the respective features of high altitude and aviation are available elsewhere (e.g. [4, 5]).

32.2 ­The Atmosphere and Physiological Effects of Hypobaria 32.2.1  Atmospheric Pressure The barometric pressure falls exponentially with altitude, as shown in Figure  32.1, whilst the percentage of oxygen remains constant at just under 21%. Thus the inspired partial pressure of oxygen (PIO2) falls with alti-

Atmospheric ozone is formed by ultraviolet irradiation of diatomic oxygen molecules, which dissociate into atoms. At very high altitudes all oxygen exists in the monatomic form. Lower down, some of this monatomic oxygen combines with oxygen molecules to form the triatomic gas ozone, with concentrations up to 10 parts per million (ppm). The ozonosphere normally exists between 12 200 and 42 700 m (40 000 and 140 000 ft). Below 12 200 m (40 000 ft) the irradiation is normally too weak for significant amounts of ozone to form. Concentrations of 1 ppm at sea level can cause lung irritation. However, modern passenger jet aircraft are fitted with catalytic converters in the environmental control system (ECS) which break down the ozone before it enters the pressurised cabin [8, 9]. 32.2.4  Cosmic Radiation Aircraft occupants are exposed to elevated levels of cosmic radiation of galactic and solar origin. The Sun has a varying magnetic field, which reverses direction approximately every 11 years. Near the reversal, at ‘solar minimum’, there are few sunspots and the Sun‘s magnetic field extending throughout the solar system is relatively weak. At solar maximum there are many sunspots and other manifestations of magnetic turbulence. The Earth’s magnetic field has a larger effect than the Sun’s magnetic field on cosmic radiation approaching the atmosphere. The protective effect is greatest at the equator and least at the magnetic poles. At jet aircraft operating altitudes, galactic cosmic radiation is 2.5–5 times more intense in Polar regions than near the equator. The Earth’s surface is shielded from cosmic radiation by the atmosphere, the ambient radiation decreasing with altitude by approximately m (2000  ft) 15% for each increase of around 600 

32.2  The Atmosphere and Physiological Effects of Hypobaria

Figure 32.1  Diagram showing the composition of alveolar gas in unacclimatised subjects under conditions of reduced barometric pressure. The upper line indicates the barometric pressure; this falls exponentially with altitude, the pressure halving every 5.5 km (18 000 ft). The distances between the other lines represent the approximate partial pressures of the alveolar gases. Conditions at sea level are represented by the horizontal lines; the line for oxygen intercepts the curve for barometric pressure at about 10 300 m (34 000 ft). The alveolar oxygen tension of a subject who is breathing oxygen at this altitude is equal to that which obtains while breathing air at sea level. Source: Committee for Medical Research. Handbook of Respiratory Data in Aviation. Washington DC, 1944.

Altitude (1000 ft) 0

10

20

30

100 700 90

600

80

70

Pressure (kPa)

60 400 50

40

Pressure (mmHg)

500

Total barometric pressure

300

Nitrogen

30 200 20

102 mmHg

13.6 kPa Oxygen

10

0

(dependent on latitude). The effect of ionising radiation depends not only on the dose absorbed, but also on the type and energy of the radiation and the tissues involved. These factors are taken into account in arriving at the dose equivalent measured in sievert (Sv). However doses of cosmic radiation are so low that ­figures are usually quoted in microsievert (μSv) (millionth of a sievert) or millisievert (mSv) (thousandth of a sievert). Calculated and measured doses are well within the International Commission on Radiological Protection (ICRP) recommended limits for occupational exposure of a 5 year average effective dose of 20 mSv per year, with no more than 50 mSv in a single year. The annual limit for the general public is 1 mSv. Whilst it is known that there is no level of ionising radiation exposure below which effects do not occur, current epidemiological evidence indicates that the probability of airline crew members or passengers ­suffering any abnormality or disease as a result of exposure to cosmic radiation is very low [8, 10, 11].

100

4.9 kPa

Carbon dioxide

37 mmHg

6.3 kPa

Water vapour

47 mmHg

3

6 Altitude (1000 m)

9

12

32.2.5 O2 and CO2 Partial Pressures at Altitude The main physiological effect of hypobaria is a lowering of the partial pressure of oxygen (PIO2) in the inspired air. In the lungs the PIO2 is also influenced by the presence of carbon dioxide and water vapour. The water vapour pressure is a function of body temperature and is effectively constant at about 6.3 kPa (47 Torr). The partial pressure of carbon dioxide (PCO2) is not constant but decreases during hypoxaemia as a result of the increased chemoreceptor drive raising the level of ventilation. A further reduction in PCO2 occurs as a result of acclimatisation. The average composition of alveolar gas in relation to barometric pressure and altitude in unacclimatised subjects is shown in Figure 32.1. 32.2.5.1  Respiratory Control and Ventilatory Acclimatisation

The immediate effect of hypoxaemia is to augment the peripheral drive to breathe from the carotid bodies [12]. Subsequently, this effect is enhanced during acclimatisation

619

620

32 Hypobaria

when the responsiveness of the glomus cells increases [13]. It is independent of the cerebral effect of hypoxia, which, normally, is to depress ventilation [14]. Hypoxaemia also initially lowers cerebral vasomotor tone and increases the local blood flow, an effect that diminishes as acclimatisation progresses. As a result the oxygen tension in some parts of the brain is increased [15]. The changes can modify central chemoreceptor drive and the responsiveness of the respiratory motor centres [16]. The onset of hyperventilation increases excretion of carbon dioxide from the lungs and blood and hence raises plasma pH; this partly inhibits the chemoreceptor drive to respiration (Chapters 20, 21). Subsequently, the respiratory alkalosis is corrected by action of the kidneys. The reabsorption of sodium from the renal tubules is reduced, whilst the reabsorption of (Cl–) is increased (Chapter 20); the urine becomes more alkaline in consequence. The plasma (H+) then rises gradually and attains the normal sea level value in about a month at high altitude. At this stage the hypoxic drive to breathe is no longer restrained by hypocapnia. Instead, it contributes to a progressive rise in ventilation. The additional ventilation raises the alveolar oxygen tension compared with the unacclimatised state. The correction of the respiratory alkalosis also restores the ventilatory response to carbon dioxide (Chapter 21). The effect of hypercapnia is mainly to stimulate central chemoreceptors, but this action is modulated by the acid–base changes in cerebrospinal fluid (CSF) being out of phase with those in blood [17, 18]. The differences are due to the choroid plexuses actively modifying the composition of CSF and preventing ingress of lactic acid. In addition, the local cerebral blood flow can be reduced by the hypocapnia increasing cerebral vasomotor tone. Some of these actions are opposed by an increase in non‐ aerobic glycolysis leading to lactacidaemia. 32.2.5.2  Blunting of the Hypoxic Response

The driving force for acclimatisation is hypoxia, which initiates the hypoxic ventilatory response (HVR). This normally persists for many years but can eventually decline to a degree. This was first observed in Andean high‐altitude natives amongst whom blunting of the HVR usually occurs in early adolescence. In adults who move to high altitude the feature is common after about 20 years. It is reversed by return to sea level. Blunting is also not invariable, e.g. Tibetans are less susceptible than mountain people in the Andes to this blunting of their HVR [19]. This suggests that the pattern of adaptation can differ between ethnic groups, reflecting their genetic differences [20]. Blunting of the HVR is accompanied by relative hypercapnia and other features of chronic mountain sickness (CMS), including gross polycythaemia and is more

c­ommon in Andean than Tibetan populations. One study [21] suggested that there is a genetic difference between those Andean highlanders with and without CMS. The syndrome resembles other conditions of chronic hypoventilation, including sleep apnoea/hypopnoea syndrome (Chapter 31). However, the associations do not extend to the Himalayan bar‐headed goose and the South American llama. These animals, when raised at high altitude, exhibit blunting yet are very tolerant of their situation [22, 23]. 32.2.5.3  Alveolar Gas Exchange

The mechanisms whereby during normoxia the blood leaves the alveolar capillaries almost fully saturated with oxygen are described in Chapters 16, 17. The most important aspect is that gas exchange is completed on the flat part of the oxyhaemoglobin dissociation curve (Figure 32.2). By contrast, during altitude hypoxia, the arterial PO2 lies on the steeper part of the O2 curve. This disadvantage is mitigated by a pronounced shift to the left of the oxygen dissociation curve through the combined effects of respiratory alkalosis, reduced PCO2, and increased production of 2,3‐diphosphoglycerate [24, 25] (Figure  32.2). Significant left shift becomes important at extreme altitude where the reduction in PCO2 and alkalosis is profound. However, the displacement only goes a small way to compensate for the marked reduction in inspired oxygen tension that occurs at high altitude. Transfer of gas is further disadvantaged by an initial increase in cardiac output that is a consequence of hypoxia. The increased flow shortens the time spent by individual red cells in the alveolar capillaries during which exchange of gases can take place. Also, and more importantly, there is a great reduction in driving pressure of oxygen across the alveolar–capillary membrane (PA,O2–Pa,O2), resulting in a slower rate of oxygen diffusion into the blood. These ­features result in blood leaving the alveolar capillaries without achieving partial pressure equilibrium with the alveolar gas (Figure  32.3). At an altitude of 3000 m the alveolar to end‐capillary gradient is small at rest, but increases during exercise, when it can exceed 1.3 kPa (10 mmHg). Larger values occur at higher altitudes. Hence, the oxygenation of arterial blood is diffusion limited, especially at very high altitudes [26]. Uneven distribution of ventilation with respect to perfusion may also occur, but its contribution to hypoxaemia is small except when there is alveolar interstitial oedema [27]. At altitude, in healthy persons, the changes described above combine to maximise the transfer of oxygen from atmospheric air to arterial blood and also in the tissues from capillary blood to mitochondria in skeletal muscle fibres [26, 28]. However, the levels of oxygenation that are achieved are not the same in everybody because of inherited or acquired differences in components of the

32.2  The Atmosphere and Physiological Effects of Hypobaria

Figure 32.2  Dissociation curve for oxyhaemoglobin at sea level, with data for subjects acclimatised to high altitude superimposed. The displacements facilitate gas exchange (see text); they are in the direction of those associated with fetal haemoglobin. 2,3‐DPG, 2,3‐ diphosphoglycerate; P50O2, partial pressure of oxygen at which haemoglobin is 50% saturated with oxygen. Source: [52, 126].

Tension of O2 (mmHg) 100 90 80

Saturation (%)

70 60

10

0

20

30

40

50

60

70

80

90

100

pH increased to 7.5 or temp decreased to 35 °C or (2,3-DPG) decreased Normal curve pH 7.4 and 37°C

Newborn baby pH 7.4 and 37 °C

19 000 ft Summit of Everest

P50O2

50 40 30 20 10

Lungs 2

3

4

5

6 7 8 Tension of O2 (kPa)

9

10

11

12

13

20 Inspired

140

Dm,O2 = 40 ml min–1 mmHg–1

120

Sea level (Pb 101.3 kPa)

VO2 = 300 ml min–1

Alveolar

100

End-capillary

80

10

60

Inspired Alveolar

Mixed venous 40 Summit of Everest (Pb 33.7 kPa)

20 0

End-capillary

Mixed venous B 0

0.2

PO2 (kPa)

Figure 32.3  Time course for uptake of oxygen by blood during transit through an alveolar capillary. The upper curve is for conditions at sea level. The lower curve extends the diagram to conditions at high altitude. Here the oxygen tensions are all lower, gas exchange takes place on the steep part of the oxygen dissociation curve (Figure 32.2), and blood leaves the capillary still containing oxygen at a lower tension than in alveolar gas. See text. Source: Redrawn from [52].

1

PO2 (mmHg)

0

0.4

0.6

0

Time along capillary (s)

oxygen transport chain. High‐altitude populations appear to have optimal oxygen transport while unacclimatised newcomers must first acclimatise to begin to optimise their oxygen transport.

32.2.5.4  Circulatory System

Acute exposure to hypoxia stimulates aortic chemoreceptors to cause an almost immediate increase in cardiac  frequency; the tachycardia can be experienced as

621

32 Hypobaria

palpitations. These are often severe in persons who take exercise soon after a rapid ascent to high altitude by train or car and can cause acute distress. Concurrently, there are increases in vasomotor tone [29] and in cardiac output relative to uptake of oxygen. The changes have the effect of increasing the delivery of oxygen to tissues. Less oxygen is then extracted per unit volume of circulating blood, so the tissue and mixed venous oxygen tensions are not as low as would otherwise be the case. If the exposed person discontinues or slows down their ascent the circulatory changes are normally reversed as part of the process of acclimatisation. As acclimatisation progresses, the relationship of cardiac output to oxygen uptake then reverts to that at sea level [30, 31]. However, the relative contributions to the cardiac output of cardiac frequency and stroke volume remain abnormal, with the frequency increased and the stroke volume reduced. During the first 2–3 days at altitude there is usually a diuresis and the plasma volume is decreased. Haemoglobin (Hb) concentration is increased, which is beneficial, but this reduction in plasma volume, by reducing venous return, probably is the major cause of the reduction in  stroke volume [32]. Hypoxic depression of the myocardium appears not to be a contributory factor [27]. 32.2.5.5  Haemopoiettic System

An immediate and persistent effect of hypoxaemia is to increase the production by the renal cortices of the hormone erythropoietin, mediated by upregulation of the nuclear transcription factor hypoxia‐inducible factor (HIF) 1 [33–35]. The rise in erythropoietin levels stimulates the bone marrow to increase the production of erythrocytes. The response is heralded within a few hours by a rise in turnover of iron and, subsequently, by an increase in blood reticulocytes. There is also a reduction in plasma volume. As a result of these changes the red cell count and blood oxygen capacity rise progressively over a period of some 5 weeks [36]; the final improvement can exceed 30% [34]. Iron supplementation can be important: women who take supplemental iron at high altitude approach the haematocrit values of men at altitude [37]. The polycythaemia is beneficial in that it increases the capacity of blood to deliver oxygen to tissues. This effect persists for a few weeks after return to sea level, so can be a help to athletes who train at altitude for an event at a lower altitude. However, as well as benefits there are also disadvantages to using altitude for training. For example, the maximal intensity of exercise is less at altitude, so the muscle strength is not developed to the extent that it would be when training at sea level. In addition, the polycythaemia is accompanied by a rise in blood viscosity and increased rouleaux formation amongst the red blood cells. The changes can predispose to vascular thromboses, impair gas exchange in both lungs and tissues, and be a

cause of heart failure. These complications affect particularly those persons in whom the erythropoietic response is at the upper end of the normal distribution. Such people are at increased risk of CMS (Monge syndrome). 32.2.5.6  Maximal Ventilation

In addition to lowering the PIO2, hypobaria has several additional effects including changing the density of respired gas, and some effects of unknown cause that distinguish hypobaric hypoxia from normobaric hypoxia. Hypobaria lowers the density and viscosity of the respired gas. Both changes reduce the Reynolds number for air flowing through tubes, including those in the lungs, and hence the extent to which the flow of air is turbulent. Similar changes occur in instruments for measuring flow [38, 39]. In the lungs the airway resistance is reduced and the maximal voluntary ventilation (MVV) is increased (Figure 32.4). The increase enables a person to ventilate the lungs with a greater volume of ambient air at high altitude than is possible at sea level. 32.2.6  Exercise at Altitude Maximal oxygen consumption drops dramatically on ascent to high altitude (for recent reviews, see [40, 41]). Maximal oxygen uptake (VO2max) falls from sea level by approximately 10% for each 1000 m of altitude gained above 1500 m. Those with the highest sea‐level VO2max

1

140

2 MVV as % of sea-level value

622

130

120

110

100 1.0

0.8

0.6

0.4

0.2

Lung gas density relative to sea level

Figure 32.4  Maximal voluntary ventilation under conditions of reduced barometric pressure; mean values for 16 healthy young men. Curves 1 and 2 are the theoretical relationships based on gas density, with and without allowance for gas viscosity. The curve that allows for viscosity describes the data better. Source: [127].

32.3  Effects of Altitude on Lung Function in Lowlanders

values have the largest decrement in VO2max at high altitude, but overall performance at high altitude is not consistently related to sea‐level VO2max [42, 43]. In fact, many of the world’s elite mountaineers have quite average VO2max values, in contrast to other endurance athletes [42]. Acclimatisation at moderate altitudes enhances submaximal endurance time, but not VO2max [40, 41]. During exercise at high altitudes the maximal exercise ventilation more nearly approaches the MVV than is usually the case at sea level [44]. The increased ventilation leads to work hypertrophy of the respiratory muscles and affects mountaineers on expeditions to high altitude [45]. The changes in ventilation augment the maximal volume of air that can be inhaled per minute, but the energy cost of ventilation rises with minute volume and can exceed the quantity of extra oxygen that is delivered [46]. Exercise is then ‘ventilation limited’ and the capacity for aerobic exercise is reduced [47, 48]. Performance at some athletic and cycling sprint events may nonetheless be increased at altitudes of up to about 3000 m (10 000 ft), since the reduced gas density lowers the resistance offered by air to forward movement of the body [49, 50]. Depending on the altitude, the level of training, and the intensity of the exercise, unacclimatised persons can develop unpleasant symptoms. These are due to the hypoxaemia driving the ventilation and cardiac frequency to cause intense breathlessness and tachycardia. During very strenuous exercise, the tachycardia can c­onvert to bradycardia on account of a rise in parasympathetic tone [51]. The hyperventilation leads to respiratory alkalosis, which in turn reduces cerebral blood flow to cause giddiness and even loss of consciousness. In acclimatised subjects the ventilation can be sustained and the associated breathlessness becomes the principal factor that limits exercise. However, the breathlessness is due to hyperventilation, which in turn is due to increased respiratory drive from hypoxaemia secondary to gas transfer limitation. As a result, in acclimatised subjects at high altitude, exercise is ‘diffusion limited’. At altitudes above the level of acclimatisation, for example at near the summit of Mount Everest, a climber can only manage the mildest exercise, stopping to pant every step or two [52] (Table 32.2). 32.2.6.1 Sleep

At altitude the hypoxaemia and resulting hypocapnia modify the changes that occur during normal sleep at sea level (Chapter 31, Breathing during sleep and its investigation). On the first night following the ascent there is an increase in very light sleep (stage 1), a reduction in slow wave sleep (stages 3 and 4), and an increased number of arousals. The associated arousals are largely secondary to periodic breathing, which is a common complaint at high altitude. Low‐dose acetazolamide effectively

Table 32.2  Exercise limitation in a climber under conditions simulating those on the summit of Mt Everest compared with that of a male patient with chronic lung disease who had the same maximal oxygen uptake at sea level. Patient (male) with COPD

Subject

Climber (CP)

Age (year)

31

62

Conditions

6300 m, breathing 14% O2 in N2, bicycle

Sea level, air, treadmill

1.06 (3.4)

1.06 (2.0)

 2 max (l STPD min−1) VO MVV (l BTPS min−1) V Emax,ex (l BTPS min−1)

190a (150)

37b (94)

166 (24)

42 (24)

as % of estimated MVV

87%

114%

RER and HbO2 (Saturation, %)

1.12 45%

0.77 97%

Cardiac frequency (min−1) 135 (95)  2 1.06 at VO l STPD min−1

109 (108)

Reference values are in brackets; those for cardiac frequency include an allowance for physical fitness. a  assuming normal value at sea level. b  estimated from FEV1. Note. Both men were limited on exercise by breathlessness caused by a high exercise ventilation relative to MVV. In CP this was due to hypoxic hyperventilation with high RER and low saturation. In the patient it was due to airways obstruction with normal RER and saturation, but raised ventilation due to VA/ Q  inequality. Cardiac frequency was normal  2 in the patient but increased in CP on account of hypoxia; both for VO values were submaximal. Sources: [126 and unpublished].

obliterates the periodic breathing, improving sleep while also stimulating ventilation [53]. During periodic breathing a cycle is established where the Pa,O2 falls and Pa,CO2 rises; the changes initiate the next cycle of hyperventilation followed by hypoventilation. Control theory predicts that, in a control system with a negative feedback loop, a high gain in the system promotes hunting, which is what we have in periodic breathing. Most lowlanders have a high enough gain, i.e. their HVR, to trigger periodic breathing. However, many high‐ altitude residents do not. A study carried out at Everest Base Camp (5300 m) confirmed this prediction [54].

32.3 ­Effects of Altitude on Lung Function in Lowlanders 32.3.1  Peak Expiratory Flow With the advent of peak flow meters and hand‐held electronic spirometers it became easy to measure peak

623

624

32 Hypobaria

expiratory flow in studies at altitude on treks or expeditions. However the Wright mini peak flow meter [55], being a variable orifice device, is affected by air density, so it under‐reads at altitude [56]. In using this meter a correction for this change of density can be applied; for instance at Everest Base Camp, where the barometric pressure is approximately half that at sea level, a correction factor of 6.5% should be added. Electronic spirometers with fixed orifices do not suffer this change of calibration. Mason et al. [57] showed that, at Everest Base Camp, peak expiratory flow (PEF) increased by 16%. This is marginally below the value expected due to the reduction in air density alone (~20%). 32.3.2  Bronchoconstriction, Hypoxia, and Hypocapnia An early effect of ascent to altitude is to cause hypoxaemia, which stimulates ventilation via the HVR. This causes a degree of hypocapnia. Fifty years ago, acute hypoxia and hypocapnia were shown to cause bronchoconstriction in animals and chronic hypobaric hypoxia in calves had the same effect [58, 59]. In humans, patients with chronic hypoxaemic lung disease, bronchoconstriction was relieved by oxygen administration [60, 61]. However, in trekkers acclimatised to the altitude of Everest Base Camp (5300 m), Mason et  al. [57] found no significant change in forced vital capacity (FVC) or forced expiratory volume 1 s (FEV1) on giving oxygen to breathe for 5 min raising the oxygen saturation from 81% to 94%. PEF, however, fell significantly, by 2.3%; possibly due to a concomitant rise in PCO2. Compared with sea‐level control measurements, values on arrival at Base Camp showed lower FVC, which was not related to either acute mountain sickness (AMS) scores or oxygen saturation. The authors discuss various causes for this finding but conclude that subclinical pulmonary oedema is the most likely. 32.3.3  Subclinical Oedema The possibility and frequency of subclinical oedema on ascent to altitude was addressed by Cremona et al. [62] in a study at 4559 m in the Alps. These authors studied clinically 262 climbers who had climbed to this altitude in a few days with lung function tests and radiography. One subject had overt high‐altitude pulmonary oedema (HAPE) and 15% had chest rales or interstitial oedema on chest radiograph. Of these 92% had increased closing volumes. Of the majority of subjects without overt signs of oedema, 74% had increased

closing volumes. There was no change in FVC or FEV1. The conclusion was that, following an ascent of this sort, three out of four climbers will have mild subclinical oedema. However, another study, at the same location, found no evidence of such widespread subclinical oedema. Dehnert and colleagues [63] carried out extensive lung function testing, including body plethysmography, chest radiographs, closing volume, CO transfer factor, lung compliance, as well as standard spirometry. Thirty‐four subjects were studied daily for 2 days after rapid ascent to 4559 m. Four subjects developed HAPE with evidence of increased lung water, which was mild, compared with their gas exchange problems. In other subjects, with and without AMS, no evidence of subclinical oedema was found. It seems possible that the increase in closing volume found in the earlier study was due to the exercise in getting to the Margherita Hut, as argued by Bärtsch et al. [64]. 32.3.4  Lung Diffusing Capacity Barcroft et  al. [1] measured the carbon monoxide diffusing capacity of the lung (Dl,CO) in themselves on their expedition to Cerro de Pasco (4300 m) and found a small rise in three subjects but no change in one and a slight fall in the fifth subject, with acclimatisation. Since then studies have shown either no change or a small (65%) climbing from sea level to 4300 m on Mount Rainier develop AMS [98]. In contrast, a climber acclimatised to 4000 m would be immune to AMS at the same altitude, even after some time at low altitude [99, 100]. With the advent of air travel and burgeoning interest in outdoor activities, AMS has become common throughout the world’s high mountains. The main symptoms of AMS are headache, nausea, vomiting, light‐headedness, and shortness of breath [53]. Predisposing factors include a relatively young age and exertion soon after the ascent. The level of habitual activity (physical fitness) or a reduced ventilatory response to hypoxia appear not to be relevant. AMS usually occurs within 6–12 h after arrival at altitude. The causes of AMS are currently incompletely understood. A popular theory, waiting for experimental testing, is that hypoxemia causes elevated intracranial pressure, which in turn causes stretching of the pain‐ sensitive meninges of the brain, accounting for the headache of AMS [54, 101]. One study in normobaric hypoxia h of measured lumbar opening pressure before 18  hypoxia, and afterwards [102]. During the experiment subjects took paracetamol (acetaminophen) to relieve the headache of AMS. The spot measurements of intracranial pressure showed no difference in intracranial pressure between AMS and no AMS subjects. Concerns are that spot measurements may miss short periods of

increased pressure during the onset of AMS that may trigger the pathophysiological processes leading to high‐ altitude headache and AMS. Thus, at present, the potential role of elevated intracranial pressure in AMS remains an open question. 32.6.2  High‐Altitude Cerebral Oedema In the absence of treatment AMS may progress to cerebral oedema but this is much less common than AMS, with an incidence of about 1–2% of trekkers to high altitude. The patient will experience ataxia and impaired consciousness. At this stage, the risk is of progression to coma and death. Immediate descent, treatment with oxygen, and/or steroids can be lifesaving. For a detailed review of HACE that is beyond the scope of this chapter see [104]. 32.6.3  High‐Altitude Pulmonary Oedema Onset of marked breathlessness and unusual physical fatigue on the second or third day after an ascent to or above 2500 m is likely to be due to onset of pulmonary oedema [87]. The oedema is initially interstitial, when it can be accompanied by a dry (unproductive) cough. The closing volume can be increased [62]. Later, there can be frank pulmonary oedema, with bloodstained sputum, mild fever, and crackles on auscultation of the chest. The associated hypoxaemia can initiate or aggravate AMS or HACE. Fortunately, HAPE usually responds well to immediate treatment. Left untreated patients with HAPE can rapidly deteriorate and die, with an estimated 50% mortality for those left untreated. The pulmonary oedema is non‐cardiogenic and is caused by exaggerated hypoxic pulmonary vasoconstriction and abnormally high pulmonary artery pressure and capillary pressure [88]. These high pressures lead to a non‐inflammatory and haemorrhagic alveolar capillary leak that may later evoke an inflammatory response [103]. Inflammation is not an initiating event [103–105]. Precipitating factors include injudicious ascent to altitude, high‐intensity exercise, and exposure to cold. The susceptibility is increased by recent lower respiratory infection and by the possession of an enhanced vasopressor response to hypoxia [106, 107]. 32.6.3.1  Prevention and Treatment

The possibility of mountain sickness should be borne in mind on all journeys to altitudes in excess of about 1500 m. However, it is unlikely below about 3000 m. For ascents to greater altitudes, time should be allowed for acclimatisation. This can usually be achieved by not ascending more than 300–400 m per day, having a pause after every third day, and including unallocated days in

627

628

32 Hypobaria

the itinerary. On general grounds, abstaining from smoking, alcohol, and sedative drugs is likely to be beneficial. For prophylaxis against AMS and HACE [108, 109] acetazolamide (Diamox) in the dosage 125 mg two times per day (BID) is effective up to about 4500 m [110] (Figure 32.5). At higher altitudes or with faster ascents, higher doses may be needed [111]. For those intolerant to acetazolamide, dexamethasone at 4 mg three times per day offers an alternative [112]. For prophylaxis against HAPE drugs that lower pulmonary artery pressure are generally effective. Studies in persons with a history of HAPE have shown nifedipine in a slow release formulation at a dose of 30 mg BID [113, 114], tadalafil (a phosphodiesterase‐5 inhibitor) at a dose of 10 mg BID, and dexamethasone at a dose of 8 mg BID [108] appear to be similarly effective in reducing the incidence of HAPE from approximately 70% to 10% [113]. These substances do not mimic acclimatisation; they block the development of HAPE. Once medication is stopped, one may be again vulnerable to high‐altitude illness. The mild form of AMS is usually self‐limiting, so an appropriate treatment is rest and/or medication. Consideration can be given to temporary descent to a lower altitude. More significant symptoms, including those of impending HACE or HAPE, are indications for immediate descent towards sea level. Meanwhile, oxygen and appropriate remedies as given above should be administered [101]. If equipment is available, the inspired oxygen tension can be raised by pressurisation using a pressure bag or portable chamber (Gamow Bag) [115–117]. Morphine or diuretic drugs are not normally recommended. Recovery from AMS is usually rapid and complete but in a few cases may be delayed [118]. Placebo Respiratory pattern 100 Sa,O2 (%)

80 60 40 Acetazolamide

Respiratory pattern 100 Sa,O2 (%)

80 60 40

32.7 ­Physiology and Medicine of Flight Physiological adaptations to hypoxaemia secondary to hypobaria are considered earlier in this chapter. How they can be reconciled with the requirements of aviation and the additional problems associated with barotrauma, atelectasis, and decompression illness are examined here. Other aspects of the aviation environment are also briefly considered. 32.7.1  The Aircraft Cabin To allow a margin of safety, the maximum certified cabin altitude in pressurised passenger aircraft is 2440  m (8000  ft), at which barometric pressure is 75.1  kPa (565 mmHg) and arterial oxygen pressure is around 55 mmHg and venous oxygen pressures have fallen only by 1–2 mmHg. The minimum cabin pressure of 75.1 kPa (565 mmHg) (2440 m [8000 ft]) will bring a healthy individual’s arterial PO2 along the plateau of the oxyhaemoglobin dissociation curve, at the top of the steep part, still almost fully saturated. However, people with respiratory disease may have arterial oxygen pressures at ground level as low as 8 kPa (55–60 mmHg). Their further fall in arterial PO2 can be reversed completely by the administration of oxygen; 30% oxygen at 2440 m (8000 ft) being equivalent to breathing air at ground level. Given prior notice, most airlines can provide a personal oxygen supply for such passengers. Aircraft operating below 3000  m (10  000  ft) do not require oxygen equipment to be installed. Most commercial passenger aircraft fly higher and maintain cabin pressure Figure 32.5  Respiratory patterns and arterial oxygen saturation (Sa,O2) with placebo and with acetazolamide, in two sleep studies of a subject at 4299 m. Note the pattern of periodic breathing during placebo treatment, which is abolished with acetazolamide. Reprinted with permission of the American Thoracic Society. Copyright (c) 2014 American Thoracic Society. Modified from Hacket et al. [6]. Official Journal of the American Thoracic Society.

32.8  Fitness to Fly as a Passenger

above 565 mmHg (2440 m [8000 ft]), providing an environment in which the occupants breathe cabin air. The cabin air supply is bled from the outside air entering the aircraft engine, or may be supplied via electrically driven compressors. The mixed conditioned air is distributed via overhead ducts and grills, circulated around the cabin and extracted through vents at floor level. Recirculated air is passed through high‐efficiency particulate air (HEPA) filters of the same specification used in hospital operating theatres, 99.99% efficient in the removal of physical contaminants; it is bacteriologically cleaner than the air in buildings, trains, or buses. An emergency oxygen supply is provided in the event of failure of the pressurisation system. High‐performance military aircraft have low‐ differential pressurised cabins, giving effective cabin altitudes up to 5500 m (18 000 ft) at high operational altitudes. Protection from hypoxia is provided by the use of personal oxygen breathing equipment. Some unpressurised aircraft, particularly high‐ performance military combat helicopters, may operate at altitudes up to 4572 m (15 000 ft) without supplementary oxygen for the crew. It has been shown that cognitive and psychomotor deficits due to hypoxic impairment may occur, although the only consistent finding is that of visual degradation, i.e. impairment of night vision. 32.7.2  Mechanical Effects of Pressure Change The climb to cruise altitude in a passenger airliner takes about 30 min and involves a maximum fall of about 200 mmHg (26.6 kPa) in cabin pressure (to the equivalent of 2440 m [8000 ft]). Descent to land takes much the same time. Body fluids and tissues generally are virtually incompressible and do not alter shape to any important extent when such pressure changes are applied. The same is true of cavities such as the lungs, gut, middle ear, and facial sinuses that contain air, provided that they can vent easily. Gas‐containing spaces that cannot vent easily behave differently [119]. The thoraco‐abdominal wall can develop transmural pressures of +100 mmHg or so briefly, but is normally flaccid and has a transmural pressure of a few millimetres of mercury. Gas within will usually be at a pressure very close to that outside, and must follow Boyle’s law. mmHg) to 2440  m Ascent from ground level (760  (8000 ft) (565 mmHg) will expand a given volume of trapped gas by about 35%. This may cause slightly uncomfortable gut distension in healthy people but it is not an important problem. Even very diseased lungs can vent themselves over a minute or so. In consequence, the risk of lung rupture in normal flight is extremely rare.

The cavity of the middle ear vents easily, but sometimes fails to fill because the lower part of the Eustachian tube behaves as a non‐return valve, especially when it is inflamed. As a result, the cavity equilibrates easily on ascent but does not refill on descent, and the eardrum bows inwards, causing pain that can be severe (otic barotrauma) [7, 9, 105–107]. 32.7.3  Assessment of Aircrew The purpose of medical assessment is to ensure flight safety by reducing the risk of in‐flight incapacitation. Different medical requirements apply to the various classes of flying licence defined by the International Civil Aviation Organisation, an agency of the United Nations. Class 1 medical certification is required by airline transport and commercial pilots and class 2 by private pilots and flight instructors. Standards are applied in the UK by the Aeromedical Section of the Civil Aviation Authority on behalf of the European Aviation Safety Agency, using requirements originally formulated by the European Joint Aviation Authorities [120]. Routine medical examinations are performed by independent aeromedical examiners, the periodicity being related to class of medical and age. For pilots, aeromedical disposition following any injury or disease is predicated on the so‐called 1% rule, which is an attempt to quantify risk assessment. By seeking to ensure that no individual with an incapacitation risk of over 1% per annum operates as a pilot, it aims to achieve a target fatal accident rate of 0.1 fatal accidents per 1 million flying hours for commercial aviation. For non‐commercial private aviation, the acceptable risk is greater and an arbitrary 2% risk of incapacitation may be acceptable.

32.8 ­Fitness to Fly as a Passenger Flying as a passenger should be no problem for the fit, healthy, and mobile individual. But for the individual with certain pre‐existing conditions, the cabin environment may exacerbate underlying problems [110–112, 121]. Although many problems relate to the physiological effects of hypoxia and expansion of trapped gases, the complex airport environment can be stressful and challenging to the passenger before even getting airborne. Whilst passengers with medical needs require medical clearance from the airline, passengers with disabilities do not. Disabled passengers do need to notify the requirement for special needs, such as wheelchair assistance, and this should be done at the time of booking. The objectives of medical clearance are to provide advice to passengers and their medical attendants on fitness to fly, and to prevent delays and diversions of the flight as a result

629

630

32 Hypobaria

of deterioration in the passenger’s well‐being. It depends upon self‐declaration by the passenger, and upon the attending physician having an awareness of the flight environment and how this might affect the patient’s condition. Most major airlines provide services for those passengers who require extra help, and most have a medical advisor to assess the fitness for travel of those with medical needs. Individual airlines work with their own guidelines, but these are often based on those  published by the Aerospace Medical Association [119] or the International Air Transport Association (IATA) [11]. IATA publishes a recommended Medical Information Form (MEDIF) for use by member airlines, available on individual airline websites. The MEDIF should be completed by the passenger’s medical attendant and passed to the airline, or travel agent, at the time of booking to ensure timely medical clearance. Medical clearance is required when: ●●

●●

fitness to travel is in doubt as a result of recent illness, hospitalisation, injury, surgery, or instability of an acute or chronic medical condition special services are required (e.g. oxygen, stretcher, or authority to carry or use accompanying medical equipment such as a ventilator or a nebuliser).

Medical clearance is not required for carriage of an invalid passenger outside these categories, although special needs (such as a wheelchair) must be reported to the airline at the time of booking. Cabin crew are not able to provide individual special assistance to invalid passengers beyond the provision of normal in‐flight service. It is vital that passengers remember to carry with them any essential medication, and not pack it in their checked baggage (particularly important for passengers requiring bronchodilator inhalers, insulin, or nitrates). Deterioration on holiday or on a business trip of a previously stable condition or accidental trauma may give rise to the need for medical clearance for the return journey. A stretcher may be required, together with medical support, and this can incur considerable cost. It is important to have adequate travel insurance, which includes provision for the use of a specialist repatriation company to provide the necessary medical support. In determining the passenger’s fitness to fly, a basic knowledge of aviation physiology and physics can be applied. Any trapped gas (including from eye or ear surgery, as well as thoraco‐abdominal procedures) will expand in volume by up to 35% during flight, and consideration must be given to the effects of the relative hypoxia encountered at a cabin altitude of up to 2434 m (8000 ft) above mean sea level. The altitude of the destination airport may also need to be taken into account. The passenger’s exercise tolerance can provide a useful guide on cardiorespiratory fitness to fly; if unable to walk

a distance greater than about 50 m without developing dyspnoea, there is a risk that the passenger will be unable to tolerate the relative hypoxia of the pressurised cabin. More specific guidance can be gained from knowledge of the passenger’s baseline sea‐level blood gas levels and Hb value and application of a hypoxic challenge test [11, 110, 113, 114]. A good source of guidance is provided by the websites of the Aerospace Medical Association (www.asma.org) and the British Thoracic Society (https://www.brit‐ thoracic .org .uk/quality‐improvement/clinic al‐ statements/air‐travel). 32.8.1  On‐Board Oxygen In addition to the main gaseous system, all commercial aircraft carry an emergency oxygen supply for use in the event of failure of the pressurisation system or during emergencies such as fire or smoke in the cabin. The passenger supply is delivered via drop‐down masks from chemical generators or an emergency reservoir, and the crew supply is from oxygen bottles strategically located within the cabin. This emergency supply has a limited duration. In addition, sufficient first‐aid oxygen bottles are carried to allow the delivery of oxygen to a passenger in case of a medical emergency inflight, but there is insufficient to provide a premeditated supply for a passenger requiring it continuously throughout a journey. If a passenger has a condition requiring continuous (‘scheduled’) oxygen for a journey, this needs pre‐notification to the airline at the time of booking the ticket. Many airlines make a charge to contribute to the cost of its provision [108]. All equipment used onboard must meet regulatory standards; the specification for aviation oxygen is higher than that for normal medical oxygen in terms of permissible water content (to prevent freezing of valves and regulators at high altitude). The supplementary or scheduled oxygen provided for use by the sick passenger may be delivered from gaseous bottles, or it may be delivered on some aircraft by tapping into the ring‐main system. Some carriers provide molecular sieve concentrators, or allow passengers to use their own, subject to the concentrator having regulatory authority (US Federal Aviation Authority or European Union Aviation Safety Agency) approval. Those airlines that do provide oxygen usually do so only in flight; if oxygen is required on the ground, e.g. at an airport of transit, the passenger is probably unfit to fly. 32.8.2  Deep Vein Thrombosis and Venous Thromboembolism Long‐haul travel is associated with prolonged periods of immobility, a recognised risk factor for deep vein thrombosis (DVT). However, there have been concerns

32.9  Altitude‐Induced Decompression Illness

as to whether there are other factors specific to air travel which further increase the risk. But the mild degree of hypoxia does not seem to increase the risk [109]. In the general population DVT occurs in 1–3 per 1000 people per year, of which 20% give rise to pulmonary embolism. Increasing age is known to be a strong risk factor, possibly due to decreased mobility and reduced muscular tone [109]. The pathogenesis of thrombosis still relies on the basic premise of Virchow, who in 1856 identified circulatory stasis, hypercoagulability, and endothelial injury as the risk factors. Several clinical studies have shown [109, 122] an association between air travel and the risk of DVT, with the risk of venous thromboembolism (VTE) in travellers increasing with the distance travelled. It has been shown that all modes of travel increase the risk of venous thrombosis about twofold, with an absolute risk of one thrombosis per 6000 journeys [109]. It has been found that combinations of risk factors synergistically increase the risk of thrombosis. In individuals with factor V Leiden, the risk of thrombosis after flying was increased about 14‐fold, and in women using oral contraceptives it was increased around 20‐fold. It has also been shown that the risk rises with the number of flights taken in a short time frame as well as with the duration of the flight. The majorities of these clots are asymptomatic and disperse naturally. Thus even though the overall risk of venous thrombosis after air travel is only moderately increased, clear subgroups can be identified in whom the risk is higher. The increased risk of DVT after major surgery is very much higher than even flights of over 8 h. The low humidity of the aircraft cabin does not in itself lead to dehydration. Excessive alcohol consumption may cause dehydration, but there is no evidence that this is a significant risk factor leading to DVT. Studies of reduced oxygen partial pressure with non‐ hypoxic control groups found no evidence of coagulation. There is no evidence that hypoxia or the hypobaric environment of an aircraft cabin is a significant risk factor for the development of DVT [109]. Although there is good evidence for the value of aspirin in preventing arterial thromboembolic disease, its role in the prevention of venous thromboembolic disease is much less clear. The side‐effect profile is significant. There is no evidence to support the use of aspirin in preventing the development of DVT during flight. For those travellers at medium to high risk of DVT, there is evidence that the use of compression stockings appears to substantially lower the risk of asymptomatic DVT, but it remains unclear as to whether this reduction is clinically significant. One study has shown that, for 20–40% of travellers, the commercially available stockings

do not fit adequately. It is essential for stockings to be correctly fitted so as to provide adequate compression to stimulate venous return. Although the use of low molecular weight heparin for the prevention of DVT in the aviation setting is not supported by direct evidence, in a high‐risk traveller consideration may be given to a single prophylactic dose prior to flying. While the relative risk of developing venous thrombosis when flying is statistically significant, the absolute risk of developing symptomatic DVT is very low. The absolute risk of developing a pulmonary embolus during or after a flight between the UK and the east coast of the USA has been calculated as less than 1 in a million [7, 13]. Medical practitioners need to be circumspect in advising any preventive measures, taking careful account of efficacy and risk profile of the preventive method.

32.9 ­Altitude‐Induced Decompression Illness Decompression illness results from nitrogen in solution appearing in the form of bubbles as a result of the air pressure round the subject being reduced suddenly [112]. The situation resembles that in a bottle of sparkling water when the cap is removed. The condition can occur after rapid ascent (decompression) from sea level to high altitude or a high simulated altitude in a decompression chamber. The risk is related to the pressure gradient for nitrogen between the tissues and the lungs. If ambient pressure falls quickly to less than half its original value, the gas dissolved in blood and tissue fluids may come out of solution precipitously, forming bubbles and obstructing flow in small blood vessels leading to neurological impairment. The time symptoms take to develop varies widely between individuals and shortens markedly as the altitude of exposure rises. Atmospheric pressure halves at 5500 m (18 000 ft) and decompression illness rarely occurs, if at all, below this altitude. It is very rare below 7600 m (25 000 ft) and therefore is normally of no concern at passenger aircraft cabin altitudes, which normally do not exceed 2434 m (8000 ft). However, it does occasionally occur in those passengers who have been exposed to a hyperbaric environment prior to flight, such as divers and tunnel workers. Sub‐aqua divers can be at risk and are advised to allow a minimum of 12 h to elapse between diving and flight, or 24 h if the dive required decompression stops, with 17 h being the optimal for scuba‐divers [107, 123]. The military U‐2 aircraft exposes its pilots to cabin m (29  500  ft), placing pressures equivalent to 8992  them  at  risk for decompression illness. The pilots wear a  full pressure suit designed to inflate during loss of

631

632

32 Hypobaria

­ ressurisation or bailout to maintain the pilot at a presp sure equivalent of 10 668 m (35 000 ft). Prior to take‐off, U‐2 pilots perform a minimum of a 60 minute pre‐breathe of 100% oxygen to wash out nitrogen from their body tissues and continue to breathe 100% throughout the flight. Despite these preventative measures, the risk for decompression sickness amongst US air force pilots has tripled from 2006, probably due to more frequent and longer periods of exposure to prolonged high altitude as a result of wartime operational needs. The decompression illness risk per flight was 0.076% from 1994 to 2005 but increased to 0.23% from 2006 to 2010 [124]. A 2013 study describes a neurological analysis of 102 pilots who fly U‐2 reconnaissance aircraft at an altitude of some 21 000 m (70 000 ft). These U‐2 pilots demonstrated an increase in volume (394%; P = 0.004) and number (295%; P