Cough: Causes, Mechanisms andTherapy

Edited by

Kian Fan ChungMD DSc FRCP

National Heart and Lung Institute

Imperial College and Royal Brompton & Harefield NHS Trust

London

UK

John G. WiddicombeMA DPhil DM FRCP

University of London and

116 Pepys Road

London

UK

Homer A. BousheyMD

Department of Medicine

University of California San Francisco

San Francisco

California

USA

Cough: Causes, Mechanisms and Therapy

Cough: Causes, Mechanisms andTherapy

Edited by

Kian Fan ChungMD DSc FRCP

National Heart and Lung Institute

Imperial College and Royal Brompton & Harefield NHS Trust

London

UK

John G. WiddicombeMA DPhil DM FRCP

University of London and

116 Pepys Road

London

UK

Homer A. BousheyMD

Department of Medicine

University of California San Francisco

San Francisco

California

USA

© 2003 by Blackwell Publishing LtdBlackwell Publishing, Inc., 350 Main Street, Malden, Massachusetts 02148-5020, USABlackwell Publishing Ltd, 9600 Garsington Road, Oxford OX4 2DQ, UKBlackwell Publishing Asia Pty Ltd, 550 Swanston Street, Carlton, Victoria 3053, Australia

The right of the Authors to be identified as the Authors of this Work has been assertedin accordance with the Copyright, Designs and Patents Act 1988.

All rights reserved. No part of this publication may be reproduced, stored in aretrieval system, or transmitted, in any form or by any means, electronic, mechanical,photocopying, recording or otherwise, except as permitted by the UK Copyright,Designs and Patents Act 1988, without the prior permission of the publisher.

First published 2003

Library of Congress Cataloging-in-Publication Data

Cough : causes, mechanisms, and therapy / edited by Kian Fan Chung, John G. Widdicombe,Homer A. Boushey.

p. ; cm.Includes bibliographical references and index.

ISBN 1-4051-1634-X1. Cough.

[DNLM: 1. Cough. WF 143 C854 2003] I. Chung, K. Fan, 1951– II. Widdicombe, John G. III. Boushey, Homer A.

RC741.5.C68 2003616.2 —dc21 2003011068

ISBN 1-4051-1634-X

A catalogue record for this title is available from the British Library

Set in 9/111/2 pt Sabon by SNP Best-set Typesetter Ltd., Hong KongPrinted and bound in Great Britain at CPI Bath, Bath

Commissioning Editor: Maria KhanManaging Editor: Rupal MaldeProduction Editor: Nick MorganProduction Controller: Kate Charman

For further information on Blackwell Publishing, visit our website:http://www.blackwellpublishing.com

v

Contents

Contributors, viiPreface, xi

Section 1: Introduction1 The clinical and pathophysiological challenge of cough, 3

Kian Fan Chung

2 Epidemiology of cough, 11Alyn H. Morice

3 A brief overview of the mechanisms of cough, 17John G. Widdicombe

Section 2: Cough in the Clinic4 Clinical assessment of cough, 27

Lorcan P.A. McGarvey

5 Measurement and assessment of cough, 39Kian Fan Chung

6 Cough sensitivity: the use of provocation tests, 49Rick W. Fuller

7 Causes, assessment and measurement of cough in children, 57Anne B. Chang

8 The quality of life in coughers, 75Richard S. Irwin, Cynthia L. French & Kenneth E. Fletcher

Section 3: Clinical Conditions with Cough9 Cough in lower airway infections, 83

Wee-Yang Pek & Homer A. Boushey

10 Cough and gastro-oesophageal reflux, 97Alvin J. Ing

11 Cough in postnasal drip, rhinitis and rhinosinusitis, 107Bruno C. Palombini & Elisabeth Araujo

12 Cough and airway hyperresponsiveness, 115Paul M. O’Byrne

13 Cough in chronic obstructive pulmonary disease, 125Kian Fan Chung & Peter M.A. Calverley

14 Cough in suppurative airway diseases, 137Robert Wilson

15 Cough in cancer patients, 147Sam H. Ahmedzai & Nisar Ahmed

CONTENTS

Section 4: Pathophysiology16 Sensory pathways for the cough reflex, 161

Stuart B. Mazzone, Brendan J. Canning & John G. Widdicombe

17 Neurogenesis of cough, 173Donald C. Bolser, Paul W. Davenport, Francis J. Golder, David M. Baekey,

Kendall F. Morris, Bruce G. Lindsey & Roger Shannon

18 Plasticity of vagal afferent fibres mediating cough, 181Marian Kollarik & Bradley J. Undem

19 Motor mechanisms and the mechanics of cough, 193Giovanni A. Fontana

20 Mucus hypersecretion and mucus clearance in cough, 207W. Michael Foster

21 Animal models of cough, 217Maria G. Belvisi & David J. Hele

Section 5: Therapy22 Mechanisms of actions of centrally acting antitussives —electrophysiological and

neurochemical analysis, 225Kazuo Takahama

23 Pharmacology of peripherally acting antitussives, 237Sandra M. Reynolds, Domenico Spina & Clive P. Page

24 Current and potential future antitussive therapies, 247Peter V. Dicpinigaitis

25 Placebo effects of antitussive treatments on cough associated with acute upper respiratory tract infection, 259Ronald Eccles

26 Mucoactive agents for the treatment of cough, 269Bruce K. Rubin

27 Management of cough, 283Kian Fan Chung

Index, 299

vi

Nisar AhmedAcademic Palliative Medicine Unit, Division ofClinical Sciences, Royal Hallamshire Hospital,Glossop Road, Sheffield S10 2JF, UK

Sam H. AhmedzaiAcademic Palliative Medicine Unit, Division ofClinical Sciences, Royal Hallamshire Hospital,Glossop Road, Sheffield S10 2JF, UK

Elisabeth AraujoOtorhinolaryngology and Respiratory Diseases,Universidade Federal do Rio Grande do Sul, 90020-090 Porto Alegre, RS Brazil

David M. BaekeyDepartment of Physiology and Biophysics, College ofMedicine, University of South Florida, Tampa, FL33612, USA

Maria G. BelvisiRespiratory Pharmacology Group, CardiothoracicSurgery, National Heart & Lung Institute, Faculty ofMedicine, Imperial College of Science, Technologyand Medicine, London SW3 6LY, UK

Donald C. BolserDepartment of Physiological Sciences, College ofVeterinary Medicine, University of Florida, PO Box100144, Gainesville, FL 32610-0144, USA

Homer A. BousheyProfessor of Medicine, Department of Medicine,University of California San Francisco, 505 ParnassusAvenue, San Francisco, CA 94143, USA

Peter M.A. CalverleyDepartment of Medicine, University of Liverpool, andUniversity Hospital Aintree, Liverpool, UK

Brendan J. CanningJohns Hopkins Asthma and Allergy Center, 5501Hopkins Bayview Circle, Baltimore, MD 21224, USA

Anne B. ChangPaediatric Respiratory Consultant, Department ofRespiratory Medicine, Royal Children’s HospitalFoundation, Herston, and Associate Professor ofPaediatrics, University of Queensland, Queensland4029, Australia

Kian Fan ChungNational Heart & Lung Institute, Imperial Collegeand Royal Brompton & Harefield NHS Trust,Dovehouse Street, London SW3 6LY, UK

Paul W. DavenportDepartment of Physiological Sciences, College ofVeterinary Medicine, University of Florida,Gainesville, FL 32610, USA

Peter V. DicpinigaitisAssociate Professor of Clinical Medicine, AlbertEinstein College of Medicine, and Director, IntensiveCare Unit, Jack D. Weiler Hospital of the AlbertEinstein College of Medicine, Bronx, New York, NY10461, USA

Ronald EcclesCommon Cold Centre, Cardiff School of Biosciences,Cardiff University, Cardiff CF10 3US, UK

Kenneth E. FletcherAssociate Professor of Psychiatry and the GraduateSchool of Nursing, University of MassachusettsMedical School, 55 Lake Avenue North, Worcester,MA 01655, USA

vii

Contributors

CONTRIBUTORS

Giovanni A. FontanaDipartimento di Area Critica Medico Chirurgica,Unità Funzionale di Medicina Respiratoria, Universitàdegli Studi di Firenze, Firenze, Italy

W. Michael FosterPulmonary and Critical Care Medicine, MSRBRoom #275, Research Drive, Duke UniversityMedical Center, Durham, NC 27710, USA

Cynthia T. FrenchUniversity of Massachusetts Memorial Health Care,55 Lake Avenue North, Worcester, MA 01655, USA

Rick W. FullerDirector of Science Funding, The Wellcome Trust, 183Euston Road, London NW1 2BE, UK

Francis J. GolderDepartment of Physiological Sciences, College ofVeterinary Medicine, University of Florida,Gainesville, FL 32610, USA

David J. HeleRespiratory Pharmacology Group, CardiothoracicSurgery, National Heart & Lung Institute, Faculty ofMedicine, Imperial College of Science, Technologyand Medicine, London SW3 6LY, UK

Alvin J. IngThoracic Physician, Clinical Senior Lecturer,University of Sydney, Concord Hospital, Concord,NSW 2139, Australia

Richard S. IrwinProfessor of Medicine, University of MassachusettsMedical School, 55 Lake Avenue North, Worcester,MA 01655-0330, USA

Marian KollarikJohns Hopkins Asthma and Allergy Center, 5501Hopkins Bayview Circle, Baltimore, MD 21224, USA

Bruce G. LindseyDepartment of Physiology and Biophysics, College ofMedicine, University of South Florida, Tampa, FL33612, USA

Stuart B. MazzoneHoward Florey Institute of Experimental Physiologyand Medicine, University of Melbourne, Melbourne,Victoria 3010, Australia

Lorcan P.A. McGarveySenior Lecturer/Consultant Physician, Department ofMedicine, The Queen’s University of Belfast, Belfast,Northern Ireland

Alyn H. MoriceHead of Academic Department of Medicine, CastleHill Hospital, Castle Road, Cottingham, EastYorkshire HU16 5JQ, UK

Kendall F. MorrisDepartment of Physiology and Biophysics, College ofMedicine, University of South Florida, Tampa, FL33612, USA

Paul M. O’ByrneFirestone Institute for Respiratory Health, St Joseph’sHospital, 50 Charlton Avenue East, Hamilton,Ontario L8N 4A6, Canada

Clive P. PageThe Sackler Institute of Pulmonary Pharmacology, 5thFloor Hodgkin Building, GKT School of BiomedicalSciences, Guy’s Campus, King’s College London,London SE1 1UL, UK

Bruno C. PalombiniPavilhao Pereira Filho, Irmandade da Santa Casa deMisericordia de Porto Alegre, Universidade Federal doRio Grande do Sul, Rua Prof. Annes Dias 285, 90020-090 Porto Alegre, RS Brazil

Wee-Yang PekVisiting Postgraduate Fellow in Pulmonary andAllergy/Immunology, Department of Medicine,University of California San Francisco, 505 ParnassusAvenue, San Francisco, CA, 94143, USA

viii

CONTRIBUTORS

Sandra M. ReynoldsThe Sackler Institute of Pulmonary Pharmacology, 5thFloor Hodgkin Building, GKT School of BiomedicalSciences, Guy’s Campus, King’s College London,London SE1 1UL, UK

Bruce K. RubinProfessor and Vice-Chair of Pediatrics, Professor ofBiomedical Engineering, Physiology andPharmacology, Department of Pediatrics, Wake ForestUniversity School of Medicine, Medical CenterBoulevard, Winston-Salem, NC 27157-1081, USA

Roger ShannonDepartment of Physiology and Biophysics, College ofMedicine, University of South Florida, Tampa, FL33612, USA

Domenico SpinaThe Sackler Institute of Pulmonary Pharmacology, 5thFloor Hodgkin Building, GKT School of BiomedicalSciences, Guy’s Campus, King’s College London,London SE1 1UL, UK

Kazuo TakahamaDepartment of Environmental and Molecular HealthSciences, Graduate School of Pharmaceutical Sciences,Kumamoto University, 5-1 Oe-honmachi, Kumamoto862-0973, Japan

Bradley J. UndemJohns Hopkins Asthma and Allergy Center, 5501Hopkins Bayview Circle, Baltimore, MD 21224, USA

John G. WiddicombeUniversity of London and 116 Pepys Road, LondonSW20 8NY, UK

Robert WilsonConsultant Physician, Royal Brompton Hospital, andReader, National Heart & Lung Institute at ImperialCollege of Science, Technology and Medicine, SydneyStreet, London SW3 6NP, UK

ix

xi

Preface

Cough is the most common symptom of airway andlung disease. More money is spent at the pharmacy on‘coughs and colds’ than on any other symptom exceptperhaps ‘aches and pains’. It can be a presenting symp-tom of more than 100 clinical conditions of the respira-tory system. At international meetings of, for example,the American Thoracic Society and the European Res-piratory Society, cough is one of the most frequentitems listed in the Proceedings’ indices. Surprisingly, ithas seldom been comprehensively reviewed, althoughmost textbooks of respiratory medicine contain a shortchapter on the subject, as do texts on respiratory pharmacology and therapeutics. Physiology textbookslargely ignore it.

In 1970, Salem and Aviado published a three-volumetext on Antitussive Agents [1] which has become a clas-sic that, while of limited scope, badly needs updating.In 1974, Korpas and Tomori published Cough andOther Respiratory Reflexes [2], a valuable text dealingmainly with animal studies from distinguished researchcentres in Eastern Europe. There is an excellent shortbook (published in 2001 but in Italian!) La Tosse Fisiopatologia e Clinica by Fontana, Lavorini, Panta-leo and Pistolesi [3]. There have been a number of pub-lished symposium proceedings dedicated to cough[4–9]. All these publications are somewhat restricted intheir approach and most need to be brought up to date.Two of the symposia [7,8], based on the 1st and 2nd International Symposia on Cough held in London in1996 and 2001, attracted much support and interest;this response was in large part the stimulus to the present book.

We therefore believe that this book is the first com-prehensive review, by internationally distinguished au-thors, of a subject of great importance to respiratorybasic science and medicine, and to health care. We havetried to make it inclusive, but are aware that there maybe gaps. We have particularly tried to review the sig-nificance of cough in the wider context of clinical med-icine, a subject which has seen impressive advances inunderstanding in recent years; and in the basic physio-logical mechanisms of cough. Understanding of the latter has been transformed recently by studies of the

plasticity of cough mechanisms at peripheral and cen-tral nervous levels, research that has important rele-vance to what happens in acute, subacute and chronicairways’ disease. Basic pharmacological studies arealso having a significant impact on the therapy ofcough, an influence which is likely to be dramaticallyextended in the near future. With such an approach andcoverage, we hope that our book will be helpful to awide readership of clinicians (particularly general practitioners), respiratory and paediatric physicians,respiratory physiologists and scientists interested incough and lung reflexes, and pharmacologists in searchof better treatments for cough.

In compiling this volume, we have ourselves learnt alot more but have also realized that there are large gapsboth at the clinical and basic level of understanding andtreatment of this common symptom. We hope that thereaders will be similarly challenged.

We are grateful for the cooperation, diligence and expertise of our many contributors; for the valuable advice and support of our colleague Tim Higenbottamat AstraZeneca UK; and for the skill and efficiency ofour publishers, Blackwell Publishing, in particularMaria Khan, Rupal Malde and Nick Morgan.

Fan ChungJohn Widdicombe

Homer Boushey

References

1 Salem H, Aviado DM, eds. Antitussive Agents. Interna-tional Encyclopedia of Pharmacology and Therapeutics,section 27, Volumes I–III. Oxford: Pergamon Press, 1970.

2 Korpas J, Tomori Z. Cough and other respiratory reflexes.In: Progress in Respiratory Research, Vol. 12. Basel: Karger,1978.

3 Fontana GA, Lavorini F, Pantaleo T et al., eds. La Tosse Fisiopatologia e Clinica. Pisa: Primula, 2001.

4 Berglund E, Nilsson BS, Mossberg B et al. Cough and ex-pectoration. Eur J Respir Dis 1980; 61 (Suppl. 110): 1–262.

5 Widdicombe JG, Korpas J, Salat D, eds. The cough reflex.Bull Eur Physiopath Respir 1987; 23 (Suppl. 10): 11S–76S.

PREFACE

6 Korpas J, Widdicombe JG, eds. Cough and related phenom-ena. Respir Med 1991; 85 (Suppl. A):1–68.

7 Widdicombe J, ed. Cough: methods and measurements.Pulm Pharmacol 1996; 9: 261–392.

8 Widdicombe J, Chung F, eds. Cough: pharmacology andtherapeutics. Pulm Pharmacol Ther 2002; 15: 185–338.

9 Korpas J, Widdicombe JG, eds. Cough: recent advances inunderstanding. Eur Respir Rev 2002; 85: 221–82.

xii

SECTION 1

Introduction

3

Introduction

Cough is a symptom that has been experienced by everyhuman and is an essential protective and defensive actwhose action secures the removal of mucus, noxioussubstances and infections from the larynx, trachea and larger bronchi. Coughing is the most efficientmechanism for clearing the upper airways, and can beconsidered to be an innate inbuilt defence mechanism.Impairment or absence of the coughing mechanism canbe harmful and even fatal in disease. On the other hand,cough may be the first overt sign of disease of the air-ways or lungs, when it represents more than a defencemechanism, and by its persistence becomes a helpfulpointer for both patient and physician of potential disease. Nearly all conditions affecting the respiratorysystem and some extrapulmonary conditions maycause cough, but to the physician it is most important to exclude the most serious conditions that needprompt treatment. Cough may be persistent in incur-able diseases such as in terminal lung cancer, whenother chronic symptoms are also present such as dys-pnoea and pain. Cough may be the most prominentsymptom complained of by patients with chronic respi-ratory disease such as asthma [1]. Excessive persistentcough may also be present in association with chronicnon-malignant disease with or without excessivemucus production. The effect of persistent cough itselfmay be harmful and deleterious to the patient by inter-fering with breathing, social activities and sleep, and bycausing deterioration in the quality of life and socialembarrassment, not to mention syncopal episodes, uri-nary incontinence, muscle ache, insomnia and fatigue.Thus, cough is a symptom with many facets: a protec-

tive mechanism for the lungs, a warning sign of disease,and a detrimental symptom when persistent.

With these several aspects of the problem of cough,the challenge to understand and adequately treat coughhas been and remains daunting, from both the investi-gational and clinical angles. Our pathophysiologicalunderstanding of the genesis of cough is incomplete,particularly with reference to disease. This volume willfocus on all the clinical, pathogenetic and treatment aspects of cough. To approach the challenge presentedby cough, this first chapter will discuss the followingthemes: (i) the normal cough; (ii) the significance ofcough in the community; (iii) the spectrum of coughpresenting to the clinician; (iv) the mechanisms of increased cough sensitivity; and (v) the treatment ofcough.

The normal cough

What is a normal cough pattern? How often do healthypeople cough? There are no clear data. Cough is notnecessarily an ‘abnormal’ symptom with clinical sig-nificance. But when does a normal pattern of cough be-comes an abnormal one? One presumes that a cough isnecessary to clear the airways of the mucus and fluid secretion from the airways (estimated to be 20–30mLover 24h), and the amount of airway secretion may berelated to the amount of exposure to daily irritants.City dwellers exposed to increasing pollution such asparticulate materials may cough more. The fact thathealthy people need to cough may be judged from theoccurrence of cough during large gatherings for a lec-ture or concert or during intervals of concerts, and a

3

The clinical and pathophysiologicalchallenge of coughKian Fan Chung

1

CHAPTER 1

rate of 2.5 coughs/min has been quoted for a gatheringof 100 people in a lecture room [2]. In a small group ofhealthy people in whom cough was monitored using aportable cough counter, the frequency of cough over a24-h period was found to be less than 16 coughs [3] and11 bursts of cough per 24h (range 1–34) in children [4],or a range of 0–141 (median of 10) coughs per day ifchildren with respiratory infections are included [5].We do not know whether the normal protective coughreflex occurs during sleep, particularly during rapid eyemovement sleep; presumably it does, since there is noother obvious protective mechanism that can be in-voked. The smoker with a chronic morning cough usu-ally considers his or her cough to be normal, necessaryto clear the airways of the excessive secretions inducedby cigarette smoke. However, to study the normality ofcough, it is important to exclude the possibility of anenhanced cough reflex response which may be inducedby cigarette smoking or even perhaps daily exposure topollutants. A cough may be provoked by a tussivestimulus, and the likelihood of this cough occurring isincreased by the presence of an enhanced cough reflex.The definition of normality of cough experience needsto be studied from the point of view of the event, thetype of cough and the cough sensitivity reflex, so thatthe significance of cough prevalence studies can be determined.

Prevalence of cough in the community

In several epidemiological surveys, persistent cough isreported as a symptom that affects a large proportionof the general population, being prevalent for examplein 18% of the US population, in up to 16% of a south-east English population and in 11% of the Swedishpopulation [6–8]. In these surveys, there has been no es-timate of how many of these reported symptoms havereceived medical attention, and whether the personconsiders the cough to be a ‘normal’ symptom. In inves-tigations of the Swedish part of the European Commu-nity Respiratory Health Survey (ECRHS) [9,10], ahigher prevalence of nocturnal and non-productivecough was reported in females than in males. In a morerecent trans-European survey of ECRHS, about one-third of subjects reported that they had been woken upby an attack of cough in the last 12 months, and about20% reported a non-productive or productive coughduring the winter months [11]. There was again a pre-

ponderance of females with nocturnal and non-productive cough, and nocturnal cough was related toasthma, tobacco smoking, exposure to environmentaltobacco smoke and obesity. However, there has been no detailed community survey of the potential diseasesunderlying the cough reported.

Exposure to pollutants or environmental irritantsappears to be important, judging from the epidemio-logical information linking dry or productive coughwith outdoor air pollution, particularly in children. Inadults and schoolchildren, productive cough or chronicnocturnal dry cough has been associated with levels ofthe PM10 particulates [12,13]. Increases in levels ofPM10 are also related with reductions in peak expirat-ory flowsandwithincreasedreportingof cough,sputumproduction and sore throat in children with or withoutasthma [14]. Living close to heavy traffic may be associ-ated with increased asthma symptoms and longstand-ing cough compared to not living close to heavy traffic[15]. In the Italian Po Valley district, the increase in airpollution has been associated with an increase in coughincidence amongst females but not males [16]. Noctur-nal cough in relation to indoor exposure to cat allergenswas observed not only in sensitized but also in non-sensitized subjects [17]. It is possible that populationsurveys are picking up more cough in the communityfrom subjects being exposed to environmental pollu-tants and allergens, but we do not know whether thesefactors induce or sensitize cough. It would seem mostimportant to determine the prevalence of sensitivity ofthe cough reflex in the community and relate this to thepresence of cough, and thus determine levels of normalvs. abnormal cough in the community. For example, exposure of workers to capsaicin in a chilli processingfactory was associated with an increase in self-reporting of cough but no increase in capsaicin coughresponsiveness in these workers, indicating a ‘normal’response to exposure to capsaicin [18].

When evaluating any community surveys, it is worthremembering that up to 25–30% of the population areusually cigarette smokers. In a survey in southeast Eng-land, up to 16% of 9077 responders had cough everyday or half of the days of the year, and up to 13.2% hadsputum every day or on half the days of the year; in thiscohort, 54% were current cigarette smokers [7]. Thechronic cough of cigarette smokers is a well-known fea-ture, and is accompanied by hypersecretion of mucusand possibly by slowing of mucociliary clearance.Chronic bronchitis has been defined as the expectora-

4

THE CLINICAL AND PATHOPHYSIOLOGICAL CHALLENGE

tion of sputum on most days during at least three con-secutive months per year over two successive years.Apart from mucus stimulation of cough, chronic smok-ing may increase airway sensitivity to capsaicin [19],and patients with chronic obstructive airways diseasecaused by cigarette smoking show an increased tussiveresponse to capsaicin [20]. Cigarette smokers withchronic cough do not usually seek medical help unlessthe pattern of their cough changes.

More detailed analysis of the factors that may causecough, its severity and an assessment of the distributionof cough responsiveness is needed in epidemiologicalsurveys. This indicates the need to have better methodsof recording the severity of cough, objective measure-ments and a rapid easy and safe method of measuringthe cough sensitivity reflex. There are no epidemiologi-cal data on the distribution of cough responsiveness inthe population, and this is important to know.

Cough presenting to the clinician

Cough is one of the most frequent reasons for seeking a consultation with a doctor. In the US, a national medical case survey reported that in 1991 cough wasthe most common complaint for which patients sought medical attention, and the second most com-mon reason for a general medical consultation [21].Much of this is likely to be due to cough occurring withthe common cold, which is a self-limiting disease. In theUS, for a chest specialist practice, patients with persis-tent chronic cough probably account for 10–38% ofoutpatient practice [22,23]. In addition, treatment of cough is a substantial proportion of health care ex-penditure. In the UK, about 3 million prescriptions forcough preparations are given annually by general prac-titioners, representing a cost of £1.9 million [24]. Thisis, however, an underestimate of the use of coughpreparations since cough mixtures can be obtainedover the counter.

Faced with a patient with a chronic cough, utilizationof an anatomical diagnostic protocol has been advo-cated [25], and application of such a protocol oftenleads to a cause identified in 88–100% of cases, withtreatment success rates of between 84 and 98%[22,23]. Such figures are perhaps overoptimistic, de-pendent on the referral practice of the clinic. Othercough clinics report that in 12–31% of their patients no underlying cause of the cough can be found

despite investigation and empirical treatment for gastro-oesophageal reflux, postnasal drip and asthma[26–28]. In patients referred with a chronic cough inwhom the diagnosis of cancer has been excluded, a pro-portion will respond to the institution of inhaled corti-costeroid therapy, on the basis of which one could beconfident of the diagnosis of asthma, cough-variantasthma or eosinophilic bronchitis as underlying thecough [29]. Whether the eosinophil is the pathogenicfactor in the induction of the cough remains unclear.Other prevalent diagnoses that are reported to ‘cause’cough are chronic rhinosinusitis often with a postnasaldrip, gastro-oesophageal reflux, chronic bronchitis,bronchiectasis or taking an angiotensin-converting enzyme inhibitor medication [30]. There appears to be little value in separating chronic dry cough fromchronic cough with excessive sputum production interms of the diagnostic yield since the frequency of thecauses of both symptoms appear to be similar [31].Postnasal drip, asthma, gastro-oesophageal reflux andbronchitis were the most frequent cause of a chronicproductive cough. When a ‘cause’ is established, one as-sumes that the specific treatment of that cause shouldrelieve the cough, and it sometimes does not. This raises the issue as to whether the ‘cause’ is really in-volved in the genesis of cough, or whether the treatmentis really effective for the ‘cause’.

The role of postnasal drip in causing cough meritssome discussion [32]. Postnasal drip is defined as thepresence of secretions at the back of the throat associ-ated with frequent throat-clearing. How could it trig-ger cough when there is no vagal afferent innervation ofthe throat and there is no evidence for aspiration of se-cretions from the sinuses into the lungs in patients withsinusitis [33]? Patients with gastro-oesophageal refluxand cough may not respond to treatment with protonpump inhibitors that suppress acid production, per-haps because other factors in the refluxate such as thecontent of pepsin may be important in the pathogenesisof cough. In spite of these uncertainties, there is a groupof chronic coughers in whom there appears to be no as-sociated cause. There are other causes that remainunidentified. For example, a condition of eosinophilicbronchitis without bronchial hyperresponsiveness hasrecently been established as a cause of chronic cough[34].Bordetella pertussis infection has been flagged as apotential cause, when nasopharyngeal aspirates wereexamined for the presence of Bordetella pertussis bypolymerase chain reaction [35].

5

CHAPTER 1

The increased tussive response

Many patients with a chronic cough have a sensitizedcough reflex response, and patients whose cough disap-pears or is controlled with appropriate treatments nor-malize their cough reflex response [36,37]. A sensitizedcough reflex is also described in many other conditionssuch as asthma, pulmonary fibrosis and chronic ob-structive pulmonary disease (COPD) [20,38,39]. How-ever, the prevalence of sensitized cough reflex in thepopulation and its association with different types ofcough are still to be determined. It is not necessary forthe cough to be ‘dry’ in order for the cough to be sensi-tized since patients with COPD with sputum produc-tion can demonstrate sensitization.

The sensitized cough reflex has various clinical con-notations. Often, there is a history that the cough wasfirst triggered following an upper respiratory virus infection or by a cold, and that the cough persisted, despite the clearing of the infective period, which indicates the notion of a central cough sensitizationprocess. Such patients usually present with a history ofpersistent cough which may have lasted for many years.The cough is described as an irritation in the throat orupper chest, ranging from a feeling of wanting to coughto severe episodes of violent coughing that cannot bestopped. The severity of the cough may be variable overperiods of months with mild or asymptomatic periodsinterspersed with periods of severe symptoms. Typi-cally, the cough may be triggered by stimuli such aschanges in temperature of inhaled air, taking a deepbreath, talking over the telephone, laughing, eatingcrumbly food, certain smells or perfumes and lying in asupine posture. Often, any suspected associated causehas been treated with no success in controlling thecough.

Pathophysiology of cough

Understanding of the cough response and of howcough may become persistent is of the utmost impor-tance. The anatomy of the cough reflex has been dis-sected extensively, with the airway afferents andreceptors, a central pathway and efferent pathways de-fined. The cough reflex is subserved mainly by vagalprimary afferent nerves such as bronchopulmonaryrapidly adapting receptors (RARs) which can beevoked by mechanical stimulation and deformation of

the airway epithelium such as particulate matter ormucus, and by airway smooth muscle contraction in-duced by constrictor agents [40,41]. These are pre-dominantly present in the larynx, trachea and carina.Activation of bronchopulmonary C fibres by chemicalssuch as bradykinin and capsaicin can evoke cough inconscious animals and humans [42], although thesemay also activate RARs. C fibres and RARs project todifferent subnuclei in the nucleus tractus solitarius inthe brainstem, considered to be a cough centre, al-though these are not anatomically discrete, and theconcept of a cough centre is not universally accepted.Various second-order neurones project to other nucleiassociated with the regulation of breathing. Integrationof the various inputs occur centrally. For example,slow-adapting receptor afferent input may have a facil-itatory effect on the genesis of cough [43].

This work has paved the way for an understanding ofthe cough response, but there is more to be understoodregarding the cough reflex in the clinical situation. Onecould take the cough associated with gastro-oesophageal reflux as an example of how one couldwork out how the refluxate could induce cough [44].Protons within the refluxate or other components of the refluxate could directly stimulate cough receptorsin the upper airways; alternatively, there could be a dis-tal oesophageal–tracheobronchial reflex mechanismwhich would be mediated through the usual pathwayof the cough receptor. The biochemical counterparts ofthe cough reflex arc remain to be elucidated, and suchinformation may provide therapeutic targets.

One defining concept of the chronic cougher is thatof an enhanced cough reflex in patients with persistentcough as being the fundamental abnormality. Throughunderstanding how this occurs, more targeted specificsuppressants of the cough response, irrespective of thecause, can be devised. The process of cough ‘sensitiza-tion’ remains unclear, and this may invoke both ‘peri-pheral’ (i.e. in the airways) and/or ‘central’ (i.e. in thebrain) mechanisms, according to our current under-standing of the cough reflex pathways. Central sensiti-zation may occur by integration from various sensorynerve subtypes in the central nervous system to initiateexaggerated reflexes and sensation [45]. Substance Pmay be involved as an important central mechanism forsensitization of the cough reflex, and its persistence. Ina model of allergic inflammation, neuroplastic changesin the response of vagal primary afferent neurones weredescribed, such that Ad mechanosensor RARs released

6

THE CLINICAL AND PATHOPHYSIOLOGICAL CHALLENGE

substance P, when under normal conditions, they donot [46]. Substance P in the nucleus tractus solitariuscan increase bronchopulmonary C fibre reflex activity[47]. Peripheral mechanisms that can heighten coughreflex mechanisms have been mainly envisaged as an effect of altered environment of the cough receptorsuch as the release of inflammatory mediators such asprostaglandins or bradykinin that could enhance theresponse of the cough receptor [48,49]. Alternatively,there may be direct interactions between inflammatorycells and the cough receptor such that the threshold ofthe cough receptor is altered.

It has not been possible to determine whether thecough receptor itself is abnormal in terms of the transduction of the stimulatory signals. Examinationof neural afferent profiles in the airways of persistentcoughers has been limited. A recent study of airwaybiopsies reported an increase in total intraepithelialnerve density together with augmented staining for theneuropeptide calcitonin gene-related peptide (CGRP)but not for substance P [27]. However, the significanceof the increase in intraepithelial nerves is unclear sincewe do not know whether any of these profiles are in-deed cough receptors. Detailed localization of cough fi-bres and their ultrastucture, particularly in an airwaythat demonstrates augmented tussive response, is cru-cial to undertake.

Approach to the treatment of cough

The treatment of cough would be more efficacious ifone could understand the mechanisms of the phenome-non of cough sensitivity, since suppression of theheightened responses would control the urge to cough.The search for new antitussives has progressed alongseveral lines, from studies of airway afferent nerve ac-tivity in vitro (with the disadvantage that the activityobserved is not necessarily that of a ‘cough’ receptor) tostudies in whole organisms using tussive challenges,and more rarely using patients with enhanced cough re-flex or with chronic cough. The predictive value of anti-tussives tested in animals to their effectiveness inhumans is not known. Some investigators study ro-dents such as mice that cannot cough, while the bestmodel for studying cough is the guinea-pig which pro-duces the explosive noise similar to the human cough.The use of the capsaicin or citric acid challenge in nor-mal subjects cannot be predictive of antitussive activity

since normal volunteers do not possess an enhancedcough reflex. Even if we do not currently have an un-derstanding of the enhanced cough reflex, it is impera-tive that novel antitussives be tested in patients with anenhanced cough reflex. The aim would be to inhibit theenhanced component of the cough reflex to maintainthe ‘normal’ cough reflex. This task would be easier ifthe enhanced cough reflex was superimposed on thenormal reflex by entirely different mechanisms ratherthan representing an amplification of ‘normal’ mechanisms.

Many targets can be defined if one considers treatingthe ‘cause’ of the cough first. Treatment of the causes ofcough may include anti-inflammatory approaches suchas the cough associated with eosinophilic inflammationin the airways as in asthma, cough-variant asthma oreosinophil bronchitis [29]. Inhaled corticosteroids areparticularly effective here, although the eosinophil as acausation of the cough is not entirely proven. Coughand enhanced cough with a neutrophilic inflammation[20] which do not respond to inhaled corticosteroids,and approaches that suppress the neutrophilic re-sponse or neutrophil activation could be helpful. Thetreatment of cough associated with gastro-oesophagealreflux with acid suppressants such as histamine H2-receptor antagonists or proton pump inhibitors, or associated with rhinosinusitis with antihistamines orwith nasal steroid drops, is not always successful. Thismay mean either that these conditions are not the causeof the cough or that a central mechanism of cough sen-sitization has occurred.

Partly because there is no apparent cause associatedwith a persistent cough in many patients, it is clear that the development of effective antitussives (i.e. drugsthat specifically block cough whatever the cause) isneeded. Currently, the most commonly used anti-tussives are the centrally acting opiates such as codeine,dihydrocodeine or pholcodeine. Sometimes for the intractable cough in terminal disease, morphine or diamorphine is used, with the advantage that these alsopossess analgesic properties. However, at their effectivedoses, these antitussives cause drowsiness, nausea andvomiting, and constipation, and often cause physicaldependence. Development of new antitussives has oc-curred on a number of fronts. An improved under-standing of the cough reflex pathways and of themechanisms by which the reflex is enhanced has led tothe identification of potential targets [50]. Finally, theassessment of antitussives in clinical studies has also

7

CHAPTER 1

been improved with the use of cough reflex challengeprotocols such as capsaicin or citric acid, or of measur-ing the cough itself in ambulatory patients [51], but ex-perience needs to be gathered.

It would be useful in clinical practice to divide the an-titussive drugs into those that are acute relievers ofcough such as the opiates; and those that can preventcough (‘preventors’) when taken on a regular basis,such as corticosteroid therapy for eosinophil-associated cough or proton pump inhibitors for gastro-oesophageal reflux-associated cough (these preventorswould be specific to the cause of cough).

Conclusion

This chapter has provided an overview of the clinicaland pathophysiological challenges of cough, and high-lighted areas of unmet needs and of future inves-tigations. Because of the high prevalence of cough,physicians commonly encounter patients presentingwith a cough problem for diagnosis and management.The potentially wide diagnostic possibilities make itnecessary for a systematic protocol to be used to ensurethat diagnoses are not missed. New causes of chronicpersistent cough will certainly be discovered. The toolsto determine the severity of a chronic cough are beingdeveloped, and these can be applied to the clinic as wellas to the evaluation of new antitussive therapies. Incough pathophysiology, an understanding of the path-ways that enhance the cough reflex is needed, and whatbiochemical or structural abnormalities constitute ahypertussive cough receptor should be determined. Weneed to apply the knowledge obtained from studyingthe cough reflex to clinical practice, which includes thediscovery of new treatments for suppressing cough.

References

1 Osman LM, McKenzie L, Cairns J, Friend JA, Godden DJ,Legge JS etal. Patient weighting of importance of asthmasymptoms. Thorax 2001; 56: 138–42.

2 Loudon RG. Cough in health and disease. Current re-search in chronic obstructive lung disease. In: Proceedingsof the 10th Emphysema Conference. US Department ofHealth, Education & Welfare, 1967: 41–53.

3 Hsu J-Y, Stone RA, Logan-Sinclair R, Worsdell M, BusstC, Chung KF. Coughing frequency in patients with persis-

tent cough using a 24-hour ambulatory recorder. EurRespir J 1994; 7: 1246–53.

4 Munyard P, Bush A. How much coughing is normal? ArchDis Child 1996; 74: 531–4.

5 Chang AB, Phelan PD, Robertson CF, Newman RG,Sawyer SM. Frequency and perception of cough severity. JPaediatr Child Health 2001; 37: 142–5.

6 Barbee RA, Halonen M, Kaltenborn WT, Burrows B. Alongitudinal study of respiratory symptoms in a commu-nity population sample. Correlations with smoking, al-lergen skin-test reactivity, and serum IgE. Chest 1991; 99:20–6.

7 Cullinan P. Persistent cough and sputum: prevalence andclinical characteristics in south east England. Respir Med1992; 86: 143–9.

8 Lundback B, Nystrom L, Rosenhall L, Stjernberg N. Ob-structive lung disease in northern Sweden: respiratorysymptoms assessed in a postal survey. Eur Respir J 1991;4: 257–66.

9 Bjornsson E, Plaschke P, Norrman E, Janson C, LundbackB, Rosenhall A etal. Symptoms related to asthma andchronic bronchitis in three areas of Sweden. Eur Respir J1994; 7: 2146–53.

10 Ludviksdottir D, Bjornsson E, Janson C, Boman G. Habitual coughing and its associations with asthma, anxiety, and gastroesophageal reflux. Chest 1996; 109:1262–8.

11 Janson C, Chinn S, Jarvis D, Burney P. Determinants ofcough in young adults participating in the European Com-munity Respiratory Health Survey. Eur Respir J 2001; 18:647–54.

12 Zemp E, Elsasser S, Schindler C, Kunzli N, Perruchoud AP,Domenighetti G etal. Long-term ambient air pollutionand respiratory symptoms in adults (SAPALDIA study).The SAPALDIA Team. Am J Respir Crit Care Med 1999;159: 1257–66.

13 Braun-Fahrlander C, Wuthrich B, Gassner M, Grize L,Sennhauser FH, Varonier HS etal. Validation of a rhinitissymptom questionnaire (ISAAC core questions) in a population of Swiss school children visiting the schoolhealth services. SCARPOL team. Swiss study on child-hood allergy and respiratory symptom with respect to airpollution and climate. International study of asthma andallergies in childhood. Pediatr Allergy Immunol 1997; 8:75–82.

14 Vedal S, Petkau J, White R, Blair J. Acute effects of ambi-ent inhalable particles in asthmatic and nonasthmatic chil-dren. Am J Respir Crit Care Med 1998; 157: 1034–43.

15 Montnemery P, Bengtsson P, Elliot A, Lindholm LH, Nyberg P, Lofdahl CG. Prevalence of obstructive lung diseases and respiratory symptoms in relation to living environment and socio-economic group. Respir Med2001; 95: 744–52.

8

THE CLINICAL AND PATHOPHYSIOLOGICAL CHALLENGE

16 Viegi G, Pedreschi M, Baldacci S, Chiaffi L, Pistelli F, Modena P etal. Prevalence rates of respiratory symptomsand diseases in general population samples of North andCentral Italy. Int J Tuberc Lung Dis 1999; 3: 1034–42.

17 Gehring U, Heinrich J, Jacob B, Richter K, Fahlbusch B,Schlenvoigt G etal. Respiratory symptoms in relation toindoor exposure to mite and cat allergens and endotoxins.Indoor Factors and Genetics in Asthma (INGA) StudyGroup. Eur Respir J 2001; 18: 555–63.

18 Blanc P, Liu D, Juarez C, Boushey HA. Cough in hot pep-per workers. Chest 1991; 99: 27–32.

19 Bergren DR. Enhanced lung C-fiber responsiveness in sen-sitized adult guinea pigs exposed to chronic tobaccosmoke. J Appl Physiol 2001; 91: 1645–54.

20 Doherty MJ, Mister R, Pearson MG, Calverley PM. Capsaicin responsiveness and cough in asthma and chronic obstructive pulmonary disease. Thorax 2000; 55:643–9.

21 Schappert SM. National ambulatory medical care survey: 1991: Summary. In: Vital and Health Statistics,Publication No. 230. US Department of Health andHuman Services, 1993: 1–20.

22 Irwin RS, Curley FJ, French CL. Chronic cough: the spec-trum and frequency of causes, key components of the diag-nostic evaluation, and outcome of specific therapy. AmRev Respir Dis 1990; 141: 640–7.

23 Irwin RS, Carrao WM, Pratter MR. Chronic persistentcough in the adult: the spectrum and frequency of causesand successful outcome of specific therapy. Am Rev RespirDis 1981; 123: 413–17.

24 British Thoracic Society. The Burden of Lung Disease.London: BMJ Publishing, 2001.

25 Irwin RS, Madison JM. The diagnosis and treatment ofcough. N Engl J Med 2000; 343: 1715–21.

26 Poe RH, Harder RV, Israel RH, Kallay MC. Chronic persistent cough. Experience in diagnosis and outcomeusing an anatomic diagnostic protocol. Chest 1989; 95:723–8.

27 O’Connell F, Springall DR, Moradoghli-Haftvani A,Krausz T, Price D, Fuller RW etal. Abnormal intraepithe-lial airway nerves in persistent unexplained cough? Am JRespir Crit Care Med 1995; 152: 2068–75.

28 McGarvey LP, Heaney LG, Lawson JT, Johnston BT, Scally CM, Ennis M etal. Evaluation and outcome of patients with chronic non-productive cough using a comprehensive diagnostic protocol. Thorax 1998; 53:738–43.

29 Chung KF. Assessment and measurement of cough: thevalue of new tools. Pulm Pharmacol Ther 2002; 15:267–72.

30 Chung KF, Lalloo UG. Diagnosis and management ofchronic persistent dry cough. Postgrad Med J 1996; 72:594–8.

31 Smyrnios NA, Irwin RS, Curley FJ. Chronic cough with ahistory of excessive sputum production. The spectrum andfrequency of causes, key components of the diagnosticevaluation, and outcome of specific therapy. Chest 1995;108: 991–7.

32 Campanella SG, Asher MI. Current controversies: sinusdisease and the lower airways. Pediatr Pulmonol 2001;31: 165–72.

33 Bardin PG, Van Heerden BB, Joubert JR. Absence of pulmonary aspiration of sinus contents in patients withasthma and sinusitis. J Allergy Clin Immunol 1990; 86:82–8.

34 Brightling CE, Pavord ID. Eosinophilic bronchitis: an im-portant cause of prolonged cough. Ann Med 2000; 32:446–51.

35 Birkebaek NH, Kristiansen M, Seefeldt T, Degn J, MollerA, Heron I etal. Bordetella pertussis and chronic cough inadults. Clin Infect Dis 1999; 29: 1239–42.

36 Choudry NB, Fuller RW. Sensitivity of the cough reflex in patients with chronic cough. Eur Respir J 1992; 5:296–300.

37 O’Connell F, Thomas VE, Pride NB, Fuller RW. Capsaicincough sensitivity decreases with successful treatment ofchronic cough. Am J Respir Crit Care Med 1994; 150:374–80.

38 Lalloo UG, Lim S, DuBois R, Barnes PJ, Chung KF. Increased sensitivity of the cough reflex in progressive systemic sclerosis patients with interstitial lung disease.Eur Respir J 1998; 11: 702–5.

39 Doherty MJ, Mister R, Pearson MG, Calverley PM. Capsaicin induced cough in cryptogenic fibrosing alveoli-tis. Thorax 2000; 55: 1028–32.

40 Widdicombe JG. Neurophysiology of the cough reflex.Eur Respir J 1995; 8: 1193–202.

41 Karlsson J-A, Sant’Ambrogio G, Widdicombe J. Afferentneural pathways in cough and reflex bronchoconstriction.J Appl Physiol 1988; 65: 1007–23.

42 Fox AJ. Modulation of cough and airway sensory fibres.Pulm Pharmacol 1996; 9: 335–42.

43 Hanacek J, Davies A, Widdicombe JG. Influence of lungstretch receptors on the cough reflex in rabbits. Respira-tion 1984; 45: 161–8.

44 Ing AJ, Ngu MC. Cough and gastro-oesophageal reflux.Lancet 1999; 353: 944–6.

45 Canning BJ. Interactions between vagal afferent nervesubtypes mediating cough. Pulm Pharmacol Ther 2002;15: 187–92.

46 Myers AC, Kajekar R, Undem BJ. Allergic inflammation-induced neuropeptide production in rapidly adapting afferent nerves in guinea pig airways. Am J Physiol LungCell Mol Physiol 2002; 282: L775–L781.

47 Mutoh T, Bonham AC, Joad JP. Substance P in the nucleusof the solitary tract augments bronchopulmonary C fiber

9

CHAPTER 1

reflex output. Am J Physiol Regul Integr Comp Physiol2000; 279: R1215–R1223.

48 Nichol GM, Nix A, Barnes PJ, Chung KF. Enhancement of capsaicin-induced cough by inhaled prostaglandin F2a: modulation by beta-adrenergic agonist and anti-cholinergic agent. Thorax 1990.

49 Fox AJ, Lalloo UG, Bernareggi M, Belvisi MG, Chung KF,

Barnes PJ. Bradykinin and captopril-induced cough inguinea-pigs. Nature Med 1996; 2: 814–7.

50 Chung KF. Cough: potential pharmacological devel-opments. Expert Opin Investig Drugs 2002; 11:955–63.

51 Chung KF. Methods of assessing cough and antitussives inman. Pulm Pharmacol 1996; 9: 373–7.

10

11

Introduction

Cough is a universal experience common to us all. It is also the commonest symptom for which medical advice is sought [1]. For the purpose of classificationcough may be divided into defined, acute, self-limitingepisodes and chronic persistent cough. This distinctionis clinically useful since the aetiology of the two syn-dromes is very different. An arbitrary cut-off of 8 weeksis taken to separate acute from chronic cough.

Acute cough

The majority of consultations with acute cough are dueto viral respiratory tract infections. The UK MorbidityStatistics from General Practice survey [2] suggests anaverage of two consultations per year for every adultand six consultations for each child under 4. Of totalprimary care workload Morrell [3] found cough to bethe presenting complaint in 527 per 1000 consultationswith new symptoms. Even this staggering statistic, thatover half of new consultations are due to cough, under-estimates the true socioeconomic impact of of thissymptom. Many of us do not consult a physician forviral cough, and if driven to treatment self-medicatewith over-the-counter products of dubious efficacy.The market in cough and cold products in the US is esti-mated at over 2 billion dollars.

This enormous morbidity is caused by a wide arrayof viral pathogens including influenza, parainfluenza,rhinovirus, adenovirus, respiratory syncytial virus andthe respiratory corona virus [4]. All of these virusesshare a common short incubation period of between 1

and 4 days. Because of its ability to undergo antigenicshift and genetic recombination influenza occurs in epi-demics. Why these epidemics occur during the wintermonths is unknown but may be related to cooling of theairway epithelium decreasing host defence [5]. Parain-fluenza viruses differ from influenza viruses in that theyare much more antigenically stable and of the four sub-types types one and three are particularly important,causing serious lower respiratory tract infections,croup and tracheobronchitis in infants and young chil-dren. In all, parainfluenza viruses are thought to be re-sponsible for a fifth of all non-bacterial respiratorytract disease in childhood. Since immunity to reinfec-tion is only transient it is one of the commonest causesof the typical infective cough which plagues familieswith small children.

Respiratory syncytial virus (RSV) is the single mostimportant aetiological agent in cough in infancy [6].RSV occurs in epidemics, usually between the autumnand early spring. Each outbreak lasts between 2 and 3months and can involve as many as half of all familieswith children. There is also a second peak of infectivityin the elderly. Unfortunately, immunity to RSV infec-tion is transient so reinfection is common.

Of the over 100 serotypes of adenovirus eight are im-portant in the production of a syndrome of severe acutecough. Immunity following infection is good but sinceso many serotypes are associated with cough the likeli-hood of repeated adenovirus-induced cough is high.The common cold or rhinovirus again represents over100 serotypes and is the major cause of mild upper res-piratory tract infection and cough in both children andadults. Unsurprisingly, rhinovirus immunity is verypoor. Coronaviruses are similar in clinical presentation

11

Epidemiology of coughAlyn H. Morice2

CHAPTER 2

to rhinoviruses and may be responsible for between 5and 10% of human coughs and colds.

At the age of 15 a gender difference appears in consultation rates reported in the morbidity statistics in general practice. During the reproductive years consultations are twice for women what they are formen. After the age of 55 consultation rates in both menand women become equal again. Whilst this could beexplicable by societal differences in that women mayconsult more, there are intriguing clues to the fact that women may have a heightened cough reflex com-pared with men. Women exposed to cough challengewith protussive agents cough twice as much as men or,conversely, cough the same amount at a lower dose[7,8]. This could be an artefact of inhaling drug into a smaller lung volume but women are also over-represented in those patients who develop angiotensin-converting enzyme (ACE) inhibitor cough [9]. In themajority of cough clinics women outnumber men andwhen tested with cough challenge they exhibit a height-ened cough response [10]. It is possible that womenhave an intrinsically heightened cough reflex comparedwith men.

Chronic cough

Several surveys have attempted to quantify the inci-dence of chronic cough within populations [11–14].The objective of most of these surveys is to assess thesymptomatology associated with cigarette smokingand so questions are directed towards discovering theprevalence of chronic bronchitis. This leads to confu-sion in the data analysis between cough and chronicsputum production. Despite these methodologicalshortcomings it is clear that chronic cough is a commonproblem with reported prevalence varying between 40and 6%. In the UK four general practices undertook apostal and telephone survey of over 11000 patients.Cough was reported every day or on over half of thedays of the year by 14% of males and 10% of females.A worldwide study from 16 countries surveyed 18277subjects aged between 20 and 48. Nocturnal cough waspresent in 30%, productive cough in 10%, and non-productive cough in 10%. There was a clear dose-related effect of cigarette smoking on cough. The veryhigh prevalence of nocturnal cough arose because apositive response to the question ‘Have you beenwoken by an attack of coughing at any time in the last

12 months?’ was taken as indicating the symptom.However, since acute cough could also lead to a positiveresponse, significant nocturnal cough was probablyoverestimated in this study. Another problem is in thechoice of age group in which to study cough. Chroniccough is present in older subjects as may be seen fromthe average age of patients seen in cough clinics (Table2.1). Whatever the failings of individual surveys chronic cough is clearly a very common symptomwhich although associated with considerable morbid-ity goes largely unheeded.

Those patients presenting to specialist cough clinics,however, represent a subgroup of this population.Most smokers assume that their morning cough reflectstheir smoking habit and do not consult. Conditionssuch as chronic bronchitis are therefore grossly under-represented, even though they cause considerable mor-bidity within the population. Similarly, such tertiaryreferral clinics are unlikely to represent the true preva-lence of conditions such as asthma as a cause of chroniccough, since at least in European practice, a therapeutictrial of antiasthma medication is usually performed bythe primary physician. If such therapy is successful the patient remains in primary care and the prevalenceof the condition is hidden from the tertiary referral centre.

The three common causes of chronic cough

All of the reported series from tertiary referral centresidentify the same three common causes of cough. Thisdiagnostic triad underlies the vast majority of chroniccough seen within the population. The problem of thehigh morbidity from chronic cough is the failure of doc-tors, both generalists and specialists, to recognize thatcough as an isolated symptom may be generated fromany of three anatomical areas. The relative incidence ofasthma or asthma-like syndrome, gastro-oesophagealreflux and rhinitis is illustrated in Table 2.1 which liststhe reported prevalence.

The individual reports from cough clinics illustrate awide variety in the prevalence of each syndrome. Thisvariation may represent different patient populations,or the different prevalence of the underlying diseasessuch as asthma in each individual population and dif-ferent diagnostic methods. A further source of error isthe criteria for diagnosis. Some clinics only accept diag-nosis when a therapeutic trial of appropriate treatmenthas been successful [15]. Other clinics report a positive

12

EPIDEMIOLOGY OF COUGH

diagnosis as a positive result during investigation, lead-ing to apparent multiple diagnoses [16].

There is no doubt that a positive result in terms of di-agnosis and therapy can suggest multiple causes ofcough in individual patients. I believe, however, thatthis illustrates the plasticity of the cough reflex. Low-grade, subclinical, cough may be present in an individ-ual but only become apparent when an additionalprovoking feature occurs. A good example of this phe-nomenon is ACE inhibitor cough, where ACE in-hibitors alter the cough reflex sensitivity as shown by shift in the capsaicin dose–response curve [27] andmay reveal previous low-grade cough due to gastro-oesophageal reflux or rhinitis [28].

Cough-predominant asthmaThe term cough-predominant asthma has been intro-duced to illustrate that cough may be one facet of anasthma syndrome which is variously represented in in-dividual patients. In classic asthma where bronchocon-striction, and conversely bronchodilator response, canbe demonstrated cough may be an additional and important feature. However, cough as an isolatedsymptom without bronchoconstriction or breathless-ness, but with the characteristic pathological featuresof asthmatic airway inflammation, is the other end of

the spectrum [29,30]. This so-called cough variantasthma is merely one end of a continuum. The termcough-predominant asthma may be preferred since thisterminology includes patients in whom the major prob-lem is cough but who also illustrate some or all of theother features of classic asthma [31]. Why individualsshould vary so much in the expression of cough as asymptom of asthma is at present unclear. An elegant se-ries of papers by Brightling has explored the extremeend of this spectrum, eosinophilic bronchitis. Here airways inflammation is present but there are no fea-tures of classic asthma, including bronchial hyper-responsiveness. However there are subtle differences in the cellular composition and location of the inflam-mation, particularly concerning mast cells and theirmediators [26,32].

Between a quarter and a third of patients presentingto a tertiary referral centre with chronic cough will be suffering from cough-predominant asthma (Table2.1). This rate of detection probably does not reflect the prevalence of cough-predominant asthma sincemany patients, particularly those who have features ofclassic asthma, are diagnosed and treated in the com-munity. Indeed it is unusual for patients with chroniccough to be seen in tertiary clinics who have not had anunsuccessful trial of inhaled medication. The reasons

13

Table 2.1 Commonest causes of chronic cough in patients investigated in specialist clinics.

Author Mean age in years Diagnosis (% of total)(range)

Asthma syndrome GOR Rhinitis Most common other causes (%)

Irwin et al. 1981 [18] 50.3 (17–88) 25 10 29 Chronic bronchitis (12)Poe et al. 1982 [20] ? (15–89) 36 0 8 Postinfectious (27)Poe et al. 1989 [21] 44.8 (19–79) 35 5 26 Idiopathic (12)Irwin et al. 1990 [19] 51 (6–83) 24 21 41 Chronic bronchitis (5)Hoffstein 1994 [22] 47 25 24 26 Postinfectious (21)O’Connell et al. 1994 [23] 49 (19–83) 6 10 13 Idiopathic (22)Smyrnios et al. 1995 [24] 58 (18–86) 24 15 40 Chronic bronchitis (11)Mello et al. 1996 [17] 53.1 (15–83) 14 40 38 Bronchiectasis (4)Marchesani et al. 1998 [25] 51 14 5 56 Chronic bronchitis (16)McGarvey et al. 1998 [15] 47.5 (18–77) 23 19 21 Idiopathic (18)Palombini et al. 1999 [16] 57 (15–81) 59 41 58 Bronchiectasis (18)Brightling et al. 1999 [26] * 31 8 24 Postviral (13)

GOR, gastro-oesophageal reflux.* No figures given for the total sample but mean age of 12/91 patients with eosinophilic bronchitis given as 52 (28–76) years.

CHAPTER 2

for failure of therapy, even when the underlying diag-nosis is of cough-predominant asthma, are all thoseusually associated with poor asthma control: compli-ance, poor inhaler technique, inappropriate choice ofdevice, etc. In addition there are other features ofcough-predominant asthma, which unless recognized,lead to failure of therapy. Clearly the usual diagnosticmeasures of reversibility testing or home peak flowmonitoring are frequently unhelpful. Even metha-choline challenge may not identify patients who re-spond adequately to corticosteroid therapy since thosewith eosinophilic bronchitis are not hypersensitive[33]. Whilst sputum examination in expert hands clearly has a role the methodological difficulties obviate its routine use. Ultimately, the diagnosis andtherefore prevalence of cough-predominant asthmarests on the use of a therapeutic trial of antiasthma medication. Here again the differences between cough-predominant asthma and classic asthma may lead toconfusion. Since bronchospasm may only be a minorfeature or even absent, add-on therapy with long-acting b-agonists rarely proves successful andleukotriene antagonists may be the preferred add-ontherapy [34]. The response to leukotriene antagonistsmay illustrate the hypothesized role of lipoxygenaseproducts in the direct modulation of the putative VR1cough receptor [35]. Ultimately, diagnosis of cough-predominant asthma may rely on the demonstration ofa response to parenteral steroids.

The oesophagus and coughA considerable portion of patients presenting withchronic cough have a disorder of the oesophagus. It is clearly poorly recognized by many physicians, yetcough as the sole presentation of gastro-oesophagealreflux has been well described [36,37]. In addition toreflux it is becoming increasingly clear that a number ofoesophageal disorders, broadly classified as dysmoti-lity and including abnormal peristalsis and abnormallower oesophageal sphincter tone, may give rise tocough [38]. That acid reflux alone is not the cause ofcough in oesophageal disease explains the partial re-sponse seen in many patients with even high doses ofproton pump inhibitors.

As with other causes of cough, diagnosis may be dif-ficult because there can be few clues from the history.However, whilst there is some disagreement [17], in in-dividual patients there may be a strong association withother symptoms, particularly heartburn. More unusual

characteristics such as an association with hoarseness,choking sensation and postnasal symptoms are increas-ingly recognized as being part of a reflux phenomenonby ENT specialists. Indeed, a striking reduction ofcough during sleep, which initially may be thought tocount against a diagnosis of oesophageal cough, mayindicate an oesophageal origin. Lower oesophagealsphincter pressure increases physiologically in recum-bency preventing reflux in the early stages of the disease[39]. The clues to the diagnosis of cough of oesophagealorigin may be obtained by looking for associations be-tween food, eating and cough.

Rhinitis and postnasal dripThere is marked geographical variation in the incidenceof rhinitis and postnasal drip in the reported series ofpatients presenting to cough clinics. Patients in theAmericas present with symptoms of postnasal drip inup to 50% of cases, whereas rhinitis is reported in approximately 10% in most European experience. The difference for this may be in part societal in that patients from North America are far more likely to describe upper respiratory tract symptoms as postnasaldrip [18,19]. In addition, the diagnosis of postnasaldrip or rhinitis is frequently accepted because of a response to ‘specific therapy’ with broad-spectrum,centrally acting antihistamines and systemic deconges-tants [40]. Of course, such treatment is anything butspecific. Such therapy may act in upper airway diseaseand in asthma. Centrally acting antihistamines maywork either on the central pathways of the cough orsimply through a sedating mechanism unrelated to theanatomical site of cough generation.

Until such problems in the definition of postnasaldrip and its subsequent specific diagnosis are resolved,rhinitis or rhinosinusitis is probably the preferred termdescribing this syndrome. One further problem is thenumerous observations from animal species, whichpoint to the absence of afferent cough sensory neuroneswithin the territory of the glossopharyngeal andtrigeminal nerves [41]. It is possible that the symptomsof rhinitis and postnasal drip are epiphenomena associ-ated with inflammation in the territory of the vagus.

Other rare causes of cough

Whilst the rarer causes of cough may not matter in theoverall epidemiology of cough, they are extremely im-portant to the individual patient since prolonged suf-

14

EPIDEMIOLOGY OF COUGH

fering may result because of a lack of firm diagnosis.The whole panoply of disorders within the territory ofthe vagus nerve must be considered in patients with ob-scure cough. The ear (from the nerve of Arnold), the oe-sophagus, and even the heart may be the seat of thegeneration of cough.

Some of the more important causes of chronic cough,which lead to considerable diagnostic and social conse-quences are:• Inhaled foreign body. Whilst this can occur in any age group, typically it affects boys aged 3 [42]. Theremay be little or no clue from the history or chest radiog-raphy. Foreign bodies can be radiolucent and thosewith a lumen can produce wheezing without distal atelectasis and may be frequently misdiagnosed as asthma.• Habit cough, a syndrome almost exclusively re-stricted to children and young people. The physicalcharacteristics of the cough are unlike those of an ‘organic’ cough and the sound produced has been described as a ‘honk’. The cough characteristically dis-appears when the patient is asleep. Whilst most paedia-tricians treat habit cough with simple measures such asreassurance and breathing exercises there remains ahard core of patients who are resistant to therapy. Inthese habit cough has many of the characteristics of a tic and may on close observation be associated withother mannerisms. Cough is a feature of Gilles de laTourette syndrome in approximately 10% of cases andtherapy with haloperidol or pimozide may be highly effective.• Postviral cough. Whilst most cough associated with upper respiratory tract viral infection abates within 1 week, in some patients coughing is prolonged and may take several months to settle. Such patientshave a heightened cough reflex and may have some sub-clinical cause of cough which is exacerbated by virus-induced vagal hypersensitivity. There is no specifictherapy [43].• ACE inhibitor cough. Similar to postviral cough,ACE inhibitor cough causes a shift in the cough sensi-tivity to tussigenic agents [44]. This alteration in thecough reflex may take several months to settle. Thus,although patients usually improve within a week of ces-sation of ACE inhibitors, a number of individuals con-tinue coughing weeks and months after the cessation oftreatment. As with postviral cough such individualsfrequently have a low-grade or subclinical coughcaused by other aetiologies.

References

1 Schappert KT. National Ambulatory Medical Care Survey: 1991, Summary. Advance Data 93 A.D., Number230. US Department of Health and Human Services, National Center for Health Statistics, 1993.

2 Office of Population Censuses and Surveys. MorbidityStatistics from General Practice: 4th National Study1991–1992. London: HMSO, 1995: Series MB5, 3.

3 Morrell DC. Symptom interpretation in general practice. JR Coll General Pract 1972; 22: 297–309.

4 Ison MG, Mills J, Openshaw P, Zambon M, Osterhaus A, Hayden F. Current research on respiratory viral infec-tions: Fourth International Symposium. Antiviral Res2002; 55 (2): 227–78.

5 Eccles R. An explanation for the seasonality of acute upperrespiratory tract viral infections. Acta Otolaryngol 2002;122 (2): 183–91.

6 Law BJ, Carbonell-Estrany X, Simoes EA. An update onrespiratory syncytial virus epidemiology: a developedcountry perspective. Respir Med 2002; 96 (Suppl. B):S1–S7.

7 Fujimura M, Kasahara K, Kamio Y, Naruse M, Hashimoto T, Matsuda T. Female gender as a determinantof cough threshold to inhaled capsaicin. Eur Respir J1996; 9: 1624–6.

8 Dicpinigaitis PV, Allusson VRC, Baldanti A, Nalamati JR.Ethnic and gender differences in cough reflex sensitivity.Respiration 1901; 68 (5): 480–2.

9 Yeo WW, Maclean D, Richardson PJ, Ramsay LE. Coughand enalapril: assessment by spontaneous reporting andvisual analogue scale under double-blind conditions. Br JClin Pharmacol 1991; 31 (3): 356–9.

10 Kastelik JA, Thompson RH, Aziz I, Ojoo JC, RedingtonAE, Morice AH. Sex-related differences in cough reflexsensitivity in patients with chronic cough. Am J Respir CritCare Med 2002; 166: 961–4.

11 Cullinan P. Persistent cough and sputum: prevalence andclinical characteristics in south east England. Respir Med1992; 86 (2): 143–9.

12 Janson C, Chinn S, Jarvis D, Burney P. Determinants of cough in young adults participating in the EuropeanCommunity Respiratory Health Survey. Eur Respir J1991; 18 (4): 647–54.

13 Littlejohns P, Ebrahim S, Anderson R. Prevalence and di-agnosis of chronic respiratory symptoms in adults. Br MedJ 1989; 298: 1556–60.

14 Boezen HM, Schouten JP, Postma DS, Rijcken B. Relationbetween respiratory symptoms, pulmonary function and peak flow variability in adults. Thorax 1995; 50:121–6.

15 McGarvey LP, Heaney LG, Lawson JT, Johnston BT, Scally CM, Ennis M, Shepherd DR, MacMahon J. Evalua-

15

CHAPTER 2

tion and outcome of patients with chronic non-productivecough using a comprehensive diagnostic protocol. Thorax1998; 53 (9): 738–43.

16 Palombini BC, Villanova CA, Araujo E, Gastal OL, Alt DC, Stolz DP, Palombini CO. A pathogenic triad inchronic cough: asthma, postnasal drip syndrome, and gastroesophageal reflux disease. Chest 1999; 116 (2):279–84.

17 Mello CJ, Irwin RS, Curley FJ. Predictive values of thecharacter, timing, and complications of chronic cough indiagnosing its cause. Arch Intern Med 1996; 156 (9):997–1003.

18 Irwin RS, Corrao WM, Pratter MR. Chronic persistentcough in the adult: the spectrum and frequency of causesand successful outcome of specific therapy. Am Rev RespirDis 1981; 123 (4 Part 1): 413–17.

19 Irwin RS, Curley FJ, French CL. Chronic cough. The spectrum and frequency of causes, key components of thediagnostic evaluation, and outcome of specific therapy.Am Rev Respir Dis 1990; 141 (3): 640–7.

20 Poe RH, Israel RH, Utell MJ, Hall WJ. Chronic cough:bronchoscopy or pulmonary function testing? Am RevRespir Dis 1982; 126 (1): 160–2.

21 Poe RH, Harder RV, Israel RH, Kallay MC. Chronic persistent cough. Experience in diagnosis and outcomeusing an anatomic diagnostic protocol. Chest 1989; 95(4): 723–8.

22 Hoffstein V. Persistent cough in nonsmoker. Can Respir J1994; 1: 40–7.

23 O’Connell F, Thomas VE, Pride NB, Fuller RW. Capsaicincough sensitivity decreases with successful treatment ofchronic cough. Am J Respir Crit Care Med 1994; 150:374–80.

24 Smyrnios NA, Irwin RS, Curley FJ. Chronic cough with ahistory of excessive sputum production. The spectrum andfrequency of causes, key components of the diagnosticevaluation, and outcome of specific therapy. Chest 1995;108 (4): 991–7.

25 Marchesani F, Cecarini L, Pela R, Sanguinetti CM. Causesof chronic persistent cough in adult patients: the results ofa systematic management protocol. Monaldi Arch ChestDis 1998; 53 (5): 510–14.

26 Brightling CE, Ward R, Goh KL, Wardlaw AJ, Pavord ID.Eosinophilic bronchitis is an important cause of chroniccough. Am J Respir Crit Care Med 1999; 160 (2): 406–10.

27 Morice AH, Lowry R, Brown MJ, Higenbottam T. An-giotensin converting enzyme and the cough reflex. Lancet1987; ii: 1116–18.

28 Ojoo JC, Kastelik JA, Morice AH. Duration of an-giotensin converting enzyme inhibitor (ACEI) inducedcough. Thorax 1902; 56: 89.

29 Glauser FL. Variant asthma. Ann Allergy 1972; 30 (8):457–9.

30 Corrao WM, Braman SS, Irwin RS. Chronic cough as thesole presenting manifestation of bronchial asthma. N EnglJ Med 1979; 300 (12): 633–7.

31 Pratter MR, Bartter T, Akers S, Dubois J. An algorithmicapproach to chronic cough. Ann Intern Med 1993; 119(10): 977–83.

32 Brightling CE, Ward R, Woltmann G, Bradding P, ShellerJR, Dworski R, Pavord ID. Induced sputum inflammatorymediator concentrations in eosinophilic bronchitis andasthma. Am J Respir Crit Care Med 2000; 162 (3 Part 1):878–82.

33 Gibson PG, Dolovich J, Denburg J, Ramsdale EH, Hargreave FE. Chronic cough: eosinophilic bronchitiswithout asthma. Lancet 1989; 1: 1346–8.

34 Dicpinigaitis PV, Dobkin JB. Effect of zafirlukast on coughreflex sensitivity in asthmatics. J Asthma 1999; 36 (3):265–70.

35 Hwang SW, Cho H, Kwak J, Lee SY, Kang CJ, Jung J, Cho S, Min KH, Suh YG etal. Direct activation of cap-saicin receptors by products of lipoxygenases: endoge-nous capsaicin-like substances. Proc Natl Acad Sci USA2000; 97 (11): 6155–60.

36 Irwin RS, Zawacki JK, Curley FJ, French CL, Hoffman PJ.Chronic cough as the sole presenting manifestation of gas-troesophageal reflux. Am Rev Respir Dis 1989; 140 (5):1294–300.

37 Ing AJ, Ngu MC. Cough and gastro-oesophageal reflux.Lancet 1999; 353: 944–6.

38 Kastelik JA, Aziz I, Thompson R etal. Gastroesophagealdysmotility as a cause of chronic persistent cough. Thorax2003 (in press).

39 Mittal RK, Balaban DH. The esophagogastric junction. NEngl J Med 1997; 336 (13): 924–32.

40 Irwin RS, Madison JM. Anatomical diagnostic protocol in evaluating chronic cough with specific reference to gastroesophageal reflux disease. Am J Med 2000; 108(Suppl. 4a): 126S–30S.

41 Widdicombe JG. Afferent receptors in the airways andcough. Respir Physiol 1998; 114 (1): 5–15.

42 Hoeve LJ, Rombout J, Pot DJ. Foreign body aspiration inchildren. The diagnostic value of signs, symptoms and pre-operative examination. Clin Otolaryngol 1993; 18 (1):55–7.

43 Ojoo JC, Kastelik JA, Morice AH. A boy with a disablingcough. Lancet 2003; 361: 674.

44 Morice AH, Brown MJ, Higenbottam T. Cough associ-ated with angiotensin converting enzyme inhibition. JCardiovasc Pharmacol 1989; 13 (Suppl. 3): S59–S62.

16