proceedings of a workshop on - havemeyer …proceedings of a workshop on inflammatory airway...

Proceedings of a Workshop on

INFLAMMATORY AIRWAY DISEASE:DEFINING THE SYNDROME

30th September – 3rd October 2002Boston, USA

Editors: A. Hoffman, N. E. Robinson and J. F. Wade

Havemeyer Foundation

Havemeyer Foundation Monograph Series No. 9

© 2003 by R & W Publications (Newmarket) LimitedSuites 3 & 4, 8 Kings Court, Willie Snaith Road, Newmarket, Suffolk CB8 7SG, UK

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, without the prior permission of the copyright owner.Authorisation to photocopy items for internal or personal use, or the internal or personal use of specific clients, isgranted by R & W Publications (Newmarket) Limited for libraries and other users registered with the CopyrightClearance Center (CCC) Transactional Reporting Service, provided that the base fee of £0.02 per copy (no additionalfee per page) is paid directly to CCC, 21 Congress Street, Salem, MA 01970. This consent does not extend to otherkinds of copying, such as copying for general distribution, for advertising or promotional purposes, for creating newcollective works, or for resale.

First published 2003

ISSN 1472-3158

Published by R & W Publications (Newmarket) Limited

Printed in Great Britain by Quality Print Services (Anglia) Limited

iii


CONTENTS

EDITORS’ FOREWORD .....................................................................................................................Page v

SESSION 1: CLINICAL EVIDENCE

Inflammatory airway disease: a clinician’s view from North AmericaB. R. Rush ..........................................................................................................................................Page 3

Inflammatory airway disease: European clinician’s perspectiveP. M. Dixon, B. C. McGorum and R. S. Pirie ....................................................................................Page 7

Inflammatory airway disease: effect of athletic disciplineM. Mazan and A. Hoffman..................................................................................................................Page 9

Mucus, cough, airway obstruction and inflammationN. E. Robinson, C. Berney, H. DeFeijter-Rupp, A. M. Jefcoat, C. Cornelisse, V. Gerber and F. J. Derksen ..............................................................................................................................Page 13

Relationship between coughing and airway inflammation in young racehorsesD. R. Hodgson, R. M. Christley, J. L. N. Wood, S. W. J. Reid and J. L. Hodgson ...........................Page 16

Poor performance: the trainer’s perspective (sport horse)C. Platz ............................................................................................................................................Page 19

The diagnostic approach to chronic cough in peopleA. J. Ghio ..........................................................................................................................................Page 23

SESSION 2: AETIOLOGY

Aetiological agents: indoor environment and endotoxinB. C. McGorum and R. S. Pirie .......................................................................................................Page 27

Aetiological agents: outdoor environment and airwaysA. J. Ghio ........................................................................................................................................Page 29

Aetiological agents: allergy and related mediatorsJ. P. Lavoie .......................................................................................................................................Page 31

Aetiological agents: viruses and inflammatory airway diseaseJ. L. N. Wood, J. R. Newton, K. C. Smith and D. J. Marlin ............................................................Page 33

Natural history of equine influenzaH. G. G. Townsend............................................................................................................................Page 37

Aetiological agents: bacteriaJ. R. Newton, J. L. N. Wood, K. C. Smith, D. J. Marlin and N. Chanter .........................................0

Dysregulation of inflammationF. Bureau and P. Lekeux ...................................................................................................................Page 45

iv

Inflammatory Airway Disease

SESSION 3: DIAGNOSTIC MEASURES OF INFLAMMATION

Significance of tracheal inflammationJ. L. Hodgson ....................................................................................................................................Page 49

Significance of bronchoalveolar cytology in inflammatory airway disease of horsesL. Viel ................................................................................................................................................2

Cytology of inflammatory airway disease K. C. Smith, J. R. Newton, S. M. Gower, S. M. Cade, D. J. Marlin and C. M. Deaton ...................Page 55

Quantifying and characterising mucus in the airwaysV. Gerber, A. M. Jefcoat, J. A. Hotchkiss, M. King and N. E. Robinson..........................................Page 59

Breath condensate measures of airway inflammationD. J. Marlin, C. M. Deaton, J. R. Newton and K. C. Smith .............................................................Page 62

SESSION 4: FUNCTIONAL SIGNIFICANCE

Lung function for beginnersF. J. Derksen .....................................................................................................................................Page 69

Airway obstruction and hyper-reactivity in horses with signs of inflammatory airway diseaseA. Hoffman and M. Mazan................................................................................................................Page 71

Inflammatory airway disease and gas exchangeG. Nyman ..........................................................................................................................................Page 75

Functional imaging of inflammatory airway diseaseD. Votion ...........................................................................................................................................Page 81

Inflammatory airway disease and clinical exercise testingL. Couëtil ..........................................................................................................................................Page 84

WORKSHOP SUMMARY ...................................................................................................................Page 89

LIST OF PARTICIPANTS ...................................................................................................................Page 92

AUTHOR INDEX ...............................................................................................................................Page 93

EDITORS’ FOREWORD

The nomenclature of equine non-infectiousairway disease is a source of puzzlementto many. The terms ‘chronic obstructive

pulmonary disease (COPD)’, ‘inflammatoryairway disease (IAD)’, ‘chronic obstructivebronchitis (COB)’, ‘reactive airway disease’,‘allergic bronchitis’, ‘heaves’, ‘broken wind’ and‘small airway disease’ confuse vets and otherbiomedical scientists. The plethora of terminologyreflects a poor understanding of the pathogenesisof equine airway disease and leads to confusion asto the best way to treat and prevent airway disease.

For years, ‘equine COPD’ was used to describeany accumulation of neutrophils and mucus in theairways in the absence of active infection, eg themature horse with severe heaves and the youngeranimal with reduced performance, increased airwayneutrophils and mucus. However, the 2 conditionsare very different and there is no evidence that theformer is a consequence of the latter. A furthersource of confusion is the incongruous use of theterm in human and veterinary medicine. In humanmedicine, COPD is a disease of aged smokers andthe term would not be used to describe a youngathlete with a cough, increased mucus productionand reduced exercise ability. In the latter individual,smoking history, allergies, work environment andinfection history would all be investigated; COPDwould not be considered as a diagnosis. Equine andhuman COPD differ in many aspects. For example,the airway obstruction of human COPD is scarcelyreversible whereas the obstruction of severe equineCOPD is reversed by a change in environment,bronchodilator drugs or corticosteroids.

To address this confusion, a conference washeld at Michigan State University in 2000.Delegates decided to eliminate the term ‘COPD’and recommended the term ‘heaves’ to describethe severe but reversible airway obstruction seenin mature horses. Because of the naturalrecurrence of airway obstruction in heaves, and its

reversibility with drugs, ‘recurrent airwayobstruction (RAO)’ was also deemed anappropriate term. By exclusion, all other forms ofnon-infectious airway disease are referred to asIAD, the topic of this workshop.

The Havemeyer Workshop series bringstogether investigators from a range of disciplinesand locations to discuss topic of specific concernto the horse industry. IAD is such a topic and wetherefore invited clinicians, epidemiologists,pathologists, microbiologists and physiologists toshare knowledge with each other. Severalimportant questions must be addressed before wecan provide solid answers about IAD. First, wemust decide what each of us means by the termIAD. Is the syndrome seen in young racehorsessimilar to that of the older ‘cougher withoutheaves’? How important is infection? Does theculture of organisms from the trachea mean thatthese organisms are causing functional changes inthe airways? Is inflammation regionalised in thelung? Does inflammation and mucus in thetrachea indicate lower airway inflammation?What are the functional consequences of IAD andthe clinical consequences to the horse? Finally,what recommendations for treatment andprevention can we make? Addressing thesequestions will require interdisciplinary studies anditt is our hope that this Workshop will stimulatethe necessary interactions to make such studies apossibility. We are extremely grateful to Mr GenePranzo, President of the Havemeyer Foundation,for his support of this workshop and to RachelPepper at R&W Publications for the excellentorganisational arrangements.

Ed RobinsonMichigan State University

Andy HoffmanTufts University

Havemeyer Workshop Organisers

v


vi


HAVEMEYER SCIENTIFIC WORKSHOPS

1981 First International Workshop on Lymphocyte Alloantigens of the HorseOctober - New York City, USAOrganiser: Dr D. F. Antczak

1982 Second International Workshop on Lymphocyte Alloantigens of the HorseOctober - Cornell University, Ithaca, New York, USAOrganiser: Dr D. F. Antczak

1983 Third International Workshop on Lymphocyte Alloantigens of the HorseApril - New Bolton Center, University of Pennsylvania, USAOrganiser: Dr D. F. Antczak

1984 First International Symposium on Equine Embryo TransferOctober - Cornell University, Ithaca, New York, USAOrganisers: Drs D. F. Antczak and W. R. Allen

1985 Fourth International Workshop on Lymphocyte Alloantigens of the HorseOctober - University of Kentucky, USAOrganisers: Drs D. F. Antczak and E. Bailey

1986 Workshop on Corynebacterium equi Pneumonia of FoalsJuly - University of Guelph, CanadaOrganiser: Dr J. F. Prescott

1987 Fifth International Workshop on Lymphocyte Alloantigens of the HorseOctober - Louisiana State University, USAOrganisers: Drs D. F. Antczak and J. McClure

1989 Second International Symposium on Equine Embryo TransferFebruary - Banff, Alberta, CanadaOrganisers: Drs D. F. Antczak and W. R. Allen

1990 International Workshop on Equine SarcoidsApril - Interlaken, SwitzerlandOrganisers: Dr D. F. Antczak and Professor S. Lazary

1992 Workshop on Equine Neonatal MedicineJanuary - Naples, FloridaOrganisers: Drs D. F. Antczak and P. D. Rossdale

Third International Symposium on Equine Embryo TransferFebruary - Buenos Aires, ArgentinaOrganisers: Drs D. F. Antczak, W. R. Allen, J. G. Oriol and R. Pashen

vii


1995 Equine PerinatologyJuly - Cambridge, EnglandOrganiser: Dr P. D. Rossdale

Second International Equine Leucocyte Antigen WorkshopJuly - Lake Tahoe, California, USAOrganisers: Drs D. F. Antczak, P. Lunn and M. Holmes

First International Workshop on Equine Gene MappingOctober - Lexington, Kentucky, USAOrganisers: Drs D. F. Antczak and E. Bailey

Erection and Ejaculation in the Human Male and Stallion: A Comparative StudyOctober - Mount Joy, Pennsylvania, USAOrganiser: Dr S. M. McDonnell

Bone Remodelling WorkshopOctober - Corcord, Massachusetts, USAOrganiser: Dr H. Seeherman

1997 Second International Workshop on Equine Gene MappingOctober - San Diego, California, USAOrganisers: Drs D. F. Antczak and E. Bailey

Maternal Recognition of Pregnancy in the MareJanuary - Dominican RepublicOrganisers: Drs W. R. Allen and T. A. E. Stout

Uterine ClearanceMarch - Gainesville, Florida, USAOrganiser: Dr M. M. LeBlanc

Trophoblast DifferentiationSeptember - Edinburgh, ScotlandOrganisers: Drs D. F. Antczak and F. Stewart

1998 Third International Genome WorkshopJanuary - San Diego, California, USAOrganisers: Drs D. F. Antczak and E. Bailey

Third International Workshop on Perinatology: Genesis and Post Natal Consequences of Abnormal Intrauterine Developments: Comparative AspectsFebruary - Sydney, AustraliaOrganiser: Dr P. D. Rossdale

Horse Genomics and the Genetic Factors Affecting Race Horse Performance March - Banbury Center, Cold Spring Harbor, New York, USAOrganisers: Drs D. F. Antczak, E. Bailey and J. Witkowski

Allergic Diseases of the HorseApril - Lipica, SloveniaOrganisers: Drs D. F. Antczak, S. Lazary and E. Marti

viii


Equine Placentitis WorkshopOctober - Lexington, Kentucky, USAOrganisers: Drs D. F. Antczak, W. R. Allen and W. Zent

Septicemia II WorkshopNovember - Boston, Massachusetts, USAOrganiser: Dr M. R. Paradis

1999 Equine Genome ProjectJanuary - San Diego, California, USAOrganisers: Drs D. F. Antczak and E. Bailey

Third International Equine Genome WorkshopJune - Uppsala, SwedenOrganisers: Drs D. F. Antczak, E. Bailey and K. Sandberg

Fourth International Meeting of OIE and WHO Experts on Control of Equine InfluenzaAugust - Miami, Florida, USAOrganiser: Dr J. Mumford

European Equine Gamete WorkshopSeptember - Lopuszna, PolandOrganisers: Drs W. R. Allen and M. Tischner

Fetomaternal Control of PregnancyNovember - Barbados, West IndiesOrganisers: Drs T. Stout and W. R. Allen

2000 Equine Genome ProjectJanuary - San Diego, California, USAOrganisers: Drs D. F. Antczak and E. Bailey

Uterine Infections in Mares and Women: A Comparative StudyMarch - Naples, Florida, USAOrganiser: Dr M. M. LeBlanc

5th International Symposium on Equine Embryo TransferJuly - Saari, FinlandOrganiser: Dr T. Katila

2001 USDA International Plant & Animal Genome ConferenceJanuary - San Diego, California

Equine Immunology in 2001January - Santa Fe, New MexicoOrganiser: Dr D. P. Lunn

Asthma and Allergies II April - HungaryOrganisers: S. Lazary and E. Marti

ix


From Elephants to Aids June - Port Douglas, AustraliaOrganiser: Professor W. R. Allen

International Equine Gene Mapping July - Brisbane, AustraliaOrganiser: K. Bell

Second Meeting of the European Gamete Group (EEGG) September - Loosdrecht, The NetherlandsOrganiser: Dr T. A. E. Stout

Foal Septicemia IIIOctober - Tufts University European Center, Talloires, FranceOrganiser: M. R. Paradis

Infectious Disease Programme for the Equine Industry and Veterinary PractitionersOctober - Marilyn duPont Scott Medical Center, Morvan Park, Virginia, USAOrganisers: Drs J. A. Mumford and F. Fregin

From Epididymis to EmbryoOctober - Fairmont Hotel, New Orleans, USAOrganiser: Dr L. H-A. Morris

2002 USDA International Plant & Animal Genome ConferenceJanuary - San Diego, California

Comparative Neonatology/PerinatologyJanuary - Palm Springs, CaliforniaOrganiser: P. Sibbons

Stallion Behavior IVJune - Reykjavik, IcelandOrganisers: S. McDonell and D. Miller

Rhodococcus Equi IIJuly - Pullman, WashingtonOrganiser: J. Prescott

Equine Orthopaedic InfectionAugust - Dublin, IrelandOrganiser: E. Santschi

Inflammatory Airway DiseaseSeptember - Boston, USAOrganiser: Dr E. Robinson

x


HAVEMEYER MONOGRAPH SERIES

The following are monographs available to date at a cost of £9.95 each.

Series No 1PROCEEDINGS OF THE FIRST MEETING OF THE EUROPEAN EQUINE GAMETE GROUP (EEGG)Editors: W. R. Allen and J. F. Wade5th–8th September 1999 Lopuszna, Poland

Series No 2PROCEEDINGS OF A WORKSHOP ON FETOMATERNAL CONTROL OF PREGNANCY

Editors: T. A. E. Stout and J. F. Wade14th–16th November 1999Barbados, West Indies

Series No 3PROCEEDINGS OF THE 5TH INTERNATIONAL SYMPOSIUM ON EQUINE EMBRYO TRANSFER

Editors: T. Katila and J. F. Wade6th–9th July 2000Saari, Finland

Series No 4PROCEEDINGS OF A WORKSHOP ON EQUINE IMMUNOLOGY IN 2001 Editors: D. P. Lunn and J. F. Wade24th–28th January 2001Santa Fe, New Mexico

Series No 5PROCEEDINGS OF THE SECOND MEETING OF THE EUROPEAN GAMETE GROUP (EEGG) Editors: T. A. E. Stout and J. F. Wade26th–29th September 2001Loosdrecht, The Netherlands

Series No 6PROCEEDINGS OF A WORKSHOP ENTITLED FROM EPIDIDYMIS TO EMBRYO

Editors: L. H-A. Morris and J. F. Wade18th–21st October 2001New Orleans, USA

xi


Series No 7FOURTH INTERNATIONAL MEETING OF OIE AND WHO EXPERTS ON CONTROL OF EQUINE

INFLUENZA

Editors: J. A. Mumford and J. F. Wade3rd–5th August 1999Crowne Plaza Hotel, Miami, Florida USA

Series No 8COMPARATIVE NEONATOLOGY /PERINATOLOGY

Editors: Dr P. Sibbons and J. F. Wade13th – 15th March 2002Palm Springs, California, USA

Series No 9INFLAMMATORY AIRWAY DISEASE: DEFINING THE SYNDROME

Editors: A. Hoffman, N. E. Robinson and J. F. Wade30th September – 3rd October 2002Boston, USA

If you wish to order copies, please contact R & W Publications Ltd, Suites 3 & 4, 8 Kings Court, WillieSnaith Road, Newmarket, Suffolk CB8 7SG, UK, Tel: +44 1638 667600, Fax: +44 1638 667229, e-mail: [email protected].

xii



1

SESSION I:

Clinical evidence

Chairman: N. E. Robinson


2


3

INFLAMMATORY AIRWAY DISEASE: A CLINICIAN’S VIEW FROM NORTH AMERICA

B. R. Rush

Kansas State University, Manhattan, Kansas, USA

Inflammatory airway disease (IAD) describes aheterogeneous group of inflammatory conditions ofthe lower respiratory tract that appear to be primarilynon-infectious. Poor exercise performance andexcessive tracheal exudates on endoscopicexamination are consistent clinical findings.Cytologic evaluation of bronchoalveolar lavage(BAL) fluid in horses with IAD will reveal one ofthe following inflammatory profiles: 1) mixedinflammation with high total nucleated cells, mildneutrophilia (15% of total cells), lymphocytosis, andmonocytosis; 2) increased metachromatic cells(mast cells >2% of total cells); or 3) eosinophilicinflammation (5–40% of total cells). Therapeuticrecommendations will be based on results of theBAL cytologic evaluation. The majority of caseswill be treated with an anti-inflammatorypreparation, bronchodilating agents, and avoidanceof environmental irritants. Depending on cytologicalfindings, immunostimulant or immunosuppressivetherapy may be indicated to reduce pulmonaryinflammation.

MAST CELL -RICH AND EOSINOPHILICINFLAMMATION

Mast cell stabilising drugs and aerosolisedcorticosteroids are recommended for treatment ofmast-cell rich IAD. Nebulisation of sodiumcromoglycate (80–200 mg) improves clinical signsof respiratory disease and reduces histaminecontent in equine mast cells (Hare et al. 1994).Mast cell stabilising drugs also inhibit the releaseof inflammatory mediators from eosinophils andmay be of potential benefit to horses witheosinophilic IAD, although therapeutic benefit hasnot been reported. Aerosolised corticosteroidadministration improves clinical signs of diseaseand BAL findings in horses with mast-cell rich

IAD. Horses with eosinophilic pulmonaryinflammation may have pulmonary parenchymalgranulomas, peripheral eosinophilia, and evidenceof extrapulmonary eosinophilic infiltration (skin,GI, urinary), necessitating systemic corticosteroidadministration.

Bronchodilators do not address the primaryinflammatory process. However, bronchodilatortherapy is an important component of treatment ofIAD because it provides immediate relief ofairway obstruction and protects against irritant-induced bronchoconstriction. Administration ofbronchodilators prior to exercise may preventexercise-induced bronchoconstriction and improveperformance of horses with mild to moderateairway obstruction. In addition, short-actingbronchodilators can be administered prior toadministration of topically active mast cellstabilising drugs and corticosteroid preparations toimprove pulmonary drug distribution and preventirritant cough.

MAST CELL STABILISING DRUGS

Sodium cromoglycate (SCG) and nedocromilsodium (NS) inhibit mast cell degranulation andprevent the release of potent inflammatory mediatorsincluding histamine, leukotrienes and cytokines. Inaddition, nedocromil sodium has been shown toinhibit the release of inflammatory mediators fromseveral other inflammatory cells (eosinophils,neutrophils, macrophages), and can block immediatevagal-mediated bronchoconstriction caused byirritant stimuli and exercise. Mast cell degranulationproducts may have an autocrine effect to induce mastcell hyperplasia, thus the proposed mechanism forreduced histamine content in equine mast cellsrecovered from horses with clinical IAD is reductionof histamine production with attenuated histamine


4

release. The therapeutic benefit of NS and SCGappears to increase with prolonged administration(Hoffman 1997). In humans, NS is approximately 10times more potent than SCG with equivalent efficacyat equipotent doses. Coughing, throat irritation, andbronchoconstriction may be observed afteradministration, which is attenuated by precedentadministration of a bronchodilator (Hoffman 1997).Neither sodium cromoglycate nor nedocromilsodium has direct bronchodilator or antihistamineproperties. Therefore, these anti-inflammatory drugsare considered prophylactic medications and arelabelled for the preventive management of asthma inhuman patients. Neither of these drugs has an effecton normal immunological defence mechanisms, norany known systemic activity and there are no reportsof toxicity.

CORTICOSTEROIDS

Aerosolised corticosteroids are effective in horseswith mild to moderate airway obstruction withclinical signs ranging from exercise intolerance tohorses with moderate increased effort of respirationat rest. Aerosolised corticosteroid preparations areshort-acting glucocorticoids so prolongedtherapeutic benefit is not anticipated from short-term administration. After the initial reduction ininflammation, administration once a day may beeffective to maintain disease remission. There are 3aerosolised corticosteroid preparations available in MDI formulation for administration to horses: fluticasone propionate, beclomethasonedipropionate, and flunisolide. The relative potencyof these surface-active corticosteroids is fluticasone(relative potency = 18.0) > beclomethasone (13.5) >flunisolide (1.9; Barnes et al. 1998).

Fluticasone is the most potent and the mostexpensive of the aerosolised corticosteroids.Fluticasone is highly lipophilic and, consequently,has the longest pulmonary residence time. Becauseof its low oral bioavailability (<2%) and extensivefirst-pass metabolism (99%), fluticasone has theleast potential for adverse systemic effects and themost favourable therapeutic index of all of theaerosolised corticosteroids. In heaves-affectedhorses, fluticasone (2,000 µg, BID, EquineAeroMask) reduces pulmonary neutrophilia,improves parameters of pulmonary function andreduces responsiveness to histamine challengeduring an episode of airway obstruction (Viel et al.1999).

Beclomethasone (500–1,500 µg, BID, Equine

Aerosol Delivery Device) reduces pulmonaryinflammation, improves parameters of pulmonaryfunction and improves ventilation imaging ofhorses with recurrent airway obstruction (Rush etal. 1998a,b) There is no immediate (15 min)therapeutic effect but clinical signs and pulmonaryfunction begin to improve within 24 h ofadministration (Rush et al. 1999). Using theEquine AeroMask, administration ofbeclomethasone (3,750 µg, BID) to horses withheaves improves parameters of pulmonaryfunction and arterial oxygen tension for a 2-weektreatment period (Ammann et al. 1998).

The timing of administration once a day has apivotal effect on the safety profile andconsequently the risk/benefit ratio of inhaledcorticosteroids (Meibohm et al. 1997). Maximumadrenal suppression occurs with administration ofaerosolised corticosteroids in the early morninghours, whereas endogenous cortisol production isleast disrupted by administration in the afternoon.The longer the terminal elimination half-life of thedrug, the earlier in the afternoon it should beadministered. For example, the adrenosuppressiveeffects of flunisolide (t1/2 = 1.5 h) are minimisedwhen a single daily dose is administered at 7 pm,whereas the optimum time for administration offluticasone propionate (t1/2 = 6 h) is 4 pm. Inaddition to safety concerns, afternoon drugadministration provides superior control of theclinical signs of nocturnal asthma. The safety andefficacy of administration of aerosolisedcorticosteroids once a day (afternoon/ evening) inhorses has not been evaluated.

BRONCHODILATORS

Although horses with IAD do not have overtevidence of bronchoconstriction on physicalexamination, affected horses are hyper-responsiveto irritant stimuli and may suffer from low-gradebronchoconstriction that impairs exerciseperformance. Short-acting bronchodilators provideimmediate relief of airway obstruction.Administration prior to exercise may preventexercise-induced bronchoconstriction and improveperformance of horses with mild to moderateairway obstruction. In addition, short-actingbronchodilators can be administered prior toadministration of anti-inflammatory medicationsto improve pulmonary drug distribution andprevent cough. Albuterol sulphate (360–900 µg,) isa potent, short-acting β2 adrenergic agent with


5

rapid onset of bronchodilation (5 min) in horses(Derksen et al. 1999). The duration of effectivebronchodilation is approximately 1–3 h. Albuterolimproves parameters of pulmonary function byapproximately 70% in horses with heaves duringan episode of airway obstruction.

Long-acting bronchodilators are recommendedfor maintenance bronchodilation in horses withIAD (Hoffman 1997). Salmeterol is a long-actingβ2 adrenergic agent that is a chemical analogue ofalbuterol. In human patients, administration ofsalmeterol twice a day provides superior control ofbronchoconstriction compared to regular (QID) orPRN administration of albuterol. In comparison toalbuterol, salmeterol has higher lipophilicity, β2affinity, β2 selectivity, and potency (10-fold). Inaddition to the bronchodilatory activity, salmeterolprevents irritant-induced bronchoconstriction andhas anti-inflammatory properties such as inhibitionof leukotiene release, prevention of histaminerelease, and reduction of eosinophil activity.Salmeterol (210 µg via Equine AeroMask)provides relief of clinical signs of airwayobstruction for 6–8 h in heaves-affected horses(Henrikson and Rush 2001).

Ipratropium bromide is a synthetic,anticholinergic compound that producesbronchodilation, inhibits cough, and protectsagainst bronchoconstrictive stimuli. Like atropine,ipratropium bromide is nonselective (M1, M2, M3)muscarinic antagonist, and bronchodilation resultsfrom blockade of the M3 receptor. Because of itsquaternary ammonium structure, ipratropium ispoorly absorbed from the respiratory system (6%)or gastrointestinal tract (2%). Therefore,ipratropium does not inhibit gastrointestinalmotility and has minimal systemic adverse effects.In addition, ipratropium does not cause drying ofrespiratory secretions and does not inhibitmucociliary clearance. Ipratropium has a greaterbronchodilatory effect on larger, more centralairways. The onset of bronchodilation isapproximately 15–30 min, and the effect lastsapproximately 4–6 h.

MIXED INFLAMMATORY (NEUTROPHILIC )INFLAMMATION

Horses with neutrophilic inflammation often havea history of viral respiratory infection prior to theonset of IAD. Interferon-α (IFNα) hasimmunomodulatory and antiviral activity, andreduces pulmonary inflammation in horses with

poor race performance and mixed inflammatoryBAL fluid. The pathophysiology of mixedinflammatory IAD may be triggered by viralrespiratory infection or environmental irritatants(ozone). Low-dose (50–150 IU) natural, humanIFNα reduces exudate in the respiratory tract,lowers total cell counts in BAL fluid, and convertsthe differential cell count to a non-inflammatorycytological profile (Rush Moore et al. 1996).Interferon-α administration is not effective inhorses with mast cell-rich or eosinophilicbronchoalveolar lavage. In addition, IFNα is notbeneficial in the treatment of acute, fulminate viralrespiratory infection in horses. Oral administrationof low-dose (0.22–2.2 IU/kg bwt) rIFNα-2a doesnot diminish the severity of clinical disease orduration of viral shedding in horses withexperimental equine herpesvirus-1 infection(Seahorn et al. 1990).

IMMUNOSTIMULANT THERAPY

The indications for immunostimulant therapy arerelatively specific; these compounds are notintended to treat a broad-spectrum of respiratoryconditions. Propionibacterium acnes isrecommended for treatment of chronic (probablyinfectious) respiratory disease in weanlings andyearlings that is unresponsive or transientlyresponsive to antibiotics. It is also recommendedfor prophylactic administration prior to stressfulevents that may impair pulmonary defencemechanisms, including weaning and long-distancetransport (Evans et al. 1988; Kilmczak 1992).Propionibacterium acnes is considered adjunct toantibiotic therapy and is labelled for ivadministration every 2–3 days for 3 treatments(Vail et al. 1990). In addition to treament ofrespiratory disease, Propionibacterium acnes isbeneficial in the treatment of refractorypapallomas and may provide adjunct therapy fortreatment of sarcoid skin tumor.

Administration of P. acnes to healthy, yearlinghorses using the recommended dosage regimenincreases the number of CD4+ lymphocytes,enhances lymphokine-activated killing activity, andincreases expression of IFNγand NK lysin (Flaminoet al. 1998). Stimulation of systemic immunity canbe documented for 4–5 days after administration,however, prolonged immunostimulant activity is notanticipated (Cox 1988). Fever, anorexia, andlethargy may occur 12–24 h after administration ofthe first or second injection, presumably due to


6

increased interleukin-1 production. Therefore,administration is not recommended immediatelyprior to an athletic event.

REFERENCES

Ammann, V.J., Vrins, A.A. and Lavoie, J.P. (1998)Effects on inhaled beclomethasone dipropionate onrespiratory function in horses with chronicobstructive pulmonary disase (COPD). Equine vet. J.30, 152-157.

Barnes, P.J., Pedersen, S. and Busse, W.W. (1998)Efficacy and safety of inhaled corticosteriods. Newdevelopments. Am. J. Resp. Crit. Care Med. 157, 1-53.

Cox, W.I. (1988) Examining the immunologic andhematopoietic properties of an immunostimulant.Vet. Med. 6, 424-428.

Derksen, F.J., Olszewski, M.A. and Robinson, N.E.(1999) Aerosolised albuterol sulfate used as abronchodilator in horses with recurrent airwayobstruction. Am. J. vet. Res. 60, 689-693.

Evans, D.R., Rollins, J.B., Huff, G.K, Hartgrove, T.B.and Van Kampen, K.R. (1988) InactivatedPropionbacterium acnes (Immunoreguli) as adjunctto conventional therapy in the treatment of equinerespiratory diseases. Equine Pract. 10, 17-21.

Flamino M.J.B.F., Thompsen M., Rush B.R. and ShumanW. (1998) Immunologic function in horses afternon-specific immunostimulant administration. Vet.Immun. Immunopath. 63, 303-315.

Hare J.E, Viel L. and O’Byrne P.M. (1994) Effect ofsodium cromoglycate on light racehorses withelevated metachromatic cell numbers onbronchoalveolar lavage. J. vet. Pharmac. Ther. 17,237.

Henrikson, S.L. and Rush, B.R. (2001) Efficacy ofsalmeterol xinafoate in horses with recurrent airwayobstruction. J. Am. vet. med. Ass. 12, 1961-1965.

Hoffman, A. (1997) Inhaled medications andbronchodilator use in the horse. Vet. Clin. N. Am.

equine Pract. 13, 519-530.Klimczak, C. (1992) Immunostimulant quickly aids

weanling ERDC cases. Equine vet. Sci. 12, 68-69.Meibohm, B., Hochhaus, G., Rohatoagi, S. and

Mollman, S. (1997) Dependency of cortisolsuppression on the administration time of inhaledcorticosteroids. J. clin. Pharmac. 37, 704-710.

Rush, B.R., Flamino, M.J. and Matson, C.J. (1998a)Cytologic evaluation of bronchoalveolar lavage fluidin horses with recurrent airway obstruction afteraerosol and parenteral administration ofbeclomethasone dipropionate and dexamethasone,respectively. Am. J. vet. Res. 59, 1033-1038.

Rush, B.R., Raub, E.S., Matson, C.J. and Hakala, J.E.(1999) Incremental doses of aerosolizedbeclomethasone in horses with recurrent airwayobstruction. Proc. 17th Symp. Comp. Resp. Soc. 95, pp 95

Rush B.R., Raub E.S. and Rhoads W.S. (1998b)Pulmonary function in horses with recurrent airwayobstruction after aerosol and parenteraladminstration of beclomethasone dipropionate anddexamethasone, respectively. Am. J. vet. Res. 59,1039-1043.

Rush Moore, B.R., Krakowka, S., Cummins, J.M. andRobertson, J.T. (1996) Changes in airwayinflammatory cell populations in Standardbredracehorses after interferon-alpha administration. Vet.Immun. Immunopath. 49, 347-358.

Seahorn, T.L., Martens, J.G. and Carter, G.K. (1990)Effects of human-alpha interferon on experimentallyinduced equine herpesvirus-1 infection in horses.Am. J. vet. Res. 51, 2006-2010.

Vail, C.D., Nestved, A.J., Martins, J.B., Huff, G.K. andHargrove T.B. (1990) Adjunct treatment of equinerespiratory disease complex (ERDC) with thePropionibacterium acnes, immunostimulant,EqStim. J. equine vet. Sci. 10, 399-403.

Viel, L., Celly, C. and Staempfli, H. (1999) Therapeuticefficacy of inhaled fluticasone propionate in horseswith chronic obstructive pulmonary disease. Am.Ass. equine Pract. 45, 306-307.


7

INFLAMMATORY AIRWAY DISEASE: EUROPEAN CLINICIANS’ PERSPECTIVE

P. M. Dixon, B. C. McGorum and R. S. Pirie

Department of Veterinary Clinical Studies, University of Edinburgh, UK

The term inflammatory airway disease (IAD) isnot used widely in Britain at present because thereis much uncertainty about which equinepulmonary disorder(s) this term describes. TheInternational Workshop on Equine AirwayDisease, held in Michigan State University in2000, described IAD as a non-septic airwaydisease, particularly of young horses, which hasan unknown relationship to heaves (Robinson2001).

To date, IAD has been used by some cliniciansas a ‘dustbin’ term for many pulmonary disordersthat cannot be categorised into a more clear-cutdiagnosis. The term IAD lends itself to this vaguerole, as almost all equine pulmonary diseasesinvolve the airways and it is unclear how anyairway disease could be anything exceptinflammatory in its pathogenesis, whether it be ofan infectious, inhaled-irritant or hypersensitiveaetiology. Even the extremely rare neoplasticequine pulmonary diseases have an inflammatorycomponent.

In contrast to a North American referralcaseload, in Britain cases of pleuropneumonia orother types of bacterial pulmonary disease in adulthorses are seen relatively seldom and, thus, mostof the referred caseload will be of chronic (>2months duration) lower grade, diffuse pulmonarydisease, with affected horses having excessivemucopurulent tracheal respiratory secretionspresent on bronchoscopy. At Edinburgh, workershave moved away from pulmonary function testsin the diagnosis of clinical cases of pulmonarydisease, because of their insensitivity in cases withlower grade pulmonary disease. Historical,clinical, bronchoscopic and especially, trachealrespiratory secretions and bronchoalveolar lavage(BAL) fluid cytology findings are currently themainstay for diagnosis.

The limitations of clinical examinations in thediagnosis of lower grade equine pulmonarydisease cannot be overemphasised. Although anaccurate history, and especially bronchoscopy, canconfirm the presence of pulmonary disease,pulmonary cytology forms a mainstay fordiagnosing the specific chronic pulmonary diseasepresent; using previously described diagnosticcriteria (Dixon et al. 1995). The use of bothtracheal respiratory secretions (RS) and BAL fluidcytology is most helpful – eg in heaves (COPD)affected horses, both samples will have apermanent neutrophilia (>5% neutrophils in BALfluid and usually 3–4 times this ratio of neutrophilsin the tracheal RS). With infectious (postinfectious) pulmonary diseases affected cases willhave a transient (<2 weeks) BAL fluidneutrophilia, but a longer-term (ie for manymonths) tracheal RS neutrophilia, usually with anormal BAL fluid cytology.

After establishing a definite diagnosis in asmany pulmonary cases as is possible, a significantnumber of horses are always left where nodefinitive diagnosis can be made, using currentunderstanding and the available ancillarydiagnostic techniques. Rather than artificiallytrying to manipulate these cases into some otherdiagnostic group, to date they have been simplyclassified as undifferentiated pulmonary disease.In many respects, including clinically (ie thepresence of chronic, diffuse, low grade pulmonarydisease); bronchoscopically (inflamed airways andexcessive mucopurulent tracheal RS) andcytologically (tracheal RS neutrophilia and normalBAL fluid cytology) these undifferentiated casesare similar to the ‘post infectious’ pulmonarydisease group - except that they have no history ofimmediately antecedent respiratory infection.Using current knowledge, this group of horses


8

with undiagnosed pulmonary disease could wellbe classified as suffering from IAD, as perhapscould some of the ‘post infectious’ pulmonarydisease group – especially those that do not resolvewith 6 weeks or so of environmental control, as domost of their cohorts.

Over the last 15 years, largely due to the workof Sasse (1971) and McPherson et al. (1978), therehas been a major change in the use of ensiled grassversus hay for feeding indoor horses and to aslightly lesser extent, in the use of non-strawversus straw bedding in the UK. Almost all casesreferred to Edinburgh are currently maintained inhay and straw-free environments. Thesemanagement changes have resulted in a decreasein the incidence of classic heaves and severelyaffected (dyspnoeic) cases are rarely seen atpresent, in contrast to 15–20 years ago. Whilstthese management changes have decreased theinhaled challenge to indoor horses, they have noteliminated it. It is interesting to speculate whetherhorses (possibly susceptible to heaves), whichexperience lung disease due to lower degrees ofenvironmental inhalation challenge, could beclassified as suffering from IAD?

As the authors’ caseload is a selected referralpopulation from a geographically distinct area, theopinion of other clinicians with different caseloadsin other parts of Britain were also sought. P.Ramzan and M. Shepperd working with 2–3-year-old flat racehorses in Newmarket are nowtentatively diagnosing IAD in this group of horses,especially in individual horses that develop a morechronic pulmonary disease following apparentrespiratory infectious disease. Horses recentlyintroduced to the indoor environment at the start oftraining appear to be at higher risk.

Tim Mair an equine practitioner in Kent,dealing primarily with sports and leisure horses,defined IAD as a non-heaves, chronic pulmonarydisease of indoor horses of all ages, but did notfrequently recognise this disorder in his patients.

IAD also appeared to occur in younger (1–2-year-old) horses in indoor environments and these casesappeared to respond to environmental control. TimBrazil working in an equine practice in Oxfordalso recognised the disorder, including in NationalHunt racehorses when first brought indoors fortraining at 4–5 years of age. The disorder wascharacterised by poor work, variable respiratoryclinical signs, and the bronchoscopic detection ofexcessive mucopurulent tracheal secretions ofneutrophilic character. The stress of competitivework, antecedent respiratory infection andadaptation to new indoor environments werebelieved to be risk factors for this syndrome.

Although no consensus was present, apparentagreement between the UK veterinarians was thatIAD is:• A disease of indoor horses.• A diffuse, lower grade pulmonary disease.Affected horses have:• Increased volumes of mucopurulent secretions

in their trachea (detected bronchoscopically).• Tracheal respiratory secretions are neutrophilic

in character.• Antecedent respiratory infection is a risk factor.• First coming indoors to training is a risk factor.• Performing high speed exercise is a risk factor.

REFERENCES

Dixon, P.M., Railton, D.I. and McGorum, B.C. (1995)Equine pulmonary disease: a case controlled studyof 300 referred cases. Part 1: Examinationtechniques, diagnostic criteria and diagnoses.Equine vet. J. 27, 416-421.

McPherson, E.A., Lawson, G.H.K., Murphy, J.R.,Nicholson, J.M., Fraser, J.A., Breeze, R.G. and Pirie,H.M. (1978) Chronic obstructive pulmonary disease(COPD): Identification of affected horses. Equinevet. J. 10, 47-53.

Robinson, N.E. (2001) Proceedings of the InternationalWorkshop on Equine Airway Disease, MichiganState University. Equine vet. J. 33, 5-19.

Sasse, H.H.L. (1971) Some Pulmonary Function Tests inHorses. Ph.D Thesis: State University of Utrecht. pp15-230.


9

INFLAMMATORY AIRWAY DISEASE: EFFECT OF ATHLETIC DISCIPLINE

M. Mazan and A. Hoffman

Lung Function Testing Laboratory, Tufts University School of Veterinary Medicine, 200 Westboro Road,North Grafton, Massachusetts 01536, USA

Despite an increasing awareness of the importanceof inflammatory airway disease (IAD) in limitingperformance, the definition of the disease remainsfluid. There appears to be a consensus that horseswith IAD are: generally young, although middle-aged horses may be affected (Hare et al. 1994) andhave impaired performance that may go unnoticedat rest or during light work. (Hare et al. 1994;Hoffman et al. 1998; Couëtil and Denicola 1999).Clinical signs which are reported are variable,including cough, nasal discharge and abnormal lungsounds (Morris 1991; Vrins et al. 1991; Viel 1997)as well as endoscopic evidence of tracheobronchialmucus accumulation (MacNamara et al. 1990;Vrins et al. 1991; Dixon et al. 1995a). Horses withIAD invariably have mild to moderate airwayinflammation, which may involve neutrophils(Fogarty and Buckley 1991; Moore et al. 1995)mast cells (Vrins et al. 1991; Hare et al. 1994;Hoffman et al. 1998; Hoffman and Mazan 1999),eosinophils (Viel 1997) or lymphocytes (Moore etal. 1995). The horses may have normal lungfunction at rest, but show evidence of airway hyper-reactivity on exposure to non-specific agents, suchas histamine (Klein and Deegen 1986; Hoffman etal. 1998), or evidence of flow limitations on forcedexpiratory manoeuvres (Couëtil et al. 2001).Although IAD is often thought of as a disease ofyounger horses which are expected to performathletically, and as such is distinct from the pictureof the dyspnoeic horse with overt recurrent airwaydisease (RAO), several workers have reported IADin older athletic horses (Persson and Lindberg 1991;Hoffman et al. 1998; Hoffman and Mazan 1999;Couëtil et al. 2001).

One of the clinical findings most commonlyreported in IAD is exercise intolerance, ordecreased performance, with or without overtsigns of respiratory disease. Horses appear to have

remarkable respiratory reserve; thus subclinicalairway disease may simply result in mildabnormalities on auscultation and endoscopy andoccasional coughing (Bracher et al. 1991). Theprevalence of coughing is hard to estimate, asmany studies use this as an inclusion criterion(Vrins et al. 1991; Christley et al. 2000; Couëtil etal. 2001), whereas others report coughing less than16–50% of the time (McKane et al. 1993; Mooreet al. 1995). Other clinical signs which have beenreported include prolonged respiratory recoveryrate (Hare and Viel 1998), nasal discharge afterexercise (Hare and Viel 1998), respiratoryembarrassment at exercise (Dixon et al. 1995a;Hare and Viel 1998), worsening of signs duringhot, humid weather and inability to perform workduring collection (Hoffman et al. 1998).

The reported effect of IAD on performance hasalso varied. It seems to depend on the type of horseand the methods used by the individual investigator.Researchers looking at Standardbred racehorses(MacNamara et al. 1990) found that those withexcessive tracheal mucus performed at a lower levelthan those with no mucus on endoscopy. Couëtiland Denicola (1999) and Nyman et al. (1999)found, in separate studies, that horses with IADundergoing a treadmill stress test demonstratedmore severe impairment of gas exchange duringpeak exercise than normal horses. Persson andLindberg (1991) looking at Standardbred trottersand saddle horses with what was effectively IAD,found a different manifestation in racehorses tosaddle horses. Not only were the saddle horsessignificantly older when airway disease wasdetected, but racehorses had decreased minutevolume and an increased red blood cell tobodyweight ratio, whereas the affected saddlehorses manifested tachycardia during exercise.Other studies have found that horses with obvious


10

evidence of airway inflammation do not necessarilyhave a history of exercise intolerance (Sweeney etal. 1992). These differences in physiological andclinical response to exercise probably relate to thedemands placed on racehorses compared to ridinghorses. This may also reflect the difficulty ofdiagnosing low-grade respiratory impairment andthe trainer’s failure to recognise poorer performancethan nature intended, rather than the benign natureof the underlying disease.

The authors hypothesised that IAD is aprogressive disease and, therefore, will presentdifferently according to aerobic demands. Clinicalrecords were reviewed of horses presented to theLarge Animal Hospital at the Tufts UniversitySchool of Veterinary Medicine from 1996–2002with a primary complaint of lowered performance,and an eventual diagnosis of IAD based on abnormalbronchoalveolar lavage (BAL) cytology (>5%neutrophils (PMNs), >2% mast cells or >0.5%eosinophils; Hoffman et al. 1998). Horses were

excluded if they had signs of heaves at rest, a historyof recent respiratory infection or recurrent increasedrespiratory effort when exposed to hay or pasture,compatible with RAO. Horses were placed into 2categories: racehorses (RH, n=41) or other sporthorses (SH, n=66), to determine if IAD presenteddifferently in these 2 groups. The data are shown inTables 1–5. The following variables were comparedbetween the 2 groups: 1) signalment (age, sexdistribution and breed); 2) history (access to turnout,hay feeding, effect of hot, humid weather, durationof signs and history of abnormal respirations atwork, cough, lethargy, quitting or exercise-inducedpulmonary haemorrhage); 3) physical examination(resting respiratory rate, abnormal tracheal or lungauscultation, nasal discharge or cough); 4) BAL(percentage of BAL cells classified as macrophages,

TABLE 1: Signalment

Variable Racehorse Sport horse

Age (years) 4 ± 2.1 11.0 ± 4.9*Thoroughbred

(% of racehorses) 58Standardbred

(% of racehorses) 42Gelding (% of total) 46 62Mare (% of total) 41 27Stallion (% of total) 9 6

*Significantly different from racehorses, P<0.05

TABLE 2: History

Variable Racehorse Sport horse (% positive responses)

Has turnout 33 86*Eats hay 98 85Affected primarilyby hot, humid weather 69 59Signs present > 6 months 20 77*History of abnormal

respirations at work 56 77*History of cough 14 50*History of lethargy 43 59History of quitting 93 3*History of EIPH or

BAL evidence 35 7*


TABLE 3: Physical examination

Variable Racehorse Sport horse(% positive response)

Resting respiratory rate > 20/min 24 47*

Abnormal tracheal auscultation 21 64*

Abnormal lung sounds 24 54*Nasal discharge present

at physical examination 21 59*Cough heard or elicited

on PE 33 76*


TABLE 4: BAL

Variable Racehorse Sport horse (mean ± SD) (mean ± SD)

% neutrophils 5.1 ± 7.1 10.4 ± 11.8*% mast cells 4.25 ± 2.8 3.4 ± 2.2% macrophages 47.3 ± 17.9 43.1 ± 14.93% lymphocytes 39.9 ± 15.9 43.0 ± 15.2% eosinophils 0.59 ± 1.9 0.47 ± 1.7


TABLE 5: Lung function

Variable Racehorse Sport horse (mean ± SD) (mean ± SD)

RRS 2HZ(cmH2O/L/s) 0.62 ± .31 0.59 ± 0.21

PC100RRS(mg/ml histamine) 3.56 ± 2.2 4.28 ± 4.05


11

lymphocytes, neutrophils, mast cells and eosinophilsby counting 500 cells stained with Wright-Giemsa[1,000 Xmag]); 5) lung function (baseline RRS at 2Hz and PC100 RRS2Hz [provocative concentration ofhistamine that doubled RRS at 2 Hz; Mazan et al.1999]).

INFORMATION OBTAINED FROM THE STUDY

Signalment: There was a difference in age betweenthe 2 groups, with SH being significantly olderthan RH (Table 1). There were no significantdifferences in sex distribution. History (Table 2):As expected, few RH had access to turnout, andmost horses, both RH and SH, were given hay asroughage. As organic dust associated with a barnenvironment and hay feeding have been associatedwith lower airway inflammation (Sweeney et al.1992; Holcombe et al. 2001) this finding may beimportant. A significant difference was foundbetween the 2 groups with respect to duration ofsigns, with most RH having signs for less than 6months, whereas most SH had signs for over 6months. This may well reflect the level of exercisenecessary to force a diagnosis through signs ofexercise intolerance and, perhaps, the greaterlikelihood of a racehorse being examinedendoscopically. It was also found that asignificantly greater proportion of SH showedabnormal respiration at work and cough than didRH. Racehorses, on the other hand, were mostlikely to ‘quit’ during hard work. Again, SH areless likely to experience overt exercise intoleranceas a consequence of IAD, due to the decreasedaerobic demands of the sport. Both groups seemedto have more difficulty in hot, humid weather withequal frequency. This may reflect an increasedgrowth of moulds and bacteria with consequentelevations in endotoxin levels, the effects ofassociated air pollution, or greater effects ofinertance in moving heavier, more saturated air.Conversely, it may also reflect the greaterlikelihood of horses being exercised maximallyduring the warmer seasons.

When considering physical examinationfindings (Table 3), a significantly greaterproportion of SH had an abnormally high restingrespiratory rate, nasal discharge, cough orabnormal tracheal or lung auscultation, althoughthese may be variably found in RH as well. Thismay be due to the progressive nature of IAD: SHtend to be older, have had signs for longer and havea lower athletic demand. It is likely that SH owners

fail to notice incipient signs of IAD as it has a lessprofound impact on the work performed: onlywhen more obvious signs appear, such as coughand nasal discharge, are these horses brought to aveterinarian’s attention. Interestingly, cough washeard on physical examination of only 33% of RH,whereas 93% had quit at the 3/4 mark. Thisunderscores the need for more detailed methods ofexamination, including BAL and lung functiontesting, to confirm a diagnosis in these horses.

BAL findings varied significantly betweengroups (Table 4). Both groups had an elevatedmast cell percentage, with no significant differencebetween them. There was, however, a significantdifference between neutrophil percentages, withSH having double the percentage of the RH. This,rather than reflecting a fundamental differencebetween groups, is more likely to reflect aprogressive inflammatory insult in the older horsesof the SH group. Finally, although both groups hadbaseline RRS2Hz within normal limits, they bothexperienced airway hyper-reactivity (Table 5). Itshould be noted that data at 2Hz were used in orderto include the maximum number of IAD horses inthis study, as some horses had missing data fromRRS1-3.

In summary, SH differ from RH in that theyare older, have more obvious clinical signs and agreater degree of airway inflammation; probablyreflecting a progression of disease. Racehorsespresent more of a diagnostic challenge as they areless likely to have easily detectable clinicalabnormalities and more likely to present as an‘unknown’ case of exercise intolerance. Indeed, ifhorses with cough and nasal discharge absent onhistory and physical examination are isolated, astriking difference is found between the 2 groups.If occult disease is defined as no cough or nasaldischarge on history or physical examination, IAD

TABLE 6: Occult disease

Variable Racehorse Sport horse(% positive response)

Occult disease – no nasal discharge or cough on PE or physical examination 54 17*

Occult disease – no nasal discharge, cough, or auscultation of lungs or trachea 40 10*



12

was occult in 54% of RH, but only 17% of SH(Table 6). If the definition is more stringent andincludes the absence of abnormal chest or trachealauscultation, occult disease still occurs in 40% ofRH, but only 10% of SH. In addition toabnormalities on BAL, IAD in RH is most reliablydetected through bronchoprovocation todemonstrate airway hyper-reactivity andunderlying airway obstruction. This is discussed inmore detail by Hoffman and Mazan (2003).

Inflammatory airway disease has thus farproduced more questions than answers, and muchwork remains to be done. If the mechanism ofairway hyper-reactivity, the stimulus forinflammation, and the role that genetics andenvironment plays in IAD can be discovered, itmay ve possible to modify the development ofdisease in young horses and prevent itsprogression. The suspicion that IAD, leftuntreated, will eventually develop into thecrippling disease, RAO, only heightens concern,and provides the stimulus to solve the manymysteries of IAD.

REFERENCES

Bracher, V., von Fellenberg, R., Winder, C.N., Gruenig,G., Hermann, M. and Kraehenmann, A. (1991) Aninvestigation of the incidence of chronic obstructivepulmonary disease (COPD) in random populationsof Swiss horses. Equine vet. J. 23,136-141.

Christley, R.M., Rose, R.J. and Hodgson, D.R. (2000)Attitudes of Australian veterinarians about the causeand treatment of lower-respiratory-tract disease inracehorses. Prev. vet. Med. 46, 149-159.

Couëtil, L.L. and Denicola, D.B. (1999) Blood gas,plasma lactate and bronchoalveolar lavage cytologyanalyses in racehorses with respiratory disease.Equine vet J. Suppl. 30,77-82.

Couëtil, L.L., Rosenthal, F.S., DeNicola, D.B. andChilcoat, C.D. (2001) Clinical signs, evaluation ofbronchoalveolar lavage fluid, and assessment ofpulmonary function in horses with inflammatoryrespiratory disease. Am. J. vet. Res. 62, 538-546.

Dixon, P.M., Railton, D.I. and McGorum, B.C. (1995a)Equine pulmonary disease: a case control study of300 referred cases. Part 3: Ancillary diagnosticfindings. Equine vet. J. 27, 428-435.

Fogarty, U. and Buckley, T. (1991) Bronchoalveolarlavage findings in horses with exercise intolerance.Equine vet. J. 23 434-437.

Hare, J.E., Viel, L., O’Byrne, P.M. and Conlon, P.D.(1994) Effect of sodium cromoglycate on lightracehorses with elevated metachromatic cell numberson bronchoalveolar lavage and reduced exercisetolerance. J. vet. Pharmac. Ther. 17, 237-244.

Hare, J.E. and Viel, L. (1998) Pulmonary eosinophilia

associated with increased airway responsiveness inyoung racing horses. J. vet. Intern. Med. 12, 163-170.

Hoffman, A.M., Mazan, M.R. and Ellenberg, S. (1998)Association between bronchoalveolar lavagecytologic features and airway reactivity in horseswith a history of exercise intolerance. Am. J. vet.Res. 59, 176-181.

Hoffman, A.M. and Mazan, M.R. (1999) Programme oflung function testing horses suspected with smallairway disease. Equine vet. Educ. 11, 322-328.

Hoffman, A. and Mazan, M. (2003) Airway obstrction andhyper-reactivity in horses with signs of inflammatoryairway disease. Workshop on ‘Inflammatory AirwayDisease: Defining the Syndrome’. HavemeyerFoundation Monograph Series No 9, Eds: A.Hoffman, N. E. Robinson and J.F. Wade, R&WPublications (Newmarket) Ltd, pp 71-74.

Holcombe, S.J., Jackson, C. and Gerber, V. (2001)Stabling is associated with airway inflammation inyoung Arabian horses. Equine vet. J. 33, 244-249.

Klein, H.J. and Deegen, E. (1986) Histamine inhalationprovocation test: method to identify nonspecificairway reactivity in equids. Am. J. vet. Res. 47, 796-800.

MacNamara, B., Bauer, S. and Lafe, J. (1990) Endoscopicevaluation of exercise-induced pulmonary hemorrhageand chronic obstructive pulmonary disease inassociation with poor performance in racingStandardbreds. J. Am. vet. med. Ass. 196, 443-445.

Mazan, M.R., Hoffman, A.M. and Manjerovic, N. (1999)Comparison of forced oscillation with the conventionalmethod for histamine bronchoprovocation testing inhorses. Am. J. vet. Res. 60, 174-180.

McKane, S.A., Canfield, P.J. and Rose, R.J. (1993)Equine bronchoalveolar lavage cytology: survey ofthoroughbred racehorses in training. Aust. vet. J. 70,401-404.

Moore, B.R., Krakowka, S., Robertson, J.T. andCummins, J.M. (1995) Cytologic evaluation ofbronchoalveolar lavage fluid obtained fromstandardbred racehorses with inflammatory airwaydisease. Am. J. vet. Res. 56, 562-567.

Morris, E. (1991) Application of clinical exercise testingfor identification of respiratory fitness and disease inthe equine athlete. Vet. Clin. N. Am. Equine Pract. 7,383-401.

Nyman, G., Bjork, M. and Funkquist, P. (1999) Gasexchange during exercise in standardbred trotterswith mild bronchiolitis. Equine vet. J. 30, 96-101.

Persson, S. and Lindberg, R. (1991) Lung biopsy pathologyand exercise tolerance in horses with chronicbronchiolitis. Equine Exer. Physiol. 3, 157-164.

Sweeney, C.R., Humber, K.A. and Roby, K.A. (1992)Cytologic findings of tracheobronchial aspiratesfrom 66 thoroughbred racehorses. Am. J. vet. Res.53, 1172-1175.

Viel, L. (1997) Small airway disease as a vanguard forchronic obstructive pulmonary disease. Vet Clin. N.Am. Equine Pract. 13, 549-560.

Vrins, A., Doucet, M. and Nunez-Ochoam L. (1991) Aretrospective study of bronchoalveolar lavagecytology in horses with clinical findings of smallairway disease. Zentbl. VetMed. A. 38, 472-479.


13

MUCUS, COUGH, AIRWAY OBSTRUCTION AND INFLAMMATION

N. E. Robinson, C. Berney, H. DeFeijter-Rupp, A. M. Jefcoat, C. Cornelisse, V. Gerber, and F. J. Derksen

Pulmonary Laboratory, Department of Large Animal Clinical Sciences, College of VeterinaryMedicine, Michigan State University, East Lansing, Michigan 48824, USA

Coughing, mucus accumulation and airwayobstruction are features of inflammation of theairways, which occur to varying degreesdepending on the pathogenesis of theinflammation and the functional and structuralchanges in the lungs. For example, in humanasthma, much of the obstruction is due tobronchospasm, whereas in human chronicobstructive pulmonary disease (COPD),bronchospasm is much less important and mucusis the major cause of obstruction (Jeffery 2000).

In horses, cough is a highly specific sign ofpulmonary disease. Coughing occurred in 71% of300 horses referred to the equine clinic atEdinburgh University with suspected respiratorydisease (Dixon et al. 1995a). In racehorses, coughis strongly associated with the presence of airwayinflammation and accumulated secretions in thetrachea (Burrell et al. 1996; Christley et al.2001a,b). However, many racehorses haveaccumulated secretions in the absence of cough.For this reason, Burrell et al. (1996) concludedthat coughing is a highly specific, but relativelyinsensitive sign of airway inflammation. Cough,which clears foreign materials from the largerairways, is initiated by activation of coughreceptors located under and within the airwayepithelium of the pharynx, larynx, trachea andbronchi (Widdicombe 1996). Therefore, althoughcough is a highly specific sign of airway disease, itis not specific for certain types of disease. Horseswith airway foreign bodies, pulmonary oedema,heaves, infectious disease and lungworms allcough. Coughing is less common in horses withexercise-induced pulmonary haemorrhage (EIPH),a problem located within the alveoli and small,rather than large, airways.

Accumulation of airway secretions is also anon-specific sign of airway disease, and more than

a few drops of mucus within the trachea isabnormal (Dixon et al. 1995b). Accumulatedsecretions are well documented in heaves-affectedhorses, both during clinical exacerbations and inremission (Dixon et al. 1995b; Jefcoat et al. 2001).Secretions also accumulate in the presence ofinflammatory airway disease, infectious disease,and lungworms, but are a less consistent sign ofEIPH (Dixon et al. 1995b). Secretions arecomposed of mucus and inflammatory cells.Mucus is produced and secreted by goblet cells inresponse to release of secretagogues such asneutrophil elastase.

Airway obstruction is clinically obvious inhorses with heaves, in which it is associated withmigration and accumulation of neutrophils withinthe airways (Robinson et al. 1996). Althoughobstruction is not clinically obvious in otherairway diseases, measurements of lung functionhave demonstrated airway obstruction in horseswith inflammatory airway disease in the absenceof clinical signs. The majority of the resistance tobreathing is located in the larger (nasopharynx,trachea, bronchi) rather than the smaller(bronchioles) airways. For this reason, there mustbe extensive bronchiolar obstruction before horsesshow respiratory distress which is obviousclinically. Small airway obstruction can lead toabnormal distribution of ventilation within the gasexchange region before the horse appearsclinically ill. This abnormal ventilationdistribution impairs oxygen exchange and maylead to poor performance.

The association among cough, accumulatedsecretions, airway obstruction and inflammationhas not been extensively studied in horses. Dixonet al. (1995b) noted that there were greatervolumes of tracheal secretions in coughing than innon-coughing horses. However, as noted above,


14

racehorses can have accumulated airwaysecretions containing increased numbers ofneutrophils and not cough. In racehorses,coughing is associated with the presence ofinflammatory cells, especially neutrophils inairway secretions.

Evaluation of the severity of airwayinflammation is generally based on the proportionof inflammatory cells, especially neutrophils, in thetracheal secretions or in bronchoalveolar lavage(BAL) fluid (McGorum et al. 1993). The range ofneutrophil ratios (NR) in the tracheal wash ofhorses without clinical signs of airway disease isvery large but the range is smaller in BAL fluid.Cough and airway obstruction are generallyassociated with an increase in NR, but some stabledhorses with no clinical signs of disease can haveNR as high as those of some horses with heaves(Tremblay et al. 1993). However, depending on thetotal number of inflammatory cells in the airways,a given NR may be associated with a very high ora much lower total number of neutrophils. Totalneutrophil count and neutrophil activity have neverbeen investigated in association with cough,mucus, and airway obstruction in horses.

The relationship among cough frequency(counted over 4 h), accumulated trachealsecretions (range 0–5), airway obstruction(measured as maximal change in pleural pressureduring tidal breathing [�Pplmax]), and BAL fluidcytology in 12 heaves-affected and 5 controlhorses was investigated recently (Robinson et al.2003). When heaves-affected horses were stabled,cough frequency increased over 3 days. Coughfrequency also increased transiently on Day 1 ofstabling in control horses but then subsided. Inheaves-affected animals, cough was initiallysporadic with periods of several hours with nocough and then multiple coughs within a 15 minperiod. For this reason, it is important to countcough for long periods when attempting toevaluate cough severity. The judgement of ownersor trainers on cough severity will be greatlycoloured by when they were in the stable: duringbouts of coughing or between bouts. Evaluation ofthe data from this goup of horses produced thefollowing conclusions:

1. Horses that cough have accumulated airwaysecretions, but horses with accumulatedsecretions may not cough.

2. Horses that have measurable airwayobstruction have accumulated secretions, but

horses with accumulated secretions may nothave measurable airway obstruction.

3. Horses that cough have measurableabnormalities of lung function, and mosthorses with abnormal lung function cough.

In heaves-affected horses, BAL fluid NR, total cellcount and total neutrophil count all increasedduring stabling. In control horses, NR alsoincreased during stabling, but the increase in totalneutrophil count was much less than in heaves-affected animals. The relationship between BALfluid and either cough frequency, accumulatedsecretions, or airway obstruction had a largeamount of scatter. Whereas in heaves-affectedanimals, a BAL fluid NR greater than 10% wasassociated in general with the presence of cough,accumulated secretions and airway obstruction,greater than 10% neutrophils in control horses wasnot associated with cough, secretions or airwayobstruction.

The relationship between inflammation andlung functions became much more apparent whencough, accumulated secretions, and airwayobstruction were examined as a function of BALfluid total neutrophil count. When neutrophilcount was less than 30 cells/µl, there was verylittle accumulated airway secretion. Secretionsincreased between 30 and 100 cells/µl andappeared to plateau above 100 cells/µl. There wasno measurable airway obstruction when neutrophilcount was less than 50 cells/µl and obstructionincreased between 100 and 1000 cells/µl.Coughing began when neutrophil count exceeded100 cells/µl. These data explain why horses canhave secretions and not cough but not vice versa.Accumulation of secretions begins at less severelevels of inflammation than those required toinitiate cough. Cough is therefore a sign of quitesevere airway inflammation. It is important torealise that the relationships described herein arederived from heaves-affected and control animals.It would be interesting to determine if similarrelationships exist in other inflammatory airwaydiseases of the horse.

REFERENCES

Burrell, M.H., Wood, J.L.N., Whitwell, K.E., Chanter,N., Mackintosh, M.E. and Mumford, J.A. (1996)Respiratory disease in thoroughbred horses intraining: the relationships between disease andviruses, bacteria and environment. Vet. Rec. 139,308-313.


15

Christley, R.M., Hodgson, D.R., Rose, R.J., Hodgson,J.L., Wood, J.L. and Reid, S.W. (2001a) Coughing inthoroughbred racehorses: risk factors and trachealendoscopic and cytological findings. Vet. Rec. 148,99-104.

Christley, R.M., Hodgson, D.R., Rose, R.J., Wood, J.L.,Reids, S.W., Whitear, K.G. and Hodgson, J.L.(2001b) A case-control study of respiratory diseasein Thoroughbred racehorses in Sydney, Australia.Equine vet. J. 33, 256-264.

Dixon, P.M., Railton, D.I. and McGorum, B.C. (1995a)Equine pulmonary disease: a case control study of300 referred cases. Part 2: details of animals andhistorical and clinical findings. Equine vet. J. 27,422-427.

Dixon, P.M., Railton, D.I. and McGorum, B.C. (1995b)Equine pulmonary disease: a case control study of300 referred cases. Part 3: Ancillary diagnosticfindings. Equine vet. J. 27, 428-435.

Jefcoat, A.M., Hotchkiss, J.A., Harkema, J.R., Basbaum,C.B. and Robinson, N.E. (2001) Persistent mucinglycoprotein alterations in equine recurrent airwayobstruction. Am. J. Physiol. Lung cell. mol. Physiol.281, 704-712.

Jeffery, P.K. (2000) Comparison of the structural and

inflammatory features of COPD and asthma. Giles F.Filley Lecture. Chest 117, 251-260.

McGorum, B.C., Dixon, P.M., Halliwell, R.E.W. andIrving, P. (1993) Comparison of cellular andmolecular components of bronchoalveolar lavagefluid harvested from different segments of theequine lung. Res. vet. Sci. 55, 57-59.

Robinson, N.E., Berney, C., Eberhart, S., deFeijter-Rupp,H., Jefcoat, A.M., Cornelissse, C.J., Gerber, V.M.and Derksen, F.J. (2003) Coughing, mucusaccumulation, airway obstruction, and airwayinflammation in control horses and horses affectedwith recurrent airway obstruction. Am. J. vet. Res.64, 550-557.

Robinson, N.E., Derksen, F.J., Olszewski, M.A. andBuechner-Maxwell, V.A. (1996) The pathogenesisof chronic obstructive pulmonary disease of horses.Br. vet. J. 152, 283-306.

Tremblay, G.M., Ferland, C., Lapointe, J.-M., Vrins, A.,Lavoie, J.P. and Cormier, Y. (1993) Effect of stablingon bronchoalveolar cells obtained from normal andCOPD horses. Equine vet. J. 25, 194-197.

Widdicombe, J.G. (1996) Sensory neurophysiology of thecough reflex. J. Allergy Clin. Immunol. 98, 84-89.


16

RELATIONSHIP BETWEEN COUGHING AND AIRWAYINFLAMMATION IN YOUNG RACEHORSES

D. R. Hodgson, R. M. Christley, J. L. N. Wood*, S. W. J. Reid† and J. L. Hodgson

Faculty of Veterinary Science, University of Sydney, New South Wales 2567, Australia; *Centre forPreventative Medicine, Animal Health Trust, Lanwades Park, Kentford, Newmarket, Suffolk CB8 7UU,UK; †Veterinary Informatics and Epidemiology, University of Glasgow, Glasgow G61 1QH, Scotland

INTRODUCTION

Coughing, an important problem in youngracehorses, is a relatively specific indicator of lowerairway disease. Potential causes in this group ofhorses include pneumonia and pleuritis, viral andbacterial respiratory infections, exercise inducedpulmonary haemorrhage, chronic obstructivepulmonary disease and pharyngeal lymphoidhyperplasia. However, the most common cause ofcoughing among Thoroughbred racehorses is likelyto be inflammatory airway disease (IAD), asyndrome resulting in mucopurulent trachealdischarge containing predominantly neutrophils.Despite this, risk factors for coughing in racehorsesare unknown. Young racehorses represent a uniquepopulation, and differ from the general horsepopulation in ways that may be relevant to theepidemiology of coughing. As well as the agefactor, they tend to be involved in strenuous activity,are transported regularly, mix with other horses and,in general, are housed in stables.

Tracheal endoscopy and cytological evaluationof lower airway secretions are performedcommonly in investigations of airway disease inhorses. However, the value of the cytologicalassessment of tracheal wash samples, as anindication of lower airway pathology, has beenquestioned due to the wide variability in relativenumbers of inflammatory cells in apparentlynormal horses. Therefore, a study was undertakenwith 2 major aims: 1) to determine if coughing is auseful indicator of lower airway inflammation byinvestigating the association between coughing andtracheal inflammation, as measured by trachealendoscopy and cytology; 2) to evaluate a number ofpotential risk factors for coughing in Thoroughbredracehorses in Sydney, Australia.

MATERIALS AND METHODS

To this end a matched case control study wasconducted using 100 horses that coughed at least 4times within minutes of commencing exercise and148 controls, the latter being free of clinical signsof respiratory tract disease. Horses were identifiedfrom participating stables at 5 racetracks aroundSydney by daily contact with the trainers and theirconsulting veterinarians. Cases were defined asThoroughbred racehorses in training that coughedat least 4 times during a 10 min period of exercise.The case definition was kept simple because caseidentification relied on the observations of thetrainer and the staff of the stable.

Controls were selected randomly from amongnon-cases in the stable from which the case wasidentified. A horse was defined as a non-case if ithad not had a recognised episode fulfilling theselection criteria within the past 2 weeks.

Information relating to the recent history ofeach case and control was collected using aquestionnaire. The investigators completed thequestionnaire during discussion with the trainer ortrainer’s representative. Details recorded includedthe name, age and sex of the horse, the number ofdays since last transportation (1–7 days, 8–14days, more than 14 days) and racing (never raced,1–7 days, more than 7 days), and the stage oftraining. Stage of training was categorised as early(mostly trotting and cantering), mid (increasingfast exercise, up to 80% of top speed, usually overshort distances), late (training included increasingamounts of exercise top speed gallop) or racing(has raced). Early, mid and late training phasesusually lasted 4–5 weeks each.

Endoscopy of the upper and lower airways,and tracheal fluid collection were performed using


17

Darien Microbiology Aspiration Catheter using 12ml of phosphate buffered saline (PBS). Prior toinstallation of PBS, the trachea was inspected anda score given for the following; nasopharyngealmucus, pharyngeal lymphoid hyperplasia, lowerrespiratory tract mucus.

Slide preparations of tracheal aspirates weremade by cytocentrifugation with 1 ml of undilutedsample. Following staining with Dif-Quik, anestimate of the relative numbers of alveolarmacrophages, haemosiderophages, lymphocytes,neutrophils, giant cells and eosinophils was madeby making counts of 100 cells in 3 different areasof the slide. The relative number ofhaemosiderophages was expressed as a proportionof all inflammatory cells.

As cases and controls were selectedconcurrently, each case was matched with itscontrol horses on the basis of time and trainingstable. Therefore, analyses had to take account ofthis matching. Crude odds ratios were calculated forall variables using conditional univariable logisticregression, with coughing as the dependentvariable. Dummy variables were generated for anycategorical variable with more than 2 levels.Multivariable conditional logistic regression modelswere constructed through backward eliminationusing variables considered, a priori, to be potentialconfounders, and those which were significant atP<0.25 in the univariable screening. Variablesremained in the model if removal of that variableresulted in a significant change (P<0.05) in the

likelihood ratio statistic, or if exclusion of thevariable altered the estimate of effect of othervariables by 5% or more. Biologically meaningful2-way interaction terms between the independentvariables were examined after identification of thereduced set main effects. The fit of the multivariablemodel was assessed by examining the delta-betavalues. Horses with the greatest delta-beta valuesfor each variable were sequentially excluded fromthe models. Models were judged as stable androbust when these exclusions did not causesignificant effects. Conditional logistic regressionwas performed using LogXact version 2.1 (CytelStatistical Software, Massachusetts, USA).

Correlation coefficients between stage oftraining and time since transportation wereexamined, following stratification by time since lastrace, to examine the relationships between thesevariables. Correlation coefficients were calculatedusing Statistica (StatSoft Inc, Oklahoma, USA).

RESULTS AND CONCLUSIONS

The hallmark of IAD is mucopurulent trachealexudate containing neutrophils. Therefore, theresults of the current study indicate that a majorityof horses that cough during exercise have evidenceof IAD. In this study, coughing was stronglyassociated with increases in tracheal mucus scoreand percent neutrophils in tracheal wash samples;80% of cases had more than 20% neutrophils intheir tracheal wash, compared to 20% of controls.

TABLE 1: Factors identified by conditional logistic regression model as being associated with therisk of coughing in Thoroughbred racehorses

Cases Controls Unadjusted OR OR* 95% CI*

Age (years)1 or 2 62 49 1* 1†

3 21 34 0.3 0.5 0.2 1.74 10 40 0.1 0.2 0.04 0.7>4 7 25 0.1 0.2 0.04 0.9

Recent transport (days)1–7 19 50 1* 1†

8–14 18 33 1.9 3.5 0.9 13.6>14 63 65 3.6 5.5 1.7 17.8

Stage of trainingEarly 49 30 1* 1†

Mid 18 20 0.8 0.6 0.2 1.6Late 11 16 0.3 0.1 0.04 0.6Trialed or racing 22 82 0.1 0.05 0.009 0.2

Previous racingNever 57 41 1* 1†

1–7days 13 32 0.3 11.1 1.6 76.5>7days 30 75 0.2 1.2 0.4 3.9

*Adjusted for age, stage of training and previous racing; †reference category


18

TABLE 2: Multivariable conditional logistic regression models of tracheal wash cytological findingsafter controlling for age. Only the results for the explanatory variable in each model are shown

Univariable MultivariableOR OR* 95% CI*

Neutrophils (%) P<0.0010–20 1† 1†

21–40 9.2 10.7 3.0 37.641–60 98.4 140.6 12.8 1546.761–80 109.8 142.8 8.7 2346.681–100 67.1 103.7 8.6 1254.0

Haemosiderophages (%) P=0.30 1† 1†

1–10 0.5 0.8 0.4 1.711–20 0.2 0.8 0.2 3.5>20 0.1 0.3 0.05 1.3

Eosinophils (%) P=0.20 1† 1†

1–2 1.1 1.0 0.5 2.13–10 0.7 0.7 0.2 2.7>10 8.4 7.3 0.7 73.7

Mast cells (%) P=0.070 1† 1†

1–2 0.6 0.7 0.3 1.43–4 0.3 0.2 0.05 0.8>5 1.1 0.8 0.3 2.5

Ciliated epithelial cells (per ml) P<0.0010–20 1† 1†

21–50 0.1 0.2 0.05 0.551–80 0.05 0.06 0.01 0.281–100 0.01 0.01 0.002 0.06

Bacteria P<0.001None 1† 1†

Extra-cellular only 2.0 2.6 0.7 9.4A few intra-cellular 10.3 24.2 2.1 285.3Many intra-cellular 28.6 41.7 4.2 411.7

*Adjusted for age; †Reference category

Similarly, 94% of cases had at least some trachealmucus, compared to 45% of controls. Because ofthe design of this study, the sensitivity of coughingas an indicator of IAD could not be calculated. Asa result, the proportion of ‘true’ IAD casesidentified using coughing as the only case criterioncould not be determined. Assuming that thesensitivity of coughing, as an indicator of IAD,was as low as reported by others, a proportion ofhorses with IAD in the current study would nothave been identified. Consequently, these horsesmay have been included in the current study ascontrols, which would tend to bias the measure ofeffect toward the null. Therefore, the associationbetween IAD and clinical, endoscopic andclinicopathological findings may be greater thansuggested by the current study.

Variables identified by multivariableconditional logistic regression as ‘significantlyassociated with coughing’ included age (decreased

risk with increasing age), stage of training (horses inearly training at greatest risk), time since last race(horses that had never raced at greatest risk) andtime since transportation (horses transported over14 days previously were more likely to cough thanthose transported in the last week; Table 1). Horseswith cough were significantly more likely to haveincreased upper and lower tracheal mucus scoresand increased pharyngeal lymphoid hyperplasia. Inaddition, the tracheal aspirates of coughing horseshad increased odds of neutrophilic infiltration andwere more likely to have bacteria observed intra-cellularly when compared to control horses (Table2). However, despite this up to 20% of controlhorses had evidence of airway inflammation.

In summary, this study demonstrates a strongrelationship between airway inflammation andcoughing in racehorses during exercise.Management factors are the most likelycontributors to this inflammatory process.


19

POOR PERFORMANCE: THE TRAINER’S PERSPECTIVE(SPORT HORSE)

C. Platz

PO Box 1240, Auburn, Maine 02411, USA

Although I am an equine veterinary practitioner, Iwear a number of other hats. I am also a rider andtrainer of dressage horses through the FederationEquestré Internationale (FEI) level and am oneexamination away from being United StatesDressage Federation certified instructor. As aresult of these endeavours, I share the perspectiveof both a consumer as well as a provider of equineveterinary services.

Most non-veterinary professional horsemenand women understand 4 different categories ofrespiratory disease:

1. Mechanical problems: eg laryngeal hemiplegia(roarer).

2. Infectious problems: pneumonia, influenza.

3. Heaves: an end-stage problem usuallyassociated with old horses.

4. Allergies: a loose term applied to a horse thatcoughs a lot, sometimes for no apparentreason.

The idea that some horses have an innaterespiratory sensitivity to environmentalprovocation that can worsen over time, theprogression of which can be modified bymanagement and medication, and which is oftenexpressed by the horse in very subtle ways is notgenerally known by the majority of trainers andriders. This represents a significant gap betweencurrent veterinary science and the application ofthat science to the population of horses we serve.

ALTERNATIVE MANIFESTATIONS OFRESPIRATORY COMPROMISE

Most trainers and riders of dressage horses thinkthat a horse expresses respiratory distress bycoughing a lot. They do not become concerned

unless they hear the horse coughing in his stall orthroughout his work session. As a result, manyhoses in the early stages of inflammatory airwaydisease are not identified and, therefore, do notbenefit from prophylactic management that canhelp preserve airway function and prolong athleticusefulness.

While it is true that some horses withinflammatory airway disease do cough significantly,many do not. There are many other indications thata horse is experiencing some degree of respiratorydistress. Early identification of affected individualswould benefit both the horse and those who have aninterest in its well-being.

For example, some buyers of dressage horsesknow to avoid an individual that emphaticallywants to carry his head ‘long and low’. While thisposture may indicate a desire to stretch the topline(which is a good thing in the dressage mount), itmay also be a manifestation of the horse’s need tostraighten his airway to maximise airflow and,thus, an indicator of respiratory compromise (abad thing in a dressage mount).

Other behaviour manifested by horses withinflammatory airway disease may be similarlymisinterpreted as training issues, when in fact maybe related to respiratory impairment. Horses arequite creative in developing strategies to evade thatwhich causes them discomfort or stress. Horsesthat resist longitudinal flexion of the poll may doso in order to avoid respiratory distress. Bykeeping their throatlatch open, they can maximiseairflow by maximising airway resistance.

Such horses may be quite soft and willing toyield their poll to the rein during warm-up andeasier parts of their work session, but as the degreeof difficulty increases, may abruptly jerk theirheads and yank the reins from the rider’s hands.By bringing themselves ‘above the bit’ (meaning


20

the profile of the face is above perpendicular to theground) or carrying themselves ‘long and low’, thehorses prevent the rein from decreasing the angleof neck and jaw, so the rider can no longer affectthe configuration of their airway. After a period ofrecovery the horse may again yield to the bit, butwhen the physical demand re-escalates, the horserepeats its resistance by taking the reins away fromthe rider.

In at least one scenario of which I ampersonally aware, a non-coughing horse withsevere inflammatory airway disease was sold to abuyer who was not informed of the horse’scondition (selling the horse without details of hismedical history and Aeromask no doubt increasedthe purchase price significantly). The gelding wasplaced in training with a professional. Withoutproper management the horse was unable toperform even lower level dressage with his head inthe desired position. Since he did not cough, hewas not identified as having respiratory disease.

The horse was by nature an extremelygenerous and accommodating individual butconsistently refused to ‘go on the bit’, and wasperceived to have a bad attitude. Various straps,gadgets and gimmicks were used to try to fix theproblem to no avail and the horse eventually endedup in an Olympic rider’s barn. By this time, thehorse would deliberately run himself into a wall toavoid having to flex his poll, a behaviourintimidating to his junior rider. Hind limblameness eventually resulted from the horse’sstruggle to keep his airway as open as possible.Ultimately the horse’s mind and body were sodamaged that he became unrideable at the age of11, in addition to becoming a respiratory cripple.

Because of my experience with this and otherhorses with inflammatory airway disease, I think itis important that veterinarians, riders and trainersconsider respiratory distress as part of thedifferential diagnosis for dressage horsesexhibiting the following behaviours:

1. Exaggerated tendency to prefer long and lowcarriage of head and neck.

2. Reluctance to flex poll and yield in the jaw.

3. Inconsistency of willingness to flex in the polland yield to the bit. This may vary fromsession to session (depending onenvironmental provocation of airwayinflammatory) or within the work session(related to increased demands of collection orathletic effort).

4. Horse going behind the bit: that is, adoptingfalse flexion that prevents the rider fromasking the horse for genuine athletic effort.Usually over-flexing at the C-3/C-4 junctionrather than at the poll/throatlatch.

5. Horse behind the leg, behind the aids or notthrough – strategies whereby the horse avoidsusing himself athletically and hence notincreasing respiratory demand.

6. Horse exhibiting tension when athletic orflexion demands are increased; may bemanifested by various means – tail swishing,teeth grinding, kicking at the leg, bucking andnot going forward, stiffening in the back, neck,poll, etc.

7. Horse goes to the bit and then when asked toflex more in poll, backs off.

8. Lack of progress in training. While a drop inperformance may be the indicator in somecases, some horses which hit a glass ceiling inperformance may have reached the limit oftheir respiratory comfort zone. The demandsof collection and flexion may be too great forwhat the horse can accommodate and stillbreath comfortably, so the horse does notprogress in its work. These horses may givethe impression of a benign resistance toworking harder, eg ‘lazy’. Rather than become‘air-hungry’, horses may keep their athleticefforts within the limits of what they can dowithout running low on air.

As is shown by the list above, manifestations ofrespiratory compromise in dressage horses caneasily be misunderstood or misinterpreted. Thereis a list of alternative explanations for theseobservations, including, but not limited to:

1. teeth and bitting issues;

2. vertebral column problems;

3. lameness;

4. saddle fit;

5. lack of expertise of rider, trainer.

Note: that lameness and back/neck pathology caneasily result if the horse uses his body incorrectlyin order to avoid respiratory stress. In myexperience dressage horses will choose to evade,rather than work in a state of respiratorycompromise.

In fairness to all concerned, it must be notedthat determining the cause(s) of training and


21

soundness problems in horses is a complicatedprocess; both knowledge and creativity arerequired. Back soreness, from incorrectpositioning of the head and neck in an effort tomaximise air flow, may provide and explanationfor poor performance, which will recur unless theunderlying respiratory condition is improved.

Not all, or even most horses with trainingproblems, have airway disease, but some do andshould be identified as early in the disease processas possible.

CLINICAL MANIFESTATIONS OF REACTIVEAIRWAY DISEASE –TRAINER ’S PERSPECTIVE

Most non-veterinary equine professionals considercoughing to be the major indicator of respiratoryproblems. Most consider some coughing to benormal. When told that normal healthy horsesgenerally do not cough without obvious reason,they are quite surprised. Other ways in whichhorses with reactive airways may manifest theircondition include the following:

1. None – some horses with early documentedairway inflammation do not cough at all. Theygive no indication of respiratory distress. Theymay not show increased respiratory rate orabdominal lift. Training problems may be theonly indication that the horse is affected.

2. Excess snorting and blowing during warm-up,wanting to extend head and neck to do this.

3. Long, soft growling or moaning noises orgutteral grunting in rhythm to breathing at theonset of work, especially if not allowed toextend head and neck.

4. Horse sounds like its trying to ‘clear its throat’,culminates in part cough, part sneezing sound.

5. Few light coughs (1–12) which do not allresemble a COPD cough, but are light andsound like an irritation in the throat, ratherthan originating from the lungs.

6. All above signs are not usually randomthroughout the work, but are associated with theonset of a new level of activity, ie trot, canter,increased degree of engagement of hindquarters(which increases metabolic workload, henceincreased demand on respiratory capacity).

7. Above signs may be associated with change ofcarriage of the horse, ie alteration in positionof head and neck and, therefore, configurationof upper airway.

8. Nasal discharge – may be none, but may beunilateral or bilateral, greyish-white and eggwhite consistency, transient and episodicrather than copious, produced after trailering,demanding work session or on days whenenvironmental provocation is particularlyhigh. Appearance may be preceded by episodeof snorting, groaning and attempting to ‘clearthe throat’, sometimes coughing.

9. Sharp coughs after sudden exercise at liberty.Horses with affected airways are often quitetranquil and do not exert themselves much inturnout. They may gallop vigorously whenprovoked but may have a brief harsh coughingspell after such exertion. Does not sound likelight cough under saddle (note: most normalhorses WILL cough briefly after rolling).

10. Change in demeanour or behaviour in anunfamiliar allergenic environment – affectedhorses stabled at an indoor show or transportedto a different part of the country may not showsignificant respiratory signs, but may seemsubdued or even depressed. If in a competitionsituation, they may not perform with normalenergy, sometimes misinterpreted as shippingfatigue or homesickness.

11. In my experience, horses that coughexcessively when eating dusty hay do notnecessarily have inflammatory airway disease.I have an unaffected horse who coughs easilyfrom overt dust in his environment or feed,while his affected stable-mates do not respondto this challenge at all. It is almost as thoughthe affected horses have an inappropriateresponse to appropriate stimuli!

It should be noted that subtle changes indemeanour, coughing or other mild respiratorysigns may be indicators of early inflammatoryairway disease. It is important to be aware of suchindicators, because by the time more prominentsigns are evident, the opportunity to intervene withthe progression of airway disease may be lost.

MANAGEMENT STRATEGIES

The time-honoured approach of increasingventilation and reducing dust and mould in theenvironment is the basis for managing horses withinflammatory airway disease. Not all horses willrespond to increased access to turnout if theprovocation for their symptoms comes from


22

outdoors – eg tree pollen. However, improvingenvironmental air quality will decrease therespiratory burden on even pollen-sensitiveindividuals and should be a major goal of non-medical management.

Bedding

In addition to no bedding at all, peat, shreddedpaper, etc, new pelleted products such as ‘WoodyPet’ have been helpful for many horses. Lack ofsuccess with this product can usually be attributedto the fact that users do not understand how to useit, often because they do not read the directions onthe bag. Keeping the proper amount of moisture inthe bedding is essential. Woody Pet is excellent fortrailering horses when normally shavings would beused. Straw is often too dusty or mouldy for idealbedding.

Forage

Traditionally, Denji has been considered the goldstandard for feeding forage to horses withinflammatory airway disease, but it has itsdrawbacks. It is expensive and contains molassesand alfalfa which may be problematic for somehorses. Sometimes Denji is mishandled, storedimproperly or has become contaminated and sobecomes mouldy in some parts of the bale. Themanufacturer has developed a new product called‘Totally Timothy’ which avoids the alfalfa problem.

Hay cubes do not work well for aged horseswith bad teeth, usually contain alfalfa, and arereported to case choke unless well-soaked.

I find soaked hay is the best forage for affectedhorses. Simply telling an owner to ‘wet’ the hay isoften insufficient to remove dust and spores.Inadequately moistened hay may grow mouldrather than protect the horse from allergens. It alsohas been know to cause gas colic.

Hay should be submerged for a prolongedperiod of time before feeding or, alternatively,should be fed submerged (such as in a tub of waterwith a weight on the hay to keep it below thewater). Horses readily accept soaked hay and alsoadapt well to pulling the hay from underneathshallow water. I have fed hay that has been soakedup to 24 h in 90° weather in New England without

any problems. Water used to soak hay should bereplaced daily.

In cold climates hay can be stuffed by thesection into heated water buckets, the bucket filledwith water and then placed in the horse’s stall. Thishas the added benefit of increasing waterconsumption in cold weather. Other strategies areto fill plastic wheeled bins with a bale of hay, coverwith water and place a Styrofoam block on the topas insulation. The hay is then fed out in sections.This has been used in Massachusetts for a numberof years.

Other dietary considerations

Horses with airway inflammation sometimes havemulti-systemic sensitivities. Simplifying the dietby eliminating commercially formulated productsin favour of simple grains with judicious use ofsupplements has resulted in apparent improvementin behavioural, digestive and skin problems insome horses. Alfalfa and molasses are especiallytroublesome for certain individuals.

SUMMARY

A gap exists between the scientific and laycommunity in the identification, understandingand management of reactive airway disease inhorses. Most dressage trainers and owners are notenthusiastic about long-term medical treatment ofrespiratory conditions in their horses, but are quitereceptive to management changes which will allowthem to prolong their horses’ working lives.

Horses affected with inflammatory airwaydisease should be identified as early as possible intheir disease process to optimise successfulintervention. As veterinarians and trainers becomemore aware of some of the subtle indications thata horse is experiencing respiratory impairment,affected individuals can be recognised andcorrective measures taken to help protect theirairways from further damage.

More work needs to be done to aid the non-veterinary equine professional in recognising earlyindications of inflammatory airway disease inhorses, and in the development of devices for moreconvenient methods of soaking feed hay, as well asmore satisfactory non-allergenic forage alternatives.


23

THE DIAGNOSTIC APPROACH TO CHRONIC COUGHIN PEOPLE

A. J. Ghio

Human Studies Division, United States Environmental Protection Agency, Chapel Hill, North Carolina,USA

Cough is cited frequently as the fifth mostcommon symptom of patients presenting formedical care and accounts for 30 million officevisits per year. Causes of cough and, therefore, thediagnostic approach, will vary with practicesetting and the age of the patient population.Private practitioners will evaluate a differentspectrum of disease compared to clinicians at atertiary care centre. Similarly, paediatricians willdiagnose asthma as the basis for cough with agreater frequency relative to physicians providingcare to an adult population.

Cough is defined as an explosive expirationproviding a protective mechanism for clearingforeign materials and tracheobronchial secretions.The afferent limb for the signal will involve cranialnerves V, IX, X and cervical nerves 2 and 3originating from both rapidly adapting pulmonarystretch receptors and unmyelinated C fibres.Slowly adapting pulmonary stretch receptors mayenhance the cough. The ‘cough centre’ is thoughtto be localised in the medulla. The efferentpathway includes the X and cervical nerves 3 and5 to laryngeal and tracheal muscles, thediaphragm, and respiratory muscles.

There are several different criteria used todefine an acute and a chronic cough. An acutecough is frequently defined as being less than 3weeks in duration. Cough can be that symptompersisting longer than 3, 6 or 8 weeks. Irwin andMadison (2000) included a subchroniccategorisation of cough as that between 3 and 8weeks. An acute cough is most likely to be theresult of infection including upper respiratoryinfection (eg pharyngitis), acute bacterial sinusitis,and pertussis. An acute cough can also reflect anexacerbation of chronic obstructive pulmonarydisease, allergic rhinitis, asthma, congestive heartfailure, pneumonia, aspiration, and environmental

exposures (tobacco smoke being most prevalent).A cough persisting for 3–8 weeks most commonlyis post infectious, but can also support a diagnosisof sinusitis and asthma.

The differential of cough persisting for weeks(ie the chronic cough) includes a much largernumber of diseases. However, 3 diagnoses canaccount for 90% of all such patients: post nasaldrip syndrome, asthma, and gastro-oesophagealreflux. In one study of non-smokers not on an ACEinhibitor having a normal chest X-ray, this numberwas 99.4%. Other diagnoses to be aware of arecough following a viral infection, a pertussiveinfection, and cough with ACE inhibitor therapy.Chronic cough can also be a presenting symptomof chronic bronchitis, interstitial lung disease,abscess, bronchiectasis, neoplasm and foreignbody. Rare causes are psychogenic, impactedcerumen/hair, and foreign body. Approximately25% of all patients presenting with cough willhave more than one cause accounting for thissymptom.

The most common cause of chronic cough inadults is the post nasal drip syndrome observed inallergic rhinitis, perennial non-allergic rhinitis,vasomotor rhinitis, nasopharyngitis and sinusitis.Secretions in the upper airway stimulate cough viastimulation of receptors within the pharyngeal/laryngeal mucosa. Common symptoms seen withthis cough are nasal discharge, the sensation ofliquid in the back of the throat and clearing of thethroat. However, it may also be ‘silent’, which is acough without any accompanying symptoms.

The second most common cause of chroniccough in adults is asthma. This can be noted withepisodic wheezing and dyspnea but might also be‘silent’. Pulmonary function testing might reflectreversible airflow obstruction, but can be normal.If the latter occurs, airway hyper-reactivity can be


24

demonstrated using methacholine challenge.However, such a positive bronchoprovocation testdoes not prove asthma and can be falsely positivein 22% of these patients (bronchitis andbronchiectasis). To confirm the diagnosis,appropriate therapy must be employed withimprovement resulting. Gastro-oesophageal refluxis cited frequently as the third most common causeof chronic cough, but was the most common in arecent study by Poe and Kallay (2003).Accompanying symptoms are heartburn and a sourtaste in the mouth. Comparable to post nasal dripand asthma, reflux can be ‘silent’. Cough resultsfrom a stimulation of receptors in the larynx,aspiration of gastric contents, and reflux of acidinto the distal oesophagus. The most sensitivemethod for diagnosis is pH monitoring.

Cough following viral infections can persist formore than 8 weeks. Secretions stimulate receptorsaccounting for the symptom. Anthihistamine/decongestant may assist in some relief of the coughbut, more frequently, it is related to airwayinflammation and hyper-reactivity. A chroniccough with pertussis may be more common thanthought in adults. Approximately one of 5 patientsdemonstrated positive serology for pertussis butsuch testing is not available to clinicians currently.Cough with ACE inhibitors ranges from 3–20% ofall patients using the medication. It follows anaccumulation of bradykinin. Onset can range fromone week to 6 months while resolution requires 4days to 4 weeks. Employment of an alternativeACE inhibitor does not help. Therapy is switchingto an angiotensin II receptor antagonist.

The history of the patient is the first and themost important tool employed in the diagnosticapproach to chronic cough. The existence of othersymptoms and use of ACE inhibitors should bedetermined. Questions are directed at thecharacterising the cough: the onset, duration,frequency and alleviating and exacerbating factors.Smoking and occupational histories can be helpful

in diagnosis. The character and timing of thecough is queried, but usually is not helpful.Similarly, sputum production is of little assistancein differentiating the cause of chronic cough. Apast medical history of heart, lung, thyroid, andgastrointestinal disease can contribute toward adiagnosis. Cobblestoning, secretions, stridor,rhonchi, wheezing, crackles and observing thequality of cough during the examination can alsoassist in diagnosis. Spirometry, both with andwithout bronchodilators, should also be obtained.If support for a specific diagnosis is found, therapyshould be provided. If no support for a specificdiagnosis is found or there is no improvement withinitial therapy, an antihistamine, a decongestant,and a nasal steroid are provided. The patientshould be re-evaluated in 2 weeks. If there isimprovement, the post nasal drip syndrome isdiagnosed. If there is none, sinus and chest X-raysshould be considered. Chest X-rays mightcontribute to a diagnosis in cough with cancer,tuberculosis, sarcoidosis, and interstitial lungdisease. In addition, methacholine challenge isdone. If the last is positive, the patient should betreated for asthma with a combination of inhaledsteroids and bronchodilators. The patient is againre-evaluated in 2 weeks. If there is noimprovement on inhalers, the patient is placed onpH monitoring. If abnormal, therapy is directedtoward reflux. Finally, if there is no improvementon this regimen of medication, bronchoscopy isconsidered. This procedure is infrequently of anyvalue except in diagnosing the rare cancer.

REFERENCES

Irwin, R.S. and Madison, J.M. (2000) The diagnosis andtreatment of cough. N. Engl. J. Med. 343, 1715-1721.

Poe, R.H. and Kallay, M.C. (2003) Chronic cough andgastroesophageal reflux disease: experience withspecific therapy for diagnosis and treatment. Chest123, 679-684.


25

SESSION 2:

Aetiology

Chairman: J. P. Lavoie


26

27


AETIOLOGICAL AGENTS: INDOOR ENVIRONMENTAND ENDOTOXIN

B. C. McGorum and R. S. Pirie

Department of Veterinary Clinical Studies, Easter Bush Veterinary Centre, University of Edinburgh,Roslin, Midlothian EH25 9RG, UK

Airborne dust in horse stables may contain highlevels of bacterial endotoxins, over 50 species ofmoulds, large numbers of forage mites, plantdebris and inorganic dusts. High levels of toxicgases such as ammonia are also present in somestables. Respirable and total airborneconcentrations of organic dusts, endotoxin(McGorum et al. 1998; Malikides et al. 2000; Pirieet al. 2001) and β-D-glucan (an index of mouldexposure; Malikides, unpublished data) in thebreathing zones of horses in conventional horsestables may exceed the thresholds for induction ofinflammation and hyper-responsiveness in healthyhumans, and may exceed the air hygiene standardsrecommended for occupational exposure in man.

Inhaled organic dust and, in particular, inhaledendotoxin and moulds, have a well accepted role inthe aetiology of heaves. Inhaled endotoxin is alsoan important cause of inflammatory airway diseasein humans and, indeed, the airborne endotoxinconcentration is the most important occupationalexposure associated with the development andprogression of airway disease in humanagricultural workers (reviewed by Schwartz 2001).The pro-inflammatory properties of these airborneagents indicate that they also have the potential tocontribute to the induction and/or persistence ofinflammatory airway disease (IAD) in stabledhorses. Consistent with this possibility, findingsfrom a recent study into the role of airborne dust,endotoxin and β-D-glucan in IAD in youngracehorses in Sydney, Australia indicate thatinhalation of high concentrations of respirableendotoxin, but not β-D-glucan, is a strong riskfactor for tracheal aspirate neutrophilia.Additionally, Burell et al. (1996) demonstratedthat the duration of IAD, and the frequency ofcoughing, were significantly greater in horses thatwere housed in loose boxes and bedded on straw

(presumed high dust exposure) compared withthose housed in American barns and bedded onpaper (presumed low dust exposure).

Several other studies have also reported arelationship between stabling and inflammation ofthe upper and lower respiratory tracts in non-heaves-affected horses (Derksen et al. 1985;Clarke et al. 1987; Tremblay et al. 1993;Holcombe et al. 2001). The relationship betweenthis stable-induced airway inflammatory responseand IAD remains to be established. It could beargued that some of the horses in some of thesestudies may have had low-grade heaves (ie dustinduced neutrophilic airway inflammation withoutsignificant airway dysfunction). However, thispossibility could be eliminated in the study thatincluded only horses considered to be too young todevelop heaves (Holcombe et al. 2001).Interestingly, the ratios of neutrophils reported inairway samples from horses with stable-inducedairway inflammation often exceeded 10%, andoccasionally exceeded 30%, ratios that arecommonly considered to support a diagnosis ofIAD or heaves. While stabling was associated withairway neutrophilia, concomitant airwaydysfunction or airway mucus hyper-secretion wasnot detected in these studies. The latter findingcontrasts with IAD and heaves, which are bothdisorders characterised by significant airwaymucus hyper-secretion.

Additional evidence supporting involvementof organic dust in IAD stems from short durationexperimental inhalation challenges with hay dustsuspension or soluble endotoxin (Pirie et al. 2001,2002). These challenges induced a dose-dependentairway neutrophilia, in both heaves and controlhorses. The threshold exposures of dust orendotoxin required to induce airway neutrophiliain controls was higher than that for heaves-affected


28

horses, with control horses developing significantairway inflammation only following very highdust or endotoxin exposures. While theseexperimental challenges induced airwayneutrophilia in control horses, mucus hyper-secretion, airway dysfunction and airway hyper-reactivity were not detected. It is possible thatthere may be significant variability amongindividual non-heaves-affected horses in theirresponsiveness to inhaled endotoxin, which wasnot identified in the aforementioned study, giventhe limited number of horses investigated.Consistent with this possibility, healthy, non-asthmatic humans have a broad range of stable andreproducible physiological responses to inhaledendotoxin, with only approximately 10–15% ofthe subjects developing either airflow obstructionafter inhaling minimal amounts of endotoxin, orhaving a negligible airway response to high dosesof inhaled endotoxin (Kline et al. 1999). Potentialvariability in responsiveness to inhaled agentscomplicates the investigation of these agents inhorses, where subject numbers are limited.Investigation of the role of organic dust in IAD iscomplicated further by the additive or synergisticeffect of co-exposure with other pro-inflammatoryagents (Williams 1997). The pro-inflammatoryeffect of inhaled agents is also augmented by thepresence of pre-existing airway inflammation,such as that induced by microbial agents.Additionally, as IAD is commonly observed inexercising horses, the pulmonary oxidative stressassociated with exercise may augment thepulmonary response to these inhaled agents.

Horses with heaves show a consistentresolution of pulmonary neutrophilia andsignificant improvements in lung functionfollowing a reduction in organic dust exposure. Incontrast, the authors are aware of some horses withIAD that had persistent pulmonary inflammationand associated clinical signs, despite beingmaintained fully at pasture for several months.This anecdotal observation suggests that factorsother than inhaled organic dust are, at least in part,involved in perpetuating the pulmonaryinflammatory response in horses with IAD.Nevertheless, the bulk of current evidencesuggests that horses with IAD should bemaintained in low dust environments.

REFERENCES

Burrell, M.H., Wood, J.L., Whitwell, K.E., Chanter, N.,MacKintosh, M.E. and Mumford, J.A. (1996)Respiratory disease in thoroughbred horses intraining: the relationships between disease, viruses,bacteria and environment. Vet. Rec. 139, 308-313.

Clarke, A.F., Madelin, T.M. and Allpress, R. (1987) Therelationship of air hygiene in stables to lowerairway disease and pharyngeal lymphoidhyperplasia in two groups of Thoroughbred horses.Equine vet. J. 19, 524-530.

Derksen, F.J., Scott, J.S., Miller, D.C., Slocombe, R.F.and Robinson, N.E. (1985) Bronchoalveolar lavagein ponies with recurrent airway obstruction(heaves). Am. Rev. Resp. Dis. 132, 1066-1070.

Holcombe, S.J., Jackson, C., Gerber, V., Jefcoat, A.,Berbey, C., Eberhardt, S. and Robinson, N.E.(2001) Stabling is associated with airwayinflammation in young Arabian horses. Equine vet.J. 33, 244-249.

Kline, J., Cowden, J.D., Hunninghake, G., Schutte, B.,Watt, J., Wohlford-Lenane, C., Powers, L., Jones,M. and Schwartz, D. (1999) Variable airwayresponsiveness to inhaled lipopolysaccharide. Am.J. Resp. Crit. Care Med. 160, 297-303.

Malikides, N., Pike, A., Kane, K., Duhs, L. andHodgson, J.L. (2000) Endotoxin concentrations inrespirable dust over time in horses bedded on strawversus sawdust. Proc. 18th symp. Comp. Resp. Soc.Nov 6-9th, University of Melbourne, Australia, p 51.

McGorum, B.C., Ellison, J. and Cullen, R.T. (1998)Total and respirable airborne dust endotoxinconcentrations in three equine managementsystems. Equine vet. J. 30, 430-434.

Pirie, R.S., Collie, D.D. and McGorum, B.C. (2002)Evaluation of nebulised hay dust suspensions(HDS) for the diagnosis and investigation of heaves;2: Effects of inhaled HDS in control and heaveshorses. Equine vet. J. 34, 337-342.

Pirie, R.S., Dixon, P.M., Collie, D.D. and McGorum,B.C. (2001) Pulmonary and systemic effects ofinhaled endotoxin in horses with heaves. Equinevet. J. 33, 311-318.

Schwartz, D.A (2001) Does inhalation of endotoxincause asthma? Am. J. Resp. Crit. Care Med. 163,305-306.

Tremblay, G.M., Ferland, C., Lapointe, J.M., Vrins, A.,Lavoie, J.P. and Cormier, Y. (1993) Effect ofstabling on bronchoalveolar cells obtained fromnormal and COPD horses. Equine vet. J. 25, 194-197.

Williams, D.L. (1997) Overview of (1-3)-ß-D-glucanimmunobiology. Med. Inflamm. 6, 247-250.

29


AETIOLOGICAL AGENTS: OUTDOOR ENVIRONMENTAND AIRWAYS

A. J. Ghio

United States Environmental Protection Agency, Chapel Hill, North Carolina, USA

Environmental agents that can potentially injure theairways of animals include ozone, nitrogen, sulphurdioxide and particles, the latter being of greatestconcern. Most of the US population is exposed toozone concentrations that exceed current standards.This is responsible for induction of inflammation inthe airways, acute loss of pulmonary function,elevated airway reactivity, increased infection ratein the lung, decreased particle clearance and alteredlung structure. Environmental sources of ozoneinclude motor vehicles, power-generating plantsand industrial combustion processes. Nitrogenoxides can cause shortness of breath, coughing andhyper-reactivity. Sulphur dioxide is produced bypower plants and industrial processes.The highestlevels, occurring in summer, can cause cough,phlegm and hyper-reactivity.

Air particles are a major environmentalchallenge to human airways. Inhalation ofsuspended particulate matter (PM) has challengedthe lower respiratory tract in man for thousands ofyears. In the past century, episodes of extremelyhigh levels of ambient air pollution particles inEurope and the USA have corresponded with acuteelevations in human morbidity and mortality. Thiswas instrumental in bringing about widespreadmonitoring and regulation of air quality from about1970. However, approximately 10 years ago,several studies using time-series analysis linkedexposure to ambient air pollution particles at levelscurrently observed in cities worldwide with indicesof acute human morbidity and mortality (Samet etal. 2000). These findings were met with somescepticism but both re-evaluation of the initialstudies and many new investigations confirmedtheir validity (Gamble and Lewis 1996; Krewski etal. 2000). The initial assumption was that thismorbidity and mortality resulted from a lung injury,which was likely to be inconsequential in healthy

individuals but significant in susceptible subjects. Descriptions of haemorheological changes and

acute elevations of morbidity and mortalityattributable to cardiovascular disease have suggestedthat the biological effects of PM are not restricted tothe lung. Also, although morbidity and mortalityassociated with ambient air PM levels is morecommon in the very young or old, and those withchronic cardiopulmonary disease, populations of allages and health status seem at risk. In composition,ambient air pollution PM can range from crustalsilicates and plant debris to carbonaceous productsof combustion. Ambient air PM levels are relativelylow usually (10–100 g/m3), although this canapproach 1,000 g/m3 in particularly urban sites,representing constant exposure to what can bepredominantly fine and ultra-fine particles.

Increased ambient air PM levels have beenassociated with a higher prevalence of respiratorysymptoms, as well as hospitalisations for bronchitis.Increased incidence of wheeze and asthma havebeen correlated with elevated ambient air PM ashave increased infections. An acute, reversibledecrement in pulmonary function has been describedfollowing exposure to ambient air PM, although notin all studies (Goren et al. 1999). A chronic loss ofpulmonary function has not been reported afterexposure to elevated ambient PM. The presence ofmany particles can affect measurements of bronchialhyper-reactivity, and a challenge inhalation withparticles has been proposed as an alternative tomethacholine in the diagnosis of asthma (Cloutier etal. 1992). Investigations have not yet defined theeffect of ambient air PM on bronchial hyper-reactivity. Inhalation of ambient air particles hasbeen associated with an acute influx of neutrophilsto the lower respiratory tract (Ghio et al. 2000). Thisis comparable to other particle-associated injuries,including cigarette smoking and diesel exposure.


30

The 2 chronic pathological processes noted onmicroscopic inspection of lungs from individualschronically exposed to particles are emphysemaand parenchymal fibrosis. These have distinctclinical and pathological characteristics and havebeen considered to be separate disorders but theyhave features in common and may be linked by acommon mechanistic pathways such that an agentwhich normally produces a fibrotic injury can bemodified to be associated with emphysema.

Collagen deposition and fibrosis in the humanlung correlate with exposure to ambient PM(Pinkerton et al. 2000) which can elicit severalchanges in peripheral blood, including decreasedred cells, elevated white blood cell counts andincreased C-reactive protein, fibrinogen and bloodviscosity. The last 2 changes contribute potentiallyto the association of ambient PM with thromboticevents. Currently, plasma fibrinogen levels appearthe most sensitive systemic marker of exposure toambient PM. Elevated plasma concentrations ofthis protein occur after exposure to many differentparticles, but the response is best described insmokers. There appears to be some dose-responserelationship between particle exposure and theplasma fibrinogen level. High levels of plasmafibrinogen accurately reflect the severity ofatherosclerosis and a hypercoagulable state.Elevations of this protein correspond to incidenceof cardiovascular disease, vascular occlusivedisease and failure of saphenous vein andprosthetic grafts. Therefore, increased levels offibrinogen after particle exposures may predict anincrease in cardiovascular disease. Inhalation ofambient air PM increases heart rate, decreases heartrate variability, and elevates arrhythmia rates.Hospital admissions for cardiovascular diseases arealso associated with PM, and the incidence ofmyocardial infarction can be increased within 2 hof PM elevation and remaine elevated for 24 h.

Investigations have suggested carcinogenicity ofambient air PM but results can be conflicting as isalso noted in investigations of cancer induction bymineral oxide particles (eg silica). It has beensuggested that cancer may be triggered by changesin local pH as a result of acidic properties of theparticle, contamination by biological agentsincluding endotoxin and worsening of atopic diseaseby organic components. In addition, a fundamentalproperty of many of these particles is the capacity topresent an oxidative stress to the lung. There arenumerous proposed pathways in which particles candirectly support electron exchange with radical

formation. Diminished concentrations ofcompounds with an ability to scavenge free radicalshave also been reported and can affect oxidativestress after particle exposure (Al-Humadi et al.2002). Finally, numerous particles effect adisequilibrium in iron metabolism in a tissuecomparable to the effect of fibres (eg asbestos). Acommon feature found on light microscopy of thelung exposed to any particle is a ferruginous body.This histologic abnormality is a particle (or fibre)with accumulations of protein and iron and providesevidence of a local disruption of normal metalmetabolism. The ferruginous body can reflectincreased catalytically active iron in a tissue andtherefore suggests oxidative stress. Features of theclinical presentation, physiological changes andpathology of individuals exposed to ambient air PMare common to many of the particle-related injuries.

REFERENCES

Al-Humadi, N.H., Siegel, P.D., Lewis, D.M., Barger,M.W., Ma, J.Y., Weissman, D.N. and Ma, J.K.(2002) Alteration of intracellular cysteine andglutathione levels in alveolar macrophages andlymphocytes by diesel exhaust particle exposure.Environ. Health Perspect. 110, 349-353.

Cloutier, Y., Lagier, F., Cartier, A. and Malo, J.L. (1992)Validation of an exposure system to particles for thediagnosis of occupational asthma. Chest. 102, 402-407.

Gamble, J.F. and Lewis, R.J. (1996) Health andrespirable particulate (PM10) air pollution: a casualor statistical association? Environ. Health Perspect.104, 838-850.

Ghio, A.J., Kim, C. and Devlin, R.B. (2000)Concentrated ambient particles induce a neutrophiliclung inflammation in healthy volunteers. Am. J.Respir. Crit. Care Med. 162, 981-998.

Goren, A., Hellmann, S., Gabbay, Y. and Brenner, S.(1999) Respiratory problems associated withexposure to airborne particles in the community.Arch. Environ. Health 54, 165-171.

Krewski, D., Burnett, R.T., Goldberg, M.S., Hoover, K.,Siemiatycki, J., Jerrett, M., Abrahamowicz, M. andWhite, W.H. (2000) Reanalysis of the Harvard SixCities study and the American Cancer Society studyof particulate air pollution and mortality. A specialreport of the Institute’s Particle EpidemiologyReanalysis Project. Cambridge, MA: Health EffectsInstitute.

Pinkerton, K.E., Green, F.H.Y., Saiki, C., Vallyathan, V.,Plopper, C.G., Gopal, V., Hung, D., Bahne, E.B.,Lin, S.S., Menache, M.G. and Schenker, M.B.(2000) Distribution of particulate matter and tissueremodeling in the human lung. Environ. HealthPerspect. 108, 1063-1069.

Samet, J.M., Dominici, F., Curriero, F.C., Coursac, I. andZeger, S.L. (2000) Fine particulate air pollution andmortality in 20 US cities, 1987-1994. N. Engl. J.Med. 343, 1742-1749.

31


AETIOLOGICAL AGENTS: ALLERGY AND RELATED MEDIATORS

J. P. Lavoie

Faculty of Veterinary Medicine, University of Montréal, Quebec, Canada

Allergy is defined as a hypersensitive stateacquired through exposure to a particular allergen.Allergic asthma (extrinsic) in humans ischaracterised by an early allergic response (EAR)that occurs within minutes of exposure to anallergen and the late allergic response (LAR),which develops 6–9 h later. EAR is initiated by theactivation of cells bearing allergen-specific IgE,primarily mast cells, through IgE cross linkage andactivation of the high affinity (FcεRI) receptors.Mast cell activation results in the release ofpreformed mediators, together with the synthesisand release of new mediators and cytokines. Therelease of preformed mediators is responsible forthe initial contraction of airway smooth muscle,mucus secretions, vasodilation and microvascularleakage. Together, these changes result in thenarrowing of the airway lumen and airflowobstruction characteristic of the EAR.

There is currently no evidence that horses withinflammatory airway disease (IAD) develop theacute respiratory signs, which would be expectedwith an EAR. However, the finding that a subset ofhorses with small airway disease have an increasednumber of mast cells in their bronchoalveolarlavage (BAL; Vrins et al. 1991), which isnegatively correlated with airway hyper-reactivity(Hoffman et al. 1998b) and responds to mast cellstabilisers (Hare et al. 1994), supports theimplication of these cells in IAD. Additionalsupport for allergy being implicated in IAD, is thefinding that blood basophils of horses with mildairway diseases exposed to various allergens hadan increased basophil degranulation index andhistamine release (Dirscherl et al. 1993).

The synthesis and secretion of mast-cellmediators, such as eicosanoid (leukotrienes,prostaglandins, PAF) and cytokines (IL-8, IL-5,RANTES, TNF-α), promotes the chemotaxis and

activation of inflammatory cells, includingeosinophils and neutrophils, into lung tissue,which amplify and sustain the airway obstructiontaking place during the LAR. Tissue and BALneutrophilia is commonly found in horses withIAD. BAL eosinophilia, however, is found only ina subset of horses with IAD (Moore et al. 1995;Hare and Viel 1998). Nevertheless, consideringthat tissue and BAL eosinophilia are not correlatedin horses with inflammatory lung diseases (Winderet al. 1991), it is possible that infiltration ofeosinophils in lung tissues is more common thansuspected, based on BAL cytology results.

Cysteinyl leukotrienes (LTC4, LTD4 and LTE4)produced by mast cells and eosinophils can inducebronchial smooth muscle contraction, increasedvascular permeability and increased mucusproduction. On a molar basis, LTD4 in horsesbreathing spontaneously is 305–970 times morepotent than methacholine in inducing smoothmuscle contraction (Marr et al. 1998a). In apreliminary report, horses with IAD had increasedlevels of LTC4 in BAL (Hoffman et al. 1998a).Cysteinyl leukotrienes do not appear to play animportant role in heaves, as suggested by theinefficacy of a FLAP antagonist (Robinson et al.1998), 5-lipoxygenase inhibitor (Marr et al. 1998b)and LTD4 receptor antagonists (Stark et al. 2001;Lavoie et al. 2002). It remains to be determinedwhether these agents will also be ineffective in IAD.

Previously, it was believed that the inflammatoryresponse, triggered by the mast cell release ofinflammatory mediators, was a pre-requisite for thedevelopment of the LAR. However recent findings,using molecular tools and murine animal models,have highlighted the complex interactions involvedin the modulation of allergic lung disease (Watanabeet al. 1997). Most cells present within the lung tissueincluding epithelial cells, myocytes, and nerve


32

endings, in addition to traditional inflammatorycells, contribute to this modulation. Among these, Tcells (CD4+ cells, more specifically of Th2 cells)play a central role in allergic airway inflammation.Cytokines produced by Th2 cells, such as interleukin(IL)-4, IL-5 and IL-13 have been implicated inallergic inflammation. Th1 cells are important forcell-mediated immunity by their production of INF-γ and other cytokines. Using in situ hybridisation, ithas been found that BAL cells from horses withheaves fed hay over a long period of time (months)have increased expression of IL-4 and IL-5 mRNAand decreased expression of INF-γ compared tocontrols (Lavoie et al. 2001). These findings areconsistent with a predominant Th2 type cytokineresponse in heaves. In a follow-up study, it wasfound that the development of the Th2-type responsecorresponded in time with the development ofairway obstruction and inflammation (Cordeau et al.2002). IL-4 promotes the development and thegrowth of Th2 cell phenotype and is essential for theinduction of B-cell isotypes switching to IgEantibody production. IL-5 production is typicallyassociated with tissue eosinophil migration, which isnot a common feature in heaves but has been foundin some horses with IAD. Equine neutrophilsexpress receptors for Th2-type cytokines, includingIL-5 (Al-Dewachi et al. 2002) that provide amechanisms by which IL-5 may contribute toneutrophil activation. However, using RT-PCR onBAL cells of horses with heaves (Giguere et al.2002) or summer pasture associated COPD (Beadleet al. 2002), a significant increase in IL-4 was alsofound, but IL-5 mRNA was not detected in thesestudies. The contribution of Th2-type cytokine inIAD is currently being investigated.

REFERENCES

Al-Dewachi, O., Joubert, P. and Taha, R. (2002)Expression of IL-5 and IL-9 receptors on neutrophilsin horses with heaves. Am. J. Resp. Crit. Care Med.165, A310.

Beadle, R.E., Horohov, D.W. and Gaunt, S.D. (2002)Interleukin-4 and interferon-gamma gene expressionin summer pasture-associated obstructive pulmonarydisease affected horses. Equine vet. J. 34, 389-394.

Cordeau, M.E., Joubert, P. and Hamid, Q. (2002)Temporal mRNA expression of Th2-type cytokine inan animal model of chronic asthma. Am. J. Resp. Crit.Care Med. 165, A.

Dirscherl, P., Grabner, A. and Buschmann, H. (1993)Responsiveness of basophil granulocytes of horsessuffering from chronic obstructive pulmonary diseaseto various allergens. Vet. Immunol. Immunopathol. 38,217-227.

Giguere, S., Viel, L., and Lee, E. (2002) Cytokineinduction in pulmonary airways of horses with heavesand effect of therapy with inhaled fluticasonepropionate. Vet. Immun. Immunopath. 85, 147-158.

Hare, J.E., Viel, L. and O’Byrne, P.M. (1994) Effect ofsodium cromoglycate on light racehorses withelevated metachromatic cell numbers onbronchoalveolar lavage and reduced exercisetolerance. J. vet. Pharmac. Ther. 17, 237-244.

Hare, J.E. and Viel, L. (1998) Pulmonary eosinophiliaassociated with increased airway responsiveness inyoung racing horses. J. vet. int. Med. 12, 163-170.

Hoffman, A.M., Lilly, C.M. and Umkauf, L. (1998a)Elevated lvels of leukotriene C4 in BAL from horseswith cough and exercise intolerance. World equineAirw. Symp. CD-Rom Proceedings.

Hoffman, A.M., Mazan, M.R. and Ellenberg, S. (1998b)Association between bronchoalveolar lavagecytologic features and airway reactivity in horses witha history of exercise intolerance. Am. J. vet. Res. 59,176-181.

Lavoie, J.P., Leguillette, R. and Pasloske, K. (2002)Comparison of dexamethasone and the LTD4Receptor Antagonist L-708,738 on lung function andairway cytology in heaves. Am. J. vet. Res. 63, 579-585.

Lavoie, J.P., Maghni, K. and Desnoyers, M. (2001)Neutrophilic airway inflammation in horses withheaves is characterised by a Th2-type cytokineprofile. Am. J. Resp. Crit. Care Med. 164, 1410-1403.

Marr, K.A., Lees, P. and Page, C.P. (1998a) Inhaledleukotrienes cause bronchoconstriction andneutrophil accumulation in horses. Res. vet. Sci. 64,219-224.

Marr, K.A., Lees, P. and Page, C.P. (1998b) Effect of the5-lipoxygenase inhibitor, fenleuton, on antigen-induced neutrophil accumulation and lung functionchanges in horses with chronic obstructive pulmonarydisease. J. vet. Pharmac. Ther. 21, 241-246.

Moore, B.R., Krakowka, S. and Robertson, J.T. (1995)Cytologic evaluation of bronchoalveolar lavage fluidobtained from standardbred racehorses withinflammatory airway disease. Am. J. vet. Res. 56, 562-567.

Robinson, N.E., Boehler, D. and Berney, C. (1998) Failureof a FLAP antagonist to prevent airway obstruction inheaves-suceptible horses. World equine Airw. Symp.31, CD-Rom Proceedings.

Stark, G., Schmid, R. and Riedelberg, K. (2001) Clinicalefficacy of a leukotrien receptor antagonist((Montelukast) in equine obstructive pulmonarydisease (COPD). A pilot study. ACVIM, CD-RomProceedings.

Vrins, A., Doucet, M. and Nunez-Ochoa, L.A. (1991)Retrospective study of bronchoalveolar lavagecytology in horses with clinical findings of smallairway disease. Zentbl. VetMed. A. 38, 472-479.

Watanabe, A., Mishima, H. and Kotsimbos, T.C. (1997)Adoptively transferred late allergic airway responsesare associated with TH2-type cytokines in the rat. Am.J. Respir. Cell Mol. Biol. 16, 67-74.

Winder, N.C., Grunig, G. and Hermann, M. (1991)Comparison of bronchoalveolar lavage andrespiratory secretion cytology in horses withhistologically diagnosed pulmonary disease.Schweizer Archiv fur Tierheilkunde 133, 123-130.

33


AETIOLOGICAL AGENTS: VIRUSES ANDINFLAMMATORY AIRWAY DISEASE

J. L. N. Wood, J. R. Newton, K. C. Smith and D. J. Marlin

Animal Health Trust, Lanwades Park, Kentford, Newmarket, Suffolk CB8 7UU, UK

WHAT IS THE DISEASE DEFINITION USEDIN THIS PAPER ?

Inflammatory airway disease (IAD) is a poorlyunderstood phenomenon and data from only a fewpopulation based studies are published. Directcomparison between studies is hampered by a lackof clear case definition and the absence ofpublished studies comparing the usefulness oftracheal and bronchoalveolar lavage (BAL) in itsdiagnosis. IAD is defined here as a diseasesyndrome with neutrophilic inflammationdetectable by tracheal or BAL in animals under 5years old. Often, neutrophilia is accompanied byincreased amounts of mucus visibleendoscopically in the trachea shortly afterexercise, although this is not regarded as essentialto case definition. Although not specificallyexcluded in every report, classic heaves, orrecurrent airway obstruction (RAO) is not includedin this discussion.

STUDIES OF IAD AFFECTED HORSES

Workers at the Animal Health Trust have studiedthis disease syndrome in racehorses, in young ponyfoals and through experimental infections for over20 years (eg Burrell 1985; Blunden et al. 1994;Burrell et al. 1996; Wood et al. 1997; Newton et al.2001). Detailed clinical investigations ofrespiratory disease and loss of performance inBritish racing yards have demonstrated that IAD ispresent in over 65% of outbreaks, raising thesuspicion that an infection, particularly withviruses, plays a key role in the aetiology. Either ahistory of prior IAD or IAD as a concurrent findingis common in young racehorses presented for lossof performance investigation within this clinic.

However, interpretation of data from case andoutbreak investigations is complicated by the highincidence of infections in these young animals andthe lack of appropriate control populations.

Detailed, prospective, longitudinalobservational epidemiological studies have beenundertaken to study respiratory disease in youngBritish Thoroughbred racehorses. These studiesgive a basic description of the epidemiology. Theyalso show how the disease incidence andprevalence vary between training yard, year anddecrease with age (which is closely correlated withduration of stay in the training yard environment).The variation in prevalence between ages andcalendar months is shown by Newton et al. (2003)and Wood et al. (1999).

One key finding from this research was that theaverage duration of disease incidents was around 9weeks, but with a very long tail to the distribution.Some young racehorses, particularly in their ‘2-year-old’ year, were affected for over 8 months,whereas the disease was not detected at themonthly examinations in 20% (all of whichremained healthy during their ‘3-year-old’ year).Such an effect will only be apparent in prospectivestudies following horses over long periods (Fig 1).

Although in these studies, horses wereclassified as having or not having IAD, animportant consideration is that different clinicalfindings might be expected in a horse sampled in itsfirst week of being affected with acute disease,compared to those in an ‘average’ horse in its 9thweek or a badly affected animal in its 6th month ofdisease (perhaps in its 1st year in training). This isa important when comparing studies of naturallyaffected animals, particularly in relation to referralbias. Certainly in the UK, and probably elsewhere,trainers are much less likely to present horses forveterinary examination when in their 1st month of


34

disease. Such an effect will be exaggerated by thefact that the disease, overall, is sub-clinical for>60% of the time and the likelihood of horsescoughing is higher in the 2nd than the 1st month ofdisease (Burrell et al. 1996).

Studies referred to in the remainder of this paperare largely restricted to those in horses (mostlyracehorses) and ponies less than 6 years of age thatwere affected naturally with the disease syndrome.Because there are few published studies that haveused IAD as the case definition, studies of coughing(Christley et al. 2001), which is a fairly specific, ifinsensitive (Burrell et al. 1996), measure of IAD inyoung racehorses have also been considered.

Whenever multidisciplinary studies of IAD havebeen undertaken, they have provided clear evidenceof a multifactorial aetiology (eg Burrell et al. 1996;Wood 1999). Considerable care is needed, therefore,to ensure that results from single factor studies (eg ofstabling) are not confounded by unmeasured effects,such as of viral infection, which might be associatedwith the stresses of housing grazing animals.

HOW COMMON ARE THE DISEASE ANDVIRAL INFECTIONS ?

IAD is common in young racehorses. In a largescale study Wood et al. (1999) estimated the overallmonthly prevalence to be 13.8% and the incidenceto be 8.9 cases/100 horses/month. Criteria for

assigning causality in natural disease are reviewedby Newton et al. (2003), but it is clear that for viralinfections to play a role in the aetiology of IAD,they must: a) occur in affected animals; and b) theirpresence must be associated with occurrence ofdisease. Few studies have assessed the role of viralinfections in IAD and those that have beenidentified are shown in Table 1. For prospective, orlongitudinal studies, incidence rates are presented,but for case control studies, prevalences ofinfection in cases are included. It is clear from thedata in the table that incidence rates of differentviral infections were generally lower than theincidence of IAD and the prevalence proportions incases (where reported) was low (<10%). Theexception was the reported association betweenlung disease in foals and active equine herpesvirus-2 (EHV-2) in the lung (Murray et al. 1996). On thebasis of these results, it seems unlikely that theseviruses are key to the aetiology of IAD in youngracehorses.

WHAT ARE THE ASSOCIATIONS BETWEEN

VIRAL INFECTION AND IAD?

In the study of Burrell et al. (1996), no viralinfection was associated with IAD, although upperrespiratory tract disease was statisticallyassociated with EHV-1/4 infection. However,considerable care is needed when interpreting

1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33

18

16

14

12

10

8

6

4

2

0

Fre

quen

cy

Number of monthly samples collected

Fig 1: Number of months that individual horses were sampled over a 3 year study period (Wood et al. 1999).


35

statistically non-significant findings, as the lack ofsignificance can occur due to lack of effect, orinsufficient sample size to detect any effect.

In a larger study undertaken in 7 training yardsover 3 years, EHV (1 and 4) was the most commonviral infection, with an overall incidence of 4.4cases/100 horses/month (Wood et al. 1999). Thiswas the only viral infection statistically associatedwith IAD and horses with EHV-1 or EHV-4infections that month (diagnosed serologically)were around 5 times as likely to be affected withIAD than those that had not experienced it (oddsratio=4.9, 95% confidence interval 1.4–17.4).However, it only occurred at 7.5% of samplingpoints, and was detected in only 7.5% of incidentcases (Wood 1999). A similar prevalenceproportion for EHV-1 or EHV-4 was reported byHoffman et al. (1993: 4%) and Christley et al.(2001: 10%) and neither of these authors foundEHV to be associated with disease.

These and other case control studies ofrespiratory disease or coughing in British orAustralian racehorses (Newton et al. 2001) foundno association of any virus with disease, other thana minor one for influenza in the British horses(Newton et al. 2001).

In contrast, Moore et al. (1996) reported thatthe BAL fluid differential cytology in horses withIAD supported the hypothesis of persistent viralrespiratory tract infection as a potential cause ofIAD, but no specific viral investigations wereundertaken. However, in the authors’ experiencethe inflammatory response and cytological picturein IAD associated with bacteria, allergy andviruses may be very difficult to differentiate.Subsequent investigations demonstrated thathuman interferon alpha had significant efficacy intreating IAD in young racehorses, after a meanduration of 3.1 months (Moore et al. 1996, 1997).

The results of this work are hard to evaluate interms of the primary aetiology of IAD, but theyprovide an interesting insight into the morechronic disease.

Contrasting results were also produced forfoals by Murray et al. (1996), who found a strongstatistical association between the presence ofEHV-2 in tracheal aspirates and clinical signs oflower respiratory disease (although no mention wasmade of bacteriological or cytological evaluation).The virus was isolated from 20/30 cases but only1/20 controls (although it was present in the bloodof almost all cases and all controls).

DISCUSSION

Studies of the association between acute viralinfections and IAD have largely been consistent indemonstrating that known equine viruses do notplay a substantial role in the aetiology orpathogenesis of IAD in the racehorse. In particular,there is evidence that EHV-1 and EHV-4 (whichserologically cross react) play only a small role incausing the disease. The size of this role reflects theincidence of infection in the horse populations inquestion. There is consistency in the effect thatEHV-1/4 appear to have (when they do occur) inthat a transient neutrophilia was induced by EHV-1infection in one reported experimental study (Kyddet al. 1996) and during some of the authors’(unpublished) investigations into outbreaks ofrespiratory disease in racehorses, a TW neutrophilia(often in the absence of increased mucus) hasfrequently been observed in outbreaks of EHVrespiratory disease.

The disease that the influenza virus causes insusceptible animals undoubtedly involves airwayinflammation but, other than in outbreaks in wellvaccinated populations, usually occurs in distinct

TABLE 1: Summary of rates of viral infections in studies of naturally occurring IAD

Rate of infection (/100 horses/month or % if incidence)Study Influenza EHV ERV-1 ERV-2 Adenovirus EHV-2

Burrell et al. (1996) 0 10.9 5.5 N/A 0.8 N/AWood (1999) 0.8 4.4 2.2 0.8 1.5 N/AThomson (1978) N/A 3.3 N/A N/A N/A N/AHoffman et al. (1993) 0% 4% N/A N/A 0% N/AChristley et al. (2001) 0% 9% 8% 2% N/A N/AMurray et al. (1996) N/A N/A N/A N/A N/A 100% in blood,

66% in lungMoore et al. (1995) N/A N/A N/A N/A N/A N/A

N/A: Not assessed or not reported


36

outbreaks and is not confused with the endemicIAD that affects most groups of young racehorsesin training.

If strong statistical associations between viralinfection and IAD had been demonstrated by manyworkers, as for bacterial infections, it would benecessary to consider whether or not theassociations were confounded by other effects.However, the lack of evidence of significantcausation means that confounding should not beconsidered a major issue in judging whether or notthe role of acute viral infection is important.

There are questions that remain around the roleof EHV-2 infection, and perhaps that of other,unidentified viral infections, in IAD. One study hasdemonstrated a strong association betweendetection of EHV-2 in tracheal aspirates and diseasein young foals, but we have been unable to detectthis virus in similar samples collected from Britishracehorses, using appropriate cultural methods(unpublished observations). More advancedmolecular diagnostic techniques might help toresolve the role that this ubiquitous infection playsin several disease syndromes, including IAD.

It is essential in all future studies of theaetiopathogenesis of IAD in horses, that the stage ofdisease in each animal be considered before over-optimistic statements are made about cause and role.

CONCLUSIONS

To conclude, there is a growing body of evidenceto suggest that the role of viral infection in theaetiology of IAD is small, although further workneeds to be done on chronic, low grade infectionssuch as EHV-2.

REFERENCES

Blunden, A.S, Hannant D., Livesay, G. and Mumford, J.A.(1994) Susceptibility of ponies to infection withStreptococcus pneumoniae (capsular type 3) Equinevet. J. 26,22-28.

Burrell, M.H. (1985) Endoscopic and virologicalobservations on respiratory disease in a group ofyoung Thoroughbred horses in training. Equine vet. J.17,99-103.

Burrell, M.H., Wood, J.L., Whitwell, K.E., Chanter, N.,Mackintosh, M.E. and Mumford, J.A. (1996)Respiratory disease in thoroughbred horses in training:the relationships between disease and viruses, bacteriaand environment. Vet. Rec. 139, 308-313.

Christley R.M, Hodgson D.R, Rose R.J, Wood J.L.N.,Reid S.W.J., Whitear K.G. and Hodgson, J.L. (2001)A case-control study of respiratory disease inThoroughbred racehorses in Sydney, Australia.Equine vet. J. 33, 256-264.

Hoffman, A.M., Viel, L., Prescott, J.F., Rosendal S. andThorsen, J. (1993) Association of microbiologicflora with clinical, endoscopic, and pulmonarycytologic findings in foals with distal respiratorytract infection. Am. J. vet. Res. 54, 1615-1622.

Kydd, J.H., Hannant, D. and Mumford, J.A. (1996)Residence and recruitment of leucocytes to theequine lung after EHV-1 infection. Vet. Immun.Immunopath. 52, 15-26.

Moore, B.R., Krakowka, S., Cummins, J.M. andRobertson, J.T. (1996) Changes in airwayinflammatory cell populations in standardbredracehorses after interferon-alpha administration. Vet.Immun. Immunopath. 49, 347-358.

Moore, B.R., Krakowka, S., Mcvey, D.S., Cummins J.M.and Robertson J.T. (1997) Inflammatory markers inbronchoalveolar lavage fluid of standardbredracehorses with inflammatory airway disease:response to interferon-alpha. Equine vet. J. 29, 142-147.

Moore, B.R., Krakowka, S., Robertson, J.T. andCummins, J.M. (1995) Cytologic evaluation ofbronchoalveolar fluid obtained from Standardbredracehorses with inflammatory airway disease.Equine vet. J. 29, 142-147.

Murray, M.J., Eichorn, E.S., Dubovi, E.J., Ley, W.B. andCavey, D.M. (1996) Equine herpesvirus type2:prevalence and seroepidemiology in foals. Equinevet. J. 28, 432-436.

Newton, J.R., Wood, J.L.N. and Chanter, N. (2001) Acase control study of factors and infectionsassociated with clinically apparent respiratorydisease in UK racehorses. Proc. Soc. vet. epid. Prev.Med. Eds. S.W.J. Reid & F. Menzies pp 127-140.

Newton, J.R., Wood, J.L.N., Smith, K.C., Marlin, D.J.and Chanter, N. (2003) Aetiological agents: bacteria.Workshop on ‘Inflammatory Airway Disease:Defining the Syndrome’. Havemeyer FoundationMonograph Series No 9, Eds: A. Hoffman, N. E.Robinson and J.F. Wade, R&W Publications(Newmarket) Ltd, pp 40-44.

Thomson, G.R. (1978) The Role of Equid Herpesvirus 1in Acute Respiratory Disease of Horses and the Useof Inactivated Antigens to Immunize Against theInfection. PhD Thesis, University of London.

Wood, J.L.N. (1999) An Epidemiological Investigation ofRespiratory Disease in Racehorses. PhD Thesis.Open University.

Wood, J.L., Newton, J.R., Chanter, N., Mumford, J.A.,Townsend, H.G.G., Lakhani, K.H., Gower, S.M.,Burrell, M.H., Pilsworth, R.C., Shepherd, M., Hopes,R., Dugdale, D., Herinckx, B.M.B, Main, J.P.M,Windsor, H.M. and Windsor, G.D. (1997) Alongitudinal epidemiological study of respiratorydiseases in racehorses. Proc. 8th int. Conf. equine inf.Dis. Eds: U. Wernery, J.F. Wade, J.A. Mumford andO-R. Kaaden, R&W Publications (Newmarket), pp64-70.

Wood J.L.N., Newton J.R., Chanter N., TownsendH.G.G., Lakhani, K.H., Mumford J.A., Sinclair, R.,Burrell, M.H., Pilsworth, R.C., Shepherd, M.,Dugdale, D., Herinckx, B.M.B., Main, J.P.M.,Windsor, H. and Windsor, D. (1999) A longitudinalepidemiological study of respiratory disease inracehorses: disease definitions, prevalence andincidence. Proc. 8th int. Conf. equine inf. Dis. Eds:U. Wernery, J.F. Wade, J.A. Mumford & O. Kaaden.

37


NATURAL HISTORY OF EQUINE INFLUENZA

H. G. G. Townsend

Department of Large Animal Clinical Studies and Veterinary Infectious Disease Organisation,University of Saskatchewan, Saskatoon, Saskatchewan, Canada S7N 5B4

EPIDEMIOLOGY

Equine influenza (EI) is a cause of inflammationof the respiratory tract and is the most commonlydiagnosed viral respiratory disease of the horse.The virus is highly infectious and contagious withtransmission occurring via direct, aerosol andfomite exposure. Outbreaks occur most often ingroups of racehorses, with those 2–3 years of agesuffering the highest morbidity (Newton andMumford 1995; Morley et al. 2000a). Subclinicalinfection is probably common in horses older thanone year of age (unpublished data).

CLINICAL SIGNS

The signs of EI seen most frequently are fever,cough and mucopurulent nasal discharge. Affectedhorses may also experience malaise, inappetance,dyspnea, weight loss and loss of athleticperformance (Gross et al. 1998). In general, signsof fever and coughing resolve by about Day 10post challenge, but close examination will revealevidence of mucopurulent nasal discharge for atleast 21 days in some animals. Serial endoscopy ofexperimentally infected foals (Sutton et al. 1997)provided visual evidence of inflammation of therespiratory tract, including findings consistentwith pharyngeal lymphoid hyperplasia,hyperaemia and oedema of the mucosa of the largeairways and mucopurulent debris in the trachea.Detailed serial ultrasonographic examination of 8horses following experimental challenge showedevidence of pulmonary consolidation, oedema andfluid filled airways by Day 7 post challenge, withresolution of these signs by Day 14 (Gross et al.1998).

PATHOLOGY

Infection of equine fetal tracheal explants isfollowed by viron budding at the base of microvilliof infected tracheal epithelium 12 h post infection.Virons were visible in the interstitial space by 24 h,scattered loss and clumping of cilia were evident insamples taken between 24 and 72 h post infection(O’Niell et al. 1984). Logical consequences ofthese lesions are decreased rate of mucociliaryclearance (Willoughby et al. 1992) and secondarybacterial infections of the respiratory tract(Sarasola et al. 1992; Newton et al. 1999).

Following experimental infection, the virusreplicates in the epithelial cells lining therespiratory tract. Peak death and shedding of thesecells occurs around Day 2 and continues for up to8–10 days post challenge. Viral antigen associatedwith cellular debris can be found inbronchoalveolar lavage (BAL) fluid samples 21days post infection (Sutton et al. 1997).Subsequent to experimental infection of ponyfoals, immunoperoxidase staining was used todetect the presence of viral antigen in cellscollected from the nasopharynx, trachea andmainstem bronchus using cytological brushespassed through an endoscope, and in samplesobtained by bronchoalveolar lavage (Sutton et al.1997). The antigen was particularly obvious inexfoliated ciliated columnar epithelial cells thatwere found in BAL fluid samples.

Numbers of neutrophils in BAL fluid sampleswere significantly increased in samples collectedon Days 3 and 7 after experimental challenge ofpony foals, with half the animals having highneutrophil counts in BAL fluid until Day 21(Sutton et al. 1997). Sequential BAL fluid samples


38

following experimental challenge of horsesshowed increases in the percentage of neutrophilsuntil the end of data collection 13 and 28 days postchallenge (Gross et al. 1998; Kastner et al. 1999).The results of complete blood counts performedduring both of these studies were also consistentwith a systemic inflammatory response (increasedneutrophil count and plasma fibrinogenconcentration) persisting for at least 21 days postchallenge.

Gross et al. (1998) observed intra-cellularbacteria in 25% of transtracheal aspirates fromchallenged horses on Days 4, 8 and 14 postchallenge. Kastner (1999) isolated bacteria fromBAL fluid samples up to 13 days post challengeand in another study large numbers ofStreptococcus zooepidemicus were isolated fromprimary culture of all nasal swabs obtained from30 influenza challenged horses 10 and 14 postchallenge (H.G.G. Townsend, unpublished data).

Equine influenza has been described as aninflammation of the upper respiratory tract.However, infection and death of epithelial liningcells of the respiratory tract and alveolarmacrophages provides evidence for widespreadinfection of the lower respiratory tract. Clumpingand loss of cilia occurs and this provides a logicalexplanation for the measured decreases inmucociliary clearance and proliferation of bacteriain the respiratory tract. Exposure of the basallamina occurs along the length of the trachea andbronchi. Infection and death of epithelial cells andsubsequent proliferation of respiratory tractbacteria explain the signs of inflammatory airwaydisease observed following natural andexperimental equine influenza infections.Although the clinical signs of disease may resolvewithin a few days, there is ample evidence to showthat inflammation of the respiratory tract maypersist in many animals for at least 21 days and insome for more than 28 days. Detailed studiesfollowing animals beyond 28 days post challengehave not been published. Information from suchstudies would be helpful in assessing any longterm consequences of this disease.

RELATIONSHIP BETWEEN INFLUENZA ANDOTHER IMPORTANT CAUSES OF IAD

Clearly, infection with influenza virus causesinflammation of the respiratory tract. In mostsituations the epidemiology, clinical history,clinical signs and laboratory findings of EI should

allow differentiation of this disease from IAD(inflammatory airway disease) and heaves.Although influenza and IAD are both commoncauses of airway inflammation in 2–3- year-oldhorses, there is good epidemiologic evidenceshowing that most cases of IAD of youngracehorses are not directly associated withprevious EIV infections (Christley et al. 2001). Itis possible that pre-existing IAD may influence theclinical severity of concurrent EIV infections.However, there appear to be no published reportsaddressing this issue.

Thorsen et al. (1983) reported high levels ofhaemagglutination inhibiting activity againstinfluenza A1, but not A2, in mucus samples fromhorses with chronic obstructive pulmonarydisease. However, well-designed studies aimed atdetecting a causal relationship between EI virusinfection and heaves have not been published.Although there is insufficient data to discount thishypothesis, there are arguments against anyassociation between EI and the subsequentdevelopment of heaves. Importantly, the peakincidence of influenza occurs in young horses(Morley et al. 2000b) several years before the peakincidence of heaves. Also, there are no publishedreports of increased incidence or prevalence ofheaves following natural outbreaks of influenza,nor are there any reports of increased incidence ofheaves among the thousands of horses that havebeen experimentally infected with influenza virus.

Although it appears that viral respiratorydisease of horses is not a pre-disposing factor inthe development of IAD or heaves, it is possiblethat both of these diseases could be exacerbated byconcurrent infection with influenza virus.Presently, there appear to be no reports of field orexperimental studies on the impact of acute viralinfections in animals with pre-existing IAD orheaves, even though there is no reason why thisshould not occur on a regular basis. An explorationof the clinical effects of influenza in animals withpre-existing IAD or heaves, could prove veryenlightening.

REFERENCES

Christley, R.M., Hodgson, D.R., Rose, R.J., Wood, J.L.,Reids, S.W., Whitear, K.G. and Hodgson, J.L.(2001) A case-control study of respiratory disease inThoroughbred racehorses in Sydney, Australia.Equine vet. J. 33, 256-264.

Gross, D.K., Hinchcliff, K.W., French, P.S., Goclan,S.A., Lahmers, K.K., Lauderdale, M., Ellis, J.A.,


39

Haines, D.M., Slemons, R.D. and Morley, P.S.(1998) Effect of moderate exercise on the severity ofclinical signs associated with influenza virusinfection in horses. Equine vet. J. 30, 489-497.

Kastner, S.B., Haines, D.M., Archer, J. and Townsend,H.G. (1999) Investigations on the ability ofclenbuterol hydrochloride to reduce clinical signsand inflammation associated with equine influenzaA infection. Equine vet. J. 31, 160-168.

Morley, P.S., Townsend, H.G., Bogdan, J.R. and Haines,D.M. (2000a) Descriptive epidemiologic study ofdisease associated with influenza virus infectionsduring three epidemics in horses. J. Am. vet. med.Ass. 216, 535-544.

Morley, P.S., Townsend, H.G.G., Bogdan, J.R. andHaines, D.M. (2000b) Risk factors for diseaseassociated with influenza virus infections duringthree epidemics in horses. J. Am. vet. med. Ass. 216,545-550.

Newton, J.R. and Mumford, J.A. (1995) Equineinfluenza in vaccinated horses. Vet. Rec. 137, 495-496.

Newton, J.R., Verheyen, K., Wood, J.L.N., Yates, P.J. andMumford, J.A. (1999) Equine influenza in theUnited Kingdom in 1998. Vet. Rec. 145, 449-452.

O’Niell, F.D., Issel, C.J. and Henk, W.G. (1984) Electronmicroscopy of equine respiratory viruses in organcultures of equine fetal respiratory tract epithelium.Am. J. vet. Res. 45, 1953-1960.

Sarasola, P., Taylor, D.J., Love, S. and McKellar, Q.A.(1992) Secondary bacterial infections following anoutbreak of equine influenza. Vet. Rec. 131, 441-442.

Sutton, G.A., Viel, L., Carman, P.S. and Boag, B.L.(1997) Study of the duration and distribution ofequine influenza virus subtype 2 (H3N8) antigens inexperimentally infected ponies in vivo. Can. J. vet.Res. 61, 113-120.

Thorsen, J., Willoughby, R.A., McDonell, W., Valli, V.E.,Viel, L. and Bignell, W. (1983) Influenzahemagglutination inhibiting activity in respiratorymucus from horses with chronic obstructivepulmonary disorders (heaves syndrome). Can. J.comp. Med. 47, 332-335.

Willoughby, R., Ecker, G., McKee, S., Riddolls, L.,Vernaillen, C., Dubovi, E., Lein, D., Mahony, J.B.,Chernesky, M. and Nagy, E. (1992) The effects ofequine rhinovirus, influenza virus and herpesvirusinfection on tracheal clearance rate in horses. Can. J.vet. Res. 56, 115-21.

40


AETIOLOGICAL AGENTS: BACTERIA

J. R. Newton, J. L. N. Wood, K. C. Smith, D. J. Marlin and N. Chanter*

Animal Health Trust, Lanwades Park, Kentford, Newmarket, Suffolk CB8 7UU *Intervet UK Ltd,Walton Manor, Walton, Milton Keynes, Buckinghamshire MK7 7AJ, UK

Inflammatory airway disease (IAD), characterisedby mucus and a neutrophilic inflammatoryresponse in the distal airways, is an important formof respiratory disease in both young Thoroughbredracehorses and Welsh Mountain ponies, despitethe marked differences in the management of these2 equine populations.

The role of bacteria in the aetiology of IADremains controversial due to the potentially mixednature of infections, and the possibility ofcontamination of the lower airways from bacteriaof upper respiratory tract origin duringendoscopy. However, there is increasing evidencefor a causal involvement of bacteria in IAD inyoung horses.

A study in racehorses in the UK demonstrated,after allowing for the effects of trainer, age, season,equine herpesvirus (EHV) infection, disease theprevious month and random variation betweenhorses, that IAD was still significantly associatedwith lower airway infection with a limited numberof bacterial species, particularly Streptococcuszooepidemicus, Actinobacillus/Pasteurella spp. andStreptococcus pneumoniae (Wood 1999). Theseasonal variation in prevalence of IAD amongracehorses linked to the introduction of yearlings inthe autumn and the decreasing amount of diseasewith increasing age and time in training, are allconsistent with acquisition of immunity against aninfectious aetiology (Fig 1).

0.2550.2350.2150.1950.1750.1550.1350.1150.0950.075

1 2 3 4 5 6 7 8 9 10 11 12Age (month)

Pre

vale

nce

0.3

0.25

0.2

0.15

0.1

0.05

Pre

vale

nce

1 2 3 4 5Age (year)

0.425

0.375

0.325

0.275

0.225

0.175

0.125

0.075

Pre

vale

nce

<1 1–3 3–6 6–12 >12

Time in training (months)

Yearlings

Fig 1: Variation in prevalence of IAD in youngracehorses with: a) month of the year; b) age; and c)time in training.

a) b)

c)


41

The traditional use of Koch’s postulates(recovery of organism from natural cases,reproduction of signs with inoculation andrecovery of organism from induced lesions) toconfirm a causal association between disease andbacteria is problematical for IAD. This is becauseof the multiple nature and high prevalence ofnatural infections and absence of specific pathogenfree animals. However, use of the 9 Bradford-Hillcriteria (Hill 1965) to assess whether theepidemiological associations between thesebacteria and IAD are likely to be causal, doessuggest that bacteria cause lower respiratory tractdisease in horses (Wood and Chanter 1994).

Several studies have confirmed a reasonablestrength of association (measured as relative risk)and a significant biological gradient for thesebacteria and IAD, with increasing numbers ofbacteria having a greater strength of associationwith disease (Burrell et al. 1986, 1996; Wood et al.1993; Wood and Chanter 1994; Wood 1999;Chapman et al. 2000). Examination of trachealwash data from young Thoroughbred racehorses,using a 0 to 9 ordinal scoring of airwayinflammation, demonstrates a significant biologicalgradient for increasing mean log colony formingunits per ml of S. zooepidemicus andActinobacillus/Pasteurella spp. and increasinginflammation score (Fig 2). There was no suchgradient for Staphylococcus spp.

The associations with IAD are also specific forthe three bacterial species of Streptococcuszooepidemicus, Actinobacillus/Pasteurella spp. andStreptococcus pneumoniae and are consistentbetween different studies (Burrell et al. 1986, 1996;Wood et al. 1993; Wood and Chanter 1994; Wood1999; Chapman et al. 2000). Bacterial infections ascauses of IAD in young horses are biologicallyplausible, are coherent as the results do not conflictwith current knowledge, and are analogous with

similar and identical bacteria being the cause ofrespiratory disease in other species. Furthermore, S.equi, a subtype of S. zooepidemicus (Chanter et al.1997), causes ‘strangles’ in horses and is generallyaccepted as a primary bacterial equine pathogen.There is limited experimental evidence that intra-tracheal inoculation of S. pneumoniae causes IADand pneumonia in young pony foals, with theorganism being recovered from the trachea andpneumonic lesions (Blunden et al. 1994). Atemporal relationship between bacterial infectionand IAD is difficult to confirm due to thepossibility of an earlier pre-disposing viralinfection. However, results of a longitudinal studyof IAD failed to show that the association betweenbacteria and IAD is dependent on prior infectionwith any of the known equine viruses (Wood 1999).

The use of endoscopy to examine and samplethe distal airways has been criticised forintroducing upper respiratory tract bacteria,consequently precluding meaningful evaluation oftracheal wash bacteriology results and theirassociation with IAD. However, a large proportion(62%; Wood et al. 1993) of endoscopicallyrecovered samples remain bacteriologically sterileduring routine examinations. There is a highlysignificant trend towards lower proportions ofsterile washes with increasing airwayinflammation (P<0.00001; Wood et al. 1993;Newton 2002). Furthermore, analyses ofassociations with IAD restricted only to horseswith certain bacteria recovered from thenasopharynx (Table 1) show highly significanttrends in association with increasing inflammationspecifically for S. zooepidemicus, Actinobacillus/Pasteurella spp. and S. pneumoniae (P<0.0006).There are no such trends for other potentialpathogens found in the nasopharynx such asStaphylococcus and Acinetobacter spp. (P>0.6).

Novel approaches have been used to

7

6

5

4

3

2

1

00 1 2 3 4 5 6 7 8 9

Inflammation score

Mea

n lo

g ba

cter

ia (

cfu/

ml)

� S. zooepidemicus

� Pasteurella spp

Fig 2: Variation in meanlog10 cfu/ml of 3 bacterialspecies with inflammationscore in young racehorses.


42

TABLE 1: Linear trend in proportions of tracheal washes positive for different bacterial speciesamong only horses from which the same organism was isolated from the upper respiratory tract

Case/control Inflammation TWs (%) positive Total TWs Odds ratioscore for bacterial spp.* examined (100%)

Streptococcus zooepidemicusControl 0 21 (28%) 74 1.00 (ref)Case 0 3 (30%) 10 1.08Control 1 28 (45%) 62 2.08Case 1 6 (60%) 10 3.79Control ≥2 30 (83%) 36 12.6Case ≥2 20 (71%) 28 6.31χ2 for linear trend = 33.25 (P≤0.00001)

Pasteurella/Actinobacillus spp.Control 0 19 (22%) 85 1.00 (ref)Case 0 4 (31%) 13 1.54Control 1 25 (32%) 77 1.67Case 1 6 (43%) 14 2.61Control ≥2 16 (80%) 20 13.9Case ≥2 13 (76%) 17 11.3χ2 for linear trend = 30.28 (P≤0.00001)

Streptococcus pneumoniaeControl 0 7 (24%) 29 1.00 (ref)Case 0 1 (33%) 3 1.57Control 1 9 (32%) 28 1.49Case 1 2 (67%) 3 6.29Control ≥2 11 (69%) 16 6.91Case ≥2 7 (70%) 10 7.33χ2 for linear trend = 11.85 (P=0.00058)

Staphylococcus spp.Control 0 110 (43%) 253 1.00 (ref)Case 0 8 (28%) 29 0.50Control 1 88 (45%) 197 1.05Case 1 15 (45%) 33 1.08Control ≥2 32 (48%) 67 1.19Case ≥2 12 (29%) 42 0.52χ2 for linear trend = 0.238 (P=0.63)

Acinetobacter spp.Control 0 8 (12%) 67 1.00 (ref)Case 0 1 (14%) 7 1.23Control 1 7 (11%) 63 0.92Case 1 1 (11%) 9 0.82Control ≥2 10 (14%) 74 1.15Case ≥2 1 (5%) 21 0.37c2 for linear trend = 0.07 (P=0.80)

Case = horse showing clinical signs (nasal discharge, coughing or pyrexia)Control = horse NOT showing clinical signs (nasal discharge, coughing or pyrexia)

investigate variability, both in susceptibility tobacterial infection among horses and in infectionby different S. zooepidemicus subtypes (Newton2002).

Transferrin, an iron-binding host protein, mayact as a source of iron for pathogenic respiratorybacteria such as Actinobacillus spp. and Pasteurellaspp., which have been shown to chelate iron fromtransferrin using transferrin-binding proteins(Ratledge and Dover 2000). To this end the authors

have examined whether different equine transferrinhaplotypes are associated with differences inmeasures of disease and infections of the respiratorytract for 2 distinct forms of respiratory disease in 2separate equine populations.

Analyses examining the effects of genetichaplotypes of iron-binding transferrin on IAD inracehorses demonstrated a significantly protectiveeffect of the D haplotype. In transferrin haplotypeD positive horses there was a reduced risk of


43

Fig 3: Proportions of ponies positive by week for tracheal wash (TW – solid line) and nasopharyngeal (NP – hatchedline) isolates of 4 subtypes of S. zooepidemicus.

40

35

30

25

20

15

10

5

0

% p

onie

s po

sitiv

e

1 2 3 4 5 6 7 8 9 10Week

A1 HV1

A1 HVu

H1 HV3 C1 HV3

IAD and reduced counts of S. zooepidemicus, S. pneumoniae, S. equisimilis and Mycoplasmaequirhinis. In addition, horse-level analysesdemonstrated that horses with haplotype Dsuffered a significantly lower overall prevalence ofIAD during months of repeated examination in alongitudinal study, and that this was primarily dueto significantly shorter mean duration of IADepisodes. These results were consistent with thedirection of effect of this haplotype for clinicalrespiratory disease and S. zooepidemicus infectionin Welsh Mountain pony foals.

Pony-level analyses from a blinded,randomised, controlled trial of a bacterial vaccineshowed good correlation between clinical scoresand mean S. zooepidemicus counts (R2=0.46,P<0.001), although there were no significantdifferences in clinical measures or infectious scoresbetween genders or vaccine groups (Newton 2002).There were, however, significant differences inclinical (0.0002<P<0.025) and S. zooepidemicusinfection measures (0.0007<P<0.019) betweenanimals possessing different transferrin haplotypes.Ponies with transferrin D haplotype demonstrated

significantly less clinical disease and S.zooepidemicus infection than those that did notpossess this transferrin type. Ponies with the F2haplotype, had significantly higher measures ofdisease and infection.

Further work is required to investigate whetherthe effects of transferrin observed in these 2separate equine populations are directlyattributable to differences in transferrin or are dueto a genetically closely linked mechanism.

To estimate the prevalence of different S.zooepidemicus types and investigate theassociation of types with respiratory disease,tracheal and nasopharyngeal isolates from WelshMountain pony foals and Thoroughbredracehorses were typed by 2 different PCR assays(Chanter et al. 1997; Walker and Timoney 1998).PCR of the 16S-23S RNA gene intergenic spacerof S. zooepidemicus could identify 8 distinct types(A1&2, B1&2, C1&2, D1&2; Chanter et al.1997), and PCR of the M-protein hypervariableregion could classify 5 types (HV1-5) 12, giving40 total possible types.

A group of 29 recently weaned Welsh

40

35

30

25

20

15

10

5

0

% p

onie

s po

sitiv

e

1 2 3 4 5 6 7 8 9 10Week

40

35

30

25

20

15

10

5

0

% p

onie

s po

sitiv

e

1 2 3 4 5 6 7 8 9 10Week

40

35

30

25

20

15

10

5

0

% p

onie

s po

sitiv

e

1 2 3 4 5 6 7 8 9 10Week


44

Mountain ponies on the bacterial vaccine trialwere assessed clinically and sampled weekly bytracheal wash and nasopharyngeal swab for 10consecutive weeks, and again 16 weeks later(n=319 samplings). S. zooepidemicus was isolatedfrom >90% of all pony samples (Newton 2002).From 538 typed isolates, 39 different types wereidentified including previously unidentified HVtypes (HVu) and polymorphic forms (egA1/D1HV1/2). Of typed isolates, 89% were of 12types and 52% were of only 4 types (A1HV1,A1HVu, A1HV3, C1HV3). More types wereisolated from the trachea than nasopharynx and24% of HV types were polymorphic compared to6% of intergenic spacer types. At the ponypopulation-level the prevalence of types variedover time, with evidence for clonal succession asnasopharyngeal types predicted tracheal types well(Fig 3; Newton 2002). There was a significantpositive correlation between numbers of coloniesof specific types and clinical respiratory score.

In addition, Thoroughbred racehorses in 3Newmarket training yards were sampledapproximately monthly between February andSeptember 2000 with 761 tracheal washes (3yards) and 225 nasopharyngeal swabs (2 yards)submitted. Of these, 22% of samples were positivefor S. zooepidemicus. From 217 typed isolates, 23different types were identified and >50% oftracheal and nasopharyngeal isolates were of only5 and 3 types, respectively.

In conclusion, despite issues over samplingtechniques and the current lack of experimentalreproduction of disease, there remainsconsiderable credible evidence for a causalassociation between IAD and a limited number ofspecific bacterial infections of the lowerrespiratory tract in young horses. New approachesare providing better insights into the complexinterplay of host and bacteria. This may lead to arealistic means of targeted control of IAD in theseyoung animals.

REFERENCES

Blunden, A.S., Hannant, D., Livesay, G. and Mumford,J.A. (1994) Susceptibility of ponies to infection withStreptococcus pneumoniae (capsular type 3). Equinevet. J. 26, 22-28.

Burrell, M.H., Mackintosh, M.E. and Taylor, C.E. (1986)Isolation of Streptococcus pneumoniae from therespiratory tract of horses. Equine vet. J. 18, 183-186.

Burrell, M.H., Wood, J.L.N., Whitwell, K.E., Chanter,N., Mackintosh, M.E. and Mumford, J.A. (1996)Respiratory disease in thoroughbred horses intraining: the relationships between disease andviruses, bacteria and environment. Vet. Rec. 139,308-313.

Chanter, N., Collin, N., Holmes, N., Binns, M. andMumford, J.A. (1997) Characterisation of theLancefield group C streptococcus 16S-23S RNAgene intergenic spacer and its potential foridentification and sub-specific typing. Epid. Inf. 118,125-135.

Chapman, P.S., Green, C., Main, J.P.M., Taylor, P.M.,Cunningham, F.M., Cook, A.J.C. and Marr, C.M.(2000) Retrospective study of the relationshipsbetween age, inflammation and the isolation ofbacteria from the lower respiratory tract ofThoroughbred horses. Vet. Rec. 146, 91-95.

Hill, A.B. (1965) The environment and disease:association or causation? Proc. Royal Soc. Med. 58,295-300.

Newton, J.R. (2002) Epidemiological Studies ofInflammatory Airway Disease in Horses. PhDThesis. Open University.

Ratledge, C. and Dover, L.G. (2000) Iron metabolism inpathogenic bacteria. Ann. Rev. Microbiol. 54, 881-941.

Walker, J.A. and Timoney, J.F. (1998) Molecular basis ofvariation in protective SzP proteins of Streptococcuszooepidemicus. Am. J. vet. Res. 59, 1129-1133.


Wood, J.L.N., Burrell, M.H., Roberts, C.A., Chanter, N.and Shaw, Y. (1993) Streptococci and Pasteurellaspp. associated with disease of the equine lowerrespiratory tract. Equine vet. J. 25, 314-318.

Wood, J.L.N. and Chanter, N. (1994) Can washing helpkeep the lungs clean? Equine vet. Educ. 6, 220-222.


45

DYSREGULATION OF INFLAMMATION

F. Bureau and P. Lekeux

Department of Functional Sciences, Faculty of Veterinary Medicine, University of Liege, Bat B42,Sart Tilman, 4000 Liege, Belgium

Chronic or recurrent airway disorders, such asheaves and inflammatory airway disease (IAD),are directly or indirectly related to an exaggeratedinflammatory reaction. It is important, therefore,to understand why the inflammatory process isdysregulated in these diseases.

Chronic or recurrent airway inflammations areassociated with overexpression of inflammatoryproteins involved in both immune andinflammatory responses. Atopic asthma is ahuman inflammatory disorder of the airwayswhich resembles heaves from an aetiological andpatho-physiological point of view (Snapper 1986;Robinson et al. 1996). Reports state that asthmaticinflammation is characterised by overexpressionof many inflammatory genes (Barnes 1996). Thesegenes encode: 1) pro-inflammatory cytokines,including interleukin-1β (IL-1β) and tumournecrosis factor-α (TNF-α), which amplifypulmonary inflammation; 2) chemokines, such asinterleukin-8 (IL-8), macrophage inflammatoryprotein-1α (MIP-1α), macrophage chemotacticprotein-3 (MCP-3), RANTES (regulated onactivation normal T- cell expressed and secreted)and eotaxin, which are chemotactic for leukocytes;3) adhesion factors, including inter-cellularadhesion molecule-1 (ICAM-1), vascular celladhesion molecule-1 (VCAM-1), and E-selectin,which play a cardinal role in leucocyterecruitment, margination, diapedesis andtransepithelial migration; and 4) inflammatoryenzymes, such as cytosolic phospholipase A2(cPLA2), inducible nitric oxide synthase (iNOS),which generate inflammatory mediators (Barnesand Adcock 1998). Protein overexpressiondepends on increased gene transcription,suggesting that activation of some transcriptionfactors underlies asthma pathogenesis. Most of theinflammatory genes overexpressed in asthma

contain κB sites for Nuclear Factor-κB (NF-κB)within their promoter (Baeuerle and Baichwal1997). This suggests that these genes arecontrolled predominantly by NF-κB, which couldbe of particular importance in the initiation and theperpetuation of allergic airway inflammation.

The NF-κB family is composed of 5 structurallyrelated DNA-binding proteins: p50, p52, p65/RelA,c-Rel/Rel, and RelB (Siebenlist et al. 1994). Themost common form is a heterodimer composed ofp50 and p65 subunits, although the different familymembers can associate in various homo- orheterodimers through a highly conserved N-terminal sequence, called the Rel homologydomain. Dimerisation of various NF-κB subunitsproduces complexes with different DNA-bindingspecificities and transactivation potentials. In mostcell types, inactive NF-κB complexes are associatedwith inhibitory proteins of the IκB family, whichsequester NF-κB in the cytoplasm. The members ofthe IκB family are IκB-α, IκB-β, IκB-ε, p100, p105and Bcl-3; the most common IκB protein is IκB-α(Siebenlist et al. 1994; Whiteside et al. 1997). p105and p100 are the precursors of p50 and p52,respectively. Following various stimuli, eg viruses,bacteria, pro-oxidants, and pro-inflammatorycytokines, IκB proteins are phosphorylated,ubiquitinated and rapidly degraded by theproteasome, allowing NF-κB nuclear translocationand transcriptional initiation of NF-κB-dependentgenes (Karin and Ben-Neriah 2000).

Increased NF-κB activity has beendemonstrated in macrophages of induced sputumand in bronchial epithelial cells of patients withstable asthma compared to normal subjects. Thisreinforces the assumption that persistence of NF-κB activity could be particularly important in thepathogenesis of chronic inflammation in asthmaand suggests that increased NF-κB activity could


46

play a role in other allergic airway disorders suchas heaves (Hart et al. 1998).

Studies of molecular mechanisms responsiblefor the inappropriate inflammatory reactionassociated with heaves, showed that NF-κB is moreactive in the bronchi of heaves-affected horses thanhealthy horses (Bureau et al. 2000a). Furthermore,NFκB activity was correlated closely to thepulmonary dysfunction associated with the diseaseand to the expression of ICAM-1 in the airways(Bureau et al. 2000a). Bureau et al. (2000a) alsoobserved that 21 days after removal of the allergen,NF-κB activity is maintained at high or moderatelevels in the bronchi of most heaves-affected horses,suggesting that perpetuation of NF-κB activitycould contribute to the persistence of bronchialinflammation when the allergen is evicted.

NF-κB induces: i) expression of pro-inflammatory cytokines which, in turn, activate NF-κB, initiating positive autoregulatory feedbackloops; and ii) expression of inhibitory proteins ofthe inhibitor of NF-κB family, thus initiatingnegative autoregulatory feedback loops. Bureau etal. (2000b) showed that positive autoregulatoryfeedback loops involving IL-1β and TNF-α developin the airways of heaves-affected horses followingallergen-induced crisis. Usually, negative loopsdominate, allowing transient, rather than persistent,NF-κB activity. Therefore, the authors of thepresent paper investigated whether such negativefeedback loops were present in the airways ofheaves-affected horses. It was observed that IκB-α,the most common inhibitor of NF-κB, is notexpressed in the bronchial cells of healthy orheaves-affected horses. Surprisingly, the prominentIκB protein present in the airways of healthy orheaves-affected horses was IκB-β, which is notunder the dependence of NF-κB for its expression.IkB-ß expression was fainter in the airways ofheaves-affected horses compared to healthysubjects. These observations indicate that negativeregulatory mechanisms are deficient in bronchialcells of heaves-affected horses and provide anexplanation for the persistent NF-κB activity foundin the bronchi of these horses.

These studies also showed that apoptosis issignificantly delayed in lung neutrophils fromheaves-affected horses. This delay is dependent onautocrine production of granulocyte/monocytecolony-stimulating factor (GM-CSF) and couldalso be involved in the persistence of inflammationobserved in heaves (Turlej et al. 2001).

Although some cases of IAD either resolvespontaneously or respond to antibiotic treatment, aproportion of IAD-affected horses show persistentinflammation, sometimes in the absence ofdetectable pathogens. The molecular mechanismsby which inflammation persists have not yet beenelucidated. Resolution of whether aberrant NF-κBactivity and/or increased granulocyte survival areinvolved in the maintenance of inflammation inhorses with IAD should provide valuableinformation regarding pathogenesis and couldassist the development of rational IAD therapies.

REFERENCES

Baeuerle, P.A. and Baichwal, V.R. (1997) NF-kappaB asa frequent target for immunosuppressive and anti-inflammatory molecules. Adv. Immunol.65, 111-137.

Barnes, P.J. (1996) Pathophysiology of asthma. Br. J.Clin. Pharmacol. 42, 3-10.

Barnes, P.J. and Adcock, I.M. (1998) Transcriptionfactors and asthma. Eur. Resp. J. 12, 221-234.

Bureau, F., Bonizzi, G., Kirschvink, N., Delhalle, S.,Desmecht, D., Merville, M.P., Bours, V., and Lekeux,P. (2000a) Correlation between nuclear factor-kappaB activity in bronchial brushing samples andlung dysfunction in an animal model of asthma. Am.J. Resp. Crit. Care Med. 161, 1314-1321.

Bureau, F., Delhalle, S., Bonizzi, G., Fievez, L., Dogne,S., Kirschvink, N., Vanderplasschen, A., Merville,M.P., Bours, V. and Lekeux P. (2000b) Mechanismsof persistent NF-kappa B activity in the bronchi of ananimal model of asthma. J. Immun. 165, 5822-5830.

Hart, L.A., Krishnan, V.L., Adcock, I.M., Barnes, P.J. andChung, K.F. (1998) Activation and localization oftranscription factor, nuclear factor-kappaB, in asthma.Am. J. Resp. Crit. Care Med. 158, 1585-1592.

Karin, M. and Ben-Neriah, Y. (2000) Phosphorylationmeets ubiquitination: the control of NF-kappaBactivity. Ann. Rev. Immun. 18, 621-663.

Robinson, N.E., Derksen, F.J., Olszewski, M.A. andBuechner-Maxwell, V.A. (1996) The pathogenesisof chronic obstructive pulmonary disease of horses.Br. vet. J. 152, 283-306.

Siebenlist, U., Franzoso, G. and Brown, K. (1994)Structure, regulation and function of NF-kappaB.Ann. Rev. Cell. Biol. 10, 405-455.

Snapper, J.R., (1986) Large animal models of asthma.Am. Rev. Resp. Dis. 133, 351-352.

Turlej, R.K. Fievez, L., Sandersen, C.F., Dogne, S.,Kirschvink, N., Lekeux, P. and Bureau, F. (2001)Enhanced survival of lung granulocytes in an animalmodel of asthma: evidence for a role of GM-CSFactivated STAT5 signalling pathway. Thorax. 56,696-702.

Whiteside, S.T., Epinat, J.C., Rice, N.R. and Israel, A.(1997) I kappa B epsilon, a novel member of the Ikappa B family, controls RelA and cRel NF-kappaB


47

SESSION 3:

Diagnostic measures of inflammation

Chairman: J. L. Hodgson


48


49

SIGNIFICANCE OF TRACHEAL INFLAMMATION

J. L. Hodgson

University Veterinary Centre Camden, Faculty of Veterinary Science, University of Sydney, New South Wales 2006, Australia

Although collection of tracheal aspirates (TAs) isconsidered a routine procedure for evaluation ofthe respiratory tract in young performance horses,a number of controversial issues remain associatedwith their interpretation. Most notable are therelative numbers of inflammatory cells that arenormally found in TAs, the significance ofincreased amounts of mucus and numbers ofinflammatory cells, especially in relationship totheir effect on performance and the interpretationof bacterial cultivation.

Inflammation in the lung can be regionallylocalised and may be restricted to the largeairways, extend into smaller airways, or involvethe pulmonary parenchyma. Tracheal aspiratescollect secretions, cells and debris that accumulatein the distal trachea and bronchi, but which mayalso be derived from the more distal airways andalveoli. As such, they provide a non-homogenoussample and are not representative of any onesegment of the lung. In contrast, a bronchoalveolarlavage (BAL) samples cells and secretions fromthe distal, smaller airways and commonly from theright caudo-dorsal lung lobes. There is nocorrelation between cytological findings betweenthese 2 methods of sample collection. Therefore, acytological diagnosis of inflammatory airwaydisease (IAD) will be influenced by the method ofsample collection.

A number of factors may influence results ofTA including long distance transportation orexercise prior to sample collection, coughing duringsample collection and the sampling technique.These factors must be taken into considerationwhen evaluating samples and when comparingstudies in which TAs were used to determine thepresence of lower airway inflammation.

Interpretation of the amount of mucus within aTA is best performed in conjunction with

endoscopic evaluations of the lower airways. Thiswill facilitate accurate estimation of the amount ofmucus present in the airways, as opposed to theamount of mucus collected by TA. The amount ofmucus in TAs increases when pulmonary irritationoccurs, such as in cases of IAD. However, thepoint at which mucus accumulation and/or airwayinflammation become clinically significant,especially for different levels of performance, iscurrently contentious. This is particularly the casein horses not exhibiting overt signs of respiratorydisease.

The information derived from TAs mostcommonly used for diagnosis of IAD is thedifferential cell count and, in particular, therelative % of neutrophils. The dilemma withinterpretation of the relative numbers ofneutrophils is to determine what value (if any)represents a significant change. Large variations inthe relative percentage of neutrophils in TAsobtained from apparently healthy horses have beenobserved within and between studies. In addition,poor correlation between the relative number ofneutrophils in TA and pulmonary histopathologyhas been described. As a result of this variation theclinical usefulness of cytological evaluation of TAsamples, especially using a relative percentage ofcells, in the diagnosis of lower airwayinflammation has been questioned. However, instudies of younger, more homogeneouspopulations of horses, smaller variations inneutrophil ratios in normal horses have beenfound. For example, studies investigating airwayinflammation in young racehorses demonstratedthat in 73–80% of clinically normal racehorses,neutrophils do not exceed 20% of theinflammatory cells found in TA samples (Sweeneyet al. 1992; Christley et al. 2001). Furthermore, ina study where 1,235 TA samples were obtained


50

from 724 horses in race training, almost 90% ofhorses sampled had relative percent neutrophilvalues <10%. A strong statistical associationbetween presence of >20% neutrophils in TAspecimens and signs of respiratory disease (iecoughing) in young racehorses and the likelihoodof isolating significant numbers of bacteria hasbeen reported (Newton et al. 1999; Chapman et al.2000; Christley et al. 2001). Therefore, it is likelythat that >20% neutrophils in TAs from youngracehorses is abnormal, representing a significantinflammatory process and a strong risk for clinicalrespiratory disease.

The absolute numbers of cells present (TNCC)should also be taken into account wheninterpreting the significance of the relativeproportions of neutrophils in TA as they indicatemore accurately a shift in cell populations.Unfortunately, the absolute number of cells isfrequently not determined for TAs, reflecting thedifficulties that may be encountered duringenumeration of cells due to the entrapment of cellswithin mucus, the presence of clumps of epithelialcells, and the variable dilution of secretions bysaline during collection. Estimates of thecellularity of samples have been advocated to helpovercome this deficiency and may be incorporatedin an inflammation score to further evaluate shiftsin cell populations.

Other inflammatory cell types may also beevaluated in TAs from young performance horsesto determine the presence of lower airwayinflammation. However, less is known about thesignificance of the relative changes of these celltypes in relation to IAD. Pulmonary alveolarmacrophages (PAM) are the most abundant type ofinflammatory cell present in TAs from normalhorses and their presence is a prerequisite to assessthe adequacy of a TA. Although PAMs are themost common inflammatory cell type in normalhorses, increased numbers are rare in horses withIAD and their significance is unknown.

Lymphocytes are present in low numbers innormal TAs and may be difficult to differentiateaccurately from other cell types present such assmall macrophages, and epithelial cells. Thenumbers of lymphocytes may increase in cases ofrespiratory tract disease, but this is variable and nocorrelation has been made between cytologicalobservations of this cell population in TA andspecific disease processes.

Eosinophils are usually present in very lownumbers in TA from normal horses. Large

increases in the relative numbers of eosinophilsmay be observed in cases of ascarid migration andlung worm infestation, whereas smaller elevations(>1%) are interpreted as evidence of a type Ihypersensitivity response to inhaled allergens.

Mast cells are rare, or present in low numbers,in TAs from normal horses. This is in contrast tosamples obtained by BAL, in which higher numbersof mast cells may be observed. There is littleinformation on alterations in mast cell numbers inTAs. Furthermore, in a study investigating therelative numbers of mast cells in TAs and BALsobtained sequentially from racehorses it was foundthat BALs had significantly higher proportions ofmast cells than TAs (Hughes et al. 2002). Therefore,TA may be less sensitive than BAL for detection ofalterations in relative numbers of mast cells.

Epithelial cells may also be evaluated whenassessing TAs. The epithelial cells present in TAsare predominantly ciliated columnar epithelialcells from the trachea and bronchi, but cuboidalcells from the smaller airways should also beobserved. Squamous epithelial cells should not bepresent in TAs from normal horses and, ifobserved, represent oropharyngeal contamination.Mild changes to epithelial cells may be seen innormal horses including nuclear changes and theloss of some cilia. These changes probablyrepresent normal wear and tear or turnover of cells.Pathological changes to epithelial cells (epithelialatypia) result from inflammation. However, thereare many causes of airway inflammation and thereare no epithelial changes that are specific for aparticular diagnosis or aetiology. The exception isciliocytophthoria, which has been reported inhorses with acute respiratory viral infection.

Results of microbial cultivation assist in theinterpretation of tracheal inflammation, butmisinterpretation is common in cases of IAD. Theisolation of bacteria from TAs may representinfection, a transient lower airway population orcontamination of the TA at the time of sampling. Itis essential for appropriate management of thesecases to differentiate between these scenarios.Clinical signs consistent with bacterialinvolvement may help in this differentiation andinclude fever, anorexia, coughing and nasaldischarge. However, absence of these signs doesnot preclude bacterial infection, especially inmilder cases of IAD.

Cytological evaluation of TAs should precedemicrobial cultivation. Tracheal aspirates fromhorses with bacterial lower respiratory tract


51

infections will have increased mucus, increasedtotal cell counts, and increased relative andabsolute neutrophil counts with possiblydegenerative neutrophils and intra-cellularbacteria. There is no indication for cultivation of asample without this cytological evidence ofinflammation. In addition, it is preferable not toculture samples with large numbers of squamousepithelial cells, even when there are manyneutrophils present, as this is evidence oforopharyngeal contamination.

Quantitative cultures, which determine thenumber of colony forming units for each bacterialspecies, provide additional information. Aspiratescollected in an appropriate fashion from healthyhorses, or from horses with airway inflammationwithout a bacterial aetiology, usually cultivate<103 bacteria (cfu)/ml and, frequently, no bacteriaat all. If >103 cfu/ml are cultivated, it is likely thatthese bacteria are contributing to the diseaseprocess and identification of the bacteria involvedwill assist in interpretation of their significance.Bacteria that are most commonly isolated fromIAD include Streptococcus spp., Pasteurella spp.,Actinobacillus spp and occasionally Bordatellabronchiseptica and Mycoplasma spp. Escherichiacoli, Klebsiella and strictly anaerobic bacteria arerarely isolated from uncomplicated cases of IAD,but may be isolated if antimicrobial therapy hasbeen initiated or pleuropneumonia or lungabscessation has evolved. Pathogenic bacteria thatrarely cause disease in the lower airways, but arecommon contaminants during sampling includecoagulase positive Staphylococcus spp.,Pseudomonas spp. and Proteus spp. Care withinterpretation of these isolates must be made asthey are rarely significant. Isolation of non-pathogenic bacteria (coagulase negativeStaphylococcus spp., Corynebacterium spp.,Bacillus spp.) indicates contamination at the time

of sampling.The significance of changes in cytological

variables of TAs in relation to performance is notknown, and further studies are required to correlatethese alterations with variables that measureperformance. Elevated percentages of neutrophilsare associated with signs of respiratory disease(coughing) and increased numbers of bacteria. Inaddition, elevated percentages of neutrophils are acommon finding in racehorses presented for poorperformance. However, there have not been anystudies which have evaluated directly thesignificance of cytological changes in TAs withany measurable variable of lung function.

REFERENCES

Chapman, P.S., Green, C., Main, J.P.M., Taylor, P.M.,Cunningham, F.M. and Cook, A.J.C. (2000)Retrospective study of the relationships betweenage, inflammation and the isolation of bacteria fromthe lower respiratory tract of thoroughbred horses.Vet. Rec. 146,91-95.

Christley, R.M., Hodgson, D.R., Rose, R.J., Wood,J.L.N., Reid, S.W.J. and Whitear, K.G. (2001) Acase-control study of respiratory disease inThoroughbred racehorses in Sydney, Australia.Equine vet. J. 33, 256-264.

Hughes, K.J., Malikides, N., Dart, A.J., Hodgson, D.R.and Hodgson, J.L. (2002) Comparison of trachealaspirates and bronchoalveolar lavage in racehorses1: evaluation of cytological stains and the relativepercentage of mast cells and eosinophils. Aust. vet.J. Submitted.

Newton, J.R. and Wood, J.L.N. (1999) Summary of acase control study of acute respiratory disease inyoung Thoroughbred racehorses. British EquineVeterinary Association Congress pp 190-191.

Sweeney, C.R., Humber, K.A. and Roby, K.A. (1992)Cytologic findings of tracheobronchial aspiratesfrom 66 Thoroughbred racehorses. Am. J. vet. Res.53, 1172-1175.


52

SIGNIFICANCE OF BRONCHOALVEOLAR CYTOLOGYIN INFLAMMATORY AIRWAY DISEASE OF HORSES

L. Viel

Department of Clinical Studies, University of Guelph, Guelph, Ontario N1G 2W1, Canada

INTRODUCTION

Based on limited, large epidemiological studies ofthe racetrack environment, respiratory conditionsare the second most important factor contributingto poor performance in horses (Bailey et al. 1999).Known causes often associated with reducedperformance are upper airway obstruction,possibly exercise-induced pulmonary hemorrhage,and lower airway inflammation. Young athletichorses suffering from non-infectious inflammatoryairway disease (IAD) often demonstrate a gradualdecline in their level of performance accompaniedby variable clinical signs including cough, nasaldischarge and prolonged recovery post exercise.IAD is currently believed to be distinct from theHeaves syndrome (Robinson 2001), and ischaracterised by the presence of inflammatorycells in the bronchiolar airways, with minimalaccumulation of mucopurulent secretions. Thepresence of inflammatory cells is accompanied byairway hyper-reactivity in response toenvironmental stimuli.

BRONCHOALVEOLAR LAVAGE (BAL)

In human asthma, it was originally believed thatcells collected by sputum samples mirrored theinflammatory cells trafficking the airways.However, Busse and Wenzel (2000) suggested thatinflammatory cells may appear in the airwaylumen in response to certain stimuli but then returnto the subepithelial milieu. The latter wasspeculated because active remodelling observed atthe level of the small airways is often not presentin the larger cartilaginous airways. Therefore, cellscollected from the lower trachea may not be anaccurate reflection of inflammatory activity in the

bronchiolar airways. Collection of cells directlyfrom the peripheral (bronchiolar) airways by BALmay thus provide a more representative sample forthe study of IAD (O’Byrne and Postma 1999).

For the BAL procedure in horses, an endoscope(10–13 mm outer diameter with a length of 1.8–2m or greater) or a specialised naso-tracheal tube isused. The infusion of 60–100 ml pre-warmedlidocaine solution (0.3%) to desensitise the irritantcough receptors in the lower trachea and largebronchi is recommended to ensure animal comfortand reduced stimulation of excessive coughing.Following secure wedging of the endoscope in asub-segmental bronchus, 250–300 ml pre-warmed(37°C) sterile 0.9% saline or phosphate bufferedsaline is instilled into the lung as 1–3 smallerboluses. Each bolus is followed by aspiration of thefluid. The volume of lavage fluid retrieved usuallyranges from 40–60% of the original infusate, butthis volume may decrease with increasing severityof airway inflammation. The gross appearance(colour, turbidity, presence of flocculent debris) ofthe pooled fluid retrieved contributes to the clinicalpathologist’s overall interpretation. Samples ofcollected fluid must be kept on ice if processing forcytology is not possible within 1 h after collection.Routine processing of BAL fluid consists ofdetermining the total number of leucocytesrecovered as well as the differential cell count. Thetotal cell count can be obtained using either aNeubauer, Malassez or Fuchs-Rosenthalhaemocytometer or an automated counting methodsuch as a Coulter counter. In the latter, special careshould be taken particularly when the samplescontain clumps of mucus, which can obstruct theaperture causing substantial underestimation of thetotal cell number. The total cell count is usuallyexpressed as either the total number of cellsrecovered from the lavage yield or as the


53

concentration of cells per millilitre of recoveredlavage fluid. With the known total cell count andthe percent cell differential, the absolute number ofeach cell type recovered per unit volume of fluidcan be quantified. However, the accuracy andreliability of this method requires standardisedvolumes of infusate and lavage fluid recovery. Ifavailable, cytospins can be made to facilitate thedifferential cell count. Following centrifugation ofsamples (600–800 g for 15 min), air-dried smearsshould also be made. Slides can be stained with aroutine cytological stain such as Wright-Giemsafor the cellular differential count.

BAL FLUID CYTOLOGY

Bronchoalveolar lavage (BAL) fluid may display avariety of distinct inflammatory cytologicalprofiles which differ significantly from those ofhealthy horses. One or more inflammatory celltypes may be elevated in the BAL fluid, includingmast cells, globule leucocytes, eosinophils,neutrophils, lymphocytes, macrophages orexfoliated epithelial cells (Table 1). For example,BAL fluid recovered from young horses with poorperformance may show a significant increase in themast cell population with very few eosinophils, ormast cells accompanied by a marked increase ofthe eosinophil population (Hare et al. 1994;Hoffman et al. 1998). Recent studies in such horseshave demonstrated a strong correlation betweenincrease of either eosinophils and/or mast cells in

the BAL fluid and airway hyper-reactivity (Hare etal. 1998; Hoffman et al. 1998). BAL fluid withlarge numbers of non-toxic neutrophils tends tocontain more flocculent debris representing clumpsof mucopus, and occasionally casts of inspissatedmucous plugs from the terminal airways(commonly known as Curschmann’s spirals) areseen (Fogarty and Buckley 1991; Rush Moore et al.1995; Couëtil and Denicola 1999). More recently,an additional feature of BAL fluid of some poorlyperforming young horses is an elevated proportionof lymphocytes with no clinical indication of anactive infectious airway process (Doucet 1994; Vielet al. 2000). Lastly, BAL fluid of young athletichorses with respiratory viral infection shows amassive exfoliation of mucosal columnar epithelialcells accompanied by numerous free cilia anddetached ciliated tufts. Sutton et al. (1998) clearlydemonstrated that this characteristicciliocytophtheria includes broken ciliated platescontaining numerous virus particles, supportingevidence of virally-induced epithelial injury. Insuch horses, the total number of alveolarmacrophages and lymphocytes are significantlyelevated during the active infectious process andare accompanied by a gradual increase in theneutrophil population from Days 3–15 postinfection. This probably corresponds with the onsetof secondary bacterial infection. Further,significant airway hyper-reactivity persists for aslong as 8 weeks post infection in horses with eitherinfluenza or herpesvirus, which may well be a

TABLE 1: BAL cytology of normal horses (italics) and young athletic horses exhibiting signs of poorperformance (typical cytological abnormalities are in bold text)

Age (years) Mac (%) Lymph (%) PMN (%) MC (%) EO (%) Author

2.7–3.5 60–67 28–32 0.5–7 0.2–1 0–1 Hare et al. (1994)(64) (30) (3.4) (0.5) (0.3) Hare and Viel (1998)

Fogarty and Buckley (1991)Rush Moore et al. (1995)

4.3 ± 1.9 64 ± 15 23 ± 11 13 ± 12 0.3 ± 0.7 0.1 ± 0.3 Fogarty and Buckley (1991)3.7 ± 0.3 48 ± 2 36 ± 2 10 ± 1 2 ± 1 4 ± 1 Rush Moore et al. (1995)

55 ± 11 34 ± 11 12 ± 7.7 Couëtil and Denicola (1999)

3.3 ± 0.7 33 ± 10 60 ± 17 4 ± 1 2 ± 0.5 0.2 ± 0.07 Doucet (1994)3.4 ± 1.1 22 ± 10 73 ± 10 1.5 ± 1.3 1 ± 0.8 1.4 ± 3.8 Viel et al. (2000)

3.4 ± 1.6 57 ± 12 36 ± 14 4 ± 3 4 ± 2 0.5 ± 0.3 Hare et al. (1994)8 ± 0.3 46 ± 1 49 ± 2 3 ± 0.3 3 ± 0.4 0.2 ± 0.05 Hoffman et al. (1998)

3.7 ± 0.3 48 ± 2 36 ± 2 10 ± 1 2 ± 1 4 ± 1 Rush Moore et al. (1995)2.6 ± 0.9 59* 26* 0.8* 1.4* 12* Hare and Viel (1998)

Mac = macrophage, Lymph = lymphocyte, PMN = neutrophils, MC = mast cell, EO = eosinophils*Data expressed as median value


54

consequence of the effects of sustained elevationsin neutrophil and mast cell populations.CONCLUSIONS

There is no doubt that recent literature on BALfluid cytology, which has been collected usinggenerally standardised BAL techniques, haspermitted a more reliable comparison of thevarious cytological profiles involved in IAD inyoung athletic horses with poor performance.Specifically, IAD appears to be distinct from thetypical neutrophilic airway inflammation observedin recurrent airway obstruction (heaves). From theexamples provided above, it can be hypothesisedthat the variety of inflammatory cell profiles inIAD may ultimately reflect different aetiologiesand immune responses to airway insult. Therefore,IAD should not be viewed as a distinct entity, butrather as a syndrome resulting from a myriad ofpossible causes.

RECOMMENDATION

• Bronchoalveolar lavage should be adopted asthe method of choice to obtain representativecell populations from the lower airways.

• BAL sampling procedures should continue tobe standardised (endoscope/tubing size,volume of infusate, volume of recovered fluid)in order to permit comparison between studies.

• Control (normal) BAL total cell counts andpercent cell differential counts should notexceed 5% for neutrophils, 2% for mast cellsand less than 1% for eosinophils. Themacrophage:lymphocyte ratio should beapproximately 60:30%.

• Studies on BAL cytology in young athletichorses should report total cell counts, celldifferential (%) counts and absolute celldifferential counts.

• It is paramount that some form of qualitycontrol on cell differential interpretation beestablished to allow reliable comparisonbetween studies.

• Epidemiological longitudinal studies need tobe performed to determine the significance ofthe various cytological profiles on pulmonaryhealth and performance.

• Contribution of inflammation to the structural

remodelling of airways should be investigatedand correlated with BAL cytology.

REFERENCES

Bailey, C.J., Reid, S.W., Hodgson, D.R. and Rose, R.J.(1999) Impact of injuries and disease on a cohort oftwo- and three-year-old thoroughbreds in training.Vet. Rec. 145, 487-493.

Busse, W.W. and Wenzel, S.E. (2000) Large and smallairway dysfunction in asthma: rationale for treatingthe entire airway. In: Asthma and the SmallAirways. Eds: S.P. Peters, S.E. Wenzel. Am. Thor.Soc. cont. med. educ. Monograph Series, pp 17-23.

Couëtil, L.L and Denicola, D.B. (1999) Blood gas,plasma lactate and bronchoalveolar lavage cytologyanalyses in racehorses with respiratory disease.Equine vet. J. 30, 77-82.

Doucet, M.Y. (1994) Clinical Findings, AirwayResponsiveness and Effect of Furosemide in Horseswith Exercise-Induced Pulmonary Hemorrhage(EIPH). DVSc thesis, University of Guelph,Canada. 105,210.

Fogarty, U. and Buckley, T. (1991) Bronchoalveolarlavage findings in horses with exercise intolerance.Equine vet. J. 23, 434-437.

Hare, J.E., Viel, L., O’Byrne, P.M. and Conlon, P.D.(1994) Effect of sodium cromoglycate on lightracehorses with elevated metachromatic cellnumbers on bronchoalveolar lavage and reducedexercise tolerance. J. vet. Pharmac. Ther. 17, 237-244.

Hare, J.E. and Viel, L. (1998) Pulmonary eosinophiliaassociated with increased airway responsiveness inyoung racing horses. J. vet. int. Med. 12, 163-170.

Hoffman, A.M., Mazan, R.M. and Ellenberg, B.S.(1998) Association between bronchoalveolar lavagecytologic features and airway reactivity in horseswith a history of exercise intolerance. Am. J. vet.Res. 59, 176-181.

O’Byrne, P.M. and Postma, D.S. (1999) The many facesof airway inflammation: asthma and chronicobstructive pulmonary disease. Am. J. resp. crit.Care Med. 159, 41-66.

Robinson, N.E. (2001) International workshop onequine chronic airway disease, Michigan StateUniversity, June 2000. Equine vet. J. 33, 5-19.

Rush Moore, B., Krakowka, S., Robertson, J.T. andCummins, J.M. (1995) Cytologic evaluation ofbronchoalveolar lavage fluid obtained fromStandardbred racehorses with inflammatory airwaydisease. Am. J. vet. Res. 56, 562-567.

Sutton, G.A., Viel, L., Carman, P.S. and Boag, B.L.(1998) Pathogenesis and clinical signs of equineherpesvirus-1 in experimentally infected ponies invivo. Can. J. vet. Res. 62, 49-55.

Viel, L., Hewson, J., Parsons, D., Staempfli, H., Kenney,D. and Baird, J. (2000) Tracheal aspirates and


55

CYTOLOGY OF INFLAMMATORY AIRWAY DISEASE

K. C. Smith, J. R. Newton, S. M. Gower, S. M. Cade, D. J. Marlin, C. M. Deaton and J. L. N. Wood

Animal Health Trust, Lanwades Park, Kentford, Newmarket, Suffolk CB8 7UU, UK

When inflammatory airway disease (IAD) isdefined as described by Wood et al. (2003), thecondition is the most important form of respiratorydisease in young racehorses in training (Burrell etal. 1996; Wood 1999). The condition affects25–30% of horses in training in the UnitedKingdom and United States of America.Epidemiological studies indicate a strongassociation with infection by Streptococcuszooepidemicus, Streptococcus pneumoniae andActinobacillus/Pasteurella spp. (Wood et al. 1993;Wood 1999). The incidence and prevalence of IADdecline with age, suggesting the development ofacquired immunity to infection (Chapman et al.2000; Newton et al. 2003).

Endoscopic tracheal wash (TW) is the mainmethod used by the authors for diagnosis of IAD,and is used routinely by racehorse practitioners inNewmarket and Lambourn. The method is lessinvasive than trans-tracheal aspiration, allowsvisualisation of the tracheal lumen, and isgenerally well tolerated by young horses intraining (Whitwell and Greet 1984). EndoscopicTW samples can also be used for quantitativebacteriology, and the fact that a large proportion ofendoscopically recovered samples are sterilesuggests that contamination of samples from theupper respiratory tract is not a common problem(Wood et al. 1993). TW samples should becollected after exercise, and the horse should notbe fed prior to endoscopy. It is essential to ensurethat the endoscope and catheter are sterile, and tominimise the procedure time. Once the endoscopeis positioned in the trachea approximately 10 cmfrom the carina, and the appearance of the mucosaand quantity of mucopus have been recorded, thetechnique involves infusion of 30 ml normal orphosphate buffered saline, which is thenwithdrawn. The tracheal aspirate is mixed

thoroughly then aliquoted into an unfixed samplefor quantitative bacteriology and a fixed samplefor cytology (10% neutral buffered formalin).

Normal TW fluid contains a predominance ofnon-degenerate ciliated and non-ciliated epithelialcells and macrophages. Mucus is scanty and notinspissated. Neutrophils generally comprise lessthan 20% of the nucleated cell population (Beech1975). The principal endoscopic and cytologicalfindings in IAD are the presence of visible trachealmucopus on endoscopy, and the presence ofneutrophilic inflammation on TW cytology.Degenerate neutrophils generally predominate inbacterial infection, which accounts for a highproportion of IAD cases in young horses, whereasnon-degenerate neutrophils usually predominate inhypersensitivity disorders of the recurrent airwayobstruction (RAO) group seen in older horses. It iscritical that tracheal wash cytology, with particularreference to recognition of airway neutrophilia, iscorrelated with microbiology to distinguishpathogenic from commensal or contaminatingmicro-organisms. Toxic degenerative changes andphagocytosed bacteria can often be found inneutrophils in cases of bacterial infection, andmacrophages may also contain phagocytosedmicro-organisms in some cases. Microbiologicaland serological investigations may reveal evidenceof infection with other pathogens, such as virusesor Mycoplasma spp., in a small minority of IADcases (Wood 1999; Wood et al. 2003).

While neutrophils are the most importantinflammatory cell type encountered in IAD, othernon-epithelial cell types that may be found in TWfluid include macrophages, eosinophils,lymphocytes, mast cells and plasma cells (Beech1975; Whitwell and Greet 1984). Macrophages arethe predominant non-epithelial cell type in normalTW samples, and may be resting or activated. The


56

nature of any phagocytosed material is importantin establishing a diagnosis of infection (bacterialor fungal infection in particular), and in assessingthe grade and chronicity of airway haemorrhage.Eosinophils are rare in normal airways, but mayoccur in small to moderate numbers in TWsamples collected from horses withhypersensitivity reactions or endoparasitism. Thecells may be intact or degranulated. Studies inCanada using bronchoalveolar lavage (BAL)samples from young horses suggest thatpulmonary eosinophilia and airway hyper-responsiveness represent early subclinical eventsin progression to RAO (Hare and Viel 1998), andindicate the importance of long-term monitoringof young racehorses with TW eosinophilia.

Lymphocytes may increase in number in viralairway disease, with concurrent epithelial celldegeneration (Freeman et al. 1993) but viralinfection that has not become complicated bybacterial infection is rarely sampled diagnostically.

Mast cells increase in number in cases of type 1hypersensitivity, but are more easily visualised inequine BAL than TW samples, where correctsample fixation and processing is critical inpreserving intact mast cell granules. Plasma cellsare recovered occasionally from TW samples ofhorses with chronic airway inflammation. Moreusual cytological changes associated withincreasing duration of IAD are a progressiveincrease in the ratio of macrophages toneutrophils; appearance of multinucleatedmacrophages; development of mucous casts(Curschmann’s spirals); and hyperplasia ordysplasia of epithelial cells (Whitwell and Greet1984).

TW scoring systems should permit: 1) rapidscreening of large batches of samples; 2)assessment of response to therapy; and 3) patternrecognition (Freeman et al. 1993). A 3-pointscoring system was introduced in Newmarket inthe 1980s (Whitwell and Greet 1984). More

0 1 2 3 4 5 6 7 8 9

40

35

30

25

20

15

10

5

0

Inflammation score

Num

bers

of

obse

rvat

ions

Fig 1a: Application of 9-point tracheal wash scoringsystem to horses previously scored as 3 on 3-point scale.

0 1 2 3 4 5 6 7 8 9

60

50

40

30

20

10

0

Num

bers

of

obse

rvat

ions

Inflammation score

Fig 1b: Application of 9-point tracheal wash scoringsystem to horses previously scored as 2 on 3-point scale.

0 1 2 3 4 5 6 7 8 9

Inflammation score

250

200

150

100

50

0

Num

bers

of

obse

rvat

ions

Fig 1c: Application of 9-point tracheal wash scoringsystem to horses previously scored as 1 on 3-point scale.

500

450

400

350

300

250

200

150

100

50

0

Num

bers

of

obse

rvat

ions

0 1 2 3 4 5 6 7 8 9

Inflammation score

Fig 1d: Application of 9-point tracheal wash scoringsystem to horses previously scored as 0 on 3-point scale.

recently, this scoring system has been refined intoa 9-point score (authors’ unpublished data). The3-point score is derived as follows: moderate orprofuse tracheal mucopus 1 point; total nucleatedcell count over 1,000 cells/mm3 1 point; andneutrophils predominant or present in moderatenumbers 1 point. TW 9-point scoring involves: 1)assessment of tracheal mucopus: score 0 (absent)to 3 (profuse); 2) assessment of smear celldensity: score 0 (low) to 3 (high); and 3)assessment of neutrophil proportion: score 0(absent or occasional) to 3 (predominant). Usingthe 9-point score, scores of 0–2 are interpreted asnormal, or within acceptable limits. Scores of 3–9are interpreted as indicating inflammatory diseaseof increasing severity.

The 3-point scoring system is simple androbust, but may underestimate mild inflammation.The 9-point scoring system is more sensitive andparticularly useful in diagnosing mildinflammation or assessing changes in degree ofinflammation over time. Figure 1 illustrates theincreased sensitivity of the 9-point over the 3-point score, through application of both scoringsystems to a large series of TWs collectedprospectively from young Thoroughbredracehorses in 7 training yards in Newmarket,Lambourn and Epsom that were sampledregularly between 1993 and 1996, generating atotal of 1,689 samples. Both TW scoring systemsdescribed in this abstract concentrate on changesin mucus quantity and neutrophil proportions. The9-point score correlates well with total andspecies-specific bacterial counts (Fig 2, Newton etal. 2003), and with other measures of airwayinflammation such as hydrogen peroxide in breath

condensate (C. Deaton, unpublished data; Marlinet al. 2003).

BAL is not used routinely in youngThoroughbreds in Newmarket, so most large scalescreening by this group has been based on TW.Studies in Australia have established parametersfor normal BAL cytology in Thoroughbred horsesin training (aged 1–7 years of age) with particularreference to diagnosis and grading of execise-induced pulmonary haemorrhage (McKane et al.1993). Prospective collection of paired TW andBAL samples from young racehorses is likely tobe valuable in establishing the relative utility of the2 techniques in diagnosis and prognosis of IAD.Lung biopsy is rarely used in racehorse practice,but may provide additional information onpathology and prognosis (Savage et al. 1998).Previously published post mortem data suggest anincomplete correlation between tracheobronchialcytology and pulmonary histopathology (Larsonand Busch 1985), although detailed post mortemstudies of the respiratory system of young horsesare rare (Blunden and Gower 1999).

CONCLUSION

1) IAD is characterised cytologically by moderateto marked airway neutrophilia; 2) use of a 9-pointscoring system incorporating assessment oftracheal mucopus, smear cell density andproportion of neutrophils permits IAD to begraded, and response to therapy evaluated; and 3)concurrent evaluation of TW cytology andquantitative bacteriology indicates a strongassociation with bacterial infection in younghorses.


57

0 1 2 3

Mea

n lo

g to

tal b

acte

ria (

cfu/

ml)

Inflammation score

Fig 2a: Correlation of total bacterial count with 3-pointtracheal wash score.

Fig 2b: Correlation of total bacterial count with 9-pointtracheal wash score.

�

�

�

�

7

6

5

4

3

2

1

00 1 2 3 4 5 6 7 8 9

Mea

n lo

g to

tal b

acte

ria (

cfu/

ml)

Inflammation score

��

� �

7

6

5

4

3

2

1

0

��

�

�

�

�


58

REFERENCES

Beech, J. (1975) Cytology of tracheobronchial aspiratesin horses. Vet. Path. 12, 157-164.

Blunden, A.S. and Gower, S.M. (1999) A histologicaland immunohistochemical study of the humoralimmune system of the lungs in young Thoroughbredhorses. J. comp. Path. 120,347-356.

Burrell, M.H., Wood, J.L., Whitwell, K.E., Chanter, N.,Mackintosh, M.E. and Mumford, J.A. (1996)Respiratory disease in thoroughbred horses intraining: the relationships between disease andviruses, bacteria and environment. Vet. Rec. 139,308-313.

Chapman, P.S., Green, C., Main, J.P.M., Taylor, P.M.,Cunningham, F.M., Cook, A.J.C. and Marr, C.M.(2000) Retrospective study of the relationshipsbetween age, inflammation and the isolation ofbacteria from the lower respiratory tract ofThoroughbred horses. Vet. Rec. 146, 91-95.

Freeman K.P., Roszel J.F., McClure J.M., Mannsman R.,Patton P.E. and Naile S. (1993) A review ofcytological specimens from horses with and withoutclinical signs of respiratory disease. Equine vet. J.25, 523-526.

Hare, J.E. and Viel, L. (1998) Pulmonary eosinophiliaassociated with increased airway responsiveness inyoung racing horses. J. vet. Int. Med. 12, 163-170.

Larson, V.L. and Busch, R.H. (1985) Equinetracheobronchial lavage: comparison of lavagecytologic and pulmonary histopathologic findings.Am. J. vet. Res. 46, 144-146.

Marlin, D.J., Deaton, C.M., Newton, C.M. and Smith,K.C. (2003) Breath condensate measures of airwayinflammation. Workshop on ‘Inflammatory AirwayDisease: Defining the Syndrome’. Havemeyer

Foundation Monograph Series No 9, Eds: A.Hoffman, N. E. Robinson and J.F. Wade, R&WPublications (Newmarket) Ltd, pp 62-65.

McKane, S.A., Canfield, P.J. and Rose, R.J. (1993)Equine bronchoalveolar lavage cytology: survey ofthoroughbred racehorses in training. Aus. vet. J. 70,401-404.

Newton, J.R., Wood, J.L.N., Smith, K.C., Marlin, D.J.and Chanter, N. (2003) Aetiological agents: bacteria.Workshop on ‘Inflammatory Airway Disease:Defining the Syndrome’. Havemeyer FoundationMonograph Series No 9, Eds: A. Hoffman, N. E.Robinson and J.F. Wade, R&W Publications(Newmarket) Ltd, pp 40-44.

Savage, C.J., Traub-Dargatz, J.L. and Mumford, E.L.(1998) Survey of the large animal diplomates of theAmerican College of Veterinary Internal Medicineregarding percutaneous lung biopsy in the horse. J.vet. Int. Med. 12, 456-464.

Whitwell, K.E. and Greet, T.R.C. (1984) Collection andevaluation of tracheobronchial washes in the horse.Equine vet. J. 16, 499-508.

Wood, J.L.N., Burrell, M.H., Roberts, C.A., Chanter, N.and Shaw, Y. (1993) Streptococci and Pasteurellaspp. associated with disease of the equine lowerrespiratory tract. Equine vet. J. 25, 314-318.


Wood, J.L.N., Newton, J.R., Smith, K.C. and Marlin,D.J. (2003) Aetiological agents: viruses and IAD.Workshop on ‘Inflammatory Airway Disease:Defining the Syndrome’. Havemeyer FoundationMonograph Series No 9, Eds: A. Hoffman, N. E.Robinson and J.F. Wade, R&W Publications(Newmarket) Ltd, pp 33-36.


59

QUANTIFYING AND CHARACTERISING MUCUS IN THE AIRWAYS

V. Gerber, A. M. Jefcoat, J. A. Hotchkiss* M. King† and N. E. Robinson

Pulmonary Laboratory, Department of Large Animal Clinical Sciences, Michigan State University,East Lansing, Michigan 48824, USA; *Toxicology, Environmental Research and Consulting, The Dow Chemical Company, Midland, Michigan 48674, USA; †Pulmonary Research Group,University of Alberta, Edmonton, Alberta, Canada

Mucus accumulation is a hallmark of recurrentairway obstruction (RAO) and inflammatoryairway disease (IAD; Burrel 1985; MacNamara etal. 1990; Moore 1996). Secretions mayaccumulate due to increased production/secretionand/or decreased clearance, but little is known ofthe mucus apparatus in IAD. This paper reviewsmethods of quantifying and characterising mucus,describes findings in the horse and proposesquestions for investigation. The terms ‘mucus(accumulation)’ and ‘(airway) secretions’ are usedinterchangeably to describe gel-like secretions andtheir constituents in airways.

HOW CAN WE QUANTIFY GROSS MUCUSACCUMULATION ?

In one study, the amount of secretions aspiratedtranstracheally (without prior infusion of saline)was measured directly. Four control horses had no,or <1 ml, aspirable mucus and 18 of 22 horses withvarying degrees of lower airway disease had 1– >5ml of tracheal secretions (Schatzmann et al. 1972).

Various systems have been used to grade theamount of secretions in the trachea endoscopically.Although ‘no mucus or a few single mucus droplets’in the trachea can be regarded as normal, largevariations are observed in asymptomatic horses.Horses with IAD usually have mild to intermediateaccumulations of mucus in their trachea. Theamount varies and can overlap with observations inboth asymptomatic and RAO-affected horses(Dieckmann 1987; Gerber et al. 2001; V. Gerber,unpublished results). In a large clinical study, Dixonet al. (1995) demonstrated the non-specific nature ofmucus accumulation, which increased in allcategories of lower airway disease. They also notedthat mucus scores were significantly higher incoughing than non-coughing horses, an associationfound in other populations (Hodgson et al. 2003;Robinson et al. 2003). Interestingly, mucus

accumulation is only weakly associated with airwayinflammation assessed by bronchoalveolar lavage(BAL) fluid, does not increase with stabling inasymptomatic horses (but BAL fluid neutrophils do)and is not increased in older compared with youngerasymptomatic horses (Gerber et al. 2001). Increasedaccumulation of mucus (not correlated withexercise-induced pulmonary haemorrhage) isassociated with poor performance in racehorses(MacNamara et al. 1990). Sport horses withincreased mucus, however, may perform well indressage and show-jumping (Gerber et al. 2001). Itis unclear what intensity and duration of exercisemay increase endoscopically visible mucusaccumulation and for how long (Luft 1987).

It is difficult to compare studies because ofinherent variability and subjectivity (eg a horsemay cough and clear its trachea from mucus, and areport of ‘a few’ vs ‘many’ mucous flakes maydepend on the observer). Therefore, the authorshave developed a standardised grading system andare testing its repeatability both within anindividual horse and between observers. However,as it is possible only to assess the amount andappearance of secretions in the large airways, thelimitations of endoscopic observation requiresother, complementary measures of mucus.

HOW CAN WE CHARACTERISE ANDQUANTIFY THE COMPONENTS OF MUCUS ?

Mucus is a physiologically heterogeneous mix ofwater (~95%), electrolytes (1%), lipids (1%),proteins (2–3%) and variable amounts of epithelialand inflammatory cells. Cells may contributedirectly to mucus volume and their contributioncould be estimated by methods assessing airwayinflammation. However, mucins (high molecularweight glycoproteins) constitute the network of themucus gel and can greatly influence the totalvolume by drawing and retaining water in the gel.


60

The major gel-forming mucins expressed inhuman and rodent airways are MUC2, MUC5ACand MUC5B. They contain non-repetitive cystein-rich regions, believed to allow intra- andintermolecular cross-linking through disulphidebond formation. MUC5AC is mainly produced ingoblet cells, whereas submucosal glands secreteMUC5B (in 2 distinct glycoforms). Together theyform the bulk of the ‘normal’ human mucus-gel.MUC2 is predominantly an intestinal mucin.

Gerber et al. (2002) identified 2 equinehomologues of gel-forming mucin genes anddemonstrated by RT-PCR that MUC5AC, but notMUC2, is expressed in horse airways, andMUC5AC mRNA up-regulation is a potentialprimary mechanism for mucus hypersecretion inRAO. These findings are complemented by thosebased on reactivity with antibodies raised againsthuman MUC5B and MUC5AC, demonstratingequine versions of these mucins in respiratory tractsecretions (Walley et al. 2001) and MUC5AC ingoblet cells (V. Gerber, unpublished data).

Jefcoat et al. (2001), to investigate carbohydrateside chains of mucins, used carbohydrate-specificenzyme-linked lectin assays (ELLA). Of the oligo-saccharides tested, α-1, 2 fucose discriminates bestbetween control and RAO-affected horses (even inremission), is associated with mucus cells and can bemeasured in BAL fluid by ELLA using Ulexeuropaeus agglutinin I. Therefore, it may be possibleto use α-1, 2 fucose levels as an indirect measure ofmucin accumulation in BAL fluid. α-1, 2 fucose isassociated with secreted mucins produced by airwaygoblet cells. Sialyl Lewis X (a tetrasaccharidestructure associated with membrane bound mucins)has been identified in equine BAL fluid and mayimpart specific binding activity toward structuressuch as bacterial adhesin molecules.

Mucus is mainly produced by goblet cells, sincesubmucosal glands are rare in horse airways (Kaupet al. 1990). Goblet cell hyper- and metaplasia is ahistopathological characteristic of RAO (Kaup et al.1990) and one of the first changes in chronic smallairway disease. Surprisingly, initial morphometricmeasurements of the volume of stored mucins in theepithelium showed no difference between controland heaves-affected animals (Hotchkiss 1998).These contradictory findings highlight the problemof defining control groups. However, they may pointto functional, rather than structural, changes inmucus apparatus. Because the orientation of theairway epithelium must be maintained,morphometric studies require large biopsy or postmortem samples.

HOW TO INVESTIGATE PHYSICALPROPERTIES, RHEOLOGY, CLEARABILITYAND CLEARANCE OF MUCUS ?

Studies on healthy horses have shown thatmucociliary clearance rate is influenced more byphysical properties of mucus than by variations ofciliary beat frequency. Based on physical appearanceonly, Schatzmann et al. (1972) proposed that mucusviscosity is proportional to severity of clinical signsin horses with lower airway disease, and Pietra et al.(2000) linked mucus viscosity with the proportion ofneutrophils in tracheobronchial secretions. Gerberet al. (2000) used magnetic micro-rheometry onsamples obtained by bronchial brushing to measurerheology of mucus from airways of animals with noobvious mucus accumulation. Viscoelasticity wasnormal in RAO-horses in remission, but duringexacerbation increased 3-fold compared withhealthy controls. Predicted mucociliary and coughclearability in RAO-affected horses were reducedsignificantly. Using the same method, horses withIAD showed surprisingly low average viscoelasticity(V. Gerber, unpublished data). Based on theseincomplete data, as no direct comparison of IAD vs.healthy or RAO horses has been performed, it isproposed that mucus clearance could be normal oreven improved in IAD.

It is important to note that other physicalproperties of mucus, such as spinnability andadhesivity, also influence its clearability. Thephospholipid content and composition of mucusaffect its adhesivity. In addition, surfactantincreases mucus transport velocity, is present inamounts sufficient to significantly reduce surfacetension in the trachea of healthy horses, but may bedeficient in RAO, and after long transports.

Finally, mucociliary clearance rates can bemeasured by endoscopic and scintigraphic markermethods. Decreased clearance in RAO has beenreported (Coombs and Webbon 1987; Turgut andSasse 1989), but others found no differencebetween healthy and RAO horses (Willoughby etal. 1991). Although no studies have investigatedmucociliary clearance rate in IAD it is interestingthat viral infections (influenza and herpes, but notrhinovirus) can compromise mucociliary clearance.

CONCLUSIONS AND QUESTIONS

Currently, few of the available tools have been usedto investigate mucus in IAD. The amount ofsecretion in the trachea increases in some horses


61

with IAD, but this is a non-specific finding. Theauthors propose that the functional significance ofmild to moderate mucus accumulations (and thesyndrome of IAD) depends on the degree ofperformance expected from the horse. This alsopoints to the difficulty of defining IAD and controlgroups of ‘healthy’ horses. Another question to beaddressed is the association of large vs. smallairway inflammation with mucus accumulation. Toaddress such questions, a standardised endoscopicscoring system and complementary tools such asmucin markers in BAL fluid are needed.Measurements of mucin production, storage andsecretion are available to investigate functional vs.structural changes of the mucus apparatus leadingto the cause(s) of increased mucus accumulation inIAD. Decreased clearability seems less likely toplay a major role, because IAD affected horsesappear to have favourable mucus rheology.However, viscoelasticity remains to be directlycompared to normal horses, and complemented byother measures of mucus clearability.

REFERENCES

Burrel, M.H. (1985) Endoscopic and virologicalobservations on a group of young racehorses intraining. Equine vet. J. 17, 99-103.

Coombs, S.L. and Webbon, P.M. (1987) Observations ontracheal mucociliary clearance in horses. Tierarztl.Prax. Suppl. 2, 5-9.

Dieckmann, M.P. (1987) Zur Wirksamkeit vonAmbroxolhydrochlorid (Mukovent) beinlungenkranken Pferden-klinische, Funktionelle undzytologische Untersuchungen. Dissertation,Tieraerztliche Hochschule Hannover, Germany.

Dixon, P.M., Railton, D.I. and McGorum, B.C. (1995)Equine pulmonary disease: a case control study of300 referred cases. Part 2: Details of animals andhistorical and clinical findings. Equine vet. J. 27,422-427.

Gerber, V., King, M., Schneider, D.A. and Robinson, N.E.(2000) Tracheobronchial mucus viscoelasticity duringenvironmental challenge in horses with recurrentairway obstruction. Equine vet. J. 32, 411-417.

Gerber, V., Robinson, N.E., Rawson, J., Venta, P.J.,Jefcoat, A.M. and Hotchkiss, J.A. (2002) Mucingenes in horse airways: MUC5AC but not MUC2may play a role in recurrent airway obstruction.Equine vet. J. In press.

Gerber, V., Straub, R., Schott, H.C. and Robinson, N.E.(2001) Is mild airway inflammation and mucusaccumulation really clinically significant? SecondHavemeyer Symposium on Allergic Diseases of theHorse, Hungary, R & W Publications (Newmarket).pp 10 (abstr).

Hodgson, D.R., Christley, R., Wood, J.L.N., Reid, S.W.J.and Hodgson, J.L. (2003) Coughing and airwayinflammation. Workshop on ‘Inflammatory AirwayDisease: Defining the Syndrome’. HavemeyerFoundation Monograph Series No 9, Eds: A.Hoffman, N. E. Robinson and J.F. Wade, R&WPublications (Newmarket) Ltd, pp 16-18.

Hotchkiss, J.A. (1998) Airway mucous synthesis,storage, and secretion. Proc. World equine AirwaysSymposium, Guelph, Canada.

Jefcoat, A.M., Hotchkiss, J.A., Harkema, J.R., Basbaum,C.B. and Robinson, N. E. (2001) Persistent mucinglycoprotein alterations in equine recurrent airwayobstruction. Am. J. Phys: Lung, Cell Molec. Phys.281, 704-712.

Kaup, F.-J., Drommer, W., Damsch, S. and Deegen, E.(1990) Ultrastructural findings in horses withchronic obstructive pulmonary disease (COPD) II:pathomorphological changes of the terminal airwaysand the alveolar region. Equine vet. J. 22, 349-355.

Luft, J. (1987) Der Einfluss koerperlicher Belastung aufdie Zytologie des Tracheobronchialsekretes beimPferd [the effect of exercise of the composition oftracheobronchial secretions in the horse.Dissertation, Tieraerztliche Hochschule Hannover,Germany.

MacNamara, B., Bauer, S. and Iafe, J. (1990) Endoscopicevaluation of exercised-induced pulmonaryhemorrhage and chronic obstructive pulmonarydisease in association with poor performance inracing standardbreds. J. Am. vet. med. Ass. 196, 443-445.

Moore, B.R. (1996) Lower respiratory tract disease. In:Exercise intolerance. Vet. Clin. N. Am. equine Pract.12, 457-472.

Pietra, M., Guglielmini, C., Forni, M. and Cinotti, S.(2000) In vitro mucolytic activity of recombinanthuman deoxyribonuclease on equinetracheobronchial mucus. Vet. Rec. 147, 627-629.

Robinson, N.E., Berney, C., DeFeijter-Rupp, H., Jefcoat,A.M., Cornelisse, C., Gerber, V. and Derksen, F.J.(2003) Mucus, cough, airway obstruction andinflammation. Workshop on ‘Inflammatory AirwayDisease: Defining the Syndrome’. HavemeyerFoundation Monograph Series No 9, Eds: A.Hoffman, N. E. Robinson and J.F. Wade, R&WPublications (Newmarket) Ltd, pp 13-15.

Schatzmann, U., Straub, R. and Gerber, H. (1972)Bronchialsekretaspiration beim Pferd (Transtrachealaspiration in the horse). Schweizer Archiv fuerTierheilkunde 114, 395-403.

Turgut, K. and Sasse, H.H. (1989) Influence ofclenbuterol on mucociliary transport in healthyhorses and horses with chronic obstructivepulmonary disease. Vet. Rec. 125,526-530.

Walley, E.A., Thornton, D.J., Corfield, A.P., Carrington,S.D. and Sheehan, J.K. (2001) Characterisation ofequine respiratory tract mucins. World equineAirways Symp. Edinburgh, WEAS.

Willoughby, R.A., Ecker, G.L., McKee, S.L. andRiddolls, L.J. (1991) Use of scintigraphy for thedetermination of mucociliary clearance rates in


62

BREATH CONDENSATE MEASURES OF AIRWAYINFLAMMATION

D. J. Marlin, C. M. Deaton, J. R. Newton* and K. C. Smith*

Centres for Equine Studies and *Preventive Medicine, Animal Health Trust, Lanwades Park, Kentford,Newmarket, Suffolk CB8 7UU, UK

There are a wide range of techniques andapproaches for the diagnosis and investigation ofpulmonary disease, including auscultation,radiography, scintigraphy, conventional pulmonaryfunction testing and exercise tolerance. In humanmedicine and research, endoscopy of the respiratorytract is considered a highly invasive procedure andis only undertaken in circumstances where there is aclear and specific indication. In contrast, in clinicalveterinary medicine, endoscopy of the respiratorytract and, in particular, the trachea with collection oftracheal wash, is probably the most commondiagnostic approach now used. Bronchoalveolarlavage (BAL) has also been used extensively inresearch studies into the causes and mechanisms ofpulmonary disease in horses such as recurrentairway obstruction (RAO) and exercise-inducedpulmonary haemorrhage (EIPH).

Endoscopy of the respiratory tract has manyadvantages over other techniques. It allows directvisualisation of the larger airways and specificsampling of different regions of the respiratorytract (eg tracheal wash, BAL, transbronchialbiopsy, bronchial brushing). These samples maythen be used for a wide range of ex vivo analyses,including cytological examination, investigation ofgene expression, biochemical and immunologicalassay and bacterial culture. The fact that samplingis regionally specific can be both an advantage anda disadvantage, depending on the specific diseaseprocess or research question. For example, in thecase of diffuse lung conditions such as RAO, asingle BAL sample from the dorso-caudal lungmay be representative of the lung as a whole.However, a positive bacterial culture in a trachealwash may occur in parallel with negative culture inBAL, depending on the precise region of lungaffected in a focal infection. BAL must also beconducted under sedation and the use of local

anaesthesia of the airways is desirable to reducecoughing. This precludes the use of BAL close tocompetition in many racing or competition horses.BAL also causes changes locally including atransient inflammatory response, which lasts up toaround 48 h (Sweeney et al. 1994) and changes inairway lung fluid biochemistry (Deaton et al.,unpublished data). Whilst it is possible to sampledifferent regions of the lung in a sequential order,there is no clear evidence that the lung respondshomogeneously to all challenges or insults, or thatthe lung is always homogenous at the start of anexperiment.

The fact that endoscopy of the upper airwaymay be harmful in man led to the development ofless invasive techniques to monitor airwayinflammation, such as sputum induction (Jones1968), which has only gained significant popularityas a diagnostic technique in human respiratorymedicine in the last 10–15 years. For example, in2002, there were over 200 papers publishedrelating to the use of this technique. Sputuminduction does result in a mixed sample whichincludes saliva and secretions and cells from alllevels of the airway. Furthermore, whilst sputuminduction has been shown recently to be bothdiagnostically useful and safe in adults (Vlachos-Mayer et al. 2000), there is evidence that it may beassociated with a greater incidence of adverseeffects in children (Jones et al. 2001) and isunsuitable for use in neonates and infants.

More recently, even less invasive techniqueshave been applied to investigate pulmonaryinflammation. The technique of analysing gases (egethane, pentane, carbon monoxide, nitric oxide) orsubstances dissolved in moisture within the breath(exhaled breath condensate or EBC) such ashydrogen peroxide has become widely used. Thisapproach offers the advantage of being totally non-


63

invasive, does not require administration of drugs(eg sedative, local anaesthetic) and is repeatablemore often than techniques such as endoscopy.Analysis of exhaled gases and mediators in EBCalso has the advantage that measurements can berepeated on a frequent basis to follow the timecourse of inflammation, for example, following aninhaled allergen challenge. It has the disadvantageor advantage, depending on the precisecircumstances, of being global rather than regional.In humans, contamination of exhaled breath bynasal or oral secretions or gases has beenproblematic (Griese et al. 2002).

Commercial systems for collecting exhaledbreath condensate from human subjects have beendeveloped (eg Ecoscreen from Jaeger). However,custom systems have also been developed. In allsystems described, the principle involves directingexhaled breath only by use of a mask, mouthpieceor nosepiece and valve system through a cooledcollecting vessel. As the warm and saturated

exhaled breath is cooled, its capacity to hold watervapour is reduced and droplets of EBC form on thesurface of the vessel.

The interest in analysis of markers ofinflammation in EBC for diagnosing andmonitoring airway inflammation in humanmedicine and in animal studies has increasedmarkedly in recent years (Fig 1).

The presence of a wide range of substanceshas been reported in EBC of man and differentanimals, including hydrogen ions, nitrite (NO2),nitrate (NO3), NH4

+, H2O2, electrolytes, 3-nitrotyrosine, S-nitrosothiols, LTB4, LTC4, LTD4,LTE4, LTF4, cys-LTs, PGE4, 8-isoprostane, IL-6,IL-4, IL-8, IL-10, IFN, HGF, MDA, TBARS andarachidonic acid. Of these different mediators, themost commonly measured is hydrogen peroxideand the rest of this review will, therefore,concentrate on the measurement of H2O2 in EBC.

There is clear evidence that EBC H2O2 iselevated in human pulmonary inflammatoryconditions such as asthma, adult respiratory distresssyndrome (ARDS) and human chronic obstructivepulmonary disease (COPD; Table 1). In addition,EBC H2O2 has been shown to be increasedfollowing cigarette smoking (Nowak et al. 1996),during acute upper and lower respiratory tractinfections (Dohlman et al. 1993) and inbronchiectasis (Loukides et al. 1998), cystic fibrosis(Jobsis et al. 2000) and sepsis (Sznajder et al. 1989).

As noted above, the measurement of mediatorsin EBC has been contentious. There have beenmarked differences of opinion as to the source ofmany substances present in EBC. It is clear that theformation of breath condensate from evaporationof water from the lung surface can be

Fig 1: Number of publications on exhaled breathcondensate by year.

40

30

20

10

0Num

ber

of p

aper

s

1996 1997 1998 1999 2000 2001 2002

Year

TABLE 1: Examples of hydrogen peroxide concentration in exhaled breath condensate (EBC) ofcontrols and patients with asthma, COPD and adult respiratory distress syndrome (ARDS)

Author Condition Control (µmol/l) Disease (µmol/l) Ratio

Jobsis et al. (1997) asthma 0.15 0.60 4Antczak et al. (1997) asthma 0.01 0.26 26Antczak et al. (1999) asthma 0.01 0.18 18Emelyanov et al. (2001) asthma 0.02 0.13 7Loukides et al. (2002) asthma 0.20 0.95 5Dekhuijzen et al. (1996) COPD (human) 0.03 0.21 7Dekhuijzen et al. (1996) COPD (human) 0.03 0.60 20Nowak et al. (1998) COPD (human) 0.05 0.55 11Nowak et al. (1999) COPD (human) 0.04 0.48 12De Benedetto et al. (2000) COPD (human) 0.12 0.50 4Baldwin et al. (1986) ARDS 0.34 1.68 5Sznajder et al. (1989) ARDS 0.72 2.34 3Wilson et al. (1992) ARDS 0.01 0.55 55Kietzmann et al. (1993) ARDS 0.25 1.50 6


64

‘contaminated’ with larger droplets of airwaysecretions that are aerosolised during breathing.The type of collection system employeddetermines to what extent contamination of thefinal ‘EBC’ sample occurs. For example, incollection systems with a short conducting tubegoing directly into the condensing and collectingchamber and valves distal to the collecting vessel,aerosol contamination is likely to occur.

To study the time course of inflammation inexperimental studies a system for collecting EBCfrom conscious, unsedated horses at rest has beendeveloped and described previously (Deaton et al.2001). The system consists of a facemask and non-rebreathing valve. Expired breath only is directedalong a 250 cm long, 5 cm diameter heated tube intoa 5-litre u-shaped stainless steel condensing tubewhich is held in a reservoir of ice and water at ~4°C.After the horse breathes through the system foraround 10 min, ~4 ml of EBC can be recoveredfrom the steel collecting tube. The authors have alsosuccessfully applied a modified EBC collectiontechnique for use in conscious, unsedated cats.

A technique involving sedation and intubationof the upper airway and trachea has also beendescribed (Fey et al. 2001) LTB4 and bradykininwere measured in airway secretions collectedusing this technique. EBC has also been collectedfrom calves (Reinhold et al. 2000).

Hydrogen peroxide, hydrogen ions, nitrite,nitrate and electrolytes are all detectable in theexhaled breath condensate of horses. Healthy,normal horses at pasture have EBC H2O2 in therange of 0.1–0.8 µmol/l EBC. The coefficient of

variation for 3 sequential repeated measurements isless than 7%. The technique appears to be sensitiveand can detect an increase in inflammation in stabledhorses compared to horses maintained at pasture. InRAO affected horses with marked inflammation butwithout clinical signs of dyspnoea, EBC H2O2 isincreased to around 2–3 µmol/l. In the human studieson asthma, COPD and ARDS cited in Table 1, themean increase in the patients over the controlsranged from 4–55-fold. However, the medianincrease for all studies was 7-fold, which is similarto the increase seen in RAO horses. High ratios inhuman studies appear to be due most commonly tolow H2O2 concentrations in EBC of controls ratherthan excessively high concentrations in patients.

EBC H2O2 concentration is closely correlatedto BAL neutrophil numbers in the horse (Deaton etal. unpublished data; Fig 2) and is also reducedfollowing treatment of non-infectious airwayinflammation with inhaled corticosteroids (Fig 3).

The concentration of H2O2 in EBC is alsoincreased in naturally occurring airwayinflammation (allergic or bacterial). In acutechallenges in experimental situations with eitherozone or allergen, there is an increase in EBCH2O2 within 1–2 h of exposure, which does notoccur following nebulisation of saline.

In conclusion, measurements of mediatorssuch as H2O2 in EBC are not a replacement forendoscopy of the respiratory tract, but allowmonitoring on a more frequent basis withoutcausing disturbance to the airways and aid in theidentification of appropriate sampling times withmore invasive procedures such as BAL.

0 10 20 30 40 50

Bronchoalveolar lavage neutrophils (/µl)

3

2

1

0

Fig 2: EBC H2O2 concentration as a function of BALneutrophil count in healthy control horses and in horseswith mild to moderate airway inflammation. Dotted linesrepresent 95% confidence interval. Deaton et al.unpublished data.

R=0.74, P<0.00013.0

2.5

2.0

1.5

1.0

0.5

0

H2O

2µm

ol/l

Pre 7days Tx End Tx +7 days post

Treatment period

Fig 3: Mean concentrations of hydrogen peroxide(H2O2) in expired breath condensate (EBC) prior totreatment in horses with inflammation (pre), after 7 (7days Tx) and 14 days of treatment (End Tx) with inhaledfluticasone propionate (3.5 µg/kg SID) and 7 days afterthe termination of treatment (+7 days post). Columnswith different letters are significantly different (P<0.05).Marlin et al. (2002).

H2O

2µm

ol/l

a

a

b b


65

REFERENCES

Antczak, A., Nowak, D., Bialasiewicz, P. and Kasielski, M.(1999) Hydrogen peroxide in expired air condensatecorrelates positively with early steps of peripheralneutrophil activation in asthmatic patients. Arch.Immunol. Ther. Exp. (Warsz) 47, 119-126.

Antczak, A., Nowak, D., Shariati, B., Krol, M., Piasecka,G. and Kurmanowska, Z. (1997) Increased hydrogenperoxide and thiobarbituric acid-reactive products inexpired breath condensate of asthmatic patients. Eur.Resp. J. 10, 1235-1241.

Baldwin, S.R., Simon, R.H., Grum, C.M., Ketai, L.H.,Boxer, L.A. and Devall, L.J. (1986) Oxidant activity inexpired breath of patients with adult respiratorydistress syndrome. Lancet 1, 11-14.

Deaton, C.M., Marlin, D.J., Roberts, C.A., Smith, N.,Harris, P.A. and Schroter, R.C. (2001) Monitoringairway inflammation by expired breath condensate.Proc. World equine Airways Symposium, Edinburgh,p 15.

De Benedetto, F., Aceto, A., Dragani, B., Spacone, A.,Formisano, S., Cocco, R. and Sanguinetti, C.M.(2000) Validation of a new technique to assess exhaledhydrogen peroxide: results from normals and COPDpatients. Monaldi. Arch. Chest. Dis. 55, 185-188.

Dekhuijzen, P.N., Aben, K.K., Dekker, I., Aarts, L.P.,Wielders, P.L., van Herwaarden, C.L. and Bast, A.(1996) Increased exhalation of hydrogen peroxide inpatients with stable and unstable chronic obstructivepulmonary disease. Am. J. Resp. Crit. Care Med. 154,813-816.

Dohlman, A.W., Black, H.R. and Royall, J.A. (1993)Expired breath hydrogen peroxide is a marker of acuteairway inflammation in pediatric patients with asthma.Am. Rev. Resp. Dis. 148, 955-960.

Emelyanov, A., Fedoseev, G., Abulimity, A., Rudinski, K.,Fedoulov, A., Karabanov, A. and Barnes, P.J. (2001)Elevated concentrations of exhaled hydrogen peroxidein asthmatic patients. Chest 120, 1136-1139.

Fey, K., Schack, S. and Sasse, H.H.L. (2001) LeukotrieneB4 and bradykinin in breath condensate of horses withrecurrent airway obstruction. Proc. 19th Symp. Vet.Comp. Resp. Soc. Edinburgh, July 19th-23rd, 2001,p18.

Griese, M., Noss, J. and Bredow Cv, C. (2002) Proteinpattern of exhaled breath condensate and saliva.Proteomics 2, 690-696.

Jobsis, Q., Raatgeep, H.C., Hermans, P.W. and de Jongste,J.C. (1997) Hydrogen peroxide in exhaled air isincreased in stable asthmatic children. Eur. Resp. J. 10,519-521.

Jobsis, Q., Raatgeep, H.C., Schellekens, S.L., Kroesbergen,A., Hop, W.C. and de Jongste, J.C. (2000) Hydrogenperoxide and nitric oxide in exhaled air of childrenwith cystic fibrosis during antibiotic treatment. Eur.Resp. J. 16, 95-100.

Jones, F.L., Jr. (1968) Aerosol induction of sputum. Avaluable aid in pulmonary diagnosis. Postgrad Med.43, 100-104.

Jones, P.D., Hankin, R., Simpson, J., Gibson, P.G. andHenry, R.L. (2001) The tolerability, safety, and success

of sputum induction and combined hypertonic salinechallenge in children. Am. J. Resp. Crit. Care Med.164, 1146-1149.

Kietzmann, D., Kahl, R., Muller, M., Burchardi, H. andKettler, D. (1993) Hydrogen peroxide in expiredbreath condensate of patients with acute respiratoryfailure and with ARDS. Int. Care Med. 19, 78-81.

Loukides, S., Bouros, D., Papatheodorou, G., Panagou, P.and Siafakas, N.M. (2002) The relationships amonghydrogen peroxide in expired breath condensate,airway inflammation, and asthma severity. Chest 121,338-346.

Loukides, S., Horvath, I., Wodehouse, T., Cole, P. J. andBarnes, P. J. (1998) Elevated levels of expired breathhydrogen peroxide in bronchiectasis. Am. J. Respir.Crit. Care Med. 158, 991-994.

Marlin, D. J., Deaton, C. M., Smith, N., Newton, R. andSmith, K. (2002) Treatment of airway inflammationreduces expired breath condensate H2O2 and smallairway but not tracheal inflammation. Proc. 20thSymp. Vet. Comparative Resp. Soc., Boston, October4th-6th, 2002, p 80.

Nowak, D., Antczak, A., Krol, M., Pietras, T., Shariati, B.,Bialasiewicz, P., Jeczkowski, K. and Kula, P. (1996)Increased content of hydrogen peroxide in the expiredbreath of cigarette smokers. Eur. Resp. J. 9, 652-657.

Nowak, D., Kasielski, M., Antczak, A., Pietras, T. andBialasiewicz, P. (1999) Increased content ofthiobarbituric acid-reactive substances and hydrogenperoxide in the expired breath condensate of patientswith stable chronic obstructive pulmonary disease: nosignificant effect of cigarette smoking. Resp. Med. 93,389-396.

Nowak, D., Kasielski, M., Pietras, T., Bialasiewicz, P. andAntczak, A. (1998) Cigarette smoking does notincrease hydrogen peroxide levels in expired breathcondensate of patients with stable COPD. Monaldi.Arch. Chest Dis. 53, 268-273.

Reinhold, P., Becher, G. and Rothe, M. (2000) Evaluationof the measurement of leukotriene B4 concentrationsin exhaled condensate as a noninvasive method forassessing mediators of inflammation in the lungs ofcalves. Am. J. vet. Res. 61, 742-749.

Sweeney, C.R., Rossier, Y., Ziemer, E.L. and Lindborg,S.R. (1994) Effect of prior lavage on bronchoalveolarlavage fluid cell population of lavaged and unlavagedlung segments in horses. Am. J. vet. Res. 55, 1501-1504.

Sznajder, J.I., Fraiman, A., Hall, J.B., Sanders, W.,Schmidt, G., Crawford, G., Nahum, A., Factor, P. andWood, L.D. (1989) Increased hydrogen peroxide in theexpired breath of patients with acute hypoxemicrespiratory failure. Chest 96, 606-612.

Vlachos-Mayer, H., Leigh, R., Sharon, R.F., Hussack, P.and Hargreave, F.E. (2000) Success and safety ofsputum induction in the clinical setting. Eur. Resp. J.16, 997-1000.

Wilson, W.C., Swetland, J.F., Benumof, J.L., Laborde, P.and Taylor, R. (1992) General anesthesia and exhaledbreath hydrogen peroxide. Anesthesiol. 76, 703-710.

66



67

SESSION 4:

Functional significance

Chairman: F. J. Derksen


68


69

LUNG FUNCTION FOR BEGINNERS

F. J. Derksen

College of Veterinary Medicine, Michigan State University, Veterinary Medical Centre, East Lansing,Michigan 48824-1314, USA

The primary function of the lung is gas exchange.Air moves in and out of the lung with a respiratoryfrequency and tidal volume appropriate for theanimal’s level of activity. This results in an arterialoxygen tension of about 100 mmHg and an arterialcarbon dioxide tension of 40 mmHg. Tests ofpulmonary function can be divided in tests of gasexchange and tests of pulmonary mechanics (Leffand Schumacher 1993).

Blood gas tensions reflect the gas exchangeefficiency of the lung. In the diseased lung,ventilation and perfusion are poorly matched, aportion of the cardiac output is shunted through thelung, and gas diffusion is impaired. All thesefactors result in hypoxaemia. However, arterialblood gas tensions are also affected by factorsunrelated to the gas exchange efficiency of thelung. These factors include hypoventilation andinhalation of low oxygen tension gases, such asoccurs at higher altitudes. Therefore, investigatorsoften report alveolar-arterial oxygen differences(Wagner and Gillespie 1989; Nyman and Bjork1999). The difference between the alveolar andarterial oxygen tensions truly reflects the gasexchange efficiency of the lung.

For efficient gas exchange to occur, blood flowto the lung and ventilation must be carefullymatched. Tests of ventilation-perfusion matching,such as the multiple inert gas eliminationtechnique (Nyman and Bjork 1999) and lungscintigraphy (Votion and Ghafir 1999), areparticularly useful because with these tests themechanisms of hypoxaemia can be determined.

Ventilation is achieved when respiratorymuscles generate pressures in the thoracic cavitythat result in expansion or contraction of the lung.When the lung is diseased, so that airways arenarrowed or the lungs are stiffer, more pressure isneeded to ventilate the system. Accordingly,

measurement of intra-thoracic pressure is a simpletest of lung mechanics. Intra-thoracic pressure isbest approximated by pleural pressure (Derksenand Robinson 1980).

The pressure that moves air in and out of therespiratory system is used to overcome resistanceto air flow and to stretch the lung. In a normalstanding horse, respiratory system resistanceresides primarily in the upper airways. Indeed,approximately 50% of the total respiratory systemresistance resides in the nose, and another 30% inthe larynx and trachea. Only 20% of the totalrespiratory system resistance resides within thelung (Art and Serteyn 1988). Inside the lung, themajority of the resistance to flow occurs in thetrachea and large bronchi. The small airwaysprovide little resistance to air flow. Therefore, testsof resistance are sensitive to upper and majorlower airway obstruction, but insensitive to smallairway obstruction.

Small airway obstruction causes redistributionof ventilation and hypoxaemia. Therefore, smallairway obstruction can be affected by tests of gasexchange and by tests of pulmonary mechanicsthat are sensitive to redistribution of ventilation,such as dynamic compliance. Another test ofpulmonary mechanics, forced expiratory flow-volume curves, is also sensitive to small airwayobstruction. Forced expiratory flow volume curvescan be generated in standing horses (Couëtil andRosenthal 2001). In horses with recurrent airwayobstruction, flows are reduced, especially duringthe last portion of the forced vital capacity.

Normal airways contract in response toirritants. When airways are inflamed, this normalresponse is exaggerated, ie the airways are hyper-responsive. Airway hyper-responsiveness is acharacteristic feature of human asthma, recurrentairway obstruction and inflammatory airway


70

disease of horses (Derksen and Robinson 1985;Hoffman and Mazan 1998; Sterk 2002). It can bedetected by measuring airway function as horsesbreathe increasing concentrations of irritants suchas histamine.

The ultimate effect of airway obstruction ispoor oxygen exchange, which leads tohypoxaemia. Normal horses exercising vigorouslybecome hypoxaemic and hypercapnic. Exercise-induced hypoxaemia occurs because bloodtraverses the lung so rapidly that there isinsufficient time for complete gas exchange(Wagner and Gillespie 1989). Airway obstruction,is likely to exaggerate exercise-inducedhypoxaemia, by causing decreased minuteventilation and mismatching of ventilation andperfusion.

REFERENCES

Art, T. and Serteyn, D. (1988) Effect of exercise on thepartitioning of equine respiratory resistance. Equinevet. J. 20, 268-273.

Couëtil, L.L. and Rosenthal, F.S. (2001) Clinical signs,evaluation of bronchoalveolar lavage fluid, andassessment of pulmonary function in horses with

inflammatory respiratory disease. Am. J. vet. Res.62, 538-546.

Derksen, F.J. and Robinson, N.E. (1980) Esophageal andintrapleural pressures in the healthy conscious pony.Am. J. vet. Res. 41, 1756-1761.

Derksen, F.J. and Robinson, N.E. (1985) Airwayreactivity in ponies with recurrent airwayobstruction (heaves). J. appl. Physiol. 58, 598-604.

Hoffman, A.M. and Mazan, M.R. (1998) Associationbetween bronchoalveolar lavage cytologic featuresand airway reactivity in horses with a history ofexercise intolerance. Am. J. vet. Res. 59, 176-181.

Leff, A.R. and Schumacher, P.T. (1993) RespiratoryPhysiology Basics and Applications. Philadelphia,W.B. Saunders.

Nyman, G. and Bjork, M. (1999) Gas exchange duringexercise in standardbred trotters with mildbronchiolitis. Equine vet. J. 30, 96-101.

Sterk, P.J. (2002) Airway hyperresponsiveness: usingbronchial challenge tests in research andmanagement of asthma. J. aerosol Med. 15, 123-129.

Votion, D. and Ghafir, Y. (1999) Analysis ofscintigraphical lung images before and aftertreatment of horses suffering from chronicpulmonary disease. Vet. Rec. 144, 232-236.

Wagner, P.D., and Gillespie, J.R. (1989) Mechanism ofexercise-induced hypoxemia in horses. J. appl.Physiol. 66, 1227-1233.


71

AIRWAY OBSTRUCTION AND HYPER-REACTIVITY IN HORSES WITH SIGNS OF INFLAMMATORY AIRWAY DISEASE

A. Hoffman and M. Mazan

Lung Function Testing Laboratory, Tufts University School of Veterinary Medicine, 200 Westboro Road,North Grafton, Massachusetts 01536, USA

FUNCTIONAL DISTURBANCES OFINFLAMMATORY AIRWAY DISEASE (IAD)

The last decade has seen exciting newdevelopments in the understanding of early lungdisease in the horse. Investigators of IAD havedisclosed the clinical, endoscopic, radiographic,and cytological changes associated with thisloosely defined syndrome (Mazan 2002).However, the evidence that horses with the signs ofIAD, possess a functional disturbance has onlyrecently surfaced. One hint that IAD is associatedwith lung dysfunction came from Klein andDeegen (1986) and later Doucet and Vrins (1991),who demonstrated greater airway reactivity inhorses exhibiting mild respiratory signs, withoutheaves. Before these studies were performed, itwas assumed that only horses in crisis from heavesexhibit airway obstruction and airway hyper-reactivity (Derksen and Robinson 1985;Armstrong and Derksen 1986; Hoffman 2002a).Now it is clear that either: 1) highly sensitivemethods of lung mechanical measurements; or 2)provocation tests are necessary to detect thefunctional disturbances associated with thesyndrome of IAD.

FORCED EXPIRATORY MANOEUVRES FORTHE DETECTION OF FLOW LIMITATION INIAD

Forced manoeuvres was first employed in thehorse by Gillespie (1974) and later refined for usein standing, sedated horses by Couëtil andRosenthal (2001). Couëtil showed that horses witha history of IAD (chronic cough, exerciseintolerance) had significantly lower values forforced expiratory flows (at 75–95% exhaled vital

capacity) compared to controls and to horses withheaves in remission. In the same horses, they wereable to see higher pulmonary resistance byconventional methods (oesophageal pressure-flowdata), confirming that horses with IAD hadsignificant flow limitation. Accompanying thisfunctional data, was bronchoalveolar lavage(BAL) cytology showing higher neutrophil % andtotal cell count (TCC) in the IAD group(polymorphonuclear [PMN] = 20. 4 ± 9.3%, TCC= 535 ± 187 cells/µl) vs controls (PMN = 6.8 ±2.7%, TCC = 321 ± 100 cells/µl), although thesewere not significantly different due to the smallnumbers of horses with IAD (n=5). Forcedexpiratory manoeuvres causes an effort-independent constriction of small airways bysudden imposition of a gradient between intra-bronchial (negative) and alveolar (less negative)pressure (Couëtil and Rosenthal 2000). Otherstudies from Couëtil and Denicola (1999) haveshown that horses similarly defined as IAD haveexercise intolerance, so there is a clear associationbetween flow limitation, airway inflammation, andexercise intolerance.

OSCILLOMETRY FOR THE DETECTION OFAIRWAY OBSTRUCTION PATTERN ANDAIRWAY REACTIVITY

Oscillometry is the measurement of the pressure-flow relationship (impedance) derived from imposedoscillatory perturbations of the respiratory system,rather than measurement of endogenous pressure-flow data. The magnitude and phase relationshipbetween pressure and flow dictate impedance.Oscillometry reveals how airway obstruction isdistributed anatomically in the lung (peripheral vscentral, homogeneous vs. heterogeneous peripheraldistribution) and serves to gauge its severity


72

(Lutchen and Hantos 1988; Young and Tesarowski1994; Lutchen and Suki 1996; Lutchen and Gillis1997). Using this technique, low-grade airwayobstruction was demonstrated in horses with signs ofIAD with concurrent abnormal BAL cytology, withvalues for resistance of the respiratory systemintermediate between controls and horses with ahistory of heaves (Hoffman 1999).

For the purpose of the IAD workshop, theauthors re-examined oscillometry andbronchoalveolar lavage (BAL) cytologic data fromoutpatients, submitted to the laboratory for testingbetween 1997–2001 (Table 1). Horses (n=107) werecategorised according to BAL cytology as follows:control, n=23 (normal BAL cytology), IAD, n=55(>10% PMN, >2% mast cells, or >0.5%eosinophils), or heaves (history of clinical heaves,but currently in remission). The cutpoint for PMNthat distinguished IAD from controls, was increasedfrom >5 to >10%. This was done in light of previouswork (Hoffman and Mazan 1998; Hoffman 1999)

that failed to show a relationship between PMN(<25%) and airway obstruction or airway reactivity,and with regard to previous observations that PMNincreases even in normal horses following exposureto mouldy hay (Tremblay and Ferland 1993). Theauthors hypothesised that using this new system forgrouping horses, there would be significantdifferences between IAD and controls with regard torespiratory system resistance at various oscillatoryfrequencies (as before).

The following variables were comparedbetween control, IAD, and heaves (duringremission): age, sex distribution, resistance -- RRS(from 1–5 Hz), reactance -- XRS (1–5Hz), slope ofRRS between 1–3 Hz assuming a 3rd orderpolynomial, PC100RRS (provocative concentrationof histamine that doubled RRS at 1 Hz),log10PC100RRS, percentage of BAL cells classifiedas macrophages, lymphocytes, PMN, mast cells,and eosinophils by counting 500 cells stained withWright-Giemsa (1,000 xmag).

TABLE 1: Clinical signs, pulmonary function data and bronchoalveolar lavage cytology from controlhorses with IAD or RAO (in remission)

Variable Control (n=23) IAD (n=55)* RAO (n=29)†

Age 5.6 ± 0.68 6.3 ± 0.65 16.3 ± 1.0 a,b

Age range (2–14 years) (2–24 years) (5–29 years)Sex distribution (Mare,Gld,Stln) 47.8, 21.7, 30.4 49, 34, 17 41.4, 58.6, 0 a,b

OscillometryRRS

1 Hz (cm H2O/l/s) 0.581 ± 0.047 0.719 ± 0.054 0.962 ± 0.103 a,b

2 Hz (cm H2O/l/s) 0.503 ± 0.033 0.602 ± 0.039 0.773 ± 0.069 a,b

3 Hz (cm H2O/l/s) 0.524 ± 0.035 0.600 ± 0.036 0.744 ± 0.058 a,b

4 Hz (cm H2O/l/s) 0.570 ± 0.040 0.622 ± 0.034 0.698 ± 0.053 a5 Hz (cm H2O/l/s) 0.685 ± 0.056 0.650 ± 0.045 0.763 ± 0.099 a,b

XRS1 Hz (l/cm H2O) -0.330 ± 0.021 -0.369 ± 0.019 -0.466 ± 0.079 a,b

2 Hz (l/cm H2O) -0.067 ± 0.014 -0.092 ± 0.016 -0.227 ± 0.061 a,b

3 Hz (l/cm H2O) -0.101 ± 0.015 0.071 ± 0.016 0.074 ± 0.0574 Hz (l/cm H2O) 0.228 ± 0.015 0.199 ± 0.015 0.097 ± 0.0395 Hz (l/cm H2O) 0.240 ± 0.30 0.223 ± 0.036 0.073 ± 0.056

Resonant frequency (Hz) 2.77 ± 0.1 2.84 ± 0.08 3.67 ± 0.29 a,b

Slope RRS from 1–3 Hz 0.098 ± 0.018 0.163 ± 0.022 0.236 ± 0.081 aAirway reactivityPC100RRS (mg/ml histamine) 11.31 ± 1.84 3.622 ± 0.462 c 3.21 ± 0.497 clog10PC100RRS (mg/ml histamine) 0.942 ± 0.08 0.419 ± 0.050 c 0.377 ± 0.078 cBAL

Mac (%) 46.4 ± 2.6 44.1 ± 1.5 28.2 ± 2.5 cLymph (%) 49.3 ± 2.6 46.6 ± 1.4 36.0 ± 3.3 cNeutrophils (%) 2.91 ± 0.4 5.24 ± 0.81 33.1 ± 5.1 cMast cells (%) 1.37 ± 0.15 3.39 ± 0.27 c,d 2.12 ± 0.28Eosinophils (%) 0.05 ± 0.02 0.47 ± 0.13 a 0.22 ± 0.09

*IAD defined as clinical history of cough and/or exercise intolerance/decline in performance, not heaves andabnormal BAL cytology (>10% PMN, >2% mast cells, >0.5% eosinophils). †In remission from ‘heaves’ (RAO)as defined clinically by Robinson (2000). One Way ANOVA; significant difference (P<0.05) from acontrol and/orbIAD group; significantly different (P<0.001) from ccontrol or dheaves. Values expressed as mean ± SEM.


73

The following observations were made. Therewere no significant age differences between IADand control horses, but horses with a history ofheaves were significantly (P<0.005) older. The agerange of controls was narrower than for IAD orheaves, with some horses exhibiting only cough orexercise intolerance and abnormal BAL (ie IAD)as old as 24 years, but as young as 2 years old.Stallions were missing from the heaves-in-remission group, but more prevalent in the controland IAD group that contained racehorses. Whethersome of the horses in the RAO group raced at anearlier time, was not disclosed by the study.

The RRS of the IAD group was higher (between1–4 Hz) but only a trend (P=0.055) was found. Thisdiffers from a previous study (Hoffman 1999)where significant (P<0.01) increases in RRS wereobserved in horses with IAD vs controls. Thedifference lies in the re-categorisation of IAD(>10% PMN vs >5% in earlier studies). Hence, theexact criterion used to group horses clearly affectsthe outcome. The heaves group had significantlygreater RRS across frequencies, and XRS wassimilarly elevated, a feature not observed in IAD.Hence, airway obstruction (increased RRS) andairway wall shunting or increased dynamicelastance (increased XRS) are features of heaves(Young and Tesarowski 1997), but many horseswith IAD do not exhibit these abnormalities atbaseline if measured with oscillometry. Resonantfrequency, the frequency where all impedance isdue to resistance, was shifted to a significantlygreater value in horses in remission from heaves.This is further evidence that the degree of airwayobstruction is significantly more severe in heaveshorses in remission vs controls, or even horses withIAD. This data is supported by Couëtil andRosenthal (2001).

Frequency dependence of RRS was quantifiedas the slope of the 3rd order polynomial (ie fitting

a parabola) that modelled RRS from 1–5 Hz. Avalue above zero denoted frequency dependence,and a value of zero a lack of frequency dependence,which is considered normal from 1–5 Hz. Horsespresented in remission from heaves but presentedfor coughing and/or exercise intolerance, hadsignificantly greater frequency dependence. Valuesfor IAD were intermediate between controls andheaves. Frequency dependence denotes either: 1)heterogeneity of ventilation (time constants) in thelung; and/or 2) generalised increase in peripheralairway disease. This supports the notion that smallairways were preferentially affected in IAD andheaves, although large airways were alsoobstructed in some cases.

The most important difference in lung functiontests between IAD and controls, was airwayreactivity to histamine. The dose of histamine thatdoubled RRS (1 Hz) was significantly (P<0.001)lower in IAD and in the heaves group. Thisexpresses a fundamental difference between IADand controls with respect to the tendency ofairways to constrict. Either histamine evokedairway constriction of greater magnitude (eg dueto pre-existing airway wall inflammation andmucus blockage), or the pattern of constrictionevoked by histamine was more heterogeneous(Lutchen and Jensen 2001). The histamineresponse results in greater frequency dependenceas well as a pattern of increased RRS at allfrequencies (Mazan and Hoffman 1999),indicating that small and large airways areconstricting simultaneously. Peripheral airwaysconstrict to a greater extent than central airways inresponse to histamine, hence oscillometry coupledwith provocation can give insight into the relativeresponse of the small vs large airways.

With the importance of airway reactivity to thepathophysiology of IAD, efforts have beenfocused on the development of a non-invasive fieldtest of non-specific airway reactivity (Hoffman etal. 2001). This system will allow the practitionerto screen horses with suspected IAD on the basisof airway reactivity, and permit large scale testingin order to study risk factors that lead to thiscondition (Hoffman 2002b).

Finally, a subset of horses presented to TuftsUniversity (n=10) with chronic cough and found tohave airway hyper-reactivity (PC100RRS mean2.08 mg/ml, range 0.57–5.03 mg/ml) and elevatedBAL mast cell % (>2%) were retested after acourse of treatment with oral prednisone (1 mg/kgBID 7 days, 0.8 mg/kg BID 7 days, 0.6 mg/kg

12

8

4

0PC

100R

RS

(mg/

ml h

ista

min

e)

1 2

Pre- vs post corticosteroid

Fig 1: Pre- and post corticosteroid treatment (30 days).


74

BID 7 days and 0.4 mg/kg BID 7 days) withoutchange in their environment or feeding practices.Figure 1 shows the change in PC100RRS for thesehorses. Whether it was an inadvertent change ofseasons, environment, or whether it was the effectof the drug prednisone, these horses had asignificant (P=0.002) increase in PC100RRS from2.08–5.23 mg/ml, or just over one log2 dose ofhistamine. However, only 7 out of 10 horsesshowed improvement. This finding is interpretedas an indication that airway hyper-reactivity is inpart an indication of an inflammatory process, butthere are other factors that cause airway hyper-reactivity, which may be mechanical in nature. Thedata also support the notion that horses with IADexhibiting cough, are presented with sub-optimallung function. Re-testing makes it possible toeventually disclose a horse’s best lung function.

In conclusion, these data delineate a clearpattern of airway hyper-reactivity in horsesreferred to our hospital with persistent signs ofIAD, such as cough, exercise intolerance, andmucus in the airways. In contrast to previousstudies (Hoffman 1999), a change in groupingcriterion for IAD (from >5% to >10% neutrophilson BAL) resulted in a loss of statisticalsignificance for airway obstruction. This suggeststhat neutrophilic inflammation criteria are relevantto airway obstruction, and a lower cut-point for‘normal’ (<5%) is a useful refinement. In thefuture, field and referral programs for diagnosingIAD should include pulmonary function tests inconjunction with BAL or other cytologicalmethods. After a suitable period of treatment,pulmonary function tests can be used to gauge theprogress of individuals in the context of theirclinical and performance trends.

REFERENCES

Armstrong, P.J. and Derksen, F.J. (1986) Airway responsesto aerosolised methacholine and citric acid in ponieswith recurrent airway obstruction (heaves). Am. Rev.Resp. Dis. 133, 357-361.

Couëtil, L.L. and Denicola, D.B. (1999) Blood gas, plasmalactate and bronchoalveolar lavage cytology analysesin racehorses with respiratory disease. Equine vet. J.30, 77-82.

Couëtil, L.L. and Rosenthal, F.S. (2000) Forced expiration:a test for airflow obstruction in horses. J. appl. Physiol.88, 1870-1879.

Couëtil, L.L. and Rosenthal, F.S. (2001) Clinical signs,evaluation of bronchoalveolar lavage fluid, andassessment of pulmonary function in horses withinflammatory respiratory disease. Am. J. vet. Res. 62,538-546.

Derksen, F.J. and Robinson, N.E. (1985) Airway reactivityin ponies with recurrent airway obstruction (heaves). J.appl. Physiol. 58, 598-604.

Doucet, M.Y. and Vrins, A.A. (1991) Histamine inhalationchallenge in normal horses and in horses with smallairway disease. Can. J. vet. Res. 55, 285-293.

Gillespie, J.R. (1974) The role of the respiratory systemduring exertion. J. S. Africa vet. Ass. 45, 305-308.

Hoffman, A.M., and Mazan, M.R. (1998) Associationbetween bronchoalveolar lavage cytologic featuresand airway reactivity in horses with a history ofexercise intolerance. Am. J. vet. Res. 59, 176-181.

Hoffman, A. (1999) Programme of lung function testinghorses with suspected small airway disease. Equinevet. Educ. 11, 322-328.

Hoffman A., Riedelberger K., Kupcinskas R. and MiscovicM. (2001) Flow-metric comparison of respiratoryinductance plethysmography and pneumotachographyin horses. J. appl. Physiol. 91, 2767-2775.

Hoffman, A. (2002a) Newest Diagnostic Methods forInflammatory Airway Disease (IAD). Am. Ass. equinePract. Proc. 48, 208-217.

Hoffman, A. (2002b) The prescience to measure airwayreactivity in horses without heaves. Pferd. 18, 1-3.

Klein, H.J. and Deegen, E. (1986) Histamine inhalationprovocation test: method to identify nonspecificairway reactivity in equids. Am. J. vet. Res. 47, 1796-1800.

Lutchen, K.R. and Gillis, H. (1997) Relationship betweenheterogeneous changes in airway morphometry andlung resistance and elastance. J. appl. Physiol. 83,1192-2001.

Lutchen, K. R. and Hantos, Z. (1988) Importance of low-frequency impedance data for reliably quantifyingparallel inhomogeneities of respiratory mechanics.IEEE Trans Biomed Eng 35, 472-481.

Lutchen, K.R. and Jensen, A. (2001) Airway constrictionpattern is a central component of asthma severity: therole of deep inspirations. Am. J. Resp. Crit. Care Med.164, 207-215.

Lutchen, K. and Suki, B. (1996) Understanding pulmonarymechanics using the forced oscillations technique:emphasis on breathing frequencies. Bioengineeringapproaches to pulmonary physiology and medicine.Khoo. NY, Plenum Press 227-253.

Mazan, M.R. (2002) Inflammatory airway disease- currentknowledge. Proc. 20th Am Coll. vet. Int. Med., Texas,USA pp 707-709.

Mazan, M.R. and Hoffman, A.M. (1999) Comparison offorced oscillation with the conventional method forhistamine bronchoprovocation testing in horses. Am. J.vet. Res. 60, 174-180.

Robinson, N.E. (2000) Report of an InternationalWorkshop on Chronic Airway Disease. Equine vet. J.33, 5-19.

Tremblay, G.M. and Ferland, C. (1993) Effect of stablingon bronchoalveolar cells obtained from normal andCOPD horses. Equine vet. J. 25, 194-197.

Young, S.S. and Tesarowski, D. (1994) Respiratorymechanics of horses measured by conventional andforced oscillation techniques. J. appl. Physiol. 76,2467-2472.

Young, S.S. and Tesarowski, D. (1997) Frequencydependence of forced oscillatory respiratorymechanics in horses with heaves. J. appl. Physiol. 82,983-987.


75

INFLAMMATORY AIRWAY DISEASE AND GASEXCHANGE

G. Nyman

Department of Large Animal Clinical Sciences, Faculty of Veterinary Medicine, Swedish University ofAgricultural Sciences, Uppsala, Sweden

Impairment of pulmonary gas exchange andoxygenation of the arterial blood accompanyrespiratory disease in horses. The commoncontributors to alveolar-arterial oxygen tensiondifferences are alveolar hypoventilation, shunt,uneven distribution of alveolar ventilation andperfusion (VA/Q) and/or alveolar-capillarydiffusion limitation of oxygen. A possibility tostudy pulmonary gas exchange and particularly theventilation-perfusion distribution in more detail isto use the multiple inert gas elimination technique(MIGET), developed by Wagner et al. (1974a).From the results of using this technique it wassuggested that an uneven distribution of VA/Qratios was the mechanism of hypoxaemia inhumans with asthma or chronic obstructivepulmonary disease (Wagner et al. 1977). As thetechnique has been used for evaluation ofanaesthesia-induced hypoxemia in horses inUppsala it was interesting to determine the gasexchange in horses with inflammatory airwaydisease.

The results presented and discussed in thisabstract are based on research in riding horses andStandardbred trotters with chronic obstructivebronchiolitis (recurrent airway obstruction, RAO)or inflammatory airway disease (IAD). Thedisease was diagnosed by clinical examination andverified with lung biopsy taken through the 10thintercostal space about 20 cm above the ventrallung border. Since bronchoalveolar lavage (BAL)was not introduced in the clinic at the time of thestudies, lung biopsies were used to provide amorphological diagnosis of the case material.Tracheal washes were taken to exclude thepresence of bacterial infections. Pulmonary gasexchange was assessed by conventional blood gasvariables, ie arterial and mixed venous bloodgases, and the VA/Q distribution was estimated

with MIGET adapted for use in the standing horse(Hedenstierna et al. 1987) and during exercise inStandardbred trotters (Nyman et al. 1995)

When MIGET is performed, 6 gases (sulphurhexafluoride, ethane, cyclopropane, enflurane,diethyl ether and acetone) inert in the sense ofbeing chemically inactive in blood, are dissolvedin isotonic sodium chloride solution and infusedinto the jugular vein at a rate of 30 ml/min in thestanding horse and 120–550 ml/min duringexercise. When a steady state condition isrecorded, arterial and mixed venous blood samplesare drawn, and mixed expired gas is collected froma heated mixing box connected to a nose mask.

Gas concentrations in the blood samples andexpirate are measured by the method of Wagner etal. (1974b). The arterial/mixed venous and mixedexpired/mixed venous concentration ratios(retention and excretion, respectively) arecalculated for each gas and their solubilitycoefficient in blood are measured in each horse bya 2-step procedure. These data are then used forderiving the distribution of ventilation and bloodflow in a 50-compartment lung model with eachcompartment having a specific VA/Q ratio rangingfrom 0 to infinity. Ventilation and blood flow inhealthy subjects have a log normal distributionagainst VA/Q ratios. This means that thedistributions of ventilation and of blood flowagainst VA/Q ratios cross each other at a VA/Q of1, and that perfusion exceeds ventilation in regionswith VA/Q ratios below one, and that ventilationexceeds perfusion in compartments with VA/Qratios above one. For the information obtainedconcerning the VA/Q distribution, data arepresented for the mean and standard deviation ofthe blood flow log distribution (Q mean and logSDQ, respectively), shunt (Qs; perfusion of lung


76

regions with VA/Q<0.005), the mean and standarddeviation of the ventilation log distribution(Vmean and log SDV, respectively). Allsubdivisions of blood flow and ventilation areexpressed in percent of cardiac output and expiredminute ventilation, respectively. Diffusionlimitation can be calculated since the calculationof PaO2 from the VA/Q distribution assumesdiffusion equilibration between end-capillaryblood and alveolar gas.

THE HEALTHY HORSE

In the healthy horse, a good matching betweenventilation and perfusion results in an overall goodoxygenation of the blood (Fig 1). The narrowdistribution of perfusion, with absence of lowVA/Q regions but a minor shunt is measured. Onthe other hand, a high VA/Q mode is frequentlyseen. This high VA/Q appears to be connected withlow pulmonary artery pressure, which suggestspoor perfusion of upper lung regions. It has beensuggested that the regulation of the perfusiondistribution is highly efficient and that collateralventilation is less important for the optimalmatching of the ventilation and perfusion in theresting horse.

PULMONARY GAS EXCHANGE IN HORSESWITH MODERATE TO SEVERE RAO

Eight adult horses (4–15 years, 382–490 kg) withclinical signs of moderate to severe chronicbronchiolitis, were studied with regard to

ventilation/perfusion relationships and lungmorphology. On biopsy, all horses showedprominent inflammatory, hyperplastic andmetaplastic bronchiolitis. In addition, diffuseacinar hyperinflation was evident at necropsyperformed in 4 of the horses. There was asignificant agreement between the extent ofbronchiolar epithelial hyperplasia in necropsyspecimens of lungs and the degree of ventilation ofhigh VA/Q regions and dead space.

The arterial oxygen tension (PaO2) wassignificantly decreased despite an increasedrespiratory rate and minute ventilation.Maintenance of normal arterial carbon dioxidetension was achieved by the considerable increasein ventilation. Cardiac output was in the normalrange, but pulmonary arterial pressure wassignificantly increased.

The major gas exchange disturbance was awidened major VA/Q mode producing a certainamount of hypoxaemia (examples shown in Fig 2).The increased scatter of VA/Q ratios was causedby an uneven distribution of alveolar ventilationand capillary blood. The considerably increaseddead space ventilation and ventilation of ‘highVA/Q regions’ (VA/Q >10) caused by a relativeincrease in alveoli, which are overventilated incomparison to their perfusion, created a loss or‘wasted ventilation’ of about 75% of the totalminute ventilation. The ventilation of high VA/Qregions and dead space might be explained byhyperinflation, with subsequent compression ofthe pulmonary vascular bed and impeded regionalperfusion. This produces high VA/Q regions ifventilation is normal or is less affected thanperfusion. High VA/Q regions have been shown tobe caused by a Zone I, ie a zone in the lung wherethe alveolar pressure exceeds capillary pressureand compresses the vessels, except for residualblood flow in capillaries at the junction betweenalveolar septa (corner vessel blood flow;Hedenstierna et al. 1979). These vesselsparticipate in gas-exchange, as indicated by highVA/Q ratios. A similar finding of high VA/Q -regions has been made in asthmatic children, andassumed to be due to regional hyperinflation andimpeded perfusion (Freyschuss et al. 1984).Neither shunting (VA/Q=0) nor perfusion of verypoorly ventilated lung regions (low VA/Q) wasfound. In view of the abundance of luminal mucusand epithelial hyperplasia in bronchioles, likely tocause small airway obstruction, the minimal shuntmay appear unexpected. However, it could be an

8

6

4

2

00.01 0.1 1 10 100V

entil

atio

n (�

) an

d bl

ood

flow

(�

) l/m

in

Horse 4 = Awake, standing

PaO2 = 12.9 kPa

VD/VT = 56.7%

VA/QQs/Qt = 1.6%

Fig 1: VA/Q distribution in a healthy standing horse.Note the narrow major VA/Q mode centered on a ratio ofone, and a small mode within high VA/Q regions.Ventilation-perfusion distribution (VA/Q) alveolarventilation as open circles (VA) and perfusion as filledcircles (Q); shunt in per cent of cardiac output (Qs/Qt);arterial oxygen tension in kPa (PaO2); dead spaceventilation (VD/VT).


77

indication of compensatory mechanisms, such ascollateral ventilation, or efficient hypoxicpulmonary vaso-constriction distributing the bloodflow away from poorly ventilated or non-ventilated lung regions. Diffusion limitation can beexcluded as a cause of hypoxaemia since themeasured and the calculated PaO2 were similar.

Interestingly, some human patients withobstructive pulmonary disease showing normalPaCO2 and ‘high VA/Q-regions’, classified byWagner et al. (1977), analogous to the horses inthis study. Wagner et al. (1977) also showed thatother human COPD patients were characterised byhigh PaCO2 and large amounts of ‘low VA/Q -regions’. Although no increase in venousadmixture (shunt and low VA/Q -regions) has beenmeasured in RAO horses at the author’sdepartment, analogous classifications may exist inequine RAO patients, which might explain thedifferent findings of shunt fraction reported bydifferent authors.

From these findings it is hypothesised that themajor functional disturbance is hyperinflation of

the lung, increasing the burden both on circulationand ventilation.

PULMONARY GAS EXCHANGE AND LUNGPATHOLOGY IN HORSES WITH RAOBEFORE AND AFTER CONTROLLEDENVIRONMENT

Pulmonary gas exchange and lung pathology inseven client owned horses with RAO wereevaluated at crisis (C) and in clinical remission (R;own unpublished data). No medical treatment wasgiven to the 7 horses during the study period. Afterthe change from a conventional managementsystem to paper bedding, grass silage and housingin a well ventilated stable a clinical improvementwas noted in all horses. After 3–12 months, therespiratory rate was significantly reduced andstable on a day to day basis, cough was no longerpresent and the tracheal mucus was minor or nolonger present. At this stage pulmonary gasexchange, MIGET and lung biopsies wererepeated.

6

5

4

3

2

1

0

l/min

Horse 5

0.001 0.01 0.1 1 10 100

Qs/Qt=4.5%

VA/Q

PaO2= 9.3 kpaVD/VT=64.3%

VA/Q

Qs/Qt=1.5%

PaO2=8.6 kpaVD/VT=70.7%

VA/QQs/Qt=0.0% Qs/Qt=1.0% VA/Q


Fig 2: VA/Q distribution in RAO Horses 5–8. Note the much broader distributions, and larger mode within high VA/Qregions than in healthy horses (cf Fig 1). Ventilation-perfusion distribution (VA/Q); alveolar ventilation as open circles(VA) and perfusion as filled circles (Q); shunt in percent of cardiac output (Qs/Qt); arterial oxygen tension in kPa(PaO2) and dead space ventilation (VD/VT).

PaO2=10.8 kpaVD/VT=41.9%

6

5

4

3

2

1

0

l/min

0.001 0.01 0.1 1 10 100

Horse 7

4

3

2

1

0

l/min

0.001 0.01 0.1 1 10 100

Horse 6 5

4

3

2

1

0

l/min

0.001 0.01 0.1 1 10 100

Horse 8


78

The total minute ventilation and respiratoryrate were significantly reduced in all horsescompared to C. Epithelial hyperplasia and gobletcell metaplasia seen on lung biopsy, had abated in4 of 7 horses after a period of 3–6 months. After 6months of treatment, 2 horses were euthanised andone horse was re-evaluated after 12 months. At theend of the study, bronchiolitis, seen as epithelialhyperplasia and goblet cell metaplasia, wasreversed in 5 of 7 horses.

The pulmonary arterial mean blood pressurewas significantly reduced but as no differences inheart rate and cardiac output were measuredduring environmental control, a reducedvasoconstriction is assumed. The majorimprovement in pulmonary gas exchange in R wasa reduced scatter of VA/Q ratios (Fig 3). Log SDQwas significantly decreased and a significantincrease of perfusion to normal VA/Q regions wasseen in R. Minimal shunt and no ‘low VA/Qregions’ were seen both before and after the horseshad been stabled in a controlled environment. Aslight improvement in arterial oxygenation wasmeasured but there were individual variationswithin the group.

An interesting and unexpected observationwas that the relative distribution of ventilation toregions with high VA/Q and dead space did notchange simultaneously with the decrease in totalminute ventilation. Thus, ventilation of dead spaceand high VA/Q regions remained on average 66%of the total minute ventilation. Pulmonaryresistance has been reported to be reducedfollowing administration of bronchodilatators to

RAO horses, but is nevertheless greater than incontrol horses. These findings indicate a residualcause of obstruction and Robinson et al. (1996)suggested that airway obstruction persisted inperipheral airways due to mucus and inflammatorychanges in the airway wall. The findings describedin the present abstract indicate that a VA/Qmismatch still exists in horses with RAO in R.Whether regional hyperinflation and impededperfusion of the lung persisted in the asymtomaticRAO horses, despite the decrese in total minuteventilation, remains to be answered. Anotherpossible explanation of the remaining VA/Qmismatch is reduced regional perfusion caused bydestruction of lung tissue, with loss of pulmonarycapillary network in areas of emphysema(Gillespie and Tyler 1969).

From these findings it is hypothesised thateven though a clinical remission was seen inhorses with RAO, distribution of ventilationremained altered compared to lung-healthy horsesand pathological lesions were not improved in allhorses after six months in a low dust environment.These findings suggest that destruction ofpulmonary capillaries and/or emphysema may stillbe present in RAO horses in R.

GAS EXCHANGE DURING EXERCISE INSTANDARDBRED TROTTERS WITH IAD

Most investigations on RAO in horses have beencarried out on middle-aged to older horses duringrest. Although IAD is quite common in youngperformance animals, and possibly can represent

Qs/Qt=1.0%



Qs/Qt=1.0%

Fig 3: The major improvement in pulmonary gas exchange after the horses had been stabled in controlled environmentwas a reduced scatter of VA/Q ratios. Log SDQ, the logarithmic standard deviation of the perfusion distribution,decreased and a significant increase of perfusion to normal VA/Q regions was seen. Minimal shunt and no ‘low VA/Qregions’ were seen either in RAO crisis or in clinical remission. Ventilation-perfusion distribution (VA/Q), alveolarventilation as open circles (VA,) and perfusion as filled circles (Q); shunt in percent of cardiac output (Qs/Qt); arterialoxygen tension in kPa (PaO2) and dead space ventilation (VD/VT).

5

4

3

2

1

0

l/min

Horse 8

0.001 0.01 0.1 1 10 100

VA/Q

5

4

3

2

1

0

l/min

6 months after change of environment

0.001 0.01 0.1 1 10 100 1000

VA/Q

an early stage of RAO, the significance inpulmonary function and gas exchange still remainsunclear.

Standardbred trotters, 3–7 years old, wereexamined during a graded exercise with thehighest workload at 95% of VO2max. All horseshad mucus present in the airways and amorphological diagnosis of mild bronchiolitis(considered as IAD) with either peribronchiolitisor bronchiolitis with luminal mucus and/or gobletcell metaplasia, and bronchioloar epithelialhyperplasia was determined in one horse. TheStandardbred trotters with IAD could achieve anadequate pulmonary gas exchange during a gradedexercise test compared to healthy Standardbreds(Nyman et al. 1999). In line with several studies,exercise-induced hypoxemia and hypercapniadeveloped at the highest work load. From thesefindings it is suggested that the airflow wasrelatively unaffected in trotters with mildbronchiolitis during exercise, but as no data onrespiratory mechanics was evaluated in this studyan increased respiratory resistence can not be ruledout. Although there was no difference in log SDQ

between healthy and IAD horses during exercise(indicated by ANOVA), a tendency to an increaseddistribution of perfusion was seen at the highestworkload in horses with IAD. The questionwhether the increased total red cell volume andhaematocrit in IAD compared to healthyStandardbreds was a compensatory mechanism foran insufficient oxygen delivery in relation to tissuedemand, or reflects that horses with IAD were invery good racing condition, remains unanswered.Recently, Harmegneis et al. (2002) usedscintigraphy to study pulmonary perfusiondistribution in ‘heavey’ horses during exercise.They reported an impaired gas exchange and moreheterogeneous perfusion distribution at the highestworkload in these horses, either if they wereclinically affected or in remission.

In line with previous work, the exercise-inducedarterial hypoxemia was induced by a considerablediffusion limitation of oxygen, some ventilation-perfusion mismatch and only minorhypoventilation, responsible for 61%, 35% and 4%,respectively, of the alveolar-arterial oxygen tensiondifference at the highest work load (Table 1).


79

TABLE 1: General cardiopulmonary variables and inert gas data in horses at rest with recurrentairway obstruction (RAO) or Standardbred trotters during exercise with inflammatory airway disease(IAD)

Healthy RAO RAO RAO Healthy IADcrisis remission

Number of horses 8 8 7 7 7Speed m/s Rest Rest Rest 8.8 ± 0.3 8.7 ± 0.4Hb g/l rest, Hct % 131.1 ± 18.1 119.6 ± 8.2* 53.7 ± 2.6 55.5 ± 5*VE l/min 65.2 ± 10.1 118.6 ± 32.8* 133.1 ± 26.5 90.9 ± 10.3*VA ml/ (min x kg) 2460 ± 574 2328 ± 370RR breaths/min 13.6 ± 4.5 29.1 ± 8.2* 25.1 ± 8.3 16.4 ± 2.3* 109 ± 8 108 ± 5Qt l/min 40.2 ± 5.6 35.8 ± 2.8 41.2 ± 6.8 36.2 ± 8.2Qt ml/ (min x kg) 633 ± 53 642 ± 136HR beats/min 41.3 ± 8.0 34.6 ± 4.1 43.4 ± 6.8 40.3 ± 7.0 212 ± 8 212 ± 13PAP mmHg 26.3 ± 5.4 33.6 ± 7.9* 33.1 ± 6.6 25.6 ± 5.6* 83 ± 11 98 ± 15*PaO2 mmHg 94.5 ± 9.0 73.5 ± 10.5* 79.5 ± 13.5 89.3 ± 10.5* 60.8 ± 3.2 59.3 ± 9.0PaCO2 mmHg 42.7 ± 4.5 43.5 ± 2.3 45.7 ± 2.3 44.3 ± 1.5 50.3 ± 5.0 50.3 ± 7.3Mean VAQ for Q 0.77 ± 0.22 0.70 ± 0.13 3.47 ± 0.66 3.51 ± 0.71Log SDQ 0.41 ± 0.13 0.74 ± 0.19* 0.59 ± 0.23 0.38 ± 0.09* 0.46 ± 0.16 0.62 ± 0.25Mean VA/Q for V 1.58 ± .96 2.90 ± 1.37* 4.31 ± 1.16 4.34 ± 0.6Log SDV 1.10 ± 0.72 1.62 ± 0.29 0.7 ± 0.6 1.0 ± 0.5 0.39 ± 0.08 0.44 ± 0.10Shunt, % of Qt 1.2 ± 1.1 1.8 ± 1.5 1.1 ± 0.1 0.9 ± 0.2 0.5 ± 0.3 0.3 ± 0.2High VA/Q, %of V 6.0 ± 7.6 10.8 ± 7.1 2.5 ± 1.2 ±Dead space, %of V 48.4 ± 14.9 63.3 ± 10.0* 63.2 ± 4.9 65.3 ± 4.3

Statistical differences * from the previous value are indicated for rest (t-test) and during exercise (ANOVA).All values are presented as mean ± SD. Haemoglobin concentration (Hb), haematocrit (Hct), minuteventilation (VE), alveolar ventilation (VA), cardiac output (Qt), heart rate (HR), mean pulmonary arterialpressure (PAP), arterial oxygen and carbon dioxide tensions (PaO2, PaCO2)


80

REFERENCES

Freyschuss, U., Hedlin, G. and Hedenstierna, G. (1984)Ventilation-perfusion relationships during exercise-induced asthma in children. Am. Rev. Resp. Dis. 130,888-894.

Gillespie, J.R. and Tyler, W.S. (1969) Chronic alveolaremphysema in the horse. Adv. vet. Sci. 13, 59-99.

Harmegnies, N.F., Duvivier, D.H., Vandenput, S.N., Art,T., Lekeux, P.M. and Votion, D.M. (2002) Exercise-induced pulmonary perfusion redistribution inheaves. Equine vet. J. 34, 478-484.

Hedenstierna, G., White, F. C., Mazzone, R. and Wagner,P. D. (1979) Redistribution of pulmonary blood flowin the dog with PEEP ventilation. J. appl. Physiol.46, 278-287.

Hedenstierna, G., Nyman, G., Kvart, C. and Funkquist,B. (1987) Ventilation-perfusion relationships in thestanding horse: An inert gas elimination study.Equine vet. J. 19, 514-519.

Nyman, G., Bjork M., Funkquist, P., Persson, S.G.B. and

Wagner P.D. (1995) Ventilation-perfusionrelationships during graded exercise in theStandardbred trotter. Equine vet. J. 18, 63-69.

Nyman, G., Bjork, M., Funkquist, P. (1999) Gasexchange during exercise in standardbred trotterswith mild bronchiolitis. Equine vet. J. 30, 96-101.

Robinson, N.E., Derksen, F.J., Olszewski, M.A. andBuechner-Maxwell, V.A. (1996) The pathogenisis ofchronic obstructive pulmonary disease of horses. Br.vet. J. 152, 283-306.

Wagner, P.D., Saltzman, H.A. and West, J.B. (1974a)Measurement of continuous distribution ofventilation-perfusion ratios: Theory. J. Appl.Physiol. 36, 588-599.

Wagner, P.D., Naumann, P.F. and Laravuso, R.B. (1974b)Simultaneous measurement of eight foreign gases inblood by gas chromatography. J. appl. Physiol. 36,600-605.

Wagner, P.D., Dantzker, D.R., Dueck, R., Clausen, J.L.and West, J.B. (1977) Ventilation-perfusioninequality in chronic obstructive pulmonary disease.J. clin. Invest. 59, 203-216.


81

FUNCTIONAL IMAGING OF INFLAMMATORY AIRWAY DISEASE

D. Votion

Department of Functional Sciences, Faculty of Veterinary Medicine, University of Liege, Sart TilmanB42, 40000 Liege, Belgium

INTRODUCTION

In recent years, lung scintigraphy has centred onthe functional modifications encountered inheaves-affected horses. Several functional testswere found to be valuable for: 1) the earlydetection of the condition (Votion et al. 1999a); 2)the assessment of environmental and medicalmanagement of the disease (Votion et al. 1999a,b);and for 3) the study of the disease process at thecellular level (Fairbairn et al. 1993; Fairbairn et al.1996). However, these scintigraphical tests havenever been performed on horses affected withinflammatory airway disease (IAD) and might alsocontribute to a diagnosis of the condition, to assesstreatment and to improve knowledge about thepathogenesis of IAD.

Following a brief introduction to the nuclearfunctional imaging, this paper describes severalpulmonary function studies that may be of interestin IAD-affected horses.

NUCLEAR FUNCTIONAL IMAGING

Scintigraphy is a nuclear imaging technique thatuses gamma-emitting radioactive tracers tovisualise and quantify functions of the body.Depending on the pulmonary function to bestudied, the tracer may be inhaled by the horse orgiven by iv injection. The distribution of theradioactive compound within the body ismonitored with a gamma-camera. This gamma-camera is connected on-line to a data-processingsystem that builds-up a 2-dimensional image of the3-dimensional tracer distribution. In this image,the colours vary proportionally to regionalradioactivity.

PULMONARY FUNCTION STUDIES

A consensual definition of IAD is not yet adopted.Nevertheless, it is agreed in general that mildpulmonary inflammation and decreasedperformance characterise this respiratory tractdisease. The scintigraphical tests that enabledetection and follow up of the inflammatoryprocess will be reviewed and it will be shown howscintigraphy might contribute to the understandingof exercise intolerance.

Study of inflammatory process

Several scintigraphical possibilities exist for thestudy of the inflammatory process: scintigraphywith radioactively labelled cells of the blood,scintigraphy with radioactively labelled antibodiesand, the scintigraphical determination of thealveolar-capillary barrier permeability.

Blood cell labelling: Sites of inflammation can bedetected using radiolabelled white blood cells. Inhuman medicine, the use of labelled granulocyteshas become an established means of diagnosing avariety of inflammatory conditions in whichneutrophil migration is a prominent pathologicalfeature. With small doses of radioactivity,granulocytes are not damaged and the cells keeptheir physiological properties. Lymphocytes,however, are strongly sensitive to radiations.Because lymphocytes are long-lived cells, the riskof malignant transformation cannot be excludeddue to chromosomal damage. Therefore, thelabelling of lymphocytes is not a commonprocedure in human medicine even if thetechnique is feasible. On the other hand, it might


82

be tested on experimental horses suffering fromIAD.

In heaves-affected horses, labelled leukocyteswere used to follow the time course accumulationof inflammatory cells into the lung of allergichorses exposed to dusty hay. Using this technique,the contribution of neutrophils, eosinophils andplatelets to the pathogenesis of heaves wasinvestigated (Fairbairn et al. 1993, 1996). Sub-populations of blood cells were separated bydensity centrifugation and the ex-vivo labelledcells were administered iv to heaves-affectedhorses exposed to mouldy hay. Scintigraphyconfirmed that airway inflammation in heaves ischaracterised by a predominant neutrophilicinflammation (Fairbairn et al. 1993). In addition,the active migration of neutrophils from the bloodcompartment to the air space can be demonstratedby collecting tracheal secretions. Indeed, thetracheal aspirate of heaves-affected horsesshowing clinical signs of the disease wasradioactive, whereas no radioactivity was found incontrol horses or in heaves-affected horses duringclinical remission of the disease (Votion et al.2000).

Labelled antibodies: Monoclonal antibodiesdirected against a specific sub-population ofleukocytes have become commercially available.Using a monoclonal antibody directed againstneutrophils, it was possible to demonstrate theactive migration of neutrophils in heaves-affectedhorses showing clinical signs of the disease(Votion et al. 2000).

A novel approach to the detection ofinflammation is based on the use of radiolabelledmonoclonal antibodies directed against adhesionmolecules. Adhesion molecules promote themigration of leukocytes into inflamed tissue byallowing leukocytes to be successfully attached tothe endothelium. Activation of these adhesionmolecules is a very early event in inflammation;consequently, these radioactive antibodies mightbe used as an early radionuclide detector of acuteinflammation.

Alveolar-capillary barrier permeability: Thescintigraphical study of the alveolar-capillarybarrier permeability enables 2 other events of theinflammatory process to be followed: 1) thevascular leakage due to increased permeability ofthe endothelium; and (2) the modification of thealveolar epithelium permeability that increases

because the inflammatory reaction damages thisepithelium.

Endothelial permeability: During an acuteinflammatory process, local vasodilatation andincreased endothelial permeability result in theextravasation of plasma. This vascular leakagemay be measured using a double isotope dilutiontechnique. In the double isotope method, 2 tracersare injected iv:

• the first tracer is made up of radioactiveproteins that have a molecular weight near thatof albumin;

• the second consists of radioactive red bloodcells labelled with another radioactiveelement. The lung radioactivity obtainedfollowing administration of both tracers iscompared. Presuming that red blood cells stayin the vascular bed, additional radioactivityfound in the lung compartment is attributed topulmonary oedema.

Epithelial permeability: The alveolar epitheliumpermeability is measured by studying theclearance from lung to blood of a smallhydrophilic compound labelled with radioactivity.Hydrophilic compounds deposited into alveoli bynebulisation slowly diffuse through theintercellular junctions of alveolar cells. Once inthe interstitium, they are cleared rapidly by theblood flow because the intercellular junctions ofthe endothelial cells are wider. When aninflammatory process alters the alveolarepithelium, the clearance of the compound fromthe alveoli is accelerated. Therefore, a more rapidclearance from the lung indicates alteration of thealveolar epithelium.

The scintigraphical determination of alveolarclearance rate demonstrated a significantlyaccelerated clearance of the radioactive tracerfrom alveoli when heaves-affected horses withclinical signs of the disease were compared tohealthy horses. After 2 months at pasture, theseheaves-affected horses were free of signs of thedisease. During this remission, horses had normalpulmonary function tests (ie mechanics ofbreathing and arterial blood gas analysis).Furthermore, the mean alveolar clearance rate wassimilar to that of healthy horses. After 2 monthsstabled in a controlled environment (ie poor inallergens responsible for clinical signs), heaves-affected horses in remission remained free of


83

symptoms and presented normal pulmonaryfunction tests. However, they showed an meanalveolar clearance rate intermediate betweenhealthy and clinically-affected horses thatsuggested the existence of mild inflammation.Scintigraphy was more sensitive than pulmonaryfunction tests in the diagnosis of this mildrespiratory condition (Votion et al. 1999a). Theseresults suggest that alveolar clearance ratedetermination might be of value for IAD detection.

Regional lung function

Definition of the ventilation-perfusion relationshipand determination of the exercise-inducedpulmonary perfusion redistribution (EIPPR) mightcontribute to the understanding of decreasedperformance in IAD-affected horses.

Ventilation-perfusion relationship: Withscintigraphy, regional analysis of the ventilation-perfusion relationship may be performed. For thatpurpose, it is necessary to compute an image of theratio between both parameters. To this aim, theventilation image is divided by the perfusion one.From the computed ventilation to perfusion ratio(V/Q) image, the lung regions ‘more ventilatedthan perfused’, ‘more perfused than ventilated’ or‘equally ventilated and perfused’ may be identifiedclearly (Votion et al. 1997). This V/Q imageenables determination of the normal V/Q gradientas well as identification of local dysfunction.

Using V/Q images analysis, it wasdemonstrated that heaves induced respiratoryfunction impairment and that respiratorydysfunction may be still present following medicaltreatment despite clinical recovery (Votion et al.1999b).

Exercise-induced pulmonary perfusionredistribution: The creation of exercising toresting perfusion ratio images enables thevisualisation of the EIPPR. These imagesdemonstrated that exercise redistributes blood flowto the dorso-caudal regions of the lungs. Thisparticular redistribution may contribute to locationof bleeding sites in exercise-induced pulmonaryhaemorrhage (O’Callaghan et al. 1987). Variationin EIPPR was found to occur depending on the

clinical status: higher heterogeneity in blood flowdistribution from rest to exercise was associated tothe heaves condition (Harmegnies et al. 2002).

CONCLUSIONS

The use of scintigraphy to scrutinise cellular andphysiological changes induced by IAD might helpto improve knowledge of the disease and may leadto a better definition of the syndrome.

ACKNOWLEDGEMENTS

The authors gratefully acknowledge the ‘FondsNational de la Recherche Scientifique’ of Belgiumfor supporting this work.

REFERENCES

Fairbairn, S.M., Page, C.P., Lees, P. and Cunningham,F.M. (1993) Early neutrophil but not eosinophil orplatelet recruitment to the lungs of allergic horsesfollowing antigen exposure. Clin. exp. Allergy 23,821-828.

Fairbairn, S.M., Marr, K.A., Lees, P., Cunningham, F.M.,and Page, C.P. (1996) Effects of platelet activatingfactor on the distribution of radiolabelled leucocytesand platelets in normal horses and asymptomatichorses with chronic obstructive pulmonary disease.Res. vet. Sci. 61, 107-113.

Harmegnies, N.F., Duvivier, D.H., Vandenput, S.N., Art,T., Lekeux, P.M. and Votion, D.M. (2002) Exercise-induced pulmonary perfusion redistribution inheaves. Equine vet. J. In press.

O’Callaghan, M.W., Pascoe, J.R., Tyler, W.S. andMason, D.K. (1987) Exercise-induced pulmonaryhaemorrhage in the horse: results of a detailedclinical, post mortem and imaging study. II. Grosslung pathology. Equine vet. J. 19, 389-393.

Votion, D., Ghafir, Y., Vandenput, S., Duvivier, D.H.,Art, T. and Lekeux, P. (1999b) Analysis ofscintigraphical lung images before and aftertreatment in horses suffering from chronicpulmonary disease. Vet. Rec. 144, 232-236.

Votion, D.M., Harmegnies, N.F. and Lekeux, P.M. (2000)Feasability of monitoring neutrophils’ migrationinto the lung with immunoscintigraphy: preliminarystudy. Eur. J. Nucl. Med. 27, 1089.

Votion, D., Vandenput, S., Duvivier, D.H., Art, T. andLekeux, P. (1997) Analysis of equine scintigraphicallung images. Vet. J. 153, 49-61.

Votion, D., Vandenput, S., Duvivier, D.H., Lambert, P.,Van Erck, E., Art, T. and Lekeux, P.M. (1999a)Alveolar clearance in horses with chronicobstructive pulmonary disease. Am. J. vet. Res. 60,495-500.


84

INFLAMMATORY AIRWAY DISEASE AND CLINICAL EXERCISE TESTING

L. Couëtil

Purdue University, West Lafayette, Indiana 47907, USA

The purpose of exercise testing is to evaluate thephysiological response of the horse in a controlledway to detect and, if possible, quantify the effect ofinflammatory airway disease (IAD) onperformance. The latter may be assessedobjectively in racehorses by comparing finishingposition, racing times, and earnings in affected andcontrol horses. However, many confoundingfactors may influence race performance such astrack condition and design, jockey or driver, andlength of the race.

In other athletic horses, the effect of IAD onperformance is more subjective. The negativeimpact of IAD on performance is suggested by astudy of Standardbred racehorses (n= 965)revealing that horses finishing in the last twopositions were 5.8 times more likely to havemucopurulent exudate during post-race endoscopyof the trachea than horses finishing first or second(P<0.0001; MacNamara et al. 1990). Anotherstudy found that Thoroughbred racehorsesexhibiting marked decrease in performance had asignificantly increased percentage of neutrophilsin bronchoalveolar lavage (BAL) fluid (Fogartyand Buckley 1991). Furthermore, 41% of horseswith IAD returned to previous level ofperformance after improving ventilation anddecreasing the amount of air born dust in thehorses’ stall.

Horses with IAD pose a diagnostic challengefor the clinician because they are clinicallyasymptomatic at rest, except for increasedrespiratory exudate visible by endoscopy in themajority of cases and mild intermittent cough inonly a 3rd of the cases. Poor performance may alsoresult from a variety of other causes, such as upperairway obstruction, lameness, exertionalrhabdomyolysis, and cardiac and neurologicdiseases. More importantly, it is common to

diagnose several concomitant problems in the samehorse (Morris and Seeherman 1991). Clinicalexercise testing should be tailored to the horse’sfitness level and type of activity to allow a safe andconsistent evaluation of respiratory, cardiovascular,and musculoskeletal function. The key isstandardisation, whereby test results from onehorse may be compared to results from healthycontrols matched for age and fitness level.Alternatively, test results collected from one horsecan be compared before and after therapy. Suchtesting may be conveniently conducted on a high-speed treadmill. Some field tests have also beenused. However, they only allow a limited range ofmeasurements and currently, are less suitable thantreadmill exercise testing for evaluation ofrespiratory function. Based on test results, theclinician needs to assess if the degree of IAD islikely to affect performance and to eliminate otherpossible causes of poor performance.

The main role of the respiratory system is totransport oxygen through the conducting airwaysto the alveolar-capillary gas-exchanging surface ofthe lung and to allow diffusion of oxygen into theblood. In return, carbon dioxide will be transportedin the opposite direction. Several studies haveshown that healthy fit horses develop hypoxemiaand hypercapnia during strenuous exercise mainlydue to diffusion limitation and ventilation/perfusion mismatch and, to a lesser extentinsufficient ventilation (Wagner et al. 1989;Nyman et al. 1995; Hopkins et al. 1998).Consequently, even mild respiratory diseaseaffecting conducting airways or gas exchangeareas of the lung may significantly impair oxygendelivery to exercising muscles and result indecreased performance.

A clinical exercise testing study ofStandardbred racehorses found that horses with


85

IAD exercising submaximally maintained lowerarteriovenous oxygen content difference andexhibited increased red cell volume (RCV/kg bwt)and pulmonary artery pressure in comparison tohealthy controls (Nyman et al. 1999). Thesefindings suggested a compensatory response toexercise-induced hypoxemia (Persson 1967;Persson 1983) even though significant differencesin PaO2 between IAD and control horses were notfound. The elevated pulmonary artery pressurewas thought to result from increased vascularresistance. Elevation in RCV/kg BW or packedcell volume has been shown to correlate withincrease in pulmonary blood pressure and vascularresistance (Funkquist et al. 1995; Davis andManohar 1988). Also, horses with more severeairway disease such as heaves have significantlyelevated pulmonary artery pressure (Eberly et al.1966; Nyman et al. 1991). Others have shown thatracehorses with IAD exhibit a more pronouncedexercise-induced hypoxemia than healthy controlsduring a standardised run to fatigue (Fig 1; Couëtiland DeNicola 1999). These findings are consistentwith data in horses with heaves where exercise-induced hypoxaemia was worse during periods ofmaximum airway obstruction (disease crisis) thanduring periods of disease remission where airwayobstruction was minimum (Art et al. 1998). Thedegree of exercise-induced hypoxemia is morepronounced as the level of training and VO2maxincrease (Christley et al. 1997). Therefore,assessment of the significance of exercise-inducedhypoxemia is dependent on control data matchedfor horses’ age and fitness level.

Another investigation evaluating gas exchangesand lung biopsy parameters in Standardbredracehorses showed that horses with IAD had

impaired performance potential compared tohealthy controls, based on lower minute ventilationat a speed corresponding to a heart rate of 200 bpm(VE-200/kg bwt) and increased RCV/kg BW(Persson and Lindberg 1991). In addition, theseverity of lung biopsy score was negativelycorrelated with VE-200/kg bwt, suggesting thatbronchial epithelial hyperplasia of the smallairways was probably causing airflow obstruction.These findings are consistent with the fact thatsmall airway obstruction is evident in horses withIAD when using sensitive methods such as forcedexpiration or forced oscillatory mechanics (Fig 2;Hoffman and Mazan 1999; Couëtil et al. 2001).Also, a significant negative correlation has beenreported between forced expiration indices andpercentage of neutrophils in BAL fluid, indicatingthat as airway inflammation becomes more severe,small airway obstruction is more pronounced(Couëtil et al. 2001).

In athletic horses with IAD other thanracehorses, the degree of airflow obstruction isusually more pronounced. This may be evidentclinically as a mildly increased respiratory rate andefforts, and sometimes wheezes are heard duringthoracic auscultation with a rebreathing bag. Clinicalexercise testing is rarely performed on theseindividuals because of a lack of reference values forhorses with different fitness levels and aerobiccapacities. In such cases, BAL fluid cytology isvaluable and the degree of airflow obstruction maybe quantified by lung function testing.

Clinical exercise testing is also a useful way ofassessing response to IAD treatment. In

110

100

90

80

70

60

502 4 6 8 10 12 14

PaO

2 (m

mH

g)

Treadmill speed (m/s)

Fig 1: Partial pressure of oxygen in arterial blood(PaO2) of healthy racehorses (control) and horses withinflammatory airway disease (IAD) during astandardised treadmill test. *Data significantly differentfrom control values (P<0.05).

✦ Control� IAD

25

20

15

10

5

0Control IAD RAO rem RAP crisis

FE

F95

% (

l/s)

*

***

Fig 2: Mean forced expiratory flow after 95% of the vitalcapacity has been exhaled (FEF95%) in healthy horses(control), horses with inflammatory airway disease(IAD), and horses with recurrent airway obstruction orheaves during period of disease exacerbation (RAOcrisis) or clinical remission (RAO rem). *Datasignificantly different from control values (P<0.001).**Data significantly different from control and IADvalues (P<0.01).

racehorses, an improvement in PaO2 may be seenduring a standardised treadmill test in conjunctionwith normalisation of BAL fluid cytology (Fig 3).In horses with detectable airflow obstruction,resolution of flow limitation is commonlyobserved within 1–2 weeks of therapy. Resolutionof airway inflammation and hyper-responsivenessusually takes more time.

In conclusion, IAD in athletic horses results inmild impairment of lung function in comparison toheaves. However, it may lead to poor performanceby compromising delivery of oxygen to the blood.The degree of lung dysfunction may be objectivelyassessed during clinical exercise testing and alsoserve as a baseline to evaluate response totreatment.

REFERENCES

Art, T., Duvivier, D., Votion, D., Anciaux, N., Vandenput,S. and Bayly, W. (1998) Does an acute COPD crisismodify the cardiorespiratory and ventilatoryadjustments to exercise in horses? J. appl. Physiol. 84,845-852.

Christley, R.M., Hodgson, D.R., Evans, D.L. and Rose,R.J. (1997) Effects of training on the development ofexercise-induced arterial hypoxemia in horses. Am. J.vet. Res. 58, 653-657.

Couëtil, L. and DeNicola, D. (1999) Blood gas, plasmalactate, and bronchoalveolar lavage cytology analysesin racehorses with respiratory disease. Equine vet. J.30, 77-82.

Couëtil, L., Rosenthal, F., DeNicola, D. and Chilcoat, C.(2001) Clinical signs, evaluation of bronchoalveolar

lavage fluid, and assessment of pulmonary function inhorses with inflammatory respiratory disease. Am. J.vet. Res. 62, 538-546.

Davis, J. and Manohar, M. (1988) Effect of splenectomy onexercise-induced pulmonary and systemichypertension in ponies. Am. J. vet. Res. 49, 1169-1172.

Eberly, V., Tyler, W. and Gillespie, J. (1966)Cardiovascular parameters in emphysematous andcontrol horses. J. appl. Physiol. 21, 883-889.

Fogarty, U. and Buckley, T. (1991) Bronchoalveolar lavagefindings in horses with exercie intolerance. Equine vet.J. 23, 434.

Funkquist, P., Nyman, G. and Persson, S.G.B. (1995)Hemodynamic responses to exercise in trotters withred cell hypervolemia and exercise induced pulmonaryhemorrhage. Newmarket, R&W Publications. Proc.10th int. Conf. racing Anal. Vet. Eds: P. Kallings, U.Bondesson and E. Houghton. R&W Publications,Newmarket, pp 165–167.

Hoffman, A. and Mazan, M. (1999) Programme of lungfunction testing horses suspected with small airwaydisease. Equine vet. Educ. 11, 322-328.

Hopkins, S., Bayly W., Slocombe, R., Wagner, H. andWagner, P. (1998) Effect of prolonged heavy exerciseon pulmonary gas exchange in horses. J. appl. Physiol.84, 1723-1730.

MacNamara, B., Bauer S. and Iafe, J. (1990) Endoscopicevaluation of exercise-induced pulmonary hemorrhageand chronic obstructive pulmonary disease inassociation with poor performance in racingStandardbreds. J. Am. vet. med. Ass. 196, 443-445.

Morris, E.A. and Seeherman, H.J. (1991) Clinicalevaluation of poor performance in the racehorse: theresults of 275 evaluations. Equine vet. J. 23, 169-174.

Nyman, G., Bjork, M. and Funkquist, P. (1999) Gasexchange during exercise in Standardbred trotters withmild bronchiolitis. Equine vet. J. 30, 96-101.

Nyman, G., Bjork, M., Funkquist, P., Persson, S. andWagner, P. (1995) Ventilation-perfusion relationshipsduring graded exercise in the Standardbred trotter.Equine vet. J. 18, 63-69.

Nyman, G., Lindberg, R., Weckner, D., Bjork, M., Kvart,C. and Persson, S. (1991) Pulmonary gas exchangecorrelated to clinical signs and lung pathology inhorses with chronic bronchiolitis. Equine vet. J. 23,253-260.

Persson, S. (1967) On blood volume and working capacityin horses. Studies of methodology and physiologicaland pathological variations. Acta vet. scand. Suppl. 19,1-189.

Persson, S. (1983) Evaluation of exercise tolerance andfitness in the performance horse. In: Equine exercisephysiology. Eds: D. Snow, S. Persson & R. RoseGranta Editions, Cambridge, pp 441-457.

Persson, S.G.B. and Lindberg, R. (1991) Lung biopsypathology and exercise tolerance in horses withchronic bronchiolitis. Eds: S.G.B. Persson, A.Lindholm, & L.B. Jeffcott, ICEEP Publications,Davis, California, 3, 457-464.

Wagner, P., Gillespie, J., Landgren, G., Fedde, M., Jones,B. and DeBowes, R. (1989) Mechanism of exercise-induced hypoxemia in horses. J. appl. Physiol. 66,1227-1233.


86

Fig 3: Partial pressure of oxygen in arterial blood(PaO2) of healthy racehorses (control), horses withinflammatory airway disease (IAD), and a horse (Horse1) with IAD before and 2 weeks after therapy (post Tx)during a standardised treadmill test.

•

••

•

•

•

•

•

•

•

••

•

••

ControlHorse 1Horse 1 (post Tx)

110

100

90

80

70

60

500 2 4 6 8 10 12 14

Treadmill speed (m/s)

PaO

2(m

mH

g)

87


WORKSHOP SUMMARY


88


WORKSHOP SUMMARY

In July 2000 a workshop addressed the confusingissue of terminology used to describe equine lowerairway disease (ie tracheobronchial disease)(Robinson 2001). At that workshop the syndromeof ‘heaves’ (also known as recurrent airwayobstruction or RAO) was defined. Participants inthat workshop recognised that many horses have achronic, lower grade, diffuse form of airwayinflammation that is not accompanied by clinicalsigns of airway obstruction, and they agreed toname that syndrome inflammatory airway disease(IAD). The present Havemeyer Workshop wasarranged to review the current understanding ofIAD. The following statements represent thegeneral consensus about IAD reached byparticipants in the Havemeyer Workshop.

CLINICAL DEFINITION OF THE SYNDROME

• The clinical presentation of IAD may includecough, accumulation of secretions in thetrachea, nasal discharge, poor performance,and prolonged recovery after exercise.

• Affected animals do not have increasedrespiratory effort at rest (ie they are notdepressed, inappetant, or febrile).

• Affected horses are not systemically ill.

• The CBC, blood chemistry, and fibrinogenconcentration of these horses are withinnormal limits.

• Horses with IAD have increased proportionsof inflammatory cells, in particularneutrophils, mast cells, and/or eosinophils inairway secretions.

CLINICAL FINDINGS

• Horses with IAD can have airwayinflammation and lung dysfunction withoutovert clinical signs.

• Horses with IAD may have accumulatedsecretions in the trachea without a cough.

• Cough is an insensitive indicator of IAD.

• In racehorses, IAD is most common in younghorses (2–4 years of age) and diminishes withtime in the training environment.

• IAD in racehorses has been shown to beassociated with poor performance.

ENDOSCOPIC OBSERVATIONS

• The clinical significance of mucusaccumulation depends on the level ofperformance expected from the horse.

• In order to chart the progression of the diseaseand response to therapies, gross observationsof mucosal appearance and the thickening andblunting of the bronchial bifurcations shouldbe noted, and the amount of tracheal secretionsshould be quantified using a standard 5-gradescoring system.

EPIDEMIOLOGY

• IAD can occur in horses of all ages anddisciplines.

• In the majority of young racehorses, IADresolves but it remains persistent in a minority.

• IAD in racehorses has been studied mostextensively in young animals in the UK andAustralia, where up to 80% of horses areaffected at some point in the first year oftraining.

• In non-racehorses, IAD is usually diagnosed atan older age than in racehorses.

• Studies in sport horses maintainedpermanently indoors indicate that 70% ofhorses or greater may be affected at one time.

89


• Non-racehorses presenting for evaluation atreferral centers tend to have more obviousclinical signs of respiratory disease than doyoung racehorses. This is probably becauseracehorses put great aerobic demands on theirlungs and so dysfunction is more readilyapparent.

AETIOLOGY

• The aetiology of IAD probably involves manyfactors that act either synergistically orsequentially to initiate and/or prolong airwayinflammation.

• The air that horses breathe in many conventionalstables contains particle levels known to causeairway inflammation in other species.

• Conventional stabling is associated withairway inflammation in apparently healthy(non-heaves-affected) horses.

• Respirable endotoxin concentration is a riskfactor for increased neutrophils in airwaysecretions.

• Viral infection has been detected in only asmall proportion of cases of IAD.

• In some of the racehorses with IAD, thepresence of detectable tracheal secretions isassociated with the presence of bacteria thatare likely to be pathogenic in the horse.

• In the UK and Australia, increased numbers ofbacteria (predominantly Streptococcus,Actinobacillus and Pasteurella) are frequentlyisolated from the trachea of young racehorseswith IAD.

• The role of allergy in IAD is unknown.

• The relationship between IAD and heaves iscurrently unknown.

LABORATORY FINDINGS

• The CBC, blood chemistry, and fibrinogenconcentration of these horses are withinnormal limits.

• Both tracheal aspirates and bronchoalveolarlavage fluid should be collected to fullyevaluate the lower respiratory tract.

• Tracheal aspirates and bronchoalveolar lavage

sample different areas of the lower respiratorytract and, therefore, their cytologicalinterpretations are not interchangeable.

• The presence of neutrophils in trachealsecretions or BALF does not per se signify thepresence of infection.

TRACHEAL ASPIRATES

• In young healthy racehorses, currentinformation suggests that there should be<20% neutrophils and <1% eosinophils intracheal aspirates.

• In 2–3-year-old performance horses, thepresence of >20% neutrophils in trachealaspirates is associated with increased mucusand increased risk of coughing.

• Tracheal aspirates for bacterial isolationshould be collected endoscopically with aguarded catheter using 10–15 ml of sterilesaline 1–2 h after exercise and a minimum of24 h after transportation.

BRONCHOALVEOLAR LAVAGE

• In clinically healthy horses, currentinformation suggests that there are on average60% macrophages, 35% lymphocytes, <5%neutrophils, <2% mast cells, <0.1%eosinophils and occasional or no epithelialcells in bronchoalveolar lavage (BAL) fluid.

• An increased percentage of mast cells oreosinophils in BAL fluid is associated withairway hyper-reactivity.

• Bronchoalveolar lavage fluid should becollected and processed according to therecommendations of the InternationalWorkshop on Equine Chronic Airway Disease(Robinson 2001).

LUNG FUNCTION

• Conventional measurements of pulmonaryfunction (maximal change in pleural pressureduring tidal breathing (�Pplmax), pulmonaryresistance, and dynamic compliance) areunaltered in horses with IAD.

• IAD-affected horses evaluated at referralcentres for cough or poor performancecommonly have airway hyper-reactivity and/or

90


airway narrowing.RECOMMENDATIONS FOR FUTUREINVESTIGATIONS

• Standardise the scoring systems for thevariables commonly used to characterise IADin order to enable comparison betweenstudies.

• Determine the prevalence of IAD in differentpopulations of horses around the world.

• Investigate the pathological changes in horseswith IAD and determine the relationshipbetween airway remodelling, BAL fluidcytology, and lung function.

• Determine the relationships between tracheallavage cytology, bronchoalveolar lavagecytology, mucus accumulation, pulmonaryfunction at rest and during exercise and

performance.

• Conduct controlled studies of currentlyrecommended therapies of IAD.

• Determine the effect of treatment on theairway inflammation, mucus accumulation,pulmonary dysfunction, and exerciseintolerance present in IAD.

• Evaluate the effects of particulate andendotoxin exposure in young horses.

• Characterise the immunological mechanismsof IAD.

• Investigate the possible genetic predispositionto IAD.

REFERENCE

Robinson, N.E. (2001) Chairperson’s introduction:International Workshop on Equine Chronic Airway

91


LIST OF PARTICIPANTS

Laurent Couëtil, United States

Frederik Derksen, United States

Paddy Dixon,United Kingdom

Vincent Gerber, United States

Andrew Ghio, United States

David Hodgson,Australia

Jennifer Hodgson,Australia

Andrew Hoffman , United States

Jean-Pierre Lavoie,Canada

Nicholas Malikides,Australia

David Marlin, United Kingdom

Melissa Mazan,United States

Bruce McGorum, United States

Richard Newton, United Kingdom

Gorel Nyman, Sweden

Rachel Pepper,United Kingdom

Candice Platz,United States

Gene Pranzo,United States

Ed Robinson,United States

Bonnie Rush,United States

Ken Smith, United Kingdom

Hugh Townsend,Canada

Laurent Viel, Canada

Dominque Votion, Belgium

James Wood,United Kingdom

Back row (left-right): Richard Newton, Hugh Townsend, Ed Robinson, Jean-Pierre Lavoie, James Wood, Fred Derksen,Andy Hoffman, Dominique Votion, David Marlin, David Hodgson, Laurent Couëtil, Chris Deaton, Bruce McGorum

Front row (left-right): Rachel Pepper, Ken Smith, Nick Malikides, Jenny Hodgson, Melissa Mazan, Bonnie Rush,Laurent Viel, Vince Gerber, Paddy Dixon.

Dr Gorel Nyman also attended the Workshop but had to leave before the picture was taken.

92


AUTHOR INDEX

BERNEY, C. see ROBINSON, N.E.

et al.

BUREAU, F. and LEKEUX, P., 45

CADE, S.M. see SMITH, K.C. et al.

CHANTER, N. see NEWTON, J.R.

et al.

CHRISTLEY, R. see HODGSON,

D.R. et al.

CORNELISSE, C. see ROBINSON,

N.E. et al.

COUËTIL, L., 84

DEATON, C.M. see MARLIN, D.J.

et al.; SMITH, K.C. et al.

DeFEIJTER-RUPP see ROBINSON,

N.E. et al.

DERKSEN, F.J., 69 and see

ROBINSON, N.E. et al.

DIXON, P.M. et al., 7

GERBER, V. et al., 59 and see

ROBINSON, N.E. et al.

GHIO, A.J., 23, 29

GOWER, S.M. see SMITH, K.C.

et al.

HODGSON, D.R. et al., 16

HODGSON, J.L., 49 and see

HODGSON, D.R. et al.

HOFFMAN, A. and MAZMAN, M.,

71 and see MAZAN, M. and

HOFFMAN, A.

HOTCHKISS, J.A. see GERBER, V.

et al.

JEFCOAT, A.M. see GERBER, V.

et al.; ROBINSON, N.E. et al.

KING, M. see GERBER, V. et al.

LAVOIE, J.P., 31

LEKEUX, P. see BUREAU, F. and

LEKEUX, P.

MALIKIDES, N. see McGORUM,

B.C. et al.

MARLIN, D.J. et al., 62 and see

NEWTON, J.R. et al.; SMITH,

K.C. et al.

MAZAN, M. and HOFFMAN, A., 9

and see HOFFMAN, A. and

MAZMAN, M.

McGORUM, B.C. et al., 27 and see

DIXON, P.M. et al.

NEWTON, J.R. et al., 40 and see

MARLIN, D.J. et al.; SMITH,

K.C. et al.

NYMAN, G., 75

PIRIE, R.S. see DIXON, P.M.

et al.; McGORUM, B.C.

et al.

PLATZ, C., 19

REID, S.W.J. see HODGSON,

D.R. et al.

ROBINSON, N.E. et al., 13

and see GERBER, V.

et al.

RUSH, B.R., 3

SMITH, K.C. et al., 55 and see

MARLIN, D.J. et al.; NEWTON,

J.R. et al.

TOWNSEND, H.G.G., 37

VIEL, L., 52

VOTION, D., 81

WOOD, J.L.N., 33 and see

HODGSON, D.R. et al.;

NEWTON, J.R. et al.;

SMITH, K.C. et al.

93

proceedings of a workshop on - havemeyer …proceedings of a workshop on inflammatory airway...

Documents