Hemostatic Studies in Racing Standardbred Horses withExercise-induced Pulmonary Hemorrhage. Hemostatic Parameters

at Rest and After Moderate Exercise

I.B. Johnstone, L. Viel, S. Crane and T. Whiting

ABSTRACT

The purpose of this study was todetermine whether a defect in hemo-stasis might be a factor in the etiologyof exercise-induced pulmonary hemor-rhage (EIPH). Hemostatic parameterswere evaluated in 22 EIPH-positiveand ten EIPH-negative racing horseswhile in a rested state. Nineteen EIPH-positive and ten EIPH-negative horseswere further evaluated just before andimmediately after a 15 min exerciseperiod on a 260 m oval track. WhenEIPH-positive and EIPH-negativehorses were compared at rest, therewas no significant difference in any ofthe coagulation and fibrinolyticparameters studied. There was how-ever, a significant difference in plateletfunction as assessed by aggregometry.The platelets from affected horses weresignificantly less responsive than thosefrom nonaffected horses when exposedin vitro to the platelet agonistsadenosine diphosphate, collagen andplatelet activating factor. Exercisetended to increase the packed cellvolume and factor VHI/von Wiflebrandfactor and to decrease platelet aggrega-tion responses to low concentrations ofadenosine diphosphate. These effectsof exercise however were quantitativelysimilar in both EIPH-positive andEIPH-negative horses. Reducedplatelet function may therefore be acontributing factor in the bleedingcharacteristic of horses with EIPH.

RESUME

L'objectif de cette etude etait deverifier si un trouble dans la cascademenant a l'hemostase pouvait jouer unrole dans 1'etiologie des hemorragiespulmonaires induites par 1'exercice(HPIE). Les parametres de la coagu-lation ont ete evalues en pe'riode derepos chez 32 chevaux de course dont22 etaient HPIE-positifs et dix etaientHPIE-negatifs. Dix-neuf chevauxHPIE-positifs et dix HPIE-negatifsfurent aussi evaluis immediatementavant et apres un exercice d'une dureede 15 minutes sur une piste ovale de260 m. Lors de la comparaison deschevaux HPIE-positifs et negatifs,aucune difference significative n'a eteconstatee au niveau des parametresimpliques dans les processus de fibri-nolyse et de coagulation, alors qu'unedifference significative a ete observeeau niveau de ceux de 1'agregation pla-quettaire. Lorsque les plaquettes6taient exposees in vitro a des agonistes,tels I'adenosine diphosphate, le colta-gene, et le facteur d'activation plaquet-taire, la reponse des plaquettes deschevaux souffrant d'HPIE etait signi-ficativement moins forte que celle desautres chevaux. L'exercice avait ten-dance A faire augmenter l'hematocrite,les facteurs VIII et von Willebrand eta faire diminuer l'agregation plaquet-taire en presence de faibles concentra-tions d'adenosine diphosphate. Leseffets de l'exercice etaient quantitati-

vement semblables chez les chevauxHPIE-positifs et negatifs. La diminu-tion de la fonction plaquettaire pour-rait donc contribuer au saignementcaractiristique observe chez des che-vaux prisentant le syndrome d'HPIE.(Traduit par D' Andre Vrins)

INTRODUCTION

Exercise-induced pulmonary hemor-rhage (EIPH) has been reported inalmost all types of equine athletes. Theincidence in these reports has generallyexceeded 40'7o of the competitivepopulation. Potential risk factors suchas age, sex, and intensity of exercisehave not been particularly useful inelucidating the pathogenesis of thiscondition (1-5).Although the lung is clearly the

source of the bleeding in horses withEIPH, the actual cause of the bleedingremains unknown. O'Callaghan et al(6) suggested that pathological lungchanges were likely preceded byinflammatory disease (bronchiolitis),and that small airway disease stimu-lated bronchial vascular proliferationleading to increased bronchial bloodflow and predisposing to hemorrhageduring exercise. Using ventilation/perfusion scintigraphy, perfusiondeficits were shown to be prominent inhorses experiencing EIPH (7).Although there is strong evidence to

support the hypothesis that small air-

Department of Biomedical Sciences (Johnstone,of Guelph, Guelph, Ontario NIG 2W1.

Crane) and Department of Clinical Studies (Viel, Whiting), Ontario Veterinary College, University

This research was supported by grants from the Canadian Veterinary Research Trust Fund and the Ontario Ministry of Agriculture and Food.

Submitted May 25, 1990.

Can J Vet Res 1991; 55: 101-106 101

way disease and vascular dysfunctionplay a role in the pathogenesis ofEIPH, it remains unclear whether anactual defect in the hemostatic mech-anism also contributes. Johnson et al(8) demonstrated significantly reducedplatelet counts in rested "bleeder"Thoroughbred horses when comparedto "nonbleeders", but this observationwas not confirmed by Bayly et al (9).Bayly et al (9,10) did however demon-strate an exercise-induced suppressionof platelet reactivity to adenosinediphosphate in "nonaffected" and"bleeder" Thoroughbred horses andsuggested that this inhibition ofplatelet aggregation induced by exer-cise might result in a delayed sealingof damaged vessels and an increase involume of hemorrhage. The beneficialeffect of the drug furosemide wasattributed to the prevention of thisinhibitory effect (11). Coagulationactivity in EIPH positive and negativehorses has been evaluated in previousstudies using intrinsic and extrinsiccoagulation screening tests (partialthromboplastin time and prothrombintime respectively), and no differencesbetween "bleeder" and "nonbleeder"horses have been noted (8,9). Plasmalevels of specific coagulant/anti-coagulant proteins have not beenassessed.The purpose of this study was to

perform a detailed hemostatic evalua-tion on Standardbred racing horsesidentified as either having, or not hav-ing, EIPH. Horses were studied whenat rest, and before and after moderateexercise. This evaluation included spe-cific procoagulant and anticoagulantfactors as well as fibrinolytic andplatelet function parameters.

MATERIALS AND METHODS

CLASSIFICATION OF THE HORSES

The Standardbred horses studiedwere mature racing animals of eithersex classified by one author (LV) asbeing EIPH-positive or EIPH-negative,based on historical, clinical and labo-ratory examination of each horse. AnEIPH-positive horse met one or moreof the following criteria: 1) epistaxisafter racing or training reported by thetrainer; 2) blood observed by the trackveterinarian in the trachea on endo-scopic examination after racing or

training; 3) a history of poor perfor-mance in the last quarter of the race.These horses underwent a thoroughrespiratory examination with a subse-quent bronchoalveolar lavage (BAL).The area lavaged was consistently theright and left diaphragmatic lunglobes. A horse was classified as EIPH-positive when the lavage fluid collectedwas bright red in color and cytologi-cally had more than 50070 of thealveolar macrophage populationheavily laden with hemosiderin. Thehorse must have presented one or moreof these criteria in the two weeks priorto being included in this study. Exercise-induced pulmonary hemorrhage-negative horses were healthy racinghorses that had a BAL and endoscopicexamination and met none of thesecriteria. Cytologically, these horseshad less than 5 70 of the alveolarmacrophages containing hemosiderinparticles in the cytoplasm. None of thehorses had received any drugs duringthe two weeks prior to study. Thestudy was carried out with informedowner/trainer consent and coopera-tion, and in accordance with theguidelines of the Guide to the Care andUse of Experimental Animals of theCanadian Council on Animal Care.

BLOOD COLLECTION

Blood samples were collected from22 EIPH-positive horses (average age3.9 yr) and ten EIPH-negative horses(average age 3.1 yr) while at rest. Theblood (45 mL) was drawn by cleanvenipuncture and immediately mixedwith 5 mL of 3.8% trisodium citratein a plastic container. A 2 mL volumeof native blood was allowed to clot in aglass fibrin degradation product (FDP)vacutainer (Wellcome Reagents Ltd.,Beckenham, England) to obtain serumfor FDP measurements. Twenty mLof citrated blood were centrifuged at4°C and 7000 g for 20 min to obtainplatelet-free plasma (PFP). Aliquots ofthis plasma were immediately frozen to- 700C. A small volume of citratedblood was used for hematocrit andplatelet count determinations. Theremaining 20 mL of citrated bloodwere centrifuged at room temperatureat 120 g for 15 min to obtain platelet-rich plasma (PRP) for platelet functionstudies. After careful removal of thePRP supernatant, the remainder of thecontents was centrifuged for a further

15 min at 2500 g to obtain platelet-poor plasma (PPP). This PPP wasused to adjust the platelet concentra-tion of the PRP to 150 x 109/L forplatelet aggregation studies.

HEMOSTATIC TESTING

Hematocrit determinations andplatelet counts were performed usinga microhematocrit technique and amanual count with phase contrastmicroscopy procedure respectively(12). Prothrombin times (PT), acti-vated partial thromboplastin times(PTT), thrombin clotting times (TCT)and fibrinogen concentrations weredetermined as previously described(13). Prothrombin times, PTT, TCT,and fibrinogen were each expressed asa patient/control ratio; that is, theratio of the patient's clotting time orfibrinogen concentration to that for anormal reference plasma. This refer-ence plasma was a species-specific con-trol plasma prepared by poolingplasmas from at least ten clinically nor-mal mature horses.

Soluble fibrin monomer (SFM) titersand fibrin(ogen) degradation producttiters were evaluated by protaminesulfate and latex agglutination pro-cedures respectively (13).Plasma factor VIII procoagulant

activity (FVIII:C) was quantitated bya differential one-stage clotting assaywhile plasma von Willebrand factorantigen (vWF:Ag) was quantitated byelectroimmunoassay as previouslydescribed (14). Plasma antithrombinIII (ATIII) was measured by a chromo-genic assay technique (15). All specificfactors were expressed as a percentageof the activity/amount in the referenceplasma.

Platelet aggregation profiles weregenerated using a single channelplatelet aggregation profiler (ModelPAP-2, Biodata Corp., Willow Grove,Pennsylvania). Fifty AL of plateletagonist were added to 450,L of PRP(platelet concentration 150 x 109/1L)and the response recorded for at least3 min. The lag phase (LP) was anestimate of the time (in s) from theaddition of the platelet agonist to theinitiation of an aggregation response(i.e. first evidence of a increase in lighttransmission). The velocity of aggrega-tion (Av) was determined from thesteepest slope of the aggregation pro-file and was expressed as chart

102

units/min. The degree of aggregationwas determined by measuring theincrease in light transmission (height ofthe aggregation curve) in chart units.The maximum degree of aggregation(Am) was defined as the greatestincrease in light transmission duringthe 3 min following addition of theplatelet agonist, expressed as a percentof the maximum possible increase inlight transmission (16).

EFFECTS OF EXERCISE

Blood samples were collected from19 EIPH-positive horses and tenEIPH-negative horses immediatelybefore exercise and immediately aftera 15 min run on a 260 m oval track atan average speed of 10 m/s. Sampleswere processed and assayed as des-cribed above.

STATISTICAL ANALYSIS

Comparisons between EIPH-positive and EIPH-negative horseswere made using an unpaired Student'st-test. Comparisons between preexer-cise and postexercise data were madeusing a paired t-test. A p value < 0.05was considered statistically significant.

RESULTS

RESTED HORSES

The hematological and coagulationparameters in resting EIPH-positiveand EIPH-negative horses are shownin Table I. There were no statisticallysignificant differences in any of theparameters studied; however plateletcounts were slightly lower and FVIII:Cactivity slightly higher in the EIPH-positive horses when compared to theEIPH-negative animals. Soluble fibrinmonomer titers were similar in bothgroups (data not shown); all werenegative (< 1:10) except for one EIPH-negative horse and two EIPH-positivehorses which had weak positive titersof 1:20. Fibrin degradation producttiters (data not shown) were negative(<1:5) in all EIPH-negative horses;however one EIPH-positive horse hada weak positive titer of 1:10.

Comparisons of platelet aggregationresponses in PRP from rested EIPH-positive and EIPH-negative horses areshown in Table II. In general plateletsfrom affected horses responded morepoorly to platelet agonists. The dif-

TABLE I. Hematological and coagulation parameters in rested exercseinduced pulmonary hemor.rhage (EIPH)-positive and EEPH-negative horses

EIPH-positive EIPH-negativeParameter horses (n = 22) horses (n =10)

Packed cell volume (Li) 0.36 ± 0.04a 0.34 ± 0.04Platelet count (x 109/L) 136.0 ± 45.8 157.0 ± 42.0Prothrombin timeb 0.99 ± 0.05 0.98 ± 0.05Partial thromboplastin timeb 1.10 ± 0.15 1.14 ± 0.21Thrombin clotting timeb 0.97 ± 0.14 1.04 ± 0.12FibrinogenC 1.20 ± 0.26 1.16 ± 0.20Factor VIII coagulant activityd 126.6 ± 33.7 104.6 ± 27.8Von Willebrand factor antigend 127.8 ± 32.1 126.6 ± 29.2Antithrombin IIld 114.3 ± 18.1 108.5 ± 12.9

aMean ± standard deviationb/c ratio = patient's time/control plasma timeCP/c ratio = patient's concentration/control plasma concentrationd070 of activity in a species-specific normal reference plasma

TABLE H. Platelet aggregation parameters in rested exercise-induced pulmonary hemorrhage (EIPH)-positive and EIPH-negative horses

EIPH-positive EIPH-negativeAggregating agent horses horses

Adenosine diphosphate LP 13.6 ± 3.7(20)a ** 9.6 ± 3.2(10)(10-5 M)b Av 33.7 ± 14.2 63.6 ± 19.6

m 60.6 ± 18.2 ** 83.5 ± 23.9

Adenosine diphosphate LP 13.8 ± 3.9(20) * 10.3 ± 3.1(10)(10-6 M)b Av 27.6 ± 14.5 53.4 ± 19.9

Am 37.4 ± 18.7 *** 71.2 ± 27.8

Collagen LP 54.4 ± 18.8(20) 54.8 ± 15.3(10)(1:1OOO)b Av 47.0 ± 19.6 * 66.7 ± 27.3

Am 70.4 ± 23.8 89.0 ± 22.7

Platelet activating LP 19.4 ± 4.3(13) * 14.7 ± 4.3(10)factor A, 63.9 ± 21.8 * 87.5 ± 17.5(0.5 ng/mL)b Am 73.0 ± 21.0 90.3 ± 18.4

aMean ± standard deviation (number of horses)bFinal concentrationLP = lag phase (in seconds)AV= initial velocity of aggregation (units/min)Am = maximum aggregation in three minutes (No of maximum possible)* p < 0.05, ** p < 0.01, *** p < 0.001

ferences were most pronounced withthe agonist adenosine diphosphate(ADP). Platelets from EIPH-positivehorses took longer to respond to ADP,and the rate and degree of responsewas significantly less than for plateletsfrom nonaffected horses. The dif-ferences in responses to the agonistscollagen and platelet activating factor(PAF) were less pronounced but stillstatistically significant. Plateletsderived from EIPH-positive horsesshowed a significantly slower responseto these agonists than did plateletsfrom horses deemed to be free of thiscondition.

EFFECTS OF EXERCISE

The effects of exercise onhematological and coagulation param-

eters in EIPH-positive and EIPH-negative horses are shown in Table III.Packed cell volume increased in bothgroups following exercise (p < 0.001).Thrombin clotting times, fibrinogenconcentrations, and ATIII concentra-tions were not affected by exercise ineither group. Although the bloodplatelet counts in both groups showeda tendency to increase following exer-cise, these increases were not statis-tically significant. The PTT ratiostended to increase and the PT ratiostended to decrease folfowing exercise.However, these changes were onlystatistically significant in the EIPH-positive horses. Plasma FVIII:C activ-ity and vWF:Ag increased followingexercise, however the changes inFVIII:C activity and vWF:Ag were

only statistically significant in EIPH-negative horses and EIPH-positivehorses respectively.The effects of exercise on platelet

aggregation responses are shown inTable IV. The major effect of exercisewas to suppress the platelet responsesto the agonist ADP in both groups ofhorses, but only at the low concentra-tion of agonist (10-6 M final concen-tration). Exercise had no significanteffect on platelet responses to thehigher concentration of ADP, or tocollagen or platelet activating factor.Soluble fibrin monomer titers andFDP titers were not altered by exercisein either EIPH-positive or EIPH-negative horses (data not shown).To quantitatively compare the

hemostatic responses of EIPH-positiveand EIPH-negative horses to exercise,change in each parameter was assessedas a percentage of the preexercisevalue. For all parameters studied therewas no statistically significant dif-ference in the degree of change inducedby the exercise.

DISCUSSION

The results of this study indicatedthat platelets derived from horses withEIPH were significantly less responsiveto platelet agonists than plateletsderived from horses deemed to be freeof this condition. The reason for thedecreased responsiveness of plateletsderived from rested EIPH-positivehorses is not known. Drug-inducedeffects were ruled out because of thethorough screening of these horsesbefore entering the study. AlthoughBayly et al (11) showed a "tendency"for platelets from five rested EIPH-positive horses to be less responsive toADP than control platelets, weobserved a statistically significantreduction in response in platelets fromaffected horses, not only to ADP butto other platelet agonists. Although theeffect on ADP-induced aggregationwas most pronounced, responses tocollagen and PAF were also signifi-cantly slower with platelets derivedfrom EIPH-positive horses. This slug-gish response of platelets from EIPH-positive horses may be a contributingfactor to the increased blood loss seenin these subjects.

TABLE III. Comparison of hematological and coagulation parameters in exercise-inducedpulmonary hemorrhage (EIPH)-positive and EIPH-negative horses before and after exercise

EIPH-positive (n = 19) EIPH-negative (n = 10)Parameter Preexercise Postexercise Preexercise Postexercise

Packed cell volume (L/L) 0.36 ± o.04a*** 0.45 ± 0.06 0.34 ± 0.04*** 0.44 ± 0.05Platelet count (x 109/L) 134.3 ± 48.9 141.4 ± 58.5 157.0 ± 42.0 163.1 ± 56.0

Prothrombin timeb 0.99 ± 0.05 ** 0.96 ± 0.04 0.98 ± 0.05 0.96 ± 0.03Partial thromboplastin timeb 1.07 ± 0.12 * 1.17 ± 0.21 1.14 ± 0.21 1.17 ± 0.23Thrombin clotting timeb 0.95 ± 0.11 0.99 ± 0.14 1.04 ± 0.12 1.01 ± 0.08Fibrinogenc 1.20 ± 0.27 1.19 ± 0.29 1.16 ± 0.20 1.09 ± 0.27

Factor VIII coagulant activityd 128.4 ± 35.6 137.8 ± 33.4 104.6 ± 27.8** 127.1 ± 26.8Von Willebrand factor antigend 133.2 ± 28.0 *** 149.9 ± 30.0 126.6 ± 29.2 165.9 ± 51.6Antithrombin IlId 115.1 ± 18.2 122.9 ± 17.2 108.5 ± 12.9 114.4 ± 31.3aMean ± standard deviationbP/c ratio = patient's time/control plasma timeCP/c ratio = patient's concentration/control plasma concentrationd'7o of activity in a species-specific normal reference plasma*p < 0.05 ** p < 0.01 *** p < 0.001

TABLE IV. Comparison of platelet aggregation responses in platelets taken from exercise-induced pulmonary hemorrhage (EIPH)-positive and EIPH-negative horses before and afterexercise

EIPH-positive (n = l7)a EIPH-negative (n = 10)Aggregating agent Preexercise Postexercise Preexercise Postexercise

Adenosine diphosphate LP 13.9 ± 4.1 13.9 ± 3.6 9.6 ± 3.2 9.3 ± 3.6(10-5 M) AV 34.9 ± 15.1 31.6 ± 9.6 63.6 ± 19.6 54.7 ± 14.6

Am 63.4 ± 19.1 59.1 ± 14.6 83.5 ± 23.9 79.5 ± 22.2

Adenosine diphosphate LP 13.9 ± 4.2 ** 17.5 ± 5.0 10.3 ± 3.1 12.0 ± 3.6(10-6 M) AV 27.1 ± 15.3 22.8 ± 7.4 53.4 ± 19.9 38.7 ± 14.0

Am 37.4 ± 20.3 * 28.0 ± 10.6 71.2 ± 27.8 * 48.3 ± 24.7

Collagen (1:1000) LP 54.9 ± 20.2 59.0 ± 16.6 54.2 ± 15.3 49.1 ± 8.5AV 46.1 ± 20.8 43.0 ± 14.4 66.7 ± 27.3 54.8 ± 14.2Am 68.9 ± 25.7 67.7 ± 20.9 89.0 ± 22.7 91.5 ± 17.9

Platelet activating LP 19.6 ± 4.6 18.5 ± 4.5 14.9 ± 4.3 13.1 ± 4.5factor - (0.5 ng/mL) AV 62.1 ± 23.3 66.7 ± 21.8 87.5 ± 17.5 86.7 ± 17.6

Am 74.6 ± 22.5 73.2 ± 20.9 90.3 ± 18.4 92.3 ± 17.0an = 10 for platelet activating factorLP = lag phase (in seconds)AV = initial velocity of aggregation (units/min)Am = maximum aggregation in three minutes ('* p < 0.05 ** p < 0.01

Platelets are the first line of defensein preventing blood loss from the cir-culation when vascular damage occurs;poor responsiveness of platelets toagonists would therefore predispose toincreased blood loss. Vascular injuryas a result of perfusion/pressurechanges associated with exercise is alikely event in any horse, but particu-larly in a horse with concomitant lungpathology as has been described inEIPH (17).

Bayly et al (10) observed thatplatelets derived from clinically normalThoroughbred horses failed to aggregate

%7o of maximum possible)

as readily to ADP after the animalshad been exercised. They noted how-ever that the aggregation response toa more potent agonist (arachidonicacid) was not affected by exercise. Itwas postulated in that study thatexercise-induced release of prostacyclinfrom the vascular endothelium wasresponsible for the inhibitory effect onplatelet reactivity. This inhibitoryeffect was detected with ADP as theagonist since it is a weak platelet acti-vator and therefore "a more sensitiveindicator of changes in the responsivecapacity of platelets". Our post-

104

exercise studies on both EIPH-positiveand EIPH-negative Standardbredhorses are in agreement with these pre-vious observations. In the presentstudy, exercise suppressed the responseof platelets to low dose ADP. Res-ponses to more potent agonists (i.e.higher concentration of ADP, or col-lagen or PAF) were not affected byexercise.

Like previous researchers we wereunable to demonstrate any significantdifferences in coagulant activity (asassessed by PT/PTT screening tests),or fibrinolytic activity as assessed byquantitation of FDP titers, in restedEIPH-positive and EIPH-negativehorses (8,9,17). Although affectedStandardbred horses in this studytended to have a lower platelet countthan nonaffected horses, this dif-ference was not significant. Inhumans, strenuous exercise has beenshown to induce a significant increasein fibrinolytic activity (19). Bayly et al(11) found no evidence of enhancedfibrinolysis in Thoroughbred horsesexercised at racing speed for 1200 mand then evaluated 5 min after exer-cise. Ferguson et al (20) found noevidence of excessive fibrinogenolysis/fibrinolysis in horses competing in a157 km endurance race. In the presentstudy, we found no detectable evidenceof enhanced fibrinolysis in exercisedclinically normal Standardbred horses.Moreover, EIPH-positive horses alsoshowed no evidence of enhanced post-exercise fibrinolysis. Whether theapparent lack of enhanced postexercisefibrinolysis in horses represents aspecies difference in human/equineresponses, or is related to inadequateduration and/or intensity of exercise,is not known. Our study suggests how-ever that alterations in fibrinolyticactivity associated with exercise and/ordifferences in fibrinolytic activitybetween EIPH-positive and EIPH-negative horses are not likely factorsin the pathogenesis of the bleedingassociated with this disease.Antithrombin III is a major natural

inhibitor of blood coagulation. Thehypothesis that abnormally high ATIIIactivity might be a factor in EIPH wasdisproven as similar plasma levels werefound in both EIPH-positive andEIPH-negative horses. Likewise, exer-cise had no effect on ATIII levels ineither group.

The plasma factor VIII/vonWillebrand factor (FVIII/vWF) com-plex was studied because vascularinjury has been suggested as a basiccomponent in the pathogenesis ofEIPH. The protein vWF is a productof vascular endothelium but FVIII isnot. Although the plasma levels ofFVIII and vWF often tend to fluctuatein parallel, a significant increase invWF without a concomitant increasein FVIII (i.e. increased vWF:Ag/FVIII:C ratio) has been described inhumans in association with primarydiseases affecting the vasculature(21,22). Disproportional increases invWF:Ag (increased vWF:Ag/FVIII:Cratios) were reported in children withacute bronchiolitis and were attributedto pulmonary endothelial damage dueto hypoxia. Ratios became normalduring the recovery stage (22). Othershave proposed that the vWF:Ag/FVIII:C ratio be used as a means ofmonitoring vascular injury (21). Nosimilar studies however have beenreported in horses. We observed nor-mal plasma concentrations of FVIII:Cand vWF:Ag in both EIPH-positiveand EIPH-negative horses in thisstudy, and similar vWF:Ag/FVIII:Cratios at rest. This may reflect a lackof "active" concurrent lung/vascularpathology and vascular damage inEIPH-positive horses at the time oftesting, although tissue injury mayhave occurred earlier.

Increases in plasma FVIII:C activ-ity and vWF:Ag following exercisehave been extensively documented inhumans (23,24). A similar effect inhorses was evident in this study as bothplasma FVIII:C and vWF:Ag increasedfollowing exercise. There was no dif-ference in the degree of increase whenEIPH-positive and EIPH-negativehorses were compared. Since theincreases in FVIII:C and vWF:Ag weresomewhat in parallel, the vWF:Ag/FVIII:C ratios in both groups ofhorses were not significantly altered byexercise.

In conclusion this study of restedhorses with EIPH points towards anunderlying platelet function abnor-mality as a possible contributing fac-tor in the pathogenesis of EIPH. It canbe postulated that this underlyingplatelet defect may be exacerbated byexercise (as may the clinical expressionof this defect), when the requirement

for good platelet function is high.Bleeding may be prolonged andtherefore blood volume loss increasedby delayed sealing of injured micro-vessels. The present study also indicatesthat alterations in coagulant andfibrinolytic activities are probably notimportant in the pathogenesis of EIPH.

ACKNOWLEDGMENTS

We wish to acknowledge the technicalsupport provided by Ms. Vivian Collett,Mrs. Jennifer Keen and Ms. DebraGaon during their respective summerstudentships.

REFERENCES

1. PASCOE JR, FERRARO GL, CANNONJH, ARTHUR RM, WHEAT JD. Exercise-induced pulmonary hemorrhage in racingThoroughbreds: A preliminary study. Am JVet Res 1981; 40: 701-706.

2. MASON DK, COLLINS EA, WATIaNS KL.Exercise-induced pulmonary hemorrhage inhorses. In: Snow DH, Persson SGB, Rose BJ,eds. Equine Exercise Physiology. Cambridge:Granta Editions, 1983; 57-63.

3. SWEENEY CR, SOMA LR. Exercise-inducedpulmonary hemorrhage in horses after differentcompetitive exercises. In: Snow DH, PerssonSGB, Rose BJ, eds. Equine Exercise Physiol-ogy. Cambridge: Granta Editions, 1983; 51-56.

4. PASCOE JR. Exercise-induced pulmonaryhemorrhage. J Equine Vet Sci 1989; 9:198-200.

5. O'CALLAGHAN MW, PASCOE JR,TYLER WS. Exercise-induced pulmonaryhaemorrhage in the horse: results of a detailedclinical, postmortem and imaging study. I.Clinical profile of horses. Equine Vet J 1987;19: 384-388.

6. O'CALLAGHAN MW, PASCOE JR,TYLER WS. Exercise-induced pulmonaryhaemorrhage in the horse: results of a detailedclinical, postmortem and imaging study. VIII.Conclusions and implications. Equine Vet J1987; 19: 428-434.

7. O'CALLAGHAN MW, PASCOE JR,TYLER WS. Exercise-induced pulmonaryhaemorrhage in the horse: results of a detailedclinical, postmortem and imaging study. VII.Ventilation/perfusion scintigraphy in horseswith EIPH. Equine Vet J 1987; 19: 423427.

8. JOHNSON JH, GARNER HE, HUTCHIN-SON DP, MERRIAM JG. Epistaxis. Proc AmAssoc Equine Pract 1973; 19: 115-121.

9. BAYLY WM, MEYERS KM, KECK MT,HUSTON LJ, GRANT BD. Effects of exer-cise on the hemostatic system of Thoroughbredhorses displaying post-exercise epistaxis. JEquine Vet Sci 1983; 3: 191-193.

10. BAYLY WM, MEYERS KM, KECK MT,HUSTON LJ, GRANT BD. Exercise-inducedalterations in haemostasis in Thoroughbredhorses. In: Snow DH, Persson SGB, Rose BJ,

105

eds. Equine Exercise Physiology. Cambridge:Granta Editions, 1983; 336-343.

11. BAYLY WM, MEYERS KM, KECK MT,HUSTON UJ, GRANT BD. Effects offusosemide on exercise-induced alterations inhaemostasis in Thoroughbred horses exhibitingpost-exercise epistaxis. In: Snow DH, PerssonSGB, Rose BJ, eds. Equine ExercisePhysiology. Cambridge: Granta Editions, 1983;64-70.

12. JAIN NC. Schalms Veterinary Hematology.4th ed. Philadelphia: Lea & Febiger, 1986.

13. JOHNSTONE IB, CRANE S. Hemostaticabnormalities in equine colic. Am J Vet Res1986; 47: 356-358.

14. JOHNSTONE IB, CRANE S. The effects ofdesmopressin on plasma factor VIII/vonWillebrand factor activity in dogs with vonWillebrand's disease. Can J Vet Res 1987; 51:189-193.

15. JOHNSTONE IB, PETERSEN D, CRANES. Antithrombin III (ATIII) activity in plasmasfrom normal and diseased horses, and in nor-

106

mal canine, bovine and human plasmas. VetClin Pathol 1987; 16: 14-18.

16. JOHNSTONE lB. Comparative effects ofphenylbutazone, naproxen and flunixinmeglumine on equine platelet aggregation andplatelet factor 3 availability in vitro. Can JComp Med 1983; 47: 172-179.

17. O'CALLAGHAN MW, PASCOE JR,TYLER WS. Exercise-induced pulmonaryhaemorrhage in the horse: results of a detailedclinical, post-mortem and imaging study. V.Microscopic observations. Equine Vet J 1987;19: 411418.

18. COOK WR. Epistaxis in the racehorse. EquineVet J 1974; 6: 45-58.

19. OGSTON D, FULTERTON HW. Changes infibrinolytic activity produced by physical activ-ity. Lancet 1961; ii: 730-733.

20. GUSON EW, BERNIER LL, SHAUGH-NESS GP, BOUCHER JH. Fibrinolytic activ-ity with fibrinogenolysis during long-distanceracing in horses. J Appl Physiol: Respir

Environ Exercise Physiology 1981; 50:245-249.

21. CORDA R, ALBERTI M, COACCI L,PUTZOLV G, MANNUCCI PM. Anincreased Factor VIII antigen as an indicatorof endothelial damage in measles. Thromb Res1979; 14: 805-810.

22. MAZZONE D, FICHERA A, PRATICO G,SCIACCA F. Factor VIII activity and factorVIII-related antigen in infants with viral bron-chiolitis. N Engl J Med 1984; 311: 1257-1258.

23. BENNETT B, RATNOFF OD. Changes inantihemophilic factor (AHF, Factor VIII) pro-coagulant activity and AHF-like antigen in nor-mal pregnancy and following exercise andpneumencephalography. J Lab Clin Med 1972;80: 256-263.

24. STIBBE J. The effects of exercise on FVIII-complex: Proportional increase of ristocetincofactor (von Willebrand factor) and FVIII-Agn, but disproportional increase of FVIII-AHF. Thromb Res 1977; 10: 163-168.