Pre- and post-race serum cardiac troponin T concentrations in Standardbred racehorses

(1)

Pre- and post-race serum cardiac troponin T concentrations in Standardbred racehorses

I.R. Hellings

^a,

*, R. Krontveit

^b

, M. Øverlie

^c

, A. Kallmyr

^c

, T. Holm

^d

, C. Fintl

^a

aNorwegianUniversityofLifeSciences,FacultyofVeterinaryMedicineandBiosciences,DepartmentofCompanionAnimalClinicalSciences,PostBox8146 Dep.,0033Oslo,Norway

bNorwegianMedicineAgency,PostBox240Skøyen,0213Oslo,Norway

cRikstotoklinikkenBjerke,PostBox194Økern,0510Oslo,Norway

dEvidensiaLørenskogDyreklinikk,Solheimveien56,1473Lørenskog,Norway

ARTICLE INFO

Keywords:

CardiactroponinT Equine

Exercise Horse

Referenceinterval

ABSTRACT

Elevated cardiactroponin T(cTnT) concentrationsmay provide evidenceof myocardial injurybut physiologicalpost-exercisereleasealsooccurs.Referenceintervalsarenotfullyestablishedinhorses makinginterpretationdifficult.Theaimsofthisstudyweretoestablishanupperreferencelimitfor serumcTnT,comparepre-andpost-raceserumcTnTconcentrations,andtoevaluatefactorsthatmay influence thesein apopulation of healthy,race-fit Standardbred racehorses.Serum samples were collectedpre-(n=108)and1–2hpost-racing(n=101)andanalysedusingahighsensitivity-cTnTassay.

Referencelimitswith90%conﬁdenceintervals(CI)werecalculatedbynon-parametricmethodsusingthe bootstrapmethod.Effectsofsex,age,racingspeed,distance,placingsandtracksurfacewereassessedby ﬁttinggeneralizedlinearmodelswithanidentitylinkfunctionandinverseGaussiandistribution.

TheupperreferencelimitforserumcTnTconcentrationwas27.4ng/L(90%CI13.1–32.0).Themedian serumcTnTconcentrationwassigniﬁcantlyhigher1–2hpost-racingcomparedtopre-racing(P<0.001).

AgeandsexdidnotsigniﬁcantlyaffectserumcTnTconcentrationspre-racing(P=0.5andP=0.11).

CardiactroponinTconcentrationsweresignificantlyhigherpost-racinginfemales(P=0.018).Racing speedandplacingshadnoeffectonserumcTnTconcentrationspost-race(P=0.71andP=0.66).The studycontributedtowardsestablishinganupperreferencelimitforserumcTnTconcentrationsina populationofrace-fitStandardbredsandevaluatedfactorsthatmayhaveinfluencedtheresultsobtained.

Introduction

Cardiacdiseaseisincreasinglyrecognizedasa causeof poor performanceaswellasofsuddendeathinhorses(Brownetal., 1988;Bodenetal.,2006;Lyleetal.,2011;Reefetal.,2014).While echocardiographyandexercisingelectrocardiographyarewidely availableinequinepractice,thesetechniquesarenotsensitivein detectingsubclinicalcardiacdiseaseincludingmyocardialinjury (Reefetal.,2014).

Cardiac troponins (cTn) are a group of cardiac regulatory proteins unique to the myocardium (Sharma et al., 2004). In humans,bothcardiactroponinI(cTnI)andtroponinT(cTnT)have beenshowntobeusefulmarkersindiagnosingmyocardialinjury (Sharmaetal.,2004;Thygesenetal.,2010).However,releaseofcTn duringandafterexerciseinapparentlyhealthyhumansisalsowell

documented(Vilelaetal.,2014;Klinkenbergetal.,2016)andvary with ﬁtness level, type and duration of exercise and certain physiological factors (Shave et al., 2010). Exercise-associated increases in cTnT typically peak within few hours (2 5 h) (GresslienandAgewall,2016)post-exerciseandreturntobaseline within24handdifferfromtheprolongedelevatedconcentrations typically seen in humans with myocardialinjury (Shave et al., 2010; Donnellan and Phelan, 2018). After a marathon, 78% of runnershadcTnTabove0.01ug/L(10ng/L)while36%hadcTnT concentrationsabovethecutoffvaluesforacutemyocardialinfarct inhumansof5ng/L(Shaveetal.,2007).

TheuseofcTnIinequineclinicalpracticeiswellestablished with normal reference intervals (RIs) available (Nostell and Haggstrom, 2008; Slack et al., 2012; Van Der Vekens et al., 2015a). Furthermore, cTnI concentrations have been shown to increase or remain unchanged after exercise (Nostell and Haggstrom,2008;Slacketal.,2012;Rossietal.,2019).However, different assays for measuring cTnI exist, and reference values indicativeofmyocardialinjuryvarybetweenassays(Sharmaetal.,

* Correspondingauthor.

E-mailaddress:[email protected](I.R. Hellings).

http://dx.doi.org/10.1016/j.tvjl.2020.105433

ContentslistsavailableatScienceDirect

The Veterinary Journal

j o u r n a l h o m e p a g e : w w w . e l s ev i er . c o m / l o c a t e/ tv j l

(2)

2004; Rossi et al., 2014; Van Der Vekens et al., 2015b). Also, differentcTnIassayshavedifferentsensitivitiesandmaytherefore be unsuitable to detect subtle myocardial injury necessary in clinicalsportsmedicine(Phillipset al.,2003;Beggetal.,2006;

Rossietal.,2014;VanDerVekensetal.,2015a).

To address these limitations, a high sensitivity troponin T (cTnT-hs)assay has beendevelopedand was recently validated for use in horses (Shields et al., 2016). Using this assay, cTnT concentrationsfromhealthyleisurehorses(VanDerVekensetal., 2015b;Shieldsetal.,2016)andracingThoroughbreds(Shieldsetal., 2016)havebeenpublished.Usingthesameassay,cTnTconcen- trationsinracingThoroughbredswere signiﬁcantlyhigheratall samplingpointsbetween2and6hpost-racecomparedtopreracing, peakedat3handreturnedtobaselinebetween12to24h(Shields etal.,2017).There is currentlynopublisheddataexploring the normalRIforcTnTinStandardbredracehorses,andtheeffectof racingoncTnTconcentrationsinthisbreed.

The speciﬁc aims of this study were toestablish an upper reference limit for serum cTnT, to compare pre-and post-race serumcTnTconcentrations,andtostudythepotentialeffectofsex, age, racing speeds and distance, placings and track surface conditions on serum cTnT concentrations in a population of healthyrace-ﬁtStandardbredracehorsesusingacTnT-hsassay.

Materialsandmethods Studydesign

Thestudywasaprospective,observationalclinicalstudywithconvenience bloodsamplingofrace-ﬁtStandardbredracehorsespre-andpost-racingcollected atasingleracetrackinOslo,NorwayduringofﬁcialracesheldbytheNorwegian RacingAuthority.DatawerecollectedbetweenDecember2017andMarch2018and thestudywasapprovedbytheethicscommitteeofTheNorwegianFoodAuthority Approvalnumber,14,190;Approvaldate,24November2017andbytheNorwegian RacingAuthorityApprovaldate,24October2017.

Horses

Informedconsentwasobtainedfromtheownersorownerrepresentativesprior toexaminationandsamplecollection.HealthyandactivelyracingStandardbred racehorsesofanyageandsexwereeligibletoenterthestudy.

Tobeincluded,horseshadtobedeemedfittoracebythetrainerbasedontheir training performance and general health status. Cardiac auscultation and assessmentofqualityandregularityoftheperipheralpulsebypalpationofthe facialarterywasperformedpriortoandaftertheraceinallhorsesbyexperienced equineveterinarians. Writtenrecords were madeonheart rate, rhythm and characterizationofanycardiacmurmurspresent.Horseswithmurmursofgrade 3/6wereexcluded.Normalcardiacrhythmwasdefinedasaregularrhythmatrest orasaregularlyirregularrhythmthatdisappearedfollowingexercise.Horseswith arrhythmiasotherthanthosedefinedwereremovedfromthestudyandfurther evaluationrecommended.

Racingspeed(averagetime/km),distance(1609mor2100m)andtracksurface conditionatthedayofracing(firm orheavy)aswellasplacing intherace wererecordedforeachhorse.Thetracksurfaceconditionwasdefinedbasedon theofficialdescriptiongivenbytheracetrackauthoritiesonanygivenraceday.

Samplecollection

Bloodsampleswerecollectedbyvenepunctureofthejugularveinintoserum tubeswithseparationgelbeforeracing,1–2hfollowingracingandinasubsetof horses,24(22–24)hfollowingracing.Followingcollectionandcoagulationofthe samples,tubeswerecentrifugedat3000gfor10min.Thetubeswereimmediately refrigeratedandtransportedchilledtothelocalhospitalforsamedayanalysis.

Laboratorymethods

SampleswereanalysedusingtheRocheElecsys troponinT-high-sensitivity (TnT-hs)assay(RocheDiagnostics)ontheCobase801Analyser(RocheDiagnostics) complyingwiththemanufacturer’sinstructionsandtheClinicalandLaboratory StandardsInstitute(CLSI)aspreviouslyevaluatedinhorses(Shieldsetal.,2016).The degreeofhaemolysiswasassessedusingtheserumindicesontheRocheCobase801 Analyser(RocheDiagnostics).Nosamplesreachedthresholdlevelsforexclusion.Daily calibrationandqualitycontrolprocedureswereperformedpriortosampleanalysis.

Usingthismethod(9minapplication),thelowestlevelofquantiﬁcation(LoQ)was 5.0ng/Landthelowestlevelofdetection(LoD)was3.0ng/L.

Statisticalanalysis

As the data was non-Gaussian distributed and not symmetrical a non- parametricmethodwasusedtodeterminetheupper97.5percentilereferencelimit forserumcTnT.The90%CIaroundtheupperreferencelimitwascalculatedusing thebootstrapmethodbyReferenceValueAdvisorfreewarev2.1¹.inaccordance withrecommendations(CLSI(ClinicalandLaboratoriesStandardsInstitute),2010;

Friedrichs et al., 2012; Geffre et al., 2011; Ozarda, 2016). Sera with cTnTconcentrations<5.0ng/Lwereforthepurposeofstatisticalanalysissetas 4.9 ng/L. Visual inspection of histograms was used to assess distribution.

HistogramswerealsousedalongwithclinicalexperienceandthestatisticsTukey and Dixon to assess outlying observations. As the data were not normally distributed,mediansarepresented.

DifferencesinmedianserumcTnTconcentrationspre-andpost-racewere assessedusingWilcoxonmatched–pairssigned–rankstests.Anyeffectsofsex,age, racingspeed,racingdistance,tracksurfaceandraceplacings(first,second,thirdand fourthplace)onserumcTnTconcentrationswereunivariablyevaluatedfitting generalized linear models with an identity linkfunction and an appropriate distribution.Evaluation ofthedistribution oftheAnscomberesidualsandthe distributionof thecTnTconcentrationswereboth appliedintheselection of distribution.InverseGaussiandistributionwasselected.Entiremalesandgeldings were groupedtogether andcompared to females.Continuous variables were assessed usinglowesscurvesand categorizedif deviationfrom linearitywas detected.Age,racingspeedandtimingofthepost-racesample(minutespassed from race to sampling) were evaluated as continuous variables. Statistical significancewassetatP<0.05.

ThevariablesweretestedforcollinearitybyGoodmanandKruskal’sgammafor ordinal and dichotomous variables and pairwise correlations for continuous variables.Associations>0.7or< 0.7wereconsideredevidenceofcollinearity.

VariableswithaP0.20,providedtherewerenocollinearitybetweenthem,were thenconsideredforfurtheranalysisinamultivariablegeneralizedlinearmodelto assesstheeffectonpost-raceserumcTnTconcentration.Biologicalplausibilityof signiﬁcantﬁndingswasevaluatedbyprioriknowledgeandcausaldiagrams.The multiple Wald test was usedto evaluate differences between categories of categoricalvariables.VariableswerekeptinthemodelifP<0.05.

StatisticalanalyseswereperformedusingStata15(StataCorp).

Results

Onehundred-andtenhorsesfrom54differenttrainingyards primarilylocatedinthesouth-eastofNorwaywereenrolledinthe study.Onehorsewasexcludedduetoaclinicalhistoryofrecurrent rhabdomyolysisandabovenormalRIconcentrationsforaspartate aminotransferaseandcreatininekinase(CK)concentrations.One horsewasdeﬁnedasanoutlierandremovedfromthestatistical analysis because it had a pre-race serum cTnT concentration (76ng/L)morethantwicethemaximumvalueofothersamples fromthegroup.

The pre-race samples therefore included serum from onehundred and eight Standardbred racehorses of which 40 were femalesand68males,thereof53geldings.Themedianageforall horseswas6(range3–11),forfemales6(range3–11)andformales 6(range3–11)years,respectively.

One-hundred and one horses, thereof 37 females and 64 males were sampled 1–2 h after racing and 18 horses were additionallysampledat22–24hpost-racing.Ofthe101horses sampledpost-racing,15horsesracedovertheshorter(1609m) distance, while 86 raced over the longer (2100 m) distance.

Sixty-eight horses raced when track surface conditions were classiﬁed as ﬁrm, while the remaining 33 horses raced on a heavy track surface. Descriptive statistics for serum cTnT concentrations pre-race, 1–2 h post-race and 22–24 h post- raceandupperlimitforRIwith90%CIforthepre-andpost-race concentrationsareshowninTable1.

Overall,medianserumcTnTconcentrationsweresigniﬁcantly higher 1–2 h post-racing compared to pre-racing (P < 0.001) (Fig.1).ThemedianserumcTnTconcentration22–24hpost-racing was not signiﬁcantly different to the pre-racing concentrations

1 See: Reference Value Advisor freeware v2.1direction insert. http://www.

biostat.envt.fr/reference-value-advisor/(AccessedJanuary15,2020)

(3)

(P=0.50).OnehorsewithaserumcTnTconcentrationof75ng/L24h post-racehadpre-raceand2hpost-raceserumcTnTconcentrations of<5.0and9.0ng/L,respectively.

Table 2 shows the univariable results from the generalized linearmodelsforpre-andpost-raceserumcTnTconcentrations.

Sex and age did not have a significant effect on serum cTnT concentrations pre-racing (P = 0.20 and P = 0.38 respectively) (Table2).Sexhadasignificanteffectonpost-racingmeasurements, with increased serum cTnT in females compared to males (P =0.018).Cardiac troponinTconcentrations werenotsignifi- cantlyhigherpost-racingwhenracingonaheavycomparedtoona firm surface track (P = 0.14), and when racing over 1609 m comparedto2100m(P=0.11)(Table 2).Racingspeedorplace positionhadnoeffectonserumcTnTconcentrationspost-racing (P=0.71andP=0.66respectively).SerumcTnTconcentrationwas notsignificantlyaffectedbythetimingofsamplecollectionwithin the1 2hintervalfollowingracing(P=0.82;Table2).Onlysexwas significantinmultivariablemodellingandhenceamultivariable modelwasnotbuilt.

Discussion

Thecurrentstudyhasprovideddatawhichcontributestowards establishinganupperreferencelimitforserumcTnTconcentrations in an apparently healthy population of race-ﬁt Standardbred racehorses.Signiﬁcantpost-raceincreasesinserumcTnTconcen- trationsalsooccurredinapparentlyhealthyhorses.

The 97.5th percentile upper reference limit for serum cTnT concentrationswas27.4ng/Landwasthushigherthanpreviously published 95th and 99th percentile upper reference limits in plasmafromnon-competitionhorses(6.8and16.2ng/L),butmore similartothosereportedfromchuckwagonracingThoroughbred geldings(14.0and23.2ng/L)(Shieldsetal.,2016).Thisindicates thatitmaybeappropriatetoapplydifferentreferenceintervalsin differentpopulations of horses.In humans, theabsoluteserum concentrationspre-andpost-exercisedependonindividualfactors suchas ageand sexas wellas ﬁtnesslevel, training,type and durationofexercise,timingofsamplingandlaboratoryanalytical factors(Shaveetal.,2010;Rossi etal., 2014)and mayapplyto horsesaswell.

CardiactroponinTconcentrationsweresigniﬁcantlyhigher1–2 hpost-racingcomparedtopre-racingandindicatesreleaseofcTnT intothecirculationduring,orimmediatelyfollowingracingThisis in agreement with a previous study in racing Thoroughbreds documenting signiﬁcantly higher serum cTnTconcentrations in horsessampled2–6hpost-race(Shieldsetal.,2017).

Arecentkineticstudyutilisingahigh-sensitivitycTnIassayin StandardbredracehorsesfoundpeakplasmacTnIconcentrations Table1

DescriptivestatisticsforcardiactroponinT(cTnT)inasampleofrace-ﬁtStandardbredracehorsespre-race,1–2hpost-raceand22–24hpost-raceincludingthe97.5%upper referencelimitwith90%conﬁdenceinterval(90%CI)forthepre-raceandpost-racecTnTconcentrations.

cTnT

Timesampled (n) Medianng/L Maximumng/L 97.5%upperreference

limit(90%CI)ng/L

Pre-race Allhorses(108) 4.9 32.0 27.4(13.1–32.0)

1–2h post-race

Allhorses(101) 6.0 24.0 23.5(18.4–24.0)

Males(64) 6.0 20.0 –

Females(37) 7.5 24.0 –

22–24h post-race

Allhorses(18) 4.9 75.0 –

Table2

Resultsfrom univariablegeneralizedlinear models ofpre–raceandpost–race cardiactroponinT(cTnT)concentrationsin asampleofrace-ﬁtStandardbred racehorses.

Variable Coefﬁcient(SE) P 95%CI^a

Pre–racecTnT(n=108) Sex

male baseline – –

female 1.22 0.20 0.64,3.08

Age(years) 0.18 0.38 0.59,0.22

Post-racecTnT(n=101) Sex

Male Baseline – –

Female 2.18 0.018 0.38,3.99

Age(years) 0.19 0.39 0.25,0.63

Racedistance

1609m Baseline – –

2100m 0.01 0.11 0.01,-0.001

Tracksurfaceconditions

Firm Baseline – –

Heavy 1.42 0.14 0.47,3.31

Racingspeed(min/km) 0.07 0,71 0.40,0.27

Raceposition 0.66^a

Firstplace Baseline – –

Secondplace 0.88 0.51 1.75,3.511

Thirdplace 1.91 0.22 1.14,4.95

Fourthplace 0.92 0.38 1.17,3.04

Minutespassedfromracetosample 0.004 0.82 0.04,0.03 95%CI,95%conﬁdenceinterval.

aOverallPvalueforcategoricalvariable,Waldtest.

Fig.1.BoxplotshowingconcentrationsofcardiactroponinT(cTnT)inasampleof race-ﬁtStandardbredracehorsespre-raceand1–2hpost-race.

(4)

alsooccur2–6hpost-exercise(Rossietal.,2019).Theseresults indicatethatcTnTandcTnImayhavesimilarreleasekineticspost- exerciseinhorses,inagreementwithapreviousstudyinhumans comparingcTnTandcTnIpost-exercise(Souetal.,2016).Further, post-exercisereleaseofbothcTnTandcTnIreturnedtopre-race concentrationswithin12–24hin healthyhorses(Shieldsetal., 2017;Rossietal.,2019).Althoughthecurrentstudydidnotinclude asufﬁcientnumberofhorsestoperformstatisticalanalysisforthe 24hsamples,medianserumcTnTconcentrationwasnotdifferent tothemedianpre-raceserumcTnTconcentration. Together,this may suggest that post-race increases in serum cTn may be expectedto be short-lived in the majority of horses, although individualvariationsmayoccur.Post-exercisephysiological cTn release in humans typically peak within 2 5 h(Gresslien and Agewall,2016)post-exerciseandreturntobaselinewithin24 h andmoderatereleaseofcTnmaybeseenasearlyas30mininto sustainedendurance exercise (Donnellan and Phelan,2018). In acute coronary syndrome on the other hand, an initial rapid increase in serum cTn is followed by a delayed and sustained releaseofcTnoccurring12 24h(Sharmaetal.,2004;Shaveetal., 2010;Rossietal.,2014)andtypicallyremainelevatedforatleast 4–7days(DonnellanandPhelan,2018).

Itiscurrentlyunknownwhetherexercise–inducedcTnrelease reﬂectsreversibleorirreversiblechangesinthecardiacmyocytes (Wu, 2017). However, the rapid clearance of Tn post-exercise suggestsa mechanism unrelated to myocyteinjury or necrosis (Shave et al., 2010) being a normal physiological response (Sabatine et al., 2009; Weippert et al., 2016). During exercise,catecholaminereleasemayinducetransientmyocardial ischemia,whileincreasedheartratesandbloodvolumemaycause mechanical stretching of the cardiac myocyte and dehydration andacid-baseimbalancesmaycausecelldamage(Weippertetal., 2016) resulting in a short-lived elevated circulating cTn concentrations.

Thepost-raceserumcTnTconcentrationsinthecurrentstudy were significantly higher in females compared to males. The populationofhorsesinthepreviousstudyofcTnT(Shieldsetal., 2017)comprisedgeldingsonlyandthereforenocomparisonswere madebetweenthedifferentsexes.However,inhumanathletes, bothage(Tianetal.,2012)andsex(Kongetal.,2017)influence post-exercise cTnT concentrations; male runners having larger increasesincTnTconcentrationsfollowingexercisecomparedto females.Thereason forthedifferencein exercise-inducedcTnT releasebetweensexhasnotbeenestablished,butdifferencesin hormones and cardiac muscle mass has beensuggested (Kong etal.,2017).Confounderssuchasrace-distanceandtracksurface weresuspectedbutonlyweaklychanged(reduced)theeffectof female. Whether the difference in post-race serum cTnT concentrationsbetweenfemalesandmalesinthecurrentstudy wascausedbysex-relateddifferencesinresponsetoexerciseor due to some other unidentified exercise-related confounding factor,isnotknown.

Inthecurrentstudy,therewasnocorrelationbetweenserum cTnT concentration and age, in agreement with previously reportedcTnIvalues inthe horse(Slacket al.,2012).However, thehorsesparticipatinginStandardbredracesaretypicallywithin a narrow age bracket, and to investigate whether there is a correlationbetweenageandcTnTconcentrationsastudyincluding awiderrangeofagesisneeded.Thiscouldbeachallengegiventhe naturalwastageofhorsesintrainingastheygetolderandalsothe upperagelimitsetbymostracingjurisdictionforwelfarereasons.

In thecurrentstudy, nosigniﬁcantdifferenceinserumcTnT concentrationswasfoundbetweenthetworace-distancestested (1609and2100m).Bothdistancesinvolveshortburstsofsprint exercisetypicallylasting80–140s.Thereisconﬂictingevidencein humans regarding which type of exercise provokes the largest

increaseincTnandmoststudieshavebeenperformedafterlong distancerunningsuchashalforfullmarathons(Vilelaetal.,2014).

Asracehorsestypicallyexerciseatmaximumcapacityforonlya fewminutes,directcomparisontohumansmaynotbeapplicable (Slacketal.,2012).

Onehorsewithamarkedlyelevated24hpost-raceserumcTnT concentrationof75ng/Lneverthelessperformedaccordingtothe trainer’s expectations. The owner declined further diagnostic work up and it is not possible torule out underlying cardiac pathology causing the increasein serumcTnT 24 hpost-race.

However, thehorsesubsequently wona raceonly 7days later making this less likely. Prolonged increased concentrations of cTnT in humans are generally attributable to irreversible myocardial damage, particularly ischemic myocardial injury (Wu,2017).However,humanpatientswithmyocardialischemia hadsigniﬁcantlyhigherconcentrationsofbloodcardiactroponins both before and after exercise compared to patients without ischemia(Souetal.,2016)anditisrecommendedthatsequential bloodsamplesaretakentofullyelucidateunderlyingcausesof increasesincTnTinresponsetoexerciseasasustainedelevated cTnTconcentrationsmaybeexpectedincardiacpatients(Shave etal.,2010;DonnellanandPhelan,2018;Thygesenetal.,2018).

Theabove-mentionedhorsehadapre-racecTnTbelowLoDof5.0 ng/Landbelowthepublishedrecommendedcut-offlevelof6.6 ng/L (Van Der Vekens et al., 2015b)proposed to differentiate primary myocardial disease from normal myocardium in the horse(VanDerVekensetal.,2015b).However,thiscut-offvalueis within thereference range described in thecurrent studyand maynotbeappropriateforthecurrentpopulation.Further,using thiscut-offvalue,14horsesinthecurrentpopulationwouldhave beenclassifiedashavingprimarymyocardialdisease.Thehorses includedintheoriginalstudybyVanderVekensetal.(2015b), werenotinactiverace-training, andfactorsknowntoincrease cTnT concentrations in humans, such as fitness levels and dehydration,were not evaluatedand mayhave influenced the results. Importantly, 21 of the 23 horses defined as primary myocardial disease had increased CK and suspect atypical myopathy (Van Der Vekens et al., 2015b). Although cTnT reportedly is cardiac specific in mammals (Katrukha, 2013), cross-reactivitytoskeletaltroponinT(sTnT)hasbeenreported with first generation cTnT assays but use of more specific antibodiesin the newgenerationscTnT-hs assaysshould limit this effect. However, in humans with nonspecific muscle pathology, damaged skeletalmuscle expressed cTnT using the detection antibodies in the hs-cTnT assay (Jaffe et al., 2011).

The authorsthereforeconcluded thatincreased cTnTvaluesin non-suspect cardiac cases should be interpreted with caution especiallywhenoccurringsimultaneouslywithskeletalmuscle pathology. It is possible that underlying skeletal muscle pathology may have interfered with interpretation in the referencedstudyabove(VanDerVekensetal.,2015b).

The current study was limited by the numberof animals includedinthestudy.Accordingtorecommendedguidelinesfor establishing RIs (CLSI (Clinical and Laboratories Standards Institute),2010;Friedrichsetal.,2012),samplingofmorethan 120individualsisrecommended.Whilethemajorityofhorsesin the current study(89/108) hadserum cTnT pre-race concen- trationsbelowtheLoQoftheassay(5.0ng/L),fourhorseshad serumcTnTconcentrationsabove20ng/L(32.0,31.0,26.0and 22.0 ng/Lrespectively) andwere initiallyregardedasoutliers However,astherewerenoobviousreasonsforthehighserum cTnTconcentrations otherthan probable randomvariation in this ﬁeld study, they were included in the ﬁnal analysis followingpreviouslypublishedguidelines(Geffreetal.,2011).

Thisresultedinskeweddataandadegreeofuncertaintyaround the upper reference limit. The upper reference limit for the

(5)

serum cTnTconcentrations therefore needs to be interpreted with caution. A complete screening for underlying cardiac diseaseorotherhealthconcernswerenotwithinthescopeof this studyand it ispossible that individuals with subclinical diseasemayhavebeen included.Physiological factorssuchas fitness levels and dehydration status influence cTnT concentrations(Shaveetal.,2010)andmayalsohaveinfluenced theresultsinthecurrentinvestigation.Alargerstudyandwitha more extensive pre-race screening may help eliminate such animals and help firmly establish the serum cTnT upper referencelimits.

Beingaﬁeldstudy,itwasnotpossibletostandardisethetiming of post-raceblood samplingsince it was often not possible to persuadeowners/trainertostaybehindafterracinghadﬁnished.

However,theactualtiminginminutesdidnotsigniﬁcantlyaffect post-race serum cTnT concentrations within the maximum predeﬁned intervalof 2h,but thismayhavebeentooshortto detect peak serum cTnT concentrations according to human studies (Donnellan and Phelan, 2018). Further studies are necessaryto determine thekineticsof serum cTnTrelease and eliminationpost-exerciseindifferentpopulationsofhorses.

Differentcut-offvaluesforreportingminimumcTnTconcen- trationsareappliedbydifferentlaboratoriesandstudiesinhorses (Van DerVekens et al.,2015a, b;Shieldset al.,2017).TheLoQ denotestheconcentrationabovewhichquantitativeresultsmay beobtainedwithaspecificdegreeofconfidenceand shouldbe higherthanthelowerlimitofdetection(LoD;CLSI(Clinicaland LaboratoriesStandardsInstitute),2010;Shieldsetal.,2016).Inthe current study, a LoQ of 5.0 ng/L was applied based on the manufacturer’sstandardsonhumansamples,whiletheLoQof3.0 ng/LreportedbyShieldsetal.(2016)wasvalidatedinhorses.In the validation study, 43.2 and 11.2% of non-competitive and competitive horses respectivelyhad cTnTconcentrations below theLoQof3.0ng/L(Shieldsetal.,2016).Thisimpliesthatmany horseshaveacTnTconcentrationthatisbelowtheLoQofthisassay of3.0ng/Lsincelowerconcentrationscouldnotbereporteddueto manufacturer’sdefinedlimitsoftheassay.Inthecurrentstudy, 89horses(82%)hadserumcTnTconcentrationsbelowtheLoQof 5.0 ng/L pre-race and a true lower reference limit cannot be determined. In addition, useof anti-coagulants in plasma may resultinlowerconcentrationsofcTnmeasuredwhileincomplete coagulationand freefibrin in serum maynon-specificallybind assay-antibodies and falselyelevatecTn measured(Rossi etal., 2014). Correct sample collection and handling are therefore paramount.

Conclusions

The current study has contributed towards establishing an upper reference limit for serum cTnT concentrations in a populationof race-fitStandardbredsand evaluated factorsthat mayinfluencethese.Thefindingsmayprovideclinicalguidelines when collectingand interpreting serumcTnTconcentrations in horses.

Conﬂictofinterest None.

Acknowledgements

TheauthorswouldliketoacknowledgeLailaFureforassistance withthelaboratoryanalysis.Preliminaryresultswerepresentedas anAbstractatthe11thInternationalCongressofEuropeanCollege ofEquineInternalMedicineCongress,Ghent7-10November2018.

References

Begg,L.M.,Hoffmann,K.L.,Begg,A.P.,2006.SerumandplasmacardiactroponinI concentrationsinclinicallynormalThoroughbredsintraininginAustralia.

AustralianVeterinaryJournal84,336–337.

Boden,L.A.,Anderson,G.A.,Charles,J.A.,Morgan,K.L.,Morton,J.M.,Parkin,T.D., Slocombe,R.F.,Clarke,A.F.,2006.Riskoffatalityandcausesofdeathof ThoroughbredhorsesassociatedwithracinginVictoria,Australia:1989-2004.

EquineVeterinaryJournal38,312–318.

Brown,C.M.,Kaneene,J.B.,Taylor,R.F., 1988.Suddenandunexpecteddeathinhorses andponies:ananalysisof200cases.EquineVeterinaryJournal20,99–103.

CLSI(ClinicalandLaboratoriesStandardsInstitute),2010.Deﬁning,Establishingand VerifyingReferenceIntervalsintheClinicalLaboratory;ApprovedGuideline- thirdEdition,3rded Wayne,PA,USA.

Donnellan,E.,Phelan,D.,2018.Biomarkersofcardiacstressandinjuryinathletes:

whatdotheymean?CurrentHeartFailureReports15,116–122.

Friedrichs,K.R.,Harr,K.E.,Freeman,K.P.,Szladovits,B.,Walton,R.M.,Barnhart,K.F., Blanco-Chavez,J.,2012.ASVCPreferenceintervalguidelines:determinationof denovoreferenceintervalsinveterinaryspeciesandotherrelatedtopics.

VeterinaryClinicalPathology41,441–453.

Geffre,A.,Concordet,D.,Braun,J.P.,Trumel,C.,2011.Referencevalueadvisor:anew freewaresetofmacroinstructionstocalculatereferenceintervalswith MicrosoftExcel.VeterinaryClinicalPathology40,107–112.

Gresslien,T.,Agewall,S.,2016.Troponinandexercise.InternationalJournalof Cardiology221,609–621.

Jaffe,A.S.,Vasile,V.C.,Milone,M.,Saenger,A.K.,Olson,K.N.,Apple,F.S.,2011.

Diseasedskeletalmuscle:anoncardiacsourceofincreasedcirculating concentrationsofcardiactroponinT.JournaloftheAmericanCollegeof Cardiology58,1819–1824.

Katrukha,I.A.,2013.Humancardiactroponincomplex.Structureandfunctions.

Biochemistry(Mosc)78,1447–1465.

Klinkenberg,L.J.,Luyten,P.,vanderLinden,N.,Urgel,K.,Snijders,D.P.,Knackstedt,C., Dennert,R.,Kietselaer,B.L.,Mingels,A.M.,Cardinaels,E.P.,etal.,2016.Cardiac troponinTandIreleaseaftera30-kmrun.AmericanJournalofCardiology118, 281–287.

Kong,Z.,Nie,J.,Lin,H.,George,K.,Zhao,G.,Zhang,H.,Tong,T.K.,Shi,Q.,2017.Sex differencesinreleaseofcardiactroponinTafterenduranceexercise.

Biomarkers:BiochemicalIndicatorsofExposure,Response,andSusceptibility toChemicals22,345–350.

Lyle,C.H.,Uzal,F.A.,McGorum,B.C.,Aida,H.,Blissitt,K.J.,Case,J.T.,Charles,J.T., Gardner,I.,Horadagoda,N.,Kusano,K.,etal.,2011.Suddendeathinracing Thoroughbredhorses:aninternationalmulticentrestudyofpostmortem ﬁndings.EquineVeterinaryJournal43,324–331.

Nostell,K.,Haggstrom,J.,2008.RestingconcentrationsofcardiactroponinIinﬁt horsesandeffectofracing.JournalofVeterinaryCardiology10,105–109.

Ozarda,Y.,2016.Referenceintervals:currentstatus,recentdevelopmentsand futureconsiderations.BiochemiaMedica26,5–16.

Phillips,W.,Giguere,S.,Franklin,R.P.,Hernandez,J.,Adin,D.,Peloso,J.G.,2003.

CardiactroponinIinpasturedandrace-trainingThoroughbredhorses.Journal ofVeterinaryInternalMedicine17,597–599.

Reef,V.B.,Bonagura,J.,Buhl,R.,McGurrin,M.K.,Schwarzwald,C.C.,vanLoon,G.,Young,L.

E.,2014.Recommendationsformanagementofequineathleteswithcardiovascular abnormalities.JournalofVeterinaryInternalMedicine28,749–761.

Rossi,T.M.,Pyle,W.G.,Maxie,M.G.,Pearl,D.L.,Physick-Sheard,P.W.,2014.Troponin assaysintheassessmentoftheequinemyocardium.EquineVeterinaryJournal 46,270–275.

Rossi,T.M.,Kavsak,P.A.,Maxie,M.G.,Pearl,D.L.,Pyle,W.G.,Physick-Sheard,P.W., 2019.Post-exercisecardiactroponinIreleaseandclearanceinnormal Standardbredracehorses.EquineVeterinaryJournal51,97–101.

Sabatine,M.S.,Morrow,D.A.,deLemos,J.A.,Jarolim,P.,Braunwald,E.,2009.

Detectionofacutechangesincirculatingtroponininthesettingoftransient stresstest-inducedmyocardialischaemiausinganultrasensitiveassay:results fromTIMI35.EuropeanHeartJournal30,162–169.

Sharma,S.,Jackson,P.G.,Makan,J.,2004.Cardiactroponins.JournalofClinical Pathology57,1025–1026.

Shave,R.,George,K.,Gaze,D.,2007.Theinﬂuenceofexerciseuponcardiac biomarkers:apracticalguideforcliniciansandscientists.CurrentMedicinal Chemistry.14,1427–1436.

Shave,R.,Baggish,A.,George,K.,Wood,M.,Scharhag,J.,Whyte,G.,Gaze,D., Thompson,P.D.,2010.Exercise-inducedcardiactroponinelevation:evidence, mechanisms,andimplications.JournaloftheAmericanCollegeofCardiology 56,169–176.

Shields,E.,Seiden-Long,I.,Massie,S.,Passante,S.,Leguillette,R.,2016.Analytical validationandestablishmentofreferenceintervalsfora’high-sensitivity’

cardiactroponin-Tassayinhorses.BMCVeterinaryResearch12,104.

Shields,E.,Seiden-Long,I.,Massie,S.,Leguillette,R.,2017.24-hourkineticsof cardiactroponin-Tusinga"high-sensitivity"assayinThoroughbred chuckwagonracinggeldingsafterraceandassociatedclinicalsampling guidelines.JournalofVeterinaryInternalMedicine32,433–440.

Slack,J.,Boston,R.C.,Soma,L.,Reef,V.B.,2012.CardiactroponinIinracing standardbreds.JournalofVeterinaryInternalMedicine.26,1202–1208.

Sou,S.M.,Puelacher,C.,Twerenbold,R.,Wagener,M.,Honegger,U.,Reichlin,T., Schaerli,N.,Pretre,G.,Abacherli,R.,Jaeger,C.,etal.,2016.Directcomparisonof cardiactroponinIandcardiactroponinTinthedetectionofexercise-induced myocardialischemia.ClinicalBiochemistry49,421–432.

(6)

Thygesen, K., Mair, J., Katus, H., Plebani, M., Venge, P., Collinson, P., Lindahl, B., Giannitsis, E., Hasin,Y.,Galvani,M.,etal.,2010.Recommendationsfortheuseofcardiactroponin measurementinacutecardiaccare.EuropeanHeartJournal31,2197–2204.

Thygesen,K.,Alpert,J.S.,Jaffe,A.S.,Chaitman,B.R.,Bax,J.J.,Morrow,D.A.,White,H.

D.,2018.Fourthuniversaldeﬁnitionofmyocardialinfarction(2018).Circulation 138,618–651.

Tian,Y.,Nie,J.,Huang,C.,George,K.P.,2012.Thekineticsofhighlysensitivecardiac troponinTreleaseafterprolongedtreadmillexerciseinadolescentandadult athletes.JournalofAppliedPhysiology113,418–425.

VanDerVekens,N.,Decloedt,A.,Sys,S.,Ven,S.,DeClercq,D.,vanLoon,G.,2015a.

EvaluationofassaysfortroponinIinhealthyhorsesandhorseswithcardiac disease.VeterinaryJournal203,97–102.

VanDerVekens,N.,Decloedt,A.,Ven,S.,DeClercq,D.,vanLoon,G.,2015b.Cardiac troponinIascomparedtotroponinTforthedetectionofmyocardialdamagein horses.JournalofVeterinaryInternalMedicine29,348–354.

Vilela,E.M.,Bastos,J.C.,Rodrigues,R.P.,Nunes,J.P.,2014.High-sensitivity troponinafterrunning–asystematicreview.NetherlandsJournalof Medicine72,5–9.

Weippert,M.,Divchev,D.,Schmidt,P.,Gettel,H.,Neugebauer,A.,Behrens,K., Wolfarth,B.,Braumann,K.M.,Nienaber,C.A.,2016.CardiactroponinTand echocardiographicdimensionsafterrepeatedsprintvs.moderateintensity continuousexerciseinhealthyyoungmales.ScientiﬁcReports6,24614.

Wu,A.H.B.,2017.Releaseofcardiactroponinfromhealthyanddamaged myocardium.FrontiersinLaboratoryMedicine1,144–150.