雙葉機械瓣膜流固耦合仿真

AnalysisofFlowFieldinMechanicalAorticBilea?etHeartValves

UsingFiniteVolumeMethod

FengZhou1.YuanYuanCui1.LiangLiangWu1.JieYang2.LiLiu3.

Received:16October2014/Accepted:8April2015

ΘTaiwaneseSocietyofBiomedicalEngineering2016

AbstractUnderphysiologicalconditions,theopening

andclosingofthelea?etsofanimplantedarti?cialheartvalve(AHV)affectsthebloodcomponentsandthereforemaycausevariouscomplicationstothepatientsuchashemolysisorplateletactivation.Inthispaper,acomputa-tional?uidmodelispresented.Theregionaldistributionof?owshearstressinanAHVisanalyzedusingcomputa-tional?uiddynamicsandAHVperformanceisevaluatedintermsofthevariationof?owvelocityandpressurewhenbloodpassesthelea?etsintheaorticvalve.TheresultssuggestthatforthedesignofamechanicalAHV,themaximumopeningangleandinternalori?cediametershouldbeincreasedtoimprovethe?uidstructureinterac-tionanddecreasethepossibilityofdamagetobloodcomponents.Finally,the?uidstressdistributionoftheAHVlea?etstructurewascalculatedandanalyzedunderpulsating?owconditions.

1

2

3

4

5

KeyLaboratoryofAdvancedTechnologiesofMaterials,

ChineseEducationMinistry,SchoolofMaterialsScienceandEngineering,SouthwestJiaotongUniversity,

Chengdu610031,China

MechanicalEngineeringSchool,SouthwestJiaotongUniversity,Chengdu610031,China

NationalInstitutefortheControlofPharmaceuticalandBiologicalProducts,Beijing10050,China

LeibnizInstituteforPolymerResearch,MaxBergmannCenterforBiomaterials,01069Dresden,Germany

LawrenceBerkeleyNationalLaboratory,Berkeley,CA94704,USA

publishedonline:10February2016

Keywords：Bilea?etmechanicalvalves.Computational?uiddynamics.Blooddamage

1Introduction

Heartvalvesmaintainunidirectionalblood?owbytheopeningandclosingoftheirleaves,whichdependonthepressuredifferenceonthetwosidesofthevalvelea?ets.Ifanaturalheartvalveisdamagedbyseveredeformityordisease,itshouldbereplacedbyanarti?cialheartvalve(AHV),eithermechanicalorbiological,torecoverbio-logicalfunctionofblood?owtransport.Currently,over55%ofimplantedAHVsaremechanicalvalves[

HYPERLINK\l"bookmark1"

1

].Mechanicalvalvesconsistofmechanicalcomponentswithgoodmechanicalbehaviorandbiocompatibility.Amongthevarioustypesofmechanicalvalve,bilea?etmechanicalvalvesarethemostwidelyappliedclinically.

A?uid–structureinteractionanalysiscomputationmethodhasbeenappliedtoAHVs[

HYPERLINK\l"bookmark2"

2

].DeHartetal.[

HYPERLINK\l"bookmark3"

3

]analyzedthemovementvariationoflea?etswithtimeusinglaserDopplervelocimetryandahigh-speedcamera.Pelliccionietal.[

HYPERLINK\l"bookmark4"

4

]establishedatwo-dimensional(2D)digitalmodeltoanalyzesingle-andbi-lea?etvalvesinaorticpositionusingadiscretelatticeBoltzmannmethod.Zhangetal.[

HYPERLINK\l"bookmark5"

5

]analyzedtheopeningandclosingofabilea?etmechanicalheartvalvewiththe?niteelementmethod(FEM)byapplyingacouplingalgorithmbasedonthearbitraryLagrangian–Eulerian(ALE)method.Zhangetal.[

HYPERLINK\l"bookmark6"

6

]analyzedtherelationshipbetweenavalveanda2D?ow?eldusingANSYSsoftware.Mostarti?cialmechanicalheartvalvedesignsandanalyseshaveadopted2Dmodelsandsteady-state?owinthevalves,newmethodsandmeasurestoimprovethevalvestructurewouldbeenproposed.

②…

Table1Damagethresholdofbloodcomponents[

HYPERLINK\l"bookmark7"

10

–

HYPERLINK\l"bookmark8"

12

]by?uidshearstress

Erythrocytes(dyn/cm2)

500

Leukocytes(dyn/cm2)

300

Bloodplatelets(dyn/cm2)

100–500

Vascularendothelium(dyn/cm2)

400

Thepresentstudyanalyzestheimpactofthevalvestructureonblood?ow,includinghemodynamicaspects.Thedestructionofbloodcomponentsandplateletactiva-tionbytheshearstressgeneratedbythecoupledinteractionbetweenthevalveand?owarediscussed.ThecomputationusestheFEMwithacouplingalgorithmbasedontheALEmethod.

Thebehaviorofblood?owthroughanAHVisrelatedtotheAHVgeometry.The?uidvelocityishighwhenthejetpassesthevalve,whichmaycauseeddiesandinduceahighshearstressintheblood?owregion.Wurzingeretal.[

HYPERLINK\l"bookmark9"

7

]foundthatturbulentstresslevelsof100–1000dyn/cm2canactivateplatelets,andthatthehemolysisthresholdisabove8000dyn/cm2[

HYPERLINK\l"bookmark10"

8

,

HYPERLINK\l"bookmark11"

9

].Krolletal.estimateddamagethresholdsofbloodcomponentsby?uidshearstress.TheirresultsareshowninTable

HYPERLINK\l"bookmark12"

1

.

Inthispaper,avolumemethodisappliedtocharacterizetheblood?ow?eld.The?uiddomainisdiscretizedtogeneratea?niteelementspatiallattice.Inthismethod,differentialequations(e.g.,Navier–Stokesequationandk-εequation)areusedtocontrolvolumeintegration,resultinginasetofdiscreteequationsforanalyzingthe?ow?eldprocess.

2Methods

Whenbloodistransportedthroughanaturalheartvalve(Fig.

HYPERLINK\l"bookmark13"

1

a),the?owiscentral-?eld.Incontrast,blood?owthroughamechanicalbilea?etvalveisthroughthreechannelsandisapproximatelycentral-?eld.Makhijanetal.[

HYPERLINK\l"bookmark14"

13

]consideredthe?uid?owthroughanaturalheartvalvetobeapulsatinglaminar?ow.Experimentsrevealedthatthe?owislaminarintheearlyaccelerationphase[

HYPERLINK\l"bookmark15"

14

],andthusathree-dimensional(3D)numericalsimulationofthe?uid–structureinteractionintheinitialperiodcanbebasedonlaminar?ow.

Inthepresentwork,a25-mmbilea?etmechanicalvalveischosenastheexperimentalmodelbecauseitiscurrentlythemostfrequentlyclinicallyappliedmechanicalheartvalve[

HYPERLINK\l"bookmark16"

15

].Thetwolea?etsarelinkedtothevalveringbyapivotandbutter?ypit.Thelea?etsarerotationallyopenedbythebloodstream?owingthroughthevalve.Aschematicdiagramoftheblood?owthroughtheAHVisshowninFig.

HYPERLINK\l"bookmark13"

1

b.

Theinteractionofthebloodwiththelea?etsandori?ceoftheAHVstructurewassimulatedusingtheFEM.TheprocessisshownschematicallyinFig.

HYPERLINK\l"bookmark17"

2

.Meshesweregeneratedusingacomputational?uiddynamics(CFD)pre-treatmentprograminintegratedcomputerengineeringandmanufacturing(ICEM)andFluentsoftware(AnalysisInc.),combinedwiththeFEM.Thevalvemodelwasthenanalyzed.

Whenthevalveisfullyopened,blood?owsthroughtheori?cealongthelea?ets.Thelea?etscanimpedetheblood?owthroughthevalveinthevalval?eld,increasingthespeedofthejetstream.Theinteractionbetweentheblood?owandthevalvecausesahighshearstress,whichincreasesthepossibilityofplateletactivationorredbloodcelldamage[

HYPERLINK\l"bookmark18"

11

,

HYPERLINK\l"bookmark19"

16

–

HYPERLINK\l"bookmark20"

18

].

2.1ModelingDesignandSimulation

Fluid?owcontrolbyanAHVisshownschematicallyinFig.

HYPERLINK\l"bookmark13"

1

c,d.Thevalvelea?etsareopenedbytheblood?owandclosedbycouplingforces.Theforcecalculationpro-cessisasfollows.Whenblood?owsthroughthevalve,thelea?etsareopened,andwhenbloodre?uxes,theyarequicklyclosed.Theblood?owisregardedassteadywhenthevalveisfullyopened,asitisassumedthatthe?owconditions,suchasvelocity,pressure,andtemperature,areconstant.Adoptingthe?nitevolumemethod,thevelocity?eldcanbesolvedusingmomentumequations,thepres-sure?eldcanbesolvedusingcontinuityandmomentumequations,andtimeintegrationcanbecarriedoutusingtheimplicitmomentumequationandthepressureequation.

A3DgeometricmodelofamechanicalheartvalveattheaorticpositionwasconstructedasshowninFig.

HYPERLINK\l"bookmark21"

3

.

Computersimulationwasperformedunderpulsatingblood?owconditions,withthein?owsetas4L/min[

HYPERLINK\l"bookmark22"

19

].Theinteractionbetweenthevalveandtheblood?owwascalculatedandanalyzedforanopenvalve.

Inthevalvesystem,blood?owchangesperiodically.Thevalvemotionisdividedintothreephases[

HYPERLINK\l"bookmark19"

16

],asshowninFig.

HYPERLINK\l"bookmark21"

3

a.InphaseI,0<tB0.06s,thevalveisclosed;inphaseII,0.06s<tB0.35s,thevalveisopen(includesopeningandclosingofthevalve);inphaseIII,0.35s<tB0.8s,thevalveisclosed.Thevalveisfullyopenattime0.09s,andthemaximumvelocityreaches0.5m/sat0.14s.Undertheconditionofnore?uxinthethirdphase,theexperimentwasmainlycarriedoutfortheperiod0.12–0.30s(AHVfullyopened).

Thebasicassumptionsofthemodelareasfollows:(i)theblood?uidisaviscous,incompressible,isothermalhomogeneousNewtonian?uid;(ii)deformationoftissuearoundtheAHVisignored,andfrictionlessinterfacesbetweenthe?uidandsolidareassumed;(iii)thereisno?uidreturnattheoutputend;(iv)thetemperatureis

②…

AnalysisofFlowFieldinMechanicalAorticBilea?etHeartValvesUsingFiniteVolumeMethod

②…

Fig.1Schematicdiagramofvalve?ow?eldandmodel.aNaturalvalve?ow?eld,bbilea?etmechanicalvalve?ow?eld,cmechanicalvalveinstalledataorticposition,dandAHVinopenedandclosedcon?gurations

Fig.2Numericalsimulation

process.1CFDprocess,2FEM

operation,and3numerical

software

constant;(v)gravitationalforceisneglected.Accordingtotheaboveassumptions,laminarsteady-state?owisselec-tedonthein?owsideinthecomputation.

A3DmodeloftheAHVisshowninFig.

HYPERLINK\l"bookmark21"

3

bandde?nedinTable

HYPERLINK\l"bookmark23"

2

.Theexperimentalconditionsweresetasfol-lows.Thediameterdoftheinletsideclosetotheheart

F.Zhouetal.

Fig.33DcomputationalmodelofphysicalstructureofaortacontainingAHV.aAorticpulsating?owvelocityimposedatin?owsection[

HYPERLINK\l"bookmark19"

16

],bparametersettings,andcboundaryconditions

ventricleis22.5mm.Thein?owconditionissetasu=0.32m/s,whereuisthein?owvelocity.Theblooddensity(p)is1.056g/cm3.Thekineticviscosity(g)ofbloodis3.5cpat37。C[

HYPERLINK\l"bookmark20"

18

],andtheReynoldsnumber(Re)is3186,whichisde?nedbyRe=udp/g.Reislocatedinthesubcriticalregion(thecriticalReynoldsnumberrangeisgenerally2100–4000inatube).Vortexturbulencecanforminthelea?etwake.Fortheoutletboundarycondition,theinitialoutletpressurePin=0Pa.Ano-slipboundaryconditionisimposedfortheinteractionbetweentheblood?owandthevalve.Theentranceboundaryconditionisgivenbytheinitialspeed;theboundaryconditionofthepressureontheexitsideissetaszero(Pout=0Pa)andthereisnoreturnofthe?uid.

Structuralboundaryconditionsweresetasfollows.Thereisnotranslationalmotionbetweenthepivotandlea?ets.Theboundarylayersbetweenthe?owandthevalveareassumedtohavenon-slidingmeshboundaryconditions[

HYPERLINK\l"bookmark24"

22

].

Meshgeneration:ThecomputationalmodelusestheALEmethod,whichreliesonthere-generationofa

Table2PhysicalparametersforFig.

HYPERLINK\l"bookmark21"

3

b

Position

Size(mm)

Position

Size(mm)

Ventricleend

D1[

HYPERLINK\l"bookmark25"

20

]

Lea?etthickness

0.8

Valvularsinus

D2[

HYPERLINK\l"bookmark25"

20

]

L2

150

Arterialend

D3[

HYPERLINK\l"bookmark25"

20

]

L3

80

L1

42

Openingangle

θ[

HYPERLINK\l"bookmark26"

21

]

continuouslydeformingmeshcapableofmaintaininggoodmeshquality.Themeshroughnessissetbythetypeandsizeofthegrid.Thecomplex?uidblocksaremeshedbytetrahedralandtriangulargrids.

Toincreasethesolutionaccuracyandreducecalculationtime,the?owmodelwasdividedmainlyintohexahedral,bothforthecylindrical?uidrunnerandtransitionalcellsofthemodel(Fig.

HYPERLINK\l"bookmark27"

4

a).Intheannularregions,the?ow?eldwasfurthersubdividedintotetrahedralelements.Fromgridqualitydetection,alltheelementshadmaximumskewnesslessofthan0.8.Thesmallertheskewnessvalue,thebetteristhemeshquality,andthusthecalculationsconvergemoreeasily.Thecheckedmeshqualityshouldbeaccept-ableto?niteelementcalculation[

HYPERLINK\l"bookmark28"

23

].Finally,thecellmodelwastranslatedintopolyhedralmeshgridsinthecomputationaldomaintoincreasecomputationalef?-ciency.Themeshconsistedofthreeblockswithatotalof59105cells.ThemeshmodelisshowninFig.

HYPERLINK\l"bookmark27"

4

.

The?ow?eldoftheAHVatcompletelyopenstateswasanalyzed.Theinteractionforceofthe?owonthebloodcomponentsinthecompletelyopenconditionwasevalu-ated.Thetransvalvularpressureandthevelocityofthe?owwereinvestigated,andthetransient?ow?eldwasanalyzedforanentirelyopenedvalve(Fig.

HYPERLINK\l"bookmark27"

4

b).Thevalveperformancewasevaluatedthrough?ow?eldanalysistodeterminetheinteractionforceonbloodcomponents.

Geetal.[

HYPERLINK\l"bookmark29"

24

]foundthattheunsteadyReynolds-aver-agedNavier–Stokes(URANS)equationsaresuitablefordetached-eddysimulation.Therefore,URANSisadoptedforexperimentalsimulation,andatwo-order-accuracyinthek-εequationisapplied,wherek,theturbulentkineticenergy,equals5.234910-4m2/s2;andtheturbulentdis-sipationrateεequals8.517910-4m2/s3.Pulsating?owisselectedonthein?owsideinthecomputation.Thesimu-lationcomputationadoptsaweakcouplingmethodtoanalyzetheinteractionsbetweentheAHVandblood?ow.

3ResultsandDiscussion

3.1FluidVelocityFieldDistribution

Figure

HYPERLINK\l"bookmark30"

5

showsthe?owvelocitycontoursandstreamlinesexpressingtheforward?owinthevelocity?eld.Thevalval

②…

AnalysisofFlowFieldinMechanicalAorticBilea?etHeartValvesUsingFiniteVolumeMethod

Fig.4Computationalmeshfor

simulation(polyhedralmesh

grids)

?eldconsistsofaturbulentregion,aneddyzone,andarecirculationzone.Thevelocityisobviouslygreaterbehindthevalve,uptoamaximumvelocityof0.59m/sandrangingfromtwicetothreetimestheinitialvelocity.Thedownstream?owisrelativelysymmetric.Inthemiddlesectionofthe?owstream,aturbulent?owregionisgen-erated,whicheasilyformsazonewithhighshearstressturbulenceastheReynoldsshearstress(RSS)[

HYPERLINK\l"bookmark31"

25

].RSSisanimportantfactorintheformationofbloodhemolysisandanirreversibleaggregationofplatelets.Inthewakeofthelea?ets,vorticesappearnearthephysiologicsinuses.Thesevorticespushblood?owdownwardtothecentralregion.Astheventriclecontracts,theaorticsinusbecomesthemainlyimpactingpointofblood?ow,whichthenturnsrightintotheaorticarch[

HYPERLINK\l"bookmark32"

26

].

3.2WallShearStress

FlowshearstressisanimportantcharacteristicforthehemodynamicsofAHVs.Figure

HYPERLINK\l"bookmark33"

6

showsthepressuredistributioninthe?uid?eld,displayingtheabsolutevalueoftheZ-directionshearstressinthe?ow?eld.

Theshearstressisexpressedasτ=μ(du/dz)[

HYPERLINK\l"bookmark34"

27

].The?owstatewasanalyzedfortheheartvalveatthemax-imumopeningangle.ThemaximumRSSoccursinanarrowchannel,nearthelea?etpivotpartandattheedgesofthevalve.

Yoganathanetal.[

HYPERLINK\l"bookmark35"

28

]foundthatashearstressof100–500dyn/cm2greatlyenhancestheabilityofplateletstoaggregate,furtherleadingtothrombosisanddirectlyendangeringthepatient’slife.Additionally,Oosterbaanetal.[

HYPERLINK\l"bookmark36"

29

]reportedthatachangeofwallshearstresscanin?uencethegrowthofsurroundingtissuecells.Figure

HYPERLINK\l"bookmark33"

6

showsthatthewallshearstressofvalvalpivotandwallpositionissomewhatgreaterthanelsewhere,withamax-imumvalueabove16Pa(160dyn/cm2),whichmaytrig-gerplateletactivation[

HYPERLINK\l"bookmark37"

30

,

HYPERLINK\l"bookmark38"

31

].Moreover,ahighvelocitygradientmaybegeneratedbetweenthere?uxandtheswirl.

3.3ComparisonofValveDesignsBasedonFlowFieldCalculations

Tooptimizetheparametersofthestructureandreducethemaximumtransvalvularpressuredifference(TPD),various

②…

F.Zhouetal.

②…

Fig.5Flowvelocitycontourandstreamlinesinvalve?ow?eld.aVelocity(m/s)cloudofforward?owforfullyopenedbilea?etvalveandbskeletondiagramof3D?owvelocitystreamlines

Fig.6X–Ysection(Z-axisdirection,Fig.

HYPERLINK\l"bookmark27"

4

b)wallshearstress(Pa,1dyn/cm2=0.1Pa)

maximumopeninganglesandannularcalibersofthevalveswerecompared.

Figure

HYPERLINK\l"bookmark39"

7

showsthevelocitydistributionatmaximumopeninganglesθ=80。,85。,and90。.ResultsfortheTPDatvariousmaximumopeningangles,atalocation2.5timesthevalveori?cediameterbetweenthefrontandbackofthevalveori?ce[

HYPERLINK\l"bookmark40"

32

],areshowninFig.

HYPERLINK\l"bookmark39"

7

b.Theresultsrevealthatincreasingthemaximumopeninganglefrom80。to

90。decreasedtheaveragepressuredifferenceΔPacrossthevalvesigni?cantly.Theaveragepressuredifferencedecreaseswithincreasingmaximumopeningangle.

TheTPDandthevelocitydistributionforori?ceinternaldiametersof20.4mm(St.JudeMedicalInc.valvemodel),22.5mm(intermediatediameter),and23.4mm(ON-Xvalvemodel)wereanalyzedforamaximumopeningangleof85。.

AnalysisofFlowFieldinMechanicalAorticBilea?etHeartValvesUsingFiniteVolumeMethod

Fig.7Comparisonof

propertiesofvalvular?ow

undervariousmaximum

openingangles.aFlowvelocity

states(m/s)andbtransvalvular

pressuredifference(Pa)

Figure

HYPERLINK\l"bookmark41"

8

showsthatthedifferencesinthe?owvelocity?elddecreasedwithincreasinginternalvalvediameter.Figure

HYPERLINK\l"bookmark41"

8

bshowstheeffectofinternalvalvediameterontheaveragedifferentialpressure.Whentheinternaldiam-eterincreasedfrom20.4to23.4mm,theaveragepressuredifferenceΔPdecreasedfrom109to52Pa.

Theseresultsindicatethathighermaximumopeningangleandinternalori?cediameterarebene?cialforloweraveragepressuredifferenceandthusalsoforlowershearstressintheblood?ow,andthusbene?cialfordecreasingthepossibilityofdamagetobloodcomponents.

4TransientPulsatingFlow

UsingtheFEM,thedragofthebilea?etAHVstructurewascalculatedandanalyzedunderpulsating?owconditions.Thein?owpeakvelocitywas0.5m/s,thelea?etthicknesswas0.8mm,andthemaximumopeninganglewas85。.

Figure

HYPERLINK\l"bookmark42"

9

ashowsthetransientvelocitydistributioninthe?ow?eldforthreetimepoints:atthehighestin?ow

velocity(0.59m/s)(I),duringthemiddleperiod,whenthe?owvelocitydistributionofthecurvedlea?et?uid?ow?eldisuniform(II),andwhenthevalvebeginstoclose(maximum?owvelocityis0.32m/s)(III).

The?owvorticityofthebilea?etAHVwascalculatedandanalyzedinthetransient?ow?eld.TheresultsareshowninFig.

HYPERLINK\l"bookmark42"

9

b.Whentheinlet?owvelocityreachesthepeakofthepulsating?ow(0.5m/s)at0.15s,thevorticitydistributionismainlyonthelea?etsandannulussurface,withthemaximumvalueatthefrontofthelea?ets.Thiscomputingresultwillhelptoimproveblood?owcharacterandreducetheimpactofhighshearstressonbloodcom-ponentsbyeddyformationinthevalve?ow?eld.

5ComparisonBetweenExperimentalandComputationalData

Table

HYPERLINK\l"bookmark43"

3

showsacomparisonoftheexperimentalresultsforthemaximumopeningangleandtherelatedeffectiveopenarea(EOA)forvariousbilea?etmechanicalheart

②…

Fig.8Comparisonof

propertiesofvalvular?ow

undervariousinternalori?ce

diameters.aFlowvelocity

states(m/s)andbtransvalvular

pressuredifference(Pa)

valves,withEOA=Q/(44.39ΔP)(whereQistheblood?owthroughthevalveandΔPistheaverageTPD).AhigherEOAisbene?cialfordecreasingtheTPD[

HYPERLINK\l"bookmark44"

33

].Table

HYPERLINK\l"bookmark43"

3

datasupportthesimulationcomputationresultsdeducedfromFig.

HYPERLINK\l"bookmark41"

8

,eventhoughdifferentbilea?etmechanicalheartvalvestructuresmayimpacttheblood?owbehavior.

Table

HYPERLINK\l"bookmark43"

3

showstheinvitrohemodynamicexperimentalresultsforvariousannulusinnerdiametersforAHVsprovidedbySt.JudeMedicalInc.

HYPERLINK\l"bookmark45"

[39

].TheseresultsindicatethatincreasingtheannulusinnersizeincreasestheEOAandreducestheaveragetransvalvularpressure.Thus,thesedatafurthersupportthesimulationresultsofFig.

HYPERLINK\l"bookmark41"

8

.

TheexperimentalresultslistedinTables

HYPERLINK\l"bookmark43"

3

and

HYPERLINK\l"bookmark46"

4

showthatthesimulationcomputationinthisworkisreliable.Theresultsnotonlyrevealtendenciesinhemodynamicbehavior,asshowninFigs.

HYPERLINK\l"bookmark39"

7

and

HYPERLINK\l"bookmark41"

8

,butalsoindicatethatattentionshouldbepaidtospeci?cgeometricpoints,suchasthepeakpositioninFig.

HYPERLINK\l"bookmark33"

6

andthesmalledgeareashowninFig.

HYPERLINK\l"bookmark42"

9

b(pointA).

6Conclusion

Thisstudyinvestigateda3Dnumericalmodeltoanalyzethe?ow?eldofabilea?etvalveatthemaximumopeningangle.Theformationanddistributionofthe?uid?ow,includingthe?uidtrajectory,velocity,pressure,shearstress,andstresseddy,wereanalyzedforsteady-state?owandpulsating?owconditions.Theresultsshowthatwhenthevalveisunderjetconditions,the?owisclosetoasubcriticalturbulentstateandvorticesforminthelea?etwake.Thesinuscanslowthe?owvelocity,andthelea?etsintroduceaminorshearforceinthe?ow?eld.Thereisagreatershearstressatthepivotposition.

Thevalvularhemodynamicperformanceatthemaxi-mumvalveopeninganglewasanalyzedusingtheFEMunderbothsteadyandthepulsating?owconditions.Finally,themaximumopeningangleandannulusdiameterrangewereinvestigated.Somedetailedvalvularhemody-namicperformancecharacteristics,includingthe?owpatternandtheexact?owpressureandshearstressdis-tribution,werealsocomputedandanalyzed.

②…

AnalysisofFlowFieldinMechanicalAorticBilea?etHeartValvesUsingFiniteVolumeMethod

Fig.9Propertiesofvalvular

?owintransient?ow?eld.

aTransientvelocity?ow?eldin

fullyopenpositionatstartof

closingphaseandbtransient

vorticitydiagramat0.15s

Table3Valvularhemodynamicparametersatvariousmaximumopeningangles[

HYPERLINK\l"bookmark28"

23

,

HYPERLINK\l"bookmark47"

34

–

HYPERLINK\l"bookmark48"

38

]

Bilea?etvalve

Maximumopeningangle(。)

EOA(cm2)

CarboMedics-25

78

0.79

JiuLing-25

81

1.41

St.Jude-25

85

2.1

ON-X-25

90

2.4

The?ow?eldofthebilea?etvalvehasauniformdis-tributionintheaorticposition,whichminimizestheresistanceofthevalvetothe?ow?eld.Inordertooptimizethevalvestructure,thetransprostheticdifferenceinshearstressshouldbereducedbyrationaldesignofthevalvelea?etpro?leandstructure,suchasincreasingthemaxi-mumopeningangleandinternalori?cediameter,which

Table4Hemodynamicparametersforvariousvalvularmodels[

HYPERLINK\l"bookmark49"

40

,

HYPERLINK\l"bookmark50"

41

]

AHV

Model(mm)

Annulusinnerdiameter(mm)

Averagetransvalvularpressure(mmHg)

EOA(cm2)

St.JudeMedical

21

23

25

27

16.7

18.67

20.5

22.6

17±5.7

17±7.6

13±6.9

9±2.4

1.4±0.21

1.6±0.32

2.1±0.64

2.6±0.33

②…

F.Zhouetal.

canleadtoamoreuniform?ow?eldanddecreasethepossibilityofdamagetobloodcomponents.

References

1.Vongpatanasin,W.,Hillis,L.D.,&Lange,R.A.(1996).Medicalprogress—Prostheticheartvalves.NewEnglandJournalofMedicine,335,407–416.

2.Yeh,H.H.,Grecov,D.,&Karri,S.(2014).Computationalmodellingofbilea?etmechanicalvalvesusing?uid–structureinteractionapproach.JournalofMedicalandBiologicalEngi-neering,34,482–486.

3.DeHart,J.,Peters,G.W.M.,&Schreurs,P.J.G.(2000).Atwo-dimensional?uid–structureinteractionmodeloftheaorticvalue.JournalofBiomechanics,33,1079–1088.

4.Pelliccioni,O.,Cerrolaza,M.,&Sur

EQ\*jc3\*hps16\o\al(\s\up0(′),o)

s,R.(2008).Abio?uiddynamiccomputercodeusingthegenerallatticeBoltzmannequation.AdvancesinSoftwareEngineering,39,593–611.

5.Jianhai,Z.,Dapeng,C.,&Shengquan,Z.(1996).ALE?niteelementanalysisoftheopeningandclosingprocessofthearti-?cialmechanicalvalve.AppliedMathematicsandMechanics,17,403–412.

6.Zhang,L.R.(2007).Asteadynumericalanalysisofa?uid–solidcouplingmodelinaorticarti?cialbilea?etheartvalves.JournalofClinicalRehabilitativeTissueEngineeringResearch,11(40),8107–8110.

7.Wurzinger,L.J.,Opitz,R.,&Wolf,M.(1984).Shearinducedplateletactivation—Acriticalreappraisal.Biorheology,22,399–413.

8.Hellums,J.D.(1994).1993Whitakerlecture:Biorheologyinthrombosisresearch.AnnalsofBiomedicalEngineering,22,445–455.

9.Lu,P.C.,Lai,H.C.,&Liu,J.S.(2001).Areevaluationanddiscussiononthethresholdlimitforhemolysisinaturbulentshear?ow.JournalofBiomechanics,34,1361–1364.

10.Harrison,P.(2005).Plateletfunctionanalysis.BloodReviews,19,111–123.

11.Kroll,M.H.,Hellums,J.D.,McIntire,L.V.,Schafer,A.I.,&Moake,J.L.(1996).Plateletsandshearstress.Blood,88,1525–