reportedhighfrequency

M.
NelleE.
P.
ZilowO.
LinderkampEffectsofhigh-frequencyoscillatoryventilationoncirculationinneonateswithpulmonaryinterstitialemphysemaorRDSReceived:15May1996Accepted:31January1997ThisstudywassupportedinpartbytheGermanResearchFoundation(DFGre-searchgrant;Li291/4)M.
Nelle())E.
P.
ZilowO.
LinderkampDivisionofNeonatology,DepartmentofPediatrics,UniversityofHeidelberg,ImNeuenheimerFeld150,D-69120Heidelberg,GermanyFAX:+49(6221)564388AbstractObjective:Mechanicalventilationmayimpaircardiovascu-larfunctionifthetranspulmonarypressurerises.
Studiesontheeffectsofhigh-frequencyoscillatoryventi-lation(HFOV)oncardiovascularfunctionshaveyieldedconflictingresults.
Thisstudywasdonetocom-parealterationsinleftventricularoutputandbloodflowvelocitiesintheanteriorcerebralartery,internalcarotidartery,andceliacarteryus-ingaDopplerultrasounddivicebe-foreand2hafterinitiatingHFOVinneonateswithrespiratorydistresssyndrome(RDS)orpulmonaryin-terstitialemphysema(PIE).
Design:Prospectiveclinicalstudy.
Setting:Neonatalintensivecareunitinaperinatalcenter.
Patients:18criticallyillinfants(postnatalage47±12h;mean±SD)werestudiedbeforeandduringHFOV(pistonoscillator).
Indica-tionsforHFOVweresevererespi-ratoryfailureduetoPIE(n=10)andseveresurfactantdeficiency(RDS,n=8).
IntheRDSgroup,gestationalagewas27±6weeks(range26–31weeks)andbirth-weight1620±380g(range850–1970g).
InthePIEgroup,gesta-tionalagewas28±2weeks(range26–36weeks)andbirth-weight1740±470g(range890–2760g).
Measurementsandmainresults:DuringHFOV,meanairwaypres-surewasmaintainedatthesamelevelasduringintermittentmanda-toryventilationinbothgroups(RDS,12±2cmH2O;PIE,10±2cmH2O).
Comparedtointer-mittentmandatoryventilation,sev-eralofthe12parametersstudiedchangedsignificantly(p60mmHgandpeakinspiratorypressureabove28cmH2O.
Allinfantshadumbilicalorradialarte-rialcathetersforbloodsamplingandbloodpressuremeasure-ments.
Moreover,thetranscutaneouspartialpressuresofoxygenandcarbondioxidewerecontinouslymeasuredinallpatients.
VentilatorymanagementConventionalandoscillatoryventilationweredonebymeansofaStephanModelHF300ventilator(Stephan,Gackenbach,Ger-many),whichwasequippedwithahigh-frequencymechanicalpis-tonoscillator(StephanSHF-Generator3000).
TheStephanventi-latorispneumaticallygenerated,time-cycled,andpressurelim-ited.
Inspirationandexpirationtime,gasflow,FIO2,andpositiveend-expiratorypressure(PEEP)wereadjustedattheventilator.
PEEPandpeakinspiratoryandmeanpressureweremeasuredelectronicallyattheproximalendoftheintratrachealtube.
BeforeinitiationofHFOV,concentionalventilationwascarriedoutwithafrequencyof60perminute,aPEEPof4–7cmH2O,aninspirationtimeof0.
4s,andapeakinspiratorypressureof18–28cmH2O.
TheconventionalventilatorgeneratesthePEEPandprovidesin-termittentmandatoryventilationduringHFOVwithaventilationrateof3–5perminute,aninspirationtimeof0.
3–0.
5s,andapeakinspiratorypressureof16–21cmH2O.
InbaboonswithhyalinemembranediseaseandsubsequentPIE,HFOValoneresultedindiffuseatelectasis,hypoxemia,andhypercarbia[11].
Therefore,combinedHFOVandintermittentmandatoryventilationwithafrequencyof3–5/minwaschosentodecreasetheriskofatelectasis[12].
Frequencyandoscillatorpressureamplitudeoftheoscillatorwereadjustedaccordingtoclinicalobservationsofchestwallmovements.
Arterialbloodgasesweremeasuredpriortoand60and120minaftercommencingHFOV.
Thestartingfrequencyofoscillationwas20Hzperminute.
Oscillationfrequencywasad-justedaccordingtotheresponseoftheinfants(14–22.
5Hz).
MeanairwaypressureduringHFOVwaskeptataboutthesamelevelasduringconventionalventilation.
Oxygenationismainlyin-fluencedbymeanairwaypressure,andcarbondioxideeliminationdependsmainlyontheoscillationamplitude[1].
TranscutaneousPaO2andPaCO2werecontinuouslymeasuredinallpatients.
Aboutthesamemeanairwaypressurewasappliedduringinter-mittentmandatoryventilationandduringthefirst2hofHFOV.
GeneralneonatalsupportAllinfantshadperipherallyintroducedcentralsilasticvenouscathetersforparenteralnutritionanddrugs.
Centralvenouspres-surecouldnotbemeasuredviathesefinecatheters.
Noinfantre-672increasedsignificantlybutdidnotchangeintheRDSgroup.
Conclusions:TheresultsshowthatHFOVasusedinthisstudy,im-provesoxygenation,CO2elimina-tion,andcirculationininfantswithRDSandPIE.
However,systemic,cerebral,andintestinalcirculationimprovedmoreinneonateswithPIEthaninthosewithRDS.
ThismaybeduetohigherpulmonarycomplianceininfantswithPIEwhencomparedtothosewithRDS.
KeywordsDopplerultrasoundCardiacoutputCerebralbloodflowvelocityGastrointestinalbloodflowvelocityHigh-frequencyoscillatoryventilationIntermittentmandatoryventilationPulmonaryinterstitialemphysemaRespiratorydistresssyndromeceivedmusclerelaxants.
Hypotension,asdefinedbyVersmoldetal.
[13],wastreatedwithvolumeexpansion(serumorplasma)andinotropicagentsatthediscretionoftheneonatalteam.
Threeinfantsreceiveddopamineinfusion(6mg/kgpermin),nonere-ceivedvolumeexpansionduringthestudyperiod.
Duringthestudyperiod,nochangesoftreatmentwerecarriedout.
Hemoglobinconcentrationwaskeptabove14g/dlinallinfants.
Atthetimeofstudy,noneoftheinfantshadsignsofhypovolemia(e.
g.
,arterialhypotension,poorcutaneousperfusion).
CardiovascularparametersEchocardiographicexaminationwasdonebeforeHFOVtoruleoutcardiacmalformation.
Bloodflowvelocitiesandleftventricu-laroutputweremeasuredbythesameinvestigator(M.
N.
)imme-diatelybeforeinitiationand2hafterthebeginningofHFOV,us-inganInterspecXLpulsedDopplerultrasoundsystem(Interspec,Conshohocken,Penna.
,USA)aspreviouslyreported[14,15].
Dur-ingtwo-dimensionalcontrol,thediameteroftheaorticvalveannu-luswasfocusedandthenthediameteroftheaorticvalveannuluswasdeterminedbyM-modeechocardiographyusingtheparaster-nallong-axisview.
Thediameterwasmeasuredusingtheleadingedgemethodfromtheanterioraorticwalltotheanteriorboundaryoftheposterioraorticwallinlatediastoleoverfiveconsecutivecy-cles.
Aorticcross-sectionalareawascalculatedasp*r2.
Fromanapicalfour-chamberview,thepulsedDopplersamplevolumewasplacedattheleveloftheaorticvalveannulus.
Aorticvelocityinte-gralswererecordedwithamechanical5.
0-MHztransducerusingtheduplexmodeinanattempttoobtainthemaximumspectralen-velopes.
Dopplerwave-formswereanalyzedbythesoftwareoftheultrasoundsystemforpeakvelocity,meanvelocity,andaveragetimevelocityintegral.
Strokevolumewascalculatedastheproductoftheaveragetimevelocityintegralandcross-sectionalareaoftheaorta.
Leftventricularoutputwascalculatedastheproductofstrokevolumeandheartrate.
Heartratewasmeasuredasthedif-ferencebetweenthepeaksofQRScomplexesbythesystemsoft-ware.
Ataconstantheartrate(i.
e.
,sinusrhythm),theaverageoffiveconsecutive,homogeneousflowwavesweretakeninallmea-surementsafter60constantflowwaveswererecorded.
Whenbeat-to-beatcoefficientofvariationwaslessthan5%,Dopplerre-cordingsweretakenasstable[4].
Fractionalshortening(FS%)wascalculatedfromthediffer-encebetweenleftventricularend-diastolic(LVED)andend-sys-tolicdiameters(LVES):FS%=100(LVEDLVES)/LVED.
ThediametersweremeasuredbyM-modeechocardiographyintheshortparasternalaxisfromtheendocardialleftseptalsurfacetotheendocardialsurfaceoftheposteriorfreewall.
Fractionalshorteningwasusedasanindicatorofleftventricularcontractility.
Fractionalshorteningisdecreasedasaresultofleftventricularfail-ureorhighintrathoracicpressureandincreasedasaresultofvol-umeoverloadaslongasthereisnocardiacfailure[14].
Cerebralbloodflowvelocitiesintheanteriorcerebralarteryandrightandleftinternalcarotidarteriesweremeasuredusinga5.
0-MHzpulsedDopplertransducerfromacoronalscanviatheanteriorfontanelle.
Thearterieswereidentifiedbyduplexscanmode.
Thesystemsoftwarewasusedtocalculatemaximalsys-tolic,maximalend-diastolic,andmeanaverageflowvelocityfromfiveconsecutive,homogeneousflowwaves.
Therewerenosignificantdifferencesbetweenthetwointernalcarotidarteries.
Therefore,themeanresultsofbothcarotidarteriesareshowninFig.
1[14].
Theceliacarterywaslocalizedbyultrasoundfromalongitudi-nalabdominalsection.
Bloodflowvelocityintheceliacarterywasdeterminedclosetotheoriginofthearteryfromtheabdominalaorta.
TherewereonlysmallanglesofinsonancebetweentheDop-plerbeamandtheceliacartery.
Nevertheless,anglecorrectionsofflowvelocitiesweremadeinallcases[14].
Ahighpassfilterwallfilter(180–364Hz)wasusedinallmea-surements.
ThesafetyofDopplersonographicinstrumentsde-pendsonenergyoutput.
Theoutputpowervaluesoftheultra-sounddeviceusedinthisstudy(InterspecXL,2-DandM-Mode)arewithintheFoodandDrugAdministrationguidelinesforfetaluseatallcontrolsettings.
The5.
0-MHzprobehasamaximumacousticoutputinthepulsedwavemodeof13.
5(W/cm2)insitu.
Duringourmeasurementsweusedthelowestpossibleenergyout-putlevel.
Intraobservervariabilitieswere:cardiacoutput9.
8%,bloodflowvelocityintheinternalcaroticartery6.
0%,anteriorcerebralartery7.
1%,andceliacartery10.
6%[14].
Meansystolicanddiastolicarterialbloodpressuresweremea-suredcontinuouslyfromindwellingarterialcatheters(eitherum-bilicalorradial)anddisplayedonamonitor.
Systemicflowresis-tancewascalculatedasmeanpressuretoleftventricularoutputra-tio.
Arterialbloodsamplesweretakenhourlyforbloodgasmea-surements(PaO2,PaCO2).
StatisticalanalysesApairedt-testwasusedtocompareHFOVwithconventionalme-chanicalventilation.
Multiplecomparisonsarerequiredtocorrectthesignificancelevelforthenumberofcomparisons.
Since12pa-rameterswerecompared,p<0.
004wasconsideredsignificant.
Dataarepresentedasmean±1SD.
ResultsTable1showstheeffectsofHFOVseparatelyforin-fantswithRDSandwithPIE.
DuringHFOV,meanair-waypressurewasmaintainedatthesamelevelasduringconventionalventilationinbothgroups(RDS,12±2cmH2O;PIE,10±2cmH2O).
IntheRDSgroup,thePaO2/FIO2ratioincreasedby57±18%(p<0.
004)andPaCO2decreasedby27±5%(p<0.
004)duringHFOV.
InthePIEgroup,thePaO2/FIO2ratioincreased673Fig.
1Individualdataforbloodflowvelocityintheinternalca-rotidarteriesinthepatientswithpulmonaryinterstitialemphy-semaPIEandrespiratorydistresssyndromeRDSbeforeanddur-ingHFOVby14±7%(p<0.
004)andPaCO2decreasedby35±8%(p<0.
004).
InthePIEgroup,heartratede-creasedandmeanarterialpressureincreasedsignifi-cantly(p<0.
004)duringHFOV,whereastheseparame-tersdidnotchangeintheRDSgroup.
Leftventricularoutputincreasedby10±7%intheRDSgroup(p<0.
05)andby17±7%(p<0.
004)inthePIEgroup.
Systemicresistanceandshorteningfractiondidnotchangeineithergroup.
Meanbloodflowvelocitiesintheinternalcarotidartery(+59%inthePIEvs19%intheRDSgroup),anteriorcerebralartery(+65%vs+19%),andceliacartery(+45%vs+17%)increasedsignificantlyinthePIEgroup(p<0.
004),butnotintheRDSgroup(p<0.
05)duringHFOV.
OnepatientwithRDSdevelopedintraventricularhemorrhagegradeIIbeforeHFOVwithnoprogressiononHFOV.
ApatientwithPIEandpneumopericardiumdied18hafterthestudyduetoprogressivecardiacfail-ure.
DiscussionInagreementwithpreviousstudies,weobservedconsid-erableimprovementinoxygenationandcarbondioxideeliminationafterinitiationofHFOVininfantswithre-spiratoryfailureduetoPIEandRDS[1,2,5,8,9,16–18].
However,oxygenationimprovedmoreinthein-fantswithRDSandcarbondioxideeliminationim-provedmoreinthePIEgroup(Table1).
Moreover,ar-terialbloodpressure,leftventricularoutput,andmeanbloodflowvelocitiesincerebralarteriesandtheceliacarteryincreasedmoreintheinfantswithPIEthanintheinfantswithRDS(Table1).
Severalstudieshaveshownthatahighmeanairwaypressuremayimpaircardiacoutputandbloodflowtovariousorgans[9,19–23].
Thiscanbeexplainedbyris-ingalveolarandintrathoracicpressure[18,20].
Thefractionoftheintra-alveolarpressuretransmittedtotheintrathoracicspaceriseswithincreasingcompliance,increasingmeanairwaypressure,anddecreasingtho-raciccompliance.
Amarkedriseinintrathoracicpres-suremaydecreasesystemicandpulmonaryvenousre-turnandcardiacoutputandincreaseextra-andintra-pulmonaryright-to-leftshunts[7,9].
Atagivenproxi-malmeanairwaypressure(measuredattheproximalendoftheendotrachealtube),intra-alveolarpressuremaybelowerduringHFOVcomparedtointermittentmandatoryventilation[1,17,18].
Inourstudy,significantincreasesinstrokevolumeandcardiacoutputwerefoundduringHFOVinthegroupwithPIE.
Strokevolumecanbeincreasedasare-sultofdecreasedleftventricularafterload,increasedpreload,orincreasedmyocardialcontractility.
In-creasedmyocardialcontractilityisunlikelybecausetheshorteningfractionoftheleftventricledidnotchange[13,14].
Unchangedsystemicresistancesuggestsun-674Table1EffectsofintermittentmandatoryventilationIMVandhigh-frequencyoscillatoryventilationHFOVinneonateswithrespira-torydistresssyndromeRDSandpulmonaryinterstitialemphysemaPIE.
VariablesRDS(n=8)PIE(n=10)IMVHFOVIMVHFOVMeanairwaypressure(MAP;cmH2O)12.
5±1.
712.
0±1.
99.
5±1.
4**9.
1±1.
0**PaO2/FIO256±986±7*63±872±7*,**PaCO2(torr)49±435±3*63±7**40±5*Meanbloodpressure(BP,mmHg)45±348±343±451±4*Heartrate(min1)142±16139±14135±15115±14*,**Strokevolume(ml/kg)1.
7±0.
61.
8±0.
51.
5±0.
52.
1±0.
4*Leftventricularoutput(LVO;ml/kgpermin)225±46248±47210±34245±36*Flowresistance(R=103*BP/LVO)209±15196±13209±20214±18Fractionalshortening(%)29±728±627±928±7A.
carotisinterna(m/s)0.
17±0,050.
19±0.
040.
13±0,04**0.
22±0.
05*A.
cerebrianterior(m/s)0.
17±0.
050.
19±0.
040.
15±0.
040.
24±0.
05*,**Celiacartery(m/s)0.
27±0.
080.
30±0.
090.
22±0.
07**0.
32±0.
09**p<0.
004comparedwithresultsbeforestartingHFOV,pairedt-test;mean±1SD**p<0.
004comparedtotheRDSgroup,unpairedt-test;mean±1SDFig.
2IndividualdataforPaO2/FIO2ratiointhepatientswithpul-monaryinterstitialemphysemaPIEandrespiratorydistresssyn-dromeRDSbeforeandduringHFOVchangedleftventricularafterload.
Wethereforespecu-latethatpreloadtotheleftventricleincreasedinthein-fantswithPIEduringHFOV.
DuringHFOV,alveolarpressuremayhavedecreasedinspiteofunchangedmeanairwaypressure[18],therebyimprovingvenousreturnandincreasingleftventricularpreloadandleftventricularoutput[9].
Pulmonarycompliancewasnotmeasuredinourstudy.
InRDS,pulmonarycomplianceisusuallymarkedlydecreased[24],whereasinPIE,pul-monarycomplianceislessaffected[19,20].
Thismayex-plainwhyweobservedamorepronouncedeffectofHFOVoncirculationininfantswithPIE.
Moreover,thehighermeanairwaypressuremayhavehinderedariseincardiacoutputintheRDSgroup[9].
Inhealthyanimals,HFOVimprovescardiacoutputiflowermeanairwaypressuresareusedthanduringinter-mittentmandatoryventilation[8].
Bohnetal.
[25]foundnoeffectofcombinedconventionalandHFOVoncar-diacoutputinanesthetizedapneicbeagledogs,whencomparedtoconventionalventilation.
Inadultpatientswithobstructivelungdisease,HFOVresultedinare-ductionofcalculatedpulmonaryright-to-leftshuntwithoutachangeincardiacoutput[17].
Previousstudiesshowednoeffectofhigh-frequencyjetoscillatoryventi-lationoncardiacoutputinprematurehumaninfants[7]orofHFOVinbabooninfants[8].
Theeffectofchangesinlungcomplianceontherelationshipbetweenmeanairwaypressureandcardiacoutputhasbeenstudiedinpigletswithnormalandreducedlungcomplianceduringconventionalventilation[20].
Whencompliancewasnormal,alineardecreaseofcardiacoutputwasfound.
Whencompliancewasreduced(i.
e.
,RDS),cardiacout-putdecreasedonlyathighermeanairwaypressure.
Wefoundthatheartratedecreasedsignificantlydur-ingHFOV.
ThiseffectwasmorepronouncedinthePIEgroupandmaybeexplainedbytheincreaseinstrokevolumen.
InthePIEgroup,meancerebralbloodflowvelocityintheinternalcarotidarteryandanteriorcerebralar-teryincreasedsignificantlyinspiteofsignificantlylow-eredPaCO2.
ReducedPaCO2usuallydecreasescerebralbloodflowvelocity[6,26].
Anincreaseincerebralbloodflowvelocitymaybedueeithertoincreasedflowortovasoconstrictionatconstantflow.
Sincebotharte-rialbloodpressureandcardiacoutputincreasedsignifi-cantlyinthePIEgroupduringHFOV,itappearslikelythattheriseincerebralbloodflowvelocityresulted,atleastinpart,fromanincreasingcerebralbloodflow.
Asaresultofimmaturecerebralautoregulationinneo-nates,changesinbloodpressuremayhaveastrongerin-fluencethanchangesinPaCO2oncerebralbloodflow[26].
Anincreaseincerebralperfusionpressuremayalsoexplainwhyanincreasedincidenceofintraventric-ularhemorrhageasaresultofHFOVhasbeenreported[5,27].
However,noneofourneonatesdevelopedintra-ventricularhemorrhageafterbeginningHFOV.
Bloodflowvelocityintheceliactrunkwasmeasuredasanindicatorofintestinalbloodflow[28].
InthePIEgroup,bloodflowvelocitiesintheceliactrunkincreasedmorethanthecardiacoutput.
Similiarresultswereob-tainedinprematurebaboonswithhyalinemembranedisease[8].
Wehavenotevaluatedpatencyoftheductusarterio-susintheneonates.
Thus,theriseinleftventricularout-putafterinitiationofHFOVcouldbeduetoamarkedleft-to-rightductalshunt,therebythwartingariseintheactualsystemiccardiacoutput.
However,concomi-tantincreasesincardiacoutputandbloodflowveloci-tiesincerebralandgastrointestinalarteriesinthePIEgroupindicateanactualincreaseinsystemiccardiacoutputinthisgroup.
Inconclusion,HFOVusedasrescuetherapyincriti-callyillneonateswithsevererespiratoryfailureunre-sponsivetoconventionalventilationresultedinim-provedoxygenationandimprovedcarbondioxideelim-ination.
HFOV,asusedinourstudy,mayalsoincreasesystemic,cerebral,andintestinalperfusion,therebyin-creasingoxygensupplytotheseorgans.
ThiseffectwasmorepronouncedintheinfantswithPIEthaninthosewithRDS.
675References1.
FrantzID,WerthammerJ,StarkAR(1993)High-frequencyventilationinprematureinfantswithlungdisease:ad-equategasexchangeatlowtrachealpressure.
Pediatrics71:4832.
MarchakBE,ThompsonWK,DuffyPetal(1981)TreatmentofRDSbyhigh-frequencyoscillatoryventilation:apre-liminaryreport.
JPediatr99:2873.
ToutantSM,ToddMM,DrummondJCetal(1983)Cerebralbloodflowduringhigh-frequencyventilationincats.
CritCareMed11:7124.
RajuTNK,BravermanB,NadkarnyUetal(1983)Intracranialpressureandcardiacoutputremainstableduringhigh-frequencyoscillation.
CritCareMed11:8565.
HIFIStudyGroup(1989)High-fre-quencyoscillatoryventilationcom-paredwithconventionalmechanicalventilationinthemanagementofrespi-ratoryfailureinpreterminfants.
NEnglJMed320:886.
CowanF,ThoresenM(1987)Theef-fectsofintermittentpositivepressureventilationoncerebralarterialandve-nousbloodvelocitiesinthenewbornin-fant.
ActaPaediatr76:2397.
AlversonDC,FrippR,CorlewS,BackstromC,AngelusP,WernerS(1990)Hemodynamicimpactofhighfrequencyjetventilation(HFJV)vsconventionalventilation(VS)inneo-nateswithpulmonaryinterstitialem-physema(PIE).
PediatrRes27:295A6768.
KinsellaJP,GerstmannDR,ClarkRHetal(1990)High-frequencyoscillatoryventilationvsintermittentmandatoryventilation:earlyhemodynamiceffectsintheprematurebaboonwithhyalinemembranedesease.
PediatrRes28:3679.
TraverseJH,KorvenrantaH,AdamsEM,GoldthwaitDA,CarloWA(1988)Impairmentofhemodynamicswithin-creasingmeanairwaypressureduringhigh-frequencyoscillatoryventilation.
PediatrRes23:62810.
CampbellRE(1970)Intrapulmonaryinterstitialemphysema:acommoncom-plicationofhyalinemembranedisease.
AmJRoentgenol110:44911.
AckermannNB,CoalsonJJ,KuehlTJ,StoddardR,MinnickLetal(1984)Pul-monaryinterstitialemphysemaintheprematurebaboonwithhyalinemem-branedisease.
CritCareMed12:51212.
Blum-HoffmannE,KopoticRJ,Man-ninoFL(1988)High-frequencyoscilla-toryventilationcombinedwithinter-mittentmandatoryventilationincriti-callyillneonates:3yearsofexperience.
EurJPediatr147:39213.
VersmoldHT,KittermanJA,PhibbsRH,GregoryGA,TooleyWH(1981)Aorticbloodpressureduringthefirst12hoursoflifeininfantswithbirthweight610to4220grams.
Pediat-rics67:60714.
NelleM,Ho¨ckerC,ZilowEP,Linder-kampO(1994)Effectsofredcelltrans-fusiononcardiacoutputandbloodflowvelocitiesincerebralandgastrointesti-nalarteriesinprematureinfants.
ArchDisChild71:F4515.
NelleM,ZilowEP,LinderkampO(1995)EffectofLeboyerchildbirthoncardiacoutput,cerebralandgastroin-testinalbloodflowvelocitiesinfull-termneonates.
AmJPerinatol12:21216.
BaylenB,MeyerRA,KorfhagenJ,BenzingG,BubbME,KaplanS(1977)Leftventricularperformanceinthecrit-icallyillprematureinfantwithpatentductusarteriosusandpulmonarydis-ease.
Circulation55:18217.
ButlerWJ,BohnDJ,BryanACetal(1980)Ventilationbyhigh-frequencyoscillationinhumans.
AnesthAnalg59:57718.
GerstmannDR,FoukeJM,WinterDC,TaylorFetal(1990)Proximal,tracheal,andalveolarpressuresduringhigh-fre-quencyoscillatoryventilationinanor-malrabbitmodel.
PediatrRes28:36719.
CarloWA,MartinRJ,ShivpuriCR,FanfaroffAA(1984)Decreaseinair-waypressureduringhigh-frequencyjetventilationininfantswithrespiratorydistresssyndrome.
JPediatr104:10120.
MirroR,BusijaD,GreenR,LefflerC(1987)Relationshipbetweenmeanair-waypressure,cardiacoutput,andorganbloodflowwithnormalanddecreasedrespiratorycompliance.
JPediatr111:10121.
JardinF,FarcotJC,BoisanteL,Cu-rienN,MargairazA,BourdariasJP(1981)Influenceofpositiveend-expira-torypressureonleftventricularperfor-mance.
NEnglJMed304:38722.
BraunwaldE,BinionJT,MorganWL,SarnofSJ(1957)Alterationsincentralbloodvolumeandcardiacoutputin-ducedbypositivepressurebreathingandcounteractedbymetaraminol(Ara-mine).
CircRes5:67023.
FewelJE,AbendscheinDR,Carl-sonCJ,MurrayJF,RapaportE(1980)Continuouspositive-pressureventila-tiondecreasesrightandleftventricularend-diastolicvolumesinthedog.
CircRes46:12524.
ClarkRH,GerstmannDR,NullDM,deLemosRA(1992)Prospectiveran-domizedcomparisonofhigh-frequencyoscillatoryandconventionalventilationinrespiratorydistresssyndrome.
Pedi-atrics89:525.
BohnDJ,MiyasakaK,MarchakBEetal(1980)Ventilationbyhighfrequencyoscillation.
JApplPhysiol48:71026.
FentonAC,FieldDJ,WoodsKL,EvansDH,LeveneMI(1990)Circula-toryeffectsoffastventilatorratesinpreterminfants.
ArchDisChild65:66227.
HIFI-Studygroup(1990)High-fre-quencyoscillatoryventilationcom-paredwithconventionalintermittentmechanicalventilationinthetreatmentofrespiratoryfailureinthepretermin-fant:neurodevelopmentralstatusat16to24monthofposttermage.
JPed-iatr117:93928.
LeidigE(1989)Doppleranalysisofsu-periormesentericarterybloodflowinpreterminfants.
ArchDisChild64:476