Low cerebral blood flow after cardiac arrest is not associated with anaerobic cerebral metabolism

(1)

ContentslistsavailableatScienceDirect

Resuscitation

j o ur na l h o me pa g e:ww w . e l s e v i e r . c o m / l o c a t e / r e s u s c i t a t i o n

Clinical

paper

Low

cerebral

blood

ﬂow

after

cardiac

arrest

is

not

associated

with

anaerobic

cerebral

metabolism

夽

Cornelia

W. Hoedemaekers

a,∗

_,

_Philip

_N.

_Ainslie

b

_,

_Stijn

_Hinssen

c

_,

_Marcel

_J.

_Aries

d

_,

Laurens

L. Bisschops

a

_,

_Jeannette

_Hofmeijer

c

_,

_Johannes

_G.

_van

_der

_Hoeven

a

a_Department_of_Intensive_Care_Medicine,_Radboud_University_Medical_Center,_Nijmegen,_The_Netherlands b_Centre_for_Heart,_Lung_and_Vascular_Health,_University_of_British_Columbia,_British_Columbia,_Canada

c_Department_of_Neurology,_Rijnstate_Hospital,_Arnhem_and_department_of_Clinical_{Neurophysiology,}_University_of_Twente,_Enschede,_The_Netherlands d_Department_of_Intensive_Care,_University_of_Maastricht,_Maastricht_University_Medical_Center,_Maastricht,_The_Netherlands

a

r

t

i

c

l

e

i

n

f

o

Articlehistory:

Received14June2017

Receivedinrevisedform10August2017 Accepted20August2017

Keywords: Cardiacarrest Cerebralbloodﬂow Lactate

Metabolism

Post-cardiacarrestsyndrome

a

b

s

t

r

a

c

t

Aimofthestudy:Estimationofcerebralanaerobicmetabolisminsurvivorsandnon-survivorsaftercardiac

arrest.

Methods:Weperformedanobservationalstudyintwentycomatosepatientsaftercardiacarrestand

19healthycontrolsubjects.Wemeasuredmeanﬂowvelocityinthemiddlecerebralartery(MFVMCA)

bytranscranialDoppler.Arterialandjugularbloodsampleswereusedforcalculationofthejugular

venous-to-arterialCO2/arterialto-jugularvenousO2contentdifferenceratio.

Results: After cardiac arrest, MFVMCA increased from 26.0[18.6–40.4]cm/sec on admission to

63.9[48.3–73.1]cm/secafter72h(p<0.0001),withnosigniﬁcantdifferencesbetweensurvivorsand

non-survivors(p=0.4853).TheMFVMCAincontrolswas59.1[52.8–69.0]cm/sec.Theoxygenextraction

fraction(O2EF)was38.9[24.4–47.7]%onadmissionanddecreasedsigniﬁcantlyto17.3[12.1–26.2]%at

72h(p<0.0001).ThedecreaseinO2EFwasmorepronouncedinnon-survivors(p=0.0173).O2EFinthe

controlgroupwas35.4[32.4–38.7]%.Thejugularbulb-arterialCO2toarterial-jugularbulbO2content

differenceratiowas>1atalltimepointsaftercardiacarrestanddidnotchangeduringadmission,with

nodifferencesbetweensurvivorsandnon-survivors.Valuesincardiacarrestpatientsweresimilarto

thoseinnormalsubjects.

Conclusions:Inthisstudy,lowCBFaftercardiacarrestisnotassociatedwithanaerobicmetabolism.

Hypoperfusionappearstobetheconsequenceofadecreaseofneuronalfunctioningandmetabolicneeds.

Alternatively,hypoperfusionmaydecreasecerebralmetabolism.Subsequently,metabolismincreases

insurvivors,consistentwithresumptionofneuronalactivity,whereasinnon-survivorslastinglow

metabolismreﬂectsirreversibleneuronaldamage.

license(http://creativecommons.org/licenses/by/4.0/).

Introduction

Cardiacarrestisaleadingcauseofdeathinwesterncountries, andmostpatientsdiefromneurologicalinjury[1].Abnormal cere-bralbloodﬂow(CBF)isakeyfeatureofpost-anoxicbraininjury. Incomatosepatientsaftercardiacarrest,CBFisinitiallylowand graduallyrestorestowardsnormalvaluesduringtheﬁrst72hafter returnofspontaneouscirculation(ROSC)[2–4].This

hypoperfu-夽 ASpanishtranslatedversionoftheabstractofthisarticleappearsasAppendix intheﬁnalonlineversionathttp://dx.doi.org/10.1016/j.resuscitation.2017.08.218. ∗ Correspondingauthorat:RadboudUniversityNijmegenMedicalCentre, Depart-mentofIntensiveCare,POBox9101,6500HBNijmegen,TheNetherlands.

E-mailaddress:Astrid.Hoedemaekers@radboudumc.nl(C.W.Hoedemaekers).

sionaftercardiacarrestcanpotentiallycauseamismatchbetween cerebraloxygendemandandsupply.Previousstudiesreporteda decreaseinthecerebralmetabolicrateofoxygen,secondarytothe cardiacarrest,providingateleologicalmechanismtomatchthis decreasedsupplyofoxygenandnutrientstothebrain[2–4].Itis yetunknown,whetherthisdecreaseinmetabolicrateis propor-tionaltothedecreaseinCBF:disproportionaladaptationofCBFto metabolismmayresultinhypoperfusionandischemiaor hyper-perfusionandhyperemia.

Recently, the mixed venous–arterial CO2 (Cv-aCO2)

to arterial–venous O2 (Ca-vO2) content difference ratio

(Cv-aCO2/Ca-vO2), has been suggested as a surrogate marker

for the balance between oxygen consumption (VO2) and CO2

production (VCO2) in sepsis patients. Under normal aerobic

http://dx.doi.org/10.1016/j.resuscitation.2017.08.218

(2)

conditions,O2 deliverymatchesCO2 production,andtheCa-vO2

approximatestheCv-aCO2.Althoughhypoperfusiondecreasesboth

O2consumptionandaerobicCO2production,anaerobicglycolysis

andATPhydrolysisincreaseanaerobicCO2production,leadingto

anincreasedCv-aCO2relativetoCa-vO2[reviewedin[5]].

Oxygen-derivedorCO2-derivedparametersasasingle

parame-tercorrelatepoorlywithanaerobicmetabolism.TheCa-vO2cannot

discriminatebetweentruetissuehypoxiaandstatesofareduced demandwithouthypoxia.AnaerobicCO2productionwillresultin

anincreasedCv-aCO2 [6].Atthesametime,duetotheso-called

CO2stagnationphenomenon,slowmicrocirculatorybloodﬂowwill

resultinincreasedtransferofCO2fromthetissuetothe

microcir-culation,resultinginanincreasedCv-aCO2intheabsenceoftissue

hypoxia.Insepsis,theCv-aCO2/Ca-vO2ratioisstrongmarkerforthe

detectionofanaerobicmetabolismandmorereliablethan conven-tionalparameterssuchashyperlactatemiaormixedvenousoxygen saturation[6–8].AnincreasedCv-aCO2/Ca-vO2ratiowasassociated

withapooroutcomeandseverityoforgandysfunctioninsepsis patients[9].

CBFis low aftercardiacarrest and maypotentiallyresult in increasedanaerobicmetabolism,despitenormaljugularbulb lac-tateconcentrations and normaljugular bulboxygensaturation. Theaimofthepresent studywastofurtherelucidatethe cou-plingbetweencerebraloxygendeliveryand demandduringthe post-cardiacarrestsyndromebydeterminationofparametersof cerebralanaerobicmetabolism.Wehypothesizedthatthechanges inCBFaftercardiacarrestwererelatedtochangesinmetabolism. In analogytothe Cv-aCO2/Ca-vO2 ratio,we calculated the

jugu-larvenous-to-arterial CO2/arterial to-jugularvenousO2 content

differenceratio(Cjb-aCO2/Ca-jbO2)asameasureofanaerobicCO2

generationinsurvivorsandnon-survivorsaftercardiacarrest.For theﬁrsttime,inordertoobtainnormativevalues,thesemetrics werealsocalculatedinhealthycontrolsubjects.

Materialandmethods Studypopulation

Weperformedasecondaryanalysisofprospectivelycollected datafrom two differenceobservational studiesin 20 comatose patientsafterout-of-hospitalcardiacarrestand19healthy con-trols.Thelocalinstitutionalreviewboardsapprovedtheoriginal studies.

Datafromthecardiacarrestpatientswereprospectively col-lected from two observational studies that studied the effect ofviscosity(n=10patients) andprolongedhypothermia (n=10 patients)onCBFaftercardiacarrest[10,11].Maininclusion cri-teriaforbothstudieswereGlasgowComaScale≤6afterROSCand age>18years.Exclusioncriteriaincludedpregnancy,thrombolytic therapy,contraindicationfortherapeutichypothermia,andchronic renalorhepaticfailure.

Dataof19normalhealthycontrolswerederivedfrom2 observa-tionalstudiesontheeffectsofmanipulationofPaO2andPaCO2on

theCBFandmetabolism[12,13].Thevolunteerswerenon-smokers, hadnohistoryofcardiovasculardiseases,andwerenottakingany medications.Onlydataobtainedatsealevelwereusedfor this study.

Post-cardiacarrestmanagement

CardiacarrestpatientswereadmittedtotheICUandtreated withmildhypothermiaat33◦Cfor24or72h,followedbypassive rewarming.Patientsweresedatedwithpropofoland/ormidazolam andsufentanil.Incaseofshivering,patientswereparalyzedusing intravenousbolusinjectionsofrocuronium.

Allpatientswereintubatedandmechanicallyventilated, aim-ingataPaO2>75mmHgandaPaCO2 between34and41mmHg.

Alpha-statwasusedforpHmaintenance.Monitoringofblood pres-sureandarterialbloodsamplingwasperformedwiththeuseof acatheterintheradialorfemoralartery.Meanarterialpressure (MAP)wasmaintainedbetween80and100mmHgwithadiuresis of>0.5mL/kg/hr.Patientsweretreatedwithvolumeinfusionand dobutamineornorepinephrine,ifnecessary.Bloodfromthejugular bulbwassampledfroma7-Frsingle-lumenjugularbulbcatheter. Datacollection

Post-cardiacarrestpatients

Demographicand clinicaldatawerecollected.Hemodynamic variables,temperature,andSaO2weremeasuredcontinuously.

Transcranial Doppler (TCD) of the middle cerebral artery (MCA)wasperformedtroughthetemporalwindowwitha2MHz probe(SonositeM-Turbo,SonoviewNederlandBV,Rijswijk,The Netherlands)onadmissionandat12,24,48and72hthereafter.

Arterialandjugularbloodsampleswerecollectedforbloodgas analysis,lactateandhemoglobinmeasurementsuponadmission andat12,24,48and72h.Outcomeaftercardiacarrestwasassessed uponICUdischarge.

Datacollectionhealthyvolunteers

TheleftMCA blood velocitywasmeasuredbyTCD (Spencer Technologies,Seattle,WA,USA)usinga2-MHzpulsedprobe.A20G arterialcatheter(Arrow,Markham,Ontario,Canada)wasinserted intotheleftradialartery,andajugularbulbcatheter(Edwards Pedi-aSatOximetrycatheter,Irvine,CA,USA)wasplacedintothejugular bulb.

Dataanalysis

Metabolicparameterswerecalculatedasfollows:

Contentofarterial(CaO2)andvenous(CvO2)oxygenwere

cal-culatedusingtheequations: CaO2

ml.dl−1

= [Hb] ·1.36·SaO2(%) 100 +0.003·PaO2 CvO2

ml.dl−1

= [Hb] ·1.36·SvO2(%) 100 +0.003·PvO2

Where1.36istheafﬁnityforoxygentohemoglobinforagiven arterialsaturation,and0.003isthepercentageofoxygendissolved intheblood.

Oxygenextractionfraction(O2EF)wascalculatedby:

O2EF (%)=

CaO2−CvO2

CaO2 ·

100%

WherearterialandjugularvenousO2contentdifferencesareequal

toCaO2andCvO2,respectively.

Thecontent ofcarbondioxide in thearterialenjugular bulb venousbloodsampleswascalculatedaccordingtoDouglasetal [14]. CCO2

mldl−1

=plasmaCCO2× (1− 0,0289×Hb (3,352−0,456×SO2)× (8,142−pH)

, withplasmaCCO2=2,226×S×PCO2×

1+10pH−pK

. SandpK’aretheplasmaCO2solubilityandapparentpK,

(3)

Table1

Demographicdatapost-cardiacarrestpatients.

Characteristics Men,n(%) 17(85%) Age(yr) 66[59.5–73] BMI(kg/m2) 26[24.8–26.5] Primaryrhythm,n(%) Shockable 10(50%) Non-shockable 10(50%) Timecollapse-ROSC 30[25–54] SAPS2 65.5[50.5–71.8] APACHEII 26[18.8–29.5]

pHuponhospitaladmission 7.19[7.07–7.27] BEuponhospitaladmission(mmol/l) −9.15[−15.8to−5.6] Lactateuponhospitaladmission(mmol/l) 6.8[3.6–10.9] Patientsdied,n(%) 9(45%) Dataareexpressedasmedian[interquartilerange]orasabsolutenumbers (per-centage).

BMI:Bodymassindex.

ROSC:returnofspontaneouscirculation. SAPS2:SimpliﬁedAcutePhysiologyScore.

APACHEII:AcutePhysiologyandChronicHealthEvaluationII.

calculatedas:

S=0,0307+0,00057× (37−T) +0,00002× (37−T)2

and

6,086+0,042× (7,4−pH) + (38−T) × (0,00472

+0,00139× (7,4−pH))

Thejugularbulb-arterialCO2contentdifferencewascalculated

as:

Cjb-aCO2(mldl−1)= CjbCO2−CaCO2.

Thejugularbulb-arterialCO2toarterial-jugularbulbO2content

differenceratiowasdeﬁnedas

Cjb-aCO2/Ca-jbO2.

Lactateextractionfraction(LacEF)wascalculatedby:

LacEF (%)= Laca−Lacv

Laca ·

100%

WhereLacaandLacvequalthearterialandjugularvenous

concen-trationoflactate,respectively.

Statistics

StatisticalanalysiswasperformedusingGraphPadPrism

ver-sion5.0(GraphPadSoftware,LaJolla,CA).Normaldistributionof

thedatawasveriﬁedusingtheKolmogorov-Smirnovtest.Datain

thetextarepresentedasmedianwith25thand75thpercentile.

Changes over time wereanalyzed with the repeated-measures

test(onewayANOVA).Differencesbetweensurvivorsand

non-survivorsintimewereanalyzedwithtwo-wayanalysisofvariance.

TheStudent’st-testortheWilcoxonsigned-ranktestwasused

forthecomparisonbetweengroups,dependingonthedistribution

ofthedata.Ap-valueof<0.05wasconsideredtoindicate

signiﬁ-cance.

Results

Demographicandclinicaldata

Weincluded20patientsaftercardiacarrest.Demographicand

clinicaldataaresummarizedinTable1.Ninepatientsdiedinthe

ICU:5 patients becauseof severe post-anoxic brain damage, 3

Fig.1. Meanﬂowvelocityinthemiddlecerebralarteryinsurvivors(whitebars) andnon-survivors(greybars)duringtheﬁrst72haftercardiacarrest.Stripedbar representsvaluesinnormalcontrolsubjects.Dataareexpressedasmean±SD.

Fig.2.Oxygenextractionfractioninsurvivors(whitebars)andnon-survivors(grey bars)duringtheﬁrst72haftercardiacarrest.Stripedbarrepresentsvaluesinnormal controlsubjects.Dataareexpressedasmean±SD.

patientsbecauseofcirculatoryfailureandin1patientactive treat-ment waswithdrawn becauseof severepreexisting pulmonary disease.PaO2andPaCO2werestableduringadmissionwithno

sig-nificantchangesbetweensurvivorsand non-survivors(datanot shown).Hemoglobin concentrationdecreasedsignificantly from 13.4[11.8–15.2]g/dLuponadmissionto11.3[9.4–11.8]g/dlat72h (p<0.0001),withnostatisticallysignificantdifferencesbetween survivorsandnon-survivors(p=0.54).Useofdosesofsedatives, inotropesandvasopressorsdidnotdifferbetweensurvivorsand non-survivors.

19subjects(17males,2females)wereincludedinthecontrol group,withamedianageof26[23.5–31]yrs.ValuesofPaO2and

PaCO2 wereessentiallyequaltothoseincardiac arrestpatients

(datanotshown). Thehemoglobinconcentrationinthecontrol groupwas14.8[14.0–15.1]g/dl.

Cerebralbloodﬂow

Inthepost-cardiacarrestgroup,theMFVMCA increasedfrom

26.0[18.6–40.4] cm/sec on admission to 63.9[48.3–73.1]cm/sec after72h(p<0.0001),withnosigniﬁcantdifferencesbetween sur-vivorsandnon-survivors(p=0.49)(Fig.1).TheMFVMCAinhealthy

controlswas59.1[52.8–69.0]cm/sec. Metabolicvariables

TheO2EFwas38.9[24.4–47.7]%onadmissionanddecreased

sig-niﬁcantlyto17.3[12.1–26.2]%at72h(p<0.0001).Thedecreasein O2EFwassigniﬁcantlymorepronouncedinnon-survivors(p=0.02)

(Fig.2).TheO2EFinthecontrolgroupwas35.4[32.4–38.7]%.

TheCjb-aCO2 decreasedfrom7.68[4.04–12.1]uponadmission

to3.32[1.56–4.70]ml/dlat72h(p=0.08).Thedecreasein jugu-larbulb-arterialCO2contentdifferencewassigniﬁcantlystronger

innon-survivorscomparedtosurvivors(p=0.0061)(Fig.3).The Cjb-aCO2inthecontrolgroupwas7.88[6.96–9.09]ml/dlandequal

(4)

Fig.3. Thejugularbulb-arterialCO2contentdifferenceinsurvivors(whitebars)

andnon-survivors(greybars)duringtheﬁrst72haftercardiacarrest.Stripedbar representsvaluesinnormalcontrolsubjects.Dataareexpressedasmean±SD.

Fig.4.Thearterial-jugularbulbO2contentdifferenceinsurvivors(whitebars)

andnon-survivors(greybars)duringtheﬁrst72haftercardiacarrest.Stripedbar representsvaluesinnormalcontrolsubjects.Dataareexpressedasmean±SD.

tovaluesofcardiacarrestpatientsuponadmissiontotheICU.The decreaseinCjb-aCO2 waspredominantlycausedbyadecreasein

jugularbulbCO2content(Table2,Supplementarydigitalcontent).

TheCa-jbO2 decreasedfrom6.0[4.08–8.43]uponadmissionto

1.85[1.48–3.63]ml/dlat72h(p<0.0001).TheCa-jbO2waslowerin

patientsafter72hcompared tohealthycontrols(normalvalues 7.18[6.33–7.97]ml/dl, p<0.0001). The Ca-jbO2 decreased

signiﬁ-cantlyintheﬁrst3days afteradmission, withalargerdecrease innon-survivorscomparedtosurvivors(p<0.0001),(Fig.4).This decreaseinCa-jbO2wasmainlyexplainedbyadecreasein

arte-rialoxygencontentduringtheadmission(Table2,Supplementary digitalcontent).

TheCjb-aCO2/Ca-jbO2was>1atalltimepointsinpatientsafter

cardiacarrestanddidnotchangesigniﬁcantlyduringadmission: (1.28[0.98–1.46]atadmissionand1.27[0.77–1.79]at72h),with nodifferencesbetweensurvivorsandnon-survivors(Fig.5).Inthe healthycontrolgrouptheratiowas1.16[1.01–1.25].

Arterialandjugularbulblactateconcentrationsdecreased sig-nificantlyinthefirst3daysafteradmissionfrom3.15[1.95–5.93] to 1.65[1.40–2.18]mmol/l and from 3.55[2.05–5.78] to 1.70[1.50–2.48]mmol/l,respectively (p<0.0001). Survivorsafter cardiacarresthadlowerarterialandjugularbulblactate concen-trationsthannon-survivors(datanotshown).Arterialandjugular bulblactateconcentrationsofcontrolsubjectsweresignificantly loweratalltime points(0.8[0.6–1.05]and 0.9[0.6–1.05]mmol/l respectively, p<0.0001) (data not shown). The lactate EF was 4.35[0.0–13.8]% upon admission and 0.0[−10.5–10.5]% at 72h (p=0.40).Therewasnosignificantdifferencebetweensurvivors andnon-survivors(p=0.48)(datanotshown).

Themetabolicvariableswerenotdifferentbetweenthegroup treatedwith24or72hofhypothermia(Tableelectronic supple-ment).

Fig.5.Thejugularbulb-arterialCO2toarterial-jugularbulbO2contentdifference

ratioinsurvivors(whitebars)andnon-survivors(greybars)duringtheﬁrst72h aftercardiacarrest.Stripedbarrepresentsvaluesinnormalcontrolsubjects.Data areexpressedasmean±SD.

Discussion

Intheﬁrsthoursaftercardiacarrest,MVFmcawaslow,while O2EFremainedwithininthenormalrange.Despitethisapparent

mismatch,wefoundnoevidenceoftissue hypoxia,indicating a well-adjustedbalancebetweenoxygendeliveryandconsumption inbothsurvivorsandnon-survivors.Thesedatastronglyindicate thatcerebralmetabolismisdecreased,especiallyintheﬁrsthours afterthearrest.ReductionsinaerobicmetabolismandO2EFwere

more apparentin non-survivors aftercardiac arrest, and likely reﬂectiveofirreversibleneuronaldamage.

The Cjb-aCO2 decreased during admission, due to an

abso-lutedecreaseinvenousjugularbulbCO2content.Thisdecreased

Cjb-aCO2 is most consistent with a decrease in cerebral

gly-colysis and CO2 production afterROSC, suggesting a decreased

metabolism.Thisdecrease wasstrongestin non-survivors, sug-gestinglessCO2productioninirreversiblydamagedbraintissue.

ThejugularbulbCO2 contentgraduallyrestoredtowardsnormal

valuesinpatientswithagoodoutcome,indicatingarestorationof metabolisminthesecells.ThechangesinjugularCO2contentwere

accompaniedbyacontinuingdecreaseinO2EFinpatientswith

apooroutcome,whereas agradualrestorationofO2EFtowards

normal values occurred in surviving patients. Taken together, thesemetabolicchanges after cardiacarrest are bestexplained byrestorationofneuronalfunctioninginpatientsthateventually recover,andirreversiblelossoffunctionalcerebraltissuein non-survivors.

Adecreaseincerebralmetabolismisinagreementwith pre-viousstudiesinhumansandanimalmodels[15–20].Ontheone hand,thelowCBFstateinthefirsthoursaftercardiacarrest ren-dersthebrainatriskforischemia.Alternatively,thehypoperfusion followsthe inactivity ofthe brain.Low energy supply leadsto anabruptdiscontinuationofvariousneuronalfunctions,mainly synapticneurotransmission[21].Thisreductionin neurotransmis-sionlowersmetabolismquickly,and iswidelyassumedtobea compensatorymechanism[22].Inturn,thisleadstoareductionof metabolismandaconsequentreductionofperfusion.This hypothe-sisisinagreementwithourrepeatedobservationswithcontinuous EEG:cardiacarrestleadstoiso-electricpatternswithin10–40sin allpatients,reflectinganabruptstoppingofcorticalsynaptic trans-mission.Inrecoveringpatients,rhythmsrestorewithin12–24h. Otherwise,patternsremaindisturbedinpatientsthatinsufficiently recoverneurologically[23].

Thevenous-arterialCO2toarterial-venousO2 content

differ-ence ratio as a parameter for anaerobic metabolism was not signiﬁcantlydifferentbetweenpatientsaftercardiacarrestandthe controlgroup,andindependentofoutcome.Thesedatasuggest

(5)

thatcerebralmetabolismaftercardiacarrestismainlyaerobicin nature,eveninpatientswithapoorneurologicoutcome.

Todate,moststudiesoncerebralischemia-reperfusioninjury aftercardiacarresthavefocusedonoxygen-derivedparameters todeterminethebalancebetweenoxygensupply anddemand. By calculation of a more speciﬁc parameter,we demonstrated thatcerebralmetabolismaftercardiacarrestismainlyaerobicin nature,eveninpatientswithapooroutcome.Assupplyof oxy-genwasnotalimitingfactorinthisstudy,itseemsunlikelythat enhancementofCBFinthesepatientswillimprovethepost-anoxic encephalopathy.Treatmentwithhypothermia at33◦Cdoesnot conferbeneﬁtcomparedtotreatmentat36◦C,neitherdoes pro-longedhypothermiafor48hcomparedto24himproveoutcome aftercardiacarrest[24,25].Thereductioninmetabolicactivityafter cardiacarrestismuchstrongerthancanbeinducedby temper-aturechangesintherangeof 32–36◦Corbyuseofsedationin clinicallyrelevant dosages.Thissuggeststhat interventions tar-getingtemperaturemanagement, or bloodpressure toimprove outcomeinthesepatients,havealowprobabilityofeffect.More likely,theregulationofCBFandmetabolismisdirectlyorindirectly undercontrolofpathophysiologicalprocessesthatdetermine neu-ronalsurvival.Lossoffunctionalneuraltissueaftercardiacarrestis relatedtoalargenumberofmechanisms,includingexcitotoxicity, disruptedcalciumhomeostasis,freeradicalformation,pathological proteasecascades,andactivationofcell-deathsignalingpathways [26].Interventionstudiesaimingatmanipulationofoneormoreof thesepathwaysofinjurymaybemoreeffectivethanenhancement ofCBFduringthepost-cardiacarrestsyndrome.

Thisstudyhasa number of limitations. Althoughdatawere prospectivelycollected,itisaretrospectiveanalysisofdataina rel-ativelysmallsamplefromonesinglecenter.Wefoundnosignsof anaerobicmetabolism,usingaprotocolaimingatrelativelyhigher meanarterialpressures>80mmHg.Theserelativelyhigh perfu-sionpressuresmayprovidesufﬁcientCBFforaerobicmetabolism, even inpatientswitha disturbed autoregulation. We foundno evidenceofanaerobicmetabolism,usingmethodsthatmeasure globalmetabolism.Wecannotexcludethepossibilitythatregional ischemiamightoccur.

Allpatientsweresedatedand treated withhypothermia for 24–72h.Theeffectsofsedationandhypothermiamayhaveaffected ourresults.It isgenerallyassumedthatcoolingreducescortical activity,however,this isnot a majorfactor inthetemperature rangesthatareusedinthesepatients[27,28].Propofolinduced changesarewellknown.InthedosagesthatwereusedintheICU, corticalactivityremainscontinuous[29].Evenifdiscontinuityis induced,burstsareheterogeneousandsuppressionsareshort[30] Thisisaphysiologicalresponseofarelativelyhealthybrainto seda-tionandcontrastswiththeabruptdiscontinuationofallneuronal activitywithin10–40safterinductionofhypoxia[31].Thepatients inthisstudyweretreatedwith24or72hofhypothermia(and con-comitantsedation).NosigniﬁcantdifferenceswerefoundinCBF ormetabolismdatawerefoundbetweenthe24and72htreated groups.ThesedatasupportthefactthatthechangesinCBFand metabolismismainlyrelatedtothepost-cardiacarreststate,rather thanahypothermiaorsedationeffect.

DerangementsinpHandPaCO2probablyinﬂuencedourresults.

Hypercapniaand/oracidosis caninducea reduction in cerebral metabolism The pH-dependent activity of phosphofructokinase (the enzyme responsible for the phosphorylation of fructose 6-phosphateinglycolysis)providesmechanisticsupportfor reduc-tions in CMRO2 with hypercapnia. Indeed, an accumulation of

glucose6-phosphate andfructose 6-phosphateis showninrats exposed to acute hypercapnia [32]. Additionally, hypercapnia depressescorticalactivitybyacidosis-inducedadenosinereceptor modulation[33–35].

Conclusion

Inthisstudy,lowcerebralperfusionaftercardiacarrestwasnot associatedwithanaerobicmetabolism.Itisunknownifthis hypop-erfusionisthecauseorconsequenceofasubstantialdecreaseof neuronalfunctioningandmetabolicneeds.Metabolismincreases inrecoveringpatients–consistentwithresumptionofneuronal activity–whereasinpatientswithapooroutcome,lowmetabolism reﬂectsirreversibleneuronaldamage.

Thetherapeuticandprognosticpotentialofthesenew parame-tersremaintobeestablished.

Acknowlegdements None.

AppendixA. Supplementarydata

Supplementarydataassociatedwiththisarticlecanbefound, intheonlineversion,athttp://dx.doi.org/10.1016/j.resuscitation. 2017.08.218.

References

[1].LaverS,FarrowC,TurnerD,NolanJ.Modeofdeathafteradmissiontoan inten-sivecareunitfollowingcardiacarrest.IntensiveCareMed2004;30:2126–8. [2].BisschopsLL,HoedemaekersCW,SimonsKS,vanderHoevenJG.Preserved

metaboliccouplingandcerebrovascularreactivityduringmildhypothermia aftercardiacarrest.CritCareMed2010;38:1542–7.

[3].BuunkG,vanderHoevenJG,FrolichM,MeindersAE.Cerebralvasoconstriction incomatosepatientsresuscitatedfromacardiacarrest?IntensiveCareMed 1996;22:1191–6.

[4].LemialeV,HuetO,VigueB,MathonnetA,SpauldingC,MiraJP,etal.Changesin cerebralbloodﬂowandoxygenextractionduringpost-resuscitationsyndrome. Resuscitation2008;76:17–24.

[5].Ospina-Tascon GA, Hernandez G, Cecconi M. Understanding the venous-arterialCO2toarterial-venousO2contentdifferenceratio.IntensiveCareMed 2016;42:1801–4.

[6].Mekontso-DessapA,CastelainV,AnguelN,BahloulM,SchauvliegeF,Richard C,etal.CombinationofvenoarterialPCO2differencewitharteriovenousO2 contentdifferencetodetectanaerobicmetabolisminpatients.IntensiveCare Med2002;28:272–7.

[7].Ospina-TasconGA,Bautista-RinconDF,UmanaM,TafurJD,GutierrezA,Garcia AF,etal.Persistentlyhighvenous-to-arterialcarbondioxidedifferencesduring earlyresuscitationareassociatedwithpooroutcomesinsepticshock.CritCare 2013;17:R294.

[8].Ospina-TasconGA,UmanaM,BermudezWF,Bautista-RinconDF,ValenciaJD, MadrinanHJ,etal.Canvenous-to-arterialcarbondioxidedifferencesreﬂect microcirculatoryalterationsinpatientswithsepticshock?IntensiveCareMed 2016;42:211–21.

[9].Ospina-TasconGA,UmanaM,BermudezW,Bautista-RinconDF,HernandezG, BruhnA,etal.Combinationofarteriallactatelevelsandvenous-arterialCO2 toarterial-venousO2contentdifferenceratioasmarkersofresuscitationin patientswithsepticshock.IntensiveCareMed2015;41:796–805.

[10].BisschopsLL,PopGA,TeerenstraS,StruijkPC,vanderHoevenJG,Hoedemaekers CW.Effectsofviscosityoncerebralbloodﬂowaftercardiacarrest.CritCareMed 2014;42:632–7.

[11].BisschopsLL,vanderHoevenJG,Hoedemaekers CW.Effects ofprolonged mildhypothermiaoncerebralbloodﬂowaftercardiacarrest.CritCareMed 2012;40:2362–7.

[12].WillieCK,MacLeodDB,SmithKJ,LewisNC,FosterGE,IkedaK,etal.The contribu-tionofarterialbloodgasesincerebralbloodﬂowregulationandfuelutilization inmanathighaltitude.JCerebBloodFlowMetab2015;35:873–81.

[13].AinsliePN,ShawAD,SmithKJ,WillieCK,IkedaK,GrahamJ,etal.Stabilityof cerebralmetabolismandsubstrateavailabilityinhumansduringhypoxiaand hyperoxia.ClinSci2014;126:661–70.

[14].DouglasAR,JonesNL,ReedJW.CalculationofwholebloodCO2content.JAppl Physiol1988;65:473–7.

[15].Beckstead JE, Tweed WA, Lee J, MacKeen WL. Cerebral blood ﬂow and metabolisminmanfollowingcardiacarrest.Stroke;JCerebCirc1978;9:569–73. [16].BeinB,CavusE,StadlbauerKH,TonnerPH,SteinfathM,ScholzJ,etal.Monitoring ofcerebraloxygenationwithnearinfraredspectroscopyandtissueoxygen par-tialpressureduringcardiopulmonaryresuscitationinpigs.EurJAnaesthesiol 2006;23:501–9.

[17].EdgrenE,EnbladP,GrenvikA,LiljaA,ValindS,WiklundL,etal.Cerebral bloodﬂowandmetabolismaftercardiopulmonaryresuscitation:a pathophysi-ologicandprognosticpositronemissiontomographypilotstudy.Resuscitation 2003;57:161–70.

(6)

[18].MortbergE,CummingP,WiklundL,RubertssonS.Cerebralmetabolicrateof oxygen(CMRO2)inpigbraindeterminedbyPETafterresuscitationfromcardiac arrest.Resuscitation2009;80:701–6.

[19].MortbergE,CummingP,WiklundL,WallA,RubertssonS.APETstudyofregional cerebralbloodﬂowafterexperimentalcardiopulmonaryresuscitation. Resusci-tation2007;75:98–104.

[20].SchaafsmaA,deJongBM,BamsJL,Haaxma-ReicheH,PruimJ,ZijlstraJG.Cerebral perfusionandmetabolisminresuscitatedpatientswithseverepost-hypoxic encephalopathy.JNeurolSci2003;210:23–30.

[21].HofmeijerJ,vanPuttenMJ.Ischemiccerebraldamage:anappraisalofsynaptic failure.Stroke;JCerebCirc2012;43:607–15.

[22].AttwellD,LaughlinSB.Anenergybudgetforsignalinginthegreymatterofthe brain.JCerebBloodFlowMetab2001;21:1133–45.

[23].HofmeijerJ,BeerninkTM,BoschFH,BeishuizenA,Tjepkema-CloostermansMC, vanPuttenMJ.EarlyEEGcontributestomultimodaloutcomepredictionof postanoxiccoma.Neurology2015;85:137–43.

[24].KirkegaardH,SoreideE,deHaasI,PettilaV,TacconeFS,ArusU,etal.Targeted temperaturemanagementfor48vs24hoursandneurologicoutcomeafter out-of-Hospitalcardiacarrest:arandomizedclinicaltrial.JAMA2017;318:341–50. [25].NielsenN,WetterslevJ,CronbergT,ErlingeD,GascheY,HassagerC,etal. Targetedtemperaturemanagementat33degreesCversus36degreesCafter cardiacarrest.NEnglJMed2013;369:2197–206.

[26].NeumarRW,NolanJP,AdrieC,AibikiM,BergRA,BottigerBW,etal.Post-cardiac arrestsyndrome:epidemiology,pathophysiology,treatment,and prognosti-cationaconsensusstatementfromtheinternationalliaisoncommitteeon resuscitation(Americanheartassociation,australianandNewZealandcouncil onresuscitation,europeanresuscitationcouncil,heartandstrokefoundationof Canada,InterAmericanheartfoundation,resuscitationcouncilofasia,andthe resuscitationcouncilofsouthernafrica);theamericanheartassociation

emer-gencycardiovascularcarecommittee;thecounciloncardiovascularsurgery andanesthesia;thecounciloncardiopulmonary,perioperative,andcritical care;thecouncilonclinicalcardiology;andthestrokecouncil.Circulation 2008;118:2452–83.

[27].HofmeijerJ,Tjepkema-CloostermansMC,vanPuttenMJ.Outcomepredictionin postanoxiccomawithelectroencephalography:thesoonerthebetter. Resusci-tation2015;91:e1–2.

[28].SivarajuA,GilmoreEJ,WiraCR,StevensA,RampalN,MoellerJJ,etal. Prognos-ticationofpost-cardiacarrestcoma:earlyclinicalandelectroencephalographic predictorsofoutcome.IntensiveCareMed2015;41:1264–72.

[29].HindriksR,vanPuttenMJ.Meanﬁeldmodelingofpropofol-inducedchangesin spontaneousEEGrhythms.Neuroimage2012;60:2323–34.

[30].ReddyRV,MoorthySS,MatticeT,DierdorfSF,DeitchJrRD.An electroencephalo-graphiccomparisonofeffectsofpropofolandmethohexital.Electroencephalogr ClinNeurophysiol1992;83:162–8.

[31].vanDijkJG,ThijsRD,vanZwetE,TannemaatMR,vanNiekerkJ,BendittDG,etal. Thesemiologyoftilt-inducedreﬂexsyncopeinrelationto electroencephalo-graphicchanges.Brain:JNeurol2014;137:576–85.

[32].FolbergrovaJ,NorbergK,QuistorffB,SiesjoBK.Carbohydrateandaminoacid metabolisminratcerebralcortexinmoderateandextremehypercapnia.J Neu-rochem1975;25:457–62.

[33].ZappeAC,UludagK,OeltermannA,UgurbilK,LogothetisNK.Theinﬂuenceof moderatehypercapniaonneuralactivityintheanesthetizednonhuman pri-mate.CerebCortex2008;18:2666–73.

[34].ThesenT,LeontievO,SongT,DehghaniN,HaglerJrDJ,HuangM,etal. Depres-sionofcorticalactivityinhumansbymildhypercapnia.HumBrainMapp 2012;33:715–26.

[35].DullaCG,DobelisP,PearsonT,FrenguelliBG,StaleyKJ,AdenosineMasinoSA. ATPlinkPCO2tocorticalexcitabilityviapH.Neuron2005;48:1011–23.