Evaluation of the neurotoxicity of dmso infused into the carotid artery of rat

Evaluation

of

the

neurotoxicity

of

DMSO

infused

into

the

carotid

artery

of

rat

Bulent

Bakar

*

,

Emine

Arzu

Kose

,

Sevilay

Sonal

,

Aslıhan

Alhan

,

Kamer

Kilinc

,

Ismail

Semih

Keskil

a_{KırıkkaleUniversity,SchoolofMedicine,DepartmentofNeurosurgery,Kirikkale,Turkey}

b_{KirikkaleUniversity,SchoolofMedicine,DepartmentofAnaestesiologyandReanimation,Kirikkale,Turkey} c

KirklareliStateHospital,DepartmentofPathology,Kirklareli,Turkey

d

UfukUniversity,SchoolofArtsandSciences,DepartmentofStatistics,Ankara,Turkey

e

HacettepeUniversity,SchoolofMedicine,DepartmentofBiochemistry,Ankara,Turkey

Introduction

Dimethylsulfoxide(DMSO)isanamphipathicmoleculewitha highlypolardomainandtwoapolargroups,makingitsolubleinboth aqueousandorganicmedia.So,DMSOisaveryefficientsolventfor water soluble compounds and is a hydrogen-bound disrupter.1

DMSO which is commonly used in several human therapeutic situations, such as drug-delivery systems, cryopreservation of autologousperipheralbloodstemcells,andembolisationofcerebral aneurysmsorarteriovenousmalformations(AVM)hasavarietyof biologicalactionsthathavemadeittargetofnumerousstudies.2,3

Although,it hasoccasionallybeenproposedtobe neuropro-tective,oxidativeandinducebehaviouralalterations,its mecha-nismsofactionarestillunclear.4Ithasbeenreportedtoalterthe

permeabilityofcellwallandfacilitatethetransportofsubstances acrossmembranes.4_{Besidesallofthepreviouslyaforementioned}

pharmacological applications for the treatment of different pathologies,severalsystemicsideeffectsofDMSOhavealsobeen reported.1_{Additionally,rapidintra-arterialinjectionspresumably}

causingfatalsolventrelatedadverseeffects(suchasvasospasm, angionecrosis, endothelial denuding, internal elastic lamina disruption, subarachnoid haemorrhage, stroke, and death) have also been documented.2,5 _{In animal studies, Sampei et al.}

suggested that during the infusion of DMSO into the cerebral arteries(suchasembolisationofthecerebralaneurysmorAVMs), careshouldbetakennottodamagenormalvesselsandbraintissue aroundthelesion.6_{Chaloupkaetal.showedthatDMSOtendstobe}

angiotoxic and neurotoxic.5,7 Furthermore, recently published studies point to the neurotoxicity associated with DMSO-preservedhematopoietic progenitorcellinfusion in human.8–10

Thisissuemaybeexplainedbyitshighsolubilityandpermeability. Despitetheexplanationsputforthinmanystudiesregarding histopathologicalevidenceofinflammatorychangesinthevessel

ARTICLE INFO Articlehistory:

Accepted17August2011 Keywords:

DMSO

Internalcarotidartery Neurotoxicity

ABSTRACT

Introduction: Despitetheexplanationsputforthinmanystudiesregardinghistopathologicalevidenceof theinflammatorystagerelatedwiththeinfusionofdimethylsulfoxide(DMSO)inthevesselwallandits lumen,therehasbeennoresearchtoevaluateitsneuraltoxicitywhenitisinfusedviatheintracarotid route.ThisstudywasdesignedtoevaluatethepossibleneurotoxiceffectsofDMSOonthecloserand distantbraintissueandcarotidarterywhenitwasslowlyinfusedintotheinternalcarotidarteriesofthe rats.

Methods: Therightcommoncarotidarterybifurcationwasexposedthroughamidlineneckincision,and thenexceptthoseofthecontrolgroupanimals(n=5),theexperimentalmaterial(normalsaline,n=5or anhydrousDMSO,n=10)wasinfusedintotheinternalcarotidarteryoftheWistaralbinorats.Afterthe experimentalmaterialswereadministeredintra-arterially,braintissueswereharvestedfor histopatho-logicalandbiochemicalstudiesat72hforinvestigationoftheacutestagechangesandon10thdayfor investigationofthechronicstagechanges.Internalcarotidarteriesofbothsideswerealsoremovedfor histopathologicalevaluation.Duringsacrificationoftherats,wholebodybloodofthemarecollectedfor biochemicalevaluation.

Results: Therewasnostatisticallysignificantdifferencebetweenthegroupsregardingcomparisonofthe meanvaluesofthehippocampalneuronalcellcountsandthecarotidarterydiametersinbothacuteand chronicstages.Also,meanvaluesofthelipidperoxidationlevelsofharvestedbraintissuesandserumsof thecollectedbloodsweresimilarincontrol,salineandDMSOgroups.

Conclusion: ThisexperimentalstudysuggestedthatDMSOhasnotoxiceffectontheneuralandarterial tissuesofratswhenitisslowlyinfusedintothecarotidartery.

ß2011ElsevierLtd.Allrightsreserved.

*Correspondingauthorat:KirikkaleUniversity,SchoolofMedicine,Department ofNeurosurgery,71100Kirikkale,Turkey.Tel.:+903182252485.

E-mailaddress:bulentbanrs@yahoo.com(B.Bakar).

ContentslistsavailableatSciVerseScienceDirect

Injury

j ou rna l h ome p a ge : w ww . e l se v i e r. co m/ l oc a te / i n j ury

0020–1383/$–seefrontmatterß2011ElsevierLtd.Allrightsreserved. doi:10.1016/j.injury.2011.08.021

wallandthelumen,2,5,6_{tothebestofourknowledgethere has}

been no research on the neural toxicity of DMSO infused via intracarotidroute.Inourpreviousstudy,wespeculatedthatDMSO usedinintra-arterialembolisationproceduresmayalsoenterthe circulation;andmayrapidlymoveawayfrombloodintocerebral tissueand/orcerebrospinalfluidcirculatingaroundthesevessels. Ourstudy demonstratedthat DMSOproduced someneurotoxic effectswhen itwasinfused intothesubarachnoid spaceof the rabbit.Furthermore,thesetoxiceffectswerenegativelycorrelated withthedistanceandconcentrationoftheinfusedexperimental material.11_{Regardingourpreviousstudyresults,this studywas}

designedtoevaluatethepossibleneurotoxiceffectsofDMSOtothe closeanddistantbraintissueandcarotidarterywhenitwasslowly infusedintotheinternalcarotidarteriesoftherats.

Materialsandmethods

Materials

AnhydrousDMSO(DMSO,MicroTherapeuticsInc.,Irvine,CA), andnormal salinesolutionwereusedin thisstudy. Anhydrous liquidDMSOdensityisapproximately1.10mg/ml,itsintravenous LD50is5.2–8.1g/kgfortherat,and itshalflifein theserumis

approximately60–72h.12

Anaesthesia wasperformed withintraperitoneal administra-tionof40mg/kgketamineHCl(Ketalar1

;Pfizer Inc., USA),and 5mg/kg xylazine HCl (Rompun1

2%; Bayer HealthCare AG, Germany).

TwentyWistar albinoratsof250–350mgweightwereused, whichwererandomlydividedintotwomaingroupsusedforthe acutestageinvestigation(72haftertheinjectionsofexperimental materials) and chronic stage investigation (ten days after the injection).

Theacutestagegroupwasrandomlydividedintofourgroups listedasbelow:

-Control(nochemicalmaterialwasinfused)(n:5)

-Saline(normalsalinesolutionwasinfusedinavolumeof0.1ml) (n:5)

-DMSO-A(anhydrousDMSOwasinfusedinavolumeof0.1ml; therighthemisphereofthebrain)(n:5)

-DMSO-AF(anhydrousDMSOwasinfusedinavolumeof0.1ml; thelefthemisphereofthebrain)(n:5)

Thechronicstagegroupwasrandomlydividedintotwogroups listedasbelow:

-DMSO-C(anhydrousDMSOwasinfusedinavolumeof0.1ml; therighthemisphereofthebrain)(n:5)

-DMSO-CF(anhydrousDMSOwasinfusedinavolumeof0.1ml; thelefthemisphereofthebrain)(n:5)

Methods

Thisexperimentalstudywasperformedinaccordancewiththe guidelinesfortheuseoflaboratoryanimalsubjectsinresearchset bytheEthicalCommitteeofKırıkkaleUniversity(Date:May27th, 2009;number:09/29).

All animals were placed under sedational anaesthesia with intraperitonealketamineHCl40mg/kgandxylazineHCl6mg/kg duringspontaneousrespirationat roomtemperature.Under an operating microscope, bifurcation of theright common carotid artery wasexposed through a midline neck incision,and then exceptthecontrolgroup,experimentalmaterial(normalsalineor anhydrous DMSO) was administered slowly (within 30s) at a volumeof0.1mlthroughtherightinternalcarotidartery(ICA)of theratsusinga 26G needle.7,13–18_{Afterthis procedure, allrats}

wereremovedfromsedationalanaesthesiaspontaneouslyunder theblanket.Theywerethenkeptatnormalroomtemperature,and examinedbyaneurosurgeontwiceadayforthedevelopmentof any neurological deficit. Seventy two hours later, all animals exceptthoseofDMSO-Cgroup;andtendayslaterallanimalsofthe DMSO-Cgroupwerere-sedatedwithintraperitonealketamineHCl 40mg/kgandxylazineHCl6mg/kgforsacrification.For sacrifica-tion, the whole body blood was collected from the vena cava inferior,andthentheratsweredecapitated(Fig.1).

Allratbrainsweredividedintotwohemisphereswitha cut fromsulcuscentralisandthenthehippocampalformationsofboth hemispheresweredissectedandstoredin10%buffered formalde-hyde solutionatroomtemperaturefor future histopathological examination. In addition, the ICA of the both sides was also resectedforfuturehistopathologicalevaluationofangiotoxicity.

Forbiochemicalexaminationfrontalregionssamplesdissected frombothhemispheresandtheserumofthecollectedwholebody bloodwereimmediatelystoredat 308Cindryair.Theremaining rightbraintissueswereusedforhistopathologicalandbiochemical evaluationofthepossibleneurotoxiceffectsofDMSOtothecloser braintissueswhilstleftbraintissueswereusedfortheevaluation ofitsdistanteffects.

Histopathologicalanalysis

For histopathological examination, all tissue samples were fixatedin10%bufferedformaldehydeandprocessedaccordingto theroutinelightmicroscopictissueprocessingtechnique.Serial sections of 5

m

Fig.1.Figuresdemonstrate:(a)thedissectedrightcarotidarteryoftherat,and(b)themacroscopicappearanceofthewholecerebrum,cerebellum,andcervicalspinalcord followingadministrationofDMSO.

CX-41Microscope.Thenumberofcellsineachgroupwascounted and calculated separately in three areas per section of the hippocampalCA1region.Thenthenumberoftheneuronecells wascalculatedasanaverageperrat.Thevesselwallsandlumens oftherightandleftinternalcarotidarterieswerealsomeasured fordiameterandthicknessbyusingacomputerimageanalysis system(Aperioimagescope,version10.1.3.2028).The investiga-torwhoperformedthesemeasurementswasblindtothestudy groups.

Biochemicalanalysis

Thebiochemicaldatawasobtainedfromtheconcentrationsof malonyldialdehyde(MDA)levelsinbraintissue.TheMDA(asan important indicatorof lipidperoxidation)levels were measured accordingtoamethodofMiharaetal..Theprincipleofthemethod wasbasedonthespectrophotometricmeasurementofthecolour thatoccurredduringthereactionofthiobarbituricacidwithMDA. Theconcentrationofthiobarbituricacidreactivesubstances(TBARS) wascalculatedbytheabsorbancecoefficientofmalondialdehyde– thiobarbituricacid complex.19_{In this study, allspecimens were}

evaluatedbyanexperiencedbiochemistblindedtothestudygroups, andexperimental material. Biochemicalanalyses were performedby thiobarbituric acidapplicationandthen532nm, spectrophotometry (Shimadzu1

UV-120-02Spectrophotometer)wasusedfor measur-ingthelipidperoxidationlevelsinnanomolespergramofwettissue.

Statisticalanalysis

The study results consisted of hippocampal neuronal cell counts, carotid artery diameters and serum and tissue lipid peroxidationlevels.Thevalueswerenotnormallydistributedand variationswerenothomogenousbetweenallgroups.Therefore, ANOVA(one-wayanalysisofvariance)testwasnotperformedand allresultswerestatisticallyanalysedbyusingtheKrusskal–Wallis

MultipleVariantAnalysis(thenon-parametricanalogueofANOVA) test,andpvalueslessthan0.05wereconsideredtobesignificant. Furthermore, Wilcoxon Signed Ranks Test (the non-parametric analogueofthepairedsamplesttest)wasperformedtodetermine thestatisticaldifferencesbetweentheresultsofrightandleftbrain hemispheresforeachgroup.

pvalueslessthan0.05wereconsideredtobesignificant.20,21

Results

Neurologicalexamination

Three animals displayed mild ataxia and fatigue after the injectionofDMSO;however,theywereincludedinthestudy.The other animals displayed normal neurological and physiological function.

Lightmicroscopy

In the control and saline groups neuropil in-between the neuronswitheuchromaticnucleishowedanormalarchitecture. Nooedema,infiltrationorstasiswasseenintheneuraltissueof thesegroups(Fig.2aandb).

ThespecimensoftheDMSO-AandDMSO-AFgroupsshowedno degeneration or destruction of the neuraltissue, haemorrhage, oedemaorinflammatorycellinfiltration(Fig.2candd).

AlsotheDMSO-CandDMSO-CFgroupshadnodegenerationof the neural tissue, haemorrhage, oedema or inflammatory cell infiltration(Fig.3aandb).

Furthermore, all resected internal carotid arteries appeared grossly normal withno angionecrosis. Histopathologic signs of angionecrosis, fibrin deposition, haemorrhage, disruption of internalelasticlamina,oracuteand/orchroniccellular inflamma-tion were not observed in any of the specimens investigated (Figs.4and5).

Fig.2.(aandb)Inthecontrolandsalinegroups,neuropilin-betweentheneuronswitheuchromaticnucleishowedanormalarchitectureinthecontrolandsalinegroups.No oedema,infiltrationorstasiswereseenintheneuraltissue.(candd)ThespecimensoftheDMSO-AandDMSO-AFgroupsshowednodegenerationordestructionoftheneural tissue,haemorrhage,oedemaorinflammatorycellinfiltration,respectively(HEx40).

Fig.3.(aandb)TheDMSO-CandDMSO-CFgroupshadnodegenerationoftheneuraltissue,haemorrhage,oedemaorinflammatorycellinfiltration(HEx40).

Fig.4.Allresectedcarotidarteriesofthecontrol(a),saline(b),DMSO-A(c),andDMSO-AF(d)groupappearedgrosslynormalwithnoangionecrosis(HEx40).

Fig.5.Histopathologicsignsofangionecrosis,fibrindeposition,haemorrhage,disruptionofinternalelasticlamina,orchroniccellularinflammationwerenotobservedinany ofthespecimensoftheDMSO-C(a)andDMSO-CF(b)groups(HEx40).

Hippocampalneuronalcellcountresults

There wasno statisticallysignificantdifference between the groupswhichwereevaluatedattheacutestageandbetweenthe groupswhichwereevaluatedatthechronicstageregardingthe meanvaluesoftheneuronalcellcounts.Furthermore,meanvalues ofthe acutestage groupswerenot differentfromthose of the chronicstagegroups(X2_{=9.575;p=0.088)(Table1).}

Inaddition,whenmeanvaluesoftheneuronalcellcountwere comparedforeachgroup,resultsoftherighthemisphereswerenot significantlydifferentfromthoseofthelefthemispheresregarding the acute (Z= 0.135; p=0.893) or chronic stage (Z= 0.271; p=0.786).

Inconclusion,itcanbesaidthatanhydrousDMSOgiven intra-arteriallydidnot produceanyneurotoxiceffectin theratbrain eitherattheacuteorchronicstage(Fig.6).

Carotidarterydiameters

There wasno statisticallysignificantdifference between the groupswhichwereevaluatedattheacutestageandbetweenthe groupswhichwereevaluatedatthechronicstageregardingthe meanvaluesofvesselwallthickness.Furthermore,meanvaluesof theacutestagegroupswerenotdifferentfromthoseofthechronic stagegroups(X2_{=1.573;p=0.905)(Table1).}

Additionally,whenmeanvaluesofthevesselwallthicknesswere comparedforeachgroup,resultsoftherightcarotidarterywerenot significantlydifferentfromthoseoftheleftcarotidarteryatacute (Z= 0.135;p=0.893)orchronicstage(Z= 1.753;p=0.080).

Eventually,it canbesaidthat anhydrousDMSO given intra-arteriallydidnotproduceanyvasospasmorangiotoxiceffectinthe internalcarotidarteryofratbothattheacuteandchronicstage (Fig.7).

Biochemicalanalysis

Therewasnostatisticallysignificantdifferenceeitherbetween thegroupswhichwereevaluatedattheacutestageorbetweenthe

groupswhichwereevaluatedatthechronicstageregardingthe comparison of mean values of the serum and tissue lipid peroxidationlevels.Furthermore,meanvaluesoftheacutestage groupswerenotdifferentfromthoseofthechronicstagegroups (X2_{=4.529; p=0.476and X}2_{=7.004; p=0.220, respectively for}

lipidperoxidationlevelsofthebraintissueandserum)(Table1). Additionally, when the mean values of the tissue lipid peroxidation levels were compared for each groupat different stages, results of the right hemispheres were not significantly differentfromthoseofthelefthemispheres(Z= 0.674;p=0.500 andZ= 0.405;p=0.686fortheacuteandchronicstagegroups, respectively)

Table1

Thisdescriptivetableshowsthemeanvaluesoftheneuronalcellcounts,thewallthicknessofthecarotidartery,andserumandtissuelipidperoxidationlevelsofthe DMSO-A,DMSO-AF,DMSO-C,DMSO-CF,saline,andcontrolgroups.

Group N Minimum Maximum Mean Std.deviation

Control HNC 5 40.66 64.33 57.32 9.58 CAT 5 74.26 138.60 93.74 26.03 BLPO 5 0.330 0.416 0.388 0.03 SLPO 5 0.044 0.070 0.054 0.01 Saline HNC 5 59.66 72.33 64.59 5.04 CAT 5 69.51 95.24 83.19 10.69 BLPO 5 0.372 0.450 0.412 0.02 SLPO 5 0.045 0.075 0.059 0.01 DMSO-A HNC 5 50.66 82.00 62.06 12.79 CAT 5 70.34 104.20 89.59 13.11 BLPO 5 0.320 0.434 0.393 0.04 SLPO 5 0.033 0.066 0.054 0.01 DMSO-AF HNC 5 30.00 66.33 56.19 15.19 CAT 5 81.28 98.59 88.91 6.26 BLPO 5 0.320 0.570 0.410 0.09 DMSO-C HNC 5 63.33 69.00 67.06 2.25 CAT 5 55.07 93.71 81.07 15.73 BLPO 5 0.350 0.406 0.372 0.02 SLPO 5 0.032 0.062 0.045 0.01 DMSO-CF HNC 5 61.33 71.33 66.53 3.59 CAT 5 68.42 100.00 86.01 11.35 BLPO 5 0.284 0.410 0.373 0.05

CAT:thewallthicknessofthecarotidartery;HNC:hippocampalneuronalcellcount;SLPO:serumlipidperoxidationlevel;BLPO:lipidperoxidationlevelofthebraintissue; N:numberofanimal.

Fig.6.ThemeanvaluesoftheneuronalcellcountsoftheDMSO-A,DMSO-AF, DMSO-C,DMSO-CF,saline,andcontrolgroups.Eacherrorbarshowstheminimum andmaximumofthecellcountvalues.

ThesefindingsshowthatanhydrousDMSOdidnotaffectthe lipidperoxidationlevelsatbothstages(Figs.8and9).

Discussion

DMSOhasbeenusedformanyyearsmainlyasasolventand tissuepreservative.Recently,manyclinicalreportspointedouta varietyofdosedependentadverseeffectssuchasnausea,vomiting, flushing, fever, chills, dyspnoea, cardiac symptoms, transient hypertension,hypotension,anaphylaxis,encephalopathy,amnesia andseizures.9_{However,Muelleretal.pointedoutintheirclinical}

studythatsevereadversereactionsingeneralandneurotoxicityin

particularuponinfusionofDMSO-preservedautologous peripher-albloodstemcells(DMSO-PBSC)occurredwithalowincidence (2%).TheyalsodiscussedthatDMSOseemsnottoenrichinbrain tissue,butaffectthevascularsystemresultinginvasospasmwhich thenmayleadtoalteredneuronalfunction(suchasareversible leukoencephalopathy syndrome). Finally, they concluded that patients with cerebral disease were at higher risk for such neurotoxicitybutapossiblepathogeneticrolewasnotsupported bytheliterature;andtheysaidthatinfusionofDMSO-PBSCcanbe performed safely in patients with preexisting cerebral disease despitetherareoccurrenceofneurotoxicity.22

Denko et al. reported in their animal model that DMSO administeredviaintraperitonealrouterapidlylodgedinsoftand hardtissues.Inthatstudy,higherlevelsofDMSOwerefoundin softtissues(suchasspleen,stomach,lung,vitreushumour,brain, kidney,heartandaorta,respectively)thanincartilageorbone.23

Kayeet al.observedin theirstudy thatDMSO wasrapidlyand extensivelydistributedthroughthetissuesandbraininmicewhen itwasinfusedviaintravenousroute,buttheydeducedthatbrain tissuescontainedthelowestconcentrationofDMSOamongstthe tissuesstudied.Theyalsoreportedthatintravenousinjection of DMSO had not produced any histological evidence of central nervoussystemchanges.24_{Sampeietal.demonstratedthatwhen}

it was administered via the intracarotid route, DMSO led to irreversible degenerativechangesincludingvacuolisation ofthe endothelialcellsandpartialnecrosisofvesselwallsmoothmuscle. Onlightmicroscopy,theyalsoobservedscartyareasofbackground andpalestaininginareascorrespondingtobraintissuesdueto exudatedEvansblueinacutestageofthestudy(5min).Onthe other hand, they observed no obvious abnormality and no exudationof Evans blue in anygroup ofthe chronicstage (10 days).However,theydidnotevaluateindetailtheneurotoxicityof the DMSO. At the end of their study, they suggested that catheterisation should be strictly superselective when DMSO was used at high concentrations.6 _{Moreover, Chaloupka et al.}

showedintheiranimalmodelthatDMSOproducedamononuclear andneutrophilpreponderantinflammatorycellinfiltrationinthe vesselwallandcontiguousneuraltissue.Theyalsoconcludedthat rapidinfusion(approximately15sorless)oftheanhydrousDMSO into the swine retemirable wasassociated withsubarachnoid

Fig.7.ThemeanvaluesofarterialwallthicknessoftheDMSO-A,DMSO-AF, DMSO-C,DMSO-CF,saline,andcontrolgroups.Eacherrorbarshowstheminimumand maximumofthethicknessvalues.

Fig.8.ThemeanvaluesofthetissuelipidperoxidationlevelsoftheDMSO-A, DMSO-AF,DMSO-C,DMSO-CF,saline,andcontrolgroups.Eacherrorbarshowsthe minimumandmaximumofthelipidperoxidationlevelvalues.

Fig.9.ThemeanvaluesoftheserumlipidperoxidationlevelsoftheDMSO-A, DMSO-C,saline, andcontrolgroups.Eacherrorbarshows theminimumand maximumofthelipidperoxidationlevelvalues.

haemorrhage,severevasospasmresultedincerebralinfarctionand death.5 Furthermore, theypointedout in 1999that lowerdose rates of slower (approximately 30s or more) superselectively infused anhydrous DMSO was associated with no significant clinicalhaemodynamicorneurologicsequale.7,25

Inourstudy,wedidnotobserveanyearlyorlatevasospasm, angionecrosis, arterial wallrupture, granulomatousobliterative angiitisorintimalhyperplasiainanyhistopathologicalsliceofthe carotidartery.So,weagreewiththeauthorswhoadvocateslow infusionofDMSOinordertoavoiditsangiotoxiceffectsduring intra-arterialadministration.Additionally, histopathologicaland biochemicalresultsofourstudyrevealedthatDMSOdestroyedthe architectureofneither brainparenchymanorvesselwallofthe carotidartery.Therewasnoinflammatorycellinfiltrationinany slice ofthebrainorcarotid artery, andtheneuronal cellcount resultsofallgroupswerealmostsimilar.Thesefindingssuggested thatDMSOdoesnothaveanyangiotoxicityorneurotoxicitywhen itisslowlyadministeredviatheintracarotidroute.Also,itcanbe suggested that DMSO cannot alter the blood–brain barrier permeability when slowly infused via the intra-arterial route sinceit dissolvesrapidlyin thewater contentofthebloodand moved away from the neural tissue by arterial circulation.6,24

Although the blood–brain barrier disruption developed by the DMSOwasnotevaluatedbyinjectionoftheEvansblue,itmaybe reasonablyhypothesisedthatDMSOcouldpassthroughtheblood– brain barrier with difficulty. Therefore, it could not reach the concentrationofitstoxicdoseinneuraltissue. Keaneetal.and Broadwelletal.reportedintheirstudiesthatDMSOdidnotalter themorphologyofendothelialcellsorbrainparenchyma,andthe permeabilityofbothblood–brainbarrierandskeletalmuscle.26,27

Moreover,Murayamaetal.showedinaswinemodelthatthebrain slicesoftherapid(0.5ml/5s)DMSOinjectiongroupdisclosedlow gradeencephalitis witha mixed infiltrateof acute and chronic inflammatory cells, acutesubarachnoid haemorrhage, and focal acute necrosis in subpial locations. However, they observed minimum or no angiographic vasospasm, minimal adventitial inflammatoryresponse,andnoclinicalcomplicationsintheslow (0.5ml/30–120s)injectionDMSOgroup.25_{Asaconclusion,these}

datadiscussedabovealsosupportouraforementionedspeculation aboutadvantagesoftheslowinfusionratesofDMSO.

DMSOwaspreviouslydescribedasafreeradicalscavengerand antioxidant.3,28_{Inthisstudy,theserumandtissuelipid}

peroxida-tionlevelswerefoundsimilarinallstudygroups.Thisresultmay beexplained withthehypothesis ofDMSO asnot causing free radicalproductionfromthemembrane phospholipidsof neural tissue. Another explanation may be related with the high permeabilityofDMSO.ItcanbesuggestedthatDMSOcouldnot showitstoxic effectson neuraltissueand vascularwallof the carotidarterysinceitwasrapidlymovedawayfromthecerebral tissue by the arterial circulation. On the other hand, growing clinicalexperience in thetreatment of cerebral aneurysmsand arteriovenousmalformationsisgainedwithOnyx(Micro Thera-peutics,Irvine, California)which iscurrentlytheonly commer-ciallyavailablenonadhesiveliquidembolicagentusingDMSOas the carrier solvent.2 _{During such intra-arterial embolisation}

proceduresDMSO mayalsoenter intocerebral tissue fromthe circulatingblood;andmayleadtotoxiceffects.11Tothebestofour knowledge,thereisstillnoinvestigationinordertodemonstrate anylocalisedneuraltissuetoxicityofDMSOwheninfusedintoa selectedcerebralartery.Wesuggestthatthetoxicityprofileofthis solventshouldbeevaluatedinfuturestudiesinwhichDMSOis infused intoselected cerebral arteries (suchas middle cerebral artery,anteriorcerebralartery).

Thispreliminarystudyhasa fewpitfalls. First,althoughthis evaluationmaypavethewayforfuturestudiesonthissubject,it doesnotcontainresultsofmorespecificbiochemicalanalysesfor

othercytotoxic pathwaysof DMSO intheacute and/orchronic stages.Additionally,thisstudyshouldbesupportedwithelectron microscopic findings which can show whether there are any ultrastructural findings of an inflammatory response and/or neuronal necrosis. Second, this preliminary study is far from explaining the toxic effects of DMSO in the very acute stage because of the selected time period. Also, it is far from demonstrating the toxic effects of rapid infusion of DMSO on carotid artery and cerebral tissue either histopathologically or biochemically.Third,weshouldevaluatethepossibleneurotoxic effects of theDMSO in a rat model withcerebral disease (e.g. subarachnoid haemorrhage model, hypoxia/reperfusion model etc.).But,inliterature,manyreportspointedoutthatavarietyof dose dependent adverse effects (such as nausea, vomiting, flushing, fever, chills, dyspnoea, cardiac symptoms, transient hypertension,hypotension,anaphylaxis,encephalopathy,amnesia andseizures)mayappearwhenDMSOusedasasolventandtissue preservative is given through intravenous route. And, we have expectedthatwecouldalsoexplainthisconfusedsituationwith thisstudybyusinghealthyratmodel.9_{Additionally,wethought}

thatthepossibleneurotoxiceffectsoftheDMSOmaybeconfused withtheprimercerebraldiseasefindings whensuchananimal modelisused.So,wedidnotprefersuchananimalmodel.Forth, thisstudycouldnotdemonstratethedirecttoxiceffectsofDMSO onneuraltissuesbecauseithasnotbeendirectlyinjectedintothe animalbrainparenchyma.Fifth,sincetheinternalcarotidarteryis usedforinfusion,thisstudyisfarfromdemonstratingthelocalised neural tissue toxicity of DMSO when infused into a selected cerebralartery.Sixth,thisstudycouldbesupportedbyinjectionof Evans blue which would demonstrate the blood–brain barrier disruptiondevelopedbyDMSO.

Conclusion

ThisstudydemonstratedthatDMSOisnottoxiconthecarotid arteryandneuraltissueofratsintheacuteandchronicstageinrat whenitisslowlyadministeredviatheintra-carotidroute.

Conflictofintereststatement

Theauthorsdeclarethattheyhavenoconflictofinterest.

References

1.SantosNC,Figueira-CoelhoJ,Martins-Silva J,SaldanhaC.Multidisciplinary utilization ofdimethylsulfoxide: pharmacological, cellular,and molecular aspects.BiochemPharmacol2003;65:1035–41.

2.DudeckO,JordanO,HoffmannKT,etal.Organicsolventsasvehiclesfor precipi-tatingliquidembolics:acomparativeangiotoxicitystudywithsuperselective injectionsofswineretemirabile.AJNRAmJNeuroradiol2006;27:1900–6. 3.JacobSW,HerschlerR.PharmacologyofDMSO.Cryobiology1986;23:14–27. 4.NasrallahFA,GarnerB,BallGE,RaeC.Modulationofbrainmetabolismbyvery

low concentrationsofthecommonlyuseddrugdelivery vehicledimethyl sulfoxide(DMSO).JNeurosciRes2008;86:208–14.

5.ChaloupkaJC,Vin˜ uelaF,VintersHV,RobertJ.Technicalfeasibilityand histo-pathologicstudiesofethylenevinylcopolymer(EVAL)usingaswine endovas-cularembolizationmodel.AJNRAmJNeuroradiol1994;15:1107–15. 6.SampeiK,HashimotoN,KazekawaK,etal.Histologicalchangesinbraintissue

andvasculatureafterintracarotidinfusionoforganicsolventsinrats. Neurora-diology1996;38:291–4.

7.ChaloupkaJC,HuddleDC,AldermanJ,etal.Areexaminationoftheangiotoxicity ofsuperselectiveinjectionofDMSOintheswinereteembolizationmodel.AJNR AmJNeuroradiol1999;20:401–10.

8.Bauwens D, Hantson P,Laterre PF,et al. Recurrentseizureand sustained encephalopathyassociatedwithdimethylsulfoxide-preservedstemcell infu-sion.LeukLymphoma2005;46:1671–4.

9.Ju´niorAM,ArraisCA,SaboyaR,etal.Neurotoxicityassociatedwith dimethyl-sulfoxide-preserved hematopoietic progenitor cell infusion. Bone Marrow Transplant2008;41:95–6.

10.WindrumP,MorrisTC.Severeneurotoxicitybecauseofdimethylsulphoxide followingperipheralbloodstemcelltransplantation.BoneMarrowTransplant 2003;31:315.

11.BakarB,OruckaptanHH,HazerBD,etal.EvaluationofthetoxicityofOnyx comparedwithn-Butyl2-Cyanoacrylateinthesubarachnoidspaceofarabbit model:anexperimentalresearch.Neuroradiology2010;52:125–34.

12.MottuF,LaurentA,RufenachtDA,DoelkerE.Organicsolventsfor pharmaceu-ticalparenteralsandembolicliquids:areviewoftoxicitydata.PDAJPharmSci Technol2000;54:456–69.

13.CekirgeHS,SaatciI,OzturkMH,etal.Lateangiographicandclinicalfollow-up resultsof100consecutiveaneurysmstreatedwithOnyxreconstruction:largest single-centerexperience.Neuroradiology2006;48:113–26.

14.MolyneuxAJ,CekirgeS,SaatciI,Ga´lG.CerebralAneurysmMulticenter Euro-peanOnyx(CAMEO)trial:resultsofaprospectiveobservationalstudyin20 Europeancenters.AJNRAmJNeuroradiol2004;25:39–51.

15.MounayerC,HammamiN,PiotinM,etal.Nidalembolizationofbrain arterio-venous malformationsusing Onyxin 94patients. AJNRAmJ Neuroradiol 2007;28:518–23.

16.PamukAG,SaatciI,CekirgeHS,AyparU.Acontributiontothecontroversyover dimethylsulfoxidetoxicity:anesthesiamonitoringresultsinpatientstreatedwith Onyxembolizationforintracranialaneurysms.Neuroradiology2005;47:380–6. 17.vanRooijWJ,SluzewskiM,BeuteGN.BrainAVMembolizationwithOnyx.AJNR

AmJNeuroradiol2007;28:172–8.

18.WeberW,KisB,SiekmannR,KuehneD.Endovasculartreatmentofintracranial arteriovenousmalformationswithOnyx:technicalaspects.AJNRAmJ Neuror-adiol2007;28:371–7.

19.MiharaM,UchiyamaM.Determinationofmalonaldehydeprecursorintissues bythiobarbituricacidtest.AnalBiochem1978;86:271–8.

20.NieNH,HullCH,JenkinsJG.SPSS:statisticalpackageforsocialscience.NewYork: McGrawHillInc.;1975.

21.TabachnickBG,FidellLS.Usingmultivariatestatistical.NewYork:HoperCollins CollagePublishers;1996.

22.MuellerLP,TheurichS,ChristopeitM,etal. Neurotoxicityuponinfusion ofdimethylsulfoxide-cryopreservedperipheralbloodstemcellsinpatients withand withoutpre-existingcerebral disease.Eur JHaematol2007;78: 527–31.

23.Denko CW,Goodman RM,MillerR, Donovan T.Distribution ofdimethyl sulfoxide-35Sintherat.AnnNYAcadSci1967;141:77–84.

24.KayeTS,EgorinMJ,RiggsJrCEetal.Theplasmapharmacokineticsandtissue distributionofdimethylsulfoxideinmice.LifeSci1983;33:1223–30. 25.MurayamaY,Vin˜ uelaF,UlhoaA,etal.Nonadhesiveliquidembolicagentfor

cerebralarteriovenousmalformations:preliminaryhistopathologicalstudiesin swineretemirabile.Neurosurgery1998;43:1164–75.

26.BroadwellRD,SalcmanM,KaplanRS.Morphologiceffectofdimethylsulfoxide ontheblood–brainbarrier.Science1982;217:164–6.

27.KeaneDM,GrayI,PanuskaJA.Ineffectivenessofdimethylsulfoxideinaltering thepermeabilityoftheblood–brainbarrier.Cryobiology1977;14:592–7. 28.SwansonBN.Medicaluseofdimethylsulfoxide(DMSO).RevClinBasicPharm