The evaluation of the effects of steroid treatment on the tumor and peritumoral edema by DWI and MR spectroscopy in brain tumors

(1)

Original

research

article

The

evaluation

of

the

effects

of

steroid

treatment

on

the

tumor

and

peritumoral

edema

by

DWI

and

MR

spectroscopy

in

brain

tumors

Cahit

Kural

a

,

Gokce

Kaan

Atac

b

,

Ozkan

Tehli

a

,

Ilker

Solmaz

a

,

Caglar

Temiz

a

,

Irgen

Hodaj

a

,

Yusuf

Izci

a,

*

a

DepartmentofNeurosurgery,UniversityofHealthSciences,GulhaneEducation andResearchHospital,Ankara,

Turkey

b_Department_of_Radiology,_Ufuk_University,_Ankara,_Turkey

*Correspondingauthorat:DepartmentofNeurosurgery,UniversityofHealthSciences,GulhaneEducationandResearchHospital,06010

Etlik,Ankara,Turkey.

E-mailaddress:yusuﬁzci@yahoo.com(Y.Izci).

a

r

t

i

c

l

e

i

n

f

o

Articlehistory: Received25December2017 Accepted4March2018 Availableonlinexxx Keywords: Glioma Metastasis Steroid

Diffusionweightedimage

Spectroscopy

a

b

s

t

r

a

c

t

Objective:To investigatetheeffects ofdexamethasoneonbraintumorand peritumoral

edemabydifferentsequencesofmagneticresonanceimaging(MRI).

Materialsandmethods: MRIwasperformedin28patientswithbraintumor.Patientswere

dividedintothe3groupsbasedonthehistologicaldiagnosis;GroupI:high-gradeglialtumor,

GroupII: low-gradeglialtumor, andGroupIII: brainmetastasis.The measurementsof

peritumoraledemavolumeandapparentdiffusioncoefﬁcient(ADC)valueswereperformed

whilethepeakareasofcerebralmetabolitesweremeasuredbyspectroscopyingroupsIand

II. The changes in edema volumes, ADC values and cholin/creatine peak areas were

compared.

Results:ThevolumeofperitumoraledemawasdecreasedingroupsIandII,butincreasedin

groupIIIafterdexamethasonetreatment.Thesechangeswerenotstatisticallysigniﬁcantfor

3groups.ADCvaluewasdecreasedingroupIandincreasedingroupsIIandIII.Changesin

ADCvalueswerestatisticallysigniﬁcant.Cholin/creatinepeakareasweredecreasedafter

dexamethasoneingroupsIandII,butthesechangeswerealsonotsigniﬁcant.

Conclusion: Dexamethasonehasnosigniﬁcanteffectonthevolumeofperitumoraledemain

glialtumorandmetastasis.Moreover,dexamethasoneincreasestheﬂuidmovementsinlow

gradegliomasandmetastases,decreasesinhighgradegliomas.However,more

compre-hensiveclinicalstudiesareneededtoshowtheeffectsofdexamethasoneonbraintumors

andperitumoraledema.

Available

online

at

www.sciencedirect.com

ScienceDirect

journalhomepage:http://www.elsevier.com/locate/pjnns

https://doi.org/10.1016/j.pjnns.2018.03.002

(2)

1. Introduction

About 80%of brain tumors are primary,metastatictumors

contribute tothe remaining 20%. Brainis thesecond most

common site for metastasis and accounted for 15% of all

metastases.Braintumorsmayoccuratanyage,buttheyare

observedmorecommonbetweentheagesof55and65years.

Mostcommonprimarybraintumorsaregliomas[1–4].

Glialtumorsarethewidestgroupofallintracranialtumors

(40–45%)[5].Gliomasmaybesolitaryormulticentric[2].These

tumorsaredividedinto2groupssuchaslow-gradegliomaand

high-gradegliomabasedonhistologicalexaminationofthe

tumor specimen. Subependymal giant cell astrocytoma,

pilocyticastrocytoma,ganglioglioma,anddiffuseastrocytoma

areexamplesoflow-gradeglialtumors.Anaplastic

astrocyto-ma,anaplastic oligodendrogliomaandglioblastoma arethe

examplesofhigh-grade glialtumors.Althoughtheaccurate

diagnosisofgliomasismadebyhistologicalexaminationof

the tumor tissue, many studies attempted to reveal some

biomarkersofbraingliomasinbloodandinotherbodyﬂuids

[3].

Metastatictumors ofthebraingenerateabout50%ofall

supratentorialbraintumors.Breast,lung,malignant

melano-ma,andgastrointestinaltractmalignanciesfrequently

metas-tasizetothebrain[1,5,6].

Braintumorsareusuallydiagnosedwithimaging

techni-ques.Currently,magneticresonanceimaging(MRI)iswidely

acceptedallovertheworldasthebestimagingstudyforthe

detectionofbraintumors [5].Thecharacteristicsof solitary

metastasisandprimarygliomasarenon-speciﬁcin

conven-tionalMRIstudiesandcannotbereliablydistinguishedbythis

examination.Contrastagentuptakeoftumorcellscanbeseen

bothoftumorsandvaryingdegreesofperitumoraledemacan

beobservedinMRIofpatients.Themostimportantcriteriafor

histologicalgradingofgliomasarevascularproliferationand

thedegreeofcellularity.Contrast-enhancedMRIscansmay

provideinformationaboutthevascularityoftumor.

Diffusion-weighted and diffusion-tensor imaging may be useful to

provideinformationaboutcelldensityoftumor.Metastases

andhighgradegliomascausedifferenttypesofperitumoral

edemainthebrain.Inﬁltrativeedemaisobservedingliomas,

while metastases form pure vasogenic edema. Apparent

diffusioncoefﬁcient(ADC)measurementsareusedtoseparate

thesetwotypesofedemainthebrain[6].Magneticresonance

spectroscopy (MRS) is a method that distinguishes tissue

metabolitesusingdifferentresonancepeaks[7].High

cellu-larity and cell-cycle secondary choline (Cho) increase are

usuallyseeningliomasandN-acetylaspartate(NAA)

reduc-tiondrawattentionwhenneuronswerereplacedbymassor

normalneuronstakedamage.Chosignalsarehigherinhigh

gradegliomascomparedtolowgradegliomas.

Dexamethasoneisamainglucocorticoidagentthatisused

totreatbrainedemasecondarytotumors.Itwasbeguntouse

intheearly1960sanditwaspreviouslyshownthatthe

pre-operative dexamethasone administration reduces

peritu-moraledemaandsomortality.Dailydosemayrangeabout

4–100mg/day.Itisalsoreportedthatthemostpowerfuleffect

of steroids begins within 24–72h of treatment [8,9].

Dexa-methasonetreatmentdecreasesbrainedemawithoutdistinct

absorptioneffect.Although thereare alotof articlesabout

dexamethasone'seffectonreductionoftumorsizeinaddition

todecreasingbrainedema,thesestatementsarenotwidely

acceptedbyscientists[8,10].

Thepurposeof thisstudyistoinvestigatetheeffectsof

dexamethasone on the intensity of primary tumor and

peritumoral edema using advanced MRI techniques.

Diffu-sion-weightedimaging(DWI),T2-weighted-MRIandMRSwere

usedforthispurpose.

2. Materials

and

methods

Afterreceivingapprovalfromournationalethicscommittee

(Approval no: 25.04.2012/i B.10.4.ISM.4.06.68.49), 28 patients

wereenrolledinthisstudy.Allofthepatientswereoverthe

18-years old and signed written consent form for this study.

Pregnant women and patients who previously underwent

brain tumor surgery had been excluded from this study.

Seventeenof28patientsweremaleand11werefemale.Mean

agewas46 years(46.0018.33)for malepatientsand 54.45

(54.4513.32)yearsforfemalepatients.Basedonhistological

diagnosis,thepatientsweredividedinto3groups:

Group1:High-gradetumors(n=11)

Group2:Low-gradetumors(n=10)

Group3:Metastatictumors(n=7)

Thediagnosiswashighgradeglialtumorin11patients,low

grade glialtumors in10patientsand metastasisin7of 28

patients.Oneofthehighgradetumorswasgliosarcoma,while

the others were glioblastoma. One of the low grade glial

tumors wasganglioglioma,2werepleomorphic

xantoastro-cytomas, 2 were oligodendrogliomas and 5 were diffuse

astrocytomas.Oneofthemetastatictumorswasbreastcancer

metastasis,theotherswerelungcancermetastasis.Locations

oftumorswerefrontal,temporal,parietalandoccipitallobes

respectivelyinorderoffrequency.

Themaincomplaintsofpatientswereheadache,fatigue,

armorlegweakness,seizureandspeechdisorder.Complaints

weremuchmoreinthepatientswithmetastatictumorsand

high grade gliomas. Eleven of 28 patients had normal

neurological examination; neurological deﬁcit was present

in remaining 17 patients at different levels. Patients with

neurologicaldeﬁcitswere94.5%ofmetastasisandhighgrade

gliomapatients.

All patients with the diagnosis of brain tumor were

screened using 3T magnet (Achieva 3 T, Philips Medical

Systems,TheNetherlands)preoperatively.Besideswithaxial

T1-andT2-weightedspin-echoimagings,T2-Aﬂuid

attenuat-edinversionrecovery(FLAIR),diffusion-weightedaxial

echo-planar,and post-contrastaxial,coronaland sagittalimages

were also obtained. Spectroscopy sequences were also

performed.Magneticresonancesequenceswhichwereused

inthisstudyareshowninTable1.Eight-channelheadcoilwas

used during the cranial MRI. Tenmilliliters of intravenous

gadoteratemeglumine(Dotarem®_,_Guerbet)_was_administered

in contrast-enhanced studies. MRI's were performed

pre-treatmentand48haftertheinitiationofsteroidtreatment.A

(3)

steroid treatment. MRI cross-sections were prepared for

measurementandtransferredasDICOMformattoaseparate

computer(MacPowerBookG4,Apple,USA).Evaluationswere

performed using a program that used open source DICOM

processingandvisualization (OsiriX32-bit,Pixmeo,Geneva,

Switzerland).

Thefollowing formulaswereused tocalculate the total

volumeofperitumoraledema.

1. Cross-sectional area of the edema of the tumor=

cross-sectionoftumor(slicethickness+cross-sectional

thick-ness)

2. Totalvolumeofedema:tumorvolumeofﬁrstcross-section

+tumorvolumeofsecondcross-section+tumorvolumeof

n.section

Duringtheassessment,areaoftheperitumoraledemawas

determinedmanuallyinT2axialslicesinordertoincludethe

tumor.Thehyper-intenseareathatbelongstotheedemawas

drawnmanually incircumferential fashion for each

cross-section(Fig.1).Volumeof edema(Fig.2)for eachslicewas

calculatedbymultiplyingthesumof cross-sectionaledema

areaandthecrosssectionslicethickness.Thetotalvolumeof

edemawascalculatedbyadditionofeachslicevolume(Fig.3).

Theareawascalculateddigitally.Thedatawastransferredtoa

database (Excel 2007, Windows 2007, Seattle, USA). Total

volumeof edema andtotal tumorvolume beforeand after

steroidtreatmentweredeterminedbythesecalculations.

ADCmapswerecreatedfromdiffusionweightedimagesby

ADC Map application on the same computer. A group of

diffusion weighted cross-section of twenty image withthe

valuesof b=0 andb=1000 processed by the programand

createdamap.Fromthemap,about2mm2_wide_area_created

fromtumor,edemaareaandatotherhemispheresymmetrical

oftumorbutfromnormal-appearingarea.Thisprocesswas

repeated threetimesand resultedwithcreating anarea of

interest (ROI=region of interest)withminimal ADCvalues

wereaveraged.ADCvaluesandratiosofaveragewerecreated

with Excel program as described previously. The ratio of

tumor-edema area, tumor-normal appearing area and

nor-mal-edemaareawascalculatedseparately.

Cho, Cr and NAA values generated from tumor and

peritumoralareasinMRSthatwerecreatedwithmulti-voxel

spectroscopicimageswrittenseparatelyandwerecalculated

usingthesamedatabase(Excel2007)beforeandaftersteroid

treatment. SPSS 15.0program (Standard version, SPSS Inc.,

USA) was used for statistical evaluation. The changes in

peritumoraledemavolumebeforeandaftersteroidtreatment

and numerical changes in diffusion images related to the

treatmentwereassessedusingWilcoxonandFriedmantest.

Neurological examination ﬁndings before and after

dexa-methasonetreatmentwerecomparedusingChi-squaretest.

p-Valuelessthan0.05wasconsideredsigniﬁcant.

3. Results

The mean volume of peritumoral edema before and after

steroid treatment for all patientsisshown inTable 2. The

meanvolumeofperitumoraledemain28patientsbeforethe

treatment with dexamethasone was 1790.8489mm3 _(41.73_–

4945.34). Mean volume of peritumoral edema after the

treatment was 1733.8743mm3 (35.83–4890). Despite there

was adecrease inthe volumeof edema aftertreatment, it

wasnotstatisticallysigniﬁcant(p=0.56).

Themeanperitumoraledemavolumeof11patientswith

highgradeglialtumors(Group1)was3014.6173mm3_(ranged

between 1771.73 and 4945.34) before the treatment,

2959.5561mm3 _(ranged _between _1896.32 _and _4890.52) _after

thetreatment.Thechangeinperitumoraledemavolumewas

not statistically signiﬁcant (p=0.76) (Fig. 4). The mean

peritumoraledemavolumeof10caseswithlowgradegliomas

(Group2)was635.554mm3(rangedbetween41.73and2014.08)

inpre-treatmentperiod,503.277mm3(rangedbetween35.83

and 1678.7)after thetreatment.Thechange inperitumoral

edema volumewas not statistically signiﬁcant(p=0.73)in

Group2.Themeanperitumoraledemavolumein7patients

withmetastaseswas1518.2057mm3_(ranged_between_174.87

and3006.93)beforethetreatmentand1565.7943mm3(ranged

between156.72and3062.02)afterthetreatment.Thechangein

peritumoral edema volume was notstatistically signiﬁcant

(p=0.7)inmetastasisgroup(Table3andFig. 5).Themean

volumeofperitumoraledemainGroup2waslessthanthe

Table1–TheparametersandMRIsequencesthatwereusedinthisstudy.

Sequence TR TE Slicethickness(mm) Slicegap(mm) Flipangle Resolution

T1 536 14 4.5 6.2 90 512512

T2 2790 80 5 6.5 150 512512

FLAIR 11,000 120 4.5 6.2 90 512512

Diffusion 2690 68 4.0 5.0 90 490539

Fig.1–Hand-drawnimageforthetumoredemasinglecross

sectionareaandtherepresentativeimageof

(4)

othergroups.Inthisgroup,2patientswithgangliogliomaand

pleomorphic xantoastrocytoma did not have signiﬁcant

peritumoral edema. Based on our data, dexamethasone

treatment does not modify signiﬁcantly the volume of

peritumoraledemainglialtumorsandmetastatictumors.

DWIwas performed inall cases and ADC valuesinthe

tumor were calculated before and after dexamethasone

treatment(Table4).ThehighestADCincreasewasdetected

inmetastatictumors(Group3).ADCvaluedecreasedinhigh

gradeglialtumors(Group1).ADCincreaseinlowgradeglial

tumors(Group2)wassimilartothoseofmetastatictumors.

ADCvaluesinperitumoraledemabeforeandafter

dexameth-asonetreatmentareshowninTable5.ADCvaluesfornormal

brainparenchymabeforeandafterdexamethasonetreatment

aregiveninTable6.ThehighestADCincreasewasobservedin

Group3(Table6).ADCvaluesdecreasedinhighgrade glial

tumors. The increase in ADC values of normal brain

parenchymainGroup2wasalsoclosetoADClevelinGroup

3.

The effect of dexamethasone treatment on cerebral

metabolitesinthetumoralandperitumoralareaof21glioma

cases wasalsoinvestigated inthisstudy.Cho/Crquantiﬁed

peakareasareshowninTable7.InGroup1,themeanCho/Cr

peakareawas170mm2_(8.9_–280)_before_the_{dexamethasone}

treatmentand167mm2(8.7–267)afterthetreatment.

Dexa-methasonetreatmentdidnotcauseastatisticallysigniﬁcant

changeinMRSofhighgradeglialtumorgroup(p=0.59).In

Group 2, the mean Cho/Crpeak area was 126mm2₍₇_–146)

beforethe treatmentand125mm2_(7.1_–148) _after_the

treat-ment.Dexamethasonetreatmentdidnotcauseastatistically

signiﬁcantchangeinMRSoflowgradeglialtumorsanddidnot

decreaseintensityofthetumoralcellsatperitumoralregion

(p=0.59)(Fig.6).

Someneurologicalﬁndingssuchasfullorpartialextremity

weakness, slurred speechandconsciousnessweredetected

beforedexamethasonetreatmentin17(60.71%)of28patients

withthediagnosisofbraintumors.Neurologicalexamination

was innormalrangesin11patients.Neurological

improve-mentwasseenin9(52.94%)casesafterthedexamethasone

treatment,butneurologicalﬁndingsremainedunchangedin8

patients.Therewasnoneurologicalworseninginanypatient

afterthetreatment.Theeffectofdexamethasonetreatment

on the patient's neurological condition was statistically

signiﬁcant(p<0.001).

4. Discussion

Braintumorisasigniﬁcanthealthproblemnowadaysandmay

causemortalityeveniftreatedsurgically.Themostcommon

primarybraintumorisglialtumororglioma.Althoughitcan

Fig.2–T2-weightedcrosssectionsofeachpatient'stumor(A).Slicethicknessmultipliedbysumofcross-sectionalthickness

ofthespaceandsovolumeofedemacanbecalculated(volumeoftheslice=aTh)(B).

Fig.3–Tumoredemavolumewasmeasuredforeach

(5)

bedetectedatanyagegroup,themeanageis62years.Itis40%

more common in males than females. Mosttypes of glial

tumors are malignant and the average life expectancy of

patientsrangedfrom12to24months.Metastasesarethemost

common brain tumors in adults. It has been previously

reportedthat25–40%ofpatientswithsystemiccancerhave

brainmetastasis[7,11].

MRI is the gold standard imaging technique for the

diagnosis and follow-up of brain tumors [2,3,12]. MRI was

ﬁrstdescribedin1946byBlochandPurcell.In1980,Hawkens

revealedmultiplane(multiplanar)featureofMRIand

identi-ﬁed ﬁrst brain tumor with this technique. Contrast agent

(gadolinium) was used for MRI for the ﬁrst time in 1984.

Advanced MRI techniques, such as DWI, Diffusion Tensor

Imaging(DTI),perfusionMRIandMRSprovidemore

informa-tionbeyondtheanatomicalknowledge[12].

Diffusion-weight-edsequenceofMRIisdescribedbyStejskalandandTannerin

1965. Diffusion of H2O (water) molecules in the tissue

contributeslessthequalityofimagingatconventionalMRI.

Butitispossibletoviewthemovementofwatermoleculesina

verystrongmagnetic ﬁeldgradientatdiffusion MRI[6].DTI

techniquesareusedfordifferentiationandgradingofgliomas,

Table2–Thechangesofmeanperitumoraledemavolumeafterdexamethasonetreatmentinallpatients.

n Meanperitumoraledemavolume(mm3₎_Standard_deviation_Minimum_Maximum _p*

Beforethedexamethasonetreatment 28 1790.8489 1312.48019 41.73 4945.34 0.56 Afterthedexamethasonetreatment 28 1733.8743 1341.95685 35.83 4890.52 * _Friedman_test.

Fig.4–72years-oldfemalepatientwithhighgradeglioma(gliosarcoma).(A)T2axialsectionsedemawasmarked.(B)A

slightlydecreaseinthevolumeofedemawasobservedtwodaysaftersteroidtreatment.(C)MRSimagesafterthetreatment

showsignificantincreaseinpeakNAA.(D)Inthepost-treatmentMRI,T1-axialcrosssectionsshowedhypointenseedema

(6)

metastasis,lymphomaandmeningiomas. MRSisamethod

that distinguishes tissue metabolites using different

reso-nance peaks. The ﬁrst brain spectroscopy results were

obtainedbyBeharandcolleaguesatYaleUniversityin1983

[7].Firstmedicalapplicationsaremadeonthebodyﬂuidsand

secretions.MRS,perfusionMRIandDWIsequencesarealsoin

useforthispurpose.Inthisstudy,weusedconventionalMRI,

DWI and MRS techniques in 28 patients with glioma and

metastatictumorsandwecomparedtheresultsfortheeffects

of dexamethasone treatment on tumor and peritumoral

edema.

Metastasescausesigniﬁcantperitumoraledemalike

high-grade glial tumors [8]. Solitary metastasis and primary

characteristicsof high-gradeglialtumors arenonspeciﬁcin

conventionalMRIanditisnotalwayspossibletodifferentiate

these tumors. Both of tumors show variable degrees of

contrastenhancementandformsperitumoraledema.T2ﬂair,

DWI,MRSandMRIimageswereusedinourstudy.

Thecausesofperitumoraledemaarenotwellunderstood,

butitisassumedthatedemaissecondarytoexcessﬂuidbuild

upintheextravascularspacesurroundingthetumor.Brainis

unabletoremovethisﬂuidduetothedisruptedblood-brain

Table4–ThechangesofADCvaluesintumoralmassforeachgroupafterthedexamethasonetreatment.

Group n MeanADCvalue Standarddeviation Minimum Maximum p*

Highgradeglialtumor

Beforethetreatment 11 608.6 259.26492 361 986 0.047

Afterthetreatment 11 564.8 313.07395 104 922

Lowgradeglialtumor

Afterthetreatment 10 703 141.55211 711 966

Metastasis

* _Friedman_test.

Fig.5–47-Years-oldmalepatientwithleftparietalmetastatictumor.(A)PeritumoraledemaareainT2axialslicebeforethe

steroidtreatment.(B)TheperitumoraledemaareaisincreasedinT2axialsliceafterthesteroidtreatment.(C)ThesagittalT2

sliceofthepatientshowstheperitumoraledemaafterthesteroidtreatment.

Table3–Thechangesofmeanperitumoraledemavolumeforeachgroupafterthedexamethasonetreatment.

Group n Meanperitumoraledemavolume(mm3₎ _Standard_deviation _Minimum _Maximum _p*

Highgradeglialtumor

Beforethetreatment 11 3014.6173 823.36699 1771.73 4945.34 0.76 Afterthetreatment 11 2959.5561 885.86281 1896.32 4890.52

Lowgradeglialtumor

Beforethetreatment 10 635.554 680.77382 41.73 2014.08 0.73 Afterthetreatment 10 503.277 498.092 35.83 1678.7

Metastasis

Beforethetreatment 7 1518.2057 932.63096 174.87 3006.93 0.7 Afterthetreatment 7 1565.7943 1054.80822 156.72 3062.02 * _Friedman_test.

(7)

barrier [9]. Dexamethasoneis widely used for the medical

treatmentofperitumoraledemainthebraintumorsformany

years[13,14].Accordingtothecurrentknowledge,

dexameth-asone treatmentreduces peritumoraledema and improves

neurologicalﬁndings.Manystudieshavebeenperformedon

thisissue,andsimilarresultswerereported[15,16].However,

noneofthesestudiesfocusedonperitumoraledemavolume

and none of them used spectroscopic examinations to

investigate the correlation between the dexamethasone

treatment and clinical condition of the patients. In 1982,

Hatam et al. [17]followed three cases withserial headCT

scans.Fanetal.showedthatdexamethasoneinhibitsglioma

cellgrowth.Inaddition,ithasneuroprotectiveeffectsinbrain

andreducestumor-inducedangiogenesis[18].Andersenetal.

[19], in a series of 23 cases, investigated the effect of

dexamethasone on peritumoral edema in 1994. Andersen

got MRI on the ﬁrst, 3rd, and 7th days of dexamethasone

treatmentandfoundthatperitumoraledemadecreased4.6%

at1stdayand13.5%at7thdaysoftreatment[19].Gasparetal.

[15]publishedanarticlein2000andclaimedthattheusageof

dexamethasone 4–8mg/day reduce peritumoral edema of

metastases radiologically and bring about a signiﬁcant

improvementinpatient'sclinicalcondition.In2006,Sofﬁetti

et al.[20] revealed that dexamethasone treatment reduces

cerebraledemasigniﬁcantlyandcauseclinicalimprovement

in75%ofpatientswithin24–72h.Thecommonpointofthese

studies is the use of dexamethasone reduces peritumoral

edema and intracranial pressure temporarilyandimproves

the patient's clinical condition [19,20]. In our study, we

examined 28 cases of glioma and metastatic tumor and

compared the volume of peritumoral edema using MRI.

Although there is a slight reduction in the volume of the

Table5–ThechangesofADCvaluesinperitumoraledemaforeachgroupafterthedexamethasonetreatment.

Highgradeglialtumor

Lowgradeglialtumor

Metastasis

Beforethetreatment 7 386 59.37295 296 441 0.025

* _Friedman_test.

Table6–ThechangesofADCvaluesinnormalbraintissueforeachgroupafterthedexamethasonetreatment.

Highgradeglialtumor

Lowgradeglialtumor

Metastasis

* _Friedman_test.

Table7–ThechangesinCho/Crpeakareasforhighandlowgradeglialtumorsafterthedexamethasonetreatment.

Group n MeanCh/Crpeakarea(mm2₎ _Standard_deviation _Minimum _Maximum _p*

Highgradeglialtumor

Beforethetreatment 11 170 21.439 8.9 280 0.59

Afterthetreatment 167 19.97 8.7 267

Lowgradeglialtumor

Beforethetreatment 10 126 19.26 7 146 0.62

Afterthetreatment 125 18.22 7.1 148

(8)

post-treatment peritumoral edema, this reduction is not

statisticallysigniﬁcant.Dexamethasonedoesnotsigniﬁcantly

reduce the overall volume of peritumoral edema. So, our

resultsaredifferentfromthoseoftheliterature.

Each of 3 groups was examined separately in order to

evaluatehowdexamethasoneeffectgroupsindividually.The

changeofperitumoraledemavolumeinallof3groupsafter

thedexamethasonetreatmentwasnotstatisticallysigniﬁcant

(p>0.05forallgroups).Basedonthedataof28cases,wecan

suggestthatthedexamethasonetreatmentdoesnotmodify

theperitumoraledemasigniﬁcantly.

Althoughtheuseofdexamethasonedoesnotreducethe

volumeofperitumoraledema,interestinglydexamethasone

treatmentcausedneurologicalimprovementsin9(52.9%)of17

patientswithneurologicalﬁndings.Althoughourstudyvaries

fromcurrentliteratureintermsofdexamethasone'seffecton

peritumoraledema,itspositiveimpactonclinicalsituationis

similar to those. This signiﬁcant improvement in clinical

conditionmaynotbesecondarytodecreaseat peritumoral

edema.Ourstudyisoneofthelargeststudiesonthisissueasa

radiological study, but more comprehensive studies with

largerseriesareneeded.

DWIwasﬁrstusedinthediagnosisofcerebralischemia,

butitmaybeusedforthediagnosisoftraumaticbraininjury,

demyelinating diseases, and determination of the tumoral

cellularity [6]. ADC values may be used for differential

diagnosis of intracranial tumors and differentiation of

peritumoraledema fromtumorarea[4].Thereareveryfew

clinical studies on the use of DWI for the evaluation of

dexamethasonetreatmentinbraintumors.Sinha etal.[10]

publishedapaperonthechangesofADCvalueofperitumoral

edema after dexamethasone treatment in15 patients with

brain tumor. In this study, 7 patients had glioblastoma, 4

patientshadmetastasisand4patientshadmeningioma.They

revealedsigniﬁcantreductionsinADCvaluesofallpatients

withdexamethasonetreatmentafter 48–72h.They

empha-sized that dexamethasone reduces the extracellular water

movementsofperitumoralbrainedemaandstrengthensthe

densityofwater[10].Luetal.[21]comparedADCchangesafter

dexamethasonetreatmentin12patientswithgliomaand12

patientswithbrainmetastasesbuttheydidnotﬁnd

statisti-callysigniﬁcantresultsontheeffectofdexamethasone.Bastin

et al. [22] performed the same study but they showed

signiﬁcant decreaseinADC valuesofjust onlyonecaseof

glioblastomapatient.Inourstudy,wemeasuredADCvalues

beforeand48hafterdexamethasonetreatmentof11patients

withhighgradeglioma,10patientswithlowgradegliomaand

7patientswithbrainmetastases.ADCvaluesoftumoralmass,

peritumoral edema, and normalcerebral tissuebefore and

after dexamethasone treatment wereevaluated separately.

Statistically signiﬁcant changes in mean ADC values of

tumoralmassweredetectedin3groupsafterdexamethasone

treatment (p<0.05 for each group). These changes in

peritumoral edema were also statistically signiﬁcant in 3

groups after dexamethasone treatment (p<0.05 for each

group). The changes in normal brain tissue were also

statisticallysigniﬁcantafterthetreatment(p<0.05foreach

group).Basedonourresults,ADCvaluesinhighgrade glial

tumorsdecreasewithdexamethasonetreatment.Thismaybe

due toriseof internal water movementsof low grade glial

tumorsandbrainmetastaseswithdexamethasonetreatment.

Clinicalimprovementinpatientsprobablymaybesecondary

to the movement of water. This study is still the most

comprehensive and detailed assessment and contains

dis-crepancieswithearlierpublications.Therearealotofdifferent

resultsinthecurrentliteraturereview.DWIstudiesandADC

measurements of brain tumors after dexamethasone

treat-mentarestillamysterioussubjectandwebelievethatthereis

a need for investigation and evaluation in a larger tumor

groups.

There was a decrease in NAA peak at MRS, signiﬁcant

increaseinCholevelsandamoderatedecreaseofCrlevels

(sometimesnotchange)inglialtumors.NAAdecreaseandCho

increasearemoreevidentinhighgradeglialtumorsthanlow

gradeglialtumors[23].In1994,Kamadaetal.[23]andin1996

Preul et al. [24] analyzed the values of NAA, Cho, Cr, etc.

metabolites and published their results. They found a

signiﬁcantCho peak,NAAand Crreductionat glialtumors

intwostudies[23,24].In1997,Chumasetal.[9]gave12mg/day

dexamethasoneto9patientswithbraintumorsfortheirstudy

Fig.6–39-Years-oldfemalepatientwithlowgradeglioma(grade2astrocytoma).(A)Pre-treatmentMRSexaminationdidnot

showsignificantdifferencebetweenNAA,cholineandcreatinepeaksforbeforeandaftertreatment.(B)Thesamepatient's

(9)

andtheyanalyzedpre-treatmentexchangevalueofcerebral

metaboliteslikeNAA,ChoandCr.Theyobservedno

signiﬁ-cantincreasesordecreasesofcerebralmetabolitesdepending

ondexamethasonetreatment[9].Inourstudy,weexamined

changes in MRS ﬁndings related to the treatment with

dexamethasone in 21 patients with glioma. Patients with

brainmetastaseswerenotincludedintheassessmentwith

MRS.WeexaminedtheChoandCrvaluesintumormassand

peritumoralarea andtheir relativeproportions with

multi-voxeltwo-dimensionalimaging beforeand 48hafter

dexa-methasone treatment. The mean Cho/Cr peak area was

decreased after dexamethasone treatment in high grade

glioma group but this is not statistically signiﬁcant. There

wasalsodecreaseinmeanCho/Crpeakareaafter

dexameth-asonetreatmentinlowgradegliomagroupandthisisalsonot

statisticallysigniﬁcant.OurresultswithMRSaresimilarwith

thoseofthepreviousstudies.

Thereare2limitationsofthisstudy.Firstisthelownumber

ofpatientforapreciseoutcomefortheeffectofsteroidson

braintumors.Secondisthe limitednumberof tumortypes

wasincludedinthisstudy.Onlyglialtumorsandmetastases

wereanalyzedfortheeffectsofsteroids.Becausethesetumors

aremostlytreatedinneurosurgeryclinic,weselectedthese

typesoftumorsandperformedMRIwithdifferentsequences

inordertoreachamostaccurateoutcome.

5. Conclusion

Steroidtreatmentimprovesclinicalsignsofglialtumorsand

brain metastases. But this effect is not secondary to the

reductionofswellingaroundthetumororchangesincerebral

metabolites.Itisprobablyduetowatermovementsintumoral

mass, peritumoral edema or normal brain parenchyma.

However,moredetailedstudieswithlargerseriesshouldbe

doneformoreaccurateknowledgeonthistopic.

Conﬂict

of

interest

Nonedeclared.

Acknowledgement

and

ﬁnancial

support

Nonedeclared.

r

e

f

e

r

e

n

c

e

s

[1] StevensonKL.Pediatricbraintumors.JNeurosci2004;1: 10–20.

[2] IzciY,GurkanlarD,TimurkaynakE.Multicentricgliomas: stillremainsacontroversialissue.TurkNeurosurg2005;15 (2):71–5.

[3] IzciY.Biomarkersforbraingliomas. In:BarhD,CarpiA, VermaM,GunduzM,editors.Cancerbiomarkers:minimal andnoninvasiveearlydiagnosisandprognosis.BocaRaton, USA:CRCPress,Taylor&FrancisGroup;2014.p.199–226.

[4] IgnjatovićJ,StojanovD,ZivkovićV,LjubisavljevićS, StojanovićN,StefanovićI,etal.Apparentdiffusion coefﬁcientintheevaluationofcerebralgliomas malignancy.VojnosanitPregl2015;72(10):870–5.

[5] SugaharaT,KorogiY,KochiM,UshioY,TakahashiM. Perfusion-sensitiveMRimagingofgliomas:comparison betweengradient-echoandspin-echoecho-planarimaging techniques.AJNRAmJNeuroradiol2001;22:1306–15.

[6] GelalF,CallıC,KitisÖ,YüntenN.Diffusion-weighted magneticresonanceimaging.JNeurolSci2001;18(2):25.

[7] BarkerPB,BreiterSN,SoherBJ,ChathamJC,ForderJR, SamphilipoMA,etal.Quantitativeprotonspectroscopyof caninebrain:invivoandinvitrocorrelations.MagnReson Med1994;32:157–63.

[8] HempenC,WeissE,HessCF.Dexamethasonetreatmentin patientswithbrainmetastasesandprimarybraintumors: dothebeneﬁtsoutweightheside-effects?SupportCare Cancer2002;10:322–8.

[9] ChumasP,CondonB,Oluoch-OlunyaD,GrifﬁthsS,Hadley D,TeasdaleG.Earlychangesinperitumorousoedemaand contralateralwhitematterafterdexamethasone:astudy usingprotonmagneticresonancespectroscopy.JNeurol NeurosurgPsychiatry1997;62:590–5.

[10] SinhaS,BastinME,WardlawJM,ArmitagePA,WhittleIR. Effectsofdexamethasoneonperitumouraloedematous brain:aDT-MRIstudy.JNeurolNeurosurgPsychiatry 2004;75:1632–5.

[11] GreenbergMS.Handbookofneurosurgery.5thed.New York:Thieme;2001.p.463–9.

[12] ButowskiNA,ChangSM.Glialtumors:thecurrentstateof scientiﬁcknowledge.Clinicalneurosurgery.Philadelphia: LippincottWilliams&Wilkins;2006.p.106–13.

[13] PitterKL,TamagnoI,AlikhanyanK,Hosni-AhmedA, PattwellSS,DonnolaS,etal.Corticosteroidscompromise survivalinglioblastoma.Brain2016;139(Pt5):1458–71.

[14] IzciY,AkayKM,GurkanlarD,DeveciMS.Radiation-induced glioblastomamultiformefollowingsurgeryfor

medulloblastomainachildwithneuroﬁbromatosis-1:case report.TurkNeurosurg2005;15(1):36–9.

[15] GasparLE,GutinPH,RogersL,SchneiderJF,LarsonD, BloomerWD,etal.Pre-irradiationevaluationand managementofbrainmetastases.AmericanCollegeof Radiology.ACRappropriatenesscriteria.Radiology2000;215 (Suppl.):1105–10.

[16] WolfsonAH,SnodgrassSM,SchwadeJG,MarkoeAM,Landy H,FeunLG,etal.Theroleofsteroidsinthemanagementof metastaticcarcinomatothebrain.Apilotprospectivetrial. AmJClinOncol1994;17:234–8.

[17] HatamA,YuZY,BergstromM,BergrenBM,GreitzT.Effect ofdexamethasonetreatmentonperitumoralbrainedema: evaluationbycomputedtomography.JComputAssist Tomogr1982;6:586–92.

[18] FanZ,SehmT,RauhM,BuchfelderM,EyupogluIY, SavaskanNE.Dexamethasonealleviatestumor-associated braindamageandangiogenesis.PLOSONE2014;9(4):e93264.

[19] AndersenC,AstrupJ,GyldenstedC.Quantitationof peritumouraloedemaandtheeffectofsteroidsusingNM relaxationtimeimagingandblood–brainbarrieranalysis. ActaNeurochirSuppl(Wein)1994;60:413–5.

[20] SofﬁettiR,CornuP,DelattreJY,GrantR,GrausF,GrisoldW, etal.EFNSGuidelinesondiagnosisandtreatmentofbrain metastases:reportofanEFNSTaskForce.EurJNeurol 2006;13:674–81.

[21] LuS,AhnD,JohnsonG,ChaS.Peritumoraldiffusiontensor imagingofhighgradegliomasandmetastaticbrain tumors.AJNRAmJNeuroradiol2003;24:937–41.

[22] BastinME,DelgadoM,WhittleIR,CannonJ,WardlawJM. Theuseofdiffusiontensorimaginginquantifyingthe

(10)

effectofdexamethasoneonbraintumours.Neuroreport 1999;10:1385–91.

[23] KamadaK,HoukinK,IwasakiY,AbeH,KashiwabaT. In-vivoprotonmagneticresonancespectroscopyformetabolic changesofhumanbrainedema.NeurolMedChir(Tokyo) 1994;34:676–81.

[24] PreulMC,CaramanosZ,CollinsDL,VillemureJ-G,Leblanc R,OlivierA,etal.Accurate,noninvasivediagnosisof humanbraintumorsbyusingprotonmagneticresonance spectroscopy.NatMed1996;2:323–5.