ORIGINAL ARTICLEPredicting the Probability of
MeningiomaRecurrence in the Preoperative and EarlyPostoperative
Period: A MultivariateAnalysis in the Midterm Follow-UpFarukIldan,
M.D.,1TahsinErman, M.D.,1A. Iskender Gcer, M.D.,1MetinTuna,
M.D.,1Hseyin Bagdatoglu, M.D.,1Erdal C etinalp, M.D.,1and Refik
Burgut, Ph.D.2ABSTRACTWereviewedtheclinical, radiological,
surgical, andhistopathologicalfeatures of patients with meningiomas
to identify factors that can predict tumorrecurrenceafter
microscopictotal removal, toimprovepreoperativesurgicalplanning,
andtohelpdeterminetheneedforcloseradiological
observationatshorterintervalsortheneedforradiotherapyasanadjuvanttreatmentintheearlypostoperativeperiod.
Clinical data, magneticresonanceimagingstudies,angiographic data,
operative reports, and histopathological ndings were
exam-inedretrospectivelyin137patientswithameningiomatreatedmicrosurgicallyand
with no evidence of residual tumor on postoperative MR images.
Based
onunivariateanalysis,tumorsize,amushroomshape,proximitytomajorsinuses,edema,osteolysis,corticalpenetration,signalintensityonT2-weightedMRIs,pial-cortical
arterial supply, presence of a brain-tumor interface in
surgery,Simpsons criteria, and histopathological classication were
signicant predictorsfor recurrence. However, age, gender, location
of tumor, dural tail, calcication,signal
intensityonT1-weightedimages, andhistopathologicsubtypes
inthebenign group were not signicant predictors. By Cox regression
analysis the mostimportant variables
relatedtothetimetorecurrenceweremushroomshape,osteolysis, dural
tail, and proximity to major sinuses. Aggressive surgical
therapywithwiderduralremovalshouldbeconsideredinthepresenceofthepreoper-ative
predictors of a recurrence. Close radiological observation at
shorter
intervalsorradiotherapyshouldbeconsideredasadjuvanttherapyinhigh-riskpatients1Department
of Neurosurgery, 2Department of Biostatistics, C ukur-ova
University, School of Medicine, Adana, Turkey.Address for
correspondenceandreprint requests: FarukIldan,M.D., Department of
Neurosurgery, C ukurova University, Schoolof Medicine,
Balcal-Adana, 01330, Turkey (e-mail: [email protected]).Skull
Base 2007;17:157172. Copyright #2007 by ThiemeMedical Publishers,
Inc., 333SeventhAvenue, NewYork, NY10001, USA. Tel:
+1(212)5844662.Received: December 28, 2004. Accepted after
revision: February24, 2006. Published online: March 23, 2007.DOI
10.1055/s-2007-970554. ISSN 1531-5010.157based on surgical ndings
predicting recurrence related to the brain-tumorinterface,
Simpsonscriteria, andhistopathological
ndingsintheearlypost-operative period.KEYWORDS: Angiography,
magnetic resonance imaging, meningioma, tumorrecurrenceEven after
complete removal,
meningiomashavebeenestimatedtorecurin10to32%ofthecaseswithin10years.1,2Except
forthedegreeoftumor removal, other risk factors for recurrence
arenot well understood in the cases with benign men-ingioma.
Regional multicentricity has been sug-gested as a cause of
recurrences.3Surgicalcleavabilityofmeningiomasfromtheadjacentpa-renchymaltissuehasbeenreportedasasignicantprognosticfactor4andcanbepredictedfrompre-operativemagneticresonanceimaging(MRI)andangiographic
studies.5The relation between tumorrecurrenceandradiological
features onMRI andangiography indicating a brain-tumor
interface(surgical cleavability) has not yet been
studied.Wethereforeinvestigatedtherelationshipsbetweentumor
recurrence andthe followingpa-rameters: age,gender, tumorsizeand
shape,prox-imity to major sinuses, edema, bone changes,calcication,
cortical penetration, dural tail, signalintensity on T1- and
T2-weighted MRIs, pial-cortical arterial supply, brain-tumor
interface atsurgery, Simpsons criteria, and pathology.
Ourmaingoalwastoidentifyhigh-riskgroupspreop-eratively, to improve
surgical planning, and todecide whether to perform close
radiological obser-vationat shorter intervals or radiotherapy as
anadjuvant therapybyestimatingtheriskof recur-rence viasurgical and
histopathologicalndings inthe early postoperative period.CLINICAL
MATERIALS AND METHODSBetween1989and2000, 201patients
underwenttreatment for intracranial meningiomas at our
Neu-rosurgical Department.
Sufcientdataforanalysiswereavailablefrom169patients.Of169patients,14
were excluded fromthe study due to radio-graphic evidence of
residual tumors. At least1 year of follow-upwas deemednecessary
sincethe volume doubling time in rapidly growing men-ingiomas,
includinginatypical or anaplastic
andbenignforms,isfaster6thanwouldbeexpectedinthe average
meningioma68and to calculate time torecurrence correctly. Eighteen
patients were ex-cluded because their follow-up period was
lessthan1year long. Inthis group, 2patients
diedpostoperativelyand4patientsdiedfromunrelateddiseasesduringthefollow-upperiodwithnoevi-denceoftumorrecurrence.
Theremaining12pa-tientswerelosttofollow-upwithlessthan1yearwithoutrecurrence.All
together,137 caseshadnoevidence of residual tumor andsuitable data
foranalysis. Their medical charts and radiographic leswere reviewed
retrospectively for clinical, radio-graphic, operative,
andpathologicaldata. Noneofthe patients had undergone a previous
operation orhad received prior radiotherapy.MRI was performedmainly
witha 1.5Tmachine (General Electric, Milwaukee, WI,
USA).Precontrast T1-weighted MRIs were obtained with600800/20/12
(repetition time/echo time/excita-tion) and T2-weighted images with
2800/90/1.Typically, slice thickness was 5mmwitha 2.5-mminterval
between slices. Gadolinium-DTPA(0.1mmol/kg)wasusedin92cases.
T1-weightedMRIs were obtained in various planes depending onthe
location of the tumor. Tumors were
categorizedaccordingtotheirlocation: parasagittal, falx,
con-vexity, anterior fossa, middle fossa, or posteriorfossa. They
were also categorized into a near
amajorsinus(NS)groupandadistantfrommajor158 SKULL BASE/VOLUME 17,
NUMBER 3 2007sinus (DS) group. In the former group, the attach-ment
margin of the tumor was located within 1
cmofthemajorsinuses,includingsagittal,transverse,sigmoid,
andcavernous sinuses. The size of
thetumorswascategorizedaslarge(>4cm)orsmall(
