Effectiveness of percutaneous vertebroplasty in cases of vertebral metastases

(1)

ORIGINAL

ARTICLE

/Oncology

Effectiveness

of

percutaneous

vertebroplasty

in

cases

of

vertebral

metastases

O.F.

Nas

a,∗

,

M.F.

Inecikli

b

,

E. Kacar

c

,

R. Buyukkaya

d

,

G. Ozkaya

e

,

O. Aydın

f

,

C. Erdogan

a

,

B. Hakyemez

a

a_Department_of_Radiology,_Faculty_of_Medicine,_Uludag_University,_Bursa,_Turkey

b_Department_of_Radiology,_Faculty_of_Medicine,_Recep_Tayyip_Erdogan_University,_Rize,_Turkey c_Department_of_Radiology,_Faculty_of_Medicine,_Afyon_Kocatepe_University,_{Afyonkarahisar,} Turkey

d_Department_of_Radiology,_Faculty_of_Medicine,_Duzce_University,_Duzce,_Turkey e_Department_of_{Biostatistics,}_Faculty_of_Medicine,_Uludag_University,_Bursa,_Turkey f_Radiology_Clinic,_Ali_Osman_Sonmez_Oncology_Hospital,_Bursa,_Turkey

KEYWORDS

Percutaneous vertebroplasty(PV); Vertebralmetastases; Visualanaloguescale (VAS);

Polymethylmethacry-late(PMMA)

Abstract

Purpose andobjectives:To assess theeffectiveness ofpercutaneousvertebroplasty (PV)in patientswithvertebralcollapseduetometastases.

Materialsandmethods:PVproceduresperformedon95vertebrasin52patientswithprimary malignancywereretrospectivelyevaluated.Vertebralmetastases,primarymalignanciesofthe patients,pain beforeandafter PVonavisual analoguescale(VAS),amountof polymethyl-methacrylate(PMMA)cementappliedtothevertebralbodyduringPV,PMMAcementleakage andvertebralapproacheswereevaluated.

Results:VASscores of43patients (intotal 79vertebras)were evaluated.Median VASscores of patients declined from 8(4—10) before PVto 3 (0—7) withinone day after the proce-dure,to2(0—9)oneweekaftertheprocedureandeventuallyto2(0—9)3monthsafterthe procedure(p<0.001).PMMAamountappliedtothevertebralbodyduringPVvariedbetween 1.5—9mL(average±SD4.91±1.61).Therewasnosigniﬁcantstatisticalcorrelationbetween PMMAamountsandVASscoreswithinonedayafter,1weekafter and3monthsafter thePV procedure(p>0.05).

∗_{Corresponding}_author.

E-mailaddresses:[email protected](O.F.Nas),[email protected](M.F.Inecikli),[email protected](E.Kacar), [email protected](R.Buyukkaya),[email protected](G.Ozkaya),[email protected](O.Aydın),[email protected] (C.Erdogan),[email protected](B.Hakyemez).

http://dx.doi.org/10.1016/j.diii.2015.05.001

(2)

Conclusion:PVisasimple,effective,reliable,easytoperformandminimallyinvasiveprocedure inpatientswithpainfulvertebralmetastases.

Metastatic disease of a primarymalignancy is most com-monlyseenintheskeletalsystemafterlungsandliver,and vertebrasarethemostcommonlyinvolvedbones[1,2].Pain isanimportantsymptominpatientswithspinalmetastasis. Progressionof spinalmetastasis can resultin compression ofthespinalcolumncausedbyfractureofvertebralbodies

[3].

Percutaneous vertebroplasty (PV) is a minimally inva-siveprocedure withpolymethylmethacrylate(PMMA) bone cementinjection intothe vertebralbody [4]. Itis one of the modalities accepted for treating vertebral fractures caused by osteoporosis, malignancies and trauma [5]. PV wasﬁrstdescribedin1987 byGalibertetal.intreatment of a painful vertebralhemangioma [6]. Two main indica-tionsofPVinspinalmetastasistreatmentareanalgesiaand vertebralcolumnstabilization[7].The superiorityofPVin malignantspinalinvolvementisthatitislessinvasivethan opensurgeryandfasterinrelievingpainthanradiotherapy andother conventionaltreatment methods[5].PVeffects occurquicklyanditisasupplementaryprocedureto radio-therapyorchemotherapyinpatientswithmalignantspinal involvement[8].Balloonkyphoplasty (BK),radiofrequency (RF)kyphoplastyandplasma-mediatedradiofrequency abla-tion(coblation)canbeusedtotreatvertebralfracturesin additiontoPV. Bonemetastases canalso betreated with percutaneousablation[9].

Fromour study,outcomesof PVprocedures performed in our department on patients with vertebral metastasis andknownprimarymalignanciesoverasix-yearperiodare presented.Thepurposeofourstudywastoassessthe effec-tivenessof PV in patients with vertebral collapse due to metastases.

Materials

and

methods

Protocol

PVproceduresperformedon95vertebrasin52patientswith primarymalignancybetweenApril2008andApril2014were retrospectivelyevaluated.Theonlyindicationof vertebro-plastywasseverebackpain,andthispaingenerallylimited bodymovementsof patientsand wasunresponsiveto dif-ferentpainkillers. Mostof our patients wereusing level3 (moderatetoseverepain)painkillers.Magneticresonance (MR) imaging was performed on all patients before the procedure,in ordertoassess thelocalizationof vertebral metastasis and theextent of tumoral inﬁltration intothe spinalcanalandparavertebraltissue.Conventionalsagittal

T1-weighted, T2-weightedandSTIRimageswereacquired ona3T(AchievaTX,Philips,Best,Netherlands)ora1.5T scanner (Magnetom Vision plus, Siemens, Erlangen, Ger-many)usingaspinecoil.Sagittalpost-contrastT1-weighted imageswereacquiredafteradministrationof0.1mmol/kg MRcontrastmediawhennecessary.TheMRsequence param-etersusedareindicatedinTable1.Metastaticinvolvement of the vertebras was detected clinically and radiologi-callyinpatientswithknownprimarymalignancies.Twelve of 52patients were treated with chemotherapy, 2 with radiotherapy and 38 with both therapies before or after theprocedure.Biopsieswereperformedonvertebraswith suspicious metastatic ﬁndings. The extent of metastatic involvement in vertebral bodies was assessed using the semi-quantitativevisualassessmentindexshowingvertebral deformity developed by Genant etal. [10]. Inthis index, heightlossintheanterior,middleand/orposteriorcolumns of a vertebralbody is deﬁnedas, grade 0:normal, grade 1:20—25%mild,grade2:25—40%moderate,grade3:>40% severe.Ourstudywasapprovedbytheethicalcommittee.

Operative

technique

PVwasperformedwithsterileconditions,under sedoanal-gesia(midazolam0.03mg/kgi.v.and/orfentanyl1␮/kgi.v. and/or ketamine 1mg/kg i.v. or propophol3—5mg/kg i.v. and/or pethidine1mg/kgi.m.),ina biplane,flat-paneled angiography unit (AXIOM Artis FD Biplane Angiosuite, SiemensMedicalSolutions,Erlangen,Germany).Ampicillin 1000mg-sulbactam 500mg combination was administered i.v. before the procedure for antibioprophylaxis. Patients laiddownontheangiographytableinproneposition.During theprocedure,11,13or14gaugesingleuseonlybonebiopsy needles were used. Biopsy needles were placed into the vertebralbodythroughthevertebralpediclewiththe guid-anceofanteroposteriorand/orlateralfluoroscopicimaging. Left transpedicular, right transpedicular and bipedicular approacheswereusedtoreachthevertebralbodies. Biop-sieswereperformedonvertebraswithsuspiciousmetastatic findings. PMMA (Cemento Fixx, Optimed, San Possidonio, Italy)bonecementwasappliedtothevertebralbodyunder fluoroscopyinaslowandcontrolledwaymanuallyorusing aninjectiongun.Thecementwasappliedpreferentiallyto thelyticzoneofthevertebra.Norupture intheposterior wallofthevertebralbodywasobserved.PVwasperformed on2,3oreven4or5vertebrallevelsinthesamesession. Patientswerefollowedintheobservationroomfor3hours aftertheprocedureandconsequentlydischargedwithinthe sameday.

(3)

Table1 MRsequenceparameters.

Sequences MR(T) TR(ms) TE(ms) TI(ms) Matrix NSA Slice thickness (mm) Slicegap (mm) FOV Time(s) Sagittal T2-weighted 3 3500 110 — 200×286 2 4 0.4 300×160 190 1.5 4200 90 — 192×512 2 3 0.3 340×200 150 Sagittal T1-weighted 3 460 8 — 200×300 2 4 0.4 300×160 110 1.5 550 15 — 192×256 2 3 0.3 340×200 180 Sagittal STIR 3 3000 80 210 124×169 2 4 0.4 300×160 160 1.5 4000 90 160 192×512 3 3 0.3 340×200 210

STIR:shorttauinversionrecovery;MR:magneticresonance;TR:timetorepetition;TE:timetoecho;TI:inversiontime;NSA:number ofsignalaverages;FOV:ﬁeldofview;T:Tesla;mm:millimeter;ms:millisecond;s:second.

Pain

assessment

Visualanaloguescale(VAS)wasusedtoassesspatients’pain scores the day before, within one day after, 1week and 3months after the procedure. VAS involves the standard pain scale from 0 to 10 (0=no pain, 10=intolerable, the most severe painever feltin a patient’s life) in orderto determine thelevel ofpainobjectively.VASscores of the patientswereassessedbytalkingfacetofaceorbyphone callsbefore,withinonedayafter,1weekand3monthsafter theprocedure.

Statistical

analysis

AllstatisticalanalyseswereperformedwiththeSPSS22.01 statisticalpackageprogram.Descriptivevaluesofvariables are expressed as means±standard deviation or medians (minimum-maximum).Shapiro-WilkTestwasusedto deter-minethenormaldistributionofdata.WilcoxonSignedRank Testwasusedforgroupcomparisons.Relationshipsbetween variables were reviewed using Spearman’s Rank Correla-tionCoefficient.Thelevelsofsignificanceweredefinedas ␣=0.05.

Results

In total, 52patients (24 men [46.2%] and 28 women [53.8%],agerange:21—86years[mean63.98±12.97])were enrolled.Theprimarytumorlocationsindescendingorder of frequency were the breast (n=14/52; [26.9%]), lung (n=9/52; [17.3%]), prostate (n=7/52; [13.5%]), kidney (n=5/52; [9.6%]), stomach (n=5/52; [9.6%]) and others (n=12/52; [23.1%]).Others weremalignancies likecolon, lymphoma, pancreas and larynx. Biopsy was performed during the procedure on15 (n=15/95; [15.8%]) vertebras withcertainprimarymalignancybutsuspiciousmetastases. Metastatic carcinomawas detected in6 (n=6/95; [6.3%]) andnon-specificfindingssuchasbloodelements,fibrinand trabecularparticlesweredetectedin9(n=9/95;[9.5%])of thesevertebras.

PV was performed on 95 vertebras (54lumbar [56.8%] and41thoracic[43.2%]).PVwasmostfrequentlyperformed at theL1 level(n=17/95; [17.9%]) for lumbar andat the T12level(n=13/95;[13.7%])forthoracicvertebras.A sin-glevertebrawasinvolved in 25patients (n=25/52;[48%]) andmorethanone vertebraswereinvolved in 27patients (n=27/52; [52%]) (in 15patients: 2, in 9patients: 3, in 2patients: 4,and in 1patient:5). According tothe semi-quantitativevisual assessmentindexdevelopedby Genant et al. [8], height losses of vertebras were grade0: 7 (n=7/95;[7.4%]),grade1:30(n=30/95;[31.6%]),grade2: 31(n=31/95;[32.6%])andgrade3:27(n=27/95;[28.4%]). Left transpedicular (n=63/95; [66.3%]), right transpedic-ular(n=7/95;[7.4%]) andbipedicular (n=25/95; [26.3%]) approacheswereusedforPVprocedures.

Theefﬁcacyofpercutaneousvertebroplastywasassessed byconsideringVASscoresbeforeandafterPV,PMMAamount appliedtothevertebralbodyandPMMAleakage.

ThePV procedurewasperformed on95vertebrasin52 patients.Ninepatientscouldnotbereachedbyphone.VAS scoresof79vertebrasin43patientswereassessed.Median VAS scores of patients declined from 8 (4—10) before PV to3 (0—7) within one day after, to2 (0—9) 1week after and eventually to 2 (0—9) 3months after the procedure (p<0.001)(Fig.1).Therewasasigniﬁcantstatistical differ-encebetweenaverageVASscoreswithinonedayafter,and 1weekaftertheprocedure(p<0.001);withinonedayafter and 3months after the procedure (p<0.001); and 1week afterand3monthsaftertheprocedure(p=0.002).

PMMA amount applied to the vertebral body varied

between1.5and9mL(average±SD4.91±1.61).Therewas nosigniﬁcantstatisticalcorrelationbetweenPMMAamounts and VAS scores within one day after, 1week after and 3monthsaftertheprocedure(p>0.05).

Nocomplicationwasobservedin52vertebras(n=52/95; [54.7%]).PMMAleakagewaspresentinatotalof43 verte-bras(n=43/95;[45.3%])(intradiscalleakagein20vertebras (n=20/95;[21.1%]),leakageintotheepiduralor paraverte-bralveinsin20 vertebras(n=20/95;[21.1%])andleakage intoboth disc and the epidural or paravertebral veins in 3 vertebras (n=3/95; [3.1%]). No neurological deﬁcits or

(4)

Figure1. VASscoresbefore,withinonedayafter,1weekafter and3monthsafterthePVprocedure(medianmin—max).

clinical symptoms wereobserved as a resultof this leak-age.

Table 2 shows the detailed analysis of 52patients and

their primary malignancies. Case examples are shown in

Figs.2and3.

Discussion

This study demonstrates that PV decreases the pain of patientswithvertebralmetastaseswhohaveexcessivepain and/orsurgeryhasgreatrisks.Vertebralmetastasisisoften asymptomatic,butitcanbecomesymptomaticdueto ver-tebraldestructioncausedbypathologicalfracturesresulting indevelopmentofspinalinstabilityandneuralcompression. It has been proven that chemotherapy, radiotherapy and hormonaltreatmentsreduceosteolyticdestructionand neu-rologicaldamage.Butthesetreatmentmodalitiesmaynot helpwithspinalstability,cordcompressionandpain reduc-tion.Surgerycanhelpwithspinalstability,cordcompression and painreduction but it presents greatrisks for critical patients[11].

PVisaminimallyinvasiveprocedurewherePMMAbone cementisinjectedintothevertebralbodytotreatthepain duetoa vertebralcompression fracture.Cementprovides structuralstabilization and apain reductioneffectin the vertebralbody[12].ThemainindicationforPVinmetastatic patients is tocontrol local pain [13].It can also be used tostabilizeboneshavinglyticmetastasesandhighfracture risk[14].Weilletal.[15]succeededinreducingpainin24 outof33proceduresinspinalmetastasispatientsusingPV. Mikami et al. [13] state that the preoperative mean VAS scorewasreduced from7.3 topostoperative 1.9 withPV performedon141metastaticvertebrasof69patients.Tseng etal.[4]showedthatthepreoperativemeanVASscoreof8.1 decreasedto3.8onedayafterandto2.8sixmonthsafter PV.Alvarezetal.[11]reportedthatthe preoperativeVAS scoreof9.1decreasedto3.2immediatelyafterandto2.8 threemonthsafterPV.Barragán-Camposetal.observeda

Table2 Assessmentof52patientswithprimary malig-nancywhohadthePVprocedure.

Sex(men/women) 24/28

Averageage 63.98±12.97(21—86)

MedianVASscorebeforePV 8(4—10) MedianVASscorewithinone

dayafterPV

3(0—7)

MedianVASscore1week afterPV

2(0—9)

MedianVASscore3months afterPV 2(0—9) Primarymalignancy 52 Breast 14/52 Lungs 9/52 Prostate 7/52 Renal 5/52 Stomach 5/52 Others 12/52 Localization 95 Thoracic 41/95 T6 2/95 T7 5/95 T8 4/95 T9 2/95 T10 8/95 T11 7/95 T12 13/95 Lumbar 54/95 L1 17/95 L2 13/95 L3 8/95 L4 8/95 L5 8/95 PVapproaches 95 Lefttranspedicular 63/95 Righttranspedicular 7/95 Bipedicular 25/95

Vertebrallossofheight 95

Grade0 7/95 Grade1 30/95 Grade2 31/95 Grade3 27/95 Complicationsonvertebral level 43/95

Leaksintothedisc 20/95 Leaksintotheepiduralor

paravertebralvein

20/95

Leaksintothediscand epiduralorparavertebral vein

(5)

Figure2. HypointensemetastaticinvolvementonsagittalT1-weightedimageonL3vertebralbody(largearrow)(a)andhyperintense metastaticinvolvementonSTIRsagittalimage(largearrow)(b)ina45-year-oldwomanwithprimarybreastmalignancybeforePV.Increase inmetabolicactivityonL3vertebralbodyisshownonbonescintigraphyscan(smallarrow)(c,d).PMMAbonecementleakageintothe L3-L4discandleftparavertebralveinafterPV(e,f).

signiﬁcantdecreaseinVASscoresofpatientswithvertebral metastasesofbreastcancer[16].Inourdepartment,median VASscores of 79 metastaticvertebras of 43patients with primarymalignancywas8 before,3within oneday after, 2after1weekand2threemonthsaftertheprocedure. In ourstudy,thedecreasesinVASscoresaftertheprocedure wasconsistentwiththeliteraturedata.PVprovidesaquick reliefofpaininpatientswithvertebralmetastasis.

PMMA cement prevents the collapse of vertebra by

strengtheningthevertebralbody[7].PMMAcausesdamage tonerveendingsandcytotoxiceffectsbecauseofitsheat releasing effectduringpolymerization [7,17].Decreasein tumorprogressionor recurrencecanbeobtainedfromthe antitumoraleffectofthecement[18].Inordertoachieve vertebralstiffness after PV, an average of 3.5mL volume ofPMMAissufﬁcient[4,19].To achievevertebralstiffness

afterPV, Tsengetal.used[4]5.16±1.63mL,Chewetal.

[5] used less than 5mL, and Barragán-Campos et al. [7]

used4.7±1.55mLofPMMA.TheaveragePMMAamountused in our study to achieve sufﬁcient vertebral stiffness was 4.91±1.61mLand wassimilartowhat wasusedin other studies. In addition, there was no signiﬁcant correlation betweenPMMAamountandVASscores.

Serious complications of PV are rare [12,20,21]. Seri-ous complications have been reported in approximately 1.1—1.3%of published cases [21]. Small leakage of PMMA intothepulmonary vesselsthroughepiduralor paraverte-bralveinsis usuallyclinicallyinsigniﬁcant.However,there arereportedcasesofpulmonaryembolism[7,20,22,23].The mainreasonforcomplicationsisbelievedtobetheleakage ofPMMA.Theincidenceofsymptomaticandasymptomatic cementleakagevariesfrom2to73%[13],butMousavietal.

(6)

Figure3. HypointensemetastaticinvolvementonsagittalT1-weightedimageonL3vertebralbody(largearrow)(a)andhyperintense metastaticinvolvementonSTIRsagittalimage(largearrow)(b)ina63-year-oldwomanwithprimarycolonmalignancybeforePV.X-rays showmetastaticinvolvementinL3vertebraconﬁrmedbybonebiopsyafterPV(c,d).

foundahigherratioof87.9%.Whilethereis nosigniﬁcant correlationbetweenthereductionofpainandcement leak-ageoramount,seriouscomplicationsarerelatedtocement leakage [24]. Alvarez et al. [11] did not observe serious complications in patients with vertebral metastasis after PV, but they reported cement leakage in 12 of 27 verte-bras(44.4%).Mikamietal.[13]detectedcementleakagein 49%of69patientswithvertebralmetastasis.Complications otherthanthesearevertebraltransverseprocessorpedicle fracture,paravertebralhematoma,epiduralabscess, pneu-mothorax,cerebrospinalﬂuid leakage,seizure becauseof oversedation or respiratory arrest and death [20]. In our study, PMMAleakage was present in a total of 43 verte-bras (45.3%) (intradiscal leakage in 20 vertebras [21.1%], leakageintotheepiduralorparavertebralveinsin20 verte-bras[21.1%]andleakageintobothdiscandtheepiduralor

paravertebral veins in 3 vertebras [3.1%]).But noserious complicationswereobservedinourpatients.

BK, RF kyphoplasty, percutaneous ablation (radiofre-quency or cryotherapy) and plasma-mediated radiofre-quencyablation(coblation)canbeusedtotreatvertebral fracturesinadditiontoPV.Surgicalvertebralbody cemento-plastycanbeanalternativemethodforpalliativetreatment of patients who are not suitable for PV [25]. PV and BK arethe mostfrequently usedpercutaneous interventional methods for vertebral compression fractures. RF kypho-plasty is a technique approved in the USA and Germany in2007and2009,respectively. Withthistechnique, ultra-viscous cement activated by ex-vitro radiofrequency is appliedtothevertebralbodyinaconstantandcontrolled way. Thus, destruction of the spongy microarchitecture of the vertebral body is minimized. Röllinghoff et al.

(7)

providedfastandpermanentimprovementintreatmentof 30osteoporoticvertebralfracturesbyRFkyphoplasty[26]. Deschampes et al. successfully treated 75% of vertebral metastases smaller than 3cm withpercutaneous ablation (radiofrequency) [9]. Coblation is the process of giving radiofrequency at relatively low temperatures (40—70◦C) intothetargetedtissuetodissolvemolecularbonds,makea cavityandﬁllingthecavitywithPMMA.Coblationisan effec-tiveandreliablemethodusedinhigh-riskpatientswhohave excessivepain.Prologoetal.providedfastimprovementin 14of15patientswhohadpainfulvertebralmetastases[27]. Therearesomelimitationsofourstudy.First,itisa ret-rospective study with no control group. Second, most of patients hadno long-termfollow-up because oftheir pri-marymalignancies.Third,onfollow-up,somepatientscould only be reached by phone. Face to face communication couldbemoreeffectivethanspeakingonthephone.Fourth, VASscorescouldbeassessedatamaximumof3monthsafter theprocedurebecausesomepatientsdidnotcometothe controlexaminationandcouldnotbereachedbyphone.

Conclusion

PV is a simple, effective, reliable, easy to perform and minimally invasive procedure in patients with vertebral metastases, who have excessive pain and/or surgery has greatrisks.

Disclosure

of

interest

Theauthorsdeclarethattheyhavenoconﬂictsofinterest concerningthisarticle.

References

[1]QianZ,SunZ,YangH,GuY,ChenK,WuG.Kyphoplastyfor thetreatmentofmalignant vertebral compressionfractures causedbymetastases.JClinNeurosci2011;18:763—7. [2]MesﬁnA, Buchowski JM,GokaslanZL, Bird JE. Management

ofmetastatic cervicalspinetumors.JAmAcadOrthopSurg 2015;23:38—46.

[3]Dalbayrak S, Onen MR, Yilmaz M, Naderi S. Clinical and radiographicresultsofballoon kyphoplastyfor treatmentof vertebralbodymetastasesandmultiplemyelomas.JClin Neu-rosci2010;17:219—24.

[4]TsengYY,LoYL,ChenLH,LaiPL,YangST.Percutaneous poly-methylmethacrylatevertebroplastyinthetreatmentofpain inducedby metastatic vertebral columntumor. SurgNeurol 2008;7:78—83.

[5]ChewC,RitchieM,O’DwyerPJ,EdwardsR.Aprospectivestudy ofpercutaneousvertebroplastyinpatientswithmyelomaand spinalmetastases.ClinRadiol2011;66:1193—6.

[6]GalibertP,DeramondH,RosatP,LeGarsD.Preliminarynote onthetreatmentofvertebralangiomabypercutaneousacrylic vertebroplasty.Neurochirurgie1987;33:166—8.

[7]Barragán-CamposHM,ValléeJN,LoD,CormierE,JeanB,Rose M,et al.Percutaneousvertebroplastyfor spinalmetastases: complications.Radiology2006;238:354—62.

[8]ChewC,CraigL,EdwardsR,MossJ,O’DwyerPJ.Safetyand efﬁcacyofpercutaneousvertebroplastyinmalignancy:a sys-tematicreview.ClinRadiol2011;66:63—72.

[9]DeschampsF,FarouilG,deBaereT.Percutaneousablationof bonetumors.DiagnIntervImaging2014;95:659—63.

[10]GenantHK,WuCY,VanKuijkC,NevittMC.Vertebralfracture assessmentusingasemiquantitativetechnique.JBoneMiner Res1993;8:1137—48.

[11]AlvarezL, Pérez-HiguerasA, Qui˜nonesD,CalvoE,RossiRE. Vertebroplastyinthetreatmentofvertebraltumors: postpro-ceduraloutcomeand qualityoflife.EurVertebralColumnJ 2003;12:356—60.

[12]AnselmettiGC,MarciaS,SabaL,MutoM,BonaldiG, Carpeg-giani P, et al. Percutaneous vertebroplasty: multi-centric resultsfromEVERESTexperienceinlargecohortofpatients. EurJRadiol2012;81:4083—6.

[13]MikamiY, NumaguchiY, KobayashiN, FuwaS, Hoshikawa Y, Saida Y. Therapeuticeffectsof percutaneousvertebroplasty forvertebralmetastases.JpnJRadiol2011;29:202—6. [14]DeschampsF,deBaereT.Cementoplastyofbonemetastases.

DiagnIntervImaging2012;93:685—9.

[15]WeillA,ChirasJ,SimonJM,RoseM,Sola-MartinezT,Enkaoua E. Spinal metastases: indications for and results of per-cutaneous injection of acrylic surgical cement. Radiology 1996;199:241—7.

[16]Barragán-CamposHM,LeFaouAL,RoseM,LivartowskiA,Doz M,AstagneauP,etal.Percutaneousvertebroplastyinvertebral metastasesfrombreastcancer:interestintermsofpainrelief andqualityoflife.IntervNeuroradiol2014;20:591—602. [17]SanMillánRuízD,BurkhardtK,JeanB,MusterM,MartinJB,

BouvierJ,etal.Pathologyﬁndingswithacrylicimplants.Bone 1999;25:85—90.

[18]RoedelB,Clarenc¸onF,TouraineS,CormierE,Molet-Benhamou L, Le Jean L, et al. Has the percutaneous vertebro-plasty a role to prevent progression or local recurrence in spinal metastasesof breastcancer? J Neuroradiol 2014, http://dx.doi.org/10.1016/j.neurad.2014.02.004.

[19]FarooqN,ParkJC,PollintineP,Annesley-WilliamsDJ,Dolan P.Canvertebroplastyrestorenormalload-bearingtofractured vertebrae?VertebralColumn2005;30:1723—30.

[20]Jensen ME, McGraw JK, Cardella JF, Hirsch JA. Posi-tion statement on percutaneous vertebral augmentation: a consensus statementdeveloped by theAmerican Societyof Interventional and Therapeutic Neuroradiology, Society of InterventionalRadiology, AmericanAssociation of Neurologi-calSurgeons/CongressofNeurologicalSurgeons,andAmerican SocietyofVertebralcolumnRadiology.AJNRAmJNeuroradiol 2007;28:1439—43.

[21]Barr JD, JensenME, Hirsch JA,McGraw JK, Barr RM,Brook AL,etal.Positionstatementonpercutaneousvertebral aug-mentation:aconsensusstatementdevelopedbytheSocietyof InterventionalRadiology(SIR), AmericanAssociationof Neu-rological Surgeons (AANS) and the Congress ofNeurological Surgeons(CNS),AmericanCollegeofRadiology(ACR), Amer-ican Societyof Neuroradiology (ASNR),American Societyof Vertebral column Radiology(ASSR), Canadian Interventional Radiology Association (CIRA), and the Society of NeuroIn-terventional Surgery (SNIS). J Vasc Interv Radiol 2014;25: 171—81.

[22]PadovaniB,KasrielO, BrunnerP,Peretti-VitonP.Pulmonary embolism caused by acrylic cement: a rare complication of percutaneous vertebroplasty. AJNR Am J Neuroradiol 1999;20:375—7.

[23]Calmels V, Vallée JN, Rose M, Chiras J. Osteoblastic and mixed spinal metastases: evaluation of the analgesic efﬁ-cacyofpercutaneousvertebroplasty.AJNRAmJNeuroradiol 2007;28:570—4.

[24]MousaviP,RothS,FinkelsteinJ,CheungG,WhyneC. Volumet-ricquantiﬁcation ofcementleakage followingpercutaneous vertebroplasty in metastatic and osteoporotic vertebrae. J Neurosurg2003;99:56—9.

(8)

[25]N’dri Oka D, Tokpa A, Derou L. Surgical vertebral body cementoplastyasspinalcancermetastasismanagement.BrJ Neurosurg2014;30:1—6.

[26]Röllinghoff M, Zarghooni K, Zeh A, Wohlrab D, Delank KS. Is there a stablevertebral height restoration withthenew

radiofrequencykyphoplasty?Aclinicalandradiologicalstudy. EurJOrthopSurgTraumatol2013;23:507—13.

[27]Prologo JD, Buethe J, Mortell K, Lee E, Patel I. Cobla-tion for metastatic vertebral disease. Diagn Interv Radiol 2013;19:508—15.