Aretrospectivelongitudinalradiographicstudyofpatientswithprogressivescoliosiswasconductedtodeterminetherelativeamountofwedgingbetweenvertebrae Thescoliosisdeformityisthreedimensionalandincludescurva ID: 215392
Download Pdf The PPT/PDF document "DiscandVertebralWedginginPatientsWithPro..." is the property of its rightful owner. Permission is granted to download and print the materials on this web site for personal, non-commercial use only, and to display it on your personal computer provided you do not modify the materials and that you retain all copyright notices contained in the materials. By downloading content from our website, you accept the terms of this agreement.
DiscandVertebralWedginginPatientsWithProgressiveScoliosisIanA.F.StokesandDavidD.AronssonDepartmentofOrthopaedicsandRehabilitation,UniversityofVermont,Burlington,Vermont,U.S.A. Aretrospectivelongitudinalradiographicstudyofpatientswithprogressivescoliosiswasconductedtodeterminetherelativeamountofwedgingbetweenvertebrae Thescoliosisdeformityisthree-dimensionalandin-cludescurvatureofthespineinthecoronalandsagittalplaneswithrotationintheaxialplane(1).Thelargestcomponentofthedeformityisthelateralcurvatureinthecoronalplane.Itresultsfromlateralwedgingofboththevertebraeanddiscs(2,3),buttheproportionsofwedgingthatoccurinthesetwoanatomicstructuresisunknown.Ithasbeensuggestedthatthedeformitybeginsinthediscs, ReceivedAugust10,2000;acceptedDecember21,2000.AddresscorrespondenceandreprintrequeststoDr.I.A.F.Stokes, curve,andreportedthattheriskofprogressionismostrelatedtoskeletalimmaturityandlargerscoliosiscurvemagnitude.Inpatientswithneuromuscularscoliosisasso-ciatedwithcerebralpalsy,progressionofthecurveislesspredictablebecauseofthevariableonsetofpuberty,de-greeofspasticity,andambulatorystatus.Milleretal.(7)retrospectivelyreviewedthescoliosiscurvepattern,length,magnitude,androtationin43patientswithcere-bralpalsywithspasticquadriplegia,followingthemuntilthecurvewas50°.EarlyonsetofscoliosisandincreasedrotationofthecurvewerepredictiveofprogressionoftheWeconductedthislongitudinalradiographicstudyofpatientswithdocumentedprogressivescoliosistodeter-minetherelativeamountofwedgingbetweenvertebraeanddiscsfortwodifferentcausesofthescoliosis(idio-pathic,andscoliosisassociatedwithcerebralpalsy).Eachpatientwasstudiedlongitudinallywithtworadiographstodeterminewhethertherelativeamountofvertebralanddiscwedgingchangedwithprogressionofthescoliosisdeformity,andwhethertherelativeamountofvertebralanddiscwedgingdifferedforcurveswiththeapexindifferentanatomicregionsofthespine.Twenty-sevenpatientswithidiopathicscoliosisand17patientswithscoliosisassociatedwithcerebralpalsywerestudiedafterwereviewedtheclinicalrecordsandradio-graphsofpatientswithprogressivescoliosisseenattwotertiaryreferralcenterseachservingpopulationsofap-proximately400,000people.Patientsradiographswereincludedinthisstudyiftheyhadradiographicallydocu-mentedprogressionofuntreatedscoliosis(increasegreaterthan5°Cobbangle)andwithradiographshavingadequatequalityformeasurementofwedgingangles.Weselectedtwofilmsforeachpatient:theearliestsuitableradiographandthelatestsubsequentfollow-upradiograph.Thelarg-estcurveofeachpatientwasstudied,andtheradiographswereallmadebeforeanysurgery.Patientswereclassifiedbytheanatomiclevelofthecurveapex;becauseoftherelativelysmallnumberofthoracolumbarcurves,theseweregroupedalongwithlumbarcurvesandwerecom-paredwiththethoraciccurves.Thedegreeofvertebraanddiscwedgingwasmeasuredfromtheposteroanteriorradiographsbydrawingalineacrossthesuperiorandinferiorendplatesofeachvertebrainthecurve.Thecoordinatesoftwopointsoneachlineweredigitized(GTCODigitizer,Rockville,MD,U.S.A.)andweresavedinacomputerfile.Customsoftwarecal-culatedtheangleofeachlinefromthehorizontal,usingthesecoordinates.Thevertebralwedgingwascalculatedbytakingthedifferenceinanglebetweentheadjacentlinesonavertebra.Similarly,theamountofdiscwedgingwascalculatedfromthedifferencebetweentheanglesofthetwolinesoneachdisc.Ineachcurve,theapicalver-tebraplustwovertebraeaboveandtwobelow(fivever-tebrae)andtheiradjacentdiscswereselectedforanalysis(Fig.1).Theapicalvertebrawasdeterminedbyacom-puterprogramthatselectedthevertebrawiththemaxi-mumlateraldeviationfromthepatientsspinalaxis(thelinejoiningthevertebralbodycentersofT1andS1).Thenthewedgeanglesforthevertebraeanddiscswereex-pressedasaproportionoftheCobbangle.Forcompari-sonsofwedgingbetweenvertebraeanddiscs,themeanofthetwoadjacentdiscwedgeangleswascomparedwiththatoftheinterveningvertebra.AnalysisofvariancewasusedtodeterminewhetherthewedgeanglesofthevertebraeandthediscsasaproportionoftheCobbangledifferedbyapexregiongroupingorby FIG.1.Discandvertebralwedgeangles.Wedgeanglesweremeasuredatapicalvertebraeanddiscs,andatoneandtwolevelsaboveandbelowtheapex.DISCANDVERTEBRALWEDGINGINSCOLIOSISJSpinalDisord,Vol.14,No.4,2001 diagnosis.Toexaminewhethertheseproportionschangedwithcurveprogression,themeanindividualdifferenceswereexaminedbythetest(nullhypothesiswasthatthemeandifferencesfromfirsttosecondradiographwerezero).Also,thechangesovertimeinthediscandvertebralwedgingasaproportionofCobbanglewereexaminedbyanalysisofvariancetodeterminewhethertheydifferedbythegroupingfactorsdiagnosisandcurveapexregion.Amongthe27patientswithidiopathicscoliosis,18hadacurvewiththoracicapexand9hadathoracolumbarorlumbarcurve.Amongthe17patientswithscoliosisasso-ciatedwithcerebralpalsy6hadacurvewithathoracicapexand11hadathoracolumbarorlumbarcurve.DetailsofthepatientsandthemeasurementsobtainedaregiveninTables1and2.Aconsistentfindinginbothgroupsofpatientswasthatthediscwedgeangle(asaproportionofCobbangle)waslessinthoraciccurvesthaninthelumbarandthoracolum-barcurves(p0.01).Themeanvertebralwedging(asaproportionofCobbangle)wasobservedtobegreaterinidiopathicthoraciccurvesthanintheidiopathiccurvesbelowT11,butnotinradiographsofpatientswithcerebralpalsy(Table3).ThemeanvertebralanddiscwedgingasaproportionofCobbangledidnotsignificantlychangebetweentheinitialandfollow-upradiographineitherdi-agnosticgroupofpatients.Also,therelativeproportionsofdiscandvertebralwedgingwithineachapexlevelgroupdidnotchangesignificantlywithcurveprogression.Thewedgingofbothvertebraeanddiscswasgreatestattheapexofthescoliosisdeformity.Thewedgingoftheapicalvertebrawasfoundtobegreaterrelativetoadjacentvertebraeinthepatientswithidiopathicscoliosisthanthepatientswithscoliosisassociatedwithcerebralpalsy,be-causetheaverageamountofwedgingoftheapicalverte-brawas1.4(idiopathicscoliosis)and1.15(cerebralpalsy)timesgreaterthanthemeanofthevertebraeimmediatelyaboveandbelowtheapex,and3.5(idiopathicscoliosis)and2.1(cerebralpalsy)timesgreaterthemeanofthevertebraetwoaboveandtwobelowtheapex.Thepatientswithcerebralpalsyhadscoliosiscurvesinvolvingmorevertebrae.Thedistributionbycurveapexwasreciprocal,inthatmostcurvesinpatientswithidiopathicscoliosiswereinthethoracicregion(18of27patients),whereasonly6of17patientswithcerebralpalsyhadthoraciccurves(Table1).Amongthepatientswithidiopathicscoliosiswhohadathoracicmajorcurve,thewedgingattheapexwasgreaterinthevertebraethaninthediscs,whereastheoppositewasgenerallyfoundattheapexofthemajorlumbarandtho-racolumbarscoliosiscurves.NodifferencewasfoundintherelativeamountofapicalvertebralanddiscalwedgingbetweenthepatientswithidiopathicscoliosisandthoseTABLE1.Detailsofpatientswithscoliosissecondarytocerebralpalsy no.ApexFirstradiographSecondradiographCobbangleVertebralwedgeangle(°)Discalwedgeangle(°)CobbangleVertebralwedgeangle(°)Discalwedgeangle(°)1T128.5350.83.611.4743.11.72T1110.9323.02.213.0470.56.03L38.0165.05.28.76010.011.04L25.380.83.413.85110.38.75T67.4444.85.113.74913.48.06L38.272.40.511.3378.93.87L210.26411.98.010.8737.913.98T128.4301.13.412.6557.14.89L21.6343.71.52.16010.01.510T1216.4171.24.619.1391.06.311T99.9200.81.916.4335.51.912T85.6254.51.312.4509.22.413T129.67814.90.211.210514.13.014L14.1563.114.27.17914.116.415T127.3366.10.811.4902.45.816T93.1160.10.98.3633.78.517T610.2302.30.518.1641.24.5Mean7.932.23.93.411.860.57.26.4SD3.418.83.93.44.118.64.54.2Allmeansareofabsolutevalues.Age,patientage;Cobbangle,Cobbangleoflargestcurve;vertebralwedgeangle,apexvertebrawedgeangle;discalwedgeangle,averageofwedgeanglesofthediscsadjacenttotheapexvertebra.Cobbangleisnegativeforleftconcavecurve;negativewedgeanglethinneronleft.I.A.F.STOKESANDD.D.ARONSSON319JSpinalDisord,Vol.14,No.4,2001 withscoliosisassociatedwithcerebralpalsy.Also,therelativeamountofwedgingofthevertebraeanddiscswasnotobservedtochangewithprogressionofthescoliosisdeformity.Therefore,theresultsofthecurrentstudydonotsupportthehypothesisofTaylor(2)thatthewedgedeformitybeginspredominatelyinthediscsandsubse-quently,withcurveprogression,thevertebraebecomewedged.Thedivisionofwedgingbetweenvertebraeanddiscsinthoracicandlumbarcurvesmayberelatedtothedifferentdiscthickness(relativetovertebralheight)inthesetwoanatomicregions.Totakeintoaccountthechangingcurvemagnitudes,wefocusedouranalysesontherelativecontributionsofdiscandvertebralwedgingtotheCobbangle.Havingdonethis,ourstudyindicatesthatcurvemagnitude,amountofcurveprogression,elapsedtime,anddiagnosiswerenotsignificantcontributorstotherelativeproportionsofdiscandvertebralwedging.Becausethediscsareflexible,themeasurementsofdiscwedgingmightchangewithpatientpositioningatradiog-raphy,andmeasuresofbothdiscandvertebralwedgingTABLE2.Detailsofpatientswithidiopathicscoliosis no.ApexFirstradiographSecondradiographCobbangleVertebralwedgeangle(°)Discwedgeangle(°)CobbangleVertebralwedgeangle(°)Discwedgeangle(°)1L211.0190.95.614.34714.07.62L211.6173.21.711.9310.76.13L211.7130.32.813.4292.55.44T1213.3191.33.514.8295.64.25L29.4134.21.014.2314.45.26L210.0130.23.312.7311.98.77T1210.6182.63.811.5292.97.78T811.9155.30.813.6335.83.59T811.7253.01.613.3388.00.810T913.1171.32.813.6252.41.811T86.2236.51.911.1416.05.812T87.0151.51.511.0410.25.513T913.0166.20.114.2332.35.514T1011.8244.52.412.2322.05.815T412.0165.21.115.1416.911.616T89.4210.74.010.8414.15.417T98.9215.12.513.76315.46.618T910.9232.92.112.75513.13.719T813.3279.25.110.8409.27.320T1013.2231.33.820.1485.54.621T810.42711.02.712.25815.45.122T910.5131.22.513.85517.02.923T812.0433.47.613.3573.37.924L23.9134.51.09.65410.88.625T811.282.30.514.4622.39.926L215.7315.76.517.44311.65.627T79.8325.65.312.1569.93.4Mean10.820.23.72.913.442.36.85.8SD2.47.32.61.92.111.44.92.4Allmeansareofabsolutevalues.Age,patientage;Cobbangle,Cobbangleoflargestcurve;vertebralwedgeangle,apexvertebrawedgeangle;discalwedgeangle,averageofwedgeanglesofthediscsadjacenttotheapexvertebra.Cobbangleisnegativeforleftconcavecurve;negativewedgeanglethinneronleft.TABLE3.Amountofapicalvertebralanddiscwedging(expressedasapercentageofthecorrespondingCobbangle),brokendownbycurveregion,andbymeasurementsmadefromtheinitialandfollow-upradiographs ScoliosisassociatedwithcerebralpalsyIdiopathicscoliosisThoraco-lumbarandlumbarThoracolumbarandlumbarInitialradiographVertebra8.42%(5.5)12.36%(12.8)18.09%(15.4)13.02%(13.8)Disc5.02%(5.5)15.61%(13.6)8.77%(9.4)17.99%(7.3)Follow-upradiographVertebra11.85%(9.6)12.93%(6.9)15.06%(8.5)13.03%(12.0)Disc10.04(4.3)11.31%(6.9)12.14%(6.0)18.73(4.9)Valuesarethemeanineachgroup(standarddeviationinparentheses).DISCANDVERTEBRALWEDGINGINSCOLIOSISJSpinalDisord,Vol.14,No.4,2001 mightbeinfluencedbytheradiographicprojectionused.Thesepotentialproblemswereprobablyminimizedbe-causeallstandingradiographsofpatientswithadolescentidiopathicscoliosisweremadeposteroanteriorly(PA)witha72-inch(1.8m)focus-to-filmdistance(FFD).Forthosepatientswithcerebralpalsywhowereradiographedsitting,thefilmswerePA,with72inchesFFD.Forsomeofthesepatients,theprocedurewasanAPprojection,72inchesFFD,withthepatientlyingsupineonthecassette,onthefloor.Ithasbeenrecognizedpreviouslythatevensmallsco-liosisdeformitiesincludevertebralwedging.Xiongetal.(8)measuredthewedgingofvertebrae(indegrees)anddiscs(measuredasthedifferenceindischeightbetweenconvexandconcavesides)ingirls13.3to19.3yearsold.Theyreportedthatwedgingofboththevertebraeanddiscswaspresentinsmallscoliosiscurves,andinterpretedthesefindingsasshowingaprimarydisturbanceofvertebralanddiscgrowth,becauseofextravertebralfactors.Inacross-sectionalstudy(9)of86skeletallyimmaturepatientswithCobbangleintherange12110°,radiographicmeasure-mentsoftheapicalvertebrawedgingshowedalinearcor-relationwiththemagnitudeoftheCobbangle.Whenthetotalwedgingofallvertebraeinacurvewascomparedwiththetotaldiscwedging(9,10),therewasagreatercontributionofdiscalthanvertebralwedginginsmallercurves,buttheratiobecamealmostequalinlargermag-nitudecurves.Inthecurrentstudy,weevaluatedonlytheapicalandtwoadjacentvertebraeaboveandbelowandthecorrespondingdiscs,soitispossiblethathadweincludedthemoreflexiblesegmentsofthescoliosiscurve,thediscsmighthavebeenfoundtocontributemoretothedeformi-ty.Perdriolleetal.(9)measured13anatomicspecimenswithameanCobbangleof91°andobservedthatwedgingwasconcentratedontheconcavehalfofthevertebralend-platesinallvertebraefromthemajorcurves,creatingacuneiform-shapedvertebralbody.Thecuneiformshapewasapparentlyacharacteristicthatdevelopedbysecond-aryremodelinginlargecurvesinadults,andourmeasuresoftheyoungerclinicalpopulationwouldhaveminimalchangesfromadultboneremodelingthatmightpredomi-natelaterinlife.Animalmodelsofscoliosisalsodemonstrateadistri-butionofthewedgedeformitybetweenvertebraeanddiscs,butthemethodofproducingthecurvatureisartifi-cial.Stillwell(11)followedaprogressivescoliosisdefor-mityforupto1yearsinimmaturemonkeysafterde-stabilizationoftheposteriorelementsofthespine.Thediscangulationappearedbeforeevidenceofbonywedg-ing,buteventuallythediscangulationaccountedforabouthalfofthescoliosisdeformity.Menteetal.(12)createdascoliosisdeformityinasegmentoftherattailbyimposingcompressiveforcesandalateralcurvature.Atthebegin-ningoftheexperiment,100%ofthedeformitywasinthediscs,butafter6weeksofgrowth,43%ofthescoliosiscurvewasinthevertebrae,whereas57%wasstillintheBecauseofdifferencesinthenaturalhistorybetweenidiopathicscoliosisandthatassociatedwithcerebralpalsy,itisdifficulttomakecomparisonsbasedonchrono-logicalage.Theageofthepatientswithidiopathicsco-liosisatthetimeoftheinitialradiographaveraged10.8years,anditaveraged13.4yearsatthetimeofthesub-sequentradiograph,whereasthepatientswithcerebralpalsyaveraged7.9yearsatthetimeoftheinitialradio-graphand11.8yearssubsequently.Therewerealsodif-ferencesbetweenthegroupsintheaveragecurvemagni-tudesinthetworadiographs(thepatientswithidiopathicscoliosishadsmallercurvemagnitudes).Thesefactorsmayhaveconfoundedcomparisonsbetweenthesegroupsinthelongitudinalaspectsofthisretrospectivestudy.Themeasurementsinthecurrentstudywerelimitedtofrontalplaneradiographs.Becausethereisalsowedginginthesagittalplaneassociatedwithlordosisandkyphosis,theplanewheremaximumvertebralwedgingoccursinaspinewithscoliosisliesinanintermediateplane,depend-ingontherelativeamountofwedginginsagittalandfrontalplanes(13).Theaxialrotationofvertebraeinsco-liosismayaltertheapparentwedgingseeninthefrontalplanebybringingthesagittalplanewedgingintoevidenceinthefrontalprojection.However,thiswasprobablyasmallfactorinthisstudybecausethesagittalwedgingisgenerallysmallinmagnituderelativetotheamountofcoronalplanewedgingbecauseofscoliosisatthecurveapex.Furthermore,theamountofvertebralrotationissmallrelativetothe90°rotationrequiredtomakethesagittalplanewedgingcompletelyvisible.ThemeasurementofvertebralwedgingwasdonebyamethodsimilartothemeasurementofCobbangle,andthereforeincludedsimilartechnicalerrors,exceptthatendvertebraselectiondidnotcontributetomeasurementerrorhere.Therefore,theerrorinwedgingvalueswouldhaveanexpected95%confidenceintervalof3.8°,basedonthestudyofMorrissyetal.(14).Thiswasconfirmedbycom-paringmeasurementsafterrepeatedmarking(twotimes)byeachofthreeobservers.Othererrorsresultfromvaria-tionsinpatientposture,andthesewouldaffectthemea-suresofdiscwedgingbutnotofvertebralwedging.Theresultsoftheseandotherstudiessuggestthatver-tebralbodywedgingoccursearlyinthedevelopmentofascoliosisdeformity,andthatboththediscsandvertebraedevelopanincreasingwedgedeformityinsimilarpropor-tionswithcurveprogression.Thevertebraegenerallyshowedalargerdeformitythanthediscsinthoraciccurves,andthediscsdevelopedalargerdeformityinlum-barandthoracolumbarcurves.ThesimilaritybetweentheI.A.F.STOKESANDD.D.ARONSSON321JSpinalDisord,Vol.14,No.4,2001 disc/vertebrawedgeratioseeninbothsmallandlargecurves,andinidiopathicscoliosisandscoliosisassociatedwithcerebralpalsy,suggeststhatthedeformityoccursinbothstructures,andcannotbeattributedtoeitherstructurealone.Inthemanagementofscoliosis,includingsmallcurves,itshouldberecognizedthatboththevertebraeanddiscshaveawedgingdeformity.ThisstudywassupportedbyNationalIn-stitutesofHealthgrantR55HD34460.TheauthorsthankLoriKarol,M.D.,whorecruitedmanyofthepatientsinthisstudy,andMarciJones,M.D.,PeterRonchetti,M.D.,andJohnGold-blatt,M.D.,whoidentifiedandmarkedmanyoftheradiographs.1.CruickshankJL,KoikeM,DicksonRA.Curvepatternsinidiopathicscoliosis:aclinicalandradiographicstudy.JBoneJointSurg[Br]2.TaylorJR.Growthofhumanintervertebraldiscsandvertebralbod-JAnat3.TaylorTK,GhoshP,BushellGR.Thecontributionoftheinterver-tebraldisktothescolioticdeformity.ClinOrthop4.StokesIAF,AronssonDD,UrbanJPG.Biomechanicalfactorsin-fluencingprogressionofangularskeletaldeformitiesduringgrowth.EurJExpMusculoskelRes5.GoodingCA,NeuhauserEBD.Growth&developmentofthever-tebralbodyinthepresenceandabsenceofnormalstress.AmJRoentgRadiumTherNuclMed6.LonsteinJE,CarlsonJM.Thepredictionofcurveprogressioninuntreatedidiopathicscoliosisduringgrowth.JBoneJointSurg[Am]7.MillerA,TempleT,MillerF.Impactoforthosesontherateofscoliosisprogressioninchildrenwithcerebralpalsy.JPediatrOr-8.XiongB,SevastikJA,HedlundR,etal.Radiographicchangesatthecoronalplaneinearlyscoliosis.9.PerdriolleR,BecchettiS,VidalJ,etal.Mechanicalprocessandgrowthcartilages.Essentialfactorsintheprogressionofscoliosis.10.PerdriolleR.LaScoliose.Sonétudetridimensionelle.Paris:MaloineSAEditeur,1979:556.11.StillwellDL.Structuraldeformitiesofvertebrae:boneadaptationandmodelinginexperimentalscoliosisandkyphosis.JBoneJointSurg[Am]12.MentePL,StokesIAF,SpenceH,etal.Progressionofvertebralwedginginanasymmetricallyloadedrattailmodel.13.AubinCE,DansereauJ,PetitY,etal.Three-dimensionalmeasure-mentofwedgedscolioticvertebraeandintervertebraldisks.SpineJ14.MorrissyRT,GoldsmithGS,HallEC,etal.MeasurementoftheCobbangleonradiographsofpatientswhohavescoliosis.Evalua-tionofintrinsicerror.JBoneJointSurg[Am]DISCANDVERTEBRALWEDGINGINSCOLIOSISJSpinalDisord,Vol.14,No.4,2001