a comparative study between data obtained from ...€¦ · conventional lateral cephalometry and...
TRANSCRIPT
123
A comparative study between data obtained from conventional lateral cephalometry and
reconstructed three-dimensional computed tomography images
Suseok Oh1, Ci-Young Kim2, Jongrak Hong1
1Department of Oral and Maxillofacial Surgery, Samsung Medical Center, Sungkyunkwan University School of Medicine, 2Department of Nursing, Hanyang University College of Medicine, Seoul, Korea
Abstract (J Korean Assoc Oral Maxillofac Surg 2014;40:123-129)
Objectives: The aim of this study was to verify the concordance of the measurement values when the same cephalometric analysis method was used for two-dimensional (2D) cephalometric radiography and three-dimensional computed tomography (3D CT), and to identify which 3D Frankfort hori-zontal (FH) plane was the most concordant with FH plane used for cephalometric radiography. Materials and Methods: Reference horizontal plane was FH plane. Palatal angle and occlusal plane angle was evaluated with FH plane. Gonial angle (GA), palatal angle, upper occlusal plane angle (UOPA), mandibular plane angle (MPA), U1 to occlusal plane angle, U1 to FH plane angle, SNA and SNB were obtained on 2D cephalmetries and reconstructed 3D CT. The values measured eight angles in 2D lateral cephalometry and reconstructed 3D CT were evaluated by intraclass correlation coefficiency (ICC). It also was evaluated to identify 3D FH plane with high degree of concordance to 2D one by studying which one in four FH planes shows the highest degree of concordance with 2D FH plane. Results: ICCs of MPA (0.752), UOPA (0.745), SNA (0.798) and SNB (0.869) were high. On the other hand, ICCs of gonial angle (0.583), palatal angle (0.287), U1 to occlusal plane (0.404), U1 to FH plane (0.617) were low respectively. Additionally GA and MPA acquired from 2D were bigger than those on 3D in all 20 patients included in this study. Concordance between one UOPA from 2D and four UOPAs from 3D CT were evaluated by ICC values. Results showed no significant difference among four FH planes defined on 3D CT.Conclusion: FH plane that can be set on 3D CT does not have difference in concordance from FH plane on lateral cephalometry. However, it is desir-able to define FH plane on 3D CT with two orbitales and one porion considering the reproduction of orbitale itself.
Key words: Lateral cephalography, Three-dimensional computed tomography, Frankfort horizontal plane[paper submitted 2014. 5. 16 / revised 2014. 6. 12 / accepted 2014. 6. 13]
Variousmethodologieshavebeenintroduced,andnormal
averagevalue(norm)forcorrespondinganalysishasbeen
reported2.Analysisoflateralcephalometryincludestheanaly-
sisofangleandlengthofhardtissuesandsofttissuesofthe
faceandisthemostcommonlyusedmethod3.However,two-
dimensional (2D) lateralcephalometry isassociatedwith
manyproblems.First,itisdifficulttoachieveaccurateiden-
tificationduetolandmarkoverlap4-6.Second,itisdifficultto
clearlyreflectthedifferencebetweentherightandleftsidesof
theface.Third,thereisafundamentalprobleminthatathree-
dimensional(3D)subjectisillustratedintwodimensions7.
Inaddition,weaknessesof2Dlateralcephalometryhave
beenreportedincludinglowconcordancewithresultsofac-
tualclinicalevaluation8,9.Conventionallateralcephalometry
hasanotherlimitationinthatdeficiencyintheparanasaland
I. Introduction
Accuratediagnosisisthemostcriticalassignmentinplan-
ningorthognathicsurgery,andconventionallateralcepha-
lometryhasbeenthestandardforevaluationandorthodontic
diagnosisofmaxillofacialdeformitysincetheearly1930s1.
ORIGINAL ARTICLE
Jongrak HongDepartment of Oral and Maxillofacial Surgery, Samsung Medical Center, Sungkyunkwan University School of Medicine, 81 Irwon-ro, Gangnam-gu, Seoul 135-710, KoreaTEL: +82-2-3410-2420 FAX: +82-2-3410-0038E-mail: [email protected] This is an open-access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/3.0/), which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.
CC
Copyright Ⓒ 2014 The Korean Association of Oral and Maxillofacial Surgeons. All rights reserved.
http://dx.doi.org/10.5125/jkaoms.2014.40.3.123pISSN 2234-7550·eISSN 2234-5930
ThisresearchwassupportedbyagrantoftheKoreaHealthTechnologyR&DProjectthroughtheKoreaHealthIndustryDevelopmentInstitutefundedbytheMinistryofHealth&Welfare,RepublicofKorea(HI13C1491).
J Korean Assoc Oral Maxillofac Surg 2014;40:123-129
124
theDepartmentofOralandMaxillofacialSurgery,Samsung
MedicalCenter(Seoul,Korea)in2012whounderwentlater-
alcephalometry(PlanmecaProMax;PlanmecaOy,Helsinki,
Finland)and3DCT(GELightSpeedVCTXT;General
ElectronicMedicalSystem,Milwaukee,WI,USA)imaging.
Allpatientswereskeletallymatureandnolongergrowing.In
total,20patientswereselectedamongwhomtherewere12
malesandeightfemaleswithanaverageageof23.2years.
Theareafromthenasalbonetothemandiblewasincluded
intheimage,andtheaxialslicethicknesswas1mm.During
CTscanning,thejawandteethweremaintainedincentric
occlusion.FacialboneCTwastakenwiththepatient ina
supineposition.AftercompletionofCTscanning,theim-
agesweresavedasdigital imagingandcommunicationin
medicine(DICOM)files.CTimageswerealsoreconstructed
toproduce3DimagesusingSimplantPro2011(Materialize
DentalNV,Leuven,Belgium).
Thefollowingpatientswereexcluded:thosewithsevere
congenitalmalformation,suchashemifacialmicrosomia,
thosewithpathologicallesionsofthejaw,andthosewithse-
vereinflammation.
Inthereconstructed3Dimage, lines,planes,andangles
weresetasfollows,andtheFHplanewasthehorizontalref-
erenceplaneinthisstudy.
LateralcephalometryimageswereanalyzedusingV-ceph
infraorbitalareascannotbeaccuratelydemonstrated.
3Dfacialanalysishasbeenpresentedtosolvesuchshort-
comingsandcanbeusedtoachievesimulationtreatmentfor
3Doperationandorthodonticprocesses10-13.Systemshave
beenestablishedfornormalaveragevalues,landmarks,ref-
erencelines,andreferenceplanes,allofwhichcanserveas
standardsofanalysis14.Ifalandmarkisdefinedfor3Danaly-
sis,suchlandmarksshouldbeusedfor2Danalysisinmost
cases.Thisisbecauselateralcephalometryanalysishasbeen
widelyperformed,andthemassiveamountofdataaccumu-
latedcanbeeffectivelyusedfor3Danalysis15.However,few
studieshaveanalyzedwhether3Danalysisagreeswiththe
largeamountof2Ddata.
Inthisregard,thisstudyanalyzedtheconcordanceandpat-
ternsbetweeneightanglesmeasuredusinglateralcephalom-
etryaswellas3Dcomputedtomography(CT)images.Inad-
dition,italsoaimstoidentifya3DFrankforthorizontal(FH)
planewithahighdegreeofconcordancetothe2Dplaneby
studyingfourFHplanes.
II. Materials and Methods
1. Patient selection and data collection
Thestudypopulationconsistedof20patientswhovisited
Table 1. Definitions of the angles, linear measurements, and planes
MeasurementsDefinition
2Dcephalometry Reconstructed3DCT
AngularmeasurementGonialanglePalatalangleMandibularplaneangleUpperocclusalplaneangleU1toocclusalplaneangleU1toFHplaneangleSNASNBLinearmeasurementUpperocclusallineMandibularplanelinePalatallineU1linePlanemeasurementFHplaneFHplane1FHplane2FHplane3FHplane4
AnglefromMetotheGototheArAnglebetweenpalatallineandFHlineAnglebetweenSNlineandMnplanelineAnglebetweenupperocclusallineandFHplaneAnglebetweenU1lineandupperocclusallineAnglebetweenU1lineandFHplaneAnglefromsellatothenasiontothesubspinalpointAnglefromsellatothenasiontothesupramentalpointLinethroughtipofcentralincisorandMBcuspoffirstmolarLinethroughcorpusleftandMeLinethroughANSandPNSLinethroughtipandapexofcentralincisorLinethroughOrandPo
----
AveragebetweenrightandleftgonialanglesSameas2DcephalometrySameas2DcephalometryAnglesbetweenupperocclusallineandeachFHplaneSameas2DcephalometryAnglebetweenU1lineandeachFHplaneSameas2DcephalometrySameas2DcephalometryLine through(midpointbetween#16and#26)and(midpointbetweencentralincisorstip)
Linethrough(midpointbetweenGo)andMeSameas2DcephalometryLinethroughtipandapexofcentralincisor
-PlanethroughbothOrandrightPoPlanethroughbothOrandleftPoPlanethroughrightOrandbothPoPlanethroughleftOrandbothPo
(2D:two-dimensional,3D:three-dimensional,Me:menton,Go:gonion,Ar:articulare,FH:Frankforthorizontal,SN:sellanasion,Mn:mandible,U1:upper1stincisor,MB:mesiobuccal,Or:orbitale,Po:porion)Suseok Oh et al: A comparative study between data obtained from conventional lateral cephalometry and reconstructed three-dimensional computed tomography images. J Korean Assoc Oral Maxillofac Surg 2014
Comparative study between 2D cephalography and reconstructed 3D CT
125
werenotsignificantlydifferent.TheICCdifferencesamong
thefouranglesacquiredin3Dwerealsonotsignificantdif-
ferent.(Table2)
2. Concordance evaluation between measured angles
from 2D and 3D systems
Concordanceoftheaverageofeightangleswasverified.
(Table3)ICCsofGA,palatalangle,U1OPA,andU1FHPA
were lowat0.583,0.287,0.404,and0.617,respectively.
Thusdemonstratinga lowconcordancebetween lateral
cephalometryand3DCT.Ontheotherhand,ICCsofMPA,
UOPA,SNA,andSNBwerehighat0.752,0.745,0.798,and
0.869,respectively.Additionally,GAandMPAacquiredin
2Dwerelargerthanthosein3Dforall20patientsincluded
inthisstudy.(Table4)
IV. Discussion
Two-dimensionalcephalometryhasbeenwidelyusedfor
diagnosisandoperationincraniofacialskeletaldevelopment
andorthodonticandorthognathicsurgeries13,16.Thisanalysis
hasbeenusedtoanalyzelinesandanglesonanimage.The
measuredvaluesfromthepatientscanbecomparedwith
generallyknownnormalvaluestojudgetheasymmetry17.In
lateralcephalometry,however,itisdifficulttodistinguishthe
overlappingstructures;inpatientswithanasymmetricface,
theaccuracyofanalysisislow.Consequently,itisdesirable
toalsouseotherradiographic images inorder toachieve
moreaccurateanalysis.
Toevaluatetheasymmetryofaface,panoramaradiogra-
phy18andposteroanteriorcephalometryhavebeentradition-
5.5(CyberMedInc.,Seoul,Korea).Allanalysiswasper-
formedbyasingleinvestigator.
2. Concordance evaluation between the FH planes
defined on 2D and 3D systems
TheFHplaneonlateralcephalometrywasdefinedwiththe
linespassingtheorbitale(Or)andporion(Po),andtheupper
occlusalplaneangle(UOPA)wasmeasured.FourFHplanes
(Table1)wereestablishedinthe3Dsystem,andtheUOPA
ofeachpointwasmeasured.Theanglesmeasuredin3Dand
lateralcephalometrywereevaluatedwithregardtointraclass
correlationcoefficient(ICC)withIBMSPSSStatistics21.0
(IBMCo.,Armonk,NY,USA).
3. Concordance evaluation between measured angles
from 2D and 3D systems
(1)Gonialangle(GA),(2)palatalangle,(3)mandibular
planeangle (MPA), (4)UOPA,(5)U1 toocclusalplane
angle(U1OPA),(6)U1toFHplaneangle(U1FHPA),(7)
SNA,(8)SNBweremeasuredbothinlateralcephalometry
andreconstructed3DCT(Table1)andwereevaluatedwith
regardtoICC.
III. Results
1. Concordance evaluation between FH planes defined
on 2D and 3D systems
Onlateralcephalometry,theaverageUOPAwas11.38o.
WhentheconcordanceofUOPAagainstthefourFHplanes
on3Dwasevaluated, theICCsshowedhighconcordance
with0.745,0.751,0.755,and0.754,respectively.Theresults
Table 2. Average UOPA values on 2D cephalometry and 3D CT and analysis of reliability (o)
2Dcephalometry Reconstructed3DCT ICC
UOPAUOPA1UOPA2UOPA3UOPA4
11.38----
-10.8311.8211.0711.66
-0.745*0.751*0.755*0.754*
(UOPA:upperocclusalplaneangle,2D: two-dimensional,3DCT:three-dimensionalcomputedtomography,ICC:intraclasscorrelationcoefficient)*Statisticalsignificance(P<0.05).Suseok Oh et al: A comparative study between data obtained from conventional lateral cephalometry and reconstructed three-dimensional computed tomography images. J Ko-rean Assoc Oral Maxillofac Surg 2014
Table 3. Average correlations between two groups (o)
2Dcephalometry Reconstructed3DCT ICC
GAPalatalangleU1OPAU1FHAMPAUOPASNASNB
126.802.0052.16116.4738.0411.3884.8283.21
120.240.0552.91116.5633.5610.8384.2283.64
0.5830.2870.4040.6170.7520.7450.7980.869
(2D: two-dimensional,3DCT: three-dimensionalcomputed tomo-graphy, ICC: intraclasscorrelationcoefficient,GA:gonialangle,U1OPA:U1toocclusalplaneangle,U1FHA:U1toFHplaneangle,MPA:mandibularplaneangle,UOPA:upperocclusalplaneangle)Suseok Oh et al: A comparative study between data obtained from conventional lateral cephalometry and reconstructed three-dimensional computed tomography images. J Ko-rean Assoc Oral Maxillofac Surg 2014
J Korean Assoc Oral Maxillofac Surg 2014;40:123-129
126
alsoreportedaccuracyofmeasurementon3DCT.
Inaddition,3Dlateralcephalometryanalysisabstracted
from3DCTwasintroduced,anditsaccuracyhasbeenre-
ported.vanVlijmenetal.1performedcomparativeanalysis
onvariousanglesanddistancesbetween3Dlateralcepha-
lometryandconventionallateralcephalometryandreported
thattherewasnostatisticallysignificantdifferenceinmany
parts.
It is truethat lateralcephalometryanalysishasbeenthe
majormethodtoevaluatefacialtypesfortensofyears15,25.3D
CTanalysis,averyusefulmethodindiagnosis,usesthesame
landmarksasconventionallateralcephalometry15.However,
theconcordancebetweenthemeasurementsoflateralcepha-
lometryand3DCThasnotbeensufficientlystudied.Using
therelationshipandpatternsofmeasurementsacquiredfrom
twoanalysismethods,aswellastheirconcordance,3DCT
valuescanbedeterminedfromthevaluesoflateralcepha-
lometryanalysis.Thus,accumulatedlateralcephalometry
valuescanbeconvertedto3D.However,fewcaseshaveper-
formedaclearcomparisonbetweenthetwomethods.
TheUOPAabstractedfromtheFHplanedefinedin3D
CTandthatfromlateralcephalometrywerefoundtobevery
similar.(Table2)Similarly,thefour3DFHplanesdidnot
showdifferencesinconcordancewiththoseof2Danalysis.
Insuchcases,itisdesirabletochooselandmarkswithhigher
allyusedwithlateralcephalometry.Panoramaradiography
is themostfamiliar techniquetomostdentistsandortho-
dontists,andit istherepresentativeexaminationfordental
diagnosisand treatmentplanning.This typeof imaging
enablesdentiststodetectcystsandtumors,supernumerary
teeth,missingteeth,andbonydeformitiesandtoevaluatethe
asymmetryofthelowerjawbone19,20.Updegrave21,however,
reportedthatanincorrect treatmentplanmayresultsfrom
seriousasymmetryofthemandibleramusandcondyleand
coronoidprocessesonpanoramaradiography.Otherauthors
havereporteddifficultyinanalyzingthefacewithpanorama
radiography22.Posteroanteriorcephalometryhas thesame
problemin thatpreciseperception isdifficultbecauseof
manylandmarkoverlaps13.Assuch,therearelimitationsto
theevaluationof3Dstructureson2Dradiographicimages.
However,nooverlappinglandmarksareproducedin3D
imaging,minimizingtheconfusionof left-andright-side
structures.Therefore,with3DCTimaging,clinicianscan
diagnoseallpatientsaccuratelyincludingthosewithfacial
asymmetry,easilyidentifytheshapesofcraniofacialstruc-
tures,andrecognizetheshapesofbonesandsofttissuesfrom
variousangles.According topreviousstudiesmeasuring
lengthonvolume-rendered3DCTimagesandincadavers23,
thedifferenceinlengthbetweentwogroupswasminimal,
confirmingtheaccuracyof3D-CTimages.Hildeboltetal.24
Table 4. GA and MPA obtained from all patients
PatientNo.GA(o) MPA(o)
2Dcephalometry Reconstructed3DCT 2Dcephalometry Reconstructed3DCT
1234567891011121314151617181920
Mean
121.21120.91128.54125.29118.96135.06126.54127.99115.46138.15137.88124.88109.51133.72127.27125.59133.71134.44129.16122.78126.85
117.13114.89122.70122.16115.70128.10122.09118.90109.77130.11127.64122.86104.14125.29126.17119.59119.32120.56117.72118.99120.19
38.4430.3142.4333.8637.7340.8342.9745.4429.5438.0536.3136.0122.8646.0043.1135.0937.4933.7740.7550.2838.06
35.2126.9338.1328.2732.5637.5237.2439.0625.2630.7235.9731.7521.3038.5243.1132.1831.2027.7635.5650.2733.93
(GA:gonialangle,MPA:mandibularplaneangle,2D:two-dimensional,3DCT:three-dimensionalcomputedtomography)Suseok Oh et al: A comparative study between data obtained from conventional lateral cephalometry and reconstructed three-dimensional computed tomography images. J Korean Assoc Oral Maxillofac Surg 2014
Comparative study between 2D cephalography and reconstructed 3D CT
127
on3DCT(Fig.2B)andobservingtheresulting3Dlateral
cephalometry,MemarkedonthesuperiorpositionthatMe
onconventionallateralcephalometry.(Fig.2C)Thisoccurs
becauseanatomicstructuresneartheMeareextendedinthe
lowerdirectionon3DCT.(Fig.2A)Inpatientswithsuch
structures,the3Dvaluesmaybedifferentfromthoseon2D
evenwhenMeisaccuratelymarkedon3DCT.Additionally,
whenMeismarkedatthefrontofskull,itmaybemarked
inalittlesuperiorposition(Fig.2D),anditmayleadtoer-
roneousunderestimationofGAandMPA.Thus,whenMeis
markedon3DCT,itisrecommendedtoapproachfromthe
bottomofthelowerjawbone.
Second,articulare(Ar)mayhavelargerx-axiserroron
lateralcephalometrycomparedwith3DCT.Trpkovaetal.7
confirmedthroughmeta-analysisthatArandbasionshowed
representationandreliability.InastudybyYuetal.26,two
observersmarkedlateralcephalometrylandmarksthreetimes
andcalculatedtheconcordancewithrespecttothex-axisand
y-axis.Inthisexperiment,Poshowedthelowestconcordance
among17landmarksonthex-axis,whileOrdidnotshow
asignificantdifferenceinconcordanceamongobserversor
amongtrialsofthesameobserver.Additionally,whendefin-
ingPoon3DCT,awidecurveisproduced,makingrepeti-
tionandreproductiondifficult.(Fig.1)Basedontheseresults,
itisreasonabletodefinetheFHplanein3DusingtwoOr
andonePovalues.
Accordingtotheresultsofthisstudy,GAmeasuredon2D
islargerthanthatmeasuredon3Dinallpatients.(Table4)
Therearethreepossibleexplanations.First, thedifference
maybebecauseoferrorinmenton(Me).WhenmarkingMe
Fig. 2. A. Downward extended adjacent structure on both sides (white arrows). B. Position of the menton (Me) in the inferior-superior di-rection. C. Position of the Me on three-dimensional (3D) cephalometry. D. Me positioned in the anterior-posterior view. E. Me positioned in the anterior view. F. Position of the Me on 3D cephalometry.Suseok Oh et al: A comparative study between data obtained from conventional lateral cephalometry and reconstructed three-dimensional computed tomography images. J Korean Assoc Oral Maxillofac Surg 2014
Fig. 1. A. Large, round-shaped bony edge around the porion. B. Sharp edge around the orbitale.Suseok Oh et al: A comparative study between data obtained from conventional lateral cephalometry and reconstructed three-dimensional computed tomog-raphy images. J Korean Assoc Oral Maxillofac Surg 2014
J Korean Assoc Oral Maxillofac Surg 2014;40:123-129
128
References
1. vanVlijmenOJ,BergéSJ,SwennenGR,BronkhorstEM,KatsarosC,Kuijpers-JagtmanAM.Comparisonofcephalometricradio-graphsobtainedfromcone-beamcomputedtomographyscansandconventionalradiographs.JOralMaxillofacSurg2009;67:92-7.
2. AthanasiouAE.Orthodonticcephalometry.London:Mosby-Wolfe;1997.
3. AdamsGL,GanskySA,MillerAJ,HarrellWEJr,HatcherDC.Comparisonbetweentraditional2-dimensionalcephalometryanda3-dimensionalapproachonhumandryskulls.AmJOrthodDento-facialOrthop2004;126:397-409.
4. MillerPA,SavaraBS,SinghIJ.Analysisoferrorsincephalomet-ricmeasurementof three-dimensionaldistancesonthemaxilla.AngleOrthod1966;36:169-75.
5. GravelyJF,BenziesPM.Theclinicalsignificanceoftracingerrorincephalometry.BrJOrthod1974;1:95-101.
6. MidtgårdJ,BjörkG,Linder-AronsonS.Reproducibilityofcepha-lometriclandmarksanderrorsofmeasurementsofcephalometriccranialdistances.AngleOrthod1974;44:56-61.
7. TrpkovaB,MajorP,PrasadN,NebbeB.Cephalometriclandmarksidentificationandreproducibility:ametaanalysis.AmJOrthodDentofacialOrthop1997;112:165-70.
8. WylieGA,FishLC,EpkerBN.Cephalometrics:acomparisonoffiveanalysescurrentlyusedinthediagnosisofdentofacialdefor-mities.IntJAdultOrthodonOrthognathSurg1987;2:15-36.
9. MagalhaesAE,StellaJP,EpkerBN.FacialanthropometricsversuscephalometryaspredictorsforsurgicaltreatmentinpatientswithclassIIIdentofacialdeformities.IntJAdultOrthodonOrthognathSurg1995;10:295-302.
10. SwennenGR,SchutyserF.Three-dimensionalvirtualapproachtodiagnosisandtreatmentplanningofmaxillo-facialdeformity.In:BellW,ed.Distractionosteogenesisofthefacialskeleton.Hamil-ton,ON:BCDeckerInc;2007:55.
11. MakiK,OkanoT,MorohashiT,YamadaS,ShibaskiY.Theap-plicationofthree-dimensionalquantitativecomputedtomographytothemaxillofacialskeleton.DentomaxillofacRadiol1997;26:39-44.
12. VannierMW,MarshJL,WarrenJO.ThreedimensionalCTrecon-structionimagesforcraniofacialsurgicalplanningandevaluation.Radiology1984;150:179-84.
13. TrpkovaB,PrasadNG,LamEW,RaboudD,GloverKE,MajorPW.Assessmentoffacialasymmetriesfromposteroanteriorcepha-lograms:validityofreferencelines.AmJOrthodDentofacialOr-thop2003;123:512-20.
14. WongRW,ChauAC,HäggU.3DCBCTMcNamara'scephalo-metricanalysisinanadultsouthernChinesepopulation.IntJOralMaxillofacSurg2011;40:920-5.
15. MaedaM,KatsumataA,ArijiY,MuramatsuA,YoshidaK,GotoS,etal.3D-CTevaluationoffacialasymmetryinpatientswithmaxil-lofacialdeformities.OralSurgOralMedOralPatholOralRadiolEndod2006;102:382-90.
16. ChebibFS,ChammaAM.Indicesofcraniofacialasymmetry.AngleOrthod1981;51:214-26.
17. EpkerBN,FishLC,StellaJP.Dentofacialdeformities:integratedorthodonticandsurgicalcorrection.StLouis:CVMosby;1998:29-33.
18. RaustiaAM,SalonenMA.Gonialanglesandcondylarandramusheightofthemandibleincompletedenturewearers--apanoramicradiographstudy.JOralRehabil1997;24:512-6.
19. PiedraI.TheLevandoskipanoramicanalysisinthediagnosisoffacialanddentalasymmetries.JClinPediatrDent1995;20:15-21.
20. MattilaM,KönönenM,MattilaK.Verticalasymmetryoftheman-dibularramusandcondylarheightsmeasuredwithanewmethodfromdentalpanoramicradiographsinpatientswithpsoriaticarthri-tis.JOralRehabil1995;22:741-5.
thelargestx-axiserrorsonlateralcephalometry.Thepos-
sibilityofloweraccuracyofGAcannotbeexcludedbecause
ofloweraccuracyofthereferencepoints.
Third,Ardefinedin3Disdifferentfromthedefinition
in2D.Aronlateralcephalometryisdefinedasthepointof
overlapbetweentheskullbaseandposteriorsurfaceofthe
lowerjawbonecondyle,whileArinthepresentstudyisde-
finedasthemostposteriorpointofthemandibularcondyle.
Thus,itisdifficulttoassertthatArinthisstudyisabsolutely
concordantwithAron2D.However,itisimpossibletode-
fineArin3Dusingthesamedefinitionofthatin2D.There-
fore,ifArisdefinedusingthemethodofthisstudy,GAin
2Disslightlylargerthanthatin3DCT.
MPA, just likeGA,was larger in2Dthan in3Dinall
patients.Thiscanbeexplainedbytheabovementionedprob-
lemsassociatedwithMe.
Theanglesshowinglowconcordancewerepalatalangle,
U1OPA,andU1FHPA.(Table3)Anangleismadebytwo
lines;ifthelinesareshort,themeasurementismoredifficult,
andtheriskoferrorincreases,whichlowerstheprobability
ofrepetitionandreproductionandconsequentlytheprobabil-
ityofconcordance.ThestraightpalatallineandU1lineused
todefinetheaforementionedthreeangleshadshortdistances,
whichresulted in their lowerconcordancevalues.Onthe
otherhand,MPA,UOPA,SNA,andSNBwereformedby
twolonglinesandsoshowedhigherconcordance.
V. Conclusion
Formeasurementofanglesmadeupoftwolonglines,the
concordancebetweentwomethodsishigh.Inthisstudy,we
showedhighconcordanceforMPA,UOPA,SNA,andSNB.
WhenMewasincludedintheanalysis,itwasmarkedina
different locationfromthatof2Danalysisbecauseof the
natureofthemeasurementpoints.Additionally,theFHplane
on3DCTshowedconcordancewiththeFHplaneonlateral
cephalometry, indicatingnodifferenceinconcordanceno
matterwhatisselectedamongfourFHplanesdefinedon3D
CT.However,itisdesirabletodefinetheFHplaneon3DCT
withtwoOrandonePovaluesconsideringthereproduction
ofOritself.
Conflict of Interest
Nopotentialconflictofinterestrelevanttothisarticlewas
reported.
Comparative study between 2D cephalography and reconstructed 3D CT
129
21. UpdegraveWJ.Visualizingthemandibularramusinpanoramicradiography.OralSurgOralMedOralPathol1971;31:422-9.
22. YeoDK,FreerTJ,BrockhurstPJ.Distortionsinpanoramicradio-graphs.AustOrthodJ2002;18:92-8.
23. CavalcantiMG,HallerJW,VannierMW.Three-dimensionalcom-putedtomographylandmarkmeasurementincraniofacialsurgicalplanning:experimentalvalidationinvitro.JOralMaxillofacSurg1999;57:690-4.
24. HildeboltCF,VannierMW,KnappRH.Validationstudyofskullthree-dimensionalcomputerizedtomographymeasurements.AmJPhysAnthropol1990;82:283-94.
25. MulickJF.Clinicaluseof thefrontalheadfilm.AngleOrthod1965;35:299-304.
26. YuSH,NahmDS,BaekSH.Reliabilityoflandmarkidentificationonmonitor-displayedlateralcephalometricimages.AmJOrthodDentofacialOrthop2008;133:790.e1-6;discussione1.