University of Groningen
Accuracy of fit analysis of the patient-specific Groningen temporomandibular joint prosthesis
Merema, B. J.; Kraeima, J.; Witjes, M. J. H.; van Bakelen, N. B.; Spijkervet, F. K. L.
Published in:
International Journal of Oral and Maxillofacial Surgery
DOI:
10.1016/j.ijom.2020.08.012
IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite from
it. Please check the document version below.
Document Version
Publisher's PDF, also known as Version of record
Publication date:
2021
Link to publication in University of Groningen/UMCG research database
Citation for published version (APA):
Merema, B. J., Kraeima, J., Witjes, M. J. H., van Bakelen, N. B., & Spijkervet, F. K. L. (2021). Accuracy of
fit analysis of the patient-specific Groningen temporomandibular joint prosthesis. International Journal of
Oral and Maxillofacial Surgery, 50(4), 538-545. https://doi.org/10.1016/j.ijom.2020.08.012
Copyright
Other than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons).
Take-down policy
If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim.
Downloaded from the University of Groningen/UMCG research database (Pure): http://www.rug.nl/research/portal. For technical reasons the number of authors shown on this cover page is limited to 10 maximum.
Research
Paper
TMJ
Accuracy
of
fit
analysis
of
the
patient-specific
Groningen
temporomandibular
joint
prosthesis
B.J.Merema,J.Kraeima,M.J.H.Witjes,N.B.vanBakelen, F.K.L. Spijkervet: Accuracyoffitanalysisofthepatient-specificGroningentemporomandibularjoint prosthesis. Int.J.OralMaxillofac.Surg.2021;50:538–545. ã2020TheAuthor(s). PublishedbyElsevierLtdonbehalfofInternationalAssociationofOraland MaxillofacialSurgeons.ThisisanopenaccessarticleundertheCCBYlicense(http:// creativecommons.org/licenses/by/4.0/).
B.J.Merema,J.Kraeima,
M.J.H.Witjes,N.B.vanBakelen, F.K.L.Spijkervet
DepartmentofOralandMaxillofacialSurgery, UniversityMedicalCenterGroningen, UniversityofGroningen,Groningen,The Netherlands
Abstract. Totaljointreplacement(TJR)withaprosthesiscanbeindicatedforpatients withsevere temporomandibularjoint(TMJ)dysfunction.Surgicalaccuracyis necessaryforcorrecttranslationofthepreoperativelypredictedfunctionaloutcome, wear,andbiomechanicalbehaviourofthepatient-specificTMJ-TJRprosthesis. Thisstudydescribesthefirstclinicalapplicationsofthepatient-specificTMJ-TJR prosthesisaccordingtotheGroningenprinciples(G-TMJ-TJR),whichwas developedandvalidatedinapriorhumancadaverteststudy.Theaimofthisstudy wastovalidatetheaccuracyofplacementofthepatient-specificG-TMJ-TJRinthe clinicalsetting.Itwashypothesizedthatavirtualsurgicalplan(VSP)combined withguidedplacementofthepatient-specificG-TMJ-TJRwouldbeperformedas predictablyandaccuratelyasinthepriorcadaverseries.Allpatientswhoreceiveda VSP-basedpatient-specificG-TMJ-TJRbetweenDecember2017andMarch2020 wereincludedinthisstudy.Theaccuracyanalysiswasbasedonpostoperativecone beamcomputedtomography(CBCT)data.All11prosthesescouldbeinsertedusing routine pre-auricularandretromandibularsurgicalapproaches.Analysisofthe VSPsandpostoperativeCBCTsshowedanaveragethree-dimensionaldeviationof 1.07mm(standarddeviation0.46mm,range0.33–1.91mm)forallofthefossaand mandibularcomponents.Thepatient-specificG-TMJ-TJRcanbeapplied predictablyandaccuratelyinaclinicalsetting.
Key words:TMJR; alloplastic; arthroplasty; 3D-VSP;guidedsurgery;custom; CAD/CAM; ankylosis.
Acceptedforpublication24August2020 Availableonline14September2020
Forpatientswithsevere temporomandib-ularjoint(TMJ)dysfunction,atotaljoint replacement (TJR) of the TMJ using a prosthesis may be considered. Reported indications are end-stage degenerative joint disease, recurrent ankylosis, and
congenital disorders affecting the TMJ, if conservativetreatmentorregularopen joint surgery do not suffice1. Moreover, condylarlossduetoneoplasiaortrauma, or the need for a revision of a failed alloplastic or autogenous reconstruction,
are reported indications for TMJ-TJR2. Mandibular movement is impaired for mostofthesepatients,duetoanatomical changesorsurgicallycausedscarification. Thisoften results inpain, difficulties in speech,andimpairedoralfunction.
0901-5027/040538+08 ã2020TheAuthor(s).PublishedbyElsevierLtdonbehalfofInternationalAssociationofOralandMaxillofacialSurgeons.Thisisan openaccessarticleundertheCCBYlicense(http://creativecommons.org/licenses/by/4.0/).
ReplacementofaTMJwithatotaljoint prosthesiscanhaveagreatimpactonthe biomechanicalaspectofthecontralateral jointandcanaffectthecomplex mandib-ular movements, which consist of both rotational and translational compo-nents3,4,5. Over the last few decades, a number of prostheses have been devel-oped with the aim ofreconstructing the TMJandrestoringitsphysiological move-ments, especiallyover the last 2 years6. MostofthecurrentlyavailableTMJ-TJRs areproducedusingcomputer-aideddesign and manufacturing (CAD/CAM) techni-ques, enabling virtual patient-specific modelling. Generally, designs are based on two commercially available prosthe-ses7,8consistingofbothaskulland man-dibular component, which are fixed separately totheglenoidfossa and man-dibular ramus, respectively. These parts havethefreedomtoarticulaterelativeto eachother,enablingmovementbetweena concaveandconvexbody.Thedesignsof these prostheses, however, do not seem to mimic the natural TMJ movement properlyinallcases5.
Thevalueofpatient-specificprostheses has been widely recognized and should increase treatment predictability and
improve long-term success, due to the closefitoftheprosthesis,whichimproves osseointegration9,10,11. Several CAD/ CAM prostheses have been placed with theaidofsurgicalguidestotranslatethe virtualsurgicalplan(VSP)tothe operat-ingtheatre12,13,14,15,16. However,data on the surgical accuracy of implementing TMJ-TJRswithsurgicalguidesarescarce. Between1983and1999, aunique de-vice–theGroningenTMJ-TJR (G-TMJ-TJR) –was developedat the University Medical Center Groningen (Groningen, TheNetherlands). Thisprosthesis makes use ofa loweredcentreofrotation com-paredtotheanatomicalsituation,whichis mathematically determined and mimics boththetranslationalandrotational com-ponents ofthehealthyjoint, withoutthe need for actual translational move-ment3,4,17,18,19,20. As a result, both the loadingandmovementofthecontralateral jointinthecaseofaunilateralprosthesis remain within the natural bound-aries3,4,17,18,19,20. The G-TMJ-TJR has separate rotation and translation sites, allowingforfreetranslationinboth med-iolateralandanteroposteriordirections,an aspect necessary for proper mastication. Furthermore, this separation allows for
improvedbearingsurfaces,unlikethe con-ventional concave–convex designs with slidingpointcontactsinsteadof ball–sock-etprincipleswithrelativelylargecontact surfaces,andenablestheoptimizationof boththerotationandtranslation articula-tionsandlowwearrates3,4,17,18,19,20.
Thefirstclinicalapplicationofthe G-TMJ-TJRwasin1999asastock prosthe-sis,andan8-yearclinicalfollow-upstudy wasconductedtoassessthefirstseriesof patientapplications21.Thisstudyshowed animprovementinmouthopeninganda reduced pain score,but the authorsalso mentionedthatthestockdevicewas diffi-culttopositionandfittothedamagedTMJ region thatis often found inTMJ disor-ders,andconcludedthatapatient-specific prosthesiswouldbepreferred.
Afterthepatient-specificversionofthe G-TMJ-TJR was developed, itwas vali-datedwiththecorrespondingdigital work-flowthroughahumancadavertestseries. Due to its fit, it appeared to be a good successortothestockprosthesis,with sub-millimetre accuracy22. The G-TMJ-TJR was subsequently available for use in patients and the first patient-specific clinicalapplicationoccurredinDecember 2017.Sincethen,aseriesof11prostheses
Accuracyoffitanalysisofthepatient-specificGroningentemporomandibularjointprosthesis 539
Fig.1. RenderingoftheGroningentemporomandibularjointprosthesis,illustratingtheuseoffixedsurgicalguidesfordrillingandcutting(A). Theguidesarefixedusingcompressionscrews(blue)andareprovidedwithstainlesssteeldrillsleeves.Theplannedcondylectomyisindicatedby thecranialedgeofthemandibularguide.(B)Alltheseparateprosthesiscomponentsarelabelled:F:titaniumfossacomponent;T:zirconia translationplate;D:UHMWPEneo-disc;S:zirconiacondylarsphere;M:titaniummandibularcomponent.
have been placed andare nowavailable foranalysis.
Theaimofthis study wastovalidate theaccuracyofplacementofthe patient-specific G-TMJ-TJR in the clinical set-ting. It was hypothesized that a VSP combined with guided placementofthe patient-specific G-TMJ-TJR would be performed predictably and accurately, inasimilarwaytothepreviouscadaver seriesresults.
Materialsandmethods
G-TMJ-TJRprosthesis
AllofthepatientsinneedofaTMJ-TJR underwentacomputedtomography(CT) scanpreoperatively(bonekernel,0.6-mm sliceinterval),withthemouthclosedand
inmaximumocclusion;thiswasusedfor the VSP and prosthesisdesign. The de-sign was conducted in-house and the screwpositionswerebasedonlocal cor-tical thickness and the position of the mandibular nerve. The chosen lowered pointofrotationandscrewpositionswere paramount in the prosthesis design pro-cess.TheG-TMJ-TJR,manufacturedand assembled by Xilloc Medical (Geleen, The Netherlands), consists ofa patient-specific grade23titanium fossa compo-nent,whichis fixedtotheglenoidfossa with 2-mm locking screws (Medicon, Tuttlingen, Germany), a patient-specific grade 23 titanium mandibular compo-nent, which is fixed bicortically to the mandibular ramus with 2-mm locking screws(Medicon,Tuttlingen,Germany), andanultra-high-molecular-weight
poly-ethylene (UHMWPE) neo-disc that is placedinbetweentheothertwo compo-nents. Zirconia to UHMWPE bearing couplesarerealizedbyconnectinga zir-coniatranslationplate andsphere, adhe-sive-free through press-fitting, to the fossa and mandibular components, re-spectively,duringfabrication (Fig. 1)22. The neo-disc is locked on the zirconia sphere at the mandibular component using a snap connection and can freely translatealongthetranslationplateofthe skullcomponent.Allofthefixedpartsare placed using patient-specific medical-grade polyamide surgicalguides, which arefixedtotheboneusing2-mmscrews. Thesurgicalguidesareusedtobothdrill allthe pilotscrewholesandto perform theplannedcondylectomyatthecondylar necklevel22.
Fig.2. Theexactsamedesignfileswereusedtoselectboththeplannedandpostoperativemarkerpositions.Superimposingthedesignfiles enablesacoordinatecomparison.Themidpointsofboththemedialandlateraledgesofthetranslationplatefortheskullcomponentwereselected, aswasthecentreofthemostventralscrewheadrecess.Themidpointofthesphere/neo-condylewaschosenforthemandibularcomponent, togetherwiththecentrepointsofthescrewheadrecessesinthemostventralscrewandthethirdcranialscrewalongtheposteriorborder(A).The plannedmarkerpositionsingreenandthepostoperativesuperimposedmarkerpositionsinredwereusedforpostoperativeanalysis(B).Aclearer viewofthemarkerdeviationisvisiblewhentheplannedprosthesisishidden(C).
Patients
Allofthepatientswhoreceiveda patient-specific G-TMJ-TJR prosthesis between December 2017 and March 2020 were includedinthisstudy.
Surgicalprocedure
A single-stage implantation is indicated whenthepatient’sfossaandcondyleare clearlyseparatedatthebonelevelonthe CT. When this is not the case, or the patient already has a prosthesis orother objectinsituthatdoesnotallowreliable scanningofthebonyrecipientarea,a two-stage surgical procedure isnecessary. In such cases, the fossa is freed up orthe obstructing objectisremovedinthefirst surgery.ACTscanforVSPandprosthesis design is then conducted following the firstsurgery.
Implantationwasperformedaccording toroutinepre-auricularand retromandib-ular approaches. During surgery, inter-maxillary fixation was applied to the preoperatively inserted surgical tooth bracketswithelastics.Thesurgicalguides wereinsertedandfixedindependentlyto themandibleorskullusing2.0-mm surgi-calscrewspriortodrillingthepilotscrew holes and performing thecondylectomy. Multiplefixationscrewswereplannedasa backupincaseofpoorgrip.Subsequently, the guides were removed and the pre-assembled titanium–zirconia fossa and mandibular components were aligned to thepre-drilledpilotscrewholes.The pros-thesis was aligned with the pre-drilled pilot screw holes, and stainless steel (316L)centring pinswereinserted (Xil-loc Medical, Geleen, The Netherlands) throughthescrewholesintheprosthesis intothedrilledpilotscrewholes.Sincethe
plannedscrew pathsarenotparallel, the centring pins only fit in one manner, resulting in a tight fit of the prosthesis tothebone.Next,thepinswerereplaced one byonewith2.0-mm lockingscrews (Medicon, Tuttlingen, Germany)and the UHMWPE neo-disc was snapped into placeovertheneo-condylewithadistinct popping sound, indicatinga good reten-tionofthedisc(supplementaryvideo).
Analysisofaccuracy
All of the study patients underwent a routinepostoperativeconebeam comput-ed tomography scan (CBCT) within 16 days after surgery, which was used to evaluate the accuracy of the prosthesis placement. The computer-aided design (CAD) files in STL format of the patient-specifictitaniumcomponentswere superimposed ontothe postoperativeCT
Accuracyoffitanalysisofthepatient-specificGroningentemporomandibularjointprosthesis 541
Fig.3. Panoramicradiographsobtained(A)pre-revisionand(B)post-revisionoftheright-sideTMJ-TJRinpatient1.Duetopriorsuboptimal positioningofthemandibularcomponent,thestockG-TMJ-TJRwasnotsatisfactory.Thisindicatesthedifficultythatcanpresentwithstock prosthesesand/orwithouttheuseofsurgicalguides.
data,andacomparisonwasmadewiththe planned positions. This alignment was performedbytwoobservers independent-ly(BM,JK)usingProPlanCMF3.0 soft-ware(Materialise,Leuven,Belgium);the inter-observer variation was determined forallfossa andmandibularcomponents separately.Theskulland mandiblewere segmented postoperatively and matched individuallytothepreoperative segmenta-tions, whilst moving along the matched CAD files. Matching was conducted in the 3-Matic 11.0 software (Materialise, Leuven, Belgium) usinga surface-based alignmentfunction.
In order to assess the deviation be-tween the planned and postoperative prosthesis positions, three landmarks were assigned to both the fossa and mandible parts. To analyse the fossa part, two coordinates were chosen on thelateralandmedialsidesofthe
trans-lation plate, in the exact middle of its length (anteroposterior direction). A third coordinate was assigned to the recessofthemostanteriorscrewhead, whichismeanttoreceivethecentreofa screwdriver. The landmarks used for the mandibular parts were assigned to the centre of the neo-condyle and recessesinthescrewheadsofthemost anterior screw and the most caudal screw along the posterior border of the mandible (Fig. 2). Since allof the translation plate, neo-condyle, and screw CADfiles are exactly thesame, thelandmarksarereproducibleandcan be assigned objectively.The deviation ofeachlandmarkcouplewasmeasured as a Euclidean or three-dimensional (3D) distance,andthemeanEuclidean errorandstandarddeviation(SD)were calculated for three landmark devia-tions per prosthesis.
Statisticalanalysis
Calculationoftheinter-observer variabil-itywasperformedinIBMSPSSStatistics version 23 (IBM Corp., Armonk, NY, USA).Theinter-observervariabilitywas determinedby calculatingthe inter-class correlation coefficient (ICC), wherebya reportedvalueof<0.40ispoor,0.40–0.59 isfair,0.60–0.74isgood,and0.75–1.00is excellent23.
Approval for this study was obtained from the Medical Ethics Board of the University Medical Center Groningen (METc2016/568).
Results
Overthedurationofthisstudy,atotalof 11prostheses were placed in 10 female patients who required either a primary (n=8) or secondary (n=2) TMJ-TJR.
Fig.4. Intraoperativepicturesshowingtheuseofthesurgicalguides.(A)Theseparatefossaandmandibularguidesaredesignedandassembled withdrillsleeves.(B)Thepilotscrewholesdrilledunderguidanceseenafterfossaguideremoval.(C)ThefossacomponentwiththeTJRinplace. (D)Themandibularresectionanddrillingguideinplace;notethetwofixationscrewsappliedtoensurethecorrectdrillpattern.(E)Thepilotscrew holesdrilledunderguidanceseenaftermandibularguideremoval.(F)Themandibularcomponentinplace.
The patients in the primary TMJ-TJR group suffered fromrecurring ankylosis, whichhadresultedinpoormouthopening, pain, and impaired speech and chewing abilities. Thetwopatientswhoneededa secondary TJR had received stock G-TMJ-TJR prostheses during the 1999– 2000periodbutstillhadpaincomplaints, which were assumed to be related to a suboptimalfit.OneofthesecondaryTJR patientsonlyrequiredamandibular com-ponent restoration to correct a subopti-mally positioned stock mandibular component(Fig.3).Accordingly,aVSP was designed that incorporated the new mandibularcomponent–includinganew zirconia neo-condyle – that matched the original 15-year-old stock fossa component.
Thesurgicalprocedureswere unevent-ful and all prostheses could be inserted according to the VSP using the routine pre-auricularandretromandibularsurgical approaches.Thisresultedinatotalof21 separate components (10 fossa and 11 mandibular) placed under guidance that were available for analysis. Figure 4
showstheintraoperativeuseofthe surgi-calguidesforpatient2.
The analysis of accuracy showed a mean Euclidean deviation of 1.10mm (SD 0.55mm, range 0.33–1.91mm) for thefossacomponentandof1.05mm(SD 0.38mm, range 0.45–1.60mm) for the mandibularcomponent.Whencombined, an overall mean Euclidean deviation of 1.07mm(SD0.46mm)wascalculatedfor all of the 21 separately placed compo-nents.Table1describesthedeviationof alloftheseparatecomponents.The inter-observervariationwasameanEuclidean
distance of0.20mm withan ICC (two-waymixed)of0.93,indicatingan excel-lentmatchingofthemeasurementsmade bythetwoobservers.
Discussion
Inthis firstprospective series ofclinical applications of the patient-specific G-TMJ-TJR,allprostheseswereplaced suc-cessfullywithhighpredictabilityandgood initial stability. Postoperative analysis showed highly accurate positioning of the G-TMJ-TJR prostheses using fixed surgical guidesas atool totranslatethe VSPtotheintraoperativesetting.
Only two prior studies have analysed the deviation between the VSP and the postoperative positions in guided place-ment of patient-specific TMJ prosthe-ses22,24. The first publication on this
subject was a study performed by the presentauthors,inwhichthe patient-spe-cificG-TMJ-TJRwasvalidatedinaseries ofhumancadavers22.Anoverallmean3D deviationof0.81mm(SD0.29mm)was observedinthecadaverseries,compared to1.07mm(SD 0.46mm)inthecurrent study. Thedifference, eventhough mar-ginal,couldberelatedtothefactthatthe positionsofthethreelandmarksforeach prosthesiscomponentwereadaptedinthe patients andplaced inthe extremitiesof theprostheses,wherethedeviationswere expected tobe greatest.Thiswas neces-sarybecausetheglasstracerspheresthat wereusedinthecadaverseriescouldnot beusedinpatients.
Thesecondpublication onthissubject wasbySembronioetal.24.Theyplaceda patient-specific VSP-based prosthesis
using surgical guides. Since the VSP, guides,andprostheseshadsimilaritiesin theirsetupandfunction,theresultsofthe deviation analysis were expected to be comparable.However,itappearsthatthey onlyusedasurgicalguidetopositionthe mandibularcomponentanditisnotclear to us whether they fixed their surgical guides to the bone prior to drilling. A mean value of the absolute deviations was not mentioned, but a scatter plot was presented instead. Based on the information provided in their paper, an approximate rangeof0.0to2.6mmand a standard deviation of approximately 0.8mm could only be guessed for their two-dimensional measurements. Their results,however,arenotdirectly compa-rabletothemeasurementspresentedinour previousstudy orthepresentstudy. The authors presented their measured devia-tionon themidsagittalplane.Therefore, onlytwooutofthreedeviationdirections were taken into account, meaning the actual 3D deviation would have been greaterthanthatdescribedintheirpaper, unlessthethirddirectionwas0mmforall of the measured points. Moreover, the measurements were performed by only one observer. Therefore, it remains unclear how user-dependent their mea-surementswere.
Thesurgicalaccuracyof patient-specif-ic TMJ-TJRs is required for a correct translationofthepreoperativelypredicted functional outcome, wear, and bio-mechanical behaviour.Discrepancies be-tween the planned and postoperative position of a TMJ-TJR prosthesis can affecttheaboveandtheproposedfunction ofthedevice4,andcausemalocclusions25,
Accuracyoffitanalysisofthepatient-specificGroningentemporomandibularjointprosthesis 543
Table1. Overviewofthepatientsincludedinthisstudy.Themeandeviationforeachseparateprosthesiscomponentplacedunderguidanceis presentedforallofthepatients.
Patient Age(years) Sex Indication Laterality
Fossapart (mm)
Mandibular part(mm)
1 56 F Revisionofstockprosthesis R NAa 0.87
2 51 F Ankylosis L 0.70 0.45
3 53 F Bilateralankylosis L 1.64 1.05
R 1.05 1.12
4 55 F Ankylosis R 1.91 1.55
5 72 F Condylarneckfracturemalunion R 1.64 1.52
6 68 F Ankylosis L 1.57 0.94
7 52 F Ankylosis R 0.33 1.60
8 59 F Revisionofstockprosthesis L 0.84 1.04
9 68 F Ankylosis L 0.83 0.63
10 56 F Ankylosis R 0.52 0.76
Mean 59 1.10 1.05
SD 0.55 0.38
Range 0.33–1.91 0.45–1.60
F,female;R,right;L,left;NA,notapplicable;SD,standarddeviation. a
resulting in unforeseen and increased loading conditions, especially when the devicedoesnotmakeuseofaspherically shapedcontactsetatboththeneo-condyle and fossa level7,14,26. Malpositioning a non-spherical fossa–condyle contact can initiate point contacts where two non-matching shapes meet and potentially couldaffectcondylar seatinginthe con-tralateral joint (Fig. 5). The occurring material stress in such point contacts can exceed the theoretically assessed values and, in turn, drastically increase UHMWPEwear27.Theflatslidingcontact of the G-TMJ-TJR neo-disc, one of the Groningen principles, ensures aconstant load-bearing surface toavoidsuch point contacts and maintain a relatively large surface area to reduce wear17. Studying the surgical accuracy provides valuable information regarding accurateboundary conditions and worst-case scenarios and can be used for biomechanical calcula-tions. When a prosthesis is accurately positionedaccordingtotheVSP, predict-ing the in situ loading becomes more accurateaswell.
The important factors for an accurate executionofaVSPusingsurgicalguides arecomprehensivepreparationand denu-dationofthebonesurfacesincontactwith theguidesandthefixationofthesurgical guidestotheboneusingscrews.Thelatter isnecessarytoensurethatplanned inter-screwrelationshipsaremaintainedduring pilot screw hole drilling, whichis espe-cially important when using locking screws.Intheabsenceofsurgicalguides,
the final prosthesis could be used as a template to drill through. However, not having any protective drill sleeves can resultindamagetotheprosthesis,thereby leavingmetalparticlesinthesurrounding tissuesandoff-centredscrews.
Knowing the surgical accuracy of the G-TMJ-TJRprocedureenablesthe worst-casepositioningscenariotobetakeninto account.Insuchacase,thepressureofthe mandibular component,throughthe neo-disc to the translationplate of the skull component,wouldbeinsufficientduetoa lack ofcontact.Withthisinmind, over-sizedbackupneo-discsareavailable dur-ing all implantations. In the case of differentTMJ-TJRdevices,theaccuracy measuredinthepresentstudycouldalso resultinnon-contactbetweentheskulland condylar componentduringocclusion or chewing.Suchagapwouldbemore com-plex, if not impossible, to correct. This also applies to stock TMJ-TJR devices wherenoVSP,surgicalguides,orspecific fit to the bone are incorporated in the surgicalprocedure.
Severalauthorshavereportedthe short-comingsofstockTMJ-TJRs.Accordingto Mercuri, slightly misplaced TMJ-TJR devicesrequireaposteriorstopatthefossa componenttopreventthecondylefrom dis-placingposteriorly28.Therefore,whena pa-tient-specific device can be positioned accuratelyusing,e.g.surgicalguides,a pos-teriorstopisnolongerrequired,savingspace towardstheauditorycanalandresultingina less bulky device. The bony anatomy of patients who require a TMJ-TJR is often
damaged, which makes fitting a stock TMJ-TJRcomplex and generally requires bone alterations. Using a patient-specific devicewillreduceintraoperativeboneloss, sinceapatient-specificfitcanbeobtainedby removinglessornobone29,whilstproviding a stable fixation. This was described by Wolfordetal.asimprovinglong-term out-comes30. Due to theaforementioned, the authors consider that the use of patient-specificTMJ-TJR devices combined with surgicalguidesbasedonaVSPshouldbe preferredatalltimes.
Thisstudyisnovelinvalidatingthetrue 3DaccuracyofplacingTMJ-TJR prosthe-sesina clinicalcohort, withconsistently good results.Currently, only 11 G-TMJ-TJRs have been placedin 10 patients and the clinicalresultsappearpromising.Bothpain scoresandmaximummouth opening im-proved when compared to preoperative measurements.However,itisstilltooearly todrawconclusionsabouttheclinicaland functionaloutcomesduetotherelatively short follow-up period of a maximum 2 years.Itishypothesizedthattheaccuracy ofpatient-specificTMJ-TJRs,asdescribed inthisstudy,willplaya keyrole in im-provedfunctionwhencomparedtostock G-TMJ-TJRs21,duetothehighlypredictable positioning protocol. The clinical results willbeinvestigatedintermsoffunctional outcomesandpainscoresinafuturestudy. The results ofthis study indicate that patient-specific Groningen TMJ-TJR prosthesescan beappliedaccuratelyand predictably in both one-stage and two-stage surgicalprocedures. Theworkflow Fig.5. Schematicexamplesoftwoprostheseswithanon-spherical/non-matchingcontactset(AandB)attheneo-condyletofossacomponent articulationsiteandtheG-TMJ-TJR(C)withaspherical/matchingcontactset.Thecross-sectionalviewsillustratetheeffectofamarginalshiftof theneo-condylarposition(A0,B0andC0)relativetotheneo-fossa,whichresultsinapointcontact(circle)inA0andB0,andincreasingmaterial stress.NotetheshiftinghasnoinfluenceonthecontactinsituationC0.
involvingaVSP,apatient-specific TMJ-TJR, andguidedsurgery,asdescribedin ourpriorcadaverseries,wasproventobe applicable and comparably accurate in patients.
Funding
Nofundingwasreceived. Competinginterests Nonetodeclare. Ethicalapproval
Approvalforthisstudywasobtainedfrom theMedicalEthicsBoardoftheUniversity Medical CenterGroningen(METc2016/ 568).
Patientconsent
All patients provided their written con-sent.
AppendixA. Supplementarydata Supplementarymaterialrelatedtothis articlecanbefound,intheonlineversion,
atdoi:https://doi.org/10.1016/j.ijom.2020.
08.012.
References
1. SidebottomAJ.Currentthinkingin tempo-romandibularjointmanagement.BrJOral MaxillofacSurg2009;47:91–4.
2. JohnsonNR,RobertsMJ,DoiSA,Batstone MD.Totaltemporomandibularjoint replace-ment prostheses: a systematic review and bias-adjustedmeta-analysis.IntJOral Max-illofacSurg2017;46:86–92.
3. vanLoonJP,OttenE,FalkenstromCH,de BontLG,VerkerkeGJ.Loadingofa unilat-eral temporomandibularjoint prosthesis:a three-dimensional mathematical study. J DentRes1998;77:1939–47.
4. vanLoonJP,FalkenstromCH,deBontLG, Verkerke GJ, Stegenga B.The theoretical optimalcenterofrotationfora temporoman-dibularjointprosthesis:athree-dimensional kinematicstudy.JDentRes1999;78:43–8.
5. Wojczynska A, Leiggener C, Bredell M, EltinD,ErniS,GalloL,ColomboV. Allo-plastictotaltemporomandibularjoint repla-cements:dotheyperformlikenaturaljoints? Prospectivecohort study with a historical control. Int J Oral Maxillofac Surg 2016;45:1213–21.
6. ElledgeR,MercuriLG,AttardA,GreenJ, SpeculandB.Reviewofemerging temporo-mandibularjointtotaljointreplacement
sys-tems.BrJOralMaxillofacSurg2019;57: 722–8.
7. WestermarkA.Totalreconstructionofthe temporomandibularjoint.Upto8yearsof follow-up of patients treatedwith Biomet ((R))totaljointprostheses.IntJOral Max-illofacSurg2010;39:951–5.
8. Wolford LM, Mercuri LG, Schneiderman ED,MovahedR,AllenW.Twenty-year fol-low-up study on a patient-fitted temporo-mandibular joint prosthesis: the Techmedica/TMJConceptsdevice. J Oral MaxillofacSurg2015;73:952–60.
9. MercuriLG,AnspachWE.Principlesforthe revisionoftotalalloplasticTMJprostheses. IntJOralMaxillofacSurg2003;32:353–9.
10. KanatasAN,NeedsC,SmithAB,JenkinsG, WorrallSF.Short-termoutcomesusingthe Christensen patient-specific temporoman-dibularjointimplantsystem:aprospective study.BrJOralMaxillofacSurg2012;50: 149–53.
11. WolfordL,MovahedR,Teschke M, Fim-mers R, Harvard D, Schneiderman E. Temporomandibular jointankylosiscanbe successfully treated with TMJ Concepts patient-fittedtotaljointprosthesisand autog-enous fat grafts. J Oral Maxillofac Surg 2016;74:1215–27.
12. HuY,ZhangL,HeD,YangC,ChenM,Zhang SY,LiH,EllisE.Simultaneoustreatmentof temporomandibularjointankylosiswith se-veremandibulardeficiencybystandardTMJ prosthesis.SciRep2017;7:45271.
13. GerbinoG,ZavatteroE,BerroneS,Ramieri G.Onestagetreatmentof temporomandibu-larjointcompletebonyankylosisusingtotal jointreplacement.JCraniomaxillofacSurg 2016;44:487–92.
14. ZhengJ,ChenX,JiangW,ZhangSY,Chen M,Yang C.Aninnovative total temporo-mandibularjointprosthesiswithcustomized designand3Dprintingadditivefabrication: a prospectiveclinicalstudy.J TranslMed 2019;17:4.
15. BaiG,HeD,YangC,ChenM,YuanJ,Wilson J.Applicationofdigitaltemplatestoguide total alloplastic joint replacement surgery with Biometstandardreplacementsystem.JOral MaxillofacSurg2014;72:2440–52.
16. OrthoBalticImplants.http://balticimplants. eu/patient-specific-medical-devices/ tmj-endoprosthesis/ [Accessibility verified April6,2020].
17. vanLoonJP,VerkerkeGJ,deVriesMP,de Bont LG. Design and wear testing of a temporomandibularjointprosthesis articula-tion.JDentRes2000;79:715–21.
18. vanLoon JP,deBont LG,SpijkervetFK, VerkerkeGJ,LiemRS.Ashort-termstudyin sheepwiththeGroningen temporomandibu-lar jointprosthesis.Int J Oral Maxillofac Surg2000;29:315–24.
19. van Loon JP, de Bont LG, Stegenga B, Spijkervet FK, Verkerke GJ. Groningen temporomandibularjointprosthesis.
Devel-opmentand firstclinical application.IntJ OralMaxillofacSurg2002;31:44–52.
20. Falkenstro¨mCH.Biomechanicaldesignofa totaltemporomandibularjointreplacement. UniversityofTwente;1993.Thesis.
21. SchuurhuisJM,DijkstraPU,StegengaB,de BontLG,SpijkervetFK.Groningen tempo-romandibulartotaljointprosthesis:an8-year longitudinalfollow-uponfunctionandpain. JCraniomaxillofacSurg2012;40:815–20.
22. KraeimaJ,MeremaBJ,WitjesMJH, Spij-kervetFKL.Developmentofa patient-spe-cific temporomandibular joint prosthesis according to the Groningen principle throughacadavertestseries.J Craniomax-illofacSurg2018;46:779–84.
23. CicchettiD.Guidelines,criteria,andrulesof thumbforevaluatingnormedand standard-ized assessmentinstrument inpsychology. PsycholAssess1994;6:284–90.
24. SembronioS,TelA,CostaF,IsolaM, Robi-onyM.Accuracyofcustom-fitted temporo-mandibular joint alloplastic reconstruction and virtual surgical planning. Int J Oral MaxillofacSurg2019;48:1077–83.
25. GerbinoG,ZavatteroE,BoscoG,BerroneS, Ramieri G. Temporomandibular joint recon-structionwithstockandcustom-madedevices: indicationsandresultsofa14-yearexperience.J CraniomaxillofacSurg2017;45:1710–5.
26. AcklandDC,MoskaljukA,HartC,VeeSin Lee P, Dimitroulis G. Prosthesis loading aftertemporomandibular jointreplacement surgery:amusculoskeletalmodelingstudy.J BiomechEng2015;137:041001.
27. van Loon JP, de Bont GM, Boering G. Evaluation of temporomandibular joint prostheses: review of theliteraturefrom1946to1994andimplications forfutureprosthesisdesigns.JOralMaxillofac Surg1995;53:984–97.
28. MercuriLG.Alloplastictemporomandibular joint replacement:rationalefortheuse of customdevices.IntJOralMaxillofacSurg 2012;41:1033–40.
29. ZhaoJ,ZouL,HeD,EllisE.Comparisonof boneadaptationaftermodificationinBiomet standard alloplastic temporomandibular joint prostheses. J Craniomaxillofac Surg 2018;46:1707–11.
30. Wolford LM, Dingwerth DJ, Talwar RM, PittaMC.Comparisonof2 temporomandib-ular jointtotal jointprosthesis systems.J OralMaxillofacSurg2003;61:685–90. dis-cussion690.
Address: F.K.L.Spijkervet
DepartmentofOralandMaxillofacial Surgery
UniversityMedicalCenterGroningen Hanzeplein1 POBox30.001 9700RBGroningen TheNetherlands Tel.:+31503613840 E-mail:f.k.l.spijkervet@umcg.nl
