Locking plate and fibular allograft augmentation in unstable fractures of proximal humerus

(1)

Locking

plate

and

ﬁbular

allograft

augmentation

in

unstable

fractures

of

proximal

humerus

Fabrizio

Matassi

a,

*

,

Renzo

Angeloni

b

,

Christian

Carulli

a

,

Roberto

Civinini

a

,

Leonardo

Di

Bella

b

,

Birgit

Redl

a

_,

_Massimo

_Innocenti

a

FirstOrthopedicClinic,DepartmentofSpecialSurgicalScience,UniversityofFlorence,CTO-LargoPalagi1,50139Florence,Italy

b

3rdGeneralOrthopedicUnit–AziendaOspedalieraCareggi,CTO–LargoPalagi1,50139Florence,Italy

Introduction

Displacedproximalhumeralfracturesremaindifficulttotreat, mainly because achieving stable fixation that maintains intra-operativereductionisunpredictable.Lockingplatefixationhasthe potentialtoprovidegreaterfixationintheproximalhumerusthan standard internalfixation techniques,offering agreater loadto failureandtherequirementforlesssoft-tissuedissectionalongthe humeralshaft.1_However,_some_studies_show_variable_results_with highrates ofscrewperforationofthearticularsurfaceorvarus collapse of the fracture, especially in osteoporotic bone or in fractureswithmedialmetaphysealcomminution.2–8

Adequatemechanicalsupportofthemedialcolumnmaybe obtainedbyachievingananatomicallystablereductionwitha medialcorticalcontactor,inthecaseofmedialcomminution, by placing a superiorlydirected obliquelockingscrew inthe inferomedial regionof theproximalfragment.Indeed,studies showthatfracturestreatedwith eitheranatomicreductionor screwsintheinferomedialhumeralheadforwhichnomedial column support was obtained had a high incidence of failure.3,9–12

Anintramedullaryﬁbularallograftusedtogetherwithalocking plateﬁxationhasbeenrecentlydescribedinbiomechanicalstudies toprovideadditionalmedialsupportandpreventvarus malalign-ment.13

The purpose of this study was to evaluatethe clinical and radiographic outcomesof a locking platewith ﬁbular allograft augmentation in unstable humeral fractures. We hypothesised thataﬁbularstrutgraftasanendostealimplantwouldbeasafe

Injury,Int.J.CareInjuredxxx(2012)xxx–xxx

*Correspondingauthor.Tel.:+390554376841;fax:+39055432145. E-mailaddress:fabriziomatassi@gmail.com(F.Matassi).

ARTICLE INFO Articlehistory: Accepted31July2012 Keywords:

Proximalhumeralfracture Fibularallograft Lockingplate Medialsupport Endostealimplant

SUMMARY

Introduction: Theuseofanintramedullaryfibularallografttogetherwithalockingplatefixationhasbeen recentlydescribedinbiomechanicalstudiestoprovideadditionalmedialsupportandpreventvarus malalignmentfordisplacedproximalhumeralfractureswithpromisingresults.Theaimofthisstudywas toevaluatetheclinicalandradiographicoutcomeofalockingplatewithfibularallograftaugmentationin unstablehumeralfractures.

Methods: Weprospectivelyassessedthefunctionaloutcome andcomplications in17patientswith proximalhumeralfractureswithdisruptedmedialcolumn,treatedwithalockingplateandaﬁbularstrut graft.Themedianpatientagewas62years.Postoperativeassessmentsincludedradiographicimaging, rangeofmotion,painaccordingtothevisualanaloguescale(VAS),ShortForm(SF36)HealthSurvey, Constant-MurleyandDisabilitiesoftheArm,ShoulderandHand(DASH)shoulderscoresaswellasreturn topreviousoccupationandcomplications.

Results: Nopatientswerelosttofollow-upandnomajorcomplicationswererecorded.Therewasno collapseofthehumeralheadmorethan2mm,osteonecrosisorscrewpenetration ofthearticular surface.Allfractureshealedclinicallyandradiographically.Afteranaveragefollow-upof13months,the meanConstantscorewas79points.Themeanactiveﬂexionwas1498;extension,478;internalrotation, 408;externalrotation,658;andabduction,1358.ThemedianVASpainlevelwas1point.Themedian DASHscorewas33points,andthemedianSF36was83points.

Conclusion: Lockingplatewithﬁbulargraftaugmentationisasafeandpromisingtechniquetosupport thehumeralheadandmaintainreductionintheproximalhumeralfracturewithmedialcomminution. ß2012ElsevierLtd.Allrightsreserved.

GModel

JINJ-5080;No.ofPages4

Pleasecitethisarticleinpressas:MatassiF,etal.Lockingplateandﬁbularallograftaugmentationinunstablefracturesofproximal humerus.Injury(2012),http://dx.doi.org/10.1016/j.injury.2012.08.004

ContentslistsavailableatSciVerseScienceDirect

Injury

j ou rna l h ome p a ge : w ww . e l se v i e r. co m/ l oc a te / i n j ury

0020–1383/$–seefrontmatterß2012ElsevierLtd.Allrightsreserved.

(2)

andvaluableoptiontomaintainreductioninproximalhumeral fractureswithmetaphysealcomminution.

Materialsandmethods

We prospectively followed a selected series of 17 patients (7 men and 10 women) with unilateral displaced proximal humeralfracturestreatedat ourinstitutionbetweenApril2008 andMay2010bylockingplateﬁxationwithintramedullaryﬁbular strutgraftaugmentation.

Inclusion criteriaintothestudywerecloseddisplaced three-andfour-partproximalhumeralfracturesinadultswithdisrupted medial hinge and signiﬁcant metaphyseal comminution or insufﬁcientosseouscontact.Openfractures,pathologicalfractures, unreconstructableheadand/ortuberosityfragmentsandtwo-part fractureswereexcludedfromthestudy.Themedian ageof the patientsatthetimeoftheoperationwas62years(range,54–73 years).Themechanismsofinjuryincludedafallwhilewalking(13 patients),motorcycleaccident(2),asports-relatedaccident(1)and afallfromanelevatedheight(1).

Preoperative evaluation includedplain antero-posterior (AP) andlateralX-rays aswellasacomputedtomography(CT)scan usedtoevaluatedisplacement.TherewerefourtypeA,fivetypeB andeighttypeCfractures,accordingtotheArbeitsgemeinschaft fu¨r Osteosynthesefragen (AO) classification.14 _Eleven _fractures were3-partandsixfractureswere4-part,accordingtotheNeer classification.15

Surgicaltechnique

Allprocedureswereperformedbyoneoftheauthors(RA).A complete description of this technique has been published previously.7 _Patients _were _placed _in _the _‘beach _chair’ _position on a radiolucent operating table under general anaesthesia. A deltopectoralapproachwasperformedfortheﬁrstsevencasesand ananterolateral deltoid-splitting for theremaining cases. Non-absorbablesutures(No.5)wereplacedintherotatorcufftendons toallowtractionandcontrolofthetuberosityandthehumeral headfragment.Two2.0-mmthreadedK-wireswereplacedinthe cancellous surface of the humeral head fragment acting as a ‘joystick’forinitialreduction.Inallcases,theﬁbularallograftwas usedasamedialstrutaugment.

Fibularallograftwascutwithasagittaloscillatingsawonthe backtabletoanappropriatelength.Theﬁbulawasinsertedinto thecanalthrough thelateralfracturelines and pusheddistally beyondthelevelofthehead.Itwasthenmedialisedmaximallyto the calcar region to indirectly reduce the medial column and advancedretrograde intothesubchondralbonetoliftthehead superiorly.

Thelockingplate(forninecases:ProximalHumerusInternal LockingSystemPHILOS, Synthes,Mezzovico-Vira, Switzerland; foreight cases: PeriarticularLocked Plating System PERI-LOC, Smith&Nephew,Memphis,TN, USA)wasthensecuredtothe lateralhumeralshaftdistal tothefibula. Ifthemedialcolumn reductioncouldnot berestoredandcomminutionexisted,the fibularallograftwasusedasanindirectreductiontool.Ascrew pushedthefibulamediallyuntilitapposedthemedialcortexof the humerus, and indirectly reduced the medial column. If anatomicalreductionwasachieved,lockingscrewswereplaced throughthefibulaintothehumeralheadandshafttosecureits position.Meticulousrepairoftherotatorcuffwascarriedout,and thewoundwasclosedwithadrainplacedundernegativesuction, whichwas removed after 48h.Prophylactic intravenous anti-bioticswereadministereduntilthedrainwasremoved,usually after24or48h.Postoperatively,thearmwassupportedinasling. Pendularmovementswerestartedfromthefirst postoperative

day and the shoulder was mobilised with passive assisted exercises,followedbyactiveexercisesafter3weeks.

Clinicalandradiographicfollow-upwasperformedat1month, 2months,6months,12monthsand2years,postoperatively.The minimumfollow-upwas13months(average,28months;range, 13–56months).

Clinical examinationincludedrange ofmotion (ROM) ofthe affectedextremitymadewithagoniometer;andclassiﬁcationof pain,accordingtothevisualanaloguescale(VAS),from0to10;and theShortForm36(SF36)survey.16_We_also_analysed_the_Constant– Murley shoulder scores for the operated and non-operated shouldersandtheDisabilitiesofArm,ShoulderandHand(DASH) shoulderscore.17,18_Return_to_previous_{activities/occupation}_was alsorecorded.

Radiologicfollow-upincludedAPandlateralplainX-raysand measurementoftheanglebetweenthehumeralheadandtheshaft intheAPviewafterhealing.Accordingtotherecommendationsof Jiangetal.,lessthan58varusorvalgusalignmentintheAPviewwere considered anatomic.4 _The _‘humeral _head _height’ _relative _to theplatewasmeasuredforeachradiograph,bothinitiallyandat theﬁnalfollow-up,whichallowedforsubsequentanalysisoflossof reduction.Thismeasurementwasdonebydrawingtwolines,both runningperpendiculartotheshaftoftheplate;onewasplacedatthe topedgeoftheplateandtheotherwasplacedatthesuperioredgeof thehumeralhead,andthedistancebetweenthesetwolineswas measured and designatedas the headheight.3 _A _change _in _the humeralheadheightgreaterthan3mmwasconsideredtobealoss ofreduction.Complicationswererecorded,suchascollapseofthe fracture,screwpenetration,avascularnecrosis,peri-implant frac-ture,infectionandneedforfurthersurgery.

Results

No patients were lost to follow-up. All fractures healed clinically and radiologically. The mean Constant score at ﬁnal follow-up was 79 points (range, 56–100 points), representing 83.7% of the median Constant–Murley score of the unaffected shoulder.Themedianrangeofshouldermotionwasﬂexion,1498 (range,93–1788);extension,478(range,40–528);internalrotation, 408(range,25–708);external rotation,658(range,45–908);and abduction,1358(range,80–1808).ThemedianVASpainlevelwas1 point(range,0–2points).ThemedianDASHscorewas33points (range,22–52points),andthemedianSF36was83points(range, 67–97 points). Fifteen (88.2%) patients returned to previous activities or occupations.Two patients experienced restrictions toactivitiesorassumedadifferentoccupation.

Alignment in the AP view was anatomic in 16 patients; however,slight (98)varus alignment wasnoted inone patient. Therewasnocollapseofthehumeralheadorscrewpenetrationof thearticularsurfaceforanyofthepatients.Thechangeinhumeral headheightbetweenpostoperativeandthefinalfollow-upX-rays averaged0.3mm(maximum,1.7mm).Medialcorticalcontinuity was restored in all cases and the fibular allograft showed progressiveincorporationintotheproximalhumerus(Figs.1–4). Of the minor complications, one patient had a superficial infectionduringtheearlypostoperativeperiod.Thepatientwas placed on oralantibiotics for 2weeks andthe infectionhealed withoutcomplication.Noneofthepatientspresentedwithanyof the major complications, such as subacromial impingement, neurovascular injury, loosening of implant or osteonecrosis of thehumeralhead.

Discussion

The purpose of this study is to analyse the clinical and radiographic outcome of a new technique to treat unstable

F.Matassietal./Injury,Int.J.CareInjuredxxx(2012)xxx–xxx 2

GModel

(3)

humeral fractures with locking plate and ﬁbular allograft augmentation.We foundrecoveryof goodclinicalresultsinall patients,nocollapsesofhumeralheadorscrewpenetrationofthe articularsurface and healingof thefracture withno changein humeralheadheightbetweenX-raystakenpostoperativelyandat theﬁnalfollow-up.

Thefirstclinicalexperienceofthistechniquetobereported intheliteraturebyGardneretal. showedencouragingresults, with7outof7fractureshealingwithoutalossofreductionor fixationstability.13 Later, Neviaseret al.reported lowratesof reduction loss (2.6%), screw cut-out (0%) and osteonecrosis (2.6%),aswellashighclinicaloutcomescoresinaseries of38 patientswithdisplacedproximalhumeralfracturestreatedwith lockingplatefixationandanendostealstrutaugment.19 Biome-chanicaltesting showedthatmedialsupport withan intrame-dullaryfibular graft andangular stable fixation increasedthe overall stiffness of the bone-implant construct and reduced migrationofthehumeralheadfragment,ascomparedwiththe lockingplate alone.20–23

In cadaveric specimens, Chow et al. showed how ﬁbular allograftaugmentationcouldincreasethestrengthofthelocking platetowithstandrepetitivevarusloading.Noneoftheaugmented

constructs failed prior to25,000 cycles, while 6 of the 8 non-augmented constructs failed at an average of 6604 cycles.21 Furthermore, tests under increasing loads showed that the maximum failure loads and stiffness in the locking platewith intramedullaryfibulargraftsweresignificantlyhigherthanthose inthelockingplatealone.20_Osterhoff_et_al._recently_tested_a_similar constructinsyntheticboneofaproximalhumeruslockingplate augmentedwithanintramedullaryfibularallograftstrut. Signifi-cantlylowerintercyclicfragmentmotion,loweroverallfragment motionandalowerresidualgap-distancedeformationafter400 cyclesofloadingwasrecordedforthegroupwiththefibulargraft whencomparedwiththeconventionaltechnique.Theaugmented construct was also stiffer and had a higher ultimate load to failure.23

Thelimitationsofourstudyareasfollows.Firstistheuseoftwo different surgical approaches and two types of locking plate.

Fig.3.IntraoperativeanteroposteriorXrayfollowingopenreductionandinternal ﬁxation with locking plate augmented with ﬁbular graftwith restorationof humeralhead-shaftangle.

Fig.4.Anteroposteriorradiographofshoulderat2yearspost-opshowshealingof thefracture,nochangeinhumeralheadheight.

Fig.1.AnteroposteriorX-rayprojectionoftheshoulderina68-year-oldfemalewith Neerfour-partfracture.

Fig.2.CTscanshowsmarkedrotationandvarusdislocationofhumeralhead.

GModel

(4)

However,webelievethatthiscouldnotbeaconfoundingvariable for the primary purpose of our study that was to study the mechanical support of a ﬁbular graft in maintaining proximal humeralfracturereduction.Second,wehadnocontrolgroupof similar fractures treated with alternative methods and cannot makedirectcomparisonsofothertreatmentmethods.

Ourfindingsencouragetheuseoffibularallograftsandlocking plates for the treatment of proximal humerus fractures with medialcomminutiontorestoretheintegrityofthemedialcolumn, supportthehumeralheadandmaintainreductionuntilfracture healing.Furthermore,thefibulaallograftwasusefulinreductionof the fracture as an indirect tool and at the same time as a mechanicalsupportforthehumeralhead.

Conclusion

Lockingplatewithﬁbulargraftaugmentationisareliableand promisingtechniquetosupportthehumeralheadandmaintain reductionin the treatment of proximalhumeral fractures with medial comminution. We advise the use of this technique in proximalhumeralfractureswithmedialcomminutiontorestore theintegrityofthemedialcolumn,supportthehumeralheadand maintain reduction until fracture healing. This technique may minimisethemostfrequentcomplicationsreportedwithproximal humeral locking plates, and allow for early and aggressive physiotherapy.

Conﬂictofinterest

Theauthors,theirimmediatefamilyandanyresearch founda-tionwithwhichtheyareaffiliatedhavenotreceivedanyfinancial paymentsorotherbenefitsfromanycommercialentityrelatedto thesubjectofthisarticle.Theauthorsdeclarethatthereareno outsidefundingorgrantsreceivedforthisstudy.

References

1.WalshS,ReindlR,HarveyE,BerryG,BeckmanL,SteffenT.Biomechanical comparisonofa uniquelockingplateversusa standardplateforinternal ﬁxationofproximalhumerusfracturesinacadavericmodel.Clinical Biome-chanics(BristolAvon)2006;21:1027–31.

2.BrunnerF,SommerC,BahrsC,HeuwinkelR,HafnerC,RillmannP,etal.Open reductionandinternalﬁxationofproximalhumerusfracturesusingaproximal humerallockedplate:aprospectivemulticenteranalysis.JournalofOrthopaedic Trauma2009;23:163–72.

3.GardnerMJ,WeilY,BarkerJU,KellyBT,HelfetDL,LorichDG.Theimportanceof medialsupportinlockedplatingofproximalhumerusfractures.Journalof OrthopaedicTrauma2007;21:185–91.

4.JiangR,LuoCF, ZengBF,MeiGH.Minimallyinvasiveplatingforcomplex humeral shaft fractures. Archives of Orthopaedic and Trauma Surgery 2007;127:531–5.

5.OwsleyKC,GorczycaJT.Displacement/screwcutoutafteropenreductionand lockedplateﬁxationofhumeralfractures.JournalofBoneandJointSurgery 2008;90:233–40.

6.EgolKA,OngCC,WalshM,JazrawiLM,Tejwani NC,Zuckerman JD.Early complicationsinproximalhumerusfractures(OTATypes11)treatedwith lockedplates.JournalofOrthopaedicTrauma2008;22:159–64.

7.KoukakisA,ApostolouCD,TanejaT,KorresDS,AminiA.Fixationofproximal humerusfracturesusingthePHILOSplate:earlyexperience.Clinical Orthopae-dicsandRelatedResearch2006;442:115–20.

8.RicchettiET,WarrenderWJ,AbboudJA.Useoflockingplatesinthetreatmentof proximalhumerusfractures.JournalofShoulderandElbowSurgery2010;19(2 Suppl.):66–75.

9.KrappingerD,BizzottoN,RiedmannS,KammerlanderC,HenggC,KralingerFS. Predictingfailureaftersurgicalﬁxationofproximalhumerusfractures.Injury 2011;42(11):1283–8.

10.LeeCW,ShinSS.Prognosticfactorsforunstableproximalhumeralfractures treated withlocking-plate ﬁxation.Journal of Shoulder and ElbowSurgery 2009;18:83–8.

11.HardemanF,BollarsP,DonnellyM,BellemansJ,NijsS.Predictivefactorsfor functionaloutcomeandfailureinangularstableosteosynthesisoftheproximal humerus.Injury2012;43(2):153–8.

12.LescheidJ,ZderoR,ShahS,KuzykPR,SchemitschEH.Thebiomechanicsof lockedplatingforrepairingproximalhumerusfractureswithorwithoutmedial corticalsupport.JournalofTrauma2010;69:1235–42.

13.GardnerM,BoraiahS,HelfetD,LorichD.Indirectmedialreductionandstrut supportofproximalhumerusfracturesusinganendostealimplant.Journalof OrthopaedicTrauma2008;22:195–200.

14.MullerME.Thecomprehensiveclassiﬁcationoffracturesoflongbones.Berlin, Germany:Springer-Verlag;1990.

15.Neer2nd.Displacedproximalhumeralfractures.I.Classiﬁcationand evalua-tion.JournalofBoneandJointSurgery1970;52:1077–89.

16.BullingerM,KirchbergerI.SF36questionnaireconcerninghealthstatus. Gottin-gen,Bern,Toronto,Seattle:Hogrefe;1998.

17.ConstantCR,MurleyAH.Aclinicalmethodoffunctionalassessmentofthe shoulder.ClinicalOrthopaedicsandRelatedResearch1987;214:160–4. 18.HudakPL,AmadioPC,BombardierC,TheUpperExtremityCollaborativeGroup

(UECG).Development ofanupperextremityoutcomemeasure:theDASH (disabilitiesofthearm,shoulderandhand).AmericanJournalofIndustrial Medicine1996;29:602–8.

19.NeviaserAS,HettrichCM,BeamerBS,DinesJS,LorichDG.Endostealstrut augmentreducescomplications associatedwithproximal humerallocking plates.ClinicalOrthopaedicsandRelatedResearch2011;469:3300–6. 20.Bae JH,OhJK,ChonCS, OhCW,HwangJH,YoonYC. Thebiomechanical

performanceoflockingplateﬁxationwithintramedullaryﬁbularstrutgraft augmentationinthetreatmentofunstablefracturesoftheproximalhumerus. JournalofBoneandJointSurgery–BritishVolume2011;98(7):937–41. 21.ChowRM,BegumF,BeaupreLA,CareyJP,AdeebS,BoulianeMJ.Proximal

humeralfractureﬁxation:lockingplateconstructintramedullaryﬁbular allo-graft.JournalofShoulderandElbowSurgery2011;20:1–8.

22.Mathison C, ChaudharyR, Beaupre L, Reynolds M,AdeebS, Bouliane M. Biomechanicalanalysisofproximalhumeralfixationusinglockingplate fixa-tionwithanintramedullaryfibularallograft.ClinicalBiomechanics2010;25: 642–6.

23.OsterhoffG,BaumgartnerD,FavreP,WannerGA,GerberH,SimmenHP,etal. Medialsupportbyﬁbulabonegraftinangularstableplateﬁxationofproximal humeralfractures:aninvitrostudywithsyntheticbone.JournalofShoulderand ElbowSurgery2011;20:740–6.

GModel