先天性髋关节脱位

一、正常髋关节（I型）的特点及治疗特点：Ⅰ型髋关节发育正常，其中又分Ia型和Ib型。Ia型β角小于55度，Ib型β角在55-77度，两者α角都大于60度。髋臼角30°，α角60°，β角55°。髋臼对股骨头有良好的覆盖，骨顶外缘锐利成角（Ⅰa型）或轻度变钝（Ⅰb型），臼顶软骨狭窄呈三角形1。治疗：对于Ⅰ型髋关节，通常不需要特殊治疗，只需定期进行复查，观察髋关节的发育情况。因为这是正常的髋关节发育状态，在日常的生长发育过程中，正常的活动和生活不会对其产生不良影响。但家长也需要关注孩子的一些日常行为，例如孩子的活动是否自如等，若发现异常情况应及时就医进一步检查。二、髋关节发育稍差（Ⅱ型）的特点与治疗特点：Ⅱ型髋关节发育稍差，分为Ⅱa、Ⅱb、Ⅱc、Ⅱd四个亚型。Ⅱa型指出生12周以内婴儿，α角为50°-59°，β角小于55度；Ⅱb型指12周以上婴儿髋，α角50°-59°，β角55°；Ⅱc型α角43°-49°，β角小于77度；Ⅱd型α角同Ⅱc型，β角大于77度。髋臼角在30°-43°之间，α角55°-60°，β角55°。该型髋的骨性髋臼外上缘缺少钙化，骨顶轮廓发育缺陷的部分由软骨顶增宽充填，覆盖在股骨头上25。治疗：对于Ⅱ型髋关节，尤其是Ⅱa和Ⅱb型，因为处于髋关节发育的临界状态或者骨化延迟状态，在早期通常采取观察和定期复查的策略。如果没有进一步的恶化或者有好转的趋势，可以继续观察。而对于Ⅱc和Ⅱd型髋臼发育不良的情况，可采用软式支具矫治，如Pavlik吊带等。通过支具将髋关节保持在合适的位置，促进髋臼的正常发育。在使用支具期间，需要定期进行超声检查来评估治疗效果，并且要注意支具的佩戴是否合适，避免对孩子的皮肤等造成损伤。三、髋关节发育不良（Ⅲ型）的特点与治疗特点：Ⅲ型髋关节发育不良，分为Ⅲa和Ⅲb型。Ⅲ型髋关节髋臼角43°-55°，α角40°-55°，β角55°-77°。两类髋中股骨头均向上外方脱位，Ⅲa软骨顶为无回声结构，是透明软骨成分，Ⅲb的软骨顶可见有程度不同回声，说明透明软骨可能发生了纤维化或变性改变。在单一冠状声像上，脱位的股骨头和髋臼不能同时完整显现，参考测量点难以辨认5。治疗：对于Ⅲ型髋关节发育不良，早期（6-7个月以下）可采用保守治疗，如使用Pavlik吊带、支具等。从经验和文献的报道来说，这种保守治疗对于这个年龄段的孩子成功率可以达到90%以上。在使用支具或吊带期间，需要密切关注孩子髋关节的复位情况，定期进行超声检查。如果保守治疗不成功，可以改用麻醉下石膏复位。经过复位的关节对髋臼的刺激，局部发育就会逐步的改善。同时，在治疗过程中，还需要关注孩子下肢的血液循环、神经功能等情况，避免因治疗导致其他并发症的出现。四、髋关节脱位（Ⅳ型）的特点与治疗特点：Ⅳ型髋关节脱位，髋臼角55°，α角40°，β角77°。此型髋的声学特点是股骨头脱位，表面只有薄层关节囊覆盖，髋臼唇盂和软骨顶也向原始髋臼的内下方移位。在B超下α角甚至测量不出，属于髋关节高位脱位的情况212。治疗：对于Ⅳ型髋关节脱位，因为脱位情况较为严重，一般需要积极治疗。对于6-7个月以下的婴儿，如果保守治疗（如Pavlik吊带等）效果不佳，可考虑麻醉下石膏复位。而对于年龄较大的孩子（7-18个月），可能需要进行麻醉下闭合复位或者切开复位，然后用石膏固定。在复位过程中，要确保股骨头准确复位到髋臼内，并且要关注复位后髋关节的稳定性。在术后，需要长期的康复随访，观察髋关节的发育情况以及孩子的下肢功能恢复情况，预防股骨头坏死、髋关节僵硬等并发症的发生。五、髋关节半脱位（Ⅴ型）的特点与治疗特点：Ⅴ型髋关节半脱位，髋臼角55°，α角40°，β角77°-90°。髋关节处于半脱位状态，股骨头部分脱离髋臼，关节的稳定性受到影响，髋臼对股骨头的覆盖不完全，这会影响髋关节的正常发育和功能1。治疗：对于Ⅴ型髋关节半脱位的治疗与Ⅲ型髋关节发育不良类似。在早期可尝试保守治疗，如使用Pavlik吊带或支具，并且定期复查超声，观察髋关节复位情况。如果保守治疗失败，根据孩子的年龄可选择麻醉下石膏复位（6-7个月以下）或者其他更复杂的复位手术（年龄较大时）。在整个治疗过程中，同样要关注髋关节的功能恢复和发育情况，避免出现并发症。

临界髋关节发育不良BorderlineDDH(5)：MRI上的股骨骨骺髋臼顶指数（FEAR）测量是否预测临界髋关节发育不良的不稳定性？作者：CécileBatailler,JanWeidner,MichaelWyatt,DominikPfluger,MartinBeck作者单位:CHULyonCroix-Rousse,HospicesCivilsdeLyon,Lyon,France.译者：陶可（北京大学人民医院骨关节科）摘要目的：临界髋关节发育不良型既可以表现为稳定，也可以表现为不稳定，这使得手术决策具有挑战性。虽然不稳定髋关节最好通过髋臼重新定位来治疗，但稳定的髋关节可以通过关节镜治疗。几个成像参数可以帮助确定适当的治疗方法，包括在普通X线片上测量的股骨-骨骺髋臼顶(FEAR)指数。本研究的目的是评估MRI上FEAR指数与其放射学测量相比的可靠性和敏感性。患者和方法：定义了在MRI上测量FEAR指数的技术并验证了其可靠性。一项回顾性研究评估了三组20名患者：一组不稳定的“临界发育不良型髋关节”，其外侧中心边缘角(LCEA)小于25°，通过髋臼周围截骨术成功治疗；一组稳定的“临界发育不良型髋关节”，其LCEA小于25°，通过撞击手术成功治疗；另一组无症状对照组，其LCEA在25°至35°之间。在标准X线片和MRI上均进行了以下测量：LCEA、髋臼指数、股骨前倾角和FEAR指数。结果：FEAR指数在MRI和X线片上均表现出极好的观察者内和观察者间可靠性。X线片上的FEAR指数比MRI上的更可靠。与不稳定临界组（平均7.9°（标准差6.8°））相比，稳定临界组的MRI上的FEAR指数较低。在FEAR指数截止值为2°的情况下，使用放射学FEAR指数可以正确识别90%的患者为稳定或不稳定，而使用MRI上的FEAR指数则为82.5%。与MRI相比，FEAR指数在普通X线片上更能预测不稳定性。结论：MRI上测量的FEAR指数比X线片上测量的FEAR指数更不可靠，灵敏度也更低。放射学FEAR指数的2°临界值预测髋关节稳定性的概率为90%。讨论本研究的主要结果是，在普通骨盆X线片上测量FEAR指数比在MRI上具有更高的灵敏度和可靠性。当LCEA在20°和25°之间时，区分稳定和不稳定髋关节可能具有挑战性。决定采用哪种手术治疗主要取决于髋关节的稳定性或不稳定性。通过新的影像学征象来改善诊断不仅有意义，而且必不可少。随着关节镜实践的进步，在同侧关节镜髋关节手术失败后，随后接受PAO截骨以矫正症状性髋关节发育不良的患者比例从2008年到2015年增加了1.92倍。临界髋关节发育不良患者的手术选择包括关节镜盂唇修复和关节囊闭合/折叠，有时还伴有凸轮切除术，或伴有或不伴有盂唇修复/凸轮切除术的PAO截骨术。治疗方案的选择受到多种因素的影响，包括患者年龄、患者偏好，尤其是外科医生是否认为髋关节不稳定。在骨性结构覆盖不足的情况下，进行髋关节镜检查的风险是因不稳定、髋关节脱位或关节炎进展而导致的持续疼痛。这必须与PAO截骨并发症的风险相平衡，即使是由经验丰富的外科医生进行手术，轻度发育不良患者的临床改善效果也会较低。有报道称，临界髋关节的保髋手术失败，因为确定不稳定或稳定的髋关节可能很困难，并可能导致治疗不当。如果在髋关节发育不良的同时发现潜在的FAI，那么不适当的撞击手术，不注意发育不良，可能会增加不稳定的症状。需要制定严格的患者选择标准，才能考虑对临界髋关节发育不良进行髋关节镜检查。最近的一项研究回顾了1368例髋臼发育不良髋关节镜检查，发现在某些明确和选定的临界髋关节发育不良病例中，可以考虑单独使用髋关节镜检查，但必须仔细注意保存盂唇和关节囊。然而，许多研究表明，如果发育不良为中度或重度，且患者选择不太严格，则临床结果不佳，并且存在医源性不稳定的风险。髋关节的稳定性通常使用LCEA进行评估；然而，仅凭这种测量不足以评估临界髋关节，需要使用其他评估参数。一些放射学或MRI参数直接由股骨头移位引起，因此高度提示髋关节不稳定，例如髂坐线距离增加、Shenton线断裂或后下关节间隙存在钆（髋关节造影检查）。不稳定髋关节可能存在其他放射学或MRI征象，但预测性较差，例如LCEA介于20°和25°之间、AI大于10°、髋外翻、股骨前倾或盂唇体积增大。其他参数可评估髋关节发育不良的严重程度，尤其是全髋关节置换术的实施。Wyatt等描述的放射学FEAR指数表示髋关节生长过程中股骨近端骨骺板的合力。该指数反映了髋关节在生长过程中的功能行为（稳定或不稳定）。Wyatt等报道称，如果患者出现髋关节疼痛和临界发育不良（定义为LCEA20°至25°），则放射学FEAR指数小于5°表示髋关节稳定的可能性为80%。FEAR指数的可靠性非常好，至少与LCEA或AI的可靠性一样好。FEAR指数优于LCEA的原因可能是难以定义髋臼源边缘，正如已经提到的。在本研究中，放射学FEAR指数的截止值为2°可预测稳定性，90%的患者被正确判断为稳定或不稳定。随着这种测量方法在临界髋关节发育不良的X线片上的应用越来越广泛，正常值和截止值的定义也越来越明确。有时，FEAR指数很难在X线片上测量。事实上，骨骺瘢痕的界限有时很难确定，并可能导致测量误差。因此，我们评估了FEAR指数的可靠性，并由两名独立观察员进行评估。在我们的研究和Wyatt等的研究中，X线片和MRIFEAR指数表现出极好的观察者间和观察者内可靠性，并且被证明优于AI或LCEA。FEAR指数的假阳性或阴性率主要是由于在某些情况下难以进行测量。MRI是一种非常准确的检查，尤其是带有放射状切口的MR关节造影，这是目前评估原生髋关节关节内病变的最佳术前成像研究。骨骺瘢痕在MRI上很容易被看到，这似乎是提高该参数灵敏度的有希望的途径。然而，我们的研究表明，MRI的可靠性和灵敏度低于普通的放射学测量。可以提出几种解释。在X线片上，骺板瘢痕代表每个额叶切片的不同骺板瘢痕的平均值，因此也代表生长过程中作用于整个股骨头的力量平衡的平均值。对于MRI上的FEAR指数，为了获得一个简单可靠的指数，我们选择仅在一个图像切片上测量FEAR指数。增加测量次数会增加出错的风险。与静态X线片相比，3DMRI重建的准确性会降低。然而，根据所选的额叶切片，骺板瘢痕的形状在各个切片之间可能有所不同，并且骺板瘢痕的方向可能会发生几度的变化。髋臼也可能出现同样的问题，因为它可能会根据与髋臼窝的距离而发生显著变化。此外，Wyatt等指出，FEAR指数代表髋关节的力量平衡。仅使用股骨头中部骨骺瘢痕的轴线主要考虑股骨头上部和顶部之间施加的力量。如果患者因不稳定而出现股骨头轻微半脱位，则矢状图上位于髋臼12点钟位置的切片与位于股骨头12点钟位置的切片不同（图4）。因此，骨骺瘢痕中央部分的测量可能不正确。股骨旋转或外展可能引起的变化与X线片相同。我们的研究有一些局限性。首先，纳入的患者数量很少，无法匹配某些参数（LCEA、性别）。尽管如此，有症状的临界髋关节患者并不常见，需要进行完整的影像学检查才能比较X线片和MRI。不稳定是由多种因素和手术后症状的发展决定的。因此，该诊断是根据参数关联进行的，并构成了不稳定性的功能定义。这项研究是回顾性的，因此我们可以显示关联但不能做出预测。此外，没有评估某些参数，例如全身韧带松弛、肌肉调节或体重指数。两年的随访似乎太短了。然而，当康复没有进展时，通常会在3到6个月内怀疑结果不佳。相反，1到1年半后，髋关节通常在功能和疼痛方面达到稳定状态。因此，两年的限制似乎是合理的，尽管可能更倾向于更长的随访时间。总之，MRI上测量的FEAR指数不如X线片上测量的FEAR指数可靠。此外，与MRI相比，FEA指数是X线片上不稳定性的更好的预测指标。放射学FEAR指数的2°截止值预测髋关节稳定性的概率为90%。放射学FEAR指数是辅助复杂髋关节不稳定诊断的可靠参数。需要前瞻性评估该测量值以预测渐进性不稳定。 Fig1a)Anteroposteriorpelvicradiographandb)hipMRIinfrontalandsagittalviewswithapositiveFemoro-EpiphysealAcetabularRoof(FEAR)index(anglebetweenthephysealscarandthesclerosisofthesourcil)forapatientwithaborderlinedysplastichip(lateralcentre-edgeangle=17.2°,acetabularindex=12.6°)treatedwithperiacetabularosteotomy.图1 a)前后位骨盆X线片和b)髋关节正位和矢状位MRI，股骨-骨骺髋臼顶(FEAR)指数（骨骺瘢痕与髋臼硬化之间的角度）为阳性，适用于接受髋臼周围截骨术治疗的临界髋关节发育不良患者（外侧中心边缘角=17.2°，髋臼指数=12.6°）。Fig2a)Anteroposteriorpelvicradiographandb)hipMRIinfrontalandsagittalviewswithanegativeFemoro-EpiphysealAcetabularRoof(FEAR)index(anglebetweenthephysealscarandthesclerosisofthesourcil)forapatientwithaborderlinedysplastichip(lateralcentre-edgeangle=24°,acetabularindex=9°)treatedwithimpingementsurgery.图2 a)前后位骨盆X线片和b)髋关节正面和矢状面MRI，股骨-骨骺髋臼顶(FEAR)指数（骨骺瘢痕与髋臼硬化之间的角度）为阴性，适用于接受撞击手术治疗的临界髋关节发育不良患者（外侧中心边缘角=24°，髋臼指数=9°）。Fig3Boxplotsofa)theradiologicalFemoro-EpiphysealAcetabularRoof(FEAR)indexandb)theFEARindexonMRI,comparingallthreegroups.Theboxplotsareconstitutedofseveraldata.Themedian(middlequartile)marksthemid-pointofthedataandisshownbythelinethatdividestheboxintotwoparts.Themiddleboxrepresentsthemiddle50%ofscoresforthegroup,delineatedbythelowerandtheupperquartiles.Theupperandlowerwhiskersrepresentthehighestandlowestvalueexcludingoutliers.Thetriangularplotpointsaretheoutliersortheextremes.图3 箱线图，a)放射学股骨-骨骺髋臼顶(FEAR)指数和b)MRI上的FEAR指数，比较所有三组。箱线图由多个数据组成。中位数（中间四分位数）标记数据的中点，由将箱子分成两部分的线表示。中间框代表该组中间50%的分数，由下四分位数和上四分位数划定。上下顶部代表不包括异常值的最高值和最低值。三角形图点是异常值或极值。Fig4MRIinsagittalandfrontalviewofaborderlinedysplastichip.Thefrontalslideislocatedat12o’clockontheacetabulumonthesagittalview.Thisdidnotcorrespondtothe12o’clockpositionofthefemoralhead,duetosmallsubluxationofthefemoralhead.图4 临界髋关节发育不良的矢状位和正位MRI。正面滑动位于矢状位上髋臼的12点钟位置。由于股骨头有轻微的半脱位，这与股骨头的12点钟位置不符。 FemoralosteotomyforosteonecrosisofthefemoralheadAbstractAims:Aborderlinedysplastichipcanbehaveaseitherstableorunstableandthismakessurgicaldecisionmakingchallenging.Whileanunstablehipmaybebesttreatedbyacetabularreorientation,stablehipscanbetreatedarthroscopically.Severalimagingparameterscanhelptoidentifytheappropriatetreatment,includingtheFemoro-EpiphysealAcetabularRoof(FEAR)index,measuredonplainradiographs.TheaimofthisstudywastoassessthereliabilityandthesensitivityofFEARindexonMRIcomparedwithitsradiologicalmeasurement.Patientsandmethods:ThetechniqueofmeasuringtheFEARindexonMRIwasdefinedanditsreliabilityvalidated.Aretrospectivestudyassessedthreegroupsof20patients:anunstablegroupof'borderlinedysplastichips'withlateralcentreedgeangle(LCEA)lessthan25°treatedsuccessfullybyperiacetabularosteotomy;astablegroupof'borderlinedysplastichips'withLCEAlessthan25°treatedsuccessfullybyimpingementsurgery;andanasymptomaticcontrolgroupwithLCEAbetween25°and35°.ThefollowingmeasurementswereperformedonbothstandardizedradiographsandonMRI:LCEA,acetabularindex,femoralanteversion,andFEARindex.Results:TheFEARindexshowedexcellentintraobserverandinterobserverreliabilityonbothMRIandradiographs.TheFEARindexwasmorereliableonradiographsthanonMRI.TheFEARindexonMRIwaslowerinthestableborderlinegroup(mean-4.2°(sd9.1°))comparedwiththeunstableborderlinegroup(mean7.9°(sd6.8°)).WithaFEARindexcut-offvalueof2°,90%ofpatientswerecorrectlyidentifiedasstableorunstableusingtheradiologicalFEARindex,comparedwith82.5%usingtheFEARindexonMRI.TheFEARindexwasabetterpredictorofinstabilityonplainradiographsthanonMRI.Conclusion:TheFEARindexmeasuredonMRIislessreliableandlesssensitivethantheFEARindexmeasuredonradiographs.Thecut-offvalueof2°forradiologicalFEARindexpredictedhipstabilitywith90%probability.Citethisarticle:BoneJointJ2019;101-B:1578-1584.DiscussionThemainresultofthisstudyisthattheFEARindexismeasuredwithgreatersensitivityandreliabilityonplainpelvicradiographsthanonMRI.DifferentiatingastableversusunstablehipcanbechallengingwhentheLCEAisbetween20°and25°.Decidinguponaparticularsurgicaltreatmentdependsprincipallyonthestabilityorinstabilityofthehip.5-7Improvingthediagnosisbynewimagingsignsisnotonlyinterestingbutisalsoessential.Asarthroscopypracticehasadvanced,theproportionofpatientsundergoingasubsequentPAOforthecorrectionofsymptomaticacetabulardysplasiafollowingafailedipsilateralarthroscopichipprocedure,hasincreasedby1.92-foldfrom2008to2015.8Surgicaloptionsforpatientswithaborderlinedysplastichipincludearthroscopiclabralrepairwithcapsularclosure/plication,andsometimeswithcamresection,orPAOwithorwithoutconcomitantlabralrepair/camresection.Thetreatmentchosenisinfluencedbymultiplefactorsincludingpatientage,patientpreference,andespeciallywhetherthesurgeonbelievesthehipisunstableornot.10Theriskofhiparthroscopyinthesettingofinadequatebonecoverageispersistentpainduetoinstability,hipdislocation,orarthriticprogressionofthejoint.18ThismustbebalancedwiththeriskofcomplicationswithPAO,evenwhenperformedbyexperiencedsurgeons,andlowerincrementalclinicalimprovementinpatientswithmilddysplasia.Therearereportcasesoffailureofhippreservationsurgeryonborderlinehips,becausedeterminationofanunstableorstablehipcanbedifficultandmayleadtoincorrecttreatment.8,19-21IfpotentialFAIisfoundconcurrentlywithhipdysplasiatheninappropriateimpingementsurgery,withoutattentiontothedysplasia,canincreasethesymptomsfrominstability.Stringentcriteriaforpatientselectionareneededtoconsiderhiparthroscopyforborderlinedysplastichips.22,23Arecentreview,on1368hiparthroscopiesonacetabulardysplasia,foundthattheisolateduseofhiparthroscopymaybeconsideredinsomedefinedandselectedcasesofborderlineacetabulardysplasia,whencarefulattentionispaidtolabralandcapsularpreservation.24However,manystudieshavedescribedpoorclinicaloutcomesandtheriskofiatrogenicinstability,ifthedysplasiaismoderateorsevere,andifthepatientselectionisnotveryrestricted.24StabilityofthehipisclassicallyassessedbyusingtheLCEA;18nevertheless,thismeasurementaloneisinsufficientforborderlinehipandtheuseofotherassessmentparametersisrequired.SomeradiologicalorMRIparametersaredirectlyduetofemoralheadmigrationandarethereforehighlyindicativeofhipinstability,suchasanincreaseofthedistancefromtheilioischialline,abreakinShenton’sline,orthepresenceofGadoliniumintheposteroinferiorjointspace.24OtherradiologicalorMRIsignscanbepresentinunstablehipbutarelesspredictive,suchasLCEAbetween20°and25°,AIgreaterthan10°,coxavalga,femoralanteversion,orincreasedlabralvolume.Otherparametersassesstheseverityofhipdysplasia,inparticularfortheimplementationoftotalhiparthroplasty.25TheradiologicalFEARindex,describedbyWyattetal,13representstheresultantoftheforcesacrosstheproximalfemoralphysisduringhipgrowth.Thisindexreflectsthefunctionalbehaviour(stableorunstable)ofthehipduringgrowth.Wyattetal13reportedthatifapatientpresentswithhippainandborderlinedysplasia(definedasaLCEA20°to25°),aradiologicalFEARindexlessthan5°indicatesan80%probabilitythatthehipisstable.ThereliabilityoftheFEARindexwasexcellentandatleastasgoodasthereliabilityofLCEAorAI.ThesuperiorityoftheFEARindexcomparedwithLCEAmaybebecauseofdifficultyindefiningtheedgeoftheacetabularsourcil,ashasbeenalludedto.26Inthisstudy,acut-offvalueof2°forradiologicalFEARindexpredictsstability,with90%ofpatientscorrectlyidentifiedasstableorunstable.Withthewideruseofthismeasurementonradiographsforborderlinedysplastichip,normalandcut-offvaluesarebecomingbetterdefined.Occasionally,theFEARindexisdifficulttomeasureonradiographs.Indeed,thelimitsofthephysealscararesometimeshardtodetermineandcanleadtomeasurementserrors.Accordingly,weassessedthereliabilityoftheFEARindexwasassessedwithtwoindependentobservers.TheradiographicandMRIFEARindicesshowedexcellentinter-andintraobserverreliabilityinourstudyandinthestudybyWyattetal,13andwasshowntobesuperiortotheAIorLCEA.ThefalsepositiveornegativeratesoftheFEARindexaremainlyduetothedifficultyintakingmeasurementsinsomecases.TheMRIisaveryaccurateinvestigation,particularlytheMRarthrogramwithradialcuts,whichispresentlythebestavailablepreoperativeimagingstudytoevaluateintra-articularlesionsofnativehip.27-30ThephysealscariseasilyvisualizedonMRIanditseemedapromisingroutetoimprovethesensitivityofthisparameter.Nevertheless,ourstudyhasdemonstratedthatMRIislessreliableandlesssensitivethanplainradiologicalmeasurement.Severalexplanationscanbeproposed.Onradiographs,thephysealscarrepresentsameanofdifferentphysealscarsofeachfrontalslice,andthusameanofthebalanceofforcesactingonthewholefemoralheadduringgrowth.FortheFEARindexonMRI,inordertoobtainaneasyandreliableindex,wehavechosentomeasuretheFEARindexononlyoneimageslice.Increasingthenumberofmeasurementsincreasestheriskoferrors.MRIreconstructionin3Dlosesaccuracycomparedwithastaticradiograph.However,accordingtothechosenfrontalslice,theshapeofthephysealscarcanvarybetweenslicesandtheorientationofthephysealscarcanchangebyseveraldegrees.Thesameproblemcanarisefortheacetabularsourcil,whichcanchangemarkedlyaccordingtothedistancetotheacetabularfossa.Additionally,Wyattetal13statedthattheFEARindexrepresentsthebalanceofforcesonthehip.Usingonlytheaxisofthephysealscarinthemiddleofthefemoralheadmainlyconsiderstheforcesexertedbetweenthesuperiorpartofthefemoralheadandthesuperiorpartoftheroof.Ifthepatienthasasmallsubluxationofthefemoralheadduetoinstability,theslicelocatedat12o’clockontheacetabulumonthesagittalviewisnotthesameastheslidelocatedat12o’clockonthefemoralhead(Fig.4).Thus,themeasureofthecentralpartofthephysealscarcanbeincorrect.Thevariationspotentiallyinducedbythefemoralrotationorabductionarethesameaswithradiographs.Ourstudyhassomelimitations.First,thenumberofpatientsincludedwaslow,withoutpossiblematchingofsomeparameters(LCEA,sex).Nevertheless,patientswithsymptomaticborderlinehipsareuncommonandcompleteimagingexaminationswerenecessarytocompareradiographsandMRI.Theinstabilitywasdeterminedbyseveralfactorsandbytheevolutionofsymptomsafterthesurgery.Thus,thisdiagnosiswasperformedonanassociationofparametersandconstituteafunctionaldefinitionofinstability.Thisstudywasretrospective,andthusallowsustoshowassociationsbutnottomakepredictions.Moreover,someparameterswerenotassessed,suchasthegeneralizedligamentouslaxity,muscularconditioning,orbodymassindex.Thefollow-upoftwoyearsmayseemtooshort.However,apoorresultusuallyissuspectedwithinthreetosixmonths,whenrehabilitationisnotprogressing.Incontrast,afteroneto1.5yearsthehiphasusuallyreachedasteadystateconcerningfunctionandpain.Therefore,atwo-yearlimitseemsrational,althoughalongerfollow-upmightbepreferred.Inconclusion,theFEARindexmeasuredonMRIislessreliablethantheFEARindexmeasuredonplainradiographs.Moreover,theFEARindexisabetterpredictorofinstabilityonplainradiographscomparedwithMRI.Thecut-offvalueof2°fortheradiologicalFEARindexpredictedhipstabilitywith90%probability.ThisradiologicalFEARindexconstitutesareliableparametertoaidthecomplexdiagnosisofhipinstability.Thismeasurementneedstobeassessedprospectivelyforthepredictionofprogressiveinstability.文献出处：CécileBatailler,JanWeidner,MichaelWyatt,DominikPfluger,MartinBeck.IstheFemoro-EpiphysealAcetabularRoof(FEAR)indexonMRIarelevantpredictivefactorofinstabilityinaborderlinedysplastichip?BoneJointJ.2019Dec;101-B(12):1578-1584.doi:10.1302/0301-620X.101B12.BJJ-2019-0502.R1.

临界髋关节发育不良BorderlineDDH(2)：疼痛性临界髋关节发育不良的治疗作者：MichaelCWyatt,MartinBeck.作者单位:KlinikfürOrthopädieundUnfallchirurgieLuzernerKantonsspital6004Luzern,Switzerland.译者：陶可（北京大学人民医院骨关节科）摘要在过去的几十年里，影像技术的改进和手术技术的进步使得保髋手术得到了快速发展。然而，疼痛性临界髋关节发育不良的治疗仍然存在争议。在这篇评论中，我们将确定相关问题并描述患者评估和治疗方案。我们将提供自己的建议，并确定未来的研究领域。简介在过去的几十年里，髋关节生物力学知识的提高和手术技术的进步使得保髋手术得到了快速发展。保髋手术适应范围广泛，从髋臼浅且不稳定的髋关节到髋臼深且患有股骨髋臼撞击(FAI)的髋关节。虽然人们普遍认为，不稳定髋关节发育不良的最佳治疗方法是重新定位髋臼以增加覆盖范围，但人们同样认为，必须减小过度覆盖的髋臼临界以消除撞击。所有这些髋关节都可能存在凸轮畸形，需要在手术矫正时加以解决[1]。在最极端的情况下，所需的治疗是显而易见的。然而，有一个过渡区，很难区分不稳定性和股骨髋臼撞击(FAI)。过去，这些髋关节被称为“临界”髋关节。通常，这包括外侧中心临界(LCE)角度在20°到25°之间的髋关节[2]。然而，“临界”一词是有问题的，因为它是一个放射学定义，只涉及描述髋关节稳定性的几个重要参数之一。髋臼顶倾斜角、前后覆盖和股骨前倾是应纳入髋关节稳定性分析的其他因素。髋关节发育不良与髋关节骨关节炎之间的关联已经确定[3,4]，有不稳定迹象的髋关节发育不良退化速度更快[5]。临界髋关节可能不稳定、撞击或两者兼而有之。临界髋关节发育不良的稳定性很难确定，并且容易受个人主观影响，骨科界普遍倾向于低估不稳定性，从而导致不适当的治疗。最近的研究表明，对患有临界发育不良(LCEA > 20°)的患者进行关节镜髋关节手术（包括盂唇修复和关节囊折叠缝合术）可能会在短期内带来适当的改善[3,4]。然而，有证据表明，之前错误的髋关节镜检查会对此类髋关节的治疗结果产生负面影响[6]。因此，疼痛性临界髋关节发育不良的治疗仍然是一个极具争议的问题。临界性髋关节发育不良在患有髋关节疼痛的年轻人中很常见，在选定的患者群中报告的患病率为37.6%[7]。在临界髋关节发育不良中，可能与其他不稳定原因（如韧带松弛症）有显著重叠[8]。然而，根本问题是难以正确分类潜在的病理生物力学。定义第一个问题在于定义。在前后位骨盆X线片[9](LCEA)上测量的Wiberg外侧中心边缘角传统上用于将髋关节分类为正常(LCEA >25°)、发育不良(LCEA <20°)或临界(LCEA20–25°)，尽管这些定义值在文献中差异很大[3,10]。然而，使用外侧中心边缘角(LCEA)存在两个问题。首先是测量方法。为了测量外侧中心边缘角(LCEA)，首先通过与股骨头轮廓相符的圆来定义股骨头的中心。角度的第一个分支垂直穿过旋转中心。第二个分支由股骨头的中心和股骨最外侧点定义（图1a）。重要的是不要使用髋臼的最外侧点（图1b），因为这不符合Wiberg的定义，并且会给出错误的高值(外侧中心边缘角(LCEA)偏大)[11]。Fig.1.(a)CorrectmeasurementoftheLCEAusingtheedgeofthesourcil,indicatingmoderatedysplasia.(b)IncorrectmeasurementoftheLCEAinthesamehip.Usingthisvaluewouldfalselyclassifythishipasborderline.图1(a)使用髋臼临界正确测量外侧中心边缘角(LCEA)，表明中度髋关节发育不良。(b)同一髋关节的外侧中心边缘角(LCEA)测量不正确。使用此值会错误地将此髋关节归类为临界。其次，实际术语“临界髋关节发育不良”是由Wiberg本人首次提出的，包括外侧中心边缘角(LCEA)在20°和25°之间的髋关节[2]。外侧中心边缘角(LCEA)是一种放射学测量，本身无法预测临界髋关节发育不良的稳定性，也无法完全描述股骨头覆盖范围。因此，外侧中心边缘角(LCEA)无法指导手术决策[12–14]。部分原因是外侧中心边缘角(LCEA)本身无法涵盖发育不良的精确位置，并且忽略了前后股骨头覆盖范围。此外，髋臼指数(AI)和股骨前倾等其他参数也与髋关节稳定性密切相关。如果外侧中心边缘角(LCEA)减少，AI可能正常，在这种情况下很难评估髋关节的稳定性[15]。另一方面，股骨前倾过度可能会加剧髋关节前部不稳定[16]。根本问题是什么？对于疼痛的临界髋关节发育不良，很难仅通过二维射线测量将病理机制表征为撞击（稳定）或发育不良（不稳定），尤其是仅由髋臼功能决定而不考虑股骨的测量。髋关节稳定性的功能表征对于指导手术决策至关重要。不稳定髋关节从逻辑上可以从髋臼重新定向截骨术中受益，而稳定髋关节可以从撞击手术（如股骨凸轮骨成形术）中受益。那么关于髋关节内病理学的了解有多少？应该如何评估这些患者？有哪些治疗方案？手术结果如何？这组患者的潜在隐患是什么？未来的发展方向是什么？在这篇叙述性综述文章中，我们旨在解决这些问题，并阐明这组具有挑战性的患者的处理方法。髋关节发育不良和临界髋关节不稳定的潜在病理是什么？髋关节发育不良患者的关节接触压力异常增高，股骨头（软骨损伤，导致软骨下）骨质相对暴露。髋臼通常较浅且前倾，盂唇经常有代偿性增大，但同时伴有髋臼后倾的情况也很高[17]。股骨通常呈外翻，前倾度高[10]。这些异常的解剖特征会导致病理性髋关节生物力学，表现为盂唇撕裂、软骨损伤和髋关节不稳定，这些很容易被误解为撞击。由于骨稳定性受损，软组织稳定器（即纤维软骨盂唇和髋关节囊）的重要性就凸显出来[18]。一旦软组织约束失效，髋关节就会变得不稳定。然而，我们必须明白，主要的潜在病理是缺乏骨性稳定性，这会导致髋关节失效，而不是软组织稳定性失效。半脱位髋关节发育不良的自然病史预后非常差，并且必然会导致关节退化[5]。恶化速度与半脱位严重程度和患者年龄直接相关，通常在症状出现后约10年，就会出现严重的退行性变化[19]。在没有半脱位的情况下，自然病史很难预测退化速度。临界髋关节发育不良也是如此。最近的一项研究强调了髋臼覆盖的重要性。在一项为期20年的大型女性队列研究中，研究显示，如果外侧中心边缘角(LCE)低于28°，则每降低一度，放射学OA风险就会增加13%[20]。因此，除了短期缓解症状外，还必须考虑长期可能的发展。临床表现临界髋关节发育不良的临床表现与其他年轻活跃成人髋关节疾病（如FAI综合征[21]）非常相似，因此，彻底的病史、体格检查和放射学评估对于正确诊断这些患者至关重要。病史重点记录病史。临界髋关节发育不良患者的主要症状是疼痛。这通常发生在腹股沟和髋关节外侧，但也可能发生在臀部(臀后区)。有必要记录完整的疼痛病史。寻找特定的不稳定和“避免疼痛”症状，这可能表明已经达到因缺乏骨性稳定性而需要的软组织代偿的极限。咔嗒声和卡住的症状也很常见。此外，还会询问患者是否有任何迹象表明患者已经患上髋关节炎，例如夜间疼痛。症状应结合患者的功能限制和已经接受的医疗护理，包括物理治疗、药物、其他意见和手术。检查随后应进行髋关节的合理临床检查，包括恐惧试验和撞击测试。患者通常会表现出“膝内翻”步态，同时伴有髋关节内收肌力矩增加和髋关节内旋增加，这与股骨前倾增加一致。为了功能性地增加前覆盖，可能存在前凸过度。应确定大转子处有无压痛[22]。务必记住检查患者的旋转轮廓、进行神经血管检查以及检查全身关节松弛的迹象，并使用Beighton评分对此进行量化。具体关键目标包括排除(i)晚期退化过程的存在，例如表现为固定屈曲畸形和运动范围减少，以及(ii)其他病理，例如腰椎病或L5神经根病引起的疼痛。调查诊断成像应从骨盆的标准化AP平片和股骨颈侧位片（穿桌侧位、Dunn位、假斜位）[23]开始。仔细检查这些图像以测量LCEA、AI、挤压指数、股骨颈干角和FEAR指数（见下文）。应确定骨关节炎的Tonnis等级以及是否存在凸轮形态。应仔细检查不稳定的直接迹象，这些迹象包括股骨头移位，可通过与髂坐线的距离增加、Shenton线断裂和AP视图上股骨头重新定位来识别，髋关节处于外展状态，使用MR关节造影时后关节间隙中有钆，这表明股骨头向前移位，因此不稳定。FEAR指数与不稳定性有很高的相关性（见下文）。必须精确测量和记录各种参数。有必要使用三维计算机断层扫描(CT)进行横断面成像，以获得有关骨解剖结构和发育不良位置的精确信息，包括髋关节周围囊肿的存在和位置[24-26]。此外，CT还应包括股骨前倾的评估，如果前倾过大，可能会加剧髋关节前部不稳定。磁共振成像(MR-关节造影)应遵循专门的髋关节检查方案，包括径向图像采集或重建和关节内造影剂应用[27]，以检查关节内结构和盂唇和关节软骨的病理。可以区分引起类似症状的其他原因，例如缺血性坏死、转子滑囊炎或臀肌病变。其他测量包括盂唇大小[13,28]和髂关节囊体积[29]。对于这些患者，我们还提倡进行非牵引性MR关节造影检查，以检查是否存在钆积聚，即所谓的“新月征”，这是轴向视图上不稳定的细微征兆[30]。这些测量值的价值是什么？在平片上，那些直接表明不稳定的测量值是股骨头移位，与髂坐线的距离增加，Shenton线断裂，髋关节外展时AP视图上股骨头重新定位，以及FEAR指数。在MR关节造影中，后下关节间隙中钆的存在表明股骨头移位，因此不稳定。AI、NSA、AT、高髂囊体积和盂唇体积可能存在增加，但不能预测不稳定性[30]（表1）。表1.用于评估髋关节不稳定性的各种参数概述TheFemoro-EpiphysealAcetabularRoof(FEAR)index:股骨骨骺髋臼顶指数Thefemoralneck-shaftangle(NSA):颈干角FEAR指数是最近描述的参数，似乎对预测髋关节稳定性具有很高的价值[27]。它是由髋臼顶与股骨生长板中央1/3处之间的角度形成的（图2）。其依据是：在生长过程中，股骨的骨骺生长板会垂直于髋关节的关节反作用力。股骨颈的生长和方向受股骨颈下生长板的控制[31]。Pauwels和Maquet[32]提出理论，合力作用于骨骺软骨的中心，在生长过程中，根据Heuter-Volkman原理，骨骺板会垂直于关节反作用力。Pauwels和Maquet的理论后来得到了Carter等人[33]的证实，他们通过二维有限元分析研究了髋关节负荷的影响。闭合的骨骺板的角度表示跨股骨近端骨骺[34]的力的平衡，也表示跨关节力在过去的作用方式。因此，它是一个功能参数，反映了髋关节在生长过程中长期的关节反作用力。如果FEAR<0°，则认为髋关节稳定。统计分析表明，5°的临界值预测稳定性的概率为80°。最近的研究表明，2°的临界值预测稳定性的概率为90%（Batailler等人，正在准备发表中）。使用FEAR指数的案例如图3a和b所示。Fig.2.TheFEARindex.Theangleismeasuredbetweenalineconnectingthemostmedialandlateralpointofthesourcilandalineconnectingthemedialandlateralendofthestraightpart(usuallycentralthird)ofthephysealscarofthefemoralhead.AnegativeFEARindex,withtheangleopeningmediallyasshowninFig.3a,indicatesastablehip.图2. FEAR指数。测量连接股骨最内侧和外侧点的线与连接股骨头骨骺直线部分（通常为中央三分之一）内侧和外侧端的线之间的角度。如图3a所示，角度向内侧打开的阴性FEAR指数，表示髋关节稳定。Fig.3.(a)CaseexamplesusingtheFEARindex.17-year-oldmale,LCEA20°,FEAR0°.Hipdeemedthereforestableandpatientmanagedwithhiparthroscopy.(b)CaseexamplesusingtheFEARindex.17-year-oldfemale,LCEA20°,FEAR8°.HipdeemedthereforeunstableandpatientmanagedwithPAO.图3.(a)使用FEAR指数的病例。17岁男性，LCEA20°，FEAR0°。因此髋关节稳定，患者接受髋关节镜治疗。(b)使用FEAR指数的病例。17岁女性，LCEA20°，FEAR8°。因此髋关节不稳定，患者接受PAO截骨治疗。有哪些治疗方案？治疗取决于髋关节的稳定性。疼痛性临界髋关节发育不良的治疗方案包括非手术治疗、解决关节内撞击的手术治疗（通过髋关节镜或髋关节外科脱位进行的FAI手术）和解决不稳定性的手术治疗（采用PAO和/或股骨截骨术的重新定位截骨术）（见图2）。非手术治疗包括患者教育、活动调整、简单的止痛药、非甾体抗炎药和髋关节腔内注射药物[35]。有针对性的物理治疗可以改善肌肉调节、疼痛和本体感受控制。以下段落将讨论包括关节镜和/或截骨术的临界髋关节发育不良的手术治疗方案。这组患者接受髋关节镜检查的结果如何？随着髋关节镜技术的最新发展，许多外科医生正在使用它来治疗临界髋关节发育不良，尤其是因为人们认为髋臼周围截骨术等替代技术的风险更高，术后恢复时间更长。临界髋关节发育不良的髋关节镜检查还可以让外科医生处理髋关节内病变，如盂唇撕裂或股骨凸轮畸形[3,12,36]。如果考虑使用PAO来解决骨稳定性不足的问题，那么关节镜检查不仅可以让外科医生了解髋关节的关节内状态，还可以了解患者在随后进行更大规模手术时的表现[37]。然而，关于临界髋关节发育不良的髋关节镜检查的已发表文献很少，而且短期随访也存在局限性。在Jo等的系统综述中，确定了13项关于髋关节发育不良的关节镜检查的研究[10]。这些研究各不相同，所有研究都是病例系列。仅有6项研究报告了主观和/或客观结果。关节镜检查的手术指征不明确，患者事先接受过多种非手术治疗。此外，临界髋关节发育不良的确切定义各不相同，只有两项研究使用了Byrd和Jones的定义[36]。三项研究报告了髋关节镜作为辅助工具，三项研究报告了髋关节镜作为独立治疗。盂唇撕裂的总患病率为77.3%，主要位于髋臼缘的前部或前上部。髋臼软骨病变比股骨病变更常见（59-75.2%比11-32%），并且位于盂唇病变的邻近。仅有两项研究检查了临界髋关节发育不良病例（LCEA20-25°）的关节镜检查结果，其中只有一项描述了患者报告的结果测量。后者是Byrd和Jones[36]的前瞻性临床病例系列，其中66%的髋关节（32髋）患有临界髋关节发育不良。关节镜检查后，平均改良Harris髋关节评分从50（差）改善到77（一般）。作者得出结论，髋关节镜治疗可能解决髋关节内病理而不是发育不良的放射学证据的结果。对临界髋关节发育不良进行髋关节镜检查有什么危险？临界髋关节发育不良患者进行关节镜盂唇切除术和髋臼外侧缘切除术可导致爆发性髋关节不稳定[38]。即使修复了盂唇，也必须保留髂股韧带和髋关节的其他静态稳定器，以防止不可逆的后果或导致髋关节不稳定[39–41]。没有确凿的文献支持在这些情况下进行关节囊修复，但这似乎是一种安全合理的做法[42]。关节囊复位技术可提高临界髋关节发育不良的稳定性[12]。如果髋关节在术前足够不稳定，那么仅通过髋关节镜治疗关节内病变是不够的，患者将需要进行PAO截骨术[43,44]。必须记住，髋关节的稳定性首先取决于髋骨几何形状。在轻微不稳定（临界发育不良）中，稳定性可能由次级软组织结构来确保。一旦这些结构因微创伤或大创伤而失效，髋关节就会变得不稳定。恢复软组织稳定性可能只会在短时间内改善髋关节稳定性，但软组织很可能再次磨损。因此，必须首先解决潜在的骨病理问题，才能取得良好的长期效果。最近的一份报告显示，髋关节发育不良患者在髋关节镜检查失败后，PAO的髋关节特定功能结果较差[6]。因此，对这组患者单独进行髋关节镜检查应谨慎处理。但是，对于那些由于髋关节状况不佳（即AI和股骨前倾正常）或高龄（即>40岁）而不适合进行PAO的患者，它可能有用。重新定向髋臼周围截骨术对这组患者有何影响？通过髋臼周围截骨术进行髋臼重新定向已成为髋关节发育不良最常见的治疗方法，据报道术后20多年效果良好。传统上，PAO时关节内病变的处理方法是进行前关节切开术。然而，随着PAO微创技术的发展，情况已不再如此。微创PAO技术缩短了术后恢复时间[45]。最近的一项研究表明，一些可改变的因素，例如较高的体力活动量和较高的BMI（大于30kg/m2）可导致PAO的发病年龄下降[46]。此外，患有较重发育不良程度的患者患PAO的年龄也较早：LCEA是手术年龄的独立预测因素，即LCEA较低的患者往往需要在较早的年龄接受PAO手术。但是，轻度和中度发育不良患者的PAO预后没有差异。在本研究中，轻度发育不良被归类为15-25°，这涵盖了我们对临界髋关节发育不良的定义。最近的一项多中心前瞻性队列研究检查了患者报告的PAO结果指标，结果表明，虽然总体结果良好，但临界髋关节发育不良患者和男性的改善程度低于发育较重的患者[47]。作者讨论了小范围矫正的危险，这可能导致过度矫正和医源性FAI、股骨前倾增加和软组织松弛。建议和未来方向在临界髋关节中，关键步骤是确定稳定性。关于髋关节的稳定性，只有两种情况：髋关节稳定或不稳定。没有中间状态。如果接受这个概念，治疗就会变得相对简单。不稳定可能与其他病症（如FAI或超负荷/过度使用和软骨疾病）相结合，需要同时治疗。如果髋关节不稳定，则需要髋臼重新定位。仅解决磨损的二级稳定器并不能解决潜在的生物力学问题，最多只能产生令人满意的短期结果。在稳定的髋关节中，可以进行开放或关节镜关节保留手术。然而，我们必须记住，低于28°的LCE角度每减少一度，骨关节炎的发病率就会增加13%[20]。因此，如果有疑问，为了最大限度地提高获得良好长期结果的机会，我们主张进行髋臼重新定向PAO截骨手术。重要的是要确定我们缺乏知识的领域，以指导进一步的研究。将对这些患者进行长期随访研究，比较髋臼重新定向和髋关节镜检查，理想情况下，将记录所有成像参数和Beighton评分。此外，还应获得患者报告的结果测量和恢复时间，以及包括运动在内的活动恢复时间。 TheFEARindexisarecentlydescribedparameterthatseemstohaveahighvaluetopredictstabilityofthehip[27].Itisformedbytheanglebetweentheacetabularroofandthecentralthirdofthefemoralgrowthplate(Fig.2).Itisbasedonthefactthatduringgrowththeepiphysealgrowthplateofthefemurorientsitselfperpendicularlytothejointreactingforcesofthehip.Growthandtheorientationofthefemoralneckareunderthecontrolofthesubcapitalgrowthplate[31].PauwelsandMaquet[32]theorizedthattheresultantforceactsfromthecenteroftheepiphysealcartilageandthatduringgrowth,theepiphysealplateorientsitselfperpendiculartothejointreactionforceinaccordancewiththeHeuter–Volkmanprinciple.PauwelsandMaquet’stheorylaterwasconfirmedbyCarteretal.[33]whostudiedtheinfluenceofhiploadingbybi-dimensionalfiniteelementanalysis.Theangleoftheclosedepiphysealplateindicatesthebalanceofforcesacrosstheproximalfemoralphysis[34]andindicateshowthetransarticularforcesactedinthepast.Therefore,itisafunctionalparameterthatreflectsthejointreactingforcesoveralongperiodoftimeduringgrowthofthehip.IftheFEARis <0°thehipisconsideredstable.Statisticalanalysishasshownthatacutoffvalueof5°predictsstabilitywith80°probability.Morerecentworkhasshownthatacutoffvalueof2°predictsstabilitywith90%probability(Batailleretal.,inpreparation).CaseexamplesofusingtheFEARindexareshowninFig.3aandb.ThemanagementofthepainfulborderlinedysplastichipAbstractImprovedimagingandtheevolutionofsurgicaltechniqueshavepermittedarapidgrowthinhippreservationsurgeryoverthelastfewdecades.Themanagementofthepainfulborderlinedysplastichiphoweverremainscontroversial.Inthisreview,wewillidentifythepertinentissuesanddescribethepatientassessmentandtreatmentoptions.Wewillprovideourownrecommendationsandalsoidentifyfutureareasforresearch.INTRODUCTIONImprovedknowledgeabouthipbiomechanicsandtheevolutionofsurgicaltechniqueshavepermittedarapidgrowthinhippreservationsurgeryoverthelastfewdecades.Thespectrumcoversawiderangefromhipswithshallowacetabuli,whichareunstable,tohipswithdeepacetabulithataresufferingfromfemoro-acetabularimpingement(FAI).Whilethereisageneralagreementthatthebesttreatmentfortheunstabledysplastichipisareorientationoftheacetabulumtoincreasecover,thereisequalagreementthattherimoftheover-coveringacetabulumhastobereducedtoremoveimpingement.Onallthosehipsacamdeformitymaybepresentthatneedstobeaddressedatthetimeofsurgicalcorrection[1].Atthefarendsofthespectrumtherequisitetreatmentisobvious.However,thereisatransitionzonewhereitisdifficulttodiscriminateinstabilityfromFAI.Inthepastthesehipswerereferredtoas‘borderline’hips.Usually,thisincludedhipswithalateralcenteredge(LCE)anglebetween20°and25°[2].However,theterm‘borderline’isproblematic,becauseitisaradiographicdefinitionandonlyaddressesoneofseveralparametersimportanttodescribehipstability.Acetabularroofobliquity,anteriorandposteriorcoverandfemoralantetorsionareotherfactorsthatshouldbeincludedintoananalysisofhipstability.Theassociationofhipdysplasiawithhiposteoarthritisisestablished[3,4]anddysplastichipswithsignsofinstabilitydegenerateatahigherrate[5].Aborderlinehipcaneitherbeunstable,impingingormaybeboth.Thestabilityoftheborderlineisdifficulttodetermineandsubjecttointerpretationwithageneraltendencyintheorthopaediccommunitytounderestimateinstabilitythatthenleadstoinappropriatetreatment.Recentstudiessuggestthatarthroscopichipsurgerywithlabralrepairandcapsularplicationinpatientswithborderlinedysplasia(LCEA > 20°)mayresultinappropriateshort-termimprovements[3,4].However,thereisevidencethatawronglydoneprevioushiparthroscopyhasanegativeimpactontheoutcomeonthetreatmentofsuchhips[6].Therefore,themanagementofthepainfulborderlinedysplastichiphoweverremainsanissueofgreatcontroversy.Borderlinehipdysplasiaiscommoninyoungadultswithhippainwithareportedprevalenceof37.6%inselectedpatientcohorts[7].Intheborderlinedysplastichiptheremaybesignificantoverlapwithothercausesofinstabilitysuchasconnectivetissuelaxity[8].However,thefundamentalissueisthedifficultyincorrectlyclassifyingtheunderlyingpatho-biomechanics.DEFINITIONThefirstproblemliesinthedefinition.TheLateralCentreEdgeAngleofWibergasmeasuredonanAntero-posteriorpelvicradiograph[9](LCEA)hastraditionallybeenusedtoclassifyhipsasnormal(LCEA >25°),dysplastic(LCEA <20°)orborderline(LCEA20–25°)althoughthesedefiningvaluesvarywidelyintheliterature[3,10].However,theuseoftheLCEAhastwoproblems.Firstlythemethodbywhichitshouldbemeasured.TomeasuretheLCEAthecenterofthefemoralheadisfirstdefinedbyacirclefittingthecontourofthefemoralhead.Thefirstbranchoftheanglerunsperpendicularthroughthecenterofrotation.Thesecondbranchisdefinedbythecenterofthefemoralheadandthemostlateralpointofthesourcil(Fig.1a).Itisimportantnottousethemostlateralpointoftheacetabulum(Fig.1b),becausethisdoesnotfollowthedefinitionofWiberg,andwillgivefalsehighvalues[11].Secondlytheactualterm‘Borderlinehipdysplasia’wasfirstintroducedbyWiberghimself,includinghipswithaLCEAbetween20°and25°[2].LCEAisaradiographicmeasureandpersecannotpredictstabilityintheborderlinedysplastichipnordoesfullydescribefemoralheadcoverage.ThereforetheLCEAcannotdirectsurgicaldecisionmaking[12–14].PartofthereasonisthatLCEAalonedoesnotencompassthepreciselocationofdysplasiaanddisregardsanteriorandposteriorfemoralheadcoverage.Alsootherparameterssuchasacetabularindex(AI)andfemoralantetorsionareveryrelevantforstabilityofthehip.InthepresenceofadecreasedLCEAAImaybenormalinwhichcasethestabilityofthehipisdifficulttoassess[15].Ontheotherhand,excessivefemoralanteversionmaypotentiateanteriorhipinstability[16].WHATISTHEFUNDAMENTALISSUE?Inthepainfulborderlinedysplastichipitisdifficulttocharacterizethepathologicalmechanismasimpingement(stable)ordysplasia(unstable)byatwo-dimensionalradiographicmeasurementalone,especiallyonethatissolelyafunctionoftheacetabulumandtakesnoaccountofthefemur.Thisfunctionalcharacterizationofhipstabilityisofparamountimportancetoguidesurgicaldecision-making.Anunstablehipwouldlogicallybenefitfromacetabularreorientationosteotomywhilstastablehipwouldbenefitfromimpingementsurgerysuchasfemoralcamosteoplasty.Sowhatisknownabouttheintra-articularpathology?Howshouldthesepatientsbeassessed?Whatarethetreatmentoptions?Whatarethesurgicaloutcomes?Whatarethepotentialpitfallswiththisgroupofpatients?Whatarethefuturedirections?Inthisnarrativereviewarticleweaimtoaddressthesequestionsandelucidatethemanagementofthischallenginggroupofpatients.WHATISTHEUNDERLYINGPATHOLOGYOFHIPDYSPLASIAANDUNSTABLEBORDERLINEHIPS?Inhipdysplasia,thereareabnormallyhigharticularcontactpressuresandrelativebonyuncoveringofthefemoralhead.Theacetabulumistypicallyshallowandantevertedwithanoftencompensatoryenlargedlabrum,butthereisalsoahighprevalenceofconcomitantacetabularretroversion[17].Thefemurisclassicallyinvalguswithhighantetorsion[10].Theseabnormalanatomicalfeaturescausepathologicalhipbiomechanicswhichmanifestaslabraltears,chondrallesions,andhipinstability,whichcaneasilybemisinterpretedasimpingement.Astheosseousstabilityiscompromisedtheimportanceofthesofttissuestabilisers,namelythefibrocartilaginouslabrumandthehipcapsule,isaccentuated[18].Oncethesofttissueconstraintsfailthenthehipbecomesunstable.However,onehastounderstandthattheprincipalunderlyingpathologyisthelackofosseousstability,whichleadstofailureofthehipandnotthefailingsofttissuestability.Thenaturalhistoryofthesubluxingdysplastichipisaverypoorprognosisandinvariablyleadstojointdegeneration[5].Therateofdeteriorationisdirectlyrelatedtosubluxationseverityandpatientageandusuallyabout10 yearsafteronsetofsymptomsseveredegenerativechangeshavedeveloped[19].Thenaturalhistoryintheabsenceofsubluxationismoredifficulttopredictconcerningthespeedofdegeneration.Thesameaccountsforborderlinedysplastichips.Arecentstudyhighlightstheimportanceofacetabularcover.Inalargecohortoffemales,followedfor20 years,itwasshownthateachdegreereductioninLCEbelow28°isassociatedwith13%increasedriskofradiographicOA[20].Therefore,besidesshort-termreliefofsymptoms,thelong-termpossibleevolutionhastobekeptinmind.CLINICALPRESENTATIONTheclinicalpresentationofborderlineacetabulardysplasiaisverysimilartothatofotheryoungactiveadulthipdisorders,suchasFAIsyndrome[21]soathoroughhistory,physicalexamination,andradiographicevaluationareessentialtoproperlydiagnosethesepatients.HISTORYAfocusedhistoryistaken.Theprimarysymptominpatientswithborderlinehipdysplasiaispain.Thisistypicallyperceivedingroinandlateralhipbutcanalsobeinthebuttock.Afullpainhistoryiswarranted.Particularsymptomsofinstabilityand‘givingway’aresoughtthatmayindicatethatthelimitsofsofttissuecompensationforalackofosseousstabilityhavebeenreached.Symptomsofclickingandcatchingarealsocommon.Furthermoreanyindicationsthatthepatienthasestablishedhiparthritis,suchasnightpain,areaskedfor.Thesymptomsshouldbeputintothecontextofthepatient’sfunctionallimitationsandmedicalattentionalreadyreceivedincludingphysiotherapy,medications,otheropinionsandsurgery.EXAMINATIONAlogicalclinicalexaminationofthehipshouldfollowincludingapprehensionandimpingementtests.Thepatientwilloftendisplaya‘kneeing-in’gaitinassociationwithanincreasedhipadductormomentandincreasedinternalhiprotationconsistentwithincreasedfemoralantetorsion.Hyperlordosismaybepresentinordertofunctionallyincreaseanteriorcover.Tendernessoverthegreatertrochantershouldbedetermined[22].Itiscrucialtoremembertoexaminethepatient’srotationalprofile,performaneurovascularexaminationandtocheckforsignsofgeneralizedjointlaxityandquantifythisusingBeighton’sscore.Specifickeyaimsincluderefutingthepresenceof(i)anadvanceddegenerativeprocessmanifestforexamplewithfixedflexiondeformityanddecreasedrangeofmotionand(ii)alternativepathologysuchaspainreferredfromlumbarspondylosisorL5radiculopathy.INVESTIGATIONSDiagnosticimagingshouldcommencewithstandardizedplainAPradiographofthepelvisandalateralfemoralneckviews(lateralcrosstable,Dunnview,falseprofileviews)[23].TheseimagesarescrutinizedtomeasuretheLCEA,AI,extrusionindex,femoralneck-shaftangleandFEARindex(seebelow).TheTonnisgradeofosteoarthritisshouldbedeterminedalongwithwhetherthereiscammorphology.Directsignsofinstabilityshouldbescrutinizedforandthesecomprisefemoralheadmigration,recognizedbyanincreaseddistancefromtheilioischialline,abreakinShenton’slineandrecenteringofthefemoralheadonanAPviewwiththehipinabductionandGadoliniumintheposteriorjointspacewhenusingMR-arthrography,thatindicatesanteriormigrationandthusinstabilityofthefemoralhead.TheFEARindexhasahighassociationwithinstability(seebelow).Thevariousparametershavetobemeasuredpreciselyandrecorded.Cross-sectionalimagingwiththree-dimensionalcomputerizedtomography(CT)forpreciseinformationonbonyanatomyandlocationofdysplasiaincludingthepresenceandlocationofperiarticularcystsiswarranted[24–26].FurthermoreCTshouldincludeestimationoffemoralantetorsionwhich,ifhighmaypotentiateanteriorhipinstability.Magneticresonanceimaging(MR-arthrography)shouldfollowadedicatedprotocolfortheexaminationofthehip,includingradialimageacquisitionorreconstructionandintra-articularapplicationofcontrast[27]toexamineforintra-articularstructuresandpathologyofbothlabrumandarticularcartilage.Othercausesforsimilarsymptomssuchasavascularnecrosis,trochantericbursitisorglutealpathologycanbedifferentiated.Additionalmeasurementsincludelabralsize[13,28]andiliocapsularisvolume[29].Inthesepatients,wealsoadvocatenon-tractionMRarthrographytoexamineforaaccumulationofgadoliniumknownasa‘crescentsign’whichisasubtlesignofinstabilityontheaxialview[30].WHATISTHEVALUEOFTHESEMEASUREMENTS?Onplainfilmsthosemeasurementsthataredirectsignsofinstabilityarefemoralheadmigrationwithanincreaseofthedistancefromtheilioischialline,abreakinShenton’slineandrecenteringofthefemoralheadontheAPviewwithhipsinabductionandtheFEARindex.OnMR-arthrographythepresenceofGadoliniuminthepostero-inferiorjointspaceindicatesmigrationofthefemoralheadandthusinstability.TheAI,NSA,AT,highiliocapsularisvolumeandincreasedlabralvolumemaybepresentbutarenotpredictiveofinstability[30](Table1).WHATARETHETREATMENTOPTIONS?Treatmentdependsonthestabilityofthehip.Thetreatmentalternativesforthepainfulborderlinedysplastichipincludenon-operativetreatment,surgicaltreatmenttoaddressintra-articularimpingement(FAIsurgerybyeitherhiparthroscopyorsurgicalhipdislocation)andsurgicaltreatmenttoaddressinstability(reorientationosteotomywithPAOand/orfemoralosteotomy)(seeFig.2).Non-operativemanagementincludespatienteducation,activitymodification,simpleanalgesics,non-steroidalanti-inflammatorymedications,andintra-articularinjections[35].Targetedphysiotherapycanimprovemuscularconditioning,painandproprioceptivecontrol.Thesurgicaltreatmentoptionsfortheborderlinedysplastichipwhichcomprisearthroscopyand/orosteotomywillbediscussedinthefollowingparagraphs.WHATARETHERESULTSOFHIPARTHROSCOPYINTHISGROUPOFPATIENTS?Withtherecentevolutioninhiparthroscopymanysurgeonsareusingthistoaddressborderlinedysplastichips,notleastbecauseofperceivedhigherrisksandlongerpost-operativerecoveryassociatedwithalternativetechniquessuchasperiacetabularosteotomy.Hiparthroscopyinborderlinedysplastichipspermitsthesurgeontoaddressintra-articularpathologysuchasalabraltearorfemoralcamdeformity[3,12,36].IfPAOisbeingconsideredtoaddresstheinadequatebonystabilitythenarthroscopymaygivethesurgeonvaluableinsightsnotonlyintotheintra-articularstatusofthehipbutalsohowthepatientislikelytofarewithamuchlargersubsequentoperation[37].However,thereislittlepublishedliteratureonhiparthroscopyinborderlinedysplastichipsandwhatthereislimitedbyshort-termfollow-up.InthesystematicreviewbyJoetal.,13studieslookingatarthroscopyindysplastichipswereidentified[10].Thestudieswereheterogeneousandallstudieswerecaseseries.Onlysixstudiesreportedonsubjectiveand/orobjectiveoutcomes.Thesurgicalindicationsforarthroscopywereambiguousandpatientshadreceivedvariablenon-operativemanagementapriori.FurthermoretheprecisedefinitionofborderlinehipdysplasiavariedandonlytwostudiesusedthedefinitionofByrdandJones[36].Threestudiesreportedonhiparthroscopyasanadjuvanttoolandthreeasastand-alonetreatment.Labraltearshadanoverallprevalenceof77.3%andtheseweremostlylocatedintheanteriororanterosuperiorportionoftheacetabularrim.Acetabularchondrallesionsweremorecommonthanfemorallesions(59–75.2%versus11–32%)andlocatedadjacenttothatofthelabralpathology.Therewereonlytwostudiesthatexaminedtheoutcomesofarthroscopyinborderlinehipdysplasticcases(LCEA20–25°)ofwhichonlyonedescribedpatientreportedoutcomemeasures.Thelatter,aprospectiveclinicalcaseseriesbyByrdandJones[36],had66%ofhips(32hips)withborderlinedysplasia.ThemeanmodifiedHarrisHipscoreimprovedfrom50(poor)to77(fair)followingarthroscopy.Theauthorsconcludedthatthetreatmentresponseislikelyafunctionofaddressingtheintra-articularpathologyratherthantheradiographicevidenceofdysplasia.WHATARETHEDANGERSWITHDOINGHIPARTHROSCOPYINBORDERLINEDYSPLASTICHIPS?Arthroscopiclabralresectionandremovaloflateralacetabularriminborderlinehipdysplasiacanleadtofulminantjointinstability[38].Evenifthelabrumisrepaireditisimperativetopreservetheiliofemoralligamentandotherstaticstabilizersofthehiptopreventtheirreversibleconsequencesorrenderingthehipunstable[39–41].Thereisnoconclusiveliteraturetosupportcapsularrepairinthesecasesbutthisseemsasafeandsensiblepractice[42].Capsularreductiontechniquestoimprovestabilityhavebeendescribedinborderlinedysplastichips[12].Ifthehipissufficientlyunstablepre-operativelythenaddressingtheintra-articularpathologyalonebyhiparthroscopywillbeinsufficientandthepatientwillrequireaPAO[43,44].Onehastobearinmindthatstabilityofthehipfirstlinedependsontheosseousgeometry.Insubtleinstability(borderlinedysplasia)stabilitymaybesecuredbysecondarysofttissuestructures.Oncethesefailduetomicro-ormacrotraumathehipbecomesunstable.Restoringsofttissuestabilitymayimprovehipstabilityforashortperiodoftimeonly,butitislikelythatthesofttissueswearoutagain.Thereforetheunderlyingosseouspathologyhastobeaddressedfirsttoachievegoodlong-termresults.ArecentreportshowedaninferiorhipspecificfunctionaloutcomeofPAOafterfailedhiparthroscopyinhipdysplasia[6].Hiparthroscopyaloneinthisgroupofpatientsshouldbethereforeapproachedwithcaution.However,itmayhavearoleinthosepatientswhoareeitherunsuitableforPAOeitherbecausetheirhipsareunfavourable(i.e.haveanormalAIandnormalfemoralanteversion)orbecausetheiradvancedage(i.e.>40years).WHATARETHERESULTSOFREORIENTINGPERIACETABULAROSTEOTOMYINTHISGROUPOFPATIENTS?Acetabularreorientationviatheperiacetabularosteotomyhasbecomethemostcommontreatmentforacetabulardysplasiawithgoodoutcomesreportedatover20 yearspostoperatively.Traditionallyintra-articularpathologywasaddressedatthetimeofPAObyperformingananteriorarthrotomy.HoweverwiththedevelopmentofminimallyinvasivetechniquesforPAOthisisnolongernecessarilythecase.LessinvasivePAOtechniqueshavedecreasedthetimetopostoperativerecovery[45].ArecentstudyshowedmodifiablefactorssuchashigherphysicalactivityandhigherBMIgreaterthan30 kg/m2leadtoadecreasedageofpresentationforPAO[46].FurthermorepatientsalsopresentedearlierforPAOwithworsedegreesofdysplasia:theLCEAwasindependentlypredictiveofageatsurgery,i.e.patientswithalowerLCEAtendedtorequirePAOsurgeryatanearlierage.However,therewasnodifferenceinoutcomesfollowingPAObetweenmildandmoderatedysplasia.Inthisstudymilddysplasiawasclassifiedas15–25°whichencompassesourdefinitionofborderlinehipdysplasia.Arecentmulticenterprospectivecohortstudythatexaminedpatient-reportedoutcomemeasuresofPAOshowedthat,althoughoverallresultsweregood,improvementsinborderlinehipdysplasticsandmaleswerelessthaninthosepatientswhohadmoreseveredysplasia[47].TheauthorsdiscussedthiswiththedangerofasmallcorrectionthatmayleadtoovercorrectionandiatrogenicFAI,increasedfemoralantetorsionandsofttissuelaxity.RECOMMENDATIONSANDFUTUREDIRECTIONSInborderlinehipsthecrucialstepistodefinestability.Regardingthestabilityofthehipthereareonlytwoconditions:Thehipiseitherstableorunstable.Thereisnothinginbetween.Ifthisconceptisaccepted,thetreatmentgetscomparablysimple.InstabilitymaybecombinedwithotherpathologieslikeFAIoroverload/overuseandcartilagediseasewhichneedconcomitanttreatment.Ifthehipisunstable,acetabularreorientationisnecessary.Addressingonlywornoutsecondarystabilizersdoesnotsolvetheunderlyingbiomechanicproblemandatbestwillyieldsatisfactoryshorttermresults.Instablehips,openorarthroscopicjointpreservingsurgerymaybeperformed.However,wehavetokeepinmindthateachdegreedecreaseoftheLCEanglebelow28°isassociatedwitha13%increaseofosteoarthrosis[20].Therefore,ifindoubt,inordertomaximizethechanceofgoodlong-termresults,wewouldadvocateforanacetabularreorientationoperation.Itisimportanttoidentifytheareaswherewelackknowledgeinordertoguidefurtherresearch.Longer-termfollow-upstudiescomparingacetabularreorientationandhiparthroscopyinthesepatients,ideallyinwhichallimagingparametersandBeightonscoresarerecordedwouldbeperformed.Inadditionpatient-reportedoutcomemeasuresandtimetorecoveryandresumptionofactivitiesincludingsportshouldbeattained.文献出处：MichaelCWyatt,MartinBeck.Themanagementofthepainfulborderlinedysplastichip.ReviewJHipPreservSurg.2018Apr5;5(2):105-112.doi:10.1093/jhps/hny012.

临界髋关节发育不良BorderlineDDH：什么时候、怎么判定为异常？作者：SarahDBixby,MichaelBMillis.作者单位:DepartmentofRadiology,BostonChildren'sHospital,Main2,300LongwoodAve.,Boston,MA,02115,USA.sarah.bixby@childrens.harvard.edu.译者：陶可（北京大学人民医院骨关节科）摘要临界髋关节发育不良是指髋臼形状和覆盖范围轻度低于正常范围，可能导致儿童易患机械功能障碍和不稳定。临界发育不良通常包括外侧中心边缘角(CEA)为18-24°的儿童。一些具有临界X线测量值的儿童具有正常的关节力学和功能，而其他儿童则需要进行髋臼截骨手术。虽然临界髋关节发育不良的X线检查结果可能表明不稳定，但最终诊断基于病史和体格检查以及影像学相结合。患有临界髋关节发育不良的儿童有时还需要进行其他横断面成像研究，如MRI成像，以评估不稳定的次要证据，包括沿髋臼边缘的损伤或盂唇退化和肥大。CT也有助于描绘3-D髋臼形态，以进行术前评估和规划。儿童放射科医生通常是第一个在X线片上发现边缘性或轻度发育不良的人。儿童放射科医生必须充当有效的顾问，并作为针对这一复杂患者群体的连贯多学科诊疗团队的一部分提供适当的建议。图1 一名16岁女孩的髋关节X线片显示了外侧中心边缘角(CEA)、髋臼指数和前CEA的测量方法。a标记了三个点以便进行后续测量：点1位于股骨头(旋转)中心，点2位于股骨髋关节(负重关节面的)内侧边缘，点3位于股骨髋关节(负重关节面的)外侧边缘。b右髋关节的前后位X线片展示了测量外侧CEA的方法。外侧CEA是连接点1和点3的线与垂直线之间形成的夹角（正常为25-40°）。c髋臼指数由连接点2和点3的线与水平线之间形成的夹角决定（正常为0-10°）。d右髋关节的假斜位片显示了确定前CEA的测量方法。前侧CEA由点1到点4（在髋臼前缘）的连线与垂直线之间的角度决定（正常值为25–40°）图2 一名19岁女性严重双侧髋关节发育不良。a骨盆前后位X线片显示右侧髋关节外侧中心边缘角(CEA)为5°，b髋臼指数为16°。左髋关节同样发育不良。c右髋关节假斜位X线片显示前方中心边缘角为8°。图3 一名右侧髋关节疼痛的16岁女孩的外侧中心边缘角(CEA)。a前后位(AP)骨盆X线片显示外侧CEA为19°。左侧髋关节正常。b同一女孩的假斜位X线片显示前方CEA为19°。c延迟钆增强软骨MRI(dGEMRIC)检查的冠状T1图显示软骨正常。女孩没有不稳定的迹象，接受了腰肌劳损治疗，症状在没有手术的情况下得到改善。由于腰肌劳损仍然可能与潜在的微不稳定有关，女孩在髋关节外科医生的护理下接受长期观察。图4 一名有髋关节发育不良家族史的11岁女性足球运动员的左侧髋关节疼痛。a前后位(AP)X线片显示为正常。b女孩现在13岁，仍然有左侧髋关节疼痛。双侧髋关节被认为是浅的，左侧测量的外侧中心边缘角(CEA)为16°，右侧为23°。左侧髋臼指数也被认为是升高的13°。阳性恐惧测试反映了体格检查中左侧髋关节的不稳定，右侧没有发现不稳定。C假斜位X线片显示前方覆盖充分，前方CEA为27°。d左侧髋臼周围截骨术后前后位AP骨盆X线片状态显示侧方覆盖改善。女孩报告症状有所改善。e一年后的前后位AP骨盆X线片，症状缓解并恢复正常活动。她右侧有轻微间歇性疼痛，并定期监测症状。图5 股骨骨骺髋臼顶(FEAR)指数。a与图1中相同的16岁女孩的髋关节前后位(AP)X线片。FEAR指数是沿髋臼顶连接点2和点3的线（黑线）与沿股骨骨骺线中央三分之一绘制的线（白线）之间形成的角度。阳性FEAR指数定义为向外侧倾斜的角度，顶点指向内侧。这个女孩的FEAR指数小于5°。与FEAR指数<5°相比，FEAR指数>5°与不稳定性具有更大的相关性。b一名13岁女孩的髋关节前后位X线片显示左侧髋臼外侧轻微上翻（箭头）。c一名13岁男孩的右侧髋关节疼痛的左侧髋关节正常前后位X线片显示外侧正常。黑线表示髋臼顶部，白线表示骨骺线的中央三分之一。此正常髋关节的FEAR指数显示FEAR指数，角度顶点指向外侧。图6 一名18岁女子田径运动员的右侧髋关节疼痛影像。a前后位(AP)骨盆X线片显示右侧髋关节发育不良，左髋关节正常。b右侧髋臼周围截骨术后前后位(AP)骨盆状态显示外侧覆盖增加，但该女性报告症状恶化。c术前（回顾性）进行的右髋关节矢状质子密度脂肪抑制MR图像显示前髋臼内有骨样骨瘤（箭头）。射频消融后，该患者的症状得到缓解。 Theborderlinedysplastichip:whenandhowisitabnormal?AbstractBorderlineacetabulardysplasiareferstomildlysub-normalpatternsofacetabularshapeandcoveragethatmightpredisposechildrentomechanicaldysfunctionandinstability.Borderlinedysplasiagenerallyincludeschildrenwithalateralcenteredgeangle(CEA)of18-24°.Somechildrenwithborderlineradiographicmeasurementshavenormaljointmechanicsandfunctionwhileothersbenefitfromacetabularreorientingsurgery.Althoughradiographicfindingsofborderlinedysplasiamightsuggestinstability,theultimatediagnosisisbasedonhistoryandphysicalexaminadditiontoimaging.Childrenwithborderlineacetabulardysplasiasometimesbenefitfromothercross-sectionalimagingstudiessuchasMRimagingtoevaluateforsecondaryevidenceofinstability,includingdamagealongtheacetabularrim,orlabraldegenerationandhypertrophy.CTisalsohelpfulfordepictionof3-Dacetabularmorphologyforpreoperativeassessmentandplanning.Pediatricradiologistsareoftenthefirsttoidentifyborderlineormilddysplasiaonradiographs.Itisimperativethatpediatricradiologistsserveaseffectiveconsultantsandofferappropriaterecommendationsaspartofacohesivemultidisciplinaryapproachtothiscomplexpatientpopulation.Fig.1Radiographsofthehipina16-year-oldgirldemonstratemeasurementtechniqueforlateralcenteredgeangle(CEA),acetabularindex,andanteriorCEA.aThreepointsaremarkedforsubsequentmeasurements:Point1atthecenterofthefemoralhead,Point2atthemedialedgeofthesourcil,Point3atthelateraledgeofthesourcil.BAnteroposteriorpelvicradiographconeddowntotherighthipdemonstratestechniqueformeasuringlateralCEA.ThelateralCEAistheangleformedbetweenthelineconnectingPoint1toPoint3andaverticalline(normal25–40°).cAcetabularindexisdeterminedbytheangleformedbetweenalineconnectingPoint2toPoint3andahorizontalline(normal0–10°).dFalseprofileviewoftherighthipdemonstratesmeasurementtechniquefordetermininganteriorCEA.AnteriorCEAisdeterminedbytheanglebetweenalinefromPoint1toPoint4(attheanteriormarginofthesourcil)andaverticalline(normal25–40°)Fig.2Severebilateralhipdysplasiaina19-year-oldwoman.aAnteroposteriorradiographofthepelvisdemonstratesalateralcenteredgeangle(CEA)oftherighthipof5°and(b)acetabularindexof16°.Thelefthipissimilarlydysplastic.cFalseprofileradiographoftherighthipdemonstratesanteriorcenteredgeangleof8°.Fig.3Borderlinelateralcenteredgeangle(CEA)ina16-year-oldgirlwithrighthippain.aAnteroposterior(AP)pelvisradiographrevealsborderlinelateralCEAof19°.Thelefthipisnormal.bFalseprofileradiographofthesamegirlrevealsanteriorCEAof19°.cCoronalT1mapfromdelayedgadolinium-enhancedMRIofcartilage(dGEMRIC)examinationrevealsnormalcartilage.Thegirldidnothavesignsofinstability,wastreatedforpsoasstrain,andsymptomsimprovedwithoutsurgery.Becausepsoasstraincouldstillberelatedtounderlyingmicroinstability,thegirlwasunderlong-termobservationundercareofahipsurgeon.Fig.4Lefthippaininan11-year-oldfemalesoccerplayerwithafamilyhistoryofhipdysplasia.aAnteroposterior(AP)radiographinterpretedasnormal.bGirlnowisage13years,stillwithlefthippain.Bilateralhipswereconsideredshallow,withlateralcenteredgeangle(CEA)measuredat16°ontheleftand23°ontheright.Acetabularindexontheleftwasalsoconsideredelevatedat13°.Positiveapprehensiontestreflectedinstabilityofthelefthiponphysicalexamination,withnoinstabilitynotedontheright.cFalseprofileradiographrevealsadequateanteriorcoverage,withanteriorCEAof27°.dAPpelvisradiographstatuspostleft-sideperiacetabularosteotomydemonstratesimprovedlateralcoverage.Thegirlreportedimprovedsymptoms.eAPpelvisradiograph1yearlater,afterreliefofsymptomsandreturntonormalactivity.Shehadmildintermittentpainontherightandwasbeingmonitoredperiodicallyforsymptoms.Fig.5Femoro-epiphysealacetabularroof(FEAR)index.AAnteroposterior(AP)radiographofthehipinthesame16-year-oldgirlasinFig.1.TheFEARindexistheangleformedbetweenalineconnectingPoint2andPoint3alongtheacetabularroof(blacklines),andalinedrawnalongthecentralthirdofthefemoralphysealscar(whiteline).ApositiveFEARindexisdefinedbyalaterallydirectedanglewiththeapexpointingmedially.TheFEARindexinthisgirlislessthan5°.AFEARindex>5°hasagreatercorrelationwithinstabilitycomparedtoFEARindex<5°.bAPradiographofthehipina13-year-oldgirlwithmildleftacetabularsourcildemonstratesamildlyupturnedlateralsourcil(arrow).cNormalAPradiographofthelefthipina13-year-oldboywithrighthippaindemonstratesanormallateralsourcilforcomparison.Theblacklineindicatestheacetabularroofandthewhitelineindicatesthecentralthirdofthephysealscar.TheFEARindexisdemonstratedinthisnormalhiptoillustrateanegativeFEARindexwiththeapexoftheangledirectedlaterally.Fig.6Imaginginan18-year-oldfemaletrackathletewithrighthippain.aAnteroposterior(AP)pelvisradiographreportedasborderlinerightacetabulardysplasiaandnormallefthip.bAPpelvisstatuspostrightperiacetabularosteotomydemonstratesincreasedlateralcoverage,thoughthewomanhadreportedworseningsymptoms.cSagittalproton-densityfat-suppressedMRimageoftherighthipperformedpreoperatively(retrospectively)revealsanosteoidosteomawithintheanterioracetabulum(arrow).Thewoman’ssymptomsresolvedafterradiofrequencyablation. AdvancinginvestigationTheterm“borderlinedysplasia”isalsofallingoutoffavor.Advancedimagingmodalitieshaverevealedpatternsofdysplasiathatarenotapparentonradiographs.FocalanteriorandposteriordysplasiagroupshaveanormallateralCEAontheAPradiograph[9].Refinedandupdatedradiographicmeasurementshavebeenproposedthatwouldenablebetteridentificationofchildrenwithfocaldysplasia,suchastheanteriorwallindexandposteriorwallindex[48].Earlystudiesdemonstrateddifferencesintheanteriorandposteriorwallindicesinsymptomaticdysplasticpatientscomparedtothosewithanormalacetabulum[48].Subsequentinvestigationrevealedthatevenasymptomaticpeoplehaveradiographicanteriorandposteriorwallindexmeasurementsthatoverlapthoseofpeoplewithdysplasia[49].Thisindicatesthatsomedegreeofvariationinthe3-Dmorphologyoftheacetabulumisnormal.Furthervalidationoftheseindiceswithcross-sectionalimagingandlongitudinalfollow-upisnecessarybeforethesenewreferencestandardscanbeconsideredreliableindicatorsofdisease.Inthepresenceofinstabilityrelatedtoacetabulardysplasia,thereisoftenovergrowthofsoft-tissuestructuresthatcompensatesforthedeficientbonysupport.Thisincludesenlargementoftheacetabularlabrum[50,51],evenintheabsenceoflabraltearordegeneration.Focalmuscleenlargementhasalsobeennotedinunstablepatients,specificallytheiliocapsularismuscle[52].MRImightbeusefulinidentifyingthesesecondarysignsofinstability.Still,noclearlypositivefindingsconfirmthepresenceofinstability.Femoralversionisalsoakeycomponentindetermininghipstability,asanantevertedfemurisbemoreanteriorlyuncoveredthananeutralfemur[53].Overtfeaturesofacetabularrimdamagealsosupportthediagnosisofdysplasia,includinglabraldegenerationandtearingandcartilageloss,thoughadolescentswhohaveborderlinedysplasiamightnotyethavevisiblemanifestationsofosteoarthritis,evenifinstabilityisthepaingenerator.Giventheseareasofinvestigationanduncertainty,theborderlinedysplastichiphasattractedwell-deservedattentionintheliterature.Specificconcernshavebeenraisedaroundwhethertheterm“borderlinedysplasia”isanadequatelabelandwhetherthisisasinglecondition.ItismorelikelythatchildrenwithaborderlinelateralCEAof18–24°consistofclustersofpatients,someofwhommighthavecamimpingement,andsomeofwhomhavefocalacetabulardeficiency[13].Specificpatternsofacetabulardeficiencyandfemoralmorphologyarebestcharacterizedwithcross-sectionalimagingexaminations,suchasMRIorCT,whichmightalsodetectothercausesforhippain.ItisrecommendedthatanychildinwhomthereisconcernfordysplasiaundergoanMRIaspartofacompleteevaluationbecausetheremightbeanotherfindingthatexplainsthechild’ssymptoms(Fig.6),orevidenceofintra-articulardamagethatsupportsrimloading.MRIiswellsuitedfordetectingcartilageandlabralabnormalitiesaswellasmarrowlesionsthatarepresentinthesettingofalteredbiomechanicsandearlyosteoarthritis[8].Low-dosepelvicCTisalsovaluableforpreoperativeassessmentofthehipmorphologywithprecisecharacterizationofthebonydeficienciesinthreedimensions.Itiscrucialthatthesechildrenaredirectedtoanexperiencedhipspecialistwhoisabletocontextualizetheimagingfindingswithacomprehensivephysicalexamandanappropriatehistory.Thesechildrenshouldbeinterrogatedwithrespecttothenatureandlocationoftheirpainwithspecificquestionsaroundinstability.Acomprehensivephysicalexamshouldfollow,includingattentiontothechild’sgait,pelvicpositionandrotationalprofile.ConclusionTheterm“borderlinedysplasia”referstopatternsofacetabularcoveragethatmightpredisposechildrentoinstability.Somechildrenwithborderlineradiographicmeasurementshavenormaljointmechanicsandfunction(Fig.1ShouldsayFigure3),whileothersbenefitfromacetabularreorientingsurgery(Fig.2ShouldsayFigure4)[11].Itisimportantthatradiologistsreflectthisuncertaintyintheirreportswithappropriatemanagementrecommendations.Instabilitymightbesuggestedbyradiographs,butultimatediagnosisisconfirmedonthebasisofhistoryandphysicalexamassessingforinstability.Thesechildrenshouldundergohigh-resolutionMRimagingofthehiptoevaluateforjointdamage,andCTmightbehelpfulforbetter3-Dcharacterizationofthebonyshapeandcontour.Dynamicultrasonographyhasbeenvalidatedasahelpfuladjunctinthedeterminationofinstabilitybymeasuringanteriorfemoralheadtranslationwithdynamicmaneuversreplicatingtheapprehensiontest[54].Aspediatricradiologists,weareoftenthefirsttoidentifyborderlineormilddysplasiaonthebasisofradiographs.Itisimperativethatweserveaseffectiveconsultantsandofferappropriaterecommendationsaspartofacohesivemultidisciplinaryapproachtothiscomplexpatientpopulation. 文献出处：SarahDBixby,MichaelBMillis.Theborderlinedysplastichip:whenandhowisitabnormal?ReviewPediatrRadiol.2019Nov;49(12):1669-1677.doi:10.1007/s00247-019-04468-4.Epub2019Nov4.  IntroductionDevelopmentalhipdysplasia(DDH)isoneofthemostimportantandmostcommonpediatricmusculoskeletalconditions.Whileasmanyas80%ofcasesarepresentatbirth,manyremainundiagnosed.Whenpresentininfancy,DDHmightbedetectedonthebasisofphysicalexaminationfindings(i.e.BarlowandOrtolanimaneuvers)andstaticanddynamicultrasoundfeaturesoriginallydescribedbyGraf[1,2].Thelong-termimplicationsofDDHaresignificantbecausetheconditionleadstodevelopmentofosteoarthritisin25–50%ofpatientsbytheageof50years[3].Thereducedsizeandtheincreasedobliquityoftheacetabularweight-bearingsurfacecreateshearingforcesonthearticularcartilageandcausechronicoverloadingoftheanteriorandanterolateralacetabularrim[4].Thismechanicaldysfunction,ifuncorrected,leadsinadulthoodtopain,abductorfatigueandoftensymptomsofinstability,culminatingingradualfailureofthecartilageandleadingtoprogressiveosteoarthritis.Treatmentstrategiesdependonthemechanicalstabilityofthehipandthetypeanddegreeofbonydeformity.IninfantswithmildDDH,capsularlaxityandmildacetabulardysplasiaaretheissues,andsimplepositioningofthehipsinabductionandflexioninaprotectivebraceorPavlikharnessusuallyleadstotighteningofthecapsuleandresolutionofthedysplasia.Inchildrenwithfullcongenitaldislocations,particularlyifdiagnosedafterinfancy,aformalmanipulativereductionmightberequired,withspicacastingforseveralmonths.Atanyage,treatmentisfocusedonreducingandmaintainingthefemoralheadtoaconcentricpositionwithintheacetabulum.IfbonymalalignmentispresentintheolderchildwithDDH,realignmentsurgeryisoftenneededtorestorestability.Infantsandchildrenwithdevelopmentalhipdysplasiamightbetreatedtocurebyvirtueoftheseearlystrategies,ortheymighthavepersistentsubluxationthatrequiresfurthersurgerylaterinadolescenceoryoungadulthood.DysplasiainadolescentsAdolescentsandyoungadultswhohadbeenasymptomaticwithrespecttothehipmightalsohavemildformsofacetabulardysplasiadetectedonradiographsbasedoncriteriaoriginallydefinedbyWiberg(Figs.1and2)[5].Inmanychildrentheindicationforradiographsishippain,thoughforsomechildrenradiographsareperformedforotherindications.Intheabsenceofadedicatedexaminationbyahipspecialistwhocanassessforsignsorsymptomsofinstability,itisnotknownwhethermildorsubtleradiographicabnormalitiesarethesourceofthechild’ssymptoms.Radiographicmeasurementsoffemoralheadcoverageandpositionmightsuggestthepossibilityofmechanicaldysfunctionofthehip,thoughitisthemechanicsthatdefinetheunderlyingdisease,nottheradiographs.IncontrasttoinfantileDDH,adolescentdysplasiahasahighermalepredominanceandismoreoftenbilateral[6].Thedifferenceindemographicsbetweengroupshaspromptedmanytoquestionwhetheradolescentandinfantilehipdysplasiasaretwodistinctentities.Thestandinganteroposterior(AP)radiographofthepelvisremainsthegoldstandardofimagingforadolescenthipdysplasia,supplementedbyotherviewsincludingthefalseprofileradiograph[4].Anumberofradiographicmeasurementshavebeendescribedthatdefinethedysplastichip,thethreemostfundamentalincludingthelateralcenteredgeangle(CEA),theacetabularindexandtheanteriorCEA[7].ThelateralCEAandtheacetabularindexarebothmeasuredonaproperlypositionedstandingAPradiographofthepelvis,whereastheanteriorCEAismeasuredonthefalseprofileradiograph(Figs.1and2).In1939Wiberg[5]definedanormallateralCEAasbeingover25°,anabnormalangleaslessthan20°(Fig.2)andeverythinginbetweenasuncertain.Theserangeswerevalidatedinsubsequentinvestigations[8,9].SimilarcriteriaexistfortheanteriorCEAasmeasuredonafalseprofileradiograph,wherethisangleisconsiderednormalabove25°,borderlineat20–24°anddeficientbelow20°(Fig.2)[10].Incertainchildrentheanterioracetabularroofinsufficiencyismoreseverethanthelateralroofinsufficiency,andthefalseprofileviewmighthelptoidentifythesechildren,withthecaveatthattheanteriorCEAisthemostdependentonradiographictechnique[11].Theacetabularindexisconsiderednormalat0–10°[7],thoughsomehavesuggestedthatanglesupto13°arenormal[8].Valuesabovethisareconsideredindicativeofdysplasia.BorderlinemeasurementsWhilefloridacetabulardysplasiaisincontrovertiblewhenidentifiedradiographically(Fig.2),thecorrectdiagnosisbecomesmoredifficultwhenthemeasurementanglesareonlymildlyoutofthenormalrange,leadingtothecreationofan“uncertain”or“borderline”category(Figs.3and4).Thesechildrenremainasourceofconfusionandcontroversyamongradiologistsandhipspecialists.Ultimately,achildfallingintoanuncertaincategoryhaseitherahealthyoranunhealthyhip,andthisdistinctiondependsonavarietyoffactorsuniquetoeachchildthatgobeyondasimpleradiographicmeasurement.Thelabel“borderlinedysplasia”hasbeenadoptedtodefineagroupthatfallsintoanuncertainmeasurementcategorywithlateralCEAof18–24°[12–14],inwhomfurtherevaluationisnecessarybeforeadiagnosiscanbemade.Giventhemechanicalbasisforthejointdamageinacetabulardysplasia,itisareasonableassertionthathipswithslightlydiminishedcoveragearepredisposedtojointdamagerelatedtoincreasedwearontheacetabularrim.Childrenwithmilddysplasiaareknowntohaveevidenceoflabralandcartilagedamageathiparthroscopy[15].ThisdoesnotmeanthatallchildrenwithalateralCEAof18–24°developosteoarthritis.Earlystudiesevaluatingtherelationshipbetweencenteredgeangleanddysplasiafocusedprimarilyonhipfunctionratherthanspecificevidenceofjointdamage[16],thoughitislikelythatnormallyfunctionalhipsmightovertimealsohavelabraltearsandcartilagelesions.Delayedgadolinium-enhancedMRIofcartilage(dGEMRIC)measurementsDelayedgadolinium-enhancedMRIofcartilage(dGEMRIC)wasdevelopedtoidentifyhipswithearlybiomechanicaldamagetothecartilagematrixinadvanceofmorphologiccartilageloss[17].Whenintroducedviaintravenousorintraarticularinjection,ananionicmoleculesuchasgadopentetate−2(Gd-DTPA−2)distributesovertimeincartilageinverselytotheconcentrationofnegativelychargedglycosaminoglycans.TheconcentrationofGd-DTPA−2canbeindirectlydeterminedwithmeasurementsofT1andisexpectedtobelowerinnormalcartilagecomparedtodegradedcartilagewithlossofglycosaminoglycans.Thismeasurementisreferredtoasthe“dGEMRICindex”(Fig.3).EvaluationofthedGEMRICindexinpeoplewithnoormilddysplasiarevealedthatthedGEMRICindexofmildlydysplastichipsdidnotdiffersignificantlyfromthatofnormalhips[17].ItshouldbenotedthatinthisstudypeoplewithmilddysplasiaweredefinedbyalateralCEA>15°,whichislowerthanwhatwouldnowbeconsideredthethresholdofmilddysplasia.Thesedatasuggestthatchildrenwithmildorborderlinedysplasiamightnotbeatincreasedriskofdevelopingend-stageosteoarthritis,thoughbecausethesepeoplewerenotfollowedlongitudinallyovertimeitisunknownwhethercartilagedegenerationevolvedovertime.ThedGEMRICindexalsodoesnotaddresswhetherchildrenhavesignsorsymptomsofinstability,whichmightbewhatbringsthemtomedicalattention.RadiographiclandmarksMeasurementsobtainedfromconventionalradiographsrelyonidentificationofpreciselandmarks,includingthecenterofthefemoralhead,themedialmarginoftheacetabularsourcil,andthelateralmarginoftheacetabularsourcil(Fig.1).Thesourcilisnotalwayswell-defined,especiallyinchildrenyoungerthan15years.Inyoungerchildrenitisuncommonforthelateralmarginofthesourciltoalsobethelateralmarginoftheacetabulum,promptingsomeinvestigatorstodevelopa“modifiedlateralCEA”thatincludesonlythescleroticportionoftheacetabularsourcil.ThisisincontrasttothetraditionalCEA,whichismeasuredtothelateralacetabularmargin[18].Therangeofnormaldependsontechniquebecausethemodifiedanglehasalowerstandardrange(15–20°)comparedtothetraditionalangle.Withoutstrictattentiontoradiographiclandmarksthevariabilitybetweenmeasurementscanbeextreme[19–21].Themostreliablemethodistoautomatetheprocesswithcomputer-aidedsoftware,eitherbyincorporatingacomputerizedmeasurementprogram[22]orbystandardizingtheprojectionoftheradiograph[23].Unlessrigorouscriteriaarebeingusedformeasurement,amildlyabnormalmeasurementshouldnotbeconsideredamarkofdiseaseunlessthereareothercompellingimagingandclinicalfindings.Anadhocmeasurementperformedby“eyeballing”theradiographisunlikelytobeaccurate.Itisbesttomeasuremultipletimes,especiallyintheabsenceofcomputer-aidedsystems.NormalvariantsCrossoversignRadiologistsarewisetoproceedcautiouslyaroundthehip,giventhefrequencywithwhichpreviouslyreportedradiographicmeasurementsorfindingshavebeensubsequentlydeterminedtorepresentnormalvariants.AnexampleofthisisthecrossoversignontheAPpelvisradiographs.In2007itwasdemonstratedthatthepresenceofacrossoversignwasahighlyreliableindicatorofcranial(superioracetabular)anteversionoflessthan4°[24].Forreference,thesuperioraspectoftheacetabulumisantevertedapproximately10–14°.Atangleslessthan0°,theanteriorwallislateraltotheposteriorwall,leadingtothecrossoversignonradiographswherethetwowallsoverlap.ItbecamestandardforradiologistsandhipspecialiststocommentonthepresenceofacrossoversignonAPradiographsofthepelvisassuggestiveofacetabularretroversion.Inclinicalpractice,however,thesechildrendidnotalwayshavesignsorsymptomsofretroversion,nordidcross-sectionalimagingconfirmretroversion.Overtime,theliteraturerefutedanassociationbetweencrossoversignandclinicallyconfirmedacetabularretroversion.Evenasymptomaticchildrenwithouthipdiseaseorsymptomatologydemonstratedacrossoversignonawellpositionedradiograph,reflectingvariationsinpatientpositioningaswellasthevariablemorphologyoftheanteriorinferioriliacspine[25,26].CoxaprofundaInsimilarfashion,theterm“coxaprofunda”fellinandoutoffavoralmostasquickly.Coxaprofundaisdefinedaspresentiftheflooroftheacetabularfossaliesmedialtotheilioischialline.Itisconsideredanindirectsignofacetabularovercoverageofthefemoralheadandwasproposedasanimagingfeatureofpincer-typefemoroacetabularimpingementin2007[27].Withthisawareness,radiologistsreadilyofferedthisimagingfindingasevidenceofanunderlyingcondition:acetabularover-coverage.Overtime,coxaprofundawasclaimedtobeanormalradiographicfindingthatdoesnotsupportadiagnosisofpincerimpingement[28–30].CamdeformityFinally,thedefinitionof“cam”deformityinchildrenwithfemoroacetabularimpingementhasbeenasubjectofinterestanddebateformanyyears.Camlesionsarebonyprotuberancesalongthefemoralhead/neckjunctionthatimpingeagainsttheacetabularriminhipflexion.Themostobjectivemeansofmeasuringthesizeofacamdeformityisthealphaangle,anangleformedbyalineconnectingthecenterofthefemoralheadtothecenterofthefemoralneck,andalinefromthecenterofthefemoralheadtothepointatwhichthefemoralneckfallsoutsideabest-fitcirclearoundthehead.Intheearly2000sitwasacceptedthatchildrenwithanalphaangleintherangeof50°likelyhadcam-typefemoroacetabularimpingement(FAI)[31–34].OverthelastdecadetherehasbeenincreasingawarenessthatsomepreviouslydefinedcamlesionsinpeoplewithFAImightbepresentinasymptomaticpopulationswithnohipdisease[35–38].Moreoverapositiveimpingementtest,oftenassociatedwiththepresenceofanteriorFAI,hasbeendemonstratedinhealthyyoungadultswhomightnothaveFAI[39],makingthisanunreliableindicatorofdiseaseinisolationofotherevidence.RadiographictechniqueGiventhatourinterestinidentifyingandtreatingpainfulanddebilitatingdiseaseinchildrenmightoutpaceourunderstandingofnormalanatomicalvariation,radiologistsneedguidelinesforinterpretingradiographsthatrevealanuncertaindegreeoffemoralheadcoverage.Acautiousapproachwouldbetosuggestthepossibilityofborderlinehipdysplasiaandrecommendreferraltoahipspecialist.Thisrecommendationshouldbeperformedwhentheimagingfindingshavebeendeemedreliable,whichrequiresstrictadherencetoproperimagingtechnique.ThefollowingshouldbeassessedoneveryAPradiographofthepelvis:(1)Isthepelvistiltedorrotated?Asaguideline,thedistancebetweenthesuperioredgeofthepubicsymphysisandthecoccyxshouldbe1–3cm[40].(2)Howwelldefinedarethemeasurementlandmarks?Ifthereisdoubtastowherethelandmarksarelocated,themeasurementsarelikelytobeinaccurate.(3)Aremeasurementsperformedusingelectroniccalipersorwithavalidatedcomputer-assistedprogram?Ifperformedbyhand,havetheinitialmeasurementsbeenvalidatedwitharepeatattempt?(4)Isthepatientolderthan15years,andifnotisthemodifiedlateralCEAstandardbeingemployedratherthantheclassiclateralCEA?Theanswerstothesequestionshaveagreatimpactonthereportedmeasures.Iftheimagingtechniqueisadequateandthechildstillfallsintoanindeterminatecategoryoftheborderlinedysplastic(alateralCEAthatfallsbetween18°and24°),thisisstillonlythefirststepinacomplexdiagnosticprocess.Ourunderstandingofhipdiseasehasevolvedconsiderablyoverthelastdecade.Itisimpossibletoaccuratelycharacterizeallofthedifferentpatternsofinstabilityandunder-coveragewith2-Dradiographicviews.RelyingsolelyonthelateralCEAtodeterminenormalversusdeficientcoverageassumesalldysplasiaisglobal,orprimarilyinvolvesthesuperioracetabulum.Wenowknowthatatleastthreedistinctpatternsofacetabulardeficiencyexist:anterosuperior,global,andposterosuperiorinsufficiency[41],andAPradiographsarenotdesignedtodetectafocalanteriororposteriordeficiency.Thepresenceofborderlinedysplasiaonradiographsalsodoesnotconfirmthepresenceofinstability,whichisultimatelywhatleadstosymptomsandjointdamage.Thepresenceorabsenceofinstabilityorimpingementmustbedeterminedthroughcarefulhistoryandphysicalexam,aswellasfromstaticandpossiblydynamicimaging.Thefemoroepiphysealacetabularroof(FEAR)indexhasbeenproposedasausefulradiographicmarkerofinstability(Fig.5)[42].Anotherhelpfulradiographiccluetothepresenceofinstabilityistheupslopinglateralsourcilmargin[43].Alloftheseobservations,however,requirefurthervalidationbeforetheycanbeconsideredreliablemarkersofdisease.Wiberg[5]laidtheimportantgroundworkwithhisseminalarticledescribingthe“normal”lateralcoverageofthefemoralhead,andformanydecadesthoseassertionshavenotbeendisproved;subsequentinvestigationshaveonlysubstantiatedhisoriginalfindings[44–46],thoughthelowerendofthenormalthresholdhasshiftedmoretowardthedysplasticendofthespectrum.Largerpopulation-basedstudieshaverecentlysuggestedthattheserangesaregender-specificandthatmaleandfemalepatientsshouldnotbemeasuredagainstthesamestandard.UpdatedreferencestandardsproposedbyLaborieetal.[39]suggestedthatcutoffvaluesformalepatientsshouldbe21°comparedto20°infemalepatients.Updatedupperthresholdvaluesforacetabularindexaccordingtothisstudywere15°formalesand16°forfemales,comparedtopreviouspublishedthresholdof10°forbothgroups.Accordingtothesenewcriteria,manypatientswhohadpreviouslybeencharacterizedasmildlyorborderlinedysplasticmightnowbeconsiderednormal.Additionally,age,gender,heightandbodymassindex(BMI)havebeenfoundtobefactorsinwhatareconsideredtobenormalrangesofacetabularcoverage,shiftingthelowerrangeofnormalcoveragefurtherintothedysplasticrangeforcertainpopulations[47]. 

问：髋关节发育不良可不可以保守治疗，是否必须手术？答：首先，保守治疗无非控制活动量、控制体重、吃止疼药、加强肌肉锻炼，一定程度可以缓解疼痛症状，但是根本的骨头畸形并没有改变，也就是说，保守治疗只能一定程度延缓病情进展，无法解决根本问题。手术的目的是为了缓解疼痛、延长自身关节的使用寿命，从根本上解决问题。如果本身关节不疼，或者疼痛非常非常轻，可以暂且保守治疗。但是，如果关节疼痛比较频繁或者比较重，手术可能是解决当前问题的最佳方式。 问：保髋术后我的关节能用多少年？答：这个问题很难回答。影响关节使用寿命的因素太多了，比如手术时关节软骨磨损重不重、自身的软骨耐磨程度如何、自身的关节畸形重不重、手术医生的水平好不好、术后体重控制得好不好、术后关节保养的好不好。理论上讲，通过手术纠正畸形可以让关节的使用寿命尽可能延长，最好的效果就是用一辈子，当然，少术患者也有术后几年、十几年后出现关节磨损严重，进而换关节的。总体上讲，找一个靠谱的医生，术后自己好好保养，剩下的就交给天意了。 问：手术后我应该怎么保养自身的关节？答：手术的目的还是希望大家回归正常的生活。有的极端的患者，为了减少关节负重会走极端，比如坐轮椅，甚至少穿衣服，其实大可不必，该干嘛干嘛。如果可以的话，适当避免长时间重体力劳动或者剧烈运动。当然，如果你觉得运动是生命中不可缺少的一部分，那也不用刻意压抑，这一点国外是比较积极的，很多患者手术就是为了后续运动时不疼。当然，如果能把体重控制在理想的区间肯定是最好的。 问：我想手术了，术前应该做哪些准备？答：1、异地就医，提前进行医保备案，具体需要询问当地医保部门；2、准备一副拐杖，肘拐腋拐都可以，调整拐杖高度，练习拄拐单腿走路；3、术前可以按医生的建议进行功能锻炼，改善肌力，加速术后康复；4、带着之前拍的片子及病历；5、酌情准备个人生活物品。 问：髋臼周围截骨手术风险高不高？答：这个手术确实难度很大，被誉为骨科的珠穆朗玛，手术的入门门槛很高，学习曲线很长，目前全国只有为数不多的医生可以做这类手术。记得我在美国学习的时候，看过两个医生做这个手术，一个医生平均需要三四个小时，另一个医生需要6-8个小时。对于我们来说，绝大多术的手术可以在1小时出头的时间完成，手术不但做得快，质量也是绝对有保证。 问：手术需要输血吗？答：这个手术的出血确实偏多，但是随着手术技术的提高和相关药物的应用，再加上手术中使用血液回收设备（可以将出血量的大概一般进行重新回收利用），目前在我中心手术的患者，90%以上的患者不需要异体输血。而且，我们中心现在术前不需要常规备自体血。 问：术后恢复期大概多久？答：手术中我们需要将骨头截断，调整好位置后进行固定，截断的骨头长好需要大概3个月的时间。所以，术后3个月内需要小心保护自己的髋关节，不要摔，一定要拄双拐，拄双拐，拄双拐！一般我会让患者术后6-8周内术腿不负重，6-8周后从0开始逐渐逐渐增加踩地的重量，注意，是匀速逐渐的增加，到3个月的时候可以负重身体重量1/3-1/2，具体以医生通知为准。过早扔拐，过早过多负重可能导致骨头移位，影响手术效果。3个月后门诊复查，评估骨头生长情况。 问：术后如何进行康复锻炼？答：康复锻炼很重要，锻炼不好，走路十有八九会瘸。我的患者我一般会给每人一个康复计划，由于每个人的手术不一样，畸形不一样，骨头质量不一样，所以方案不会完全一样，大家按照自己的方案去做锻炼即可。大家认真阅读锻炼资料，保证动作做对，一旦动作做错，就可能练错肌肉。3个月复查时人要过来，很重要，我会根据查体结果和骨头愈合情况调整康复方案。复查方式参考：保髋术后门诊复查注意事项

发育性髋关节发育不良(DDH)伯尔尼髋臼周围截骨术(PAO):从其在当地的开始到在世界范围内的采用（2023）Berneseperiacetabularosteotomy(PAO):fromitslocalinceptiontoitsworldwideadoption GanzR,LeunigM.Berneseperiacetabularosteotomy(PAO):fromitslocalinceptiontoitsworldwideadoption[J].JOrthopTraumatol,2023,24(1):55.. 转载文章的原链接1：https://pubmed.ncbi.nlm.nih.gov/37917385/ 转载文章的原链接2：https://jorthoptraumatol.springeropen.com/articles/10.1186/s10195-023-00734-2 AbstractThedevelopmentoftheBerneseperiacetabularosteotomy(PAO)isbasedonastructuredapproachstartingwithananalysisofthepreexistingprocedurestoimprovethecoverageofthefemoralheadandwasfollowedbyalistofadditionalgoalsandimprovements.Cadavericdissectionswithadetaileddescriptionofthevascularsupplyofacetabulumandperiacetabularbonesetthestageforanintrapelvicapproach,whichofferedthelargestacetabularcorrectionpossiblecombinedwithsafeintracapsularaccess.Thefinalcompositionofosteotomiesrequiredthedevelopmentofseveralinstrumentsandcuttingdevicesbeforethefeasibilitycouldbetestedonaseriesofcadaverichips.伯尔尼髋臼周围截骨术（PAO）的发展基于一种结构化的方法，首先对现有的手术方法进行分析，以改善股骨头的覆盖范围，然后列出了其他目标和改进措施。通过对髋臼和周围骨的血管供应进行详细描述的Cadaveric解剖，为内盆入路奠定了基础，这种入路可提供最大的髋臼矫正范围，同时确保安全的关节囊内入路。在对一系列cadaveric髋关节进行可行性测试之前，需要开发几种器械和切割装置来完成所需的截骨。Whilethesequenceoftheosteotomiesremainedlargelyunchangedovertime(exceptforthepubicandischialosteotomies),severalpropositionsforaneasier/lessinvasiveapproachhavebeendiscussed;somemadeitintostandardpractice.Effortswereundertakentooptimizethelearningcurveandminimizefailuresusingvideo-clips,hands-oncourses,fellowships,publications,andongoingmentoringprograms.Inretrospect,withalmost40yearsofexperience,sucheffortshavepromotedaworldwideadoptionoftheBerneseperiacetabularosteotomy.虽然骨盆截骨的顺序在长期内变化不大（除了耻骨和坐骨截骨），但人们还是讨论了一些更简单、更微创的方法。其中一些方法已纳入了常规实践。人们还通过视频剪辑、动手课程、研讨会、出版物和持续的导师项目等方式，努力优化学习曲线，减少失败。回顾过去，在近40年的经验基础上，这些努力促进了伯尔尼髋臼周围截骨术在全球范围内的采用。 KeywordsHipjointpreservationsurgery,Periacetabularosteotomy,PAO,BernesePAO,Pelvicosteotomyhistory SimilarcontentbeingviewedbyothersResultsofPeriacetabularOsteotomy(PAO)Chapter©2017BerneseperiacetabularosteotomythroughadoubleapproachArticle10August2018Mini-IncisionPeriacetabularOsteotomyChapter©2017 IntroductionAttemptstoimprovecoverageofthefemoralheaddatebacktothebeginningoflastcentury,whenAlessandroCodivilla,directorofthefamousRizzoliOrthopedicInstituteinBologna,Italy,proposedamethodtotreattheestablisheddislocationofthehip[1,2].Hesuggestedtocoverthefemoralheadcompletelybythejointcapsuleandtointroduceitintoadeepenedacetabulumattheanatomiclevel.Thetechniquewaspopularized30yearslaterbyColonnaandremainsknownunderhisname[3].Theprocedurewasexecuteduntilthelate1990s,whenBoardmanandMoseleyintheirfollow-uppaperconcluded:“wedonotsupportrevivalofthisnowobscureprocedure”[4].Nevertheless,betterunderstandingofthebloodsupplytothehipregionfavoredarevivalofamodifiedversion,nowcalledcapsulararthroplasty[5]. AcetabularaugmentationClassicaugmentationproceduresofthedeficientacetabulumaretheshelf-arthroplasty[6,7]andtheChiariosteotomy[8].Likethecapsulararthroplasty,theyrelyontransformationoftheinterposedcapsulartissuetofibrocartilage.Comparedwithhyalinecartilagethemechanicalqualityoffibrocartilageisinferior,nevertheless,thesurvivalofsucha“neo-joint”canlastuptoseveraldecades.Theexecutionoftheacetabularshelfaugmentationiscomparativelyeasy.Overtimethistechniqueunderwentseveralminormodifications;itisstillinuseinFranceandincountrieswithFrenchorthopedicinfluence[9].ThetransversejuxtaarticulariliacosteotomyofChiariistechnicallysomewhatmoredemanding.Itwasfirstdescribedintheearly1950s[8]andgainedinternationalacceptancefordecades.Despitemultiplearticlesreportinggoodlong-termresults[10,11,12,13],theprocedurelostpopularitywiththeextensionoftotalhipreplacement(THR)indicationtoyoungerpatientsandwiththeemergenceofreorientationprocedures. AcetabularreorientationIntheseprocedurestheentireacetabularboneandhyalinecartilageisredirectedoverthefemoralhead;theglidingsurfaceofthenewarticularrelationwiththefemoralheadremainsallhyalinecartilage.Experiencehasproventhatthereisnearlyalwayssufficientacetabularcartilageavailabletoallowcorrectreorientation.LeCoeurwasthefirsttoexecutesuchanosteotomyin1939,butdidnotpublishitbefore1965[14].Today,severalareofhistoricalvalue,themajorityisusedinthecountryoftheinventor,andfewhavereachedinternationalacceptance. SingleosteotomiesTheSalterosteotomyistheonlyonewherewithatransverseandcompletesingleosteotomyofthesupraacetabulariliumisusedtoreorienttheacetabulum[14].Itispreferredbypediatricorthopedicsurgeonsforitsrathereasyexecution.However,withitspivotpointnearthepubicsymphysis,correctionislimitedandinterlinkedwithsidecorrectionssuchaslateral,anterior,andcaudadshiftofthejoint.Thehighestclinicalrelevance,however,hasthetendencytocreateretroversionoftheacetabulum,whichinhipswithlowornoanteversionmaycreateimpingementproblemsinearlyadulthood.ThePemberton[15]andDega[16]osteotomiesareincompletesupraacetabularosteotomiescorrectingtheroofonly.ThecutissimilarasforaSalterosteotomy[14]butdoesnotseparatetheiliumcompletely.WhilethePembertoncutisstraight,theDegacutiscurvedfollowingtheacetabularroof.Thedeeperthecutpenetrates,theeasiertheacetabularpartofthebonecanbebenddownwards,decreasingtheenlargedacetabularradius.Bothtechniquesareusedalmostexclusivelyinprematurehipswithashallowacetabulum.BotharetechnicallymoredemandingcomparedwiththeSalterosteotomy[17].Comparinglong-termresults,thePembertonosteotomyseemstocreateslightlybetterresultscomparedwiththeDegaandSalterosteotomies[18,19]. DoubleosteotomiesTwodoubleosteotomiesforacetabularcorrectionwithdifferentestimateshavebeendescribedfromSutherlandandGreenfield[20]andfromHopf[21].Thefirstiscuttingthroughthebonenearthesymphysis,thesecondiscuttingthroughtheemptydistalpartofthepaleo-acetabulumincasesofseveresubluxationandsecondaryacetabulum. TripleosteotomiesTheycanbedividedintodifferenttypes:thoseexecutedatadistancefromthejoint,whichimpliesthatthesacrospinalandsacrotuberalligamentsinsertattheacetabularfragmentandthereforelimitthecorrection,andthoseexecutedclosetothejoint,whichimpliestheligamentsarenotattachedtotheacetabularfragmentandthereforeallowahigherspatialcorrection[14,22,23,24]. SphericalosteotomiesTheseareperformedveryclosetothejoint,allowingextensivecorrectionexceptformedialshiftingoftheacetabularfragment.Theexecutionisratherdemanding,andduetotheacetabularvicinityoftheosseouscuts,thefragmentperfusionislimitedtotheobturatorandcapsularvessels.Simultaneousarthrotomyisthereforelimited[25,26,27]. TheBerneseperiacetabularosteotomy(PAO)ThemostrecentlyintroducedreorientationprocedureisthePAO[28].Incontrasttopreviousosteotomies,itisexecutedfromanteriorandfromtheinsideofthepelvis.Theprocedureistheresultofaresearchprojectwithanalysisofthelimitationsoftheexistingprocedures,followedbyextensivecadavericdissectionsofthevascularsupplyandbytrialosteotomiesforbestdirectionoftheosteotomycuts.Finally,thedefinitiveversionoftheprocedurewastestedon25cadaverhipswiththefocustotestasetofnewinstruments.Traditionally,pelvicsurgeryperformedbyorthopedicsurgeonsapproachedtheacetabularareaalmostexclusivelyfromthelateralside.Withtheprogressinsurgicaltreatmentoftraumatotheacetabulumandpelvisinthelate1970sandearly1980sbyLetournel[29],aswellaswithbetterunderstandingofthevascularsupplyofthepelvis,theinnersideofthepelvisbecameaninterestingnewsurgicalapproacharea.Additionally,knownlimitationsofthepreexistingaugmentationandreorientationprocedures(Table1[14,15,20,21,22,23,24,25,26,27,28])raisedtheambitionforamoreversatileprocedureofacetabularreorientationthatshouldfulfillseveralgoals(Table2).  Table1Characteristicsofacetabularreorientationprocedures  Table2Criteriatomeetforanewreorientationprocedure  Thefirststepoftheprojectwasacadavericstudyofthevasculartopographytothepelvisrelevanttoperiacetabularosteotomies.Hereitbecameobviousthatthemostcrucialaspectwasthepreservationofthefragmentvascularityduringtheosteotomiesandduringwideningthegapsrequiredforcorrection.Itwasbestachievedifosteotomieswereperformedfromtheinnersideofthepelvis[28](Fig.1).AfterfirstclinicalintroductionofthePAO,intraoperativelaserDopplerflowmetryoftheacetabularfragmentwasusedanddemonstratedthatthesignalsdropduringthereorientationbutreturntonormalvaluesafterseveralminutes[30].Overall,theseteststookup1year,mainlyduetolimitedavailabilityofsuitablecadavers.Forthefinalsequenceandorientationofthefivenecessaryosteotomysteps,JeffreyMast,aclinicalfellowfromtheUSAatthetime,wasaningeniouscontributortothedevelopmentofthePAO(Figs.2,3).Finally,wepracticedonseveralcadaverhipstobecomefamiliarwithtechniqueandanewsetofinstruments,whichlaterbecamecommerciallyavailable(Fig.4)fromseveralproviders.  Fig.1Periacetabularvascularsupplycadaverstudyonvascularsupplyoftheperiacetabularbone.Righthip,osteotomiesfrompelvicinside.Thebranchesfromsuperiorandinferiorglutealarterycanbepreservedduringosteotomy(left)andfragmentcorrection(right)  Fig.2PAOosteotomycutsarrangementofthefiveosteotomycuts,visibleontheleftlateralview,inthemiddleanteriorview,andontherightmedialviewofarighthemipelvis,namelyincompleteischialcut(first),pubiscut(second),supraacetabularcut(third),retroacetabularcut(fourth),andinfraacetabularcutcompletingtheischialcut(fifth)  Fig.3Pelvisplasticmodel:plasticmodelshowingseparationoftheacetabularfragmentfromposteriorcolumnincludingsacrospinalandsacrotuberalligaments.Top:anteriorview;bottom:posteriorview  Fig.4PAOinstrumentationsetofinstrumentswithspecialosteotomesandretractors  Thefirsthipwasoperatedonon13March1984.Wewerewaitingseveralweeksforan“easy”case,butDrMastcouldnolongerpostponehisflightbacktotheUSA;thiswasthereasonwhywedecidedtotreatthemosturgentcaseonthelist,whichwasaproximalfemoralfocaldeficiency(PFFD)hipwithcomplexdeformity.Thesurgeryofthis13-year-oldgirltook4hand10minandincludedafemoralvalgusosteotomytocompensateforanearliervarusosteotomy.Atotalof3monthsaftersurgery,thehipdislocatedposteriorly,acomplicationthatwassuccessfullytreatedwithaposteriorshelf.Atthattime,wedidnotknowthatPFFDhipshavesevereacetabularretroversionasacharacteristicpartofthemalformation.Thehipfunctionedwellduring33yearsbeforeTHRbecamenecessary(Fig.5).Theresultsofthefirst75hipswithratherheterogeneousformsofdysplasia,multipleprecedentsurgeries,andosteoarthrosisstageincludingTönnisgrade2werepublished4yearslater[28];thesuccessrateofthisgroupdroppedfrom80%after10yearsto60%after20yearsandto30%after30years[31,32].Meanwhile,morehomogeneousgroupsofhipdysplasia,youngerageatsurgery,exclusionofadvancedosteoarthritis,aswellasaconcurrentarthrotomyforfemoralhead–neckshaping,haveledtofurtherincreaseofsurvival[33,34].  Fig.51984:firstoperatedcase—13-year-oldfemalewithPFFDofthelefthip.aVarusproximalfemurafterintertrochantericosteotomy3yearsearlier.bPeriacetabularosteotomyfollowedbyintertrochantericrevalgisationforbetterjointcongruity.Goodlateralcoverageofthehead.Increasedretroversionoftheacetabulumledtoposteriorsubluxation3monthsaftersurgery,whichwastreatedwithaposteriorshelf.cProgressivejointdegenerationafter36years.Plateforshelffixationstillinplace.dFollow-upradiographyofTHR2yearslater.Thepatientregainedanormalgaitpattern  TechniqueThepatientisinsupineposition,thelegtobeoperatedonisdrapedmobile.Generalanesthesiaisstandardasisbloodsalvage(cellsaver).TheapproachisamodificationofSmithPetersen’sprimarydescription[35].Thesequenceofthefivepartialcutsisdescribedelsewhereindetail[28].Someperformthepubiccutpriortothefirstischialcut,especiallywhenarectussparingapproachisused.Theclassicexposurestartswithosteotomyoftheinsertionofthesartoriusoriginattheanteriorsuperioriliacspine(ASIS).Mainlyinyoungerpatients,thesartoriusinsertionincludingtheinguinalligamentarereleasedfromtheASISwithoutanosteotomy.Byincisingthefasciaofthetensorfasciaelataeandmovingawaythemuscletothelateralside,thejointcapsuleisreached.Bydissectingtheiliacussubperiosteally,themedialflapcanbemobilized.Intheclassicapproachthereflectedandthedirectheadoftherectusfemorismusclearedividedandtheiliocapsularismuscleisdissectedfromthecapsule.Followingtheanteroinferiorcapsuletotheinferior,thegapbetweencapsuleandpsoastendonfurtherdowntotheobturatorexternusmuscleiswidenedtoreachtheischium.Thedeepbranchofthemedialfemoralcircumflexartery(MFCA)runsonthecaudadsurfaceofthismuscle,whichneedstoberespected.Initially,foranincompletecutoftheischiumattheinfracotyloidaxilla,aspeciallyangulatedosteotomeisused.Positionandpropagationcanbecontrolledwithfluoroscopy.Intherectussparingapproach,theischiumisapproachedmedialtotherectusfemoris,whichisnotreleased.Thedeepfasciaisincised,andtheischialosteotomyisperformedquitesimilartotheclassictechnique.Incomparisonwiththeclassictechniquetakingdowntheinsertionoftherectus,therectussparingapproachisnotreallyinternervous,becausethefirstbranchofthefemoralnerveiscrossingtheapproachdistally.Thepubiccut,intheclassicversionsecond,isacompleteseparationofthepubisaftersubperiostealdissection;theobturatorneurovascularbundleisprotectedusingtwobluntretractors.Thethirdcutisasupraacetabularhorizontalseparationoftheiliacbone,startingjustdistaltothesartoriusmuscleorigin.Thecutisperformedfromtheanteriorandinnersideoftheiliumwithanoscillatingsaw.Duringthisosteotomytheabductormusculatureisprotectedwithabluntretractor.Thecutstopsabout10mmbeforethepelvicbrim.Forthefourthcut,aretroacetabularcutstartsfromthereandisdirecteddownwardsatanangleof110–120°posteriortothesupraacetabularcut.Again,fluoroscopyhelpstocorrectlypropagatethiscutusinga10mmosteotome.Attheangulationbetweenthethirdandfourthcut,acurvedosteotomeisusedtocuttheoutsidecortex.Atthispointabonespreaderisinsertedjustdistaltotheangulationtoopenupthegap.Witharetractorheldagainstthequadrilateralsurfaceneartheischialspine,theareaforthefifthosteotomycanbeopened.Thecutisexecutedwithaspecialosteotomeandcutsthefinalosseousbridgebetweenthefourthandthefirstcut.Itseparatesthesacropelvicligamentsfromtheacetabularfragment(Fig.3).Whenallcutsareproperlyaccomplished,theacetabularfragmentcanbeseparatedwithacounter-rotatingmoveofthespreaderandaSchanzscrewinsertedintotheiliacpartoftheacetabularfragment.Thedesiredcorrectionistemporarilyfixedwithtwo(−3)Kirschnerwiresandiscontrolledwithfluoroscopyorastandardanteroposteriorradiographoftheentirepelvis.Thelatterispreferredbecauseitalsoguaranteesthenecessaryorthogradepositionofthepelvis. ModificationsoftheapproachovertimeSoonaftertheinitialexperiencewith75PAOs,thedissectionoftheabductormusculaturewasabandonedandsufficientprotectionofthesemusclesduringthethirdcutcouldberealizedwithtunnelingandplacementofabluntretractor[36].Animportantinsightcamefromfollow-upstudies,showingthatsomehipsdevelopedimpingementsymptomsafteracorrectlyorientedacetabulum.Acloserlookrevealedthatconcomitantfemoralhead–neckdeformityisrelativelyfrequentandwasnotproducingsymptomsbeforethePAO-improvedanteriorcoverage.Theproblemcanbedetectedbyassessinginternalrotationinhipflexionduringsurgeryandsolvedwithadditionalarthrotomyforosteochondroplastyoftheanteriorhead–neckjunction.SomeauthorsrecommendarthroscopybeforePAO[34],whichmaycompensateformissingorinadequateMRIandmayeveninfluencetheindicationforPAO.IncomplexdeformitiesnecessitatingthecombinationofaPAOwithafemoralprocedureweareinclinedtostartwiththefemoralprocedureandapproachtheinfracotyloidaxillabetweenmuscles.gemellusinferiorandobturatorexternus.Forbetterdiscriminationitisadvantageoustoexposethisgapbeforethefemoralprocedure;itallowstoperformthefirstischialcutunderdirectviewwiththesciaticnervebeingheldaside[37](Fig.6).  Fig.6Ischialcutviafemoralapproachinhipswithcombinedacetabularandfemoralsurgery.Ontop,accesstotheinfracotyloidgroovebetweenobturatorinternusandgemellusinferiormuscles.Bottom:itallowsoptimalprotectionoftheischialnerveduringexecutionofthecut  Amorerecentmodificationoftheoriginaltechniqueistherectussparingapproach,whichallowsthefirstcuttobeperformedbyapproachingtheischiummedialtotheintactrectustendons.Acadavericstudydescribesthefeasibilityofthismodification;however,intraarticularinspectionandrevisionoftheanteriorinferioriliacspine(AIIS)mightbesomewhatrestrictedandthemostproximalbranchofthefemoralnervecanbeoverstretched[38].Anumberofproposalsdealwithmodificationsoftheapproachwithoutchangingsequenceandconfigurationoftheosteotomy.TheattemptstoexecutethePAOusingmini-invasivetechniques,includingthetranssartoriousapproach[39,40],maybeapplicableforminimalcorrectionsbuthavelimitationswithcomplexdeformities;consequentlythewidespreaduseremainslimited.AnotherpropositionistheuseofamodifiedStoppaorapararectusapproaches;bothhavebeensuccessfullyperformedforanterioracetabularfractures[41];however,theusefulnessfortheexecutionofaPAOhasnotyetbeendemonstrated.Aninguinalextensionoftheapproachmayincreasetheviewonthepubicosteotomyarea,butwasnearlycompletelyabandonedafterseveralcaseshadpostoperativedeepveinthromboses(DVTs)ofthefemoralvein.Fewareusingatwo-incisiontechnique[42],similartotheTönnistripleosteotomy[24].Aprimaryconceptionoftheoriginaltechniquewastoavoidfluoroscopyanduseonlylocallandmarksfortheexecution.However,mostsurgeonstodayusestandardizedfluoroscopycontrolforsomeorevenallosteotomies.Computerassistancehasbeenproposedseveraltimes,butlikeforTHR,sofarithasnotfounditswayintogeneralpractice.Amostrecentversion[43]confirmsthatitmaybehelpfulforthelessexperiencedsurgeon,aconclusionwegainedalreadymorethan25yearsagousingasimilarapproach;itwasgivenupafterashortclinicaltestingperiod[44]. SpectrumoftheprocedureThefollowingselectedcasesmaydemonstratethespectrumofindicationandthereforeversatilityofthePAO.Whiletheprocedurewasoriginallyintendedasamethodforprimaryacetabulardysplasiafromdevelopmentaldislocation(DDH),itbecameusefulsoonafterasanadditionalsurgeryforothercomplexdeformitieswithacetabularparticipation,mainlyPerthesorPerthes-likedisease,butalsoincaseswithfemoralexostosisorsepticarthritiswithfemoralheadnecrosisintherareepiphysealdysplasiaandoccasionallyinearlystageofprotrusionandposttraumaticperiacetabulardeformities.Specialchallengewasconstitutedincasesafterprevioussurgery,suchasretroversionafterSalterosteotomyorwhenare-PAOwasnecessary. Case1A16-year-oldfemalesufferingfromseverebilateralacetabulardysplasiaandcoxavalgawithhighfoveacapitishadmorepainontherightside.Therewasbilateralsubluxationwithfatiguefractureofthelateralacetabularrimontherightside(Fig.7).Thelateralviewoftherighthipshowedsubstantialintraosseousganglionformationintheroof,whiletheabductionviewfeaturedrecenteringofthefemoralheadandsomereductionoftheacetabularfragment.ThecomplexdeformityonbothhipsmadeitnecessarytocombineaPAOwithafemoralvarusosteotomy.Surgerystartedonthefemorallevelinlateraldecubitus.Beforeexecutingtheintertrochantericosteotomy,thefirstischialcutofthePAOwasperformedthroughthisapproachunderdirectview.FortheremainingPAOcuts,thepatientwasturnedintosupineposition.Theacetabularrimfracturereducedspontaneouslyanddidnotneedadditionalfixation.Thelefthipwasoperatedon6monthslater.Healingoftheosteotomies,includingtherimfracture,wasuneventful.Painofbothhipsdisappearedshortlyaftersurgery,allowingforunrestrictedage-appropriateactivities.Theprominentbendofthefemoralimplantcreatedsomediscomfort,andthiswasthereasonwhypartialmetalremovalwasexecuted2yearsaftersurgery.Afollow-upradiographyat5-yearfollow-upshowedaperfectlyreducedandhealedrimfragmentanddisappearingacetabularcysts.  Fig.7Bilateralsevereacetabulardysplasiaof16-year-oldfemale;morepainontherightside.aSeveralsubchondralbonecysts,bestvisibleinalateralview(blackarrow).Displacedrimfracturewithsomereductioninabduction(whitearrow).bPAOcombinedwithfemoralvarusosteotomy.A6-monthintervalbetweensurgeryofthehipswasobserved(rightsidefirst).SlightovercorrectionofthePAOontherightsideforbetterunloadingofthedamagedrimareawasperformed.Partialmetalremovalafter2yearswasperformed.Radiologicalresultobservedafter5years.cCloserlooktothecriticaljointareaoftherighthipbeforesurgeryandaftersugery(d)showingreducedandhealedrimfragment(doublearrow)  Case2A13-year-oldmalewithPerthesdisease(Fig.8).AfterSalterosteotomypaindidnotsubsideandlimpingincreased;bothcouldbeexplainedwithpersistentsubluxation,widenedacetabulum,followedbyadductioncontracture.Computersimulationshowedreasonablereductionofthesubluxatedfemoralheadwithintracapitalosteotomy[45]andPAO.Thepreciseexecutionofthefemoralheadosteotomywasfacilitatedbyacuttingtemplateprefabricatedonthebasisofthecomputersimulationdata.Healingoftheosteotomieswasuneventful.Painsubsidedcompletely;abductortraininghelpedtoregainawalkingpatternwithoutlimping.At1yearaftersurgery,partialmetalremovalhelpedtosettlepainfromprojectingscrewheads.At2yearsaftersurgerytheconfigurationofthehipwasclosetonormal,rangeofmotion(ROM)wasslightlylesscomparedwiththeoppositesideandfunctionwasunrestricted.  Fig.8Perthesdiseaseintherighthipofa13-year-oldmale,operatedonwithSalterosteotomy.aPersistentsubluxationandadductionwithextrusionofthehealthylateralpillarandloadingofthenecroticarea.Adaptivewideningoftheacetabularcavity.Ontheright,computersimulationshowingrelocatedheadafterresectionofthenecroticarea(redportion)andoptimalcoveragewithPAO.bIntraoperativepicturesshowingtheresectionofthenecroticcentralpartoftheheadwiththetemplateinplaceandthefinalsizeoftheheadafterscrewfixationofthemobilelateralpartoftheheadwithtwoscrews.cPostoperativesequenceswiththenewheadhealedwithoutnecrosis  Case3A21-year-oldverygracilefemalewithmultipleexostosesaroundbothhipsandconcomitantacetabulardysplasia(Fig.9).Sheonlyhadpainontheleftsideanddescribeditassudden,happeningfrequentlyduringexternalrotationinfullextension.Thephenomenoncouldbedemonstratedbyultrasoundasimpingementbetweenaposteromedialneckexostosisandtheinfracotyloidischium;thecontactwasfollowedbyfemoralheadsubluxation.Ultrasoundalsorevealedthatthecausativetumorwasbiggerthanitsradiographicappearance.Bothhipsshowedahighcoxavalgaandacetabulardysplasia.Thesurgicalplanwastoremovethenecktumor,correctthehighneckvalguswithavarusosteotomy,andtoimproveacetabularcoveragewithaPAO.Preoperativeplanningbroughtoutthatintertrochantericosteotomywouldnotsufficientlyincreasethepelvifemoralclearance.Ontheotherside,theriskofdamagetothefemoralheadvascularsupplywasestimatedtobehighduringanattempttoexcisethebigtumorviaanteriorand/orposteriorretinaculardissection.Basedontheknowledgethattheposteriorneckisfreefrombloodvessels,itwasdecidedtoexecutetheresectionthroughtheneckosteotomy,whichcouldalsobeusedtoperformtheplannedvarusosteotomy.Again,surgerystartedwithdislocationofthehip.Subperiostealanteriorandposteriordissectionoftheneck,containingthevesselstothehead,wasperformedtowardthebaseoftheexostosis.Itwasfollowedbyamediocervicalosteotomyandwideningofthegapusingaspreader.Theviewwassufficienttoallowsubperiostealpiecemillresectionofthetumorunderconstantobservationofbleedingofthehead–neckfragment.Varuspositionoftheneckwasstabilizedwithtwoscrews.PAOwasexecutedasdescribedearlier.Subluxationofthehipdisappearedimmediatelyaftersurgery.The10-yearresultshowsareasonablehipjointwithawidejointspace.ThepatientispainfreeandthehiphasanormalROM.Sheisseenregularlybyherlocaldoctorwhoconfirmedthattheotherhipisstillfunctioningwell.  Fig.9Acetabulardysplasia,multipleexostoses,andsubluxationmultipleexostosesnearbothhipswithacetabulardysplasiainaverygracile21-year-oldfemale.aPainanddiscomfortofthelefthipduringsubluxation,palpablewithrotationinfullextension.Lateralview(right)showingthecausativeexostosisattheposteriorneck.bPostoperativeresultaftersurgicaldislocation,femoralneckvarusosteotomy,andremovaloftheexostosisthroughtheneckosteotomy,followedbyPAO.cThe10-year-result.Removalofthefemoralscrewsforlocalpainsoonaftersurgery.Bothhipsarepain-free  Case4A19-year-oldrugbyplayerwithpainoftherighthip,renderinghimunabletoparticipateinhisfavoredsport(Fig.10).Historyrevealedthathewasrunoverbyatruckwhenhewas3yearsold.Hesurvivedmultiplefracturesincludingacrushinjurytothepelvicring,alltreatedconservatively.Afterfullrecoveryhehadnofurthercontroluntilrecentlywhenheexperiencedincreasingpainintherighthipduringhissportiveactivities.Radiographyrevealedatypicalposttraumaticdysplasia,apparentlyaconsequenceofthecrushinjuringthetriradiatecartilage.Thisraretypeofdysplasiaischaracterizedbythickeningoftheinnerwall,deformityofthehemipelvis,andretrotorsionoftheacetabulum.Lateralandcaudadgrowthofthephysisleadstolengtheningoftheleg.Thelargeacetabularfragmentinthiscaseisaconsequenceofchronicoverloadoftherimarea.Thesurgicalplanwastocombinetheacetabularreorientationwithexceptionalmedialshiftingofthelateralizedacetabularfragment.Concernsaboutsufficientunloadingofthefragmentinfluencedthedecisiontofixitseparatelywithtwoscrews.Thesupra-andretroacetabularosteotomycutswerelaboriousduetotheunusuallylargediameterofthesupra-andretroacetabularbone.Optimalmedializationoftheacetabulumledtominimalbonycontactbetweenacetabulumandhemipelvis,whichwasthereasonwhythecorrectionofversionwaslessthandesired.Nevertheless,consolidationwasuneventful.Thepatientresumedhisrugbyactivitiesafter9months,butalthoughpainfree,didnotregainthenecessaryaggressivityforthistypeofsportandthereforegaveitup.The2-yearresultshowedperfectconsolidationofosteotomyandrimfragmentwithalargeandcongruentjointspace.  Fig.10Posttraumaticdysplasiaina19-year-oldmale.aTypicalposttraumaticdysplasiaoftherighthipafteracaraccidentatage3.Largeacetabularrimfragment.Laboriousexecutionoftheosteotomyduetothethicksupra-andretroacetabularbone.Screwfixationofthelargerimfragment.PAOwithmaximalpossiblemedialdisplacementoftheacetabularfragment.bRadiologicalresultafter2yearswithhealedfractureandosteotomies  Case527-year-oldfemalewithosteogenesisimperfectaandbilateralprotrusio(Fig.11).Shehadincreasingandconstantpainfromglobalpincerimpingementintherighthip,whiletheleftsideremainednearlypainfree.Radiographicallythejointspaceontherightsidewasgloballynarrowed.ThepatientfavoredjointpreservationaftershehadlearnedthatthelifetimeofTHRwithsuchbonequalityislimited.Finally,sheunderstoodthatjointpreservationforherhipwouldbeelaborateandthatagoodandlastingresultcannotbeguaranteed.Thedirectionofreorientationbeingreversedtotheclassiccorrectionwouldrequirealargerimtrimmingtostartwith.Surgicaldislocationconfirmedareasonablecartilagelayerofthefemoralheadwithsomeosteophytes.Circumferentialtrimming,especiallyattheposteroinferiorrimwascombinedwithsomeosteochondroplastyattheheadneckjunction.WhileexecutionofthePAOcutswaseasy,valguscorrectionoftheacetabulumwasratherdifficult,mainlyduetotheosteoporoticbone.Tobridgethelargesupraacetabularstepasaresultofthefinalposition,astaircase-shapedplatehadtobeused.Healingoftheosteotomywasuneventful;consolidationofthefatiguefracturecouldbeobservedatthe6-weekcontrol.At2yearsaftersurgery,thepatientwaspleasedwiththeimprovedandpain-freeROM.Radiographicwideningofthejointspacewasinterpretedassignofongoingimprovement.  Fig.11Acetabularprotrusiowasobserved,withabilateralprotrusionina21-year-oldladywithosteogenesisimperfecta.Constantpainontherightsidewhiletheleftsidewaspain-freeduringmostdailyactivities.PatientrefusedtogetTHR.Onthetop,preoperativeradiographyshowingfatiguefracturethroughthebottomofthejoint(whitearrow).bResult1yearafterexcessiverimtrimmingandPAOdecoveringthehead.Fatiguefracturehealed(blackarrow)withsubstantiallyreducedpain  Case6A23-year-oldfemalewithsevereunilateralnarrowingofthepelviccavityafteraconservativelytreatedcomplexperiacetabularfractureinchildhood(Fig.12).Thejointlookedrathernormalandhippainorrestrictedmotionwereminimal;herproblemwasconnectedtothedesperatedesiretogetpregnantandastatementofhergynecologistwhoexpressedconcernswhethertheasymmetricnarrowingcouldleadtoadeviationoftheincreasinguteruswithincreasdriskofabortion.Athree-dimensional(3D)modelofthepelvisdidnotgiveaconclusiveanswertothepossibilityofuterusdeviation,butshowedthatthebirthcanalwassomewhatnarrowforanaturaldelivery.Fromanearliercasewithpelvicdeformationandsuccessfuluseofadistractor,itseemedpossibletousesuchasystemforthenecessarylateralanddistalshiftingofthemedializedjointcomplex.TestsurgeryonaplasticmodelofthedeformedpelvisrevealedthatamodifiedPAO,separatingtheposteriorcolumnproximally,wouldallowtolateralizetheacetabularfragmentsufficientlyandrecreateanormalcurvatureoftheinnerpelviccontour.Fortheverticalizedpubicramusanadditionalosteotomynearthesymphysiswasnecessary.ThebestplacefortheSchanz’screwsofthedistractorweretheoppositeiliacbonenearthesacroiliacjointandattheipsilateralfemoralhead.Bestfixationwaspossiblewithareconstructionplateplacedalongthepelvicbrim.Anilio-inguinalapproachwouldallowexposureoftheentirehemipelvis.FortheSchanz’screwneartheoppositesacroiliacjoint,ashortincisionattheiliaccrestwouldbesufficient.Thethreecutsaroundthejointandtheadditionalcutofthepubisnearthesymphysiswereexecutedastestedonthemodel.Manualtractiondidnotdispalcetheacetabularfragmentmuch.Instrumenteddistraction,however,allowedslowbutconstantdisplacementwhilealignmentcouldbeassistedwithinstruments.Theprebendedplatewasfirstfixednearthesymphysisandonthemobilepubissegment.Proximaloff-standingoftheplatecouldbesteadilyconvergedtothestableiliacbonebyalternatedrivinginthescrews.Thistechniqueallowedanoptimalfinalpositioningoftheacetabularfragment.Osteotomyhealingwasuneventful;shortlyaftershebecamepregnant.Shefinallyhadanaturaldeliveryanddidso2yearsaftersurgerywithasecondchild.Atotalof16yearsaftersurgery,thepatient,nowamotherof5children,ispainfreeandlikesoutdooractivitiesverymuch.  Fig.12Posttraumaticprotrusioobservedinpatient.aPosttraumaticprotrusioanddeformationofthehemipelvisina23-year-oldfemale.Hergynecologistdiscussedtheriskofabortionduetodeviationoftheuterus,whichwasthereasonwhyshewasaskingforcorrection.AmodifiedPAOwasfirsttestedonaplasticmodel.Correctionofprotrusioundpelvicdeformationwasassistedbyadistractor.bThenewpositionwasstabilizedwithalongrecoplateplacedalongthepelvicbrim.Ultrasoundwasperformedjustbeforenaturaldelivery.cAt16yearsaftersurgery,sheisnowthemotheroffivechildrenandisveryactiveinoutdooractivities.Thestraightlegistheoperatedone  Case7Thiscase(Fig.13)wasoperatedbyProfessorPauloRego,aformerfellow.Itispresentedtoshowthatthetechniquesarelearnablebyothers.  Fig.13Severedeformity(courtesyProf.P.Rego,HospitaldaLuz,Lisbon,Portugal)oftherighthipina16-year-oldmaleafteropenreductionattheageof1.5years,followedbyavascularnecrosis.aVarusosteotomyattheageof14years,whichdidnotimprovelimpingandpain.bPostoperativeradiographyaftercomplexsurgerywithrelativelengtheningoftheneck,intracapitalosteotomytoreducethesizeofthehead,andsubtrochantericderotationosteotomyfollowedbyPAO.cThe4-yearfollow-upresultwithgoodandpain-freefunction  Complexdeformityoftherighthipina16-year-oldmalewasobserved.HehadopenreductionforDDHattheageof1year,whichwasfollowedbyincompletenecrosisoftheepiphysis.Togetherwithongoingsubluxation,theacetabulumbecameshallowoveraPerthes-likedeformedfemoralhead.Painandlimpingdidnotimproveafterintertrochantericvarusosteotomyattheageof14years.MRIshowedareasonableconditionofthejointcartilageandcomputersimulationrevealedanimprovedheadsphericitywithintracapitalosteotomy[45].Thepreoperativeevaluationalsoexposedahighfemoralneckanteversion.Surgerywasstartedwiththefemoralside:thefirstischialcutofthePAOwasfollowedbyrelativenecklengtheningandintracapitalosteotomyandfinallysubtrochantericderotation.ThePAOwascompletedafterturningthepatientintosupineposition.Thesurgeoncommentedthattheentiresurgerywasdemandingandtimeconsuming;however,theradiographicresultwasconvincing.The6-yearfollow-upradiographyshowedaresultofcorrectionclosetoanormalhip.Thepatientispain-free,hasanormalwalkingpatternandanearlysymmetricalROM(courtesyProf.P.Rego,HospitaldaLuz,Lisbon,Portugal). ComplicationsTechnicalcomplicationsdevelopedfirstandforemostduringthelearningcurve,whichinturnisdependentonthecaseloadovertime.Inaliteraturereviewfrom2006,inabout13publications,theincidenceofmajorcomplicationsrangedfrom6%to37%[46].Inourfirstseriesof500cases,themostfrequentcomplicationwassevereunder-,over-,andintraarticularcorrectioncountingfornearly2%ofcomplications[47].Injurytooneofthemainnerves(femoral,sciatic,orobturator)follows,makinguplessthan1%ofcomplications[48],anumberthatmayhavebeeninfluencedbystrictobservationofrecommendationsfromacadaverstudyaboutsurgicalcircumstancesleadingtosuchinjury[48].In1760casesoftheANCHORgroup,complicationincidencewas2.1%,50%(17outof36)ofwhichweretransient[49].NeuromonitoringduringPAOcanidentifythesurgicalmaneuverendangeringtraumatoanerve,butcannotpersepreventdamagewhentheblowisheavyenoughtocausealaceration[50]. InternationaladoptionoftheBernesePAOThisisprimarilytheresultofmultiplepublicationscomingfromdifferentcentersovernearly40yearssinceitsintroduction.Furtherpromotioncamefromaseriesofinstructionalcoursesindifferentcountriesandfromfellowships,mainlyorganizedinSwitzerlandandtheUSA.Also,theongoingmentorshipforcolleaguesduringtheirlearningperiodshouldberemembered[51].SwitzerlandhasprobablythehighestconcentrationoforthopedicdepartmentsperformingtheBerneseperiacetabularosteotomy,withmostoftherelevantpublicationscomingfromtheBerneseGroup[52].NextaretheUSAandCanada,whereseveralinstitutionsworktogetherintheANCHORgrouptostudyallaspectsofsurgerypreservingthenativehipjoint[53].Oneormorecentersarepracticing,andsomedoreportabouttheirresultswiththeBerneseperiacetabularosteotomyinGermany,England,Denmark,Italy,Spain,Portugal,Chile,Iran,andfinallyChina,fromwherethemostrecentreportwaspublished[54]. FutureopportunitiesandchallengesTheindicationfor(oragainst)PAOisstilloneofthemostdifficultdecisionstobemade.Thereisatendencytoperformthesurgeryevenwhenthetriradiategrowthplateisstillopen,butlittleisknownabouthowfaronecangowithoutriskingsurgery-relatedmalformationafterinjuringthegrowthplate.Interestinglyenough,deepeningoftheimmatureacetabulumforcapsulararthroplastydoesnotseemtoproduceamalformation.Maybemolecularstudiesofthetriradiatecartilagecangivesomeideasaboutitsgrowthbalance.TheindicationlimitforoldercandidatesisindirectlydefinedbytherecognitionthatPAOresultsaftertheageof40yearsarestatisticallylessfavorable.However,maybethedirectreasoncouldbetheincreaseofcartilagedegenerationwithage.However,tousedetailedinformationaboutthestatusofthejointcartilageasoftwarewouldberequiredtoquantifyandlocatethecartilagedamageoffemoralheadandacetabulumseenonanMRI.Inanycase,todaytheTönnisclassificationofhipjointarthrosisaloneisnotagoodparameterfordecision-makingandfollow-upconclusions.WhilethegeometryofthePAOcutsdidnotchange,theapproachalreadywentthroughsomemodifications.Somerepresentaclearadvantage;oneshowedabenefittogetherwithonlyoneminorhandicapduringcadavertesting.Itwouldbehelpfultohavesuchorsimilarstudiesofcomparison,includingthepreoperativeexpenditurefortheminimallyinvasivetechniques.Furtherpropositionsofcomputerassistanceshouldbeanalyzedforexpenditureversusimpact.Progresscanbeexpectedwhenthisdemandingsurgeryisconcentratedinfewcenterswithahighcaseloadandwhenseveralsuchcentersworktogetherandcollecttheirmaterialinacentralregistry.

儿童发育性髋关节发育不良（DDH）是小儿骨科最常见的髋关节畸形，严重影响儿童的健康成长。18月龄以内的DDH患儿由于重塑形的潜能大，通常可以通过吊带或闭合复位、石膏固定的方式有效治疗。随着年龄的增大，18月龄及以上的DDH患儿病理改变进一步加重，通常需要更大矫正幅度的截骨手术。对于学龄前（小于5-6岁）的低龄儿童髋关节畸形，Salter骨盆截骨术(SalterInnominateOsteotomy,SIO)，仍是目前使用最为广泛的手术方式。作为小儿骨科的关键手术技术之一，其对于儿童的各种复杂髋关节畸形，具有强大的矫正能力，可以实现髋臼对股骨头的良好包容、稳定髋关节。该类截骨手术被广泛应用于治疗DDH等各种原因导致的儿童发育性髋关节发育不良，是目前全世界公认的低龄儿童髋关节畸形最有效的手术方法之一。传统Salter骨盆截骨术RobertSalter于1961年发明了Salter骨盆截骨术（salterinnominateosteotomy，SIO）治疗发育性髋关节脱位。SIO以耻骨联合为铰链，可使髋臼向前、下、外恢复覆盖，骨盆截骨处可通过骨块和克氏针固定，从而得到一个稳定复位的髋关节。SIO至今已经沿用了50余年，治疗了大量的发育性髋关节脱位病例，获得了良好的效果！改良原创（Improveandoriginal）国内著名儿童骨科专家李旭教授带领汕头大学广州华新骨科医院儿童骨科团队在传统的Salter骨盆截骨术（SIO）基础上做了进一步的改良，并推广应用于临床工作中，我们将此技术命名为Salter-Li骨盆截骨术（Salter-LiInnominateOsteotomy，SLIO)。区别于传统SIO手术，优势明显华新儿童骨科团队2015年开始，受瑞士TheddySlongo教授的Bernese骨盆截骨术影响，李旭教授带领团队对小儿骨科的经典手术—Salter骨盆截骨术进行了大幅改良，具体改进内容包括：（1）无需劈开髂骨骨骺，避免了远期髂嵴的发育性畸形风险；（2）仅从髂骨内板行骨膜外剥离，提前处理骶髂关节旁滋养血管，大大减少了术中失血；（3） 术中无需剥离髂骨外板的肌肉软组织附着，避免了远期因臀肌损伤导致的步态异常，及髋臼缘出现缺血性改变的风险；（4）特殊的倒“L”形髂骨截骨设计，无需进行髂骨取骨即可获得较经典手术更优的截骨稳定性；（5）特殊的髂骨截骨设计，可以使远端骨块获得更大旋转幅度而不影响固定的稳定性，从而获得较经典手术更大的髋臼指数的矫正和头臼覆盖改善；（6）由于手术操作大大简化，切口长度、手术时间、术中失血量等均明显优于经典Salter手术，手术更加微创，学习曲线亦显著缩短。SLIO相比于传统SIO更加微创，该术式不需要劈开髂骨骨骺，不需要植骨，不需要剥离髂骨外板的肌肉组织，不会破坏臀肌以及损伤髋臼周围的血供。实现在截骨端近端和远端之间的两点接触，对髋臼畸形有更强大的矫正能力。更重要的是术中切口非常微创，不超过3cm，这也是迄今为止同类手术中文献所报道使用的最小切口！！！颠覆你的想象：中国最微创的儿童髋关节开放复位、Salter骨盆截骨术——李旭教授改良Salter骨盆截骨术（SLIO）临床病例Case1:冯2岁，女，右侧发育性髋关节脱位，双下肢不等长手术方式：右侧髋关节开放复位、Salter骨盆截骨矫形内固定、髋人字石膏固定术、右侧阔筋膜张肌、内收肌、腰肌松解手术时间：1.5小时手术出血量：小于100mlCase2:陆2岁，女，左侧发育性髋关节脱位，双下肢不等长手术方式：左侧髋关节切开复位关节腔清理、Salter骨盆截骨矫形内固定、左侧内收肌经皮松解、髂腰肌松解、石膏固定手术时间：1.5小时手术出血量：80ml中国最微创的儿童髋关节开放复位、Salter骨盆截骨手术李旭儿童骨科团队近年来开展改良原创的（微创）骨盆截骨术——Salter-Li骨盆截骨术并大力推广应用于临床，充分体现了团队不断探索、创新的发展理念。上面两个病例在手术中都使用了团队原创改良的Salter-Li骨盆截骨术，手术时间短，术中出血量小，手术切口小于3cm，这也是迄今为止同类手术中文献所报道使用的最小切口，术后患者恢复快，临床症状改善明显，满意率高！“Salter骨盆截骨术”作为小儿骨科的关键手术技术之一，对于儿童的复杂髋关节畸形，具有强大的矫正能力，但由于手术难度大、学习曲线长，目前在国内尚只在具有儿童骨科专科的医院能够独立开展，大多数骨科医师尚未能完全掌握和理解这一手术技术。这些年我们赋能提质，髋关节手术逐步向微创化、创新化、精准化高质量发展，团队对该手术技术已经处于国际领先水平。不再“大刀阔斧”，我院骨科手术迈入微创化时代工欲善其事，必先利其器。华新李旭儿童骨科团队在李旭教授的带领下，始终秉持“复杂手术简单化、常规手术微创化、疑难手术个性化”的创新发展理念，一路披荆斩棘、技术革新，科室医疗技术得到全新的突破，从“大刀阔斧”到“精雕细琢”，华新李旭儿童骨科手术迈入微创化时代，为中国患儿提供更有质量的医疗。迎难而上的勇气，是源于团队加强技术积累和敢于业务攻关的实力和底气。上述技术的开展，只是我们在微创化道路上的缩影，为了更好的服务广大患者，我们将继续迎难而上，往高精尖技术的道路上大步迈进！汕头大学广州华新骨科医院儿童骨科，前身为原南方医科大学第三附属医院儿童骨科团队，由李旭博士于2011年7月份创建成立。2018年12月，李旭博士带领儿童骨科团队迁移至广州华新骨科医院，学科建设迈上新台阶，与国际间的学术交流合作日益增多。2020年5月，美国哥伦比亚大学终身名誉教授——纽约摩根斯坦利儿童医院前小儿骨科主任DavidP.RoyeJr.教授加盟华新儿童骨科团队，以谋求为中国患儿提供更有质量的医疗，和培训中国儿童骨科医生加速走向国际化。广州华新骨科医院李旭儿童骨科团队，为中国医师协会骨科分会儿童骨科学组副主任委员单位，和广东省医师协会骨科分会儿童骨科学组主任委员单位。学科技术实力雄厚、特色鲜明；除常见的儿童创伤和畸形诊治外，李旭儿骨团队对传统国内骨科界鲜少涉足的领域，如脑瘫等神经肌肉型疾病所涉及的四肢、脊柱畸形和运动功能障碍的治疗，可提供世界级水平的高质量服务；2020年9月，国内一流的3D步态实验室也于我院正式落成。医学博士后，主任医师，硕士研究生导师；汕头大学广州华新骨科医院院长；中国著名儿童骨科专家。学术任职：亚太小儿骨科学会（APPOS）理事会中国执委；华裔小儿骨科学会首任主席；中国骨科医师协会第一、二届小儿骨科专委会副主任委员；中华医学会骨科分会小儿创伤与矫形学组委员；广东省骨科医师协会常务委员暨小儿骨科学组主任委员；SICOT中国部小儿骨科专委会副主任委员；中国康复医学会修复重建外科专委会四肢先天畸形学组副组长；Depuy-Synthes南方小儿骨科培训中心主任；AO中国小儿骨科讲师。专业擅长：儿童骨关节系统各种严重复杂创伤和后遗症的手术治疗；儿童先天性髋关节脱位(DDH)、脑瘫、马蹄内翻足、多并指（趾）等各种儿童先天及后天性复杂肢体畸形；各种骨病后遗症、儿童股骨头坏死(Perthes病)、肌性斜颈、骨与软组织肿瘤、各种肢体少见病及疑难杂重症。关注我➕了解更多儿童骨科相关知识！

髋关节发育不良（Developmentaldysplasiaofthehips，DDH）是儿童最常见的肌肉骨骼先天缺陷之一，各地报道的DDH发病率在1‰~3.4‰。DDH是导致髋关节骨关节炎的重要原因，延误诊治或治疗不当，都将严重的影响成年后生活质量。既往文献报告，近40%接受髋关节置换病人和“髋关节发育不良”有关。近年来，全髋关节置换术（THA）是当今最成功的骨科手术之一。多种原因导致的髋部疼痛时，THA都可缓解疼痛、恢复功能并提高生存质量。但是，有个重大的问题是技术本身克服不了的，那就是，“全髋关节置换”的“年限”和初次关节置换时年龄密切相关，与之相反的是，“翻修手术”难度大风险高，但效果反而不如初次置换。因此，对于髋关节发育不良这样长病程的疾患，“保髋”是“全生命周期治疗”中不可或缺的一个重要环节。青少年、儿童期严重型髋关节发育不良/脱位，通过规范手术治疗可以获得满意的中期效果。对于行走期（walkingage）及以前发现的髋关节脱位孩子，可以通过“闭合复位石膏裤固定”保守治疗的方法，不过，存在一定的“残余髋关节发育不良”的几率，部分病例需行二期手术矫正。和很多与发育相关的疾病一样，髋关节脱位同样需要尽可能早的发现、诊断，及时复位，让髋臼-股骨头维持良好对位，才能让儿童髋关节沿着正常“轨道”发育，从而“治愈”它。而早期针对髋关节脱位/发育不良的筛查是解决“早期诊断”的唯一出路。回顾看来，最早从1879年WilhelmRoser、后来Ortolani等医生更多是通过体格检查的方法检查出儿童期“脱位”的髋关节。在我国，以上海新华医院吴守义教授等为代表的小儿骨科前辈也陆续开展了针对髋关节脱位的筛查工作。自上世纪80年代，以奥地利Graf、美国Harcke、挪威的Rosendahl、Terjesen医生为代表，开始利用髋关节超声技术来评估髋关节形态，并逐渐形成了早期髋关节超声检查技术及分型系统。从而，将针对DDH的早期检出时间大大提前。并使基于这项技术的、针对人群的DDH筛查成为可能。奥地利是最早针对DDH实行全民筛查（universalscreeningprogram）的国家之一，他们的数据表明，早期全民筛查可以大大降低后期需手术的比例。从卫生经济学角度来说，整体上也大大降低了政府的医疗支出。我国已从政策层面高度重视，国家卫健委在《健康儿童行动提升计划（2021-2025年）》中再次强调，要“逐步将先天性髋脱位等疾病纳入筛查病种”，从而整体提升儿童骨骼发育健康。然而，西京医院严亚波主任的一项针对儿科、儿保、妇产科、小儿骨科等7个专业医生，共466份有效问卷的调查发现，会对婴幼儿常规进行髋关节查体的比例为37.9%，而常规进行髋关节超声的比例仅为20%。理想是丰满的，而现实是骨感的。目前，以上海、北京为代表的大城市，实行的是基于危险因素的“选择性筛查”或“区域性全民筛查”，不过，仍面临诸如政策、运营协调等多方面的问题亟待解决。康复医学会修复重建外科专委会保髋学组张洪教授、罗殿中教授、程徽和杨劼教授等，在西藏地区克服重重困难、开展针对髋关节脱位的系统性早期筛查工作，星星之火已被点燃。在以天津医院小儿骨科杨建平主任、天津市妇女儿童保健中心刘功姝主任为代表的儿保、小儿骨科医生持续努力下，在上海第六人民医院陈博昌主任等的技术支持下，新生儿期针对髋关节发育不良的早期筛查，和先心病、白内障等一样被纳入“早期筛查项目”。天津市自2008年开始实行全市范围内“全民筛查”，将针对髋关节发育不良的超声筛查纳入到“天津市儿童保健手册”、“天津市预防接种手册”（小红本）内，基于信息化管理，采用“2+1”的模式，确保了筛查覆盖率及良好的质量控制。天津市妇女儿童保健中心潘蕾主任总结了天津市2013~2020年的“天津髋筛经验”，题为“天津市婴儿发育性髋关节发育不良的筛查结果及危险因素分析”的文章发表于2022年《中华骨科杂志》上。新生儿期检出，及时规范治疗，是从临床、影像学上“治愈”髋关节脱位/发育不良的前提。有意思的是，天津市的人群数据显示，经超声确诊髋关节发育不良的孩子中，仅有不足12%存在例如臀位、家族史等“危险因素”。上海第六人民医院陈博昌主任、扬州市妇幼保健院王加宽主任积极探索利用AI技术，提高髋关节超声检查的效率、精准度和同质化。针对髋关节发育不良的早期筛查，意义重大，但仍需多方持续努力、协作。不让每一名孩子输在起跑线上，保髋，真的需要大家一起努力、从娃娃抓起啊。

臀中肌锻炼(2)：渐进式臀中肌抗阻力康复训练，髋臼发育不良DDH/股骨头坏死——保髋截骨术后康复指导作者：JayREbert,PeterKEdwards,DanielPFick,GregoryCJanes.作者单位:SchoolofSportScience,ExerciseandHealth,UniversityofWesternAustralia,Crawley,Perth,WesternAustralia.译者：陶可（北京大学人民医院骨关节科）摘要背景：鉴于臀中肌在骨盆和下肢稳定性中的作用，以及臀中肌无力与许多下肢疾病之间已知的联系，臀中肌康复至关重要。目的：系统地回顾文献并提出一系列循证的渐进式臀中肌负荷练习。证据获取：2016年1月进行了系统文献检索，以确定报告康复锻炼期间臀中肌活动占最大等长收缩(MVIC)百分比的研究。其中包括调查无受伤参与者的研究。对锻炼的类型或方式没有限制，但排除了无法在独立环境中准确复制或进行的锻炼。未将肌电活动标准化为侧卧MVIC的研究被排除。根据运动类型和%MVIC对运动进行分层：低（0%至20%）、中（21%至40%）、高（41%至60%）和极高（＞61%）。证据综合：本次综述纳入了20项研究，报告了33项练习（以及同一练习的一系列变体）的结果。俯卧、四足和双侧桥式练习通常产生低或中等负荷。据报告，特定的髋部外展/旋转练习为中等、高或极高负荷。存在对侧肢体运动的单侧站立练习通常是高负荷或极高负荷的活动，而一系列功能性负重练习则存在高变异性。结论：这篇综述概述了康复环境中常用的一系列练习，根据运动类型和臀中肌激活程度进行分层。这将有助于临床医生从术后早期到康复后期为患者量身定制臀中肌负荷方案。ASystematicReviewofRehabilitationExercisestoProgressivelyLoadtheGluteusMediusAbstractContext:Gluteusmediusrehabilitationisofcriticalimportancegivenitsroleinpelvicandlowerlimbstability,andtheknownlinkbetweengluteusmediusweaknessandmanylowerlimbconditions.Objective:Tosystematicallyreviewtheliteratureandpresentanevidence-basedgraduatedseriesofexercisestoprogressivelyloadgluteusmedius.Evidenceacquisition:AsystematicliteraturesearchwasconductedinJanuary2016toidentifystudiesreportinggluteusmediusmuscleactivityasapercentageofmaximalvolitionalisometriccontraction(MVIC),duringrehabilitationexercises.Studiesthatinvestigatedinjuryfreeparticipantswereincluded.Norestrictionswereplacedonthetypeormodeofexercise,thoughexercisesthatcouldnotbeaccuratelyreplicatedorperformedwithinanindependentsettingwereexcluded.StudiesthatdidnotnormalizeelectromyographicactivitytoasidelyingMVICwereexcluded.Exerciseswerestratifiedbasedonexercisetypeand%MVIC:low(0%to20%),moderate(21%to40%),high(41%to60%),andveryhigh(＞61%).Evidencesynthesis:20studieswereincludedinthisreview,reportingoutcomesin33exercises(andarangeofvariationsofthesameexercise).Prone,quadruped,andbilateralbridgeexercisesgenerallyproducedlowormoderateload.Specifichipabduction/rotationexerciseswerereportedasmoderate,high,orveryhighload.Unilateralstanceexercisesinthepresenceofcontralaterallimbmovementwereoftenhighorveryhighloadactivities,whilehighvariabilityexistedacrossarangeoffunctionalweight-bearingexercises.Conclusions:Thisreviewoutlinedaseriesofexercisescommonlyemployedinarehabilitationsetting,stratifiedbasedonexercisetypeandthemagnitudeofgluteusmediusmuscularactivation.Thiswillassistcliniciansintailoringgluteusmediusloadingregimenstopatients,fromtheearlypostoperativethroughtolaterstagesofrehabilitation.Jiànjìnshìtúnzhōngjīfùhèkāngfùxùnliàn文献出处：JayREbert,PeterKEdwards,DanielPFick,GregoryCJanes.ASystematicReviewofRehabilitationExercisestoProgressivelyLoadtheGluteusMedius.ReviewJSportRehabil.2017Sep;26(5):418-436.doi:10.1123/jsr.2016-0088.视频资料来源:Youtube，Googleimage.