➢ The goal of treatment of arthrogrypotic foot deformities should be to achieve a plantigrade, pain-free, and functional platform that is suitable for standing, walking, or at least standard shoe wear.
➢ Treatment should be started as early as possible, before adaptive contractures and bone deformities increase the overall stiffness of the feet.
➢ Manual corrections and casting based on the Ponseti principles, followed by minimal additional operative treatment, can lead to successful results in patients with all types of arthrogrypotic foot deformities and should be the standard initial treatment.
➢ Recurrent deformities causing obstacles to everyday function should be addressed with use of the minimum operative intervention, depending on the kind of deformity.
➢ Individually adjusted bracing as well as intensive physical therapy are recommended to maintain correction.
The term arthrogryposis is used to describe >300 conditions of differing etiologies and clinical manifestations that are characterized by generally nonprogressive congenital joint contractures of more than two different body areas, which occur in one per 3000 live births1,2. So-called classic arthrogryposis (amyoplasia or arthrogryposis multiplex congenita) constitutes about 30% of all congenital contractures and usually presents with severe, multiple upper and/or lower-extremity contractures and muscle weakness of variable intensity; in distal arthrogryposis—a subgroup of autosomal dominant conditions without underlying general neuromuscular pathology—the hands and feet are mainly affected, whereas contractures around proximal joints are absent or limited in their presentation3. Foot deformities are common in all types of arthrogryposis, especially amyoplasia (prevalence, 80% to 100%), usually presenting as severe contractures, which require complex, and often repeated, orthopaedic treatment. If untreated, foot deformities constitute an obstacle to achieving independent standing and/or functional walking and may become painful, fixed deformities causing difficulty with shoe fitting and other everyday care problems4-9.
The general aim of treatment of arthrogryposis multiplex congenita in children focuses on improvement in the quality of life by increasing abilities in independent functioning. To achieve this goal, individualized and holistic treatment strategies—based on the triad of conservative treatment (e.g., manual corrections and casting, physical and occupational therapy), a broad array of operative methods, and orthotic management preventing recurrence of deformities and/or improving joint passive range of motion— are needed to address contractures of both upper and lower extremities.
Foot deformities in patients with arthrogryposis multiplex congenita are characterized by stiffness, resistance to obtaining successful correction, and a tendency to recur. There is a high rate of unsuccessful results necessitating repeat operative procedures6-8.
With the introduction of the method of Ponseti et al. for the initial treatment of idiopathic clubfoot, the results improved substantially worldwide, making this protocol the treatment of choice—the so-called gold standard—for the treatment of classic idiopathic clubfoot as well as so-called neglected or recurrent complex deformities10-12.
Starting in 2008, there have been reports of promising early results of the application of the Ponseti method for the treatment of foot deformities associated with arthrogryposis multiplex congenita13-19. Those reports may potentially change the traditional orthopaedic approach to the treatment of this condition.
This review will provide an update on contemporary orthopaedic knowledge regarding foot deformities associated with arthrogryposis multiplex congenita and recommendations for their treatment.
Prevalence and Treatment Aims
Clubfoot remains the most common congenital deformity of the foot in children, constituting approximately 70% of all abnormalities of the foot in arthrogrypotic syndromes and 98% of those in classic arthrogryposis multiplex congenita7-9,20. Other conditions related to arthrogryposis include vertical talus (prevalence, 2% to 12%), which is more frequently noted in patients with syndromic arthrogryposis, and, rarely, isolated equinus (prevalence, 1%), metatarsus adductus, cavovarus, equinovalgus, and calcaneovalgus5-9,20,21.
The aim of treatment of arthrogrypotic clubfoot was clearly described in 1970 by Lloyd-Roberts and Lettin as conversion of the deformed foot into a plantigrade, braceable, and painless platform that enables the patient to stand and, if possible, permits functional walking22. This goal of treatment remains true today; however, there are important recent developments—specifically, the need to consider patient satisfaction and the need to limit the number and extent of operative interventions. D’Souza et al. underlined that, in cases of arthrogrypotic clubfoot, the presence of mild equinus and mild forefoot deformity are still compatible with a good result23. Cassis and Capdevila included patients’ opinions in the final assessment of treatment and reported that a pain-free foot with the possibility of wearing standard shoes and overall patient satisfaction were enough to merit a good result despite the presence of mild residual deformity (up to 5° of equinus on clinical examination)24. Therefore, in patients with arthrogryposis multiplex congenita, residual mild or moderate foot deformities do not necessarily constitute an indication for further operative treatment if the foot is pain-free and functional and the patient remains satisfied. In some patients, this allows the avoidance or delay of repeated operative interventions, which are known to have adverse effects, including increased scarring and stiffness5-8. To avoid these adverse consequences of excessive operative procedures, van Bosse et al. postulated that orthopaedic surgeons should attempt to achieve the desired outcomes with as few procedures as possible and with use of the least ablative procedures possible16.
Lessons from the Past
For decades, it was believed that manual correction of the feet and casting were prolonged and ineffective methods for the treatment of arthrogryposis multiplex congenita. Therefore, it was suggested that foot treatment should be delayed until after knee or hip contractures had been corrected, at which time aggressive operative treatment of the foot, including extensive soft-tissue releases or primary talectomies, was recommended14. Long-term follow-up studies showed that the results of this strategy were often disappointing: the number of recurrences was high, with a mean of 4.9 procedures per foot being necessary to achieve a final satisfactory outcome. For example, Drummond and Cruess reported recurrence in seventeen of twenty-three clubfeet after posterior release and in six of eleven clubfeet after primary talectomy5. After primary posteromedial release, Niki et al. reported recurrence in thirty of forty-one clubfeet8, Segal et al. reported recurrence in eleven of sixteen clubfeet25, and Carlson et al. reported recurrence in nineteen of twenty-six patients26. Final clinical examination often revealed stiff and painful feet, and high complication rates were reported.
In long-term follow-up studies, the rate of satisfactory and permanent results after primary soft-tissue releases has ranged from 21% to 75%. Specifically, Lloyd-Roberts and Lettin reported a satisfactory result for three of fourteen feet22, Drummond and Cruess reported a satisfactory result for six of twenty-three feet5, Niki et al. reported a satisfactory result for eleven of forty-one feet8, Södergård and Ryöppy reported a satisfactory result for sixteen of fifty-two feet7, Widmann et al. reported a satisfactory result for nine of twelve feet27, and Carlson et al. reported a satisfactory result for seven of twenty-six children26. Higher rates of successful treatment of arthrogryposis multiplex congenita deformities were reported after aggressive circumferential soft-tissue releases, with tendon resections being performed before walking age27-29. In studies of primary talectomy with more than five years of follow-up, satisfactory outcomes were achieved in only about half of the patients. Specifically, Hsu et al. reported a satisfactory outcome for nine of fifteen feet30, D’Souza et al. reported a satisfactory outcome for eleven of twenty-one feet23, and Green et al. reported a satisfactory outcome for nineteen of thirty-four feet31. In conclusion, neither primary soft-tissue releases nor primary talectomies have proved to be superior, and final long-term results after both techniques are far from satisfactory.
Initial Treatment with the Ponseti Method
Manipulations, Casts, and Achilles Tendon Release
Before any treatment is begun, an initial assessment of the severity of the arthrogrypotic clubfoot deformity should be performed according to the scoring systems of Diméglio et al.32 or Pirani et al.33. Such an assessment should also be done during every clinical review in the outpatient clinic to provide proper monitoring of clubfoot correction.
To date, we are aware of seven reports presenting the relatively short-term results of the Ponseti technique for the treatment of a total of 119 clubfeet in patients with amyoplasia and/or distal arthrogryposis13-19. The authors of those reports supported the thesis that conservative treatment of arthrogrypotic clubfoot should be started as soon after birth as possible, before the occurrence of adaptive changes of bones and periarticular soft-tissue contractures8. Despite severe initial deformities, all of the arthrogrypotic clubfeet in those reports responded to the Ponseti manipulation and casting method, allowing for correction of all clubfoot components except equinus. However, the number of casting sessions needed to obtain correction was higher among patients with arthrogryposis multiplex congenita than among those with idiopathic clubfeet (mean, 7.4 compared with 5.5)16.
Percutaneous Achilles tendon release was necessary in almost all arthrogrypotic patients (115 of 119 clubfeet), but, again, post-tenotomy ankle dorsiflexion was less among those patients than among those with idiopathic clubfeet (mean, 5° compared with 20°) (Fig. 1)12,14,15,19. In cases in which equinus persisted, postoperative weekly casting sessions were proposed, instead of classic three-week immobilization, to gradually increase ankle dorsiflexion14. Morcuende et al. noticed that ankle dorsiflexion that remains limited after Achilles tendon release may improve over time with use of a foot abduction brace14. On the basis of those promising short-term results, the authors concluded that the method of Ponseti manipulations and castings followed by Achilles tendon release should be considered as the recommended initial treatment for clubfoot deformities in patients with amyoplasia.
Modification of the Initial Ponseti Treatment
Van Bosse et al. proposed that, for selected patients, percutaneous Achilles tendon release should be performed with local anesthesia initially, at the very first casting session in the outpatient clinic16. This maneuver should unlock the calcaneus from the posterior aspect of the tibia and allow for easier subsequent classic Ponseti manual corrections. If this maneuver and subsequent manual corrections resulted in satisfactory foot positioning (i.e., the heel in valgus, the forefoot abducted and not supinated, and the ankle possibly passively dorsiflexed beyond a plantigrade position), the last cast was applied for three weeks. If any equinus remained after the manual corrections, repeated Achilles tendon release was performed, followed by immobilization in a long-leg cast for three weeks. In the study by van Bosse et al., this approach was used for the treatment of nineteen clubfeet in arthrogrypotic patients, and repeated Achilles tendon release was necessary in ten16.
Recurrences During the First Twelve Months After Initial Treatment
In cases of early recurrences, a supplemental, individually tailored treatment approach is recommended34, usually beginning with repeated Ponseti casting sessions and subsequent repeated Achilles tenotomy if passive ankle dorsiflexion cannot be achieved with conservative treatment. Van Bosse et al. proposed that, if Achilles tendon release alone is not sufficient to achieve satisfactory ankle dorsiflexion, percutaneous posterior ankle release can be performed by inserting a hemostat clamp into the ankle through a stab incision and spreading it in order to disrupt the posterior capsule16. However, it should be noted that more thorough posterior release—from the tip of the medial malleolus to the tip of the lateral malleolus—might be difficult with this maneuver; therefore, in cases of persistent equinus, one should consider performing open posterior or posteromedial releases14-16.
Bracing and Physical Therapy After Ponseti Casting
Properly fitted orthoses that are worn for walking and during the night can decrease the recurrence rate8; however, the method of post-tenotomy bracing in arthrogrypotic patients remains a subject of debate.
Morcuende et al. and Boehm et al. advocated the use of foot abduction braces11,13,14. In contrast with idiopathic clubfeet, 70° of external rotation of the shoes can rarely be achieved in patients with arthrogryposis multiplex congenita; therefore, less abduction (50° on average) is required, along with less pronation, because of decreased ankle dorsiflexion13. The disadvantages of the foot abduction brace for patients with arthrogryposis multiplex congenita were emphasized by van Bosse et al.16 and Kowalczyk and Lejman15, with the foot abduction brace being considered as an obstacle during physical therapy applied to other contracted areas. The use of such braces was often difficult because of coexisting hip and knee contractures. Moreover, passive stretching of the gastrocnemius-soleus muscle resulting from active hip and knee movement cannot be expected in cases of amyoplasia. Finally, noncompliance with the foot abduction brace may lead to recurrence16. Because of those inconveniences, individually tailored unilateral orthoses, such as solid or dorsiflexing ankle-foot orthoses or knee-ankle-foot orthoses allowing for better rotational control and simultaneous correction of concomitant mild to moderate knee flexion contractures, combined with stretching exercises to increase passive ankle range of motion, may be considered for arthrogrypotic patients (Fig. 2).
Exercises for improving ankle dorsiflexion should be started as soon as immobilization is discontinued; these exercises should be part of the general physical therapy program and should be performed not only by the therapist but also daily by well-informed parents15. In addition, educational and support programs designed for parents of children with arthrogryposis multiplex congenita, provided by trained nurses, seem to be beneficial for improving patient care and are reported to diminish the rate of noncompliance with foot bracing13.
Treatment of Residual and Recurrent Deformities
When planning any treatment to correct deformities due to arthrogryposis multiplex congenita, one should keep in mind the goal of treatment—specifically, the achievement of plantigrade, braceable, pain-free feet with the least intervention. Recurrent or persistent deformity causing pain and/or obstacles in everyday function should be considered for treatment; however, operative treatment can be avoided or delayed in selected cases involving mild deformities (mild equinus, forefoot abduction, or supination) that do not disturb function, bracing, and/or walking.
Moderate recurrences caused by soft-tissue contractures can be corrected or diminished by repeated casting sessions and subsequent adequate bracing14,15. Tendon transfers are not performed routinely in patients with amyoplasia as the muscles are usually too weak to function or are completely inactive5; however, Morcuende et al. proposed tibialis anterior muscle transfer from the first metatarsal to the third cuneiform in cases of persistent flexible forefoot supination as well as in cases of idiopathic clubfoot; this procedure was believed to balance the foot and to prevent further recurrences14. Posterior or posteromedial releases, accompanied by tendon releases rather than lengthenings, remain standard treatment for moderate equinus or equinovarus deformities (Fig. 3)3,5,8,22. On the basis of our experience, tendon lengthenings in arthrogrypotic clubfeet are considered only in cases of mild deformities, and only if active muscle movements are present preoperatively.
In patients with more complex deformities associated with stiff joints, even extensive soft-tissue releases may not be sufficient to provide satisfactory correction, and therefore osseous procedures should be considered.
Excision of cancellous bone from the talus and cuboid, followed by forceful manipulation to correct foot deformity by collapsing the remaining cortical shell of the bones, is known as the Verebelyi-Ogston procedure. Gross reported that this procedure led to satisfactory outcome in four of four arthrogrypotic clubfeet35, and Spires et al. reported that it led to a satisfactory outcome in eleven of thirteen arthrogrypotic clubfeet36. In 2011, Iskandar et al. described a similar method, modified by the addition of decancellation of the anterior aspect of the talus, for the treatment of recurrent arthrogrypotic clubfeet37. The authors reported excellent and good results in twelve of fifteen clubfeet after a minimum duration of follow-up of three years. Despite those positive results after relatively short follow-up (mean, fifty-seven35, forty-five36, and forty months37), a possible disadvantage of this method should be underlined: specifically, collapse of the tarsal bones disturbs their anatomy and the congruency of adjacent joints, which may result in severe osteoarthritic changes in the long term. Moreover, in two of these series36,37, residual deformities often persisted, necessitating repeat surgery; in the study by Spires et al., for example, repeat surgery was necessary in six of thirteen feet.
In the past, as a result of disappointing results following soft-tissue releases and frequent recurrences of the deformity, complete excision of the talus (talectomy, astragalectomy) via a lateral approach gained popularity both as a primary procedure and as a secondary salvage procedure for the treatment of arthrogrypotic clubfeet5,6,22-25,30,31,38,39 (Fig. 4). Most authors agreed that this method was effective, especially in cases of recurrent deformity, with the rate of satisfactory results reported to be as high as 70%23,24,30,31,38. Mild recurrent equinus or forefoot adduction was observed after some talectomies, but these recurrences rarely caused trouble in everyday living and did not constitute an indication for further operative treatment23,24,30,31. Poor outcomes after talectomy generally were considered to be the result of operative technical errors. To avoid such outcomes, complete excision of the talus, possibly in one piece, should be performed; otherwise, if a large fragment remains in the ankle mortise, the fragment will grow with time, leading to recurrent deformity23,30. Small residual, osteocartilaginous talar pieces are not disabling and do not require treatment24. Soft-tissue release, especially Achilles tendon resection, should be an adjunct to the main procedure23. It is important to translate the calcaneus posteriorly in the mortise in order to create normal contouring of the heel that will allow for standard footwear23,30. D’Souza et al. recommended that the tip of the medial malleolus should be positioned on the navicular level medially and that the tip of the lateral malleolus should be situated close to the calcaneocuboid joint laterally23. Dias and Stern reported that medial soft-tissue scarring and constriction after previous operations led to recurrences and advocated excision of medial constricting scars followed by skin-grafting38. Recurrent forefoot abduction and supination after talectomy may be caused by a fibrotic and constricted tibialis anterior tendon, which should be resected at an early stage of recurrence; otherwise, forefoot osteotomies may become necessary to correct forefoot abduction and/or supination23.
Interestingly, in long-term follow-up studies, spontaneous osseous fusion of tibiotarsal joints has been observed after talectomy in as many as 47% of cases; this finding was noted in eleven of thirty-eight feet in the report by Green et al.31, seven of fifteen feet in the report by Hsu et al.30, and four of twenty-one feet in the report by D’Souza et al.23 (Fig. 4, C). This phenomenon did not prevent a good final result. Moreover, it was considered beneficial as it prevented further recurrence30. On this basis, abrasion of the articular surfaces of the tibiocalcaneal articulation was proposed to promote ankylosis23.
As is true for any operative technique, the disadvantages of talectomy should be noted as well. Talectomy is an ablative procedure, is unable to correct forefoot deformities, and increases the risk of future arthritic changes. The need for further operative treatment has been reported in up to 67% of patients in long-term follow-up studies. Finally, there are limited possibilities to treat failed talectomies, with wedge midtarsal or metatarsal resections, creation of tibiocalcaneal fusion in older children, and soft-tissue releases accompanied by navicular excision in younger patients constituting the main salvage options30,31,38-40.
Currently, as a result of improvements in the initial treatment of arthrogrypotic clubfeet according to the method of Ponseti et al. as well as improvements in bracing, physical therapy, and soft-tissue releases, there are fewer indications for talectomy. Talectomy remains justified only as a salvage procedure in selected arthrogrypotic patients (e.g., patients with severe neglected deformities).
Multiple operative procedures to correct arthrogrypotic clubfeet may result in stiff deformities, and osseous procedures may be required to treat recurrence. Precise planning should be performed before any osseous procedure, and the deformity should be analyzed both clinically and radiographically. The apex points of all components should be identified, and operative treatment should be directed toward converting a so-called deformed foot into a stiff but plantigrade platform. The methods include metatarsal osteotomies, such as Japas, Akrome dome, or double first-ray osteotomies, to correct cavus and adductus, which occasionally may be supplemented by soft-tissue releases, such as medial and/or plantar release, if needed41-43. Lateral closing wedge osteotomy is recommended to correct hindfoot varus deformity43. Excision of the navicular bone with dorsal cuboid closing-wedge osteotomy, which creates a joint between the talar head and the cuneiforms, was described as an effective salvage operation for the treatment of severe residual cavus in arthrogrypotic patients (Fig. 5)44. Triple arthrodesis alone or in combination with forefoot osteotomies usually is associated with good results when performed as a salvage treatment but should be reserved for children older than twelve years of age5. Tibiocalcaneal arthrodesis has been reported to be an effective salvage procedure for the treatment of a painful tibiocalcaneal junction after talectomy. Nicomedez et al., in a report on seven feet, noted functional improvement in four feet and diminished pain in two; however, osteoarthritic changes were noted in the adjacent knee and midtarsal joints40. Supramalleolar osteotomies are relatively easy procedures to perform and usually result in satisfactory correction, with the foot deformity remaining accepted and with the correction of equinus, valgus-varus, and rotation deformities proximal to the ankle, but at a cost of distal tibial angulation45. Nelman et al. evaluated the results of this procedure in a study of eighteen patients (including five with arthrogrypotic clubfeet) and noted that, despite the overall satisfactory results in the series, all arthrogrypotic patients required further operative treatment46. Handelsman and Weinberg suggested that very severe deformities, i.e., severe equinus deformities, are not suitable for supramalleolar osteotomies and that durable results are achieved when osteotomy is performed close to the time of skeletal maturity45. Supramalleolar osteotomies are indicated when the center of deformity remains within the tibial metaphysis; this occasionally may happen following iatrogenic injuries to the physis during operative correction of arthrogrypotic clubfeet.
The foot deformity also can be corrected with an external fixator (Fig. 6). External fixators for clubfoot correction include the Ilizarov device and the Taylor Spatial Frame device. Both techniques can be used to provide gradual correction involving soft tissues only or can be combined with tarsal osteotomies, depending on the initial deformity, with good short-term results47-49. Choi et al. reported that eleven of twelve recurrent clubfeet remained plantigrade and painless after a mean of thirty-five months of follow-up47. Eidelman and Katzman reported on ten clubfeet that were initially corrected with the Taylor Spatial Frame device, with two recurrences during follow-up48. It should be stressed that these methods require a team of surgeons and nurses experienced in external fixation techniques, that correction is achieved slowly, and that the deformity may still recur47,48.
Vertical Talus and Other Foot Deformities
Vertical talus (congenital convex pes valgus) is a rare entity that constitutes about 2% to 12% of arthrogrypotic foot deformities and is noted more frequently in patients with distal arthrogryposis or other syndromic forms of arthrogryposis5-9,20,21,49. Vertical talus is characterized by a rigid, irreducible dorsal dislocation of the talonavicular complex, with the calcaneus in fixed equinus and the talus severely plantar flexed. Joint deformities are complicated by peroneus longus and tibialis posterior tendon dislocations and soft-tissue contractures20. The treatment of vertical talus depends on the prognosis for further development, including walking ability; in patients prone to independent walking (e.g., in patients with amyoplasia), operative correction to achieve a painless plantigrade foot should be performed before walking age, whereas for children with syndromic contractures, especially those with neurologic deficits and a poor prognosis for walking, the goals of treatment include the possibility of reasonable footwear use and aid in transportation. These goals should be achieved without major operative intervention or solely with individually fitted orthoses20.
Traditionally, vertical talus was treated with initial weekly casting to diminish dorsal contractures, followed by wide soft-tissue releases in one or two-stage surgical procedures, including posterior ankle and subtalar, talonavicular, and calcaneocuboid releases and extensor, peroneal, and tibialis anterior lengthenings. Direct talonavicular reduction and Kirschner-wire fixation, accompanied by tendon transfers to the talar neck (i.e., split tibialis anterior or peroneus brevis transfers), also were recommended if needed50,51. This method of treatment, despite short-term success, was associated with a number of complications, including wound necrosis, undercorrections, stiffness of the ankle and subtalar joints, and an increased rate of degenerative joint disease50. Consequently, Alaee et al. and Dobbs et al. proposed an alternative minimally invasive strategy for the treatment of idiopathic or syndromic vertical talus, based on the Ponseti concept for the treatment of idiopathic clubfeet (Fig. 7)51,52. The treatment should be started during the first weeks of life with increasing forefoot plantar flexion and inversion while the talus is stabilized medially. Talonavicular reduction should be achieved in four to six weeks, followed by Achilles tendon release and percutaneous or minimally open Kirschner-wire fixation of the talonavicular joint; fractional lengthenings of the extensor digitorum communis and/or peroneus brevis can be added if needed. After completion of the procedure, a long-leg cast is applied for eight weeks, followed by nighttime bracing and passive stretching exercises to increase the ankle range of movement. This method achieved satisfactory early results when used for the treatment of idiopathic vertical talus52,53 and perhaps could also be suitable for syndromic vertical talus, but further studies are required to confirm its efficacy for arthrogrypotic vertical talus.
Isolated equinus, equinovalgus, or calcaneovalgus, which are occasionally observed in patients with arthrogryposis multiplex congenita, should be treated according to a similar philosophy. Specifically, attempts to achieve gradual decrease of the deformities should be started as early as possible with weekly sessions of manipulations and casting. Next, limited a la carte surgery should be performed to correct rigid contractures. Finally, bracing should be performed with use of individually adjusted orthoses. To maintain correction, range-of-motion exercises should be regularly performed.
The authors wish to dedicate this manuscript to Dr. Tadeusz Lejman, MD, PhD (1947 to 2014)—previous Chairman of our Orthopedic Department and the teacher and unquestionable leader of our team—for all of his educational efforts, for discussions on this and other manuscripts, and for commencement of the treatment for children with arthrogrypotic deformities in our institution. They also thank Mr. Christopher Bradish, MD, for his productive help with preparation of this manuscript.
Source of Funding: No external funds were received in support of this study.
Investigation performed at the Orthopedic Department, University Children’s Hospital of Krakow, Krakow, Poland
Disclosure: None of the authors received payments or services, either directly or indirectly (i.e., via his or her institution), from a third party in support of any aspect of this work. None of the authors, or their institution(s), have had any financial relationship, in the thirty-six months prior to submission of this work, with any entity in the biomedical arena that could be perceived to influence or have the potential to influence what is written in this work. Also, no author has had any other relationships, or has engaged in any other activities, that could be perceived to influence or have the potential to influence what is written in this work. The complete Disclosures of Potential Conflicts of Interest submitted by authors are always provided with the online version of the article.
- Copyright © 2015 by The Journal of Bone and Joint Surgery, Incorporated