➢ With the advent of modern tissue-processing and rigorous donor-screening techniques, the use of allograft tissue in reconstructive procedures in foot and ankle surgery has steadily increased.
➢ Allografts can be used to correct deformity, to fill bone or osteochondral voids, to provide collagen in massive unreconstructable tendon defects, and, in many cases, to provide a salvage option when no other viable surgical alternative exists.
➢ A review of the available data suggests that allograft transplantation can result in predictable relief of pain and restoration of functional capacity in many difficult surgical salvage scenarios. However, the quality of the data is only fair, and better evidence is needed to support the extensive use of allograft transplantation in foot and ankle surgery.
Allografts are commonly used during reconstructive procedures around the foot and ankle1. Compared with autograft tissue, allografts offer the advantage of unlimited tissue sources, decreased operative time, and lack of potential donor-site morbidity1. Allografts can be classified into three broad categories: (1) osteochondral grafts, (2) osseous/structural grafts (which can be subclassified as either bulk grafts or cortical/cancellous chips), and (3) allograft tendons. The use of these grafts has increased dramatically in the past decade, but there is concern that enthusiasm surrounding their use has outpaced the published evidence supporting their efficacy and safety. This article will critically review the available evidence regarding outcomes related to the use of allografts in foot and ankle surgery.
Symptomatic osteochondral defects at any anatomic site present a substantial treatment challenge. The goals of operative treatment include the resolution of pain and/or mechanical symptoms, the restoration of articular surface congruity, the prevention of articular surface collapse, and, ultimately, the prevention of degenerative arthritis of the apposing joint surfaces2. In foot and ankle surgery, osteochondral allografts are most commonly used for the treatment of talar lesions. The treatment of symptomatic osteochondral lesions of the talus can be challenging, and allograft transplantation (Fig. 1) can be a particularly useful option for the reconstruction of very large defects or in situations in which preferred initial treatments such as arthroscopic debridement, anterograde or retrograde drilling, microfracture, and/or autologous cartilage implantation or osteochondral autograft transfer have failed2.
Cadaveric allografts can be either kept fresh (stored at 2°C to 4°C) or cryopreserved (fresh-frozen). Cryopreservation may result in a substantial decline in chondrocyte viability, with only 20% to 30% of cells remaining at one year after transplantation3,4. Conversely, fresh osteochondral allografts have been shown to contain viable chondrocytes seventeen years after transplantation5-7. Fresh osteochondral allografts that are stored for more than fourteen days, however, undergo substantial decreases in chondrocyte viability, density, and metabolic activity8,9. Consequently, fresh allografts are preferred to fresh-frozen grafts, and use within fourteen days after graft harvest is preferable.
In a prospective series, Haene et al.9 reported the outcomes for sixteen patients (seventeen ankles) who underwent fresh osteochondral allograft transplantation for the treatment of uncontained osteochondral lesions of the talus. All talar lesions involved the medial or lateral gutter, and all but one had at least one dimension measuring >15 mm. After a mean duration of follow-up of 4.1 years, patients showed significant improvements in terms of both the average Ankle Osteoarthritis Score (AOS) (p = 0.003) and the average American Academy of Orthopaedic Surgeons (AAOS) Foot and Ankle Module scores (p = 0.002), both of which are validated outcome instruments10-12; the AOS pain score and the Short Form-36 (SF-36) scores trended toward improvement (p = 0.14 and 0.09, respectively) (Table I). Five of the seventeen ankles were deemed failures (two allografts failed to incorporate postoperatively, one patient withdrew from the study with ongoing symptoms, and two ankles required arthrodesis). Ten of the seventeen ankles were considered to have a good or excellent outcome.
Adams et al.5 performed a retrospective review of eight patients who underwent fresh talar shoulder allograft transplantation for the treatment of osteochondral lesions of the talus over a seven-year period. After a mean duration of follow-up of four years, patients demonstrated a significant decrease in the visual analog score (VAS) for pain (p < 0.05) and a significant increase in the Lower Extremity Functional Scale (LEFS) score (p < 0.05) (Table I). At the time of the latest follow-up, the mean American Orthopaedic Foot & Ankle Society (AOFAS) Ankle-Hindfoot Scale score was 84 points. Four patients (50%) required an additional procedure. No failures, defined as either tibiotalar arthrodesis or total ankle arthroplasty, were reported.
Raikin13 reported on a prospective series of fifteen patients with particularly large symptomatic osteochondral lesions of the talus (mean volume, 6059 mm3) who underwent fresh allograft transplantation. The mean duration of follow-up was fifty-four months. The measured AOFAS Ankle-Hindfoot Scale scores improved significantly, with an average increase of 45 points per patient (p < 0.05), with greatest improvements noted in the pain, activity, and walking subscales (Table I). The mean VAS pain score decreased significantly by 5.2 points (p < 0.05). Postoperative radiographs showed evidence of some collapse or resorption of grafts in ten of fifteen patients, and there was narrowing of the joint space overlying the graft in nine patients that was concerning for early degenerative change. No association between radiographic findings and clinical outcome could be identified. Two patients underwent ankle arthrodesis and were considered to have had a failure. Of note, all fifteen patients, including the two who underwent ankle arthrodesis because of graft failure, were pleased that they underwent the allograft procedure and stated that they would choose to do it again. Thus, after intermediate-term follow-up, the author claimed an 87% survival rate with 73% of patients reporting good or excellent results at two years.
Several authors14,15 also have described bipolar osteochondral allografting for the treatment of tibiotalar joint arthritis in young active patients. Meehan et al.14 reported on a series of eleven patients with a mean age of forty-three years who underwent osteochondral grafting of the tibiotalar joint, nine of whom had bipolar grafting. After a mean duration of follow-up of thirty-three months, six of the eleven patients were considered to have a successful result, with the remaining five patients having a failure that required repeat allografting, total ankle replacement, or patient-selected nonoperative management. The average AOFAS Ankle-Hindfoot Scale score was significantly improved (p = 0.01) (Table I). Despite the relatively high rates of failure (45%) and reoperation (64%), the authors claimed that the success rate and the improvements in terms of pain relief were comparable with those for young patients who undergo total ankle replacement and suggested that bipolar osteochondral grafting offers an alternative to arthrodesis or arthroplasty in young, active patients with advanced tibiotalar arthritis14. Jeng et al.16 reported a similarly high failure rate in a series of twenty-nine patients who underwent bipolar tibiotalar allografting, with twenty (69%) of twenty-nine patients regarded as having radiographic failure at two years and fourteen requiring revision grafting, total ankle arthroplasty, or arthrodesis.
Other similarly designed case series, both retrospective17,18 and prospective19, have corroborated these findings that fresh osteochondral allograft transplantation is a viable reconstructive procedure that results in significant improvements in functional outcome measures and pain scores in patients with osteochondral lesions of the talus, even particularly large lesions13 and those involving the talar shoulder5. Those series showed similar rates of secondary procedures and graft failure. However, there remains a lack of long-term data regarding the outcome of these procedures, and the literature is limited to small, retrospective case series20. It also should be noted that many of those studies utilized the AOFAS Ankle-Hindfoot Scale, which is not a validated instrument, as a key outcome measure11; as a consequence, the improvements should be interpreted with caution. Those Level-IV studies do not provide sufficient evidence (Grade-C recommendation) (Table II) to conclusively support the use of osteochondral allograft for the treatment of an osteochondral lesion of the talus of any size, location, grade, stability, displacement, or containment, or for the use of bipolar tibiotalar allografting20.
Many choices of allograft and bone substitute are available for surgery on the foot and ankle. Bone grafts often serve a combined mechanical and biologic function21. Allograft bone (i.e., fresh, fresh-frozen, and freeze-dried bulk grafts) may be structural and may support a largely mechanical function, whereas cancellous allograft croutons for use at sites of impaction of host cancellous bone and/or intended arthrodesis can provide biologic stimulus for new bone formation with little or no mechanical function1,21.
Dolan et al.22 reported the results of a prospective, controlled, randomized trial in which autogenous tricortical iliac crest autograft was compared with allograft for lateral column lengthening procedures that were performed for reconstruction in patients with adult-acquired pes planus deformity. The trial included a total of thirty-three feet in thirty-one patients with stage-II posterior tibial tendon insufficiency who had had a failure of nonoperative treatment. Eighteen feet (55%) were prospectively randomized to treatment with freeze-dried tricortical (the allograft group), and fifteen feet (45%) were randomized to treatment with bone harvested from the ipsilateral iliac crest (autograft group); all patients were followed to and beyond the point of union. Union was determined on the basis of postoperative radiographs and clinical evidence of healing. By twelve weeks, all patients in both groups had achieved graft union. Seventeen (94.4%) of the eighteen feet in the allograft group had achieved union by eight weeks, compared with nine (60%) of the fifteen feet in the autograft group (p = 0.03). The average time to union was 8.9 weeks (Table III). At the time of the twelve-month follow-up, there were no documented cases of delayed union, nonunion, graft collapse, or hardware failure. Of note, two patients in the autograft group continued to have donor-site hip pain three months postoperatively.
Myerson et al.23 reported on a retrospective series of seventy-five feet in seventy-three patients who underwent various foot and ankle arthrodesis procedures that were performed with a fresh-frozen femoral head structural allograft. Structural allografts were utilized to restore normal dimensions of the foot and ankle following surgery or trauma and to treat arthritis or deformity in situations in which conventional cancellous graft would not have provided sufficient mechanical support, such as subtalar arthrodesis (twenty-eight feet) and calcaneal osteotomy (eleven). No standardization of internal fixation devices was performed, with a variety of implant choices being utilized on the basis of the size and shape of the bone graft, the adjacent bone structure, and the requirements for obtaining stability for each specific procedure. Union was determined on the basis of postoperative radiographs and clinical evidence of healing. At the time of the four-month follow-up, sixty-nine (92%) of the seventy-five feet demonstrated successful union, with an average time to union of four months (range, two to ten months) (Table III). Delayed union (defined as union more than four months after surgery) occurred in twenty-two (29%) of the seventy-five feet. These patients were managed conservatively with prolonged immobilization and/or an external bone stimulator, and all went on to achieve union. Four (67%) of the six feet with nonunion were treated with revision arthrodesis with cancellous autograft; all went on to union, although one patient required a second revision procedure and conversion from subtalar to triple arthrodesis with autograft and a direct-current internal bone-stimulator to achieve union. At the time of follow-up at a mean of 3.5 years, there was no evidence of graft resorption or collapse. Although there were a limited number of nonunions in that study, the prevalence of delayed union was high; the authors contended that this finding was a result of the patients being from a relatively high-risk population rather than a factor inherent to the structural allograft as the literature has shown similar success rates with structural autograft harvested from the iliac crest in these complex cases23.
Garras et al.24 reported on a consecutive prospective series of twenty-two patients (twenty-four feet) who underwent subtalar bone-block arthrodesis with interpositional frozen femoral head structural allograft. The cohort comprised patients with subtalar arthritis, considerable hindfoot collapse, loss of heel height, and symptomatic anterior ankle impingement. Of note, at the time of the operation, a platelet-rich product (Symphony; DePuy/Johnson & Johnson, Warsaw, Indiana), derived from the patient’s own blood, was used in seven (33%) of twenty-one feet that were noted to have avascular bone at the arthrodesis site. Union was determined on the basis of postoperative radiographs and clinical evidence of healing. Twenty patients (twenty-one feet) were available for follow-up at a mean of 35.8 months. Union was achieved in nineteen of twenty-one feet at a mean of 15.5 weeks. The mean AOFAS Ankle-Hindfoot Scale score and seven objective radiographic parameters were significantly improved (Table III). Jeng et al.25 presented a series of thirty-two patients who underwent tibiotalocalcaneal arthrodesis with use of bulk femoral head allograft to fill segmental osseous defects resulting from failed ankle arthroplasty, osteonecrosis of the talus, or trauma (Fig. 2). Sixteen patients (50%) achieved radiographic union, and an additional seven patients had an asymptomatic nonunion; the authors thus claimed a 71.9% rate of functional limb salvage. Of note, all nine patients who had diabetes mellitus had a nonunion, and six patients required a below-the-knee amputation; the authors contended that tibiotalocalcaneal arthrodesis with femoral head allografts is a difficult salvage procedure with a high rate of complications25.
Demineralized bone matrix represents a commonly available source of allograft bone that has been shown to be osteogenic in human models26. Michelson and Curl26 performed a prospective study of fifty-five patients who underwent hindfoot fusion augmented by either autogenous iliac crest bone graft or demineralized bone matrix allograft bone over a five-year period. Eleven patients (20%) underwent isolated subtalar fusion (three with iliac crest bone graft, eight with demineralized bone matrix) and forty-four (80%) underwent triple arthrodesis (fifteen with iliac crest bone graft and twenty-nine with demineralized bone matrix). Fifty-two (95%) of the fifty-five patients demonstrated complete radiographic union, two showed clinical evidence of nonunion, and one showed radiographic evidence of nonunion without clinical symptoms (Table III). There was no significant difference in terms of the rate of union associated with the two types of grafts in either the subtalar group (with union occurring in three of three patients managed with iliac crest bone graft and in seven of eight patients managed with demineralized bone matrix) or the triple fusion group (with union occurring in thirteen of fifteen patients managed with iliac crest bone graft and in twenty-nine of twenty-nine patients managed with demineralized bone matrix) (p > 0.05). The time to clinical union ranged from 2.7 to 3.7 months in the iliac crest bone graft group, compared with 3.0 to 3.4 months in the demineralized bone matrix group. In addition, the estimated blood loss in patients receiving demineralized bone matrix was significantly lower than that in patients receiving iliac crest bone graft (p < 0.05) (Table III).
Other small, well-designed series27-29 have demonstrated similar outcomes following the use of allograft bone for arthrodesis in the foot and ankle, with high union rates equivalent to those associated with autograft. These studies suggest that both structural and cancellous allograft transplantation can achieve similar rates of fusion compared with autograft bone while eliminating donor-site morbidity and that structural allografts provide useful options in situations involving complex reconstructions and in salvage situations involving considerable bone defects that are difficult to reconstruct with autograft. It should be noted, however, that most of the discussed studies22-24 utilized postoperative radiographs as the sole imaging modality to assess union. Coughlin et al.30, in a series of hindfoot arthrodeses, demonstrated that standard radiographs substantially overestimate the rate of hindfoot fusion when compared with computed tomography (CT) scans and that CT scans are significantly more reliable for determining osseous union. Given that radiographs were used as the sole imaging modality to assess union in each of the discussed series, the true rate of union may have been overestimated.
Despite this caveat, several high-quality Level-I and II studies have provided sufficient strong evidence (Grade-A recommendation) to support the use of allografts in certain procedures (Table II). However, many Level-IV studies evaluating structural allografts have not provided sufficient evidence (Grade-C recommendation) to conclusively support the use of allograft in other scenarios.
Allografts play an important role in tendon and ligament reconstruction, particularly in patients with insufficient competent native tissue, those with massive irreparable tears, and those for whom autologous tendon transfers are not desirable. All types of grafts have been shown to lose some strength after implantation, and allografts have been shown to have lost a substantial proportion of strength by six to nine months after implantation31. This loss of strength may result from detrimental effects of graft processing (fresh-freezing, freeze-drying, or irradiation to reduce immunogenicity) on the biomechanical properties of the graft32-34. These findings may not have clinical importance, however, as longitudinal studies evaluating the outcomes of knee ligament reconstruction with use of allograft tendons have demonstrated no significant change in knee laxity or knee scores after three and seven years of follow-up32,35.
Horibe et al.36 first described reconstruction of lateral ligaments of the ankle with use of fresh-frozen allograft tendons in a series of thirteen patients, with toe extensor and flexor tendons being used to reconstruct the anterior talofibular and calcaneofibular ligaments. At the time of the two-year follow-up, nine patients were rated as having an excellent outcome and four were rated as having a good outcome, with no clinical instability being noted by patients or on stress radiographs (Table IV). Jung et al.37 performed a retrospective review of twenty-three patients (twenty-four ankles) who underwent anatomic lateral ankle ligament reconstruction for the treatment of chronic ankle instability with use of semitendinosus tendon allograft to reconstruct the anterior talofibular ligament and calcaneofibular ligament. After a mean duration of follow-up of nineteen months, the mean AOFAS Ankle-Hindfoot Scale scores, VAS pain scores, and Karlsson-Peterson ankle instability scores were all significantly improved (Table IV). Stress radiographs demonstrated significant improvements in the median talar tilt angle (p < 0.05) and anterior talar translation (p < 0.05) during the anterior drawer test. Twenty-one (88%) of twenty-four ankles were rated as satisfactory or very satisfactory by patients.
Ellis et al.38 performed a similar retrospective study of eleven patients (twelve ankles) who underwent lateral ankle ligament reconstruction with use of anterior tibial tendon allografts. After a mean duration of follow-up of 3.5 years, ten of the eleven patients rated the outcome as good or excellent. The patients had high mean Foot and Ankle Outcome Scores (FAOS), SF-36 scores, and Karlsson scores, with low VAS pain scores, at the time of the latest follow-up, although no preoperative measures were reported (Table IV). Stress radiographs showed significant improvement in the median tibiotalar tilt angle (p < 0.01) (Table IV). Four patients reported no activity restrictions, six reported mild restrictions, and one reported moderate restrictions. All but four patients were able to return to their sport of interest; the authors did not specify if return to sports activity correlated with self-reported level of activity restriction.
Nakata et al.39 presented a series of twenty patients who underwent anatomic lateral ankle ligament reconstruction with use of allogenic fascia lata. In seven of the twenty patients both the anterior talofibular ligament and calcaneofibular ligament were reconstructed, whereas in the other thirteen patients the anterior talofibular ligament was reconstructed and the calcaneofibular ligament was repaired by means of proximal advancement. After a mean duration of follow-up of 4.2 years, nineteen (95%) of the twenty patients rated the outcome as good or excellent. The mean talar tilt and anterior talar translation during the anterior drawer test were significantly improved (p < 0.05 for both) (Table IV). The authors joined Ellis et al.38, Horibe et al.36, and Jung et al.37 in concluding that allograft tendons can be used to safely and effectively reconstruct incompetent lateral ligaments of the ankle with satisfactory clinical outcomes (Level of Evidence IV) (Table II).
Allograft tendons also may be used for reconstruction of large intercalary segment defects of native tendons. Bulk intercalary allograft transplantation has been described as a salvage operation for large chronic or neglected Achilles tendon ruptures, particularly in cases in which direct end-to-end anastomosis of the ruptured tendon ends is impossible and common soft-tissue augmentation techniques such as advancement flaps and autologous tendon transfers will prove insufficient40-44.
Cienfuegos et al.44 reported the case of a single patient in whom a bulk Achilles tendon allograft was used to reconstruct a neglected Achilles tendon rupture with associated 12-cm tendinous defect. The authors used interference screws to secure either end of the tendon to the calcaneus and sutured the graft to the native musculotendinous unit. At one year of follow-up, the patient was able to walk without limitations, could perform a bilateral heel rise, and had returned to his normal activities. No objective outcome measures were assessed either preoperatively or postoperatively. Lepow and Green45 described the case of a single patient in whom a freeze-dried Achilles tendon allograft was used to reconstruct a neglected Achilles rupture with an associated 6-cm gap. By ten weeks postoperatively, the patient had normal strength and range of motion in a non-weight-bearing examination and could perform bilateral heel rise without difficulty; she could not, however, perform an isolated heel rise on the injured side. At one year of follow-up, the patient had assumed preinjury functional capacity; no objective outcome measures were assessed either preoperatively or postoperatively. Two other similar case reports46,47 in the literature have documented similar results in association with the use of bulk Achilles allograft tendon for the reconstruction of a chronic Achilles tendon rupture.
An operative technique for the reconstruction of large intercalary peroneus brevis tendon defects with use of peroneal tendon allografts has been reported48, and authors at our institution49 recently reported on a series of fourteen patients with irreparable peroneus tendon tears (eleven brevis tears, two longus tears, and one combined brevis/longus tear) that were treated with allograft peroneal tendon reconstruction (Fig. 3). In ten of the twelve cases of brevis tears, there was sufficient distal tendon stump on the fifth metatarsal to secure the tendon allograft to native tendon; in the other two cases, the distal tendon was completely avulsed from the metatarsal and suture anchors were required for distal fixation. The average length of intercalary segment reconstructed was 10.8 cm (range, 6.0 to 20.0 cm). After a mean duration of follow-up of seventeen months, the mean VAS pain score (p = 0.005), SF-12 score (p = 0.02), LEFS score (p < 0.001), and eversion strength (p < 0.001) were significantly improved (Table IV). All patients returned to their preoperative activity levels.
The authors of those series agreed that allograft tendon transplantation is a viable reconstructive procedure in salvage situations in which the patient has insufficient native tissue and/or massive irreparable tendon tears and that this procedure results in a reliable decrease in pain and satisfactory patient-reported outcomes. There remains, however, a lack of objective and long-term data evaluating the outcome of this procedure. These Level-IV and V studies do not provide sufficient evidence (Grade-C or I recommendation) (Table II) to conclusively support the use of tendon allograft in these scenarios.
A concern with the use of allograft tissue is the possibility of disease transmission, but a review of available data suggests that the risk of disease transmission for properly prepared and handled musculoskeletal allografts is acceptably low22. There have been documented cases of transmission of hepatitis B, hepatitis C, and human immunodeficiency virus (HIV) through transplanted fresh-frozen allograft bone50-52, but the last documented case of HIV transmission from bone graft occurred in 198524,52. With the implantation of strict guidelines by the American Association of Tissue Banks (AATB) and the United States Food and Drug Administration (FDA), the rate of blood-borne pathogen transmission approaches zero for both fresh and freeze-dried grafts24,53,54. Most reported cases of disease transmission have involved transplanted bone rather than tendon; the estimated risk of disease transmission from an allograft tendon procured from an adequately-screened donor is one in 1.5 million, less than that of an allogeneic blood transfusion32. Since 1951 there have been >1,000,000 recipients of freeze-dried grafts, and there have been no known cases of HIV or hepatitis virus transmission through irradiated or freeze-dried bone or tendon grafts55.
Cases of bacterial infection resulting from allograft implantation have been documented, but the risk remains very low. As of 2002, twenty-six cases of bacterial disease transmission had been identified; of note, thirteen were associated with Clostridium species infection, and fourteen of the twenty-six were associated with a single tissue- processor22.
To our knowledge, there have been no well-designed studies examining the cost-effectiveness of allografts compared with autografts in foot and ankle surgery. The cost of individual allografts can be high, with price estimates at our institution for 10 cc of allograft cancellous chips, an Achilles tendon allograft, and a fresh whole cadaveric talus ranging from $100 to $200, from $1800 to $2100, and from $4000 to $5000, respectively. As Dolan et al.22 noted, however, the cost of allograft is relatively fixed, and the cost of iliac crest bone autograft must be considered in terms of additional anesthesia time, additional operating room resources, surgeon fees, increased postoperative pain management costs, the possibility of longer hospital stays, and the costs of dealing with second operative site complications in both the long and short terms. Given the numerous benefits attributable to the use of allografts, detailed cost analyses comparing allografts and autografts are needed.
Allografts provide a useful and versatile adjunct in foot and ankle surgery. They can be used to correct deformity, to fill bone or osteochondral voids during reconstructive surgery, to provide collagen in cases of massive unreconstructable tendon defects, and, in many cases, to provide a salvage option in cases in which no other viable surgical alternative exists. Importantly, the use of allografts can reduce operative time and cost and eliminates the risk of donor-site morbidity associated with autograft harvest procedures, which in one study were shown to result in chronic pain in 26% of 134 patients who underwent iliac crest bone graft harvest for spinal arthrodesis56. With the implementation of rigid screening and preservation guidelines by multiple regulatory agencies, modern allograft implantation is a safe procedure associated with minimum risk of disease transmission.
Several high-quality Level-I and II studies have provided sufficient evidence (Grade-A recommendation) (Table II) to support the use of allografts in certain procedures, including tricortical allografts in lateral column lengthening procedures and demineralized bone matrix in hindfoot arthrodesis procedures. There is insufficient evidence, however, to provide higher than a Grade-C recommendation in other clinical scenarios (Table II). It may prove difficult to achieve higher levels of evidence, however, as in instances in which large bone or tissue voids are not reconstructable with autograft tissue, allografts remain the only viable salvage option; consequently, the development of large controlled trials may not be possible. Furthermore, procedures such as talar allograft transplantation are relatively rare and most studies evaluating outcomes will be limited with respect to sample size. Regardless, when possible, larger, prospective, randomized trials are needed to further support the use of allograft tissue in foot and ankle surgery.
Source of Funding: No external funding was utilized for this investigation.
Investigation performed at the Department of Orthopaedic Surgery, Duke University Medical Center, Durham, North Carolina
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. One or more of the authors, or his or her institution, has had a financial relationship, in the thirty-six months prior to submission of this work, with an entity in the biomedical arena that could be perceived to influence or have the potential to influence what is written in this work. 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 © 2013 by The Journal of Bone and Joint Surgery, Incorporated