➢ Controversy permeates the use of prophylaxis against venous thromboembolic disease (VTED) after foot and ankle surgery.
➢ The potential clinical consequences of VTED as well as national guidelines that increasingly view deep-vein thrombosis and pulmonary embolism as preventable events must be weighed against the risks of chemoprophylaxis, including not only hemorrhagic complications but also potentially life-threatening manifestations such as immunologic heparin-induced thrombocytopenia.
➢ Certain patient populations may be at higher risk for clinically relevant VTED, including patients over the age of forty years with acute Achilles tendon ruptures, patients over the age of fifty years with ankle fractures, patients with diabetes mellitus, patients with connective-tissue inflammatory diseases, and patients with a history of VTED. Other factors to consider may include oral contraceptive use, smoking history, recent air travel, and a familial history of VTED events.
➢ To further confound matters, the use of chemoprophylaxis against VTED in foot and ankle surgery may not invariably lower the incidence of VTED events.
➢ Large-scale, prospective, randomized trials are necessary to better guide foot and ankle surgeons in their decision-making processes regarding thromboprophylaxis after surgery.
Controversy permeates the use of prophylaxis against venous thromboembolic disease (VTED) after foot and ankle surgery. Underpinning these debates are the unique perspectives offered by several specialty societies, the disparate incidences of VTED among varied orthopaedic patient populations, inconsistency in the quality of relevant research, and uncertainty about the applicability of broader governmental quality measures. As a result, foot and ankle clinicians continue to be forced to make decisions on the basis of incomplete information and contradictory guidelines. The inaccuracy inherent in projecting hip and knee arthroplasty studies onto the foot and ankle population—which comprises a vast array of demographics, diagnoses, and procedures—amplifies this uncertainty.
Specialty societies, including the American College of Chest Physicians (ACCP) and the American Academy of Orthopaedic Surgeons (AAOS), have issued recommendations regarding the appropriate use of prophylaxis against VTED in the perioperative period1,2. The strength of these sometimes diverging recommendations is based on literature of variable quality and applicability. There is no agreement on the appropriate framework for a risk-benefit analysis of those issues. Furthermore, unlike the ACCP and AAOS, the American Orthopaedic Foot & Ankle Society (AOFAS) has repeatedly expressed its inability to provide any guidance due to a paucity of available information3.
American College of Chest Physicians 2008 Recommendations
In 2008, the ACCP published the eighth edition of its clinical practice guidelines for postoperative prophylaxis against VTED, with explicit recommendations extending to orthopaedic procedures such as elective hip and knee arthroplasty1. These guidelines strongly recommended the use of low-molecular-weight heparin, fondaparinux, or warfarin for at least ten days postoperatively, and explicitly recommended against the use of aspirin as a solitary agent in all patients. The studies that were used to support these recommendations4 often employed ultrasound or venography-documented deep-vein thrombosis (DVT) as a surrogate for pulmonary embolism (PE), independent of whether patients diagnosed with DVTs were clinically symptomatic or went on to develop a PE. The proponents of these recommendations underscored a historical relationship between DVTs and PEs found on imaging studies as well as a commensurate decrease in both DVT and PE rates in clinical studies in which antithrombotic agents were administered5,6.
American Academy of Orthopaedic Surgeons
In 2011, the AAOS updated its own evidence-based guidelines regarding VTED in patients undergoing hip and knee arthroplasty2. Notably, these guidelines focused on so-called critical outcome measures, including symptomatic PE, major bleeding events, reoperation within ninety days, and all-cause mortality. In contrast to the ACCP guidelines, the AAOS perceived a poor correlation between acute DVT and acute PE and therefore did not use DVTs diagnosed with ultrasound as a surrogate for PE7. The strength-of-recommendation ratings were strong, moderate, weak, inconclusive, or consensus, with the last being used for group opinions not rooted in reliable evidence. The AAOS committee was only able to issue a single strong recommendation, advising against the use of routine ultrasound postoperatively. Moderate recommendations included stopping antiplatelet agents such as aspirin or clopidogrel preoperatively as well as using either pharmacological or mechanical prophylaxis against VTED in patients undergoing elective arthroplasty without an otherwise elevated risk of VTED. It did not specify which pharmacological agents were most appropriate, leaving aspirin as a potential alternative to those agents highlighted by the ACCP.
American College of Chest Physicians 2012 Recommendations
Adding to this confusion, the ninth edition of the ACCP guideline on prevention of VTED, issued in 2012, tempered many of the prior recommendations. The ACCP assigns its recommendations a Grade of 1 or 2 on the basis of whether its recommendations are strong or weak8. It also designates the quality of evidence supporting a given recommendation as “A,” “B,” or “C” on the basis of whether the evidence is of high, moderate, or low quality. In its latest guidelines, the ACCP rescinded all previous Grade-1A recommendations and instead assigned a Grade-1B rating to its recommended use of pharmacological prophylaxis against VTED after hip or knee arthroplasty9. Notably, aspirin was included among the potential agents, and a residual ACCP preference for low-molecular-weight heparin over aspirin was rated as weak in strength and based on low-quality evidence (Grade 2C). The use of pneumatic compression boots also was strongly recommended, albeit on the basis of weak evidence (Grade 1C). In light of these new recommendations, the Surgical Care Improvement Project (SCIP) updated its measures in 2014 to include aspirin as a viable option for prophylaxis against postoperative VTED10.
Postoperative prophylaxis against VTED is fundamentally an exercise in risk mitigation, which necessitates an explicit understanding of baseline risk. Historically quoted rates of DVT after hip or knee arthroplasty have ranged from 40% to 84%, and, more critically, 0.1% to 0.7% of patients may sustain a fatal PE11. It is unclear whether modern surgical techniques and earlier patient mobilization have impacted these rates. In addition, a number of DVTs diagnosed on imaging are not overtly symptomatic, although there has yet to be substantive research that has evaluated long-term patient-reported outcomes or impact on quality of life after a venous thromboembolic event, which is a major shortcoming of the current literature12,13.
However, risk mitigation invariably comes with a cost. Perioperative use of low-molecular-weight heparin has carried a combined risk of minor and major bleeding events of as high as 10% in some studies as well as a 0.2% risk of an immunogenic form of heparin-induced thrombocytopenia14,15. While infrequent, the latter is an antibody-dependent process that can result in an amputation rate of 22% and a mortality rate of 11%16. Indeed, one documented case report described a patient death as a result of heparin-induced thrombocytopenia after a single dose of low-molecular-weight heparin following a foot and ankle procedure16. Furthermore, even minor bleeding events precipitated by the use of blood thinners may be clinically consequential. One study examining 17,714 total hip replacements that had been performed at 128 New York state hospitals demonstrated that greater hospital compliance with SCIP measures, which included pharmacological prophylaxis against VTED, was linked to increased infection rates17. A separate study examining the use of chemoprophylaxis after 2237 primary total hip or knee arthroplasty procedures demonstrated prolonged wound drainage among patients receiving low-molecular-weight heparin as compared with aspirin or warfarin18. Moreover, prolonged drainage was associated with extended hospital stays and an increased risk of wound infections (odds ratio [OR] = 1.42; 95% confidence interval [CI] = 1.18 to 1.71; p < 0.001), with each day of drainage translating into a 42% increase in wound infection risk. A recent meta-analysis similarly identified a 53% lower rate of bleeding events when patients receiving aspirin were compared with a cohort of pooled patients receiving heparin, enoxaparin, danaparoid, or warfarin (3.8% versus 8.0%)19. Some of the newer oral agents also may contribute to bleeding events. A recent study comparing the use of subcutaneous enoxaparin and oral rivaroxaban (Xarelto; Bayer Schering Pharma, Berlin, Germany) for prophylaxis against VTED in a total of 13,123 patients undergoing total hip or knee arthroplasty demonstrated a higher rate of wound complications among patients receiving the oral agent (2.81% compared with 3.85%; OR = 0.72; 95% CI = 0.58 to 0.90; p = 0.005).
On the other hand, curtailing the use of prophylaxis against VTED has its own consequences. When a PE does occur, mortality rates may exceed 15% in the first three months after diagnosis20. Broader population studies have highlighted that the first clinical manifestation of PE may be sudden death in 25% of patients21,22. Moreover, while prophylaxis against VTED has its own clinical consequences, once a clinically relevant PE is diagnosed, the higher anticoagulant doses inherent to therapeutic treatment must be taken into consideration. Among patients undergoing total joint arthroplasty who experienced postoperative PEs requiring therapeutic-dose low-molecular-weight heparin, the associated complication rate was 33%, including a 10% rate of major hemorrhagic complications23. Along this line of thinking, some authors have expressed concern about a surge in overdiagnosis and overtreatment of clinically unimportant PEs24,25.
Foot and Ankle Surgery
Compared with hip and knee arthroplasty, there is a paucity of literature examining the rates of VTED after foot and ankle surgery. Furthermore, the recommendations of specialty societies diverge somewhat from those of broader governmental regulatory agencies that have identified VTED as a global quality-improvement measure. The most recent ACCP guidelines suggested that prophylaxis against VTED in patients with foot and ankle conditions was unnecessary, but it rated this recommendation as being weak in strength and based on low-quality evidence (Grade 2C)9. In its own position statement, the AOFAS noted that there were insufficient data to make a recommendation for or against the use of VTED prophylaxis with respect to type, duration, or patient risk factors that would compel its use3. Partly at issue is that patients undergoing hip or knee arthroplasty are not a surrogate for patients undergoing foot and ankle surgery, who have a broad range of diagnoses and for whom operative procedures remain even more diverse. Furthermore, whereas the hip and knee rarely undergo prolonged immobilization, prolonged splint, cast, or brace immobilization below the knee is commonplace after foot or ankle surgery. While validated risk-assessment models for VTED in individual patients do exist, they unfortunately have often been validated for non-orthopaedic procedures such as general surgery, urology, or vascular surgery26,27. For example, one of the most popular models, purported by Caprini, utilizes a point-based system of forty potential risk factors in patients undergoing general surgical procedures28,29. Given that this scoring system considers any patient over the age of forty years (1 point) with a body mass index (BMI) of >25 kg/m2 (1 point) who undergoes minor surgery (1 point) to be of “higher risk” (3 to 4 points total), it is difficult to know how to extend such risk-assessment models to the field of foot and ankle surgery. On closer inspection, using such stratification would obligate the vast majority of this population to receive postoperative prophylaxis.
In contrast to specialty societies, the National Institute for Health and Care Excellence recommended that prophylaxis against VTED should be considered for all admitted hospital patients with risk factors, which it defined to include reduced mobility for more than three days, reduced mobility for any length of time in patients over sixty years of age, a BMI of >30 kg/m2, or medical comorbidities such as heart disease30. The SCIP guidelines do not explicitly recommend the use of prophylaxis against VTED for orthopaedic procedures (other than hip or knee arthroplasty and procedures performed for the surgical treatment of hip fractures), but they do penalize hospitals for DVTs that are diagnosed in inpatients who are not being managed with a prophylactic regimen31.
Incidence and Risk Factors
Multiple studies have suggested that the incidence of VTED after foot and ankle surgery is probably lower than that after hip and knee arthroplasty. A retrospective study that examined the rates of readmission due to VTED in 57,183 patients undergoing operative repair of ankle fractures in California demonstrated readmission rates of 0.05% and 0.34% for DVT and PE, respectively32. An age of greater than fifty years (OR = 3.90) and peripheral vascular disease (OR = 3.36) were identified as relative risk factors for VTED. In a comparable study of 4481 patients undergoing calcaneal fracture repair, the rate of readmission for VTED was 0.25%33. Still, these large, population-based studies likely underestimated the incidence of VTED by capturing only patients with VTED events severe enough to warrant hospital admission or for whom outpatient management was unfeasible.
While a focus on readmissions potentially underestimates the rate of VTED, a recent study exploring complications after 25,999 foot and ankle procedures performed by 2970 surgeons seeking certification by the American Board of Orthopaedic Surgery (ABOS) demonstrated a 0.19% incidence of DVT and a 0.28% incidence of PE34. Similarly, an expansive study examining diagnostic codes in the National Health Service in England demonstrated DVT and PE rates of 0.12% and 0.17%, respectively, after 45,949 ankle fracture repairs; 0.01% and 0.02%, respectively, after 33,626 first metatarsal osteotomies; and 0.03% and 0.11%, respectively, after 7033 hindfoot arthrodeses35. The limitations of that study included the inability to ascertain whether prophylaxis against VTED was used in any given patient. The study did, however, demonstrate an elevated risk of VTED among patients managed for ankle fractures if they were over fifty years of age and had at least two major comorbidities; the risk was markedly elevated if one of these comorbidities was non-insulin-dependent diabetes (OR = 14.6; 95% CI = 9.67 to 21.9). Another retrospective study demonstrated a 4% incidence of symptomatic VTED for 602 patients undergoing foot and ankle procedures and suggested that inflammatory connective-tissue diseases such as rheumatoid arthritis and a history of VTED were significant risk factors (p = 0.04 and 0.02, respectively)36.
Cast immobilization also has been implicated in the occurrence of VTED events. A recent meta-analysis demonstrated that the incidence of DVT ranged from 4.3% to 40% after casting37. In that same meta-analysis, patients receiving low-molecular-weight heparin had a decreased incidence of DVT, ranging from 0% to 37% (OR = 0.49; 95% CI = 0.37 to 0.72), and the authors recommended the use of low-molecular-weight heparin for all patients undergoing casting. Of the six prospective, randomized studies included in that meta-analysis, however, three failed to demonstrate a protective effect afforded by pharmacological prophylaxis38-40. All studies involved the use of ultrasound or venography to screen for DVT, independent of whether patients were symptomatic. While not all studies were explicit about weight-bearing status, some indicated that patients were mobilized with crutches. A separate study that involved the use of ultrasound for the evaluation of 2761 patients who presented to the emergency room with below-the-knee injuries demonstrated an overall rate of DVT of 6.4%, with rigid immobilization being a significant risk factor (OR = 2.70; 95% CI = 1.66 to 4.38; p < 0.0001)41. The study also showed that non-weight-bearing status (OR = 4.11; 95% CI = 1.72 to 9.86; p = 0.0015), an age of greater than fifty years (OR = 3.14; 95% CI = 2.27 to 4.33; p < 0.0001), and greater injury severity (OR = 1.88; 95% CI = 1.34 to 2.62; p = 0.0002) correlated with an increased risk of DVT. However, only 1% of the patients developed symptomatic DVT and only 0.04% (one patient) developed a symptomatic PE. Furthermore, only 0.2% (five patients) developed proximal DVTs. The use of pharmacological prophylaxis against VTED did not appear to offer a protective benefit, although it was generally used for higher-risk patients. A separate study counterintuitively showed higher rates of DVT among patients undergoing foot and ankle surgery who received low-molecular-weight heparin, likely for analogous confounding reasons in that selective chemoprophylaxis was preferentially used in higher-risk patients36. The study also showed that casting, recent air travel, prior VTED, and rheumatoid arthritis were risk factors. Weight-bearing status was not an independent risk factor, although the authors did note that a subgroup of patients who were allowed partial weight-bearing had higher rates of DVT (p = 0.003). Thus, casting or other forms of immobilization may increase the risk of VTED events after foot and ankle injury, but many studies have shown contradictory findings and the role of chemoprophylaxis remains unclear.
Patients with acute Achilles tendon ruptures also may have a heightened risk of VTED events, with reported rates of DVT in the literature ranging from 0.4% to 34% without the use of chemoprophylaxis42. A prospective case series in which 100 patients with acute Achilles tendon rupture underwent ultrasound screening of the lower extremity demonstrated a 34% rate of DVT, irrespective of whether patients were managed operatively or nonoperatively43. Retrospective case series examining the rate of symptomatic VTED events after Achilles tendon rupture have demonstrated a lower but nonetheless broad range of incidences. Three retrospective series demonstrated rates of symptomatic DVT ranging from 4% to 6.3% and of symptomatic PE ranging from 1.1% to 1.4%44-46. The authors of one of those studies, however, explicitly noted a near-fatal PE46. In contrast, another retrospective series showed a much higher rate of symptomatic DVT (23.5%)47. In that study, one-third of DVTs occurred preoperatively and an age of greater than forty years was found to be a significant risk factor (p < 0.0026 for an age of forty to fifty-nine years and p < 0.0014 for an age of sixty to seventy-nine years). The largest retrospective study of which we are aware included 1172 patients presenting to a large hospital system and demonstrated a lower rate of symptomatic DVT of 0.43% and a PE rate of 0.34%42. While age was not a significant risk factor in that series (p = 0.068), none of the DVTs occurred in patients younger than forty years of age. Confounding the potential role of chemoprophylaxis, one study randomized 105 patients undergoing surgical repair of acute Achilles tendon ruptures to receive either placebo or dalteparin, with ninety-one patients completing the study40. Screening ultrasound examinations, performed three and six weeks postoperatively, demonstrated relatively high rates of VTED but showed no protective effect afforded by dalteparin over placebo (34% compared with 36%; p = 0.8).
The use of total ankle arthroplasty has increased in the past two decades48. Accordingly, a recent qualitative review study examined thirty articles published between 1999 and 2013 to assess the rates of VTED after total ankle arthroplasty49. The use of pharmacological prophylaxis against VTED was common but inconsistent among the studies, and the authors of the review did not perform a formal meta-analysis. If one manually tabulates the rates of VTED among the quoted studies, one finds that, among 3613 patients undergoing 3826 total ankle arthroplasties, the overall rate of DVT was 1.3% and one patient (0.03%) presented with a symptomatic PE, but it is impossible to draw conclusions because a vast number of these patients were receiving chemoprophylaxis and bleeding complications were not reliably assessed.
Efficacy of Prophylaxis Against Venous Thromboembolic Disease
Any risk-benefit analysis obligates one to define the potential benefits. While studies are limited, many of those that have examined the efficacy of pharmacological prophylaxis against VTED after foot and ankle surgery have suggested that its use may not invariably lower the incidence of VTED. One study compared the rate of symptomatic VTED in 1078 patients who received aspirin (75 mg) after elective foot and ankle surgery with that in 1576 patients who did not receive chemoprophylaxis and demonstrated that the overall incidence of VTED was 0.42% (0.27% for DVT and 0.15% for PE), with no protective effect being attributable to aspirin50. It is unclear how the dose of aspirin (75 mg) was chosen, and previous studies in the arthroplasty literature have involved total daily dosages ranging from 325 to 650 mg51,52. As previously discussed, a prospective trial of 105 patients undergoing acute Achilles tendon rupture repair followed by six weeks of immobilization also demonstrated no decrease in the rate of VTED when patients who received dalteparin were compared with those who did not40. The baseline DVT rates in that study were 35% across all study participants, but all participants underwent ultrasonography, irrespective of whether they were symptomatic for VTED. Among the 180 patients who were excluded from the study for various reasons, the rate of symptomatic DVT was only 5.6% and four of ten of these DVTs occurred despite the use of postoperative dalteparin. A similar prospective trial involving 272 patients undergoing ankle fracture repair with postoperative ultrasound screening of all participants also demonstrated no protective effect afforded by dalteparin, with DVT rates of 21% and 28%, respectively, among patients who did and did not receive chemoprophylaxis39. In another study involving 665 patients (701 total ankle arthroplasties) who received postoperative dalteparin, the rate of symptomatic DVT was 3.9%, with the vast majority (84.6%) of DVTs being distal53. Obesity (OR = 6.54), prior VTED (OR = 5.43), and limited weight-bearing (OR = 3.57) were significant risk factors. A prospective, randomized study of 300 patients who were managed with at least three weeks of casting and who underwent venography on cast removal demonstrated an almost 20% incidence of DVT; this rate was not significantly different among patients who had received prophylaxis with low-molecular-weight heparin38. Yet another recent study demonstrated no difference in the rate of clot progression among seventy-six patients who received anticoagulants for the treatment of isolated soleal and gastrocnemius vein thrombosis as compared with sixty-five patients who did not receive treatment54.
Need for Treatment
Surprisingly few data exist in the current literature as to the actual impact of VTED events on patient quality of life, although the scant information that is available underscores that symptomatic thrombotic events continue to lower health status at two years, especially if patients develop post-thrombotic syndrome55,56. Proximal DVTs in the lower extremity often require treatment in order to prevent both post-thrombotic syndrome and clinical PE57. Both of these sequelae, however, may be much less applicable to DVTs that are localized distal to the popliteal fossa, which represent the vast majority of DVTs affecting foot and ankle patients41. One study examining post-thrombotic syndrome demonstrated that proximal location, and not thrombus size, underlay progression to post-thrombotic syndrome58. Among distal DVTs, progression to post-thrombotic syndrome may be very low or even negligible59. A more critical clinical fear is progression to clinical PE by virtue of a distal DVT becoming a proximal DVT. A historical review of twenty studies involving a total of 1995 patients with lower extremity DVTs demonstrated that, on average, approximately 10% (range, 0% to 20%) of distal DVTs propagate proximally, although none of the 1995 patients had development of a fatal PE60. More recent data based on the use of serial ultrasound suggested that, left untreated, distal DVTs propagate proximally only 1% to 5.7% of the time61. A lower rate of proximal progression has been observed in association with muscular vein thromboses in the gemellar and soleal veins (as opposed to peroneal or posterior tibial veins), with one study showing only a 3% rate of proximal extension without treatment62. In turn, the progression of proximal DVT to symptomatic PE is dependent on circumstances, with the risk being lower if the initial precipitating factor is a singular event that resolves, as in the case of injury, as opposed to a prolonged risk factor such as malignancy6. Furthermore, the sensitivity of computed tomography (CT) scans in detecting PE may lead to overdetection of clinically inconsequential events, such as small subsegmental emboli, a conclusion buttressed by the fact that the per capita incidence of PE in the United States has doubled since the introduction of CT scanning for the detection of PE, with no change in mortality25,63. Notably, only half of patients presenting with clinically symptomatic PE will have a positive finding on lower extremity ultrasound, possibly because of the inability of ultrasound to evaluate pelvic veins or complete embolization of the lower extremity thrombus64. All of these considerations must be balanced against the realization that a three-month period of therapeutic anticoagulant therapy itself carries a hemorrhagic risk of 0.6% to 1.2% and a fatal bleeding risk of 0.1% to 0.4%61.
The literature examining the incidence of VTED after foot and ankle surgery is scant. In spite of this scarcity of data, foot and ankle surgeons must nonetheless make clinical decisions against a backdrop of global quality initiatives that largely consider VTED to be a preventable event. Existing studies of variable quality have demonstrated a wide range of rates, partly because of study design. The reported rates of VTED vary depending on whether imaging studies were used to screen all patients irrespective of symptoms and whether only symptomatic patients were investigated. The former studies are generally prospective and count all VTED events, irrespective of their clinical relevance. The latter studies are frequently retrospective and therefore may underestimate the incidence of VTED. It is important to note that there is a substantial heterogeneity of diagnoses and treatments among the population of patients undergoing foot and ankle procedures as compared with those undergoing hip and knee arthroplasty. Certain patient populations may be at higher risk for clinically relevant VTED, including patients over the age of forty years with acute Achilles tendon ruptures or over the age of fifty years with ankle fractures, patients with diabetes mellitus, patients with connective-tissue inflammatory diseases, and patients with a history of VTED. Other factors to consider may include oral contraceptive use, smoking history, recent air travel, and a familial history of VTED events65. Unfortunately, the overall rates of VTED do not meaningfully estimate this complication, and it remains impossible to quantify the consequences of VTED or its prophylaxis until better outcomes research becomes available that employs quality-of-life measures to help characterize such events.
Most DVTs in patients undergoing foot and ankle procedures are distal to the popliteal fossa and, fortunately, these are less likely to cause post-thrombotic syndrome or PE. However, catastrophic events, including death, have been reported in association with VTED. Direct comparison studies have suggested that pharmacological prophylaxis may not always significantly alter this rate. Such findings must in turn be weighed against the risks of pharmacological prophylaxis, such as bleeding and heparin-induced thrombocytopenia. Ultimately, clinically relevant VTED events do occur among patients undergoing foot and ankle procedures and need to be more accurately characterized to improve the overall quality of care. Existing risk-stratification systems, while proven to be predictive in other specialties, are difficult to apply to patients undergoing foot and ankle procedures and do not provide insight into the potentially deleterious side effects of chemoprophylaxis around the foot and ankle. Additional prospective randomized trials are necessary to better understand how to approach and risk-stratify VTED in these patients, to determine which subsets of patients should receive routine prophylaxis (chemical or otherwise), and to determine the duration and type of therapy.
Source of Funding: No external funds were used in the preparation of this review.
Investigation performed at the Foot and Ankle Center, Massachusetts General Hospital/Newton-Wellesley Hospital, Waltham, Massachusetts
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 © 2015 by The Journal of Bone and Joint Surgery, Incorporated