Abstract
Background:
A major complication of foot and ankle arthrodesis is nonunion, which occurs in approximately 12% of cases. Various factors influence a patient’s risk for nonunion following foot and ankle arthrodesis. We surveyed international foot and ankle surgeons to determine (1) risk factors perceived most important for nonunion, (2) factors considered absolute contraindications for arthrodesis, and (3) differences among expert groups regarding perceived risk factors and their stratification.
Methods:
A questionnaire was e-mailed to members of a major foot and ankle journal editorial board and four foot and ankle society executive committees. The relative risk of 18 potential nonunion risk factors was rated from 1 to 10, using smoking 1 pack/day as a benchmark score of 5.00.
Results:
The response rate was 72% (100/139); 81% declared foot and ankle surgery encompasses >90% of their practice. The highest perceived risk factors (p < 0.001) were smoking 2 packs/day (mean score 8.69), lack of fusion site stability (8.66), and poor local vascularity (7.66). The least important risk factors (p < 0.001) were perceived to be age >60 years (mean score 2.54), rheumatoid arthritis (3.05), and osteoporosis (3.56). The most frequently cited absolute contraindications to arthrodesis surgery were local infection (46%), poor local vascularity (41%), and smoking (32%).
Conclusion:
To improve arthrodesis outcomes, resource allocation and patient and surgeon education should focus on smoking, construct stability, and local vascularity. Development of an objective nonunion risk assessment tool to identify patients at risk for nonunion using these results could help maximize the efficiency of available resources.
Introduction
A major complication of foot and ankle arthrodesis is nonunion, which may lead to failure of the procedure and require reoperation or a repeat arthrodesis. 1,2 Nonunion occurs in approximately 12% (range: 3–23%) of foot and ankle fusions 3 –6 and may occur more frequently in patients presenting with risk factors. 7,8
Various factors collectively influence a patient’s risk for nonunion following foot and ankle arthrodesis and fracture surgery. 9 However, a current concepts review of nonunion risk factors in foot and ankle arthrodesis surgery 8 (Table 1) found only fair evidence (grade B) to support smoking, 10 –13 diabetes, 14 –17 and soft tissue injury 18 –20 as risk factors for nonunion. There was little conclusive evidence (grade C or I) to implicate other risk factors for nonunion following arthrodesis or fracture surgery.
Literature review of risk factors and grades of evidence.
aGrade A indicates good clinical evidence from at least one randomized controlled trial and other good quality studies. Grade B indicates fair clinical evidence from well-conducted clinical studies but no randomized clinical trials. Grade C indicates conflicting clinical evidence. Grade I indicates insufficient clinical evidence.
Identification of these risk factors and appreciation of their relative significance enable the cognizant physician to modify a patient’s risk profile, improve fusion rates, 18 and potentially reduce patient morbidity and the associated health-care costs. Strategies for risk modification may be preoperative, intraoperative, or postoperative.
The purpose of this study was to conduct a survey of senior international orthopedic foot and ankle specialists in the English-speaking world to determine (1) current opinions on the most important risk factors for nonunion, (2) factors that are considered to be absolute contraindications for foot and ankle arthrodesis, and (3) any differences among expert groups with respect to the perceived risk factors and their stratification.
Materials and methods
A questionnaire was developed in a pilot study at a large, urban, teaching hospital in Canada. Four attending orthopedic surgeons and three orthopedic fellows, all with foot and ankle subspecialty training, were asked to identify nonunion as a mild, moderate, or severe problem in their practice, to list the 10 most important risk factors that concerned them, and to identify any risk factors that served as absolute contraindications to foot and ankle arthrodesis. Based on their responses, 13 risk factors were shortlisted. Smoking was identified as a risk factor by all (7/7) respondents in the pilot study, and 6/7 considered smoking to be one of the top three most significant risk factors. Consequently, smoking was used as the benchmark against which the other risk factors would be weighted.
The survey was then designed and sent to the same seven specialists. Based on their feedback, subtle ambiguities and risk factors that were considered “open to interpretation” were adjusted. For example, severity of diabetes was categorized by HbA1c levels, and extent of obesity was based on body mass index (BMI). The final version of the questionnaire consisted of 18 risk factors (Figure 1). Respondents were asked to rate, on a scale of 0 to 10, their perception of the relative risk of the 18 risk factors in relation to the benchmark risk factor of smoking 1 pack a day, which was assigned a risk significance of 5/10 on the scale. Respondents were also asked to declare any risk factors they would consider an absolute contraindication to arthrodesis surgery.
Survey distributed to expert orthopedic foot and ankle surgeons.
The survey questionnaire was e-mailed to individuals who, by the nature of their position, could be described as expert orthopedic foot and ankle surgeons. Nonresponders were contacted again 4 weeks after the initial survey was sent out. The survey cohort consisted of (1) all editorial members of Foot and Ankle International (FAI; n = 94); and (2) the current executive committee members of the British Foot and Ankle Society (BOFAS; n = 19), the International Federation of Foot and Ankle Surgeons (IFFAS; n = 11), the Canadian Orthopaedic Foot and Ankle Society (COFAS; n = 8), and the Australian Orthopaedic Foot and Ankle Society (AustOFAS; n = 7).
Statistical analysis
Descriptive statistics were used to summarize the responses to all questions. The two-way paired t-test for unequal variance was used to compare the benchmark risk factor (smoking 1 pack/day—5/10) to the other risk factors listed. Given the multiple comparisons, p < 0.001 was chosen to define a significant difference.
Results
Of the 139 foot and ankle experts, 100 (72%) completed the survey. The differential response rates were FAI 67/94 (72%), BOFAS 14/19 (74%), IFFAS 7/11 (64%), COFAS 6/8 (75%), and AustOFAS 6/7 (86%). Eighty-one percent (81/100) of the respondents declared that more than 90% of their practice was foot and ankle surgery, 11% declared 50–75%, and the remaining 8% declared 30–50% of their practice consisted of foot and ankle surgery.
Mean scores for each risk factor were calculated, compared against the benchmark score of 5/10 for smoking 1 pack/day, and tabulated into three categories: (1) risk factors with mean scores significantly greater than 5/10 (p < 0.001), (2) risk factors with mean scores not statistically different from 5/10 (p > 0.001), and (3) risk factors with mean scores significantly lower than 5/10 (p < 0.001; Table 2). Risk factors for nonunion considered by the experts to be more important than smoking 1 pack/day included smoking 2 packs/day, lack of fusion site stability, poor local vascularity, poor compliance with weight bearing, gaps at the fusion site, poor soft tissue envelope, and poorly controlled diabetes with HbA1c above 8%.
Risk factors compared to smoking 1 pack/day.
BMI: body mass index; NSAIDs: nonsteroidal anti-inflammatory drugs.
The response to absolute contraindications to arthrodesis surgery was heterogenous but invariably involved one of three risk factors or a combination thereof: local infection (46%), poor local vascularity (41%), or smoking (32%).
We analyzed the surveyed expert groups (i.e. FAI, BOFAS, COFAS, IFFAS, and AustOFAS) for any differences between the groups with respect to their perceptions of the significance of individual risk factors as well as the overall trend for risk factor stratification. There was no significant difference between the surveyed groups (p > 0.01 for all risk factors).
Discussion
Nonunion in arthrodesis surgery is a classic paradigm of a clinical outcome attributed to a multifactorial etiology. Given the complex interplay between different risk factors and the probability of nonunion, it is not surprising there is a paucity of data to validate the role of even the most commonly cited risk factors for nonunion. We sought to determine the opinions of international orthopedic foot and ankle specialists on the most important risk factors for nonunion and any factors considered absolute contraindications for arthrodesis.
The majority of the surveyed foot and ankle experts considered heavy smoking (i.e. 2 packs/day), lack of construct stability, poor local vascularity, poor compliance with postoperative weight-bearing protocols, and gaps at the fusion site as the most significant risk factors for nonunion. Fair evidence in the literature supports the role of smoking as a risk factor for nonunion. 21,22 While little evidence correlates heavy smoking with an increased risk for nonunion, 23 the surveyed experts thought that doubling the nicotine load translated to a significant increase in risk. The role of construct stability, driven by the AO Foundation philosophy of rigid osteosynthesis, is supported primarily by animal studies. 8 Human studies have found the use of modern internal fixation techniques and bone grafting to effectively reduce nonunion rates in triple arthrodesis 24,25 and transfibular ankle arthrodesis. 26 Conventional wisdom and traditional surgical teaching have long attributed both wound and bone healing to adequate local perfusion. 27 While a strong body of level II experimental literature indicates the angiogenic response is pivotal in bone healing, 28,29 this evidence has yet to be reflected clinically. A large interfragmentary gap has been demonstrated to contribute to poor bone healing and to affect the quality of tissue bridging the gap in animal studies, 30 but human studies to support this association are lacking. 8
Poor patient compliance with weight-bearing protocols has been considered a risk factor for nonunion. Current literature hypothesizes that this may be partly attributable to poor fixation techniques historically. 8 Indeed, there is little evidence to support patient compliance as a risk factor. 31,32 The surveyed experts, however, felt patient compliance was a notable risk factor. This perception may vary, depending on the anatomy of the joint being fused (talonavicular vs. first metatarsophalangeal joint), but this differential view was not probed by the survey. Further research on the potential effect of weight bearing on nonunion when using modern fixation techniques is warranted.
Younger age and rheumatoid arthritis were regarded as the least significant risk factors. There is insufficient evidence to implicate age as a risk factor for nonunion. 33,34 The role of rheumatoid arthritis as a risk factor is also poorly substantiated. While some evidence suggests long-term corticosteroid therapy may be detrimental to fracture healing, 35 the isolated role of rheumatoid arthritis as a risk factor for nonunion has been difficult to demonstrate clinically and remains unfounded. 36
The majority of experts chose at least one of three risk factors, local infection, poor vascularity, or smoking, as an absolute contraindication for arthrodesis. While the roles of local infection and poor vascularity in nonunion have been difficult to demonstrate clinically, they are somewhat intuitive, given their well-established role in contributing to delayed wound healing. 37,38
The survey highlights some important findings in relation to fiercely debated risk factors for nonunion. While there is insufficient evidence in the current literature for or against the role of nonsteroidal anti-inflammatory medication in nonunion, 39,40 the experts perceived its deleterious effect as less significant than smoking 1 pack a day. The perceived role of diabetes was also clearly elucidated. While the impact of well-controlled diabetes was perceived as no more severe than smoking a pack a day, a poorly controlled diabetic with an HbA1c >8% was considered significantly more at risk for nonunion than the pack a day smoker.
This survey enabled the pooling of expertise on nonunion risk factors among international foot and ankle experts. Given the paucity of literature on nonunion risk factor identification or stratification, much of the current clinical practice on preoperative patient assessment, intraoperative technical assessments, and postoperative rehabilitation needs has been subjective. Indeed, what constitutes a case with a high risk of nonunion still remains poorly defined. The development of a robust risk assessment tool that is more objective than the clinical practices currently employed would enable arthrodesis surgery to be approached in a more scientific fashion. A tailored approach to each patient based on an objective risk profile would allow focused preoperative education on modifiable risk factors, selective use of costly intraoperative resources such as locking plates and synthetic bone graft substitutes, and a postoperative regimen that anticipates the need for multidisciplinary health-care professionals.
Limitations of this study include the 28% nonresponse rate, but this is considered low for e-mail-based surveys. 41 The response rate was consistent across organizations, minimizing the likelihood of sampling bias. Formal requirements for “expertise” in foot and ankle surgery were not established for this study. However, the chosen cohort represents academic leaders well informed on the current literature and standard of practice. It is possible a minority of the surveyed members may not be full-time practicing foot and ankle surgeons, although 81% of respondents indicated more than 90% of their practice was foot and ankle surgery. There is no other international cohort of foot and ankle subspecialists that ensures a homogenous quality of expertise that can be surveyed. Finally, the reliance on a mean score to afford weighting to the significance of individual risk factors may not be adequately sensitive.
We conclude that, in terms of the current state of practice, smoking, lack of construct stability, poor local vascularity, poor patient compliance with weight-bearing protocols, and intermedullary gaps at the fusion site are perceived as the most significant risk factors for nonunion following foot and ankle arthrodesis. As such, future allocation of resources and focused education of both patients and surgeons should begin with these risk factors. The development of a robust, objective risk assessment tool would facilitate preoperative, intraoperative, and postoperative strategies customized to individual patients to reduce the risk of nonunion.
Footnotes
Acknowledgments
The author thanks Dagmar Gross for assistance with the preparation of this article.
Declaration of conflicting interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.