ORIGINAL ARTICLE
Does Syndesmotic Injury Have a Negative Effect on Functional Outcome? A Multicenter Prospective Evaluation Jody Litrenta, MD,* David Saper, MD,* Paul Tornetta III, MD,* Laura Phieffer, MD,† Clifford B. Jones, MD,‡ Brian H. Mullis, MD,§ Kenneth Egol, MD,k Cory Collinge, MD,¶ Ross K. Leighton, MD,** William Ertl, MD,†† William M. Ricci, MD,‡‡ David Teague, MD,†† and Janos P. Ertl, MD§
Objective: To evaluate the effect of syndesmotic disruption on the functional outcomes of Weber B, SE4 ankle fractures treated operatively.
Setting: Multicenter trauma hospitals. Patients: Data were prospectively gathered during a previous, multicenter randomized trial including 242 patients (136 women, 106 men) from 9 trauma centers with operatively treated Weber B SE4 ankle fractures. There were 81 patients (35%) with syndesmotic instability confirmed intraoperatively after fibula fixation. Intervention: Functional evaluations were performed postoperatively at 6, 12, 26, and 52 weeks. The presence of symptomatic hardware and peroneal tendon discomfort was evaluated with 9–12 months of follow-up.
Main Outcome Measures: Functional outcomes evaluated included Short Musculoskeletal Function Assessment (SMFA), Bother index, and American Orthopaedic Foot and Ankle Society (AOFAS) scores. The recovery curve of the 2 groups was analyzed using a mixed linear regression analysis for repeated measures and included gender and race in the model. Symptomatic hardware and
Accepted for publication January 14, 2015. From the *Department of Orthopaedic Surgery, Boston University School of Medicine, Boston, MA; †Ohio State University; ‡Orthopaedic Associates of Michigan; §Indiana University; kNYU Hospital for Joint Disease; ¶Texas Health Forth Worth; **Dalhousie University; ††University of Oklahoma; and ‡‡Barnes-Jewish Hospital. Received Consultancy fees from Smith and Nephew, Biomet, Stryker (C.C.); Exactech (K.E.); Etex (R.K.L.); Biomet (B.H.M.); Stryker, Smith and Nephew, Biomet (W.M.R.); Grants from OMeGA, OREF, Synthes (K.E.); METRC DOD, NIH (C.B.J.); Medtronic, CIHR (R.K.L.); Synthes (B.H.M.); and AONA, COTA (W.M.R.); and Payment for lectures and others: Biomet, Etex, Synthes, Smith and Nephew, Stryker, Zimmer, and Bioventus (R.K.L.); and Synthes, Smith and Nephew (B.H.M.). The remaining authors report no conflict of interest. Presented in part at the Annual Meeting of the Orthopaedic Trauma Association, October 2013, Phoenix, AZ, the American Academy of Orthopaedic Surgeons Annual Meeting, March 2014, New Orleans, LA, and at the Combined COA/AOA, June 2014, Montreal, Canada. IRB approval was obtained for this study with DUA agreements from all centers. Reprints: Paul Tornetta III, MD, Boston Medical Center, 850 Harrison Avenue, Dowling 2 North, Boston Medical Center, Boston, MA 02118 (e-mail: [email protected]). Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.
410
| www.jorthotrauma.com
peroneal tendon discomfort were compared between the 2 groups with a x2 analysis.
Results: The adjusted mean linear regression analyses demonstrated that patients without a syndesmotic injury had better functional outcomes for some outcome measures. SMFA scores at 12 weeks were statistically lower in patients without syndesmotic injury (P = 0.02), but not at other visits. AOFAS scores were significantly higher (P = 0.0006), and Bother index trended toward lower results (P = 0.07) in patients without syndesmotic injury at all time points. Isolated analyses (T-tests) at 1 year demonstrated a difference in the SMFA (P = 0.04) and Bother index (P = 0.05), but not the AOFAS (P = 0.21). Men consistently demonstrated better recovery than women for all outcomes, whereas race was not significant for any measure. Symptomatic hardware and peroneal tendon irritation was not statistically different between the groups.
Conclusions: The recovery curves after ankle fractures were different based on syndesmotic injury. However, the difference was at the limit of clinical significance. Syndesmotic injury has a slightly detrimental effect on outcomes of operatively treated Weber B SE4 fractures. Key Words: Weber B ankle fracture, syndesmotic injury, functional outcomes, trauma
Level of Evidence: Prognostic Level II. See Instructions for Authors for a complete description of levels of evidence. (J Orthop Trauma 2015;29:410–413)
INTRODUCTION Ankle fractures are some of the most commonly encountered injuries in orthopaedic practice.1 Supinationexternal rotation (SE) type injuries are the most prevalent, and the majority are Weber B injuries. Multiple authors have demonstrated that syndesmotic injury can occur in association with Weber B, SE type ankle fractures.2–6 The clinical relevance of an associated syndesmotic injury is unclear and has been suggested to contribute to poorer outcomes. In a series of unstable ankle fractures, Egol et al7 demonstrated worse functional outcomes at 1 year in patients with syndesmotic injury using the American Orthopaedic Foot and Ankle Society (AOFAS) scoring system and the Short Musculoskeletal Function Assessment (SMFA).8,9 This study included patients with both Weber B and C fibula J Orthop Trauma Volume 29, Number 9, September 2015
Copyright © 2015 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
J Orthop Trauma Volume 29, Number 9, September 2015
pattern fractures. Outcomes by Lauge–Hansen or Weber classifications were not reported separately. Thus, it is unclear if syndesmotic instability was a negative factor for Weber B, SE pattern fractures, or whether the C type injuries drove the analysis. Previous research has demonstrated that syndesmotic instability is present in as high as 30%–40% of operatively treated Weber B SE4 ankle injuries.4,6 This instability can be identified intraoperatively after lateral malleolar fixation by an external rotation stress test under fluoroscopy.4,6,10 The clinical relevance of an associated syndesmotic injury in the most common ankle fracture type (SE, Weber B) has never been evaluated in isolation, because previous work has grouped all unstable ankle fractures together. This grouping confuses the picture because high fibula fractures with syndesmotic injury are of higher energy and may portend worse outcomes than the lesser degrees of ligamentous injury associated with Weber B injuries. Thus, the purpose of this study was to evaluate functional outcomes (AOFAS, SMFA, and Bother index scores), symptomatic hardware, and peroneal tendon irritation in a large series Weber B SE ankle fractures with and without associated syndesmotic injury. Our null hypothesis was that the presence of syndesmotic instability based on stress testing after fibular fixation would not significantly affect these outcomes.
METHODS Data were prospectively gathered during a previous randomized trial evaluating posterolateral antiglide plating versus lateral-based fibula plating for SE ankle fractures.11 In this trial, it was standard to keep patients non–weight bearing for 10–12 weeks if syndesmotic fixation was used and 4–6 weeks if it was not. Patients were evaluated at 12, 26, and 52 weeks postoperatively using the AOFAS, SMFA, and Bother index, and additionally at 6 weeks for the SMFA and Bother index. There were no statistical differences in outcomes between the 2 groups in any outcome measure, thus the data were pooled to evaluate syndesmotic injury. There were 242 patients, 136 women, and 106 women from 9 centers. All patients had an operatively treated Weber B, SE pattern ankle fractures. The presence of a syndesmotic injury was determined intraoperatively by applying an external rotation stress, and any talar lateralization was considered syndesmotic disruption. Thirty-five percent of Weber B, SE ankle fractures had an associated syndesmotic injury, and this did not differ between the centers.
Scored Outcome Measures Data collected at 6, 12, 24, and 52 weeks were used for our analysis. This included AOFAS, SMFA, and Bother index. Statistical analysis using mixed linear regression models for repeated measures was performed by a PhD statistician. In addition to the presence of a syndesmotic injury, gender and race were also included in the model. Because the 1-year outcomes were thought to be important, in addition to the regression analysis, we evaluated the outcome measures at 1 year using 2-tailed unpaired T-tests. Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.
Syndesmotic Injury and Functional Outcome
Symptomatic Hardware and Peroneal Tendon Irritation Clinical data regarding symptomatic hardware and peroneal tendon irritation were collected at the 9-month to 1-year follow-up. Patients were asked if their hardware was not palpable, palpable but not painful, caused shoe-wear restrictions, activity restrictions, or required or would require removal. Hardware was considered symptomatic if there was any complaint of restrictions, or required removal. Hardware that was not palpable or did not cause irritation was considered asymptomatic. We graded peroneal tendon irritation as none, tenderness, diminished function, or irritation with activities of daily life. All patients that indicated any answer other than no irritation were considered symptomatic. Chi-square analysis was performed comparing the percentages of patients with peroneal tendon irritation and symptomatic hardware in patients with and without syndesmosis fixation.
RESULTS Mixed Linear Regression of Functional Outcomes A mixed linear regression was performed using adjusted mean values, controlling for the presence of syndesmotic injury, race, and gender for all outcome measures at 12, 26, and 52 weeks, and additionally at 6 weeks for the SMFA and Bother index. For all outcome measures, the most significant finding was the consistent improvement over time for all patients. Also, men consistently demonstrated better recovery than women (SMFA, P = 0.002; Bother index, P = 0.008; AOFAS, P = 0.0006). Race was not significant in any model (SMFA, P = 0.66; Bother index, P = 0.28; AOFAS, P = 0.71). Patients without syndesmotic injury demonstrated better outcome scores, although this did not consistently reach statistical significance for all functional outcomes. Bother index scores were lower (lower score is better) at all time points in patients without a syndesmotic injury, and this finding approached statistical significance (P = 0.07). AOFAS scores were higher (higher score is better) at all time points in patients without syndesmosis injury, and this was statistically significant (P = 0.0006). Both groups demonstrated improvement over time in Bother index and AOFAS scores, with better scores at longer follow-up (P , 0.0001). The data for these 2 analyses are seen in Tables 1 and 2. It should be noted that the data presented include adjusted mean values for the model, not the actual mean values of the outcomes scores. SMFA scores demonstrated a statistically significant interaction between postoperative visit and the presence of syndesmotic injury; therefore, these results are reported separately according to visit. SMFA scores were lower (lower score is better) at only the 12-week visit in patients without syndesmotic injury (P = 0.04). All other time points demonstrated no significant differences between the 2 groups; also, as seen in other models, both groups continued to improve over time (Table 3). In summary, the mixed linear regression demonstrated consistent improvement with time, and statistically significant www.jorthotrauma.com |
411
Copyright © 2015 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
J Orthop Trauma Volume 29, Number 9, September 2015
Litrenta et al
TABLE 1. Bother Index—Mixed Linear Regression Adjusted Mean Syndesmotic injury Yes No Gender Female Male Race African American White Latino Other Visit 6 wk 12 wk 26 wk 52 wk
TABLE 3. SMFA—Mixed Linear Regression P
21.96 18.05
0.07
Adjusted Mean
P
Yes No Yes No Yes No Yes No
36.49 35.86 28.83 22.68 15.66 14.79 13.51 10.96
0.76
22.76 17.25
0.008
18.49 22.50 21.14 17.88
0.28
30.63 23.24 14.88 12.91
,0.0001
Week 12 Week 26 Week 52
differences based on the presence of syndesmotic injury in AOFAS scores at all time points and the SMFA at the 12week visit only.
One-year Functional Outcomes When evaluating the effect of syndesmotic injury using 1-year results in isolation, without controlling for gender or race, the actual mean scores of the SMFA and Bother index were statistically lower in patients without a syndesmotic injury, whereas the AOFAS demonstrated no difference (Table 4). T-tests demonstrated that patients without syndesmotic injury had lower SMFA scores than those with it (11.60 vs 17.19, P = 0.04). Similarly, patients without a syndesmotic injury also had lower Bother index scores at one year (12.06 vs 19.36, P = 0.05). Although patients without syndesmotic injury had higher AOFAS scores, this was not significant (85.89 vs 80.58, P = 0.21). These results are also represented graphically in Figure 1.
TABLE 2. AOFAS—Mixed Linear Regression Adjusted Mean Syndesmotic Injury Yes No Gender Female Male Race African American White Latino Other Visit 12 wk 26 wk 52 wk
412
Syndesmotic Injury Week 6
| www.jorthotrauma.com
P
82.96 90.08
0.0006
83.01 90.03
0.0006
84.70 87.10 86.74 87.54
0.71
83.72 91.19 84.65
,0.0001
0.02 0.73 0.32
Adjusted Mean
P
24.77 19.35
0.002
21.10 23.37 22.48 21.30
0.66
Gender Female Male Race African American White Latino Other
Symptomatic Hardware and Peroneal Tendon Irritation Although a greater number of patients with syndesmotic injury complained of peroneal tendon irritation and symptomatic hardware, these differences did not reach significance. Among patients with syndesmotic fixation, hardware was symptomatic in 17% of patients compared with 10% of those without syndesmotic fixation (P = 0.28). Peroneal tendon irritation was reported in 14% of patients with syndesmotic fixation as compared with 8% of those without (P = 0.24).
DISCUSSION Our series included a 35% rate of syndesmotic injury, which is consistent with previous reports.4,6 We evaluated these groups using 2 methods: a mixed linear regression and a standard comparison at 1 year to allow an evaluation of the recovery curve over time and the differences in those with and without a syndesmotic injury. The most dominant finding in the regression analysis was that both groups showed significant improvement with time. In addition, patients with a syndesmotic injury in association with a Weber B, SE ankle fractures have worse functional outcome scores than those without. In the mixed linear regression analysis, AOFAS scores at all time points and SMFA scores at the 12-week postoperative visit were statistically better in patients without a syndesmotic injury using adjusted mean values that TABLE 4. One-Year Functional Outcomes Group
SMFA*
Bother*
AOFAS
17.19 11.60 0.04
19.36 12.06 0.05
80.58 85.89 0.21
Syndesmotic injury No Syndesmotic injury P *Indicates statistical significance.
Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.
Copyright © 2015 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
J Orthop Trauma Volume 29, Number 9, September 2015
Syndesmotic Injury and Functional Outcome
FIGURE 1. Recovery curves of the SMFA, Bother index, and AOFAS scores. Editor’s note: A color image accompanies the online version of this article.
controls for gender and race. It is important to note that patients without a syndesmotic injury were treated with 4–6 weeks of non–weight bearing compared with 12 weeks in patients with a syndesmotic injury, and this protocol may explain the isolated difference seen in the SMFA at the 12week postoperative visit. Finally, we compared the AOFAS, SMFA, and Bother index of patients with and without syndesmotic injury at their final 1-year follow-up using t tests. These results demonstrated that patients without syndesmotic injury had statistically better results in both the SMFA and Bother index, but not the AOFAS scores. More importantly, the use of the T-tests demonstrated that the value of the differences between the groups was relatively small, and therefore may be of limited clinical significance. The absolute differences in the mean values of these groups was less than half of the reported SD for these instruments, which is considered below the level of clear clinical importance. Egol et al7 demonstrated worse SMFA and AOFAS scores at 1-year follow-up in patients with syndesmotic injury. However, this study included all unstable ankle fractures, including Weber B and C ankle fractures. The reason for poorer functional outcomes in patients with syndesmotic injury could be due to a number of factors, including fracture pattern. Weber C fractures may represent a higher energy injury and this could possibly have had a role in the outcomes. Furthermore, the accuracy of syndesmotic reduction may be different in Weber C compared to B type fractures. Sagi et al12 demonstrated worse functional outcomes in ankle fractures with a malreduced syndesmosis. In this study, 15% of patients that underwent open reduction and 44% with a closed reduction of the syndesmosis were determined postoperatively to be malreduced, and these patients had significantly worse SMFA scores. Therefore, differences in technique and quality of syndesmotic reduction between different fracture types may contribute to functional outcome, strengthening our study methodology of including only Weber B SE injuries that were open-reduced in association with plating. Our series did not demonstrate significant differences in peroneal tendon irritation or symptomatic hardware between those with and without syndesmotic fixation. Previous research has shown that as many as 31% of patients may have symptomatic hardware after lateral malleolar fixation.13 The presence of these symptoms may affect recovery. In our analysis, we included data from 9 months and 1 year to ensure that the maximum number of symptomatic patients was captured. Therefore, we feel confident that symptomatic patients only represent a reasonably small percentage of both groups.
Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.
We also demonstrated that women had a worse recovery curve than men for all outcome measures. This is consistent with previous report showing worse SF-36 and SMFA scores in women at a mean follow-up of 27 months after sustaining an unstable ankle fracture.14 Based on our analysis, the presence of a syndesmotic injury in a Weber B, SE ankle fracture contributes to slightly worse result in operatively treated fractures. In contrast to previous study, we showed these are relatively small differences when only Weber B SE injuries are evaluated. Syndesmotic injury does contribute to the recovery process but does not dominate patient outcomes. REFERENCES 1. Bucholz R, Heckman J, Court-Brown C, et al. Rockwood and Green’s Fractures in Adults. 7th ed. Philadelphia, PA: Lippincott Williams & Wilkins; 2005. 2. Xenos JS, Hopkinson WJ, Mulligan ME, et al. The tibiofibular syndesmosis. Methods of fixation and radiographic assessment. J Bone Joint Surg Am. 1995;77:847–856. 3. Ebraheim NA, Elgafy H, Padanilam T. Syndesmotic disruption in low fibular fractures associated with deltoid ligament injury. Clin Orthop Relat Res. 2003;409:260–267. 4. Jenkinson RJ, Sanders DW, Macleod MD, et al. Intraoperative diagnosis of syndesmosis injuries in external rotation ankle fractures. J Orthop Trauma 2005;19:604–609. 5. Weening B, Bhandari M. Predictors of functional outcome following transsyndesmotic screw fixation of ankle fractures. J Orthop Trauma. 2005;19;102–108. 6. Stark E, Tornetta P III, Creevy WR. Syndesmotic instability in weber B ankle fractures: a clinical evaluation. J Orthop Trauma. 2007;21: 643–646. 7. Egol KA, Pahk B, Walsh M, et al. Outcome after unstable ankle fracture: effect of syndesmotic stabilization. J Orthop Trauma. 2010;24:7–11. 8. Kitaoka HB, Alexander IJ, Adelaar RS, et al. Clinical rating systems for the ankle-hindfoot, midfoot, hallux, and lesser toes. Foot Ankle Int. 1994; 15:349–353. 9. Agel J, Obremsky W, Kregor P, et al. Administration of the short musculoskeletal function assessment: impact on office routine and physicianpatient interaction. Orthopedics. 2003;26:783–788. 10. Tornetta P III, Axelrad TW, Sibai TA, et al. Treatment of the stress positive ligamentous SE4 ankle fracture: incidence of syndesmotic injury and clinical decision making. J Orthop Trauma. 2012;26:659–661. 11. Tornetta P III, Phieffer L, Jones C, et al. Posterolateral antiglide versus lateral plating for SE pattern ankle fractures: A multicenter randomized control trial [Abstract]. OTA Paper #61. 2011. 12. Sagi HC, Shah AR, Sanders RW. The functional consequence of syndesmotic joint malreduction at a minimum 2-year follow-up. J Orthop Trauma. 2012;26:439–443. 13. Brown OL, Dirschl DR, Obremskey WT. Incidence of hardware-related pain and its effect on functional outcomes after open reduction and internal fixation of ankle fractures. J Orthop Trauma. 2001;15: 271–274. 14. Obremskey WR, Brown O, Driver R, et al. Comparison of SF-36 and short musculoskeletal functional assessment in recovery from fixation of unstable ankle fractures. Orthopedics. 2007;30:145–151.
www.jorthotrauma.com |
413
Copyright © 2015 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Does Syndesmotic Injury Have a Negative Effect on Functional Outcome? A Multicenter Prospective Evaluation Jody Litrenta, MD,* David Saper, MD,* Paul Tornetta III, MD,* Laura Phieffer, MD,† Clifford B. Jones, MD,‡ Brian H. Mullis, MD,§ Kenneth Egol, MD,k Cory Collinge, MD,¶ Ross K. Leighton, MD,** William Ertl, MD,†† William M. Ricci, MD,‡‡ David Teague, MD,†† and Janos P. Ertl, MD§
Objective: To evaluate the effect of syndesmotic disruption on the functional outcomes of Weber B, SE4 ankle fractures treated operatively.
Setting: Multicenter trauma hospitals. Patients: Data were prospectively gathered during a previous, multicenter randomized trial including 242 patients (136 women, 106 men) from 9 trauma centers with operatively treated Weber B SE4 ankle fractures. There were 81 patients (35%) with syndesmotic instability confirmed intraoperatively after fibula fixation. Intervention: Functional evaluations were performed postoperatively at 6, 12, 26, and 52 weeks. The presence of symptomatic hardware and peroneal tendon discomfort was evaluated with 9–12 months of follow-up.
Main Outcome Measures: Functional outcomes evaluated included Short Musculoskeletal Function Assessment (SMFA), Bother index, and American Orthopaedic Foot and Ankle Society (AOFAS) scores. The recovery curve of the 2 groups was analyzed using a mixed linear regression analysis for repeated measures and included gender and race in the model. Symptomatic hardware and
Accepted for publication January 14, 2015. From the *Department of Orthopaedic Surgery, Boston University School of Medicine, Boston, MA; †Ohio State University; ‡Orthopaedic Associates of Michigan; §Indiana University; kNYU Hospital for Joint Disease; ¶Texas Health Forth Worth; **Dalhousie University; ††University of Oklahoma; and ‡‡Barnes-Jewish Hospital. Received Consultancy fees from Smith and Nephew, Biomet, Stryker (C.C.); Exactech (K.E.); Etex (R.K.L.); Biomet (B.H.M.); Stryker, Smith and Nephew, Biomet (W.M.R.); Grants from OMeGA, OREF, Synthes (K.E.); METRC DOD, NIH (C.B.J.); Medtronic, CIHR (R.K.L.); Synthes (B.H.M.); and AONA, COTA (W.M.R.); and Payment for lectures and others: Biomet, Etex, Synthes, Smith and Nephew, Stryker, Zimmer, and Bioventus (R.K.L.); and Synthes, Smith and Nephew (B.H.M.). The remaining authors report no conflict of interest. Presented in part at the Annual Meeting of the Orthopaedic Trauma Association, October 2013, Phoenix, AZ, the American Academy of Orthopaedic Surgeons Annual Meeting, March 2014, New Orleans, LA, and at the Combined COA/AOA, June 2014, Montreal, Canada. IRB approval was obtained for this study with DUA agreements from all centers. Reprints: Paul Tornetta III, MD, Boston Medical Center, 850 Harrison Avenue, Dowling 2 North, Boston Medical Center, Boston, MA 02118 (e-mail: [email protected]). Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.
410
| www.jorthotrauma.com
peroneal tendon discomfort were compared between the 2 groups with a x2 analysis.
Results: The adjusted mean linear regression analyses demonstrated that patients without a syndesmotic injury had better functional outcomes for some outcome measures. SMFA scores at 12 weeks were statistically lower in patients without syndesmotic injury (P = 0.02), but not at other visits. AOFAS scores were significantly higher (P = 0.0006), and Bother index trended toward lower results (P = 0.07) in patients without syndesmotic injury at all time points. Isolated analyses (T-tests) at 1 year demonstrated a difference in the SMFA (P = 0.04) and Bother index (P = 0.05), but not the AOFAS (P = 0.21). Men consistently demonstrated better recovery than women for all outcomes, whereas race was not significant for any measure. Symptomatic hardware and peroneal tendon irritation was not statistically different between the groups.
Conclusions: The recovery curves after ankle fractures were different based on syndesmotic injury. However, the difference was at the limit of clinical significance. Syndesmotic injury has a slightly detrimental effect on outcomes of operatively treated Weber B SE4 fractures. Key Words: Weber B ankle fracture, syndesmotic injury, functional outcomes, trauma
Level of Evidence: Prognostic Level II. See Instructions for Authors for a complete description of levels of evidence. (J Orthop Trauma 2015;29:410–413)
INTRODUCTION Ankle fractures are some of the most commonly encountered injuries in orthopaedic practice.1 Supinationexternal rotation (SE) type injuries are the most prevalent, and the majority are Weber B injuries. Multiple authors have demonstrated that syndesmotic injury can occur in association with Weber B, SE type ankle fractures.2–6 The clinical relevance of an associated syndesmotic injury is unclear and has been suggested to contribute to poorer outcomes. In a series of unstable ankle fractures, Egol et al7 demonstrated worse functional outcomes at 1 year in patients with syndesmotic injury using the American Orthopaedic Foot and Ankle Society (AOFAS) scoring system and the Short Musculoskeletal Function Assessment (SMFA).8,9 This study included patients with both Weber B and C fibula J Orthop Trauma Volume 29, Number 9, September 2015
Copyright © 2015 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
J Orthop Trauma Volume 29, Number 9, September 2015
pattern fractures. Outcomes by Lauge–Hansen or Weber classifications were not reported separately. Thus, it is unclear if syndesmotic instability was a negative factor for Weber B, SE pattern fractures, or whether the C type injuries drove the analysis. Previous research has demonstrated that syndesmotic instability is present in as high as 30%–40% of operatively treated Weber B SE4 ankle injuries.4,6 This instability can be identified intraoperatively after lateral malleolar fixation by an external rotation stress test under fluoroscopy.4,6,10 The clinical relevance of an associated syndesmotic injury in the most common ankle fracture type (SE, Weber B) has never been evaluated in isolation, because previous work has grouped all unstable ankle fractures together. This grouping confuses the picture because high fibula fractures with syndesmotic injury are of higher energy and may portend worse outcomes than the lesser degrees of ligamentous injury associated with Weber B injuries. Thus, the purpose of this study was to evaluate functional outcomes (AOFAS, SMFA, and Bother index scores), symptomatic hardware, and peroneal tendon irritation in a large series Weber B SE ankle fractures with and without associated syndesmotic injury. Our null hypothesis was that the presence of syndesmotic instability based on stress testing after fibular fixation would not significantly affect these outcomes.
METHODS Data were prospectively gathered during a previous randomized trial evaluating posterolateral antiglide plating versus lateral-based fibula plating for SE ankle fractures.11 In this trial, it was standard to keep patients non–weight bearing for 10–12 weeks if syndesmotic fixation was used and 4–6 weeks if it was not. Patients were evaluated at 12, 26, and 52 weeks postoperatively using the AOFAS, SMFA, and Bother index, and additionally at 6 weeks for the SMFA and Bother index. There were no statistical differences in outcomes between the 2 groups in any outcome measure, thus the data were pooled to evaluate syndesmotic injury. There were 242 patients, 136 women, and 106 women from 9 centers. All patients had an operatively treated Weber B, SE pattern ankle fractures. The presence of a syndesmotic injury was determined intraoperatively by applying an external rotation stress, and any talar lateralization was considered syndesmotic disruption. Thirty-five percent of Weber B, SE ankle fractures had an associated syndesmotic injury, and this did not differ between the centers.
Scored Outcome Measures Data collected at 6, 12, 24, and 52 weeks were used for our analysis. This included AOFAS, SMFA, and Bother index. Statistical analysis using mixed linear regression models for repeated measures was performed by a PhD statistician. In addition to the presence of a syndesmotic injury, gender and race were also included in the model. Because the 1-year outcomes were thought to be important, in addition to the regression analysis, we evaluated the outcome measures at 1 year using 2-tailed unpaired T-tests. Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.
Syndesmotic Injury and Functional Outcome
Symptomatic Hardware and Peroneal Tendon Irritation Clinical data regarding symptomatic hardware and peroneal tendon irritation were collected at the 9-month to 1-year follow-up. Patients were asked if their hardware was not palpable, palpable but not painful, caused shoe-wear restrictions, activity restrictions, or required or would require removal. Hardware was considered symptomatic if there was any complaint of restrictions, or required removal. Hardware that was not palpable or did not cause irritation was considered asymptomatic. We graded peroneal tendon irritation as none, tenderness, diminished function, or irritation with activities of daily life. All patients that indicated any answer other than no irritation were considered symptomatic. Chi-square analysis was performed comparing the percentages of patients with peroneal tendon irritation and symptomatic hardware in patients with and without syndesmosis fixation.
RESULTS Mixed Linear Regression of Functional Outcomes A mixed linear regression was performed using adjusted mean values, controlling for the presence of syndesmotic injury, race, and gender for all outcome measures at 12, 26, and 52 weeks, and additionally at 6 weeks for the SMFA and Bother index. For all outcome measures, the most significant finding was the consistent improvement over time for all patients. Also, men consistently demonstrated better recovery than women (SMFA, P = 0.002; Bother index, P = 0.008; AOFAS, P = 0.0006). Race was not significant in any model (SMFA, P = 0.66; Bother index, P = 0.28; AOFAS, P = 0.71). Patients without syndesmotic injury demonstrated better outcome scores, although this did not consistently reach statistical significance for all functional outcomes. Bother index scores were lower (lower score is better) at all time points in patients without a syndesmotic injury, and this finding approached statistical significance (P = 0.07). AOFAS scores were higher (higher score is better) at all time points in patients without syndesmosis injury, and this was statistically significant (P = 0.0006). Both groups demonstrated improvement over time in Bother index and AOFAS scores, with better scores at longer follow-up (P , 0.0001). The data for these 2 analyses are seen in Tables 1 and 2. It should be noted that the data presented include adjusted mean values for the model, not the actual mean values of the outcomes scores. SMFA scores demonstrated a statistically significant interaction between postoperative visit and the presence of syndesmotic injury; therefore, these results are reported separately according to visit. SMFA scores were lower (lower score is better) at only the 12-week visit in patients without syndesmotic injury (P = 0.04). All other time points demonstrated no significant differences between the 2 groups; also, as seen in other models, both groups continued to improve over time (Table 3). In summary, the mixed linear regression demonstrated consistent improvement with time, and statistically significant www.jorthotrauma.com |
411
Copyright © 2015 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
J Orthop Trauma Volume 29, Number 9, September 2015
Litrenta et al
TABLE 1. Bother Index—Mixed Linear Regression Adjusted Mean Syndesmotic injury Yes No Gender Female Male Race African American White Latino Other Visit 6 wk 12 wk 26 wk 52 wk
TABLE 3. SMFA—Mixed Linear Regression P
21.96 18.05
0.07
Adjusted Mean
P
Yes No Yes No Yes No Yes No
36.49 35.86 28.83 22.68 15.66 14.79 13.51 10.96
0.76
22.76 17.25
0.008
18.49 22.50 21.14 17.88
0.28
30.63 23.24 14.88 12.91
,0.0001
Week 12 Week 26 Week 52
differences based on the presence of syndesmotic injury in AOFAS scores at all time points and the SMFA at the 12week visit only.
One-year Functional Outcomes When evaluating the effect of syndesmotic injury using 1-year results in isolation, without controlling for gender or race, the actual mean scores of the SMFA and Bother index were statistically lower in patients without a syndesmotic injury, whereas the AOFAS demonstrated no difference (Table 4). T-tests demonstrated that patients without syndesmotic injury had lower SMFA scores than those with it (11.60 vs 17.19, P = 0.04). Similarly, patients without a syndesmotic injury also had lower Bother index scores at one year (12.06 vs 19.36, P = 0.05). Although patients without syndesmotic injury had higher AOFAS scores, this was not significant (85.89 vs 80.58, P = 0.21). These results are also represented graphically in Figure 1.
TABLE 2. AOFAS—Mixed Linear Regression Adjusted Mean Syndesmotic Injury Yes No Gender Female Male Race African American White Latino Other Visit 12 wk 26 wk 52 wk
412
Syndesmotic Injury Week 6
| www.jorthotrauma.com
P
82.96 90.08
0.0006
83.01 90.03
0.0006
84.70 87.10 86.74 87.54
0.71
83.72 91.19 84.65
,0.0001
0.02 0.73 0.32
Adjusted Mean
P
24.77 19.35
0.002
21.10 23.37 22.48 21.30
0.66
Gender Female Male Race African American White Latino Other
Symptomatic Hardware and Peroneal Tendon Irritation Although a greater number of patients with syndesmotic injury complained of peroneal tendon irritation and symptomatic hardware, these differences did not reach significance. Among patients with syndesmotic fixation, hardware was symptomatic in 17% of patients compared with 10% of those without syndesmotic fixation (P = 0.28). Peroneal tendon irritation was reported in 14% of patients with syndesmotic fixation as compared with 8% of those without (P = 0.24).
DISCUSSION Our series included a 35% rate of syndesmotic injury, which is consistent with previous reports.4,6 We evaluated these groups using 2 methods: a mixed linear regression and a standard comparison at 1 year to allow an evaluation of the recovery curve over time and the differences in those with and without a syndesmotic injury. The most dominant finding in the regression analysis was that both groups showed significant improvement with time. In addition, patients with a syndesmotic injury in association with a Weber B, SE ankle fractures have worse functional outcome scores than those without. In the mixed linear regression analysis, AOFAS scores at all time points and SMFA scores at the 12-week postoperative visit were statistically better in patients without a syndesmotic injury using adjusted mean values that TABLE 4. One-Year Functional Outcomes Group
SMFA*
Bother*
AOFAS
17.19 11.60 0.04
19.36 12.06 0.05
80.58 85.89 0.21
Syndesmotic injury No Syndesmotic injury P *Indicates statistical significance.
Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.
Copyright © 2015 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
J Orthop Trauma Volume 29, Number 9, September 2015
Syndesmotic Injury and Functional Outcome
FIGURE 1. Recovery curves of the SMFA, Bother index, and AOFAS scores. Editor’s note: A color image accompanies the online version of this article.
controls for gender and race. It is important to note that patients without a syndesmotic injury were treated with 4–6 weeks of non–weight bearing compared with 12 weeks in patients with a syndesmotic injury, and this protocol may explain the isolated difference seen in the SMFA at the 12week postoperative visit. Finally, we compared the AOFAS, SMFA, and Bother index of patients with and without syndesmotic injury at their final 1-year follow-up using t tests. These results demonstrated that patients without syndesmotic injury had statistically better results in both the SMFA and Bother index, but not the AOFAS scores. More importantly, the use of the T-tests demonstrated that the value of the differences between the groups was relatively small, and therefore may be of limited clinical significance. The absolute differences in the mean values of these groups was less than half of the reported SD for these instruments, which is considered below the level of clear clinical importance. Egol et al7 demonstrated worse SMFA and AOFAS scores at 1-year follow-up in patients with syndesmotic injury. However, this study included all unstable ankle fractures, including Weber B and C ankle fractures. The reason for poorer functional outcomes in patients with syndesmotic injury could be due to a number of factors, including fracture pattern. Weber C fractures may represent a higher energy injury and this could possibly have had a role in the outcomes. Furthermore, the accuracy of syndesmotic reduction may be different in Weber C compared to B type fractures. Sagi et al12 demonstrated worse functional outcomes in ankle fractures with a malreduced syndesmosis. In this study, 15% of patients that underwent open reduction and 44% with a closed reduction of the syndesmosis were determined postoperatively to be malreduced, and these patients had significantly worse SMFA scores. Therefore, differences in technique and quality of syndesmotic reduction between different fracture types may contribute to functional outcome, strengthening our study methodology of including only Weber B SE injuries that were open-reduced in association with plating. Our series did not demonstrate significant differences in peroneal tendon irritation or symptomatic hardware between those with and without syndesmotic fixation. Previous research has shown that as many as 31% of patients may have symptomatic hardware after lateral malleolar fixation.13 The presence of these symptoms may affect recovery. In our analysis, we included data from 9 months and 1 year to ensure that the maximum number of symptomatic patients was captured. Therefore, we feel confident that symptomatic patients only represent a reasonably small percentage of both groups.
Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.
We also demonstrated that women had a worse recovery curve than men for all outcome measures. This is consistent with previous report showing worse SF-36 and SMFA scores in women at a mean follow-up of 27 months after sustaining an unstable ankle fracture.14 Based on our analysis, the presence of a syndesmotic injury in a Weber B, SE ankle fracture contributes to slightly worse result in operatively treated fractures. In contrast to previous study, we showed these are relatively small differences when only Weber B SE injuries are evaluated. Syndesmotic injury does contribute to the recovery process but does not dominate patient outcomes. REFERENCES 1. Bucholz R, Heckman J, Court-Brown C, et al. Rockwood and Green’s Fractures in Adults. 7th ed. Philadelphia, PA: Lippincott Williams & Wilkins; 2005. 2. Xenos JS, Hopkinson WJ, Mulligan ME, et al. The tibiofibular syndesmosis. Methods of fixation and radiographic assessment. J Bone Joint Surg Am. 1995;77:847–856. 3. Ebraheim NA, Elgafy H, Padanilam T. Syndesmotic disruption in low fibular fractures associated with deltoid ligament injury. Clin Orthop Relat Res. 2003;409:260–267. 4. Jenkinson RJ, Sanders DW, Macleod MD, et al. Intraoperative diagnosis of syndesmosis injuries in external rotation ankle fractures. J Orthop Trauma 2005;19:604–609. 5. Weening B, Bhandari M. Predictors of functional outcome following transsyndesmotic screw fixation of ankle fractures. J Orthop Trauma. 2005;19;102–108. 6. Stark E, Tornetta P III, Creevy WR. Syndesmotic instability in weber B ankle fractures: a clinical evaluation. J Orthop Trauma. 2007;21: 643–646. 7. Egol KA, Pahk B, Walsh M, et al. Outcome after unstable ankle fracture: effect of syndesmotic stabilization. J Orthop Trauma. 2010;24:7–11. 8. Kitaoka HB, Alexander IJ, Adelaar RS, et al. Clinical rating systems for the ankle-hindfoot, midfoot, hallux, and lesser toes. Foot Ankle Int. 1994; 15:349–353. 9. Agel J, Obremsky W, Kregor P, et al. Administration of the short musculoskeletal function assessment: impact on office routine and physicianpatient interaction. Orthopedics. 2003;26:783–788. 10. Tornetta P III, Axelrad TW, Sibai TA, et al. Treatment of the stress positive ligamentous SE4 ankle fracture: incidence of syndesmotic injury and clinical decision making. J Orthop Trauma. 2012;26:659–661. 11. Tornetta P III, Phieffer L, Jones C, et al. Posterolateral antiglide versus lateral plating for SE pattern ankle fractures: A multicenter randomized control trial [Abstract]. OTA Paper #61. 2011. 12. Sagi HC, Shah AR, Sanders RW. The functional consequence of syndesmotic joint malreduction at a minimum 2-year follow-up. J Orthop Trauma. 2012;26:439–443. 13. Brown OL, Dirschl DR, Obremskey WT. Incidence of hardware-related pain and its effect on functional outcomes after open reduction and internal fixation of ankle fractures. J Orthop Trauma. 2001;15: 271–274. 14. Obremskey WR, Brown O, Driver R, et al. Comparison of SF-36 and short musculoskeletal functional assessment in recovery from fixation of unstable ankle fractures. Orthopedics. 2007;30:145–151.
www.jorthotrauma.com |
413
Copyright © 2015 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
