Fixing the Syndesmosis: Suture Button Versus Screws
Dr. Nicole Stevens and Dr. Philipp Leucht comment on the results of a randomized controlled trial comparing anatomic reduction of the syndesmosis with these 2 methods of fixation.
Nicole Stevens, MD, and Philipp Leucht, MD
Anderson MR, Frihagen F, Helund JC, Madsen JE, Figved W. Randomized trial comparing suture button with single syndesmotic screw for syndesmosis injury. J Bone Joint Surg Am. 2018; 100:2-12.
Syndesmotic injuries occur in 10% to 13% of ankle fractures. As one of the primary stabilizers of the tibiofibular relationship, anatomic fixation of the syndesmosis is critical for ankle mortise congruity.
The surgeon has a myriad of options for stabilizing the injured syndesmosis. Historically, stabilization meant the use of screws of various permutations, including quadricortical screws, tricortical screws, a singular screw, and multiple screws. More recently, the suture button has come into favor, chiefly due to the slight plasticity of the construct and the lower incidence of hardware removal.
The goal of the study by Anderson et al was to compare use of a single quadricortical syndesmotic screw fixation with the use of suture button fixation of syndesmosis injury. The primary outcome was the American Orthopedic Foot and Ankle Society (AOFAS) ankle-hindfoot score. Other quality measures included the Olerud-Molander Ankle (OMA) score, visual analog scale (VAS), and EuroQol-5D (EQ-5D) index. CT scans of both ankles were compared at 2 weeks, 1 year, and 2 years postoperatively to measure anatomic reduction of the syndesmosis. Follow-up was recorded for 2 years.
This 2-center study included patients age 18 to 70 years who underwent operative fixation of OTA/AO type 44-C ankle fractures. Patients were randomized to suture button or syndesmotic screw fixation preoperatively, 46 patients in the suture button group and 45 in the screw fixation group. At the time of randomization, the screw fixation group had a higher number of patients with a combination of medial and posterior fractures, as well as osteochondral lesions, which was statistically significant.
Surgery was performed by the on-call orthopedist, frequently with assistance by a senior resident. Reduction of the syndesmosis was performed via manual reduction or clamp fixation, and then the chosen fixation strategy was employed. Patients were non-weight-bearing for 2 weeks, partial weight-bearing for the following 4 weeks, and then transitioned to weight-bearing as tolerated.
Postoperatively, the median AOFAS score was higher in the suture button group than in the screw fixation group at all time points, which was statistically significant at 6 months (89 vs 87, respectively; P=0.008), 1 year (96 vs 87, respectively; P< 0.001), and 2 years (96 vs 86, respectively; P=0.001). This was also true in the OMA score starting at 6 months. Furthermore, patients in the suture button group had less pain with walking and activities.
Due to the imbalance in randomization of the fracture patterns, a subgroup analysis was performed, and it demonstrated that suture button patients had higher AOFAS scores at 1 year and 2 years than screw fixation patients, even with posterior and medial malleolus fractures. .
Radiographically, patients in the screw fixation group were 3.4 times more likely to have talar osteophytes than patients in the suture button group. On CT scans, the tibiofibular distance – a marker of syndesmotic reduction – was similar between the 2 groups at 2 weeks postoperatively, but it increased in the screw group at 1 year and 2 years postoperatively. When comparing the injured and the uninjured ankles, more patients in the suture button group had adequate anatomic reduction at all time points than patients in the screw fixation group, and this became significant at 1 year (28% vs 58% of patients with malreduction, respectively; P= 0.008) and 2 years (20% vs 50%, respectively; P=0.009) postoperatively.
Seven patients in the screw fixation group were diagnosed with recurrent syndesmotic diastasis, 5 of whom underwent further surgery. No patients in the suture button group had recurrent diastasis (P=0.005). Of note, 6 patients in the suture button group (6%) had the suture button removed due to lateral-side discomfort. The number of patients with staged removal of the syndesmotic screw was not reported.
Anatomic restoration of the syndesmosis is paramount to good outcomes following ankle fracture fixation. Malreduction has been shown to cause increased pain and disability and subsequent development of arthritis.
The article by Anderson et al demonstrates better clinical and radiographic results in suture button fixation versus screw fixation of the syndesmosis. This phenomenon may be due to the slight plasticity of the suture button, which is more similar to the functionality of the ligament than a rigid screw. Furthermore, the suture button may allow the fibula to appropriately reduce into the incisura regardless of manual reduction maneuver in the operating room, thereby decreasing user error.
Interestingly, the authors found that the syndesmosis widened over time in the screw fixation group. This seemingly correlated with a plateau in patients’ functional scores, while the suture button group continued to improve clinically from 6 months to 1 year. The authors hypothesized that this could be due to staged implant removal and subsequent deformity, but they noted that it was difficult to define when exactly the diastasis occurred because CTs were not obtained until 1 year postoperatively.
One of the drawbacks of screw fixation of the syndesmosis is the frequent rate of hardware removal. Although prior studies have shown no difference in outcomes following screw removal, many surgeons prefer to remove the screw to prevent breakage or to facilitate range of motion of the syndesmosis. One of the advertised benefits of the suture button is that it does not require removal. This study had a 6% rate of hardware removal in the suture button group, however, which is consistent with other reports.
A limitation of this study is the outcome scores used, as no prior literature has reported the clinical significance of variations in AOFAS scores. The authors postulated that based on standard deviation, a difference in greater than 6 would be clinically significant, but this has not been corroborated by other groups. Therefore, although statistically significant differences were noted in this study, the differences may not be clinically relevant.
In conclusion, due to their slight plasticity, suture buttons likely provide a more anatomic reduction of the syndesmosis over time and yield better patient outcomes. Future studies should characterize whether the statistically significant outcomes reported here have clinical relevance. Given the cost differential of the implants and rates of subsequent hardware removal, a cost benefit analysis of usage may also be prudent.
Nicole Stevens, MD, is a resident in the Department of Orthopaedic Surgery and Philipp Leucht, MD, is an Assistant Professor of Orthopaedic Surgery and Cell Biology at NYU School of Medicine and NYU Langone Orthopedic Hospital, New York, New York.
The authors have no disclosures relevant to this article.