➢ Anatomic restoration of fibular length, alignment, and rotation is essential for accurate syndesmotic reduction.
➢ Accuracy of syndesmotic reduction can be improved by reduction and fixation of the posterior malleolus and open reduction of the syndesmosis.
➢ Postoperative or intraoperative computed tomography (CT) is the most reliable method of assessing the accuracy of syndesmotic reduction.
➢ There is no consensus regarding many aspects of syndesmotic stabilization, including the method of reduction, the device used for stabilization (metal or titanium screws, suture-button device, or bioabsorbable implant), number of screws, number of cortices engaged, and retention or removal of screws.
➢ Syndesmotic malreduction is common and if left uncorrected is associated with inferior functional outcomes when compared with those of an anatomic syndesmotic reduction.
The distal aspects of the tibia and fibula as well as the associated stabilizing ligamentous structures are collectively known as the ankle syndesmosis. The syndesmosis functions to maintain normal ankle kinematics and provide talar stability through a physiologic range of motion. Disruption of the ligamentous structures can occur in isolation or in the setting of an ankle fracture. The need for recognition and stabilization of syndesmotic instability following ankle trauma is well established. Recently, investigators have reevaluated best practices with regard to the assessment, reduction, and fixation of syndesmotic injuries. The effects of syndesmotic malreduction and failed stabilization have also been examined to a greater degree, resulting in new literature on this topic. The purpose of this article is to present the current evidence on syndesmotic injuries in the setting of an ankle fracture.
The ankle syndesmosis comprises four ligamentous structures between the distal aspects of the tibia and fibula. The anterior inferior tibiofibular ligament originates from the anterolateral (Chaput) tubercle of the distal part of the tibia and inserts into the anterior (Wagstaffe) tubercle of the distal part of the fibula. The interosseous tibiofibular ligament is continuous with the interosseous membrane between the tibia and fibula and runs from medial to lateral in a distal and anterior direction. The ligament terminates approximately 1 cm above the ankle joint. The posterior inferior tibiofibular ligament originates on the posterior (Volkmann) malleolus and runs transversely to insert on the posterior tubercle of the fibula. The transverse tibiofibular ligament, also known as the posterior oblique ligament, has been described both as its own distinct entity and as a deep component of the posterior inferior tibiofibular ligament and runs in the same horizontal orientation1.
The osseous relationship of the distal parts of the tibia and fibula at the incisura fibularis tibiae is inconsistent2,3. The concavity and depth of the incisura tibialis are variable. Ebraheim et al. described two groups, those with a concave surface and those with a shallow concave surface4. This variation may have clinical implications with regard to the ability to obtain, assess, and maintain reduction of the syndesmosis. Individual syndesmotic anatomic variation is best observed with use of computed tomography (CT)5,6.
There have been multiple biomechanical studies of the contributions of the syndesmotic ligaments. The deltoid ligament on the medial side of the ankle is a primary stabilizer of the ankle while the syndesmotic ligaments are considered secondary stabilizers. Cadaveric studies have shown that, in the setting of an intact deltoid ligament, complete sectioning of the syndesmotic ligaments results in only minimal syndesmotic widening7. Marked instability was noted following deltoid transection. The key element in mechanical instability of the ankle is therefore deltoid incompetence in the setting of a syndesmotic injury. Ogilvie-Harris et al. found that, when the syndesmotic ligaments were sequentially cut, the anterior inferior tibiofibular ligament and transverse tibiofibular ligament contributed the greatest amount of resistance (35% and 33%, respectively) to lateral displacement of the fibula8. Incremental sectioning of syndesmotic ligaments results in increasing instability9.
Mechanism of Injury
The primary mechanism of injury to the syndesmosis is believed to be an external rotation force to the foot. The talus abducts and externally rotates, placing stress onto the anterior inferior tibiofibular ligament. A purely ligamentous injury can occur if the ligamentous structures fail before a malleolar fracture occurs.
The location of a fibular fracture can indicate syndesmotic competence. A proximal fibular fracture associated with a complete syndesmotic disruption and injury to the deep deltoid ligament is known as a Maisonneuve injury10. Not all proximal fibular fractures result in ankle instability. If the syndesmotic injury is incomplete or the deep deltoid ligament remains intact, a proximal fibular fracture following a rotational injury will not result in ankle instability. Direct trauma to the proximal part of the fibula can also result in a proximal fibular fracture without syndesmotic disruption.
Fibular fractures below the level of the tibial plafond (Danis-Weber type A) are rarely associated with a disruption of the deltoid complex or syndesmotic instability11,12. These are commonly classified as supination-adduction injuries according to the Lauge-Hansen system13,14. Fibular fractures that originate at the level of the ankle joint (Danis-Weber type B) may or may not be associated with disruption of the syndesmosis. When a patient presents with this type of injury, preoperative or intraoperative stress examination is required to determine syndesmotic stability. Fractures of the fibula above the level of the plafond (Danis-Weber type C) are the result of pronation-external rotation injuries and, compared with more distal fractures, are associated with a higher prevalence of syndesmotic disruption. As noted, instability is dictated by the competence of the deltoid, the injury mechanism, and the degree of syndesmotic injury.
Anteroposterior and lateral radiographs of the entire tibia and fibula and ankle should be obtained. An internal rotation mortise view of the ankle should also be made. If a posterior malleolus fracture is identified or suspected, and further information or detail regarding the fracture is needed, a CT scan can be considered.
Multiple radiographic findings have been described to identify a syndesmotic injury. Tibiofibular clear space is measured 1 cm above the plafond and is the distance between the lateral border of the posterior aspect of the tibia and the medial border of the fibula. This distance in a normal ankle should be <6 mm on both the anteroposterior and the mortise radiograph of the ankle15,16. Tibiofibular overlap is also measured 1 cm proximal to the plafond and is defined as the overlap of the lateral malleolus and the anterior tibial tubercle. It should measure >6 mm on the anteroposterior radiograph and >1 mm on the mortise radiograph of the ankle15,16. Medial clear space is defined as the distance between the medial border of the talus and the lateral border of the medial malleolus. This distance should be equal to or less than the superior clear space between the talus and plafond15-17. Any measured value outside of these normal radiographic parameters may indicate syndesmotic injury. In a patient with an intact medial malleolus, deltoid incompetence will result in an increased medial clear space18.
Syndesmotic disruption cannot be reliably identified with static radiographs and is poorly predicted by practitioners on the basis of fracture pattern19. Stress external rotation radiographs are needed to assess syndesmotic injury and are performed by obtaining a mortise radiograph of the ankle, stabilizing the tibia, and gently externally rotating a neutral plantigrade foot. Lateral displacement of the fibula indicates a positive finding. A lateral radiograph obtained during this examination will demonstrate posterior fibular displacement in the setting of a syndesmotic injury9,20.
Intraoperative Evaluation of Syndesmotic Stability
Two examinations are commonly used for intraoperative evaluation of the syndesmosis following fracture fixation. The Cotton test21 and the external rotation stress examination22 are performed following stable anatomic fracture fixation (Figs. 1 and 2).
Stoffel and colleagues performed a cadaveric study to evaluate these two examinations and found that the modified lateral stress (Cotton) test with examination of the tibiofibular clear space more clearly identified syndesmotic injury than the external rotation stress test in specimens with a simulated Weber type-C injury. Medial clear space widening was seen both in specimens with an isolated deltoid disruption as well as in specimens with an isolated disruption of the anterior inferior tibiofibular ligament23. The authors of a cadaveric study of simulated supination-external rotation injuries concluded that the lateral stress test with examination of the tibiofibular clear space should be the preferred examination of syndesmotic injury; they found increased medial clear space following an external rotation stress test in the setting of an intact syndesmosis with deltoid deficiency24.
Pakarinen et al. conducted a prospective clinical study to assess both examinations in a series of 140 patients with an unstable supination-external rotation ankle fracture25. They found that, although the intraobserver reliability and specificity of both examinations were excellent, the sensitivity of the Cotton test (0.25) and the external rotation stress test (0.58) was poor. Interestingly, the sensitivity did not improve when both tests were used.
A systematic review was recently conducted to evaluate the available literature on the diagnosis of syndesmotic injuries. Eight clinical tests were investigated. Three papers were analyzed, and evidence did not show superiority of a single diagnostic examination. The authors concluded that practitioners cannot rely on a single examination to determine syndesmotic instability and that additional modalities, such as magnetic resonance imaging (MRI), should be used26.
Intraoperative Syndesmotic Reduction
Anatomic syndesmotic reduction during ankle fracture fixation is difficult and begins with restoration of fibular length, alignment, and rotation. Normal radiographic relationships between the fibula and distal part of the tibia, the fibula and talus27, as well as the fibula alone28 can be used to avoid malreduction. Unfortunately, radiographs and fluoroscopy alone cannot reliably demonstrate syndesmotic malreduction that occurs in isolation or with malleolar malreduction29. Evaluation of syndesmotic reduction by CT following stabilization has shown that the malreduction rate may be >50%30.
Clamp reduction of the syndesmosis is frequently performed. Iatrogenic syndesmotic malreduction can occur as a result of clamp or screw placement31. Phisitkul et al. described clamp placement (in a neutral position across the syndesmosis, along the lateral malleolar ridge and the center of the anteroposterior width of the tibia medially, 1 cm above the joint line) to accurately reduce the syndesmosis in a cadaveric model32.
Open reduction of the syndesmosis has been described. Miller et al. reported a malreduction in twenty-four (16%) of 149 ankles following open reduction, debridement, and postoperative CT evaluation of the syndesmosis33. This was an improvement compared with a malreduction rate of 52% in a control group of twenty-five ankles that underwent postoperative CT evaluation after intraoperative fluoroscopy alone had been utilized for reduction33. Improved syndesmotic reduction with direct visualization of the joint has led some authors to advocate the routine use of open reduction3,34.
Intraoperative imaging can be used to assess syndesmotic reduction. Radiographic relationships, as described above, should be routinely used to assess length, alignment, and rotation of the fibula within the tibial incisura. Grenier et al. proposed use of the lateral fluoroscopic view and the anteroposterior tibiofibular (APTF) ratio as a radiographic measurement to help guide syndesmosis reduction35. Intraoperative CT has been described to assess the syndesmosis in an effort to eliminate postoperative identification of syndesmotic malreduction29,36. Franke et al. reported the use of intraoperative three-dimensional (3D) imaging to assess syndesmotic reduction in 251 consecutive cases29. Imaging resulted in a surgical change of the syndesmotic reduction or the implant position in eighty-two (32.7%) of the cases. The authors concluded that surgical treatment of any syndesmotic injury should include intraoperative 3D imaging or a postoperative CT scan. The necessity, cost effectiveness, and role of advanced imaging techniques, such as intraoperative or postoperative CT, in the treatment of syndesmotic injuries has not been well studied and has yet to be established.
Surgical stabilization of syndesmotic injuries is technically challenging, with numerous potential pitfalls. Anatomic fibular reduction is a prerequisite for accurate reduction and stabilization of the syndesmosis7,37. There are numerous controversies regarding syndesmotic fixation. Debate continues with respect to the number, size, location, and material of screws as well as the number of cortices that should be engaged. In addition, some authors advocate routine screw or implant removal while others do not.
Number of Screws
Transsyndesmotic stabilization can be performed with one or more screws. The manual of the Arbeitsgemeinschaft für Osteosynthesefragen (AO) group described placement of a single screw between the fibula and tibia for syndesmotic stabilization38 (Fig. 3). Others have described the use of multiple screws, particularly for Maisonneuve-type injuries (Fig. 4)39. A recent study evaluating the functional outcomes of patients with a Maisonneuve-type injury demonstrated no differences regardless of whether stabilization had been performed with one or two screws39. Another study evaluating cases of failed syndesmotic stabilization showed no association with the number of screws used40. One randomized controlled trial compared the use of a single 4.5-mm quadricortical screw with the use of two 3.5-mm tricortical screws; higher functional scores and less pain were found with two screws at three months, but these differences disappeared by one year41.
Size of Screws
The core diameter of a screw determines its bending strength, which varies according to the radius to the fourth power42. However, syndesmotic screws are subject to stresses besides bending alone. Biomechanical studies comparing standard 3.5-mm and 4.5-mm screws have produced conflicting results43,44. The use of 3.2-mm locking screws may provide additional resistance to torque45.
In a review of syndesmotic fixation, 4.5-mm screws were used in twelve of 137 patients46. No loss of reduction was reported in any of those twelve patients, whereas eight patients treated with 3.5 or 4.0-mm screws had loss of reduction. Although that difference was not significant, the authors recommended using 4.5-mm screws. As noted, no functional difference was identified in the randomized controlled trial that compared a single 4.5-mm quadricortical screw with two 3.5-mm tricortical screws41.
Internal fixation commonly is performed with stainless-steel or titanium screws. A biomechanical study comparing stainless-steel with titanium 3.5 mm screws in a simulated Maisonneuve injury showed no failures with either type of screw under simulated partial weight-bearing47.
Some authors have proposed using a bioabsorbable screw for syndesmotic fixation to obviate the need for screw removal48-51. Polylevolactic acid (PLLA) is one such material. A single 5-mm PLLA screw has been reported to be as strong as a single 5-mm stainless-steel screw50. A study of twenty-three patients who had syndesmotic stabilization with use of a single quadricortical 4.5-mm PLLA screw demonstrated no loss of reduction and no screw-related complications at two years48. Two randomized controlled trials demonstrated no difference in the range of motion or complications when a PLLA screw was compared with a 4.5-mm quadricortical screw48,50. Interestingly, some patients treated with a PLLA screw still required removal of symptomatic nonabsorbed fragments50.
Alternatives to Transsyndesmotic Screws
Suture-button devices have been utilized for syndesmotic stabilization. Several limited studies on this technique have been reported52-54. In a recent prospective cohort, a suture-button construct was compared with a single 3.5-mm or 4.5-mm quadricortical screw for syndesmotic fixation; there was no difference in American Orthopaedic Foot & Ankle Society (AOFAS)55 or Foot and Ankle Disability Index (FADI)56 scores, although there was a higher prevalence of syndesmotic malreduction in the screw group57. Although there is some literature on the technique, further investigation is needed.
Posterior Malleolar Fixation
In ankle fractures with a syndesmotic injury and associated posterior malleolar fragment, the posterior inferior tibiofibular ligament typically remains intact. Anatomic reduction of the posterior malleolus therefore reestablishes the normal osseous anatomy of the tibial incisura and restores the posterior inferior tibiofibular ligament. Reduction and fixation of the posterior malleolar fracture has been reported to contribute greater stiffness than fixation with a syndesmotic screw58. Displaced posterior malleolar and fibular fractures can be addressed through a posterolateral approach59.
Selecting a Construct
Most authors have advocated placing syndesmotic screws as position screws to maintain syndesmotic relationships without syndesmotic overcompression. A cadaveric study by Tornetta et al. demonstrated that placement of a lag screw did not result in diminished ankle dorsiflexion60. Another cadaveric study demonstrated that use of 3.5-mm or 4.5-mm lag screws increased maintenance of syndesmotic reduction compared with that seen with a tricortical 3.5-mm position screw after clamp reduction61. We are not aware of any clinical data comparing position with lag screws.
Number of Cortices Engaged
Syndesmotic screws may be placed in tricortical or quadricortical fashion depending on engagement of the medial tibial cortex. The AO manual describes a quadricortical 4.5-mm cortical screw for syndesmotic fixation38. Other authors have recommended tricortical screws62. One theorized advantage of a quadricortical screw is the ability to remove the screw from the medial tibial cortex if the screw breaks. A biomechanical study comparing a tricortical 3.5-mm screw with a quadricortical 3.5-mm screw demonstrated no failures with either construct under simulated partial weight-bearing47.
Long-term outcomes are similar with both constructs. One randomized controlled trial of 3.5-mm screws placed in a tricortical or quadricortical fashion showed no difference in the techniques with regard to the rates of loss of reduction, screw breakage, or need for implant removal63. Another randomized controlled trial comparing a single 4.5-mm quadricortical screw (with routine removal) with two 3.5-mm tricortical screws demonstrated significantly higher functional scores (p = 0.025) and less pain (p = 0.017) in the tricortical screw group at three months; however, these differences disappeared at one year with no difference in range of motion41.
Eccentric transsyndesmotic screw placement can result in malreduction31,64. A screw should ideally be directed along the transmalleolar axis. However, the optimal screw location relative to the plafond is less clear. It has been theorized that a suprasyndesmotic screw position permits overtightening of the syndesmosis or causes fibular deformity. More distal screw placement necessitates screw passage through cartilage at the incisura. Biomechanical studies have demonstrated conflicting results65,66. The authors of one study reported less syndesmotic widening with a screw at 2 cm than was seen when the screw was 3.5 cm above the plafond65; another demonstrated increased fixation strength and less widening with fixation at 5 cm compared with when it was 2 cm above the plafond66. Finite element models have shown that 3.5-mm and 4.5-mm screws experience the lowest stress between 30 and 40 mm above the plafond whereas the least widening occurs with placement at 20 to 30 mm above the plafond67.
Available clinical data are similarly contradictory. In a study comparing transsyndesmotic fixation with suprasyndesmotic fixation in patients with a Weber type-C ankle fracture, there were no differences between groups with respect to pain, restriction of range of motion, talar shift or tilt, osteoarthritis, activity restriction, or overall satisfaction68. In another series, of 102 patients with a syndesmotic injury, placement of the syndesmotic screw at 41 to 60 mm above the plafond was associated with worse pain (as reported on a visual analog scale [VAS]) than placement at 21 to 40.99 mm above the plafond (7.5 versus 8.5 points, p = 0.024)69.
There is no clear evidence or evidence-based recommendations regarding the best postoperative care after treatment of syndesmotic injuries. Various postoperative protocols, representing surgeon preference, patient characteristics, and injury severity, have been reported. A recent survey of orthopaedists revealed that most surgeons did not allow immediate weight-bearing following acute fixation of syndesmotic injuries70. In a systematic review comparing early-motion protocols with immobilization for a minimum of six weeks, the authors found no difference in functional outcomes between the two groups but acknowledged the lack of evidence on the topic71. Finally, some authors have supported screw removal before weight-bearing is initiated46,72.
Screw Removal and Duration of Fixation
There is no consensus regarding the necessity or timing of syndesmotic screw removal. Some advocate routine syndesmotic screw removal between six weeks and four months postoperatively72-74. Screw removal is theorized to reduce discomfort, allow physiologic motion at the distal tibiofibular joint, and prevent screw breakage. Potential loss of reduction as well as complications and expense related to a second surgical procedure are cited as reasons for screw retention. Although there may be some loss of tibiofibular reduction with removal75, the talar position and medial clear space frequently remain relatively unchanged75,76.
Clinical evidence does not support either screw removal or retention. Although few investigators have specifically analyzed screw removal or retention, improvement in clinical outcome scores after syndesmotic screw removal74 or in those who have had broken screws76,77 has been reported. Manjoo et al. compared functional outcomes among patients who had retained, broken, loose, or removed syndesmotic screws76. The Lower Extremity Measure78 and Olerud-Molander79 ankle score were significantly higher in those with broken, loose, or removed screws compared with those with retained screws. Hamid et al. reported significantly higher outcome scores in patients with broken syndesmotic screws, whereas they found no difference between those with intact and those with removed screws77. In the only randomized controlled trial on the topic of which we are aware, the authors compared routine removal of a 4.5-mm quadricortical screw with retention of a tricortical 3.5-mm screw in a series of sixty-four patients41. Those with a retained 3.5-mm tricortical screw had superior three-month outcomes and pain scores but no difference at one year. However, the groups differed with respect to the size and number of screws and the engaged cortices.
In other series, screw removal has not been found to be beneficial72,73,80. One study showed a trend toward improved Olerud-Molander79 ankle scores in those with retained screws, although the authors did not report details of their fixation72. In another series, no difference in the Baird and Jackson81 ankle score or complication rate was found between those with retained screws and those with removed screws73. Egol et al. reported no difference in pain, range of motion, or function between patients with retained syndesmotic screws and those with broken or removed syndesmotic screws80.
Failed Syndesmotic Fixation
Syndesmotic malreduction is associated with worse functional outcomes34. If syndesmotic or fibular malreduction is recognized, or the fixation fails, revision fixation should be considered. Revision surgery should be carried out through an open approach to directly visualize the fracture or syndesmosis before revision fixation is performed.
Chronic instability or delayed presentation of syndesmotic injuries presents unique challenges. Authors of various case reports and case series have proposed internal fixation82,83, ligamentous reconstruction84,85, advancement of the anterior inferior tibiofibular ligament through a tibial bone block86, and distal tibiofibular arthrodesis87. Olson et al. reviewed the results of salvage of failed syndesmotic treatment with distal tibiofibular arthrodesis in ten patients and found a significant improvement in AOFAS hindfoot scores and restoration of talocrural relationships87.
The long-term clinical and radiographic outcomes following syndesmotic fixation are difficult to generalize. A wide spectrum of injuries involve syndesmotic disruption. In a recent report39, >90% of patients with a pronation external rotation injury (predominantly Maisonneuve-type) treated with syndesmotic screws alone had good-to-excellent AOFAS and Foot and Ankle Ability Measure (FAAM)56 functional scores at more than twenty years. Egol et al. compared one-year functional outcome data between patients with an isolated malleolar fracture and those with an associated syndesmotic disruption. At one year, both the Short Musculoskeletal Function Assessment (SMFA)88 dysfunction index and AOFAS ankle-hindfoot scores were worse in the patients with an associated syndesmotic injury80. A more severe injury may be associated with a worse functional outcome but, more importantly, anatomic fracture reduction has been shown to improve clinical outcome and aid proper reduction of the syndesmosis37.
Syndesmotic malreduction has been correlated with poor functional outcomes. Weening and Bhandari found that proper syndesmotic reduction (as seen on radiographic examination alone) predicted improved SMFA (functional) and Olerud-Molander scores; thirty patients (59%) had removal of syndesmotic fixation89. Sagi et al. reported inferior SMFA (functional) and Olerud-Molander scores in patients with a syndesmotic malreduction (as seen on CT examination) as compared with those with an anatomic reduction; functional assessment was performed at two years, and all screws were removed at four to six months34.
There is limited evidence to support the belief that screw removal reverses the functional consequences of malreduction. Song et al. reported that eight of nine patients with syndesmotic malreduction experienced reduction of the syndesmosis following removal of transsyndesmotic screws at three months90. The authors did not report pain or functional scores in their study but hypothesized that improved syndesmotic reduction may improve overall clinical outcome.
The distal tibiofibular syndesmosis of the ankle can be disrupted in the setting of an ankle fracture. The orthopaedic surgeon must understand the anatomy, diagnosis, radiographic evaluation, and treatment of syndesmotic instability. Anatomic malleolar reduction is essential in the setting of an ankle fracture. Iatrogenic intraoperative syndesmotic malreduction can occur following application of a clamp or stabilization device. Syndesmotic reduction is improved following formal open reduction; utilization of recently described radiographic relationships or advanced imaging modalities can also assist in accurate intraoperative reduction. There is no consensus on the best method of syndesmotic reduction or mode of fixation or whether the fixation should be removed or left in place. A postoperative CT scan is a sensitive tool for assessing reduction, but its use should be left to the surgeon’s discretion. Available evidence suggests that nonanatomic syndesmotic reduction is associated with poor long-term functional results, regardless of management. In conclusion, there is a lack of consensus in many areas of syndesmosis management and this topic must continue to be an area of study (Table I).
Source of Funding: No external financial support was provided for the creation of this manuscript.
Investigation performed at the University of California, Irvine, Orange, California; University of Southern California, Los Angeles, California; and University of Washington-Harborview Medical Center, Seattle, Washington
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 © 2014 by The Journal of Bone and Joint Surgery, Incorporated