Introduction

Injury to the syndesmotic ligaments, including the anteroinferior tibiofibular ligament (AITFL), interosseous ligament (IOL), and posteroinferior tibiofibular ligament (PITFL), occurs in an estimated 18% of all ankle sprains [10, 25, 28]. Severe ankle sprains can combine lateral ligament injuries with syndesmotic injuries, affecting long-term recovery and longer return to sports compared to isolated lateral ankle ligament injuries [19, 35]. Critical to the successful clinical management of syndesmotic injuries, however, is not necessarily the identification of the injury, but rather the effective diagnosis of any resultant laxity of the distal tibiofibular articulation. If left unaddressed, syndesmotic instability can lead to severe residual pain and irreversible osteoarthritic changes [19]. While the diagnosis of syndesmotic laxity is relatively straightforward in the setting of frank distal tibiofibular diastasis, it becomes increasingly difficult in the setting of more subtle laxity [32].

Ankle arthroscopy currently represents the gold standard for diagnosing subtle syndesmotic laxity by virtue of its ability to directly visualize the distal tibiofibular articulation [6, 11, 20, 21, 31, 32]. A previous arthroscopic, cadaveric study by Lubberts et al. found that the syndesmosis is destabilized only after transection of all three syndesmotic ligaments (AITFL, IOL, and PITFL) [15]. On the other hand, partial injuries to the AITFL and IOL can still precipitate syndesmotic laxity if associated with an injury to the deltoid ligaments (DL), [23] presumptively due to the deltoid’s ability to tether the fibula via its attachment through the talus and lateral ankle ligaments. These studies, however, only focused the association between three syndesmosis ligaments and deltoid ligament injury, which in the so-called clinical practice corresponds to pronation-external rotation or pronation-abduction injury. The effect of lateral ankle ligament injury on the stability of syndesmosis has not been investigated.

While the presence of combined lateral ankle and syndesmotic ligamentous injuries is not uncommon, the contribution of the lateral ligaments toward stabilizing the syndesmosis remains unclear. Therefore, this study aim was to detect the impact of lateral ankle ligaments injury on syndesmotic laxity when evaluated arthroscopically. In addition, this result can be an important criterion in deciding how severe ankle ligament injury should be considered to fix syndesmosis as clinical relevance. The null hypothesis was that lateral ankle ligament injury does not affect the stability of syndesmosis.

Materials and methods

This study was approved by the Institutional Review Board of Massachusetts General Hospital (USA), Protocol number, 2016P001295.

Specimen preparation

Sixteen nonpaired fresh-frozen, above-knee amputated cadaveric specimens were used, that were purchased from MedCure.inc (RI, USA). The mean age at the time of death was 66 (31–90) years. Thirteen were males and three were females. Specimens were thawed at room temperature 24 h before the start of the experiment. Prior to the experiment, each ankle was evaluated by fluoroscopy (OrthoScan FD Pulse C-Arm, OrthoScan, Scottsdale, AZ). Specimens were excluded if there were any signs of ankle laxity, previous surgeries, fractures, ligamentous injuries, or pre-existing ankle osteoarthritis, and all specimens in this sample were ultimately included. Subsequently, specimens were secured to a wooden board using 5-mm Schanz pins inserted anteroposteriorly into the proximal, middle, and distal aspect of the tibia.

Sequential transection of ligaments

To randomize the intact cadaveric specimen into two groups with equal sample size, a block randomization method was used. None of the raters had been involved in the randomization process. In both groups, arthroscopic assessment (Arthrex, Naples, FL) of the distal tibiofibular joint was performed first with all syndesmotic and ankle ligaments intact and then after a progressive sequence of ligamentous transections. Group 1 underwent sequential transection of the lateral three ankle ligaments first to identify the effects of lateral ligament injury (1) anterior talofibular ligament (ATFL), (2) calcaneofibular ligament (CFL), (3) posterior talofibular ligament (PTFL), followed by the syndesmotic ligaments, (4) AITFL, (5) IOL, and (6) PITFL. Group 2 underwent sequential transection of the (1) AITFL, (2) ATFL, (3) CFL, (4) IOL, (5) PTFL, and (6) PITFL. (Fig. 1) The sequence of ligament transection in Group 2 was based on the most commonly understood progression of injury severity among lateral ankle ligament injuries with associated high ankle sprain [7, 27, 31]. At that time, the ligaments tend to be injured in sequence from the anterolateral to the posterolateral element in the varus ankle position.

Fig. 1
figure 1

The lateral ankle ligaments and the other ligaments that configure syndesmosis. ATFL anterior talofibular ligament, CFL calcaneofibular ligament, PTFL posterior talofibular ligament, AITFL anteroinferior tibiofibular ligament, IOL interosseous ligament, PITFL posteroinferior tibiofibular ligament

Full size image

Arthroscopic assessment

Arthroscopic evaluation of the distal tibiofibular joint was performed using a 30-degree 2.7-mm arthroscope through standard anterolateral and anteromedial portals. To test laxity of the distal tibiofibular joint at each sequence of ligament transection, stress was applied to the distal fibula using a bone hook. A small incision was made lateral to the fibula 5 cm proximal to the ankle joint, a bone hook was placed around the distal fibular diaphysis, and a 100-N standardized force was applied in the lateral direction to assess coronal and sagittal plane laxity. The 100-N force used in this study was based on a cadaveric study performed by Stoffel et al. who found that forces of more than 100-N did not show a substantial increase in syndesmotic displacement [30]. An electronic force gauge was used to standardize and measure the exact amount of applied force. Halo digital goniometer (Halo Medical Devices, Australia) was used to assure that the applied forces were applied parallel (coronal plane) and perpendicular (sagittal plane) to the ground surface [23].

Coronal plane tibiofibular diastasis was measured at the anterior third of the incisura immediately posterior to the anterior lip, as well as at the posterior third of the incisura immediately anterior to the posterior lip. The measurements were performed using precision-manufactured, spherical arthroscopic probes with diameter increments of 0.2 mm. Displacement in the sagittal plane was assessed by measuring translation of the fibula in both the anterior to posterior (AP) and posterior to anterior (PA) direction relative to the fixed tibia. To do so, a bony mark was made at the center of the intra-articular, medial fibula. An intra-articular probe with a scaled laser mark was used to measure the translation distance between the laser mark and the bony mark with application of a sagittal stress (Fig. 2). Arthroscopic images were captured during the experiment and the precise amount of sagittal translation was measured using imaging software (Image J; NIH 2019). During arthroscopic measurements, the ankle was manually kept in neutral position and was not towed due to a prior study, highlighting that the applied tension on the ankle ligaments, capsular elements, and surrounding soft tissues can mask syndesmotic laxity [16].

Fig. 2
figure 2

Arthroscopic assessment of the syndesmosis in the sagittal plane under a 100 N hook force directed from anterior to posterior (AP) (a, b) or from posterior to anterior (PA) (c, d). The force was measured and standardized using an electronic force gauge and digital goniometer. One-headed arrows represent the direction of force applied. Two-headed arrows represent the distances from the bony landmark at the fibula (Fib) relative to the laser mark at the arthroscopic probe (P) (mm)—before and after applied stress. Fib Fibula, Tib Tibia, P Probe

Full size image

Two orthopaedic fellowship-trained foot and ankle surgeons performed all arthroscopic probes measurements and sagittal translation measurements with image software independently in two randomly selected specimens to assess interobserver agreement using the intraclass correlation coefficient (ICC) through a two-way mixed-effects model with absolute agreement. The interobserver agreement for the arthroscopic coronal plane translation measurements at the anterior and the posterior third of syndesmosis was thus calculated based on 28 observations, while the interobserver agreement for the arthroscopic AP and PA sagittal plane translation measurements was calculated using 14 observations. Absolute agreement in an ICC assesses how much each measurement performed per observer differs from the other observer. Interpretation of the ICC values was carried out according to the guidelines proposed by Shrout as follows: 0.00–0.10, virtually none; 0.11–0.40, slight; 0.41–0.60, fair; 0.61–0.80, moderate; and 0.81–1.00, substantial [29].

Statistical analysis

Arthroscopic measurements were summarized using means and standard deviations (SD). In each group, Wilcoxon signed-rank test were used to determine differences between the intact stage and the six stages of ligamentous transection. A p value of < 0.05 was considered statistically significant.

With reference to a previous study by Bart Lubberts et al. as the primary outcome, [18] the fibula translational cut off was set to 2 mm for both anterior–posterior and posterior-anterior directions. Then, to achieve 80% statistical power with an overall two-tailed Type-1 rate of 2.5%, the moderate effect of ligamentous transection (partial eta-squared = 0.03), and met sphericity assumption (Nonsphericity correction E = 1) for one-way repeated analysis of variance. It was detected that eight specimens were needed in each group. The sample size calculation was performed using G*Power Version 3.1.9.4. All analyses were performed using Stata 14.2 (StataCorp LP, College Station, TX).

Results

In Group 1, compared to the intact ligamentous stage, the distal tibiofibular joint remained stable in both the coronal and sagittal plane after transection of all the lateral ankle ligaments (ATFL, CFL, and PTFL) and the AITFL. However, after additional transection of the IOL, the syndesmosis became unstable in both the coronal and sagittal planes. Coronal plane diastasis increased at both the anterior and posterior third of the incisura (p < 0.001 both), as well as with both AP and PA translation of the fibula in the sagittal plane (p < 0.001 both (Table 1). This laxity worsened further with subsequent transection of the PITFL.

Table 1 Summary of arthroscopic measurements in Group 1, the distal tibiofibular coronal plane joint space at the anterior and posterior third of syndesmosis, and sagittal plane fibular translation from anterior to posterior or posterior to anterior direction
Full size table

In Group 2, the syndesmosis remained stable in the coronal plane after transection of the AITFL, ATFL, and CFL. However, after additional transection of the IOL, the syndesmosis similarly became unstable in both the coronal plane sagittal planes. Coronal plane diastasis increased at both the anterior and posterior third of the incisura (p < 0.001 and p < 0.001, respectively), as well as with both AP and PA translation of the fibula in the sagittal plane (p = 0.008 and p = 0.028, respectively) (Table 2). This laxity worsened further with subsequent transection of the PITFL.

Table 2 Summary of arthroscopic measurements in Group 2, the distal tibiofibular coronal plane joint space at the anterior and posterior third of syndesmosis, and sagittal plane fibular translation from anterior to posterior or posterior to anterior direction
Full size table

The interobserver reliability for the arthroscopic coronal plane translation measurement at the anterior (0.99; 95% confidence interval [CI] 0.95–1.0) and the posterior third of syndesmosis (0.97; 95% CI 0.84–0.99) were substantial. The interobserver reliability for the AP and PA sagittal translation measurements was substantial, as well (0.86; 95% CI 0.79–0.92 and 0.85; 95% CI 0.73–0.92, respectively).

Discussion

This study highlights that, under arthroscopic assessment, the lateral ankle ligaments injury indirectly contributes to syndesmotic stability if the damage extends to the IOL injury. In other words, stable partial syndesmotic ligamentous injuries can be rendered unstable if the lateral ankle ligaments are also injured. As stated anatomically, a high ankle sprain with injury to the AITFL and IOL that would otherwise have been stable can be destabilized if the ATFL and CFL are also ruptured.

This result has taken a great step in prior studies, which have highlighted the association between lateral ankle sprains and high ankle sprains. The concomitant presence of both injuries did not hitherto delineate whether the lateral ankle ligaments actually contribute to syndesmotic stability, or whether both may simply be impacted by the same injury mechanism [7, 24, 27, 33].

The critical role of the syndesmosis toward maintaining the ankle mortise is empirically highlighted by numerous studies demonstrating the significant morbidity associated with leaving syndesmotic instability unaddressed [6, 26, 34]. In turn, the need for the talus to be bracketed by the fibula laterally and tibia medially intuits the need to preserve the distal tibiofibular relationship, historically understood to be maintained by the syndesmotic ligaments that directly bind the fibula to the tibia. In the same manner that walls of a house may be bound to each other not only by the roof but also through the foundation, the fibula and tibia are bound together not only through the syndesmotic ligaments above the ankle joint but also through their respective attachments to the talus below. The lateral ankle ligaments and the multiple components of the deltoid ligament therefore directly contribute toward maintaining the distal tibiofibular relationship by virtue of bridging their respective bony origins to the same bone—the talus. Thus, partial syndesmotic injuries to the AITFL and IOL that have been shown to be stable in isolation may not predicate that the intact PITFL alone can stabilize the syndesmosis, but rather that the residual attachment of the fibula to the tibia through the talus can continue to prevent abnormal motion at the distal tibiofibular articulation. In such scenarios, additional injury to the deltoid ligament has been shown to destabilize the syndesmosis [23]. This study highlights that the mirrored scenario is also true. Concomitant lateral ankle ligament injury can also destabilize otherwise stable syndesmotic injuries.

It is critical to differentiate injury patterns from mechanisms of injury. Calder et al. found that athletes with AITFL ruptures who also had a deltoid injury were 11-fold more likely to have an unstable syndesmosis than when injuring the AITFL alone, and that lateral ligament injury to the ATFL was actually protective, making it less likely that the syndesmosis was destabilized [5]. Part of this is structural, wherein the deltoid has been shown to contribute to syndesmotic stability as discussed above. But overwhelmingly this likely relates to mechanism of injury, wherein external rotation injuries are not only effective at injuring the syndesmosis, but also at tearing the multiple components of the deltoid ligament. Inversion injuries are highly effective at tearing the lateral ankle ligaments but much less effective at tearing the syndesmotic ligaments to the point of instability. On the other hand, this does not predicate that the lateral ankle ligaments do not contribute to syndesmotic stability, simply that the injury mechanism is less effective at destabilizing the distal tibiofibular articulation. When an inversion moment is severe enough, so that the AITFL and IOL are torn, alongside the ATFL and CFL, this study highlights that the distal tibiofibular joint is indeed rendered unstable.

From a practical point of view, how much fibular motion at the distal tibiofibular articulation should a clinician consider to be abnormal under arthroscopic stress examination? Prior studies have highlighted that approximately 3 mm of coronal plane diastasis and a total of 2 mm of total AP and PA sagittal plane motion can be considered reasonable thresholds for diagnosing syndesmotic laxity [12, 13, 17]. The practicality of these thresholds is heightened by the fact that while the precision tools used to conduct measurements in this study are not readily available in the operating room, 2 mm and 3 mm probes are standard to many instrumentation kits. The values measured in this study and displayed in Tables 1 and 2 corroborate these threshold values.

There are some limitations to this study. First, this study evaluated the translation of the fibula in the coronal and sagittal directions as the laxity of syndesmosis, but not in fibula rotation. Markus et al. demonstrated that an unstable AITFL should be repaired and augmented, as it represents an important stabilizer of external rotation of the distal fibula [22]. Given those ligament injuries of the ankle generally occur from the anterolateral side in clinical situation, fibula laxity in the external rotation can occur in the early stages of ligament injury, compared to the appearance of coronal and sagittal laxity. Second, the diagnosis of syndesmotic laxity was a statistical one, wherein laxity was judged to be present when distal tibiofibular motion after ligamentous transection was significantly greater than the intact state. It remains to be seen whether the threshold values discussed directly translate into clinical significance. Third, there may be variation in normal distal tibiofibular motion between individuals. Juan et al. demonstrated that measuring parameters indicating syndesmosis alignment should be compared to the contralateral ankle [14]. While arthroscopy allows direct evaluation of the distal tibiofibular relationship, it is invasive and does not allow a contralateral comparison in vivo. In a cadaveric model, one benefits from being able to examine the intact state and thereafter create the injury via controlled ligamentous transections. In the clinical setting, threshold values are useful, but a contralateral comparison arguably serves as a better internal control. Therefore, noninvasive, dynamic studies such as weight bearing CT with volumetric measurements of the distal tibiofibular space [1,2,3, 8, 9] or bedside ultrasound may better overcome some of these challenges [13]. Additional studies are necessary to better delineate their role.

Conclusions

The findings from the present study underscore that syndesmosis is not directly destabilized by injury to the lateral ankle ligaments. However, when combined with IOL injuries, syndesmosis can be destabilized. As a clinical relevance, clinicians must be aware the combination of IOL injury is the key to make clinicians consider the fixation of syndesmosis.