Outcomes of suture button fixation versus embrace fixation for syndesmotic injury

Abstract

Background

This study aimed to compare the clinical and radiological outcomes of the embrace fixation (EF) technique with those of suture button fixation (SF) for syndesmotic injuries in ankle fractures, with a focus on postoperative CT-assessed reduction quality, functional scores, and complication rates.

Methods

A retrospective cohort study was conducted on 148 patients (86 EF and 62 SF) treated between January 2021 and January 2023. Preoperative and postoperative assessments included radiographies and 3D-CT scans, American Orthopedic Foot & Ankle Society (AOFAS) hindfoot scores, Olerud-Molander Ankle (OMA) scores, and visual analog scale (VAS) pain scores. The primary outcome was the AOFAS hindfoot score at the final follow-up.

Results

Both groups demonstrated excellent clinical outcomes, with no significant differences in AOFAS, OMA, or VAS scores. Postoperative CT analysis revealed low malreduction rates. The complication rates were comparable and primarily involved minor soft tissue irritation.

Conclusion

Both the embrace and suture button techniques demonstrated comparable accuracy and effectiveness for syndesmotic stabilization in ankle fractures and yielded excellent radiological and clinical outcomes. Notably, the embrace technique yielded equivalent outcomes compared with suture button fixation for managing syndesmotic injuries.

Level of evidence

Level III, retrospective case-control study, therapeutic.

Introduction

Syndesmotic complex rupture with subsequent tibiofibular instability represents a characteristic finding in rotational ankle fractures, occurring in approximately 13–20% of all ankle injuries [1–4].This critical joint disruption requires simultaneous surgical stabilization along with fracture fixation, as misdiagnosis or inadequate treatment may result in chronic pain, functional limitations, and accelerated osteoarthritis development [5–7].

Current fixation approaches for syndesmotic injuries have varying efficacy profiles [8–11]. Although screw fixation is the most popular intervention, it carries substantial risks, including hardware failure (loosening/breakage) and soft tissue irritation [12]. Postoperative CT evaluations reveal particularly concerning malreduction rates ranging from 20.8% to 52% with this method [13].

Dynamic stabilization alternatives such as suture button constructs have emerged to address these limitations, theoretically permitting more physiological joint motion while maintaining anatomical reduction [14–16]. Nevertheless, clinical data indicate persistent malreduction rates of approximately 17% with these devices [17]. The recently developed embrace technique represents a paradigm shift, achieving postoperative CT-assessed malreduction rates of merely 2.99% through its innovative stabilization mechanism [18]. Despite this improvement, no studies have compared the embrace technique with suture button fixation, To date, no comparative studies have directly assessed the performance of dynamic fixation methods for syndesmotic injuries.

This is the first study aimed at clinically comparing suture button fixation versus embrace fixation for syndesmotic diastasis, evaluating both clinical and radiological outcomes, with a particular emphasis on postoperative CT measurements and complication profiles. Our hypothesis posited that suture button fixation would demonstrate comparable postoperative outcomes to embrace fixation in managing syndesmotic diastasis.

Methods

Study group selection

A retrospective cohort study was conducted through medical record review and radiographic database analysis of 148 patients with ankle fractures complicated by syndesmotic diastasis. We performed surgical interventions on all patients at our institution between January 2021 and January 2023. Approval (QYFYWZLL29985) was obtained from the institutional review board. The detailed inclusion/exclusion criteria are presented in Fig. 1.

Fig. 1.

Flow chart of study group selection by means of the inclusion and exclusion criteria

The participants chose to enroll after receiving a detailed explanation of both fixation modalities. Patients were presented with a detailed explanation of both fixation modalities, including potential benefits and risks. The sample size was determined based on the actual number of patients we had. Ultimately, 86 patients were included in the embrace fixation (EF) group, and 62 patients were included in the suture button fixation (SF) group. Patients provided written consent prior to the surgeries.

Preoperative assessment

Preoperative radiographic imaging included standard anteroposterior (AP) and lateral view plain films along with three-dimensional computed tomography (3D-CT) scans of the bilateral ankle joints. The syndesmosis width was quantitatively measured at approximately 1 cm proximal to the tibial lateral plafond using standardized radiographic landmarks. Syndesmotic diastasis was defined according to established diagnostic criteria as a ≥ 2 mm increase in syndesmotic interval width compared with the contralateral uninjured ankle.

Operative technique

Embrace fixation (EF group)

All patients underwent ankle fracture and syndesmotic fixation under general or regional anesthesia with tourniquet control. Patients were placed in the supine position. Lateral malleolar fractures were anatomically reduced and stabilized by locking plates. Medial malleolar fractures were fixed with two cannulated screws, and deltoid ligament injuries were repaired using suture anchors (Smith & Nephew, Mansfield, OH, USA). Thin posterior malleolar fragments may not require fixation, whereas large triangular fragments should be fixed. Haraguchi type I/II fragments require fixation, but type III does not. For fixation, use a posterolateral approach.Then, syndesmotic stability was evaluated via the Cotton test or external rotation at the surgeon’s discretion.

Posterior malleolar fractures involving more than 15 ~ 20% of the tibial plafond articular surface were reduced and fixed with cannulated screws. Anterior bone tunnels were drilled at the anterolateral edge of the tibial syndesmotic notch, 2.0 cm above the plafond (Fig. 2a and b). A second posterior tunnel was created at the posterolateral edge of the syndesmotic notch. The fiber wires were passed medially through the anterior tunnel and subcutaneously to the posterior tunnel. Manual syndesmosis reduction was performed. Fluoroscopic assessment of reduction was performed using the medial clear space and tibiofibular overlap as indicators of appropriate reduction. The wire was tied over the fibula or plate using NICE knots (Fig. 2c–f).

Fig. 2.

Embrace fixation technique

Suture button fixation (SF group)

The patient position and surgical incisions mirrored those used in the EF group. Following anatomical fracture reduction and fixation, stabilization of the syndesmosis was achieved using a dynamic TightRope system (Arthrex, Naples, FL). Surgeons are recommended to fix the syndesmosis at a level just proximal to the inferior tibiofibular joint [19]. A 3.5-mm osseous tunnel was created at the same height from the tibial platform as in the EF group. A guide needle with pull-through sutures was advanced through the tunnel from the lateral aspect. The oblong button was flipped to rest on the medial cortex of the tibia. The sutures were tightened until the lateral button fit firmly on the cortex or on the plate when present. Fluoroscopic assessment of reduction was performed using the medial clear space and tibiofibular overlap as indicators of appropriate reduction. Three half-hitches secured the fixation. Postoperative management included immobilization in a cast for a 15-day period to optimize soft tissue recovery.

Postoperative rehabilitation

All patients followed the same protocol postoperatively. Cotton gauze dressings secured with elastic bandages were applied, followed by a standardized protocol for anti-inflammatory therapy, edema control, and analgesia. An early-phase rehabilitation protocol involving active ankle range-of-motion exercises was initiated at 2 weeks postsurgery. Progressive resistance training was incorporated as tolerated. Protected weight-bearing commenced at 6 weeks postinjury, with gradual advancement to full weight-bearing ambulation under clinical guidance.

Assessment

The primary outcome was the American Orthopedic Foot & Ankle Society (AOFAS) hindfoot scale at the last follow-up. The AOFAS scale comprises three domains—pain, function, and alignment—integrating both subjective and objective parameters into a 0–100 scoring system (100 indicates optimal outcomes). The secondary outcome measures included the Olerud–Molander Ankle (OMA) scale [20], and VAS for pain during daily activities. The OMA scale is a self-administered functional assessment tool ranging from 0 to 100 points, with 100 representing full functional recovery. The OMA scale was validated by correlating patient-reported linear analog scales with loaded ankle range-of-motion measurements and radiographic evaluations for osteoarthritis or dislocation in postsurgical ankle fracture patients. Moreover, the results of types B and C of the DW classification were compared. Complications were systematically documented, and follow-up assessments were performed by two study-participating orthopedic surgeons.

Radiologic outcomes were evaluated via anterior‒posterior and lateral radiographs of the injured ankle acquired immediately postsurgery, followed by weight-bearing radiographs at 3 months, 12 months, and the final follow-up. 3D-CT scans of both ankles were performed within 2 weeks postsurgery and at 12 months.

All syndesmosis parameters were analyzed on axial CT slices parallel to the talotibial joint, positioned 1 cm proximal to the midpoint of the tibial plafond [21, 22]. Two junior physicians independently conducted the blinded measurements, with the mean value recorded as the final result. Standardized landmarks were utilized for quantification: the tibiofibular distance at two predefined anatomical points (Fig. 3a and b), anterior displacement (Fig. 3c), and fibular rotation (Fig. 3d). The syndesmosis width discrepancy between the injured and contralateral sides was calculated. Malreduction was defined as a ≥ 2 mm difference in syndesmosis width, which is consistent with prior criteria [18]. This CT-based measurement protocol demonstrated high reproducibility in earlier studies, as supported by intraclass correlation coefficients [23].

Fig. 3.

Parameters on CT: A anterior difference (AD); B posterior difference (PD); C fibular translation (FT); D fibular rotation (FR)

Additionally, postoperative radiographic parameters were analyzed by an orthopedic foot and ankle fellow. Each parameter was measured in triplicate, with the mean value documented for analysis. In assessing postoperative complications, we will stratify them into minor and major categories according to the requirement for operative reintervention.

Statistical analysis

The Kolmogorov‒Smirnov (K-S) test was applied to evaluate data normality. Normally distributed measurement data are expressed as the means ± standard deviations. Independent samples t tests were utilized to compare CT parameter changes (injured side: preoperative vs. postoperative; normal side) between groups. Categorical variables (sex, side, DW class) were analyzed using chi-square tests. Group comparisons of AOFAS-hindfoot scores, OMA scores, and VAS pain scores were performed through independent samples t tests. To account for potential confounding effects, linear regression analysis was performed to adjust for covariates including age, side, time, sex, DW class, PD, FT, FR, and AD. Additionally, analysis of variance (ANOVA) with time as a covariate was conducted to eliminate the potential influence of different follow-up periods on patient outcomes. Statistical significance was defined as P < 0.05. All the statistical analyses were performed using SPSS (version 26.0; Armonk, NY, USA).

Results

A total of 148 patients were enrolled in the study, with the EF group comprising 86 patients and the SF group consisting of 62 patients. The baseline demographic and clinical characteristics are comprehensively summarized in Table 1.

Table 1.

Patient demographics

Variables	EF, n (%)	SF, n (%)	P, n (%)
Age (y), mean ± SD	31.15 ± 10.35	32.35 ± 8.90	0.461
Sex			0.099
Male	52 (60.47)	29 (46.77)
Female	34 (39.53)	33 (53.23)
Side			0.975
Left	46 (53.49)	33 (53.23)
Right	40 (46.51)	29 (46.77)
DWC			0.71
B	28 (32.56)	22 (35.48)
C	58 (67.44)	40 (64.52)
Follow-up (mo), mean ± SD	29.45 ± 4.55	30.03 ± 5.77	0.497

DWC, Dannis-Webber classification

Clinical outcomes

We planned to recruit 180 participants (90 per group) to account for potential attrition.

However, during the study period, 32 participants were lost due to voluntary withdrawal (n = 16), missing experimental data (n = 9), and loss to follow-up (n = 7). Consequently, the final analysis included 148 participants. At surgery, the mean age was 31.15 ± 10.35 years (range: 18–56) in the EF group and 32.35 ± 8.90 years (range: 19–52) in the SF group. The mean follow-up duration was 29.45 ± 4.55 months (range: 24–40) for the EF group and 30.03 ± 5.77 months (range: 20–48) for the SF group. The demographic characteristics and preoperative injury patterns were not significantly different between the two groups.

The primary outcome, AOFAS hindfoot score at the final follow-up, was 94.58 ± 6.69 (range: 84–100) and 95.15 ± 5.69 (range: 85–100) for the EF and SF groups, respectively.At the final follow-up, the EF group had an OMA score of 95.47 ± 5.92 (range: 80–100) versus 94.92 ± 5.77 (range: 85–100) in the SF group. VAS scores were 0.66 ± 1.01 (range: 0–3) in the EF group and 0.56 ± 0.74 (range: 0–3) in the SF group. Comparative analysis revealed no statistically significant differences between groups in OMA (P > 0.05), AOFAS (P > 0.05), or VAS (P > 0.05) scores (Table 2). The data indicated that both methods achieved satisfactory clinical scores. Comparative analysis of DW classification revealed no statistically significant differences between groups in OMA, AOFAS, or VAS scores (Table 3).

Table 2.

Comparison of the cores between the 2 groups

	EF	SF	95% CI	P
OMA score	95.47 ± 5.92	94.92 ± 5.77	− 1.38 to 2.47	0.577
VAS score	0.66 ± 1.01	0.56 ± 0.74	− 0.20 to 0.37	0.549
AOFAS hindfoot score	94.58 ± 6.69	95.15 ± 5.69	− 2.63 to 1.91	0.591

Table 3.

Comparison of the mean CT parameters between the 2 groups

	Type B				Type C
	EF	SF	95% CI	P	EF	SF	95% CI	P
OMA score	96.78 ± 4.94	95.32 ± 5.92	− 1.62 to 4.56	0.345	94.83 ± 6.28	94.7 ± 5.75	− 2.35 to 2.60	0.919
VAS score	0.25 ± 0.75	0.36 ± 0.49	− 0.49 to 0.26	0.543	0.844 ± 1.07	0.68 ± 0.83	− 0.21 to 0.55	0.38
AOFAS score	97.46 ± 4.91	96.36 ± 4.92	− 1.72 to 3.92	0.436	96.20 ± 7.01	94.48 ± 6.02	− 3.92 to1.34	0.334

Radiologic outcomes

Quality of reduction assessment: 3D-reconstructed CT at the final follow-up was used, with an incongruence criterion defined as a ≥ 2 mm difference on axial CT scans. In the SF group, 4 patients (6.45%) exhibited ≥ 2 mm differences in anterior difference (AD) and posterior difference (PD) postoperatively compared with normal parameters. In the EF group, 3 patients (3.49%) had similar differences. No significant intergroup differences were detected (P > 0.05).

The comparison of CT parameters between the preoperative injured side and the postoperative injured side revealed P < 0.05. Additionally, the comparison between the preoperative injured side and the normal side yielded P < 0.05. However, there were no significant differences in the CT parameters for the postoperative injured side versus the normal side (P > 0.05) (Fig. 4). Statistical comparisons of the mean CT parameters between the two groups were performed. There was no significant difference between the two groups in any of the parameters (Table 4; Fig. 5).

Fig. 4.

Comparison of mean CT Parameters Among the preoperative and postoperative values for the injured side and the normal side

Table 4.

Comparison of the mean CT parameters between the 2 groups

	Preoperative value, injured side, mm				Postoperative value, injured side, mm				Normal side, mm
	EF	SF	95% CI	P	EF	SF	95% CI	P	EF	SF	95% CI	P
AD	9.79 ± 3.30	9.32 ± 3.05	− 0.60 to 1.52	0.388	4.23 ± 0.96	4.26 ± 0.91	− 0.34 to 0.28	0.838	4.17 ± 0.99	3.95 ± 0.91	− 0.10 to 0.53	0.18
PD	10.19 ± 3.25	9.86 ± 3.65	− 0.81 to 1.47	0.57	6.10 ± 1.66	6.06 ± 1.50	− 0.49 to 0.56	0.899	6.07 ± 1.62	5.74 ± 1.63	− 0.21 to 0.87	0.23
FT	2.22 ± 1.42	1.91 ± 1.26	− 0.14 to 0.76	0.171	1.75 ± 0.89	1.73 ± 0.85	− 0.27 to 0.31	0.971	1.57 ± 0.94	1.62 ± 0.89	− 0.35 to 0.26	0.763
FR	11.42 ± 5.22	11.26 ± 5.65	− 1.63 to 1.95	0.862	12.79 ± 5.05	11.66 ± 4.70	− 0.49 to 2.76	0.17	12.83 ± 4.6	11.65 ± 4.26	− 0.29 to 2.66	0.115

The units of FRwere degrees

Fig. 5.

. Comparison of mean CT parameters (AD, PD, FD, and FR)between the groups

CT parameter comparisons: (1) Preoperative vs. postoperative (injured side): Significant differences (P < 0.05). (2) Preoperative (injured side) vs. normal side: significant differences (P < 0.05). (3) Postoperative (injured side) vs. normal side: There were no significant differences (P > 0.05) (Fig. 4) (4) The mean CT parameters were not significantly different between the groups (Table 4; Fig. 5).

Complications

Overall, there was no significant difference in complication rates between the two groups (EF group: 9 patients [10.47%] vs. SF group: 5 patients [8.06%], P > 0.05). No major complications occurred in either group. Minor complications in the EF group included lateral pain due to wire knot irritation (3 patients, 3.49%), which resolved in 2 patients after knot removal under local anesthesia at 3–6 months postsurgery. Patients in the EF group also experienced medial pain from fiber wire irritation (5 patients, 5.81%), with symptoms subsiding after percutaneous wire cutting in the outpatient clinic. The SF group included patients with syndesmosis pain (2 patients) and implant irritation (1 patient, 1.61%), which resolved after hardware removal at 3 months postsurgery. Superficial wound infections (EF: 1 patient; SF: 2 patients) were managed successfully with oral antibiotics and wound dressings.

Discussion

To our knowledge, no prior study has directly compared these two flexible fixation methods for syndesmotic injuries. This is the first comparative study demonstrating that the embrace technique achieves satisfactory clinical and radiological outcomes in syndesmosis management, with results that are noninferior to those of suture button fixation. Both groups had comparable clinical and radiological outcomes at the final follow-up, with no statistically significant differences. These findings confirm our study hypothesis. However, the long-term clinical relevance of these findings requires further validation through extended follow-up studies.

The syndesmosis, a complex articulation between the distal tibia and fibula, permits critical motion during gait, including 1–2 mm of mortise widening, 3–5° of external rotation, and 2–3 mm of proximal‒distal fibular migration [24, 25]. Syndesmotic injuries frequently result from traumatic forces applied to the ankle, particularly in varus or supinated positions [26]. Additional mechanisms include excessive dorsiflexion–abduction forces. These injuries are commonly associated with pronation–abduction (stages II–III), pronation–external rotation (stage IV), and supination–external rotation fracture patterns [27]. Malreduction remains the sole modifiable risk factor for posttraumatic ankle arthrosis in patients with fractures requiring syndesmotic stabilization [5]. Consequently, precise anatomic restoration of the syndesmosis followed by biomechanically appropriate fixation is paramount for optimal outcomes. Both fixation methods demonstrated excellent reduction rates in the current study.

The optimal fixation technique for syndesmotic injuries remains controversial. While syndesmotic screws have traditionally been the mainstay of treatment, they have been criticized for their rigid stabilization and association with intraoperative malreduction [28, 29]. To address these limitations, suture button fixation has emerged as an alternative. Growing evidence suggests potential advantages of this technique. Zhang et al. [30] demonstrated an improved range of motion and an earlier return to work with suture button fixation. In a randomized trial with a 2-year follow-up, Andersen et al. [9]. reported superior AOFAS scores, reduced pain, and better radiological outcomes in the suture button group. Ræder et al. [31] reported sustained benefits at > 5 years of follow-up, including higher AOFAS scores and lower rates of posttraumatic osteoarthritis. Needle arthroscopy has been shown to allow real-time assessment of syndesmotic stability and to identify associated chondral lesions that may not be evident on fluoroscopy or CT [32, 33]. While future investigations will focus on advancements in this domain, it should be noted that the present study did not incorporate this technique as a standardized component of the methodological approach. Both the suture button and the embrace technique are flexible fixations and have achieved high clinical and radiological results. How about the comparison between these two techniques? The primary outcome and the secondary outcome, Our data showed that there were no statistically significant differences, among the OMA (P = 0.577), AOFAS (P = 0.549) and VAS (P = 0.591) scores at the last follow-up. Furthermore, we assessed postoperative distal tibiofibular syndesmotic reduction and compared four parameters (AD, PD, FD, and FR) between the EF and SF groups as judged by CT scan, according to the method described by Abdelaziz [17]. There was no statistically significant difference between the two groups (Figs. 4 and 5; Table 4). On the basis of the results of this study, we obtained satisfactory clinical and radiological outcomes with the use of the embrace technique in managing syndesmosis injuries. There was no significant difference between the 2 groups in terms of clinical or radiological outcomes at the final follow-up. Therefore, no distinct superiority was found between the 2 groups, but longer follow-up may be needed to assess the posttraumatic arthritis rate.

Both the suture button and the embrace technique are flexible fixation methods that produce excellent clinical and radiological outcomes. Our comparative analysis revealed that there were no statistically significant differences in the functional outcomes based on OMA (P = 0.577), AOFAS (P = 0.549), or VAS (P = 0.591) scores at the final follow-up, and there were no differences between the different types of DW classification. Radiographic assessment (CT using Abdelaziz’s method²⁰) revealed comparable syndesmotic reduction across four parameters (AD, PD, FD, and FR) between the groups (Figs. 4 and 5; Table 3). These findings indicate that the embrace technique yields equally satisfactory clinical and radiological results for syndesmosis injuries. The study confirmed our hypothesis, demonstrating no clear superiority of either fixation method at the final follow-up. From a biomechanical perspective, the embrace technique generates a circumferential force that centers the fibula within the tibial incisura. Unlike a point-specific action, this multidirectional resultant force may be closer to the anatomical state and contribute to a higher error tolerance rate, but a biomechanical study is needed to prove this. Additionally, the embrace technique offers practical advantages: it requires no specialized instrumentation and avoids the use of a medial metal plate. This feature makes it particularly valuable for patients with medial soft tissue injuries.

The complication rates were not significantly different between the groups (EF group: 9 patients [10.47%] vs. SF group: 5 patients [8.06%], P > 0.05). In the EF group, early cases with knots tied on the fibular plate resulted in palpable subcutaneous knots that frequently caused lateral discomfort during cross-legged sitting; this issue was prevented in later cases by rotating the knots anterior to the plate. In the SF group, two patients experienced syndesmotic pain potentially due to overtightening, while one patient developed implant irritation; all symptoms resolved following hardware removal at three months postsurgery. These findings are [20]. report of five suture button cases requiring implant removal due to lateral irritation (p = 0.2).

The retrospective study design has several well-known limitations, and we cannot rule out selection bias. In the current study, the long-term follow-up results are uncertain because the clinical relevance requires longer-term follow-up verification. Compared with plain radiography, CT provides increased sensitivity for the detection of malreduction. However, the inability to obtain weight-bearing CT images represents a limitation, as syndesmoses may reduce under physiologic loads. Additionally, the surgical technique lacked strict standardization regarding the placement of suture buttons and bone tunnels in the embrace fixation method. In the suture button fixation group, the number of TightRope devices used was not strictly controlled, with a few cases employing two devices rather than one. Due to data volume, fracture type data were not classified, studied, or evaluated. Additionally, the lack of inter - observer reliability indicators limited the quality of this paper.

Conclusion

Both the embrace technique and the suture button technique demonstrated comparable accuracy and effectiveness for syndesmotic stabilization in ankle fractures and yielded excellent radiological and clinical outcomes. Notably, the embrace technique yielded equivalent outcomes compared with suture button fixation for managing syndesmotic injuries.

Author contributions

Zhen Lu Cao, Zhen Hua Fang and Chen Han Wang wrote the main manuscript text Jian Wang, Tian Yu Li, Xiao Heng Ding perform statistical data analysis and prepared figures, TablesQuan Yu Dong Coordinate and guide the thesis writing.

Funding

This work was Supported by Qingdao Key Medical and Health Discipline Project（2025-WJKY179）; The Clinical Medicine + X Research Projects of the Affiliated Hospital of Qingdao University in 2024(QDFY+X2024143)

Data availability

No datasets were generated or analysed during the current study.

Declarations

Competing interests

The authors declare no competing interests.