In the present study, we retrospectively examined 63 patients who underwent suture-button fixation for ankle syndesmosis with TightRope (Arthrex, Naples, FL, USA) from January 2014 to February 2019. This study had 3 inclusion criteria adopted from previous studies: 1) presence of preoperative diastasis of tibiofibular clear space (TFCS) >6 mm as measured in the standard anteroposterior (AP) view or mortise view, 2) preoperative MRI image with increased signal in T2-weighted imaging of interosseous space presenting disruption of the anteroinferior tibiofibular ligament (AITFL) or posteroinferior tibiofibular ligament (PITFL), and 3) occurrence of intraoperative movement >3 mm after bony fixation by pulling of the fibula laterally with a towel clip (9). Patients who meet either of the above criteria were included. Exclusion criteria were patients under 18 years old, occurrence or presence of an open fracture, inflammatory arthritis (rheumatoid or psoriatic), multiple fractures that prevented patients from immediate ambulation, and a concomitant head injury or neurologic deficit. The study was approved by Taipei Medical University – Joint Institutional Review Board.
Surgical technique and rehabilitation
Surgery was performed under general or spinal anesthesia in the supine position with a bump under the hip. If any fracture is present, open reduction and internal fixation with either plate or screws was done, followed by intraoperative evaluation of the syndesmosis by lateral pull of fibula. After syndesmosis instability was confirmed, syndesmosis fixation was performed through following steps. First, open reduction of the syndesmosis was done under direct visualization and temporarily fixed with a k-wire. Thereafter, we used a guide pin to locate the optimal insertion site and trajectory. The site of insertion was 2–4 cm above the ankle joint aimed parallel to the joint and 20° to 30° anteriorly in the coronal plane from the fibula to tibia. After the guide pin was passed through the ideal tract, reaming was performed to open the canal for suture-button passage. The suture-button was passed through the guide needle with pull-through sutures, and then, the oblong button was flipped to rest on the medial cortex of the distal tibia. The position and reduction were determined under fluoroscopy, and then, the suture was tightened until the lateral button rested on the distal fibular cortex or lateral malleolar plate if present. (Figure 1)
After surgery, all patients were immobilized with a below-knee splint and maintained a nonweight bearing status for 2 weeks and then protected weight bearing for another 4 weeks. Full weight bearing was allowed 6 weeks after surgery. A monthly outpatient follow-up was performed for at least 6 months. Implants were removed only if local discomfort was claimed or requested by the patient.
Both AP, lateral, and mortise views of the ankle were taken at each time points, namely pre-operative, post-operative, and final follow-up. The pre-operative and post-operative X-rays were taken in supine position, and the final follow-up X-rays were taken under weight-bearing condition. The imaging techniques were standardized as follows.
For non-weight bearing AP views, the patient lay supine on examination table, with a pillow under head and leg fully extended. The foot was placed vertically on the image receptor, in neutral extended position. Slight foot pronation may be required for true AP position. For non-weight bearing mortise views, the patient positioned same as AP view, with the ankle internally rotated 15 to 20 degrees until the intermalleolar line was parallel to the image receptor. For AP and mortise views, the central ray was perpendicular to the receptor and directed at a point midway between the malleoli (22, 23). For non-weight bearing lateral views, the patient lay in lateral recumbent position with the affected side down, flex the knee approximately 45 degrees and the opposite leg behind the injured leg. Place support under the knee as needed to place the leg and foot in a true lateral position, with the lateral malleolus about 1cm posterior to the medial malleolus. Dorsiflex the foot so the plantar surface was at a right angle to the leg or as far as the patient could tolerate. The central ray was perpendicular to the receptor and directed to medial malleolus (22, 23).
For weight bearing AP and lateral views, have the patient stand erect with weight evenly distributed on both feet. The feet should be directed straight ahead and parallel to each other. The placement of image receptor and direction of the central ray was same as non-weight bearing views. For weight bearing mortise views, have the patient stand erect with weight evenly distributed on both feet. Place the image receptor behind the feet, internally rotate the injured leg by 15 to 20 degrees, until the intermalleolar line was parallel to the image receptor (22, 23).
For all above images, the source to image distance were set to 102cm (40 inches) (22, 23).
Radiographic measurements were done at both 3 time points mentioned above. The tibiofibular clear space (TFCS) was measured as the distance between the medial border of the fibula and the incisura fibularis on a line parallel and 1cm above the tibia plafond on AP view. The tibiofibular overlap (TFO) was measured as the maximum amount of overlap on mortise view (24, 25). The medial clear space (MCS) was measured as the distance between the medial border of the talus and the lateral border of the medial malleolus on a line parallel and 5mm below the talar dome (25, 26). All measurements were recorded to the nearest 0.1mm. (Figure 2)
Patients were grouped based on the injury pattern, which included pure syndesmotic injury, Weber-B type fracture, and Weber-C type fracture (27). The paired t-test was used to compare preoperative, postoperative, and final follow-up measurements. Then, the one-way analysis of variance test and Welch test were used for comparing final follow-up measurements between the groups. All analyses were performed using SPSS, version 19.0 (SPSS, Chicago, IL, USA).