Clinical outcomes of arthroscopic and navigation-assisted two tunnel technique for coracoclavicular ligament augmentation of acute acromioclavicular joint dislocations

Thirty-five consecutive patients with acute ACJ dislocations who were treated via arthroscopic and navigation-assisted DT procedures from 2013 to 2018 were included in this retrospective study [22, 23]. All interventions were performed by two experienced joint surgeons using the same suture system and our navigation technique. All operations where done within a period of 2 weeks after the injury. This study was approved by the local institutional ethics committee and follows the principles of the Declaration of Helsinki. Informed written consent was obtained from each patient.

A standard diagnostic arthroscopy of the shoulder joint was performed under general anesthesia with the patient in a beach chair position (see Additional file 1). In cases of concomitant glenohumeral injuries, these injuries were treated first. Following arthroscopy, an anterior-inferior working portal was created just above the subscapularis tendon and a lateral viewing portal was created using the outside-in technique. Next, the subcoracoid space and the base of the coracoid were prepared using a radiofrequency ablation device inserted through the anterior-inferior portal.

A 5-cm sagittal saber incision was made across the clavicle approximately 1.5 cm medial to the ACJ. The deltotrapezial fascia was identified, and a T-shaped incision of the fascia was made over the ACJ and lateral clavicle. The ACJ was freed from the wrapped soft tissue of the AC capsule and AC ligaments. If severely damaged, the articular disc was excised. After an open reduction of the ACJ, AC transfixation was performed using a K-wire. The correct reduction of the joint was evaluated using an image intensifier.

Two holes were drilled in the clavicle for the reconstruction of the CC ligament. The first followed the course of the conoid ligament with a 2.4-mm K-wire from the clavicle to the coracoid, 5 mm medial to the isometric point of the clavicle, as defined by Rios [24]. The target zone was the posterior side of the coracoid base, 5 mm lateral to the medial boundary. The second hole followed the course of the trapezoid ligament, beginning 5 mm lateral to the isometric point of the clavicle. The target zone for the second coracoid tunnel was 10 mm anterior to the conoidal tunnel and 5 mm medial to the medial edge of the coracoid, with the intention of leaving a bony bridge of at least 10 mm between the tunnels [12, 25].

An established optoelectronic system with a corresponding software module was used for navigation (Trauma 2D 3.1 software, produced by Brainlab AG, Munich, Germany). Reflective markers were attached to the pointer to determine its position and to the drill sleeve to navigate the drilling direction. A 3D camera enabled real-time tracking of the drill sleeve in relation to the pointer. The movement of the two instruments was controlled in three projections (front, top, and overview), which were displayed on a touchscreen. A virtual red line marked the tips of both instruments and showed the target trajectory. The corresponding drilling direction of the sleeve was then compared with the red line to reach the corresponding target point. An autopilot was used to orient the navigated instruments, similar to its recent application in trauma software [18, 20].

After the instruments had been calibrated, the tip of the pointer was positioned at the subcoracoid target area through the anterior-inferior portal under arthroscopic control. The inserted K-wires were drilled over with a cannulated drill (4.0 mm). Two suture cerclage systems (TightRope, Arthrex, Inc., Naples, Florida, USA) were introduced into the CC tunnel with an insertion aid (Application Sleeve & Pusher, Arthrex, Inc., USA) until the oval buttons could be anchored under the coracoid arch under arthroscopic control. The thread systems were tensioned by alternating tension between the two implants. Finally, the threads were tied proximally.

The detached deltoid and trapezius fascia were anatomically reattached to the lateral clavicle using transosseous sutures (#2 FiberWire, Arthrex, Inc., Naples, Florida, USA), and the complete closure of the fascia was accomplished using a fascia suture (Vicryl size 1, Ethicon, Norderstedt, Germany). The initial T-shaped incision of the AC capsule was closed using a 3.5-mm suture anchor (Arthrex, Naples, Florida, USA) on the lateral clavicle. Finally, the temporary ACJ K-wire transfixation was removed using the image intensifier. The upper incision was sutured in layers, and the arthroscopic portals were closed using conventional methods.

During the immediate postoperative period, the shoulder was immobilized via an internal rotation sling (shoulder immobilization support, Medi GmbH & Co. KG, Bayreuth, Germany). Patients were permitted to perform passive movement exercises up to a flexion and abduction of 45° for the first 3 weeks postoperatively and up to 90° during the subsequent 3 weeks. Active movement exercises were permitted beginning in postoperative week 7. Patients were advised to avoid exercises that stressed the ACJ, such as grasping, pushing, and pulling during that time.

The clinical evaluation (Fig. 1) consisted of a complete physical examination of both shoulders and several shoulder function evaluations, including the Constant-Murley score (CMS), subjective shoulder value (SSV), Taft score (TF), and acromioclavicular joint instability (ACJI) score [11, 26‐28]. The TF was described by Taft et al. 1987 [26]. It grades results after conservative and surgical treatment of AC joint dislocations. Subcategories are “subjective”(=pain; 4 points), “objective” (=range of motion and strength; 4 points) and “radiologic” (4 points). So, the maximum score is 12 points. Points can be subtracted for different symtoms: tenderness to palpation of the AC-joint, bad cosmetic results, or crepitation. Force measurements were performed on both shoulders with the aid of an isometric dynamometer (Isobex TM dynamometer, MDS Medical Device Solutions AG, Burgdorf, Switzerland). Quality of life was assessed using the EuroQol-5D (EQ-5D) questionnaire [29, 30].

Panoramic images were obtained with an axial load of 10 kg (Fig. 1). The CC distance was measured as the distance between the upper edge of the coracoid and the lower cortex of the clavicle. The measurement was performed on both shoulders preoperatively, during the immediate postoperative period, and during follow-up.

Horizontal stability was assessed via bilateral Alexander views on radiographs obtained during follow-up (Fig. 1C) [31‐34]. Patients were classified into three groups according to the ACJI score: stable ACJ, partially stable ACJ, and unstable ACJ [35].

The mean CMS on the injured side was 90 (range: 56–100) and on the contralateral side was 95 (range: 89–100) (p = 0.53). The mean SSV was 92% (range: 80–100). The mean TF was 10 (range: 4–12). The mean ACJI score was 86 (range: 34–100) (Fig. 2). The mean EQ-5D was 86 (range: 2–100) (Table 2).

The mean preoperative CC distance of the affected shoulder was 22.6 mm (range: 13.4–35.7 mm), and the mean postoperative CC distance was 8.8 mm (range: 4.4–14.9 mm). The mean CC distance of the healthy shoulder was 10.3 mm (range: 5.8–14.4 mm; p = 0.8). The mean CC difference during the follow-up period was 4 mm (range: 1.9–9.1 mm), which was not significantly different than that of the healthy shoulder (p = 0.06) (Fig. 3).

The Alexander view of postoperative radiographs revealed 1 patient (3.1%) with complete instability of the ACJ and 11 patients (34.4%) with partial instability. Of the patients with Rockwood type V injuries, none had complete instability and 10 (40%) had partial instability.

This study is not without limitations. The follow-up duration of 31 months is relatively short and no conclusions regarding the long-term effects or the development of osteoarthritis can be made. Unfortunately, a comparison group was not possible in the study setup. For this reason only the comparison with the literature was carried out. Additionally, a direct measurement of the drill holes via computed tomography to verify their position was not performed due to the radiation exposure and the young age of the patients. A complete clinical and radiological examination was performed at the final time point shown. Interim examinations at 6 and 12 months were performed irregularly and were not included in the analysis. The influence of open reconstruction of the AC ligaments, capsule, and fascia cannot be reliably differentiated from the anatomical configuration. At minimum, this study shows comparable or better outcomes for horizontal instability (based on the Alexander views and ACJI scores) compared to other clinical studies on the reconstruction of the ACJ [11, 13, 14].