Treatment Outcomes of Anterior Sternoclavicular Joint Dislocation using An Acromioclavicular Joint Hook Plate Versus a Locking Plate: A Retrospective Study.

Background: We aimed to compare the ecacy and functional outcomes of using an acromioclavicular joint hook plate (AJHP) versus a locking plate (LP) in the treatment of anterior sternoclavicular joint dislocation. Methods: Seventeen patients with anterior sternoclavicular joint dislocation were retrospectively analyzed from May 2014 to September 2019. Six patients were surgically treated with an AJHP, and 11 were surgically treated with an LP. Five male and one female patients comprised the AJHP group, and nine male and two female patients comprised the LP group. The mean age of all patients was 49.5 years. Results: Reduction and xation were performed with AJHP or LP in all 17 patients. All patients were followed up for a mean duration of 14.4 months. There were no reported complications, wound infections, or instances of plate or screw breakage. The mean operative blood loss, operative time, and length of incision in the AJHP group were signicantly better than those in the LP group. Shoulder girdle movement of the AJHP group was signicantly better than that of the LP group. Conclusions: This study revealed that AJHP facilitated glenohumeral joint motion, reduced the risk of rupture of mediastinal structures, required a shorter incision, and had less blood loss and a shorter duration of operation compared with LP. However, some deciencies require further improvement.


Background
The sternoclavicular joint is the only bony articulation between the upper extremity and axial skeleton [1][2][3]. The ligamentous structures surrounding the sternoclavicular joint maintain its stability and make it a constricted joint [4]. Due to the presence of these stable structures surrounding the sternoclavicular joint, sternoclavicular joint dislocation is infrequent, representing only 3% of all dislocations in the shoulder girdle treated clinically [5,6]. Dislocation can be divided into anterior dislocation and posterior dislocation by different injury mechanisms, dislocation directions, and clinical manifestations. The incidence of anterior dislocation is approximately 90% [7]. Lateral compressive force on the shoulder girdle can cause the anterior capsule and costoclavicular ligament to rupture, which results in anterior dislocation. These broken structures make the joint unstable, and manual reduction is di cult to maintain, leading to frequent cases of re-dislocation. Therefore, surgery is recommended for reducing the occurrence of unstable dislocation.
There are many different surgical methods that have been designed for the treatment of anterior sternoclavicular joint dislocation, such as Kirschner wires, FiberWire, two screws and a strong suture, Tplate, locking plate (LP), and acromioclavicular joint hook plate (AJHP) [8][9][10][11][12][13]. Important thoracic structures, such as the trachea, esophagus, brachiocephalic vein, brachiocephalic artery, and brachial plexus, exist behind the sternoclavicular joint, and surgical treatment may be required to prevent accidental injury, especially on the sternal manubrium [3,14,15]. As a classical surgical method, LPs have been used for the reduction and xation of the sternoclavicular joint [16,17]. We have reported the safety and e cacy of using an AJHP for the treatment of anterior sternoclavicular joint dislocation [13].
However, we did not evaluate the surgical methods and functional outcomes of AJHP compared with other kinds of internal xation.
This study aimed to compare the surgical methods and functional outcomes of using an AJHP to LPs in a series of 17 patients who were followed up for a minimum of 3 months. The Ethics Committee of Tongji Medical College, Huazhong University of Science and Technology approved this study.

General data
From May 2014 to September 2019, 17 patients with anterior sternoclavicular joint dislocation were admitted into our study. Of these, six patients were surgically treated with an AJHP, and 11 were surgically treated with an LP.
The exclusion criteria were as follows: (1) patients with brain injury or other serious illnesses that caused increased risk throughout surgery or anesthesia; (2) patients who demanded conservative treatment even if the closed reduction was useless; (3) patients with spinal cord injury or peripheral nerve injury that affected upper limb movement; and (4) patients with fracture, scapulohumeral periarthritis, or rotator cuff tear of the same side upper limb that affected upper limb movement. There were ve male and one female patients in the AJHP group and nine male and two female patients in the LP group. The mean age of all patients was 49.5 years. One patient in the AJHP group experienced bilateral dislocation of the sternoclavicular joints. Four patients had rib fractures, one had an ankle fracture, one had a lumbar vertebra fracture without spinal cord injury, and one had a thoracic vertebra fracture without spinal cord injury. The mechanisms of injury that caused the dislocation varied. Nine patients were injured in car accidents, ve experienced accidental falls from heights, two were injured by falling from driving a motorbike, and one was injured by beating. All patients underwent the standard preoperative assessment, including preoperative history, physical examination, radiography, and computed tomography. Closed reduction was attempted for all patients but was found to be unproductive, and surgery was chosen as the best mode of treatment. The interval between injury and surgery ranged from 1 to 15 days.

Surgical technique
All patients were placed in the supine position on the operating table and were administered with general anesthesia. For patients who underwent AJHP surgery, an anterosuperior straight incision was made, extending from the medial part of the clavicle to the middle of the superior aspect of the sternal manubrium, and the sternoclavicular joint, sternal manubrium, and medial clavicle were exposed (Fig. 1). The incarcerated soft tissue of the sternoclavicular joint was cleaned, and the broken sternoclavicular joint cartilage plate was replaced or cleaned. The pointed end of the AJHP was inserted into the dorsal osteal surface of the sternal manubrium, and the lever effect was used to press the proximal end of the clavicle down to achieve reduction. When reduction was successfully performed, three or four bicortical screws were inserted into the clavicle. The broken anterior sternoclavicular and costoclavicular ligaments were repaired using absorbable sutures.
For patients who underwent LP surgery, an anterosuperior curved incision was made extending from the medial part of the clavicle to the upper part of the mesosternum. After the incarcerated soft tissue and broken sternoclavicular joint cartilage plate were cleaned, the proximal end was pressed to reduce the dislocation, and a Kirschner wire was used for temporary xation. The LP was bent and placed on the surface of the clavicle and sternum, 3-4 bicortical screws were inserted into the clavicle, and 3-4 unicortical screws were xed into the sternum. The broken ligaments were then repaired.

Postoperative management
In the rst 4 weeks postoperatively, the shoulder was immobilized with a sling, and easy exercises of the glenohumeral joint in the range of 0°-90° abduction were authorized. After 4 weeks, the range of motion was increased in accordance with each patient's healing course. Within the rst 3 months, it was recommended that sporting activities be avoided.

Follow-up
All patients were followed up for a mean duration of 14.4 (range, 3-28) months. Radiography and range of motion measurement of the glenohumeral joint were performed on the rst postoperative day; after 4, 8, and 12 weeks; and then once every 6 months.

General data
In the AJHP group, one patient underwent bilateral operation, ve underwent unilateral operations, and a total of seven sternoclavicular joints underwent surgery. For the seven joint operations, the mean operative blood loss of each joint was 52.9 (range, 35-80) mL, and the mean operative time was 0.73 (range, 0.5-1.08) h. The mean incision length was 8.21 (range, 6.5-9.5) cm.
In the LP group, 11 patients underwent unilateral operations. For the 11 joint operations, the mean operative blood loss of each joint was 83.2 (range, 60-150) mL, and the mean operative time was 1.68 (range, 0.83-2.25) h. The mean incision length was 17.5 (range, 15-20) cm. There were no respiratory or circulatory abnormalities reported in any of the patients considered. No postoperative incision infection and no complications, such as joint re-dislocation, vascular rupture, or vital organ injury, were observed. One patient in the LP group had pneumothorax after the operation, but the patient also underwent same-side rib fracture internal xation in the same operation. Associated injuries were treated effectively. There was no plate breakage or screw breakage reported at the nal follow-up.

Radiography, motion range measurement, and physical function
Postoperative radiography showed that all dislocated joints were successfully treated, and the location and angle of the plates were suitable. No re-dislocation occurred during the period of follow-up. In the AJHP group, the mean postoperative abduction angle of the glenohumeral joint was 164.3° (range, 150°-175°), the mean posterior extension angle of the glenohumeral joint was 45.7° (range, 40°-55°), and the mean external rotation angle was 71.4° (range, 50°-80°). In the LP group, the mean postoperative abduction angle of the glenohumeral joint was 146.8° (range, 130°-160°), the mean posterior extension angle of the glenohumeral joint was 32.7° (range, 20°-50°), and the mean external rotation angle was 57.7° (range, 45°-70°). The shoulder girdle movement of patients in the AJHP group was signi cantly better than that of patients in the LP group (Table 1).

Discussion
The ligaments that surround the sternoclavicular joint guarantee its stability; however, the joint is inherently unstable [2,3]. A direct force on the medial part of the clavicle could lead to posterior dislocation of the joint [15]. Indirect lateral compressive forces that affect the shoulder could rupture the anterior capsule and ligaments via the lever effect of the clavicle and can potentially result in anterior sternoclavicular joint dislocation [7,15]. Closed reduction is the rst choice for the treatment of anterior sternoclavicular joint dislocation, but it is usually unsatisfactory due to ligamentous rupture, which makes the joint unstable. The sternoclavicular joint is the only bony articulation between the axial skeleton and upper extremity, and movement at the joint, which is produced by the transmission of the movements of the scapula on the chest wall, can occur passively in three planes. During shoulder abduction, the sternoclavicular joint can move in the coronal and anteroposterior planes. Therefore, preservation of micromotion in the sternoclavicular joint has a signi cant effect on the postoperative range of glenohumeral joint motion [18]. According to our research, patients in the LP group had limited movement of the sternoclavicular joint; therefore, glenohumeral joint motion was signi cantly limited relative to that of those in the AJHP group.
There are many important thoracic structures, such as the esophagus, trachea, brachiocephalic veins, brachiocephalic artery, phrenic nerve, superior vena cava, and aorta, that are located posterior to the sternoclavicular joint [14,15]. The fact that these complex structures surround the sternoclavicular joint implies that dislocation causes serious trauma, such as rupture of blood vessels, nerve injury, pleura rupture, lung rupture, and mediastinum organ injury. Additionally, operations on the joint require increased caution. In the LP group, drilling and screwing on the sternum could increase the risk of rupture or injury to mediastinal structures, although such complications were not observed in this study.
Comparing the incisions used to surgically treat both groups, in the AJHP group, a small part of the sternal manubrium was exposed, which was su cient for plate insertion. However, in the LP group, most of the sternal manubrium, and even the mesosternum, was exposed to facilitate xation of the plate and screws. Due to the shorter incision and smaller number of screws needed, blood loss in the AJHP group was signi cantly lower than that in the LP group. In addition, LP bending, drilling, and screwing on the sternum prolonged the duration of the operation in the LP group.
In our research, AJHP and LP were used for the treatment of sternoclavicular joint dislocation. Some advantages of AJHP treatment were as follows: (1) AJHP facilitates micromotion within a certain range between the hook and manubrium of the sternum, which is bene cial for glenohumeral joint motion; (2) the hook structure behind the sternal manubrium could reduce the risk of mediastinal structure rupture during drilling and screwing; and (3) the AJHP operation method requires a shorter incision, produces less blood loss, and has a shorter operation time.
However, there are some disadvantages of AJHP treatment, such as the following: (1) the structure of the AJHP is not suitable for the anatomical structure of the sternoclavicular joint, and the plate is not in line with the joint; (2) because of the sharpness of the hook of the plate, some risk of mediastinal structure rupture still exists; and (3) there is no special screw hole available for sternoclavicular joint dislocation with medial clavicle fracture. Based on these disadvantages, we designed a new hook plate that has obtained a Chinese patent. The new hook plate conforms to the anatomical structure of the sternoclavicular joint, has a blunt hook to reduce the risk of rupturing mediastinal structures, and has some screw holes next to the hook to facilitate xation of medial clavicle fractures. This new hook plate has also been produced and used in the treatment of some patients with sternoclavicular joint dislocation, and its clinical e cacy should be evaluated further in future studies.

Conclusions
A comparison of AJHP and LP revealed that AJHP provided increased glenohumeral joint motion, reduced the risk of rupture of mediastinal structures, required a shorter incision, incurred less blood loss, and required a shorter operation time. However, AJHP still has some disadvantages. Further studies are needed to improve the design of the new hook plate in the future.  Figure 2 The patient has a left anterior sternoclavicular joint dislocation and is treated with an AJHP (A) The preoperative radiograph of the patient (B) The preoperative CT scan shows dislocation of the left anterior