Three-view plain radiographs, including true anteroposterior of the glenohumeral joint, scapular Y (scapular lateral), and Velpeau axillary [43] views, are the mainstay of imaging in the setting of traumatic anterior instability. The latter view is crucial to obtain, as the first two alone do not allow to exclude a dislocation. The goal is to confirm the direction of dislocation and to evaluate associated lesions (Fig. 17.5). Once reduced, further imaging studies in the setting of an associated fracture (computed tomography (CT)), suspicion of rotator cuff injury (ultrasonography or MRI), or vascular impairment (injected CT) may be warranted.

The first step is to analyze, if available, plan radiographs with the shoulder out of joint to confirm the direction of instability. Plain radiographs including anteroposterior in neutral, internal, and external rotations, scapular Y, and Bernageau [44] views are then obtained. Bone loss, static instability, post-dislocation arthropathy, and coracoid nonunion (if a Latarjet procedure is planned) have to be estimated. Magnetic resonance imaging (MRI) arthrogram is useful to assess for an anterior labral tear and Hill-Sachs lesion which confirms the instability. Associated intra-articular pathology such as SLAP, HAGL, and rotator cuff lesions or a paralabral cyst is also assessed (Fig. 17.6) [45]. The evaluation is completed by a 3D CT arthrogram in the setting of recurrent instability in which there is primary concern for bone loss. The extent of both glenoid bone loss and Hill-Sachs lesions is best assessed by CT scan and is used to determine the need for Latarjet as opposed to arthroscopic Bankart repair (Fig. 17.7).

The first step, whenever possible, is to obtain a complete set of radiographs before attempting a reduction. This will allow an assessment of the type of dislocation and associated bone injuries. Attempting to reduce a fracture dislocation can have troublesome clinical and legal consequences (Fig. 17.8). Exceptions are an impossibility to have reasonably fast access to radiology or a patient with neurological impairment. Because of the possible association of nerve injuries [47] and, to a lesser extent, vascular injuries (Fig. 17.9) [48], an essential part of the physical examination is an assessment of the neurovascular status of the upper extremity before reduction. There are numerous appropriate methods of reduction that have been described [49–58]. The second step is to use the technique of closed reduction which is mastered by the doctor who will perform the maneuver. The glenohumeral joint should be reduced as gently and expeditiously as possible. In the case of fracture dislocation, the reduction is best performed under general anesthesia to have adequate muscle relaxation. After reducing the dislocation, plain radiographs are obtained to verify the adequacy of the reduction. Results concerning conservative treatment are still debatable [59]. A stable shoulder is obtained at 10 years in only half of the patients with conservative treatment [60]. However, recurrence rate is highly dependent on age and activity of the patient; studies have reported a 72–95 % recurrence in patients under 20 years of age and 70–82 % recurrence between the ages of 20 and 30 years [61–66] and only 30 % in those over 30 years of age [67]. Many patients above the age of 30 would consequently undergo unnecessary surgery if proposed after the first dislocation. Conservative treatment after the first traumatic anterior dislocation may be thus recommended for patients who are not actively engaged in sports, above the age of 30 years old [61], with a low functional demand, with an associated humeral fracture [60], or for the athlete with an in-season shoulder dislocation [4]. For the latter situation, athletes are allowed to attempt to return to competition provided there is enough time left in the season to permit adequate rehabilitation with progression to sport-specific drills. Rehabilitation including return of range of motion and strengthening of dynamic stabilizers may facilitate return to sport within several weeks. Motion-limiting braces that prevent extreme shoulder abduction, extension, and external rotation are often prescribed as it may reduce the risk of recurrence. However, such braces are not well tolerated in patients who must complete certain overhead tasks such as throwing. Moreover, a second in-season shoulder dislocation should lead to removal from sport and proceed with stabilization so as to avoid further glenohumeral damage.

A number of studies have compared nonoperative treatment and arthroscopic stabilization. Overall, these studies report a sevenfold reduction in the risk of recurrent instability after arthroscopic stabilization, when compared with nonoperative treatment for the first-time dislocator [59]. A Cochrane review concluded that early surgical intervention is warranted in young adults aged less than 30 years engaged in highly demanding physical activities [68]. Consequently, for patients who are actively engaging in a collision or contact or overhead sport, who risk their life in case of a new dislocation (e.g., firemen, proponents of extreme sports like base jumping and climbing), with associated glenoid fracture, static anterior subluxation, an interposed tissue, or a nonconcentric reduction, or patients with rotator cuff avulsion, conservative measures are usually inadequate and prompt surgery is indicated.

Recurrent dislocation is not trivial. Each episode creates new lesions and increases the risk of developing dislocation arthropathy. The concept of early operative surgical management of the first-time dislocator has consequently been introduced to address the high recurrence rate in the young athletic population. A surgery should be proposed, as having the ultimate aim to achieve a pain-free stable shoulder while preserving range of motion. The surgical approach is based on the extent of bone loss and patient-specific risk factors for recurrence.

Boileau et al. proposed a simple 10-point scale scoring system (instability severity index score (ISIS)) based on factors derived from a preoperative questionnaire, physical examination, and anteroposterior radiographs to determine the risk of treatment failure following isolated arthroscopic Bankart repair (Table 17.1) [69]. In this model an ISIS of 3 or less was associated with a 5 % rate of recurrence, an ISIS of 4–6 was associated with a 10 % rate of recurrence, and an ISIS over 6 was associated with a 70 % rate of recurrence. Although it has imperfections, this score, validated since [70], has merit to easily remind the clinician of factors that are important to consider when evaluating a patient.

The aim of a Bankart repair is to restore anatomy by reattaching the labrum to the glenoid and tighten the inferior glenohumeral ligament by shifting from inferior to superior. It was previously believed that a minimum of three double-loaded suture anchors had to be used [71]. However, a recent study demonstrated that one to two anchors are enough [72]. One must take caution with the latter as the position of the anchors is probably the most important factor; in other words, the placement of three anchors likely reflects the fact that the surgeon is adequately placing inferior anchors and obtaining an inferior to superior shift of the pathological lesion. Although this surgery can be performed in an open manner, the advantage of an arthroscopic approach is that it preserves the subscapularis and allows assessment of associated pathology [73–75]. The literature demonstrates that patients with low risk of recurrence will benefit from either an anatomic open or arthroscopic repair with an acceptable rate of recurrence [76].

Remplissage has been described by Connoly [77] and may be used as an adjunct to arthroscopic Bankart repair in the setting of a large Hill-Sachs lesion with glenoid bone loss of <25 %. This technique consists of a posterior capsulodesis and infraspinatus tenodesis that fills the Hill-Sachs lesion. The purpose is to render the Hill-Sachs lesion extracapsular, avoiding its engagement. Wolf and Arianjam and Boileau et al. recently presented encouraging mid- to long-term results of arthroscopic remplissage and concomitant anterior Bankart repair [13, 78]. However, the indication for remplissage has not been well defined. Recently, DiGiacomo et al. introduced the concept of an “on-track vs. off-track” Hill-Sachs lesion [27]. This view incorporates the concept of the glenoid track by Yamamoto et al. which describes the contact of the posterior humeral head with the glenoid during abduction [79]. Based on study in normal individuals, this zone of contact or “glenoid track” averages 83 % of the glenoid width. In the proposal of DiGiacomo et al., the normal glenoid diameter is determined via CT scan of the contralateral glenoid or arthroscopic assessment (by doubling the distance from the glenoid bare area to the posterior glenoid rim). Then, the glenoid track is determined by multiplying the normal glenoid diameter by .83 and subtracting the anterior glenoid bone defect. Finally, the distance from the medial margin of the infraspinatus to the most medial aspect of the Hill-Sachs lesion (or Hill-Sachs interval) is measured. If the Hill-Sachs interval is greater than the remaining glenoid track, the lesion is considered “off-track,” and they recommend a remplissage in addition to an arthroscopic Bankart repair. However, this elegant method has to be clinically confirmed and factors like ligament status (HAGL lesions, laxity, and translation that render all Hill-Sachs lesions engaging) or ability to obtain intraoperatively reliable measures are not taken into account in the evaluation.

In the setting of glenoid bone loss ≥25 % of the glenoid diameter, an arthroscopic Bankart repair has an unacceptably high rate of recurrence. Burkhart and DeBeer reported a 4 % recurrence rate for arthroscopic Bankart repair when glenoid bone loss was <25 %. However, with glenoid bone loss ≥25 %, the recurrence rate was 67 % with an arthroscopic approach [25]. They subsequently recommended a Latarjet procedure in the population with substantial glenoid bone loss. In 1954, Latarjet reported a coracoid transfer procedure in which the inferior aspect of the coracoid was secured to the anterior glenoid. The excellent stability of this procedure is obtained by a triple effect first proposed by Patte and Debeyre [80]: (1) the sling effect of the conjoint tendon when the arm is abducted and externally rotated, (2) the “bony effect” that increases or restore the glenoid anteroposterior diameter, and (3) the retensioning of inferior capsule to the stump of coracoacromial ligament, rendering the coracoid extra-articular. The Latarjet procedure is associated with a very low recurrence rate even in the setting of substantial glenoid bone loss and has become the gold standard of treatment in such settings. In addition, this nonanatomic method of anterior glenohumeral stabilization has progressively expanded and is actually the primary technique of choice for many European surgeons, as it prevents recurrent anterior instability in approximately 95–99 % of cases [81, 82]. This procedure is also favored by some as a first choice in many contact athletes [83]. While traditionally an open procedure, the technique can now be performed arthroscopically [10, 11, 84, 85]. To date, however, there is actually no proven benefit of an arthroscopic approach to the procedure, and the risk of complication remains high with a large learning curve [86].

Finally, some authors have recommended autogenous iliac crest or tibial allograft as a means of restoring glenoid bone loss. However, it has been demonstrated that the most important effect of the Latarjet is provided by the conjoint tendon at both the end-range and the mid-range arm positions [87]. We therefore believe that the indication for iliac crest or tibial grafting should be limited to revision procedures such as following a failed Latarjet reconstruction.