- ▪
Shoulder instability can be unidirectional, bidirectional, or multidirectional.
- ▪
Unidirectional anterior instability is the most common pattern of shoulder instability.
- ▪
Instability occurs along a spectrum and can affect patients with and without hyperlaxity of the joints.
- ▪
The glenohumeral joint is inherently unstable and relies on static and dynamic stabilizers whose function is integrated to produce smooth, painless shoulder range of motion; these stabilizers individually are not sufficient to stabilize the shoulder joint.
- ▪
Treatment choices are nonoperative or operative, either arthroscopic or open, and depend on the age and occupational pursuits of the patient, the details of the injury, the direction of instability, associated bony injuries, and the presence of or potential for recurrence.
The spectrum of shoulder instability is wide, manifesting itself as unidirectional, bidirectional, or multidirectional, with or without fracture. Shoulder instability can be recurrent, causing chronic symptoms, and posing a difficult challenge to the surgeon. Unidirectional anterior instability is the most common pattern. Unidirectional posterior or inferior instability can occur, but these are less common presentations. Multidirectional instability has been increasingly recognized since Neer and Foster popularized an understanding of its etiology and treatment. Instability occurs along a spectrum and may present as subluxation or frank dislocation, and can occur in patients with and without hyperlaxity of the joints. Instability can be painful and may limit the pursuit of athletics and sometimes even the ability to perform activities of daily living.
Anterior Glenohumeral Instability
The mechanism of injury for traumatic anterior dislocations is usually falls, trauma, athletic injuries, or domestic or work injuries, typically with the arm in an abducted and externally rotated position ( Fig. 91-1 ). Recurrent instability is the most common complication after traumatic anterior shoulder dislocation. The rate of recurrence has been reported to range from less than 25% to approaching 100%, and is related to patient age at the time of dislocation, patient sex, the degree of trauma and capsulolabroligamentous injury, bony lesions including Hill–Sachs lesions (impaction of the posterosuperior humeral head), anteroinferior glenoid rim fractures, greater tuberosity fractures, rotator cuff tears, and the presence of neurologic injury ( Fig. 91-2 ). In one study, 21 patients who were younger than 13 years all had recurrent anterior dislocations, suggesting that younger age at the time of the initial dislocation correlated with a higher recurrence rate of persistent instability. A longer duration of immobilization after the initial dislocation has not decreased the recurrence rate of persistent instability.
Recurrent instability in the form of subtle subluxation to frank dislocation can be detrimental to the glenohumeral joint and can lead to pain, glenoid and humeral head bone loss, chondral defects, arthritic change, and other labral and ligamentous injuries. The incidence of osteoarthritis in unoperated unstable shoulders has been reported to be between 9.2% and 20.0%. The prevalence of osteoarthritis in these shoulders was related to patient age, time from injury to surgery, lack of external rotation, and presence of osseous glenoid rim fractures. Therefore, it is imperative to diagnose and properly treat recurrent anterior shoulder instability.
Anatomic and Biomechanical Considerations
The factors that contribute to persistent instability and recurrence can be understood through a complete understanding of the normal anatomy. The shoulder girdle is composed of the scapulothoracic, glenohumeral, and acromioclavicular articulations. The glenohumeral articulation is one of the most inherently unstable joints in the human body and it has the widest range of motion of any joint. Therefore, it can be easily understood why the glenohumeral joint is the most frequently dislocated joint.
Glenohumeral joint stabilizers are categorized as static or dynamic. Static stabilizers consist of the bony articulation, capsulolabroligamentous structures, and negative intra-articular pressure, which produces a suction effect of the humeral head in the glenoid. Dynamic stabilizers consist of the rotator cuff muscles, biceps tendon, thoracohumeral musculature, and periscapular musculature. The function of the static and dynamic stabilizers is integrated to produce a smooth, painless, and wide range of motion. Individually, they are not sufficient to stabilize the glenohumeral joint.
The bony anatomy provides little inherent stability. Although there is a congruent articulation between the glenoid and humeral head, there is disparity in their surface areas. Only one fourth of the humeral head articulates with the glenoid at any given time. In addition, the glenoid is shallow, permitting translation of the humeral head in multiple directions. A more flattened glenoid allows a greater degree of translation of the humeral head. The socket, however, is deepened by the glenoid labrum by approximately 50%, increasing the surface contact area up to 75%. Other bony considerations include the anteversion or retroversion of the glenoid. Increased anteversion faces the glenoid more anteriorly with respect to the plane of the scapula, increasing the potential for anterior instability. Conversely, a retroverted glenoid is oriented more posteriorly with respect to the scapula, increasing the potential for posterior instability. Protraction or retraction of the scapula also changes the relative anteversion or retroversion of the glenoid, respectively, affecting the propensity for instability.
The mechanisms of adhesion–cohesion and negative intra-articular pressure are other passive mechanisms maintaining stability. Adhesion–cohesion is based on the wettable surfaces of the humeral and glenoid articular cartilage. The surfaces adhere to each other because of the adhesive and cohesive properties of water molecules, permitting easy gliding of the surfaces while preventing distraction of the joint. The glenohumeral joint also contains negative intra-articular pressure, creating a suction effect to help maintain a centered humeral head.
The role of the dynamic stabilizers is one of compression of the humeral head into the glenoid. The dynamic stabilizers include the rotator cuff, long head of the biceps, deltoid, teres major, and latissimus. The unique muscles of the rotator cuff provide compression of the humeral head into the glenoid at varying positions of the arm, thus increasing the stability of the glenohumeral joint. The rotator cuff forms a force couple around the glenohumeral joint. Coactivation of the agonist and antagonist muscles increases the control of motion, whereas coordinated activation of the agonist muscles and inhibition of the antagonist muscles produces motion of the joint. Injury to the rotator cuff can disrupt the force couple and alter joint forces such that translation rather than rotation occurs. Both the size and the location of the tear influence the pattern of instability.
The capsule of the glenohumeral joint contains thickened areas that form the glenohumeral ligaments ( Fig. 91-3 ). These ligaments play an important role in stability. The capsuloligamentous structures confer stability at extreme ranges of motion. In midranges of motion, these structures are lax, allowing joint translation. Stability in midranges of motion is provided by the dynamic stabilizers, adhesion–cohesion effect, negative intra-articular pressure, articular congruity, and labrum. At extreme positions, the muscular stabilizers are unable to generate force and the ligaments come under progressively greater tension to provide stability.
The main ligaments are the superior, middle, and inferior glenohumeral ligaments, as well as the coracohumeral ligament. These ligaments play a differential role with respect to their function as a glenohumeral stabilizer. The coracohumeral ligament and the superior glenohumeral ligament limit external rotation and inferior translation in adduction and posterior translation in flexion. The middle glenohumeral ligament limits external rotation and inferior translation in adduction and anterior translation in midabduction. The inferior glenohumeral ligament limits anterior, posterior, and inferior translation in abduction. The inferior glenohumeral ligament is designed with bands in the anterior and posterior aspects of the inferior aspect of the joint, with a patulous capsule in between that mimics a hammock.
The key to a normal, stable articulation of the glenohumeral joint is to maintain the humeral head centered on the glenoid during motion. This stability is provided by the mechanism of concavity compression and the concept of scapulohumeral balance.
Concavity compression refers to the stability afforded a convex object that is pressed into a concave surface. In the glenohumeral joint, this exists as a balanced interplay between the static and dynamic stabilizers. Therefore, any compromise of these structures, including ligament and labral tears, glenoid fractures, chondral lesions, rotator cuff tears, and injuries to the outer muscular sleeve (deltoid, pectoralis major, and latissimus), can contribute to the development of instability.
Scapulohumeral balance refers to the principle that the humeral head is balanced on the glenoid if the net joint reaction force passes through the glenoid fossa. The scapula must be positioned such that the net joint reaction force is coaxial with a line through the center of the glenoid fossa. Thus, motion at the scapulothoracic joint helps to maintain a centered glenohumeral articulation.
Historical Perspective
Uncomplicated anterior dislocations typically produce a soft tissue injury with or without a Hill–Sachs defect and anteroinferior glenoid rim fracture. Bankart originally described this traumatic disruption in 1938 as a tear of the fibrocartilaginous labrum from almost the entire anterior half of the rim of the glenoid cavity, including a tear of the capsule and periosteum from the anterior surface of the neck of the scapula. The torn labrum was termed the Bankart lesion, and it was believed that this alone was the “essential” lesion ( Fig. 91-4 ). Rowe et al. reported its presence in more than 90% of operative cases. Bankart believed that repair of this capsulolabral injury anatomically was the only rational approach.
Biomechanical studies, however, proved the importance of the glenohumeral ligaments as restraints to dislocation. A Bankart lesion alone does not allow enough translation of the joint to result in dislocation. Elongation of the inferior glenohumeral ligament is also required for dislocation. Therefore, the inferior glenohumeral ligament and anterior labrum collectively contribute to stability in the abducted and externally rotated position of dislocation. Rowe et al. confirmed these assertions when reviewing the causes of failure of initial surgical repair. Bankart lesions were found in 84% of shoulders with recurrence after operation, and excessive capsular laxity was found to be a contributing factor in 83% of shoulders.
Open Procedures
The procedures described to manage anterior instability can be grouped as anatomic and nonanatomic. Anatomic repairs consist of reattaching the capsulolabral structures to the glenoid as well as reestablishing the proper tension in the capsule and ligaments. Bankart and Thomas and Matsen held that only repair of the labrum and its attached capsule was necessary, without reefing of the anterior capsuloligamentous structures. Capsular reefing and subscapularis advancement was applicable when a labral tear was not present, indicating that a patulous capsule was the cause of instability. Others described medial reefing of the capsule or advancement of the subscapularis as part of the Bankart repair. Wirth et al. explained that reducing the capsular volume was as important as the Bankart repair. An inferior capsular shift was performed that was analogous to that originally described by Neer and Foster. The anterior capsular shift is designed to reduce the circumferential capsular volume. This is done by incising the capsule and advancing the capsular flaps to stiffen and reduce the volume of the circumferential capsular tissue.
Nonanatomic surgical procedures were designed to limit external rotation and provide an anterior buttress to dislocation. These included the Putti-Platt procedure, which involved shortening and plication of the subscapularis; the Magnuson-Stack procedure, which is an advancement of the subscapularis; the Eden-Hybbinette procedure, which is iliac crest bone grafting of the anterior glenoid ; and the Bristow-Latarjet procedure, which is a transfer of the coracoid with the attached conjoint tendon to the anteroinferior glenoid through or around the subscapularis ( Fig. 91-5 ).
Complications
Complications of open anterior stabilization procedures include recurrent instability, stiffness, subscapularis deficiency, arthrosis, neurovascular injuries, and hardware problems.
The rate of recurrent instability after a Bankart repair in most short-term studies has generally been low. The rate of recurrent instability for nonanatomic reconstructions has also been low. Long-term recurrence rates have varied and are secondary to traumatic and atraumatic causes, but the rate has been lower for Bankart repairs and the Bristow-Latarjet procedure than for the Putti-Platt and Magnuson-Stack procedures. Late dislocation rates after the Putti-Platt and Magnuson-Stack procedures are between 10% and 20% in some studies. Some studies, such as one by Hovelius et al., compared open Bankart repair with the Bristow-Latarjet procedure and found no major differences after 16- to 20-year follow-up.
Factors related to recurrent instability after an open procedure include misdiagnosis of posterior or multidirectional instability, incomplete correction of the anatomic lesion, a bony defect with loss of glenoid concavity, a Hill–Sachs lesion, and anterior capsular deficiency. Other authors have stated that recurrent dislocations can be associated with younger age, bilateral involvement, positive family history, severe reinjury, and scarring of the subscapularis, and may be more indicative of inherent hyperlaxity of the joint or multidirectional instability. Reoperation can be successful and should be considered if a correctable cause of the recurrent instability can be determined. However, it is more important to identify patients with multidirectional instability before the index procedure. These patients require a different approach to management. Revision surgery is also associated with poorer outcomes. Levine et al. showed a recurrence rate of 17% after one operation and 44% after multiple open procedures.
Incomplete repair of the anatomic lesion may occur despite making the correct diagnosis. These technical failures include unrepaired Bankart lesions, unrepaired capsular injury, and failure to detect and treat an anterior bony glenoid rim defect or a humeral avulsion of the glenohumeral ligament tear. These factors are preventable and require appropriate preoperative evaluation.
Complications specific to the Bristow-Latarjet procedure (coracoid transfer) include recurrent instability, articular cartilage injury, nonunion of the coracoid to the scapula, resorption, screw loosening and hardware problems, neurovascular injury, and posterior instability. Young and Rockwood did not recommend the use of the Bristow-Latarjet procedure because it is a nonanatomic operation that is technically demanding and can be associated with multiple complications. In addition, they reported that the revision operation was difficult, and the rate of success of other anterior reconstructions was similar. If the operation is done technically well, however, the long-term results of the Bristow-Latarjet procedure are acceptable and in some reports superior to those of other reconstructions, particularly in the contact athlete. Allain et al. found only a 7% rate of complications after 14 years of follow-up. Hovelius et al. found good results with low complication rates at 15 years of follow-up. Another study comparing the Bristow-Latarjet procedure and the Bankart repair reported comparable results at 15 to 20 years of follow-up.
Loss of external rotation was once believed to be beneficial to help prevent recurrence. Open anterior capsulorraphy procedures, mainly the Putti-Platt and Magnuson-Stack procedures, often significantly limit external rotation and result in pain, osteoarthritic changes, and even posterior dislocation. Conversely, increased external rotation after anterior stabilization procedures can be indicative of subscapularis failure, resulting in postoperative disability. Current surgical reconstructions strive to restore stability as well as painless full range of motion.
Long-term results of the open Bankart repair have been reported to be favorable, with low recurrence rates, nearly full range of motion, and high rates of patient satisfaction. The Bankart repair has gained wide acceptance and has become the procedure of choice. Currently, most open repairs are anatomic reconstructions with a combination of Bankart repair and tensioning of the capsule.
Arthroscopy
The advent of arthroscopy brought intrigue and new challenges to the management of anterior shoulder instability. Morgan and Bodenstab described the first arthroscopic technique to repair the anterior labrum using a pull-through suture. Twenty-five patients were followed for an average of 17 months, and all achieved painless, full range of motion without recurrence of instability. Others also reported excellent results after Bankart repair. However, other reports showed high failure rates, with significant recurrence of glenohumeral instability. High failure rates were linked to inadequate retensioning of the capsule and ligaments. Others found high failure rates in younger patients who participated in contact sports.
Results of arthroscopic repair changed when capsular laxity was recognized as a problem. Pagnani and Warren described the “drive-through sign” in 1993 to determine the presence of capsular laxity. The ability to pass the arthroscopic cannula from the superior aspect of the joint to the inferior pouch easily after labral repair indicates capsular laxity. These patients were candidates for open procedures because arthroscopic techniques to restore tension to the capsule had not yet been developed. Reports emerged showing low recurrence rates comparable to those associated with open repair when a capsular shift was performed. Studies of open versus arthroscopic Bankart repairs also yielded comparable results. A recent meta-analysis found inferior results with arthroscopic compared with open repair in preventing recurrent instability. However, the study design and specific arthroscopic repair technique used had a substantial effect on the results. Arthroscopic approaches were found to result in better overall function.
An arthroscopic approach has several advantages over the open approach. There is superior visualization of intra-articular pathology, limited dissection obviating the release and repair of the subscapularis, reduction in postoperative scarring and loss of strength, decreased rehabilitation time, decreased loss of motion, and reduced postoperative pain and bleeding. In addition, the procedure can be performed in an outpatient setting. Disadvantages of the open technique include subscapularis failure, loss of external rotation, and difficulty in achieving strength and returning to athletic participation.
Arthroscopic repair in the high-demand patient (young patients, those with military participation, contact and overhead throwing athletes) has been shown to be effective in the recent literature ( Fig. 91-6 ). In a comparative study of open and arthroscopic treatment of collision athletes, however, open treatment resulted in lower rates of failure.
Arthroscopic failures are related to persistent or recurrent Bankart lesions, large Hill–Sachs lesions, glenoid bone loss, the number of dislocations, the number of revision surgeries, concomitant capsular laxity, and the postoperative immobilization period. Bone loss of greater than 25% of the anteroinferior glenoid, either developmentally or acquired, is generally considered a contraindication to arthroscopic repair.
In the past, anterior stabilization procedures were nonanatomic, with significant long-term complications. Historically, management focused on stable reduction and prevention of recurrent dislocation. Patients today are more active, seeking to maintain athletic ability into older age. Contemporary standards of success are higher, especially in the younger athlete, with maintenance of full motion and strength and return to functional activities, including competitive sports, in addition to the restoration of stability. Current arthroscopic procedures permit patients to return to their previous level of sporting or daily activity. Proper patient selection is imperative for improved outcomes with an arthroscopic technique.
Management
The management of traumatic anterior glenohumeral dislocation is variable. Many factors must be considered, including the following:
- •
Patient age
- •
Activity level, including participation in collision or overhead sports
- •
First dislocation or recurrent
- •
Glenoid or humeral bony involvement
- •
Other instability patterns, such as posterior, inferior, or multidirectional
Patients with first-time dislocations typically present after a reduction was performed in the emergency room. The affected area may be swollen and painful. Evaluation should rule out neurovascular injuries, especially to the axillary nerve and artery. Pre- and post-reduction radiographs taken in the emergency room should be reviewed for fracture. Younger patients may have a concomitant greater tuberosity or surgical neck fracture and a Hill–Sachs lesion. Patients older than 40 years are more likely to injure the rotator cuff.
The evaluation involves assessment of anterior instability with the apprehension test. This is performed by abducting and externally rotating the arm, which causes apprehension in unstable patients. Other instability patterns are also determined. Laxity testing, however, can often be misleading. A significant amount of capsular laxity can be present in normal individuals without symptoms. Hyperlaxity does not equal instability. Pathologic instability exists when instability testing reproduces the patient’s symptoms, avoiding inaccurate diagnoses and unnecessary surgical treatment. After several weeks, the pain and swelling from the acute dislocation may be diminished and the rotator cuff can be examined for weakness to assess whether a rotator cuff tear is suspected.
Patients with first-time dislocation are mainly treated nonoperatively. Rehabilitation starts with a 3-week period of immobilization in a sling with the shoulder in neutral to 10 degrees of external rotation. A study by Etoi et al. suggested that immobilization in more external rotation may better reapproximate the capsulolabral tissues and allow for healing. In addition to the sling immobilization, a supervised therapy program is initiated for pendulum exercises and elbow range of motion. From 3 to 6 weeks, the range of motion is progressively increased. Passive and active assisted forward flexion to 90 degrees with full internal rotation is begun as tolerated. External rotation with the arm at the side is limited to 30 degrees. Abduction of the shoulder beyond 90 degrees is also avoided. Preventing an abducted and externally rotated position allows the soft tissues to heal in a more anatomic position and may reduce the rate of recurrence. Isometric strengthening of the rotator cuff and deltoid is started at 6 weeks and progressed to resistive isokinetic strengthening by 3 months. After 3 months, unlimited activities are allowed, except for collision or overhead sports, which are begun at 6 months.
Factors that predict recurrence include young age and participation in contact, collision, or overhead sports. These patients may benefit from early operative intervention. Therefore, management of the patient with first-time dislocation must be individualized. Patients must be counseled to ensure quality care and achieve excellent outcomes. Surgery is offered in these high-demand patients. In the more sedentary patient, a trial of nonoperative treatment is appropriate.
Recurrent dislocation is an indication for surgery. Associated injuries, such as glenoid rim fractures and rotator cuff tears, may be present, and a high index of suspicion is required. Surgery may be necessary in these cases, and proper evaluation may necessitate advanced imaging. MRI may show a Bankart lesion or rotator cuff tear ( Fig. 91-7 ). A computed tomography (CT) scan of the shoulder more accurately shows the osseous structures, allowing better assessment of glenoid rim fractures and Hill–Sachs lesions, and permits more accurate planning of surgical reconstruction.
