Introduction

For carefully selected patients with primary anterior shoulder instability after traumatic dislocation, Bankart repair, open or arthroscopic, has been shown to have a high success rate. In a recent, well-designed meta-analysis of primary open and arthroscopic techniques, Lenters and colleagues¹ showed better return to work/sport with decreased risk of recurrent instability in open versus arthroscopically treated cases of recurrent anterior instability. Interestingly, Rowe scores were better in the arthroscopically treated group, despite the fact that instability contributes to approximately half of the outcome score. This finding suggests that function and motion were possibly better in the arthroscopically treated group. The authors also acknowledged that they did not perform sub-group analyses on arthroscopically treated shoulders in the settings of glenoid bone loss and contact sports, which may have contributed to the higher recurrence rates reported.

Multiple factors may contribute to failure of surgical stabilization and recurrent anterior shoulder instability. In broad categories, etiologies for recurrent instability after Bankart repair include patient age, technical error, missed injuries at index surgery, significant glenoid bone loss, large humeral head defects, capsular attenuation tissues, early return to contact sport, and inappropriate rehabilitation.

Revision surgery may be performed by open or arthroscopic techniques, although specific injury patterns or concomitant injuries may dictate the surgical tactic. With careful preoperative evaluation and planning, revision anterior stabilization is a successful surgery with a low rate of recurrent instability and dislocation.

Anatomy

The stabilizers of the glenohumeral joint have traditionally been classified as static and dynamic; shoulder stability is conferred by the complex interplay of these anatomic structures. The shallow dish of the glenoid is effectively deepened by the dimensions of the articular cartilage, which deepens at the periphery and the labrum, which deepens glenoid concavity and improves adhesive contact with the head of the humerus.² The labrum receives its blood supply from its peripheral synovial junction where branches from the suprascapular, posterior humeral circumflex, and circumflex scapular arteries give branches. The anterosuperior portion of the labrum has the most tenuous blood supply, which may predispose to superior labral anteroposterior (SLAP) tears.³ Without the restraint of the labrum, the glenohumeral joint is destabilized by at least 20%.⁴ Tearing or avulsion of the labrum, particularly at the insertion of the inferior glenohumeral ligament, is often described as the essential instability lesion of the shoulder.^5,6 Cracks or tears in the labrum at the posterior glenoid rim have also been shown to destabilize the glenohumeral joint.⁷

The glenohumeral ligaments contribute to the centering of the humeral head in the glenoid at the extremes of motion, protecting the rotator cuff muscles from excessive stretch.² The three principle capsular thickenings are the superior, middle, and inferior glenohumeral ligaments. The superior glenohumeral ligament centers the humeral head when the arm is held in adduction, the middle resists anterior translation in the mid-range of abduction. The inferior glenohumeral ligament has an anterior and posterior band that stabilize and center the joint when the shoulder is in the abducted and externally rotated position, as in the cocking phase of throwing.

The glenohumeral joint lacks bony congruity, but the stability conferred by the intact glenoid is highlighted by cases of glenoid fracture, ie, a bony Bankart lesion or glenoid insufficiency (the pear-shaped glenoid). Studies have demonstrated that glenoid deficiency >21% leads to recurrent shoulder instability, even after effective Bankart repairs.^8,9 Joint congruency may also be adversely affected by large defects in the articular surface of the humeral head, such as Hill-Sachs lesions formed by traumatic dislocations of the shoulder joint. Studies have shown the incidence of Hill-Sachs lesions after anterior dislocation to be between 30% and 50%.¹⁰ These lesions may engage the lip of the glenoid at the limits of rotation and lever the humerus out of the glenoid. Recent studies have also demonstrated a correlation between excessive glenoid retroversion and glenohumeral instability.⁷

The dynamic stabilizers of the glenohumeral joint are the muscles of the rotator cuff and their tendinous insertions on the greater and lesser tuberosities of the humerus and the scapulothoracic and thoraco-humeral muscles. Biomechanical studies have shown that rotator cuff tears destabilize the glenohumeral joint without disruption of the capsule or labrum. Tears of the supraspinatus extending into the infraspinatus tendon result in loss of resistance to inferior translation.¹¹

The rotator cuff muscles are often described as force couplers, coordinating muscle action about the shoulder to maintain a compressive force that centers the humeral head on the glenoid. The subscapularis, supraspinatus, infraspinatus, and teres minor act in coordination to compress the humeral head through the full range of shoulder motion, in contrast with other scapulothoracic muscles like the pectoralis major, pectoralis minor, and latissimus dorsi, which contribute only in specific positions.²

The smooth control of torque across the joint occurs by co-activation, or simultaneous contraction of agonist and antagonist cuff muscles, and by coordinated activation and inhibition of the agonist and antagonist.¹¹ For example, in the late cocking phase of the throwing cycle, when the arm is abducted and maximally externally rotated, the subscapularis and the infraspinatus/teres minor muscles are simultaneously firing to center the humeral head.¹¹

The long head of the biceps tendon is thought to contribute glenohumeral stability by depressing the humeral head and resisting anterior and posterior translation in external rotation.¹² The long head of the biceps is considered a primary stabilizer of the shoulder in the presence of a Bankart lesion.¹¹ Significant tears of the biceps anchor and the superior labrum can lead to increased instability in multiple planes, further destabilizing the glenohumeral joint.¹

History and Physical Examination

Many patients with recurrent shoulder instability are able to accurately describe their symptoms and the movements and/or positions that provoke them and, therefore, the diagnosis seems evident. Others may present with more subtle and complicated symptoms. If possible, a copy of the operative report of the primary surgery should be obtained and reviewed. The surgeon should develop a clear sense of the timeline of recurrent symptoms. Did the patient experience recurrent instability early after the primary surgery? It is also necessary to determine whether the recurrent shoulder instability followed a traumatic event or evolved slowly over time. How frequent are the patient's instability episodes? Can the patient relocate his or her shoulder without assistance? Can the patient voluntarily dislocate his or her glenohumeral joint?

After a thorough history of the instability complaints, the physical examination should provide the orthopedic surgeon with an opportunity to confirm his or her sense of the diagnosis. The shoulder examination should always be preceded by a thorough cervical spine examination, including the provocative maneuvers of Spurling and Lhermitte. This is followed by a careful observation of the shoulder through a normal range of motion. Is there notable muscle atrophy or scapular dyskinesis present? Examination should begin with the asymptomatic shoulder. Range of motion in all planes is tested with the patient standing and supine, and deficits and/or asymmetries are documented. Standard examination of the acromioclavicular joint and subacromial impingement tests are performed and rotator cuff strength is documented, with particular attention to the lift-off test for subscapularis strength and integrity. As with any case of shoulder instability, particular attention should be paid to the scapula, noting scapulohumeral rhythm, any muscle atrophy (especially of the infraspinatus), and any dyskinesis. A wall push-up as well as a standing examination with the arms slowly elevated provides crucial scapular information.

Anterior shoulder instability is tested first. The apprehension test is performed with the patient supine: the arm is abducted and externally rotated until the patient describes a sense of impending instability. The examiner then performs the relocation test by placing his hand over the anterior aspect of the glenohumeral joint and depressing the humeral head. The relocation test is positive when the patient's apprehension is relieved. Posterior instability is examined by performing a posterior stress test. With the patient supine, the arm is loaded posteriorly in a flexed and internally rotated position, and the examiner feels for posterior subluxation of the humeral head. The jerk test is performed with the patient seated and the arm similarly positioned. If subluxation occurs with posterior loading, then the humeral head reduces with a jerk as the arm is extended. The jerk test is positive if the relocation occurs with a painful jerk. The patient is then asked to roll onto his or her side for a modified load and shift test. In this position, the examiner grasps the humeral head with one hand while supporting the weight of the upper arm in the other hand. This allows for the most sensitive detection of anterior and posterior translation, which is graded minimal translation (grade 0), humeral head translation to the glenoid rim (grade 1), humeral head translates over the rim but spontaneously relocates (grade 2), and humeral head translates over the rim but requires manual reduction (grade 3).

Humeral head translation testing may also be tested in the lateral decubitus position, documenting any aspect of symptomatic instability in the anterior, posterior, or inferior direction. The direction of instability that reproduces the patient's symptoms should be documented. Keep in mind that anterior instability is not purely in an anterior direction, but rather anteroinferior, so a large inferior component of symptomatic instability should be documented as failure to address this surgically may lead to failure.

The sulcus test is performed with the patient seated, the elbow flexed 90°, while the examiner applies an inferiorly directed force to the humerus attempting to translate the humeral head inferiorly. The test is positive when a visible groove can be created beneath the lateral lip of the acromion. If the sulcus is positive, then the test should be repeated with the arm externally rotated, noting if the sulcus is obliterated with external rotation. Some surgeons have suggested that, in the setting of a persistent sulcus with external rotation, a rotator interval lesion is present and warrants surgical closure to improve stability.¹³ This is a controversial topic, however, rotator interval closure has been shown to biomechanically improve the anterior stability of the shoulder, much more so than inferior stability.

Imaging Studies

In the case of high-energy traumatic anterior shoulder dislocations, a plain radiograph shoulder series may demonstrate the classic findings of the bony Bankart, or antero-inferior glenoid lip fracture, and the Hill-Sachs deformity, or posterior humeral head impaction fracture. In more subtle instability cases, no such findings may be present. All instability cases should be imaged with an axillary lateral view to depict humeral head centering on the glenoid and an anteroposterior view in the line of the scapula. Additional radiographs, such as the apical oblique and West Point View may provide improved visualization of the glenoid and offer evaluation of the glenoid bone stock.

Magnetic resonance imaging (MRI) provides the most sensitive and specific depiction of shoulder anatomy. Magnetic resonance imaging, especially with the addition of contrast injection or magnetic resonance arthrography (MRA), gives clear imaging of tears of the glenoid labrum and concomitant injuries to the biceps tendon or its anchor, the glenohumeral ligaments, or the rotator cuff.

In cases of recurrent instability where radiographs or MRI/MRA indicate significant glenoid fractures or bone loss, computed tomography (CT) scans with 3-dimensional reconstructions and humeral head subtraction may guide surgical decision making.⁴

Risk Factors for Recurrence of Instability after Primary Arthroscopic Stabilization

One of the most common technical errors resulting in failed primary anterior shoulder stabilization is the failure to restore the labrum to its anatomic height on the glenoid rim. This can be the result of an insufficient release of the torn labral cartilage, which often settles level or medial to the glenoid rim. A failure to reduce a glenoid lip fracture with its attached labrum is also a common cause of recurrent instability. In the case of anterior labral periosteal sleeve avulsion lesions, the labrum and the attached periosteum must be identified medially on the neck of the glenoid, liberated, and restored to its anatomic height and position.

Capsular imbrication and shift plays a critical role in the restoration of stability in the revision setting. Studies have shown that capsular hyperlaxity alone can be the cause of instability.^14-16 Inadequate capsular tensioning by open or arthroscopic technique may lead to recurrent instability. The devastating complication of capsular attenuation or deficiency after thermal capsulorrhaphy has also been described.⁴ Complex capsular reconstructions by an open approach may be necessary in this setting.¹⁷

Another common cause of failure is missed concomitant pathology at the time of primary repair. This may include posterior labral tears, SLAP tears, long head biceps tendinopathy, and/or rotator cuff tears. Failure to recognize and address such injuries can lead to recurrent instability and/or diminished functional outcomes.

Boileau and colleagues¹⁸ prospectively examined 91 consecutive patients who underwent primary arthroscopic stabilization for recurrent anterior traumatic shoulder instability. Nearly half of the patients were involved in collision sports. At 3-years' follow-up, 15% of patients had recurrent instability. The risk of postoperative recurrence was significantly related to glenoid bone defects, Hill-Sachs lesions, inferior shoulder hyperlaxity and/or anterior shoulder hyperlaxity, and the use of <3 suture anchors. The combination of glenoid bone loss and inferior hyperlaxity led to a 75% recurrence in their series.

Arthroscopic Versus Open Revision Bankart Surgery

At the inception of arthroscopic shoulder stabilization, recurrence rates were unacceptably high, nearing 50%, when compared with open Bankart revision rates. While open Bankart surgery has traditionally been considered the gold standard in the revision setting, there is a growing body of evidence that equivalent results may be achieved with revision arthroscopic Bankart surgery.^18,19 Prior open surgical repair does not necessitate a revision open approach. Authors have described revision arthroscopic repairs with good results, even in the setting of a prior open procedure.¹⁹

Indications to consider an open approach for revision shoulder stabilization include significant glenoid bone loss (especially >20%), subscapularis insufficiency, capsular attenuation, and engaging Hill-Sachs deformities. Open procedures best address the primary pathology related to the recurrence of instability after surgery.⁴

Addressing Glenoid Bone Loss

Glenoid bone loss remains a challenge and may be difficult to recognize preoperatively. The best imaging modality to assess glenoid bone loss is a 3-dimensional CT scan with digital subtraction of the humeral head. The CT scan should be considered if a patient has a long history of shoulder instability (with documented traumatic instability), an initial instability episode of trivial trauma (suggesting glenoid hypoplasia), multiple prior dislocations and reduction attempts, and an easy ability to dislocate or sublux the shoulder. Oftentimes, the diagnosis of bone loss will not be made until the arthroscopy or open procedure.

To best visualize anterior glenoid bone loss, the method of Lo¹⁴ and Burkhart²⁰ is used with the arthroscope in the anterior superior portal. A calibrated probe is inserted from the posterior portal to measure bone loss from the bare spot of the glenoid. A glenoid bare spot to anterior bone loss area of the glenoid should be 12 mm normally, and a 6-mm loss (a 6-mm measurement from the bare spot) corresponds to a 25% glenoid bone loss. The authors' consider proceeding with arthroscopic repair if the glenoid bone loss is <20% to 25%, and transition to bony augmentation procedures if there is a higher grade of loss. In the setting of glenoid bone loss, results have been less favorable than for those without bone loss, however, a reasonable retensioning of the capsulolabral structures may lead to a predictably stable shoulder joint.^20,21 Surgeons should discuss with patients that the overall failure risk is higher in patients with bone loss treated with an arthroscopic procedure. Bony augmentation procedures, such as the Latarjet, have been shown to work well in the setting of anterior glenoid bone loss and are generally recommended with losses >20%, although this is controversial.

Rotator Interval Closure

Closure of the rotator interval as an adjunct to improve anterior stability (and inferior) stability of the shoulder remains controversial. Although Harryman¹³ did not show improvement in anterior stability with open imbrication of the rotator interval structures, others have shown that arthroscopic rotator interval closure may improve anterior stability of the shoulder, possibly by the combination of a barrier effect, tensioning of the middle glenohumeral ligament, and improving capsular tension anteriorly.^22-24 External rotation loss may be a factor if rotator interval closure is performed. However, this may be minimized if the shoulder is placed in 30° to 45° of external rotation during the rotator interval closure. Although rotator interval closure is generally not a routine part of primary arthroscopic anterior shoulder surgery, it may be a useful adjunct in the revision setting. More evidence is necessary to draw definitive conclusions regarding the use of rotator interval closure in a revision situation.

Conclusion

Revision Bankart surgery, by open or arthroscopic technique, may yield high rates of success when care is taken to identify the root causes of failure. The patient with previously unrecognized hyperlaxity may pose additional challenges in the revision setting. Care must be taken to diagnose and address Hill-Sachs lesions and glenoid bone loss, as well as possible other contributing causes of instability, such as missed rotator cuff tears.

References

1. Lenters TR, Franta AK, Wolf FM, Leopold SS, Matsen FA. Arthroscopic compared with open repairs for recurrent anterior shoulder instability: A systematic review and meta-analysis of the literature. J Bone Joint Surg Am. 2007; 89:244-254.
   
    
2. Matsen FA, Chebli C, Lippitt S. Principles for the evaluation and management of shoulder instability. J Bone Joint Surg Am. 2006; 88:648-659.
   
    
3. Cooper DE, Arnoczky SP, O'Brien SJ, Warren RF, DiCarlo E, Allen AA. Anatomy, histology, and vascularity of the glenoid labrum. An anatomical study. J Bone Joint Surg Am. 1992; 74:46-52.
   
    
4. Millett PJ, Clavert P, Hatch III GFR, Warner JJP. Recurrent posterior shoulder instability. J Am Acad Ortho Surg. 2006; 14:464-476.
   
    
5. Rowe C, Patel D, Southmayd W. The Bankart procedure: A long-term end-result study. J Bone Joint Surg Am. 1978; 60:1-16.
   
    
6. Itoi E, Newman SR, Kuechle DK, Morrey BF, An KN. Dynamic anterior stabilizers of the shoulder with the arm in abduction. J Bone Joint Surg Br. 1994; 76:834-836.
   
    
7. Kim SH, Noh KC, Park JS, Ryu BD, Oh I. Loss of chondrolabral containment of the glenohumeral joint in atraumatic posteroinferior multidirectional instability. J Bone Joint Surg Am. 2005; 87:92-98.
   
    
8. Itoi E, Lee SB, Berglund LJ, Berge LL, An KN. The effect of a glenoid defect on anteroinferior stability of the shoulder after Bankart repair: A cadaveric study. J Bone Joint Surg Am. 2000; 82:35-46.
   
    
9. Burkhart SS, De Beer JF. Traumatic glenohumeral bone defects and their relationship to failure of arthroscopic Bankart repairs: Significance of the inverted-pair glenoid and the humeral engaging Hill-Sachs lesion. Arthroscopy. 2000; 16:677-694.
   
    
10. Miniaci A, Gish MW. Management of anterior glenohumeral instability associated with large Hill-Sachs defects. Techniques in Shoulder & Elbow Surgery. 2004; 5:170-175.
    
     
11. Abboud JA, Soslowsky LJ. Interplay of the static and dynamic restraints in glenohumeral instability. Clin Orthop Relat Res. 2002; 400:48-57.
    
     
12. Hawkins RJ, Schutte JP, Janda DH, Huckell GH. Translation of the glenohumeral joint with the patient under anesthesia. J Shoulder Elbow Surg. 1996; 5:255-262.
    
     
13. Harryman DT 2nd, Sidles JA, Harris SL, Matsen FA 3rd. The role of the rotator interval capsule in passive motion and stability of the shoulder. J Bone Joint Surg Am. 1992; 74:53-66.
    
     
14. Lo IK, Parten PM, Burkhart SS. The inverted pear glenoid: An indicator of significant glenoid bone loss. Arthroscopy. 2004; 20:169-174.
    
     
15. Bigliani LU, Kurzweil PR, Schwartzbach CC, Wolfe IN, Flatow EL. Inferior capsular shift procedure for anterior-inferior shoulder instability in athletes. Am J Sports Med. 1994; 22:578-584.
    
     
16. Turkel SJ, Panio MW, Marshall JL, Girgis FG. Stabilizing mechanisms preventing anterior dislocation of the glenohumeral joint. J Bone Joint Surg Am. 1981; 63:1208-1217.
    
     
17. Warner JJ, Venegas AA, Lehtinen JT, Macy JJ. Management of capsular deficiency of the shoulder. A report of three cases. J Bone Joint Surg Am. 2002; 84:1668-1671.
    
     
18. Boileau P, Villalba M, Henry JY, Balg F, Ahrens P, Neyton L. Risk factors for recurrence of shoulder instability after arthroscopic Bankart repair. J Bone Joint Surg Am. 2006; 88:1755-1763.
    
     
19. Kim S. Arthroscopic revision Bankart repair for failed anterior shoulder instability. JSES. 2002; 3:292-298.
    
     
20. Burkhart SS, Debeer JF, Tehrany AM, Parten PM. Quantifying glenoid bone loss arthroscopically in shoulder instability. Arthroscopy. 2002; 18:488-491.
    
     
21. Sugaya H, Moriishi J, Kanisawa I, Tsuchiya A. Arthroscopic osseous Bankart repair for chronic recurrent traumatic anterior glenohumeral instability. J Bone Joint Surg Am. 2005; 87:1752-1760.
    
     
22. Plausinis D, Bravman JT, Heywood C, Kummer FJ, Kwon YW, Jazrawi LM. Arthroscopic rotator interval closure: Effect of sutures on glenohumeral motion and anterior-posterior translation. Am J Sports Med. 2006; 34:1656-1661.
    
     
23. Yamamoto N, Itoi E, Tuoheti Y, et al. Effect of rotator interval closure on glenohumeral stability and motion: A cadaveric study. J Shoulder Elbow Surg. 2006; 15:750-758.
    
     
24. Provencher MT, Dewing CB, McCormick F, Stanley M. An analysis of the rotator interval in patients with anterior, posterior, and multidirectional shoulder instability. Paper presented at: Annual Meeting of the American Academy of Orthopaedic Surgeons; February 2007; San Diego, Calif.