Posterior instability: Open and bone augmentation surgery

History of open posterior shoulder surgery

Posterior glenohumeral instability includes a variety of clinical presentations, ranging from traumatic posterior glenohumeral dislocation to recurrent microinstability that frequently presents as nondescript posterior shoulder pain. ^, Since the first case of posterior glenohumeral dislocation documented in the medical literature by Sir Astley Cooper in 1822 and later posthumously published within an American edition in 1851, ^, our understanding and treatment of posterior glenohumeral instability have significantly evolved. However, controversies regarding appropriate treatment of this increasingly recognized, albeit rare, clinical entity continue to exist.

The earliest described operative techniques for treatment of both acute and chronic posterior glenohumeral dislocation included open humeral head resection. ^, This morbid procedure was later replaced with humeral head–preserving open reduction techniques at the start of the 20th century. However, it was not until the mid-20th century that concomitant osseous pathology associated with chronic or recurrent posterior glenohumeral instability, including posterior glenoid bone loss, glenoid dysplasia, and reverse Hill-Sachs lesions, were adeptly recognized by treating surgeons. This led to the development of anatomy-restoring open surgical techniques, including the McLaughlin procedure, posterior bone block procedures, and glenoid osteotomy procedures. Many of these original open procedures, some of which have been subtly modified, continue to serve a role in practice today.

As imaging techniques progressed from glenohumeral axillary arthrotomography to computed tomography (CT) arthrograms and magnetic resonance (MR) arthrograms, clinicians gained a clearer understanding of the osseous and soft tissue pathology associated with posterior glenohumeral instability. Although first described in 1931, glenohumeral arthroscopy for diagnostic purposes gained traction throughout the 1980s and 1990s, ^, further improving our understanding of the pathoanatomy associated with posterior glenohumeral instability. This concurrent evolution in imaging techniques and diagnostic arthroscopy provided the rationale behind open labral repair and capsular shift techniques performed throughout the 1980s and 1990s.

Open versus arthroscopic posterior shoulder stabilization

A discussion of the history of open posterior shoulder surgery would be incomplete without acknowledging the technological advancements and increased familiarity with arthroscopic techniques. While early arthroscopic techniques simply allowed for debridement of the pathological labrum, ^, the late 1990s and early 2000s provided a growing number of reports on arthroscopic posterior capsulolabral repair. As arthroscopic techniques and skills continued to advance, procedures that were once only performed through open techniques, including posterior bone block procedures and subscapularis remplissage, ^, have now become part of the arthroscopist’s armamentarium. This evolution is not unique to the management of posterior glenohumeral instability, as utilization of arthroscopic techniques for management of all directions of glenohumeral instability has increased in recent years. ^, However, the relative rarity of posterior glenohumeral instability has limited any high-level comparative studies, and much of what is known has been extrapolated from anterior shoulder instability data. Proponents of arthroscopic techniques frequently cite the minimally invasive nature of the procedure, limited soft tissue dissection, and superior outcomes compared with open posterior stabilization techniques when evaluating metadata. Still, as has been appreciated from similar transitions in anterior shoulder stabilization surgery, these complex arthroscopic procedures have a prolonged learning curve, high degree of variability, and higher associated costs. Certainly, both open and arthroscopic procedures will continue to serve a role for management of posterior glenohumeral instability, particularly as arthroscopic posterior bone augmentation techniques continue to evolve. ^, ^, While the best available evidence to guide when and where open versus arthroscopic posterior techniques are best utilized remains at the expert opinion level, understanding risk factors for recurrence and failure, while drawing from experiences with anterior shoulder instability, can help guide educated treatment decisions.

This chapter reviews the approach, indications, techniques, and outcomes of open posterior glenohumeral stabilization procedures and bone augmentation procedures for posterior glenohumeral instability. As noted above, arthroscopic-assisted bone augmentation techniques have recently been described while maintaining the principles of traditional open procedures. As techniques continue to evolve, clinicians treating posterior glenohumeral instability should continue to be familiar with open posterior techniques to appropriately treat a clinical entity with such diverse clinical presentations and associated pathoanatomy. In light of the limited clinical evidence, a thorough understanding of patient and anatomical risk factors is necessary to appropriately utilize open posterior techniques and recognize when bone augmentation may be necessary to optimize patient outcomes.

Posterior glenohumeral surgical approaches

Several approaches have been described for management of posterior glenohumeral instability. To address significant humeral head bone loss in the setting of large reverse Hill-Sachs lesions ^, or, less frequently, open anterior capsular plication to address primary posterior instability, a standard deltopectoral approach is utilized as described in Chapter 39 . Arthroscopic techniques utilize a similar approach to that described in Chapter 35 , with additional accessory portals or enlarged portals often required to introduce graft. This can be performed with the patient in the lateral decubitus or beach chair position.

The open posterior approach can be performed in the beach chair, lateral decubitus, or prone position depending on surgeon preference and concomitant procedures. Original posterior approaches described subperiosteal takedown of the posterior and middle heads of the deltoid to provide a rather extensive posterior exposure of the rotator cuff and glenohumeral joint. ^, For open treatment of posterior glenohumeral instability, this has been supplanted by a deltoid-splitting approach with a longitudinal incision centered over the posterior glenohumeral joint line starting just inferior to the scapular spine and extending distally toward the axillary fold. The posterior deltoid fascia is incised in line with the incision and the obliquely oriented deltoid muscle is bluntly dissected in line with the fibers and retracted, exposing the underlying infraspinatus and teres minor near the myotendinous junction. The deltoid split should not extend inferior to the teres minor to prevent injury to the axillary nerve ( Fig. 36.1 ). The deltoid is retracted with blunt retractors and the posterior rotator cuff exposed. Early techniques described infraspinatus or teres minor tenotomy to expose the underlying posterior capsule. ^, ^, However, access can reliably be achieved in the internervous interval between the infraspinatus and teres minor or between the muscle bellies of the bipennate infraspinatus in line with its muscle fibers. ^, This interval is often distinctly defined by a fat stripe extending beyond the myotendinous junction. Both options preserve the native rotator cuff attachment on the humerus, with the split of the infraspinatus theoretically providing easier retraction without significant morbidity and direct access to the posterior joint line at the optimal craniocaudal level. Careful dissection of the posterior capsule from the deep surface of the posterior rotator cuff and careful retractor placement aids with planning and completing the capsulotomy. The location and orientation of the capsulotomy is dependent on the procedure being performed, including glenoid osteotomy, bone block, humeral head bone grafting, or simply a posterior capsular shift with or without a formal labral repair. The posterior capsular shift is useful to imbricate and shift the often-redundant posterior capsule and utilizes a glenoid or humeral head–based T-shaped capsulotomy that aids intra-articular exposure prior to capsular closure. ^, ^,

Indications

Indications for open posterior glenohumeral stabilization and bone augmentation procedures remain relative and continue to evolve with technological advances and surgeon familiarity with complex arthroscopic techniques. Anecdotally, open posterior soft tissue techniques, including posterior labral repair or posterior capsular shift, can be considered in the setting of failed arthroscopic surgery with recurrent posterior instability, excessive glenoid retroversion, or in situations where there is clinically significant posterior glenoid bone loss. Both open and arthroscopic bone augmentation techniques have been described for clinically significant posterior glenoid bone loss and/or reverse Hill-Sachs lesion or in patients with severe glenoid dysplasia.

The biggest challenge in defining specific indications for open posterior procedures or bone augmentation has been determining which factors predict failure after primary posterior capsulolabral repair or capsular plication and what defines clinically significant posterior glenoid bone loss or bipolar bone loss, respectively. While patients with posterior labral tears and posterior instability symptoms have consistently demonstrated increased glenoid retroversion, outcomes of arthroscopic repair techniques in the primary setting do not appear to be significantly influenced by the magnitude of glenoid retroversion. However, decreased glenoid width, which in our opinion may serve as a surrogate measure for subclinical posterior glenoid bone loss or clinically unrecognized glenoid dysplasia or hypoplasia, has been associated with inferior outcomes and increased risk of revision surgery after primary arthroscopic posterior capsulolabral repair ^, ( Fig. 36.2 ). Some authors have recommended that a bone augmentation procedure or glenoid osteotomy be considered in the primary setting when glenoid retroversion exceeds 20 to 25 degrees. ^, While distinct cutoffs for guiding treatment remain debatable, open posterior or bone augmentation procedures or osteotomy should be thoughtfully considered in patients with short glenoid bone width or retroversion in excess of 20 degrees, particularly in those who have not responded to a previous arthroscopic procedure. Determining whether glenoid bone augmentation procedures, designed to address glenoid bone loss, or glenoid osteotomy, designed to correct glenoid retroversion, is better suited in specific situations is complicated by the complex interplay between glenoid bone loss and glenoid version. The subtleties of specific procedure selection in these scenarios depend on the treating surgeon’s comfort and familiarity with these procedures.

While there has been a great deal of literature in recent years focused on clinically significant anterior glenoid bone loss and bipolar bone loss in the setting of anterior glenohumeral instability, there is a relative paucity of literature on bone loss in patients with posterior glenohumeral instability. Hines et al. described a mean posterior glenoid bone loss of 7.3% in patients with posterior glenohumeral instability. In this population, 69% of patients had measurable glenoid bone loss, with 22% of patients having posterior glenoid bone loss exceeding 13.5%, a bone loss value that has previously been shown to result in inferior clinical outcomes after arthroscopic capsulolabral repair in patients with anterior glenohumeral instability. While there are limited clinical data to suggest that this bone loss threshold similarly affects patients with posterior glenohumeral instability, several cadaveric biomechanical studies do provide some insight. Nacca et al. determined that shoulders with posterior glenoid bone loss in excess of 20% remained unstable even after posterior capsulolabral repair, suggesting that a glenoid bone augmentation procedure may serve a role in this setting ( Fig. 36.3 ). Moroder et al. addressed the issue of bipolar bone loss in their CT-based study, determining that incremental posterior glenoid bone loss may lead to engagement of noncritical reverse Hill-Sachs lesions, a concept that is in line with the popularized glenoid track model described in anterior shoulder instability. Their subsequent algorithm would suggest that bone augmentation of the posterior glenoid and/or humeral head should be considered in the setting of bipolar bone loss.

In summary, the indications for open posterior shoulder stabilization and bone augmentation remain relative. Particular attention should be paid to prior procedures performed, glenoid version, glenoid bone width, and glenoid and humeral head bone loss. In addition to standard radiographic series, advanced imaging, including magnetic resonance imaging (MRI) and CT, can help aid in decision making. Three-dimensional MRI and CT reconstructions can be particularly useful when assessing glenoid or bipolar bone loss, as they have been shown to allow more accurate measurement of these bony defects. After thoughtful preoperative planning, conservative indications for open posterior stabilization or bone augmentation include recurrent posterior instability after a prior arthroscopic capsulolabral repair with residual patulous posterior capsule, posterior glenoid bone loss exceeding 20%, bipolar bone loss with concern for an engaging reverse Hill-Sachs lesion, short glenoid bone width, and glenoid retroversion in excess of 20 degrees. Although less frequently encountered, cavitary lesions of the glenoid vault secondary to ganglion cysts that do not necessarily change the dimensions of the glenoid but do cause concerns with the structural integrity of the underlying subchondral bone can be addressed through open posterior techniques and bone grafting ( Fig. 36.4 ).

Procedures and outcomes

Open posterior soft tissue repair

The goal of open posterior soft tissue repair is to restore the anatomy and function of the posterior labrum while concomitantly addressing and imbricating the patulous posterior and posteroinferior capsule in order to restore glenohumeral stability and reduce pain. This is performed through the previously described posterior approach, and the capsule exposed through the split in the bipennate infraspinatus muscle or in the internervous plane between infraspinatus and teres minor. In situations where the posterior capsule is particularly thin or adherent to the overlying rotator cuff, the infraspinatus can be taken down in whole or in a layered fashion as originally described in order to add tissue to the repair. ^, The joint is typically exposed through a T-capsulotomy and repair of the glenoid labrum performed with suture anchors. The capsulotomy is then closed, shifting the inferior leaflet superiorly to reduce the volume of the posteroinferior capsule and completing the posterior capsular shift. ^, ^, ^,

Neer’s original report on capsular shift for involuntary multidirectional instability, which included a combined cohort of both anterior and posterior approaches, described only one failure in 40 patients. Tibone et al. shared less optimistic results in an isolated posterior instability cohort treated with a glenoid-based medial staple capsulorrhaphy, reporting a 30% rate of recurrent postoperative instability. Hawkins et al. would later provide more encouraging results using a so-called capsulotenodesis of the posterior capsule and infraspinatus, with no recurrent instability at mean follow-up of more than 3 years and minimal pain reported. More modern techniques utilizing rotator cuff–preserving approaches and suture anchor–based repairs of the glenoid labrum combined with a formal capsular shift report recurrent posterior instability rates less than 20% and good or excellent patient reported outcomes in more than 85% of cases. ^, ^, ^, ^, In Bigliani’s series of 35 shoulders, nearly all of the treatment failures following posterior capsular shift occurred in the revision setting (six of seven shoulders), again highlighting the importance of recognizing and addressing bone loss and deformity. As has been previously reported in patients with anterior and multidirectional glenohumeral instability, patients with voluntary or habitual posterior instability provide a particularly difficult cohort to treat successfully with soft tissue repair.

Glenoid osteotomy

Opening wedge posterior glenoid osteotomy, or glenoplasty, has been described as an open posterior procedure to correct excessive glenoid retroversion contributing to posterior glenohumeral instability. The original technique, an osteotomy of the posterior glenoid neck, was first reported by Kretzler and Blue for the treatment of posterior dislocation in the setting of cerebral palsy. The technique was further defined and illustrated by Scott in 1967 for the treatment of recurrent posterior dislocation. Relatively few modifications to the procedure have been made to date. Indications for the procedure include recurrent posterior instability with a flat or retroverted glenoid. Absolute retroversion cutoffs are limited within the available literature, with Lacheta et al. recommending retroversion greater than 20 degrees as a relative indication. The opening wedge correction is maintained with an interference fit autologous bone graft, most frequently harvested from the acromion or iliac crest. ^, ^, Results are limited to case reports or small, retrospective case series. Graichen et al. provides the largest series to date, reporting retrospectively on 32 shoulders that underwent glenoid osteotomy. They reported good or excellent results in 81% of patients at 5 years with recurrent instability occurring in 12.5% of patients. They additionally noted that 25% of patients showed glenohumeral degenerative changes at last follow-up. Whether this is the result of intra-articular penetration of the osteotomy, as Johnston and colleagues reported, or the natural history of the posterior wear pattern in the retroverted glenoid cannot be clearly elucidated.

While several authors have reported positive outcomes with respect to recurrent instability, ^, ^, the lack of clarity with respect to indications, morbidity of the procedure, and high rate of clinically significant complications has limited the widespread adoption of the procedure. As previously mentioned, intra-articular fracture or extension of the osteotomy is a reported complication with the theoretical potential to accelerate degenerative changes ( Fig. 36.5 ). Additionally, anteriorization of the humeral head may result in iatrogenic anterior instability ^, or pain due to subcoracoid or coracohumeral impingement, as described by Gerber and colleagues. Although not reported, both the suprascapular and axillary nerve are at risk during the exposure and osteotomy. Due to the risk of these potentially severe complications, Lacheta and colleagues have rightfully recommended that the procedure only be considered and performed by experienced surgeons aware of the technical considerations and complications of the procedure.

Posterior glenoid bone augmentation

Posterior glenoid bone augmentation, frequently referred to as the posterior bone block procedure, has been described to address recurrent posterior glenohumeral instability in the setting of posterior glenoid bone loss, severe glenoid dysplasia, excessive glenoid retroversion, or hypoplasia of the posteroinferior glenoid ( Fig. 36.6 ). The procedure was originally described as an open procedure, but a variety of arthroscopic techniques for the posterior bone block procedure have evolved in parallel with the arthroscopic Latarjet procedure for anterior glenohumeral instability. ^, ^, ^, The procedure aims to increase the glenoid width, thus serving to lengthen the sliding distance prior to posterior glenohumeral subluxation or dislocation.

The procedure may be performed as an open or arthroscopic technique. In the open technique, the capsule is managed to allow a concomitant capsular shift that can be performed to make the graft intra- or extra-articular depending on the graft used and surgeon preference. The graft is typically secured with two or three 3.5 or 4.0 mm cannulated screws directed from posterior to anterior under fluoroscopic guidance ( Fig. 36.7 ). Lafosse was the first to describe the arthroscopic procedure in detail, utilizing commercially available guides for the Latarjet procedure to help deliver and fix the iliac crest bone graft through an extended posterior portal and capsulotomy. In Lafosse’s original description, the capsule is repaired back to the native glenoid, keeping the bone block extra-articular. While autologous iliac crest bone graft is the most frequently used graft in the literature, a variety of graft options exist. Autologous sources include tricortical iliac crest, distal clavicle, and scapular spine/acromion bone graft, while allograft sources include distal tibial allograft and glenoid allograft. ^, ^, While all graft sources have been shown to restore glenohumeral stability in cadaveric models, distal tibial allograft provides the advantage of being an osteoarticular allograft with potentially viable chondrocytes when obtained as a fresh bulk allograft while also optimizing joint congruity. ^,

The majority of the clinical outcomes data following posterior bone block procedures exist as retrospective case series. To date, there are no comparative series of open and arthroscopic techniques. In the largest published series, Servien et al. reported recurrent instability in only three of 21 (14.3%) patients, with more than 70% of patients returning to sport at their preinjury level at minimum 2-year follow-up using autologous iliac crest bone graft and an open technique. Meuffels et al. provide the longest follow-up at a median of 18 years, reporting significant concerns in the long term due in part to the 36% rate of recurrent instability as well as the ubiquitous degenerative changes noted in their cohort treated with autologous iliac crest bone graft and an open technique. Early outcomes with arthroscopic techniques have been promising. In a series of 24 shoulders, Wellman et al. reported a 12.5% recurrent instability rate at mean follow-up of 26 months. However, it will be important to monitor the learning curve associated with such advanced arthroscopic procedures as well as the long-term outcomes given the deteriorating results reported by Meuffels and colleagues with long-term follow-up of open posterior bone block techniques. ^, Complications following the posterior bone block procedure include symptomatic hardware requiring hardware removal, nonunion of the graft or resorption of the graft, intra-articular screw penetration, and progression or development of degenerative changes. As with any posterior approach requiring extensive exposure of the posterior glenoid and glenoid neck, injury to the axillary and suprascapular nerve can occur.

Lesser tuberosity transfer

The lesser tuberosity transfer, or modified McLaughlin procedure, is designed to address moderate-sized reverse Hill-Sachs lesions typically encountered with traumatic, locked posterior dislocations as the result of high-energy trauma, seizure, electrocution injury, or chronic posterior dislocations. The procedure is a modification of the original McLaughlin procedure, which included a transfer of the subscapularis tendon into the reverse Hill-Sachs defect. Both procedures aim to prevent engagement of the lesion with the posterior glenoid rim by filling the defect. The modified procedure provides several advantages, most significantly more robust initial fixation in addition to the opportunity for osseous union as opposed to tendon-to-bone healing. The procedure is typically performed through the standard deltopectoral approach and the defect visualized through the rotator interval. The osteotomy of the lesser tuberosity is then planned, extending from the bicipital groove laterally toward the most anterior edge of the reverse Hill-Sachs defect. The osteotomy is completed with a saw and/or osteotomes and the bone block transferred to the prepared bed of the defect. Fixation is typically achieved with two anterior to posterior directed bicortical screws, although synthetic suture fixation has also been described. ^, While the modified McLaughlin procedure has not been described arthroscopically, arthroscopic remplissage of the subscapularis tendon, a procedure more akin to the original McLaughlin procedure, has been described, but with no long-term data or specific clinical indications to guide treatment. ^, ^,

Indications for the modified McLaughlin procedure are limited due to the relative rarity of moderate to large-sized reverse Hill-Sachs lesions with recurrent posterior instability as well as a limited classification system. Moroder et al. attempted to provide a classification system that would more accurately categorize reverse Hill-Sachs lesions and guide treatment, an improvement from fractional estimates of humeral head involvement. ^, However, their methodology has yet to be widely adopted clinically.

Clinical outcomes following the modified McLaughlin procedure are limited to small, retrospective case series. In their original description, Hawkins et al. described all four patients as doing well at 2- to 9-year follow-up without recurrent instability. Banerjee et al. described good or excellent outcomes in all seven patients treated with reverse Hill-Sachs lesions ranging in size from 25% to 45% of the humeral head, with no further instability noted. Finkelstein et al. similarly noted good outcomes in seven patients treated for acute reverse Hill-Sachs impaction fractures with no recurrent instability. While the procedure has been shown to offer stability, there are concerns with restricted internal rotation motion following the procedure. To date, the procedure is primarily utilized for large lesions involving 20% to 50% of the humeral head. Further research is required to determine when and where other procedures, such as the arthroscopic subscapularis remplissage, could be utilized as a potentially less morbid procedure for smaller engaging reverse Hill-Sachs lesions.

Humeral head bone grafting

Determining when humeral head bone grafting is necessary to restore stability and function in the setting of posterior instability remains an unanswered question. Moroder et al. investigated the characteristics of reverse Hill-Sachs lesions and determined that the lesions were variable in both size and location with some characteristics correlating with the chronicity of posterior instability. In a cadaveric model, they determined that both lesion size and location influence the risk of engagement with the posterior glenoid. They offered the gamma angle as a potentially useful measurement on axial imaging to help determine the likelihood of engagement and recurrent instability. They would later build upon this concept in the setting of bipolar bone loss, proposing the “gamma angle concept” to predict the risk of engagement based on the size and location of humeral head bone loss and glenoid bone loss ( Fig. 36.8 ). While useful, these findings have yet to be clinically validated. Additionally, this algorithm does not guide specific surgical indications given that engagement can be negated by extending the glenoid width (posterior bone block procedure), limiting motion by filling the humeral head defect (lesser tuberosity transfer or subscapularis remplissage), or extending the humeral head articular surface (humeral head bone grafting). Therefore the indication for humeral head bone grafting remains elusive. It does, however, afford the advantage of restoring joint congruity with an osteoarticular surface compared with lesser tuberosity transfer or subscapularis remplissage.