The Hill-Sachs lesion is a well-known entity that threatens recurrent instability, but the treatment options are multiple and the surgical indications remain undefined. The evidence for each operative technique is limited to retrospective reviews and small case series without controls. The decision of which technique to use resides with the surgeon. Older, osteopenic patients, especially those with underlying arthritis and large defects, should be managed with complete humeral resurfacing. Humeralplasty is best used in younger patients with good quality bone in an acute setting with small- to moderate-sized bone defects. Partial resurfacing and remplissage are best used with small to moderate lesions, and both require further study. Allograft humeral reconstruction is an established technique for patients with moderate to large defects, and is best applied to nonosteopenic bone. Surgeons must be able to recognize the presence of humeral bone loss via specialized radiographs or cross-sectional imaging and understand its implications. The techniques to manage humeral bone loss are evolving and further biomechanical and clinical studies are required to define the indications and treatment algorithms.
Traumatic unidirectional glenohumeral joint instability is common and typically occurs in conjunction with labral or capsular injury; however, it can also be associated with bony lesions. Successful treatment requires recognition of all aspects of simple and complex instability to determine the most appropriate treatment algorithm. Management of glenohumeral instability includes nonoperative rehabilitation, soft-tissue repair or reconstruction, osseous reconstruction, or prosthetic replacement. This review outlines treatment options to restore or reconstruct osseous anatomy and shoulder stability in patients with humeral head bone defects.
In the nineteenth century, Flower and Broca and Hartmann described a posterior superior humeral head defect sustained after glenohumeral joint dislocation. In 1941, Hill and Sachs further described the eponymous lesion of humeral articular impaction that resulted from contact with the dense cortical bone of the anterior inferior glenoid rim during anterior subcoracoid dislocation. Palmer and Widén and Burkhart and DeBeer described an engaging Hill-Sachs lesion that occurs when the glenoid falls into the humeral head defect. To truly engage, the long axis of the humeral head defect must be parallel to the anterior glenoid rim with the arm in a position of abduction and external rotation. Engagement outside of this functional arc, with the arm at the side for example, is termed nonfunctional engagement or functionally nonengaging. If the humeral defect is not parallel to the glenoid rim, it may provide a sensation of subluxation when the lesion rolls over the anterior glenoid, but the rim will not fall into the defect and lever the humeral head from the glenoid socket.
Anterior inferior glenoid bone loss and engaging Hill-Sachs lesions have been implicated as factors leading to higher failure rates following soft-tissue stabilization procedures. Although osseous lesions are recognized factors associated with shoulder instability, there is little information describing surgical indications and optimal techniques to restore osseous anatomy and joint stability.
Epidemiology
Humeral head impression fractures are found in 65% to 71% of initial glenohumeral dislocations, and in up to 100% of patients with recurrent instability. A correlation between the presence of humeral head lesions and bony Bankart lesions exists and increases with recurrent instability. In a study of 61 patients, Widjaja and colleagues found that after a first dislocation, 64% of patients with a bony Bankart lesion also had a Hill-Sachs, and 70% of those with a Hill-Sachs lesion also had a Bankart lesion. Seventy-nine percent of patients with recurrent instability and a bony Bankart lesion also had a Hill-Sachs, and 81% of those with a Hill-Sachs had a bony Bankart lesion.
Imaging and anatomy
Hill and Sachs attributed the humeral impression fracture to the impact of the posterior humeral head onto the dense cortical bone of the anterior glenoid rim. They described the compression of softer cancellous bone of the humeral head as a radiographically discernable line of increased radiodensity along the medial wall of the lesion. The best technique for visualizing the defect has been debated. Reports have evaluated the integrity of the posterolateral humeral head with an internally rotated anteroposterior radiograph of the shoulder, an apical oblique view of Garth and colleagues, a Stryker notch view, a modified Didiee view, and through use of orthographic projection as described by Ito and colleagues. The Ito technique produces an undistorted image of the posterolateral notch by positioning the patient supine with the arm in 135° of flexion and 15° of internal rotation with the cassette directly under the shoulder joint and the central x-ray beam angled vertically through the humeral head.
The size and location of the lesion are important factors to consider when assessing the degree of instability and the risk of engagement of the defect on the glenoid rim during humeral abduction and external rotation. It is theorized that the size of a lesion is related to the force of the injury and the extent of joint laxity of the shoulder. A deep defect is more common following 2 or more dislocations, whereas patients with recurrent subluxations and apprehension following a dislocation are more likely to have a shallow lesion. A possible explanation for this observation is that greater force is required to dislocate a stable shoulder and a larger defect in the humerus results. However, less force is required to precipitate instability in individuals with ligament laxity and less impaction of the humeral head occurs.
Cross-sectional imaging is routinely used to evaluate humeral bone loss. Two-dimensional axial, sagittal, and coronal oblique reconstructions help to quantify the size and location of a defect, especially if it seems to involve greater than 20% of the humeral head. In a computerized tomography (CT) study, Saito and colleagues reported that Hill-Sachs lesions exist from 0 to 24 mm from the top of the humeral head, oriented from 6:46 cephalad to 8:56 caudal on a clock face with 12:00 defined as the intertubercular sulcus. The caudal extent is not to be mistaken for the bare spot or natural humeral groove that is regularly found between 19 and 21 mm from the top of the humeral head in the same location. Yamamoto and colleagues proposed the glenoid track as the zone of glenoid contact on the posterior humeral head from inferiormedial to superiorlateral, and postulated that a Hill-Sachs lesion extending medial to this area would likely engage the glenoid rim and risk recurrent dislocation. They defined the glenoid track as the region on the humeral head that begins at the rotator cuff insertion site on the greater tuberosity and extends medially on the humeral head for a distance equal to 84% of the glenoid width. Any anterior glenoid bone loss will reduce the glenoid width, increase the likelihood that the humeral defect is located medial to the glenoid track, and increase the risk of engagement and redislocation. When the humeral lesion is located medial to the glenoid track, Bankart repair alone may be insufficient to prevent lesion re-engagement. Thus, factors affecting the influence of a Hill-Sachs lesion on glenohumeral stability include the size, location, and orientation of the humeral defect but also the amount of anterior glenoid bone loss.
Miniaci and Gish recommend three-dimensional reconstructions of the bony anatomy to better map the size, location, and especially the orientation of humeral bone loss. He notes that the humeral defects are often oblique to the axial plane and thus not well represented and often underappreciated with two-dimensional imaging.
Ultrasound has also been successfully used to image Hill-Sachs lesion. Cicak and colleagues prospectively showed sonography to be 96% sensitive, 100% specific, and 97% accurate in diagnosing Hill-Sachs lesions compared with surgical findings in 61 patients with average lesions 19.2 mm long, 16.0 mm wide, and 4.1 mm deep. However, the diagnostic value of ultrasound may be lessened by the challenges associated with accurately identifying the lesion location and orientation.
Magnetic resonance imaging (MRI) is often used as an alternative to CT to assess osseous deformity when additional soft-tissue pathologies such as rotator cuff injuries, chondral defects, labral tears, and capsular laxity are suspected.
Classification
Rowe and colleagues categorized humeral lesions according to size, as follows: mild, 2×0.3 cm; moderate, 4×0.5 cm; severe, 4×1 cm or larger. In their series of solely Bankart repairs, recurrence rates were 0/30 (0%) for mild lesions, 3/64 (4.7%) for moderate, and 1/16 (6%) for severe. From these data, Rowe and colleagues recommended that severe lesions be addressed surgically because those shoulders were at risk for recurrent dislocation.
The size of the defect can also be calculated as a percentage loss of articular length relative to the entire humeral articular surface. General guidelines for surgical treatment based on percentage of articular involvement have been developed based on these criteria.
Indications
The indications for humeral head reconstruction are controversial. Kropf and Sekiya proposed that recreating normal humeral articular length would reduce tension on anterior capsulolabral structures and protect a Bankart repair. However, many consider humeral reconstruction secondary or even superfluous to anterior inferior glenoid bone grafting for recurrent instability. Traditionally, Hill-Sachs lesions have been corrected only after failure of a primary soft-tissue stabilizing procedure. More recently, surgical reconstruction of a humeral defect is advocated as a primary procedure when patients experience ongoing symptomatic shoulder instability including painful apprehension, popping, or catching accompanied by a significant (>25% of the humeral head) engaging bone defect or a defect medial to the glenoid track. Patients at high risk for surgical failure, such as contact athletes or those with seizure disorders, may benefit from early primary humeral head reconstruction. More importantly, Grondin and Leith stress that the surgeon should be prepared to address glenoid and humeral bone loss when preoperative CT examination reveals more than 30% surface loss of each. Despite these recommendations, there have been no definitive studies defining the specific size, location, or orientation of a lesion that must be addressed surgically. Whether patients require surgical treatment of the humeral head alone, in conjunction with anterior reconstruction, or not at all for unidirectional anterior shoulder instability with associated humeral bone loss has yet to be clearly defined.
Surgical treatment options
Treatment options to reconstruct symptomatic Hill-Sachs lesions can be broadly categorized into 4 groups. The defect may be disimpacted, filled with soft tissue or allograft bone, rotated away, or partially or completely replaced.
Humeralplasty/Disimpaction
The technique for acutely elevating an impaction fracture without significantly altering normal anatomic structures has been termed humeralplasty. In a cadaveric study, Kazel and colleagues described a technique to reduce the volume of freshly created humeral defects using curved bone tamps inserted retrograde through a distal cortical window. One year later, Re and colleagues reported a humeralplasty technique performed on 4 patients in whom the lesion was localized using an anterior cruciate ligament (ACL) drill guide, elevated with retrograde tamping and filled with cancellous bone graft. Three patients were also treated with anterior capsulolabral reconstruction, and the fourth patient underwent a concomitant Latarjet procedure. There were no recurrences or other complications at 1 year follow-up. Although Re and colleagues did not specifically comment on the acuity of their cohort, humeralplasty is likely most reproducible and beneficial following an acute injury. An advantage of this technique is the ability to restore the normal proximal humeral geometry without internal fixation or extensive surgical alteration of normal anatomy ( Fig. 1 ). There is scant information on the indications and outcomes of this technique, but general guidelines for humeralplasty include defects that are less than 3 weeks old and involve less than 45% of the humeral articular surface.
Remplissage
Remplissage, which means filling in French, refers to the insetting of the infraspinatus tendon and posterior capsule into the Hill-Sachs lesion. The lesion is filled with soft tissue and converted to an extra-articular lesion incapable of engaging with the glenoid. The procedure initially involves preparation and repair of the associated anterior labral lesion. To conduct the remplissage procedure, the Hill-Sachs lesion is viewed with the camera placed through an anterior portal. Superior and inferior suture anchors, inserted percutaneously or through a posterior cannula, are placed into the bony defect. A sharp penetrating device is used to create horizontal mattress sutures through the infraspinatus and joint capsule. These sutures are then tied in the subdeltoid space, which secures the infraspinatus and posterior capsule to the Hill-Sachs lesion, thereby filling the bony defect. To facilitate suture tying, the subdeltoid space may be cleared before suture passage ( Fig. 2 ).
Related posts:
Arthroscopic Management of Anterior Instability: Pearls, Pitfalls, and Lessons Learned
Arthroscopic Bankart-Bristow-Latarjet (2B3) Procedure: How to Do It and Tricks To Make it Easier and Safe
Management of Failed Instability Surgery: How to Get It Right the Next Time
Evaluation of Glenohumeral Instability
Management of Failed Instability Surgery: How to Get It Right the Next Time
Humeral Head Bone Defects: Remplissage, Allograft, and Arthroplasty
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
